Computer Simulation of a Large-Scale Signalling …epubs.surrey.ac.uk/841058/1/Thesis Copy 10th March.pdfComputer Simulation of a Large-Scale Signalling Network Regulating Translation

Computer Simulation of a Large-Scale

Signalling Network Regulating Translation

Initiation in the Mammalian Cell

By

David James Taylor

Submitted in Partial Fulfilment of

the Requirements for the Degree of

Doctor of Philosophy

January 2017

School of Biosciences & Medicine

Faculty of Health & Medical Sciences

University of Surrey

i

Declaration of Originality

I confirm that the submitted work is my own work and that I have clearly identified and fully

acknowledged all material that is entitled to be attributed to others (whether published or

unpublished) using the referencing system set out in the programme handbook. I agree

that the University may submit my work to means of checking this, such as the plagiarism

detection service Turnitin® UK. I confirm that I understand that assessed work that has

been shown to have been plagiarised will be penalised.

ii

Abstract

The process of translation initiation in mammalian systems is complex and not fully

understood. It is regulated by an intricate network of signalling pathways and is a significant

energetic burden to the cell. Although models of initiation are available for yeast, to date,

such models do not include the regulation of this process, nor do they exist for mammalian

systems. Existing literature was used to reconstruct the process of translation initiation and

the regulatory signalling networks in the Petri Net formalism within the software Snoopy.

The final version of the model was altered to incorporate the effects of Murine Norovirus.

The model was converted to a binary form and the software QSSPN was used to run

Gillespie algorithm-based stochastic simulations. The predictive power of the model was

established by incorporating commonly used chemical inhibitors. Using the Matthews’

Correlation Coefficient, a quantitative measure of predictive power was established by

comparing the model behaviour to the effects of each inhibitor recorded in existing

literature. A qualitative model containing 584 reactions was constructed. The predictive

power of the model was raised to MCC = 0.4558 through a series of refinements. Two

predicted behaviours, an increase in eIF4E phosphorylation and a reduction of AKT

phosphorylation both in response to Rapamycin, were validated with the Immunoblotting

techniques, Western Blotting and Human Phospho-MAPK Arrays, in the murine

monocyte/macrophage RAW 264.7 cell line. The model incorporating the effects of Murine

Norovirus infection generated five testable predictions. Of these, four were verified with

the Human Phospho-MAPK Arrays. The model presented here demonstrates the value of

generating large-scale models using the binary model formalism and performing simulations

with QSSPN. The model of the regulation of translation initiation has shown that it is

capable of generating experimentally verifiable predictions. Furthermore, the incorporation

of viral effects demonstrates that the model has a range of potential future uses.

iii

Acknowledgements

For funding my work, I need to thank the University of Surrey.

I would like to thank my supervisors, Prof Andrzej Kierzek, Dr Nicolas Locker and Prof Lisa

Roberts for their help and support over the course of this project. This work would not have

been possible without their guidance and patience over the last three years. It has been a

pleasure to work with all of you.

I also need to thank everyone who has helped me over the last three years (Dr Elizabeth

Royall, Dr Nicole Doyle, Dr Margaret Carter, Dr Belinda Hall, Salmah Zaini, Majid Al-Sailawi,

Dr Melvin Leteane, Dr Nargs Alfagih, Mariam Sulaiman, Nefeli Karataraki, Joy Ogbechi, Dr

Adeola Fowotade, Katja Ebert-Keel, Dr Valerie Odon, Dr Manuel Salvador de Lara, Joe

Holley, Dr Juliet Jones, Dr Huihai Wu and Dr Mehdi Khoury to name but a few). Your

friendship and support has made the last three years so much easier.

I must also thank Dr Ben Neuman (University of Reading) for inspiring me to pursue a career

in molecular biology. Secondly, thank you to Dr Ditte Hobbs (University of Reading) for her

friendship and help during my undergraduate degree. Additional thanks must go to Dr

Alessia Ruggieri (Universität Heidelberg) for a very thought provoking conversation that

made me re-evaluate some of my positions on the experimental work undertaken in this

project.

I must thank my two best friends, Matthew Fletcher and Anthony Kendall, for all of the good

times we have had! You guys have always managed to help me make the best of even the

worst situation.

iv

My mum, Patricia, has always supported and encouraged me. Thank you for never doubting

me. You have always been there for me and I can never thank you enough. Although they

are no longer here, I must thank my Nan, Jean, and my Granddad, Joe. I miss you both very

much and hope that you would both be proud of me. The three of you made me into the

person I am today and I owe everything I achieve to you all.

v

Table of Contents

Declaration of Originality i

Abstract ii

Acknowledgements iii

Table of Contents v

List of Figures xiii

List of Tables xvii

List of Equations xviii

List of Abbreviations xx

List of Models & Variants xxviii

Chapter 1 – Introduction

Section 1.1 – Cellular Translation Initiation 2

Section 1.1.1 – Post Termination Ribosomal Complex Recycling 4

Section 1.1.2 – Ternary Complex Formation 5

Section 1.1.3 – eIF4F Components & mRNA Circularisation 6

Section 1.1.4 – Start Codon Recognition 8

Section 1.1.5 – 60S Subunit Joining 9

Section 1.2 – Key Points of Regulation & Upstream Signalling Pathways 10

Section 1.2.1 – eIF4E-Binding Proteins & The mTORC1 Signalling Pathway 12

Section 1.2.2 – eIF4E Phosphorylation and The Role of ERK1/2 & p38 20

Section 1.2.3 – Ribosomal Protein S6 Kinases 27

Section 1.2.4 – Phosphorylation of Eukaryotic Initiation Factor 2α 30

vi

Section 1.2.5 – Emerging Roles for JNK1 in the Regulation of Translation Initiation 32

Section 1.3 – Overview of the Viruses Modelled in this Project and Viral Strategies

Used to Alter Host Cell Translation Initiation

33

Section 1.3.1 – General Features & Epidemiology of Viruses Incorporated in the

Model

33

Section 1.3.1.1 – Arenaviruses & Junin Virus 34

Section 1.3.1.2 – Caliciviruses & Murine Norovirus 35

Section 1.3.1.3 – Hantaviruses & Andes Virus 37

Section 1.3.2 – General Overview of Viral Effects on Translational Regulation 38

Section 1.3.2.1 – Internal Ribosome Entry Site (IRES)-Mediated Translation & Host

mRNA Degradation

38

Section 1.3.2.2 - VPg Proteins 40

Section 1.3.2.3 – Host Factor Degradation & Viral Mimics of Host Factors 41

Section 1.3.2.4 – Alteration to Cell Signalling Pathways 41

Section 1.4 – Background to Systems Biology Techniques 42

Section 1.4.1 – Exact Stochastic Simulations with the Gillespie Algorithm 43

Section 1.4.2 – Ordinary Differential Equations (ODEs) 49

Section 1.4.3 – Flux Balance Analysis 51

Section 1.4.4 – The Petri Net Formalism 54

Section 1.4.5 – The Quasi-Steady State Petri Net Algorithm 63

Section 1.4.6 – Translation of Petri Net Terminology for a Molecular Biologist 71

Section 1.5 – Aims & Objectives and Novel Contributions to Knowledge 72

Section 1.5.1 – Aims & Objectives 72

vii

Section 1.5.2 – Novel Contributions to Knowledge 74

Chapter 2 – Model Creation & Refining

Section 2.1 – Materials & Methods 77

Section 2.1.1 – Qualitative Petri Nets 77

Section 2.1.2 – The Binary Model Formalism 78

Section 2.1.3 – Model Construction 85

Section 2.1.3.1 – General Model Construction Process 85

Section 2.1.3.2 – Specific Examples 98

Section 2.1.3.3 – Creation of the Benchmark Datasets 102

Section 2.1.4 – Simulation Method 105

Section 2.1.4.1 – Monte Carlo Simulations 105

Section 2.1.4.2 – Evaluation of the Predictive Power of the Model 107

Section 2.1.4.3 – Synchronous Simulation Methods 111

Section 2.1.4.3.1 - Case Study Identifying Differences in the Asynchronous

Simulation Approach & the Synchronous Methods

113

Section 2.1.5 – Investigation of the Interaction Between eIF4E Phosphorylation and

mTORC1 Inhibition

121

Section 2.2 – Results 122

Section 2.2.1 – Binary Model of LPS Stimulation 122

Section 2.2.1.1 – Updated Benchmark Dataset 127

Section 2.2.2 – Re-Evaluation of the Binary Model 128

Section 2.2.3 – Mechanism by Which mTORC1 Inhibition Affects eIF4E

Phosphorylation

135

viii

Section 2.2.4 – Evaluation of the Predictive Power of the Final Model 142

Section 2.3 – Discussion 151

Section 2.3.1 – Mechanism of Increasing eIF4E Phosphorylation in Response to

mTORC1 Inhibition

151

Section 2.3.2 – Predictive Power of the Final Version of the Model 152

Section 2.3.3 – Problems Encountered & Troubleshooting 155

Section 2.3.4 – Concluding Remarks 155

Chapter 3 – Experimental Validation of Model Predictions

Section 3.1 – Introduction 159

Section 3.1.1 – Rapamycin 159

Section 3.1.2 – eIF4E Phosphorylation During mTORC1 Inhibition 160

Section 3.1.3 – Interplay Between mTORC1 & AKT 161

Section 3.2 – Materials & Methods 167

Section 3.2.1 – Growth & Passage of RAW 264.7 Murine Macrophage Cells 167

Section 3.2.2 – Determination of Rapamycin Cytotoxicity 167

Section 3.2.3 – Cell Harvesting & Bicinchoninic Acid (BCA) Assay 169

Section 3.2.4 – Human Phospho-MAPK Arrays 170

Section 3.2.5 – eIF4E/Phosphorylated eIF4E & AKT/Phosphorylated AKT Western

Blot Analysis

170

Section 3.2.6 – Re-Evaluation of the Model Based Upon the Results of the Human

Phospho-MAPK Arrays

173

Section 3.2.7 – Statistical Analysis 174

ix


Section 3.3.1 – Cytotoxicity of Rapamycin 175

Section 3.3.2 – Evaluation of the Human Phospho-MAPK Arrays 177

Section 3.3.3 – Alterations to eIF4E Phosphorylation Levels 181

Section 3.3.4 – Confirmation of the Effect Rapamycin has Upon AKT1

Phosphorylation

183

Section 3.3.5 – Comparison of Model Results to Array Data 185


Section 3.4.1 – Validation of the eIF4E Phosphorylation Prediction 187

Section 3.4.2 – Validation of the AKT Phosphorylation Prediction 191


Section 3.4.4 – Potential Future Work 196


Chapter 4 – Modelling the Effects of Viral Infection


Section 4.2 – Model Changes & Simulation Methods 202

Section 4.2.1 – Junin Virus Model 202

Section 4.2.2 – Murine Norovirus Model 204

Section 4.2.3 – Andes Virus Model 205

Section 4.2.4 – Experimental Validation of the Murine Norovirus Model 207


Section 4.3.1 – Simulation of the Junin Virus Models 208

Section 4.3.2 – Simulation of the Murine Norovirus Models 211

x

Section 4.3.2.1 – Experimental Validation of the Hypotheses Resulting from

Simulations of the Murine Norovirus Model

214

Section 4.3.3 – Simulation of the Andes Virus Models 217





Chapter 5 – Integration of the Translational Control Model with a Genome Scale Metabolic

Network of a RAW 264.7 Cell


Section 5.1.1 – Linking mTORC1 via S6K1 to the Regulation of Metabolism 228

Section 5.2 – Methods 232

Section 5.2.1 – Proposed New Approaches 232


Section 5.3.1 – Current, Asynchronous, QSSPN Simulation Method 234

Section 5.3.2 – Proposed New Methods 239

Section 5.3.2.1 – Ordinary Differential Equation Method 239

Section 5.3.2.2 – Synchronous Method 243




Section 5.4.2.1 – Proposed Novel Modelling Strategies 245

Section 5.4.2.2 - Integration of the GSMN with the Translational Regulation Model 246

xi

Chapter 6 – Conclusion


Section 6.2 – Model Construction & Predictive Power Evaluation 250

Section 6.3 – Experimental Validation 254

Section 6.4 – Mechanism for Rapamycin-Induced Increases in eIF4E

Phosphorylation

256

Section 6.5 – Models of Viral Infection 257

Section 6.6 – Integration of the Translational Regulation Model with a GSMN 261

Section 6.7 – Novel Modelling Strategies 262

Section 6.8 – General Conclusion 264

References 265

Appendices 349

Supplementary Information 1 - References for the Construction of the Literature

Datasets

353

Supplementary Information 1a - Benchmark Dataset References 354

Supplementary Information 1b - Benchmark Dataset References for the Binary Model with

Lipopolysaccharide (LPS Benchmark)

383

Supplementary Information 2 - Dataset against Which the Binary Model was

Tested (Final Benchmark Dataset)

396

Supplementary Information 3 - Dataset Against Which the Model Integrated with

Metabolism was Tested (Metabolic Benchmark Dataset)

411

Supplementary Information 4 – Ingredients List for Solutions Produced In-House 415

xii

The Files Below are Saved to the USB Drive Enclosed

Supplementary Information E1 – Standardization of the Nomenclature of Model

Components

Supplementary Information E2 - List of Model Transitions & References in the

Binary Model

Supplementary Information E3 – Final Version of the Binary Model

Supplementary Information E4 – MNV Model

Supplementary Information E5 – LPS Benchmark Dataset

Supplementary Information E6 – Revised Original Benchmark Dataset

Supplementary Information E7 - Frequency Data Generated by Reachfq with Associated

99.99% Binomial Confidence Intervals

xiii

List of Figures


Figure 1.1a – Overview of Mammalian Translation Initiation 3

Figure 1.2a – Position of Translation Initiation within the Signalling Network 11

Figure 1.2.1a – Schematic of the mTORC1 Pathway & the Downstream Components

Regulating Translation Initiation

14

Figure 1.2.1b – Summary of the Role of AKT1 in the Activation of mTORC1 16

Figure 1.2.1c – Regulation of Translation Initiation by the Sequestration of eIF4F

Components

19

Figure 1.2.2a – Signalling Cascade Resulting in the Activation of ERK1/2 22

Figure 1.2.2b – Effects of ERK1/2 Signalling on Translation Initiation 24

Figure 1.2.2c – The Activation of p38α & The Effects of this Kinase on Translation and

Gene Expression

26

Figure 1.2.3a – Overview of Pathways Resulting in the Activation of the Ribosomal S6

Protein Kinases

29

Figure 1.2.4a – Cellular Stresses Leading to the Inhibition of Translation through the

Phosphorylation of eIF2α

31

Figure 1.4.1a – Schematic View of Propensity Functions 47

Figure 1.4.4a – Example of a Simple Petri Net 57

Figure 1.4.4b – Extended Petri Net Notation 60

Figure 1.4.5a – Summation of the QSSPN Algorithm 70


Figure 2.1.2a – Overview of the Binary Model Formalism 82

Figure 2.1.3.1a – Petri Net Representation of Translation Initiation 88

Figure 2.1.3.1b – Petri Net Representation of the ERK1/2 Signalling Pathway 90

xiv

Figure 2.1.3.1c – Petri Net Representation of the p38α Signalling Pathway 93

Figure 2.1.3.1d – Petri Net Representation of the JNK1 Signalling Pathway 95

Figure 2.1.3.1e – Petri Net Representation of the PI3K and mTORC1 Signalling Pathways 97

Figure 2.1.3.2 – Example of Model Construction Process 101

Figure 2.1.4.3.1a - Schematic of the Example Model of eIF4E Phosphorylation and

Availability

114

Figure 2.1.4.3.1b - Example Trajectory from the Simple Model Generated Using the

Asynchronous Simulation Approach

116

Figure 2.1.4.3.1c - Example Trajectory from the Simple Model Generated Using the

Synchronous Simulation Approach

118

Figure 2.2.1a – Example of the LPS Model Comparing the Effects of Inhibitors to a Wild

Type Model

124

Figure 2.2.1b – Overview of the Predictive Power of the Model Incorporating the Effects

of Lipopolysaccharide Signalling

126

Figure 2.2.2a – Overall Predictive Power of the Model Through the Four Stages of

Refinement

131

Figure 2.2.2b – The Model Refinement Workflow 132

Figure 2.2.2c – Evaluation of the Predictive Power of the Fifth Iteration of the Model 134

Figure 2.2.4a – Confirmation of the Binary Nature of the Model Formalism 144

Figure 2.2.4b – Components of the Model Common to Human Phospho-MAPK Array 146

Figure 2.2.4c – Evaluation of the Predictive Power of the Final Version of the Model 148

Figure 2.2.4d – Stability Analysis of the Final Model

150

xv


Figure 3.1.3a – Overview of the Feedback Mechanism between S6K1 and the Insulin

Receptor Substrate 1 (IRS-1)

163

Figure 3.1.3b – Overview of the Feedback Mechanism that Exists between S6K1 and

mTORC2

165

Figure 3.3.1a – Evaluation of the Effects of Three Hour Treatments with Differing

Concentrations of Rapamycin (Rapa) or 1% DMSO on the Murine Macrophage RAW

264.7 Cells

176

Figure 3.3.2a – Evaluation of the Broad Early Effects of 10μM Rapamycin Treatment to

the Murine Macrophage RAW 264.7 Cells

178

Figure 3.3.3a – Rapamycin-Induced Increase in Phosphorylation of eIF4E. 182

Figure 3.3.4a – Effect of Rapamycin on the Levels of AKT1 Phosphorylation (Ser-473) 184

Figure 3.3.5a – Comparison of Array Data with the Behaviour of the Binary Model 186

Figure 3.4.1a – Hypothesis Linking mTORC1 to the MNK-Dependent Stress Response 190

Chapter 4 – Modelling the Effects of Viral Infection

Figure 4.3.1a – Modelling the Effects of JUNV Infection 210

Figure 4.3.2a – Modelling the Effects of MNV Infection 213

Figure 4.3.2.1a – Effect MNV Infection has Upon CREB Phosphorylation 216

Figure 4.3.3a - Modelling the Effects of ANDV Infection

218

xvi

Chapter 5 - Integration of the Translational Control Model with a Genome Scale

Metabolic Network of a RAW 264.7 Cell

Figure 5.1.1a – Overview of the Role Played by S6K1 in Regulating Multiple Metabolic

Pathways

231

Figure 5.3.1a – Predictive Power of the Model of Translational Regulation Integrated

with a GSMN of RAW 264.7 Cell Metabolism and Simulated with the Current,

Asynchronous Method

237

Figure 5.3.2.1a – Summary of the Results of the Ordinary Differential Equation

Simulation Method

240

Figure 5.3.2.1b – Overview of the Predictive Power of the Model when Simulated with

the Ordinary Differential Equation Method

242


Figure 6.5a – Schematic of the Effects of MNV-Induced ERK1/2 and p38 Activation 259

xvii

List of Tables


Table 2.1.3.3a - Chemical Inhibitors used in the Model 104


Table 3.2.5a – Summary of the Antibody Solutions Used in Testing eIF4E and AKT

Phosphorylation

172

xviii

List of Equations

Chapter 1 - Introduction

Equation 1.4.1a 45

Equation 1.4.1b 45

Equation 1.4.1c 45

Equation 1.4.1d 48

Equation 1.4.1e 48

Equation 1.4.2a 50

Equation 1.4.3a 51

Equation 1.4.3b 52

Equation 1.4.3c 52

Equation 1.4.3d 52

Equation 1.4.4a 56

Equation 1.4.4b 56

Equation 1.4.4c 58

Equation 1.4.4d 58

Equation 1.4.4e 58

Equation 1.4.4f 58

Equation 1.4.4g 58

Equation 1.4.4h 58

Equation 1.4.4i 58

Equation 1.4.4j 58

Equation 1.4.4k 58

xix

Equation 1.4.4l 58

Equation 1.4.4m 58

Equation 1.4.4n 58

Equation 1.4.5a 65

Equation 1.4.5b 66

Equation 1.4.5c 67

Equation 1.4.5d 68


Equation 2.1.4.2a 109

Equation 2.1.4.2b 110

Equation 2.1.4.2c 110

xx

List of Abbreviations

3' UTR 3' Untranslated Region

40S 40 Svedberg Ribosomal Subunit

43S 43 Svedberg Complex

48S 48 Svedberg Complex

4E-BP1 Eukaryotic Initiation Factor 4E-Binding Protein 1



4E-BPs Eukaryotic Initiation Factor 4E-Binding Proteins

60S 60 Svedberg Ribosomal Subunit

80S 80 Svedberg Ribosome

ABCE1 ATP-Binding Cassette 1

AC Approximate Coefficient

Acetyl-CoA Acetyl-Coenzyme A

ACP Average Conditional Probability

ADP Adenosine Diphosphate

AGC

Protein Kinase A, Cyclic Guanosine Monophosphate-Dependent Protein

Kinase and Protein Kinase C

AHF Argentine Haemorrhagic Fever

AKT1 Protein Kinase B

ANDV Andes Virus

ANOVA Analysis of Variance

ATF-2 Cyclic AMP-Dependent Transcription Factor

xxi

ATP Adenosine Trisphosphate

AUG Adenosine-Uracil-Guanosine

BCA Bicinchoninic Acid

CAD

Carbamoyl-Phosphate Synthetase 2, Aspartate Transcarbamylase and

Dihydroorotase

Cas CRISPR-Associated Protein

CREB Cyclic AMP-Responsive Element-Binding Protein

CRISPR Clustered, Regularly Interspaced, Short Palindromic Repeats

CUG Cytosine-Uracil=Guanosine

DEAD Asp-Glu-Ala-Asp

DEPTOR DEP-Domain-Containing mTOR-Interacting Protein

DMSO Dimethyl Sulphoxide

DT Dynamic Transitions

EBOV Ebola Virus

EBV Epstein-Barr Virus

eIF Eukaryotic Initiation Factor

eIF1 Eukaryotic Initiation Factor 1

eIF1A Eukaryotic Initiation Factor 1A


eIF2B Eukaryotic Initiation Factor 2B

eIF2α Eukaryotic Initiation Factor 2, Subunit α

eIF2β Eukaryotic Initiation Factor 2, Subunit β

eIF2βγ Eukaryotic Initiation Factor 2, Complex of Subunits β and γ

xxii


eIF4A Eukaryotic Initiation Factor 4A

eIF4B Eukaryotic Initiation Factor 4B

eIF4E Eukaryotic Initiation Factor 4E

eIF4F Eukaryotic Initiation Factor 4F

eIF4G Eukaryotic Initiation Factor 4G

eIF4GI Eukaryotic Initiation Factor 4G, Isoform I

eIF4GII Eukaryotic Initiation Factor 4G, Isoform II

eIF4H Eukaryotic Initiation Factor 4H



eRF1 Eukaryotic Release Factor 1

eRF3 Eukaryotic Release Factor 2

ERK1/2 Extracellular Signal-Regulated Kinase 1/2

ERK5 Extracellular Signal-Regulated Kinase 5

FBA Flux Balance Analysis

FCV Feline Calicivirus

FKBP FK506-Binding Proteins

FKBP12 FK506-Binding Protein 12



FN False Negative

FP False Positive

xxiii

GAP GTPase Activating Protein

GCN2 General Control Nonderepressing 2

GDP Guanosine Diphosphate

GEF Guanine Nucleotide Exchange Factor

GPC Glycoprotein C-Terminus

GSK3 Glycogen Synthase Kinase 3

GSMN Genome-Scale Metabolic Network

GTP Guanosine Trisphosphate

HCMV Human Cytomegalovirus

HCPS Hantavirus Cardiopulmonary Syndrome

HIF-1α Hypoxia Inducible Factor 1α

HIV-1 Human Immunodeficiency Virus Type 1

hnRNP A1 Heterogenous Nuclear Ribonucleoprotein A1

hpi Hours Post Infection

HRI Haem-Regulated eIF2α Kinase

HSV Herpes Simplex Virus

HSV-1 Herpes Simplex Virus Type 1

HTLV-1 Human T-Cell Leukaemia Virus 1

HuNV Human Norovirus

IHC Immunohistochemistry

IRES Internal Ribosome Entry Site

IRS-1 Insulin Receptor Substrate 1

JNK c-Jun N-Terminal Kinase

xxiv

JUNV Junin Virus

kDa Kilo Daltons

LPS Lipopolysaccharide

MAPK Mitogen-Activated Protein Kinase

MCC Matthews' Correlation Coefficient

MEK MAPK/ERK Kinase

Meti-tRNA Methionine-Aminoacetylated Initiator tRNA

MK2 MAPK-Activated Protein Kinase-2

MKK3 Mitogen-Activated Protein Kinase Kinase 3

MKK6 Mitogen-Activated Protein Kinase Kinase 6

MKP-7 MAPK Phosphatase 7

mLST8 Mammalian Lethal with Sec13 Protein 8

MNK1 MAPK-Interacting Kinase 1

MNK2 MAPK-Interacting Kinase 2

MNV Murine Norovirus

MOI Multiplicity of Infection

mRNA Messenger Ribonucleic Acid

mSIN1 Mammalian Stress-Activated Protein Kinase Interacting Protein

MSK1 Mitogen- and Stress-Activated Kinase 1

MSK2 Ribosomal Protein S6 Kinase α-4

mTOR Mammalian Target of Rapamycin

mTORC1 Mammalian Target of Rapamycin Complex 1

mTORC2 Mammalian Target of Rapamycin Complex 2

xxv

ns Non-Significant

ODE Ordinary Differential Equation

PABP Poly(A) Binding Protein

PAM Protospacer Adjacent Motif

PBS Phosphate-Buffered Saline

PCR Polymerase Chain Reaction

PDCD4 Programmed Cell Death 4

PDK1 3-Phosphoinositide-Dependent Protein Kinase-1

PERK Double-Stranded RNA-Activated Protein Kinase-Like ER Kinase

PI3K Phosphoinositide-3-Kinase

PKC Protein Kinase C

PKCα Protein Kinase C, Isoform α

PKCβII Protein Kinase C, Isofrom β II

PKCδ Protein Kinase C, Isoform δ

PKCζ Protein Kinase C, Isoform ζ

PKCθ Protein Kinase C, Isoform θ

PKR Double-Stranded RNA-Dependent Protein Kinase

PMID PubMed Identifier

Poly(A) Poly Adenosine Tract

Poly(I:C) Polyinosinic-Polycytidylic Acid

Post-TC Post Termination Ribosomal Complex

PP2A Protein Phosphatase 2A

PP5 Protein Phosphatase 5

xxvi

PRAS40 Proline-Rich AKT Substrate 40kDa

PVDF Polyvinylidene Fluoride

QSSF Quasi-Steady State Fluxes

QSSPN Quasi-Steady State Petri Net

RAPTOR Regulatory-Associated Protein of mTOR

Rheb Ras-Homologue Enriched in Brain

Rictor Rapamycin-Insensitive Companion of mTOR

RNA Ribonucleic Acid

RSK1 p90 S6 Kinases, Isoform 1

RSK2 p90 S6 Kinase, Isoform 2

S6K1 p70 S6 Kinases, Isoform 1

SDS Sodium Dodecyl Sulphate

SDS-PAGE Sodium Dodecyl Sulphate-Polyacrylamide Gel Electrophoresis

Ser Serine

siRNA Small Interfering RNA

SLBP Stem Loop Binding Protein

SREBP1 Sterol-Regulatory Element Binding Protein 1

SREBP1/2c Proteolytic Processed form of SREBP1/2

SREBP2 Sterol-Regulatory Element Binding Protein 2

STAT3 Signal Transducer and Activator of Transcription 3

TAK1 Transforming Growth Factor-β-Activated Kinase 1

TC Ternary Complex

TCID50 50% Tissue Culture Infectious Dose

xxvii

TGF-β Transforming Growth Factor-β

Thr Threonine

TLR Toll-Like Receptor

TN True Negative

TP True Positive

Tpl2 Mitogen-Activated Protein Kinase Kinase Kinase 8

tRNA Transfer Ribonucleic Acid

Trp Tryptophan

TSC Tuberous Sclerosis Complex

TSC1 Hamartin

TSC2 Tuberin

TTP Tristetraprolin

Tyr Tyrosine

xxviii

List of Models and Variants

Case Study Model -

Comment

Petri Net Component Count This model contained only molecular interactions relating to eIF4E phosphorylation and availability. This information was taken from existing literature on mammalian translation initiation and the signalling pathways regulating this process. The interactions in this model were extracted from the large-scale model (see Final Version of the Binary Model)

Place/Molecular Species 70

Transitions/Reaction 65

Edges 132

Read Edges 50

Inhibitor Edges 43

Final Version of the Binary Model -

Comment

Petri Net Component Count This model contained information from existing literature on mammalian translation initiation and the signalling pathways regulating this process. Additionally, this model contained the model of ERK1/2 activation created by Schilling et al. (2009) and, which was treated as a binary model during the course of this project.



Edges 1236

Read Edges 515

Inhibitor Edges 545

Metabolic Model -

Comment

Petri Net Component Count This version of the Final model included information on the regulation of metabolism by the mTORC1 signalling pathway. In this regard, enzymes expression was added along with any post-translational modifications dependent upon mTORC1. Furthermore, these enzymes were also linked to the corresponding reactions in the FBA GSMN model of a RAW 264.7 murine macrophage cell line. This model was simulated with the Asynchronous method and the ODE method. This was used in Chapter 5 – Integration of the Translational Control Model with a Genome Scale Metabolic Network of a RAW 264.7 Cell.



Edges 1576

Read Edges 656

Inhibitor Edges 706

xxix

Junin Virus Model -

Comment

Petri Net Component Count Used in Chapter 4 – Modelling the Effects of Viral Infection, this variant of the Final model included extra Petri Net components to model the translational effects of JUNV infection and the effects viral factors have upon cellular signalling pathway components.



Edges 1301

Read Edges 532

Inhibitor Edges 557

Murine Norovirus Model -

Comment

Petri Net Component Count Used in Chapter 4 – Modelling the Effects of Viral Infection, this variant of the Final model was used to model MNV infection. Various Petri Net components were added to model the effects MNV has upon cellular signalling components. Furthermore, the alterations MNV induces in the cellular translational machinery were included. Such changes to the model were based upon existing literature, such as Royall et al. (2015).



Edges 1287

Read Edges 521

Inhibitor Edges 542

Andes Virus Model -

Comment

Petri Net Component Count Used in Chapter 4 – Modelling the Effects of Viral Infection, this variant of the Final model concerned the effects of ANDV infection. It was necessary to include additional Petri Net components to model how viral translation is carried out. Furthermore, the effects of ANDV on signalling factors was included.



Edges 1292

Read Edges 529

Inhibitor Edges 554

Page 1


Page 2

Section 1.1 – Cellular Translation Initiation

The process of translation initiation was central to the model produced during the course of

this project. As a consequence of this, a complete understanding of this process is

necessary.

The ability to synthesise nascent proteins is fundamental for all organisms. Translation is

the cytoplasmic process of forming complex polypeptides from amino acids using

messenger RNA (mRNA) transcripts to convey information encoded by genes within the

nucleus (Brenner et al., 1961). The process of translation can be divided into three key

stages: initiation, elongation and termination. A number of key signalling pathways act to

regulate these stages; of which initiation is the most tightly regulated. The importance of

this ‘energetically demanding process’ (Laxman et al., 2013) is highlighted by the presence

of over 30,000 protein encoding regions in the human and mouse genomes (Lander et al.,

2001; Waterston et al., 2002). Each of these genes, would in theory, need to be expressed

and translated. The process of translation, particularly in higher eukaryotic and mammalian

systems, is complex and not fully understood. The currently accepted view of translation

initiation is shown in Figure 1.1a.

Page 3

Figure 1.1a – Overview of Mammalian Translation Initiation. Adapted from Jackson et al. (2010),

this illustration gives a complete overview of the currently accepted mechanism of translation

initiation in mammalian cells.

Page 4

As can be seen in Figure 1.1a, the process of translation can be regarded as cyclic. The end

product of the termination stage is a Post-Termination Ribosomal Complex (Post-TC).

Recycling of this complex must occur to allow the loading of various eukaryotic initiation

factors (eIFs) onto the small 40 Svedberg (40S) ribosomal subunit to form the 43S complex

(Jackson et al., 2010). Prior to translation, mature mRNA is exported from the nucleus.

Such mRNA contains a 5’ methylguanosine cap and 3’ poly(A) tail with both 5’ and 3’

untranslated regions and an open reading frame (Sahin et al., 2014). Once the mRNA, and

associated eIFs, are loaded onto the the 43S complex, scanning of the mRNA to find a start

codon sequence within the 5’ region of the mRNA begins. This scanning process requires

the unwinding of the mRNA by the action of a helicase. Once this has been found, the 48S

complex is formed and this complex is committed to initiating translation at this site (Ibid).

The formation of the 80S ribosome, capable of carrying out translation, occurs through the

joining of the large, 60S ribosomal subunit (Jackson et al., 2010).

Section 1.1.1 – Post-Termination Ribosomal Complex Recycling

The first stage of initiation is the recycling of Post-TCs. These are formed by the release of

the nascent polypeptide and the action of the eukaryotic Release Factors (eRF) 1 and 2

(Alkalaeva et al., 2006). Although only eIF3 displayed any inherent activity towards carrying

out Post-TC recycling, and the subsequent formation of the 40S and 60S subunits, eIF1 and

eIF1A were required for efficient removal of deactetylated tRNA and mRNA (Pisarev et al.,

2007). The release of the nascent polypeptide from the 80S ribosome is mediated by the

actions of eRF1 and eRF3 (Alkalaeva et al., 2006) both of which are retained on the Post-TC

(Pisarev et al., 2007). However, it is possible that the retained eRF1 plays a later role in the

recycling of the Post-TC (Pisarev et al., 2010). The ATP-Binding Cassette, ABCE1, has been

Page 5

shown to mediate recycling of Post-TCs in a manner that requires the presence of eRF1

(Ibid).

Section 1.1.2 – Ternary Complex Formation

The Ternary Complex (TC) consists of an amino-acetylated tRNA charged with a Methionine

residue (Meti-tRNA) and eIF2 bound to Guanosine Diphosphate (GDP) and an inorganic

phosphate group (Algire et al., 2005). Prior to a new round of translation initiation, the GDP

molecule is exchanged for a Guanosine Trisphosphate (GTP) molecule through the

conserved action of eIF2B in both yeast and mammals (Cigan et al., 1993; Williams et al.,

2001).

Given that eIF2 has around a 400-fold greater affinity for GDP than GTP (Panniers et al.,

1988), eIF2B can only catalyse this exchange if GTP is in excess relative to GDP ((Dholakia &

Wahba, 1989; Oldfield & Proud, 1992; Nika et al., 2000 (As cited by Williams et al., 2001)).

eIF2B functions in the formation of the TC by promoting the release of eIF2 from the

complex with eIF5 (Jennings et al., 2013). The context of this complex between eIF2 and

eIF5 will be discussed in Section 1.1.5 – 60S Subunit Joining. Of the five eIF2B subunits, the

ε-subunit is required for the exchange of GDP for GTP (Fabian et al., 1997).

eIF2 consists of α, β and γ subunits (Hashem et al., 2013). The γ-subunit of eIF2 may be

responsible for interacting with both the Meti-tRNA and the 40S ribosomal subunit (Shin et

al., 2011). The γ-subunit of eIF2 links the TC to the small ribosomal subunit through eIF3

within the 43S Complex (Chaudhuri et al., 1997; Chaudhuri et al., 1999). The eIF2 β-subunit

mediates an interaction between eIF2 and eIF2B (Kimabll et al., 1998). The ability of eIF2B

to exchange GDP for GTP is partially dependent upon phosphorylation of eIF2β by Protein

Kinase A (Ibid). This interaction between eIF2β and eIF2B appears to be altered under

Page 6

conditions in which eIF2α is phosphorylated (Kimball et al., 1998; Krishnamoorthy et al.,

2001). The role of the α-subunit of eIF2, and the mechanism by which it is capable of

inhibiting translation initiation, will be discussed in Section 1.2.4 – Phosphorylation of

Eukaryotic Initiation Factor 2α.

Section 1.1.3 – eIF4F Components & mRNA Circularisation

The eIF4F complex is comprised of eIF4A, eIF4E and eIF4G and mediates the interaction

between the mRNA and the assembling ribosome. The components of the eIF4F complex

anchor mRNA to the 43S complex and allow mRNA scanning. eIF4F carries out the function

of binding mRNA through the cap binding protein, eIF4E, and the unwinding of mRNA during

scanning by the helicase, eIF4A (Jackson et al., 2010).

eIF4G is a scaffold protein of around 200kDa (Yan et al., 1992; Hernández et al., 1998). Two

functional isoforms of eIF4G, designated eIF4GI and eIF4GII, are found in human cells (Gradi

et al., 1998) and display 46% similarity at the amino acid level with with some regions of the

middle domain being completely conserved (Ibid). These two isoforms may be required for

targeting specific mRNAs for preferential expression under certain environmental conditions

(Gradi et al., 1998).

eIF4G recruits mRNA to the 48S complex through an interaction with eIF3 (Korneeva et al.,

2000; LeFebvre et al., 2006; Villa et al., 2013). In addition to multiple initiation factors,

eIF4G can interact with several additional proteins, including the Mitogen-Activated Protein

Kinase (MAPK)-Interacting Kinase (MNK) 1 (Pyronnet et al., 1999) and a protein required for

the translation of histone mRNAs, Stem Loop Binding Protein (SLBP)-Interacting Protein 1

(Cakmakci et al., 2007).

Page 7

A second component of the eIF4F complex is primarily responsible for scanning mRNA in

order to identify start codon sequences. This component is the DEAD (Asp-Glu-Ala-Asp)-Box

RNA helicase, eIF4A (Rogers Jr. et al., 1999; Rogers Jr. et al., 2001a). The helicase activity of

eIF4A is dependent upon ATP binding (Ray et al., 1985; Abramson et al., 1987; Blum et al.,

1992). However, binding of either ATP or mRNA to eIF4A, in the first instance, does not

preclude the other from interacting with the helicase (Lorsch & Herschlag, 1998a). The

conformational changes in eIF4A occurring upon an interaction with either ATP or mRNA

result in an increase in affinity for the other (Lorsch & Herschlag, 1998b). Furthermore, ADP

binding results in lower affinity binding to mRNA suggesting that a succession of ATP

hydrolysis events enable further mRNA unwinding (Ibid).

Two additional initiation factors, namely eIF4B (Abramson et al., 1987) and eIF4H (Richter et

al., 1999; Rogers Jr. et al., 2001b; Rozovsky et al., 2008) are required to achieve optimal

eIF4A activity. While a combination of both factors has been shown to increase the activity

of eIF4A by around 30% (Richter-Cook et al., 1998), the effects of eIF4B and eIF4H are not

simply additive with the increase being more dependent upon eIF4B (Özeş et al., 2011).

eIF4B, together with eIF4G, couples the unwinding of mRNA to ATP hydrolysis by

maintaining eIF4A in a state in which both ATP and mRNA are bound (Andreou &

Klostermeier, 2014). Furthermore, the amount of ATP required for unwinding mRNA is

markedly reduced in the presence of eIF4B (Harms et al., 2014).

The final component of eIF4F is the 5’ cap binding protein, eIF4E. This protein acts to

anchor the 5’ end of mRNA to the eIF4F complex and the 43S complex. As discussed later in

Section 1.2.1 – eIF4E-Binding Proteins & The mTORC1 Signalling Pathway, eIF4E availability

Page 8

is tightly regulated by the 4E-Binding Proteins (4E-BPs) which function to sequester eIF4E

and prevent an interaction with eIF4G (Marcotrigiano et al., 1999).

The mRNA 3’ Poly(A) tail interacts with the Poly(A) Binding Protein (PABP) (Sachs et al., 1986

(as cited by Tarun Jr. & Sachs, 1996); Görlach et al., 1994). PABP has also been shown to

interact with human and yeast eIF4G via the the N-terminal domain (Tarun Jr. & Sachs,

1996; Imataka et al., 1998; Kessler & Sachs, 1998; Wakiyama et al., 2000). Given these

interactions, it has been postulated that the mRNA, linked by eIF4E and PABP, and bridged

by eIF4G, at the 5’ and 3’ ends, respectively, becomes circularised (Wells et al., 1998).

Section 1.1.4 – Start Codon Recognition

The start codon refers to an AUG (Adenosine-Uracil-Guanosine) nucleotide sequence found

within the 5’ region of mRNA. Kozak (1987) established that in addition to this AUG

sequence, several other nucleotides, both upstream and downstream, are required to

ensure that this sequence is decoded by the 48S complex, only when it occurs in an

appropriate context. The optimal and conserved context consists of the sequence: AXXaugG

where X represents any nucleotide (Kozak, 1987; Kozak, 1997; Rogozin et al., 2001).

In addition, alternative initiation sequences, such as CUG (Cytosine-Uracil-Guanosine), are

commonly used. It has been shown that the use of ‘Kozak Sequences’ as initiation sites

maybe a low as a third of initiation sequences (Lee et al., 2012).

Following the eIF4A-mediated scanning, recognition of the mRNA start codon can occur.

This stage of translation initiation has not been well studied in mammalian systems and, as a

consequence, much of the summary here relies on work conducted in yeast. eIF1 and eIF1A

both act to ensure the recognition of a start codon (Saini et al., 2010) by maintaining the

Page 9

ability of the 48S complex to scan through mRNA for the start codon (Passmore et al., 2007;

Hussain et al., 2014). When a start codon is recognised, eIF1 is displaced from the codon-

anti-codon binding site and this converts the 48S complex to a state where scanning is no

longer possible (Hussain et al., 2014).

Consistent with this view, in mammalian systems eIF1, eIF1A and eIF5 interact through the

C-terminal domain of eIF5 and N-terminal domain of eIF1A (Luna et al., 2013). This

interaction prevents eIF5 from accessing eIF2 in advance of start codon recognition. Once

the Meti-tRNA and start codon interact, this interaction between eIF1A and eIF5 is

prevented and scanning ceases, allowing eIF1 to dissociate (Ibid).

Section 1.1.5 – 60S Subunit Joining

Upon recognition of the start codon, eIF5, a GTPase-Activating Protein (Paulin et al., 2001),

mediates eIF2-driven GTP hydrolysis to release the tRNA. It has been proposed that eIF5

interacts with eIF2β which allows the release of the inorganic phosphate molecule from eIF2

(Luna et al., 2013). eIF5B subsequently mediates the 60S subunit joining through a second

GTP hydrolysis (Pestova et al., 2000) and may be dependent upon eIF1A (Acker et al., 2006),

while the bound eIFs are released (Unbehaun et al., 2004).

To prevent re-association of the 40S and 60S subunits after recycling, eIF6 acts by binding to

the 60S subunit (Valenzuela et al., 1982; Raychaudhuri et al., 1984; Si et al., 1997; Wood et

al., 1999). The ability of eIF6 to prevent the formation of the 80S complex is regulated by

PKC β-mediated phosphorylation (Ceci et al., 2003). The yeast Elongation Factor-Like 1

protein may also regulate the interaction between eIF6 and the 60S subunit (Senger et al.,

2001).

Page 10

Section 1.2 – Key Points of Regulation & Upstream

Signalling Pathways

The regulation of initiation occurs through the coordinated and complex activity of a

number of cellular signalling pathways, including the Mitogen-Activated Protein Kinase

(MAPK) pathways (Extracellular Signal-Regulated Kinase (ERK) 1/2, p38, c-Jun N-Terminal

Kinase (JNK1) and ERK5), the Phosphoinositide-3-Kinase (PI3K) and the mammalian Target of

Rapamycin Complex 1 (mTORC1) pathways, as shown in Figure 1.2a.

Page 11

Figure 1.2a – Position of Translation Initiation within the Signalling Network. The cellular

signalling network is intimately involved with the regulation of translation initiation and while

the initiation phase of translation is key to the synthesis of nascent proteins it is only a small part

of the model produced during the course of this project, as shown by the area within the red

box. This signalling network must interact with the translational machinery at key points in

order to regulate the process of initiation. The aim of this project was to produce a qualitative

model of this system in which translation initiation is regulated by a signalling network.

Page 12

In order to accurately model the regulation of the initiation phase of translation it is

necessary to understand the points at which these signalling pathways converge on the

translational machinery. Furthermore, the mechanisms by which translation initiation is

affected by these pathways must also be known in order to correctly model this process. A

further level of complexity is added by the large degree of cross-talk and feedback

mechanisms existing in and between these pathways resulting in complex interaction

networks. This section will therefore focus upon introducing the key signalling pathways

and how they affect the translational machinery.

Section 1.2.1 – eIF4E-Binding Proteins & The mTORC1 Signalling Pathway

As introduced in Section 1.1.3 – eIF4F Components & mRNA Circularisation, the activity of

a key initiation factor, eIF4E, is regulated by the phosphorylation of 4E-BP1, which is

controlled by the mammalian Target of Rapamycin (mTOR) Complex 1 (mTORC1).

Two discrete mTOR complexes are found in mammalian cells, each with distinct biological

functions (Laplante & Sabatini, 2009). mTORC1 is composed of mTOR, Regulatory-

Associated Protein of mTOR (RAPTOR), DEP-Containing mTOR-Interacting Protein (DEPTOR),

mammalian Lethal with Sec13 Protein 8 (mLST8), and the Proline-Rich AKT1 Substrate of

40kDa (PRAS40) (Ibid). By comparison, mTOR Complex 2 (mTORC2) contains mTOR, the

Rapamycin-Insensitive Companion of mTOR (Rictor), mammalian Stress-Activated Protein

Kinase Interacting Protein (mSIN1), DEPTOR and mLST8 (Leplante & Sabatini, 2009). The

activation on mTORC1 is not fully understood but several discrete mechanisms have been

observed.

Upstream of mTORC1 lies the Phosphoinosite 3-Kinase (PI3K) pathway component Protein

Kinase B (AKT1) (summarised in Figure 1.2.1a). This Ser/Thr kinase is a member of the AGC

Page 13

(Protein Kinase A, Cyclic Guanosine Monophosphate-Dependent Protein Kinase and Protein

Kinase C) kinase family and expressed in three isoforms: AKT1, AKT2 & AKT3 (Hanks &

Hunter, 1995; Scheid & Woodgett, 2001; Zinda et al., 2001). AKT1 is phosphorylated and

activated in response to several stimuli, including insulin (Alessi et al., 1996; Kulik et al.,

1997; Hermann et al., 2000), by 3-Phosphoinositide-Dependent Protein Kinase-1 (PDK1)

(Alessi et al., 1997b) and mTORC2 (Hresko & Mueckler, 2005; Sarboassov et al., 2005;

Guertin et al., 2006; Ikenoue et al., 2008). Feedback mechanism within this pathway are

discussed in Section 3.1.3 – Interplay Between mTORC1 & AKT1.

Page 14

Figure 1.2.1a – Schematic of the mTORC1 Pathway & the Downstream Components Regulating

Translation Initiation. Upon binding of Insulin to the Insulin Receptor, the Insulin Receptor Substrate

1 (IRS-1) becomes phosphorylated (Gual et al., 2005). Downstream of IRS-1 is Phosphatidylinositol 3-

Kinase (PI3K) (Shaw, 2001; Khamzina et al., 2003). Downstream of PI3K lies the 3-Phosphoinositide-

Dependent Protein Kinase (PDK1) and Protein Kinase B (AKT1) (Cantley, 2002; Hemmings & Restuccia,

2015). A further signalling component downstream of PI3K is the Mammalian Target of Rapamycin

Complex (mTORC) 2 (Liu et al., 2015). The activation of AKT1 requires phosphorylation by both PDK1

and mTORC2 (Alessi et al., 1997b; Hresko & Mueckler, 2005; Sarbassov et al., 2005). Once activated,

AKT1 serves to activate, via several mechanisms, mTORC1 (discussed later in this Section).

Downstream of mTORC1, two proteins involved in the regulation of translation initiation are

phosphorylated, p70 Ribosomal S6 Kinase 1 (S6K1) (Seufferlein & Rozengurt, 1996; Burnett et al.,

1998) and eIF4E-Binding Protein 1 (4E-BP1) (Beretta et al., 1996; von Manteuffel et al., 1996).

Page 15

The first mechanism of mTORC1 activation involves the phosphorylation, and inhibition of

the protein Tuberin (TSC2) by AKT1 (Inoki et al., 2002; Potter et al., 2002; Cai et al., 2006)

and several other key signalling components including, ERK1/2 (Arvisais et al., 2006; Winter

et al., 2011), the p90 Ribosomal S6 Protein Kinase (RSK1) (Roux et al., 2004) and the p38

substrate MAPK-Activated Protein Kinase-2 (MK2) (Li et al., 2003). TSC2, along with

Hamartin (TSC1), composes the Tuberous Sclerosis Complex (Carbonara et al., 1994; Green

et al., 1994 (as cited by Tee et al., 2003)). In the unphosphorylated state, TSC2 interacts

with the GTPase protein, Ras-Homologue Enriched in Brain (Rheb) and the hydrolysis of

Rheb-bound GTP occurs, which maintains mTORC1 in an inactive state (Garami et al., 2003;

Inoki et al., 2003; Zhang et al., 2003). However, once phosphorylated, the GTPase Activating

Protein (GAP) activity of TSC2 is inhibited and the ability of Rheb to hydrolyse bound GTP is

diminished (Inoki et al., 2002; Potter et al., 2002).

AKT1 further activates mTORC1 through the phosphorylation of PRAS40 (Kovacina et al.,

2003; Nascimento et al., 2006; Vander Haar et al., 2007) (as shown in Figure 1.2.1b).

PRAS40 interacts, via RAPTOR, with mTORC1 with PRAS40 preventing mTORC1 substrates

interacting with RAPTOR (Sancak et al., 2007; Wang et al., 2007c). Such inhibition is

prevented by AKT1-mediated phosphorylation of PRAS40 (Sancak et al., 2007).

Page 16

Figure 1.2.1b – Summary of the Role of AKT1 in the Activation of mTORC1. AKT1 is known to mediate

the activation of mTORC1 through two distinct, complementary mechanisms. The first is the

phosphorylation of the Tuberin (TSC2) component of the Tuberous Sclerosis Complex 1/2 complex.

AKT1-dependent TSC2 phosphorylation is believed to occur on multiple Serine residues (Inoki et al.,

2002). The result of this phosphorylation is suppression of the GTPase Activating Protein (GAP) activity

of TSC1/2 toward Ras-Homologue Enriched in Brain (Rheb) (Garami et al., 2003). This results in the

activation of mTORC1 via a mechanism that is not clearly defined. The second mechanism concerns

the direct phosphorylation of Proline-Rich AKT Substrate 40kDa (PRAS40) (Kovacina et al., 2003). In

the unphosphorylated form, PRAS40 prevents substrate recognition by the Regulatory-Associated

Protein of mTOR (RAPTOR) (Wang et al., 2007a). Once PRAS40 is phosphorylated the mammalian

Target of Rapamycin (mTOR) Complex 1 (mTORC1) becomes activated as RAPTOR is free to interact

with mTORC1 substrates, such as 4E-BP1 and S6K1. The two mechanisms are believed to act in a

synergistic manner (Sancak et al., 2007).

Page 17

A further mechanism requires the localisation of mTORC1 to the lysosome (Sancak et al.,

2008). This localisation is mediated through the actions of four small heterodimeric GTPase

proteins, termed RagA through to RagD, and a five protein complex, termed Ragulator,

which serves as a Guanine Nucleotide Exchange Factor (GEF) for the Rag GTPases (Sancak et

al., 2008; Sancak et al., 2010; Bar-Peled et al., 2012). The GATOR1 complex opposes the GEF

activity of the Ragulator complex by providing Rag-directed GAP activity (Bar-Peled et al.,

2013). The activation of mTORC1 by the Ragulator complex requires amino acids and the

lysosomal vacuolar H+/ATPase (Zoncu et al., 2011; Bar-Peled et al., 2012; Jewell et al., 2015).

This mechanism of activation is further complicated by the fact that different amino acids

activate mTORC1 through different Rag proteins and that additional proteins are required to

allow Rag GTPases to respond to specific amino acids (Jewell et al., 2015; Resamen et al.,

2015; Wang et al., 2015).

Both mechanisms of activation serve to allow mTORC1 to respond to a diverse array of

stimuli and may in fact work to complement each other. In support of this idea, it is worth

noting that Rheb is localised to the lysosomal membrane (Saito et al., 2005 (as cited by

Groenewald & Zwartkruis, 2013)).

In a tissue-specific manner, vertebrates encode three proteins that sequester eIF4E from an

interaction with eIF4G: 4E-BP1, 4E-BP2 and 4E-BP3 (Marcotrigiano et al., 1999; Graber et al.,

2013). The sequence to which eIF4E interacts with 4E-BP1, or eIF4G, is Tyr-x-x-x-x-Leu-h (in

which x represents any amino acid whilst h denotes a hydrophobic residue) (Marcotrigiano

et al., 1999; Niedzwiecka et al., 2002). Interactions with 4E-BP1 prevent an interaction

between eIF4E and the mRNA cap structure (Tomoo et al., 2005; Volpon et al., 2006).

Hypophosphorylated 4E-BP1 interacts with eIF4E, as shown in Figure 1.2.1c. Once

Page 18

phosphorylated, by the RAPTOR subunit (Hara et al., 2002; Schalm et al., 2003) of mTORC1

(Beretta et al., 1996; von Manteuffel et al., 1996; Burnett et al., 1998; Gingras et al., 1998;

Marcotrigiano et al., 1999), 4E-BP1 dissociates from eIF4E and allows the eIF4F complex to

form. Multiple resides on 4E-BP1, Thr-37, Thr-46, Thr-70 and Ser-65, are targetted for

phosphorylation (Gingras et al., 2001) with phosphorylation of positions 37 and 46 occuring

first (Gingras et al., 1999).

Page 19

Figure 1.2.1c – Regulation of Translation Initiation by the Sequestration of eIF4F Components. The

eIF4F complex consists of the scaffold protein, eIF4G, the cap binding protein eIF4E and the mRNA

helicase, eIF4A. The availability of eIF4E and eIF4A are tightly regulated by interactions with the

binding proteins 4E-BP1 and Programmed Cell Death 4 (PDCD4), respectively. Once sequestered,

these initiation factors are unable to interact with eIF4G to form the eIF4F complex. Consequently,

the 43S complex is then unable to form. However, the activity of these binding proteins is tightly

regulated by the mTORC1 and PI3K signalling pathways. Once these pathways are activated, 4E-BP1

becomes phosphorylated by mTORC1 (Beretta et al., 1996) and PDCD4 is phosphorylated by AKT1

(Palamarchuk et al., 2005) and S6K1 (Dorrello et al., 2006). These post-translational modifications

have an inhibitory effect on the interactions with either eIF4E or eIF4A (as shown by the red round-

headed arrows) and, thus, these factors are free to form the eIF4F complex.

Page 20

Unlike the regulation of 4E-BP1, the interaction between eIF4E and 4E-BP2 is only partially

regulated by mTORC1-dependent phosphorylation. 4E-BP2, within the brain of mammals,

interacts with RAPTOR at higher affinity when two Asparagine residues, not conserved in 4E-

BP1 or 4E-BP3 at positions 99 and 102, are deamidated (Bidinosti et al., 2010). The

regulation of 4E-BP3 is not fully understood (Kleijn et al., 2002).

Additionally, truncation of 4E-BP1 (Tee & Proud, 2002; Constantinou et al., 2008; Dennis et

al., 2011) and 4E-BP2 (Wollenhaupt et al., 2012), at the N-terminal phosphorylation sites,

results in sequestration of eIF4E. Such truncation promotes the long-term, stable repression

of translation and occurs primarily during cellular stress (Constantinou et al., 2008).

The availability of eIF4A is similarly regulated by the binding protein, Programmed Cell

Death 4 (PDCD4) (Yang et al., 2003; Zakowicz et al., 2005; Suzuki et al., 2008). As with 4E-

BP1, the phosphorylation of PDCD4 serves to negatively regulate the PDCD4-eIF4A

interaction (Palamarchuk et al., 2005; Dorrello et al., 2006).

Section 1.2.2 – eIF4E Phosphorylation and The Role of ERK1/2 & p38

eIF4E phosphorylation occurs in response to multiple stimuli, including oxidative stress (Rao,

2000; Duncan et al., 2003; Duncan et al., 2005) and viral infection (Mizutani et al., 2004),

and may contribute to the transformation process (Topisirovic et al., 2004; Bianchini et al.,

2008; Furic et al., 2010; Robichaud et al., 2015). Whilst the relevance of eIF4E

phosphorylation has been hard to elucidate (Minich et al., 1994; Scheper et al.¸2001;

Zuberek et al., 2003), recent work has shown it mediates a change in gene expression during

cellular stress responses (Andersson & Sundler, 2006; Herdy et al., 2012; Royall et al., 2015).

Page 21

Extracellular Signal-Regulated Kinase (ERK) 1 (44kDa) and ERK2 (42kDa) are two of the most

canonical signalling components in the cell and are responsible for the phosphorylation of

over 160 proteins with a variety of cellular functions (Cobb et al., 1994; Ferrer et al., 2001;

Yoon & Seger, 2006 (as cited by Zampieri et al., 2007). ERK1/2 are situated downstream of

MAPK/ERK Kinase (MEK) 1/2 and Raf-1 (Crews et al., 1992; MacDonald et al., 1993), as

shown in Figure 1.2.2a. These two protein kinases are activated by dual phosphorylation of

sites Thr-183 and Tyr-185 (Payne et al., 1991; Robbins et al., 1993; Canagarajah et al., 1997).

Page 22

Figure 1.2.2a – Signalling Cascade Resulting in the Activation of ERK1/2. Extracellular

Signal-Regulated Kinase 1/2 (ERK1/2) is a substrate of MAPK/ERK Kinase 1/2 (MEK1/2). In

order to phosphorylate MEK1/2, RAS, when bound to Guanosine Trisphosphate (GTP),

phosphorylates Raf-1 (MacDonald et al., 1993; Jelinek et al., 1996). The activation of this

pathway occurs in response to a wide variety of environmental stimuli through cell

surface receptors, including the Erythropoietin Receptor (EpoR) (Guillard et al., 2003;

Miyake et al., 2013) and the Epidermal Growth Factor Receptor (EGFR) (Correa-Meyer et

al., 2002; Pastore et al., 2005).

Page 23

Several of the 160 ERK1/2 substrates have important functions in the regulation of

translation intiation, as outlined in Figure 1.2.2b. These include the phosphorylation of

RSK1 (Gavin & Nebreda, 1999) and the phosphorylation, and activation, of the mTORC1

component, RAPTOR (Carríere et al., 2010).

Page 24

Figure 1.2.2b – Effects of ERK1/2 Signalling on Translation Initiation. Once activated, ERK1/2

induces several stimulatory effects on translation initiation. The phosphorylation of the mTORC1

component Raptor brings about the full activation of the mTORC1 pathway, resulting in the

phosphorylation of 4E-BP1. Translation is also enhanced by ERK1/2 through the RSK1-mediated

phosphorylation of helicase accessory factor, eIF4B and through the suppression of 4E-BP1

expression. The other main mechanism by which ERK1/2 can affect translation is through the

phosphorylation of eIF4E via the ERK1/2 substrate, MNK2. The phosphorylation of eIF4E is

viewed as allowing the selective translation of specific mRNAs. Through these pathways, ERK1/2

can be seen as a central regulator of translation initiation.

Page 25

Four isoforms of p38 (α, β, γ and δ) are expressed in a tissue-specific manner in mammals

and share between 45% and 75% amino acid sequence homology (Jiang et al., 1996;

Mertens et al., 1996; Goedert et al., 1997; Hu et al., 1999; Korb et al., 2006; Cuenda &

Rousseau, 2007). Three kinases have been implicated in the activation of p38: Mitogen-

Activated Protein Kinase Kinase (MKK) 3 (Derijard et al., 1995; Raingeaud et al., 1996;

Makeeva et al., 2006), MKK4 (Winston et al., 1997; Ganiatsas et al., 1998; Blüher et al.,

2009) and MKK6 (Raingeaud et al., 1996; Bode et al., 2001; Brancho et al., 2003). It is

believed that MKK3 is the main kinase responsible for activation of p38 (Galan-Moya et al.,

2011). Furthermore, MKK4 is only responsible for phosphorylating p38 in vitro (Winston et

al., 1997; Ganiatsas et al., 1998; Blüher et al., 2009).

The activation of p38 produces considerable effects on the cell and a number of substrates

have been identified as being downstream of this kinase, including MK2 (Alessi et al., 1996;

Zu et al., 1996; Krump et al., 1997), Cyclic AMP-Dependent Transcription Factor (ATF-2)

(Raingeaud et al., 1995; Waas et al., 2001; Ouwens et al., 2002) and Mitogen- and Stress-

Activated Kinase 1 (MSK1) (Deak et al., 1998; McCoy et al., 2005; van der Heide et al., 2011).

Downstream of MK2, p38 plays a role in the regulation of translation, as shown in Figure

1.2.2c, by the phosphorylation of Tristetraprolin (TTP) which has a considerable affect on

mRNA stability by negatively regulating the degradation of mRNAs containing Adenosine

and Uridine-Rich Elements (Clement et al., 2011; Tiedje et al., 2012).

Page 26

Figure 1.2.2c – The Activation of p38α & The Effects of this Kinase on Translation and Gene Expression. The

activation of p38α occurs in response to cellular stresses, such as Lipopolysaccharide (LPS) treatment. The

main cellular kinase responsible for the activation of this MAPK is MAPK Kinase 3 (MKK3) (Galan-Moya et al.,

2011). Upstream of MKK3 lies several kinases, including Transforming Growth Factor-β1-Activated Kinase 1

(TAK1) (Kim et al., 2007a; Xin et al., 2010). Other kinases also shown to be responsible for MKK3

phosphorylation include Mixed Lineage Kinase 3 (Tibbles et al., 1996; Abi Saab et al., 2012; Zhou et al., 2014),

MAPK Kinase Kinase (MEKK) 3 (Uhlik et al., 2003) and MEKK4 (Abell et al., 2007). Once phosphorylated by

MKK3, p38α is capable of not only mediating a change in gene expression, through Cyclic AMP-Dependent

Transcription Factor (ATF-2) phosphorylation, but it plays a key role in controlling translation. The first such

role involves the phosphorylation of MAPK-Interacting Kinase (MNK) 1. This kinase mediates an increase in

eIF4E phosphorylation. The second role in controlling translation is found in regulating the stability of specific

mRNA. This occurs through the p38α substrate, MK2. Once active, MK2 phosphorylates Tristetraprolin (TTP)

and the degradation of mRNAs containing Adenosine and Uridine Rich Elements.

Page 27

Despite p38 and ERK1/2 having different roles in the regulation of translation initiation,

both kinases share roles in regulating the expression of 4E-BP1 and the phosphorylation of

eIF4E. Both p38 and ERK1/2 have been shown to negatively regulate 4E-BP1 expression

through the phosphorylation of Elk-1 (GIlle et al., 1996; Cruzalegui et al., 1999; Guha et al.,

2001; Rolli-Derkinderen et al., 2003).

Both MNK1 and MNK2 have been shown to phosphorylate eIF4E at position Ser-209

(Scheper et al., 2001; Ueda et al., 2004). The activity of MNK2 is constitutive and maintains

the phosphorylation (Ueda et al., 2004), whilst MNK1 activity is inducible (Scheper et al.,

2001). These two kinases are regulated by the ERK1/2 and p38 mitogen activated protein

kinases cascades (Fukunaga & Hunter, 1997; Waskiewicz et al., 1997). MNK1 is regulated by

p38 whilst ERK1/2 is solely responsible for MNK2 phosphorylation (Fukunaga & Hunter,

1997; Waskiewicz et al., 1997). The ability of MNK1 to phosphorylate eIF4E is not

exclusively dependent upon p38. It must first interact with eIF4G, within the 43S complex,

and requires the phosphorylation of eIF4G at position Ser-1186 by Protein Kinase C (PKC) α

(Dobrikov et al., 2011; Walsh & Mohr, 2014).

Section 1.2.3 – Ribosomal Protein S6 Kinases

The activities of the kinases responsible for the phosphorylation of the ribosomal S6 protein

are a good example of convergence of biological activity. The two classes of enzyme

responsible for this have been termed the p70 Ribosomal S6 Kinases (S6K1) (Shima et al.,

1998) and RSK1 (Erikson & Maller, 1985). Both of these kinases are also able to

phosphorylate eIF4B (Shahbazian et al., 2006; van Gorp et al., 2009), thereby enhancing the

activity of eIF4A. Despite these similarities, both S6K1 and RSK1 have distinct cellular

functions. The S6K1 acts to negatively regulate mammalian Target of Rapamycin Complex 2

Page 28

(Julien et al., 2010) and provide negative feedback to Insulin receptor signalling (Tremblay et

al., 2007). RSK1 can act to stimulate mTORC1 via suppression of Tuberous Sclerosis

Complex function (Roux et al., 2004).

RSK1 is activated by ERK1/2 (Gavin & Nebreda, 1999), ERK5 (Ranganathan et al., 2006) and

the PDK1 (Jensen et al., 1999), as shown in Figure 1.2.3a. Interestingly there is some

evidence that RSK1 is able to phosphorylate ERK5, suggesting that dual regulation may exist

here (Le et al., 2013). The regulation of S6K1 is more straight-forward requiring dual

phosphorylation. Both mTORC1 (Seufferlein & Rozengurt, 1996; Burnett et al., 1998) and

PDK1 (Alessi et al., 1997a) are required for activation of this kinase by phosphorylating Ser-

412 (Isotani et al., 1999) and Thr-252 (Alessi et al., 1997a).

Page 29

Figure 1.2.3a – Overview of Pathways Resulting in the Activation of the Ribosomal S6 Protein

Kinases. The activation of the p70 Ribosomal S6 Kinase (S6K1) and p90 Ribosomal S6 Kinase

(RSK1) involves phosphorylation (Magnuson et al., 2012). While 3-Phosphoinositide-Dependent

Protein Kinase (PDK1) can phosphorylate both RSK1 and S6K1 (Alessi et al., 1997a; Jensen et al.,

1999), RSK1 can also be activated by Extracellular-Signal Regulated Kinase (ERK) 1/2 (Gavin &

Nebreda, 1999). However, in order to fully activate S6K1, an additional mTORC1-dependent

phosphorylation is also required (Seufferlein & Rozengurt, 1996; Burnett et al., 1998). In terms of

biological function, RSK1 and Protein Kinase B (AKT1) both function in the phosphorylation of

Tuberous Sclerosis Complex (TSC) 2 protein (Inoki et al., 2002; Potter et al., 2002; Roux et al.,

2004) to bring about Mammalian Target of Rapamycin Complex 1 (mTORC1). Both S6K1 and

RSK1 function to phosphorylate eukaryotic Initiation Factor 4B (not shown here) (Shahbazian et

al., 2006; van Gorp et al., 2009).

Page 30

Section 1.2.4 – Phosphorylation of Eukaryotic Initiation Factor 2α

Eukaryotic initiation factor 2α (eIF2α) phosphorylation occurs in response to a variety of

cellular stresses, including viral infection (Connor & Lyles, 2005; McInerney et al., 2005;

Krähling et al., 2009; Elbahesh et al., 2011; Welnowska et al., 2011), amino acid starvation

(Thiaville et al., 2008; Jiménez-Díaz et al., 2013) and hypoxia (Koumenis et al., 2002; Owen

et al., 2005; Zhu et al., 2009; Liu et al., 2010). Phosphorylation of the Serine residue at

position 51 (Rhoads, 1999) (Ser-51) is an efficient mechanism of inhibiting translation as this

interaction inhibits the ability of eIF2B to exchange GDP for GTP (Krishna et al., 1997;

Ramaiah et al., 1994; Kimball et al., 1998; Sudhakar et al., 2000), thus preventing a new

round of translation initiation occurring.

While different mechanisms have been proposed in yeast and mammalian cells, the

phosphorylation of eIF2α leads to a situation in which the interaction between the β- and γ-

subunits of eIF2 and eIF2B, that allows the exchange of GDP for GTP, to be removed

(Kimball et al., 1998; Krishnamoorthy et al., 2001).

This mechanism of inhibiting translation is mediated by four kinases capable of

phosphorylating eIF2α: Double-Stranded RNA-Dependent Protein Kinase (PKR) (Romano et

al., 1995; Romano et al., 1998; Ung et al., 2001; Vattem et al., 2001), Double-Stranded RNA-

Activated Protein Kinase-Like ER Kinase (PERK) (Shi et al., 1998; Harding et al., 1999; Sood et

al., 2000; Kouroku et al., 2007), Haem-Regulated eIF2α Kinase (HRI) (Crosby et al., 1994; Han

et al., 2001; Liu et al., 2008) and General Control Nonderepressing 2 (GCN2) (Dever et al.,

1992; Ramirez et al., 1992; Dever et al., 1993; Berlanga et al., 1999; Hao et al., 2005; Maurin

et al., 2005). The stresses that result in the phosphorylation of eIF2α are shown in Figure

1.2.4a.

Page 31

Figure 1.2.4a – Cellular Stresses Leading to the Inhibition of Translation through the

Phosphorylation of eIF2α. The phosphorylation of eIF2α is a powerful mechanism by

which cellular translation can be inhibited. Four cellular kinases have been shown to

phosphorylate this initiation factor on Ser-51. This phosphorylation occurs in response

to a wide variety of cellular stresses. It is worth noting that stresses resulting in the

activation of HRI and PERK are both considered to be Endoplasmic Reticulum Stresses.

Page 32

Section 1.2.5 – Emerging Roles for JNK1 in the Regulation of Translation

Initiation

JNK is expressed from three gene: jnk1, jnk2 and jnk3. Ten isoforms of JNK can be expressed

from alternative splicing of these genes (Davis, 2000). While JNK1 and JNK2 are expressed

ubiquitously, JNK3 expression is typically expressed in the brain and testes (Ibid). JNK1 is

canonically associated with stress responses and apoptosis (Dérijard et al., 1994; Minden et

al., 1994; Paraskevas et al., 1999; Te et al., 2015). Whilst not traditionally associated with

the control of translation initiation, several emerging roles have recently began to emerge.

There is some confusion of the role of JNK1 in the phosphorylation of eIF4E. Despite earlier

work revealing no role for JNK1 in the phosphorylation of this initiation factor (Fukunaga &

Hunter, 1997), later work demonstrated that eIF4E phosphorylation was inhibited by

inhibition of JNK1 (Gandin et al., 2010; Jiang et al., 2010). It is not clear how this regulation

occurs as both expression of eIF4E has shown to be reduced by JNK1 inhibition (Jiang et al.,

2010) and the eIF4E kinases, MNK1 and MNK2, have been shown to be inhibited by the

JNK1 chemical inhibitor, SP600125 (Bain et al., 2007).

The second role for JNK1 in the regulation of translation is found in regulating eIF2α

phosphorylation. Dual phosphorylated JNK1 acts as a kinase for Protein Phosphatase 1.

Such inhibitory phosphorylation prevents the dephosphorylation of eIF2α (Monick et al.,

2006). ERK1/2 prevents eIF2α phosphorylation by maintaining JNK1 inactivity (Ibid).

Page 33

Section 1.3 – Overview of Viruses Modelled in this Project

and Viral Strategies to Alter Host Cell Translation Initiation

Viruses are obligate intracellular parasites and as such usurp the host translation machinery.

To this end, viruses have evolved several strategies to divert this machinery from translating

cellular to viral mRNA, and some of these rely on manipulating the signalling pathways that

regulate translation. A major part of this project was to adapt the model of cellular

translational regulation to investigate the effects viruses have upon, the not only

translational machinery but also, the regulatory signalling network. Given this aim, it is

essential to understand the commonly used strategies by which viruses commandeer the

cellular translation initiation machinery and the points within signalling pathways which

they affect in order to inhibit host protein synthesis and carry out processes essential for

viral replication. Moreover, as a major part of this project involved modelling the effects of

viral infection upon cellular translational regulation, an introduction into the three viruses

with effects incorporated into the model is necessary: Junin Virus (JUNV), Murine Norovirus

(MNV) and Andes Virus (ANDV).

Section 1.3.1 – General Features & Epidemiology of Viruses Incorporated into

the Model

This section describes the general feature of the three viral families incorporated into the

model in Chapter 4 - Modelling the Effect of Viral Infection and provides a brief

introduction to the epidemiology and genome of the three viruses in question. In the case

of MNV, the overview of the epidemiology will focus primarily upon the medically important

Human Norovirus (HuNV) as MNV infection, at the organism level, is largely asymptomatic.

Page 34

The incorporation of the effects these viruses induced in the cellular translational and

signalling machinery into the model served to increase the value of the project.

Section 1.3.1.1 – Arenaviruses & Junin Virus

The Arenaviridae family contains 22 enveloped, ambisense RNA viruses (Emonet et al.,

2009; Emonet et al., 2011) with a broad geographical distribution including: North America

(Bear Canyon Virus, Whitewater Arroya Virus and Catarina Virus), the Caribbean (Tacaribe

Virus) and South America (Junin Virus (JUNV), Pichindé Virus and Machupo Virus) (Cajimat et

al., 2007; Cajimat et al., 2013). Phylogenetic analysis of the New World Arenaviruses reveals

three clades (A, B and C) (Bowen et al., 1997; Archer & Rico-Hesse, 2002). Clade B viruses,

such as JUNV, Machupo, Guanarito and Sabiá viruses are responsible for distinct

haemorrhagic fevers (Delgado et al., 2008).

The genome of these cytoplasmic viruses is composed of two segments: the Large segment

(7.3kb) and Small segment (3.5kb) (de la Torre, 2009; Emonet et al., 2011). The Large

segment encodes the 200kDa L protein, an RNA-dependent RNA polymerase (RdRp), and

the 11kDa Z protein, a Zinc Finger protein. The Small segment encodes the 75kDa viral

protein precursor, GPC protein and the 63kDa Nucleoprotein (de la Torre, 2009). Two stem

loop structures are found on each segment in the intergenic regions of the JUNV genome

(Ghiringhelli et al., 1991 (as cited by Palacios et al., 2010)).

JUNV is responsible for Argenitine Haemorrhagic Fever (AHF) (Arribalzaga, 1955; Parodi et

al., 1958 (as cited by Peters, 2006)). The fatality rate for untreated AHF lies between 10%

and 30% (Maizegui et al., 1975; Harrison et al., 1999). Around 70% of cases are diagnosed

by muscle weakness, redness of the eyes, presyncope and subcutaneous rupturing of

capillaries (Schwarz et al., 1970 (as cited by Harrison et al., 1999)). Other symptoms include

Page 35

hypotension, lymphadenopathy, irritability and tremors in the hand and tongue (Enria et al.,

2008). Nearly a third of cases go on to develop haemorrhage and severe neurological

complications, including convulsions and coma (Ibid). A combination of leukopenia and

thromobocytopenia characterise this disease (Harrison et al., 1999).

New World Arenaviruses are zoonotic infections (Charrell & de Lamballerie, 2010).

Serological investigations have revealed that predominant rodent species infected with

JUNV are Calomys musculinus and Calomys laucha, however several other rodent species

and mammals may also act as hosts (Mills et al., 1991; Mills et al., 1992; Mills et al., 1994).

Section 1.3.1.2 - Caliciviruses & Murine Norovirus

The Caliciviridae family contains five genera (Norovirus, Sapovirus, Lagovirus, Vesivirus and

Nebovirus (ICTV Website, 2012)) of positive-sense, single-stranded RNA viruses (Pesavento

et al., 2008). The name of this family is derived from the Chalice, or Calyx, shaped

indentations covering the viral surface (Madely, 1979). The genomes of these viruses

contain between two and four open reading frames (Alhatlani et al., 2015) with fairly

conserved lengths of between 7.4kb and 7.7kb (Meyers et al., 1991; Carter et al., 1992;

Rasschaert et al., 1994; Ward et al., 2007). As HuNV cannot be cultivated (Duizer et al.,

2004; Leung et al., 2010), several other viruses, including MNV, are used as surrogates

(Esseili et al., 2015).

The first open reading frame encodes the non-structural polyprotein (Alhatlani et al., 2015).

The proteins derived from this polyprotein, in FCV, are p5.6, p32, p39 (2C-Like Nucleoside

Trisphosphatase (NTPase)), p30, p13 (VPg) and p76 (Polymerase-Protease fusion)

(Sosnovtseva et al., 1999; Sosnovtsev et al., 2002). In contrast with FCV, MNV requires

distinct RdRp and protease proteins (Wei et al., 2001; Sosnovtsev et al., 2006). The second

Page 36

and third open reading frames encode the capsid proteins (Neill et al., 1991). The fourth

open+ reading frame of MNV encodes the virulence factor VF1, believed to regulate host

cell apoptosis (McFadden et al., 2011).

Two complimentary modes of transmission, foodbourne (Fankhauser et al., 2002;

Widdowson et al., 2005) and person-to-person (Gunn et al., 1980; Green et al., 2002;

Isakbaeva et al., 2005), help to ensure disease dissemination. Initial cases of HuNV occur

through the consumption of contaiminated water or shellfish (Saitoh et al., 2007; Lowther et

al., 2012; Pérez-Sautu et al., 2012). Then similar to FCV dissemination (Pedersen et al.,

2000; Schorr-Evans et al., 2003; Hurley et al., 2004; Clay et al., 2006), the person-to-person

spread of HuNV occurs through ingestion or inhalation of HuNV-containing fomites (Repp &

Keene, 2012).

Most cases of HuNV do not require medical intervention, causing acute, self-limiting

diarrhoea and vomiting lasting up to five days (Rockx et al., 2002; Lopman et al., 2004; Glass

et al., 2009). It is, however, the second leading cause of viral gastroenteritis requiring

medical intervention, behind Rotavirus (Medici et al., 2006; Tran et al., 2010; Lorrot et al.,

2011; Payne et al., 2013; Lu et al., 2015), and causes considerable morbidity and mortality in

people over 65 years (Harris et al., 2008; Trividi et al., 2012) and is responsible for

hospitalising over 23,000 children, under five years, annually in the United States of America

alone (Patel et al., 2008). Despite being behind Rotavirus, HuNV places an extra

CAD4,000,000 burden on healthcare costs in Canada alone (Morton et al., 2015) and a

multi-million pound burden on the National Health Service in the United Kingdom (Danial et

al., 2011).

Page 37

Section 1.3.1.3 – Hantaviruses & Andes Virus

The Hantaviruses, of the Bunyaviridae family, have a genome of segmented, negative-sense

single-stranded RNA and cause either Haemorrhagic Fever with Renal Syndrome or

Hantavirus Cardiopulmonary Syndrome (HCPS) (Plyusnin et al., 1996; Jonsson et al., 2010;

Vaheri et al., 2013). The genome segments of Andes Virus (ANDV) are termed Large (6.6kb),

Medium (3.7kb) and Small (1.9kb) (Meissner et al., 2002) and mainly encode the

Nucleoprotein, Glycoprotein Precursor (GPC) and the RdRp (Tischler et al., 2003; Vera-

Otarola et al., 2012).

Numerous Hantaviruses cause HCPS, including ANDV (López et al., 1996; López et al., 1997),

Sin Nombre Virus (SNV) (Elliott et al., 1994) and several others (Morzunov et al., 1995; Khan

et al., 1996; Johnson et al., 1997; Torrez-Martinez et al., 1998; Johnson et al., 1999; Vincent

et al., 2000). In constrast to the strictly zoonotic SNV (Calisher et al., 2007), ANDV is also

transmissible through person-to-person contact (Enría et al., 1996; Padula et al., 1998;

Martinez et al., 2005; Lázaro et al., 2007; Martinez-Valdebenito et al., 2014) with infected

excretions (Castillo et al., 2007; Godoy et al., 2009). The host reservoir of ANDV appears to

be members of the Oligoryzomys genus (Toro et al., 1998; Gonzalez Della Valle et al., 2002;

Medina et al., 2009 (as cited by Torres-Pérez et al., 2010); Andreo et al., 2011). During

ANDV infection, the host does not display symptoms, in contrast to SNV infection (Calisher

et al., 2005).

ANDV infection has a fatality rate of around 35% and most commonly occurs in men,

representing an occupational risk, with many cases being labourers in the logging or farming

industries (Castillo et al., 2001; Unidad de Vigilancia, Dpto. de Epidemiología, Gobierno de

Chile, 2014). The initial picture of ANDV infection is reminiscent of influenza. As the disease

Page 38

progresses, a cough develops and hypotension, tachycardia and tachypnea are noted

(Castillo et al., 2001). The majority go on to develop bleeding and renal damage or, in

severe cases, haemorrhage and renal failure (Castillo et al., 2001). Extracorporeal

Membrane Oxygenation has increased survival rates of HCPS in SNV infection (Dietl et al.,

2008; Wernly et al., 2011 (as cited by McNulty et al., 2013)).

Section 1.3.2 – General Overview of Viral Effects on Translational Regulation

While the effects specific viruses have upon the cellular translational machinery and the

signalling network were incorporated into the model, it is worth noting that many other

strategies are also used by other viruses. These strategies can be broadly classified as

belonging to one of several general approaches. This section describes these strategies, and

the effect each has upon the translational machinery, and aims to put the specific examples

incorporated into the model into a broader context.

Section 1.3.2.1 – Internal Ribosome Entry Site (IRES)-Mediated Translation &

Host mRNA Degradation

Cap-dependent translation is used by the majority of cellular mRNAs and requires the 5’ cap

to interact with eIF4E (Altmann et al., 1987; Carberry et al., 1989; Goss et al., 1990). In

constrast, RNA secondary structures within the 5’ Untranslated Regions of viruses, termed

Internal Ribosome Entry Sites (IRES) elements, can drive internal recruitment of the

ribosome to engage in cap-independent translation (Kanamori & Nakashima, 2001;

Nishiyama et al., 2003; Thurner et al., 2004; Willcocks et al., 2011). These elements are

found in the Picornaviridae (Jang et al., 1988; Pelletier & Sonenberg, 1988; Liu et al.,

1999)9* Flaviviridae (Tsukiyama-Kohara et al., 1992; Rijnbrand et al., 1997; Fletcher &

Page 39

Jackson, 2002) and Retroviruses (Ohlmann et al., 2000; Brasey et al., 2003; Herbreteau et

al., 2005; Camerini et al., 2008; Weill et al., 2010; Amorim et al., 2014) and allow viral

translation to continue during infection when host translation is diminished. The IRES

elements of these different virus types can be, broadly, classified based upon the

requirements for cellular factors (Kieft, 2008).

Whilst several viral families encode the enzymes responsible for encoding 5’ cap structures

(Decroly et al., 2008; Selisko et al., 2010), others do not and require cellular cap structures

to carry out viral replication. These viral families include the Bunyaviridae (Kormelink et al.,

1992; Garcin et al., 1995), Arenaviridae (Raju et al., 1990; Morin et al., 2010) and the

Orthomyxoviridae (Plotch et al., 1981; Dias et al., 2009). These viruses conduct cap-

snatching to promote viral transcription, rather than translation, (Lelke et al., 2010; Reguera

et al., 2010; Lehmann et al., 2014), while suppressing the innate interferon response of the

cell (Marcus et al., 2005; Qi et al., 2010) and more generally the translational capabilities of

the host.

Host mRNA degradation is also mediated by targeting the RNA binding proteins that affect

mRNA stability. Members of the Alphavirus family sequester the mRNA stabilising protein,

HuR, to the 3’ end of the viral genome (Sokoloski et al., 2010; Barnhart et al., 2013). This

serves to not only promote the degradation of mRNAs requiring HuR, but given the large-

number of HuR binding sites in intron sequences (Lebedeva et al., 2011; Mukherjee et al.,

2011), this also affects the alternative splicing of mRNAs.

Page 40

Section 1.3.2.2 - VPg Proteins

Cap-dependent translation does not specifically require the presence of a 5’

methylguanosine cap and can occur if a cap analogue is present. The VPg protein of the

Caliciviridae family is a good example.

The viral protein, VPg, is covalently linked to the 5’ region of the viral genome (Schaffer et

al., 1980) and acts as a 5’ cap analogue (Chaudhry et al., 2006). Whilst the presence of this

protein is not unique to the Caliciviridae, with VPg proteins being found in viral families

including the Astroviridae (Jiang et al., 1993; Fuentes et al., 2012), Picornaviridae (Flanegan

et al., 1977; Weitz et al., 1986) and Potyviridae (Riechmann et al., 1989; Puustinen et al.,

2002), the size and exact function of this protein is fairly unique. The VPg function differs

from that of the Picornaviruses. Whilst the Caliciviral VPg is required for viral protein

synthesis (Daughenbaugh et al., 2003; Goodfellow et al., 2005), the VPg of the Picornaviridae

functions only for viral genome replication (Flanegan et al., 1977; Morasco et al., 2003;

Murray & Barton, 2003; Liu et al., 2007).

Much like that of the Potyviridae (Léonard et al., 2004; Hébrard et al., 2010), the Caliciviral

VPg (Daughenbaugh et al., 2003; Daughenbaugh et al., 2006; Hosmillo et al., 2014) is

responsible for the interactions with a number of cellular factors vital for the initiation of

translation. However, unlike that of the Potyviridae, the VPg of the Caliciviridae does not

interact with the PABP (Léonard et al., 2004; Beauchemin & Laliberté, 2007).

Several interactions between VPg and eIF3 and eIF4E are known to occur in Norwalk Virus

(Daughenbaugh et al., 2003) and Feline Calicivirus (FCV) (Goodfellow et al., 2005).

Differences in the requirements of eIFs exist between Caliciviruses, but overall, imply a

mechanism by which the VPg protein is able to join the eIF4F complex to the 43S complex to

Page 41

enable cap-dependent translation (Chaudhry et al., 2006). More information regarding

interactions between the VPg and initiation factors is given in Section 4.2.2 – Murine

Norovirus Model.

Section 1.3.2.3 - Host Factor Degradation & Viral Mimics of Host Factors

Viral infection subjects the host cell to pronounced stress and requires the cell to respond

with changes in gene expression. However, viruses are also capable of manipulating this

change to confer an advantage (Villas-Bôas et al., 2009; Royall et al., 2015). An obvious viral

target is the cellular translational machinery with mechanisms including the degradation of

host initiation factors and signalling components and the mimicry of host factors.

To control eIF2 activity, several viruses, including Vaccinia Virus and members of the

Ranavirus genera, encode functional proteins mimicking eIF2α, termed K3L (Essbauer et al.,

2001) and vIF2α (Yu et al., 1999; Grayfer et al., 2015), respectively. K3L inhibits the

phosphorylation of eIF2α via an interaction between the protein and PKR (Kawagishi-

Kobayashi et al., 1997). vIF2α, however, targets the homologue of PKR in fish for

degradation (Jancovich & Jacobs, 2011). PKR is also degraded during infection with the

virulent Phleboviruses, such as Rift Valley Fever Virus (Habjan et al., 2009; Ikegami et al.,

2009; Kalveram & Ikegami, 2013; Lihoradova et al., 2013).

Section 1.3.2.4 - Alteration to Cell Signalling Pathways

In addition to the degradation of host factors, a commonly used strategy to alter the ability

of the host to synthesise proteins is to interfere with the normal functioning of cellular

signalling pathways that control translational activity (as described in Section 1.2 – Key

Points of Regulation & Upstream Signalling Pathways).

Page 42

Given that mTORC1 has a central role in regulating translation initiation, numerous viruses

target this pathway for activation, including: ANDV (McNulty et al., 2013), Human

Cytomegalovirus (HCMV) (a β-Herpesvirus) (Bai et al., 2015) and Influenza A Virus (an

Orthomyxovirus) (Denisova et al., 2014).

The mechanism by which mTORC1 is activated appears to be dependent upon the family of

virus that is infecting the cell. HCMV (Moorman et al., 2008) and Herpes Simplex Virus 1

(HSV-1) (Chuluunbaatar et al., 2010) both affect the activation of mTORC1 at the level of

TSC2 activity. The HSV-1 protein, Us3, phosphorylates TSC2 at the site normally targeted by

cellular kinases, Thr-1462 (Chuluunbaatar et al., 2010). The HCMV viral protein, pUL38, was

able to interact with TSC2 in such a way that mTORC1 activity was increased with this

interaction mimicking the phosphorylation of TSC2 (Moorman et al., 2008).

Section 1.4 – Background to Systems Biology Techniques

Literature reviewed in previous parts of this chapter show that translation is a fundamental

process in the cell. It consumes approximately 30% of the ATP produced during oxidative

phosphorylation (Buttgereit & Brand, 1995). Furthermore, this literature shows that

translation initiation is regulated by a vast, interconnected network of molecular

components which coordinate protein synthesis with other cellular processes in response to

a diverse array of environmental cues. Due to the number of components and interactions,

it is not possible to understand how the system functions by following individual scientific

articles describing individual interactions. Since the beginning of this century, the Systems

Biology field has robustly argued that cellular behaviour at the molecular level can only be

fully understood if it is studied as a whole and that computational and mathematical

modelling tools are indispensable in this process (Di Ventura et al., 2006; Chuang et al.,

Page 43

2010). The challenges of modelling biological systems relate to unprecedented complexity

and the difficulty in obtaining quantitative data have led to the development of an armoury

of computational systems biology techniques aimed at representing the system in different

levels of detail. This being said, these approaches have not yet been fully integrated into

one generally accepted methodology, with the choice of method depending upon the aim

and hypothesis of individual projects.

Throughout the course of this project, several systems biology techniques were used to

construct the computational model representing literature on the molecular interactions

network regulating translation initiation in mammalian cells. The following sections provide

the background information to these methods. The first three sections describe the three

fundamental, most commonly used approaches of computational systems biology: Exact

Stochastic Simulations with the Gillespie algorithm, Ordinary Differential Equations (ODEs)

and Flux Balance Analysis (FBA). Following these sections, Petri Nets will be introduced as a

qualitative simulation approach and a unifying framework for representing qualitative Exact

Stochastic and ODE models. Finally, Quasi-Steady State Petri Nets (QSSPN) will be

introduced as an attempt to integrate Petri Nets with FBA. In the final section, the general

terminology of Petri Nets will be translated to a vernacular more familiar to biologists.

Section 1.4.1 – Exact Stochastic Simulations with the Gillespie Algorithm

The Gillespie algorithm for Exact Stochastic Simulations was initially developed to account

for inherent, random noise in systems of coupled chemical reactions, resulting from the fact

that individual reactions occur due to random collisions between molecules. In laboratory

experiments, reactions are carried out in large volumes involving large numbers of

molecules. A consequence of this is deterministic behaviour due to the averaging of

Page 44

random effects. However, in the small volumes of living cells there may be reactions

involving just a few molecules (e.g. the binding of 10 molecules per cell of the LacR

transcription factor to the lactose operon) (Krebs et al., 2013). In this case, the considerable

random noise in the outcome of individual reactions is expected and leads to the

heterogeneity of cellular behaviours. Indeed, early simulations of prokaryotic gene

expression with the GIllesipe algorithm (McAdams & Arkin, 1997; Kierzek et al.¸ 2001)

predicted fundamental stochastic phenomena in gene expression. These were later

experimentally demonstrated (Ozbudak et al., 2002; Elowitz et al., 2002).

The Gillespie algorithm is described in detail below as it is the workhorse of this project and

used in the majority of the simulations reported here. However, the purpose of using this

algorithm is different than in the majority of computational systems biology projects. The

Gillespie algorithm is not used to perform quantitative simulations nor is it used to study

random heterogeneity in the number of molecules in individual cells. Rather, it is used to

randomly sample alternative sequences of molecular events, linking signalling network

inputs and molecular effector outputs in a qualitative model of the molecular system

regulating mammalian translation initiation. This important distinction will be revisited in

Section 1.4.4 – The Petri Net Formalism, where the Qualitative Petri Net approach will be

introduced in detail.

The Gillespie algorithm was initiatally formulated to model the dynamics of systems of

coupled chemical reactions (Gillespie, 1977). This method is frequently referred to as an

exact stochastic simulation because it proceeds by generating individual reaction events

occurring at exact times. At each iteration two questions are answered: i. Which of these

reactions will next occur? ii. What is the waiting time for the next reaction to occur?

Page 45

In order to answer these questions, Gillespie (1977) applied a Monte Carlo approach (Ulam

et al., 1947; Metropolis & Ulam, 1949). A pseudo-random number generator was used to

create two numbers (𝑟1 and 𝑟2). These numbers are, in turn, used to calculate two

additional variables, 𝜏 and 𝜇, respectively. The equations below (Equation 1.4.1a and

Equation 1.4.1b), from Gillespie (1977), describe the calculations used to ascribe values to 𝜏

and 𝜇:

𝜏 = (1 𝛼0⁄ )𝑙𝑛(1 𝑟1⁄ )

The value of 𝜇 is an integer value that satisfies the following statement:

∑ 𝛼𝜈 < 𝑟2𝛼0 ≤

𝜇 − 1

𝜈 = 1

∑ 𝛼𝜈

𝜇

𝜈=1

In both calculations, the value of 𝛼0 is equal to the sum of the propensity functions

describing all the reactions in the system. A propensity function for reaction 𝜇 can be

described as the product of the number of molecules of various substances involved in

reaction 𝜇 (ℎ𝜇) and the stochastic rate constant for the reaction 𝜇 (𝑐𝜇), as shown below

(Equation 1.4.1c):

𝑎𝜇 = ℎ𝜇𝑐𝜇

In short, the answer to the question of which reaction will occur next, can be viewed as

being a probability (Gillespie, 1977). The probability of instigating reaction 𝜇 will depend

upon the size of the propensity function for the reaction, 𝑎𝜇, as a fraction of the sum of all

propensity functions (𝛼0). As the propensity function of a particular reaction increases, the

ratio of this to 𝛼0 also increases. Consequently, it becomes more probable that this reaction

will occur. The selection of a propensity function can be exemplified with a pie chart. As

Equation 1.4.1a

Equation 1.4.1c

Equation 1.4.1b

Page 46

shown in Figure 1.4.1a, the probability of randomly selecting a reaction is proportional to

the size of the area that this function occupies.

Page 47

Figure 1.4.1a – Schematic View of Propensity Functions. The value of 𝛼0 would be

described as the sum of all the propensity functions (α1 to α

5) (Gillespie, 1977). The

larger the propensity function for a particular reaction, the more likely it is to occur.

As α3 has the largest propensity function, and so the highest fraction of α

0, it is most

likely to occur. Conversely, as α5 makes up the smallest fraction of α

0, the reaction

with propensity function α5 would occur least.

α1

α2

α3

α4

α5

Page 48

The value of 𝜏 describes the time interval taken for the next reaction to occur (Ibid) and

provides an answer to the second question given above. In other words, at 𝑡𝑛 = 𝑡𝑛−1 + 𝜏 ,

reaction 𝜇 will occur. The probability of time interval 𝜏 occurring can be described using an

exponential decay (Gillespie, 1977), and it is theoretically possible for the waiting time to

increase to infinity. However, the probability of this infinite waiting time interval becomes

infinitesimally small. The following equation (Equation 1.4.1d) (Ibid) shows how the

Gillespie algorithm calculates the probability of a waiting time interval, 𝜏, for reaction 𝜇:

𝑃(𝜏, 𝜇) = 𝛼𝜇𝑒−𝛼0𝜏

Once the identity of the next reaction and associated waiting time are determined, the state

of the system is updated by executing one elementary reaction event occurring at an exact

time. Thus, if reaction 𝜇 is:

𝐴 + 2𝐵 → 𝐶

In Equation 1.4.1e, the number of molecules of 𝐴 will be decreased by one, the number of

molecules of 𝐵 will decrease by two and the number of molecules of 𝐶 will increase by one.

The simulation time is then increased by adding the waiting time, 𝜏, to the current

simulation time. This way of updating the system can be described as asynchronous: in a

given iteration of the simulation, only one event is executed and only the species

participating in this event are updated.

The direct method of Gillespie (1977) described above, is computationally time consuming.

At every iteration, two random numbers must be generated. Gibson and Bruck (2000) have

proposed a method, termed the Next Reaction Method, which is equivalent to the Gillespie

algorithm where only one random number per iteration needs to be generated, and where

Equation 1.4.1d

Equation 1.4.1e

Page 49

the search for the next reaction is optimised. While very sophisticated, this method would

still be very computationally expensive if the number of molecules were very large.

To address this problem, the approximate stochastic techniques have also been developed,

including the τ-Leap method (Gillepsie, 2001). In this method, the discrete timestep is

introduced. In each iteration, the number of firings of each reaction within this timestep is

generated by a pseudo-random number from a Poisson distribution. Subsequently, new

states of the system are calculated by executing all reactions capable of firing. This way of

updating the system can be described as synchronous: in a given iteration of the simulation

all reactions and substances are updated simultaneously. While this algorithm is much

faster and practical for systems containing large numbers of molecules, it is no longer exact,

as arbitrary timesteps are introduced. The synchronous updates may even lead to the

generation of negative numbers of molecules if the number of reactants are not sufficiently

large. This algorithm cannot be applied to systems containing reactions with small numbers

of molecules (e.g. transcription factors binding to bacterial genes). As the τ-Leap method

has not been used in this project, it will not be described in more detail. A possible link

between these approximate stochastic methods and the exact stochastic methods is the K-

Leap Method (Cai & Xu, 2007).

Section 1.4.2 – Ordinary Differential Equations (ODEs)

Deterministic modelling involves the representation of a systems as a set of coupled

Ordinary Differential Equations (ODEs). This method is the most frequently used

representation of biological systems (Wu et al., 2008). Individual model components are

viewed as undergoing changes in concentration per unit time, as shown in Equation 1.4.2a.

Page 50

A description of a system in this way requires an in depth knowledge of the system

(Williamson et al., 2009).

𝑑[𝑋]

𝑑[𝑡]= 𝑘1[𝑌][𝑍] − 𝑘2[𝐴][𝐵]

As shown in Equation 1.4.2a, an ODE describes changes in the concentration of a substance

(𝑑[𝑋]) per unit time (𝑑[𝑡]). This rate of change is, therefore, the sum of all reactions that

both consume and produce substance 𝑋 (Wu et al., 2008; Korkola et al., 2015). In the

example given in Equation 1.4.2a, the reaction involving 𝑌 and 𝑍 and described by the rate

constant 𝑘1 produce 𝑋, whilst that described by the constant 𝑘2, and involving 𝐴 and 𝐵

degrade substance 𝑋. In a deterministic model, a series of ODEs is constructed to account

for the synthesis and degradation of each model component. In order for ODEs to be

constructed it is essential that two conditions are satisfied. Firstly, the temporal nature of

the system must be known (Rangamani & Iyengar, 2008; Cowan et al., 2012). Secondly, it is

necessary to have knowledge of either the volume of the compartment in which the

reactions occur or the initial concentrations of each species (Cowan et al., 2012). In relation

to the initial concentrations of model components, it is necessary to state that these values

should be sufficiently large so as to view them as continuous variables (Rangamani &

Iyengar, 2008).

Numerical simulations of ODE systems is one of the most developed areas in applied

mathematics, with an array of sophisticated methods having been developed since

computers became available in the second half of the 20th Century (Press et al., 2007). The

common features of all algorithms is that at each iteration of the simulation a timestep is

assumed. The ODEs are then used to calculate the change in concentration of each

Equation 1.4.2a

Page 51

substance during this timestep. Substance concentrations are then synchronously updated.

Individual, highly optimised algorithms are implemented in numerical solver libraries, such

as Sundials (Hindmarsh et al., 2005). A detailed description of these methods is beyond the

scope of this dissertation.

Section 1.4.3 – Flux Balance Analysis (FBA)

The application of ODE or exact stochastic simulations to large-scale, and potentially whole-

cell, molecular interaction networks is hindered by a lack of sufficient, quantitative data on

the values such as reaction rate constants or numbers of molecules. FBA offers a method

for the analysis of whole genome-scale metabolic networks (GSMN). The cost of using this

method is the introduction of two approximations. Firstly, the use of reaction fluxes rather

than molecular amounts. Secondly, this method only permits the analysis of the system at

steady state. As a consequence, the ODE model describing the steady state reactions

between metabolites is expressed by the following equation (Equation 1.4.3a):

𝑑𝑥

𝑑𝑡= 0 = 𝑆𝑣

In Equation 1.4.3a, the change in concentration per unit time is equal to the product of the

Stoichiometric matrix (𝑆) and the flux vector (𝑣) (Raman & Chandra, 2009; Orth et al., 2010).

The Stoichiometric matrix contains m rows and n columns denoting m metabolites and n

reactions, respectively (Covert et al., 2001; Benyamini et al., 2010). For example, the

Stoichiometric matrix 𝑆𝑖𝑗 represents the stoichiometric coefficient of the 𝑖-th metabolite in

the 𝑗-th reaction. Flux vector 𝑣 would then contain information on the metabolic reaction

fluxes of n reactions. At steady state this becomes a vector with all components equal to

zero. (Covert et al., 2001). If experimental data on some of the steady state flux values or

Equation 1.4.3a

Page 52

their ranges are available, the solution space can be further constrained by flux bounds, as

shown in Equation 1.4.3b:

𝑓𝑙 < 𝑣 < 𝑓𝑢

where the flux vector (𝑣) must fall between upper (𝑓𝑢) and lower (𝑓𝑙) flux bounds (Orth et

al., 2010). Furthermore, it is worth noting that fluxes maybe positive or negative depending

on the reversibility of the biochemical reaction being described by that particular flux (Ibid).

The boundary conditions of the FBA problem are defined by external metabolites which are

not subject to the balance constraints of Equation 1.4.3a. These metabolites are treated as

unlimited sources and sinks of metabolic fluxes. Usually, external metabolites are used to

represent nutrients available in the medium (e.g. glucose and amino acids). The bounds of

exchange reactions connecting external metabolites to the first balanced precursor in the

network can be used to quantitatively constrain nutrient availability.

Flux Balance Analysis is commonly used to study full GSMNs involving thousands of

reactions (Orth et al., 2010). In these systems it is never the case that enough fluxes are

measured to solve Equation 1.4.3a and determine values of all fluxes. However, it is

possible to determine the maximal value of any linear combination of fluxes:

max𝒇𝑙,𝒇𝑢,𝑺

(𝒄 ∙ 𝒗), 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡

𝑺 ∙ 𝒗 = 𝟎

𝑓𝑙 < 𝑣 < 𝑓𝑢

where, 𝑐 is a vector of 𝑛 coefficients set by the user. This vector defines an objective

function, which expresses the metabolic functions of the cell under investigation. A

Equation 1.4.3c

Equation 1.4.3b

Equation 1.4.3d

Page 53

common choice is biomass, for an objective function, where coefficients corresponding to

reactions producing cellular biomass components are set according to measured biomass

composition and other coefficients are set to zero. The maximisation problem defined by

Equation 1.5.3c is solved by linear programming algorithms, with the Simplex method being

most frequently used (Orth et al., 2010). This use of linear programming guarantees finding

the maximal, unique value of the objective function, but there may be many solution

vectors 𝒗 corresponding to this value. A detailed understanding of linear programming

(Kantorovich, 1939 (as cited by Bentobache & Bibi, 2012)) and the Simplex algorithm

(Dantzig, 1951; Dantzig, 1963 (as cited by Bentobache & Bibi, 2012)) are beyond the scope

of this thesis.

One of the reasons why FBA has been widely adopted is that it offers a gradually changing

level of quantitative model accuracy depending on available experimental data. If no

quantitative information is available, the maximisation of the objective function can be used

to check whether a certain metabolic function is possible. The conclusion drawn from a

maximal value of the objective function equalling zero is that the metabolic function in

question is not possible (Orth et al., 2010). Conversely, objective function values greater

than zero can be interpreted as a qualitative indication of the metabolic function being

feasible (Ibid). This is particularly useful in evaluating effects of perturbations and gene

knockouts. For example, Beste et al. (2007) used the GSMN-TB model of Mycobacterium

tuberculosis to test feasibility of growth (through a biomass objective function) for all single

gene knockouts against metabolic genes. The results of which were then compared with

transposon site mutagenesis (Ibid). Statistically significant predictive power has been

achieved with this model (Beste et al., 2007). As quantitative information about fluxes

Page 54

become available, solution space can be further constrained. For microorganisms growing

in bioreactors, quantitative predictions of metabolic yields are possible and applied in

metabolic engineering (Fisher et al., 2014).

Recently, Flux Balance Analysis has been widely applied to modelling metabolism of human

tissues (Gille et al., 2010; Thiele et al., 2013). The capability of FBA to qualitatively predict

feasible flux distributions has been further exploited in the mechanistic interpretation of

transcriptome data. For example, Leoncikas et al. (2016) have analysed the transcriptomes

of 2,000 individual breast cancer tumours in the context of a human GSMN (Thiele et al.,

2013) and have identified metabolic pathways associated with poor prognosis.

To summarise, Flux Balance Analysis is the first approach capable of mechanistically

modelling the genotype-phenotype relationship at the full genome scale. This is currently

limited to metabolic genes and metabolic phenotypes only. From this perspective, one of

the aims of this project was to extend the scope of mechanistic modelling to large scale

signalling and gene regulatory network that cannot be studied under steady state.

Section 1.4.4 – The Petri Net Formalism

This project extensively used the Petri Net framework to provide a graphical language and

tools for model building, integration of models developed in different formalisms and the

notion of qualitative dynamic modelling. The Petri Net formalism was first developed by

Carl Adam Petri in the 1962 dissertation, Kommunikation mit Automaten (Petri, 1962).

While this formalism was developed to model engineering systems (Ibid), the modelling of

chemical reactions has also been proposed. Thus, the formalism can be easily extended to

model biochemical reactions or cellular signalling events. Before a biologically meaningful

Page 55

description of this formalism is given in Section 1.4.6 – Translation of Petri Net Terminology

for a Molecular Biologist, a formal definition is necessary.

Petri Nets are bipartite graphs in which two types of nodes represent places and transitions

(Petri, 1962). Various types of edges link places to transitions and, given Petri Nets are

bipartite graphs, no two nodes of the same type can be linked together with edges. Pre-

places describe places that are linked to transitions with an edge instigated from that place,

whilst a post-place is one linked to a transition by an edge originating from that transition

(Wang, 2007; Breitling et al., 2008). Places within the model are assigned non-negative,

integer number of tokens. The assignment of tokens to places is termed ‘marking’ (Breitling

et al., 2008). The firing of transitions moves tokens from pre-places to post-places via

transitions and results in the state of the system being updated (Ibid). The consumption or

production of multiple tokens by one transition firing is possible. The number of tokens

moved upon transition firing is specified by the weight of the edges linking pre- and post-

places to transitions (Breitling et al., 2008). Each transition is enabled when the associated

firing rule, concerning the marking of pre-places with sufficient tokens to account for the

weighting of edges, is satisfied (Breitling et al., 2008).

Definitions of Petri Nets found in literature differ slightly in detailed notation (Wang, 2007;

Breitling et al., 2008). Following Wang (2007), Petri Nets can be defined as a quintuple (𝑃,

𝑇, 𝐼, 𝑂, 𝑀0):

𝑃 = {𝑝1, 𝑝2, … , 𝑝𝑛} and 𝑇 = {𝑡1, 𝑡2, … , 𝑡𝑚} are finite sets of places and transitions in

which 𝑃 ∪ 𝑇 ≠ ∅ and 𝑃 ∩ 𝑇 = ∅

Page 56

𝐼: 𝑃 × 𝑇 → 𝑁 describes an input function defining edges originating from pre-places,

with 𝑁 being a non-negative integer set

𝑂: 𝑇 × 𝑃 → 𝑁 describes an output function defining edges linking transitions to post-

places, similarly, 𝑁, is a set of non-negative integers

𝑀0: 𝑃 → 𝑁 is the initial marking.

Changing the distribution of tokens, through the firing of enabled transitions in Petri Nets,

allows the complex and dynamic behaviour of a modelled system to be represented (Wang,

2007). The transitions, which can be fired are called ‘Enabled Transitions’. The transition is

enabled if the number of tokens marking all pre-places for transitions, 𝑡, are greater than

the weighting of edges. To put this another way, for place 𝑝:

𝑀(𝑝) ≥ 𝐼(𝑡, 𝑝)

Since enabled transitions are the only ones that can fire, the markings on places can only be

non-negative integer values. When a transition fires, tokens marking pre-places are

removed by the amount equal to the weighting of the edges connecting the place to the

transition (Wang, 2007), as shown in Figure 1.4.4a. Concomitantly, tokens are added to the

post-places with the increase in marking being equal to the weighting of edges linking the

transition to the place (Ibid). The new marking, 𝑀′, can be described mathematically in the

following way for transition, 𝑡, and place, 𝑝:

𝑀′(𝑝) = 𝑀(𝑝) − 𝐼(𝑡, 𝑝) + 𝑂(𝑡, 𝑝)

Equation 1.4.4b

Equation 1.4.4a

Page 57

Figure 1.4.4a – Example of a Simple Petri Net. Taken from Wang (2007), this Petri Net contains

three transitions (𝑡1, 𝑡2 and 𝑡3) and four places (𝑝1, 𝑝2, 𝑝3 and 𝑝4). Given Petri Nets are a form of

bipartite graph, places can only be linked to transitions and vice versa. Edges are used to link

places to transitions and transitions to places. The edge linking the pre-place 𝑝1 to transition 𝑡1

has a weighting of two, meaning two tokens of species 𝑝2 are required for the transition to fire.

Once 𝑡1 fires, two tokens of the post-place, 𝑝2, and one token of post-place, 𝑝3 are produced.

Only one token of places, 𝑝2 and 𝑝3 are required to fire transitions, 𝑡2 and 𝑡3, respectively. The

firing of either of these transitions results in the movement of one token, in each iteration of the

model, to place, 𝑝4.

𝑝1

𝑝3

𝑝2

𝑝4

𝑡1

𝑡2

𝑡3

2

2

Page 58

In the following case study, taken from Wang (2007), the state of the Petri Net shown in

Figure 1.4.4a, can be described in the following way (Equations 1.4.4c – 1.4.4k):

𝑃 = {𝑝1, 𝑝2, 𝑝3, 𝑝4} ;

𝑇 = {𝑡1, 𝑡2, 𝑡3, 𝑡4} ;

𝐼(𝑡1, 𝑝1) = 2, 𝐼(𝑡1, 𝑝𝑖) = 0 for 𝑖 = 2, 3, 4 ;

𝐼(𝑡2, 𝑝2) = 1, 𝐼(𝑡2, 𝑝𝑖) = 0 for 𝑖 = 1, 3, 4 ;

𝐼(𝑡3, 𝑝3) = 1, 𝐼(𝑡3, 𝑝𝑖) = 0 for 𝑖 = 1, 2, 4 ;

𝑂(𝑡1, 𝑝2) = 2, 𝑂(𝑡1, 𝑝3) = 1, 𝑂(𝑡1, 𝑝𝑖) = 0 for 𝑖 = 1, 4 ;

𝑂(𝑡2, 𝑝4) = 1, 𝑂(𝑡2, 𝑝𝑖) = 0 for 𝑖 = 1, 2, 3 ;

𝑂(𝑡3, 𝑝4) = 1, 𝑂(𝑡3, 𝑝𝑖) = 0 for 𝑖 = 1, 2, 3 ;

𝑀0 = (2 0 0 0)𝑇

If the enabled transition 𝑡1 then fires, the new marking becomes:

𝑀1 = (0 2 1 0)

The distribution of tokens following this is such that sufficient tokens are then present to

enabled transitions 𝑡2 and 𝑡3 (Wang, 2007). If 𝑡2 fires next, the token markings become:

𝑀2 = (0 1 1 1)

Should 𝑡3 then fire, the new token marking series would become:

𝑀3 = (0 1 0 2)

Equation 1.4.4c

Equation 1.4.4d

Equation 1.4.4e

Equation 1.4.4f

Equation 1.4.4g

Equation 1.4.4h

Equation 1.4.4i

Equation 1.4.4j

Equation 1.4.4k

Equation 1.4.4l

Equation 1.4.4m

Equation 1.4.4n

Page 59

In the Standard Petri Net formulation described so far, there is only one type of an edge

linking places and transitions. In Extended Petri Nets, additional edge types are used to

introduce additional rules. To model the action of enzymes and catalysis in biochemical

systems, a read edge is used to link a pre-place to a transition (Breitling et al., 2008). If this

pre-place is marked with the necessary number of tokens, the transition is enabled. The

tokens marked on this pre-place are not consumed during the firing. Similarly, inhibitory

edges are used to express the act of inhibition (Ibid). If a pre-place is linked to a transition

through an inhibitory edge and this pre-place is marked with an adequate number of tokens

to overcome the threshold value, then the firing of the transition is inhibited. An example of

a Petri Net containing these differing types of edges is given in Figure 1.4.4b which also

illustrates how the Extended Petri Net notation can be used to construct signalling cascades.

Page 60

Figure 1.4.4b – Extended Petri Net Notation. i. The extended Petri Net notation includes read edges to

simulate the action of an enzyme. In this pre-place, linked to a transition by such an edge must be marked with

sufficient tokens to overcome the threshold to satisfy the firing rule but these tokens are not consumed during

the firing of the transition. An inhibitor edge is used to simulate the action of an inhibitor. When the inhibitor

is marked with sufficient tokens, the transition it is linked to is inhibited. When the inhibitor is absent the

transition is capable of firing. ii. Many examples of enzymatic cascades exist in molecular biology. The

Extended Petri Net notation enables the construction of enzymatic cascades, including the Mitogen-Activated

Protein Kinase cascades. In such a cascade, the most upstream kinase (here KinaseA) is activated by

phosphorylation in response to a particular environmental stimulus activating a receptor. In the

phosphorylated form, pKinaseA is able to activate via phosphorylation, a further kinase (KinaseB). In this

example phosphorylated KinaseB (pKinaseB) can phosphorylate KinaseC. Phosphorylated KinaseC (pKinaseC)

would then be free to act on downstream effectors, such as further kinases or transcription factors. Individual

reactions, or transitions in Petri Net terminology, are denoted as R1 to R6. The forward reactions (R1, R3 and

R5) are responsible for converting kinases to the phosphorylated kinases whereas reverse reactions (R2, R4 and

R6) are those that mediate the dephosphorylation of kinase. Such dephosphorylation of kinases would

normally require the action of a phosphatase.

ii

i

Page 61

Petri Net semantics refers to the algorithm used to fire enable transitions and the rules that

govern which transitions are enabled. Simulations, within Petri Net terminology, are

referred to as Token Game simulations, with the result of such a simulation being termed a

‘trajectory’ or ‘trace’ (Breitling et al., 2008). In the semantics used in the example above

(Figure 1.4.4b), if sufficient numbers of tokens are present in each pre-place, a transition is

enabled (Ibid). The simulation algorithm randomly chooses and fires one transition from a

list of enabled transitions, in each iteration. In such Petri Net simulations, the probability of

selecting an enabled transition is equal for all enabled transitions (Breitling et al., 2008) and

generates a trajectory detailing state changes but where such changes are not assigned

times. However, it is possible to determine if one transition is fired before or after another

(Breitling et al., 2008). Consequently, such a simulation is frequently termed time free.

The full dynamic simulation where integer token numbers represent physical quantities

changing in real time can be performed with Stochastic Petri Nets, where the rules of the

Gillespie algorithm are used to fire transitions and calculate waiting times. Each transition is

assigned a propensity function calculated as a product of the pre-place markings and

stochastic rate constant of the transition. The Gillespie algorithm (discussed in Section 1.4.1

– Exact Stochastic Simulations with the Gillespie Algorithm) is then used to determine

which transition will occur next and after what waiting time. Subsequently, a selected

transition is fired and time is increase by the waiting time.

It is possible to implement ODE simulations within Petri Nets. In a Continuous Petri Net

real, rather than integer, numbers of tokens are used. The rate equations are then assigned

to transitions and ODEs are constructed from the interactions in the Petri Net network

Page 62

(Breitling et al., 2008). State changes and simulations are carried out using the standard

methods of numerically integrating ODE systems (Breitling et al., 2008).

The unique feature of the Petri Net framework, which make it exceptionally suited for the

modelling of complex biological systems, is the ability to combine multiple Petri Net types in

to Hybrid Petri Net simulations (Breitling et al., 2007; Ghomri & Alla, 2007). Rather than

defining the entire model as Stochastic or Continuous, the Hybrid Petri Nets allow the

construction of a model from both stochastic and continuous transitions. Likewise,

Standard Petri Net places, with integer numbers of tokens, can be used alongside

continuous places with real number token marking. Furthermore, the Hybrid Petri Net

algorithm combines ODE simulations with the Gillespie algorithm to integrate different

types of places and transitions (Breitling et al., 2007).

Biological systems involve processes occurring at different spatial and temporal scales (Mast

et al., 2014). Different levels of cellular organisations are modelled in different formalisms.

For example, metabolism is modelled by FBA, while regulatory processes are usually

modelled at small scale by ODE systems. As demonstrated above, Petri Nets are likely to

become a commanding tool in computational systems biology.

When applied to molecular biology, Petri Nets use places to represent species such as

proteins, lipids or nucleotides, and transitions to denote the interactions and reactions

between places. Tokens denote the number of molecules and allow Token Game

simulations to be executed. In qualitative simulations, Token Game simulations represent

‘abstract quantities whose changes, over time, correlate to changes that occur in the

amounts of active proteins present in the cell’ (Ruths et al., 2008). As remarked by Ruths

and colleagues (2008), the ‘abstract’ numbers of tokens are similar to the qualitative levels

Page 63

of proteins visualised in Western Blots. Should quantitative parameters be available it is

possible to use Stochastic Petri Nets. Here the numbers of tokens marking a place represent

the number of molecules directly. For concentrations to be modelled, Continuous Petri Nets

may be used.

Section 1.4.5 – The Quasi-Steady State Petri Net Algorithm

The Quasi-Steady State Petri Net (Fisher et al., 2013) is a recent extension to the Petri Net

framework inspired by Systems Biology. The Hybrid Petri Nets are capable of integrating

Gillespie algorithm and ODE simulations. The QSSPN algorithm allows further integration of

Hybrid Petri Nets with Flux Balance Analysis, thus providing a unified framework for all three

major algorithms of Computational Systems Biology.

This project uses the QSSPN simulation engine (Fisher et al., 2013) to conduct all

simulations. While the majority of the simulations presented in this dissertation could have

been done with standard Petri Net tools, QSSPN was chosen to allow the integration of

models developed in this project with GSMNs in Chapter 5 – Integration of the

Translational Control Model with a Genome Scale Metabolic Network of a RAW 264.7 Cell.

Given that QSSPN is still a new approach and that the software is used extensively as a

simulation engine throughout this project, it is necessary to provide extensive background

to this method.

In QSSPN models, the transitions are divided into two sets: quasi-steady state fluxes (QSSF)

and dynamic transitions (DT) (Fisher et al., 2013). The method then assumes quasi-steady

state: following any state change resulting from DTs, the QSSF quickly reaches steady state.

This requires time scale separation between two parts of the model with DTs being much

slower than QSSF (Ibid). This assumption is motivated and satisfied by biological systems

Page 64

where metabolic reactions are much faster than gene regulation and signalling processes.

The assumption that following state changes in the dynamic regulatory network, the

metabolic network quickly reaches steady state has already been used to integrate Boolean

models and ODEs with FBA. The result of such integrations are rFBA (Covert et al., 2001)

and iFBA (Covert et al., 2008). In QSSPN, the DT part of the model is described by Hybrid

Petri Net semantics allowing both stochastic and continuous transitions, while QSSF is

simulated in the Flux Balance Analysis framework. This makes QSSPN the first algorithm

allowing integration of all three major frameworks of Computational Systems Biology: exact

stochastic simulations, ODEs and FBA (Fisher et al., 2013).

The integration of QSSF and DT parts of the model is achieved through the introduction of

two new types of places: Constraint Places and Objective Places (Fisher et al., 2013). The

marking of the Constraint Place is determined by a dynamic simulation. The Constraint

Place then translates the marking to FBA flux bounds of selected transitions in QSSF (Ibid).

The Objective Place requests evaluation of the objective function and then translates the

value of this objective function into a marking. For example, if an FBA model of glycolysis

were to be linked to a Petri Net signalling network, the integration could be achieved by

using a glycolytic enzyme regulated by the signalling network. A glycolytic enzyme, such as

Phosphofructokinase, featured in the Petri Net model would serve as the Constraint Place.

This place would contain information on the flux boundaries of the reaction catalysed by

Phosphofructokinase in the form of an activity table. Such a table would state that if

Phosphofructokinase is marked with one token, the flux boundary is set, for example, at

(0.0, 100,000). However, if the enzyme is not marked with tokens, the FBA reaction would

be inactive with flux bounds set at (0.0, 0.0) (Fisher et al., 2013). When the QSSPN

Page 65

simulation is carried out, the qualitative Petri Net model would mark Phosphofructokinase

with a token. This would then update the flux bounds of the FBA model and, ultimately, the

objective function representing the metabolic function of interest, such as maxiamal Acetyl

Coenzyme A (Acetyl-CoA) production. This objective can then be represented as an

Objective Place and objective function fed back to the dynamic part of the model. For

example, one could set the activity table stating that if maximal Acetyl-CoA production is

greater than zero, the Objective Place is marked with one token. Should this condition not

be satisfied then no tokens would mark the Acetyl-CoA place. Given the diverse roles of

glycolysis and Acetyl-CoA (Moussaieff et al., 2015), or metabolites in general, the integration

of FBA models with Petri Net models would allow for a more complete understanding of

how metabolic shifts are affected by cellular signalling pathways and vice versa.

QSSPN uses a general formulation of a propensity function for each transition to facilitate

qualititative simulations with thresholds (Fisher et al., 2013), as shown in Equation 1.4.5a:

𝑃𝑡 = 𝑐𝑡∏𝜇𝑖(𝑥𝑖)

𝑁

𝑖=1

Where 𝑃𝑡 and 𝑐𝑡 describes the propensity function and rate constant of transition 𝑡,

respectively. 𝑁 is the number of pre-places for transition 𝑡, and 𝑥𝑖 is the marking of pre-

place 𝑖. The variable 𝜇𝑖 describes the acitivty of pre-place 𝑖 in the transition as a function of

its marking. The activity function, 𝜇, is a look-up table (Equation 1.4.5b) defining the

contribution of a pre-place to a propensity function of a specific transition (Fisher et al.,

2013). Within this table, 𝑇 thresholds, 𝑡𝑖, and activities, 𝑎𝑖, are contained and permit the

pre-place contribution to be calculated.

Equation 1.4.5a

Page 66

𝜇(𝑥) = {

𝑥 ∈ [𝑡1, 𝑡2), 𝜇(𝑥) = 𝑎1𝑥 ∈ [𝑡𝑖, 𝑡𝑖+1), 𝜇(𝑥) = 𝑎𝑖

…𝑥 ∈ [𝑡𝑇−1, 𝑡𝑇), 𝜇(𝑥) = 𝑎𝑇

}

For example, when the marking of the pre-place 𝑖 in transition 𝑡 exceeds the threshold 𝑡2,

but is smaller than the threshold 𝑡3, the activity function contributes value 𝑎3 (Fisher et al.,

2013).

Be it stochastic, continuous or immediate, the transition type lends an interpretation of the

propensity function of the transition. Regardless of the transition type, a propensity

function equal to zero denotes a disabled transition (Fisher et al., 2013) as it cannot fire in a

stochastic simulation scenario and does not contribute any state change in continuous

simulations. In the case of stochastic transitions, the propensity function is interpreted as

the probability density of a transition firing in the next infinitesimally small timestep. For

continuous transitions, the propensity function is used as the reaction rate in the ODE

simulation part of the hybrid algorithm. Immediate transitions are only capable of firing

when the propensity function is non-zero (Ibid). Each such transition is then only fired once

(Fisher et al., 2013). Both continuous and immediate transitions are fired according to the

function ‘fireDeterministicTransitions(Δ𝑡) (Ibid). The ODE part of the hybrid algorithm has

undergone significant upgrade since the first version of QSSPN solver was published in 2013.

Technical details of an adaptive timestep ODE solver used within

‘fireDeterministicTransitions(Δ𝑡)’ function are beyond the scope of this dissertation.

Based on the marking of a constraint place, the upper- and lower-bounds of the fluxes in the

QSSF model component are set using another look-up table (Equation 1.4.5c) (Fisher et al.,

2013). Constraint Places are potentially linked to a number of fluxes within the QSSF model

Equation 1.4.5b

Page 67

component. The look-up table related to a specific Constraint Place, using the number of

tokens marking this node, sets the bounds of the fluxes (Fisher et al., 2013). In the

simplified example above, such as Constraint Place would be Phosphofructokinase. The

function (setQSSFbounds()) is used to this end (Ibid).

(𝑙𝑏(𝑥), 𝑢𝑏(𝑥)) =

{

𝑥 ∈ [𝑡1, 𝑡2), 𝑙𝑏(𝑥) = 𝑓𝑙,1, 𝑢𝑏(𝑥) = 𝑓𝑢,1𝑥 ∈ [𝑡𝑖 , 𝑡𝑖+1), 𝑙𝑏(𝑥) = 𝑓𝑙,𝑖, 𝑢𝑏(𝑥) = 𝑓𝑢,𝑖

…𝑥 ∈ [𝑡𝑇−1, 𝑡𝑇), 𝑙𝑏(𝑥) = 𝑓𝑙,𝑇 , 𝑢𝑏(𝑥) = 𝑓𝑢,𝑇}

In this look-up table, the upper and lower bounds of flux are described by the variables

𝑢𝑏(𝑥) and 𝑙𝑏(𝑥), respectively, while the 𝑥 denoted the the number of tokens, or state of

the node (Fisher et al., 2013). The variables describing the number of token thresholds are

given by 𝑡1, … , 𝑡𝑇. 𝑓𝑙,𝑖 and 𝑓𝑢,𝑖, according to the state of the node, are the lower and upper

flux bounds respectively (Fisher et al., 2013). A Boolean variable (‘UpdateRequired’) set to

TRUE is required to trigger an evaluation of the objective function of the FBA model

component if the the lower or upper flux bounds change. Once this has occurred, and prior

to the looping over of Constraint Place, the Boolean variable is re-set to FALSE (Ibid).

A QSSPN simulation, via the function evaluateObjective(), uses an Objective Place, for

example Acetyl-CoA in the simplified case given earlier in this section, to evaluate the

objective function in an FBA model. Within the QSSF model component, the objective

function may be a flux name or metabolite name. As has been stated previously, the

objective function is maximised by the Simplex algorithm (or other Linear Programming

methods) in the case of the former, or, in the latter, by maximising the sums of the fluxes

producing the objective function metabolite (Fisher et al., 2013). The most computationally

Equation 1.4.5c

Page 68

demanding stage of the QSSPN algorithm occurs when the objective function requires

evaluation by linear programming-mediated maximisation of the entire QSSF model

component (Ibid). Such evaluation of the objective function comes about through the

variable ‘UpdateRequired’ being set to TRUE. Moreover, such a variable ensures that the

evaluation only occurs when the QSSF bounds have changed (Fisher et al., 2013).

The dynamic part of the model requests information evaluation of the capabilities of steady

state metabolism using the function, updateObjectiveNode() (Fisher et al., 2013). Such a

function, based upon the value of the objective function, sets the state of an Objective

Place. The state of the Objective Place is set using another look-up table (Fisher et al., 2013)

(Equation 1.4.5d).

𝑥(𝑜) = {

𝑜 ∈ [𝑡1, 𝑡2), 𝑥(𝑜) = 𝑠1𝑜 ∈ [𝑡𝑖, 𝑡𝑖+1), 𝑥(𝑜) = 𝑠𝑖

…𝑜 ∈ [𝑡𝑇−1, 𝑡𝑇), 𝑥(𝑜) = 𝑠𝑇

}

In this, 𝑥(𝑜) is the state of the Objective Place. Variable 𝑡1, … , 𝑡𝑇 defines the objective

function thresholds while 𝑠1, … , 𝑠𝑇 are integer token numbers marking an objective node

(Fisher et al., 2013). The number of tokens marking this Place are determined by the

magnitude of the objective function. In this look-up table, the variable 𝑜 denoted the

objective function (Ibid). The calculation of the objective function value is performed by the

variable ‘evaluateObjective()’ (Fisher et al., 2013).

Boolean values are used to fire any transitions requiring a delay of (𝑡, 𝑡 + 𝑡). TRUE or FALSE

values are returned when such transitions are present or absent, respectively (Fisher et al.,

2013). When a delayed transition is fired, at the scheduled time (𝑡𝑠), the simulation

timestep (∆𝑡) is set to (𝑡𝑠 − 𝑡). In a system with multiple delayed transitions, these are fired

Equation 1.4.5d

Page 69

sequentially with the order being determined by the magnitude of the 𝑡𝑠 (Fisher et al.,

2013). The function ‘fireDelayedTransition(𝑡, 𝑡)’ is used for handling delayed transitions.

Stochastic transitions with a non-zero delay time are handled by including them with

delayed transitions using the function ‘scheduleDelayTransition(𝑚)’. The delay of stochastic

transitions is governed by the waiting time calculated using the Gillespie algorithm (Fisher et

al., 2013).

Now that basic operations within QSSPN method have been defined, it is possible to outline

the algorithm. The algorithm begins by evaluating the number of tokens marking each

Constraint Place. The state of these nodes updates the bounds of the QSSF model

component. The objective function of the Objective Place is then determined using FBA

maximisation (Fisher et al., 2013). Following this, using the processes described above,

immediate and continuous transitions followed by stochastic transitions are executed. The

firing of stochastic transitions is governed by the Gillespie algorithm (described in detail in

Section 1.4.1 – Exact Stochastic Simulations with the Gillespie Algorithm). Synchronisation

of stochastic immediate and continuous transitions is implemented using the Maximal

Timestep Method algorithm (Puchalka & Kierzek 2004). Delayed stochastic transitions are

permitted. Repeated iterations of this lead to the generation of simulation trajectories.

Such trajectories are dynamic and permit the order of transition execution to be monitored.

This allows identification of the effects of altering the initial marking of Constraint Places on

the Objective Place and the associated objective function (Fisher et al., 2013). The

algorithm is summarised in Figure 1.4.5a.

Page 70

Figure 1.4.5a – Summation of the QSSPN Algorithm. Taken from Fisher et al. (2013), this figure

summarises how the QSSPN algorithm orders various types of transitions and integrates the differing

model types that form a QSSPN model.

Page 71

Section 1.4.6 – Translation of Petri Net Terminology for a Molecular Biologist

When applied to molecular biology, Petri Nets use places to represent molecular species

such as proteins, lipids or nucleotides, and transitions to denote the interactions between

the molecules (Breitling et al., 2008). In this notation, marking of Petri Net places

represents amounts of molecular species. In Stochastic Petri Nets, the integer number of

tokens represents numbers of molecules in the volume of the cell, cellular compartment or

cell micro-environment. In Continuous Petri Nets, the marking is a real number

representing molecular concentrations. In qualitative simulations, the integer number of

tokens represents abstract, discrete activity levels. As remarked by Ruths and colleagues

(2008), the ‘abstract’ numbers of tokens are similar to the qualitative levels of proteins

visualised in Western Blots.

This project capitalised on the Extended Petri Net graphical notation. The read edges used

in the model of translational regulation created in this project denoted activation where, for

example, there is a requirement for an enzyme or molecular co-factor that is not consumed

by the reaction. Inhibitory edges are used in situations where there is a need to represent

inhibition: the reaction rate is decreased by the presence of the inhibitor and, furthermore,

the inhibitor is not consumed by the transition.

In the opinion of this author, one of the major challenges of interdisciplinary research is that

it requires researchers, working in different, well established scientific fields, to agree on

common terminology. The development of such a common language is frequently more of

a social than scientific issue: the experts in individual fields have to accept that the language

in which interdisciplinary research is conducted will be different than the frequently strict

terminology of their field of expertise. This frequently creates barriers to interdisciplinary

Page 72

work. This project is a good example of such a challenge. Should it use Petri Net

terminology which is acceptable to Petri Net experts, but difficult to read for a molecular

biologists or should it use molecular biology terminology that a Petri Net expert may find

inacceptable as a language of algorithm description? Should the variable of interest be

referred to as places or molecular species? Should mathematical terms that change

variables be called transitions or interactions?

This project will use original Petri Net terminology in the above sections (Section 1.4.4 – The

Petri Net Formalism and Section 1.4.5 – The Quasi-Steady State Petri Net Algorithm),

where Petri Net methodology is described. However, subsequent chapters will use

molecular biology terminology and will refer to the Petri Net terms in the way given at the

start of this section. This project uses published computational methods to investigate

biological problems. It does not introduce new algorithms. The results of this project are in

the biological domain and will therefore be described in the language appropriate to

molecular biology.

Section 1.5 – Aims & Objectives and Novel Contributions

to Knowledge

Section 1.5.1 – Aims & Objectives

This project encompasses two main areas: first, the modelling of mammalian translation

initiation and the upstream regulatory pathways and secondly, modelling the effects viral

infection has upon this process.

The factors involved in translation initiation and the regulation of this process are known to

play roles in the pathogenesis of medically important disease states, including heart

Page 73

diseases (Lu et al., 2014), cancer (Heikkinen et al., 2013) and neurological disease (Li et al.,

2004; Li et al., 2005a). A better understanding of this process, and also the regulatory

pathways surrounding it, are essential to understanding the pathology of human and animal

disease. To this end, the overall aim of this project can be regarded as:

Constructing and experimentally validating a literature-based, qualitative model of

translation initiation and the signalling network that regulates this process in mammalian

cells.

In terms of the second aspect of this project, viral infection, at least from a systems biology

standpoint, can be viewed as ‘equivalent to a systems-level perturbation’ to the cell

(Garmaroudi et al., 2010). Investigating the effects viral infection has upon the cellular

translational machinery has not been attempted. The second part of the project is to model

viral infection. The aim of this part of the project is to:

Introduce the host-pathogen interactions into the model of translational regulation and

establish the effects of these interactions on both the cellular translational machinery and

the regulatory signalling network.

The specific objectives of this work are:

Construct and iteratively refine a large-scale, qualitative model representing cellular

translation initiation and the signalling pathways involved in regulating this process

in a mammalian system

Validate the predictive power of this model by a comparison of model behaviours

with literature data

Page 74

Apply the model to generate novel hypotheses on the regulation of translation

initiation and experimentally validate these hypotheses

Simulate the effects of viral infection and experimentally validate predicted

behaviours

Improve the predictive power of the model through the incorporation of detailed

kinetic models of major regulatory hubs into the large-scale qualitative model

Integrate the model of translational regulation with a GSMN to facilitate the study of

the coordinated translational regulation with whole-cell metabolic reprogramming

Section 1.5.2 – Novel Contributions to Knowledge

This project produced a number of key outputs which can be regarded as novel. These

include:

The production of an experimentally validated model from over 1,100 pieces of

peer-reviewed literature of mammalian translation initiation regulated by a

large-scale signalling network

Developing a benchmark concerning the effects of chemical inhibitors on key

signalling effectors and using this to develop a method for the quantitative

analysis of the predictive power of the model

Theoretical insight into the mechanism by which Rapamycin-mediated mTORC1

inhibition can lead to an increase in eIF4E phosphorylation.

Extending and experimentally validating the model to investigate the effects of

viral infection upon the translational machinery and the regulatory signalling

network

Page 75

Constructing a model setting the foundation of work looking to link the

regulation of translation initiation to the regulation of metabolism

Page 76

Chapter 2 – Model Creation &

Refining

Page 77

The construction of the model of translation initiation regulated by a large-scale signalling

network was the main aim of this project. This chapter describes the work undertaken to

not only construct the model from existing literature, but also to evaluate the predictive

power of the model. Furthermore, detailed in this chapter is work to establish a mechanism

by which the main model prediction occurs.

Section 2.1 – Materials & Methods

Section 2.1.1 – Qualitative Petri Nets

Despite recent advances in quantitative computational biology, it is still not possible to

establish the rate constants of all the molecular interactions in large-scale models (Di

Ventura et al., 2006). A solution to this challenge, which is gaining acceptance in the

literature is to use qualitative, time-free Petri Net modelling (Ruths et al., 2008; Fisher et al.,

2013). The Petri Net methodology has been described in Section 1.4.4 – The Petri Net

Formalism. This Section and Section 2.1.2 – The Binary Model Formalism describe details

of the specific time-free Petri Net approach used to build the translation initiation model.

The terminology of the Petri Net field is translated into the vernacular more suited to

molecular biology in Section 1.4.6 – Translation of Petri Net Terminology for a Molecular

Biologist.

This dissertation presents an Extended Petri Net model encompassing the translational

machinery and surrounding signalling pathways. Places represent molecular species and

transitions denote the interactions between molecules. Read edges model the activation of

interactions by certain molecular species and inhibitory edges are used to model the act of

inhibition. Since there is an absence of sufficient quantitative information regarding the

numbers of molecules in the cell to parameterise large-scale models, discrete activity levels

Page 78

were used to represent molecular activity levels. Likewise, the time is not parameterised to

time units. Time is exclusively used to state the order in which molecular interactions

occurred during simulations. Therefore, the model is used to simulate the sequence of

molecular events changing the activity levels of molecular species. In subsequent parts of

this thesis, the sequence of molecular events will be referred to as a ‘trajectory’.

Section 2.1.2 – The Binary Model Formalism

The definition of the molecular activity levels, and the rules describing how molecular

interactions change them, is the key aspect of the qualitative modelling approach. One

possibility involves using multiple activity levels representing concentration ranges of

molecular species. Activity levels are represented by integer numbers (tokens (or discrete

activity levels)) and changed by interactions decreasing states of substrates (or pre-places in

Petri Net parlance) and increasing the states of products (post-places). Calder et al. (2005)

analysed different maximal numbers of discrete activity levels in qualitative models of

signalling pathways and concluded that experimental data can be approximated with ten

levels. Ruths and colleagues (2008) used a different strategy placing certain discrete activity

levels numbers in the initial state of the signalling network and allowing Petri Net molecular

interactions to exchange them without setting any upper limit. The number of discrete

activity levels was then considered to be proportional to the concentration of that particular

molecular species (Ibid).

The aim of this project is to build a large-scale model of the signalling network controlling

protein synthesis. The resulting model is considerably larger than previously published

discrete models (Calder et al., 2005; Ruths et al., 2008). For that reason it has been very

difficult to calibrate the maximal number of discrete activity levels, or maximal numbers of

Page 79

tokens, in the initial state. To do this, calibration would have to be repeated for every new

version of the model as substantial increases to the model, in terms of scale and complexity,

could potentially affect the choice of arbitrary discretisation of parameters such as maximal

numbers of activity levels or the numbers of tokens in the initial states. Evaluation of the

predictive power of the large-scale model is a complex exercise and it was not practical to

repeat it frequently enough to optimise discretisation during model development.

Therefore, more abstract, binary modelling formalisms have been developed, where

molecular species were considered to be active or inactive and intermediate activity states

were not considered. The original contribution of this work is the approach of setting the

rules changing active/non-active species states, in such a way that biological processes are

reflected with sufficient realism, to achieve statistically significant and applicable predictive

power.

This work employs the following binary modelling approach. Each molecular species in the

network can have either zero or one discrete activity levels. The state of zero discrete

activity levels indicates that the molecular species is not present, in a sufficient amount,

to activate interactions it initiates or is a substrate in. The state of one discrete activity

level indicates that there is a sufficient amount of a molecular species to ensure that

molecular interactions consuming this species, or are activated by this species, can take

place. While the state of one corresponds to a higher amount than a state of zero, the

proportionality between amount and discrete level is not assumed. This is justified by the

fact that many, if not most, reactions in the signalling network follow a sigmoidal

input/output relationship (Kothamachu et al., 2013).

Page 80

Having this basic notion of a molecular species present in a sufficient amount to activate a

downstream molecular process, one can proceed to the construction of the signalling

network model. The elementary step of signalling transduction is an enzyme-catalysed

covalent modification of a protein molecular species. Protein A is modified by the

attachment of a chemical moiety, such as phosphorylation, acetylation and ubiquitination.

In a modified form, Protein A can be considered to be a distinct chemical species, termed

Protein B. Enzyme E is a protein that catalyses this post-translational modification of

Protein A. In this signalling cascade, Protein B is an enzyme catalysing the next step of signal

transduction and E is a protein activated by covalent, post-translational modification during

the previous step. This signalling network could not function if the activation of B was

permenant as it would lead to the permanent activation of downstream processes, even

when the external signal initiating this cellular response, for example a cytokine, was no

longer present. In the cell, covalently modified proteins are inactivated, either by the

removal of the chemical moiety or by the degradation of a protein. Both of these processes

are catalysed by specific enzymes, for example, phosphatases or a proteasome complex. In

both cases the signalling system is returned to a basal level of activity and is ready to

respond to a new round of stimulation. If Protein B is inactivated by the removal of the

post-translational modification, then the concentration of Protein B falls and a concomitant

rise in the concentration of Protein A is observed. In this case the concentration of Protein

B returns to a basal level. However, in the case of the degradation of Protein B, the amount

of Protein A is maintained by constitutive protein synthesis. Frequently, the literature

describing signal transduction networks describes the events of covalent modification

leading to forward propagation of the signal. The details of the biological mechanisms

surrounding how a return to the basal levels of signalling components is achieved, upon

Page 81

removal of the stimulus, remain largely undefined. In order to make a model as realistic as

possible, one still needs to assume the existence of mechanisms to effectively ‘switch off’

signal propagation, and ensure the system returns to a basal state capable of reacting to

repeated exposure to the stimulus.

Figure 2.1.2a shows formalism used in the construction of the binary model, using the

information presented earlier in this chapter, to represent the basic signal transduction

step. First, the connectivity of molecular species and biological processes are modelled

using the Extended Petri Net formalism (detailed in Section 1.4.4 – The Petri Net

Formalism), as shown in Figure 2.1.2ai. In reaction r1, Protein A is covalently modified to

form Protein B. Common edges are used as this process must decrease Protein A levels in

order to increase Protein B levels. Reaction r2 is used to represent the net effect of

potentially multiple, perhaps not clearly defined, molecular mechanisms that return the

signal transduction apparatus to a basal level. This molecular interaction serves to deplete

the pool of Protein B and increase the level of Protein A and further requires the use of

common edges. Enzyme E is connected to reaction r1 by read edges to represent catalysis.

Where the mechanism underlying reaction r2 can be directly attributed to a specific

enzyme, for example a phosphatase, a read edge would also be used. The situation

presented here, and in the more common of situations, is where the mechanism of

inactivation is not known in detail.

Page 82

Figure 2.1.2a – Overview of the Binary Model Formalism. Despite the advances in computational systems biology over the past decade, it is still not possible to produce large-scale quantitative models of biological systems (Di Ventura et al., 2006). In order to construct a biologically meaningful model of mammalian translation initiation regulated by a large-scale signalling network it was necessary to introduce the Binary Model Formalism. i. In a basic binary model the amount of a molecular species present can be thought of as the amount necessary to induce a particular behaviour. Consequently, one discrete activity level of sufficient amounts of Protein A and Enzyme E are necessary within a cell to enable Protein B formation, through post-translational modification, in a reaction termed r1. As Enzyme E is merely a catalyst for r1 this molecular species is linked to the reaction via a read edge. Reaction r2 is the reverse reaction in which the pool of Protein B is returned to the basal amount of Protein A, and, in a biological system may require an enzyme such as a phosphatase. Such mechanism are, in many cases, ill-defined but necessary to include in a model of a biological system. ii. This formalism produces random oscillations which are unrealistic in a biological system as, although such systems are stochastic in nature, the majority of Protein A (shown in red) will be modified to Protein B (shown in green) until such time as Enzyme E is removed through physiological feedback mechanisms. iii. In order to eliminate these random oscillations, rules where introduced, via an inhibitory edge between Enzyme E and reaction r2, to ensure that until the stimulus (Enzyme E) is removed, the pool of Protein B is maintained to allow potential downstream signalling to occur. iv. The Token Game simulation of this model demonstrates that oscillations are removed, with the pool of Protein B (shown in green) being maintained while the amount of Protein A (shown in red) is depleted. v. It is common in biological systems to contain feedback mechanisms in order to allow only the appropriate activation of signalling pathways. The inclusion of rules in the model accounts for such mechanisms. When the pool of Enzyme E is depleted, possibly due to a feedback mechanism originating from Protein B, reaction r2 can then occur to return Protein A to a basal level. This system is then capable of responding, should the cell once again be exposed, to the environmental cue leading to Enzyme E activation. vi. Following the removal of the Enzyme E-mediated inhibition of reaction r2, the depletion of Protein B (shown in green) occurs through the removal of the post-translational modification and restores the basal level of Protein A (shown in red).

Page 83

Figure 2.1.2aii shows the trajectory resulting from a Token Game simulation of the model

presented in Figure 2.1.2ai. This starts from the state where one discrete activity level is

assigned to the molecular species representing Protein A, and no discrete activity levels are

assigned to the Petri Net molecular species representing Protein B. This dynamic has the

following interpretation in the modelling framework used over the course of this project:

The cellular pool of Protein A is depleted below the amount sufficient to allow the enzyme-

catalysed reaction r1 to occur, while the cellular pool of Protein B rises to the level required

for reaction r2 to happen. This system then returns to the initial state and the process

repeats. The resulting oscillatory nature is unrealistic. If Protein A and Enzyme E are

present in the cell, in sufficient amounts to function, the protein pool should shift towards

the covalently modified form. This situation should remain until Enzyme E is inactivated by

an upstream process, in which case the system should return to the basal state. This

behaviour is more realistic in a biological system. It can be achieved by the addition of the

following rule to the structure of the biological process: Add an inhibitory edge between

Enzyme E and reaction r2, as shown in Figure 2.1.2aiii. This rule expresses the following

biological understanding: If Protein A and Enzyme E are present in sufficiently large

amounts, the net rate of all elementary molecular processes covalently modifiying Protein A

is faster than the net rate of molecular processes returning Protein A to a basal level. The

Token Game simulation of the modified model is shown in Figure 2.1.2aiv. The system has,

therefore, moved to a state where the covalently modified form of the protein is constantly

at a sufficient amount for the activation of downstream signalling. The system will remain

in this state until Enzyme E is depleted, after such time the system will return to a basal

state, as shown in Figure 2.1.2av and Figure 2.1.2avi.

Page 84

During model reconstruction, care was taken to build the model using the basic motif

described above, or other simple sub-networks, obviously maintaining the state in {0,1}

range and connected to the remaining part of the model only by read and inhibitory edges

that do not need discrete activity levels. This does not, however, constitute the formal

proof that under no circumstances will molecular species be assigned two discrete activity

levels. Obtaining such proof would require the use of invariant analysis of Petri Net theory

implemented in tools such as Integrated Net Analyzer (INA) (Roch & Starke, 2003) or Charlie

(Heiner et al., 2015). While these tools are integrated with Snoopy, they do not allow the

analysis of networks that have read and inhibitory edges. These edges are essential for the

formulation of the transition rules, required for tackling the biological complexity, present in

the large-scale model. Additionally, invariant analysis is prone to a combinatorial explosion

challenge (Sun & Zhang, 2011) and would be difficult for a full scale model, even assuming

such a model was constructed as a standard Petri Net. Consequently, in the absence of a

practical way of formally verifying that no molecular species in the model presented in this

dissertation can assume a state of more than one discrete activity level, a feature within the

QSSPN solver has been used to enforce this condition within the model simulations (Fisher

et al., 2013). For any molecular species, the maximal number of discrete activity levels has

been set to one. If the simulator attempted to increase the number of discrete activity

levels to a value of more than one, the number of activity levels was reset to one. Due to

the way in which the model was carefully constructed, this condition was rarely, if ever,

invoked and did not influence conclusions based upon the examination of the large number

of simulation trajectories.

Page 85

Section 2.1.3 – Model Construction

Section 2.1.3.1 – General Model Construction Process

The binary formalism, described in Section 2.1.2 – The Binary Model Formalism, has been

used to represent literature on molecular events controlling translation initiation in

mammalian cells. An illustration of this is shown earlier in Figure 1.1a and again in the

information detailed in Section 1.1 – Cellular Translation Initiation and Section 1.2 – Key

Points of Regulation & Upstream Signalling Pathways, as an executable, computational

model capable of making predictions on the effects of molecular perturbations on the

initiation of protein synthesis.

A comprehensive search of the PubMed (US National Library of Medicine, Bethesda, MD

(US)) database has been conducted for the articles, describing molecular events, in the

system under investigation. The use of the PubMed database was to ensure that only peer-

reviewed work was included and that it can be referenced in a rigorous format by giving a

PubMed Identifier (PMID) numbers. Given the differences in translation that exist between

yeast and mammalian cells, such as in the composition of the Multi-Factoral Complex in

yeast (Fletcher et al., 1999; Bieniossek et al., 2006; Sokabe et al., 2012), the articles

describing the process in mammalian species have, without exception, been prioritised over

articles describing the process in yeast. However, where the process was described only in

experiments conducted in yeast, this work has been used to fill gaps in the literature. The

scope of the literature search has been limited to the signalling pathways that are directly

involved in the regulation of translation initiation in mammals. These pathways include the

PI3K, mTORC1, mTORC2 and the MAPK pathways: ERK1/2, p38α, JNK1 and ERK5. It was

decided that these signalling pathways would be extended back to a cell surface receptor

and a literature search has been conducted accordingly.

Page 86

The experimental biology literature articles, which provide evidence for individual

interactions, are used to construct the rules of the binary model. As detailed in Section

2.1.2 – The Binary Model Formalism, every effort has been made to use the basic switch

motif, in which the exchange of single discrete activity levels between forms of covalently

modified proteins is controlled by both read and inhibitory edges. However, the use of

more complex motifs has, sometimes, been needed when this was not sufficient to

represent literature data in a meaningfull and accurate way. As indicated in a number of

studies (Wang et al., 2013; Green et al., 2014), there is a growing need for standardised

nomenclature. In order to construct the model in a transparent and open way, protein,

gene and micro RNA (miRNA) components of the model are labelled with Uniprot (The

Uniprot Consortium, 2014), Entrez (NCBI Resource Coordinators, 2013) and miRBase

(Kozomara & Griffiths-Jones, 2014) database identifiers, respectively, as shown in

Supplementary Information E1 (available on the attached Flash Drive). Similarly, a list of the

papers used in the construction of the model is available in Supplementary Information E2.

The focus of this section will now turn to describing the construction of the various parts of

the model in the software Snoopy (Rohr et al., 2010; Marwan et al., 2012; Heiner et al.,

2012). Section 1.2 – Key Points of Regulation & Upstream Signalling Pathways details the

biological particulars of each key pathway included in the model and, as a result, this section

will focus on illustrating the Petri Net versions of each model pathway and on describing the

rationale behind the inclusion of various aspects of the pathways.

The model details the synthesis of nascent polypeptides and describes the assembly of 80S

ribosome (Figure 2.1.3.1a). This component of the model was constructed using

information taken from multiple peer-reviewed literature sources. As already shown in

Page 87

Figure 1.2a, this component of the model comprises only a small part of the overall model

produced during the course of this work.

Page 88

Figure 2.1.3.1a – Petri Net Representation of Translation Initiation. Produced in the software Snoopy, the

translation initiation pathway is represented using the Binary Model Formalism. This Petri Net model was

extracted from the large-scale model produced over the course of this project and extends back to include

the key points of regulation. Although this figure includes key points of regulation, it is analogous to Figure

1.1a.

Page 89

One of the pathways most associated with the regulation of translation initiation is that of

ERK1/2. As such, it was necessary to include a model of this pathway. It was decided to

attempt to use an existing model of the activation of ERK1/2. The BioModels database (Li et

al., 2010) was searched for such a model and one was selected (BioModels identifier:

BIOMD0000000270). The model constructed by Schilling et al. (2009) concerned the

activation of ERK1/2 via MEK1/2 in response to known ligand Erythropoietin (Seong et al.,

2006; Nairz et al., 2011). In order to fully integrate this model, as a qualitative component

of the larger model, it was necessary to reconstruct the model in the software Snoopy using

the Petri Net formalism. This reconstruction of the model involved expressing each reaction

given in the working version of the model in the software Copasi (Hoops et al., 2006;

Mendes et al., 2009) to a Petri Net (Figure 2.1.3.1b). To put this component in the same

notation as the rest of the model it was necessary to ensure that these molecular

interactions were expressed using the Binary Model Formalism. In keeping with the

qualitative nature of the model, all molecular amounts were replaced with a marking of one

discrete activity level, which as elsewhere in the model represented an amount of a

particular molecular species needed to provoke a biological response. Similarly, all rate

constants within the model were expressed as one.

Page 90

Figure 2.1.3.1b – Petri Net Representation of the ERK1/2 Signalling Pathway. The ERK1/2 signalling

pathway is a canonical component of the cellular signalling network and is integral to the regulation of

translation initiation. The model of the activation of ERK1/2 was taken from Schilling and colleagues

(2009). Once activated, ERK1/2 was able to phosphorylate numerous downstream components within

several other signalling pathways. Within this model, several inhibitors affecting ERK1/2 activation were

incorporated: 5Z-(7)-Oxozeanol (FR1), PD184352 (PD1) and PD98059 (PD9). This model was produced as a

Petri Net using the software Snoopy.

Page 91

Once reconstructed, the integration of this model with the full model was conducted. This

was first achieved by linking an early point within the model to an existing model

component. In this case it was achieved by linking the activation of the MEK1/2 kinase, Raf-

1 to Ras (MacDonald et al., 1993; Diaz-Meco et al., 1994; Fabian et al., 1994; Leevers et al.,

1994; Stokoe et al., 1994; Jelinek et al., 1996). The model already contained a number of

important ERK1/2 substrates, including MNK2 and RSK1, so it was possible to ensure the

model functioned within the full model. The final stage of this integration occurred during

the stages of model refinement. This included the modification of the model to take into

account the activation of ERK1/2 through additional pathways. These pathways included the

Nuclear Factor-κB-induced kinase, Tumour Progression Locus 2 (Tpl2) (Dumitru et al., 2000;

Eliopoulos et al., 2003; Rousseau et al., 2008; López-Pelaéz et al., 2012; Roget et al., 2012;

Hirschhorn et al., 2014) and PKCζ (Monick et al., 2000; Lesseux et al., 2008).

The other MAPK pathways were included in the model as they form part of the network that

regulates the translational machinery. The primary role of p38α in the regulation of

translation is given in Section 1.2.2 – eIF4E Phosphorylation and The Role of ERK1/2 & p38.

This pathway (Figure 2.1.3.1c) was reconstructed in the model from published, peer-

reviewed literature. Reconstructing this pathway necessitated including not only the kinases

involved in the activation of p38α, but also to extend this pathway to a cell-surface receptor.

In the case of p38α, this receptor was TLR4. It was not necessary to include this receptor per

se, as a basal level of p38α has been reported in a number of cell-types, including the type

used to experimentally validate the predictions generated by the model produced during the

course of this work (detailed in Chapter 3 – Experimental Validation of Model Predictions)

(Hsieh & Papaconstantinou, 2002; Carrozzino et al., 2009; Naidu et al., 2009; Bouazza et al.,

Page 92

2014). However, for completeness and to allow the effect of LPS stimulation to be studied in

future, it was included.

Page 93

Figure 2.1.3.1c – Petri Net Representation of the p38α Signalling Pathway. The p38α signalling pathway

regulates translation through controlling eIF4E phosphorylation and mRNA stability. This kinase plays a

role in regulating many other processes in the cell. This pathway is extended back to Toll-Like Receptor 4.

Key points of cross-talk between other signalling pathways are also included. These include the negative

regulation of Protein Kinase C ζ (PKCζ) and the inhibition of p38α by AKT1. This model was created in the

software Snoopy in the Petri Net notation.

Page 94

The JNK1 pathway (Figure 2.1.3.1d) was constructed in much the same way, in that the

pathway was extended back to a cell-surface receptor and that a basal level of activity

(Javelaud et al., 2003; Sabapathy et al., 2004; Carrozzino et al., 2009; Cui et al., 2009; Zhu et

al., 2016) was maintained. With regards to the activation of JNK1 the model included the

requirement, for the presence of the metabolite, ceramide (Mackichan & DeFranco, 1999;

Medvedev et al., 1999; Monick et al., 2000; Brandt et al., 2010). While not directly relevant

to the regulation of translation initiation, the requirement for ceramide was included as it

offered an opportunity to begin to understand the links between the cellular signalling and

metabolic networks.

Page 95

Figure 2.1.3.1d – Petri Net Representation of the JNK1 Signalling Pathway. JNK1 has an emerging role in

the regulation of translation initiation. The main, yet not fully elucidated, role concerns regulating eIF4E

phosphorylation. This pathway was extended back to the Interferon Receptor. As with the p38α pathway,

this receptor was not necessary as a basal level of JNK1 was maintained in the model. This model was

extracted from the large-scale model of translational regulation and was constructed in Petri Net notation

using the software Snoopy.

Page 96

As already stated, mTORC1 is fundamental to the regulation of translation initiation. This

pathway, within the model, was extended back through the PI3K pathway to the Insulin

Receptor (Figure 2.1.3.1e). While mTORC1 is activated by AKT1, it serves as a point of

integration for several other key signalling pathways. It was necessary to include these

pathways in the activation of this signalling component as it allowed the integration of a

signal to produce more complex patterns of biological behaviour. The complexity of this

pathway was further augmented by the addition of feedback loops including the S6K1-

dependent inhibition of mTORC2 (Julien et al., 2010) (such mechanisms are discussed in

more detail in Section 3.1.3 – Interplay Between mTORC1 & AKT1). The inclusion of such

feedback served to ensure that the pathway was better able to simulate a biological system

responding to an external perturbation.

Page 97

Figure 2.1.3.1e – Petri Net Representation of the PI3K and mTORC1 Signalling Pathways. Created using

the software Snoopy, this Petri Net model contains molecular interactions and species concerning the PI3K

and mTORC1 signalling pathways. Included in this model are key points of cross-talk from other signalling

pathways, including ERK1/2, p38α and JNK1. The PI3K pathway is one of the main activators of mTORC1.

Page 98

One of the main features of the model, as a whole, is the implementation of cross-talk

between the signalling pathways. The main kinases found to be responsible for cross-talk

were AKT1 and ERK1/2. With regards to AKT1, this kinase was found, through ill-defined

mechanisms, to inhibit both p38α (Gratton et al., 2001; Liao & Hung, 2003; Shi et al., 2005;

Marderosian et al., 2006; Yu et al., 2006; Taniike et al., 2008) and JNK1 (Kim et al., 2001; Lu

et al., 2002; Park et al., 2002; Aikin et al., 2004; Hui et al., 2005; Wang et al., 2006; Wang et

al., 2007b). ERK1/2 were shown to inhibit ERK5 (Sarközi et al., 2007) and JNK1 (Carter et al.,

1998; Reardon et al., 1999). These inhibitory interactions were included in the construction

of the model as such interactions are responsible for the complex patterns of behaviour

visible in biological systems.

Section 2.1.3.2 – Specific Examples

Literature-based, bottom-up mechanistic model building is a manual process, where expert

curators make decisions about how best to represent the content of literature detailing

experimental biology in a particular framework. While attempts have been made to

formalise this process in particular domains, for example in GSMN (Thiele & Palsson, 2010),

the key role of an expert modeller in making decisions remains. This section presents the

modelling approach used in this work by giving specific examples of how particular literature

articles where represented.

The bottom-up approach to model construction used in this project was achieved in three

stages: i. manual creation, ii. model validation, iii. model refinement. The first stage of

manual creation involved the reconstruction of the translation initiation machinery in

mammalian cells. Such an approach initially required an understanding of the main steps of

the process, for example the recycling of the Post-TCs, formation of the eIF4F complex and

Page 99

mRNA circularisation, and the joining of the 60S ribosomal subunit, as outlined in Figure 1.1.

This basic understanding came from review articles such as Jackson et al. (2010). In order to

represent this process as accurately as possible it was then necessary to undertake more

targeted literature searches to gain an appreciation of how the various stages are

undertaken. For example while Jackson et al. (2010) were able to describe the process of

recycling Post-TCs, the role and regulation of eIF6 in this process is largely overlooked by

this work. Consquently, it was necessary to carry out more targeted searches of existing

literature to gain an appreciation of the role of this initiation factor. As eIF6 functions to

prevent aberrant binding of the 60S subunit (Valenzuela et al., 1982; Raychaudhuri et al.,

1984; Si et al., 1997; Wood et al., 1999) to the assembling 48S complex it has a clear role in

the recycling process. However, at this stage it should be clear that some form of regulation

is necessary to ensure that this factor does not prevent the formation of the 80S ribosome

when required. To this end, it was necessary to investigate what is currently known about

mechanisms of eIF6 regulation. Given the work of Ceci et al. (2003) has shown that eIF6 is

regulated through phosphorylation by PKC βII, it was possible to represent this stage using

the basic switch motif decribed earlier.

The second stage of model construction concerned model validation. Although this stage

had little to do with the construction of the translational regulation model per se, it was

necessary to validate the model against available literature in order to further refine the

connectivity of the model (Pattyn et al., 2013). To this end, various chemical inhibitors were

incorporated into the model (as detailed in Section 2.1.4.2 – Evaluation of the Predictive

Power of the Model). The effects of these chemical inhibitors on various model

components were investigated. Documented effects were recorded into benchmark

Page 100

datasets. A key point of regulating the formation of the eIF4F complex involves the

sequestration of eIF4E by 4E-BP1 (Marcotrigiano et al., 1999). The mTORC1-dependent

phosphorylation of 4E-BP1 inhibits such interactions and allows the eIF4F complex to form

(Beretta et al., 1996; von Manteuffel et al., 1996; Burnett et al., 1998; Gingras et al., 1998;

Marcotrigiano et al., 1999). However, the model simulation results failed to account for a

link between 4E-BP1 phosphorylation and treatment with the inhibitor SP600125 (Bain et

al., 2007; Ito et al., 2011) given in the benchmark dataset, as shown in Figure 2.1.3.2. The

third stage of model construction was refinement. In order to account for this effect

recorded in the benchmark dataset, it was necessary to establish a mechanism by which the

JNK1 inhibitor, SP600125 (Wang et al., 2004), could affect 4E-BP1 phosphorylation. A more

detailed search of the literature concerning JNK1 substrates found that MSK1 was also

capable of phosphorylating 4E-BP1 (Liu et al., 2002; Aggeli et al., 2006; Teng et al., 2009).

The last two stages of model construction can be viewed as iterative as the model

refinement process precipitated a further round of model validation.

Page 101

Figure 2.1.3.2 – Example of Model Construction Process. As discussed in Section 1.1.3 – eIF4F

Components & mRNA Circularisation and Section 1.2.1 – eIF4E-Binding Proteins & The mTORC1

Signalling Pathway, the availability of eukaryotic Initiation Factor 4E (eIF4E) is regulated by the eIF4E-

Binding Protein 1 (4E-BP1). Mammalian Target of Rapamycin Complex 1 (mTORC1) phosphorylates 4E-BP1

(p4E-BP1) and consequently, eIF4E is then free to form the eukaryotic Initiation Factor 4F complex. The

dephosphorylation of p4E-BP1 is mediated by Protein Phosphatase 2A (PP2A) (Peterson et al., 1999; Pham

et al., 2000; Janzen et al., 2011). This example of an Extended Petri Net model is constructed using the

Binary Model Formalism (as described in Section 1.4.4 – The Petri Net Formalism and Section 2.1.2 – The

Binary Model Formalism). The frequency data given above was taken from the model produced during

the course of this project and calculated using the software, Reachfq. The initial model determined that,

despite mTORC1 inhibition with Rapamycin being noted as negatively regulating p4E-BP1 formation in the

Benchmark Dataset, no inhibition was present. Such a finding led other parts of the Benchmark Dataset to

be investigated. The SP600125 model was identified and enabled a more targeted literature search to be

conducted in order to elucidate a link between JNK1 and the phosphorylation of 4E-BP1. This literature

search revealed a link between JNK1 and a kinase noted to be involved in 4E-BP1 phosphorylation,

Mitogen- and Stress-Activated Kinase-1 (MSK1) (Liu et al., 2002). This process of initial construction

followed by re-evaluation of the connectivity of the model through targeted literature searches was used

throughout the project to refine the model.

Page 102

Section 2.1.3.3 – Creation of the Benchmark Datasets

Evaluation of the predictive power of the model is key for future applications. Although no

model is expected to provide absolute accuracy, there is always a risk that model

predictions are wrong. To make a model useful, it is essential to comprehensively evaluate

the predictive power to give future user’s guidance about the likelihood of obtaining correct

predictions and observing different types of errors, such as false predictions of positive or

negative outcomes. Ultimately, the model is a decision making tool with predictions being

used to guide future experimental work. From this perspective, informing the user about

the risks associated with pursuing model predictions should be regarded with equal

importance as the prediction itself. For example, the user may decide that in the absence of

hypotheses for the next step of experimental molecular biology work on a complex system,

following model predictions is useful, even if there is a high probability of a particular

prediction being false. The user may be of the opinion that predictions from a literature-

based model are more likely to be accurate than a guess. Even if such a prediction is false,

the user would still learn from the experimental outcome. On the other hand, a model with

low predictive power would not be applicable for the design of a clinical trial, where

subjects would be put at risk and where considerable financial investment was at stake.

Therefore, in this project, the development of the model was accompanied by the creation

of a benchmark dataset for the evaluation of predictive power. The literature has been

systematically examined to identify the effects of commonly used chemical inhibitors (as

shown in Table 2.1.3.3a) on the activity of different molecular species in the model. These

chemical inhibitors were chosen because usually the effects of these on cell signalling

components are well documented. For each of the inhibitors, original research articles have

been gathered, which describe whether the application of this inhibitor leads to an increase

Page 103

or decrease in the activity of a molecular species present in the model. The PMID identifier

of these articles are included in the dataset. Care was taken to ensure that data was

collected on different inhibitors, applied under similar experimental conditions. For

example, all data was collected in the absence, or presence, of lipopolysaccharide

stimulation, to ensure that a benchmark dataset was created, suitable for evaluating the

predictive power of a model. It should be noted that the systematic evalution of the

literature to create the benchmark datasets was of comparible effort to the construction of

the model itself. This datset is unique and represents a major outcome of the project. The

benchmark dataset is presented in the subsequent results sections with the Final

Benchmark dataset being given in Section 2.2.4 – Evaluation of the Predictive Power of the

Final Model.

Page 104

Inhibitor Target Reference

BIX02189 MEK5 Tatake et al., 2008

CGP57380 MNK1/2 Tschopp et al., 2000

GSK2334470 PDK1 Najafov et al., 2011

Ku0063794 mTORC1/2 García-Martínez et al., 2009

LY294002 PI3K Vlahos et al., 1994

PD184352 MEK1/2-ERK1/2 Allen et al., 2003

PD98059

MEK1/2-ERK1/2 & ERK5

Alessi et al., 1995;

Dudley et al., 1995

Rapamycin mTORC1 Sabers et al., 1995

SB203580 p38 Badger et al., 1996

SP600125 JNK1 Wang et al., 2004

XMD8-92 ERK5 Yang et al., 2010

5Z-7-Oxozeaenol

TAK1, MEK1/2

Ninomiya-Tsuji et al., 2003;

Choo et al., 2006;

Windheim et al., 2007

Acuña et al., 2012

Table 2.1.3.3a - Chemical Inhibitors used in the Model. The target given above is the marketed

cellular target and does not take into account off-target effects. Note that 5Z-7-Oxozeaenol was

only used in the LPS model and Ku0063794 was replaced by Rapamycin in the binary model.

Page 105

Section 2.1.4 – Simulation Method

Section 2.1.4.1 – Monte Carlo Simulations

Both example simulations shown in Section 2.1.2 – The Binary Model Formalism and all

simulations leading to the results presented in Section 2.2 – Results have been performed

with the QSSPN solver (described in Section 1.4.5 – The Quasi-Steady State Petri Net

Algorithm) (Fisher et al., 2013). While simulations presented in this chapter could have

been run in other tools, such as Snoopy, the QSSPN solver was used to make the resulting

model amenable to integration with the Flux Balance Analysis of GSMN (presented in

Chapter 5 – Integration of the Translational Control Model with a Genome-Scale

Metabolic Network of a RAW 264.7 Cell) – a unique feature of QSSPN not available in other

tools. The QSSPN solver performs qualitative simulations using the Gillespie algorithm

(described in Section 1.4.1 – Exact Stochastic Simulations with the Gillespie Algorithm)

with all reaction propensity functions set equal to one (Ibid). It should be noted that this

algorithm was used to generate alternative sequences of molecular interactions occurring in

the binary formalism model, rather than the modelling of the dynamics of molecular

amounts in a quantitative model. Consequently, the propensity function could only assume

two values: zero if the molecular interaction was not enabled, or one, if the molecular

interaction was enabled. Furthermore, the Gillespie algorithm time was used exclusively to

order the sequence of molecular interaction occurences i.e. to determine which molecular

interactions occurred either earlier or later. It was not interpreted as physical time.

Moreover, the Gillespie algorithm used time as a stop condition for the simulations. For

example, trajectories were run for 100 arbitrary time units. Again, this number was not

interpreted as a physical time.

Page 106

The qualitative behaviour of the model has been explored by the simulation of an ensemble

of trajectories. Each trajectory was started from the same initial state and the only

difference between individual trajectories was exclusively due to the pseudo-random

numbers generated in the Gillespie algorithm simulation. Therefore, trajectories represent

the alternative sequences of molecular interactions possible, given the literature knowledge

expressed by the rules of the binary model formalism. These alternative scenarios of model

behaviour were then analysed by two different approaches implemented in Reachfq and

Eventfq scripts (developed by A. Kierzek as part of the QSSPN project).

The Reachfq script calculates the number of trajectories where a particular molecular

species reached a state of k discrete activity levels. In simulations with the binary model, k

was set to one. This script reads every trajectory in the ensemble and moves sequentially

through the states from time 0 to the end. If this condition has been fulfilled, the script

stops analysing that particular trajectory, increases the counter, and moves to the next

trajectory. If the end of the trajectory is reached and the condition was not satisfied, the

counter remains unchanged and the script analyses the next trajectory in the ensemble.

The Eventfq script calculates, for each molecular interaction in the model, the total number

of firings in all trajectories in the ensemble. Each trajectory is read in turn. The first column

of the trajectory file indicates which molecular interaction firing changed state. The script

reads each field in this column, in turn, and whenever the reaction name is encountered,

the counter for this molecular interaction is increased. The result is the total number of

firings, across all trajectories calculated, for each molecular interaction in the model.

Page 107

Section 2.1.4.2 – Evaluation of the Predictive Power of the Model

Predictive power of the model was evaluated by simulating the application of inhibitors,

accounted for in the benchmark datasets, and comparing the simulation results with

experimentally observed effects. First, the Wild Type model, representing a situation where

no inhibitors were applied, was simulated and 1,000 simulation trajectories were collected.

The Reachfq script was then used to calculate the number of trajectories reaching each of

the key effectors from the important signalling pathways featured in the model, and

described in Section 1.2 – Key Points of Regulation & Upstream Signalling Pathways. The

activity changes of these effectors, as a result of the application of inhibitors, were recorded

in the benchmark datasets. Subsequently, for each inhibitor, another variant of the model

representing the application, was created. The Monte Carlo simulation was then run and

1,000 simulation trajectories were collected. The effect of the inhibitor on a particular

model effector has been evaluated by comparing the number of trajectories reaching this

effector, in the Wild Type model, and the model accounting for the action of a particular

inhibitor. In the simulation, the inhibitor was considered to have had an effect on a

particular effector, if the number of trajectories reaching this molecular species were

significantly different between the Wild Type and inhibitor models. A difference in the

number of trajectories reaching a particular effector was considered significantly different if

the Wild Type and inhibitor numbers displayed no overlap in the 99.99% Binomial

Confidence Intervals (Marshall & Olkin, 1985). These Intervals were calculated in the

software R (R Core Development Team, 2013) using the binconf function, which is part of

the Hmisc package (Brant et al., 1991-2016). The application of confidence intervals, in

comparing the numbers of Monte Carlo trajectories exhibiting behaviours of interest, is an

Page 108

accepted methodology in the Statistical Model Checking field of computer science

(Kwiatkowska et al., 2011).

The predictive power of the model was evaluated using the well accepted approach of

constructing Confusion Matrices. The Confusion Matrix classifies comparisons of the model

and simulation results to one of the four possible outcomes:

True Positive (TP) – Positive Experiment & Positive Model

False Positive (FP) – Negative Experiment & Positive Model

True Negative (TN) – Negative Experiment & Negative Model

False Negative (FN) – Positive Experiment & Negative Model

In these outcomes, a Positive result denotes a difference in the frequency of trajectories,

between the ‘Wild Type’ and inhibitor models, resulting in the activation of a particular

effector. A Negative outcome connotes that the incorporation of the inhibitor resulted in

no change in the frequency of Monte Carlo simulation trajectories bringing about the

activation of a particular effector. A True or False designation for each effector is given

based upon whether or not the simulation result is in agreement with the benchmark

dataset.

Page 109

Using the results of the Confusion Matrices, the Matthews’ Correlation Coefficient (MCC)

(Matthews, 1975) was calculated for each model and each inhibitor. This calculation was

done using the equation (Equation 2.1.4.2a) shown in Gowthaman & Agrewala (2008) and

Schloss &Westcott (2011):

𝑀𝐶𝐶 = ((𝑇𝑃 × 𝑇𝑁) − (𝐹𝑃 × 𝐹𝑁))

√((𝑇𝑃 + 𝐹𝑁) × (𝑇𝑁 + 𝐹𝑃) × (𝑇𝑃 + 𝐹𝑃) × (𝑇𝑁 + 𝐹𝑁))

The MCC value ranges from between +1 and -1 and has three points for interpretation (Shi

et al., 2010; Jurman et al., 2012). As the MCC value approaches +1, this indicates increasing

agreement between the model predictions and what has been experimentally shown. A

value of +1 would, therefore, indicate complete agreement. Conversely, a value

approaching -1 indicates increasing disagreement between the model predictions and the

experimental data contained within the benchmark datasets (Shi et al. 2010). An illustration

of this point can be found in a system in which a decrease in the phosphorylation of a

particular model component is predicted, while the experimental evidence demonstrates an

increase. The final point for interpretation is 0. As the MCC value approaches this, it

indicates that the predictions generated by the model are essentially random (Shi et al.,

2010; Jurman et al., 2012). One point to note with regard to MCC values is that in a system

yielding limited numbers of both FP and TP values, the resulting value tends to be high

(Baldi et al., 2000).

An overall MCC value was also calculated, based upon the total numbers, pooled from each

inhibitor model Confusion Matrix. This was necessary to ensure that predictive power was

calculated from sufficient data points to overcome the classic argument that small sample

Equation 2.1.4.2a

Page 110

sizes lead to the validity of a correlation coefficient being called into question (Quenouille,

1949). Therefore, this global MCC value was used to define the overall predictive power of

the model, while the MCC values for the individual inhibitor models were used to re-assess

the connectivity of the model, through targeted literature searches, during refinement.

In order to confirm the validity of the overall MCC values, two additional tests were also

carried out. The first involved the calculation of the Approximate Coefficient (AC) (Burset &

Guigó, 1996) and the calculation was carried out using the variables contained within

Confusion Matrices using the following equations (Burset & Guigó, 1996; Baldi et al., 2000)

(Equation 2.1.4.2b and Equation 2.1.4.2c):

𝐴𝐶𝑃 = 1

4 × (

𝑇𝑃

𝑇𝑃 + 𝐹𝑁 +

𝑇𝑃

𝑇𝑃 + 𝐹𝑃+

𝑇𝑁

𝑇𝑁 + 𝐹𝑃+

𝑇𝑁

𝑇𝑁 + 𝐹𝑁)

𝐴𝐶 = (𝐴𝐶𝑃 − 0.5)

The second additional test involved subjecting the overall Confusion Matrix of each binary

model to a Chi Square Test. This was done by comparing the observed number of Confusion

Matrix outcomes to an expected number for each outcome (Baldi et al., 2000). In this case

the expected number of TP, TN, FP and FN outcomes can be viewed as being the total

number of data points divided by four (as the number of possible outcomes). In simple

terms, if the total number of data points is 100, then it would be reasonable to assume that

each of the four possible outcomes in the Confusion Matrix would occur 25 times. If this

were the case, the resulting MCC value would be equal to zero. By performing this Chi

Square analysis, the MCC value obtained is tested for a significant difference from an MCC

Equation 2.1.4.2b

Equation 2.1.4.2c

Page 111

value of zero (Ibid). Both this and the calculation of the AC value were limited to the

evaluation of the models constructed using the binary model formalism as these were

inherently more meaningful, as discussed in Section 2.1.2 – The Binary Model Formalism.

Section 2.1.4.3 – Synchronous Simulation Methods

The Gillespie algorithm (described in Section 1.4.1 – Exact Stochastic Simulations with The

Gillespie Algorithm), which is the main simulation approach used in this work (detailed in

full in Section 2.1.4.1 – Monte Carlo Simulations), is a stochastic and asynchronous

method. This means that the firing of individual Petri Net molecular interactions is not

synchronised within the simulation timesteps. In each iteration of the algorithm, only one

molecular interaction is chosen randomly and fired. The qsspn solver implements an

alternative, synchronous simulation approach in which firing of molecular interactions is

synchronised. Rather than firing only one molecular interaction in a particular timestep, the

algorithm attempts to fire all molecular interactions, which are enabled at this timestep.

This however, frequently leads to a situation where two or more molecular interactions

attempt to update the same molecular species. If this molecular species contains only one

discrete activity level, the synchronised execution of more than one molecular interaction

would lead to negative numbers of discrete activity levels, which must not occur. This

conflict between more than one molecular interaction requesting discrete activity levels

from the same molecular species is resolved by the stochastic method. The order in which

molecular interactions connected to the same molecular species remove discrete activity

levels from that molecular species is randomised. Molecular interactions consume discrete

activity levels until none are left. In other words, if multiple enabled molecular interactions

are connected to the same molecular species, and the number of discrete activity levels

Page 112

associated with this species is smaller than the number of molecular interactions, not all of

these interactions will be fired and those which are fired are chosen randomly. Therefore,

this method is stochastic and synchronous.

The τ-Leap method (already discussed in Section 1.4.1 – Exact Stochastic Simulations with

the Gillespie Algorithm), proposed by Gillespie (2001), is a stochastic, synchronous

algorithm frequently used in Systems Biology. Briefly, in each timestep 𝜏, the number of

firings, 𝑘𝑇, is randomly generated from a Poisson distribution for each molecular

interaction, 0𝑇, in the system (Gillespie, 2001; Gillespie, 2007). Subsequently, each

molecular interaction is fired 𝑘𝑇 times. This method does not resolve the problem of

molecular interactions simultaneously accessing the same molecular species and, ultimately,

it is likely that such a situation would lead to negative numbers of discrete activity levels in

the qualitative model produced during the course of this project (Cao et al., 2006).

Consequently, this method is not applicable to the binary model produced in this project

and so will not be discussed further.

The synchronous stochastic approach is attractive because it is less computationally

demanding than the asynchronous method. Since multiple molecular interactions are

executed simultaneously, smaller number of simulation timesteps may be required to reach

a biological outcome of interest. Where multiple enabled molecular interactions compete

for a substrate, one interaction is selected at random. Furthermore, simulation trajectory

files are smaller, requiring less storage space. This reduced file size makes the analysis with

the Reachfq script faster. Given this, the synchronous stochastic approach implemented in

the qsspn solver has been considered as an alternative to the Gillespie algorithm. However,

it has been established that the asynchronous method is better in terms of generating

Page 113

biologically meaningful hypotheses and, therefore, the synchronous algorithm was not

used. A case study detailed in Section 2.1.4.3.1 – Case Study Identifying Differences in the

Asynchronous Simulation Approach & the Synchronous Methods gives an example to show

why this was the case.

Section 2.1.4.3.1 – Case Study Identifying Differences in the Asynchronous

Simulation Approach & the Synchronous Methods

A small model (Figure 2.1.4.3.1a) has been extracted from the full molecular network. This

small model contains 65 molecular interactions and 53 molecular species, all relating to

eIF4E availability and phosphorylation.

Page 114

Figure 2.1.4.3.1a - Schematic of the Example Model of eIF4E Phosphorylation and Availability. This

Petri Net model was constructed in the software Snoopy and only encompasses the molecular

interactions relating to the availability of eukaryotic Initiation Factor (eIF) 4E (eIF4E). All interactions

present in this model can be found in the large-scale model of translational regulation. This model is

constructed of 53 molecular species and 65 molecular interactions.

Page 115

The simulation of 1,000 trajectories, of which an example is shown in Figure 2.1.4.3.1b, with

the Gillespie algorithm, followed by the application of the Reachfq script yielded the

following results:

Model Effector Frequency Value

Phosphorylated eIF4E 0.472

Phosphorylated MNK1 0.472


Phosphorylated AKT1 0.511

Phosphorylated JNK1 0.556

Page 116

Figure 2.1.4.3.1b - Example Trajectory from the Simple Model Generated Using the Asynchronous

Simulation Approach. This approach uses the Gillespie algorithm to generate a sequence of enabled

molecular interactions. As this model incorporated molecular interactions only concerning eukaryotic

Initiation Factor (eIF) 4E phosphorylation and availability, this trajectory ends with eIF4E phosphorylation

(marked with the green L). This trajectory starts with the Raf-1-mediated phosphorylation of the Mitogen-

Activated Protein Kinase (MAPK) Kinase, MEK1 (marked with the green A). B concerns the activation of

ERK1. A prerequisite for MNK2 phosphorylation (D) is the activation of JNK1 through phosphorylation (C).

The phosphorylation of eIF4E occurs through MNK2 and requires the PKCα-dependent eIF4G

phosphorylation (E and F). The availability of eIF4E is regulated through sequestration by 4E-BP1. This

sequestration is negative regulated by both mTORC1 and MSK1 (K). The activation of mTORC1 (J) is

dependent upon Rheb while MSK1 activation is carried out through p38α (H). p38α phosphorylation

occurs through MKK3 (G). The model includes the phosphorylation of AKT1 (I). In this model, AKT1 serves

to inactivate JNK1. This model is shown in detail in Figure 2.1.4.3.1a

Page 117

By comparison, the simulation of 1,000 trajectories with the synchronous stochastic

approach, followed by the application of the Reachfq script, yielded the following results:

Model Effector Frequency Value

Phosphorylated eIF4E 0.515



Phosphorylated AKT1 1.000

Phosphorylated JNK1 1.000

An example trajectory yielded by this method is given in Figure 2.1.4.3.1c

Page 118

Figure 2.1.4.3.1c - Example Trajectory from the Simple Model Generated Using the Synchronous

Simulation Approach. This model is shown in detail in Figure 2.1.4.3.1a. This method fires all enabled

molecular interactions in each simulation time interval. In the trajectory shown above this is illustrated.

All molecular species shown in red are updated first. This is followed by the updating of dark blue, green,

light blue and pink molecular species. Given this trajectory, it is clear that MNK2 phosphorylation is not

permitted during model simulations with this method. While phosphorylated JNK1 (pJNK1) does form

(shown in red/green), it is removed once the dual-phosphorylated form of AKT1 (ppAKT1) forms. This is a

result of how the synchronous simulation approach removes the temporal nature of signalling pathways

imparted by the waiting time calculation of the Gillespie algorithm. Molecular species consumed during

the molecular interactions are not coloured, rather only molecular species formed during the firing of

interactions. An exception of this is pJNK1 which is central to the illustration of the limitations to this

approach.

Page 119

It can immediately be seen that the simulation with the Gillespie algorithm provided

examples of the sequence of molecular events leading to MNK2 phosphorylation, while the

synchronous stochastic approach predicted that the species cannot be phosphorylated.

According to molecular biology literature, the phosphorylation of MNK2 is possible (Ueda et

al., 2004). Thus, the synchronous stochastic approach misses biologically meaningful

trajectories of molecular events and, consequently, it should not be used in this work,

despite the advantages this approach has, in terms of compuatational performance.

Closer examination of the synchronous stochastic simulation results show that the

biologically realistic behaviour arises as a result of the feedback surrounding the kinase,

JNK1. This kinase is inactivated by AKT1 (Kim et al., 2001; Lu et al., 2002; Park et al., 2002;

Aikin et al., 2004; Hui et al., 2005; Wang et al., 2006; Wang et al., 2007b). This synchronous

approach exacerbates this feedback by removing the effects of waiting time, implemented

in the asynchronous approach via the Gillepsie algorithm. As all enabled molecular

interactions fire as each simulation timestep occurs, this method assumes that each

interaction takes the same length to occur. In actual biological systems this is not the case

as reactions have different rates. While many of these rates are not known, the

asynchronous approach allows the random sampling of alternative molecular sequences

where molecular interactions do not occur simultaneously. The example above

demonstrates that this allows the identification of a known biological outcome. Since there

are many feedback loops within the large model, this example illustrates a general problem

with the synchronous approach, rather than an example specific to this small example. As a

result of these deficiencies, the synchronous stochastic approach was taken forward but

Page 120

only in a limited capacity (detailed in Section 5.3.2 – Proposed New Approaches and

Section 5.3.2.2 – Synchronous Method).

Finally, another example of a synchronous simulation algorithm is the integration of

Ordinary Differential Equations, which is a workhorse of computational Systems Biology for

the simulation of quantitative models. As follows from the description in Section 1.4.2 –

Ordinary Differential Equations (ODEs), all molecular species in the model are updated

synchronously, but random numbers are not used to resolve competition between

molecular interactions comsuming the same species. Instead, molecular amounts are

described as concentrations rather than numbers of molecules, but competing interactions

decrease the concentrations to numbers possibly smaller than one. However, these

numbers remain positive. Very sophisticated adaptive timestep strategies are used, within

ODE solvers, to ensure that small molecular concentrations are simulated without leading to

negative numbers (Press et al., 2007). In terms of the discussion within this chapter, the

ODE simulation algorithm is therefore synchronous and deterministic. It generates only

one trajectory representing the time evolution of the system. While computationally even

more attractive than the synchronous stochastic simulation approach, the application of this

method is limited to fully parameterised, quantitative models. An attempt to run

simulations of models presented here led to a trajectory which did not have a clear

biological interpretation. As discussed earlier, quantitative parameterisation of the large-

scale molecular network model, described here, is not possible due to a lack of quantitative

measurements in existing scientific literature. However, as this approach was

computationally attractive, it was applied to the full model. The results are given in Section

5.3.2.1 – Ordinary Differential Equation Method.

Page 121

Section 2.1.5 – Investigation of the Interaction Between eIF4E Phosphorylation

and mTORC1 Inhibition

Once the binary model had been refined and the predictive power raised to a sufficient

level, it was necessary to correct the decrease in eIF4E phosphorylation noted in the binary

model (as discussed in Section 2.2.2 – Re-Evaluation of the Binary Model). Moreover, a

mechanism by which eIF4E phosphorylation is increased in response to the inhibition of

mTORC1 was necessary.

Several unsuccessful attempts were made to elucidate such a mechanism before the

necessary circumstances surrounding this increase were identified. The methods detailed in

the subsequent section were used to achieve these objectives. Both efforts involved

extracting information from trajectories, such as using Eventfq to establish the number of

times individual molecular interactions fired and mining the sequence of molecular

interactions occurring up to and including the phosphorylation of eIF4E. The latter required

a script created by D. Taylor using the Python programming language (van Rossum, 2012).

Since neither Eventfq nor the use of Invariant files yielded a clear mechanism, it was decided

that a new approach was necessary. A smaller model was constructed that only

encompassed the molecular interactions surrounding eIF4E phosphorylation and mTORC1

activity. The small scale and scope of this model allowed for easier identification of

molecular species upon which eIF4E phosphorylation is dependent. The small model was

then changed in various ways so as to find a reason for the decrease in eIF4E

phosphorylation.

Page 122

The identification of factors in the small model permitted further, more targeted, work to be

undertaken on the larger model. The critical effectors for eIF4E phosphorylation were then

identified and one-by-one, or in groups, they were subject to removal. Following the

removal of effectors, a new set of simulations was conducted.

Section 2.2 – Results

Once the model was constructed it was simulated with QSSPN calculating 1,000 trajectories.

Reachfq was then used to interrogate these trajectories in order to establish the frequency

of trajectories reaching a particular model effectors. The mode of model construction

enabled the data generated using Reachfq to be easily correlated to the components of the

model. Consequently, points of refinement were easily identified and, after going back to

exisiting literature, were honed without difficulty.

Section 2.2.1 – Binary Model of LPS Stimulation

The binary model formalism was adopted because of the arbitrary discretisation and

unrealistic oscillatory nature of the initial model in preliminary work. This behaviour was

produced by the unrealistic assumption that the switch-like activation and deactivation of

enzymes is caused by equal kinetic parameters. The binary model formalism and rationale

for using this is described above in Section 2.1.2 – Binary Model Formalism. In order to

introduce the rule-based transitions, it was necessary to further extend the granularity of

the signalling pathways contained within the model by elaborating on the annotation of

these pathways back to the receptor-ligand interactions. As several of the pathways were

related to stress responses, the LPS-mediated Toll-Like Receptor 4 signalling (Poltorak et al.,

1998; Chow et al., 1999; Hoshino et al., 1999) was one such pathway to be modified.

Page 123

Although this preliminary work is not shown, the predictive power of this initial model was

found to be MCC = 0.2354.

The following table (Figure 2.2.1a) shows an excerpt of the 99.99% Binomial Confidence

Intervals calculated for the Reachfq analyses of the frequency of activation of model

effectors in a binary model incorporating LPS stimulation.

Page 124

Effector

Wild Type PD98059

Frequency Upper Lower Frequency Upper Lower

ppERK1 1.000 1.000 0.989 0.000 0.011 0.000

ppERK2 1.000 1.000 0.989 0.000 0.011 0.000

ppMEK1 1.000 1.000 0.989 1.000 1.000 0.989

ppMEK2 1.000 1.000 0.989 1.000 1.000 0.989

pRSK1 1.000 1.000 0.989 1.000 1.000 0.989

rpS6-Pi 1.000 1.000 0.989 1.000 1.000 0.989

pMNK1 1.000 1.000 0.989 0.925 0.948 0.893

pMNK2 1.000 1.000 0.989 0.000 0.011 0.000

eIF4E-Pi 0.864 0.896 0.824 0.047 0.074 0.029

Figure 2.2.1a – Example of the LPS Model Comparing the Effects of Inhibitors to a Wild Type

Model. Inhibitors are incorporated into the model and the effects are compared to a Wild Type

model in which no inhibitors are included. Reachfq was used to interrogate the QSSPN simulations

to determine, for each model effector, the fraction of the 1,000 trajectories sampled in which this

effector is activated. 99.99% Binomial Confidence Intervals were used to identify regions of

significant difference (shown in red).

Page 125

These frequency values, and the associated 99.99% Binomial Confidence Intervals, were

then used to form Confusion Matrices (as summarised in Figure 2.2.1b). The measures

placed in the matrices are determined based upon the identification of perturbations

resulting in significant changes in the trajectory frequencies. The frequency value for a

particular model effector in an inhibitor model was considered significantly different if the

99.99% Binomial Confidence Interval for that effector if it did not overlap with the

corresponding 99.99% Binomial Confidence Interval in the Wild Type model. To form these

Confusion Matrices, the behaviour of the model was compared to the LPS Benchmark which

contained the expected behaviour of model components in response to the various

chemical inhibitors. This benchmark was composed of information taken from 122

literature articles. Supplementary Information 1b contains the bibliographic information of

the articles contained with this benchmark, whilst Supplementary Information E5 contains

the LPS Benchmark dataset itself.

Page 126

Confusion Matrix Values

Model TP TN FP FN MCC Value CGP57380 2 0 1 3 -0.447

LY294002 8 0 3 4 -0.302

PD184352 5 2 1 2 0.356

PD98059 2 1 1 6 -0.218

Rapamycin 6 0 2 3 -0.289

SB203580 2 1 0 11 0.113

SP600125 2 5 0 1 0.745

(5Z)-7-Oxozeaenol 3 0 0 5 Undefined

OVERALL 30 9 8 35 -0.007

Figure 2.2.1b – Overview of the Predictive Power of the Model Incorporating the Effects of

Lipopolysaccharide Signalling. The Petri Net model of translational regulation was interrogated with

QSSPN followed by Reachfq. The binomial confidence intervals of model effectors in the model

including a chemical inhibitor were compared with those in the Wild Type model and regions of

significant difference were identified. By comparing the results of this evaluation to the LPS

Benchmark it was possible to produce Confusion Matrices with results being characterised as either

True Positives (TP), True Negative (TN), False Positive (FP) and False Negative (FN). These qualitative

measures of predictive power were used to produce the Matthews’ Correlation Coefficient (MCC).

The overall predictive power was established by pooling the sums Confusion Matrix variables for

each inhibitor and calculating an overall MCC value. Due to the absence of TN and FP values, an

MCC value for the (5Z)-7-Oxozeaenol was not quantifiable.

Page 127

Whilst this overall MCC value suggests that the binary model leads to random predictions, it

is worth noting that there is a smaller number of data points being used. Moreover, the

maximum time limit of the simulation was such that the majority of effectors reached

frequencies of one. These results prompted a re-evaluation of the simulation protocol.

Therefore the binary model was re-tested against the initial, non-LPS dataset and the

maximal time-step of the simulations adjusted so that the Wild Type effector frequencies

should be approximately 0.50. This approach, which had already been implemented by

Fisher and colleagues (2013), allowed for the discovery of increases and decreases in the

reachability of the model effectors in response to perturbations. Despite this leading to

larger 99.99% Binomial Confidence Intervals for the Wild Type model, it enabled more

complex behaviour to be observed. The results of applying this refined protocol are shown

below in Section 2.2.2 – Re-Evaluation of the Binary Model.

One of the main outcomes of the preliminary work was the finding that eIF4E

phosphorylation was increased in response to mTORC1 inhibition. As such, it is noteworthy

that this behaviour was conserved in the LPS model with an increase in the phosphorylation

of eIF4E being seen in the Rapamycin inhibitor model.

Section 2.2.1.1 – Updated Benchmark Dataset

Proteomic differences between cell types is well established (Pontén et al., 2009;

Paulitschke et al., 2013) and the question was raised as to the validity of using data from a

variety of cell lines as to the cellular effects of inhibitors. The dataset constructed with

publications, not using LPS stimulation, contains very few studies that use macrophage cell

lines, let alone RAW 264.7 cells. It was decided that the dataset containing studies that only

use LPS stimulation would be more appropriate to evaluate this effect. As LPS stimulation is

Page 128

important in the innate immune system, more studies in this dataset use macrophage-

derived cell lines. It is therefore possible to test this assumption with the LPS dataset, the

intial results of which are given above in Section 2.2.1 – Binary Model of LPS Stimulation.

Although there are undoubtedly differences in the functioning of different macrophage-

derived cell lines, in order to test this, in the least conservative way, all macrophage-derived

cell lines will be included. By eliminating any studies with cell types other than macrophage

cell lines, the number of data points used for calculating the overall MCC and AC values was

reduced from 82 to 41. By performing a Chi Square Test (α = 1.55x10-4) it is clear that this is

a significant reduction. The information pertaining to these datasets can be found in

Supplementary Information E5.

Despite an overall increase in the global MCC value, by limiting the cell types included in the

formation of the benchmark datasets, the number of possible data points was being

severely reduced. By doing so, the relevance of even the global MCC values is being

reduced. As such, the method of including multiple cell types in the benchmark dataset was

the most appropriate. In order to further enhance the value of this dataset, verifying results

across multiple studies and cell types was best practise.

Section 2.2.2 – Re-Evaluation of the Binary Model

The adjusted maximal time-steps for each effector in the Wild Type was applied to each of

the inhibitor models. The frequency of each effector in the Wild Type model was

approximately 0.5, as a result of the maximal simulation time being limited. By ensuring the

Wild Type frequency took this approximate value, more complex patterns of behaviour

were visible. For example, if the simulation time was extended to permit an effector to

reach a frequency approaching 1.0, then it would not be possible to observe perturbations

Page 129

resulting in an increase in the formation of a particular effector. The frequencies, with

associated 99.99% Binomial Confidence Intervals, for this model are shown in

Supplementary Information E7 – Frequency Data Generated by Reachfq with Associated

99.99% Binomial Confidence Intervals (Sheet entitled First Refinement of the Model).

By changing back to an updated version of the original dataset, more points of analysis were

available against which to test the binary model. This updated version of the original

dataset, termed the Revised Original Benchmark dataset (shown in Supplementary

Information E6) contains information from 130 peer-reviewed articles on the effects of

chemical inhibitors on key model effectors. Such a change led to a slight improvement of

the predictive power of the model (MCC = -0.007 vs MCC = 0.1393 (shown in Figure 2.2.2a)

in the first round of model refinement. However the predictive power of this binary model

is lower than the initial model (MCC = 0.2354 vs MCC = 0.1393). This being said, the changes

made to the model during the transition to the binary model formalism allowed for

improvements to be made.

Further iterations of the model were made to improve the predictive power with the aim of

approaching an MCC value of approximately 0.5. The workflow process for these iterations

is shown in Figure 2.2.2b and summarises the process used to refine the model through five

iterations. The first four iterations of are summarised in Figure 2.2.2a and show the overall

MCC values increase as changes are made. Whilst global improvements were made to the

MCC values, these were subject to fluctuations when looking at individual Inhibitor Models.

This highlights the need to view the MCC values of the individual models with caution.

Furthermore, the changes to the model have made plain the importance of identifying and

Page 130

increasing cross-talk between pathways. The Revised Original Benchmark dataset was used

in each comparison of the model in order to calculate the measure of predictive power.

Of interest to the experimental work of this project, the earlier prediction concerning the

effect of mTORC1 inhibition on eiF4E phosphorylation is still observed. The frequency of

trajectories producing eIF4E phosphorylation fell from 0.520 in the Wild Type model to

0.110 when Rapamycin was included.

Page 131

First Round of Model

Refinement

Second Round of Model

Refinement

Third Round of Model

Refinement

Fourth Round of Model

Refinement

TP

31

34

39

43

TN

31

35

38

42

FP

8

6

7

5

FN

59

51

43

43

MCC

0.1393

0.2553

0.3185

0.3918

AC

0.1393

0.2553

0.3185

0.3918

Chi Square

Test α-Value

8.27x10-9

2.78x10-7

8.81x10-6

5.18x10-7

Figure 2.2.2a – Overall Predictive Power of the Model Through the Four Stages of Refinement. The

binary model of translational regulation was compared to the Revised Original Benchmark Dataset.

The comparison of the dataset to the behaviour of the model led to the construction of Confusion

Matrices with behaviours being classified as being True Positives (TP), True Negatives (TN), False

Positives (FP) and False Negatives (FN). These qualitative measures were converted to quantitative

measures through the calculation of both the Matthews’ Correlation Coefficient (MCC) and

Approximate Coefficient (AC) values. Through four rounds of model refinement, it was possible to

raise the level of predictive power from an MCC = 0.1393 to MCC = 0.3918. Each refinement stage

concerned the increase in the level of cross-talk between pathways and the elimination of FP values.

To accomplish these, it was necessary to return to existing literature.

Page 132

Figure 2.2.2b – The Model Refinement Workflow. Over four iterations of the model, it was

refined to improve the predictive power. Simulations were run using QSSPN and Reachfq once the

optimal simulation times were established for each model effector. This was necessary to ensure

that the Wild Type frequency of activation was approximately 0.5. The 99.99% Binomial

Confidence Intervals were calculated and the results were compared to the relevant Benchmark

dataset. Using these comparisons, the Matthews’ Correlation Coefficient (MCC) values were

calculated to quantitatively measure the predictive power of the model. Based upon the MCC

values for the individual inhibitor models, the connectivity of the model was refined using

additional literature. The cross-talk of the signalling pathways was the main target for refinement

so as to allow more complex patterns of behaviour.

Page 133

The fifth round of model refinement was responsible for raising the predictive power of the

model to an MCC value of approximately 0.5 (as shown in Figure 2.2.2c). The frequency

values, along with the 99.99% Binomial Confidence Intervals, are shown in Supplementary

Information E7 – Frequency Data Generated by Reachfq with Associated 99.99% Binomial

Confidence Intervals (sheet entitled Fifth Refinement of the Model). Of particular interest is

the finding that eIF4E phosphorylation was significantly altered by Rapamycin, as noted by a

deviation in the frequency of trajectories reaching one discrete activity level (0.522 in the

Wild Type model versus 0.005 in the Rapamycin model). Such an affect was also noted in

the preliminary work and the LPS containing model (outlined in Section 2.2.1 – Binary

Model of LPS Stimulation). Moreover, this alteration is associated with a perturbation in

MNK2, as noted by Stead and Proud (2013). However, in this model eIF4E phosphorylation

appears reduced despite the increase in the frequency of trajectories producing

phosphorylated MNK2. Work in subsequent sections of this chapter were aimed at

eliminating this discrepancy and defining the mechanism by which Rapamycin can increase

the phosphorylation of this initiation factor.

Page 134

i.


Model TP TN FP FN MCC Value BIX02189 1 7 0 4 0.3568

CGP57380 4 4 0 1 0.8000

GSK2334470 2 12 0 2 0.6547

LY294002 4 3 0 6 0.3651

PD184352 5 1 0 7 0.2282

PD98059 9 3 1 3 0.4472

Rapamycin 5 3 0 5 0.4330

SB203580 5 4 2 6 0.1177

SP600125 10 2 0 5 0.4364

XMD8-92 2 2 0 0 1.0000

ii.


Measures of Predictive Power

TP TN FP FN MCC Value AC Value α-Value

Overall 47 47 3 39 0.4652 0.4652 5.15x10-8

Figure 2.2.2c – Evaluation of the Predictive Power of the Fifth Iteration of the Model. (i) Once

QSSPN and Reachfq had evaluated the model and regions of significant difference had been

identified, it was possible to compare the effects of chemical inhibitors in the model to published

effects of the inhibitor on particular model effectors (within the Revised Original Benchmark

Dataset). The result of this comparison was the formation of Confusion Matrices. Consequently, the

behaviour of the model was characterised as being either a True Positive (TP), True Negative (TN),

False Positive (FP) or False Negative (FN). In order to produce a quantitative measure of predictive

power, these values were used to calculate the Matthews’ Correlation Coefficient (MCC). (ii) To

produce an overall measure of predictive power, the Confusion Matrices variables in (i) were pooled.

The MCC and Approximate Coefficient (AC) values were calculated. In order to show the predictive

power was significantly different from random, the Chi Square Test α-Value was calculated.

Page 135

Section 2.2.3 – Mechanism by Which mTORC1 Inhibition Affects eIF4E

Phosphorylation

Based upon the connectivity of the model, only four factors are directly linked to the

phosphorylation of eIF4E: MNK1, MNK2, eIF4G phosphorylation and 4E-BP1

phosphorylation. Through the construction of a smaller model, again within Snoopy, a more

detailed investigation of eIF4E phosphorylation was possible. This smaller model contained

only molecular interactions included in the original model and only molecular interactions

that are proximal to eIF4E phosphorylation. Given the reduced scale of the model, it was

not possible to include the complex cross-talk between the signalling pathways present in

the larger model. However, as the aim of this smaller model was to explain the behaviour of

the eIF4E phosphorylation seen when Rapamycin is present, much of this cross-talk was not

deemed necessary in the first instance.

Simulations with this model revealed that eIF4G phosphorylation and 4E-BP1

phosphorylation appeared to act to reduce eIF4E phosphorylation in a model incorporating

Rapamycin. This raised three possible, but not necessarily mutually exclusive, mechanisms

to explain how an increase in eIF4E phosphorylation was possible:

i. The level of 4E-BP1 in the cell must lower than that of eIF4E so as to allow

sequestration of eIF4E by hypophosphorylated 4E-BP1 and at the same time

permit an increase in eIF4E phosphorylation.

ii. mTORC2 cannot be the only kinase responsible for the activation of the eIF4G

kinase, PKCα.

Page 136

iii. The activity of a protein phosphatase may be suppressed by mTORC1 activity.

Such a phosphatase may act on either MNK1 or MNK2, eIF4G, PKCα or even

eIF4E directly.

The first hypothesis is in agreement with the work of Eckerdt and co-workers (2014) who

established that the concentration of 4E-BP1 had little influence on the increased

phosphorylation of eIF4E. Although difficult and costly, proteomic evaluation of RAW 264.7

cells could reveal if this is the case and show if and to what extent the levels of eIF4E and

4E-BP1 differ. Several studies have employed proteomic approaches with RAW 264.7 cells

(Martínez-Solano et al., 2009; Shi et al., 2009; Eichelbaum & Krijgsveld, 2014). However to

date no information regarding the relative levels of either protein have been reported in

these studies. Cai et al. (2014) and Decarlo et al. (2015) both remark that the ratio of eIF4E

to 4E-BP1 is critical in the early transformation events of cells. Quantitative immunoblotting

was not considered here as the method is often regarded as pseudo-quantitative or, in the

case of advanced quantitative immunoblotting, requires the expression of recombinant

proteins in control samples to be used to aid the quantification (Janes, 2015). Morevoer,

the quantification of Western Blots with densitometry is affected by a great number of

factors that may arise during the immunoblotting procedure, the scanning of blots or even

during the quantification itself and consequently has little in the way of scientific foundation

(Gassmann et al., 2009). Taylor and co-workers (2013) even go so far as to state that the

use of radiographic film is inferior to camera-based development systems when it comes to

quantitative immunoblotting. For these reasons, a proteomics based method would be

preferred for the quantification of eIF4E and 4E-BP1 levels.

Page 137

Whilst there is no direct evidence to support for the hypothesis that 4E-BP1 is expressed at

a lower level than eIF4E, there is reason to believe this is the case. Data taken from The Cell

Line Atlas database, within The Human Protein Atlas (Uhlén et al., 2005; Uhlén et al., 2015),

has revealed that in the majority of cases, across multiple human cell cancer lines, that the

level of eIF4E (Analysis of Expression of EIF4E Website, taken from

http://www.proteinatlas.org/ENSG00000151247-EIF4E/cell/CAB004077 [last accessed

22/05/2015]) is higher than that of 4E-BP1 (Analysis of Expression of EIF4EBP1 Website,

taken from http://www.proteinatlas.org/ENSG00000187840-EIF4EBP1/cell/CAB005032 [last

accessed 22/05/2015]). Using Immunohistochemistry (IHC) data, of the 46 cell lines tested,

only four (8.7%) displayed evidence of 4E-BP1 being present at higher levels than eIF4E. The

data shown in The Human Protein Atlas, whilst supportive of the view that eIF4E is

expressed to a higher degree than 4E-BP1, is not quantitative enough to provide conclusive

support to this hypothesis. This issue of IHC being quantifiable is summarized by Taylor and

Levenson (2006). This work notes that there is some contention over whether a correlation

exists between the amount of antigen staining and the actual amount of the antibody-

target. Samples generated by IHC are also subject to a variable interpretation by observers

and, thus, the resulting scoring of sample staining is subjective and open to vast

interpretation. As a result of this, and a lack of suitable experimental controls,

reproducibility is often lacking even when analytical software is used (Ibid).

Although mTORC2 has previously been found to be a kinase for PKCα (Guertin et al., 2006;

Facchinetti et al., 2008; Ikenoue et al., 2008), experimental work presented in Chapter 3 –

Experimental Validation of Model Predictions has demonstrated that the mTORC2

phosphorylation site of AKT1, Ser-473 (Frias et al., 2006; Jacinto et al., 2006; Breuleux et al.,

http://www.proteinatlas.org/ENSG00000151247-EIF4E/cell/CAB004077

http://www.proteinatlas.org/ENSG00000187840-EIF4EBP1/cell/CAB005032

Page 138

2009), displays a significant reduction in AKT1 phosphorylation when using the Human

Phospho-MAPK Arrays following a one hour Rapamycin treatment. It is worth re-iterating

here that whilst AKT1 phosphorylation was reduced, the p-value calculated was only

significant prior to a Holm-Sidak Correction. However, the AKT Pan level remained

significantly reduced after this correction. Furthermore, immunoblotting of lysates treated

for three hours displayed no significant difference in the level of Ser-473 phosphorylated

AKT1. It is possible that this is as a result of reduced antibody sensitivity with this method as

the signal quantified was very low. Given there is still reason to argue that AKT1

phosphorylation is reduced, along with the simulation results and the Rapamycin-mediated

increase in eIF4E phosphorylation, it is reasonable to argue that mTORC2 cannot be the only

kinase responsible for phosphorylating PKCα, a kinase known to be responsible for eIF4G

phosphorylation (Dobrikov et al., 2011).

Although speculative, PDK1 has been shown to be capable of phosphorylating PKCα, and

other PKC enzymes, in such a way as to be required for the activation of PKC (Dutil et al.,

1998; Le Good et al., 1998; Sonnenburg et al., 2001). Gao and co-workers (2001) noted

that, in PKCβII, PDK1 mediates phosphorylation of Thr-500 residue within the PKC activation

loop. Autophosphorylation of PKCβII can then occur when PDK1 has dissociated (Ibid).

Several studies have noted that PDK1 is required to maintain the level of PKCα and other

PKC isoforms expressed in cells (Collins et al., 2005; Wood et al., 2007). When taken

together with the work of Dutil et al. (1998) and Sonnenberg et al. (2001), which noted that

PDK1 was capable of phosphorylating PKCα on the activation loop, these works suggest that

PDK1 not only activates PKCα but also positively regulates the expression of this protein

kinase. Although PDK1 has been shown to be the kinase responsible for activating multiple

Page 139

PKC isoforms (including PKCβ (Helliwell et al., 2003), PKCδ (Deb et al., 2003), PKCζ (Chou et

al., 1998) and PKCθ (Bauer et al., 2001; Villalba et al., 2002) this occurs in a manner

dependent upon the PI3K pathway. It would not be unreasonable, based upon this

evidence, to hypothesise that it is probable that PKCα activation would occur in an

analogous fashion.

Testing this hypothesis could be achieved with either chemical inhibition of PDK1, CRISPR or

with transgenic mice (these are explained in Section 3.4.4 – Potential Future Work).

However, for reasons stated previously, the relevance of both to this project is somewhat

unclear. At present CRISPR has not been used in RAW 264.7 cells and so it is not clear if this

would be appropriate at this time. Transgenic mice have been successfully used to create

PDK1 knockout mice (Tawaramoto et al., 2012), however, the cells derived from these

would be likely to display behaviour different to that of RAW 264.7 cells. Chemical

inhibition of PDK1 would at this time be the best option. This being said, it would require

treating the cells with multiple inhibitors and the issue of inhibitor-inhibitor interactions

would then be raised.

With regard to the third proposed mechanism, mTORC1 is known to regulate the

functioning of several protein phosphatases, PP5 (Huang et al., 2004) and PP2A (Hartley &

Cooper, 2002). As Rapamycin has been shown to lead to an increase in the Ser-437

phosphorylation of MNK2 (Stead & Proud, 2013), it follows that if a phosphatase were

involved, that it must be suppressed during mTORC1 inhibition. This suggested that the

increase in eIF4E phosphorylation seen during Rapamycin treatment maybe a result of

alterations to protein phosphatase activity. Given that Hartley and Cooper (2002) found

that Rapamycin enhanced PP2A activity, it is unlikely this phosphatase is a viable candidate.

Page 140

Bishop and co-workers (2006) similarly noted that mTORC1 acts to limit PP2A activity.

However, Huang and co-workers (2004) established that during Rapamycin treatment, the

regulatory subunit of PP5, PR72, interacts with PP5 with reduced affinity and thus

moderates PP5 activity. This phenomenon appears to be dependent upon the expression of

4E-BP1 but no mechanism has been proposed (Ibid).

Whilst this would seem to indicate that that PP5 is possible candidate here, to date, no work

has linked PP5 to the regulation of eIF4E phosphorylation or to either MNK1 or MNK2. This

being said, it has been reported that PP5 is capable of dephosphorylating sites targeted by

PKC (Katayama et al., 2014). Whilst not conclusive, this does intimate that eIF4G may be a

substrate for PP5 in this system. Furthermore, von Kriegsheim and co-workers (2006) noted

that PP5 may serve to negatively regulate ERK1/2 signalling by dephosphorylating Raf-1

which suggests that downstream effectors, such as eIF4E, may also be affected.

Any potential work to elucidate a role for PP5 would be severely hampered by a lack of a

suitable or specific inhibitor of PP5 (Swingle et al., 2007). Studies of the function of PP5

have been aided by RNA interference against it (Messner et al., 2006) as well as the

generation of mice deficient in PP5 (Amable et al., 2011). Nevertheless, given the problems

detailed in Section 3.4.3 – Problems Encountered & Troubleshooting, these do not appear

to be promising avenues. However, both Gentile et al. (2006) and Messner et al. (2006)

have used PP5 in which the Tyr-451 is mutated to an Alanine residue. The result of this is a

considerable reduction in sensitivity to the inhibitor, Okadaic Acid. A comparison of the

effects Okadaic Acid has upon wild type PP5 and the resistant PP5 may yield information on

whether this phosphatase affects eIF4E phosphorylation or the activity of upstream

signalling components, such as MNK1, MNK2 or eIF4G.

Page 141

Whilst the only direct link between mTORC1 inhibition and PP2A appears to be one of

activation, there is some evidence that PP2A may be positively regulated by negative

feedback loops that exist between AKT1 and mTORC1 (as described in Section 3.1.3 –

Interplay Between mTORC1 and AKT1). Although AKT1 is negatively regulated by PP2A

(Hong et al., 2008; Aceto et al., 2009; Kim et al., 2009b), there is some evidence to suggest

that there is some dual-regulation of PP2A and AKT1. In response to Insulin stimulation, the

increase in AKT1 phosphorylation occurs concomitantly with an increase in PP2A

phosphorylation (Kumar & Tikoo, 2015). This phosphorylation occurs on the Tyr-307 residue

(Ibid) which is noted as suppressing the phosphatase activity of PP2A (Chen et al., 1992).

This suggests that enhancing the activity of AKT1 would increase the phosphorylation of the

inhibitory Tyrosine residue of PP2A. This could be achieved with, for example, low

concentrations of Rapamycin. Although not demonstrated, this could act to counter the

increase in PP2A activity noted by Hartley and Cooper (2002) and Bishop et al. (2006).

While both Src (Chen et al., 1992; Arif et al., 2014) and Glycogen Synthase Kinase 3β (Yao et

al., 2011) are known to phosphorylate PP2A, it is not clear whether AKT1 is involved in

either mechanism.

Whilst somewhat subjective, the first and second explanations given towards the start of

this section have been validated by keeping eIF4E at a higher level than 4E-BP1 and the

removal of a dependence on MNK1 and phosphorylated eIF4G from the model. Indeed,

analogous changes in the small model had already been shown to result in the level of eIF4E

phosphorylation increasing in response to Rapamycin. The removal of these was shown to

bring about an increase in eIF4E phosphorylation, as shown in Section 2.2.4 – Evaluation of

the Predictive Power of the Final Model. The removal of these effectors demonstrates that

Page 142

MNK2 is the kinase responsible for the increase in eIF4E phosphorylation, even in the

absence of any increase in ERK1 or ERK2 activity. Whilst in this case it is harder to ascribe a

role for the third hypothesis mentioned above, the inhibition of a phosphatase cannot be

ruled out as it is a possibility that the increase in MNK2 phosphorylation accompanying

Rapamycin treatment is associated with a reduction in phosphatase activity.

Section 2.2.4 – Evaluation of the Predictive Power of the Final Model

The final version of the Petri Net model contained 511 molecular species and 584 molecular

interactions. These species and interactions were derived from exisiting literature, detailing

the interactions between mammalian cellular signalling components regulating a variety of

eIFs and between the translation initiation machinery itself.

Before starting the next phase of the project, implementing the effects viral infection has

upon the signalling network and the cellular translational machinery (as detailed in Chapter

4 – Modelling the Effects of Viral Infection), it was necessary to re-evaluate the predictive

power of the model. This was of particular importance once eIF4E phosphorylation was

increased in response to Rapamycin. The dataset against which this model is tested is

shown in Supplementary Information 2. Furthermore, work in this section will also address

potential concerns that the model was over-trained to the Benchmark Dataset against which

the behaviour of the model was compared.

Prior to detailing the results of this final round of predictive power analysis, it is necessary to

address a point mentioned in Section 2.1.2 – The Binary Model Formalism. It must be

demonstrated that the Binary Model Formalism is rigorous and does not lead to a model

Page 143

which permits a molecular species to be loaded with more than one discrete activity level

(Figure 2.2.4a).

Page 144

Figure 2.2.4a – Confirmation of the Binary Nature of the Model Formalism. A QSSPN simulation was run

sampling 100 trajectories. Rather than interrogating the model with Reachfq to see how many trajectories

reach one discrete activity level of each molecular species, the interrogation determined how many

trajectories resulted in each model effector reaching two discrete activity levels. All model effectors in this

list were included in the model refinement process. Protein Synthesis was included here as the

overarching aim of the project was to model the regulation of nascent protein synthesis.

Page 145


99.99% Binomial Confidence Intervals (sheet entitled Final Binary Model) details the

frequency values, and associated 99.99% Binomial Confidence Intervals, calculated from the

results of the QSSPN simulations for each of the inhibitor models. The frequency values

and associated 99.99% Binomial Confidence Intervals of the common components, found in

both the model and Human Phospho-MAPK Arrays (as discussed in Section 3.3.5 –

Comparison of Model Results to Array Data) are shown in Figure 2.2.4b. The final model

file is provided in Supplementary Information E3 (as both a Snoopy model file (E3a) and an

SBML file (.xml)) (E3b).

Importantly, eIF4E phosphorylation was shown to be increased in response to Rapamycin.

This was determined through an increase in the frequency of trajectories reaching one

discrete activity level of phosphorylated eIF4E (0.492 in the Wild Type model versus 0.805 in

the Rapamycin model). Such an increase was expected based upon the findings of a number

of studies (Sun et al., 2005; Wang et al., 2007c; Jin et al., 2008a; Grosso et al., 2011; Marzec

et al., 2011; Stead & Proud, 2013; Eckerdt et al., 2014). Moreover, the continued presence

of an increase in trajectories reaching phosphorylated MNK2 in response to Rapamycin

(0.463 in the Wild Type model versus 0.761 in the Rapamycin model), is in keeping with

earlier work (Stead & Proud, 2013).

Page 146

Wild Type Upper Lower Rapamycin Upper Lower

ppERK1 0.520 0.570 0.468 0.525 0.576 0.473

ppERK2 0.500 0.552 0.448 0.512 0.564 0.460

pRSK1 0.577 0.627 0.525 0.560 0.611 0.508

ppp38a 0.537 0.588 0.485 0.544 0.595 0.492

pJNK1 0.594 0.644 0.542 0.625 0.674 0.574

mTORC1-Act 0.555 0.606 0.503 0.000 0.011 0.000

ppS6K1 0.153 0.194 0.119 0.000 0.011 0.000

ppAKT1 0.569 0.620 0.517 0.006 0.021 0.002

pMKK3 0.542 0.593 0.490 0.521 0.572 0.469

pMKK6 0.534 0.585 0.482 0.529 0.580 0.477

pMSK2 0.607 0.656 0.555 0.598 0.648 0.546

pCREB 0.576 0.626 0.524 0.573 0.623 0.521

pGSK3 0.546 0.597 0.494 0.256 0.304 0.213

Figure 2.2.4b – Components of the Model Common to Human Phospho-MAPK Array. Model

components for which the effects of Rapamycin were determined using the Human Phospho-

MAPK Arrays (the overall results of which are detailed in Section 3.3.2 – Evaluation of the

Human Phospho-MAPK Arrays) were evaluated using QSSPN. Binomial Confidence Intervals,

calculated in R, were used to find regions of significant difference, shown in red, by comparing

the frequency of activation for each effector in the Wild Type and Rapamycin model.

Page 147

Figure 2.2.4c summarises the comparison between the behaviour of the model and either

the Final Benchmark dataset of 134 literature articles concerning the effects chemical

inhibitors have upon model effectors, or the results of the Human Phospho-MAPK Arrays.

The Final Benchmark dataset is given in Supplementary Information 2, whilst the

bibliographic information contained within it is given in Supplementary Information 1a.

Page 148

i.

Confusion Matrix Values Model TP TN FP FN MCC Value

BIX02189 1 7 0 4 0.3568

CGP57380 4 4 0 1 0.8000

GSK2334470 2 12 0 2 0.6547

LY294002 4 3 0 6 0.3651

PD184352 5 1 0 7 0.2282

PD98059 9 3 1 3 0.4472

Rapamycin 6 5 0 5 0.5222

SB203580 5 4 3 6 0.0255

SP600125 10 3 0 5 0.5000

XMD8-92 2 2 0 0 1.0000

Array Data 3 6 1 3 0.3858

ii.




Overall 51 50 5 42 0.4558 0.4558 1.94x10-8

Figure 2.2.4c – Evaluation of the Predictive Power of the Final Version of the Model. (i) By

comparing the Final Benchmark dataset with the behaviour of the model, it was possible to produce

Confusion Matrices detailing a qualitative measure of predictive power. Such matrices characterise

the behaviour of the model as being either True Positive (TP), True Negative (TN), False Positive (FP)

or False Negative (FN). Additionally, under ‘Array Data’ it was possible to evaluate the behaviour of

model effectors which are also found on Human Phospho-MAPK Arrays (as detailed in Section 3.3.2

– Evaluation of the Human Phospho-MAPK Arrays). For each inhibitor and the experimental data, a

quantitative measure of predictive power is given in the form of the Matthews’ Correlation

Coefficient (MCC). (ii) The overall predictive power is found by pooling the Confusion Matrix

variables and calculating a final MCC value. Additionally, the Approximate Coefficient (AC) was also

calculated. The Chi Square Test α-Value was calculated to ensure the level of predictive power was

significantly different from random.

Page 149

Overall the final version of the binary model has considerable predictive power which is

conserved following the alterations to the model that corrected the discrepancy in eIF4E

phosphorylation following Rapamycin inclusion. Whilst the predictive power does suffer a

slight reduction (from MCC = 0.4652 (as calculated in Section 2.2.2 – Re-Evaluation of the

Binary Model) to 0.4558) this is not significant (as determined by a Chi Square Test (α =

0.2747)). A further Chi Square Test revealed that the Matthews’ Correlation Coefficient for

this model is significantly different from random (α = 1.94x10-8).

Whilst it is not feasible to reconstruct the model multiple times, nor is it possible to perturb

the parameters of a qualitative non-parametric model, it was possible to ensure that this

MCC value is representative of a model which has not been over-trained to a particular

dataset. To this end, parts of the Final Benchmark Dataset were, randomly removed and

the MCC value recalculated, as shown in Figure 2.2.4d. Whilst this is not a cross-validation

procedure in a machine learning sense, it is an analysis of the stability of the results

obtained during this work and demonstrates that the model provides stable predictions of

the perturbations induced by inhibitors.

Page 150

Figure 2.2.4d – Stability Analysis of the Final Model. In order to demonstrate that the model

can provide stable predictions of the behaviour of inhibitors, subsets of the Final Benchmark

Dataset were randomly chosen. These subsets consisted of half of the data points within the

Final Benchmark Dataset. Using the recorded Confusion Matrix outcomes (True Positive (TP),

True Negative (TN), False Positive (FP) and False Negative (FN)) of these data points, it was

possible to recalculate the predictive power of the model using the Matthews’ Correlation

Coefficient (MCC). Whilst the five subsets used each have different levels of predictive power,

the MCC values are not dissimilar to the Overall value calculated earlier in the section (shown in

Figure 2.2.4b). To demonstrate statistical significance a Chi Square Test was performed on all

data. The resulting α-value provides evidence of a model having a level of predictive power

which is statistically significant from random.

Page 151

Section – 2.3 - Discussion

Section 2.3.1 – Mechanism of Increasing eIF4E Phosphorylation in Response

to mTORC1 Inhibition

With regards to the proposed mechanism by which Rapamycin leads to an increase in eIF4E

phosphorylation, three theoretical mechanisms have been proposed to explain this.

However, it is likely that a combination of the three would be required. It seems possible in

order for any phosphorylation of eIF4E to occur during Rapamycin treatment that the cell

must contain a larger pool of this initiation factor than that of 4E-BP1. Following on from

this, experimentally a decrease in AKT1 phosphorylation on the mTORC2-dependent site is

seen, it is logical to assume that mTORC2 inhibition occurs, therefore, for eIF4E

phosphorylation to increase in response to Rapamycin, mTORC2 cannot be the only kinase

responsible for activating the eIF4G kinase, PKCα. Inhibition of an mTORC1-stimulated

protein phosphatase, for example PP5, would then occur. The result would be a system in

which eIF4E is phosphorylated by both an increase in the positive regulatory signals and a

decrease in the signals that negatively regulate eIF4E phosphorylation.

While the work presented in this chapter, offered the circumstances required for the

mTORC1-mediated increase in eIF4E phosphorylation, the finding that the phosphorylation

of this initiation factor is increased is not novel. Several studies have similarly noted this

effect (Sun et al., 2005; Wang et al., 2007c; Jin et al., 2008a; Grosso et al., 2011; Marzec et

al., 2011; Stead & Proud, 2013; Eckerdt et al., 2014). A more detailed summary of the

findings of these papers is given in Section 3.1.2 – eIF4E Phosphorylation During mTORC1

Inhibition.

Page 152

The 99.99% Binomial Confidence Intervals infer there is no effect of Rapamycin treatment

on the activation of ERK1 or ERK2 and this is in agreement with Marzec et al. (2011) and

Stead and Proud (2013). The notion of Rapamycin treatment having no effect on ERK1/2

activity (Oh et al., 2001b; Omura et al., 2005; Soares et al., 2013) is somewhat controversial,

with other studies noting an increase (Carracedo et al., 2008; Chaturvedi et al., 2009).

However, this can be explained by differences in the concentration of Rapamycin used

(Chen et al., 2010). This aside, and in agreement with the work of Stead and Proud (2013),

the model does predict that MNK2 mediates this alteration to eIF4E phosphorylation in an

ERK1/2-independent manner.

Section 2.3.2 – Predictive Power of the Final Version of the Model

The re-evaluation of the predictive power of the final version of the binary model displayed

an increase in eIF4E phosphorylation. This increase was achieved through the elimination of

molecular interactions linking eIF4G phosphorylation and MNK1 to the phosphorylation of

eIF4E. The elimination of these molecular interactions goes some way towards proving that

PKCα phosphorylation cannot rely solely upon mTORC2. The second change to the model

was to maintain a pool of eIF4E so that it was in excess over the level of 4E-BP1 and does

prove that there must be some disparity in the cellular pools of these factors. Only linking

the decrease in activity of a phosphatase was not attempted here. This was not investigated

as the nature of the relationship between mTORC1 and cellular phosphatases is not clear

and mechanistically difficult to implement.

As the final version of this model, it was important to ensure the predictive power was

maintained at a level at which the model is an accurate representation of the regulatory

network that controls translation initiation. The final version of the binary model was raised

Page 153

to a values of MCC = 0.4558. Given the probability of seeing this level of predictive power

by chance was very low (α = 1.94x10-8), the model has achieved a significant predictive

power.

In addition to measuring the predictive power of the model, it was also necessary to ensure

that the model had not been over-trained to the benchmark dataset against which it was

tested. To this end, a form of cross-validation was used. Cross-validation is a term from the

machine learning field (Bousquet & Elisseeff, 2002; Shipp et al., 2002; Arlot & Celisse, 2010).

All available data is divided into ‘training’ and ‘validation’ sets. Subsequently, the model is

automatically trained on the ‘training’ set and evaluated against the ‘validation’ set. This

division is then randomly changed and the process repeated. This methodology is not

applicable in the field of GSMN model reconstruction, as such models are constructed

manually rather than automatically. Manual reconstruction of large-scale models cannot be

repeated multiple times as it takes many months to complete by expert bio-curators (Terzer

et al., 2009; Henry et al., 2010 (as cited by McCloskey et al., 2013)). Consequently, none of

the models in the Constraint Based Modelling field is formally cross-validated in the same

way as in the machine learning field. Despite this controversy, the Constraint Based

Modelling field has achieved spectacular successes in biotechnology (Scheibe et al., 2009;

Fang et al., 2011) and recently has been gaining applications in the pharmaceutical industry

(Trawick & Schilling, 2006). This work further extends the manual, expert reconstruction

approach to the qualitative modelling of large-scale signalling networks.

With this in mind, work was conducted to demonstrate the predictive power of the model.

Firstly, the Benchmark Datasets of responses to external perturbations was constructed.

Such datasets were not were not used in the initial round of model construction.

Page 154

Consequently, the MCC of 0.2354 can be used to demonstrate the predictive power of the

model, prior to any refinement. This MCC value was found to be statistically significant.

Furthermore, the model has been used to make predictions regarding biological behaviour

which dedicated experimental work has been conducted to validate (detailed in Chapter 3 –

Experimental Validation of Model Predictions). This established that the model can be used

to make predictions which are useful in guiding future experimental work. This is in fact the

most important demonstration of the predictive power of the model.

A further indication that the model is not over-trained in a machine-learning sense is that it

is not perfect. It would be relatively easy to construct rules, such that all responses in the

Benchmark Dataset would be reproduced. This has not been done. Instead, the model has

been modified exclusively when there is peer-reviewed literature supporting the relevant

modification to a molecular interaction. This approach did not lead to a model with perfect

predictive power, but rather a model with statistically significant MCC value. Such a value

indicates that many discrepancies between the current state of knowledge on molecular

interactions and the behaviour of the model still exist. Such incongruities were left in the

model rather than being removed by arbitrary over-training.

Whilst the points listed above make a compelling argument against the model being over-

trained, the work detailed in Figure 2.2.4c provide an argument for a reader originating in a

machine-learning field. Each of the five subsets of the Benchmark Dataset were randomly

selected and yield a level of predictive power that was comparible to the overall level of

predictive power of MCC = 0.4558. Moreover, these subsets were capable of producing

levels of predictive power that were significantly better than random.

Page 155

Section 2.3.3 – Problems Encountered & Troubleshooting

The main problem encountered during the completion of work that contributed towards

this chapter occurred when trying to correct the decrease in eIF4E phosphorylation in the

Rapamycin model. Two issues became evident here. Firstly, dissecting useful data from the

large model proved difficult given that there is a high degree of cross-talk between the

signalling pathways. Secondly, the interrogation of the model by either Reachfq or Eventfq

did not readily lead to a potential mechanism.

These problems were overcome through the generation of a small scale model. This model

allowed for the easy identification of effectors critical for eIF4E phosphorylation. From here

it was possible to use the knowledge of these effectors to construct a literature-based

mechanism, as detailed in Section 2.2.3 – Mechanism by Which mTORC1 Inhibition Affects

eIF4E Phosphorylation.

Section 2.3.4 – Concluding Remarks

Results of this project are novel because, to date, no computational model of mammalian

translation has been produced. In contrast to a yeast model of translation produced by

Firczuk et al. (2013), the model presented here is entirely qualitative. Moreoever, the

model of Firczuk and co-workers (2013) is very much limited in scope with the model

encompassing only the translational machinery. The model presented here covers not only

the translational machinery but also the signalling network that regulates this process and

so is not comparable to the Firczuk et al. (2013) model. A quantitative model of the

signalling pathways regulating translation initiation could not have been constructed due to

a lack of quantitative parameters.

Page 156

The model has been subject to multiple rounds of model refinement. As a result of this

rigorous process, the predictive power has been raised from a questionable MCC = 0.2354

to a more sound MCC = 0.4558. This process has largely been achieved through increasing

the cross-talk between signalling pathways. Similarly, the main motivation of establishing

more cross-talk has been to reduce the number of FP readings whilst simultaneously

increasing the number of TP determinations. This was of considerable importance given

that these have more marked effects on the MCC values than either TN or FN numbers

(Baldi et al., 2000).

It has been argued that MCC values are inherently more meaningful than the AC values, as

the AC does not necessarily equate random predictions with zero (Baldi et al., 2000).

Furthermore, Baldi and colleagues (2000) take the view that the AC is ‘unnecessary’.

However, in this case with the MCC and AC being calculated in different ways, it was

included as a check to ensure that the calculated predictive power was significant.

The formation of a benchmark dataset for each inhibitor can be viewed as adding

considerable value to this work. It ultimately allowed for the predictive power of the model

to be quantified and improved upon. By ensuring that stimulation of cells was restricted to

the relevant dataset, it was possible to ensure that the inhibitors were the causative agents

producing the effects. Moreover, it has been shown that, although cell type differences

may be important in some cases, it is necessary to include multiple cell types in order to

ensure a large enough number of data points. The importance of cell type differences can

be reduced by verifying inhibitor effects across multiple studies using multiple cell types.

Over the course of the iterations to the model, two main hypotheses were formulated

about the regions of differences between the Wild Type and the Rapamycin models. There

Page 157

was an alteration in eIF4E phosphorylation when the Rapamycin was included in the model.

This finding was relevant as it was a prediction that linked a signalling component of the

model to the control of translation initiation and provided an opportunity to experimentally

validate the model. Secondly, the model inferred that AKT1 phosphorylation should also be

affected by mTORC1 inhibition with Rapamycin. Both of these predictions were be tested

and the results are given in Chapter 3 – Experimental Validation of Model Predictions.

While the work presented in this chapter, offered the circumstances required for the

mTORC1-mediated increase in eIF4E phosphorylation, the finding that the phosphorylation

of this initiation factor is increased is not novel. Several studies have similarly noted this

effect (Sun et al., 2005; Wang et al., 2007c; Jin et al., 2008a; Grosso et al., 2011; Marzec et

al., 2011; Stead & Proud, 2013; Eckerdt et al., 2014). A more detailed summary of the

findings of these papers is given in Section 3.1.2 – eIF4E Phosphorylation During mTORC1

Inhibition.

The 99.99% Binomial Confidence Intervals infer there is no effect of Rapamycin treatment

on the activation of ERK1 or ERK2 and this is in agreement with Marzec et al. (2011) and

Stead and Proud (2013). The notion of Rapamycin treatment having no effect on ERK1/2

activity (Oh et al., 2001b; Omura et al., 2005; Soares et al., 2013) is somewhat controversial,

with other studies noting an increase (Carracedo et al., 2008; Chaturvedi et al., 2009).

However, this can be explained by differences in the concentration of Rapamycin used

(Chen et al., 2010). This aside, and in agreement with the work of Stead and Proud (2013),

the model does predict that MNK2 mediates this alteration to eIF4E phosphorylation in an

ERK1/2-independent manner.

Page 158


Page 159

Section 3.1 – Introduction

In Chapter 2 – Model Creation & Refining two model predictions were generated,

concerning the effect Rapamyin had upon AKT1 and eIF4E phosphorylation. In the final

version of the model (detailed in Section 2.2.4 – Evaluation of the Predictive Power of the

Final Model) Rapamycin was found to increase eIF4E phosphorylation (as measured by an

increase in the frequency of trajectories leading to the production of eIF4E

phosphorylation), from 0.492 in the Wild Type model to 0.805 in the Rapamycin model.

Furthermore, AKT1 phosphorylation was shown to be decreased by the incorporation of

Rapamycin into the model, with the frequency of trajectories reaching phosphorylated AKT1

falling from 0.569 in the Wild Type model to 0.006 in the Rapamycin model. Both of these

behaviours required experimental validation to corroborate the predictions which are

detailed in this chapter.

Section 3.1.1 – Rapamycin

Rapamycin, or Sirolimus, is a macrolide agent derived from isolates of Streptomyces

hygroscopicus (Sehgal et al., 1975; Vézina et al., 1975; Rani et al., 2013). Rapamycin acts as

an immunosuppressive agent and potent antifungal agent (Vézina et al., 1975; Martel et al.,

1977; Kino et al., 1987a; Kino et al., 1987b (as cited by Abraham & Wiederrecht (1996)).

Both behaviours can be explained by Rapamycin acting to arrest the cell cycle in the G1

phase (Heitman et al., 1991; Kawamata et al., 1998; Decker et al., 2003; Zinzalla et al.,

2007).

The mechanism by which Rapamycin inhibits mTOR involves many of the 14 FK506-Binding

Proteins (FKBP) but FKBP-12 predominates (Wandless et al., 1991; Chiu et al., 1994; Fruman

et al., 1995; Kozany et al., 2009 (as cited by März et al., 2013)). Rapamycin forms a complex

Page 160

with FKBP12 which then binds to the FKBP12-Rapamycin Binding domain of mTOR located

close to the mTOR active site (Vilella-Bach et al., 1999; Banaszynski et al., 2005; Yang et al.,

2013a). As a consequence of this binding, the active site is hindered and thus, mTOR is

inhibited (Yang et al., 2013a).

Section 3.1.2 – eIF4E Phosphorylation During mTORC1 Inhibition

The general information regarding phosphorylation of eIF4E is discussed in Section 1.2.2 –

eIF4E Phosphorylation and The Role of ERK1/2 & p38. A number of studies have noted

that mTORC1 inhibition causes an increase in eIF4E phosphorylation (Sun et al., 2005; Wang

et al., 2007c; Jin et al., 2008a; Grosso et al., 2011; Marzec et al., 2011; Stead & Proud, 2013).

Stead and Proud (2013) noted that MNK2a appears to have greater inherent activity

through a reduction in the level of phosphorylation of the Serine residue at position 437 in

response to Rapamycin exposure. Furthermore, this greater basal level of activation occurs

without any input from the upstream mitogen-activated protein kinase, ERK1/2 (Ibid). This

finding is in keeping with the work by Eckerdt and co-workers (2014). These authors noted

that whilst eIF4E phosphorylation increased following Rapamycin treatment, there was no

concomitant increase in MAPK activity. Eckerdt et al. (2014) went further to find that the

intracellular 4E-BP1 concentration had little or no bearing on this level of increased eIF4E

phosphorylation.

Other studies noted that various other signalling pathways may be involved in this increase

in eIF4E phosphorylation. Both Sun et al. (2005) and Wang et al. (2007c) noted that the

PI3K pathway appears to be involved in this increase in eIF4E phosphorylation, with Wang

and colleagues (2007c) suggesting that PI3K is capable of influencing MNK activity through a

yet undefined mechanism. It is possible that this mechanism involves affecting the

Page 161

phosphorylation of the Serine residue identified by Stead and Proud (2013). However,

Marzec and colleagues (2011) found that this increase was independent of PI3K, ERK1/2 and

p38 activity while Grosso et al. (2011) hypothesised that p38 activity was responsible for the

phosphorylation of eIF4E in Rapamycin treated cells.

By interpreting the results presented in these works, one obvious and logical conclusion can

be drawn: mTORC1 inhibition leads to an increase in eIF4E phosphorylation. It would also

appear, based on these studies that the increase in activity of MNK1 or MNK2 that brings

about this increase is dependent upon neither ERK1/2 nor p38 activity. This suggests that

an, as yet, undefined but mTORC1-dependent, pathway is responsible for this increase.

Section 3.1.3 – Interplay Between mTORC1 & AKT1

The model of translational regulation requires not only an understanding of how the

mTORC1 pathway is activated, but also how feedback mechanisms regulate the upstream

component AKT1. By including such mechanisms in the model, it is then possible to better

model the complex patterns of behaviour observed in vivo. Such mechanisms include the

degradation of the Insulin Receptor Substrate (IRS)-1 (Haruta et al., 2000) and the mTORC1-

mediated inhibition of mTORC2 (Julien et al., 2010) and account for the increase in the

phosphorylation of AKT1 (Ikezoe et al., 2007; Breuleux et al., 2009; Chiarini et al., 2009),

observed after inhibition of mTORC1.

The first mechanism concerns the down-regulation of a protein responsible for the

induction of insulin receptor signalling, IRS-1 (Paz et al., 1996). Such down-regulation of IRS-

1 was initially noted as being Rapamycin-dependent (Haruta et al., 2000; Pederson et al.,

2001; Takano et al., 2001), as shown in Figure 3.1.3a, with S6K1 phosphorylating IRS-1 on

multiple amino acid residues, including Ser-307 and Ser-636 (Shah & Hunter, 2006;

Page 162

Tremblay et al., 2007; Zhang et al., 2008). This phosphorylation leads to the ubiquitin-

mediated degradation of IRS-1 (Greene et al., 2003; Xu et al., 2008; Scheufele et al., 2014).

As the activation of PI3K and AKT1 signalling is downstream of IRS-1, this mechanism limits

AKT1 signalling (Wan et al., 2007). A similar mechanism has also been shown for the

adaptor protein, Growth Factor Receptor-Bound Protein 10 (Hsu et al., 2011; Yu et al.,

2011).

Page 163

Figure 3.1.3a – Overview of the Feedback Mechanism between S6K1 and the Insulin Receptor

Substrate 1 (IRS-1). IRS-1, in response to Insulin, becomes phosphorylated, which results in the

activation of the PI3K signalling cascade and ultimately the activation of the mTORC1 pathway.

Once S6K1 is phosphorylated by mTORC1 and PDK1, this kinase is capable of phosphorylating

alternative sites of IRS-1, such as Ser-307 and Ser-636. This leads to the ubiquitination of IRS-1.

From this it is easy to see how inhibition of mTORC1 results in an increase in the activity of PI3K

signalling components including AKT1. It is worth noting that full activation of AKT1 requires both

PDK1 and mTORC2. However, for simplicity, in this case, only PDK1 has been shown.

Page 164

The second mechanism by which mTORC1 regulates AKT1, involves the negative regulation

of mTORC2. S6K1 is capable of phosphorylating the Thr-1135 residue of the Rictor

component of mTORC2, resulting in impaired activity towards mTORC2 substrates, including

AKT1 (Julien et al., 2010) as shown in Figure 3.1.3b. Additionally, S6K1 and AKT1 have been

shown to phosphorylate another mTORC2 component, mSIN1, to trigger mTORC2 inhibition

(Liu et al., 2013).

Page 165

Figure 3.1.3b – Overview of the Feedback Mechanism that Exists between S6K1 and

mTORC2. The full activation of AKT1 requires phosphorylation of two sites (Ser-473 and Thr-

308) by mTORC2 (Sarbassov et al., 2005) and PDK1 (Wan & Helman, 2003; Tsuchiya et al.,

2013; Dangelmaier et al., 2014), respectively. Downstream of AKT1 lies mTORC1 and the

kinase S6K1. mTORC1, along with PDK1, are responsible for the full activation of S6K1. A

feedback mechanism between mTORC2 and S6K1 has recently been identified by Julien et al.

(2010). In this mechanism the mTORC2 component, Rictor, is phosphorylated via S6K1 on

Thr-1135. A second, analogous mechanism may involve the S6K1- and AKT1-dependent

phosphorylation of another mTORC2 component, mSIN1 (Liu et al., 2013). Both mechanisms

would result in an inhibition of mTORC2. Phosphorylation, resulting in activation, is denoted

with a blue circle whilst those having an inhibitory effect are shown by red circles.

Page 166

Whilst both mechanisms could explain an increase in AKT1 phosphorylation during mTOR

inhibition, neither can explain a decrease. As such, it is necessary to consider that

Rapamycin may affect mTORC2. Only nascent mTOR, or mTOR from disassembled mTORC1

and mTORC2, is free to interact with the FKBP12-Rapamycin complex to be incorporated

into inhibited mTORC2 (Brown et al., 1994, Sabatini et al., 1994; Sabers et al., 1995 (as cited

by Sarbassov et al., 2006)). Diminished AKT1 phosphorylation has been noted after

prolonged or excessive Rapamycin exposure, suggesting Rapamycin is perhaps capable of

inhibiting mTORC2 in a cell-type or dose-dependent manner (Sarbassov et al., 2006; Zeng et

al., 2007; Wang et al., 2008a; Chen et al., 2010). In keeping with this, decreased AKT1

phosphorylation was seen at higher Rapamycin concentrations while increased at lower

concentrations (Chen et al., 2010). However, the ability of Rapamycin to inhibit mTORC2 is

dependent upon the ratio of FKBP12 to FKBP51 expression in cells. Cell lines with a higher

ratio of these proteins are more receptive to mTORC2 inhibition (Schreiber et al., 2015).

Alongside the aim of creating a model of the translation initiation regulatory pathways, this

project must experimentally validate predicted behaviours of the model. The information

presented in the introduction to this chapter highlights that the regulation of AKT1 and

eIF4E by mTORC1 is not fully understood. This chapter aimed to validate the predictions of

the model but not to provide a hypothesised mechanism. Work presented in Chapter 2 –

Model Creation & Refining, details a computationally elucidated mechanism by which eIF4E

phosphorylation is affected by Rapamycin.

Page 167

Section 3.2 – Materials & Methods

Section 3.2.1 – Growth & Passage of RAW 264.7 Murine Macrophage Cells

RAW 264.7 cells are an Abelson Murine Leukaemia Virus-transformed murine macrophage

cell line (Raschke et al., 1978). These cells (from European Collection of Cell Cultures,

Product Number: 91062702, Lot: 13J020) were cultured in a vented T75 flask (Nunc™,

Thermo Fisher Scientific Inc., Rockford, IL (USA)) using media containing 89% high glucose

(4.5g per litre) Delbecco’s Modified Eagle Medium, containing Sodium Pyruvate and L-

Glutamine, 10% Foetal Bovine Serum and 1% Penicillin and Streptomycin (Gibco®, Life

Technologies Corp., Carlsbad, CA (USA)). The cells were incubated at 37°C with 5% Carbon

Dioxide in a humidified Galaxy S incubator (Wolf Laboratories Ltd, York (UK)). Cells were

passaged at around 70% confluence, with each passage requiring new media, and the cells

were detached from the T75 flask using 3mm glass beads (Sigma-Aldrich Co. LLC, St. Louis,

MO (USA)).

Section 3.2.2 – Determination of Rapamycin Cytotoxicity

Given the qualitative behaviour of the model, the action of inhibitors can best be described

as ‘all or nothing’ with regard to the effect these chemicals have on the main target. To this

end it was decided that a high concentration of Rapamycin would be used. Jin et al. (2009)

had already established data concerning the viability of RAW 264.7 cells with a combination

of Rapamycin, up to a concentration of 10μM, and LPS. This paper was used to guide the

selection of Rapamycin concentrations to test and three concentrations of Rapamycin would

be tested (10μM, 1μM and 0.1μM) in the absence of LPS.

Cytotoxicity of Rapamycin was determined by measuring the concentration of Adenosine

Trisphosphate in the cell media after treatment. For this purpose the CellTiter Glo®

Page 168

Luminescent Cell Viability Assay kit (Promega UK, Promega Corp., Madison, MI (USA)) was

used. The Rapamycin (Tocris Biosciences, R&D Systems Inc., Minneapolis, MN (USA)) was

reconstituted in Dimethyl Sulfoxide (DMSO) (Sigma-Aldrich Co. LLC, St. Louis, MO (USA)).

Rapamycin was diluted to the desired concentration in RAW 264.7 cell media (the

composition of which is detailed in Section 3.2.1 – Growth and Passage of RAW 264.7

Murine Macrophage Cells). Care was taken to ensure that the final concentration of DMSO

was not greater than 1%.

At approximately 70% confluence, RAW 264.7 cells were seeded in a white-walled 96-well

plate (Corning Incorporated, Tewksbury, MA (USA)) and incubated overnight. The cells were

treated with three concentrations of Rapamycin (10μM, 1μM and 0.1μM) and two controls

of 1% DMSO and cell media. Additionally, to control for potential inherent luminescent

properties of Rapamycin, DMSO or indeed the media, some samples containing no cells

were also included. In all cases, samples were seeded in triplicate.

After three hours of incubation, the plate was left at room temperature for 30 minutes.

Reconstituted CellTiter Glo® Reagent was added to each well in a one-to-one ratio with the

amount of media already present in the well. Following a ten minute incubation at room

temperature, the luminescence of each well was recorded by a Glomax® 96 Microplate

Luminometer (Promega UK, Promega Corp., Madison, MI (USA)).

The results of this assay were analysed by subtracting the average luminosity of the wells

containing no cells from each of the readings for the corresponding treatment. The data

was then normalized to the DMSO control. Statistical analysis of this data will be discussed

in Section 3.2.7 – Statistical Analysis.

Page 169

Section 3.2.3 – Cell Harvesting & Bicinchoninic Acid (BCA) Assay

RAW 264.7 cells, at around 70% confluence in a T75 flask, were seeded into untreated 6-

well plates (Nunc™, Thermo Fisher Scientific Inc., Rockford, IL (USA)) and incubated

overnight at 37°C with 5% Carbon Dioxide. The following day, the cells were treated with

either 10μM Rapamycin or 1% DMSO for a period of either one or three hours (as stated in

the subsequent Section 3.2.4 – Human Phospho-MAPK Array and Section 3.2.5 –

eIF4E/Phosphorylated eIF4E & AKT1/Phosphorylated AKT1 Western Blot Analysis).

Prior to lysis it was necessary to remove the treatment media and ensure as much serum

was removed as possible, thus ensuring accurate measurement of cellular protein

concentration. To this end the cells were washed twice in chilled, sterile one-times

Phosphate-Buffered Saline (PBS) solution. The cells were then lysed in 70μl of Lysis Buffer 6,

taken from the Human Phospho-MAPK Array kit (R&D Systems, Inc., Minneapolis, MN

(USA)), and incubated at 4°C for 30 minutes to complete lysis. The lysate was centrifuged at

4°C for 5 minutes and the pellet discarded.

The supernatant would later be used for both Western Blot analysis and the Human

Phospho-MAPK Arrays. Prior to this however, it was necessary to determine the

concentration of protein within the lysates. This was done using a BCA Assay kit (Pierce™,

Thermo Fisher Scientific, Inc., Rockford, IL (USA)) and a 96-well Nunclon™ plate (Nunc™,

Thermo Fisher Scientific Inc., Rockford, IL (USA)). The standard curve was produced

according to the manufacturer’s instructions. 25μl of each sample was loaded, in duplicate,

onto the plate. The resulting colour change was analysed with a Victor™ X5 Multilabel Plate

Reader (Perkin Elmer, Inc., Waltham, MA (USA)). Based upon the results of this assay, the

Page 170

protein concentration in the samples could be adjusted to either 400μg or 50μg of protein

per ml, for the Human Phospho-MAPK Array or Western Blotting, respectively.

Section 3.2.4 – Human Phospho-MAPK Arrays

RAW 264.7 cells were treated for one hour with either 10μM Rapamycin or 1% DMSO and

harvested, according to the method outlined in the previous section. Following the BCA

assay, the protein concentrations in both lysates were adjusted to 400μg of protein per

millilitre with MIlli-Q® Ultrapure (EMD Millipore Corp., Merck KGaA, Darmstadt (DE)) water

and carried out in accordance with the manufacturer’s instructions. This method required

20μl of the Detection Antibody Cocktail being incubated with the lysate at room

temperature for one hour prior to incubation with the membrane overnight at 4°C. The

array was subsequently developed onto Super-RX radiographic film (Fujifilm Corporation,

Tokyo (JP)). These arrays were done in triplicate, each with new lysates.

The results of the array were analysed in the software ImageJ (Rasband, 1997-2012). This

software analyses the pixel colour density of each spot on the membrane. As each protein

is represented on the membrane in duplicate, the average is taken for each protein. A value

for the background pixel colour density is also taken and subtracted from the average value

for each protein. Normalization of all the data to the DMSO-treated controls allows for the

identification of trends.

Section 3.2.5 – eIF4E/Phosphorylated eIF4E & AKT1/Phosphorylated AKT1

Western Blot Analysis

Once determined by the results of the BCA assay, the protein concentration of each lysate

was adjusted to 50μg of protein with MIlli-Q® Ultrapure water and Red Loading Buffer (New

Page 171

England BioLabs Inc., Ipswich, MA (USA)). Protein samples were separated using 4-20%

gradient Mini-PROTEAN® TGX (Bio Rad Laboratories, Inc., Hercules, CA (USA)) Sodium

Dodecyl Sulphate (SDS) Polyacrylamide Gel Electrophoresis (SDS-PAGE) gels with an SDS-

based Running Buffer (for solutions which are made in-house, the constituents are shown in

Supplementary Information 4. Prestained Protein Marker (P7706) (New England BioLabs

Inc., Ipswich, MA (USA)) with a resolution of between 190kDa and 11kDa was included

alongside the samples. Samples were heated to 100°C for five minutes prior to loading.

Proteins were transferred from the gel onto Immobilon-P (EMD Milipore Corp, Merck KGaA,

Darmstadt (DE)) Polyvinylidene Fluoride (PVDF) membrane at 100V for 60 minutes in

Transfer Buffer solution. Ponceau staining solution was used to confirm the transfer from

the gel to the PVDF membrane. The membrane was cut to avoid the need for stripping the

membrane between primary antibodies. A blocking solution was added to the membrane

and incubated for one hour, or in the case of the primary α-Tubulin antibody, overnight at

4°C. The antibodies used in testing the model validity are shown in Table 3.2.5a.

Following the appropriate incubation period for each antibody, it was removed and washed

in PBS-Tween solution. The secondary antibody solutions were prepared in 0.5% (w/v)

Marvel® milk powder (Premier Foods PLC, St. Albans (UK)) in PBS-Tween solution. For α-

Tubulin and both eIF4E and Phospho-eIF4E antibodies, the anti-mouse (Dako Denmark A/S,

Glostrup (DK)) and anti-rabbit (Dako Denmark A/S, Glostrup (DK)) secondary antibodies,

respectively, were incubated at room temperature for one hour. All antibodies used in the

testing of the AKT1 phosphorylation prediction required anti-rabbit secondary antibodies

(Dako Denmark A/S, Glostrup (DK)).

Page 172

Antibody Diluting

Solution

Amount of

Antibody

Incubation

Period

Supplier

Rabbit Phospho-

eIF4E (Ser-209)

#9741

5% BSA in PBS-

Tween

15μl (1:667) Over Night Cell Signaling

Technology, Inc.,

Danvers, MA (USA)

Rabbit eIF4E

#9742

5% BSA in PBS-

Tween


Technology, Inc.,

Danvers, MA (USA)

Rabbit AKT1

#9272

5% BSA in PBS-

Tween


Technology, Inc.,

Danvers, MA (USA)

Rabbit Phospho-

AKT1 (Ser-473)

#4060

5% BSA in PBS-

Tween


Technology, Inc.,

Danvers, MA (USA)

eIF6

10291-1-AP

5% BSA in PBS-

Tween

10μl (1:1000) Over Night Proteintech Group, Inc.,

Chicago, IL (USA)

α-Tubulin

T9026

PBS-Tween 2μl (1:5000) One Hour Sigma-Aldrich Co. LLC,

St. Louis, MO (USA)

Table 3.2.5a – Summary of the Antibody Solutions Used in Testing eIF4E and AKT1

Phosphorylation.

Page 173

The PVDF membranes were then washed in PBS-Tween solution and incubated for one

minute in SuperSignal West Pico Chemiluminescent Substrate (Pierce®, Thermo Fisher

Scientific Inc., Rockford, IL (USA)). Each membrane was then developed onto Super-RX

radiographic film. As with the Human Phospho-MAPK Array kit, quantification of each band

was carried out in ImageJ and normalized to the DMSO lysate which acts as a control.

Section 3.2.6 – Re-Evaluation of the Model Based Upon the Results of the

Human Phospho-MAPK Arrays

Using the methods described in Chapter 2 – Model Creation & Refining the binary model

was interrogated against the data collected in the by the Human Phospho-MAPK Array kits.

Simulations were run looking at the activation of effectors that are found in both the model

and the array. The length of simulations was adjusted to ensure that the frequency of

activation of each effector was around 0.5 in the Wild Type model. Simulations of the

model containing Rapamycin was run for the same lengths of time. By comparing the

99.99% Binomial Confidence Intervals for each effector it was possible to find regions of

significant difference. As previously stated, the Binomial Confidence Intervals were

calculated in the statistical software R (R Core Development Team, 2013).

To ensure that the only rigorous differences in the activation are found, only array

constituents displaying a difference that remains significant after the Holm-Sidak Correction

(as described in Section 3.2.7 – Statistical Analysis) was considered. Given the qualitative

nature of the model, only an increase, decrease or no effect were recorded.

Page 174

Section 3.2.7 – Statistical Analysis

This section will cover the methods used to evaluate the significance of results obtained in

the cell viability assay, array or in Western Blotting. Most of the statistical analysis was

done using GraphPad Prism 6 (Version 6.05, for 64 bit Windows, GraphPad Software, Inc., La

Jolla, CA (USA)). In all cases the statistical analysis applies only to normalized data.

The CellTiter Glo® Assay generated results for five independent variables (namely the three

inhibitor concentrations, one DMSO treatment and the untreated, media-only, control).

With this in mind, it was necessary to determine the distribution of the data. Whilst it was

not possible to determine, using GraphPad Prism, if the data points for each variable were

normally distributed, it is possible to determine if the Standard Deviation of each group

displays no significant difference. This can be done using the Brown-Forsythe Test (Brown &

Forsythe, 1974) calculated as part of the Analysis of Variance (ANOVA) function. Given that

the normality of the data cannot be determined, and assuming that the Standard Deviation

of each group is not significantly different, the Kruskal-Wallis Test should be used in

preference of a one-way ANOVA (Kruskal & Wallis, 1952). No significant difference being

found between concentrations of Rapamycin or between the concentrations and the DMSO

or media controls, infers that cell viability is not affected by the presence of this inhibitor.

The data collected from both the Human Phospho-MAPK Array data and Western Blot

analysis is suitable to analysis with t-tests. Once the average intensity of each

phosphoprotein on the array is known and normalized (in triplicate), it can be analysed in

GraphPad Prism using the multiple t-test function. In order to correct for individual tests

returning significant results solely due to multiple testing, a Holm-Sidak Correction (Sidak,

1967; Holm, 1979) was applied. The Western Blot data was also corrected, but in this case

Page 175

a Bonferroni Correction was applied. The p-value, calculated using an unpaired, non-

parametric t-test, was multiplied by the number of t-tests it was possible to carry out on a

dataset (Emily et al., 2009). In this case, the p-value was multiplied by three as the results

were obtained in triplicate. This method of p-value correction is preferable as it is viewed as

very orthodox and results in ‘over-corrected’ (McIntyre et al., 2000 (As cited, and quoted, by

Klein et al., 2005)) test statistics.


Section 3.3.1 – Cytotoxicity of Rapamycin

As the behaviour of a qualitative model treats the action of inhibitors as binary, in that the

result is complete inhibition or complete activity of the target effector, it was decided that

for experimental validation of predictions a high concentration of inhibitor would be used.

To determine which concentration of Rapamycin does not have any effect on cell viability,

the CellTiter Glo® Assay was used which measures the concentration of ATP released from

cells lysed due to exposure to Rapamycin.

As shown in Figure 3.3.1a, it appears that the mTORC1 inhibitor has mild, positive effects on

cell viability. As determined by the Kruskal-Wallis test though, none of these effects are

significantly different from either the 1% DMSO treatment or the effects of media alone

with the resulting p-values never reaching below 0.5437 (comparing 0.1μM Rapamycin and

1% DMSO). Although none of the concentrations tested produced a significant effect on cell

viability, a concentration of 10μM was chosen as it represented the highest concentration

tested. Consequently, this concentration would be most in keeping with the qualitative

behaviour of the model in which the action of inhibitors is viewed in an all or nothing

manner.

Page 176

Figure 3.3.1a – Evaluation of the Effects of Three Hour Treatments with Differing

Concentrations of Rapamycin (Rapa) or 1% DMSO on the Murine Macrophage RAW 264.7 Cells.

The effects on viability were measured by changes to the amount of ATP present in the

supernatant following cell lysis. The concentration of DMSO was never higher than 1%. All data

is given as relative to the effect of 1% DMSO. The Kruskal-Wallis test was used to evaluate

significance, with ‘ns’ denoting non-significant. Error bars represent Standard Error of the Mean

values for each dataset.

ns

ns ns ns

ns ns ns ns

Page 177

Section 3.3.2 – Evaluation of the Human Phospho-MAPK Arrays

The binary model of translational control demonstrated that whilst eIF4E phosphorylation

was affected by the presence of an mTORC1 inhibitor, the MAPKs, ERK1/2 and p38, were

not. Similarly, the level of phosphorylated AKT1 is predicted to be reduced during exposure

to a high concentration of this macrolide. It was important to establish if either MAPK or

AKT1 was affected by treatment with Rapamycin. The Human Phospho-MAPK Array kits

offered a viable method to test all of these effects in a simple way. RAW 264.7 cells were

treated with either 10μM Rapamycin or 1% DMSO for one hour. The array demonstrates, as

shown in Figure 3.2.2a, that Rapamycin treatment has broad implications for MAPK

signalling in RAW 264.7 cells.

Page 178

AK

T1

AK

T2

AK

T3

AK

T P

an

CR

EB

ER

K1

ER

K2

GS

K3

/ß

GS

K3

ß

HS

P2

7

JN

K1

JN

K2

JN

K3

JN

K P

an

MK

K3

MK

K6

MS

K2

p3

8

p3

8ß

p3

8

p3

8

p5

3

S6

K

RS

K1

RS

K2

mT

OR

0 .0

0 .2

0 .4

0 .6

0 .8

1 .0

1 .2

N o rm a liz e d D a ta - P o o le d A rra y D a ta C o m p a rin g 1 0 M

R a p a m y c in T re a tm e n t (1 h r ) V s . D M S O (1 % ) (C o n tro l) T re a tm e n t (1 h r)

S p e c ie s

Re

lati

ve

Le

ve

l o

f P

ho

sp

ho

ryla

tio

n

D M S OR a p a m y c in

** ** ** ** ** ** ** ** * * * * *

Figure 3.3.2a – Evaluation of the Broad Early Effects of 10μM Rapamycin Treatment to the Murine Macrophage RAW 264.7 Cells. Following treatment of RAW 264.7

cells with either 10μM Rapamycin or 1% DMSO for one hour, cells were lysed using the Lysis Buffer 6 from the Human Phospho-MAPK Array Kit (R&D Systems Inc.,

Minneapolis, MN (USA)). After the protein concentration of each lysate was adjusted to 400μg, the array was carried out in accordance with the manufacturers’

instructions. The results were analysed in ImageJ. T-tests were carried out on each phosphoprotein and were adjusted for multiple comparisons using the Holm-Sidak

Correction. Those phosphoproteins that were significant after the t-test are denoted with *. If they remained significant after the Holm-Sidak Correction they are

denoted with **. Error bars represent the Standard Error of the Mean.

** * * * *

Kinases

Page 179

It was necessary to prove that Rapamycin was having an effect on the marketed, target

substrate. The Human Phospho-MAPK Array allows for measuring the level of mTOR

phosphorylated on the Ser-2448 residue. The phosphorylation of this Serine residue has

been linked to S6K1 and 4E-BP1 phosphorylation (Dickinson et al., 2011) and has been used

as a marker of mTORC1 activity (Altman et al., 2011). This phosphorylation site has been

shown to be responsive to Rapamycin treatment (Copp et al., 2009; Tyler et al., 2009;

Carayol et al., 2010). In this MAPK array, it was established that Rapamycin produced a

statistically significant 41% decrease (p=1.95x10-4) in the phosphorylation of Ser-2448 of

mTORC1, as shown in Figure 3.3.2a. This p-value is that calculated after a Holm-Sidak

Correction was applied.

Establishing the activity of ERK1/2 and p38 in response to Rapamycin was an important

determination with regards to experimentally validating the alterations to the level of

phosphorylated eIF4E in Rapamycin treatment. As shown in Figure 3.3.2a, neither kinase

was affected by Rapamycin treatment. The effect of Rapamycin on both ERK1/2 and p38 is

not fully understood. ERK1/2 has been shown to be subject to a diverse array of effects,

with one study noting an increase (Chaturvedi et al., 2009), another a decrease (Chen et al.,

2010) and several displaying no effect (Omura et al., 2005; Soares et al., 2013). Although

the work of Chaturvedi et al. (2009) and Chen et al. (2010) indicated that the concentration

of Rapamycin may be important in determining the direction of change, the work presented

here is more in keeping with Omura et al. (2005) and Soares et al. (2013). Several studies

have noted that a feedback mechanism exists between ERK1/2 and mTORC1 (Carracedo et

al., 2008; Melemedjian et al., 2013), however the finding that Rapamycin does not affect

Page 180

ERK1/2, after a treatment of one hour at least, suggests that this mechanism could either be

absent in RAW 264.7 cells or become important much later.

As with ERK1/2, the published view of the relationship between Rapamycin and p38 is

equally contradictory. Oh et al. (2001b) and Omura et al. (2005) both found that p38

activity was increased by Rapamycin. Despite these publications contradicting the work

presented here, several other studies are in agreement and note that Rapamycin has no

effect on p38 activity (Cuenda & Cohen, 1999; Huang et al., 2003). Whilst no definitive

explanation for this can be given, it may reflect differences in cell type.

With regards to regions of significant difference, 18 regions were noted as being different

and, of these, nine regions were found to remain significant after the Holm-Sidak Correction

was applied. These regions are denoted with * and ** respectively in Figure 3.3.2a. Of

particular interest here is the finding that both AKT1 and AKT2 display significant inhibition

of 65.7% and 50.5%, respectively. Whilst the phosphorylation of both AKT1 and AKT2 was

significantly reduced (with p = 0.018 and p = 0.014) neither were found to remain significant

following the Holm-Sidak Correction. However, the reduction (on average of 51.5%) of AKT

Pan phosphorylation, representing the total of phosphorylated AKT1, AKT2 and AKT3, in

response to Rapamycin remains significant after this correction. Other differences that are

significant in this robust sense are JNK3, MKK3, MKK6, Ribosomal Protein S6 Kinase α-4

(MSK2), p38δ, p53 and RSK2.

The results outlined in Figure 3.3.2a clearly demonstrate that Rapamycin has a pronounced

inhibitory effect on a broad range of cellular targets. This could reflect the central role of

mTORC1 in the signalling network of the cell. The work of Caron and co-workers (2010)

clearly demonstrates that mTORC1 is linked to a wide variety of cellular processes and

Page 181

pathways. It is feasible that short-term exposure to Rapamycin, as used here, would have

such pronounced effects on the signalling pathways examined by with these Human

Phospho-MAPK Arrays.

Section 3.3.3 – Alterations to eIF4E Phosphorylation Levels

Phosphorylated eIF4E was not present on the Human Phospho-MAPK Arrays and thus it was

not possible to test the effects of Rapamycin on the level of eIF4E phosphorylation using the

approach described in the previous section. In order to test the hypothesis that inhibition of

the mTOR pathway by Rapamycin alters the level of eIF4E phosphorylation, immunoblotting

was used. Murine Macrophage RAW 264.7 cells were treated with either 10μM Rapamycin

or 1% DMSO for one hour and lysed using the Lysis Buffer 6 supplied with the Human

Phospho-MAPK Array kit. After this time period, the protein concentration was adjusted to

50mg per millilitre. Although this was only conducted once, there was a mild, 18% increase

in phosphorylated eIF4E in response to Rapamycin treatment (data not shown).

In order to further investigate this effect, RAW 264.7 cells were treated for three hours with

either 10μM Rapamycin or 1% DMSO. As expected, quantification of the pixel intensity in

ImageJ revealed that the increase in treatment length resulted in a greater increase in

phosphorylated eIF4E of, on average, 44.0% (range of between 36.4% and 52.2%, p =

0.0318) in the Rapamycin-treated RAW 264.7 cell lysates (as shown in Figure 3.3.3ai-ii). The

significance of this increase was determined using a two-tailed t-test with a Bonferroni

Correction being applied to the resulting p-value. The Bonferroni Correction was applied

and the results of the two-tailed t-test were multiplied by three to account for the three

independent, biological replicates conducted.

Page 182

Figure 3.3.3a – Rapamycin-Induced Increase in Phosphorylation of eIF4E. RAW 264.7 Cells were

treated with 10μM Rapamycin or 1% DMSO for three hours. 50mg per ml of protein, from each

lysate, were separated using SDS-PAGE and phosphorylated eIF4E (25kDa), eIF4E (25kDa) and α-

Tubulin (50kDa) were detected using immunoblotting. i) A representative blot of the increase in

eIF4E phosphorylation due to Rapamycin treatment. The levels of total eIF4E and the α-Tubulin

loading control remain constant. ii) The pooled quantitative data of three independent

replicates showing the mean ± Standard Error of the Mean. This data is expressed as the level of

phosphorylated eIF4E relative to the 1% DMSO control. This quantification was tested using a

two-tailed t-test after which a Bonferroni Correction was applied to the calculated p-value.

Page 183

Section 3.3.4 – Confirmation of the Effect Rapamycin has Upon AKT1

Phosphorylation

Despite the fairly conclusive result generated in the Human Phospho-MAPK Arrays regarding

a decrease in AKT1 phosphorylation on the Ser-473 residue, the effect was further

investigated using immunoblotting. RAW 264.7 cells were treated with either 10μM

Rapamycin (in 1% DMSO) or 1% DMSO for three hours. As was the case for the

eIF4E/Phospho-eIF4E immunoblotting experiment, it was decided to attempt to accentuate

the affect by increasing the length of Rapamycin treatment to three hours.

As shown in Figure 3.3.4a, only a small, non-significant difference in the levels of Ser-473

phosphorylated AKT1 was found between the Rapamycin-treated and control lysates. On

average the level of phosphorylated AKT1 was 20.4% (range of between 3.2% and 31.5%, p =

0.4344) higher in the Rapamycin-treated lysates. Reasons for this lack of significant

difference are discussed in Section 3.4.2 – Validation of the AKT1 Phosphorylation

Prediction.

Page 184

Phospho-AKT1

(Ser-473)

Total AKT1

eIF6

p = 0.4344

i. ii.

Figure 3.3.4a – Effect of Rapamycin on the Levels of AKT1 Phosphorylation (Ser-473). RAW 264.7

Cells were treated with either 10μM Rapamycin (in 1% DMSO) or 1% DMSO for three hours. 50mg

per ml of protein from each lysate were separated using SDS-PAGE and phosphorylated AKT1

(60kDa), AKT1 (60kDa) and eIF6 (27kDa) were detected using immunoblotting. i) The blot shown is

representative of the results obtained. Only very low levels of AKT1 phosphorylation (Ser-473) were

noted. The levels of the eIF6 loading control and total AKT1 remain constant. ii) The pooled data

collected from three independent replicates ± the Standard Error of the Mean. The data shown was

quantified in the software ImageJ and is expressed as a ratio of the DMSO control. The p-value

shown was calculated using a Bonferroni Correction applied to the results of a two-tailed t-test.

Page 185

Section 3.3.5 – Comparison of Model Results to Array Data

Out of all the phosphoproteins recorded on the array, 13 were found to be present in the

binary model. As shown in Figure 3.3.5a, nine effectors displayed behaviour that was in

agreement with the model. Simulations performed to generate these results were run using

QSSPN on the final version of the binary model, the overall results of which are detailed in

Section 2.2.4 – Evaluation of the Predictive Power of the Final Model. The results shown in

this section were included in the overall analysis of the final model.

The comparison of the model to the array data was done to show that the behaviour of the

model correlates to experimental outcomes detailed in this chapter. Furthermore, the

additional testing of the model can be seen as adding further weight behind the view that

the model is capable of mimicking experimental data. The Matthews’ Correlation

Coefficient (MCC) for this comparison is calculated using the equation given in Section

2.1.4.2 – Evaluation of the Predictive Power of the Model. The MCC calculated for

available array data is 0.3958.

Page 186

Effector Behaviour of the Effector in the Array Predicted

ppERK1 No Effect TN

ppERK2 No Effect TN

pRSK1 No Effect TN

p38α No Effect TN

pJNK1 No Effect TN

mTORC1-Act Inhibition Demonstrated TP

ppS6K No Effect FP

ppAKT Inhibition Demonstrated TP

pMKK3 Inhibition Demonstrated FN

pMKK6 Inhibition Demonstrated FN

pMSK2 Inhibition Demonstrated FN

pCREB No Effect TN

pGSK3 Inhibition Demonstrated TP

Figure 3.3.5a – Comparison of Array Data with the Behaviour of the Binary Model. In

keeping with the qualitative nature of the model, the array data was analysed as

showing 10μM Rapamycin having no effect or being an inhibitor or activator of an

effector. The Binomial Confidence Intervals calculated for each effector allowed for

differences to be found between the Wild Type or Rapamycin models. Comparing the

two datasets permitted a quantitative analysis of the predictive power of the model.

Page 187

Section 3.4 – Discussion

Section 3.4.1 – Validation of the eIF4E Phosphorylation Prediction

The mTORC1 pathway is integral to growth and proliferation of cells (Laplante & Sabatini,

2013) and, as of 2010, can be viewed as including close to one thousand molecular species

(Caron et al., 2010). Despite this broad knowledge base regarding the interactions of

mTORC1 with other signalling components and pathways, the exact nature of the mTORC1

input into many processes, for example eIF4E phosphorylation, remains unclear.

In keeping with the prediction generated by the model, a number of studies have reported

an increase in eIF4E phosphorylation in response to mTORC1 inhibition (Sun et al., 2005;

Wang et al., 2007c; Jin et al., 2008a; Grosso et al., 2011; Marzec et al., 2011; Stead & Proud,

2013). Marzec and colleagues (2011) noted that this increase may be independent of

ERK1/2 or p38α activity. However, Grosso et al. (2011) argued that the remaining p38

activity following mTORC1 inhibition, may account for the apparent increase in eIF4E

phosphorylation. The former is supported by the work of Stead and Proud (2013). In this,

they argued that the interaction between ERK1/2 and MNK2 is not affected by Rapamycin-

mediated mTORC1 inhibition (Ibid).

The experimental results derived in this project have shown that the Rapamycin treatment

produces a notable increase in eIF4E phosphorylation after just three hours of exposure.

Whilst no attempt was made to distinguish whether this was through MNK1 or MNK2

(discussed in Section 3.4.3 – Problems Encountered & Troubleshooting) the activity of both

ERK1/2 and p38 were recorded in response to Rapamycin treatment. After one hour of

exposure there was no indication of the activity of either kinase increasing.

Page 188

It can be argued that the work here is in agreement with that shown by Stead and Proud

(2013). In this work it was noted that Rapamycin was able to increase eIF4E

phosphorylation through a MNK2-dependent but ERK1/2-independent manner. Here the

array data has shown no effect on the levels of phosphorylated ERK1/2 or p38α. In fact,

even one hour of Rapamycin treatment seemed to produce a reduction, albeit non-

significantly, in the levels of these phosphorylated kinases. In keeping with this concept, the

idea that at higher concentrations (at greater than 1μM) of Rapamycin there is concomitant

inhibition of mTORC1 and mTORC2 which leads to inhibition of not just ERK1/2 but also

AKT1 has been demonstrated (Chen et al., 2010). This suggests that increasing the

treatment length would produce ERK1/2 inhibition rather than activation.

As already noted in Section 1.2.2 – eIF4E Phosphorylation and The Role of ERK1/2 & p38,

eIF4E phosphorylation is now being seen as a mechanism by which genes associated with

stress responses are being preferentially up-regulated (Andersson & Sundler, 2006; Herdy et

al., 2012; Royall et al., 2015). It is also of interest to note the MNK kinases are also capable

of phosphorylating the Sprouty proteins (DaSilva et al., 2006; Sharma et al., 2012) and the

Heterogenous Nuclear Ribonucleoprotein A1 (hnRNP A1) (Buxadé et al., 2005). Both of

these have roles in the physiological response to cellular stress. The Sprouty proteins are

required for the negative regulation of Interferon-Stimulated Gene 15 and p38 (Sharma et

al., 2012), and the response is MNK-dependent. Furthermore, phosphorylated hnRNP A1 is

known to play a role in the movement of mRNA to stress granules (Guil et al., 2006).

The role of phosphorylated eIF4E in the physiological response to stress and the finding that

inhibition of mTORC1 leads to increased resting activity of MNK2, suggest that mTORC1 is

capable of suppressing the cellular stress response through modulation of MNK activity, as

Page 189

shown in Figure 3.4.1a. This hypothesis is supported, albeit indirectly, by the finding that

S6K1 is capable of inhibiting Transforming Growth Factor-β-Activated Kinase 1 (TAK1)

resulting in increased signalling through Nuclear Factor-κB (Kim et al., 2014). Additionally,

Shin et al. (2013) found that TAK1 was also capable of inhibiting S6K1 activation by

preventing an interaction with the RAPTOR subunit of mTORC1. These findings support the

hypothesis presented here by demonstrating a link between cellular stress responses and a

major pathway regulating cellular growth. These works highlight that cell growth pathways

and cell stress pathways appear to be able to negatively regulate each other as one

becomes more necessary. Upon mTORC1 activation, there is less requirement for the cell

stress responses, and thus, there is less need for eIF4E phosphorylation to facilitate the

associated gene expression change. In this scenario the basal activity of MNK2, and the

inducible MNK1 (Ueda et al., 2004), would be kept lower. However, if the stress response is

required, the level of mTORC1 activity could be decreased and bring about MNK1 induction

as well as an increase in the basal level of MNK2 activity to allow such as a response.

Page 190

Figure 3.4.1a – Hypothesis Linking mTORC1 to the MNK-Dependent Stress Response. Given that

eIF4E phosphorylation, via MNK2 (Stead & Proud, 2013), has been shown to increase during

inhibition of mTORC1, it supports the hypothesis that inhibition of cell growth induces the cellular

stress response. In such a scenario, inhibition of mTORC1 leads to an increase in the basal activity

of MNK2. As a result, the phosphorylation of eIF4E and other substrates of the MNK kinases,

hnRNP A1 and the Sprouty proteins occur. eIF4E phosphorylation would promote the translation

of mRNAs associated with the stress response. hnRNP A1 would facilitate this by promoting the

formation of Stress Granules. The Sprouty proteins may then, through a feedback mechanism,

serve to return the cell to an unstressed state once the inhibition of cell growth was removed.

Links experimentally shown, either in the literature or here, are denoted by the unbroken green

arrows, also denoting hypothesised links. The red arrow illustrates the suppressive effect active

mTORC1 has upon MNK2 to lower the level of basal activity.

Page 191

Section 3.4.2 – Validation of the AKT1 Phosphorylation Prediction

Due to the presence of a variety of feedback interactions, AKT1 can be viewed as being very

tightly regulated by mTORC1. One feedback loop is mediated by the phosphorylation, and

subsequent degradation, of the IRS-1 by S6K1 (Tremblay et al., 2007; Xu et al., 2008; Zhang

et al., 2008). In this case, AKT1 is inhibited by this phosphorylation (Shah et al., 2004; Xu et

al., 2008). The importance of this mechanism of feedback inhibition can be seen in cases of

Simian Virus 40 infection in which the proteolytic degradation of IRS-1 is inhibited. This

results in adherent activation of AKT1 and the ERK1/2 pathway components (Hartmann et

al., 2014). In the second case, S6K1 phosphorylates the Rictor subunit of mTORC2 (Julien et

al., 2010) resulting in diminished output of mTORC2. The result is a decrease in AKT1

phosphorylation (Ibid). There is also some evidence that mTORC2 may also play a role in the

degradation of IRS-1, via phosphorylation, of the Serine residue at position 83, and thus,

increasing the stability of the ubiquitin ligase Fbw8 (Kim et al., 2012).

Rapamycin is known to affect AKT1 differently depending upon the concentration of the

mTORC1 inhibitor used (Sun et al., 2005; Chen et al., 2010; Saores et al., 2013). This leads to

the conclusion that differing concentrations of Rapamycin are capable of affecting these

feedback loops in different ways.

Given the qualitative nature of the model, inhibitors behave in an all-or-nothing fashion

with complete inhibition being found when they are present and complete activity of the

target when absent. As Rapamycin is capable of inhibiting mTORC2 (Sarbassov et al., 2006;

Lamming et al., 2012; Schreiber et al., 2015), Rapamycin is viewed in the model as inhibiting

this component so as to replicate the effects of exposure to prolonged or high

concentrations of Rapamycin.

Page 192

A by-product of testing the effects of Rapamycin treatment on the activity of ERK1/2 and

p38α in RAW 264.7 murine macrophage cells was the investigation of AKT1

phosphorylation. The Human Phospho-MAPK Array data revealed that AKT1 and AKT2

displayed significant inhibition in the presence of Rapamycin compared with the 1% DMSO

control. However, only AKT Pan remained significantly inhibited following the Holm-Sidak

Correction. To further validate this prediction, immunoblotting was carried out against

AKT1 phosphorylation. During this investigation, only very low levels of AKT1

phosphorylated on the Ser-473 site was found, as shown in Figure 3.3.4a. This very low

signal resulted in accurate quantification becoming very difficult and raises the possibility

that sensitivity of the differing antibodies (between the Human Phospho-MAPK Array and

Western Blotting) may have been a contributing factor here. Indeed the low signal may

explain the relatively large Standard Error of the Mean values, as shown by the large error

bars in Figure 3.3.4a. Additionally, there is some evidence that the activation of AKT1 is

governed, at least partly, by the cell cycle (Liu et al., 2014). Given that the RAW 264.7 cells

were growing asynchronously, it is possible that the reduction in AKT1 phosphorylation

predicted by the model does occur despite not being observed in the Western Blotting work

shown in Figure 3.3.4a. Moreover, given the large change in AKT Pan phosphorylation,

shown in Figure 3.3.2a it is not unreasonable to conclude that following a one hour

treatment with 10μM Rapamycin, Ser-473 phosphorylation is discernibly reduced in a

manner that is consistent with Rapamycin may be capable of inhibiting the Rictor subunit of

mTORC2. Based solely upon the results of the immunoblotting experiments presented in

Section 3.3.4 – Confirmation of the Effect Rapamycin has Upon AKT1 Phosphorylation, it is

however, not possible to say if this trend continues or if the situation is reversed.

Page 193

From a modelling perspective both situations (of Rapamycin bringing about either an

increase or decrease in AKT1 phosphorylation) are easily explainable. If AKT1

phosphorylation is to increase, the removal of Rapamycin-mediated mTORC2 inhibition is

required. Similarly, if a decrease is observed, the inhibition can remain. Although not

shown, both cases have been observed in silico during model simulations.


The work outlined in this chapter represented one of the most demanding points of the

project in terms of the length of time taken.

Initially, the intensity of bands obtained from Western Blotting was very low. Several

reasons were suggested that may account for this: The method of cell lysis being such that

samples were degrading during the harvesting process or the cells were not sufficiently

stimulated to see background effects. Thirdly, the method of development may not be

sensitive enough to detect the relatively low signal of phosphorylated eIF4E. LPS treatment

of the RAW 264.7 cells was attempted which resulted in significant activation of multiple

signalling pathways, as measured by both Western Blotting and Human Phospho-MAPK

Arrays. However, the stimulation of RAW 264.7 cells with LPS was found to be obscuring

effects that Rapamycin was having upon the signalling pathways. The results presented in

Section 3.3.2 – Evaluation of the Human Phospho-MAPK Arrays show the effects that

Rapamycin is capable of exerting upon the cell. However during initial testing with the array

kits, far fewer regions of difference were found between the 1% DMSO and 10μM

Rapamycin samples when both were exposed to 10ng per ml LPS (from Escherichia coli

Serotype O55:B5, TLRgrade™ Enzo (Life Sciences, Inc., Farmingdale, NY (USA))).

Page 194

It would be interesting to examine the effects of Toll-Like Receptor (TLR) ligands other than

LPS. Other TLR ligands, such as Polyinosinic-Polycytidylic Acid (Poly(I:C)), are relevant to the

cellular response to viral infection. Poly(I:C) is a synthetic double-stranded RNA analogue

(Field et al., 1967; Levy et al., 1969). Consequently, the use of this TLR3 ligand would have

produced an experimental system in which signalling pathways were activated in a manner

that is similar, but not identical, to a virus-infected cell (Li et al., 2005b). Although chemical

stimulation of signalling pathways was attempted to overcome the problem of low signal

intensity, it is worth noting that at this stage of the project, it was necessary to produce a

model that accurately reproduced the cellular regulation of translation and not a system in

which viral translation is modelled.

This problem also extended to the evaluation of AKT1 phosphorylation with

immunoblotting. The very low signal generated is likely due to very low levels of AKT1

phosphorylation in cells without stimulation. This issue could have been intensified by

potentially considerable differences in antibody sensitivities. To address these two issues,

stimulating RAW 264.7 cells with either a growth factor or TLR ligand could be of benefit if

the experiments were going to be repeated. Furthermore, the role of cell cycle progression

in AKT1 phosphorylation should be investigated, possibly through the synchronisation of cell

growth.

The second potential problem was addressed by changing the lysis buffer used in harvesting

the cells. Several buffers were trialled and all produced a marked increase in the signal

found after exposure to chemiluminescent substrate. These methods included lysing the

RAW 264.7 cells directly into sample buffer, as described in Simmonds et al. (2009). It was,

however, decided that using the Lysis Buffer 6 supplied with the Human Phospho-MAPK

Page 195

Array kits would give some degree of uniformity to these experiments and thus ensure that

the cells were handled in the same way across experiments.

The method of developing the initial blots was limited to the FluorChem®Q apparatus (Alpha

Innotech, ProteinSimple Inc., San Jose, CA (USA)). Whilst this system was capable of

detecting bands of high intensity, such as loading controls, it was not capable of detecting

low intensity bands. Even replacing the SuperSignal West Pico Chemiluminescent Substrate

with SuperSignal West Femto Chemiluminescent Substrate (Pierce®, Thermo Fisher

Scientific Inc., Rockford, IL (USA)) did not alleviate this problem and resulted in even higher

levels of background signal.

It was possible to overcome these problems with a combination of a traditional radiographic

film based development process and switching to a new lysis buffer. This suggests that a

combination of signal degradation and insufficient sensitivity of the development process

were to blame for these initial setbacks. This being said, one point is still unresolved with

regards to the method used to quantify radiographic film using pixel counting (as used in

Section 3.3.2 – Evaluation of the Human Phospho-MAPK Arrays, Section 3.3.3 –

Alterations to eIF4E Phosphorylation Levels and Section 3.3.4 – Confirmation of the Effect

Rapamycin has Upon AKT1 Phosphorylation): the fact that exposure of radiographic film to

the fluorescent signal of chemiluminescence does not produce a linear response (Gassmann

et al., 2009). Consequently, it may not be the most accurate method of quantification,

although this method was recommended for use with the Human Phospho-MAPK Arrays by

the manufacturer.

One question is still unresolved with regards to Rapamycin leading to an increase in eIF4E

phosphorylation. This problem lies with the identification of the MAPK-Interacting Kinase

Page 196

responsible for this increase. Attempts have been made by other members of the research

group, as part of the Royall et al. (2015) study, looking at using small interfering RNA (siRNA)

to knock-down MNK1 and MNK2 in RAW 264.7 cells. However, this work was not taken

forward due to a lack of suitable internal controls and in some cases, knock-down of MNK1

with siRNA targeted against MNK2. Similarly, there is a lack of specific MNK1 or MNK2

antibody, thus preventing direct visualisation of specific MNK phosphorylation during

Rapamycin treatment. The potential benefit of an additional approach, Clustered, Regularly

Interspaced, Short Palindromic Repeats (CRISPR)/CRISPR-Associated Protein (Cas) (Cong et

al., 2013; Mali et al., 2013) is discussed in Section 3.4.4 – Potential Future Work.

Section 3.4.4 – Potential Future Work

eIF4E phosphorylation is now being viewed as playing a part in mediating changes in gene

expression. mRNA transcripts involved in cellular stress appear to require the

phosphorylation of the 5’ cap-binding protein in order for efficient translation initiation

(Herdy et al., 2012). Given this and that mTORC1 plays a vital role in regulating cell growth,

it would be of interest to determine if eIF4E phosphorylation occurring during mTORC1

inhibition is accompanied by global changes to the transcriptome of the cell. Moreover,

establishing if other MNK substrates, such as hnRNP A1, exhibit a Rapamycin-induced

increase in phosphorylation may help to demonstrate if mTORC1 inhibition leads to a global

stress response.

As AKT1 phosphorylation has been shown to be negatively regulated in a manner implying

mTORC2 inhibition following a one hour treatment with 10μM Rapamycin, establishing the

sensitivity of mTORC2 to Rapamycin in RAW 264.7 cells would be of importance. Schreiber

and colleagues (2015) have shown that a high ratio of relative FKBP12 to FKBP51 expression

Page 197

is an important determinant in the degree of mTORC2 inhibition in response to Rapamycin

treatment. Determining this ratio in RAW 264.7 cells could, potentially, be of interest.

Determining if MNK1 or MNK2 is responsible for mediating the Rapamycin-induced increase

in eIF4E phosphorylation, although technically demanding, would be important. If this area

were to be pursued, the use of the CRISPR/Cas technique could be considered over the

development of a transgenic mouse. Given the ease and efficiency of CRISPR/Cas, the

technique could be of benefit and has proven useful in knocking out genes on a large scale

in mammalian cells (Shalem et al., 2013; Walsh & Hochedlinger, 2013; Auer et al., 2014;

Wang et al., 2014). However, this approach does suffer from the limitation that it requires a

Protospacer Adjacent Motif (PAM), to which the Cas protein can bind (Ibid), surrounding,

what in this case would be the MNK1 or MNK2 genes. A lack of this PAM site, or insufficient

sequence homology between the Cas protein and this sequence, would result in reduced

efficiency (Fu et al., 2013; Hsu et al., 2013 (As cited by Walsh & Hochedlinger, 2013)).

Although this approach has been successfully used in vivo to create knockout mice (Wang et

al., 2013; Fujii et al., 2013), to date no such studies have used this technique on RAW 264.7

cells. Walsh and Hochedlinger (2013) noted that additional research, such as that by Hou et

al. (2013), in identifying novel Cas proteins may mean that this approach will become more

applicable here.

The use of transgenic mice is costly in the terms of financial implications, time and the high

failure (Hall et al., 2009). There are also ethical considerations. The number of animal

experiments required in any project should be minimised. Furthermore, experimental

validation of predictions in this project is aimed at improving a model which was to be used

Page 198

to mimic viral infection, as such using primary cells taken from a mouse would, most

probably, behave in a different manner to the transformed RAW 264.7 cells.


Despite experimental challenges, the effect of Rapamycin on eIF4E phosphorylation,

predicted by the model, has been shown. Although there were also problems regarding the

effect this macrolide has upon AKT1 phosphorylation, there is sufficient evidence to

conclude that a considerable reduction does occur following one hour of treatment. From

this perspective, given the qualitative nature of the model, this prediction has also been

validated.

By using more of the data from the array in the comparison to the binary model behaviour,

it has been shown that the model is in the majority of cases behaving in a manner

reminiscent of the model system, the murine macrophage RAW 264.7 cell. In doing so the

validity of the model has increased and, although the comparison was of a relatively small

number of data points, without further refinement or adjustment, the final model has

produced a positive MCC value.

The experimental validation of these two predictions was a major step forward in this

project. Although these two predictions are well documented in the literature, in

experimentally corroborating these hypothesised behaviours, it has been shown that the

model is proficient in predicting complex and non-trivial patterns of behaviour. In keeping

with this, at this stage the rationale for conducting these experiments was not to necessarily

provide solid mechanistic data regarding the link between the phosphorylation of AKT1 and

eIF4E and Rapamycin-mediated inhibition of mTORC1, but rather to provide clear evidence

that the behaviour of the model is predictable when certain, specific perturbations are

Page 199

applied. In doing so, it has been shown that the model was ready to be used in novel ways,

such as the investigation of the interactions between a virus and the host signalling network

regulating translation initiation.

Page 200

Chapter 4 – Modelling the Effects of

Viral Infection

Page 201


Viruses are pathogens which commandeer cellular machinery to prioritise the synthesis of

viral proteins. The validated model presented in this work gives a unique opportunity to

study these host-pathogen interactions. Within Chapter 2 – Model Creation & Refining and

Chapter 3 – Experimental Validation of Model Predictions, the model of translational

regulation was created and experimentally validated. The predictive power of the model

was raised significantly, to MCC = 0.4558. This improved predictive power and the

experimental validation of two predicted behaviours meant than the model was behaving in

a reliable way. In order to demonstrate the value of model, in a broad context, the

incorporation of viral effects into the model was carried out.

Controlling and combatting viral infection is a major challenge facing modern science.

Despite considerable advances in the field of virology, there is a limited spectrum of viable

treatments for viral infections. Moreover, some of these diseases are causes of

considerable morbidity and mortality and represent a major burden on the National Health

Service and other healthcare systems and public health services worldwide. With the use of

systems biology techniques to study viral infection, it is hoped that new cellular targets for

treating viral infection will emerge.

Whilst the main focus of this project has been to create a model of mammalian translation,

the possibility of using this model to mimic the effects viral infection has on the translational

machinery and upstream regulatory signalling pathways was investigated. Three viruses

have been selected on the basis that they perturb different cellular pathways and have

different effects on the translational machinery. JUNV, ANDV and Murine Norovirus (MNV)

are all considered important human pathogens or models of such pathogens. Although two

Page 202

of these viruses may not infect RAW 264.7 cells, they have been included here as they

produce different effects on the cellular signalling network and manipulate the host

translational machinery in ways that MNV does not.

Section 4.2 – Model Changes & Simulation Methods

For each virus it was necessary to run simulations with two models. The models were

identical in every respect except that one was used to represent an ‘infected’ system and

one as the ‘mock’, uninfected control. QSSPN simulations were carried out on the ‘infected’

or ‘mock’ models. The ‘mock’ model was initially tested to ensure that the simulation times

were adjusted to result in the frequency of each effector being equal to around 0.5.

Binomial Confidence Intervals, calculated in the software R, were used to identify regions of

significant difference between the ‘infected’ or ‘mock’ models. The final model presented

at the end of Chapter 2 – Model Creation & Refining and experimentally validated in

Chapter 3 – Experimental Validation of Model Predictions was used here. More model

effectors, than used in previous chapters, were monitored as it was hoped that new effects

or, in the case of MNV, testable predictions could be found.

Section 4.2.1 – Junin Virus Model

To mimic the large scale changes to the host translation factors, two features of Junin Virus

infection were considered. Firstly, JUNV contains the same conserved amino acids believed

to be found in the active site of the L endonuclease that are found in other Arenaviruses

(Morin et al., 2010). Therefore, it is reasonable to assume that a cap-snatching mechanism

is employed by JUNV. Secondly, like many of the other New World Arenaviruses, JUNV

replication occurs with the Nucleoprotein acting as the cap-binding protein in place of eIF4E

(Linero et al., 2013). The ability of the Arenaviridae to actively use eIF4E is controversial. As

Page 203

noted by Knopp and colleagues (2015), the Old World Arenaviruses, such as Lassa Fever

Virus, may differ from the New World Arenaviruses in the need for this cellular factor

(Volpon et al., 2008; Volpon et al., 2010; Kranzusch & Whelan, 2011). Furthermore,

Arenaviruses contain RNA that is not polyadenylated (Emonet et al., 2009). This would

suggest that PABP is redundant in JUNV infection.

To date JUNV infection is known to require the activation of two fundamental signalling

pathways. Although displaying an oscillatory nature, JUNV requires the activation of ERK1/2

(Rodríguez et al., 2014). Of most significance here is the fact that at 24 hours post infection

the level of ERK1/2 activation is significantly increased. This period of activation co-insides

with the finding of Linero and colleagues (2013) that viral infection and protein synthesis

were not affected by a knock-down of eIF4E. Rodríguez and co-workers (2014) posit that

the N protein mediates the increase in ERK1/2 activation. Given these events occur at the

same time point it is possible to model these events simultaneously. The second pathway

upon which JUNV infection depends is the PI3K pathway (Linero & Scolaro, 2009). The

significance of this pathway at 24 hours post infection is not clear as there is an absence of

data at this time point regarding the activation of AKT1 or another PI3K substrate. This

being said the presence of a PI3K inhibitor after 24 hours post infection did markedly reduce

viral replication (Ibid). No clear data regarding other signalling pathways is known in respect

of this virus. Boswick et al. (2007) have shown that the related Pichinde Virus induces p38

activation at earlier time points but no work has been carried out with JUNV.

Given that the model is entirely qualitative and provides no temporal information, both the

ERK1/2 and PI3K pathways were activated in the ‘infected’ model. The use of both cap

snatching and substitution of eIF4E were also applied in the model.

Page 204

The effects that JUNV is capable of inducing, in the cellular signalling pathways and

translational machinery, were incorporated into the final version of the Petri Net model.

Building on this model, 17 new molecular species were included in the model, to account for

different translational intermediates and viral proteins that effect signalling pathway

components. These new species are involved in 27 new molecular interactions, which

include viral translation and the effects of viral components on cellular signalling pathways.

Section 4.2.2 – Murine Norovirus Model

The role of the various initiation factors in MNV infection is complex and not fully

understood. The VPg protein of MNV is known to function in place of the 5’ cap structure

(Chaudhry et al.¸ 2006). However, the interactions between this protein and cellular

initiation factors are still not fully understood. Whilst an interaction between VPg and eIF4E

has been reported (Ibid) the significance is not clear. Chaudhry and co-workers (2006)

found that a reduction of cellular eIF4E or the overexpression of 4E-BP1 did not preclude

viral protein synthesis. It has since been proposed that the interaction between the viral

protein and eIF4E only functions to help circumvent a change in gene expression to bring

about an immune response (Chung et al., 2014). In terms of an integral interaction, Chung

and co-workers (2014) demonstrated that VPg must interact with eIF4G. Indeed this finding

is corroborated by the view that eIF4E depletion does not affect viral translation (Chaudhry

et al., 2006).

Whilst eIF4E is not required for the translation of MNV proteins, there is a dependence on

the phosphorylation of this initiation factor (Royall et al., 2015). This work established that

the increase in eIF4E phosphorylation serves to bring about a change in gene expression

including an increase in the expression of an inhibitor of Nuclear Factor κB (Ibid). Whilst

Page 205

both ERK1/2 and p38 phosphorylation were shown to be increased during MNV infection,

only the axis of p38α-MNK1 was required for the increase in eIF4E phosphorylation (Royall

et al., 2015). Additionally, unpublished data (generated by N. Doyle and E. Royall (University

of Surrey)) indicates that MNV-1 infection produces a marked increase in the

phosphorylation of AKT1 on the residue Ser-473. This Human Phospho-MAPK array data

also demonstrated that the signalling axes of mTORC1-S6K1 and JNK1 were unaffected by

MNV-1 infection.

The MNV model, building on the final model, included 13 new molecular species involved in

20 novel molecular interactions. These interactions were involved in viral translation and

the interactions between viral factors, and the cellular translational machinery and

signalling factors. These interactions were documented in existing literature or taken from

experimental data generated in-house.

Section 4.2.3 – Andes Virus Model

Whilst not explicitly shown to be the case for ANDV, the Hantaviruses are widely believed to

use cellular 5’ methylguanosine caps to initiate viral translation (Mir et al., 2010; Cheng &

Mir, 2012). However, the work of Heinemann and colleagues (2013) has suggested that

selective host mRNA degradation occurs in ANDV infection and is mediated by the L protein.

When taken together these works suggest that the RdRp L protein may function to

selectively degrade cellular mRNA in ANDV infection whilst the N protein may then

sequester the cleaved cap fragment to prevent further cell-mediated degradation (Cheng &

Mir, 2012; Heinemann et al., 2013). Furthermore, the N-terminal domain of the L protein,

which is thought to mediate the cleavage of mRNA, is well conserved amongst the

Hantaviruses (Reguera et al., 2010).

Page 206

There is additional evidence that suggests translation of at least one ANDV protein, the NSs

protein is encoded by a Small segment of viral RNA (Vera-Otarola et al., 2010) that does not

require the use of PABP. Rather, the untranslated region within the 3’ end of the Small

transcript is used (Vera-Otarola et al., 2012). These authors suggest that an unknown

cellular protein may mediate Small RNA translation (Ibid). It would be of interest to

compare the translation of histone mRNA to the translation of the Small genomic segment.

The 3’ UTR of histone mRNAs contains a stem-loop structure to which SLBP must interact for

translation to occur (Sànchez & Marzluff, 2002). As noted by Vera-Otarola and co-workers

(2010) no base pairing between the 5’ and 3’ structures could occur. It raises the possibility

that viral translation could occur through a mechanism that involves circularisation of viral

mRNA by SLBP interacting with the 40S subunit, possibly through eIF3 (as is the case for

histone mRNA (Ling et al., 2002)) and potentially by the N protein replacing the eIF4F

complex (Mir & Panganiban, 2008).

In addition to the effects that ANDV imposes on the translational machinery, infection with

this Hantavirus has the potential to exert large effects on the cellular signalling network.

ANDV is capable of inducing mTORC1 activation through the autophagic degradation of the

viral protein Gn (formed during cleavage of the GPC protein (Löber et al., 2001), which

serves to maintain the pool of amino acids within the lysosomes (McNulty et al., 2013).

Whilst mTORC1 activation has also been demonstrated by Gavrilovskaya et al. (2013), there

is some disagreement over the level of S6K1 activity. Gavrilovskaya and co-workers (2013)

showed that ANDV infection, in conjunction with hypoxia, resulted in increased S6K1

phosphorylation whereas no such increase in either this or 4E-BP1 phosphorylation was

noted by McNulty et al. (2013). An obvious difference in these works is the induction of

Page 207

hypoxia, however no S6K1 phosphorylation was found in hypoxic cells alone (Gavrilovskaya

et al., 2013). These works do however suggest that the result of mTORC1 activation by

ANDV is to bring about induction of a variety of mTORC1-dependent pathways such as

Hypoxia Inducible Factor 1α (HIF-1α).

Additionally, no change in the activation of the inflammatory regulator Nuclear Factor-κB

was noted in cells infected with ANDV (Khaiboullina et al., 2013). ANDV infection was

capable of increasing Nuclear Factor-κB activation in cells pre-treated with Tumour Necrosis

Factor α (Ibid).

Building on the final version of the Petri Net model of mammalian translational regulation,

the effects of ANDV infection were incorporated. To this end, 17 new molecular species

were added to the model to account for various viral factors and translational intermediate

stages, where viral factors or viral RNA were incorporated. To model viral protein synthesis

and the effects ANDV has upon the cellular signalling network, 22 additional molecular

interactions were added to the model.

Section 4.2.4 – Experimental Validation of the Murine Norovirus Model

Upon examination of the behaviour of the MNV model, the level of phosphorylated Cyclic

AMP-Responsive Element-Binding Protein (CREB) was found to be increased during MNV

infection. Using Human Phospho-MAPK Arrays, it was possible to test this behaviour.

The experimental work presented in this chapter was conducted by N. Doyle and E. Royall

using the method detailed in Royall et al. (2015). RAW 264.7 cells were cultured as per the

method given in Royall et al. (2015). These cells were then infected with MNV strain CW1 at

a Multiplicity of Infection (MOI) of ten 50% Tissue Culture Infectious Dose units and were

Page 208

incubated at 37°C with 5% Carbon Dioxide for either two or 12 hours. Uninfected control

samples for each time point were also taken. Human Phospho-MAPK Arrays were carried

out using 300μg of protein, as determined by a BCA Assay, from each lysate. These Human

Phospho-MAPK Arrays were carried out according the manufacturers’ instruction and were

developed on to radiographic film. Quantification of each phosphoprotein, also carried out

by N. Doyle, was conducted in ImageJ. As detailed in Royall et al. (2015), the values for

CREB phosphorylation were expressed as relative pixel density. Prior to this, these values

were expressed as relative values of the array control spots CREB. The statistical analysis of

the data was carried out using two-way ANOVA with Tukey’s post-hoc test in GraphPad

Prism 6.


Section 4.3.1 – Simulation of the Junin Virus Models

The effects of JUNV infection were incorporated into the final version of the binary model.

QSSPN simulations and Reachfq analyses were performed and the results, shown in Figure

4.3.1a, reveal multiple differences between the ‘infected’ and ‘mock’ models. As expected,

both ERK1/2 and PI3K activity was increased in the ‘infected’ model. Similarly, the

production of viral protein was limited to the ‘infected’ model. A significant reduction in

host protein synthesis was also observed (as seen in the reduced frequency of trajectories

resulting in protein synthesis). The activity of many substrates for either ERK1/2, such as

MAPK Phosphatase 7 (MKP-7) or RSK1, or PI3K, such as AKT1 or PDK1, also exhibited

increases.

The most unexpected results concerned the nature of feedback mechanisms between the

ERK1/2 and JNK1 signalling pathways. Viral infection appeared to negatively affect eIF4E

Page 209

phosphorylation through both MNK1 and MNK2 activity. mTORC1 signalling also appears to

be, for the most part, increased. Interestingly, transcription factors, such as HIF-1α and Elk-

1, were also up-regulated and suggest that JUNV infection would promote large scale gene

expression changes. The HIF-1α expression also suggests that the metabolic network, as

discussed in Section 5.1.1 – Linking mTORC1 to the Regulation of Metabolism, of the

infected cell would exhibit considerable metabolic changes. It would therefore, be of

considerable future interest to evaluate the nature and degree of impact on the metabolic

network by simulations of the integrated signalling network model and GSMN. Chapter 5 –

Integration of the Translational Control Model with a Genome Scale Metabolic Network of

a RAW 264.7 Cell describes the preliminary results obtained in this direction.

Page 210

Mock Upper Lower Infected Upper Lower

ppMEK1 0.521 0.572 0.469 0.496 0.548 0.444

ppMEK2 0.507 0.559 0.455 0.455 0.507 0.404

ppERK1 0.521 0.572 0.469 0.992 0.997 0.976

ppERK2 0.534 0.585 0.492 0.989 0.996 0.972

pRSK1 0.430 0.481 0.380 0.726 0.770 0.677

rpS6_Pi 0.550 0.601 0.498 0.718 0.762 0.669

pMNK1 0.540 0.591 0.488 0.147 0.188 0.114

pMNK2 0.448 0.500 0.397 0.149 0.190 0.115

eIF4E_Pi 0.480 0.532 0.429 0.134 0.173 0.102

pJNK1 0.460 0.512 0.409 0.138 0.178 0.106

cJun_Pi 0.552 0.603 0.500 0.119 0.157 0.089

ppp38a 0.541 0.592 0.489 0.552 0.603 0.500

pERK5 0.448 0.500 0.397 0.259 0.307 0.216

PI3K_Local 0.450 0.502 0.399 0.944 0.964 0.915

PIP3 0.508 0.560 0.456 0.912 0.937 0.878

PDK1_Local 0.553 0.604 0.501 0.910 0.935 0.876

pPKCz 0.490 0.542 0.438 0.768 0.809 0.721

ppAKT1 0.562 0.613 0.510 0.849 0.883 0.808

ppS6K1 0.150 0.191 0.117 0.008 0.024 0.003

p4EBP1 0.523 0.574 0.471 0.762 0.803 0.715

mTORC1_Local_Pi 0.482 0.534 0.430 0.731 0.774 0.683

pTSC12 0.489 0.541 0.437 0.895 0.923 0.859

eIF4B_Pi 0.511 0.563 0.459 0.668 0.715 0.617

Protein_Synthesis 0.515 0.567 0.463 0.324 0.374 0.277

pCREB 0.531 0.582 0.479 0.758 0.800 0.711

pMSK2 0.586 0.636 0.534 0.584 0.634 0.532

pMKK3 0.542 0.593 0.490 0.512 0.564 0.460

pMKK6 0.534 0.585 0.482 0.524 0.575 0.472

pGSK3 0.575 0.625 0.523 0.625 0.674 0.574

pMSK1 0.538 0.589 0.486 0.886 0.914 0.848

pMK2 0.533 0.584 0.481 0.510 0.562 0.458

pTTP 0.551 0.602 0.499 0.599 0.649 0.547

HIF1a 0.535 0.586 0.483 0.716 0.760 0.667

pSTAT3 0.481 0.533 0.429 0.697 0.742 0.647

pElk1 0.529 0.580 0.477 0.895 0.923 0.859

pATF2 0.534 0.585 0.482 0.496 0.548 0.444

pTpl2 0.533 0.584 0.481 0.543 0.594 0.491

ppPKCa 0.537 0.588 0.485 0.540 0.591 0.488

pMKP7 0.486 0.538 0.434 0.997 0.999 0.984

eIF4G_Pi 0.574 0.624 0.522 0.437 0.489 0.386

Viral_Protein 0.000 0.011 0.000 0.335 0.386 0.288

Figure 4.3.1a – Modelling the Effects of JUNV Infection. Comparison of the impact JUNV infection

has upon the cellular signalling network and the translation initiation pathway. Significant

differences between the models are found by comparing the 99.99% Binomial Confidence

Intervals, as calculated in R, of the QSSPN simulation results for each effector. Significant

differences are shown in purple and indicate no overlap in the confidence intervals.

Page 211

Section 4.3.2 – Simulation of the Murine Norovirus Models

MNV was chosen as it is permissive in RAW 264.7 cells, the experimental system used in this

project. The effects of MNV were incorporated into the final version of the binary model

and simulations were run with QSSPN. Reachfq was used to calculate the frequency of

1,000 trajectories which resulted in the activation or production of each effector. 99.99%

Binomial Confidence Intervals were calculated for each effector, as shown in Figure 4.3.2a.

Significant differences were identified by no overlap being present in these 99.99%

confidence intervals. The model incorporating MNV infection is given in Supplementary

Information E4 (given as both a Snoopy model file (E4a) and an SBML file (.xml)) (E4b).

In order to accurately model MNV infection, the activity of several pathways was altered.

The activity of the ERK1/2 and p38 MAPK pathways was increased in keeping with

experimental findings. Similarly, the activation of AKT1 was also promoted. This Human

Phospho-MAPK Array data also confirmed that neither JNK1 nor mTORC1 activity was

increased. One of the main findings of this model was the raised eIF4E phosphorylation.

The production of viral protein was only found in the ‘infected’ model. The ‘infected’ model

also displayed no significant reduction in cellular protein synthesis. This model prediction is

somewhat in agreement, the work of Royall et al. (2015). When using polysome profiling,

this work noted that the translational profile of the cell changes in response to infection.

This being said, the level of a housekeeping gene, Glyceraldehyde 3-Phosphate

Dehydrogenase, was not affected by MNV infection (Ibid) and whilst not conclusive proof, it

is indicative of no global reduction in host cell protein synthesis.

These simulations, the results of which are outlined in Figure 4.3.2a, have predicted

numerous molecular targets that have not, to date, previously been identified as being

Page 212

affected by MNV infection. Several transcription factors, namely CREB, Elk-1 and ATF-2,

downstream of ERK1/2 and p38 were found to exhibit increased phosphorylation. An

increase in the activation of various signalling components was also noted. The majority of

these components (MNK1, MNK2, MSK1, MSK2, MK2 and MKP-7) are, again, downstream of

either ERK1/2 or p38. This model also suggests that AU-Rich Element-containing mRNA is

stabilised as a result of increased phosphorylation of TTP. Experimental work to validate the

increase in CREB phosphorylation is detailed in Section 4.3.2.1 – Experimental Validation of

the Hypotheses Resulting from Simulations of the Murine Norovirus Model.

The perturbations to the model also led to the level of phosphorylation of several initiation

factors to be altered. eIF4B phosphorylation is strikingly increased. This affect can be linked

to the increase in the ERK1/2-mediated phosphorylation of RSK1. Despite no change in

PKCα phosphorylation, eIF4G phosphorylation suffers a slight, but significant, reduction. As

already noted, eIF4E phosphorylation is increased. The work of Royall and co-workers

(2015) determined that MNK1, via p38, is responsible for this increase. Whilst this picture is

somewhat confirmed by the behaviour of the model, this work also notes an increase in

MNK2 phosphorylation.

Page 213

Mock Upper Lower Infected Upper Lower

ppMEK1 0.538 0.589 0.486 0.504 0.556 0.452

ppMEK2 0.510 0.562 0.458 0.521 0.572 0.469

ppERK1 0.534 0.585 0.482 0.994 0.998 0.979

ppERK2 0.540 0.591 0.488 0.992 0.997 0.976

pRSK1 0.462 0.514 0.411 0.832 0.867 0.790

rpS6_Pi 0.471 0.523 0.420 0.744 0.787 0.696

pMNK1 0.000 0.011 0.000 0.931 0.953 0.900

pMNK2 0.515 0.567 0.463 0.775 0.815 0.729

eIF4E_Pi 0.511 0.563 0.459 0.827 0.863 0.784

pJNK1 0.468 0.520 0.417 0.452 0.504 0.401

cJun_Pi 0.462 0.514 0.411 0.450 0.502 0.399

ppp38a 0.444 0.496 0.393 0.882 0.912 0.844

pERK5 0.458 0.510 0.407 0.284 0.333 0.240

PI3K_Local 0.448 0.500 0.397 0.456 0.508 0.405

PIP3 0.441 0.493 0.390 0.484 0.536 0.432

PDK1_Local 0.551 0.602 0.499 0.569 0.620 0.517

pPKCz 0.460 0.512 0.409 0.462 0.514 0.411

ppAKT1 0.461 0.513 0.410 0.616 0.665 0.564

ppS6K1 0.127 0.166 0.096 0.078 0.111 0.054

p4EBP1 0.479 0.531 0.428 0.435 0.487 0.384

mTORC1_Local_Pi 0.479 0.531 0.428 0.447 0.499 0.396

pTSC12 0.507 0.559 0.455 0.730 0.774 0.681

eIF4B_Pi 0.572 0.622 0.520 0.814 0.851 0.770


pCREB 0.507 0.559 0.455 0.915 0.940 0.881

pMSK2 0.480 0.532 0.429 0.803 0.841 0.758

pMKK3 0.461 0.513 0.410 0.437 0.489 0.386

pMKK6 0.442 0.494 0.391 0.435 0.487 0.384

pGSK3 0.444 0.496 0.393 0.462 0.514 0.411

pMSK1 0.464 0.516 0.413 0.946 0.965 0.917

pMK2 0.577 0.627 0.525 0.859 0.891 0.819

pTTP 0.542 0.593 0.490 0.705 0.750 0.656

HIF1a 0.497 0.549 0.445 0.475 0.527 0.424

pSTAT3 0.477 0.529 0.426 0.447 0.499 0.396

pElk1 0.466 0.518 0.415 0.952 0.970 0.924

pATF2 0.472 0.524 0.421 0.786 0.825 0.740

pTpl2 0.526 0.577 0.474 0.510 0.562 0.458

ppPKCa 0.557 0.627 0.525 0.610 0.659 0.558

pMKP7 0.483 0.535 0.431 0.998 1.000 0.986

eIF4G_Pi 0.498 0.550 0.446 0.357 0.408 0.309

Viral_Protein 0.000 0.011 0.000 0.377 0.428 0.328

Figure 4.3.2a – Modelling the Effects of MNV Infection. Outline of the effects MNV infection

has upon the cellular signalling network and translation initiation pathway. QSSPN simulations,

followed by Reachfq, generated information on the activation or production of various model

effectors. Comparison of the Binomial Confidence Intervals revealed regions of significant

difference, as shown in red.

Page 214

Section 4.3.2.1 – Experimental Validation of the Hypotheses Resulting from

Simulations of the Murine Norovirus Model

The MNV model demonstrates a clear, significant increase in CREB phosphorylation in

response to MNV infection. Additionally, no significant differences were noted in the

phosphorylation of Glycogen Synthase Kinase 3 (GSK3), MKK3 and MKK6. The data collected

by N. Doyle and E. Royall using Human Phospho-MAPK Arrays enabled validation of this

prediction. To this end, RAW 264.7 cells were infected with MNV strain CW1 at an MOI of

10. Uninfected cells were used as a control. Both infected and uninfected cells were

incubated for two and 12 hours prior to lysis. Cells were lysed using Lysis Buffer 6 according

to the manufacturers’ instructions.

As shown in Figure 4.3.2.1a, although no difference in the phosphorylation state of CREB

was seen at 2 hours post infection (hpi), there is a significant difference between the

uninfected, mock and infected samples at 12hpi (p = 0.0019). At 12hpi, MNV infection

induces the phosphorylation of this transcription factor. Moreover, a comparison between

the infected samples at 2hpi and 12hpi reveals a significant increase in CREB

phosphorylation (p < 0.0001). No significant difference between the mock, control time

points was seen.

In the cases of GSK-3, MKK3 and MKK6, the experimental results demonstrated no

significant differences in the level of phosphorylated forms of any of these kinases at either

2hpi or 12hpi (data not shown). These findings indicate the model is capable of predicting a

diverse array of signalling behaviours.

Page 215

Only one point of difference was noted between the behaviour of the model and the

experimental data. MSK2 phosphorylation was predicted to increase however, the

experimental data failed to detect a significant increase. It is worth noting that a non-

significant increase in MSK2 phosphorylation (p = 0.1792) was noted in the Human Phospho-

MAPK Array. This increase may become significant if the time course were extended.

Page 216

p = 0.0019

p < 0.0001

Figure 4.3.2.1a – Effect MNV Infection has Upon CREB Phosphorylation.

RAW 264.7 cells were infected for either two or 12 hours. These cells

(MNV), along with uninfected samples (mock) incubated for the same

length of time, were lysed. The protein concentration was determined

using the BCA Assay. Human Phospho-MAPK Arrays using 300μg of protein

from each lysate were carried out in accordance with the manufacturers’

instructions. Quantification was carried out in ImageJ. Statistical analysis,

using GraphPad Prism 6, involved two-way ANOVA with Tukey’s post-hoc

test of significance. The data presented has been normalized to the array

control spots, as detailed in Royall et al. (2015), and therefore expressed as

Relative Pixel Density. Error bars represent the Standard Error of the Mean.

The data presented above was supplied by N. Locker and obtained by N.

Doyle and E. Royall.

Page 217

Section 4.3.3 – Simulation of the Andes Virus Models

The final virus investigated in this project was ANDV. In terms of the effects this virus has

upon the signalling network, only very few documented affects were previously noted and

consequently this virus has the smallest number of perturbations implemented into the final

version of the binary model.

Reachfq analyses were performed on the QSSPN simulation results. A comparison of the

Binomial Confidence Intervals calculated for each of the results, shown in Figure 4.3.3a, was

used to find regions of significant difference (as shown in dark purple). The only pathway

altered in this model was the mTORC1 pathway. As can be seen in Figure 4.3.3a, this

pathway was significantly increased. Surprisingly, although viral protein was produced and

this required a cap-snatching mechanism, no significant reduction in host protein synthesis

was observed. It is worth noting that the work of Jääskeläinen and co-workers (2007)

demonstrates that other Hantavirues, namely Tula Virus and Puumala Virus, do not induce a

strong inhibition of cellular protein synthesis. This work is somewhat supportive of the

finding here. The polymerase encoded by the Large genome segment has been shown to

not decrease the expression of cellular proteins in general, but it has been shown to

specifically limit the production of thoses proteins only of high molecular weight

(Heinemann et al., 2013). It could be argued, therefore, that the non-significant reduction

of cellular protein synthesis, shown in Figure 4.3.3a, is in agreement with the view that no

general suppression of protein synthesis is seen during ANDV infection.

Page 218

Mock Upper Lower Infected Upper Lower ppMEK1 0.521 0.572 0.469 0.550 0.601 0.498

ppMEK2 0.507 0.559 0.455 0.499 0.551 0.447

ppERK1 0.521 0.572 0.469 0.503 0.555 0.451

ppERK2 0.534 0.585 0.482 0.527 0.578 0.475

pRSK1 0.430 0.482 0.380 0.444 0.496 0.393

rpS6_Pi 0.466 0.518 0.415 0.491 0.543 0.439

pMNK1 0.481 0.533 0.429 0.496 0.548 0.444

pMNK2 0.478 0.530 0.427 0.467 0.519 0.416

eIF4E_Pi 0.455 0.507 0.404 0.408 0.460 0.358

pJNK1 0.516 0.568 0.464 0.555 0.606 0.503

cJun_Pi 0.511 0.563 0.459 0.491 0.543 0.439

ppp38a 0.530 0.581 0.478 0.514 0.566 0.462

pERK5 0.448 0.500 0.397 0.454 0.506 0.403

PI3K_Local 0.514 0.566 0.462 0.536 0.587 0.484

PIP3 0.508 0.560 0.456 0.516 0.568 0.464

PDK1_Local 0.464 0.516 0.413 0.460 0.512 0.409

pPKCz 0.490 0.542 0.438 0.498 0.550 0.446

ppAKT1 0.562 0.613 0.510 0.577 0.627 0.525

ppS6K1 0.156 0.197 0.122 0.242 0.289 0.200

p4EBP1 0.476 0.528 0.425 0.916 0.941 0.882

mTORC1_Local_Pi 0.568 0.619 0.516 0.996 0.999 0.982

pTSC12 0.431 0.483 0.380 0.419 0.471 0.369

eIF4B_Pi 0.511 0.563 0.459 0.513 0.565 0.461


pCREB 0.531 0.582 0.479 0.515 0.567 0.463

pMSK2 0.586 0.636 0.534 0.594 0.644 0.542

pMKK3 0.542 0.593 0.490 0.538 0.589 0.486

pMKK6 0.534 0.585 0.482 0.533 0.584 0.481

pGSK3 0.575 0.625 0.523 0.606 0.655 0.554

pMSK1 0.538 0.589 0.486 0.562 0.613 0.510

pMK2 0.581 0.631 0.529 0.584 0.634 0.532

pTTP 0.474 0.526 0.423 0.508 0.560 0.456

HIF1a 0.535 0.586 0.483 0.974 0.986 0.952

pSTAT3 0.496 0.548 0.444 0.986 0.994 0.968

pElk1 0.529 0.580 0.477 0.559 0.610 0.507

pATF2 0.500 0.552 0.448 0.505 0.557 0.453

pTpl2 0.533 0.584 0.481 0.528 0.579 0.476

ppPKCa 0.537 0.588 0.485 0.527 0.578 0.475

pMKP7 0.486 0.538 0.434 0.463 0.515 0.412

eIF4G_Pi 0.421 0.473 0.371 0.287 0.336 0.242

Viral_Protein 0.000 0.011 0.000 0.150 0.191 0.117

Figure 4.3.3a - Modelling the Effects of ANDV Infection. These simulation results show the

effects ANDV infection has upon the cellular signalling network and the translation initiation

pathway. Binomial Confidence Intervals of Reachfq analyses were calculated for each

model effector from the QSSPN simulations. Regions of significant difference, shown in

dark purple, are found where no overlap in confidence intervals is present.

Page 219


The elaboration of the binary model to include viral effects was necessary to further validate

the benefits of taking a systems approach to understanding complex biological systems.


Given that up until this point it has been possible to give a quantitative measure of the

predictive power of the model, the single biggest problem encountered in this part of the

work was the lack of a suitable benchmark dataset to establish the predictive power of the

models. Data concerning the effects of these viruses is very limited and the action of

inhibitors is not directly transferable from the datasets used to quantify the predictive

power of the cellular translational control model.

The only solution to this would be to test the effects each of the inhibitors has when

combined with infection. However, this is not feasible and so the solution implemented

here is to evaluate the effects qualitatively. For example, Rodríguez et al. (2014) noted that

ERK1/2 activity is promoted by JUNV infection and so this effect was incorporated into the

model. By comparing the simulation results for the ‘mock’ and ‘infected’ models it is

possible to qualitatively prove that the model is functioning in accordance with published

data. Furthermore, even if efforts were made to quantify these effects, such as by

comparing the literature and simulation results in a Confusion Matrix, the small number of

data points available for each model would mean that the resulting MCC value is of little

benefit.

As already noted in Section 3.4.4 – Problems Encountered & Troubleshooting, the use of

pixel counting to quantify the Human Phospho-MAPK Arrays in validating MNV predictions,

is not considered particularly accurate due to the non-linear response of radiographic film to

Page 220

a fluorescent signal (Gassmann et al., 2009). However, it was used in this work as it was the

method recommended by the array manufacturer.


Both the literature examined during the construction of the viral models and the simulation

results indicate that viral infection can potentially affect the global metabolic flux

distributions. This motivates future work which is aimed at utilising the full capabilities of

the QSSPN software to integrate the signalling network model with a GSMN. Chapter 5 –

Integration of the Translational Control Model with a Genome Scale Metabolic Network of

a RAW 264.7 Cell describes the first version of such an integrated model. However, given

the low predictive power of the integrated model, when evaluated against the Metabolic

Benchmark dataset, more work is needed before this model can be used to generate

hypotheses directing experimental work. However, examining the effects of viral infection

upon the cellular metabolic network would have added significant value to this section of

the project. Future use of this type of model and in this way should involve working

towards understanding the impact of viral infection upon the cellular metabolic network.

This would be achievable with this type of model if the predictive power were raised

considerably.

Furthermore, experimental works looking at predictions generated by the MNV model are

needed. Given that a number of these predictions involve looking at the post-translational

modification, particularly phosphorylation, of well-recognized signalling components (such

as MK2 and MSK1), it should be possible to use immunoblotting for many. A possibility also

exists to examine the effects of the increases to transcription factor activity, such as CREB

and ATF-2, through the use of the Polymerase Chain Reaction (PCR) technique targeting

Page 221

genes under the control of these transcription factors. Conducting these works would give a

broader understanding to the effects MNV infection has upon the cell.

Developing a method for quantifying the predictive power of the model should also be

considered if modelling of this type is to be used when looking at host-pathogen

interactions. Despite the difficulties noted in Section 4.4.1 – Problems Encountered &

Troubleshooting, a method based upon the one used throughout this project may be viable.

For this to be the case, increasing the number of data points used in the calculation of the

MCC value must be achieved. The increasing availability of transcriptomics and proteomics

datasets looking at viral infection, such as those generated by Aevermann et al. (2014), will

hopefully begin to provide access to large-scale datasets that would be necessary for a

quantitative measure of the predictive power of the model.


The incorporation of the various effects viral infection can induce, in both the cellular

signalling network and to the translational machinery, was carried out. Moreover all three

viruses represent understudied, important human pathogens, or in the case of MNV, a

surrogate of one. Whilst it is not being suggested that the viruses used here all infect RAW

264.7 cells, the three viruses chosen in this work were selected as they produce different

effects to the cellular signalling network and the translational apparatus. With this in mind,

the number of effectors monitored during simulations was substantially increased so as to

be able to detect more widespread effects.

The work presented here differs from other published works looking at viral infection from a

systems biology perspective. The model to which the viral effects were added was a large-

scale reconstruction of the signalling network that surrounds and regulates translation

Page 222

initiation in mammals. Other models, such as the work of Jensen and co-workers (2013),

were very limited in scope. Whilst the approach of Jensen et al. (2013) does lead to

experimentally validated findings, it does limit the future uses of the model. The main

benefit of using a literature-based reconstruction model is that the model can be adapted

with relative ease to simulate new viral or, more generally, pathological effects. Moreover,

the qualitative nature of the model means that kinetics of viral perturbations are not

necessary and do not require proper integration with the kinetic parameters of the host.

Furthermore, the scope of this model means that it is possible to monitor the potential

impact of infection upon a wider range of cellular effectors and consequently lead to the

generation of more testable behaviours.

The model of JUNV infection demonstrates that large-scale aberrations to the cellular

signalling network are possible even when only a few initial alterations are made. This

demonstrates the high degree of cross-talk between signalling pathways, and the effects

that this has upon the functionality of the model as a whole. Consequently, this highlights

just how viral infection can be thought of as a ‘systems-level perturbation’ (Garmaroudi et

al., 2010) to the host. The results of this set of simulations are suggestive of JUNV

producing effects to numerous signalling pathways adjacent to ERK1/2 and PI3K, such as

MSK1. Furthermore, these simulations suggest that the increase in ERK1/2 activity

promotes alterations to the metabolic network through mTORC1.

Of the three viruses selected, perhaps the most disappointing model was that of ANDV

infection. Both Gavrilovskaya et al. (2013) and McNulty et al. (2013) demonstrate mTORC1

activation and Cimica and co-workers (2014) have shown that Interferon signalling is

affected by ANDV infection. Despite these findings, little else is known as to the cellular

Page 223

effects of ANDV infection. It was hoped that this model would shed light on additional

pathways activated by ANDV. However, the predictions generated were both few in

number and for the most part, directly downstream of mTORC1. The reduction of eIF4G

phosphorylation was also noted. This finding is suggestive of a reduction of mTORC2

activity, possibly as a result of the increase in S6K1 activity (Julien et al., 2010). However,

with no reduction in either AKT1 or PKCα phosphorylation, which are both mTORC2

substrates in the model, it is not possible to state this with any certainty. This being said,

this model had the fewest number of well documented perturbations to input into the

model. However, the effectors with altered levels of activity are suggestive of cellular

metabolic reprogramming. The implication of this model yielding only few predictions is

redolent of both the need for more research into ANDV infection and to ensure that models

are as expanded as possible to make certain that important events can be properly placed

into context.

Given safety considerations, it is difficult to undertake work to validate hypotheses

generated about either JUNV or ANDV infection. However, work with MNV is possible and

may give valuable insight into the molecular biology of the important human pathogen,

HuNV. As already stated in Section 4.4.2 – Potential Future Work, the effectors altered

during MNV infection may lead to vast changes to the cell. Alterations to the level of

activity of transcription factors may produce large shifts in gene expression. Both CREB

(Zhang et al., 2005) and ATF-2 (Bailey & Europe-Finner, 2005) can regulate genes involved in

a diverse array of important cellular processes. To date, with the exception of Royall et al.

(2015), very little work has been embarked upon to examine the effects members of the

Norovirus genus can induce in the cellular signalling network.

Page 224

The work with a model of MNV infection has revealed that the virus may affect a large

number of important signalling pathways. Whilst the findings that ERK1/2 and p38 were

activated was expected (Royall et al., 2015), the activation of downstream effectors of these

pathways, notably Elk-1, MK2 and MKP-7, is a novel hypothesis. PKCα activity has been

linked positively, in the case of Respiratory Syncytial Virus entry (San-Juan-Vergara et al.,

2004), or negatively, in the case of Vesicular Stomatitis Virus replication (Zhu et al., 2011), to

the regulation of viral infection. However, to date, the involvement of PKCα activity has not

been investigated in MNV infection. This work suggests that PKCα activity is not altered by

MNV infection.

The ability to experimentally validate findings of the model was a significant milestone for

this section of the project. The predicted increase in CREB phosphorylation, as shown in

Section 4.3.2 – Simulations of the Murine Norovirus Models, in response to MNV infection

was experimentally verified. The unpublished work by N. Doyle and E. Royall with Human

Phospho-MAPK Arrays indicated that CREB phosphorylation was increased during MNV

infection at 12hpi (as shown in Section 4.3.2.1 – Experimental Validation of the Hypotheses

Resulting from Simulations of the Murine Norovirus Model). As detailed in Royall et al.

(2015), RAW 264.7 cells were infected with MNV at an MOI of 10. Following 2hpi or 12hpi,

the cells were harvested and the lysates used in Human Phospho-MAPK Arrays. Given that

CREB phosphorylation was not specifically targeted for perturbation when the viral effects

were added to the MNV model, this can be viewed as experimental validation of one of the

predicted behaviours. This is true despite the behaviour of CREB not being unexpected

given the position of this transcription factor downstream of both ERK1/2 (Impey et al.,

Page 225

1998; Xing et al., 1998; Ying et al., 2002) and p38 (Di Petro et al., 2007; Di Giacomo et al.,

2009).

Although no work has investigated the effect of the MNV-induced increase in CREB

phosphorylation, as noted in Section 4.4.2 – Potential Future Work, this could possibly be

achieved through the use of PCR aimed at investigating if CREB target genes exhibit

expression changes as a result of this. An increase in CREB phosphorylation on the Ser-133

site is not unique to MNV infection. Both Human Immunodeficiency Virus 1 (Gibellini et al.,

1998) and Human T-Cell Leukaemia Virus (HTLV-1) (Kim et al., 2007b) are known to induce

CREB phosphorylation on the Ser-133 residue. HTLV-1 appears to require this

phosphorylation of CREB, through the viral protein Tax, so as to enhance transcription of the

viral genome (Geiger et al., 2008) and maybe functions to regulate apoptosis (Trevisan et al.,

2004; Trevisan et al., 2006 (as cited by Geiger et al., 2008)). Given that MNV (Bok et al.,

2009), and FCV (Roberts et al., 2003; Sosnovtsev et al., 2003), have been shown to induce

apoptosis, any anti-apoptotic roles for phosphorylated CREB seem questionable but it

remains to be seen if CREB is required for viral transcription. Bok and co-workers (2009)

noted that viral RNA achieves a maximal level at 8hpi with the level being maintained at

12hpi. This is in keeping with the work presented here in which an increase in CREB

phosphorylation was noted between 2hpi and 12hpi.

The factor resulting in an increase in p38 activity during MNV infection is not known. The

results presented here and in the array data showing that neither MKK3 nor MKK6 exhibit a

concomitant increase in phosphorylation, suggest that a viral protein maybe responsible for

this increase. Many other viruses, including JUNV (Rodríguez et al., 2014) and members of

Page 226

the Flavivirus genus (Lin et al., 2006; Werme et al., 2008; Laurent-Rolle et al., 2010), use

viral factors to alter the activation of cellular signalling pathways.

The behaviour of the other model components, GSK3, MKK3 and MKK6, were also

experimentally validated. In all three cases, no significant increase in phosphorylation was

predicted. The Human Phospho-MAPK Array data confirmed this behaviour. Whilst the

model failed to predict that MSK2 phosphorylation would be unaffected by MNV infection,

the validation of four predicted behaviours so far indicates that the MNV model has an 80%

success rate and, whilst not directly comparable, is indicative of a positive MCC value.

The models presented in this chapter reveal that it is possible to mimic viral infection and

predict additional behaviours. As has been shown here, a systems approach is capable of

identifying cellular targets involved in viral infection which may not have been immediately

obvious targets of study. On the whole, all viruses examined here produce a wide variety of

cellular affects and in all cases generate findings that are not readily explainable, which is

indicative of complex, non-trivial behaviour. As such the work presented here can be seen

as providing a new method for modelling the complex network of host-pathogen

interactions that accompany infection. Whilst many of the predicted behaviours are not yet

experimentally validated, the background model of cellular translation has been validated

(as shown in Chapter 3 – Experimental Validation of Model Predictions) and demonstrated

as having a significant degree of predictive power (as shown in Chapter 2 – Model Creation

& Refining). To conclude, the simulations presented here generated plausible hypotheses

that can be used to direct future experimental research on the interactions of viruses with

the host translational machinery and signalling network.

Page 227

Chapter 5 – Integration of the

Translational Control Model with a

Genome Scale Metabolic Network of

a RAW 264.7 Cell

Page 228


This chapter covers the work aimed at linking the model of the regulation of translation to a

GSMN, Flux Balanced Analysis model of a RAW 264.7 cell. This model was developed by

Bordbar et al. (2012) and is based upon transcriptomic, proteomic and metabolomic data.

There is a growing body of literature that highlights the pronounced effects the host-

pathogen interactions have upon the metabolic network of the cell (Wang et al., 2008b;

Saric et al., 2010; Coelho et al., 2015; Milner et al., 2015), or of the pathogen (Raghunathan

et al., 2010). This motivates the integration of the translational control model with the

GSMN to study whether translation initiation is co-regulated with the biosynthesis of

metabolic precursors of protein synthesis.

In previous chapters, the level of predictive power of the model was raised to an MCC =

0.4558 and several model behaviours have been validated. From here, attention was

turned to carrying out preliminary work of investigating the dual regulation of translation

and metabolism. Additionally, as the method of simulating the model detailed in Chapter 2

– Model Creation & Refining is computationally demanding, several attempts in this

chapter, were made at developing a method for running less computationally demanding

simulations, which yield results of comparable predictive power to the existing method.

Section 5.1.1 – Linking mTORC1 via S6K1 to the Regulation of Metabolism

An emerging role for mTORC1 is being discovered in the regulation of metabolism. mTORC1

regulates anabolic metabolism by promoting the synthesis of new proteins, nucleotides and

lipids while at the same time inhibiting the cellular process of autophagy (Shimobayashi &

Hall, 2014). This regulation of anabolic metabolism occurs via either a regulation of

transcription or post-translational modification of transcription factors or enzymes central

Page 229

to a particular metabolic pathway. As shown in Figure 5.1.1a, the majority of these effects

have been shown to be dependent on the mTORC1 substrate, S6K1.

One of the core metabolic pathways regulated by mTORC1 is the glycolytic pathway. This

regulation has also been shown to be required for carrying out normal cellular functions and

immune regulation (Donnelly et al., 2014). The ability to regulate glycolysis is mediated by

the transcription factors, c-Myc and Hypoxia Inducible Factor-1α (HIF-1α) (Düvel et al.,

2010). Both of these transcription factors are capable of regulating multiple enzymes in the

glycolytic pathway (Ibid). Moreover, despite being capable of regulating the expression of

distinct enzymes, both c-Myc and HIF-1α can bind to similar nucleic acid sequences and,

despite sometimes acting in an antagonistic fashion (Mazure et al., 2002; Koshiji et al.,

2004), function to enhance the expression of many of the same glycolytic enzymes, such as

Phosphofructokinase, Hexokinase and Triose Phosphate Isomerase (Gordan et al., 2007).

Additionally, mTORC1 activity has been shown to not only up-regulate Glucose Transporter

1 expression but also promotes membrane localization of the glucose transporter

(Makinoshima et al., 2015).

S6K1 mediates HIF-1α and c-Myc transcription and translation through the phosphorylation

of Signal Transducer and Activator of Transcription 3 (STAT3) (Kim et al., 2009a; Thiem et al.,

2013; Dodd et al., 2014; Li et al., 2014; Rad et al., 2015) and eIF4B (Tandon et al., 2011; Csibi

et al., 2014; Dodd et al., 2014). Regulation of pyrimidine synthesis occurs through S6K1-

mediated phosphorylation of the rate-limiting enzyme Carbamoyl-Phosphate Synthetase 2,

Aspartate Transcarbamylase and Dihydroorotase (CAD) (Lindsey-Boltz et al., 2004; Ben-Sahra

et al., 2013; Robitaille et al., 2013). The final pathway canonically regulated by mTORC1 is

lipid and sterol biosynthesis. The transcription factors, Sterol-Regulatory Element Binding

Page 230

Proteins (SREBP) 1 and 2, are regulated at the level of expression by the mTORC1 substrate,

Retinoid X Receptor (Norrmén et al., 2014) whilst proteolytic activation of these

transcription factors is mediated by S6K1 (Düvel et al., 2010).

Page 231

Figure 5.1.1a – Overview of the Role Played by S6K1 in Regulating Multiple Metabolic

Pathways. The mTORC1 substrate S6K1 is responsible for carrying out many of the regulatory

roles ascribed to mTORC1 in the regulation of metabolism. In the phosphorylated form, S6K1

can bring about changes in gene expression, by either increasing the level of phosphorylated

eukaryotic Initiation Factor 4B (eIF4B), mediating the proteolytic processing of the Sterol-

Regulatory Element Binding Proteins (SREBP) 1/2 (SREBP1/2c), or by directly phosphorylating

Carbamoyl-Phosphate Synthetase 2, Aspartate Transcarbamylase and Dihydroorotase (CAD).

Page 232

Section 5.2 – Methods

Section 5.2.1 – Proposed New Approaches

The asynchronous stochastic method outlined in Chapter 2 – Model Creation & Refining

uses a Monte Carlo method to analyse sequences of feasible molecular interactions, given

the connectivity of the model. As described in Section 2.1.4.2 – Evaluation of the Predictive

Power of the Model, the 1,000 trajectories sampled for the Wild Type and inhibitor models

were interrogated using Reachfq and 99.99% Binomial Confidence Intervals calculated and

regions of significant difference were identified. A benchmark dataset (the Metabolic

Benchmark dataset) detailing published effects of each inhibitor, on the selected model

components, was used to validate the behaviour of the model and, quantitatively, analyse

the predictive power of the model. Both the MCC (Matthews, 1975) and the AC (Burset &

Guigó, 1996) were again used to convert the qualitative measures of predictive power, given

by the comparison of simulation behaviours to the benchmark datasets, to a quantitative

measure. The equations used to calculate these measures of predictive power are given in

Section 2.1.4.2 – Evaluation of the Predictive Power of the Model.

The asynchronous simulation is computationally more demanding, as large numbers of

molecular interactions firing need to be generated and saved to disk space for further

analysis. This creates a challenge for QSSPN simulations integrating GSMN models. In every

instance where a Constraint Place state change, the FBA of the entire GSMN needs to be run

multiple times to update the states of objective nodes. Since FBA is the most time

consuming operation in the QSSPN algorithm, this increases the computational cost when

compared with the un-integrated signalling network alone.

Page 233

In order to attempt to reduce the computational demands of the simulations, additional

method were employed. The synchronous execution of at least some molecular

interactions would lead to a reduction in the number of iterations and thus, diminish the

number of times the FBA would need to be performed. Consequently, the simulation

output files would be smaller.

The first alternative method proposed was aimed at interpreting the model as a series of

ODEs (as discussed in Section 2.1.4.3.1 – Case Study Identifying Differences in the

Asynchronous Simulation Approach & the Synchronous Methods). This deterministic

approach was selected to apply to the full model, due the computational benefits this

method has over the asynchronous stochastic method. In this approach, the reaction kinetic

parameters are set equal to one. If a molecular species is present, the species

concentration variable becomes one and the reaction is enabled. Inhibitor concentrations

are seen as taking a value of one if it is absent, and so enabling the molecular interaction, or

zero if the species is marked with a molecular activity level. This would have the effect of

reducing the solution to that particular ODE to zero. A further synchronous simulation

approach was tested, in which all enabled molecular interactions to fire in each time-step.

However, due to the deficiencies discussed in Section 2.1.4.3 – Synchronous Simulation

Methods and Section 2.1.4.3.1 – Case Study Identifying Differences in the Asynchronous

Simulation Approach & the Synchronous Methods this method was tested only with

Rapamycin incorporated into the model to establish if this approach could produce the

expected model predictions.

Page 234

Both the asynchronous stochastic and ODE methods were implemented using version 3.0 of

the software QSSPN and developed by A. Kierzek (as part of the QSSPN formalism). These

methods were run using an identical model so as to ensure the results were comparable.

The QSSPN software was used here, as it allowed the integration of a qualitative model of

the regulation of translation initiation with a GSMN of a RAW 264.7 murine macrophage.

This GSMN has been shown to be capable of reproducing experimentally derived data

(Bordbar et al., 2012), and is constraint-based. The integration is achieved by identifying

transcription factors regulated by mTORC1 and linking these transcription factors to genes

found within the GSMN. As noted by Fisher and colleagues (2013), the activation of the

constraint-based GSMN is achieved only if the activation thresholds are reached in the

qualitative model. 23 metabolic genes, from a range of metabolic pathways including the

pentose phosphate pathway, glycolysis and lipid and sterol biosynthesis, were linked to

mTORC1 activation in this way. The functioning of the integrated model was confirmed by

the production of Biomass. Biomass, as a ‘metabolite’ within the flux balanced analysis

model, is the sum of all the external metabolites produced by the model, as such Biomass

production was viewed as being required for translation to occur.

Section 5.3 - Results

Section 5.3.1 – Current, Asynchronous, QSSPN Simulation Method

Initial integration of the models was aimed at linking the mTORC1 signalling pathway to the

regulation of metabolic pathways found in the RAW 264.7 cell GSMN. This integration

resulted in 23 genes required for multiple metabolic pathways, including glycolysis, the

pentose phosphate pathway and lipid/sterol biosynthesis, being placed under the control of

mTORC1. In order to incorporate the expression of 23 new metabolic genes and the effects

Page 235

of the mTORC1 substrate, S6K1 on these genes, it was necessary to add 197 novel molecular

interactions. The expression of these 23 genes required the addition of multiple new

molecular species including several transcription factors and the stages of expressing these

genes. In total, including the repetition of molecular species to account for involvement in

multiple molecular interactions, 188 new Petri Net compentents, to model these species,

were incorporated into the model.

Before the two new methods could be tested, a point of comparison was necessary. This

point of comparison was obtained by running simulations using the method detailed in

Chapter 2 – Model Creation & Refining. The novel methods required version 3.0 of QSSPN.

Since the implementation of the synchronous approach altered the main simulation loop in

the QSSPN code, validation of the new code was necessary. Accordingly, version 1.0 of

QSSPN was also included in this new round of simulations to ensure that functionality of

QSSPN had been conserved through to version 3.0. The dataset against which the model

integrated with the RAW 264.7 cell GSMN was tested is shown in Supplementary

Information 3. This dataset was termed the Metabolic Benchmark, composed of 186

literature articles, concerned the effects of chemical inhibitors on key model effectors. The

bibliographic information concerning this Benchmark can be found in Supplementary

Information 1a. Simulations show that the results obtained with the asynchronous method

implemented in version 3.0 of the QSSPN software are identical to the results obtained with

version 1.0.

The simulations run in this section were done using the method presented and used

elsewhere in this project. These results were obtained using version 3.0 of QSSPN. These

results (shown in Supplementary Information E7 – Frequency Data Generated by Reachfq

Page 236

with Associated 99.99% Binomial Confidence Intervals (sheet entitled Asynchronous Method

Results)) were then to be used as a benchmark against which the other two, alternative

methods would be tested. The upper and lower 99.99% Binomial Confidence Intervals,

calculated in the software R, were then calculated for each identified key model effector.

As Rapamycin has been shown to affect eIF4E phosphorylation via MNK2, elsewhere in this

project, it is worth noting that the integration of the Petri Net model with the RAW 264.7

cell GSMN did not eliminate this prediction. The frequency of trajectories resulting in eIF4E

phosphorylation was reduced when Rapamcyin was introduced to the model (0.472 in the

Wild Type model versus 0.007 in the Rapamycin model). Moreover, the frequency of

trajectories leading to MNK2 phosphorylation was also altered, with an increase from 0.502

in the Wild Type model to 0.845, in the Rapamycin model.

The summary of the Confusion Matrices, shown in Figure 5.3.1ai, describe the comparison

of the results of simulations with QSSPN version 3.0, with the literature-based Metabolic

Benchmark dataset. For each inhibitor model, a summary of the predictive power is given.

The overall measure of predictive power, given in Figure 5.3.1aii, contains the pooled results

of all the matrices. This then provides enough information to make a reasonable

assessment of the overall predictive power of the model. The results of individual models

are of interest when looking at where model improvements are needed.

Page 237

i.


BIX02189 1 7 0 3 0.4183

CGP57380 3 6 0 1 0.8081

GSK2334470 2 11 1 2 0.4623

LY294002 4 3 1 9 0.0537

PD184352 6 1 0 7 0.2402

PD98059 11 3 0 3 0.6268

Rapamycin 9 3 1 4 0.3812

SB203580 5 2 4 8 -0.2630

SP600125 11 3 1 4 0.7100

XMD8-92 2 2 0 1 0.6667

ii.




Overall 54 41 8 42 0.3789 0.3762 4.4x10-7

Figure 5.3.1a – Predictive Power of the Model of Translational Regulation Integrated with a GSMN

of RAW 264.7 Cell Metabolism and Simulated with the Current, Asynchronous Method. (i)

Following the simulation and interrogation of the model with QSSPN (Version 3.0) and Reachfq, the

frequency of trajectories resulting in the activation of each model effector, were calculated for the

Wild Type and inhibitor models. By calculating the 99.99% Binomial Confidence Intervals for each

effector, and comparing the Wild Type and inhibitor models, it was possible to identify regions of

significant difference. Comparing the behaviour of the model to the Metabolic Benchmark dataset

allowed the construction of Confusion Matrices with behaviour being characterised as being either

True Positive (TP), True Negative (TN), False Positive (FP) or False Negative (FN). Calculating the

Matthews’ Correlation Coefficient (MCC) converted this qualitative measure of predictive power to a

quantitative one. (ii) The overall predictive power of the model was calculated by pooling the

variables in the Confusion Matrices of the individual inhibitors. The Approximate Coefficient (AC)

was also calculated. By conducting a Chi Square Test, using the resulting α-Value, it was possible to

confirm that the predictive power of the model was significantly different from random.

Page 238

In order to ensure the continued functioning of QSSPN, it was necessary to ensure

agreement between versions 1.0 and 3.0. The data shown in Supplementary Information E7

– Frequency Data Generated by Reachfq with Associated 99.99% Binomial Confidence

Intervals details the values obtained with version 3.0. Although the data is not shown here,

those values are identical to those obtained with version 1.0 and so demonstrates that

functionality of the QSSPN software is conserved. Integration of the model is known to be

functioning as Biomass was produced in the Wild Type model, albeit at a low level with the

frequency of trajectories reaching Biomass production being 0.044. Such a low frequency is

not unexpected given the experimentally derived lower bound of the Biomass reaction

within the RAW 264.7 cell GSMN model was set to 0.0281/h (Bordbar et al., 2012). This may

be indicative of a need for a more comprehensive integration of the GSMN with the

signalling pathways.

During this evaluation of the functioning of the model, the alteration to the level of

phosphorylated eIF4E and a reduction to phosphorylated AKT1 during Rapamycin

incorporation were maintained. This suggests that the functioning of the model was

maintained during the integration with the RAW 264.7 cell GSMN. Despite the two

experimentally validated predictions being maintained through the integration process, the

MCC value is markedly, but not significantly (as determined by a Chi Square Test, α = 0.305),

lower than it was prior to the integration with the GSMN (MCC = 0.3759 versus MCC =

0.4558 (as calculated in Section 2.2.4 – Evaluation of the Predictive Power of the Final

Model)). Given this reduction in predictive power was quite substantial, albeit not

significant, it was decided that the method that produced the smallest decrease in MCC

Page 239

value would be the best method to take forward, even if this means simulations are

relatively expensive in a computational sense.

Section 5.3.2 – Proposed New Methods

Section 5.3.2.1 – Ordinary Differential Equation Method

The first proposed method was deterministic and treated the model as a series of ODEs.

Consequently, only one trajectory was necessary. While ODEs are usually used to express

quantitative models, they were applied here to a qualitative model. All continuous variables

were allowed to vary between zero and one. All kinetic parameters are set equal to one. If

a molecular species is not marked with any discrete activity levels, then this parameter is

multiplied by zero. Conversely, if molecular species are marked with discrete activity levels,

then this is multiplied by one.

Attempts with this method revealed that the predicted behaviours of alterations to the level

of eIF4E phosphorylation and a reduction in AKT1 phosphorylation, were shown to occur, as

shown in Figure 5.3.2.1a.

Page 240

Figure 5.3.2.1a – Summary of the Results of the Ordinary Differential Equation Simulation Method. This method evaluates the model as a series of ODEs in which the kinetic parameters are set equal to one and are multiplied by one if the molecular species are marked with discrete activity levels. This subjective, yet deterministic, method is evaluated in an equally qualitative manner by comparing the Wild Type model and an Inhibitor model. If an increase of more than 10% in the amount of each effector is observed in the simulation a ‘↑’ is given. Conversely, for a decrease of more than 10% a ‘↓’ designation is listed. Where no change occurs, the result is given by ‘-‘. For marginal increases or decreases (less than 10%), combinations of these are given.

BIX02189 CGP57380 GSK2334470 LY294002 PD184352 PD98059 Rapamycin SB203580 SP600125 XMD8-92

ppMEK1 - - - ↓ ↓ ↓ ↓ - - -

ppMEK2 - - - ↓ ↓ ↓ ↓ - - -

ppERK1 - - - ↓ ↓ ↓ ↓ - - -

ppERK2 - - - ↓ ↓ ↓ ↓ - - -

pRSK1 - - - - - - - - - -

rpS6-Pi - - - - - - - - - -

pMNK1 - ↓ ↑/- ↑ ↓ ↓ ↑ ↓ ↓ -

pMNK2 - ↓ ↑/- ↑ ↓ ↓ ↑ ↓ ↓ -

eIF4E-Pi - ↓ ↑/- ↑ ↓ ↓ ↓ ↓ ↓ -

pJNK1 - - - ↑ - - ↑ - ↓ -

cJun-Pi - - - - - - - ↓ ↓ -

ppp38a - - - ↑ - - ↑ ↓ - -

pERK5 ↓ - - ↑ ↑ ↓ ↑ - - ↓

PI3K-Act - - - ↓ - - - - - -

PIP3 - - - ↓ - - - - - -

PDK1-Act - - ↓ ↓ - - - - - -

pPKCz - - ↓ ↓ - - ↓ ↑ - -

ppAKT1 - - - ↓ - - ↓ - - -

ppS6K1 ↓/- ↓ ↓ ↑ ↓ ↓ ↓ ↓ ↓ ↓/-

eIF4B-Pi - - - ↓ ↑/- ↑/- ↑ ↓ ↑/- -

p4E-BP1 - - - - ↓ ↓ ↓ - - -

mTORC1-Act - - - - ↓ ↓ ↓ - - -

pTSC2 - - - - - - - - - -

c-Myc - - - - ↓ ↓ ↓ - - -

HIF-1a - - - - ↓ ↓ ↓ - - -

SCD1 - ↓ ↓ ↑ ↓ ↓ ↓ ↓ ↓ ↓/-

PFK - - - - ↓ ↓ ↓ - - -

BIOMASS - ↓ ↑ - - - ↓ - ↓ -

Page 241

It is worth noting that the use of the word ‘subjective’, in Figure 5.3.2.1a, to describe the

results of the ODE method refers only to the fact that as rates are multiplied by one if

discrete acitivty levels are present in the molecular species serving as reactants for a

particular molecular interaction. As a consequence of this use of pseudo-kinetic parameters

the traces of molecular amounts must be regarded as subjective. Had kinetic parameters

been available concerning the concentration of particular species and the rate constants for

model reactions, the output of the ODE model would be absolute with the presence of large

molecule numbers minimising deviations from the mean (Kleinstein & Singh, 2001).

The ODE method is a deterministic method that has the potential to greatly improve the

existing methodologies of systems biology but also suffers from significant drawbacks. This

method is attractive in terms of the contribution it makes towards the aim of making

simulations computationally less costly. The need for only one trajectory both speeds up

the running of simulations and reduces the computational power required. However, the

deterministic properties of this method, whilst attractive, suffer greatly from a lack of

realistic kinetic parameters. The results of this simulation suffer from a lack of sensitivity, as

shown in Figure 5.3.2.1a. In this Figure, no observed change in the trajectory was denoted

by ‘-‘, a marginal increase or decrease (of less than 10%) is denoted by ‘↑/-‘ or ‘↓/-‘,

respectively and larger increases or decreases, of greater than 10%, are given by ‘↑’ and

‘↓’, respectively. Only gross changes are visible using this method. The summary of the

Confusion Matrices shown below, Figure 5.3.2.1b, demonstrate this in the marked reduction

in the predictive power.

Page 242

i.


BIX02189 1 7 0 3 0.4183

CGP57380 3 5 1 1 0.5833

GSK2334470 2 11 1 2 0.4623

LY294002 5 1 3 8 -0.3110

PD184352 8 0 1 5 -0.2070

PD98059 10 3 2 3 0.3508

Rapamycin 11 0 3 3 -0.2141

SB203580 6 4 2 7 0.1207

SP600125 6 4 0 9 0.3508

XMD8-92 1 2 0 2 0.4082

ii.




Overall 53 37 12 43 0.2922 0.2926 1.38x10-5

Figure 5.3.2.1b – Overview of the Predictive Power of the Model when Simulated with the

Ordinary Differential Equation Method. (i) The interactions in the Petri Net model were converted

to a series of Ordinary Differential Equations. The tool to do this was developed by A. Kierzek as part

of the QSSPN project. By comparing the traces of each effector in the Wild Type model to those in

each inhibitor model, it was possible to identify regions of difference. This behaviour of the

inhibitors in the model were then compared to the Metabolic Benchmark dataset. This behaviour

was characterised as either being a True Positive (TP), True Negative (TN), False Positive (FP) or a

False Negative (FN). These Confusion Matrix variables were then converted to a quantitative

measure of predictive power in the form of the Matthews’ Correlation Coefficient (MCC). (ii) By

pooling these variables for each inhibitor, it was possible to calculate an overall MCC value. This

overall measure of predictive power was confirmed by calculating the Approximate Coefficient (AC).

Moreover, to demonstrate that the level of predictive power was significantly different from

random, the α-Value of a Chi Square Test was calculated.

Page 243

This marked reduction in the MCC values between the current, asynchronous method (MCC

= 0.3759) and this ODE method (MCC = 0.2922), whilst appearing to be non-significant (as

determined by a Chi Square Test, α = 0.492), was of great importance in deciding which

method was to be taken forward for use in later simulations.

Section 5.3.2.2 – Synchronous Method

Although synchronous simulations demonstrated that Rapamycin was able to alter the level

of phosphorylated eIF4E and reduce AKT1 phosphorylation, it did produce some rather

questionable results from the model of translational regulation integrated with the GSMN.

As expected from Section 2.1.4.3.1 – Case Study Identifying Differences in the

Asynchronous Simulation Approach & the Synchronous Methods, MNK2 phosphorylation

was abolished by this method.

Signalling pathways require a highly coordinated response to a signal in order for signal

transduction to occur. Each interaction must occur in a tightly temporally regulated fashion

(Kholodenko et al., 2010). The synchronous approach assumes that all molecular

interactions proceed with the same velocity. While it is easy to understand how the

temporal nature of a tiered signalling pathway is essential, it may be less clear for cross-talk

between signalling pathways. In a MAPK cascade for example, in order for the next kinase

to be phosphorylated the upstream kinase must be activated (Ibid). Similarly, the temporal

nature of feedback mechanisms are clear. During negative feedback, a downstream effector

acts to inhibit the pathway and thus prevents further transduction. This allows oscillatory

behaviour to be observed (Kholodenko, 2006). However, in order to understand how cross-

talk between signalling pathways is temporally regulated, it is necessary to note that the

nature of cross-talk is complex and may only occur if sustained activation of a pathway is

Page 244

achieved (Housden & Perrimon, 2014). Moreover, even within a particular pathway the

efficacy of the interaction between the kinase and differing downstream targets is not

uniform (Loog & Morgan, 2005 (as cited by Jin et al., 2008b)).

To relate this back to the synchronous approach, if AKT1 has in excess of 70 known cellular

targets (Martelli et al., 2012) it is likely that AKT1 is able to interact with some with greater

efficiency than others. Consequently, these substrates will be phosphorylated in a much

shorter timescale than others. The synchronous approach does not take this into account

when firing all enabled molecular interaction.

While the asynchronous approach does not include exact reaction rates, it does allow for

some molecular interactions to occur before others. Gillespie algorithm simulations are

used to create an ensemble of 1,000 individual trajectories, in which different orders of the

firing of molecular interactions are tested. In this ensemble, there are examples of

trajectories in which reactions occur in the correct order for biological behaviour to be

feasible. In this way, non-parametric, qualitative simulations with the Gillespie algorithm

are capable of generating biological behaviour of interest in signalling cascades. The

qualitative synchronous approach is not capable of doing this. The only way to simulate

signalling networks in a synchronous way would be to fully parameterise the model and run

ODE simulations. Thus, examples in this dissertation show that the asynchronous

simulations with the Gillespie algorithm can be used to qualitatively model signalling

networks, while a synchronous approach is not.

Page 245



Given the reduction in predictive power that accompanied the integration of the RAW 264.7

cell GSMN, it would seem logical to attempt to increase the predictive power of the model

through considerable increases in the level of cross-talk between the signalling pathways. A

more detailed understanding of the nature of the interaction between signalling pathways,

other than mTORC1, and the regulation of metabolism would also be of benefit. As noted in

Chapter 4 – Modelling the Effects of Viral Infection, the preliminary work presented in this

chapter is of importance as it provides a model which, when the predictive power is suitably

raised, will allow the effects of viral infection upon metabolism to be modelled.


Section 5.4.2.1 – Proposed Novel Modelling Strategies

Swanson and colleagues (2003) noted that one of the key requirements in mathematical

modelling is the need for as complete an understanding as possible of the biological system

in question. The ODE method, whilst initially promising, failed to meet this requirement.

Even with a seemingly complete knowledge of the reaction kinetic parameters required to

accurately model a biological system, the fundamental problems of failing to take into

account all variables and not having parameters that translate well to other systems are

always present. When referring to the models proposed by Tracqui et al. (1995) and

Swanson et al. (2002) of gliomas, Hatzikirou and co-workers (2005) remark that the

parameters used are only likely to refer explicitly to one system and may not be

representative of all glioma systems. This example serves to illustrate that even with truly

mathematical models it is difficult to represent a biological system in a way that is both

Page 246

realistic and meaningful. The treatment of the qualitative, stochastic translational control

model as an ODE model resulted in an obvious, albeit non-significant, decrease in predictive

power. The ODE method used in this study resulted in a considerable decrease in

computational cost but was only brought about at the expense of accuracy and

verisimilitude. This being said, of the new ODE and synchronous methods, only the ODE

method was worthy of carrying forward for a complete examination of the model.

Section 5.4.2.2 – Integration of the GSMN with the Translational Regulation

Model

In addition to the trial of new modelling approaches, the integration of a GSMN with the

model of the regulation of translation initiation represented a new direction for the project.

Given that protein synthesis requires around one third of the total ATP produced by

oxidative phosphorylation (Buttgereit & Brand, 1995), the ability to model both allowed the

dynamic relationship between the two processes to be simulated. The work shown here

establishes the plausibility of connecting a GSMN to a stochastic model. Although a

significant reduction in the predictive power of the translation model was noted, the

predictive power did remain positive. A possible explanation for this is that additional

signalling pathways must be linked to the expression of genes found within the Bordbar et

al. (2012) GSMN. As shown in Figure 5.3.1a in Section 5.3.1 – Current, Asynchronous,

QSSPN Simulation Method, there is low correlation between the majority of the inhibitor

models and model effectors involved in the linking of the translational regulation model and

the GSMN.

Page 247


Page 248


This chapter summarises the findings of the project as a whole and describes future uses of

the model presented here. More detailed information on the findings can be found in the

individual chapters.

The process of synthesising nascent polypeptides is a major energetic burden to the cell,

requiring around 30% of all ATP produced (Buttgereit & Brand, 1995). The understanding of

translation in mammalian cells is far from complete with much information being inferred

from work with yeast. This being said, whilst the process in both systems is analogous there

are differences. An illustration of the differences can be seen in the structure and size of

the yeast and mammalian ribosomes (Morgan et al., 2000). Moreover, translation is

involved in the pathophysiology of a number of significant human and veterinary diseases.

Graff and co-workers (2008) even posit that the targeting of specific initiation factors may

offer therapeutic approaches for diseases, such as cancer.

The signalling pathways that regulate translation are complex. The interactions that exist

between such pathways add a further level of complexity to the cellular environment. Siso-

Nadal and co-workers (2009) noted that in order for signals to be transduced in concert, the

signalling pathways must form an interconnected network. Indeed, the connectivity within

this network is the source of the robust nature of many of the dynamic behaviours that cells

exhibit (Ibid) and functions to allow the integration of multiple signals (Hartwell et al., 1999).

A consequence of this integration is to allow cells to demonstrate complex and coordinated

patterns of behaviour (Ibid). With this in mind, any attempt to model a process, such as

translation, which affects the functioning and indeed, viability of the cell must take into

account the interconnected nature of the signalling pathways.

Page 249

In Section 1.5.1 – Aims & Objectives, the aims of the project were listed. In all cases, these

aims were achieved with various degrees of success and work investigating them was

carried out. The main aim of this project was not to just computationally simulate the

regulation of mammalian translation initiation by a large-scale reconstruction of a signalling

network, but also to carry out these simulations alongside experimental validation of the

behaviour of the model. In keeping with this view, it was possible to experimentally

validate six predicted behaviours.

The overarching aim of this work was to produce a stochastic, qualitative model of the

regulation of translation initiation by a signalling network. Over the course of this project,

several accomplishments were necessary in order to achieve this aim:

Analyse available literature of over 1,100 articles on translational

regulation and construct large-scale molecular interactions network

model

Develop a literature-based benchmark dataset and a method to

quantitatively analyse the predictive power of the model

Experimental validation of the prediction involving an increase in the

phosphorylation of eIF4E in response to Rapamycin

Experimental validation of the prediction involving a decrease in the

phosphorylation AKT1 in response to Rapamycin

Provide a mechanism to explain the increase in eIF4E phosphorylation

Incorporate the effects of viral infection into this model

Experimental validation of the prediction involving an increase in CREB

phosphorylation during MNV infection and three other behaviours

Page 250

Quantification of the model through the incorporation of a model of

ERK1/2 activation

Attempt to link metabolism to the regulation of translation

Section 6.2 – Model Construction & Predictive Power

Evaluation

In order to construct the model it was necessary to collate information published in over

1,158 peer-reviewed articles. This model was then converted to the binary model

formalism, which represents the activation conditions of each molecular interaction,

without introducing unrealistic oscillations. The final model accounted for 584 interactions

between 511 molecular species. This model differs from currently existing ones, such as

that of Firczuk et al. (2013), in several regards. The yeast model presented by Firczuk and

co-workers (2013) is considerably smaller than the one created here and does not include

the regulatory network controlling translation initiation. This difference is important as it

means that the model of regulating translation covers a much larger range of pathways.

Consequently, this model has a much wider range of future implications.

The future development of this model should focus upon linking translational regulation to

important and emerging pathways and processes. Furthermore, any continuation of this

work should also look to incorporate new pathways that are found to regulate translation,

for example the phosphorylation of eIF4B by Pim kinases (Yang et al., 2013b; Cen et al.,

2014).

Given that cellular stress, and the formation of Stress Granules, have been shown to bring

about considerable alterations to the levels of translation (Hoffman et al., 2012; Ruggieri et

Page 251

al., 2012), it seems that the incorporation of the process of forming Stress Granules should

be considered. If the translational control model were to incorporate Stress Granule

formation, it would better allow the phosphorylation of eIF2α to be modelled. This point of

translational regulation has been one of the hardest to model as the exact nature of the

input signal was hard to define and the upstream signalling pathways are very complex and

not necessarily directly linked to the regulation of translation. Moreover, the biological

relevance of eIF2α phosphorylation in RAW 264.7 cells is hard to define. Overexpression of

Monocyte Chemotactic Protein-Induced Protein 1 has been shown to attenuate the

phosphorylation of this initiation factor (Qi et al., 2011). Similarly, given the model is

amenable to the simulation of viral infection, it would be necessary to also model strategies

used by RNA viruses to limit the activation of PKR and the interferon response (Överby et

al., 2010; Uchida et al., 2014).

The cell cycle represents the second process that would be of interest to link to translation.

The process of translation has been noted to be a key regulatory point in the progression of

the cell cycle in yeast (Daga & Jimenez, 1999), human (Göpfert et al., 2003) and Xenopus

(Groisman et al., 2002) cells.

A further application for extending this work may come from combining the model of the

regulation of translation initiation with models of elongation. Attempts at this may prove

difficult given the qualitative nature of the model presented in this work. However, the

stochastic framework used in the implementation of the Totally Asymmetric Exclusion

Process models of elongation, such as that of Reuveni and co-workers (2011), may lead to a

better understanding of how translation is regulated by a combination of extrinsic (in the

Page 252

case of the signalling pathways) and intrinsic (for example at the level of the ribosome)

factors.

The conversion to the binary model formalism eliminated the random fluctuations between

the on and off states of effectors, such as kinases, in the model. Moreover, a binary model,

in which effectors are either active or inactive, removes the pseudo-kinetic parameters that

marking molecular species with multiple discrete activity levels introduces and, therefore,

the model is governed solely by the known connectivity of the molecular network.

Furthermore, the functionality of the model is reduced to simulating the conversion of

sufficient model effectors that would bring about a biological effect. This binary model

formalism represents a powerful, novel method of simulating large-scale biological systems.

One of the major outcomes of this work was the creation of a benchmark dataset and a

method for measuring the predictive power of the model. This method concerns a

comparison of the behaviour of the model with known effects. Existing literature has been

examined to compile data on the effect of 12 chemical inhibitors on 23 molecular effectors.

Using the Human Phospho-MAPK Arrays, it was possible to examine how 13 of these

effectors respond to Rapamycin in RAW 264.7 cells. From this it was possible to provide a

quantitative measure of predictive power by way of the MCC and the AC. Beck and co-

workers (1997) argued that model validity is a determination as to whether the system can

carry out a particular function in a consistent manner with the number of adverse outcomes

being minimised. To this end, this model validation approach used here provides insight

into how well existing data can be reproduced by the model and gives an indication as to

how predictive the behaviour of the model is.

Page 253

Based on the findings of this work that the MCC value is sensitive to small changes in the

model, it is recommended that this method should be taken forward. However, this in silico

predictive power analysis has a potential shortcoming. The method is only amenable to

large systems or systems where sufficient data about model components is available. For

such models, the number of data points is limited and therefore the resulting MCC value is

susceptible to large changes being introduced after relatively small changes to model

connectivity. It was shown in Section 2.2.1.1 – Updated Benchmark Dataset that the use of

data only relating to small data subsets, such as specific cell types, can also lead to a marked

reduction in data points. For this reason, the inclusion of data from all cell types was used.

Although it could be argued that this would serve to reduce the overall value of the dataset

against which to compare the model, the large size of this dataset means that variations are

potentially reduced.

Although Fisher and co-workers (2013) employed the use of MCC values to measure the

predictive power of the hepatocyte model, the use of this correlation coefficient in this

project is novel in one key regard. In order to evaluate the functioning of the translational

regulation model, no dataset against which to compare model behaviour was available.

Fisher et al. (2013) were able to compare the behaviour of the hepatocyte model to the

work of Song et al. (2009). Here 134 literature articles have been used to create a

benchmark dataset of the effects of 12 chemical inhibitors on 23 model effectors. Apart

from being applied for model validation, this dataset constitutes a comprehensive literature

review that can be used by researchers when planning future work on the effects various

inhibitors have upon key model signalling effectors.

Page 254

The standardisation of nomenclature used in the model, with links to databases through the

assignment of identifiers, adds to the quality of the project by providing an open and

transparent development process. One of the major obstacles within the field of systems

biology is the arbitrary naming of molecular species. This prevents the integration of

models developed by different groups. Whilst the Systems Biology Markup Language has

helped to standardise biochemical models (Hucka et al., 2003 (as cited by Bornstein et al.,

2008)), the links to established databases, such as Uniprot and Entrez, will help to make the

model more accessible to those outside of systems biology.

Section 6.3 – Experimental Validation

In an oversimplified way, systems biology aims to reconstruct biological systems and, using

these models, to generate predictions (Ideker et al., 2001). The usefulness of the model

constructed here was, therefore, demonstrated by the ability to generate seven predicted

behaviours that were linked to translational machinery directly and also the regulatory

signalling pathways. Furthermore, these behaviours were amenable to experimental

validation. Moreover, as stated above, the model not only consistently reproduced the

behaviour of inhibitors in multiple cell types but also displayed good agreement with

experimental data collected in RAW 264.7 cells treated with Rapamycin using the Human

Phospho-MAPK Arrays. Indeed this ability ‘to bring the theoretical predictions and

experimental data into close apposition’ is one of the fundamental principles of systems

biology (Ideker et al., 2001). Consequently, the experimental validation of six of the seven

predicted behaviours overcame a major obstacle for this work.

The ability to demonstrate that six in silico behaviours occur in a cell line adds significant

value to the model as a whole. The work undertaken in this project has produced a model

Page 255

of cellular translation that it has been experimentally validated. However, in regard to the

behaviour of the MNV model, despite several predicted behaviours being experimentally

validated, it was not possible to validate the behaviour of MSK2, in this model. This

indicates that further refinement of the model upstream of MSK2 is required.

Consequently, of the seven predicted behaviours that were experimentally tested, over the

course of this project, it was possible to confirm six of these. This corresponds to a success

rate of 85.7%.

The exact nature of eIF4E phosphorylation has only recently become known. Herdy et al.

(2012) noted that eIF4E phosphorylation allows for a change in gene expression. This view

is in keeping with the work presented here. Experimentally, Rapamycin-mediated mTORC1

inhibition was shown in Chapter 3 – Experimental Validation of Model Predictions, to cause

an increase in eIF4E phosphorylation. Given the role of mTORC1 in mediating cell growth

(Dowling et al., 2010) and the finding that cellular stress signalling is inhibited by the

mTORC1 substrate, S6K1 (Shin et al., 2013), it seems plausible that mTORC1 inhibition

should bring about a stressed state in the cell and necessitate a change in gene expression.

Whilst it is a little harder to conclusively state the effect Rapamycin has upon the

phosphorylation of AKT1, the work presented here does indicate that inhibition of this

protein kinase occurs, possibly through mTORC2 inhibition, following one hour of treatment

of cells with this macrolide. The work presented in Chapter 3 – Experimental Validation of

Model Predictions, using Human Phospho-MAPK Arrays, clearly demonstrates that the level

of phosphorylated AKT Pan is significantly reduced. This is in keeping with the work of Chen

and co-workers (2010). These authors observed that high concentrations of Rapamycin

served to reduce the phosphorylation of AKT1. Only with respect to treatment length does

Page 256

the work demonstrated here offer some disagreement with Sarbassov et al. (2006). Both of

these studies use treatment lengths of 24 hours. However, Sarbassov and co-workers

(2006) noted that the effect of diminishing AKT1 phosphorylation was cell type specific. This

specific effect correlates well with the work by Schreiber et al. (2015) in which the ratio of

FKBP12 and FKBP51 was found to be important for determining the extent of Rapamycin-

mediated inhibition of mTORC2. As noted in Chapter 3 – Experimental Validation of Model

Predictions, determining this ratio in RAW 264.7 cells may prove useful. Despite the fact

that immunoblotting with lysates of cells treated with Rapamycin for three hours failed to

demonstrate that AKT1 phosphorylation was reduced, it is still possible to argue that this

hypothesis has been, to some extent, experimentally demonstrated. Possible reasons for

the inability to corroborate the Human Phospho-MAPK Array result is given in Section 3.4.2

– Validation of the AKT1 Phosphorylation Prediction.

Section 6.4 – Mechanism for Rapamycin-Induced Increases

in eIF4E Phosphorylation

Whilst experimentally, Rapamycin has been shown to increase the level of eIF4E

phosphorylation, the first version of the binary model predicted a decrease (as noted in

Section 2.2.2 – Re-Evaluation of the Binary Model). This discrepancy motivated further

work that has led to the improvement of the model and a better understanding of the

nature of the relationship between mTORC1 and eIF4E. In order to identify a potential

mechanism by which eIF4E phosphorylation can be increased by Rapamycin, a small model

was created. This model only concerned the molecular interactions directly linked to

mTORC1 activation and eIF4E phosphorylation. Interrogation of this small model revealed

Page 257

the three points which would allow mTORC1 inhibition to positively affect eIF4E

phosphorylation. These points, as noted in Section 2.2.3 – Mechanism by Which mTORC1

Inhibition Affects eIF4E Phosphorylation, detail factors which when taken together all

contribute to eIF4E phosphorylation. These involve the relative levels of eIF4E and 4E-BP1,

an alternative kinase for PKCα and potentially alterations to the activity of a protein

phosphatase. Work to correct the direction of change in eIF4E phosphorylation was able to

provide in silico evidence for two of these points. In the model eIF4G phosphorylation was

disconnected from eIF4E phosphorylation and 4E-BP1 was maintained at a level lower than

that of eIF4E. Together both changes permitted an increase in eIF4E phosphorylation that

was dependent upon MNK2 and independent of MNK1. This is in agreement with the work

by Stead and Proud (2013) in which MNK2 was implicated in mediating a Rapamycin-

induced increase in eIF4E phosphorylation. Moreover, the simulations presented here

suggest that this increase is not dependent upon either ERK1/2 or p38, again in agreement

with earlier work (Stead & Proud, 2013; Eckerdt et al., 2014). Although numerous other

studies have also noted that Rapamycin is capable of increasing eIF4E phosphorylation (Sun

et al., 2005; Wang et al., 2007c; Grosso et al., 2011; Marzec et al., 2011), no real consensus

was found between these studies as to how mTORC1 is related to the post-translational

modification of this initiation factor.

Section 6.5 – Models of Viral Infection

In order to demonstrate the potential of the translational control model, perturbations

normally associated with viral infection were implemented into the model. Additionally, to

show the flexibility of the modelling approach, a range of viral effects were chosen. Two of

the viruses (JUNV and ANDV) are severe, emerging human pathogens about which little is

Page 258

known. The third virus, MNV, is a surrogate of the medically important HuNV. Although not

permissive in RAW 264.7 cells, ANDV and JUNV are capable of inducing changes in the

activation of different cellular signalling pathways. The changes that were made to the

three models were all taken from existing literature or, in the case of MNV, from

experimental data generated in-house.

QSSPN simulations and Reachfq analysis revealed that MNV infection induces considerable

changes to the signalling network of the cell. As shown in Figure 6.5a, the outcome of these

simulations led to the conclusion that MNV not only mediates changes in gene expression

through the promotion of eIF4E phosphorylation (Royall et al., 2015), but this also occurs

through alterations to the phosphorylation states of the transcription factors, CREB, ATF-2

and Elk-1. The experimental increase in CREB phosphorylation correlates well to the

increase seen in the simulations presented in Chapter 4 – Modelling the Effects of Viral

Infection. Despite this increase not being an unexpected finding, seeing as it is downstream

of both p38 (Di Petro et al., 2007; Di Giacomo et al., 2009) and ERK1/2 (Impey et al., 1998;

Xing et al., 1998), it was not a perturbation that was specifically induced in the model and

thus, represents the experimental validation of one viral prediction.

Page 259

Figure 6.5a – Schematic of the Effects of MNV-Induced ERK1/2 and p38 Activation. MNV induces

the activation of ERK1/2 and p38α. Both the model and Human Phospho-MAPK Array data

suggest that p38α activation is independent of MKK3 and MKK6. The work of Royall et al. (2015)

noted that the increase in eIF4E phosphorylation is dependent upon p38α and MNK1.

Experimentally, an increase in the phosphorylation of CREB has been demonstrated. Given that no

significant change in MSK2 phosphorylation was noted in the array data., it is likely that the

increase in CREB phosphorylation requires ERK1/2- (Deak et al., 1998; Sindreu et al., 2007; Brami-

Cherrier et al., 2009) or p38α- (Aggeli et al., 2006; McCoy et al., 2007) dependent MSK1

phosphorylation (Ananieva et al., 2008). The stability of AU-Rich Element-containing mRNA is also

hypothesised as being affected during MNV infection with TTP phosphorylation being increased in

a MK2-dependent manner. Hypothesised links are denoted by broken blue arrows. The solid blue

arrow is the proposed mechanism by which the increase in CREB phosphorylation may occur

during MNV infection. Red arrows denote links where the model and array data are not in

agreement or, in the case of broken, faded red arrows, where link are found in the literature but

not believed to be important based upon the array data. Green arrows represent those reactions

perturbed in the model.

Page 260

In terms of the other viral models, the effects being simulated are of considerable

importance to public health in South America as the pathogens triggering them cause

severe and often fatal human infection. The incorporation of JUNV infection into the model

produced considerable effects to the cellular signalling network. The cause of which

appears to be connected to the feedback loops linking ERK1/2 to JNK1. Despite knowing

that ERK1/2 activation is required for JUNV replication (Rodríguez et al., 2014), it is unclear

whether this feedback loop is of importance during JUNV infection.

Whilst the model of ANDV infection produced only limited testable predictions, the results

do indicate that metabolic reprogramming of the cell could occur through mTORC1

activation, as indicated by the increase in HIF-1α expression (shown in Section 4.3.3 –

Simulation of the Andes Virus Models). The model presented here indicates that S6K1 and

4E-BP1 phosphorylation occurs during ANDV infection. This is in contrast with the work of

McNulty et al. (2013) but somewhat in agreement with Gavrilovskaya et al. (2013). It is

worth noting, however, that the latter publication used cells under hypoxic conditions but

does show that S6K1 phosphorylation accompanies ANDV infection. The use of

metabolomics approaches during ANDV infection may help to elucidate whether the

activation of mTORC1 is linked to metabolism.

The work presented in Chapter 4 – Modelling the Effects of Viral Infection clearly

demonstrates that the translational regulation model is amenable to modelling other

systems. Further work must be undertaken to improve the functioning of the model

integrated with the metabolic network. This is of considerable importance given the

growing body of literature identifying the metabolic network as a target for viral infection

(Vastag et al., 2011; Grady et al., 2013).

Page 261

Despite the successful modelling of viral infection, this section of work suffers from a lack of

suitable methods to quantify the predictive power, as noted in Section 4.4.1 – Problems

Encountered & Troubleshooting. The method used in this work concerned validation of

effects by comparing expected behaviours to the reality of simulations. Given the small

number of known effects for these viruses, the use of the method used elsewhere in this

project was not deemed appropriate. Future work looking at viral infection with such

models should be aimed at developing a method, perhaps using the MCC values, for the

quantification of predictive power. This method will become appropriate as more data is

collected and particularly so if more data concerning the effects chemical inhibitors can

have on the infection process.

Section 6.6 – Integration of the Translational Regulation

Model with a GSMN

Whilst the complete integration of the ERK1/2 model did not provide any additional benefit

to the model, linking mTORC1 signalling to the regulation of metabolism was tested. The

use of FBA models is recognised as allowing the experimentally suggested alterations in

metabolism to be recognised (Quek et al., 2014). Moreover, QSSPN permits the integration

of such models of metabolism with Petri Net models (Fisher et al., 2013). Given the

intimate relationship between cell growth and anabolic metabolism, linking mTORC1 to the

control of metabolic gene expression seemed a logical way to bring about this integration.

Several models of metabolism, derived from both the human (Thiele et al., 2013) and mouse

(Sigurdsson et al., 2010) genomes, are available. Whilst of value, neither of these models

was specific enough to use in a model that would ultimately be tested in RAW 264.7 cells.

Page 262

23 genes found in the Bordbar et al. (2012) GSMN of a RAW 264.7 cell were placed, based

upon existing literature, under to control of mTORC1.

The linking of the GSMN of a RAW 264.7 cell allowed the relationship between translation

and metabolism to be modelled. The production of Biomass in this GSMN was seen as a

requirement for protein synthesis. Although the frequency of Biomass production was low,

the lower bound of the Biomass reaction, as detailed in Bordbar et al. (2012), was set to the

low value of 0.0281/h. Although this reaction rate is not directly comparable with the

Reachfq analysis, this low frequency of Biomass production in the Wild Type model (with a

value of 0.044) can in a way be seen as in keeping with the published data. Despite the

integration being accomplished successfully, the result of this incorporation was a marked,

albeit not significant, reduction in predictive power (MCC = 0.3376 versus MCC = 0.4558).

Whilst no definitive reason can be given for this reduction, it could be argued that a more

comprehensive integration of the Petri Net model, through additional signalling pathways,

with the GSMN could serve to correct this. This argument is valid, as the comparison

between the literature dataset and the simulation results demonstrated a somewhat less

than ideal correlation for those effectors involved in linking of mTORC1 to the GSMN. This

first attempt at linking metabolism to the control of translation highlighted the need for a

more complete understanding of the dual regulation of translation and metabolism by the

cellular signalling network.

Section 6.7 – Novel Modelling Strategies

The main aim of the work linking metabolism to the translational regulation model was to

improve the overall quality of the model by allowing the interplay between metabolism and

translation to be modelled. However a second aspect to the work involved the testing of

Page 263

two novel modelling strategies. As detailed in Section 5.2.1 – Proposed New Approaches,

these approaches aimed to reduce the computational cost of running simulations with little

or no detrimental effects on predictive power. These methods, termed the synchronous

simulation and the Ordinary Differential Equation Methods, were run in parallel with the

asynchronous simulation method (outlined in Chapter 2 – Model Creation & Refining). The

performance of the two novel methods further validated the method that had been used

throughout the project and led to a better understanding of the need for the use of exact

stochastic simulation methods when modelling biological systems.

An initial test of these approaches was carried out to ensure that the two experimentally

tested behaviours of the model were conserved. The synchronous method was unable to

model the Rapamycin-mediated increase in eIF4E phosphorylation. It was discovered that

this approach introduces the false assumption that within a cell, all biochemical reactions

that can occur will occur, in the same time interval. The removal of this temporal control of

reactions, as noted in Section 5.3.2.2 – Synchronous Method, limits any further use of this

method. It highlights the importance of the process of selecting the next reaction, which is

based on probability and the Propensity Function (as discussed in Section 1.4.1 – Exact

Stochastic Simulation with the Gillespie Algorithm).

The other method, the Ordinary Differential Equation Method, relies on converting the

model to a series of qualitative ODEs, in which reaction kinetics are equal to one. Although

it was possible to quantitatively analyse the predictive power of the model using the MCC

and AC formulae, this method produced results that were entirely subjective and,

consequently, the identification of regions of difference was open to observer bias. Perhaps

Page 264

because of this subjective nature, the predictive power of the model analysed in this way

was significantly reduced from MCC = 0.3376 to MCC = 0.2785.

Section 6.8 – General Conclusion

In summary, the overall aim of this project was to create a model simulating the regulation

of mammalian translation initiation by the network of cellular signalling pathways. This

work has produced such a model and validated the predictive power of this model against a

unique benchmark dataset based on a comprehensive review of published literature.

Statistically significant predictive power has been achieved and the model has produced

experimentally testable hypotheses. Experimental work directed by these findings provided

mechanistic insight into cross-talk between the mTORC1 and eIF4E phosphorylation.

Moreover, this mechanism has provided some insight into the cellular stoichiometry of

eIF4E and the interacting partner, 4E-BP1. This model of cellular translation has also been

linked to the GSMN of a RAW 264.7 cell and given much needed insight into how such

integration, with a large-scale model, can be achieved. Finally, the overall potential for

future uses of this model has been demonstrated by successfully modelling the effects

several medically important viruses can have upon the cellular signalling network and the

translational machinery. Although much future work has been suggested throughout, the

main output of this project has been to produce a model that can function within other

larger models. The potential applications for this model are not limited to just viral

infections. Areas such as cancer or cardiovascular biology may also benefit from a model of

translation which, with adjustment, could be used to help guide research in these fields.

265

References

266

ICTV Website (2013). Available

at: http://ictvonline.org/virusTaxonomy.asp?version=2013 (Accessed: 12/23).

Abraham, R. T. & Wiederrecht, G.J. (1996) Immunopharmacology of Rapamycin. Annual

Review of Immunology, 14, 483-510.

Abramson, R. D., Dever, T.E., Lawson, T.G., Ray, B.K., Thach, R.E. & Merrick,

W.C. (1987) The ATP-Dependent Interaction of Eukaryotic Initiation Factors with mRNA. The

Journal of Biological Chemistry, 262, 3826-3832.

Abi Saab, W.F., Brown, M.S. & Chadee, D.N. (2012) MLK4β Functions as a Negative

Regulator of MAPK Signaling and Cell Invasion. Oncogenesis, 1, doi: 10.1038/oncsis.2012.6.

Abell, A.N., Granger, D.A. & Johnson, G.L. (2007) MEKK4 stimulation of p38 and JNK activity

is negatively regulated by GSK3β. The Journal of Biological Chemistry, 282, 30476-30484.

Aceto, N., Bertino, P., Barbone, D., Tassi, G., Manzo, L., Porta, C., Mutti, L. & Gaudino,

G. (2009) Taurolidine and Oxidative Stress: A Rationale for Local Treatment of

Mesothelioma. The European Respiratory Journal, 34, 1399-1407.

Acker, M. G., Shin, B.S., Dever, T.E. & Lorsch, J.R. (2006) Interaction between Eukaryotic

Initiation Factors 1A and 5B is Required for Efficient Ribosomal Subunit Joining. The Journal

of Biological Chemistry, 281, 8469-8475.

Acuña, U. M., Wittwer, J., Ayers, S., Pearce, C.J., Oberlies, N.H. & De Blanco,

E.J. (2012) Effects of (5Z)-7-Oxozeaenol on MDA-MB-231 Breast Cancer Cells. Anticancer

Research, 32, 2415-2421.

Aevermann, B. D., Pickett, B.E., Kumar, S., Klem, E.B., Agnihothram, S., Askovich, P.S.,

Bankhead 3rd, A., Bolles, M., Carter, V., Chang, J., Clauss, T.R., Dash, P., Diercks, A.H.,

Eisfeld, A.J., Ellis, A., Fan, S., Ferris, M.T., Gralinski, L.E., Green, R.R., Gritsenko, M.A.,

Hatta, M., Heegel, R.A., Jacobs, J.M., Jeng, S., Josset, L., Kaiser, S.M., Kelly, S., Law, G.L., Li,

C., Li, J., Long, C., Luna, M.L., Matzke, M., McDermott, J., Menachery, V., Metz, T.O.,

Mitchell, H., Monroe, M.E., Navarro, G., Neumann, G., Podyminogin, R.L., Purvine, S.O.,

Rosenberger, C.M., Sanders, C.J., Schepmoes, A.A., Shukla, A.K., Sims, A., Sova, P., Tam,

V.C., Tchitchek, N., Thomas, P.G., Tilton, S.C., Totura, A., Wang, J., Webb-Robertson, B.J.,

Wen, J., Weiss, J.M., Yang, F., Yount, B., Zhang, Q., McWeeney, S., Smith, R.D., Waters,

K.M., Kawaoka, Y., Baric, R., Aderem, A., Katze, M.G. & Scheuermann, R.H. (2014) A

Comprehensive Collection of Systems Biology Data Characterizing the Host Response to

Viral Infection. Scientific Data, 1, doi: 10.1038/sdata.2014.33.

Aggeli, I. K., Gaitanaki, C. & Beis, I. (2006) Involvement of JNKs and p38-MAPK/MSK1

Pathways in H2O2-Induced Upregulation of Heme Oxygenase-1 mRNA in H9c2 Cells. Cellular

Signalling, 18, 1801-1812.

267

Aikin, R., Maysinger, D. & Rosenberg, L. (2004) Cross-Talk between Phosphatidylinositol 3-

kinase/AKT and c-Jun NH2-Terminal Kinase Mediates Survival of Isolated Human

Islets. Endocrinology, 145, 4522-4531.

Alessi, D. R., James, S.R., Downes, C.P., Holmes, A.B., Gaffney, P.R.J., Reese, C.B. & Cohen,

P. (1997b) Characterization of a 3-Phosphoinositide-Dependent Protein Kinase which

Phosphorylates and Activates Protein Kinase Bα. Current Biology, 7, 261-269.

Alessi, D. R., Kozlowski, M.T., Weng, Q.-P., Morrice, N. & Avruch, J. (1997a) 3-

Phosphoinositide-Dependent Protein Kinase 1 (PDK1) Phosphorylates and Activates the p70

S6 Kinase in Vivo and in Vitro. Current Biology, 8, 69-81.

Alessi, D. R., Andjelkovic, M., Caudwell, B., Cron, P., Morrice, N., Cohen, P. & Hemmings,

B.A. (1996) Mechanism of Activation of Protein Kinase B by Insulin and IGF-1. The EMBO

Journal, 15, 6541-6551.

Alessi, D. R., Cuenda, A., Cohen, P., Dudley, D.T. & Saltiel, A.R. (1995) PD 098059 is a

Specific Inhibitor of the Activation of Mitogen-Activated Protein Kinase Kinase in Vitro and in

Vivo. The Journal of Biological Chemistry, 270, 27489-27494.

Algire, M. A., Maag, D. & Lorsch, J.R. (2005) Pi Release from eIF2, Not GTP Hydrolysis, is the

Step Controlled by Start-Site Selection during Eukaryotic Translation Initiation. Molecular

Cell, 20, 251-262.

Alhatlani, B., Vashist, S. & Goodfellow, I. (2015) Functions of the 5' and 3' Ends of

Calicivirus Genomes. Virus Research, 206, 134-143.

Alkalaeva, E. Z., Pisarev, A.V., Frolova, L.Y., Kisselev, L.L. & Pestova, T.V. (2006) In Vitro

Reconstitution of Eukaryotic Translation Reveals Cooperativity between Release Factors

eRF1 and eRF3. Cell, 125, 1125-1136.

Allen, L. F., Sebolt-Leopold, J. & Meyer, M.B. (2003) CI-1040 (PD184352), A Targeted Signal

Transduction Inhibitor of MEK (MAPKK). Seminars in Oncology, 30, 105-116.

Altman, J. K., Sassano, A., Kaur, S., Glaser, H., Kroczynska, B., Redig, A.J., Russo, S., Barr, S.

& Platanias, L.C. (2011) Dual mTORC2/mTORC1 Targeting Results in Potent Suppressive

Effects on Acute Myeloid Leukemia (AML) Progenitors. Clinical Cancer Research, 17, 4378-

4388.

Altmann, M., Handschin, C. & Trachsel, H. (1987) MRNA Cap-Binding Protein: Cloning of the

Gene Encoding Protein Synthesis Initiation Factor eIF-4E from Saccharomyces

cerevisiae. Molecular and Cellular Biology, 7, 998-1003.

Amable, L., Grankvist, N., Largen, J.W., Ortsäter, H., Sjöholm, Å. & Honkanen,

R.E. (2011) Disruption of serine/threonine Protein Phosphatase 5 (PP5:PPP5c) in Mice

Reveals a Novel Role for PP5 in the Regulation of Ultraviolet Light-Induced Phosphorylation

268

of serine/threonine Protein Kinase Chk1 (CHEK1). The Journal of Biological Chemistry, 286,

40413-40422.

Amorim, R., Costa, S.M., Cavaleiro, N.P., da Silva, E.E. & da Costa, L.J. (2014) HIV-1

Transcripts use IRES-Initiation Under Conditions Where Cap-Dependent Translation is

Restricted by Poliovirus 2A Protease. PLoS ONE, 9, doi: 10.1371/journal.pone.0088619.

Ananieva, O., Darragh, J., Johansen, C., Carr, J.M., McIlrath, J., Park, J.M., Wingate, A.,

Monk, C.E., Toth, R., Santos, S.G., Iversen, L. & Arthur, J.S. (2008) The Kinases MSK1 and

MSK2 Act as Negative Regulators of Toll-Like Receptor Signaling. Nature Immunology, 9,

1028-1036.

Andersson, K. & Sundler, R. (2006) Posttranscriptional Regulation of TNFα expression Via

Eukaryotic Initiation Factor 4E (eIF4E) Phosphorylation in Mouse Macrophages. Cytokine, 33,

52-57.

Andreo, V., Glass, G., Shields, T., Provensal, C. & Polop, J. (2011) Modeling Potential

Distribution of Oligoryzomys longicaudatus, the Andes Virus (Genus: Hantavirus) Reservoir,

in Argentina. EcoHealth, 8, 332-348.

Andreou, A. Z. & Klostermeier, D. (2014) eIF4B and eIF4G Jointly Stimulate eIF4A ATPase

and Unwinding Activities by Modulation of the eIF4A Conformational Cycle. Journal of

Molecular Biology, 426, 51-61.

Archer, A. M. & Rico-Hesse, R. (2002) High Genetic Divergence and Recombination in

Arenaviruses from the Americas. Virology, 304, 274-281.

Arif, M., Kazim, S.F., Grundke-Iqbal, I., Garruto, R.M. & Iqbal, K. (2014) Tau Pathology

Involves Protein Phosphatase 2A in Parkinsonism-Dementia of Guam. The Proceedings of the

National Academy of Sciences of the United States of America, 111, 1144-1149.

Arlot, S. & Celisse, A. (2010) A Survey of Cross-Validation Procedures for Model Selection.

Statistics Surveys, 4, 40-79.

Arribalzaga, R. A. (1955) New Epidemic Disease due to Unidentified Germ: Nephrotoxic,

Leukopenic and Enanthematous Hyperthermia. El Día médico, 27, 1204-1210.

Arvisais, E. W., Romanelli, A., Hou, X. & Davis, J.S. (2006) AKT-Independent

Phosphorylation of TSC2 and Activation of mTOR and Ribosomal Protein S6 Kinase Signaling

by Prostaglandin F2α. The Journal of Biological Chemistry, 281, 26904-26913.

Auer, T. O., Duroure, K., De Cian, A., Concordet, J.P. & Del Bene, F. (2014) Highly Efficient

CRISPR/Cas9-Mediated Knock-in in Zebrafish by Homology-Independent DNA

Repair. Genome Research, 24, 142-153.

269

Avdulov, S., Herrera, J., Smith, K., Peterson, M., Gomez-Garcia, J.R., Beadnell, T.C.,

Schwertfeger, K.L., Benyumov, A.O., Manivel, J.C., Li, S., Bielinsky, A.K., Yee, D., Bitterman,

P.B. & Polunovsky, V.A. (2015) eIF4E Threshold Levels Differ in Governing Normal and

Neoplastic Expansion of Mammary Stem and Luminal Progenitor Cells. Cancer Research, 75,

687-697.

Badger, A. M., Bradbeer, J.N., Votta, B., Lee, J.C., Adams, J.L. & Griswold,

D.E. (1996) Pharmacological Profile of SB 203580, a Selective Inhibitor of Cytokine

Suppressive Binding Protein/p38 Kinase, in Animal Models of Arthritis, Bone Resorption,

Endotoxin Shock and Immune Function. The Journal of Pharmacology and Experimental

Therapeutics, 279, 1453-1461.

Bai, Y., Xuan, B., Liu, H., Zhong, J., Yu, D. & Qian, Z. (2015) Tuberous Sclerosis Complex

Protein 2 (TSC2)-Independent Activation of mTORC1 by Human Cytomegalovirus

pUL38. Journal of Virology, 89, 7625-7635.

Bailey, J. & Europe-Finner, G.N. (2005) Identification of Human Myometrial Target Genes of

the c-Jun NH2-Terminal Kinase (JNK) Pathway: The Role of Activating Transcription Factor 2

(ATF2) and a Novel Spliced Isoform ATF2-Small. Journal of Molecular Endocrinology, 34, 19-

35.

Bain, J., Plater, L., Elliott, M., Shpiro, N., Hastie, C.J., McLauchlan, H., Klevernic, I., Arthur,

J.S., Alessi, D.R. & Cohen, P. (2007) The Selectivity of Protein Kinase Inhibitors: A further

Update. The Biochemical Journal, 408, 297-315.

Baldi, P., Brunak, S., Chauvin, Y., Andersen, C.A. & Nielsen, H. (2000) Assessing the

Accuracy of Prediction Algorithms for Classification: An Overview. Bioinformatics, 16, 412-

424.

Banaszynski, L. A., Liu, C.W. & Wandless, T.J. (2005) Characterization of the

FKBP.Rapamycin.FRB Ternary Complex. Journal of the American Chemical Society, 127,

4715-4721.

Barnhart, M. D., Moon, S.L., Emch, A.W., Wilusz, C.J. & Wilusz, J. (2013) Changes in Cellular

mRNA Stability, Splicing, and Polyadenylation through HuR Protein Sequestration by a

Cytoplasmic RNA Virus. Cell Reports, 5, 909-917.

Bar-Peled, L., Schweitzer, L.D., Zoncu, R. & Sabatini, D.M. (2012) Ragulator is a GEF for the

Rag GTPases that Signal Amino Acid Levels to mTORC1. Cell, 150, 1196-1208.

Bar-Peled, L., Chantranupong, L., Cherniack, A.D., Chen, W.W., Ottina, K.A., Grabiner, B.C.,

Spear, E.D., Carter, S.L., Meyerson, M. & Sabatini, D.M. (2013) A Tumor Suppressor

Complex with GAP Activity for the Rag GTPases that Signal Amino Acid Sufficiency to

mTORC1. Science, 340, 1100-1106.

270

Bauer, B., Krumbock, N., Fresser, F., Hochholdinger, F., Spitaler, M., Simm, A., Überall, F.,

Schraven, B. & Baier, G. (2001) Complex Formation and Cooperation of Protein Kinase

Cθ and Akt1/protein Kinase Bα in the NF-κB Transactivation Cascade in Jurkat T Cells. The


Beauchemin, C. & Laliberté, J.F. (2007) The Poly(A) Binding Protein is Internalized in Virus-

Induced Vesicles Or Redistributed to the Nucleolus during Turnip Mosaic Virus

Infection. Journal of Virology, 81, 10905-10913.

Beck, M. B., Ravetz, J.R., Mulkey, L.A. & Barnwell, T.O. (1997) On the Problem of Model

Validation for Predictive Exposure Assessments. Stochastic Hydrology and Hydraulics, 11,

229-254.

Ben-Sahra, I., Howell, J.J., Asara, J.M. & Manning, B.D. (2013) Stimulation of De Novo

Pyrimidine Synthesis by Growth Signaling through mTOR and S6K1. Science, 339, 1323-1328.

Bentobache, M. & Bibi, M.O. (2012) A Two-Phase Support Method for Solving Linear

Programmes: Numerical Experiments. Methematical Problems in Engineering, 2012, doi:

10.1155/2012/482193

Benyamini, T., Folger, O., Ruppin, E. & Shlomi, T. (2010) Flux Balance Analysis Accounting

for Metabolite Dilution. Genome Biology, 11, doi: 10.1186/gb-2010-11-4-r43.

Beretta, L., Gingras, A.-C., Svitkin, Y.V., Hall, M.N. & Sonenberg, N. (1996) Rapamycin

Blocks the Phosphorylation of 4E-BP1 and Inhibits Cap-Dependent Initiation of

Translation. The EMBO Journal, 15, 658-664.

Berlanga, J. J., Santoyo, J. & De Haro, C. (1999) Characterization of a Mammalian Homolog

of the GCN2 Eukaryotic Initiation Factor 2α kinase. European Journal of Biochemistry, 265,

754-762.

Beste, D.J.V., Hooper, T., Stewart, G., Bonde, B., Avignone-Rossa, C., Bushell, M.E.,

Wheeler, P., Klamt, S., Kierzek, A.M. & McFadden, J. (2007) GSMN-TB: A Web-Based

Genome-Scale Network Model of Mycobacterium tuberculosis Metabolism. Genome

Biology, 8, doi: 10.1186/gb-2007-8-5-r89

Bianchini, A., Loiarro, M., Bielli, P., Busà, R., Paronetto, M.P., Loreni, F., Geremia, R. &

Sette, C. (2008) Phosphorylation of eIF4E by MNKs Supports Protein Synthesis, Cell Cycle

Progression and Proliferation in Prostate Cancer Cells. Carcinogenesis, 29, 2279-2288.

Bidinosti, M., Ran, I., Sanchez-Carbente, M.R., Martineau, Y., Gingras, A.C., Gkogkas, C.,

Raught, B., Bramham, C.R., Sossin, W.S., Costa-Mattioli, M., DesGroseillers, L., Lacaille, J.C.

& Sonenberg, N. (2010) Postnatal Deamidation of 4E-BP2 in Brain Enhances its Association

with Raptor and Alters Kinetics of Excitatory Synaptic Transmission. Molecular Cell, 37, 797-

808.

271

Bieniossek, C., Schutz, P., Bumann, M., Limacher, A., Uson, I. & Baumann, U. (2006) The

Crystal Structure of the Carboxy-Terminal Domain of Human Translation Initiation Factor

eIF5. Journal of Molecular Biology, 360, 457-465.

Bishop, J. D., Nien, W.L., Dauphinee, S.M. & Too, C.K. (2006) Prolactin Activates

Mammalian Target-of-Rapamycin through Phosphatidylinositol 3-Kinase and Stimulates

Phosphorylation of p70S6K and 4E-Binding Protein-1 in Lymphoma Cells. The Journal of

Endocrinology, 190, 307-312.

Blüher, M., Bashan, N., Shai, I., Harman-Boehm, I., Tarnovscki, T., Avinaoch, E., Stumvoll,

M., Dietrich, A., Klöting, N. & Rudich, A. (2009) Activated Ask1-MKK4-p38MAPK/JNK Stress

Signaling Pathway in Human Omental Fat Tissue may Link Macrophage Infiltration to Whole-

Body Insulin Sensitivity. The Journal of Clinical Endocrinology and Metabolism, 94, 2507-

2515.

Blum, S., Schmid, S.R., Pause, A., Buser, P., Linder, P., Sonenberg, N. & Trachsel,

H. (1992) ATP Hydrolysis by Initiation Factor 4A is Required for Translation Initiation

in Saccharomyces cerevisiae. The Proceedings of the National Academy of Sciences of the

United States of America, 89, 7664-7668.

Bode, J. G., Ludwig, S., Freitas, C.A., Schaper, F., Ruhl, M., Melmed, S., Heinrich, P.C. &

Haussinger, D. (2001) The MKK6/p38 Mitogen-Activated Protein Kinase Pathway is Capable

of Inducing SOCS3 Gene Expression and Inhibits IL-6-Induced Transcription. Biological

Chemistry, 382, 1447-1453.

Bok, K., Prikhodko, V.G., Green, K.Y. & Sosnovtsev, S.V. (2009) Apoptosis in Murine

Norovirus-Infected RAW264.7 Cells is Associated with Downregulation of Survivin. Journal of

Virology, 83, 3647-3656.

Bordbar, A., Mo, M.L., Nakayasu, E.S., Schrimpe-Rutledge, A.C., Kim, Y.M., Metz, T.O.,

Jones, M.B., Frank, B.C., Smith, R.D., Peterson, S.N., Hyduke, D.R., Adkins, J.N. & Palsson,

B.O. (2012) Model-Driven Multi-Omic Data Analysis Elucidates Metabolic

Immunomodulators of Macrophage Activation. Molecular Systems Biology, 8, doi:

10.1038/msb.2012.21.

Bornstein, B. J., Keating, S.M., Jouraku, A. & Hucka, M. (2008) LibSBML: An API Library for

SBML. Bioinformatics, 24, 880-881.

Bouazza, B., Debba-Pavard, M., Amrani, Y., Isaacs, L., O’Connell, D., Ahamed, S., Formella,

D. & Tliba, O. (2014) Basal p38 Mitogen-Activated Protein Kinase Regulates Unliganded

Glucocorticoid Receptor Function in Airway Smooth Muscle Cells. American Journal of

Respiratory Cell and Molecular Biology, 50, 301-315.

Bousquet, O. & Elisseeff, A. (2002) Stability and Generalization. Journal of Machine Learning

Research, 2, 499-526.

272

Bowen, M. D., Peters, C.J. & Nichol, S.T. (1997) Phylogenetic Analysis of the Arenaviridae:

Patterns of Virus Evolution and Evidence for Cospeciation between Arenaviruses and their

Rodent Hosts. Molecular Phylogenetics and Evolution, 8, 301-316.

Bowick, G. C., Fennewald, S.M., Scott, E.P., Zhang, L., Elsom, B.L., Aronson, J.F., Spratt,

H.M., Luxon, B.A., Gorenstein, D.G. & Herzog, N.K. (2007) Identification of Differentially

Activated Cell-Signaling Networks Associated with Pichinde Virus Pathogenesis by using

Systems Kinomics. Journal of Virology, 81, 1923-1933.

Brami-Cherrier, K., Roze, E., Girault, J.A., Betuing, S. & Caboche, J. (2009) Role of the

ERK/MSK1 Signalling Pathway in Chromatin Remodelling and Brain Responses to Drugs of

Abuse. Journal of Neurochemistry, 108, 1323-1335.

Brancho, D., Tanaka, N., Jaeschke, A., Ventura, J.J., Kelkar, N., Tanaka, Y., Kyuuma, M.,

Takeshita, T., Flavell, R.A. & Davis, R.J. (2003) Mechanism of p38 MAP Kinase Activation in

Vivo. Genes & Development, 17, 1969-1978.

Brandt, B., Abou-Eladab, E.F., Tiedge, M. & Walzel, H. (2010) Role of the JNK/c-Jun/AP-1

Signaling Pathway in Galectin-1-Induced T-Cell Death. Cell Death & Disease, 1,

doi:10.1038/cddis.2010.1.

Brant, R., Harrell, F. & Biggerstaff, B. (Division of Vector-Borne Infectious Diseases, Centers

for Disease Control and Prevention, Fort Collins, CO (USA)) (1991-2016) Confidence

Intervals for Binomial Probabilities. Furman University, Greenville, SC (USA). [07/12/2016]

Available from:

http://math.furman.edu/~dcs/courses/math47/R/library/Hmisc/html/binconf.html

Brasey, A., Lopez-Lastra, M., Ohlmann, T., Beerens, N., Berkhout, B., Darlix, J.L. &

Sonenberg, N. (2003) The Leader of Human Immunodeficiency Virus Type 1 Genomic RNA

Harbors an Internal Ribosome Entry Segment that is Active during the G2/M Phase of the

Cell Cycle. Journal of Virology, 77, 3939-3949.

Breitling, R., Gilbert, D., Heiner, M. & Orton, R. (2008) A Structured Approach for the

Engineering of Biochemical Network Models, Illustrated for Signalling Pathways. Briefings in

Bioinformatics, 9, 404-421.

Brenner, S., Jacob, F. & Meselson, M. (1961) An Unstable Intermediate Carrying

Information from Genes to Ribosomes for Protein Synthesis. Nature, 190, 576-581.

Breuleux, M., Klopfenstein, M., Stephan, C., Doughty, C.A., Barys, L., Maira, S.M.,

Kwiatkowski, D. & Lane, H.A. (2009) Increased AKT S473 Phosphorylation After mTORC1

Inhibition is Rictor Dependent and does Not Predict Tumor Cell Response to PI3K/mTOR

Inhibition. Molecular Cancer Therapeutics, 8, 742-753.

273

Brown, E. J., Albers, M.W., Shin, T.B., Ichikawa, K., Keith, C.T., Lane, W.S. & Schreiber,

S.L. (1994) A Mammalian Protein Targeted by G1-Arresting Rapamycin-Receptor

Complex. Nature, 369, 756-758.

Brown, M. B. & Forsythe, A.B. (1974) Robust Tests for the Equality of Variances. Journal of

the American Statistical Association, 69, 364-367.

Burnett, P. E., Barrow, R.K., Cohen, N.A., Synder, S.H. & Sabatini, D.M. (1998) RAFT1

Phosphorylation of the Translational Regulators p70 S6 Kinase and 4E-BP1. The Proceedings

of the National Academy of Sciences of the United States of America, 95, 1432-1437.

Burset, M. & Guigó, R. (1996) Evaluation of Gene Structure Prediction

Programs. Genomics, 34, 353-367.

Buttgereit, F. & Brand, M.D. (1995) A Hierarchy of ATP-Consuming Processes in Mammalian

Cells. The Biochemical Journal, 312, 163-167.

Buxadé, M., Parra, J.L., Rousseau, S., Shpiro, N., Marquez, R., Morrice, N., Bain, J., Espel, E.

& Proud, C.G. (2005) The Mnks are Novel Components in the Control of TNFα Biosynthesis

and Phosphorylate and Regulate hnRNP A1. Immunity, 23, 177-189.

Cai, S.-L., Tee, A.R., Short, J.D., Bergeron, J.M., Kim, J., Shen, J., Guo, R., Johnson, C.L.,

Kiguchi, K. & Walker, C.L. (2006) Activity of TSC2 is Inhibited by AKT-Mediated

Phosphorylation and Membrane Partitioning. The Journal of Cell Biology, 173, 279-289.

Cai, W., Ye, Q. & She, Q.B. (2014) Loss of 4E-BP1 Function Induces EMT and Promotes

Cancer Cell Migration and Invasion Via Cap-Dependent Translational Activation of

Snail. Oncotarget, 5, 6015-6027.

Cai, X. & Xu, Z. (2007) K-Leap Method for Accelerating Stochastic Simulation of Coupled

Chemical Reactions. The Journal of Chemical Physics, 126, doi: 10.1063/1.2436869.

Cajimat, M. N., Milazzo, M.L., Bradley, R.D. & Fulhorst, C.F. (2007) Catarina Virus, an

Arenaviral Species Principally Associated with Neotoma Micropus (Southern Plains Woodrat)

in Texas. The American Journal of Tropical Medicine and Hygiene, 77, 732-736.

Cajimat, M. N., Milazzo, M.L., Mauldin, M.R., Bradley, R.D. & Fulhorst, C.F. (2013) Diversity

among Tacaribe Serocomplex Viruses (Family Arenaviridae) Associated with the Southern

Plains Woodrat (Neotoma micropus). Virus Research, 178, 486-494.

Cakmakci, N. G., Lerner, R.S., Wagner, E.J., Zheng, L. & Marzluff, W.F. (2008) SLIP1, A

Factor Required for Activation of Histone mRNA Translation by the Stem-Loop Binding

Protein. Molecular and Cellular Biology, 28, 1182-1194.

274

Calder, M., Vyshemirsky, V., Gilbert, D. & Orton, R. (2005) Analysis of Signalling Pathways

using the PRISM Model Checker. Proceedings of Computational Methods in Systems Biology,

2005, 179-190.

Calisher, C. H., Root, J.J., Mills, J.N., Rowe, J.E., Reeder, S.A., Jentes, E.S., Wagoner, K. &

Beaty, B.J. (2005) Epizootiology of Sin Nombre and El Moro Canyon Hantaviruses,

Southeastern Colorado, 1995-2000. Journal of Wildlife Diseases, 41, 1-11.

Calisher, C. H., Wagoner, K.D., Amman, B.R., Root, J.J., Douglass, R.J., Kuenzi, A.J., Abbott,

K.D., Parmenter, C., Yates, T.L., Ksiazek, T.G., Beaty, B.J. & Mills, J.N. (2007) Demographic

Factors Associated with Prevalence of Antibody to Sin Nombre Virus in Deer Mice in the

Western United States. Journal of Wildlife Diseases, 43, 1-11.

Camerini, V., Decimo, D., Balvay, L., Pistello, M., Bendinelli, M., Darlix, J.L. & Ohlmann,

T. (2008) A Dormant Internal Ribosome Entry Site Controls Translation of Feline

Immunodeficiency Virus. Journal of Virology, 82, 3574-3583.

Canagarajah, B. J., Khokhlatchev, A., Cobb, M.H. & Goldsmith, E.J. (1997) Activation

Mechanism of the MAP Kinase ERK2 by Dual Phosphorylation. Cell, 90, 859-869.

Cantley, L.C. (2002) The Phosphoinositide 3-Kinase Pathway. Science, 296, 1655-2657.

Cao, Y., Gillespie, D.T. & Petzold, L.R. (2006) Efficient Step Size Selection for the Tau-

Leaping Simulation Method. The Journal of Chemical Physics, 124, doi: 10.1063/1.2159468.

Carayol, N., Vakana, E., Sassano, A., Kaur, S., Goussetis, D.J., Glaser, H., Druker, B.J.,

Donato, N.J., Altman, J.K., Barr, S. & Platanias, L.C. (2010) Critical Roles for mTORC2- and

Rapamycin-Insensitive mTORC1-Complexes in Growth and Survival of BCR-ABL-Expressing

Leukemic Cells. The Proceedings of the National Academy of Sciences of the United States of

America, 107, 12469-12474.

Carberry, S. E., Rhoads, R.E. & Goss, D.J. (1989) A Spectroscopic Study of the Binding of

m7GTP and m7GpppG to Human Protein Synthesis Initiation Factor 4E. Biochemistry, 28,

8078-8083.

Carbonara, C., Longa, L., Grosso, E., Borrone, C., Garrè, M.G., Brisigotti, M. & Migone,

N. (1994) 9q34 Loss of Heterozygosity in a Tuberous Sclerosis Astrocytoma Suggests a

Growth Suppressor-Like Activity also for the TSC1 Gene. Human Molecular Genetics, 3,

1829-1832.

Caron, E., Ghosh, S., Matsuoka, Y., Ashton-Beaucage, D., Therrien, M., Lemieux, S.,

Perreault, C., Roux, P.P. & Kitano, H. (2010) A Comprehensive Map of the mTOR Signaling

Network. Molecular Systems Biology, 6, doi: 10.1038/msb.2010.108.

Carracedo, A., Ma, L., Teruya-Feldstein, J., Rojo, F., Salmena, L., Alimonti, A., Egia, A.,

Sasaki, A.T., Thomas, G., Kozma, S.C., Papa, A., Nardella, C., Cantley, L.C., Baselga, J. &

275

Pandolfi, P.P. (2008) Inhibition of mTORC1 Leads to MAPK Pathway Activation through a

PI3K-Dependent Feedback Loop in Human Cancer. The Journal of Clinical Investigation, 118,

3065-3074.

Carríere, A., Romeo, Y., Acosta-Jaquez, H.A., Moreau, J., Bonneil, E., Thibault, P., Fingar,

D.C. & Roux, P.P. (2011) ERK1/2 Phosphorylate Raptor to Promote Ras-Dependent

Activation of mTOR Complex 1 (mTORC1). The Journal of Biological Chemistry, 286, 567-577.

Carrozzino, F., Pugnale, P., Féraille, E. & Montesano, R. (2009) Inhibition of Basal p38 or

JNK Activity Enhances Epithelial Barrier Function through Differential Modulation of Claudin

Expression. American Journal of Physiology – Cell Physiology, 297, C775-C787.

Carter, M. J., Milton, I.D., Meanger, J., Bennett, M., Gaskell, R.M. & Turner, P.C. (1992) The

Complete Nucleotide Sequence of a Feline Calicivirus. Virology, 190, 443-448.

Carter, S., Auer, K.L., Reardon, D.B., Birrer, M., Fisher, P.B., Valerie, K., Schmidt-Ullrich, R.,

Mikkelsen, R. & Dent, P. (1998) Inhibition of the Mitogen Activated Protein (MAP) Kinase

Cascade Potentiates Cell Killing by Low Dose Ionizing Radiation in A431 Human Squamous

Carcinoma Cells. Oncogene, 16, 2787-2796.

Castillo, C., Naranjo, J., Sepúlveda, A., Ossa, G. & Levy, H. (2001) Hantavirus Pulmonary

Syndrome due to Andes Virus in Temuco, Chile: Clinical Experience with 16

Adults. Chest, 120, 548-554.

Castillo, C., Nicklas, C., Mardones, J. & Ossa, G. (2007) Andes Hantavirus as Possible Cause

of Disease in Travellers to South America. Travel Medicine and Infectious Disease, 5, 30-34.

Ceci, M., Gaviraghi, C., Gorrini, C., Sala, L.A., Offenhäuser, N., Marchisio, P.C. & Biffo,

S. (2003) Release of eIF6 (p27BBP) from the 60S Subunit Allows 80S Ribosome

Assembly. Nature, 426, 579-584.

Cen, B., Xiong, Y., Song, J.H., Mahajan, S., DuPont, R., McEachern, K., DeAngelo, D.J.,

Cortes, J.E., Minden, M.D., Ebens, A., Mims, A., LaRue, A.C. & Kraft, A.S. (2014) The Pim-1

Protein Kinase is an Important Regulator of MET Receptor Tyrosine Kinase Levels and

Signaling. Molecular and Cellular Biology, 34, 2517-2532.

Charrel, R. N. & de Lamballerie, X. (2010) Zoonotic Aspects of Arenavirus

Infections. Veterinary Microbiology, 140, 213-220.

Chaturvedi, D., Gao, X., Cohen, M.S., Taunton, J. & Patel, T.B. (2009) Rapamycin Induces

Transactivation of the EGFR and Increases Cell Survival. Oncogene, 28, 1187-1196.

Chaudhry, Y., Nayak, A., Bordeleau, M.-E., Tanaka, J., Pelletier, J., Belsham, G.J., Roberts,

L.O. & Goodfellow, I.G. (2006) Caliciviruses Differ in their Functional Requirements for eIF4F

Components. The Journal of Biological Chemistry, 35, 25315-25325.

276

Chaudhuri, J., Chakrabarti, A. & Maitra, U. (1997) Biochemical Characterization of

Mammalian Translation Initiation Factor 3 (eIF3): Molecular Cloning Reveals that p110

Subunit is the Mammalian Homologue of Saccharomyces cerevisiae Protein Prt1. The Journal


Chaudhuri, J., Chowdhury, D. & Maitra, U. (1999) Distinct Functions of Eukaryotic

Translation Initiation Factors eIF1A and eIF3 in the Formation of the 40S Ribosomal

Preinitiation Complex. The Journal of Biological Chemistry, 274, 17975-17980.

Chen, J., Martin, B.L. & Brautigan, D.L. (1992) Regulation of Protein Serine-Threonine

Phosphatase Type-2A by Tyrosine Phosphorylation. Science, 257, 1261-1264.

Chen, X. G., Liu, F., Song, X.F., Wang, Z.H., Dong, Z.Q., Hu, Z.Q., Lan, R.Z., Guan, W., Zhou,

T.G., Xu, X.M., Lei, H., Ye, Z.Q., Peng, E.J., Du, L.H. & Zhuang, Q.Y. (2010) Rapamycin

Regulates Akt and ERK Phosphorylation through mTORC1 and mTORC2 Signaling

Pathways. Molecular Carcinogenesis, 49, 603-610.

Cheng, E. & Mir, M.A. (2012) Signatures of Host mRNA 5' Terminus for Efficient Hantavirus

Cap Snatching. Journal of Virology, 86, 10173-10185.

Chiarini, F., Fala, F., Tazzari, P.L., Ricci, F., Astolfi, A., Pession, A., Pagliaro, P., McCubrey,

J.A. & Martelli, A.M. (2009) Dual Inhibition of Class IA Phosphatidylinositol 3-Kinase and

Mammalian Target of Rapamycin as a New Therapeutic Option for T-Cell Acute

Lymphoblastic Leukemia. Cancer Research, 69, 3520-3528.

Chiu, M. I., Katz, H. & Berlin, V. (1994) RAPT1, a Mammalian Homolog of Yeast Tor,

Interacts with the FKBP12/rapamycin Complex. The Proceedings of the National Academy of

Sciences of the United States of America, 91, 12574-12578.

Choo, M. K., Kawasaki, N., Singhirunnusorn, P., Koizumi, K., Sato, S., Akira, S., Saiki, I. &

Sakurai, H. (2006) Blockade of Transforming Growth Factor-β-Activated Kinase 1 Activity

Enhances TRAIL-Induced Apoptosis through Activation of a Caspase Cascade. Molecular

Cancer Therapeutics, 5, 2970-2976.

Chou, M. M., Hou, W., Johnson, J., Graham, L.K., Lee, M.H., Chen, C.S., Newton, A.C.,

Schaffhausen, B.S. & Toker, A. (1998) Regulation of Protein Kinase C ζ by PI 3-Kinase and

PDK-1. Current Biology, 8, 1069-1077.

Chow, J. C., Young, D.W., Golenbock, D.T., Christ, W.J. & Gusovsky, F. (1999) Toll-Like

Receptor-4 Mediates Lipopolysaccharide-Induced Signal Transduction. The Journal of

Biological Chemistry, 274, 10689-10692.

Chuang, H.Y., Hofree, M. & Ideker, T. (2010) A Decade of Systems Biology. Annual Review

of Cell and Developmental Biology, 26, 721-744

277

Chuluunbaatar, U., Roller, R., Feldman, M.E., Brown, S., Shokat, K.M. & Mohr,

I. (2010) Constitutive mTORC1 Activation by a Herpesvirus Akt Surrogate Stimulates mRNA

Translation and Viral Replication. Genes & Development, 24, 2627-2639.

Chung, L., Bailey, D., Leen, E.N., Emmott, E.P., Chaudhry, Y., Roberts, L.O., Curry, S.,

Locker, N. & Goodfellow, I.G. (2014) Norovirus Translation Requires an Interaction between

the C Terminus of the Genome-Linked Viral Protein VPg and Eukaryotic Translation Initiation

Factor 4G. The Journal of Biological Chemistry, 289, 21738-21750.

Cigan, A. M., Bushman, J.L., Boal, T.R. & Hinnebusch, A.G. (1993) A Protein Complex of

Translational Regulators of GCN4 mRNA is the Guanine Nucleotide-Exchange Factor for

Translation Initiation Factor 2 in Yeast. The Proceedings of the National Academy of Sciences

of the United States of America, 90, 5350-5354.

Cimica, V., Dalrymple, N.A., Roth, E., Nasonov, A. & Mackow, E.R. (2014) An Innate

Immunity-Regulating Virulence Determinant is Uniquely Encoded by the Andes Virus

Nucleocapsid Protein. mBio, 5, doi: 10.1128/mBio.01088-13.

Clay, S., Maherchandani, S., Malik, Y.S. & Goyal, S.M. (2006) Survival on Uncommon

Fomites of Feline Calicivirus, A Surrogate of Noroviruses. American Journal of Infection

Control, 34, 41-43.

Clement, S. L., Scheckel, C., Stoecklin, G. & Lykke-Andersen, J. (2011) Phosphorylation of

Tristetraprolin by MK2 Impairs AU-Rich Element mRNA Decay by Preventing Deadenylase

Recruitment. Molecular and Cellular Biology, 31, 256-266.

Cobb, M. H., Hepler, J.E., Cheng, M. & Robbins, D. (1994) The Mitogen-Activated Protein

Kinases, ERK1 and ERK2. Seminars in Cancer Biology, 5, 261-268.

Coelho, C., Souza, A.C., Derengowski, L.D., de Leon-Rodriguez, C., Wang, B., Leon-Rivera,

R., Bocca, A.L., Gonçalves, T. & Casadevall, A. (2015) Macrophage Mitochondrial and Stress

Response to Ingestion of Cryptococcus neoformans. The Journal of Immunology, 194, 2345-

2357.

Collins, B. J., Deak, M., Murray-Tait, V., Storey, K.G. & Alessi, D.R. (2005) In Vivo Role of the

Phosphate Groove of PDK1 Defined by Knockin Mutation. Journal of Cell Science, 118, 5023-

5034.

Cong, L., Ran, F.A., Cox, D., Lin, S., Barretto, R., Habib, N., Hsu, P.D., Wu, X., Jiang, W.,

Marraffini, L.A. & Zhang, F. (2013) Multiplex Genome Engineering using CRISPR/Cas

Systems. Science, 339, 819-823.

Connor, J. H. & Lyles, D.S. (2005) Inhibition of Host and Viral Translation during Vesicular

Stomatitis Virus Infection. eIF2 is Responsible for the Inhibition of Viral but Not Host

Translation. The Journal of Biological Chemistry, 280, 13512-13519.

278

Constantinou, C., Elia, A. & Clemens, M.J. (2008) Activation of p53 Stimulates Proteasome-

Dependent Truncation of eIF4E-Binding Protein 1 (4E-BP1). Biology of the Cell, 100, 279-289.

Copp, J., Manning, G. & Hunter, T. (2009) TORC-Specific Phosphorylation of Mammalian

Target of Rapamycin (mTOR): Phospho-Ser2481 is a Marker for Intact mTOR Signaling

Complex 2. Cancer Research, 69, 1821-1827.

Correa-Meyer, E., Pesce, L., Guerrero, C. & Sznajder, J.L. (2002) Cyclic Stretch Activates

ERK1/2 via G Proteins and EGFR in Alveolar Epithelial Cells. American Journal of Physiology –

Lung Cellular and Molecular Physiology, 282, L883-L891.

Covert, M. W., Schilling, C.H. & Palsson, B. (2001) Regulation of Gene Expression in Flux

Balance Models of Metabolism. Journal of Theoretical Biology, 213, 73-88.

Covert, M. W., Xiao, N., Chen, T. J. & Karr, J. R. (2008) Integrating Metabolic,

Transcriptional, Regulatory and Signal Transduction Models in Escherichia coli.

Bioinformatics, 24, 204-2050

Cowan, A. E., Moraru, I.I., Schaff, J.C., Slepchenko, B.M. & Loew, L.M. (2012) Spatial

Modeling of Cell Signaling Networks. Methods in Cell Biology, 110, 195-221.

Crews, C. M., Alessandrini, A. & Erikson, R.L. (1992) The Primary Structure of MEK, a

Protein Kinase that Phosphorylates the ERK Gene Product. Science, 258, 478-480.

Crosby, J. S., Lee, K., London, I.M. & Chen, J.J. (1994) Erythroid Expression of the Heme-

Regulated eIF-2α kinase. Molecular and Cellular Biology, 14, 3906-3914.

Cruzalegui, F. H., Cano, E. & Treisman, R. (1999) ERK Activation Induces Phosphorylation of

Elk-1 at Multiple S/T-P Motifs to High Stoichiometry. Oncogene, 18, 7948-7957.

Csibi, A., Lee, G., Yoon, S.O., Tong, H., Ilter, D., Elia, I., Fendt, S.M., Roberts, T.M. & Blenis,

J. (2014) The mTORC1/S6K1 Pathway Regulates Glutamine Metabolism through the eIF4B-

Dependent Control of c-Myc Translation. Current Biology, 24, 2274-2280.

Cuenda, A. & Cohen, P. (1999) Stress-Activated Protein Kinase-2/p38 and a Rapamycin-

Sensitive Pathway are Required for C2C12 Myogenesis. The Journal of Biological

Chemistry, 274, 4341-4346.

Cuenda, A. & Rousseau, S. (2007) p38 MAP-Kinases Pathway Regulation, Function and Role

in Human Diseases. Biochimica et Biophysica Acta - Molecular Cell Research, 1773, 1358-

1375.

Cui, J., Wang, Q., Wang, J., Lv, M., Zhu, N., Li, Y., Feng, J., Shen, B. & Zhang, J. (2009) Basal

c-Jun NH2-Terminal Protein Kinase Activity is Essential for Survival and Proliferation of T-Cell

Acute Lymphoblastic Leukemia Cells. Molecular Cancer Therapies, 8, 3214-3222.

279

Daga, R. R. & Jimenez, J. (1999) Translational Control of the cdc25 Cell Cycle Phosphatase: A

Molecular Mechanism Coupling Mitosis to Cell Growth. Journal of Cell Science, 112,, 3137-

3146.

Dangelmaier, C., Manne, B.K., Liverani, E., Jin, J., Bray, P. & Kunapuli, S.P. (2014) PDK1

Selectively Phosphorylates Thr(308) on Akt and Contributes to Human Platelet Functional

Responses.Thrombosis and Haemostasis, 111, 508-517.

Danial, J., Cepeda, J.A., Cameron, F., Cloy, K., Wishart, D. & Templeton,

K.E. (2011) Epidemiology and Costs Associated with Norovirus Outbreaks in NHS Lothian,

Scotland 2007-2009. The Journal of Hospital Infection, 79, 354-358.

Dantzig, G.B. (1951) ‘Maximization of a Linear Function of Variables Subject to Linear

Inequalities’ by R.C. Koopmans (ed.) in Activity Analysis of Production and Allocation, pp.

339-347., John Wiley & Sons, Inc., New York, NY (USA).

Dantzig, G.B. (1963) Linear Programming and Extensions. Princeton University Press,

Princeton, NJ (USA).

DaSilva, J., Xu, L., Kim, H.J., Miller, W.T. & Bar-Sagi, D. (2006) Regulation of Sprouty

Stability by Mnk1-Dependent Phosphorylation. Molecular and Cellular Biology, 26, 1898-

1907.

Daughenbaugh, K. F., Fraser, C.S., Hershey, J.W.B. & Hardy, M.E. (2003) The Genome-

Linked Protein VPg of the Norwalk Virus Binds eIF3, Suggesting its Role in Translation

Initiation Complex Recruitment. The EMBO Journal, 22, 2852-2859.

Daughenbaugh, K. F., Wobus, C.E. & Hardy, M.E. (2006) VPg of Murine Norovirus Binds

Translation Initiation Factors in Infected Cells. Virology Journal, 3, doi:10.1186/1743-422X-3-

33.

Davis, R. J. (2000) Signal Transduction by the JNK Group of MAP Kinases. Cell, 103, 239-252.

de la Torre, J. C. (2009) Molecular and Cell Biology of the Prototypic Arenavirus LCMV:

Implications for Understanding and Combating Hemorrhagic Fever Arenaviruses. Annals of

the New York Academy of Sciences, 1171, E57-E64.

Deak, M., Clifton, A.D., Lucocq, L.M. & Alessi, D.R. (1998) Mitogen- and Stress-Activated

Protein Kinase-1 (MSK1) is Directly Activated by MAPK and SAPK2/p38, and may Mediate

Activation of CREB. The EMBO Journal, 17, 4426-4441.

Deb, D. K., Sassano, A., Lekmine, F., Majchrzak, B., Verma, A., Kambhampati, S., Uddin, S.,

Rahman, A., Fish, E.N. & Platanias, L.C. (2003) Activation of Protein Kinase Cδ by IFN-γ. The

Journal of Immunology, 171, 267-273.

280

Decarlo, L., Mestel, C., Barcellos-Hoff, M.H. & Schneider, R.J. (2015) eIF4E is a Feed-

Forward Translational Coactivator of TGF β Early Pro-Transforming Events in Breast

Epithelial Cells. Molecular and Cellular Biology, 35, 2597-2609.

Decker, T., Hipp, S., Ringshausen, I., Bogner, C., Oelsner, M., Schneller, F. & Peschel,

C. (2003) Rapamycin-Induced G1 Arrest in Cycling B-CLL Cells is Associated with Reduced

Expression of Cyclin D3, Cyclin E, Cyclin A, and Survivin. Blood, 101, 278-285.

Decroly, E., Imbert, I., Coutard, B., Bouvet, M., Selisko, B., Alvarez, K., Gorbalenya, A.E.,

Snijder, E.J. & Canard, B. (2008) Coronavirus Nonstructural Protein 16 is a Cap-0 Binding

Enzyme Possessing (Nucleoside-2'O)-Methyltransferase Activity. Journal of Virology, 82,

8071-8084.

Delgado, S., Erickson, B.R., Agudo, R., Blair, P.J., Vallejo, E., Albariño, C.G., Vargas, J.,

Comer, J.A., Rollin, P.E., Ksiazek, T.G., Olson, J.G. & Nichol, S.T. (2008) Chapare Virus, a

Newly Discovered Arenavirus Isolated from a Fatal Hemorrhagic Fever Case in Bolivia. PLoS

Pathogens, 4, doi: 10.1371/journal.ppat.1000047.

Denisova, O. V., Söderholm, S., Virtanen, S., Von Schantz, C., Bychkov, D., Vashchinkina, E.,

Desloovere, J., Tynell, J., Ikonen, N., Theisen, L.L., Nyman, T.A., Matikainen, S., Kallioniemi,

O., Julkunen, I., Muller, C.P., Saelens, X., Verkhusha, V.V. & Kainov, D.E. (2014) Akt

Inhibitor MK2206 Prevents Influenza pH1N1 Virus Infection in Vitro. Antimicrobial Agents

and Chemotherapy, 58, 3689-3696.

Dennis, M. D., Schrufer, T.L., Bronson, S.K., Kimball, S.R. & Jefferson,

L.S. (2011) Hyperglycemia-Induced O-GlcNAcylation and Truncation of 4E-BP1 Protein in

Liver of a Mouse Model of Type 1 Diabetes. The Journal of Biological Chemistry, 286, 34286-

34297.

Dérijard, B., Hibi, M., Wu, I.H., Barrett, T., Su, B., Deng, T., Karin, M. & Davis,

R.J. (1994) JNK1: A Protein Kinase Stimulated by UV Light and Ha-Ras that Binds and

Phosphorylates the c-Jun Activation Domain. Cell, 76, 1025-1037.

Derijard, B., Raingeaud, J., Barrett, T., Wu, I.H., Han, J., Ulevitch, R.J. & Davis,

R.J. (1995) Independent Human MAP-Kinase Signal Transduction Pathways Defined by MEK

and MKK Isoforms. Science, 267, 682-685.

Dever, T. E., Chen, J.J., Barber, G.N., Cigan, A.M., Feng, L., Donahue, T.F., London, I.M.,

Katze, M.G. & Hinnebusch, A.G. (1993) Mammalian Eukaryotic Initiation Factor 2α kinases

Functionally Substitute for GCN2 Protein Kinase in the GCN4 Translational Control

Mechanism of Yeast. The Proceedings of the National Academy of Sciences of the United

States of America, 90, 4616-4620.

281

Dever, T. E., Feng, L., Wek, R.C., Cigan, A.M., Donahue, T.F. & Hinnebusch,

A.G. (1992) Phosphorylation of Initiation Factor 2α by Protein Kinase GCN2 Mediates Gene-

Specific Translational Control of GCN4 in Yeast. Cell, 68, 585-596.

Dholakia, J. N. & Wahba, A.J. (1989) Mechanism of the Nucleotide Exchange Reaction in

Eukaryotic Polypeptide Chain Initiation. Characterization of the Guanine Nucleotide

Exchange Factor as a GTP-Binding Protein. The Journal of Biological Chemistry, 264, 546-550.

Di Giacomo, V., Sancilio, S., Caravatta, L., Rana, R.A., Di Pietro, R. & Cataldi,

A. (2009) Regulation of CREB Activation by p38 Mitogen Activated Protein Kinase during

Human Primary Erythroblast Differentiation. International Journal of Immunopathology and

Pharmacology, 22, 679-688.

Di Pietro, R., di Giacomo, V., Caravatta, L., Sancilio, S., Rana, R.A. & Cataldi, A. (2007) Cyclic

Nucleotide Response Element Binding (CREB) Protein Activation is Involved in K562

Erythroleukemia Cells Differentiation. Journal of Cellular Biochemistry, 100, 1070-1079.

Di Ventura, B., Lemerle, C., Michalodimitrakis, K. & Serrano, L. (2006) From in Vivo to in

Silico Biology and Back. Nature, 443, 527-533.

Dias, A., Bouvier, D., Crepin, T., McCarthy, A.A., Hart, D.J., Baudin, F., Cusack, S. & Ruigrok,

R.W. (2009) The Cap-Snatching Endonuclease of Influenza Virus Polymerase Resides in the

PA Subunit. Nature, 458, 914-918.

Diaz-Meco, M.T., Lozano, J., Municio, M.M., Berra, E., Frutos, S., Sanz, L. & Moscat, J.

(1994) Evidence for the in Vitro and in Vivo Interaction of Ras with Protein Kinase C ζ. The


Dickinson, J. M., Fry, C.S., Drummond, M.J., Gundermann, D.M., Walker, D.K., Glynn, E.L.,

Timmerman, K.L., Dhanani, S., Volpi, E. & Rasmussen, B.B. (2011) Mammalian Target of

Rapamycin Complex 1 Activation is Required for the Stimulation of Human Skeletal Muscle

Protein Synthesis by Essential Amino Acids. The Journal of Nutrition, 141, 856-862.

Dietl, C. A., Wernly, J.A., Pett, S.B., Yassin, S.F., Sterling, J.P., Dragan, R., Milligan, K. &

Crowley, M.R. (2008) Extracorporeal Membrane Oxygenation Support Improves Survival of

Patients with Severe Hantavirus Cardiopulmonary Syndrome. The Journal of Thoracic and

Cardiovascular Surgery, 135, 579-584.

Dobrikov, M., Dobrikova, E., Shveygert, M. & Gromeier, M. (2011) Phosphorylation of

Eukaryotic Translation Initiation Factor 4G1 (eIF4GI) by Protein Kinase Cα Regulates eIF4GI

Binding to Mnk1. Molecular and Cellular Biology, 31, 2947-2959.

Dodd, K. M., Yang, J., Shen, M.H., Sampson, J.R. & Tee, A.R. (2014) mTORC1 Drives HIF-

1α and VEGF-A Signalling Via Multiple Mechanisms Involving 4E-BP1, S6K1 and

STAT3. Oncogene, 34, 1-12.

282

Donnelly, R. P., Loftus, R.M., Keating, S.E., Liou, K.T., Biron, C.A., Gardiner, C.M. & Finlay,

D.K. (2014) mTORC1-Dependent Metabolic Reprogramming is a Prerequisite for NK Cell

Effector Function. The Journal of Immunology, 193, 4477-4484.

Dorrello, N. V., Peschiaroli, A., Guardavaccaro, D., Colburn, N.H., Sherman, N.E. & Pagano,

M. (2006) S6K1- and βTRCP-Mediated Degradation of PDCD4 Promotes Protein Translation

and Cell Growth. Science, 314, 467-471.

Dowling, R. J. O., Topisirovic, I., Alain, T., Bidinosti, M., Fonseca, B.D., Petroulakis, E.,

Wang, X., Larsson, O., Selvaraj, A., Liu, Y., Kozma, S.C., Thomas, G. & Sonenberg, N. (2010)

mTORC1-Mediated Cell Proliferation, but Not Cell Growth, Controlled by the 4E-

BPs. Science, 328, 1172-1176.

Dudley, D. T., Pang, L., Decker, S.J., Bridges, A.J. & Saltiel, A.R. (1995) A Synthetic Inhibitor

of the Mitogen-Activated Protein Kinase Cascade. The Proceedings of the National Academy

of Sciences of the United States of America, 92, 7686-7689.

Duizer, E., Schwab, K.J., Neill, F.H., Atmar, R.L., Koopmans, M.P. & Estes,

M.K. (2004) Laboratory Efforts to Cultivate Noroviruses. The Journal of General Virology, 85,

79-87.

Dumitru, C.D., Ceci, J.D., Tsatsanis, C., Kontoyiannis, D., Stamatakis, K., Lin, J.-H., Patriotis,

C., Jenkins, N.A., Copeland, N.G., Kollias, G., Tsichlis, P.N. (2000) TNF-α Induction by LPS Is

Regulated Posttranscriptionally via a Tpl2/ERK-Dependent Pathway. Cell, 103, 1071-1083.

Duncan, R. F., Peterson, H., Hagedorn, C.H. & Sevanian, A. (2003) Oxidative Stress

Increases Eukaryotic Initiation Factor 4E Phosphorylation in Vascular Cells. The Biochemical

Journal, 369, 213-225.

Duncan, R. F., Peterson, H. & Sevanian, A. (2005) Signal Transduction Pathways Leading to

Increased eIF4E Phosphorylation Caused by Oxidative Stress. Free Radical Biology and

Medicine, 38, 631-643.

Dutil, E. M., Toker, A. & Newton, A.C. (1998) Regulation of Conventional Protein Kinase C

Isozymes by Phosphoinositide-Dependent Kinase 1 (PDK-1). Current Biology, 8, 1366-1375.

Düvel, K., Yecies, J.L., Menon, S., Raman, P., Lipovsky, A.I., Souza, A.L., Triantafellow, E.,

Ma, Q., Gorski, R., Cleaver, S., Vander Heiden, M.G., MacKeigan, J.P., Finan, P.M., Clish,

C.B., Murphy, L.O. & Manning, B.D. (2010) Activation of a Metabolic Gene Regulatory

Network Downstream of mTOR Complex 1. Molecular Cell, 39, 171-183.

Eckerdt, F., Beauchamp, E., Bell, J., Iqbal, A., Su, B., Fukunaga, R., Lulla, R.R., Goldman, S.

& Platanias, L.C. (2014) Regulatory Effects of a Mnk2-eIF4E Feedback Loop during mTORC1

Targeting of Human Medulloblastoma Cells. Oncotarget, 5, 8442-8451.

283

Edwards, J. S., Covert, M. & Palsson, B. (2002) Metabolic Modelling of Microbes: The Flux-

Balance Approach. Environmental Microbiology, 4, 133-140.

Eichelbaum, K. & Krijgsveld, J. (2014) Rapid Temporal Dynamics of Transcription, Protein

Synthesis, and Secretion during Macrophage Activation. Molecular & Cellular

Proteomics, 13, 792-810.

Elbahesh, H., Scherbik, S.V. & Brinton, M.A. (2011) West Nile Virus Infection does Not

Induce PKR Activation in Rodent Cells. Virology, 421, 51-60.

Eliopoulos, A.G., Wang, C.-C., Dumitru, C.D., Tsichlis, P.N. (2003) Tpl2 Transduces CD40 and

TNF Signals That Activate ERK and Regulates IgE Induction by CD40. The EMBO Journal, 22,

3855-3864.

Elliott, L. H., Ksiazek, T.G., Rollin, P.E., Spiropoulou, C.F., Morzunov, S., Monroe, M.,

Goldsmith, C.S., Humphrey, C.D., Zaki, S.R., Krebs, J.W., Maupin, G., Gage, K., Childs, J.E.,

Nichol, S.T. & Peters, C.J. (1994) Isolation of the Causative Agent of Hantavirus Pulmonary

Syndrome. The American Journal of Tropical Medicine and Hygiene, 51, 102-108.

Elowitz, M. B., Levine, A.J., Siggia, E.D. & Swain, P.S. (2002) Stochastic Gene Expression in a

Single Cell. Science, 297, 1183-1186.

Emily, M., Mailund, T., Hein, J., Schauser, L. & Schierup, M.H. (2009) Using Biological

Networks to Search for Interacting Loci in Genome-Wide Association Studies. European

Journal of Human Genetics, 17, 1231-1240.

Emonet, S. E., Urata, S. & de la Torre, J.C. (2011) Arenavirus Reverse Genetics: New

Approaches for the Investigation of Arenavirus Biology and Development of Antiviral

Strategies. Virology, 411, 416-425.

Emonet, S. F., de la Torre, J.C., Domingo, E. & Sevilla, N. (2009) Arenavirus Genetic Diversity

and its Biological Implications. Infection, Genetics and Evolution, 9, 417-429.

Enría, D., Padula, P., Segura, E.L., Pini, N., Edelstein, A., Posse, C.R. & Weissenbacher,

M.C. (1996) Hantavirus Pulmonary Syndrome in Argentina. Possibility of Person to Person

Transmission. Medicina, 56, 709-711.

Enria, D. A., Briggiler, A.M. & Sánchez, Z. (2008) Treatment of Argentine Hemorrhagic

Fever. Antiviral Research, 78, 132-139.

Erikson, E. & Maller, J.L. (1985) A Protein Kinase from Xenopus Eggs Specific for Ribosomal

Protein S6. The Proceedings of the National Academy of Sciences of the United States of

America, 82, 742-746.

Essbauer, S., Bremont, M. & Ahne, W. (2001) Comparison of the eIF-2α homologous

Proteins of Seven Ranaviruses (Iridoviridae). Virus Genes, 23, 347-359.

284

Esseili, M. A., Saif, L.J., Farkas, T. & Wang, Q. (2015) Feline Calicivirus, Murine Norovirus,

Porcine Sapovirus, and Tulane Virus Survival on Postharvest Lettuce. Applied and

Environmental Microbiology, 81, 5085-5092.

Fabian, J. R., Kimball, S.R., Heinzinger, N.K. & Jefferson, L.S. (1997) Subunit Assembly and

Guanine Nucleotide Exchange Activity of Eukaryotic Initiation Factor-2B Expressed in Sf9

Cells. The Journal of Biological Chemistry, 272, 12359-12365.

Fabian, J. R., Vojtek, A.B., Cooper, J.A. & Morrison, D.K. (1994) A Single Amino Acid Change

in Raf-1 Inhibits Ras Binding and Alters Raf-1 Function. The Proceedings of the National

Academy of Sciences of the United States of America, 91, 5982-5986.

Facchinetti, V., Ouyang, W., Wei, H., Soto, N., Lazorchak, A., Gould, C., Lowry, C., Newton,

A.C., Mao, Y., Sessa, W.C., Qin, J., Zhang, P., Su, B. & Jacinto, E. (2008) The Mammalian

Target of Rapamycin Complex 2 Controls Folding and Stability of Akt and Protein Kinase

C. The EMBO Journal, 27, 1932-1943.

Fang, Y., Scheibe, T.D., Mahadevan, R., Garg, S., Long, P.E. & Lovley, D.R. (2011) Direct

Coupling of a Genome-Scale Microbial in Silico Model and a Groundwater Reactive

Transport Model. Journal of Contaminant Hydrology, 122, 96-103.

Fankhauser, R. L., Monroe, S.S., Noel, J.S., Humphrey, C.D., Bresee, J.S., Parashar, U.D.,

Ando, T. & Glass, R.I. (2002) Epidemiologic and Molecular Trends of “Norwalk-Like Viruses”

Associated with Outbreaks of Gastroenteritis in the United States. The Journal of Infectious

Diseases, 186, 1-7.

Ferrer, I., Blanco, R., Carmona, M., Ribera, R., Goutan, E., Puig, B., Rey, M.J., Cardozo, A.,

Viñals, F. & Ribalta, T. (2001) Phosphorylated Map Kinase (ERK1, ERK2) Expression is

Associated with Early Tau Deposition in Neurones and Glial Cells, but Not with Increased

Nuclear DNA Vulnerability and Cell Death, in Alzheimer Disease, Pick's Disease, Progressive

Supranuclear Palsy and Corticobasal Degeneration. Brain Pathology, 11, 144-158.

Field, A. K., Tytell, A.A., Lampson, G.P. & Hilleman, M.R. (1967) Inducers of Interferon and

Host Resistance. II. Multistranded Synthetic Polynucleotide Complexes. The Proceedings of

the National Academy of Sciences of the United States of America, 58, 1004-1010.

Firczuk, H., Kannambath, S., Pahle, J., Claydon, A., Beynon, R., Duncan, J., Westerhoff, H.,

Mendes, P. & McCarthy, J.E. (2013) An in Vivo Control Map for the Eukaryotic mRNA

Translation Machinery. Molecular Systems Biology, 9, doi: 10.1038/msb.2012.73.

Fisher, A.K., Freedman, B.G., Bevan, D.R. & Senger R.S. (2014) A Review of Metabolic and

Enzymatic Engineering Strategies for Designing and Optimizing Performance of Microbial

Cell Factories. Computational and Structural Biotechnology Journal, 11, 91-99.

285

Fisher, C. P., Plant, N.J., Moore, J.B. & Kierzek, A.M. (2013) QSSPN: Dynamic Simulation of

Molecular Interaction Networks Describing Gene Regulation, Signalling and Whole-Cell

Metabolism in Human Cells. Bioinformatics, 29, 3181-3190.

Flanegan, J. B., Petterson, R.F., Ambros, V., Hewlett, N.J. & Baltimore, D. (1977) Covalent

Linkage of a Protein to a Defined Nucleotide Sequence at the 5'-Terminus of Virion and

Replicative Intermediate RNAs of Poliovirus. The Proceedings of the National Academy of


Fletcher, C. M., Pestova, T.V., Hellen, C.U. & Wagner, G. (1999) Structure and Interactions

of the Translation Initiation Factor eIF1. The EMBO Journal, 18, 2631-2637.

Fletcher, S. P. & Jackson, R.J. (2002) Pestivirus Internal Ribosome Entry Site (IRES) Structure

and Function: Elements in the 5' Untranslated Region Important for IRES Function. Journal

of Virology, 76, 5024-5033.

Frias, M. A., Thoreen, C.C., Jaffe, J.D., Schroder, W., Sculley, T., Carr, S.A. & Sabatini,

D.M. (2006) mSin1 is Necessary for Akt/PKB Phosphorylation, and its Isoforms Define Three

Distinct mTORC2s. Current Biology, 16, 1865-1870.

Fruman, D. A., Wood, M.A., Gjertson, C.K., Katz, H.R., Burakoff, S.J. & Bierer,

B.E. (1995) FK506 Binding Protein 12 Mediates Sensitivity to both FK506 and Rapamycin in

Murine Mast Cells. European Journal of Immunology, 25, 563-571.

Fu, Y., Foden, J.A., Khayter, C., Maeder, M.L., Reyon, D., Joung, J.K. & Sander,

J.D. (2013) High-Frequency Off-Target Mutagenesis Induced by CRISPR-Cas Nucleases in

Human Cells. Nature Biotechnology, 31, 822-826.

Fuentes, C., Bosch, A., Pintó, R.M. & Guix, S. (2012) Identification of Human Astrovirus

Genome-Linked Protein (VPg) Essential for Virus Infectivity. Journal of Virology, 86, 10070-

10078.

Fujii, W., Kawasaki, K., Sugiura, K. & Naito, K. (2013) Efficient Generation of Large-Scale

Genome-Modified Mice using gRNA and CAS9 Endonuclease. Nucleic Acids Research, 41,

e187.

Fukunaga, R. & Hunter, T. (1997) MNK1, A New MAP Kinase-Activated Protein Kinase,

Isolated by a Novel Expression Screening Method for Identifying Protein Kinase

Substrates. The EMBO Journal, 16, 1921-1933.

Furic, L., Rong, L., Koumakpayi, I.H., Yoshida, K., Brueschke, A., Petroulakis, E., Robichaud,

N., Pollak, M., Gaboury, L.A., Pandolfi, P.P., Saad, F. & Sonenberg, N. (2010) EIF4E

Phosphorylation Promotes Tumorigenesis and is Associated with Prostate Cancer

Progression. The Proceedings of the National Academy of Sciences of the United States of

America, 107, 14134-14139.

286

Galan-Moya, E. M., de la Cruz-Morcillo, M.A., Llanos Valero, M., Callejas-Valera, J.L.,

Melgar-Rojas, P., Hernadez Losa, J., Salcedo, M., Fernández-Aramburo, A., Ramon y Cajal,

S. & Sánchez-Prieto, R. (2011) Balance between MKK6 and MKK3 Mediates p38 MAPK

Associated Resistance to Cisplatin in NSCLC. PLoS ONE, 6, doi:

10.1371/journal.pone.0028406.

Gandin, V., Brina, D., Marchisio, P.C. & Biffo, S. (2010) JNK Inhibition Arrests

Cotranslational Degradation. Biochimica et Biophysica Acta, 1803, 826-831.

Ganiatsas, S., Kwee, L., Fujiwara, Y., Perkins, A., Ikeda, T., Labow, M.A. & Zon,

L.I. (1998) SEK1 Deficiency Reveals Mitogen-Activated Protein Kinase Cascade

Crossregulation and Leads to Abnormal Hepatogenesis. The Proceedings of the National

Academy of Sciences of the United States of America, 95, 6881-6886.

Gao, T., Toker, A. & Newton, A.C. (2001) The Carboxyl Terminus of Protein Kinase c

Provides a Switch to Regulate its Interaction with the Phosphoinositide-Dependent Kinase,

PDK-1. The Journal of Biological Chemistry, 276, 19588-19596.

Garami, A., Zwartkruis, F.J., Nobukuni, T., Joaquin, M., Roccio, M., Stocker, H., Kozma,

S.C., Hafen, E., Bos, J.L. & Thomas, G. (2003) Insulin Activation of Rheb, a Mediator of

mTOR/S6K/4E-BP Signaling, is Inhibited by TSC1 and 2. Molecular Cell, 11, 1457-1466.

García-Martínez, J. M., Moran, J., Clarke, R.G., Gray, A., Cosulich, S.C., Chresta, C.M. &

Alessi, D.R. (2009) Ku-0063794 is a Specific Inhibitor of the Mammalian Target of Rapamycin

(mTOR). The Biochemical Journal, 421, 29-42.

Garcin, D., Lezzi, M., Dobbs, M., Elliott, R.M., Schmaljohn, C., Kang, C.Y. & Kolakofsky,

D. (1995) The 5' Ends of Hantaan Virus (Bunyaviridae) RNAs Suggest a Prime-and-Realign

Mechanism for the Initiation of RNA Synthesis. Journal of Virology, 69, 5754-5762.

Garmaroudi, F. S., Marchant, D., Si, X., Khalili, A., Bashashati, A., Wong, B.W., Tabet, A.,

Ng, R.T., Murphy, K., Luo, H., Janes, K.A. & McManus, B.M. (2010) Pairwise Network

Mechanisms in the Host Signaling Response to Coxsackievirus B3 Infection. The Proceedings


Gassmann, M., Grenacher, B., Rohde, B. & Vogel, J. (2009) Quantifying Western Blots:

Pitfalls of Densitometry. Electrophoresis, 30, 1845-1855.

Gavin, A. C. & Nebreda, A.R. (1999) A MAP Kinase Docking Site is Required for

Phosphorylation and Activation of p90(Rsk)/MAPKAP Kinase-1. Current Biology, 9, 281-284.

Gavrilovskaya, I. N., Gorbunova, E.E. & Mackow, E.R. (2013) Hypoxia Induces Permeability

and Giant Cell Responses of Andes Virus-Infected Pulmonary Endothelial Cells by Activating

the mTOR-S6K Signaling Pathway. Journal of Virology, 87, 12999-13008.

287

Geiger, T. R., Sharma, N., Kim, Y.M. & Nyborg, J.K. (2008) The Human T-Cell Leukemia Virus

Type 1 Tax Protein Confers CBP/p300 Recruitment and Transcriptional Activation Properties

to Phosphorylated CREB. Molecular and Cellular Biology, 28, 1383-1392.

Gentile, S., Darden, T., Erxleben, C., Romeo, C., Russo, A., Martin, N., Rossie, S. &

Armstrong, D.L. (2006) Rac GTPase Signaling through the PP5 Protein Phosphatase. The

Proceedings of the National Academy of Sciences of the United States of America, 103, 5202-

5206.

Ghiringhelli, P. D., Rivera-Pomar, R.V., Lozano, M.E., Grau, O. & Romanowski,

V. (1991) Molecular Organization of Junin Virus S RNA: Complete Nucleotide Sequence,

Relationship with Other Members of the Arenaviridae and Unusual Secondary

Structures. The Journal of General Virology, 72, 2129-2141.

Gibellini, D., Bassini, A., Pierpaoli, S., Bertolaso, L., Milani, D., Capitani, S., La Placa, M. &

Zauli, G. (1998) Extracellular HIV-1 Tat Protein Induces the Rapid Ser133 Phosphorylation

and Activation of CREB Transcription Factor in both Jurkat Lymphoblastoid T Cells and

Primary Peripheral Blood Mononuclear Cells. The Journal of Immunology, 160, 3891-3898.

Gibson, M. A. & Bruck, J. (2000) Efficient Exact Stochastic Simulation of Chemical Systems

with Many Species and Many Channels. The Journal of Physical Chemistry A, 104, 1876-

1889.

Gille, C., Bölling, C., Hoppe, A., Bulik, S., Hoffmann, S., Hübner, K., Karlstädt, A., Ganeshan,

R., König, M., Rother, K., Weidlich, M., Behre, J. & Holzhütter, H.G. (2010) HepatoNet1: A

Comprehensive Metabolic Reconstruction of the Human Hepatocyte for the Analysis of Liver

Physiology. Molecular Systems Biology, 6, doi: 10.1038/msb.2010.62.

Gille, H., Kortenjann, M., Thomae, O., Moomaw, C., Slaughter, C., Cobb, M.H. & Shaw,

P.E. (1996) ERK Phosphorylation Potentiates Elk-1-Mediated Ternary Complex Formation

and Transactivation. The EMBO Journal, 14, 951-962.

Gillespie, D.T. (1977) Exact Stochastic Simulation of Coupled Chemical Reactions. The

Journal of Physical Chemistry, 81, 2340-2361.

Gillespie, D.T. (2001) Approximate Accelerated Stochastic Simulation of Chemically Reacting

Systems. The Journal of Chemical Physics, 115, 1716-1733.

Gillespie, D.T. (2007) Stochastic Simulation of Chemical Kinetics. Annual Review of Physical

Chemistry, 58, 35-55.

Gilbert, D., Heiner, M. & Lehrack, S. (2007) A Unifying Framework for Modelling and

Analysing Biochemical Pathways using Petri Nets. Computational Methods in Systems

Biology, 4695, 200-216.

288

Gingras, A.-C., Kennedy, S.G., O'Leary, M.A., Sonenberg, N. & Hay, N. (1998) 4E-BP1, a

Repressor of mRNA Translation, is Phosphorylated and Inactivated by the Akt(PKB) Signaling

Pathway. Genes & Development, 12, 502-513.

Gingras, A. C., Gygi, S.P., Raught, B., Polakiewicz, R.D., Abraham, R.T., Hoekstra, M.F.,

Aebersold, R. & Sonenberg, N. (1999) Regulation of 4E-BP1 Phosphorylation: A Novel Two-

Step Mechanism. Genes & Development, 13, 1422-1437.

Gingras, A. C., Raught, B., Gygi, S.P., Niedzwiecka, A., Miron, M., Burley, S.K., Polakiewicz,

R.D., Wyslouch-Cieszynska, A., Aebersold, R. & Sonenberg, N. (2001) Hierarchical

Phosphorylation of the Translation Inhibitor 4E-BP1. Genes & Development, 15, 2852-2864.

Glass, R. I., Parashar, U.D. & Estes, M.K. (2009) Norovirus Gastroenteritis. The New England

Journal of Medicine, 361, 1776-1785.

Godoy, P., Marsac, D., Stefas, E., Ferrer, P., Tischler, N.D., Pino, K., Ramdohr, P., Vial, P.,

Valenzuela, P.D., Ferrés, M., Veas, F. & López-Lastra, M. (2009) Andes Virus Antigens are

Shed in Urine of Patients with Acute Hantavirus Cardiopulmonary Syndrome. Journal of

Virology, 83, 5046-5055.

Goedert, M., Cuenda, A., Craxton, M., Jakes, R. & Cohen, P. (1997) Activation of the Novel

Stress-Activated Protein Kinase SAPK4 by Cytokines and Cellular Stresses is Mediated by

SKK3 (MKK6); Comparison of its Substrate Specificity with that of Other SAP Kinases. The

EMBO journal, 16, 3563-3571.

Gonzalez Della Valle, M., Edelstein, A., Miguel, S., Martinez, V., Cortez, J., Cacace, M.L.,

Jurgelenas, G., Sosa Estani, S. & Padula, P. (2002) Andes Virus Associated with Hantavirus

Pulmonary Syndrome in Northern Argentina and Determination of the Precise Site of

Infection. The American Journal of Tropical Medicine and Hygiene, 66, 713-720.

Goodfellow, I. G., Chaudhry, Y., Gioldasi, I., Gerondopoulos, A., Natoni, A., Labrie, L.,

Laliberté, J.F. & Roberts, L. (2005) Calicivirus Translation Initiation Requires an Interaction

between VPg and eIF4E. EMBO Reports, 6, 968-972.

Göpfert, U., Kullmann, M. & Hengst, L. (2003) Cell Cycle-Dependent Translation of p27

Involves a Responsive Element in its 5'-UTR that Overlaps with a uORF. Human Molecular

Genetics, 12, 1767-1779.

Gordan, J. D., Thompson, C.B. & Simon, M.C. (2007) HIF and c-Myc: Sibling Rivals for

Control of Cancer Cell Metabolism and Proliferation. Cancer Cell, 12, 108-113.

Görlach, M., Burd, C.G. & Dreyfuss, G. (1994) The mRNA Poly(A)-Binding Protein:

Localization, Abundance, and RNA-Binding Specificity. Experimental Cell Research, 211, 400-

407.

289

Goss, D. J., Carberry, S.E., Dever, T.E., Merrick, W.C. & Rhoads, R.E. (1990) Fluorescence

Study of the Binding of m7GpppG and Rabbit Globin mRNA to Protein Synthesis Initiation

Factors 4A, 4E, and 4F. Biochemistry, 29, 5008-5012.

Gowthaman, U. & Agrewala, J.N. (2008) in Silico Tools for Predicting Peptides Binding to

HLA-Class II Molecules: More Confusion than Conclusion. Journal of Proteome Research, 7,

154-163.

Graber, T. E., McCamphill, P.K. & Sossin, W.S. (2013) A Recollection of mTOR Signaling in

Learning and Memory. Learning & Memory, 20, 518-530.

Gradi, A., Imataka, H., Svitkin, Y.V., Rom, E., Raught, B., Morino, S. & Sonenberg,

N. (1998) A Novel Functional Human Eukaryotic Translation Initiation Factor 4G. Molecular

and Cellular Biology, 18, 334-342.

Grady, S. L., Purdy, J.G., Rabinowitz, J.D. & Shenk, T. (2013) Argininosuccinate Synthetase 1

Depletion Produces a Metabolic State Conducive to Herpes Simplex Virus 1 Infection. The

Proceedings of the National Academy of Sciences of the United States of America, 110,

E5006-E5015.

Graff, J. R., Konicek, B.W., Carter, J.H. & Marcusson, E.G. (2008) Targeting the Eukaryotic

Translation Initiation Factor 4E for Cancer Therapy. Cancer Research, 68, 631-634.

Grayfer, L., De Jesús Andino, F. & Robert, J. (2015) Prominent Amphibian (Xenopus laevis)

Tadpole Type III Interferon Response to the Frog Virus 3 Ranavirus. Journal of Virology, 89,

5072-5082.

Gratton, J.P., Morales-Ruiz, M., Kureishi, Y., Fulton, D., Walsh, K. & Sessa, W.C. (2001) Akt

Down-Regulation of p38 Signaling Provides a Novel Mechanism of Vascular Endothelial

Growth Factor-Mediated Cytoprotection in Endothelial Cells. The Journal of Biological

Chemistry, 276, 30359-30365.

Green, K. Y., Belliot, G., Taylor, J.L., Valdesuso, J., Lew, J.F., Kapikian, A.Z. & Lin,

F.Y. (2002) A Predominant Role for Norwalk-Like Viruses as Agents of Epidemic

Gastroenteritis in Maryland Nursing Homes for the Elderly. The Journal of Infectious

Diseases, 185, 133-146.

Green, A. J., Smith, M. & Yates, J.R. (1994) Loss of Heterozygosity on Chromosome 16p13.3

in Hamartomas from Tuberous Sclerosis Patients. Nature Genetics, 6, 193-196.

Green, R. E., Braun, E.L., Armstrong, J., Earl, D., Nguyen, N., Hickey, G., Vandewege, M.W.,

St. John, J.A., Capella-Gutiérrez, S., Castoe, T.A., Kern, C., Fujita, M.K., Opazo, J.C., Jurka, J.,

Kojima, K.K., Caballero, J., Hubley, R.M., Smit, A.F., Smit, A.F., Lavoie, C.A., Ramakodi,

M.P., Finger Jr., J.W., Suh, A., Isberg, R., Miles, L., Chong, Y., Jaratlerdsiri, W., Gongora, J.,

Moran, C., Iriarte, A., McCormack, J., Burgess, S.C., Edwards, S.V., Lyons, E., Williams, C.,

290

Breen, M., Howard, J.T., Gresham, C.R., Peterson, D.G., Schmitz, J., Pollock, D.D., Haussler,

D., Triplett, E.W., Zhang, G., Irie, N., Jarvis, E.D., Brochu, C.A., Schmidt, C.J., McCarthy,

F.M., Faircloth, B.C., Hoffmann, F.G., Glenn, T.C., Gabaldón, T., Paten, B. & Ray,

D.A. (2014) Three Crocodilian Genomes Reveal Ancestral Patterns of Evolution among

Archosaurs. Science, 346, doi:10.1126/science.1254449.

Greene, M. W., Sakaue, H., Wang, L., Alessi, D.R. & Roth, R.A. (2003) Modulation of Insulin-

Stimulated Degradation of Human Insulin Receptor Substrate-1 by Serine 312

Phosphorylation. The Journal of Biological Chemistry, 278, 8199-8211.

Groenewoud, M. J. & Zwartkruis, F.J. (2013) Rheb and Rags Come Together at the

Lysosome to Activate mTORC1. Biochemical Society Transactions, 41, 951-955.

Groisman, I., Jung, M.Y., Sarkissian, M., Cao, Q. & Richter, J.D. (2002) Translational Control

of the Embryonic Cell Cycle. Cell, 109, 473-483.

Grosso, S., Pesce, E., Brina, D., Beugnet, A., Loreni, F. & Biffo, S. (2011) Sensitivity of Global

Translation to mTOR Inhibition in REN Cells Depends on the Equilibrium between eIF4E and

4E-BP1. PLoS ONE, 6, doi: 10.1371/journal.pone.0029136.

Gual, P., Le Marchand-Brustel, Y. & Tanti, J.-F. (2005) Positive and Negative Regulation of

Insulin Signalling Through IRS-1 Phosphorylation. Biochimie, 87, 99-109.

Guertin, D. A., Stevens, D.M., Thoreen, C.C., Burds, A.A., Kalaany, N.Y., Moffat, J., Brown,

M., Fitzgerald, K.J. & Sabatini, D.M. (2006) Ablation in Mice of the mTORC Components

Raptor, Rictor, Or mLST8 Reveals that mTORC2 is Required for Signaling to Akt-FOXO and

PKCα, but Not S6K1. Developmental Cell, 11, 859-871.

Guha, M., O'Connell, M.A., Pawlinski, R., Hollis, A., McGovern, P., Yan, S.-F., Stern, D. &

Mackman, N. (2001) Lipopolysaccharide Activation of the MEK-ERK1/2 Pathway in Human

Monocytic Cells Mediates Tissue Factor and Tumor Necrosis Factor α Expression by Inducing

Elk-1 Phosphorylation and Egr-1 Expression. Blood, 98, 1429-1439.

Guil, S., Long, J.C. & Caceres, J.F. (2006) HnRNP A1 Relocalization to the Stress Granules

Reflects a Role in the Stress Response. Molecular and Cellular Biology, 26, 5744-5758.

Guillard, C., Chrétien, S., Pelus, A.S., Porteu, F., Muller, O., Mayeux, P. & Duprez, V. (2003)

Activation of the Mitogen-Activated Protein Kinases Erk1/2 by Erythropoietin Receptor via a

Gi Protein βγ-Subunit-initiated Pathway. The Journal of Biological Chemistry, 278, 11050-

11056.

Gunn, R.A., Terranova, W.A., Greenberg, H.B., Yashuk, J., Gary, G.W., Wells, J.G., Taylor,

P.R. & Feldman, R.A. (1980) Norwalk Virus Gastroenteritis Aboard a Cruise Ship: An

Outbreak on Five Consecutive Cruises. American Journal of Epidemiology, 112, 820-827.

291

Habjan, M., Pichlmair, A., Elliott, R.M., Överby, A.K., Glatter, T., Gstaiger, M., Superti-

Furga, G., Unger, H. & Weber, F. (2009) NSs Protein of Rift Valley Fever Virus Induces the

Specific Degradation of the Double-Stranded RNA-Dependent Protein Kinase. Journal of

Virology, 83, 4365-4375.

Hall, B., Limaye, A. & Kulkarni, A.B. (2009) Overview: Generation of Gene Knockout

Mice. Current Protocols in Cell Biology, Chapter 19, Unit 19.12 19.12.1-17.

Han, A. -P., Yu, C., Lu, L., Fujiwara, Y., Browne, C., Chin, G., Fleming, M., Leboulch, P.,

Orkin, S.H. & Chen, J.-J. (2001) Heme-Regulated eIF2α Kinase (HRI) is Required for

Translational Regulation and Survival of Erythroid Precursors in Iron Deficiency. The EMBO

Journal, 20, 6909-6918.

Hanks, S. K. & Hunter, T. (1995) Protein Kinases 6. the Eukaryotic Protein Kinase

Superfamily: Kinase (Catalytic) Domain Structure and Classification. FASEB Journal, 9, 576-

596.

Hao, S., Sharp, J.W., Ross-Inta, C.M., McDaniel, B.J., Anthony, T.G., Wek, R.C., Cavener,

D.R., McGrath, B.C., Rudell, J.B., Koehnle, T.J. & Gietzen, D.W. (2005) Uncharged tRNA and

Sensing of Amino Acid Deficiency in Mammalian Piriform Cortex. Science, 307, 1776-1778.

Hara, K., Maruki, Y., Long, X., Yoshino, K.-I., Oshiro, N., Hidayat, S., Tokunaga, C., Avruch, J.

& Yonezawa, K. (2002) Raptor, a Binding Partner of Target of Rapamycin (TOR), Mediates

TOR Action. Cell, 110, 177-189.

Harding, H. P., Zhang, Y. & Ron, D. (1999) Protein Translation and Folding are Coupled by an

Endoplasmic-Reticulum-Resident Kinase. Nature, 397, 271-274.

Harms, U., Andreou, A.Z., Gubaev, A. & Klostermeier, D. (2014) eIF4B, eIF4G and RNA

Regulate eIF4A Activity in Translation Initiation by Modulating the eIF4A Conformational

Cycle. Nucleic Acids Research, 42, 7911-7922.

Harris, J. P., Edmunds, W.J., Pebody, R., Brown, D.W. & Lopman, B.A. (2008) Deaths from

Norovirus among the Elderly, England and Wales. Emerging Infectious Diseases, 14, 1546-

1552.

Harrison, L. H., Halsey, N.A., McKee, K.T.,Jr, Peters, C.J., Barrera Oro, J.G., Briggiler, A.M.,

Feuillade, M.R. & Maiztegui, J.I. (1999) Clinical Case Definitions for Argentine Hemorrhagic

Fever. Clinical Infectious Diseases, 28, 1091-1094.

Hartley, D. & Cooper, G.M. (2002) Role of mTOR in the Degradation of IRS-1: Regulation of

PP2A Activity. Journal of Cellular Biochemistry, 85, 304-314.

Hartmann, T., Xu, X., Kronast, M., Muehlich, S., Meyer, K., Zimmermann, W., Hurwitz, J.,

Pan, Z.Q., Engelhardt, S. & Sarikas, A. (2014) Inhibition of Cullin-RING E3 Ubiquitin Ligase 7

292

by Simian Virus 40 Large T Antigen. The Proceedings of the National Academy of Sciences of

the United States of America, 111, 3371-3376.

Hartwell, L. H., Hopfield, J.J., Leibler, S. & Murray, A.W. (1999) From Molecular to Modular

Cell Biology. Nature, 402, C47-C52.

Haruta, T., Uno, T., Kawahara, J., Takano, A., Egawa, K., Sharma, P.M., Olefsky, J.M. &

Kobayashi, M. (2000) A Rapamycin-Sensitive Pathway Down-Regulates Insulin Signaling Via

Phosphorylation and Proteasomal Degradation of Insulin Receptor Substrate-1. Molecular


Hashem, Y., des Georges, A., Dhote, V., Langlois, R., Liao, H.Y., Grassucci, R.A., Hellen,

C.U., Pestova, T.V. & Frank, J. (2013) Structure of the Mammalian Ribosomal 43S

Preinitiation Complex Bound to the Scanning Factor DHX29. Cell, 153, 1108-1119.

Hatzikirou, H., Deutsch, A., Schaller, C., Simon, M. & Swanson, K. (2005) Mathematical

Modelling of Glioblastoma Tumour Development: A Review. Mathematical Models and

Methods in Applied Sciences, 15, 1779-1794.

Hébrard, E., Poulicard, N., Gérard, C., Traoré, O., Wu, H.C., Albar, L., Fargette, D., Bessin, Y.

& Vignols, F. (2010) Direct Interaction between the Rice Yellow Mottle Virus (RYMV) VPg

and the Central Domain of the Rice eIF(Iso)4G1 Factor Correlates with Rice Susceptibility

and RYMV Virulence. Molecular Plant-Microbe Interactions, 23, 1506-1513.

Heikkinen, T., Korpela, T., Fagerholm, R., Khan, S., Aittomäki, K., Heikkilä, P., Blomqvist, C.,

Carpén, O. & Nevanlinna, H. (2013) Eukaryotic Translation Initiation Factor 4E (eIF4E)

Expression is Associated with Breast Cancer Tumor Phenotype and Predicts Survival After

Anthracycline Chemotherapy Treatment. Breast Cancer Research and Treatment, 141, 79-

88.

Heinemann, P., Schmidt-Chanasit, J. & Günther, S. (2013) The N Terminus of Andes Virus L

Protein Suppresses mRNA and Protein Expression in Mammalian Cells. Journal of

Virology, 87, 6975-6985.

Heiner, M., Herajy, M., Liu, F., Rohr, C., Schwarick, M. (2012) Snoopy – A Unifying Petri Net

Tool. Computer Science Department, Brandenberg University of Technology, DE.

Heiner, M., Schwarick, M & Wegener, J.-T. (2015) ‘Charlie – An Extensible Petri Net Analysis

Tool’ in Devillers, R & Valmari, A. (Eds.) Application and Theory of Petri Nets and

Concurrency. 36th International Conference, Petri Nets 2015, Brussels (BE).

Heitman, J., Movva, N.R. & Hall, M.N. (1991) Targets for Cell Cycle Arrest by the

Immunosuppressant Rapamycin in Yeast. Science, 253, 905-909.

Helliwell, P. A., Rumsby, M.G. & Kellett, G.L. (2003) Intestinal Sugar Absorption is

Regulated by Phosphorylation and Turnover of Protein Kinase C βII mediated by

293

Phosphatidylinositol 3-Kinase- and Mammalian Target of Rapamycin-Dependent

Pathways. The Journal of Biological Chemistry, 278, 28644-28650.

Hemmings, B.A. & Restuccia, D.F. (2015) The PI3K-PKB/AKT Pathway. Cold Springs Harbor

Perspectives in Biology, 7, doi: 10.1101/cshperspect.a026609.

Henry, C.S., DeJongh, M., Best, A.A., Frybarger, P.M., Linsay, B. & Stevens, R.L. (2010)

High-Throughput Generation, Optimization and Analysis of Genome-Scale Metabolic

Models. Nature Biotechnology, 28, 977-982.

Herbreteau, C. H., Weill, L., Décimo, D., Prévôt, D., Darlix, J.L., Sargueil, B. & Ohlmann,

T. (2005) HIV-2 Genomic RNA Contains a Novel Type of IRES Located Downstream of its

Initiation Codon. Nature Structural & Molecular Biology, 12, 1001-1007.

Herdy, B., Jaramillo, M., Svitkin, Y.V., Rosenfeld, A.B., Kobayashi, M., Walsh, D., Alain, T.,

Sean, P., Robichaud, N., Topisirovic, I., Furic, L., Dowling, R.J., Sylvestre, A., Rong, L.,

Colina, R., Costa-Mattioli, M., Fritz, J.H., Olivier, M., Brown, E., Mohr, I. & Sonenberg,

N. (2012) Translational Control of the Activation of Transcription Factor NF-κB and

Production of Type I Interferon by Phosphorylation of the Translation Factor eIF4E. Nature

Immunology, 13, 543-550.

Hermann, C., Assmus, B., Urbich, C., Zeiher, A.M. & Dimmeler, S. (2000) Insulin-Mediated

Stimulation of Protein Kinase Akt: A Potent Survival Signaling Cascade for Endothelial

Cells. Arteriosclerosis, Thrombosis, and Vascular Biology, 20, 402-409.

Hernández, G., del Mar Castellano, M., Agudo, M. & Sierra, J.M. (1998) Isolation and

Characterization of the cDNA and the Gene for Eukaryotic Translation Initiation Factor 4G

from Drosophila melanogaster. European Journal of Biochemistry, 253, 27-35.

Hindmarsh, A.C., Brown, P.N., Grant, K.E., Lee, S.L., Serban, R., Shumaker, D.E. &

Woodward, C.S. (2005) SUNDIALS: Suite of Nonlinear and Differential/Algebraic Equation

Solvers. ACM Transactions on Mathematical Software, 31, 363-396

Hirschhorn, J., Mohanty, S. & Bhat, N.R. (2014) The Role of Tumor Progression Locus 2

Protein Kinase in Glial Inflammatory Response. Journal of Neurochemistry, 128, 919-926.

Hofmann, S., Cherkasova, V., Bankhead, P., Bukau, B. & Stoecklin, G. (2012) Translation

Suppression Promotes Stress Granule Formation and Cell Survival in Response to Cold

Shock. Molecular Biology of the Cell, 23, 3786-3800.

Holm, S. (1979) A Simple Sequentially Rejective Multiple Test Procedure. Scandinavian

Journal of Statistics, 6, 65-70.

Hong, K., Lou, L., Gupta, S., Ribeiro-Neto, F. & Altschuler, D.L. (2008) A Novel Epac-Rap-

PP2A Signaling Module Controls cAMP-Dependent Akt Regulation. The Journal of Biological

Chemistry, 283, 23129-23138.

294

Hoops, S., Sahle, S., Gauges, R., Lee, C., Pahle, J., Simus, N., Singhal, M., Xu, L., Mendes, P.

& Kummer, U. (2006) Copasi - A Complex Pathway Simulator. Bioinformatics, 22, 3067-

3074.

Hoppe, A., Hoffmann, S. & Holzhütter, H.G. (2007) Including Metabolite Concentrations

into Flux Balance Analysis: Thermodynamic Realizability as a Constraint on Flux Distributions

in Metabolic Networks. BMC Systems Biology, 1, doi:10.1186/1752-0509-1-23.

Hoshino, K., Takeuchi, O., Kawai, T., Sanjo, H., Ogawa, T., Takeda, Y., Takeda, K. & Akira,

S. (1999) Cutting Edge: Toll-Like Receptor 4 (TLR4)-Deficient Mice are Hyporesponsive to

Lipopolysaccharide: Evidence for TLR4 as the Lps Gene Product. The Journal of

Immunology, 162, 3749-3752.

Hosmillo, M., Chaudhry, Y., Kim, D.S., Goodfellow, I. & Cho, K.O. (2014) Sapovirus

Translation Requires an Interaction between VPg and the Cap Binding Protein eIF4E. Journal

of Virology, 88, 12213-12221.

Hou, Z., Zhang, Y., Propson, N.E., Howden, S.E., Chu, L.F., Sontheimer, E.J. & Thomson,

J.A. (2013) Efficient Genome Engineering in Human Pluripotent Stem Cells using Cas9

from Neisseria meningitidis. The Proceedings of the National Academy of Sciences of the


Housden, B.E. & Perrimon, N. (2014) Spatial and Temporal Organization of Signaling

Pathways. Trends in Biochemical Sciences, 39, 457-464.

Hresko, R. C. & Mueckler, M. (2005) mTOR·RICTOR is the Ser473 Kinase for Akt/protein

Kinase B in 3T3-L1 Adipocytes. The Journal of Biological Chemistry, 280, 40406-40416.

Hsieh, C.C. & Papaconstantinou, J. (2002) The Effect of Aging on p38 Signaling Pathway

Activity in the Mouse Liver and in Response to ROS Generated by 3-Nitropropionic Acid.

Mechanisms of Aging and Development, 123, 1423-1435.

Hsu, P. D., Scott, D.A., Weinstein, J.A., Ran, F.A., Konermann, S., Agarwala, V., Li, Y., Fine,

E.J., Wu, X., Shalem, O., Cradick, T.J., Marraffini, L.A., Bao, G. & Zhang, F. (2013) DNA

Targeting Specificity of RNA-Guided Cas9 Nucleases. Nature Biotechnology, 31, 827-832.

Hsu, P. P., Kang, S.A., Rameseder, J., Zhang, Y., Ottina, K.A., Lim, D., Peterson, T.R., Choi,

Y., Gray, N.S., Yaffe, M.B., Marto, J.A. & Sabatini, D.M. (2011) The mTOR-Regulated

Phosphoproteome Reveals a Mechanism of mTORC1-Mediated Inhibition of Growth Factor

Signaling. Science, 332, 1317-1322.

Hu, M. C., Wang, Y.P., Mikhail, A., Qiu, W.R. & Tan, T.H. (1999) Murine p38δ mitogen-

Activated Protein Kinase, a Developmentally Regulated Protein Kinase that is Activated by

Stress and Proinflammatory Cytokines. The Journal of Biological Chemistry, 274, 7095-7102.

295

Huang, S., Shu, L., Dilling, M.B., Easton, J., Harwood, F.C., Ichijo, H. & Houghton,

P.J. (2003) Sustained Activation of the JNK Cascade and Rapamycin-Induced Apoptosis are

Suppressed by p53/p21(Cip1). Molecular Cell, 11, 1491-1501.

Huang, S., Shu, L., Easton, J., Harwood, F.C., Germain, G.S., Ichijo, H. & Houghton,

P.J. (2004) Inhibition of Mammalian Target of Rapamycin Activates Apoptosis Signal-

Regulating Kinase 1 Signaling by Suppressing Protein Phosphatase 5 Activity. The Journal of


Hucka, M., Finney, A., Sauro, H.M., Bolouri, H., Doyle, J.C., Kitano, H., Arkin, A.P.,

Bornstein, B.J., Bray, D., Cornish-Bowden, A., Cuellar, A.A., Dronov, S., Gilles, E.D., Ginkel,

M., Gor, V., Goryanin, I.I., Hedley, W.J., Hodgman, T.C., Hofmeyr, J.H., Hunter, P.J., Juty,

N.S., Kasberger, J.L., Kremling, A., Kummer, U., Le Novère, N., Loew, L.M., Lucio, D.,

Mendes, P., Minch, E., Mjolsness, E.D., Nakayama, Y., Nelson, M.R., Nielsen, P.F.,

Sakurada, T., Schaff, J.C., Shapiro, B.E., Shimizu, T.S., Spence, H.D., Stelling, J., Takahashi,

K., Tomita, M., Wagner, J., Wang, J. & SBML Forum (2003) The Systems Biology Markup

Language (SBML): A Medium for Representation and Exchange of Biochemical Network

Models. Bioinformatics, 19, 524-531.

Hui, L., Pei, D.S., Zhang, Q.G., Guan, Q.H. & Zhang, G.Y. (2005) The Neuroprotection of

Insulin on Ischemic Brain Injury in Rat Hippocampus through Negative Regulation of JNK

Signaling Pathway by PI3K/Akt Activation. Brain Research, 1052, 1-9.

Hurley, K. F., Pesavento, P.A., Pedersen, N.C., Poland, A.M., Wilson, E. & Foley,

J.E. (2004) An Outbreak of Virulent Systemic Feline Calicivirus Disease. Journal of the

American Veterinary Medical Association, 224, 241-249.

Hussain, T., Llácer, J.L., Fernández, I.S., Munoz, A., Martin-Marcos, P., Savva, C.G., Lorsch,

J.R., Hinnebusch, A.G. & Ramakrishnan, V. (2014) Structural Changes Enable Start Codon

Recognition by the Eukaryotic Translation Initiation Complex. Cell, 159, 597-607.

Ideker, T., Galitski, T. & Hood, L. (2001) A New Approach to Decoding Life: Systems

Biology. Annual Review of Genomics and Human Genetics, 2, 343-372.

Ikegami, T., Narayanan, K., Won, S., Kamitani, W., Peters, C.J. & Makino, S. (2009a) Rift

Valley Fever Virus NSs Protein Promotes Post-Transcriptional Downregulation of Protein

Kinase PKR and Inhibits eIF2α phosphorylation. PLoS Pathogens, 5, doi:

10.1371/journal.ppat.1000287.

Ikenoue, T., Inoki, K., Yang, Q., Zhou, X. & Guan, K.-L. (2008) Essential Function of TORC2 in

PKC and Akt Turn Motif Phosphorylation, Maturation and Signalling. The EMBO Journal, 27,

1919-1931.

Ikezoe, T., Nishioka, C., Bandobashi, K., Yang, Y., Kuwayama, Y., Adachi, Y., Takeuchi, T.,

Koeffler, H.P. & Taguchi, H. (2007) Longitudinal Inhibition of PI3K/Akt/mTOR Signaling by

296

LY294002 and Rapamycin Induces Growth Arrest of Adult T-Cell Leukemia Cells. Leukemia

Research, 31, 673-682.

Imataka, H., Gradi, A. & Sonenberg, N. (1998) A Newly Identified N-Terminal Amino Acid

Sequence of Human eIF4G Binds Poly(A)-Binding Protein and Functions in Poly(A)-

Dependent Translation. The EMBO Journal, 17, 7480-7489.

Impey, S., Obrietan, K., Wong, S.T., Poser, S., Yano, S., Wayman, G., Deloulme, J.C., Chan,

G. & Storm, D.R. (1998) Cross Talk between ERK and PKA is Required for Ca2+ Stimulation of

CREB-Dependent Transcription and ERK Nuclear Translocation. Neuron, 21, 869-883.

Inoki, K., Li, T., Zhu, T., Wu, J. & Guan, K.-L. (2002) TSC2 is Phosphorylated and Inhibited by

Akt and Suppresses mTOR Signalling. Nature Cell Biology, 4, 648-657.

Inoki, K., Li, Y., Xu, T. & Guan, L.-K. (2003) Rheb GTPase is a Direct Target of TSC2 GAP

Activity and Regulates mTOR Signaling. Genes & Development, 17, 1829-1834.

Isakbaeva, E. T., Widdowson, M.-A., Beard, R.S., Bulens, S.N., Mullins, J., Monroe, S.S.,

Bresee, J., Sassano, P., Cramer, E.H. & Glass, R.I. (2005) Norovirus Transmission on Cruise

Ship. Emerging Infectious Diseases, 11, 154-157.

Isotani, S., Hara, K., Tokunaga, C., Inoue, H., Avruch, J. & Yonezawa,

K. (1999) Immunopurified Mammalian Target of Rapamycin Phosphorylatesand Activates

p70 S6 Kinase α in Vitro. The Journal of Biological Chemistry, 274, 34493-34498.

Ito, M., Kitamura, H., Kikuguchi, C., Hase, K., Ohno, H. & Ohara, O. (2011) SP600125

Inhibits Cap-dependent Translation Independently of the c-Jun N-terminal Kinase Pathway.

Cell Structure and Function, 36, 27-33.

Jääskeläinen, K. M., Kaukinen, P., Minskaya, E.S., Plyusnina, A., Vapalahti, O., Elliott, R.M.,

Weber, F., Vaheri, A. & Plyusnin, A. (2007) Tula and Puumala Hantavirus NSs ORFs are

Functional and the Products Inhibit Activation of the Interferon-Beta Promoter. Journal of

Medical Virology, 79, 1527-1536.

Jacinto, E., Facchinetti, V., Liu, D., Soto, N., Wei, S., Jung, S.Y., Huang, Q., Qin, J. & Su,

B. (2006) SIN1/MIP1 Maintains Rictor-mTOR Complex Integrity and Regulates Akt

Phosphorylation and Substrate Specificity. Cell, 127, 125-137.

Jackson, R. J., Hellen, C.U.T. & Pestova, T.V. (2010) The Mechanism of Eukaryotic

Translation Initiation and Principles of its Regulation. Nature Reviews Molecular Cell

Biology, 10, 113-127.

Jancovich, J. K. & Jacobs, B.L. (2011) Innate Immune Evasion Mediated by the Ambystoma

Tigrinum Virus Eukaryotic Translation Initiation Factor 2α homologue. Journal of

Virology, 85, 5061-5069.

297

Janes, K. A. (2015) An Analysis of Critical Factors for Quantitative Immunoblotting. Science

Signaling, 8, doi: 10.1126/scisignal.2005966.

Jang, S. K., Krausslich, H.G., Nicklin, M.J., Duke, G.M., Palmenberg, A.C. & Wimmer,

E. (1988) A Segment of the 5' Nontranslated Region of Encephalomyocarditis Virus RNA

Directs Internal Entry of Ribosomes during in Vitro Translation. Journal of Virology, 62, 2636-

2643.

Jang, S. K., Kräusslich, H.G., Nicklin, M.J., Duke, G.M., Palmenberg, A.C. & Wimmer,

E. (1988) A Segment of the 5' Nontranslated Region of Encephalomyocarditis Virus RNA

Directs Internal Entry of Ribosomes during in Vitro Translation. Journal of Virology, 62, 2636-

2643.

Janzen, C., Sen, S., Cuevas, J., Reddy, S.T. & Chaudhuri, G. (2011) Protein Phosphatase 2A

Promotes Endothelial Survival via Stabilization of Translational Inhibitor 4E-BP1 Following

Exposure to Tumor Necrosis Factor-α. Arteriosclerosis, Thrombosis, and Vascular Biology,

31, 2586-2594.

Javelaud, D., Laboureau, J., Gabison, E., Verrecchia, F. & Mauviel, A. (2003) Disruption of

Basal JNK Activity Differentially Affects Key Fibroblast Functions Important for Wound

Healing. The Journal of Biological Chemistry, 278, 24624-24628.

Jelinek, T., Dent, P., Sturgill, T.W. & Weber, M.J. (1996) Ras-Induced Activation of Raf-1 is

Dependent on Tyrosine Phosphorylation. Molecular and Cellular Biology, 16, 1027-1034.

Jennings, M. D., Zhou, Y., Mohammad-Qureshi, S.S., Bennett, D. & Pavitt,

G.D. (2013) EIF2B Promotes eIF5 Dissociation from eIF2▪GDP to Facilitate Guanine

Nucleotide Exchange for Translation Initiation. Genes & Development, 27, 2696-2707.

Jensen, C. J., Buch, M.B., Krag, T.O., Hemmings, B.A., Gammeloft, S. & Frödin,

M. (1999) 90-kDa Ribosomal S6 Kinase is Phosphorylated and Activated by 3-

Phosphoinositide-Dependent Protein Kinase-1. The Journal of Biological Chemistry, 274,

27168-27176.

Jensen, K. J., Garmaroudi, F.S., Zhang, J., Lin, J., Boroomand, S., Zhang, M., Luo, Z., Yang,

D., Luo, H., McManus, B.M. & Janes, K.A. (2013) An ERK-p38 Subnetwork Coordinates Host

Cell Apoptosis and Necrosis during Coxsackievirus B3 Infection. Cell Host & Microbe, 13, 67-

76.

Jewell, J. L., Kim, Y.C., Russell, R.C., Yu, F.X., Park, H.W., Plouffe, S.W., Tagliabracci, V.S. &

Guan, K.L. (2015) Differential Regulation of mTORC1 by Leucine and Glutamine. Science,

347, 194-198.

Jiang, B., Monroe, S.S., Koonin, E.V., Stine, S.E. & Glass, R.I. (1993) RNA Sequence of

Astrovirus: Distinctive Genomic Organization and a Putative Retrovirus-Like Ribosomal

298

Frameshifting Signal that Directs the Viral Replicase Synthesis. The Proceedings of the


Jiang, Y., Chen, C., Li, Z., Guo, W., Gegner, J.A., Lin, S. & Han, J. (1996) Characterization of

the Structure and Function of a New Mitogen-Activated Protein Kinase (p38β). The Journal


Jiang, Y., Zhang, S.H., Han, G.Q. & Qin, C.Y. (2010) Interaction of Pdcd4 with eIF4E Inhibits

the Metastatic Potential of Hepatocellular Carcinoma. Biomedicine & Pharmacotherapy, 64,

424-429.

Jiménez-Díaz, A., Remacha, M., Ballesta, J.P. & Berlanga, J.J. (2013) Phosphorylation of

Initiation Factor eIF2 in Response to Stress Conditions is Mediated by Acidic Ribosomal

P1/P2 Proteins in Saccharomyces cerevisiae. PLoS ONE, 8, doi:

10.1371/journal.pone.0084219.

Jin, H. K., Ahn, S.H., Yoon, J.W., Park, J.W., Lee, E.K., Yoo, J.S., Lee, J.C., Choi, W.S. & Han,

J.W. (2009) Rapamycin Down-Regulates Inducible Nitric Oxide Synthase by Inducing

Proteasomal Degradation. Biological & Pharmaceutical Bulletin, 32, 988-992.

Jin, F., Liu, H., Liang, F., Rizkallah, R., Hurt, M.M., & Wang, Y. (2008b) Temporal Control of

the Dephosphorylation of Cdk Substrates by Mitotic Exit Pathways in Budding Yeast. The

Proceedings of the National Academy of Sciences of the United States of America, 105, 16177-16182.

Jin, X., Jin, H.R., Lee, D., Lee, J.H., Kim, S.K. & Lee, J.J. (2008a) A Quassinoid 6α-

Tigloyloxychaparrinone Inhibits Hypoxia-Inducible Factor-1 Pathway by Inhibition of

Eukaryotic Translation Initiation Factor 4E Phosphorylation. European Journal of


Johnson, A. M., Bowen, M.D., Ksiazek, T.G., Williams, R.J., Bryan, R.T., Mills, J.N., Peters,

C.J. & Nichol, S.T. (1997) Laguna Negra Virus Associated with HPS in Western Paraguay and

Bolivia. Virology, 238, 115-127.

Johnson, A. M., de Souza, L.T., Ferreira, I.B., Pereira, L.E., Ksiazek, T.G., Rollin, P.E., Peters,

C.J. & Nichol, S.T. (1999) Genetic Investigation of Novel Hantaviruses Causing Fatal HPS in

Brazil. Journal of Medical Virology, 59, 527-535.

Jonsson, C. B., Figueiredo, L.T. & Vapalahti, O. (2010) A Global Perspective on Hantavirus

Ecology, Epidemiology, and Disease. Clinical Microbiology Reviews, 23, 412-441.

Julien, L.-A., Carriere, A., Moreau, J. & Roux, P.P. (2010) mTORC1-Activated S6K1

Phosphorylates Rictor on Threonine 1135 and Regulates mTORC2 Signaling. Molecular and

Cellular Biology, 30, 908-921.

Jurman, G., Riccadonna, S. & Furlanello, C. (2012) A Comparison of MCC and CEN Error

Measures in Multi-Class Prediction. PLoS ONE, 7, doi: 10.1371/journal.pone.0041882.

299

Kalveram, B. & Ikegami, T. (2013) Toscana Virus NSs Protein Promotes Degradation of

Double-Stranded RNA-Dependent Protein Kinase. Journal of Virology, 87, 3710-3718.

Kanamori, Y. & Nakashima, N. (2001) A Tertiary Structure Model of the Internal Ribosome

Entry Site (IRES) for Methionine-Independent Initiation of Translation. RNA, 7, 266-274.

Kantorovich, L.V. (1939) Mathematical Methods in the Organization and Planning of

Production. Publication House of the Leningrad State University, St. Petersburg (RU).

Translated in Management Science, 6, 366-422 (1960).

Katayama, K., Yamaguchi, M., Noguchi, K. & Sugimoto, Y. (2014) Protein Phosphatase

Complex PP5/PPP2R3C Dephosphorylates P-glycoprotein/ABCB1 and Down-Regulates the

Expression and Function. Cancer Letters, 345, 124-131.

Kawagishi-Kobayashi, M., Silverman, J.B., Ung, T.L. & Dever, T.E. (1997) Regulation of the

Protein Kinase PKR by the Vaccinia Virus Pseudosubstrate Inhibitor K3L is Dependent on

Residues Conserved between the K3L Protein and the PKR Substrate eIF2α. Molecular and


Kawamata, S., Sakaida, H., Hori, T., Maeda, M. & Uchiyama, T. (1998) The Upregulation of

p27Kip1 by Rapamycin Results in G1 Arrest in Exponentially Growing T-Cell Lines. Blood, 91,

561-569.

Kessler, S. H. & Sachs, A.B. (1998) RNA Recognition Motif 2 of Yeast Pab1p is Required for

its Functional Interaction with Eukaryotic Translation Initiation Factor 4G. Molecular and


Khaiboullina, S. F., Rizvanov, A.A., Lombardi, V.C., Morzunov, S.P., Reis, H.J., Palotás, A. &

St Jeor, S. (2013) Andes-Virus-Induced Cytokine Storm is Partially Suppressed by

Ribavirin. Antiviral Therapy, 18, 575-584.

Khamzina, L., Gruppuso, P.A. & Wands, J.R. (2003) Insulin Signaling Through Insulin

Receptor Substrate 1 and 2 in Normal Liver Development. Gastroenterology, 125, 572-585.

Khan, A. S., Gaviria, M., Rollin, P.E., Hlady, W.G., Ksiazek, T.G., Armstrong, L.R., Greenman,

R., Ravkov, E., Kolber, M., Anapol, H., Sfakianaki, E.D., Nichol, S.T., Peters, C.J. & Khabbaz,

R.F. (1996) Hantavirus Pulmonary Syndrome in Florida: Association with the Newly

Identified Black Creek Canal Virus. The American Journal of Medicine, 100, 46-48.

Kholodenko, B. (2006) Cell-Signalling Dynamics in Time and Space. Nature Reviews

Molecular Cell Biology, 7, 165-176.

Kholodenko, B., Hancock, J.F. & Kolch, W. (2010) Signalling Ballet in Space and Time. Nature

Reviews Molecular Cell Biology, 11, 414-426.

300

Kieft, J. S. (2008) Viral IRES RNA Structures and Ribosome Interactions. TRENDS in

Biochemical Sciences, 33, 274-283.

Kierzek, A. M., Zaim, J. & Zielenkiewicz, P. (2001) The Effect of Transcription and

Translation Initiation Frequencies on the Stochastic Fluctuations in Prokaryotic Gene

Expression. The Journal of Biological Chemistry, 276, 8165-8172.

Kim, A. H., Khursigara, G., Sun, X., Franke, T.F. & Chao, M.V. (2001) Akt Phosphorylates and

Negatively Regulates Apoptosis Signal-Regulating Kinase 1. Molecular and Cellular

Biology, 21, 893-901.

Kim, J. H., Yoon, M.S. & Chen, J. (2009a) Signal Transducer and Activator of Transcription 3

(STAT3) Mediates Amino Acid Inhibition of Insulin Signaling through Serine 727


Kim, K. Y., Baek, A., Hwang, J.E., Choi, Y.A., Jeong, J., Lee, M.S., Cho, D.H., Lim, J.S., Kim,

K.I. & Yang, Y. (2009b) Adiponectin-Activated AMPK Stimulates Dephosphorylation of AKT

through Protein Phosphatase 2A Activation. Cancer Research, 69, 4018-4026.

Kim, S.I., Kwak, J.H., Zachariah, M., He, Y., Wang, L. & Choi, M.E. (2007a) TGF-β-Activated

Kinase 1 and TAK1-Binding Protein 1 Cooperate to Mediate TGF-β1-Induced MKK3-p38

MAPK Activation and Stimulation of Type I Collagen. American Journal of Physiology – Renal

Physiology, 292, F1471-F1478.

Kim, S. J., DeStefano, M.A., Oh, W.J., Wu, C.C., Vega-Cotto, N.M., Finlan, M., Liu, D., Su, B.

& Jacinto, E. (2012) mTOR Complex 2 Regulates Proper Turnover of Insulin Receptor

Substrate-1 Via the Ubiquitin Ligase Subunit Fbw8. Molecular Cell, 48, 875-887.

Kim, S. Y., Baik, K.H., Baek, K.H., Chah, K.H., Kim, K.A., Moon, G., Jung, E., Kim, S.T., Shim,

J.H., Greenblatt, M.B., Chun, E. & Lee, K.Y. (2014) S6K1 Negatively Regulates TAK1 Activity

in the Toll-Like Receptor Signaling Pathway. Molecular and Cellular Biology, 34, 510-521.

Kim, Y. M., Ramírez, J.A., Mick, J.E., Giebler, H.A., Yan, J.P. & Nyborg,

J.K. (2007b) Molecular Characterization of the Tax-Containing HTLV-1 Enhancer Complex

Reveals a Prominent Role for CREB Phosphorylation in Tax Transactivation. The Journal of


Kimball, S. R., Fabian, J.R., Pavitt, G.D., Hinnebusch, A.G. & Jefferson,

L.S. (1998) Regulation of Guanine Nucleotide Exchange through Phosphorylation of

Eukaryotic Initiation Factor eIF2α. Role of the α- and δ-Subunits of eIF2B. The Journal of


Kino, T., Hatanaka, H., Hashimoto, M., Nishiyama, M., Goto, T., Okuhara, M., Kohsaka, M.,

Aoki, H. & Imanaka, H. (1987a) FK-506, a Novel Immunosuppressant Isolated from a

301

Streptomyces. I. Fermentation, Isolation, and Physico-Chemical and Biological

Characteristics. The Journal of Antibiotics, 40, 1249-1255.

Kino, T., Hatanaka, H., Miyata, S., Inamura, N., Nishiyama, M., Yajima, T., Goto, T.,

Okuhara, M., Kohsaka, M. & Aoki, H. (1987b) FK-506, a Novel Immunosuppressant Isolated

from a Streptomyces. II. Immunosuppressive Effect of FK-506 in Vitro. The Journal of

Antibiotics, 40, 1256-1265.

Kleijn, M., Scheper, G.C., Wilson, M.L., Tee, A.R. & Proud, C.G. (2002) Localisation and

Regulation of the eIF4E-Binding Protein 4E-BP3. FEBS Letters, 532, 319-323.

Klein, R. J., Zeiss, C., Chew, E.Y., Tsai, J.Y., Sackler, R.S., Haynes, C., Henning, A.K.,

SanGiovanni, J.P., Mane, S.M., Mayne, S.T., Bracken, M.B., Ferris, F.L., Ott, J., Barnstable,

C. & Hoh, J. (2005) Complement Factor H Polymorphism in Age-Related Macular

Degeneration. Science, 308, 385-389.

Kleinstein, S. H. & Singh, J.P. (2001) Toward Quantitative Simulation of Germinal Center

Dynamics: Biological and Modeling Insights from Experimental Validation. Journal of

Theoretical Biology, 211, 253-275.

Knopp, K. A., Ngo, T., Gershon, P.D. & Buchmeier, M.J. (2015) Single Nucleoprotein Residue

Modulates Arenavirus Replication Complex Formation. mBio, 6, doi: 10.1128/mBio.00524-

15.

Korb, A., Tohidast-Akrad, M., Cetin, E., Axmann, R., Smolen, J. & Schett,

G. (2006) Differential Tissue Expression and Activation of p38 MAPK α, β, γ,

and δ Isoforms in Rheumatoid Arthritis. Arthritis and Rheumatism, 54, 2745-2756.

Korkola, J. E., Collisson, E.A., Heiser, L., Oates, C., Bayani, N., Itani, S., Esch, A., Thompson,

W., Griffith, O.L., Wang, N.J., Kuo, W.L., Cooper, B., Billig, J., Ziyad, S., Hung, J.L., Jakkula,

L., Feiler, H., Lu, Y., Mills, G.B., Spellman, P.T., Tomlin, C., Mukherjee, S. & Gray,

J.W. (2015) Decoupling of the PI3K Pathway Via Mutation Necessitates Combinatorial

Treatment in HER2+ Breast Cancer. PLoS ONE, 10, doi: 10.1371/journal.pone.0133219.

Kormelink, R., van Poelwijk, F., Peters, D. & Goldbach, R. (1992) Non-Viral Heterogeneous

Sequences at the 5' Ends of Tomato Spotted Wilt Virus mRNAs. The Journal of General

Virology, 73, 2125-2128.

Korneeva, N. L., Lamphear, B.J., Hennigan, F.L.C. & Rhoads, R.E. (2000) Mutually

Cooperative Binding of Eukaryotic Translation Initiation Factor (eIF) 3 and eIF4A to Human

eIF4G-1. The Journal of Biological Chemistry, 275, 41369-41376.

Koshiji, M., Kageyama, Y., Pete, E.A., Horikawa, I., Barrett, J.C. & Huang, L.E. (2004) HIF-

1α induces Cell Cycle Arrest by Functionally Counteracting Myc. The EMBO Journal, 23,

1949-1956.

302

Kothamachu, V.B., Feliu, E., Wiuf, C., Cardelli, L. & Soyer, O.S. (2013) Phosphorelays

Provide Tunable Signal Processing Capabilities for the Cell. PLoS Computational Biology, 9,

doi: 10.1371/journal.pcbi.1003322.

Koumenis, C., Naczki, C., Koritzinsky, M., Rastani, S., Diehl, A., Sonenberg, N., Koromilas,

A. & Wouters, B.G. (2002) Regulation of Protein Synthesis by Hypoxia Via Activation of the

Endoplasmic Reticulum Kinase PERK and Phosphorylation of the Translation Initiation Factor

eIF2α. Molecular and Cellular Biology, 22, 7405-7416.

Kouroku, Y., Fujita, E., Tanida, I., Ueno, T., Isoai, A., Kumagai, H., Ogawa, S., Kaufman, R.J.,

Kominami, E. & Momoi, T. (2007) ER Stress (PERK/eIF2α phosphorylation) Mediates the

Polyglutamine-Induced LC3 Conversion, an Essential Step for Autophagy Formation. Cell

Death and Differentiation, 14, 230-239.

Kovacina, K. S., Park, G.Y., Bae, S.S., Guzzetta, A.W., Schaefer, E., Birnbaum, M.J. & Roth,

R.A. (2003) Identification of a Proline-Rich Akt Substrate as a 14-3-3 Binding Partner. The


Kozak, M. (1997) Recognition of AUG and Alternative Initiator Codons is Augmented by G in

Position +4 but is Not Generally Affected by the Nucleotides in Positions +5 and +6. The

EMBO Journal, 16, 2482-2492.

Kozak, M. (1987) An Analysis of 5'-Noncoding Sequences from 699 Vertebrate Messenger

RNAs. Nucleic Acids Research, 15, 8125-8149.

Kozany, C., März, A., Kress, C. & Hausch, F. (2009) Fluorescent Probes to Characterise

FK506-Binding Proteins. Chembiochem, 10, 1402-1410.

Kozomara, A. & Griffiths-Jones, S. (2014) miRBase: Annotating High Confidence microRNAs

using Deep Sequencing Data. Nucleic Acids Research, 42, D68-D73.

Krähling, V., Stein, D.A., Spiegel, M., Weber, F. & Mühlberger, E. (2009) Severe Acute

Respiratory Syndrome Coronavirus Triggers Apoptosis Via Protein Kinase R but is Resistant

to its Antiviral Activity. Journal of Virology, 83, 2298-2309.

Kranzusch, P. J. & Whelan, S.P. (2011) Arenavirus Z Protein Controls Viral RNA Synthesis by

Locking a Polymerase-Promoter Complex. The Proceedings of the National Academy of


Krebs, J.E., Lewin, B., Kilpatrick, S.T. & Goldstein, E.S. (2013) ‘Chapter 26 – The Operon’ by

J.E. Krebs (ed.) in Lewin’s Genes XI, 11th Editions, pp. 745-776., Jones and Bartlett Publishers,

Inc., Burlington, MA (USA).

Krishna, V. M., Janaki, N. & Ramaiah, K.V. (1997) Wheat Germ Initiation Factor 2 (WG x

eIF2) Decreases the Inhibition in Protein Synthesis and eIF2B Activity of Reticulocyte Lysates

Mediated by eIF2α phosphorylation. Archives of Biochemistry and Biophysics, 346, 28-36.

303

Krishnamoorthy, T., Pavitt, G.D., Zhang, F., Dever, T.E. & Hinnebusch, A.G. (2001) Tight

Binding of the phosphorylated α subunit of Initiation Factor 2 (eIF2α) to the

regulatory subunits of Guanine Nucleotide Exchange Factor eIF2B is Required for Inhibition

of Translation Initiation. Molecular and Cellular Biology, 21, 5018-5030.

Krump, E., Sanghera, J.S., Pelech, S.L., Furuya, W. & Grinstein, S. (1997) Chemotactic

Peptide N-Formyl-Met-Leu-Phe Activation of p38 Mitogen-Activated Protein Kinase (MAPK)

and MAPK-Activated Protein Kinase-2 in Human Neutrophils. The Journal of Biological

Chemistry, 272, 937-944.

Kruskal, W. H. & Wallis, W.A. (1952) Use of Ranks in One-Criterion Variance

Analysis. Journal of the American Statistical Association, 260, 583-621.

Kulik, G., Klippel, A. & Weber, M.J. (1997) Antiapoptotic Signalling by the Insulin-Like

Growth Factor I Receptor, Phosphatidylinositol 3-Kinase, and Akt. Molecular and Cellular

Biology, 17, 1595-1606.

Kumar, S. & Tikoo, K. (2015) Independent Role of PP2A and mTORc1 in Palmitate Induced

Podocyte Death. Biochimie, 112, 73-84.

Kwiatkowska, M., Norman, G. & Parker, D. (2011) PRISM 4.0: Verification of Probabilistic

Real-Time Systems (2011 23rd International Conference on Computer Aided Verification

(CAV’11)). Snowbird, UT, USA 14th-20th July (2011). Published by Springer International

Publishing AG, Cham, ZG (CHE).

Lai, W. S., Kennington, E.A. & Blackshear, P.J. (2003) Tristetraprolin and its Family Members

can Promote the Cell-Free Deadenylation of AU-Rich Element-Containing mRNAs by Poly(A)

Ribonuclease. Molecular and Cellular Biology, 23, 3798-3812.

Lamming, D. W., Ye, L., Katajisto, P., Goncalves, M.D., Saitoh, M., Stevens, D.M., Davis,

J.G., Salmon, A.B., Richardson, A., Ahima, R.S., Guertin, D.A., Sabatini, D.M. & Baur,

J.A. (2012)Rapamycin-Induced Insulin Resistance is Mediated by mTORC2 Loss and

Uncoupled from Longevity. Science, 335, 1638-1643.

Lander, E. S., Linton, L.M., Birren, B., Nusbaum, C., Zody, M.C., Baldwin, J., Devon, K.,

Dewar, K., Doyle, M., FitzHugh, W., Funke, R., Gage, D., Harris, K., Heaford, A., Howland, J.,

Kann, L., Lehoczky, J., LeVine, R., McEwan, P., McKernan, K., Meldrim, J., Mesirov, J.P.,

Miranda, C., Morris, W., Naylor, J., Raymond, C., Rosetti, M., Santos, R., Sheridan, A.,

Sougnez, C., Stange-Thomann, N., Stojanovic, N., Subramanian, A., Wyman, D., Rogers, J.,

Sulston, J., Ainscough, R., Beck, S., Bentley, D., Burton, J., Clee, C., Carter, N., Coulson, A.,

Deadman, R., Deloukas, P., Dunham, A., Dunham, I., Durbin, R., French, L., Grafham, D.,

Gregory, S., Hubbard, T., Humphray, S., Hunt, A., Jones, M., Lloyd, C., McMurray, A.,

Matthews, L., Mercer, S., Milne, S., Mullikin, J.C., Mungall, A., Plumb, R., Ross, M.,

Shownkeen, R., Sims, S., Waterston, R.H., Wilson, R.K., Hillier, L.W., McPherson, J.D.,

304

Marra, M.A., Mardis, E.R., Fulton, L.A., Chinwalla, A.T., Pepin, K.H., Gish, W.R., Chissoe,

S.L., Wendl, M.C., Delehaunty, K.D., Miner, T.L., Delehaunty, A., Kramer, J.B., Cook, L.L.,

Fulton, R.S., Johnson, D.L., Minx, P.J., Clifton, S.W., Hawkins, T., Branscomb, E., Predki, P.,

Richardson, P., Wenning, S., Slezak, T., Doggett, N., Cheng, J.F., Olsen, A., Lucas, S., Elkin,

C., Uberbacher, E., Frazier, M., Gibbs, R.A., Muzny, D.M., Scherer, S.E., Bouck, J.B.,

Sodergren, E.J., Worley, K.C., Rives, C.M., Gorrell, J.H., Metzker, M.L., Naylor, S.L.,

Kucherlapati, R.S., Nelson, D.L., Weinstock, G.M., Sakaki, Y., Fujiyama, A., Hattori, M.,

Yada, T., Toyoda, A., Itoh, T., Kawagoe, C., Watanabe, H., Totoki, Y., Taylor, T.,

Weissenbach, J., Heilig, R., Saurin, W., Artiguenave, F., Brottier, P., Bruls, T., Pelletier, E.,

Robert, C., Wincker, P., Smith, D.R., Doucette-Stamm, L., Rubenfield, M., Weinstock, K.,

Lee, H.M., Dubois, J., Rosenthal, A., Platzer, M., Nyakatura, G., Taudien, S., Rump, A.,

Yang, H., Yu, J., Wang, J., Huang, G., Gu, J., Hood, L., Rowen, L., Madan, A., Qin, S., Davis,

R.W., Federspiel, N.A., Abola, A.P., Proctor, M.J., Myers, R.M., Schmutz, J., Dickson, M.,

Grimwood, J., Cox, D.R., Olson, M.V., Kaul, R., Raymond, C., Shimizu, N., Kawasaki, K.,

Minoshima, S., Evans, G.A., Athanasiou, M., Schultz, R., Roe, B.A., Chen, F., Pan, H.,

Ramser, J., Lehrach, H., Reinhardt, R., McCombie, W.R., de la Bastide, M., Dedhia, N.,

Blöcker, H., Hornischer, K., Nordsiek, G., Agarwala, R., Aravind, L., Bailey, J.A., Bateman,

A., Batzoglou, S., Birney, E., Bork, P., Brown, D.G., Burge, C.B., Cerutti, L., Chen, H.C.,

Church, D., Clamp, M., Copley, R.R., Doerks, T., Eddy, S.R., Eichler, E.E., Furey, T.S.,

Galagan, J., Gilbert, J.G., Harmon, C., Hayashizaki, Y., Haussler, D., Hermjakob, H.,

Hokamp, K., Jang, W., Johnson, L.S., Jones, T.A., Kasif, S., Kaspryzk, A., Kennedy, S., Kent,

W.J., Kitts, P., Koonin, E.V., Korf, I., Kulp, D., Lancet, D., Lowe, T.M., McLysaght, A.,

Mikkelsen, T., Moran, J.V., Mulder, N., Pollara, V.J., Ponting, C.P., Schuler, G., Schultz, J.,

Slater, G., Smit, A.F., Stupka, E., Szustakowski, J., Thierry-Mieg, D., Thierry-Mieg, J.,

Wagner, L., Wallis, J., Wheeler, R., Williams, A., Wolf, Y.I., Wolfe, K.H., Yang, S.P., Yeh,

R.F., Collins, F., Guyer, M.S., Peterson, J., Felsenfeld, A., Wetterstrand, K.A., Patrinos, A.,

Morgan, M.J., de Jong, P., Catanese, J.J., Osoegawa, K., Shizuya, H., Choi, S., Chen, Y.J. &

International Human Genome Sequencing Consortium (2001) Initial Sequencing and

Analysis of the Human Genome. Nature, 409, 860-921.

Laplante, M. & Sabatini, D.M. (2013) Regulation of mTORC1 and its Impact on Gene

Expression at a Glance. Journal of Cell Science, 126, 1713-1719.

Laplante, M. & Sabatini, D.M. (2009) mTOR Signaling at a Glance. Journal of Cell

Science, 122, 3589-3594.

Laurent-Rolle, M., Boer, E.F., Lubick, K.J., Wolfinbarger, J.B., Carmody, A.B., Rockx, B., Liu,

W., Ashour, J., Shupert, W.L., Holbrook, M.R., Barrett, A.D., Mason, P.W., Bloom, M.E.,

García-Sastre, A., Khromykh, A.A. & Best, S.M. (2010) The NS5 Protein of the Virulent West

Nile Virus NY99 Strain is a Potent Antagonist of Type I Interferon-Mediated JAK-STAT

Signaling. Journal of Virology, 84, 3503-3515.

305

Laxman, S., Sutter, B., Wu, X., Kumar, S., Guo, X., Trudgian, D., Mirzaei, H. & Tu,

B. (2013) Sulfur Amino Acids Regulate Translational Capacity and Metabolic Homeostasis

through Modulation of tRNA Thiolation. Cell, 154, 416-429.

Lázaro, M. E., Cantoni, G.E., Calanni, L.M., Resa, A.J., Herrero, E.R., Iacono, M.A., Enria,

D.A. & González Cappa, S.M. (2007) Clusters of Hantavirus Infection, Southern

Argentina. Emerging Infectious Diseases, 13, 104-110.

Le Good, J. A., Ziegler, W.H., Parekh, D.B., Alessi, D.R., Cohen, P. & Parker,

P.J. (1998) Protein Kinase C Isotypes Controlled by Phosphoinositide 3-Kinase through the

Protein Kinase PDK1. Science, 281, 2042-2045.

Le, N. T., Heo, K.S., Takei, Y., Lee, H., Woo, C.H., Chang, E., McClain, C., Hurley, C., Wang,

X., Li, F., Xu, H., Morrell, C., Sullivan, M.A., Cohen, M.S., Serafilmova, I.M., Taunton, J.,

Fujiwara, K. & Abe, J. (2013) A Crucial Role for p90RSK-Mediated Reduction of

ERK5 Transcriptional Activity in Endothelial Dysfunction and Atherosclerosis. Circulation,

127, 486-499

Lebedeva, S., Jens, M., Theil, K., Schwanhausser, B., Selbach, M., Landthaler, M. &

Rajewsky, N. (2011) Transcriptome-Wide Analysis of Regulatory Interactions of the RNA-

Binding Protein HuR. Molecular Cell, 43, 340-352.

Lee, S., Liu, B., Lee, S., Huang, S.-X., Shen, B. & Qian, S.-B. (2012) Global Mapping of

Translation Initiation Sites in Mammalian Cells at Single-Nucleotide Resolution. The

Proceedings of the National Academy of Sciences of the United States of America, 109,

E2424-E2432.

Leevers, S. J., Paterson, H.F. & Marshall, C.J. (1994) Requirement for Ras in Raf Activation is

Overcome by Targeting Raf to the Plasma Membrane. Nature, 369, 411-414.

LeFebvre, A. K., Korneeva, N.L., Trutschl, M., Cvek, U., Duzan, R.D., Bradley, C.A., Hershey,

J.W.B. & Rhoads, R.E. (2006) Translation Initiation Factor eIF4G-1 Binds to eIF3 through the

eIF3e Subunit. The Journal of Biological Chemistry, 281, 22917-22932.

Lehmann, M., Pahlmann, M., Jérôme, H., Busch, C., Lelke, M. & Günther, S. (2014) Role of

the C Terminus of Lassa Virus L Protein in Viral mRNA Synthesis. Journal of Virology, 88,

8713-8717.

Lelke, M., Brunotte, L., Busch, C. & Gunther, S. (2010) An N-Terminal Region of Lassa Virus L

Protein Plays a Critical Role in Transcription but Not Replication of the Virus

Genome. Journal of Virology, 84, 1934-1944.

Léonard, S., Viel, C., Beauchemin, C., Daigneault, N., Fortin, M.G. & Laliberté,

J.F. (2004) Interaction of VPg-Pro of Turnip Mosaic Virus with the Translation Initiation

306

Factor 4E and the Poly(A)-Binding Protein in Planta. The Journal of General Virology, 85,

1055-1063.

Leoncikas, V., Wu, H., Ward, L.T., Kierzek, A.K. & Plant, N.J. (2016) Generation of 2,000

Breast Cancer Metabolic Landscapes Reveal a Poor Prognosis Group with Active Serotonin

Production. Scientific Reports, 6, doi:10.1038/srep19771.

Lesseux, L., Laurent, G., Laurent, C., Rigo, M., Blanc, A., Olive, D. & Bezombes, C. (2008)

PKC ζ–mTOR Pathway: A New Target for Rituximab Therapy in Follicular Lymphoma. Blood,

111, 285-291.

Leung, W. K., Chan, P.K.S., Lee, N.L.S. & Sung, J.J.Y. (2010) Development of an in Vitro Cell

Culture Model for Human Noroviruses and its Clinical Application. Hong Kong Medical

Journal, 16, 18-21.

Levy, H. B., Law, L.W. & Rabson, A.S. (1969) Inhibition of Tumor Growth by Polyinosinic-

Polycytidylic Acid. The Proceedings of the National Academy of Sciences of the United States

of America, 62, 357-361.

Li, C., Donizelli, M., Rodriguez, N., Dharuri, H., Endler, L., Chelliah, V., Li, L., He, E., Henry,

A., Stefan, M.I., Snoep, J.L., Hucka, M., Le Novere, N. & Laibe, C. (2010) BioModels

Database: An Enhanced, Curated and Annotated Resource for Published Quantitative Kinetic

Models. BMC Systems Biology, 4, doi: 10.1186/1752-0509-4-92.

Li, H., Lee, J., He, C., Zou, M.H. & Xie, Z. (2014) Suppression of the mTORC1/STAT3/Notch1

Pathway by Activated AMPK Prevents Hepatic Insulin Resistance Induced by Excess Amino

Acids. American Journal of Physiology - Endocrinology and Metabolism, 306, E197-E209.

Li, H., Li, W.X. & Ding, S.W. (2002) Induction and Suppression of RNA Silencing by an Animal

Virus. Science, 296, 1319-1321.

Li, K., Chen, Z., Kato, N., Gale Jr., M. & Lemon, S.M. (2005b) Distinct Poly(I-C) and Virus-

Activated Signaling Pathways Leading to Interferon-β production in Hepatocytes. The


Li, X., Alafuzoff, I., Soininen, H., Winblad, B. & Pei, J.J. (2005a) Levels of mTOR and its

Downstream Targets 4E-BP1, eEF2, and eEF2 Kinase in Relationships with Tau in Alzheimer's

Disease Brain. The FEBS Journal, 272, 4211-4220.

Li, X., An, W.L., Alafuzoff, I., Soininen, H., Winblad, B. & Pei, J.J. (2004) Phosphorylated

Eukaryotic Translation Factor 4E is Elevated in Alzheimer Brain. Neuroreport, 15, 2237-2240.

Li, Y., Inoki, K., Vacratsis, P. & Guan, K.L. (2003) The p38 and MK2 Kinase Cascade

Phosphorylates Tuberin, the Tuberous Sclerosis 2 Gene Product, and Enhances its

Interaction with 14-3-3. The Journal of Biological Chemistry, 278, 13663-13671.

307

Liao, Y. & Hung, M.C. (2003) Regulation of the Activity of p38 Mitogen-Activated Protein

Kinase by Akt in Cancer and Adenoviral Protein E1A-Mediated Sensitization to Apoptosis.

Molecular and Cellular Biology, 23, 6836-6848.

Lihoradova, O. A., Indran, S.V., Kalveram, B., Lokugamage, N., Head, J.A., Gong, B., Tigabu,

B., Juelich, T.L., Freiberg, A.N. & Ikegami, T. (2013) Characterization of Rift Valley Fever

Virus MP-12 Strain Encoding NSs of Punta Toro Virus Or Sandfly Fever Sicilian Virus. PLoS

Neglected Tropical Diseases, 7, doi: 10.1371/journal.pntd.0002181.

Lin, R. J., Chang, B.L., Yu, H.P., Liao, C.L. & Lin, Y.L. (2006) Blocking of Interferon-Induced

Jak-Stat Signaling by Japanese Encephalitis Virus NS5 through a Protein Tyrosine

Phosphatase-Mediated Mechanism. Journal of Virology, 80, 5908-5918.

Lindsey-Boltz, L. A., Wauson, E.M., Graves, L.M. & Sancar, A. (2004) The Human Rad9

Checkpoint Protein Stimulates the Carbamoyl Phosphate Synthetase Activity of the

Multifunctional Protein CAD. Nucleic Acids Research, 32, 4524-4530.

Linero, F., Welnowska, E., Carrasco, L. & Scolaro, L. (2013) Participation of eIF4F Complex in

Junin Virus Infection: Blockage of eIF4E does Not Impair Virus Replication. Cellular

Microbiology, 15, 1766-1782.

Linero, F. N. & Scolaro, L.A. (2009) Participation of the Phosphatidylinositol 3-kinase/Akt

Pathway in Junin Virus Replication in Vitro. Virus Research, 145, 166-170.

Ling, J., Morley, S.J., Pain, V.M., Marzluff, W.F. & Gallie, D.R. (2002) The Histone 3'-

Terminal Stem-Loop-Binding Protein Enhances Translation through a Functional and Physical

Interaction with Eukaryotic Initiation Factor 4G (eIF4G) and eIF3. Molecular and Cellular

Biology, 22, 7853-7867.

Liu, Y., Franco, D., Paul, A.V. & Wimmer, E. (2007) Tyrosine 3 of Poliovirus Terminal Peptide

VPg(3B) has an Essential Function in RNA Replication in the Context of its Precursor Protein,

3AB. Journal of Virology, 81, 5669-5684.

Liu, P., Begley, M., Michowski, W., Inuzuka, H., Ginzberg, M., Gao, D., Tsou, P., Gan, W.,

Papa, A., Kim, B.M., Wan, L., Singh, A., Zhai, B., Yuan, M., Wang, Z., Gygi, S.P., Lee, T.H.,

Lu, K.P., Toker, A., Pandolfi, P.P., Asara, J.M., Kirschner, M.W., Sicinski, P., Cantley, L. &

Wei, W. (2014) Cell-Cycle-Regulated Activation of Akt Kinase by Phosphorylation at its

Carboxyl Terminus. Nature, 508, 541-545.

Liu, P., Gan, W., Chin, Y.R., Ogura, K., Guo, J., Zhang, J., Wang, B., Blenis, J., Cantley, L.C.,

Toker, A., Su, B., Wei, W. (2015) PtdIns(3,4,5)P3-Dependent Activation of the mTORC2

Kinase Complex. Cancer Discovery, 5, 1194-1209.

Liu, P., Gan, W., Inuzuka, H., Lazorchak, A.S., Gao, D., Arojo, O., Liu, D., Wan, L., Zhai, B.,

Yu, Y., Yuan, M., Kim, B.M., Shaik, S., Menon, S., Gygi, S.P., Lee, T.H., Asara, J.M., Manning,

308

B.D., Blenis, J., Su, B. & Wei, W. (2013) Sin1 Phosphorylation Impairs mTORC2 Complex

Integrity and Inhibits Downstream Akt Signalling to Suppress Tumorigenesis. Nature Cell

Biology, 15, 1340-1350.

Liu, S., Bhattacharya, S., Han, A., Suragani, R.N., Zhao, W., Fry, R.C. & Chen,

J.J. (2008) Haem-Regulated eIF2α kinase is Necessary for Adaptive Gene Expression in

Erythroid Precursors Under the Stress of Iron Deficiency. British Journal of

Haematology, 143, 129-137.

Liu, Y., László, C., Liu, Y., Liu, W., Chen, X., Evans, S.C. & Wu, S. (2010) Regulation of G1

Arrest and Apoptosis in Hypoxia by PERK and GCN2-Mediated

eIF2α phosphorylation. Neoplasia, 12, 61-68.

Liu, Z., Carthy, C.M., Cheung, P., Bohunek, L., Wilson, J.E., McManus, B.M. & Yang,

D. (1999) Structural and Functional Analysis of the 5' Untranslated Region of Coxsackievirus

B3 RNA: In Vivo Translational and Infectivity Studies of Full-Length Mutants. Virology, 265,

206-217.

Liu, G., Zhang, Y., Bode, A.M., Ma, W.Y. & Dong, Z. (2002) Phosphorylation of 4E-BP1 is

Mediated by the p38/MSK1 Pathway in Response to UVB Irradiation. The Journal of


Löber, C., Anheier, B., Lindow, S., Klenk, H.D. & Feldmann, H. (2001) The Hantaan Virus

Glycoprotein Precursor is Cleaved at the Conserved Pentapeptide WAASA. Virology, 289,

224-229.

Loog, M. & Morgan, D.O. (2005) Cyclin Specificity in the Phosphorylation of Cyclin-

Dependent Kinase Substrates. Nature, 434, 104-108.

López, N., Padula, P., Rossi, C., Lázaro, M.E. & Franze-Fernández, M.T. (1996) Genetic

Identification of a New Hantavirus Causing Severe Pulmonary Syndrome in

Argentina. Virology, 220, 223-226.

López, N., Padula, P., Rossi, C., Miguel, S., Edelstein, A., Ramírez, E. & Franze-Fernández,

M.T. (1997) Genetic Characterization and Phylogeny of Andes Virus and Variants from

Argentina and Chile. Virus Research, 50, 77-84.

López-Pelaéz, M., Fumagalli, S., Sanz, C., Herrero, C., Guerra, S., Fernandez, M. & Alemany,

S. (2012) Cot/tpl2-MKK1/2-Erk1/2 Controls mTORC1-Mediated mRNA Translation in Toll-

Like Receptor–Activated Macrophages. Molecular Biology of the Cell, 23, 2982-2992.

Lopman, B. A., Reacher, M.H., Vipond, I.B., Sarangi, J. & Brown, D.W.G. (2004) Clinical

Manifestation of Norovirus Gastroenteritis in Health Care Settings. Clinical Infectious

Diseases, 39, 318-324.

309

Lorrot, M., Bon, F., El Hajje, M.J., Aho, S., Wolfer, M., Giraudon, H., Kaplon, J., Marc, E.,

Raymond, J., Lebon, P., Pothier, P. & Gendrel, D. (2011) Epidemiology and Clinical Features

of Gastroenteritis in Hospitalised Children: Prospective Survey during a 2-Year Period in a

Parisian Hospital, France. European Journal of Clinical Microbiology & Infectious

Diseases, 30, 361-368.

Lorsch, J. R. & Herschlag, D. (1998a) The DEAD Box Protein eIF4A. 1. A Minimal Kinetic and

Thermodynamic Framework Reveals Coupled Binding of RNA and

Nucleotide. Biochemistry, 37, 2180-2193.

Lorsch, J. R. & Herschlag, D. (1998b) The DEAD Box Protein eIF4A. 2. A Cycle of Nucleotide

and RNA-Dependent Conformational Changes. Biochemistry, 37, 2194-2206.

Lowther, J. A., Gustar, N.E., Powell, A.L., Hartnell, R.E. & Lees, D.N. (2012) Two-Year

Systematic Study to Assess Norovirus Contamination in Oysters from Commercial Harvesting

Areas in the United Kingdom. Applied and Environmental Microbiology, 78, 5812-5817.

Lu, L., Jia, R., Zhong, H., Xu, M., Su, L., Cao, L., Dong, Z., Dong, N. & Xu, J. (2015) Molecular

Characterization and Multiple Infections of Rotavirus, Norovirus, Sapovirus, Astrovirus and

Adenovirus in Outpatients with Sporadic Gastroenteritis in Shanghai, China, 2010-

2011. Archives of Virology, 160, 1229-1238.

Lu, X., Masic, A., Li, Y., Shin, Y., Liu, Q. & Zhou, Y. (2010) The PI3K/Akt Pathway Inhibits

Influenza A Virus-Induced Bax-Mediated Apoptosis by Negatively Regulating the JNK

Pathway Via ASK1. The Journal of General Virology, 91, 1439-1449.

Lu, Z., Xu, X., Fassett, J., Kwak, D., Liu, X., Hu, X., Wang, H., Guo, H., Xu, D., Yan, S., McFalls,

E.O., Lu, F., Bache, R.J. & Chen, Y. (2014) Loss of the Eukaryotic Initiation Factor 2α kinase

General Control Nonderepressible 2 Protects Mice from Pressure Overload-Induced

Congestive Heart Failure without Affecting Ventricular Hypertrophy. Hypertension, 63, 128-

135.

Luna, R. E., Arthanari, H., Hiraishi, H., Akabayov, B., Tang, L., Cox, C., Markus, M.A., Luna,

L.E., Ikeda, Y., Watanabe, R., Bedoya, E., Yu, C., Alikhan, S., Wagner, G. & Asano,

K. (2013) The Interaction between Eukaryotic Initiation Factor 1A and eIF5 Retains eIF1

within Scanning Preinitiation Complexes. Biochemistry, 52, 9510-9518.

MacDonald, S. G., Crews, C.M., Wu, L., Driller, J., Clark, R., Erikson, R.L. & McCormick,

F. (1993) Reconstitution of the Raf-1-MEK-ERK Signal Transduction Pathway in

Vitro. Molecular and Cellular Biology, 13, 6615-6620.

310

Mackichan, M.L. & DeFranco, A.L. (1999) Role of Ceramide in Lipopolysaccharide (LPS)-

Induced Signaling. LPS Increases Ceramide Rather than Acting as a Structural Homolog. The


Madely, C. R. (1979) Comparison of the Features of Astroviruses and Caliciviruses seen in

Samples of Feces by Electron Microscopy. The Journal of Infectious Diseases, 139, 519-523.

Magnuson, B., Ekim, B. & Fingar, D.C. (2012) Regulation and Function of Ribosomal Protein

S6 Kinase (S6K) within mTOR Signalling Networks. The Biochemical Journal, 441, 1-21.

Maiztegui, J. I. (1975) Clinical and Epidemiological Patterns of Argentine Haemorrhagic

Fever. Bulletin of the World Health Organization, 52, 567-575.

Makeeva, N., Myers, J.W. & Welsh, N. (2006) Role of MKK3 and p38 MAPK in Cytokine-

Induced Death of Insulin-Producing Cells. The Biochemical Journal, 393, 129-139.

Makinoshima, H., Takita, M., Saruwatari, K., Umemura, S., Obata, Y., Ishii, G., Matsumoto,

S., Sugiyama, E., Ochiai, A., Abe, R., Goto, K., Esumi, H. & Tsuchihara, K. (2015) Signaling

through the Phosphatidylinositol 3-Kinase (PI3K)/Mammalian Target of Rapamycin (mTOR)

Axis is Responsible for Aerobic Glycolysis Mediated by Glucose Transporter in Epidermal

Growth Factor Receptor (EGFR)-Mutated Lung Adenocarcinoma. The Journal of Biological

Chemistry, 290, 17495-17504

Mali, P., Yang, L., Esvelt, K.M., Aach, J., Guell, M., DiCarlo, J.E., Norville, J.E. & Church,

G.M. (2013) RNA-Guided Human Genome Engineering Via Cas9. Science, 339, 823-826.

Marcotrigiano, J., Gingras, A.C., Sonenberg, N. & Burley, S.K. (1999) Cap-Dependent

Translation Initiation in Eukaryotes is Regulated by a Molecular Mimic of eIF4G. Molecular

cell, 3, 707-716.

Marcus, P. I., Rojek, J.M. & Sekellick, M.J. (2005) Interferon Induction and/or Production

and its Suppression by Influenza A Viruses. Journal of Virology, 79, 2880-2890.

Marderosian, M., Sharma, A., Funk, A.P., Vartanian, R., Masri, J., Jo, O.D. & Gera,

J.F. (2006) Tristetraprolin Regulates Cyclin D1 and c-Myc mRNA Stability in Response to

Rapamycin in an Akt-Dependent Manner Via p38 MAPK Signaling. Oncogene, 25, 6277-6290.

Marshall, W. A. & Olkin, I . (1985) A Family of Bivariate Distributions Generated by the

Bivariate Bernoulli Distribution. Journal of the American Statistical Association, 80, 332-338.

Martel, R. R., Klicius, J. & Galet, S. (1977) Inhibition of the Immune Response by Rapamycin,

a New Antifungal Antibiotic. Canadian Journal of Physiology and Pharmacology, 55, 48-51.

Martelli, A.M., Tabellini, G., Bressanin, D., Ognibene, A., Goto, K., Cocco, L. & Evangelisti,

C. (2012) The Emerging Multiple Roles of Nuclear Akt. Biochimica et Biophysica Acta –

Molecular Cell Research, 1823, 2168-2178.

311

Martinez, V. P., Bellomo, C., San Juan, J., Pinna, D., Forlenza, R., Elder, M. & Padula,

P.J. (2005) Person-to-Person Transmission of Andes Virus. Emerging Infectious Diseases, 11,

1848-1853.

Martínez-Solano, L., Reales-Calderón, J.A., Nombela, C., Molero, G. & Gil,

C. (2009) Proteomics of RAW 264.7 Macrophages upon Interaction with Heat-

Inactivated Candida albicans Cells Unravel an Anti-Inflammatory Response. Proteomics, 9,

2995-3010.

Martinez-Valdebenito, C., Calvo, M., Vial, C., Mansilla, R., Marco, C., Palma, R.E., Vial, P.A.,

Valdivieso, F., Mertz, G. & Ferrés, M. (2014) Person-to-Person Household and Nosocomial

Transmission of Andes Hantavirus, Southern Chile, 2011. Emerging Infectious Diseases, 20,

1629-1636.

Marwan, W., Rohr, C. & Heiner, M. (2012) Petri Nets in Snoopy: A Unifying Framework for

the Graphical Display, Computational Modelling, and Simulation of Bacterial Regulatory

Networks. Methods in Molecular Biology, 804, 409-437.

März, A. M., Fabian, A.K., Kozany, C., Bracher, A. & Hausch, F. (2013) Large FK506-Binding

Proteins Shape the Pharmacology of Rapamycin. Molecular and Cellular Biology, 33, 1357-

1367.

Marzec, M., Liu, X., Wysocka, M., Rook, A.H., Odum, N. & Wasik, M.A. (2011) Simultaneous

Inhibition of mTOR-Containing Complex 1 (mTORC1) and MNK Induces Apoptosis of

Cutaneous T-Cell Lymphoma (CTCL) Cells. PLoS ONE, 6, doi: 10.1371/journal.pone.0024849.

Mast, F.D., Ratushny, A.V. & Aitchinson, J.D. (2014) Systems Cell Biology. The Journal of Cell

Biology, 206, 695-706.

Matthews, B. W. (1975) Comparison of the Predicted and Observed Secondary Structure of

T4 Phage Lysozyme. Biochimica et Biophysica Acta, 405, 442-451.

Maurin, A. C., Jousse, C., Averous, J., Parry, L., Bruhat, A., Cherasse, Y., Zeng, H., Zhang, Y.,

Harding, H.P., Ron, D. & Fafournoux, P. (2005) The GCN2 Kinase Biases Feeding Behavior to

Maintain Amino Acid Homeostasis in Omnivores. Cell Metabolism, 1, 273-277.

Mazure, N. M., Chauvet, C., Bois-Joyeux, B., Bernard, M.A., Nacer-Chérif, H. & Danan,

J.L. (2002) Repression of Alpha-Fetoprotein Gene Expression Under Hypoxic Conditions in

Human Hepatoma Cells: Characterization of a Negative Hypoxia Response Element that

Mediates Opposite Effects of Hypoxia Inducible Factor-1 and c-Myc. Cancer Research, 62,

1158-1165.

McAdams, H.H & Arkin, A. (1997) Stochastic Mechanisms in Gene Expression. The

Proceedings of the National Academy of Sciences of the United States of America, 94, 814-

819.

312

McCloskey, D., Palsson, B.Ø. & Feist, A.M. (2013) Basic and Applied Uses of Genome-Scale

Metabolic Network Reconstructions of Escherichia coli. Molecular Systems Biology, 9, doi:

10.1038/msb.2013.18.

McCoy, C. E., Campbell, D.G., Deak, M., Bloomberg, G.B. & Arthur, J.S. (2005) MSK1

Activity is Controlled by Multiple Phosphorylation Sites. The Biochemical Journal, 387, 507-

517.

McCoy, C. E., Macdonald, A., Morrice, N.A., Campbell, D.G., Deak, M., Toth, R., McIlrath, J.

& Arthur, J.S. (2007) Identification of Novel Phosphorylation Sites in MSK1 by Precursor Ion

Scanning MS. The Biochemical Journal, 402, 491-501.

McFadden, N., Bailey, D., Carrara, G., Benson, A., Chaudhry, Y., Shortland, A., Heeney, J.,

Yarovinsky, F., Simmonds, P., Macdonald, A. & Goodfellow, I. (2011) Norovirus Regulation

of the Innate Immune Response and Apoptosis Occurs Via the Product of the Alternative

Open Reading Frame 4. PLoS Pathogens, 7, doi:10.1371/journal.ppat.1002413.

McInerney, G. M., Kedersha, N.L., Kaufman, R.J., Anderson, P. & Liljeström,

P. (2005) Importance of eIF2α phosphorylation and Stress Granule Assembly in Alphavirus

Translation Regulation. Molecular Biology of the Cell, 16, 3753-3763.

McIntyre, L. M., Martin, E.R., Simonsen, K.L. & Kaplan, N.L. (2000) Circumventing Multiple

Testing: A Multilocus Monte Carlo Approach to Testing for Association. Genetic

Epidemiology, 19, 18-29.

McNulty, S., Flint, M., Nichol, S.T. & Spiropoulou, C.F. (2013) Host mTORC1 Signaling

Regulates Andes Virus Replication. Journal of Virology, 87, 912-922.

Medici, M. C., Martinelli, M., Abelli, L.A., Ruggeri, F.M., Di Bartolo, I., Arcangeletti, M.C.,

Pinardi, F., De Conto, F., Izzi, G., Bernasconi, S., Chezzi, C. & Dettori, G. (2006) Molecular

Epidemiology of Norovirus Infections in Sporadic Cases of Viral Gastroenteritis among

Children in Northern Italy. Journal of Medical Virology, 78, 1486-1492.

Medina, R. A., Torres-Perez, F., Galeno, H., Navarrete, M., Vial, P.A., Palma, R.E., Ferres,

M., Cook, J.A. & Hjelle, B. (2009) Ecology, Genetic Diversity, and Phylogeographic Structure

of Andes Virus in Humans and Rodents in Chile. Journal of Virology, 83, 2446-2459.

Medvedev, A.E., Blanco, J.C., Qureshi, N. & Vogel, S.N. (1999) Limited Role of Ceramide in

Lipopolysaccharide-Mediated Mitogen-Activated Protein Kinase Activation, Transcription

Factor Induction, and Cytokine Release. The Journal of Biological Chemistry, 274, 9342-9350.

Meissner, J. D., Rowe, J.E., Borucki, M.K. & St Jeor, S.C. (2002) Complete Nucleotide

Sequence of a Chilean Hantavirus. Virus Research, 89, 131-143.

313

Melemedjian, O. K., Khoutorsky, A., Sorge, R.E., Yan, J., Asiedu, M.N., Valdez, A., Ghosh,

S., Dussor, G., Mogil, J.S., Sonenberg, N. & Price, T.J. (2013) mTORC1 Inhibition Induces

Pain Via IRS-1-Dependent Feedback Activation of ERK. Pain, 154, 1080-1091.

Mendes, P., Hoops, S., Sahle, S., Gauges, R., Dada, J. & Kummer, U. (2009) Computational

Modeling of Biochemical Networks using COPASI. Methods in Molecular Biology, 500, 17-59.

Mertens, S., Craxton, M. & Goedert, M. (1996) SAP Kinase-3, a New Member of the Family

of Mammalian Stress-Activated Protein Kinases. FEBS letters, 383, 273-276.

Messner, D. J., Romeo, C., Boynton, A. & Rossie, S. (2006) Inhibition of PP2A, but Not PP5,

Mediates p53 Activation by Low Levels of Okadaic Acid in Rat Liver Epithelial Cells. Journal of

Cellular Biochemistry, 99, 241-255.

Metropolis, N. & Ulam, S. (1949) The Monte Carlo Method. Journal of the American

Statistical Association, 44, 335-341.

Meyers, G., Wirblich, C. & Thiel, H.-J. (1991) Rabbit Hemorrhagic Disease Virus—Molecular

Cloning and Nucleotide Sequencing of a Calicivirus Genome. Virology, 184, 664-676.

Mills, J. N., Ellis, B.A., Childs, J.E., McKee Jr., K.T., Maiztegui, J.I., Peters, C.J., Ksiazek, T.G.

& Jahrling, P.B. (1994) Prevalence of Infection with Junin Virus in Rodent Populations in the

Epidemic Area of Argentine Hemorrhagic Fever. The American Journal of Tropical Medicine

and Hygiene, 51, 554-562.

Mills, J. N., Ellis, B.A., McKee Jr., K.T., Calderon, G.E., Maiztegui, J.I., Nelson, G.O., Ksiazek,

T.G., Peters, C.J. & Childs, J.E. (1992) A Longitudinal Study of Junin Virus Activity in the

Rodent Reservoir of Argentine Hemorrhagic Fever. The American Journal of Tropical

Medicine and Hygiene, 47, 749-763.

Mills, J. N., Ellis, B.A., McKee Jr., K.T., Ksiazek, T.G., Oro, J.G., Maiztegui, J.I., Calderon,

G.E., Peters, C.J. & Childs, J.E. (1991) Junin Virus Activity in Rodents from Endemic and

Nonendemic Loci in Central Argentina. The American Journal of Tropical Medicine and

Hygiene, 44, 589-597.

Milner, J. J., Rebeles, J., Dhungana, S., Stewart, D.A., Sumner, S.C., Meyers, M.H.,

Mancuso, P. & Beck, M.A. (2015) Obesity Increases Mortality and Modulates the Lung

Metabolome during Pandemic H1N1 Influenza Virus Infection in Mice. The Journal of

Immunology, 194, 4846-4859.

Minden, A., Lin, A., Smeal, T., Dérijard, B., Cobb, M., Davis, R. & Karin, M. (1994) C-Jun N-

Terminal Phosphorylation Correlates with Activation of the JNK Subgroup but Not the ERK

Subgroup of Mitogen-Activated Protein Kinases. Molecular and Cellular Biology, 14, 6683-

6688.

314

Minich, W. B., Luisa Balasta, M., Goss, D.J. & Rhoads, R.E. (1994) Chromatographic

Resolution of in Vivo Phosphorylated and Nonphosphorylated Eukaryotic Translation

Initiation Factor eIF-4E: Increased Cap Affinity of the Phosphorylated Form. The Proceedings


Mir, M. A. & Panganiban, A.T. (2008) A Protein that Replaces the Entire Cellular eIF4F

Complex. The EMBO Journal, 27, 3129-3139.

Mir, M. A., Sheema, S., Haseeb, A. & Haque, A. (2010) Hantavirus Nucleocapsid Protein has

Distinct m7G Cap- and RNA-Binding Sites. The Journal of Biological Chemistry, 285, 11357-

11368.

Miyake, M., Goodison, S., Lawton, A., Zhang, G., Gomes-Giacoia, E. & Rosser, C.J. (2013)

Erythropoietin is a JAK2 and ERK1/2 Effector that Can Promote Renal Tumor Cell

Proliferation under Hypoxic Conditions. Journal of Hematology & Oncology, 6,

doi:10.1186/1756-8722-6-65

Mizutani, T., Fukushi, S., Saijo, M., Kurane, I. & Morikawa, S. (2004) Phosphorylation of

p38 MAPK and its Downstream Targets in SARS Coronavirus-Infected Cells. Biochemical and

Biophysical Research Communications, 319, 1228-1234.

Monick, M.M., Carter, A.B., Flaherty, D.M., Peterson, M.W. & Hunninghake, G.W. (2000)

Protein Kinase C ζ Plays a Central Role in Activation of the p42/44 Mitogen-Activated Protein

Kinase by Endotoxin in Alveolar Macrophages. The Journal of Immunology, 165, 4632-4639.

Monick, M. M., Powers, L.S., Gross, T.J., Flaherty, D.M., Barrett, C.W. & Hunninghake,

G.W. (2006) Active ERK Contributes to Protein Translation by Preventing JNK-Dependent

Inhibition of Protein Phosphatase 1. The Journal of Immunology, 177, 1636-1645.

Moorman, N. J., Cristea, I.M., Terhune, S.S., Rout, M.P., Chait, B.T. & Shenk,

T. (2008) Human Cytomegalovirus Protein UL38 Inhibits Host Cell Stress Responses by

Antagonizing the Tuberous Sclerosis Protein Complex. Cell Host & Microbe, 3, 253-262.

Morasco, B. J., Sharma, N., Parilla, J. & Flanegan, J.B. (2003) Poliovirus Cre(2C)-Dependent

Synthesis of VPgpUpU is Required for Positive- but Not Negative-Strand RNA

Synthesis. Journal of Virology, 77, 5136-5144.

Morgan, D. G., Ménétret, J.F., Radermacher, M., Neuhof, A., Akey, I.V., Rapoport, T.A. &

Akey, C.W. (2000) A Comparison of the Yeast and Rabbit 80 S Ribosome Reveals the

Topology of the Nascent Chain Exit Tunnel, Inter-Subunit Bridges and Mammalian rRNA

Expansion Segments. Journal of Molecular Biology, 301, 301-321.

Morin, B., Coutard, B., Lelke, M., Ferron, F., Kerber, R., Jamal, S., Frangeul, A., Baronti, C.,

Charrel, R., de Lamballerie, X., Vonrhein, C., Lescar, J., Bricogne, G., Günther, S. & Canard,

315

B. (2010) The N-Terminal Domain of the Arenavirus L Protein is an RNA Endonuclease

Essential in mRNA Transcription. PLoS Pathogens, 6, doi:10.1371/journal.ppat.1001038.

Morton, V. K., Thomas, M.K. & McEwen, S.A. (2015) Estimated Hospitalizations Attributed

to Norovirus and Rotavirus Infection in Canada, 2006-2010. Epidemiology & Infection, 1-10.

Morzunov, S. P., Feldmann, H., Spiropoulou, C.F., Semenova, V.A., Rollin, P.E., Ksiazek,

T.G., Peters, C.J. & Nichol, S.T. (1995) A Newly Recognized Virus Associated with a Fatal

Case of Hantavirus Pulmonary Syndrome in Louisiana. Journal of Virology, 69, 1980-1983.

Moussaieff, A., Rouleau, M., Kitsberg, D., Cohen, M., Levy, G., Barasch, D., Nemirovski, A.,

Shen-Orr, S., Laevsky, I., Amit, M., Bomze, D., Elena-Herrmann, B., Scherf, T., Nissim-

Rafinia, M., Kempa, S., Itskovitz-Eldor, J., Meshorer, E., Aberdam, D. & Nahmias, Y. (2015)

Glycolysis-Mediated Changes in Acetyl-CoA and Histone Acetylation Control the Early

Differentiation of Embryonic Stem Cells. Cell Metabolism, 21, 392-402.

Mukherjee, N., Corcoran, D.L., Nusbaum, J.D., Reid, D.W., Georgiev, S., Hafner, M., Ascano

Jr., M., Tuschl, T., Ohler, U. & Keene, J.D. (2011) Integrative Regulatory Mapping Indicates

that the RNA-Binding Protein HuR Couples Pre-mRNA Processing and mRNA

Stability. Molecular Cell, 43, 327-339.

Murray, K. E. & Barton, D.J. (2003) Poliovirus CRE-Dependent VPg Uridylylation is Required

for Positive-Strand RNA Synthesis but Not for Negative-Strand RNA Synthesis. Journal of

Virology, 77, 4739-4750.

Naidu, S., Vijayan, V., Santoso, S., Kietzmann, T. & Immenschuh, S (2009) Inhibition and

Genetic Deficiency of p38 MAPK Up-Regulates Heme Oxygenase-1 Gene Expression via Nrf2.

The Journal of Immunology, 182, 7048-7057.

Nairz, M., Schroll, A., Moschen, A.R., Sonnweber, T., Theurl, M., Theurl, I., Taub, N.,

Jamnig, C., Neurauter, D., Huber, L.A., Tilg, H., Moser, P.L. & Weiss,

G. (2011) Erythropoietin Contrastingly Affects Bacterial Infection and Experimental Colitis by

Inhibiting Nuclear Factor-κB-Inducible Immune Pathways. Immunity, 34, 61-74.

Najafov, A., Sommer, E.M., Axten, J.M., Deyoung, M.P. & Alessi,

D.R. (2011) Characterization of GSK2334470, a Novel and Highly Specific Inhibitor of

PDK1. The Biochemical Journal, 433, 357-369.

Nascimento, E. B., Fodor, M., van der Zon, G.C., Jazet, I.M., Meinders, A.E., Voshol, P.J.,

Vlasblom, R., Baan, B., Eckel, J., Maassen, J.A., Diamant, M. & Ouwens,

D.M. (2006) Insulin-Mediated Phosphorylation of the Proline-Rich Akt Substrate PRAS40 is

Impaired in Insulin Target Tissues of High-Fat Diet-Fed Rats. Diabetes, 55, 3221-3228.

NCBI Resource Coordinators (2013) Database Resources of the National Center for

Biotechnology Information. Nucleic Acids Research, 41 (Database Issue), D8-D20.

316

Neill, J. D., Reardon, I.M. & Heinrikson, R.L. (1991) Nucleotide Sequence and Expression of

the Capsid Protein Gene of Feline Calicivirus. Journal of Virology, 65, 5440-5447.

Niedzwiecka, A., Marcotrigiano, J., Stepinski, J., Jankowska-Anyszka, M., Wyslouch-

Cieszynska, A., Dadlez, M., Gingras, A.C., Mak, P., Darzynkiewicz, E., Sonenberg, N., Burley,

S.K. & Stolarski, R. (2002) Biophysical Studies of eIF4E Cap-Binding Protein: Recognition of

mRNA 5' Cap Structure and Synthetic Fragments of eIF4G and 4E-BP1 Proteins. Journal of


Nika, J., Yang, W., Pavitt, G.D., Hinnebusch, A.G. & Hannig, E.M. (2000) Purification and

Kinetic Analysis of eIF2B from Saccharomyces cerevisiae. The Journal of Biological

Chemistry, 275, 26011-26017.

Ninomiya-Tsuji, J., Kajino, T., Ono, K., Ohtomo, T., Matsumoto, M., Shiina, M., Mihara, M.,

Tsuchiya, M. & Matsumoto, K. (2003) A Resorcylic Acid Lactone, 5Z-7-Oxozeaenol, Prevents

Inflammation by Inhibiting the Catalytic Activity of TAK1 MAPK Kinase Kinase. The Journal of


Nishiyama, T., Yamamoto, H., Shibuya, N., Hatakeyama, Y., Hachimori, A., Uchiumi, T. &

Nakashima, N. (2003) Structural Elements in the Internal Ribosome Entry Site of Plautia Stali

Intestine Virus Responsible for Binding with Ribosomes. Nucleic Acids Research, 31, 2434-

2442.

Norrmén, C., Figlia, G., Lebrun-Julien, F., Pereira, J.A., Trötzmüller, M., Köfeler, H.C.,

Rantanen, V., Wessig, C., van Deijk, A.L., Smit, A.B., Verheijen, M.H., Rüegg , M.A., Hall,

M.N. & Suter, U. (2014) mTORC1 Controls PNS Myelination Along the mTORC1-RXRγ-SREBP-

Lipid Biosynthesis Axis in Schwann Cells. Cell Reports, 9, 646-660.

Oh, C. D., Kim, S.J., Ju, J.W., Song, W.K., Kim, J.H., Yoo, Y.J. & Chun,

J.S. (2001b) Immunosuppressant Rapamycin Inhibits Protein Kinase C α and p38 Mitogen-

Activated protein kinase Leading to the Inhibition of Chondrogenesis. European Journal of


Ohlmann, T., Lopez-Lastra, M. & Darlix, J.L. (2000) An Internal Ribosome Entry Segment

Promotes Translation of the Simian Immunodeficiency Virus Genomic RNA. The Journal of


Oldfield, S. & Proud, C.G. (1992) Purification, Phosphorylation and Control of the Guanine-

Nucleotide-Exchange Factor from Rabbit Reticulocyte Lysates. European Journal of

Biochemistry, 208, 73-81.

Olmsted, E. A., O'Brien, L., Henshaw, E.C. & Panniers, R. (1993) Purification and

Characterization of Eukaryotic Initiation Factor (eIF)-2α kinases from Ehrlich Ascites Tumor


317

Omura, T., Yoshiyama, M., Izumi, Y., Kim, S., Matsumoto, R., Enomoto, S., Kusuyama, T.,

Nishiya, D., Nakamura, Y., Akioka, K., Iwao, H., Takeuchi, K. & Yoshikawa,

J. (2005) Involvement of c-Jun NH2 Terminal Kinase and p38MAPK in Rapamycin-Mediated

Inhibition of Neointimal Formation in Rat Carotid Arteries. Journal of Cardiovascular


Orth, J. D., Thiele, I. & Palsson, B.Ø. (2010) What is Flux Balance Analysis? Nature

Biotechnology, 28, 245-248.

Ouwens, D. M., de Ruiter, N.D., van der Zon, G.C., Carter, A.P., Schouten, J., van der Burgt,

C., Kooistra, K., Bos, J.L., Maassen, J.A. & van Dam, H. (2002) Growth Factors can Activate

ATF2 Via a Two-Step Mechanism: Phosphorylation of Thr71 through the Ras-MEK-ERK

Pathway and of Thr69 through RalGDS-Src-p38. The EMBO Journal, 21, 3782-3793.

Överby, A. K., Popov, V.L., Niedrig, M. & Weber, F. (2010) Tick-Borne Encephalitis Virus

Delays Interferon Induction and Hides its Double-Stranded RNA in Intracellular Membrane

Vesicles. Journal of Virology, 84, 8470-8483.

Owen, C. R., Kumar, R., Zhang, P., McGrath, B.C., Cavener, D.R. & Krause, G.S. (2005) PERK

is Responsible for the Increased Phosphorylation of eIF2α and the Severe Inhibition of

Protein Synthesis After Transient Global Brain Ischemia. Journal of Neurochemistry, 94,

1235-1242.

Ozbudak, E. M., Thattai, M., Kurtser, I., Grossman, A.D. & van Oudenaarden,

A. (2002) Regulation of Noise in the Expression of a Single Gene. Nature Genetics, 31, 69-73.

Özeş, A. R., Feoktistova, K., Avanzino, B.C. & Fraser, C.S. (2011) Duplex Unwinding and

ATPase Activities of the DEAD-Box Helicase eIF4A are Coupled by eIF4G and eIF4B. Journal of


Padula, P. J., Edelstein, A., Miguel, S.D., López, N.M., Rossi, C.M. & Rabinovich,

R.D. (1998) Hantavirus Pulmonary Syndrome Outbreak in Argentina: Molecular Evidence for

Person-to-Person Transmission of Andes Virus. Virology, 241, 323-330.

Palacios, G., Savji, N., Hui, J., Travassos da Rosa, A., Popov, V., Briese, T., Tesh, R. & Lipkin,

W.I. (2010) Genomic and Phylogenetic Characterization of Merino Walk Virus, a Novel

Arenavirus Isolated in South Africa. The Journal of General Virology, 91, 1315-1324.

Palamarchuk, A., Efanov, A., Maximov, V., Aqeilan, R.I., Croce, C.M. & Pekarsky,

Y. (2005) Akt Phosphorylates and Regulates Pdcd4 Tumor Suppressor Protein. Cancer

Research, 65, 11282-11286.

Panniers, R., Rowlands, A.G. & Henshaw, E.C. (1988) The Effect of Mg2+ and Guanine

Nucleotide Exchange Factor on the Binding of Guanine Nucleotides to Eukaryotic Initiation

Factor 2. The Journal of Biological Chemistry, 263, 5519-5525.

318

Paraskevas, S., Aikin, R., Maysinger, D., Lakey, J.R., Cavanagh, T.J., Hering, B., Wang, R. &

Rosenberg, L. (1999) Activation and Expression of ERK, JNK, and p38 MAP-Kinases in

Isolated Islets of Langerhans: Implications for Cultured Islet Survival. FEBS Letters, 455, 203-

208.

Park, H. S., Kim, M.S., Huh, S.H., Park, J., Chung, J., Kang, S.S. & Choi, E.J. (2002) Akt

(Protein Kinase B) Negatively Regulates SEK1 by Means of Protein Phosphorylation. The


Parodi, A. S., Greenway, D.J., Rugiero, H.R., Frigerio, M., de la Barrera, J.M., Mettler, N.,

Garzon, F., Boxaca, M., Guerrero, L. & Nota, N. (1958) Concerning the Epidemic Outbreak in

Junin. El Día médico, 30, 2300-2301.

Passmore, L. A., Schmeing, T.M., Maag, D., Applefield, D.J., Acker, M.G., Algire, M.A.,

Lorsch, J.R. & Ramakrishnan, V. (2007) The Eukaryotic Translation Initiation Factors eIF1

and eIF1A Induce an Open Conformation of the 40S Ribosome. Molecular Cell, 26, 41-50.

Pastore, S., Mascia, F., Mariotti, F., Dattilo, C., Mariani, V. & Girolomoni, G. (2005) ERK1/2

Regulates Epidermal Chemokine Expression and Skin Inflammation. The Journal of

Immunology, 174, 5047-5056.

Patel, M. M., Widdowson, M.-A., Glass, R.I., Akazawa, K., Vinjé, J. & Parashar,

U.D. (2008) Systematic Literature Review of Role of Noroviruses in Sporadic

Gastroenteritis. Emerging Infectious Diseases, 14, 1224-1231.

Pattyn, F., Perichon, L., Durand, G., Favier, L., Gagliardini, O., Hindmarsh, R.C.A., Zwinger,

T., Albrecht, T., Cornford, S., Docquier, D., Fürst, J.J., Goldberg, D., Gudmundsson, G.H.,

Humbert, A., Hütten, M., Huybrechts, P., Jouvet, G., Kleiner, T., Larour, E., Martin, D.,

Morlighem, M., Payne, A.J., Pollard, D., Rückamp, M., Rybak, O., Seroussi, H., Thoma, M.

& Wilkens, N. (2013) Grounding-Line Migration in Plan-View Marine Ice-Sheet Models:

Results of the Ice2Sea MISMIP3d Intercomparison. Journal of Glaciology, 59, 410-422.

Paulin, F. E. M., Campball, L.E., O'Brien, K., Loughlin, J. & Proud, C.G. (2001) Eukaryotic

Translation Initiation Factor 5 (eIF5) Acts as a Classical GTPase-Activator Protein. Current

Biology, 11, 55-59.

Paulitschke, V., Haudek-Prinz, V., Griss, J., Berger, W., Mohr, T., Pehamberger, H.,

Kunstfeld, R. & Gerner, C. (2013) Functional Classification of Cellular Proteome Profiles

Support the Identification of Drug Resistance Signatures in Melanoma Cells. Journal of

Proteome Research, 12, 3264-3276.

Payne, D. M., Rossomando, A.J., Martino, P., Erickson, A.K., Her, J.H., Shabanowitz, J.,

Hunt, D.F., Weber, M.J. & Sturgill, T.W. (1991) Identification of the Regulatory

Phosphorylation Sites in pp42/Mitogen-Activated Protein Kinase (MAP Kinase). The EMBO

Journal, 10, 885-892.

319

Payne, D. C., Vinjé, J., Szilagyi, P.G., Edwards, K.M., Staat, M.A., Weinberg, G.A., Hall, C.B.,

Chappell, J., Bernstein, D.I., Curns, A.T., Wikswo, M., Shirley, S.H., Hall, A.J., Lopman, B. &

Parashar, U.D. (2013) Norovirus and Medically Attended Gastroenteritis in U.S.

Children. The New England Journal of Medicine, 368, 1121-1130.

Paz, K., Voliovitch, H., Hadari, Y.R., Roberts Jr., C.T., LeRoith, D. & Zick,

Y. (1996) Interaction between the Insulin Receptor and its Downstream Effectors. use of

Individually Expressed Receptor Domains for structure/function Analysis. The Journal of


Pedersen, N. C., Elliot, J.B., Glasgow, A., Poland, A. & Keel, K. (2000) An Isolated Epizootic

of Hemorrhagic-Like Fever in Cats Caused by a Novel and Highly Virulent Strain of Feline

Calicivirus.Veterinary Microbiology, 73, 281-300.

Pederson, T. M., Kramer, D.L. & Rondinone, C.M. (2001) Serine/threonine Phosphorylation

of IRS-1 Triggers its Degradation: Possible Regulation by Tyrosine

Phosphorylation. Diabetes, 50, 24-31.

Pelletier, J., Kaplan, G., Racaniello, V.R. & Sonenberg, N. (1988) Cap-Independent

Translation of Poliovirus mRNA is Conferred by Sequence Elements within the 5' Noncoding

Region. Molecular and Cellular Biology, 8, 1103-1112.

Pelletier, J. & Sonenberg, N. (1988) Internal Initiation of Translation of Eukaryotic mRNA

Directed by a Sequence Derived from Poliovirus RNA. Nature, 334, 320-325.

Pérez-Sautu, U., Sano, D., Guix, S., Kasimir, G., Pintó, R.M. & Bosch, A. (2012) Human

Norovirus Occurrence and Diversity in the Llobregat River Catchment, Spain. Environmental

Microbiology, 14, 494-502.

Pervin, S., Singh, R., Freije, W.A. & Chaudhuri, G. (2003) MKP-1-Induced Dephosphorylation

of Extracellular Signal-Regulated Kinase is Essential for Triggering Nitric Oxide-Induced

Apoptosis in Human Breast Cancer Cell Lines: Implications in Breast Cancer. Cancer

Research, 63, 8853-8860.

Pesavento, P. A., Chang, K.-O. & Parker, J.S.L. (2008) Molecular Virology of Feline

Calicivirus. Veterinary Clinics of North America: Small Animal Practice, 38, 775-786.

Pestova, T. V., Lomakin, I.B., Lee, J.H., Choi, S.K., Dever, T.E. & Hellen, C.U. (2000) The

Joining of Ribosomal Subunits in Eukaryotes Requires eIF5B. Nature, 403, 332-335.

Peters, C. J. (2006) Emerging Infections: Lessons from the Viral Hemorrhagic

Fevers. Transactions of the American Clinical and Climatological Association, 117, 189-196.

Peterson, R.T., Desai, B.N., Hardwick, J.S. & Schreiber, S.L. (1999) Protein Phosphatase 2A

Interacts with the 70-kDa S6 Kinase and is Activated by Inhibition of FKBP12–Rapamycin

320

Associated Protein. The Proceedings of the National Academy of Sciences of the United


Petri, C.A. (1962) Kommunikation mit Automaten. PhD Thesis. Universität Hamburg.

Available at: http://edoc.sub.uni-hamburg.de/informatik/volltexte/2011/160/ 31/01/2011

(Accessed: 11/10/2014).

Pham, F.H., Sugden, P.H. & Clerk, A. (2000) Regulation of Protein Kinase B and 4E-BP1 by

Oxidative Stress in Cardiac Myocytes. Circulation Research, 86, 1252-1258.

Pisarev, A. V., Hellen, C.U.T. & Pestova, T.V. (2007) Recycling of Eukaryotic Posttermination

Ribosomal Complexes. Cell, 131, 286-299.

Pisarev, A. V., Skabkin, M.A., Pisareva, V.P., Skabkina, O.V., Rakotondrafara, A.M., Hentze,

M.W., Hellen, C.U. & Pestova, T.V. (2010) The Role of ABCE1 in Eukaryotic Posttermination

Ribosomal Recycling. Molecular Cell, 37, 196-210.

Plotch, S. J., Bouloy, M., Ulmanen, I. & Krug, R.M. (1981) A Unique Cap(m7GpppXm)-

Dependent Influenza Virion Endonuclease Cleaves Capped RNAs to Generate the Primers

that Initiate Viral RNA Transcription. Cell, 23, 847-858.

Plyusnin, A., Vapalahti, O. & Vaheri, A. (1996) Hantaviruses: Genome Structure, Expression

and Evolution. The Journal of General Virology, 77, 2677-2687.

Poltorak, A., He, X., Smirnova, I., Liu, M.Y., Van Huffel, C., Du, X., Birdwell, D., Alejos, E.,

Silva, M., Galanos, C., Freudenberg, M., Ricciardi-Castagnoli, P., Layton, B. & Beutler,

B. (1998) Defective LPS Signaling in C3H/HeJ and C57BL/10ScCr Mice: Mutations in Tlr4

Gene. Science, 282, 2085-2088.

Pontén, F., Gry, M., Fagerberg, L., Lundberg, E., Asplund, A., Berglund, L., Oksvold, P.,

Björling, E., Hober, S., Kampf, C., Navani, S., Nilsson, P., Ottosson, J., Persson, A.,

Wernérus, H., Wester, K. & Uhlén, M. (2009) A Global View of Protein Expression in Human

Cells, Tissues, and Organs. Molecular Systems Biology, 5, doi: 10.1038/msb.2009.93.

Potter, C. J., Pedraza, L.G. & Xu, T. (2002) Akt Regulates Growth by Directly Phosphorylating

Tsc2. Nature Cell Biology, 4, 658-665.

Press, W.H., Teukolsky, S.A., Vetterling, W.T. & Flannery, B.P (2007) Numerical Recipes 3rd

Editions – The Art of Scientific Computing. Cambridge University Press, Cambridge, Cambs

(UK).

Puchałka, J. & Kierzek, A.M. (2004) Bridging the Gap Between Stochastic and Deterministic

Regimes in the Kinetic Simulations of the Biochemical Reaction Networks. Biophysical

Journal, 86, 1357-1372.

321

Puustinen, P., Rajamäki, M.-L., Ivanov, K.I., Valkonen, J.P.T. & Mäkinen,

K. (2002) Detection of the Potyviral Genome-Linked Protein VPg in Virions and its

Phosphorylation by Host Kinases. Journal of Virology, 76, 12703-12711.

Pyronnet, S., Imataka, H., Gingras, A.-C., Fukunga, R., Hunter, T. & Sonenberg,

N. (1999) Human Eukaryotic Translation Initiation Factor 4G (eIF4G) Recruits Mnk1 to

Phosphorylate eIF4E. The EMBO Journal, 18, 270-279.

Qi, D., Huang, S., Miao, R., She, Z.G., Quinn, T., Chang, Y., Liu, J., Fan, D., Chen, Y.E. & Fu,

M. (2011) Monocyte Chemotactic Protein-Induced Protein 1 (MCPIP1) Suppresses Stress

Granule Formation and Determines Apoptosis Under Stress. The Journal of Biological

Chemistry, 286, 41692-41700.

Qi, X., Lan, S., Wang, W., Schelde, L.M., Dong, H., Wallat, G.D., Ly, H., Liang, Y. & Dong,

C. (2010) Cap Binding and Immune Evasion Revealed by Lassa Nucleoprotein

Structure. Nature, 468, 779-783.

Quek, L. E., Dietmair, S., Hanscho, M., Martínez, V.S., Borth, N. & Nielsen,

L.K. (2014) Reducing Recon 2 for Steady-State Flux Analysis of HEK Cell Culture. Journal of


Quenouille, M. H. (1949) Approximate Tests of Correlation in Time-Series. Journal of the

Royal Statistical Society. Series B (Methodological), 11, 68-84.

R Core Development Team (2013) The R Foundation of Statistical Computing. R: A language

and Environment for Statistical Computing.

Rad, E., Dodd, K.M., Thomas, L.E., Upadhyaya, M. & Tee, A.R. (2015) STAT3 and HIF1α

Signalling Drives Oncogenic Cellular Phenotypes in Malignant Peripheral Nerve Sheath

Tumors (MPNSTs). Molecular Cancer Research, 13, 1149-1160.

Raghunathan, A., Shin, S. & Daefler, S. (2010) Systems Approach to Investigating Host-

Pathogen Interactions in Infections with the Biothreat Agent Francisella. Constraints-Based

Model of Francisella tularensis. BMC Systems Biology, 4, doi: 10.1186/1752-0509-4-118.

Raingeaud, J., Gupta, S., Rogers, J.S., Dickens, M., Han, J., Ulevitch, R.J. & Davis,

R.J. (1995) Pro-Inflammatory Cytokines and Environmental Stress Cause p38 Mitogen-

Activated Protein Kinase Activation by Dual Phosphorylation on Tyrosine and Threonine. The


Raingeaud, J., Whitmarsh, A.J., Barrett, T., Derijard, B. & Davis, R.J. (1996) MKK3- and

MKK6-Regulated Gene Expression is Mediated by the p38 Mitogen-Activated Protein Kinase

Signal Transduction Pathway. Molecular and Cellular Biology, 16, 1247-1255.

Raju, R., Raju, L., Hacker, D., Garcin, D., Compans, R. & Kolakofsky,

D. (1990) Nontemplated Bases at the 5' Ends of Tacaribe Virus mRNAs. Virology, 174, 53-59.

322

Ramaiah, K. V., Davies, M.V., Chen, J.J. & Kaufman, R.J. (1994) Expression of Mutant

Eukaryotic Initiation Factor 2 α subunit (eIF-2α) Reduces Inhibition of Guanine Nucleotide

Exchange Activity of eIF-2B Mediated by eIF-2α Phosphorylation. Molecular and Cellular

Biology, 14, 4546-4553.

Raman, K. & Chandra, N. (2009) Flux Balance Analysis of Biological Systems: Applications

and Challenges. Briefings in Bioinformatics, 10, 435-449.

Ramirez, M., Wek, R.C., Vazquez de Aldana, C.R., Jackson, B.M., Freeman, B. &

Hinnebusch, A.G. (1992) Mutations Activating the Yeast eIF-2α kinase GCN2: Isolation of

Alleles Altering the Domain Related to Histidyl-tRNA Synthetases. Molecular and Cellular

Biology, 12, 5801-5815.

Rangamani, P. & Iyengar, R. (2008) Modelling Cellular Signalling Systems. Essays in


Ranganathan, A., Pearson, G.W., Chrestensen, C.A., Sturgill, T.W. & Cobb, M.H. (2006) The

MAP Kinase ERK5 Binds to and Phosphorylates p90 RSK. Archives of Biochemistry and

Biophysics, 449, 8-16.

Rani, P. B., Kumar, B.S., Rao, A.K., Sreenivasrao, S. & Narasu, M.L. (2013) Recovery and

Purification of Rapamycin from the Culture Broth of Mtcc 5681. Pakistan Journal of

Biological Sciences, 16, 219-225.

Rao, G. N. (2000) Oxidant Stress Stimulates Phosphorylation of eIF4E without an Effect on

Global Protein Synthesis in Smooth Muscle Cells. The Journal of Biological Chemistry, 275,

16993-16999.

Rasband, W.S. (1997-2012) ImageJ, U. S. National Institutes of Health, Bethesda, Maryland,

USA.

Raschke, W. C., Baird, S., Ralph, P. & Nakoinz, I. (1978) Functional Macrophage Cell Lines

Transformed by Abelson Leukemia Virus. Cell, 15, 261-267.

Rasschaert, D., Huguet, S., Madelaine, M.-F. & Vautherot, J.F. (1994) Sequence and

Genomic Organization of a Rabbit Hemorrhagic Disease Virus Isolated from a Wild

Rabbit. Virus Genes, 9, 121-132.

Ravkov, E. V., Rollin, P.E., Ksiazek, T.G., Peters, C.J. & Nichol, S.T. (1995) Genetic and

Serologic Analysis of Black Creek Canal Virus and its Association with Human Disease and

Sigmodon Hispidus Infection. Virology, 210, 482-489.

Ray, B. K., Lawson, T.G., Kramer, J.C., Cladaras, M.H., Grifo, J.A., Abramson, R.D., Merrick,

W.C. & Thach, R.E. (1985) ATP-Dependent Unwinding of Messenger RNA Structure by

Eukaryotic Initiation Factors . The Journal of Biological Chemistry, 260, 7651-7658.

323

Raychaudhuri, P., Stringer, E.A., Valenzuela, D.M. & Maitra, U. (1984) Ribosomal Subunit

Antiassociation Activity in Rabbit Reticulocyte Lysates. Evidence for a Low Molecular Weight

Ribosomal Subunit Antiassociation Protein Factor (Mr = 25,000). The Journal of Biological

Chemistry, 259, 11930-11935.

Reardon, D.B., Contessa, J.N., Mikkelsen, R.B., Valerie, K., Amir, C., Dent, P. & Schmidt-

Ullrich, R.K. (1999) Dominant Negative EGFR-CD533 and Inhibition of MAPK Modify JNK1

Activation and Enhance Radiation Toxicity of Human Mammary Carcinoma Cells. Oncogene,

18, 4756-4766.

Rebsamen, M., Pochini, L., Stasyk, T., de Araújo, M.E., Galluccio, M., Kandasamy, R.K.,

Snijder, B., Fauster, A., Rudashevskaya, E.L., Bruckner, M., Scorzoni, S., Filipek, P.A.,

Huber, K.V., Bigenzahn, J.W., Heinz, L.X., Kraft, C., Bennett, K.L., Indiveri, C., Huber, L.A. &

Superti-Furga, G. (2015) SLC38A9 is a Component of the Lysosomal Amino Acid Sensing

Machinery that Controls mTORC1. Nature, 519, 477-481.

Reguera, J., Weber, F. & Cusack, S. (2010) Bunyaviridae RNA Polymerases (L-Protein) have

an N-Terminal, Influenza-Like Endonuclease Domain, Essential for Viral Cap-Dependent

Transcription. PLoS Pathogens, 6, doi: 10.1371/journal.ppat.1001101.

Repp, K. K. & Keene, W.E. (2012) A Point-Source Norovirus Outbreak Caused by Exposure to

Fomites. The Journal of Infectious Diseases, 205, 1639-1641.

Reuveni, S., Meilijson, I., Kupiec, M., Ruppin, E. & Tuller, T. (2011) Genome-Scale Analysis

of Translation Elongation with a Ribosome Flow Model. PLoS Computational Biology, 7,

doi:10.1371/journal.pcbi.1002127.

Rhoads, R. E. (1999) Signal Transduction Pathways that Regulate Eukaryotic Protein

Synthesis. The Journal of Biological Chemistry, 274, 30337-30340.

Richter, N. J., Rogers Jr., G.W., Hensold, J.O. & Merrick, W.C. (1999) Further Biochemical

and Kinetic Characterization of Human Eukaryotic Initiation Factor 4H. The Journal of


Richter-Cook, N. J., Dever, T.E., Hensold, J.O. & Merrick, W.C. (1998) Purification and

Characterization of a New Eukaryotic Protein Translation Factor. Eukaryotic Initiation Factor

4H. The Journal of Biological Chemistry, 273, 7579-7587.

Riechmann, J. L., Laín, S. & García, J.A. (1989) The Genome-Linked Protein and 5′ End RNA

Sequence of Plum Pox Potyvirus. The Journal of General Virology, 70, 2785-2789.

Rijnbrand, R., van der Straaten, T., van Rijn, P.A., Spaan, W.J. & Bredenbeek,

P.J. (1997) Internal Entry of Ribosomes is Directed by the 5' Noncoding Region of Classical

Swine Fever Virus and is Dependent on the Presence of an RNA Pseudoknot Upstream of the

Initiation Codon. Journal of Virology, 71, 451-457.

324

Robbins, D. J., Zhen, E., Owaki, H., Vanderbilt, C.A., Ebert, D., Geppert, T.D. & Cobb,

M.H. (1993) Regulation and Properties of Extracellular Signal-Regulated Protein Kinases 1

and 2 in Vitro. The Journal of Biological Chemistry, 268, 5097-5106.

Roberts, L. O., Al-Molawi, N., Carter, M.J. & Kass, G.E. (2003) Apoptosis in Cultured Cells

Infected with Feline Calicivirus. Annals of the New York Academy of Sciences, 1010, 587-590.

Robichaud, N., Del Rincon, S.V., Huor, B., Alain, T., Petruccelli, L.A., Hearnden, J.,

Goncalves, C., Grotegut, S., Spruck, C.H., Furic, L., Larsson, O., Muller, W.J., Miller, W.H. &

Sonenberg, N. (2015) Phosphorylation of eIF4E Promotes EMT and Metastasis Via

Translational Control of SNAIL and MMP-3. Oncogene, 34, 2032-2042.

Robitaille, A. M., Christen, S., Shimobayashi, M., Cornu, M., Fava, L.L., Moes, S.,

Prescianotto-Baschong, C., Sauer, U., Jenoe, P. & Hall, M.N. (2013) Quantitative

Phosphoproteomics Reveal mTORC1 Activates De Novo Pyrimidine Synthesis. Science, 339,

1320-1323.

Roch, S. & Starke, P.H. (2003) Integrated Net Analyzer INA (Version 2.2). Lehrstuhl für

Automaten- und Systemtheorie, Institut für Informatik, Humboldt-Universität zu Berlin (DE).

http://www2.informatik.hu-berlin.de/lehrstuehle/automaten/ina/

Rockx, B., de Wit, M., Vennema, H., Vinjé, J., de Bruin, E., van Duynhoven, Y. & Koopmans,

M. (2002) Natural History of Human Calicivirus Infection: A Prospective Cohort

Study. Clinical Infectious Diseases, 35, 246-253.

Rodríguez, M. E., Brunetti, J.E., Wachsman, M.B., Scolaro, L.A. & Castilla,

V. (2014) Raf/MEK/ERK Pathway Activation is Required for Junin Virus Replication. The

Journal of General Virology, 95, 799-805.

Rogers Jr., G. W., Lima, W.F. & Merrick, M.C. (2001a) Further Characterization of the

Helicase Activity of eIF4A: Substrate Specificity. The Journal of Biological Chemistry, 276,

12598-12608.

Rogers Jr., G. W., Richter, N.J. & Merrick, W.C. (1999) Biochemical and Kinetic

Characterization of the RNA Helicase Activity of Eukaryotic Initiation Factor 4A. The Journal


Rogers Jr., W., Richter, N.J., Lima, W.F. & Merrick, W.C. (2001b) Modulation of the Helicase

Activity of eIF4A by eIF4B, eIF4H, and eIF4F. The Journal of Biological Chemistry, 276, 30914-

30922.

Roget, K., Ben-Addi, A., Mambole-Dema, A., Gantke, T., Yang, H.-T., Janzen, J., Morrice, N.,

Abbott, D. & Ley, S.C. (2012) IκB Kinase 2 Regulates TPL-2 Activation of Extracellular Signal-

Regulated Kinases 1 and 2 by Direct Phosphorylation of TPL-2 Serine 400. Molecular and


325

Rogozin, I. B., Kochetov, A.V., Kondrashov, F.A., Koonin, E.V. & Milanesi,

L. (2001) Presence of ATG Triplets in 5' Untranslated Regions of Eukaryotic cDNAs Correlates

with a 'Weak' Context of Start Codon. Bioinformatics, 17, 890-900.

Rohr, C., Marwan, W. & Heiner, M. (2010) Snoopy--a Unifying Petri Net Framework to

Investigate Biomolecular Networks. Bioinformatics, 26, 974-975.

Rolli-Derkinderen, M., Machavoine, F., Baraban, J.M., Grolleau, A., Beretta, L. & Dy,

M. (2003) ERK and p38 Inhibit the Expression of 4E-BP1 Repressor of Translation through

Induction of Egr-1. The Journal of Biological Chemistry, 278, 18859-18867.

Romano, P. R., Garcia-Barrio, M.T., Zhang, X., Wang, Q., Taylor, D.R., Zhang, F., Herring, C.,

Mathews, M.B., Qin, J. & Hinnebusch, A.G. (1998) Autophosphorylation in the Activation

Loop is Required for Full Kinase Activity in Vivo of Human and Yeast Eukaryotic Initiation

Factor 2α kinases PKR and GCN2. Molecular and Cellular Biology, 18, 2282-2297.

Romano, P. R., Green, S.R., Barber, G.N., Mathews, M.B. & Hinnebusch,

A.G. (1995) Structural Requirements for Double-Stranded RNA Binding, Dimerization, and

Activation of the Human eIF-2α kinase DAI in Saccharomyces cerevisiae. Molecular and


Rousseau, S., Papoutsopoulou, M., Symons, A., Cook, D., Lococq, J.M., Prescott, A.R.,

O’Garra, A., Ley, S.C. & Cohen, P. (2008) TPL2-Mediated Activation of ERK1 and ERK2

Regulates the Processing of Pre-TNFα in LPS-Stimulated Macrophages. Journal of Cell

Science, 121, 149-154.

Roux, P. P., Ballif, B.A., Anjum, R., Gygi, S.P. & Blenis, J. (2004) Tumor-Promoting Phorbol

Esters and Activated Ras Inactivate the Tuberous Sclerosis Tumor Suppressor Complex Via

p90 Ribosomal S6 Kinase. The Proceedings of the National Academy of Sciences of the


Royall, E., Doyle, N., Abdul-Wahab, A., Emmott, E., Morley, S.J., Goodfellow, I., Roberts,

L.O. & Locker, N. (2015) Murine Norovirus 1 (MNV1) Replication Induces Translational

Control of the Host by Regulating eIF4E Activity during Infection. The Journal of Biological

Chemistry, 290, 4748-4758.

Rozovsky, N., Butterworth, A.C. & Moore, A.J. (2008) Interactions between eIF4AI and its

Accessory Factors eIF4B and eIF4H. RNA, 14, 2136-2148.

Ruggieri, A., Dazert, E., Metz, P., Hofmann, S., Bergeest, J.P., Mazur, J., Bankhead, P., Hiet,

M.S., Kallis, S., Alvisi, G., Samuel, C.E., Lohmann, V., Kaderali, L., Rohr, K., Frese, M.,

Stoecklin, G. & Bartenschlager, R. (2012) Dynamic Oscillation of Translation and Stress

Granule Formation Mark the Cellular Response to Virus Infection. Cell Host & Microbe, 12,

71-85.

326

Ruths, D., Muller, M., Tseng, J.T., Nakhleh, L. & Ram, P.T. (2008) The Signaling Petri Net-

Based Simulator: A Non-Parametric Strategy for Characterizing the Dynamics of Cell-Specific

Signaling Networks. PLoS Computational Biology, 4, doi: 10.1371/journal.pcbi.1000005.

Sabapathy, K., Hochedlinger, K., Nam, S.Y., Bauer, A., Karim, M. & Wagner, E.F. (2004)

Distinct Roles for JNK1 and JNK2 in Regulating JNK Activity and c-Jun-Dependent Cell

Proliferation. Molecular Cell, 15, 713-725.

Sabatini, D. M., Erdjument-Bromage, H., Lui, M., Tempst, P. & Snyder, S.H. (1994) RAFT1: A

Mammalian Protein that Binds to FKBP12 in a Rapamycin-Dependent Fashion and is

Homologous to Yeast TORs. Cell, 78, 35-43.

Sabers, C. J., Martin, M.M., Brunn, G.J., Williams, J.M., Dumont, F.J., Wiederrecht, G. &

Abraham, R.T. (1995) Isolation of a Protein Target of the FKBP12-Rapamycin Complex in

Mammalian Cells. The Journal of Biological Chemistry, 270, 815-822.

Sachs, A. B., Bond, M.W. & Kornberg, R.D. (1986) A Single Gene from Yeast for both

Nuclear and Cytoplasmic Polyadenylate-Binding Proteins: Domain Structure and

Expression. Cell, 45, 827-835.

Sahin, U., Karikó, K. & Türeci, O. (2014) mRNA-Based Therapeutics - Developing a New Class

of Drugs. Nature Reviews Drug Discovery, 13, 759-780.

Saini, A. K., Nanda, J.S., Lorsch, J.R. & Hinnebusch, A.G. (2010) Regulatory Elements in

eIF1A Control the Fidelity of Start Codon Selection by Modulating tRNAiMet Binding to the

Ribosome. Genes & Development, 24, 97-110.

Saito, K., Araki, Y., Kontani, K., Nishina, H. & Katada, T. (2005) Novel Role of the Small

GTPase Rheb: Its Implication in Endocytic Pathway Independent of the the Activation of the

Mammalian Target of Rapamycin. The Journal of Biochemistry, 137, 423-430.

Saitoh, M., Kimura, H., Kozawa, K., Nishio, O. & Shoji, A. (2007) Detection and Phylogenetic

Analysis of Norovirus in Corbicula fluminea in a Freshwater River in Japan. Microbiology and


Sancak, Y., Bar-Peled, L., Zoncu, R., Markhard, A.L., Nada, S. & Sabatini,

D.M. (2010) Ragulator-Rag Complex Targets mTORC1 to the Lysosomal Surface and is

Necessary for its Activation by Amino Acids. Cell, 141, 290-303.

Sancak, Y., Peterson, T.R., Shaul, Y.D., Lindquist, R.A., Thoreen, C.C., Bar-Peled, L. &

Sabatini, D.M. (2008) The Rag GTPases Bind Raptor and Mediate Amino Acid Signaling to

mTORC1. Science, 320, 1496-1501.

Sancak, Y., Thoreen, C.C., Peterson, T.R., Lindquist, R.A., Kang, S.A., Spooner, E., Carr, S.A.

& Sabatini, D.M. (2007) PRAS40 is an Insulin-Regulated Inhibitor of the mTORC1 Protein

Kinase. Molecular Cell, 25, 903-915.

327

Sànchez, R. & Marzluff, W.F. (2002) The Stem-Loop Binding Protein is Required for Efficient

Translation of Histone mRNA in Vivo and in Vitro. Molecular and Cellular Biology, 22, 7093-

7104.

San-Juan-Vergara, H., Peeples, M.E., Lockey, R.F. & Mohapatra, S.S. (2004) Protein Kinase

C-α activity is Required for Respiratory Syncytial Virus Fusion to Human Bronchial Epithelial

Cells. Journal of Virology, 78, 13717-13726.

Sarbassov, D. D., Ali, S.M., Sengupta, S., Sheen, J.H., Hsu, P.P., Bagley, A.F., Markhard, A.L.

& Sabatini, D.M. (2006) Prolonged Rapamycin Treatment Inhibits mTORC2 Assembly and

Akt/PKB. Molecular Cell, 22, 159-168.

Saric, J., Li, J.V., Utzinger, J., Wang, Y., Keiser, J., Dirnhofer, S., Beckonert, O., Sharabiani,

M.T., Fonville, J.M., Nicholson, J.K. & Holmes, E. (2010) Systems Parasitology: Effects

of Fasciola hepatica on the Neurochemical Profile in the Rat Brain. Molecular Systems

Biology, 6, doi: 10.1038/msb.2010.49.

Sarközi, R., Miller, B., Pollack, V., Feifel, E., Mayer, G., Sorokin, A. & Schramek, H. (2007)

ERK1/2-Driven and MKP-Mediated Inhibition of EGF-Induced ERK5 Signaling in Human

Proximal Tubular Cells. Journal of Cellular Physiology, 211, 88-100.

Schaffer, F. L., Ehresmann, D.W., Fretz, M.K. & Soergel, M.E. (1980) A Protein, VPg,

Covalently Linked to 36S Calicivirus RNA. The Journal of General Virology, 47, 215-220.

Schalm, S. S., Fingar, D.C., Sabatini, D.M. & Blenis, J. (2003) TOS Motif-Mediated Raptor

Binding Regulates 4E-BP1 Multisite Phosphorylation and Function. Current Biology, 13, 797-

806.

Scheibe, T.D., Mahadevan, R., Fang, Y., Garg, S., Long, P.E. & Lovley, D.R (2009) Coupling a

Genome-Scale Metabolic Model with a Reactive Transport Model to Describe in Situ

Uranium Bioremediation. Microbial Biotechnology, 2, 274-286.

Scheid, M. P. & Woodgett, J.R. (2001) PKB/AKT: Functional Insights from Genetic

Models. Nature Reviews Molecular Cell Biology, 2, 760-768.

Scheper, G. C., Morrice, N.A., Kleijn, M. & Proud, C.G. (2001) The Mitogen-Activated

Protein Kinase Signal-Integrating Kinase Mnk2 is a Eukaryotic Initiation Factor 4E Kinase with

High Levels of Basal Activity in Mammalian Cells. Molecular and Cellular Biology, 21, 743-

754.

Scheufele, F., Wolf, B., Kruse, M., Hartmann, T., Lempart, J., Muehlich, S., Pfeiffer, A.F.,

Field, L.J., Charron, M.J., Pan, Z.Q., Engelhardt, S. & Sarikas, A. (2014) Evidence for a

Regulatory Role of Cullin-RING E3 Ubiquitin Ligase 7 in Insulin Signaling. Cellular

Signalling, 26, 233-239.

328

Schilling, M., Maiwald, T., Hengl, S., Winter, D., Kreutz, C., Kolch, W., Lehmann, W.D.,

Timmer, J. & Klingmuller, U. (2009) Theoretical and Experimental Analysis Links Isoform-

Specific ERK Signalling to Cell Fate Decisions. Molecular Systems Biology, 5, doi:

10.1038/msb.2009.91.

Schloss, P. D. & Westcott, S.L. (2011) Assessing and Improving Methods used in Operational

Taxonomic Unit-Based Approaches for 16S rRNA Gene Sequence Analysis. Applied and

Environmental Microbiology, 77, 3219-3226.

Schorr-Evans, E. M., Poland, A., Johnson, W.E. & Pedersen, N.C. (2003) An Epizootic of

Highly Virulent Feline Calicivirus Disease in a Hospital Setting in New England. Journal of

Feline Medicine & Surgery, 5, 217-226.

Schreiber, K. H., Ortiz, D., Academia, E.C., Anies, A.C., Liao, C.Y. & Kennedy,

B.K. (2015) Rapamycin-Mediated mTORC2 Inhibition is Determined by the Relative

Expression of FK506-Binding Proteins. Aging Cell, 14, 265-273.

Schwarz, E. R., Mando, O.G., Maiztegui, J.I. & Vilches, A.M. (1970) Symptoms and Early

Signs of Major Diagnostic Value in Argentine Hemorrhage Fever. Medicina, 30, 8-14.

Sehgal, S. N., Baker, H. & Vézina, C. (1975) Rapamycin (AY-22,989), a New Antifungal

Antibiotic. II. Fermentation, Isolation and Characterization. The Journal of Antibiotics, 28,

727-732.

Selisko, B., Peyrane, F.F., Canard, B., Alvarez, K. & Decroly, E. (2010) Biochemical

Characterization of the (Nucleoside-2'O)-Methyltransferase Activity of Dengue Virus Protein

NS5 using Purified Capped RNA Oligonucleotides (7Me)GpppAC(n) and GpppAC(n). The

Journal of General Virology, 91, 112-121.

Senger, B., Lafontaine, D.L., Graindorge, J.S., Gadal, O., Camasses, A., Sanni, A., Garnier,

J.M., Breitenbach, M., Hurt, E. & Fasiolo, F. (2001) The Nucle(Ol)Ar Tif6p and Efl1p are

Required for a Late Cytoplasmic Step of Ribosome Synthesis. Molecular Cell, 8, 1363-1373.

Seong, S. R., Lee, J.W., Lee, Y.K., Kim, T.I., Son, D.J., Moon, D.C., Yun, Y.W., Yoon do, Y. &

Hong, J.T. (2006) Stimulation of Cell Growth by Erythropoietin in RAW264.7 Cells:

Association with AP-1 Activation. Archives of Pharmacal Research, 29, 218-223.

Seufferlein, T. & Rozengurt, E. (1996) Rapamycin Inhibits Constitutive

p70s6k Phosphorylation, Cell Proliferation, and Colony Formation in Small Cell Lung Cancer

Cells. Cancer Research, 56, 3895-3897.

Shah, O. J. & Hunter, T. (2006) Turnover of the Active Fraction of IRS1 Involves Raptor-

mTOR- and S6K1-Dependent Serine Phosphorylation in Cell Culture Models of Tuberous

Sclerosis. Molecular and Cellular Biology, 26, 6425-6434.

329

Shah, O. J., Wang, Z. & Hunter, T. (2004) Inappropriate Activation of the

TSC/Rheb/mTOR/S6K Cassette Induces IRS1/2 Depletion, Insulin Resistance, and Cell

Survival Deficiencies. Current Biology, 14, 1650-1656.

Shahbazian, D., Roux, P.P., Mieulet, V., Cohen, M.S., Raught, B., Taunton, J., Hershey, J.W.,

Blenis, J., Pende, M. & Sonenberg, N. (2006) The mTOR/PI3K and MAPK Pathways Converge

on eIF4B to Control its Phosphorylation and Activity. The EMBO Journal, 25, 2781-2791.

Shalem, O., Sanjana, N.E., Hartenian, E., Shi, X., Scott, D.A., Mikkelsen, T.S., Heckl, D.,

Ebert, B.L., Root, D.E., Doench, J.G. & Zhang, F. (2014) Genome-Scale CRISPR-Cas9

Knockout Screening in Human Cells. Science, 343, 84-87.

Sharma, B., Joshi, S., Sassano, A., Majchrzak, B., Kaur, S., Aggarwal, P., Nabet, B., Bulic, M.,

Stein, B.L., McMahon, B., Baker, D.P., Fukunaga, R., Altman, J.K., Licht, J.D., Fish, E.N. &

Platanias, L.C. (2012) Sprouty Proteins are Negative Regulators of Interferon (IFN) Signaling

and IFN-Inducible Biological Responses. The Journal of Biological Chemistry, 287, 42352-

42360.

Shaw, L.M. (2001) Identification of Insulin Receptor Substrate 1 (IRS-1) and IRS-2 as

Signaling Intermediates in the α6β4 Integrin-Dependent Activation of Phosphoinositide 3-

OH Kinase and Promotion of Invasion. Molecular and Cellular Biology, 21, 5082-5093.

Shi, L., Campbell, G., Jones, W.D., Campagne, F., Wen, Z., Walker, S.J., Su, Z., Chu, T.M.,

Goodsaid, F.M., Pusztai, L., Shaughnessy Jr., J.D., Oberthuer, A., Thomas, R.S., Paules, R.S.,

Fielden, M., Barlogie, B., Chen, W., Du, P., Fischer, M., Furlanello, C., Gallas, B.D., Ge, X.,

Megherbi, D.B., Symmans, W.F., Wang, M.D., Zhang, J., Bitter, H., Brors, B., Bushel, P.R.,

Bylesjo, M., Chen, M., Cheng, J., Cheng, J., Chou, J., Davison, T.S., Delorenzi, M., Deng, Y.,

Devanarayan, V., Dix, D.J., Dopazo, J., Dorff, K.C., Elloumi, F., Fan, J., Fan, S., Fan, X., Fang,

H., Gonzaludo, N., Hess, K.R., Hong, H., Huan, J., Irizarry, R.A., Judson, R., Juraeva, D.,

Lababidi, S., Lambert, C.G., Li, L., Li, Y., Li, Z., Lin, S.M., Liu, G., Lobenhofer, E.K., Luo, J.,

Luo, W., McCall, M.N., Nikolsky, Y., Pennello, G.A., Perkins, R.G., Philip, R., Popovici, V.,

Price, N.D., Qian, F., Scherer, A., Shi, T., Shi, W., Sung, J., Thierry-Mieg, D., Thierry-Mieg, J.,

Thodima, V., Trygg, J., Vishnuvajjala, L., Wang, S.J., Wu, J., Wu, Y., Xie, Q., Yousef, W.A.,

Zhang, L., Zhang, X., Zhong, S., Zhou, Y., Zhu, S., Arasappan, D., Bao, W., Lucas, A.B.,

Berthold, F., Brennan, R.J., Buness, A., Catalano, J.G., Chang, C., Chen, R., Cheng, Y., Cui, J.,

Czika, W., Demichelis, F., Deng, X., Dosymbekov, D., Eils, R., Feng, Y., Fostel, J., Fulmer-

Smentek, S., Fuscoe, J.C., Gatto, L., Ge, W., Goldstein, D.R., Guo, L., Halbert, D.N., Han, J.,

Harris, S.C., Hatzis, C., Herman, D., Huang, J., Jensen, R.V., Jiang, R., Johnson, C.D., Jurman,

G., Kahlert, Y., Khuder, S.A., Kohl, M., Li, J., Li, L., Li, M., Li, Q.Z., Li, S., Li, Z., Liu, J., Liu, Y.,

Liu, Z., Meng, L., Madera, M., Martinez-Murillo, F., Medina, I., Meehan, J., Miclaus, K.,

Moffitt, R.A., Montaner, D., Mukherjee, P., Mulligan, G.J., Neville, P., Nikolskaya, T., Ning,

B., Page, G.P., Parker, J., Parry, R.M., Peng, X., Peterson, R.L., Phan, J.H., Quanz, B., Ren, Y.,

Riccadonna, S., Roter, A.H., Samuelson, F.W., Schumacher, M.M., Shambaugh, J.D., Shi, Q.,

330

Shippy, R., Si, S., Smalter, A., Sotiriou, C., Soukup, M., Staedtler, F., Steiner, G., Stokes,

T.H., Sun, Q., Tan, P.Y., Tang, R., Tezak, Z., Thorn, B., Tsyganova, M., Turpaz, Y., Vega, S.C.,

Visintainer, R., von Frese, J., Wang, C., Wang, E., Wang, J., Wang, W., Westermann, F.,

Willey, J.C., Woods, M., Wu, S., Xiao, N., Xu, J., Xu, L., Yang, L., Zeng, X., Zhang, J., Zhang,

L., Zhang, M., Zhao, C., Puri, R.K., Scherf, U., Tong, W., Wolfinger, R.D. & MAQC

Consortium (2010) The MicroArray Quality Control (MAQC)-II Study of Common Practices

for the Development and Validation of Microarray-Based Predictive Models. Nature


Shi, L., Chowdhury, S.M., Smallwood, H.S., Yoon, H., Mottaz-Brewer, H.M., Norbeck, A.D.,

McDermott, J.E., Clauss, T.R., Heffron, F., Smith, R.D. & Adkins, J.N. (2009) Proteomic

Investigation of the Time Course Responses of RAW 264.7 Macrophages to Infection

with Salmonella enterica. Infection and Immunity, 77, 3227-3233.

Shi, L., Summers, D.F., Peng, Q. & Galarz, J.M. (1995) Influenza A Virus RNA Polymerase

Subunit PB2 is the Endonuclease which Cleaves Host Cell mRNA and Functions Only as the

Trimeric Enzyme. Virology, 208, 38-47.

Shi, Y., Sharma, A., Wu, H., Lichtenstein, A. & Gera, J. (2005) Cyclin D1 and c-Myc Internal

Ribosome Entry Site (IRES)-Dependent Translation is Regulated by AKT Activity and

Enhanced by Rapamycin through a p38 MAPK- and ERK-Dependent Pathway. The Journal of


Shi, Y., Vattem, K.M., Sood, R., An, J., Liang, J., Stramm, L. & Wek, R.C. (1998) Identification

and Characterization of Pancreatic Eukaryotic Initiation Factor 2α-Subunit Kinase, PEK,

Involved in Translational Control. Molecular and Cellular Biology, 18, 7499-7509.

Shima, H., Pende, M., Chen, Y., Fumagali, S., Thomas, G. & Kozma, S.C. (1998) Disruption of

the p70s6k/p85s6k Gene Reveals a Small Mouse Phenotype and a New Functional S6

Kinase. The EMBO Journal, 17, 6649-6659.

Shimobayashi, M. & Hall, M.N. (2014) Making New Contacts: The mTOR Network in

Metabolism and Signalling Crosstalk. Nature Reviews Molecular Cell Biology, 15, 155-162.

Shin, B.-S., Kim, J.-R., Walker, S.E., Dong, J., Lorsch, J.R. & Dever, T.E. (2011) Initiation

Factor eIF2γ Promotes eIF2–GTP–Met-tRNAiMet Ternary Complex Binding to the 40S

Ribosome. Nature Structural & Molecular Biology, 18, 1227-1234.

Shin, J. H., Min, S.H., Kim, S.J., Kim, Y.I., Park, J., Lee, H.K. & Yoo, O.J. (2013) TAK1

Regulates Autophagic Cell Death by Suppressing the Phosphorylation of p70 S6 Kinase

1. Scientific Reports, 3, doi:10.1038/srep01561.

Shipp, M.A., Ross, K.N., Tamayo, P., Weng, A.P., Kutor, J.L., Aguiar, R.C.T., Gaasenbeer, M.,

Angelo, M., Reich, M., Pinkus, G.S., Ray, T.S., Koval, M.A., Last, K.W., Norton, A., Lister,

T.A., Mesirov, J., Neuberg, D.S., Lander, E.S., Aster, J.C. & Golub, T.R. (2002) Diffuse Large

331

B-Cell Lymphoma Outcome Prediction by Gene-Expression Profiling and Supervised Machine

Learning. Nature Medicine, 8, 68-74.

Si, K., Chaudhuri, J., Chevesich, J. & Maitra, U. (1997) Molecular Cloning and Functional

Expression of a Human cDNA Encoding Translation Initiation Factor 6. The Proceedings of

the National Academy of Sciences of the United States of America, 94, 14285-14290.

Sidak, Z. (1967) Rectangular Confidence Regions for the Means of Multivariate Normal

Distributions. Journal of the American Statistical Association, 62, 626-633.

Sigurdsson, M. I., Jamshidi, N., Steingrimsson, E., Thiele, I. & Palsson, B.Ø. (2010) A

Detailed Genome-Wide Reconstruction of Mouse Metabolism Based on Human Recon

1. BMC Systems Biology, 4, doi: 10.1186/1752-0509-4-140.

Sindreu, C. B., Scheiner, Z.S. & Storm, D.R. (2007) Ca2+-Stimulated Adenylyl Cyclases

Regulate ERK-Dependent Activation of MSK1 during Fear Conditioning. Neuron, 53, 79-89.

Simmonds, R. E., Lali, F.V., Smallie, T., Small, P.L. & Foxwell, B.M. (2009) Mycolactone

Inhibits Monocyte Cytokine Production by a Posttranscriptional Mechanism. The Journal of

Immunology, 182, 2194-2202.

Siso-Nadal, F., Fox, J.J., Laporte, S.A., Hébert, T.E. & Swain, P.S. (2009) Cross-Talk between

Signaling Pathways can Generate Robust Oscillations in Calcium and cAMP. PLoS ONE, 4, doi:

10.1371/journal.pone.0007189.

Soares, H. P., Ni, Y., Kisfalvi, K., Sinnett-Smith, J. & Rozengurt, E. (2013) Different Patterns

of Akt and ERK Feedback Activation in Response to Rapamycin, Active-Site mTOR Inhibitors

and Metformin in Pancreatic Cancer Cells. PLoS ONE, 8, doi: 10.1371/journal.pone.0057289.

Sokabe, M., Fraser, C.S. & Hershey, J.W. (2012) The Human Translation Initiation Multi-

Factor Complex Promotes Methionyl-tRNAi Binding to the 40S Ribosomal Subunit. Nucleic

Acids Research, 40, 905-913.

Sokoloski, K. J., Dickson, A.M., Chaskey, E.L., Garneau, N.L., Wilusz, C.J. & Wilusz,

J. (2010) Sindbis Virus Usurps the Cellular HuR Protein to Stabilize its Transcripts and

Promote Productive Infections in Mammalian and Mosquito Cells. Cell Host & Microbe, 8,

196-207.

Song, K. H., Li, T., Owsley, E., Strom, S. & Chiang, J.Y. (2009) Bile Acids Activate Fibroblast

Growth Factor 19 Signaling in Human Hepatocytes to Inhibit Cholesterol 7α-Hydroxylase

Gene Expression. Hepatology, 49, 297-305.

Sonnenburg, E. D., Gao, T. & Newton, A.C. (2001) The Phosphoinositide-Dependent Kinase,

PDK-1, Phosphorylates Conventional Protein Kinase C Isozymes by a Mechanism that is

Independent of Phosphoinositide 3-Kinase. The Journal of Biological Chemistry, 276, 45289-

45297.

332

Sood, R., Porter, A.C., Ma, K., Quilliam, L.A. & Wek, R.C. (2000) Pancreatic Eukaryotic

Initiation Factor-2α kinase (PEK) Homologues in Humans, Drosophila melanogaster and

Caenorhabditis elegans that Mediate Translational Control in Response to Endoplasmic

Reticulum Stress. The Biochemical Journal, 346, 281-293.

Sosnovtsev, S. V., Belliot, G., Chang, K.-O.K., Prikhodko, V.G., Thrackray, L.B., Wobus, C.E.,

Karst, S.M., Virgin, H.W. & Green, K.Y. (2006) Cleavage Map and Proteolytic Processing of

the Murine Norovirus Nonstructural Polyprotein in Infected Cells. Journal of Virology, 80,

7816-7831.

Sosnovtsev, S. V., Garfield, M. & Green, K.Y. (2002) Processing Map and Essential Cleavage

Sites of the Nonstructural Polyprotein Encoded by ORF1 of the Feline Calicivirus

Genome. Journal of Virology, 76, 7060-7072.

Sosnovtsev, S. V., Prikhod'ko, E.A., Belliot, G., Cohen, J.I. & Green, K.Y. (2003) Feline

Calicivirus Replication Induces Apoptosis in Cultured Cells. Virus Research, 94, 1-10.

Sosnovtseva, S. A., Sosnovtsev, S.V. & Green, K.Y. (1999) Mapping of the Feline Calicivirus

Proteinase Responsible for Autocatalytic Processing of the Nonstructural Polyprotein and

Identification of a Stable Proteinase-Polymerase Precursor Protein. Journal of Virology, 73,

6626-6633.

Stead, R. L. & Proud, C.G. (2013) Rapamycin Enhances eIF4E Phosphorylation by Activating

MAP Kinase-Interacting Kinase 2a (Mnk2a). FEBS Letters, 587, 2623-2628.

Stokoe, D., Macdonald, S.G., Cadwallader, K., Symons, M. & Hancock, J.F. (1994) Activation

of Raf as a Result of Recruitment to the Plasma Membrane. Science, 264, 1463-1467.

Sudhakar, A., Ramachandran, A., Ghosh, S., Hasnain, S.E., Kaufman, R.J. & Ramaiah,

K.V. (2000) Phosphorylation of Serine 51 in Initiation Factor 2 α (eIF2α) Promotes Complex

formation between eIF2α(P) and eIF2B and Causes Inhibition in the Guanine Nucleotide

Exchange Activity of eIF2B. Biochemistry, 39, 12929-12938.

Sun, S. Y., Rosenberg, L.M., Wang, X., Zhou, Z., Yue, P., Fu, H. & Khuri,

F.R. (2005) Activation of Akt and eIF4E Survival Pathways by Rapamycin-Mediated

Mammalian Target of Rapamycin Inhibition. Cancer Research, 65, 7052-7058.

Sun, Y. & Zhang, H. (2011) A Software Safety Analysis Method Based on S-Invariant of Petri

Net (2011 9th International Conference on Reliability, Maintainability and Safety (ICRMS

2011)). Guiyang, Guizhou Province (CN) 12th-15th June 2011. Published by IEEE, New York,

NY (USA).

Suzuki, C., Garces, R.G., Edmonds, K.A., Hiller, S., Hyberts, S.G., Marintchev, A. & Wagner,

G. (2008) PDCD4 Inhibits Translation Initiation by Binding to eIF4A using both its MA3

333

Domains. The Proceedings of the National Academy of Sciences of the United States of

America, 105, 3274-3279.

Swanson, K. R., Alvord Jr., E.C. & Murray, J.D. (2002) Quantifying Efficacy of Chemotherapy

of Brain Tumors with Homogeneous and Heterogeneous Drug Delivery. Acta

Biotheoretica, 50, 223-237.

Swanson, K. R., Bridge, C., Murray, J.D. & Alvord Jr., E.C. (2003) Virtual and Real Brain

Tumors: Using Mathematical Modeling to Quantify Glioma Growth and Invasion. Journal of

the Neurological Sciences, 216, 1-10.

Swingle, M., Ni, L. & Honkanen, R.E. (2007) Small-Molecule Inhibitors of ser/thr Protein

Phosphatases: Specificity, Use and Common Forms of Abuse. Methods in Molecular

Biology, 365, 23-38.

Takano, A., Usui, I., Haruta, T., Kawahara, J., Uno, T., Iwata, M. & Kobayashi,

M. (2001) Mammalian Target of Rapamycin Pathway Regulates Insulin Signaling Via

Subcellular Redistribution of Insulin Receptor Substrate 1 and Integrates Nutritional Signals

and Metabolic Signals of Insulin. Molecular and Cellular Biology, 21, 5050-5062.

Tandon, P., Gallo, C.A., Khatri, S., Barger, J.F., Yepiskoposyan, H. & Plas,

D.R. (2011) Requirement for Ribosomal Protein S6 Kinase 1 to Mediate Glycolysis and

Apoptosis Resistance Induced by Pten Deficiency. The Proceedings of the National Academy

of Sciences of the United States of America, 108, 2361-2365.

Taniike, M., Yamaguchi, O., Tsujimoto, I., Hikoso, S., Takeda, T., Nakai, A., Omiya, S.,

Mizote, I., Nakano, Y., Higuchi, Y., Matsumura, Y., Nishida, K., Ichijo, H., Hori, M. & Otsu,

K. (2008) Apoptosis Signal-Regulating Kinase 1/p38 Signaling Pathway Negatively Regulates

Physiological Hypertrophy. Circulation, 117, 545-552.

Tarun Jr., S. Z. & Sachs, A.B. (1996) Association of the Yeast Poly(A) Tail Binding Protein with

Translation Initiation Factor eIF-4G. The EMBO Journal, 15, 7168-7177.

Tatake, R. J., O'Neill, M.M., Kennedy, C.A., Wayne, A.L., Jakes, S., Wu, D., Kugler Jr., S.Z.,

Kashem, M.A., Kaplita, P. & Snow, R.J. (2008) Identification of Pharmacological Inhibitors of

the MEK5/ERK5 Pathway. Biochemical and Biophysical Research Communications, 377, 120-

125.

Tawaramoto, K., Kotani, K., Hashiramoto, M., Kanda, Y., Nagare, T., Sakaue, H., Ogawa,

W., Emoto, N., Yanagisawa, M., Noda, T., Kasuga, M. & Kaku, K. (2012) Ablation of 3-

Phosphoinositide-Dependent Protein Kinase 1 (PDK1) in Vascular Endothelial Cells Enhances

Insulin Sensitivity by Reducing Visceral Fat and Suppressing Angiogenesis. Molecular


334

Taylor, C. R. & Levenson, R.M. (2006) Quantification of Immunohistochemistry-- Issues

Concerning Methods, Utility and Semiquantitative Assessment II. Histopathology, 49, 411-

424.

Taylor, S. C., Berkelman, T., Yadav, G. & Hammond, M. (2013) A Defined Methodology for

Reliable Quantification of Western Blot Data. Molecular Biotechnology, 55, 217-226.

Te, J. A., Spradling-Reeves, K.D., Dillman 3rd, J.F. & Wallqvist, A. (2015) Neuroprotective

Mechanisms Activated in Non-Seizing Rats Exposed to Sarin. Brain Research, 1618, 136-148.

Tee, A. R., Manning, B.D., Roux, P.P., Cantley, L.C. & Blenis, J. (2003) Tuberous Sclerosis

Complex Gene Products, Tuberin and Hamartin, Control mTOR Signaling by Acting as a

GTPase-Activating Protein Complex Toward Rheb. Current Biology, 13, 1259-1268.

Tee, A. R. & Proud, C.G. (2002) Caspase Cleavage of Initiation Factor 4E-Binding Protein 1

Yields a Dominant Inhibitor of Cap-Dependent Translation and Reveals a Novel Regulatory

Motif. Molecular and Cellular Biology, 22, 1674-1683.

Teng, H., Ballim, R.D., Mowla, S. & Prince, S. (2009) Phosphorylation of Histone H3 by

Protein Kinase C Signaling Plays a Critical Role in the Regulation of the Developmentally

Important TBX2 Gene. The Journal of Biological Chemistry, 284, 26368-26376.

Terzer, M., Maynard, N.D., Covert, M.W. & Stelling, J. (2009) Genome-Scale Metabolic

Networks. Wiley Interdisciplinary Reviews – Systems Biology and Medicine, 1, 285-297.

The Human Protein Atlas (2015) Analysis of Expression of EIF4E Website. Available

at: http://www.proteinatlas.org/ENSG00000151247-EIF4E/cell/CAB004077 (Accessed:

05/22).

The Human Protein Atlas (2015) Analysis of Expression of EIF4EBP1 Website. Available

at: http://www.proteinatlas.org/ENSG00000187840-EIF4EBP1/cell/CAB005032 (Accessed:

05/22).

The UniProt Consortium (2014) Activities at the Universal Protein Resource

(UniProt). Nucleic Acids Research, 42, D191-D198.

Thiaville, M. M., Pan, Y.X., Gjymishka, A., Zhong, C., Kaufman, R.J. & Kilberg,

M.S. (2008) MEK Signaling is Required for Phosphorylation of eIF2α following Amino Acid

Limitation of HepG2 Human Hepatoma Cells. The Journal of Biological Chemistry, 283,

10848-10857.

Thiele, I. & Palsson, B.Ø. (2010) A Protocol for Generating a High-Quality Genome Scale

Metabolic Reconstruction. Nature Protocols, 5, 93-121.

Thiele, I., Swainston, N., Fleming, R.M., Hoppe, A., Sahoo, S., Aurich, M.K., Haraldsdottir,

H., Mo, M.L., Rolfsson, O., Stobbe, M.D., Thorleifsson, S.G., Agren, R., Bölling, C., Bordel,

http://www.proteinatlas.org/ENSG00000151247-EIF4E/cell/CAB004077

http://www.proteinatlas.org/ENSG00000187840-EIF4EBP1/cell/CAB005032

335

S., Chavali, A.K., Dobson, P., Dunn, W.B., Endler, L., Hala, D., Hucka, M., Hull, D., Jameson,

D., Jamshidi, N., Jonsson, J.J., Juty, N., Keating, S., Nookaew, I., Le Novère , N., Malys, N.,

Mazein, A., Papin, J.A., Price, N.D., Selkov Sr., E., Sigurdsson, M.I., Simeonidis, E.,

Sonnenschein, N., Smallbone, K., Sorokin, A., van Beek, J.H., Weichart, D., Goryanin, I.,

Nielsen, J., Westerhoff, H.V., Kell, D.B., Mendes, P. & Palsson, B.Ø. (2013) A Community-

Driven Global Reconstruction of Human Metabolism. Nature Biotechnology, 31, 419-425.

Thiem, S., Pierce, T.P., Palmieri, M., Putoczki, T.L., Buchert, M., Preaudet, A., Farid, R.O.,

Love, C., Catimel, B., Lei, Z., Rozen, S., Gopalakrishnan, V., Schaper, F., Hallek, M.,

Boussioutas, A., Tan, P., Jarnicki, A. & Ernst, M. (2013) mTORC1 Inhibition Restricts

Inflammation-Associated Gastrointestinal Tumorigenesis in Mice. The Journal of Clinical

Investigation, 123, 767-781.

Thurner, C., Witwer, C., Hofacker, I.L. & Stadler, P.F. (2004) Conserved RNA Secondary

Structures in Flaviviridae Genomes. The Journal of General Virology, 85, 1113-1124.

Tibbles, L.A., Ing, Y.L., Kiefer, F., Chan, J., Iscove, N., Woodgett, J.R. & Lassam, N.J. (1996)

MLK-3 Activates the SAPK/JNK and p38/RK Pathways via SEK1 and MKK3/6. The EMBO

Journal, 15, 7026-7035.

Tiedje, C., Ronkina, N., Tehrani, M., Dhamija, S., Laass, K., Holtmann, H., Kotlyarov, A. &

Gaestel, M. (2012) The p38/MK2-Driven Exchange between Tristetraprolin and HuR

Regulates AU-Rich Element-Dependent Translation. PLoS Genetics, 8, doi:

10.1371/journal.pgen.1002977.

Tischler, N. D., Fernández, J., Müller, I., Martínez, R., Galeno, H., Villagra, E., Mora, J.,

Ramírez, E., Rosemblatt, M. & Valenzuela, P.D. (2003) Complete Sequence of the Genome

of the Human Isolate of Andes Virus CHI-7913: Comparative Sequence and Protein Structure

Analysis. Biological Research, 36, 201-210.

Tomoo, K., Matsushita, Y., Fujisaki, H., Abiko, F., Shen, X., Taniguchi, T., Miyagawa, H.,

Kitamura, K., Miura, K. & Ishida, T. (2005) Structural Basis for mRNA Cap-Binding Regulation

of Eukaryotic Initiation Factor 4E by 4E-Binding Protein, Studied by Spectroscopic, X-Ray

Crystal Structural, and Molecular Dynamics Simulation Methods. Biochimica et Biophysica

Acta - Proteins and Proteomics, 1753, 191-208.

Topisirovic, I., Ruiz-Gutierrez, M. & Borden, K.L.B. (2004) Phosphorylation of the Eukaryotic

Translation Initiation Factor eIF4E Contributes to its Transformation and mRNA Transport

Activities. Cancer Research, 64, 8639-8642.

Toro, J., Vega, J.D., Khan, A.S., Mills, J.N., Padula, P., Terry, W., Yadón, Z., Valderrama, R.,

Ellis, B.A., Pavletic, C., Cerda, R., Zaki, S., Shieh, W.J., Meyer, R., Tapia, M., Mansilla, C.,

Baro, M., Vergara, J.A., Concha, M., Calderon, G., Enria, D., Peters, C.J. & Ksiazek,

336

T.G. (1998) An Outbreak of Hantavirus Pulmonary Syndrome, Chile, 1997. Emerging

Infectious Diseases, 4, 687-694.

Torres-Pérez, F., Palma, R.E., Hjelle, B., Ferrés, M. & Cook, J.A. (2010) Andes Virus

Infections in the Rodent Reservoir and in Humans Vary Across Contrasting Landscapes in

Chile. Infection, Genetics and Evolution, 10, 820-825.

Torrez-Martinez, N., Bharadwaj, M., Goade, D., Delury, J., Moran, P., Hicks, B., Nix, B.,

Davis, J.L. & Hjelle, B. (1998) Bayou Virus-Associated Hantavirus Pulmonary Syndrome in

Eastern Texas: Identification of the Rice Rat, Oryzomys palustris, as Reservoir

Host. Emerging Infectious Diseases, 4, 105-111.

Tracqui, P., Cruywagen, G.C., Woodward, D.E., Bartoo, G.T., Murray, J.D. & Alvord Jr.,

E.C. (1995) A Mathematical Model of Glioma Growth: The Effect of Chemotherapy on

Spatio-Temporal Growth. Cell Proliferation, 28, 17-31.

Tran, A., Talmud, D., Lejeune, B., Jovenin, N., Renois, F., Payan, C., Leveque, N. &

Andreoletti, L. (2010) Prevalence of Rotavirus, Adenovirus, Norovirus, and Astrovirus

Infections and Coinfections among Hospitalized Children in Northern France. Journal of

Clinical Microbiology, 48, 1943-1946.

Trawick, J.D. & Schilling, C.H. (2006) Use of Constraint-Based Modeling for the Prediction

and Validation of Antimicrobial Targets. Biochemical Pharmacology, 71, 1026-1035.

Tremblay, F., Brûlé, S., Hee Um, S., Li, Y., Masuda, K., Roden, M., Sun, X.J., Krebs, M.,

Polakiewicz, R.D., Thomas, G. & Marette, A. (2007) Identification of IRS-1 Ser-1101 as a

Target of S6K1 in Nutrient- and Obesity-Induced Insulin Resistance. The Proceedings of the


Trevisan, R., Daprai, L., Acquasaliente, L., Ciminale, V., Chieco-Bianchi, L. & Saggioro,

D. (2004) Relevance of CREB Phosphorylation in the Anti-Apoptotic Function of Human T-

Lymphotropic Virus Type 1 Tax Protein in Serum-Deprived Murine Fibroblasts. Experimental

Cell Research, 299, 57-67.

Trevisan, R., Daprai, L., Paloschi, L., Vajente, N., Chieco-Bianchi, L. & Saggioro,

D. (2006) Antiapoptotic Effect of Human T-Cell Leukemia Virus Type 1 Tax Protein Correlates

with its Creb Transcriptional Activity. Experimental Cell Research, 312, 1390-1400.

Trivedi, T. K., De Salvo, T., Lee, L., Palumbo, A., Moll, M., Curns, A., Hall, A.J., Patel, M.,

Parashar, U.D. & Lopman, B.A. (2012) Hospitalizations and Mortality Associated with

Norovirus Outbreaks in Nursing Homes, 2009-2010. The Journal of the American Medical

Association, 308, 1668-1675.

Tschopp, C., Knauf, U., Brauchle, M., Zurini, M., Ramage, P., Glueck, D., New, L., Han, J. &

Gram, H. (2000) Phosphorylation of eIF-4E on Ser 209 in Response to Mitogenic and

337

Inflammatory Stimuli is Faithfully Detected by Specific Antibodies. Molecular Cell Biology

Research Communications, 3, 205-211.

Tsuchiya, A., Kanno, T. & Nishizaki, T. (2013) PI3 Kinase Directly Phosphorylates Akt1/2 at

Ser473/474 in the Insulin Signal Transduction Pathway. The Journal of Endocrinology, 220,

49-59.

Tsukiyama-Kohara, K., Iizuka, N., Kohara, M. & Nomoto, A. (1992) Internal Ribosome Entry

Site within Hepatitis C Virus RNA. Journal of Virology, 66, 1476-1483.

Tyler, W. A., Gangoli, N., Gokina, P., Kim, H.A., Covey, M., Levison, S.W. & Wood,

T.L. (2009) Activation of the Mammalian Target of Rapamycin (mTOR) is Essential for

Oligodendrocyte Differentiation. The Journal of Neuroscience, 29, 6367-6378.

Uchida, L., Espada-Murao, L.A., Takamatsu, Y., Okamoto, K., Hayasaka, D., Yu, F.,

Nabeshima, T., Buerano, C.C. & Morita, K. (2014) The Dengue Virus Conceals Double-

Stranded RNA in the Intracellular Membrane to Escape from an Interferon

Response. Scientific Reports, 4, doi: 10.1038/srep07395.

Ueda, T., Watanabe-Fukunga, R., Fukuyama, H., Nagata, S. & Fukunaga, R. (2004) Mnk2

and Mnk1 are Essential for Constitutive and Inducible Phosphorylation of Eukaryotic

Initiation Factor 4E but Not for Cell Growth Or Development. Molecular and Cellular

Biology, 24, 6539-6549.

Uhlén, M., Björling, E., Agaton, C., Szigyarto, C.A., Amini, B., Andersen, E., Andersson, A.C.,

Angelidou, P., Asplund, A., Asplund, C., Berglund, L., Bergström, K., Brumer, H., Cerjan, D.,

Ekström, M., Elobeid, A., Eriksson, C., Fagerberg, L., Falk, R., Fall, J., Forsberg, M.,

Björklund, M.G., Gumbel, K., Halimi, A., Hallin, I., Hamsten, C., Hansson, M., Hedhammar,

M., Hercules, G., Kampf, C., Larsson, K., Lindskog, M., Lodewyckx, W., Lund, J., Lundeberg,

J., Magnusson, K., Malm, E., Nilsson, P., Odling, J., Oksvold, P., Olsson, I., Oster, E.,

Ottosson, J., Paavilainen, L., Persson, A., Rimini, R., Rockberg, J., Runeson, M., Sivertsson,

A., Sköllermo, A., Steen, J., Stenvall, M., Sterky, F., Strömberg, S., Sundberg, M., Tegel, H.,

Tourle, S., Wahlund, E., Waldén, A., Wan, J., Wernérus, H., Westberg, J., Wester, K.,

Wrethagen, U., Xu, L.L., Hober, S. & Pontén, F. (2005) A Human Protein Atlas for Normal

and Cancer Tissues Based on Antibody Proteomics. Molecular & Cellular Proteomics, 4,

1920-1932.

Uhlén, M., Fagerberg, L., Hallström, B.M., Lindskog, C., Oksvold, P., Mardinoglu, A.,

Sivertsson, A., Kampf, C., Sjöstedt, E., Asplund, A., Olsson, I., Edlund, K., Lundberg, E.,

Navani, S., Szigyarto, C.A., Odeberg, J., Djureinovic, D., Takanen, J.O., Hober, S., Alm, T.,

Edqvist, P.H., Berling, H., Tegel, H., Mulder, J., Rockberg, J., Nilsson, P., Schwenk, J.M.,

Hamsten, M., von Feilitzen, K., Forsberg, M., Persson, L., Johansson, F., Zwahlen, M., von

Heijne, G., Nielsen, J. & Pontén, F. (2015) Tissue-Based Map of the Human

Proteome. Science, 347, doi: 10.1126/science.1260419.

338

Uhlik, M.T., Abell, A.N., Johnson, N.L., Sun, W., Cuevas, B.D., Lobel-Rice, K.E., Horne, E.A.,

Dell’Acqua, M.L. & Johnson, G.L. (2003) Rac–MEKK3–MKK3 Scaffolding for p38 MAPK

Activation During Hyperosmotic Shock. Nature Cell Biology, 5, 1104-1110.

Ulam, S., von Neumann, J. & Richtmyer, R.D. (1947) Statistical Methods in Neutron

Diffusion. Available at: http://library.lanl.gov/cgi-bin/getfile?00329286.pdf (Accessed:

04/23).

Unbehaun, A., Borukhov, S.I., Hellen, C.U.T. & Pestova, T.V. (2004) Release of Initiation

Factors from 48S Complexes during Ribosomal Subunit Joining and the Link between

Establishment of Codon-Anticodon Base-Pairing and Hydrolysis of eIF2-Bound GTP. Genes &

Development, 18, 3078-3093.

Ung, T. L., Cao, C., Lu, J., Ozato, K. & Dever, T.E. (2001) Heterologous Dimerization Domains

Functionally Substitute for the Double-Stranded RNA Binding Domains of the Kinase

PKR. The EMBO Journal, 20, 3728-3737.

Unidad de Vigilancia, Dpto. de Epidemiología, Gobierno de Chile (2014) INFORME

SEMANAL Síndrome cardiopulmonar (SCPH) y enfermedad leve (EL) por hantavirus. Available

at:http://epi.minsal.cl/epi/html/bolets/reportes/Hantavirus/hantavirus_SE522013.pdf (Acce

ssed: 27/07/2015).

Vaheri, A., Henttonen, H., Voutilainen, L., Mustonen, J., Sironen, T. & Vapalahti,

O. (2013) Hantavirus Infections in Europe and their Impact on Public Health. Reviews in

Medical Virology, 23, 35-49.

Valenzuela, D. M., Chaudhuri, A. & Maitra, U. (1982) Eukaryotic Ribosomal Subunit Anti-

Association Activity of Calf Liver is Contained in a Single Polypeptide Chain Protein of Mr =

25,500 (Eukaryotic Initiation Factor 6). The Journal of Biological Chemistry, 257, 7712-7719.

van der Heide, L. P., van Dinther, M., Moustakas, A. & ten Dijke, P. (2011) TGFβ activates

Mitogen- and Stress-Activated Protein Kinase-1 (MSK1) to Attenuate Cell Death. The Journal


van Gorp, A. G. M., Brenkman, A.B., van den Broek, N., Calkhoven, C.F., Hershey, J.W.,

Burgering, B.M.T. & Coffer, P.J. (2009) AGC Kinases Control Phosphorylation and Activation

of Eukaryotic Translation Initiation Factor 4B. Oncogene, 28, 95-106.

van Rossum, G. (2012) The Python Language Reference. 3.2.3 edn., 9450 SW Gemini Drive,

Beaverton, OR 97008, USA: Python Software Foundation.

Vander Haar, E., Lee, S.-I., Bandhakavi, S., Griffin, T.J. & Kim, D.-H. (2007) Insulin Signalling

to mTOR Mediated by the Akt/PKB Substrate PRAS40. Nature Cell Biology, 9, 316-323.

http://library.lanl.gov/cgi-bin/getfile?00329286.pdf

http://epi.minsal.cl/epi/html/bolets/reportes/Hantavirus/hantavirus_SE522013.pdf

339

Vastag, L., Koyuncu, E., Grady, S.L., Shenk, T.E. & Rabinowitz, J.D. (2011) Divergent Effects

of Human Cytomegalovirus and Herpes Simplex Virus-1 on Cellular Metabolism. PLoS

Pathogens, 7, doi: 10.1371/journal.ppat.1002124.

Vattem, K. M., Staschke, K.A. & Wek, R.C. (2001) Mechanism of Activation of the Double-

Stranded-RNA-Dependent Protein Kinase, PKR: Role of Dimerization and Cellular

Localization in the Stimulation of PKR Phosphorylation of Eukaryotic Initiation Factor-2

(eIF2). European Journal of Biochemistry, 268, 3674-3684.

Vera-Otarola, J., Solis, L., Soto-Rifo, R., Ricci, E.P., Pino, K., Tischler, N.D., Ohlmann, T.,

Darlix, J.L. & López-Lastra, M. (2012) The Andes Hantavirus NSs Protein is Expressed from

the Viral Small mRNA by a Leaky Scanning Mechanism. Journal of Virology, 86, 2176-2187.

Vera-Otarola, J., Soto-Rifo, R., Ricci, E.P., Ohlmann, T., Darlix, J.L. & López-Lastra,

M. (2010) The 3' Untranslated Region of the Andes Hantavirus Small mRNA Functionally

Replaces the Poly(A) Tail and Stimulates Cap-Dependent Translation Initiation from the Viral

mRNA. Journal of Virology, 84, 10420-10424.

Vézina, C., Kudelski, A. & Sehgal, S.N. (1975) Rapamycin (AY-22,989), a New Antifungal

Antibiotic. I. Taxonomy of the Producing Streptomycete and Isolation of the Active

Principle. The Journal of Antibiotics, 28, 721-726.

Vilella-Bach, M., Nuzzi, P., Fang, Y. & Chen, J. (1999) The FKBP12-Rapamycin-Binding

Domain is Required for FKBP12-Rapamycin-Associated Protein Kinase Activity and G1

Progression. The Journal of Biological Chemistry, 274, 4266-4272.

Villa, N., Do, A., Hershey, J.W. & Fraser, C.S. (2013) Human Eukaryotic Initiation Factor 4G

(eIF4G) Protein Binds to eIF3c, -d, and -e to Promote mRNA Recruitment to the

Ribosome. The Journal of Biological Chemistry, 288, 32932-32940.

Villalba, M., Bi, K., Hu, J., Altman, Y., Bushway, P., Reits, E., Neefjes, J., Baier, G., Abraham,

R.T. & Altman, A. (2002) Translocation of PKCθ in T Cells is Mediated by a Nonconventional,

PI3-K- and Vav-Dependent Pathway, but does Not Absolutely Require Phospholipase C. The

Journal of Cell Biology, 157, 253-263.

Villas-Bôas, C. S., Conceição, T.M., Ramírez, J., Santoro, A.B., Da Poian, A.T. & Montero-

Lomelí, M. (2009) Dengue Virus-Induced Regulation of the Host Cell Translational

Machinery. Revista Brasileira de Pesquisas Médica e Biológica, 42, 1020-1026.

Vincent, M. J., Quiroz, E., Gracia, F., Sanchez, A.J., Ksiazek, T.G., Kitsutani, P.T., Ruedas,

L.A., Tinnin, D.S., Caceres, L., Garcia, A., Rollin, P.E., Mills, J.N., Peters, C.J. & Nichol, S.T.

(2000) Hantavirus Pulmonary Syndrome in Panama: Identification of Novel Hantaviruses and

their Likely Reservoirs. Virology, 277, 14-19.

340

Vinh, D. C. & Embil, J.M. (2009) Hantavirus Pulmonary Syndrome: A Concise Clinical

Review. Southern Medical Journal, 102, 620-625.

Vlahos, C. J., Matter, W.F., Hui, K.Y. & Brown, R.F. (1994) A Specific Inhibitor of

Phosphatidylinositol 3-Kinase, 2-(4-Morpholinyl)-8-Phenyl-4H-1-Benzopyran-4-One

(LY294002). The Journal of Biological Chemistry, 269, 5241-5248.

Volpon, L., Osborne, M.J. & Borden, K.L. (2008) NMR Assignment of the Arenaviral Protein

Z from Lassa Fever Virus. Biomolecular NMR Assignments, 2, 81-84.

Volpon, L., Osborne, M.J., Capul, A.A., de la Torre, J.C. & Borden, K.L. (2010) Structural

Characterization of the Z RING-eIF4E Complex Reveals a Distinct Mode of Control for

eIF4E. The Proceedings of the National Academy of Sciences of the United States of

America, 107, 5441-5446.

Volpon, L., Osborne, M.J., Topisirovic, I., Siddiqui, N. & Borden, K.L. (2006) Cap-Free

Structure of eIF4E Suggests a Basis for Conformational Regulation by its Ligands. The EMBO

Journal, 25, 5138-5149.

von Kriegsheim, A., Pitt, A., Grindlay, G.J., Kolch, W. & Dhillon, A.S. (2006) Regulation of

the Raf-MEK-ERK Pathway by Protein Phosphatase 5. Nature Cell Biology, 8, 1011-1016.

von Manteuffel, S. R., Gingras, A.-C., Ming, X.-F., Sonenberg, N. & Thomas, G. (1996) 4E-

BP1 Phosphorylation is Mediated by the FRAP-p70s6k Pathway and is Independent of

Mitogen-Activated Protein Kinase. The Proceedings of the National Academy of Sciences of

the United States of America, 93, 4076-4080.

Waas, W. F., Lo, H.H. & Dalby, K.N. (2001) The Kinetic Mechanism of the Dual

Phosphorylation of the ATF2 Transcription Factor by p38 Mitogen-Activated Protein (MAP)

Kinase α. Implications for signal/response Profiles of MAP kinase pathways. The Journal of


Wakiyama, M., Imataka, H. & Sonenberg, N. (2000) Interaction of eIF4G with Poly(A)-

Binding Protein Stimulates Translation and is Critical for Xenopus Oocyte

Maturation. Current Biology, 10, 1147-1150.

Walsh, D. & Mohr, I. (2014) Coupling 40S Ribosome Recruitment to Modification of a Cap-

Binding Initiation Factor by eIF3 Subunit e. Genes & Development, 28, 835-840.

Walsh, R. M. & Hochedlinger, K. (2013) A Variant CRISPR-Cas9 System Adds Versatility to

Genome Engineering. The Proceedings of the National Academy of Sciences of the United

States of America,110, 15514-15515.

Wan, X., Harkavy, B., Shen, N., Grohar, P. & Helman, L.J. (2007) Rapamycin Induces

Feedback Activation of Akt Signaling through an IGF-1R-Dependent

Mechanism. Oncogene, 26, 1932-1940.

341

Wan, X. & Helman, L.J. (2003) Levels of PTEN Protein Modulate Akt Phosphorylation on

Serine 473, but Not on Threonine 308, in IGF-II-Overexpressing Rhabdomyosarcomas

Cells. Oncogene, 22, 8205-8211.

Wandless, T. J., Michnick, S.W., Rosen, M.K., Karplus, M. & Schreiber, S.L. (1991) FK506

and Rapamycin Binding to FKBP: Common Elements in Immunophilin-Ligand

Complexation. Journal of the American Chemical Society, 113, 2339-2341.

Wang, R., Zhang, Q.G., Han, D., Xu, Q., Lü, Q. & Zhang, G.Y. (2006) Inhibition of MLK3-

MKK4/7-JNK1/2 Pathway by Akt1 in Exogenous Estrogen-Induced Neuroprotection Against

Transient Global Cerebral Ischemia by a Non-Genomic Mechanism in Male Rats. Journal of

Neurochemistry, 99, 1543-1554.

Wang, W., Shi, L., Xie, Y., Ma, C., Li, W., Su, X., Huang, S., Chen, R., Zhu, Z., Mao, Z., Han, Y.

& Li, M. (2004) SP600125, A New JNK Inhibitor, Protects Dopaminergic Neurons in the MPTP

Model of Parkinson’s Disease. Neuroscience Research, 48, 195-202.

Wang, H., Yang, H., Shivalila, C.S., Dawlaty, M.M., Cheng, A.W., Zhang, F. & Jaenisch,

R. (2013) One-Step Generation of Mice Carrying Mutations in Multiple Genes by

CRISPR/Cas-Mediated Genome Engineering. Cell, 153, 910-918.

Wang, J. (2007) ‘Petri Nets for Dynamic Event-Driven System Modeling’ in P.A. Fishwick (ed.)

Handbook of Dynamic System Modeling, 1st Edition, pp. 24.1-24.9., Chapman & Hall/CRC

Computer & Information Science Series, Boca Raton, FL (USA).

Wang, L., Harris, T.E., Roth, R.A. & Lawrence Jr., J.C. (2007a) PRAS40 Regulates mTORC1

Kinase Activity by Functioning as a Direct Inhibitor of Substrate Binding. The Journal of


Wang, Q., Zhang, Q.G., Wu, D.N., Yin, X.H. & Zhang, G.Y. (2007b) Neuroprotection of

Selenite Against Ischemic Brain Injury through Negatively Regulating Early Activation of

ASK1/JNK Cascade Via Activation of PI3K/AKT Pathway. Acta Pharmacologica Sinica, 28, 19-

27.

Wang, S., Tsun, Z.Y., Wolfson, R.L., Shen, K., Wyant, G.A., Plovanich, M.E., Yuan, E.D.,

Jones, T.D., Chantranupong, L., Comb, W., Wang, T., Bar-Peled, L., Zoncu, R., Straub, C.,

Kim, C., Park, J., Sabatini, B.L. & Sabatini, D.M. (2015) Lysosomal Amino Acid Transporter

SLC38A9 Signals Arginine Sufficiency to mTORC1. Science, 347, 188-194.

Wang, T., Wei, J.J., Sabatini, D.M. & Lander, E.S. (2014) Genetic Screens in Human Cells

using the CRISPR-Cas9 System. Science, 343, 80-84.

Wang, X., Yue, P., Chan, C.B., Ye, K., Ueda, T., Watanabe-Fukunaga, R., Fukunaga, R., Fu,

H., Khuri, F.R. & Sun, S.Y. (2007c) Inhibition of Mammalian Target of Rapamycin Induces

342

Phosphatidylinositol 3-Kinase-Dependent and Mnk-Mediated Eukaryotic Translation

Initiation Factor 4E Phosphorylation. Molecular and Cellular Biology, 27, 7405-7413.

Wang, X., Yue, P., Kim, Y.A., Fu, H., Khuri, F.R. & Sun, S.Y. (2008a) Enhancing Mammalian

Target of Rapamycin (mTOR)-Targeted Cancer Therapy by Preventing mTOR/raptor

Inhibition-Initiated, mTOR/rictor-Independent Akt Activation. Cancer Research, 68, 7409-

7418.

Wang, Y., Utzinger, J., Saric, J., Li, J.V., Burckhardt, J., Dirnhofer, S., Nicholson, J.K., Singer,

B.H., Brun, R. & Holmes, E. (2008b) Global Metabolic Responses of Mice to Trypanosoma

Brucei Infection. The Proceedings of the National Academy of Sciences of the United States

of America, 105, 6127-6132.

Wang, Z., Pascual-Anaya, J., Zadissa, A., Li, W., Niimura, Y., Huang, Z., Li, C., White, S.,

Xiong, Z., Fang, D., Wang, B., Ming, Y., Chen, Y., Zheng, Y., Kuraku, S., Pignatelli, M.,

Herrero, J., Beal, K., Nozawa, M., Li, Q., Wang, J., Zhang, H., Yu, L., Shigenobu, S., Wang, J.,

Liu, J., Flicek, P., Searle, S., Wang, J., Kuratani, S., Yin, Y., Aken, B., Zhang, G. & Irie,

N. (2013) The Draft Genomes of Soft-Shell Turtle and Green Sea Turtle Yield Insights into the

Development and Evolution of the Turtle-Specific Body Plan. Nature Genetics, 45, 701-706.

Ward, V., McCormick, C., Clarke, I., Salim, O., Wobus, C., Thrackray, L., Virgin IV, H. &

Lambden, P. (2007) Recovery of Infectious Murine Norovirus using Pol II-Driven Expression

of Full-Length cDNA. The Proceedings of the National Academy of Sciences of the United


Waskiewicz, A. J., Flynn, A., Proud, C.G. & Cooper, J.A. (1997) Mitogen-Activated Protein

Kinases Activate the Serine/Threonine Kinases Mnk1 and Mnk2. The EMBO Journal, 16,

1909-1920.

343

Waterston, R. H., Lindblad-Toh, K., Birney, E., Rogers, J., Abril, J.F., Agarwal, P., Agarwala,

R., Ainscough, R., Alexandersson, M., An, P., Antonarakis, S.E., Attwood, J., Baertsch, R.,

Bailey, J., Barlow, K., Beck, S., Berry, E., Birren, B., Bloom, T., Bork, P., Botcherby, M., Bray,

N., Brent, M.R., Brown, D.G., Brown, S.D., Bult, C., Burton, J., Butler, J., Campbell, R.D.,

Carninci, P., Cawley, S., Chiaromonte, F., Chinwalla, A.T., Church, D.M., Clamp, M., Clee, C.,

Collins, F.S., Cook, L.L., Copley, R.R., Coulson, A., Couronne, O., Cuff, J., Curwen, V., Cutts,

T., Daly, M., David, R., Davies, J., Delehaunty, K.D., Deri, J., Dermitzakis, E.T., Dewey, C.,

Dickens, N.J., Diekhans, M., Dodge, S., Dubchak, I., Dunn, D.M., Eddy, S.R., Elnitski, L.,

Emes, R.D., Eswara, P., Eyras, E., Felsenfeld, A., Fewell, G.A., Flicek, P., Foley, K., Frankel,

W.N., Fulton, L.A., Fulton, R.S., Furey, T.S., Gage, D., Gibbs, R.A., Glusman, G., Gnerre, S.,

Goldman, N., Goodstadt, L., Grafham, D., Graves, T.A., Green, E.D., Gregory, S., Guigo, R.,

Guyer, M., Hardison, R.C., Haussler, D., Hayashizaki, Y., Hillier, L.W., Hinrichs, A., Hlavina,

W., Holzer, T., Hsu, F., Hua, A., Hubbard, T., Hunt, A., Jackson, I., Jaffe, D.B., Johnson, L.S.,

Jones, M., Jones, T.A., Joy, A., Kamal, M., Karlsson, E.K., Karolchik, D., Kasprzyk, A., Kawai,

J., Keibler, E., Kells, C., Kent, W.J., Kirby, A., Kolbe, D.L., Korf, I., Kucherlapati, R.S.,

Kulbokas, E.J., Kulp, D., Landers, T., Leger, J.P., Leonard, S., Letunic, I., Levine, R., Li, J., Li,

M., Lloyd, C., Lucas, S., Ma, B., Maglott, D.R., Mardis, E.R., Matthews, L., Mauceli, E.,

Mayer, J.H., McCarthy, M., McCombie, W.R., McLaren, S., McLay, K., McPherson, J.D.,

Meldrim, J., Meredith, B., Mesirov, J.P., Miller, W., Miner, T.L., Mongin, E., Montgomery,

K.T., Morgan, M., Mott, R., Mullikin, J.C., Muzny, D.M., Nash, W.E., Nelson, J.O., Nhan,

M.N., Nicol, R., Ning, Z., Nusbaum, C., O'Connor, M.J., Okazaki, Y., Oliver, K., Overton-

Larty, E., Pachter, L., Parra, G., Pepin, K.H., Peterson, J., Pevzner, P., Plumb, R., Pohl, C.S.,

Poliakov, A., Ponce, T.C., Ponting, C.P., Potter, S., Quail, M., Reymond, A., Roe, B.A.,

Roskin, K.M., Rubin, E.M., Rust, A.G., Santos, R., Sapojnikov, V., Schultz, B., Schultz, J.,

Schwartz, M.S., Schwartz, S., Scott, C., Seaman, S., Searle, S., Sharpe, T., Sheridan, A.,

Shownkeen, R., Sims, S., Singer, J.B., Slater, G., Smit, A., Smith, D.R., Spencer, B.,

Stabenau, A., Stange-Thomann, N., Sugnet, C., Suyama, M., Tesler, G., Thompson, J.,

Torrents, D., Trevaskis, E., Tromp, J., Ucla, C., Ureta-Vidal, A., Vinson, J.P., Von

Niederhausern, A.C., Wade, C.M., Wall, M., Weber, R.J., Weiss, R.B., Wendl, M.C., West,

A.P., Wetterstrand, K., Wheeler, R., Whelan, S., Wierzbowski, J., Willey, D., Williams, S.,

Wilson, R.K., Winter, E., Worley, K.C., Wyman, D., Yang, S., Yang, S.P., Zdobnov, E.M.,

Zody, M.C., Lander, E.S. & Mouse Genome Sequencing Consortium (2002) Initial

Sequencing and Comparative Analysis of the Mouse Genome. Nature, 420, 520-562.

344

Wei, L., Huhn, J.S., Mory, A., Pathak, H.B., Sosnovtsev, S.V., Green, K.Y. & Cameron,

C.E. (2001) Proteinase-Polymerase Precursor as the Active Form of Feline Calicivirus RNA-

Dependent RNA Polymerase. Journal of Virology, 75, 1211-1219.

Weill, L., James, L., Ulryck, N., Chamond, N., Herbreteau, C.H., Ohlmann, T. & Sargueil,

B. (2010) A New Type of IRES within Gag Coding Region Recruits Three Initiation Complexes

on HIV-2 Genomic RNA. Nucleic Acids Research, 38, 1367-1381.

Weitz, M., Baroudy, B.M., Maloy, W.L., Ticehurst, J.R. & Purcell, R.H. (1986) Detection of a

Genome-Linked Protein (VPg) of Hepatitis A Virus and its Comparison with Other

Picornaviral VPgs. Journal of Virology, 60, 124-130.

Wells, S. E., Hillner, P.E., Vale, R.D. & Sachs, A.B. (1998) Circularization of mRNA by

Eukaryotic Translation Initiation Factors. Molecular Cell, 2, 135-140.

Welnowska, E., Sanz, M.A., Redondo, N. & Carrasco, L. (2011) Translation of Viral mRNA

without Active eIF2: The Case of Picornaviruses. PLoS ONE, 6, doi:

10.1371/journal.pone.0022230.

Werme, K., Wigerius, M. & Johansson, M. (2008) Tick-Borne Encephalitis Virus NS5

Associates with Membrane Protein Scribble and Impairs Interferon-Stimulated JAK-STAT

Signalling. Cellular Microbiology, 10, 696-712.

Wernly, J. A., Dietl, C.A., Tabe, C.E., Pett, S.B., Crandall, C., Milligan, K. & Crowley,

M.R. (2011) Extracorporeal Membrane Oxygenation Support Improves Survival of Patients

with Hantavirus Cardiopulmonary Syndrome Refractory to Medical Treatment. European

Journal of Cardio-Thoracic Surgery, 40, 1334-1340.

Widdowson, M.-A., Sulka, A., Bulens, S.N., Beard, R.S., Chaves, S.S., Hammond, R., Salehi,

E.D.P., Swanson, E., Totaro, J., Woron, R., Mead, P.S., Bresee, J.S., Monroe, S.S. & Glass,

R.I.(2005) Norovirus and Foodborne Disease, United States, 1991–2000. Emerging Infectious

Diseases, 11, 95-102.

Willcocks, M. M., Locker, N., Gomwalk, Z., Royall, E., Bakhshesh, M., Belsham, G.J.,

Idamakanti, N., Burroughs, K.D., Reddy, P.S., Hallenbeck, P.L. & Roberts,

L.O. (2011) Structural Features of the Seneca Valley Virus Internal Ribosome Entry Site (IRES)

Element: A Picornavirus with a Pestivirus-Like IRES. Journal of Virology, 85, 4452-4461.

Williams, D. D., Price, N.T., Loughlin, A.J. & Proud, C.G. (2001) Characterization of the

Mammalian Initiation Factor eIF2B Complex as a GDP Dissociation Stimulator Protein. The


Williamson, T., Schwartz, J.M., Kell, D.B. & Stateva, L. (2009) Deterministic Mathematical

Models of the cAMP Pathway in Saccharomyces cerevisiae. BMC Systems Biology, 3, doi:

10.1186/1752-0509-3-70.

345

Windheim, M., Lang, C., Peggie, M., Plater, L.A. & Cohen, P. (2007) Molecular Mechanisms

Involved in the Regulation of Cytokine Production by Muramyl Dipeptide. The Biochemical

Journal, 404, 179-190.

Winston, B. W., Chan, E.D., Johnson, G.L. & Riches, D.W. (1997) Activation of p38mapk,

MKK3, and MKK4 by TNF-Alpha in Mouse Bone Marrow-Derived Macrophages. The Journal

of Immunology, 159, 4491-4497.

Winter, J. N., Jefferson, L.S. & Kimball, S.R. (2011) ERK and Akt Signaling Pathways Function

through Parallel Mechanisms to Promote mTORC1 Signaling. American Journal of Physiology

- Cell Physiology, 300, C1172-C1180.

Wollenhaupt, K., Brussow, K.P., Albrecht, D. & Tomek, W. (2012) The eIF4E Repressor

Protein 4E-BP2 is Merely Truncated, Despite 4E-BP1 Degradation in the Porcine Uterine

Tissue during Implantation. Molecular Reproduction and Development, 79, 767-776.

Wood, C. D., Kelly, A.P., Matthews, S.A. & Cantrell, D.A. (2007) Phosphoinositide-

Dependent Protein Kinase-1 (PDK1)-Independent Activation of the Protein Kinase C

Substrate, Protein Kinase D. FEBS Letters, 581, 3494-3498.

Wood, L. C., Ashby, M.N., Grunfeld, C. & Feingold, K.R. (1999) Cloning of Murine

Translation Initiation Factor 6 and Functional Analysis of the Homologous Sequence

YPR016c in Saccharomyces cerevisiae. The Journal of Biological Chemistry, 274, 11653-

11659.

Wu, W. H., Wang, F.S. & Chang, M.S. (2008) Dynamic Sensitivity Analysis of Biological

Systems. BMC Bioinformatics, 9 Suppl 12, doi: 10.1186/1471-2105-9-S12-S17.

Xin, X., Zhou, L., Reyes, C.M., Liu, F. & Dong, L.Q. (2010) APPL1 Mediates Adiponectin-

Stimulated p38 MAPK Activation by Scaffolding the TAK1-MKK3-p38 MAPK Pathway.

American Journal of Physiology – Endocrinology and Metabolism, 300, E103-E110.

Xing, J., Kornhauser, J.M., Xia, Z., Thiele, E.A. & Greenberg, M.E. (1998) Nerve Growth

Factor Activates Extracellular Signal-Regulated Kinase and p38 Mitogen-Activated Protein

Kinase Pathways to Stimulate CREB Serine 133 Phosphorylation. Molecular and Cellular

Biology, 18, 1946-1955.

Xu, X., Sarikas, A., Dias-Santagata, D.C., Dolios, G., Lafontant, P.J., Tsai, S.C., Zhu, W.,

Nakajima, H., Nakajima, H.O., Field, L.J., Wang, R. & Pan, Z.Q. (2008) The CUL7 E3 Ubiquitin

Ligase Targets Insulin Receptor Substrate 1 for Ubiquitin-Dependent Degradation. Molecular

Cell, 30, 403-414.

Yan, R., Rychlik, W., Etchison, D. & Rhoads, R.E. (1992) Amino Acid Sequence of the Human

Protein Synthesis Initiation Factor eIF-4γ. The Journal of Biological Chemistry, 267, 23226-

23231.

346

Yang, H.-S., Jansen, A.P., Komar, A.A., Zheng, X., Merrick, W.C., Costes, S., Lockett, S.J.,

Sonenberg, N. & Colburn, N.H. (2003) The Transformation Suppressor Pdcd4 is a Novel

Eukaryotic Translation Initiation Factor 4A Binding Protein that Inhibits

Translation. Molecular and Cellular Biology, 23, 26-37.

Yang, H., Rudge, D.G., Koos, J.D., Vaidialingam, B., Yang, H.J. & Pavletich,

N.P. (2013a) mTOR Kinase Structure, Mechanism and Regulation. Nature, 497, 217-223.

Yang, J., Wang, J., Chen, K., Guo, G., Xi, R., Rothman, P.B., Whitten, D., Zhang, L., Huang, S.

& Chen, J.L. (2013b) eIF4B Phosphorylation by Pim Kinases Plays a Critical Role in Cellular

Transformation by Abl Oncogenes. Cancer Research, 73, 4898-4908.

Yang, Q., Deng, X., Lu, B., Cameron, M., Fearns, C., Patricelli, M.P., Yates III, J.R., Gray, N.S.

& Lee, J.D. (2010) Pharmacological Inhibition of BMK1 Suppresses Tumor Growth through

Promyelocytic Leukemia Protein. Cancer Cell, 18, 258-267.

Yao, X. Q., Zhang, X.X., Yin, Y.Y., Liu, B., Luo, D.J., Liu, D., Chen, N.N., Ni, Z.F., Wang, X.,

Wang, Q., Wang, J.Z. & Liu, G.P. (2011) Glycogen Synthase Kinase-3β regulates

Tyr307 Phosphorylation of Protein Phosphatase-2A Via Protein Tyrosine Phosphatase 1B but

Not Src. The Biochemical Journal, 437, 335-344.

Ying, S. W., Futter, M., Rosenblum, K., Webber, M.J., Hunt, S.P., Bliss, T.V. & Bramham,

C.R. (2002) Brain-Derived Neurotrophic Factor Induces Long-Term Potentiation in Intact

Adult Hippocampus: Requirement for ERK Activation Coupled to CREB and Upregulation

of Arc Synthesis. The Journal of Neuroscience, 22, 1532-1540.

Yoon, S. & Seger, R. (2006) The Extracellular Signal-Regulated Kinase: Multiple Substrates

Regulate Diverse Cellular Functions. Growth Factors, 24, 21-44.

Yu, Y., Nagai, S., Wu, H., Neish, A.S., Koyasu, S. & Gewirtz, A.T. (2006) TLR5-Mediated

Phosphoinositide 3-Kinase Activation Negatively Regulates Flagellin-Induced

Proinflammatory Gene Expression. The Journal of Immunology, 176, 6194-6201.

Yu, Y., Yoon, S.O., Poulogiannis, G., Yang, Q., Ma, X.M., Villen, J., Kubica, N., Hoffman,

G.R., Cantley, L.C., Gygi, S.P. & Blenis, J. (2011) Phosphoproteomic Analysis Identifies Grb10

as an mTORC1 Substrate that Negatively Regulates Insulin Signaling. Science, 332, 1322-

1326.

Yu, Y. X., Béarzotti, M., Vende, P., Ahne, W. & Brémont, M. (1999) Partial Mapping and

Sequencing of a Fish Iridovirus Genome Reveals Genes Homologous to the Frog Virus 3 p31,

p40 and Human eIF2α. Virus Research, 63, 53-63.

Zakowicz, H., Yang, H.-S., Stark, C., Wlodawer, A., Laronde-Leblanc, N. & Colburn,

N.H. (2005) Mutational Analysis of the DEAD-Box RNA Helicase eIF4AII Characterizes its

Interaction with Transformation Suppressor Pdcd4 and eIF4GI. RNA, 11, 261-274.

347

Zampieri, C. A., Fortin, J.F., Nolan, G.P. & Nabel, G.J. (2007) The ERK Mitogen-Activated

Protein Kinase Pathway Contributes to Ebola Virus Glycoprotein-Induced

Cytotoxicity. Journal of Virology, 81, 1230-1240.

Zeng, Z., Sarbassov, D.D., Samudio, I.J., Yee, K.W., Munsell, M.F., Ellen Jackson, C., Giles,

F.J., Sabatini, D.M., Andreeff, M. & Konopleva, M. (2007) Rapamycin Derivatives Reduce

mTORC2 Signaling and Inhibit AKT Activation in AML. Blood, 109, 3509-3512.

Zhang, Y., Gao, X., Saucedo, L.J., Ru, B., Edgar, B.A. & Pan, D. (2003) Rheb is a Direct Target

of the Tuberous Sclerosis Tumour Suppressor Proteins. Nature Cell Biology, 5, 578-581.

Zhang, J., Gao, Z., Yin, J., Quon, M.J. & Ye, J. (2008) S6K Directly Phosphorylates IRS-1 on

Ser-270 to Promote Insulin Resistance in Response to TNF-α Signaling through IKK2. The


Zhang, X., Odom, D.T., Koo, S.H., Conkright, M.D., Canettieri, G., Best, J., Chen, H., Jenner,

R., Herbolsheimer, E., Jacobsen, E., Kadam, S., Ecker, J.R., Emerson, B., Hogenesch, J.B.,

Unterman, T., Young, R.A. & Montminy, M. (2005) Genome-Wide Analysis of cAMP-

Response Element Binding Protein Occupancy, Phosphorylation, and Target Gene Activation

in Human Tissues. The Proceedings of the National Academy of Sciences of the United States

of America, 102, 4459-4464.

Zhu, J., Coyne, C.B. & Sarkar, S.N. (2011) PKC Alpha Regulates Sendai Virus-Mediated

Interferon Induction through HDAC6 and β-Catenin. The EMBO Journal, 30, 4838-4849.

Zhu, K., Chan, W., Heymach, J., Wilkinson, M. & McConkey, D.J. (2009) Control of HIF-

1α expression by eIF2α phosphorylation-Mediated Translational Repression. Cancer

Research, 69, 1836-1843.

Zhu, T., Wang, J., Pei, Y., Wang, Q., Wu, Y., Qiu, G., Zhang, D., Lv, M., Li, W. & Zhang, J.

(2016) Neddylation Controls Basal MKK7 Kinase Activity in Breast Cancer Cells. Oncogene,

35, 2624-2633.

Zinda, M. J., Johnson, M.A., Paul, J.D., Horn, C., Konicek, B.W., Lu, Z.H., Sandusky, G.,

Thomas, J.E., Neubauer, B.L., Lai, M.T. & Graff, J.R. (2001) AKT-1, -2, and -3 are Expressed

in both Normal and Tumor Tissues of the Lung, Breast, Prostate, and Colon. Clinical Cancer

Research, 7, 2475-2479.

Zinzalla, V., Graziola, M., Mastriani, A., Vanoni, M. & Alberghina, L. (2007) Rapamycin-

Mediated G1 Arrest Involves Regulation of the Cdk Inhibitor Sic1 in Saccharomyces

cerevisiae. Molecular Microbiology, 63, 1482-1494.

Zoncu, R., Bar-Peled, L., Efeyan, A., Wang, S., Sancak, Y. & Sabatini, D.M. (2011) mTORC1

Senses Lysosomal Amino Acids through an Inside-Out Mechanism that Requires the

Vacuolar H+-ATPase. Science, 334, 678-683.

348

Zu, Y. L., Ai, Y., Gilchrist, A., Labadia, M.E., Sha'afi, R.I. & Huang, C.K. (1996) Activation of

MAP Kinase-Activated Protein Kinase 2 in Human Neutrophils After Phorbol Ester Or fMLP

Peptide Stimulation. Blood, 87, 5287-5296.

Zuberek, J., Wyslouch-Cieszynska, A., Niedzwiecka, A., Dadlez, M., Stepinski, J.,

Augustyniak, W., Gingras, A.-C., Zhang, Z., Burley, S.K., Sonenberg, N., Stolarski, R. &

Darzynkiewicz, E.(2003) Phosphorylation of eIF4E Attenuates its Interaction with mRNA 5'

Cap Analogs by Electrostatic Repulsion: Intein-Mediated Protein Ligation Strategy to Obtain

Phosphorylated Protein. RNA, 9, 52-61.

349

Appendices

350

The follow Supplementary Information is provided:

Supplementary Information 1 - References for the Construction of the Literature Datasets

This information was used when evaluating the effect of inhibitors within the model. This

dataset was used throughout this work (Chapter 2 – Model Creation & Refining and

Chapter 5 – Integration of the Translational Control Model with a Genome Scale Metabolic

Network of a RAW 264.7 Cell)

Supplementary Information 2 - Dataset against Which the Binary Model was Tested (Final

Benchmark Dataset)

The information contained within this table should be used in conjunction with the

information contained with Supplementary Information 1. The information contained

within this dataset was used to analyse the final version of the binary model.

Supplementary Information 3 - Dataset Against Which the Model Integrated with

Metabolism was Tested (Metabolic Benchmark Dataset)

The information contained within this table should be used in conjunction with the

information contained with Supplementary Information 1. The overall dataset is similar to

that presented in Supplementary Information 3. To prevent repetition, only the effects

linked to the extra effectors are shown.

Supplementary Information 4 – Ingredients List for Solutions Produced In-House

351

The Following Can be Found on the Enclosed USB Drive

Supplementary Information E1 – Standardization of the Nomenclature of Model

Components

This files works to standardize the naming of model components by linking all proteins,

genes and miRNAs within the model to the established databases, Uniprot, Entrez and

miRBase, respectively.

Supplementary Information E2 - List of Model Transitions & References in the Binary

Model

This provides complete documentation of all model reactions with associated references.

Supplementary Information E3 – Final Version of the Binary Model

The .spept (E3a) and .xml (E3b) files of the Final version of the model. This model

represents the Wild Type model against which the behaviour is tested. The predictive

power of this model is given in Section 2.2.4 – Evaluation of the Predictive Power of the

Final Model.

Supplementary Information E4 – MNV Model

The .spept (E4a) and .xml (E4b) files contain the model used as the mock infected MNV

model against which the effects of MNV infection were tested. The results of this model are

given in Section 4.3.2 – Simulation of the Murine Norovirus Models.

352

Supplementary Information E5 – LPS Benchmark Dataset

This .xlsx file contains the information relating to the LPS Benchmark and the effect of cell

type on the level of predictive power.

Supplementary Information E6 – Revised Original Benchmark Dataset

This .xlsx file contains an updated version of the Initial Benchmark. The references

contained within this benchmark do not use LPS stimulation.


99.99% Binomial Confidence Intervals

This .xlsx file contains all frequency data generated by Reachfq and the associated 99.99%

Binomial Confidence Intervals for model refinement and predictive power analysis.

The supplementary information E1-E7 can be found on the flash drive that accompanies this

work. To open the .spept files for Supplementary Information E3 and E4, please ensure that

Snoopy is installed on your computer. To download Snoopy, copy the following into your

internet browser:

http://www-dssz.informatik.tu-cottbus.de/DSSZ/Software/Snoopy#downloads

353

Supplementary Information 1 -

References for the Construction of

the Literature Datasets

354

Supplementary Information 1a - Benchmark Dataset References

Shown below is the reference list used in the construction of the benchmark datasets concerning the

effects the inhibitors have in the absence of Lipopolysaccharide stimulation.

Note: Those references that are faded are those that were initially used but later found to have used

LPS-stimulation

BIX02189 Mock Dataset

PMID: 20075332: Kimura, T.E., Jin, J., Zi, M., Prehar, S., Liu, W., Oceandy, D., Abe, J., Neyses, L.,

Weston, A.H., Cartwright, E.J. & Wang, X. (2010) Targeted Deletion of the Extracellular Signal-

Regulated Protein Kinase 5 Attenuates Hypertrophic Response and Promotes Pressure Overload-

Induced Apoptosis in the Heart. Circulation Research, 106, 961-970.

PMID: 22521884: Kondo, M., Inamura, H., Matsumura, K. & Matsuoka, M. (2012) Cadmium Activates Extracellular Signal-Regulated Kinase 5 in HK-2 Human Renal Proximal Tubular Cells. Biochemical and Biophysical Research Communications, 421, 490-493.

PMID: 22997248: Li, J., Kritzer, M.D., Michel, J.J., Le, A., Thakur, H., Gayanilo, M., Passariello, C.L., Negro, A., Danial, J.B., Oskouei, B., Sanders, M., Hare, J.M., Hanauer, A., Dodge-Kafka, K. & Kapiloff, M.S. (2013) Anchored p90 Ribosomal S6 Kinase 3 is Required for Cardiac Myocyte Hypertrophy. Circulation Research, 112, 128-139.

PMID: 22254155: Lim, J.H. & Woo, C.H. (2011) Laminar Flow Activation of ERK5 Leads to Cytoprotective Effect Via CHIP-Mediated p53 Ubiquitination in Endothelial Cells. Anatomy & Cell Biology, 44, 265-273.

PMID: 24491810: Mirza, A.A., Kahle, M.P., Ameka, M., Campbell, E.M. & Cuevas, B.D. (2014) MEKK2 Regulates Focal Adhesion Stability and Motility in Invasive Breast Cancer Cells. Biochimica et Biophysica Acta, 1843, 945-954.

PMID: 12574153: Nadruz Jr, W., Kobarg, C.B., Constancio, S.S., Corat, P.D. & Franchini, K.G. (2003)

Load-Induced Transcriptional Activation of c-Jun in Rat Myocardium: Regulation by Myocyte

Enhancer Factor 2. Circulation Research, 92, 243-251.

PMID: 16735500: Schweppe, R. E., Cheung, T.H. & Ahn, N.G. (2006) Global Gene Expression Analysis

of ERK5 and ERK1/2 Signaling Reveals a Role for HIF-1 in ERK5-Mediated Responses. The Journal of


PMID: 18834865: Tatake, R.J., O'Neill, M.M., Kennedy, C.A., Wayne, A.L., Jakes, S., Wu, D., Kugler

Jr., S.Z., Kashem, M.A., Kaplita, P. & Snow, R.J. (2008) Identification of Pharmacological Inhibitors of

the MEK5/ERK5 Pathway. Biochemical and Biophysical Research Communications, 377, 120-125.

355

CGP57380 Mock Dataset

PMID: 20664001: Altman, J.K., Glaser, H., Sassano, A., Joshi, S., Ueda, T., Watanabe-Fukunaga, R.,

Fukunaga, R., Tallman, M.S. & Platanias, L.C. (2010) Negative Regulatory Effects of Mnk Kinases in

the Generation of Chemotherapy-Induced Antileukemic Responses. Molecular Pharmacology, 78,

778-784.

PMID: 23401599: Fortin, C.F., Mayer, T.Z., Cloutier, A. & McDonald, P.P. (2013)

Translational Control of Human Neutrophil Responses by MNK1. Journal of Leukocyte Biology, 94, 693-703.

PMID: 23269249: Gorentla, B.K., Krishna, S., Shin, J., Inoue, M., Shinohara, M.L., Grayson, J.M.,

Fukunaga, R. & Zhong, X.P. (2013) Mnk1 and 2 are Dispensable for T Cell Development and

Activation but Important for the Pathogenesis of Experimental Autoimmune Encephalomyelitis. The


PMID: 18081851: Kjellerup, R.B., Kragballe, K., Iversen, L. & Johansen, C. (2008) Pro-Inflammatory

Cytokine Release in Keratinocytes is Mediated through the MAPK Signal-Integrating Kinases.

Experimental Dermatology, 17, 498-504.

PMID: 11463832: Knauf, U., Tschopp, C. & Gram, H. (2001) Negative Regulation of Protein

Translation by Mitogen-Activated Protein Kinase-Interacting Kinases 1 and 2. Molecular and Cellular

Biology, 21, 5500-5511.

PMID: 25403230: Landon, A. L., Muniandy, P.A., Shetty, A.C., Lehrmann, E., Volpon, L., Houng, S.,

Zhang, Y., Dai, B., Peroutka, R., Mazan-Mamczarz, K., Steinhardt, J., Mahurkar, A., Becker, K.G.,

Borden, K.L. & Gartenhaus, R.B. (2014) MNKs Act as a Regulatory Switch for eIF4E1 and eIF4E3

Driven mRNA Translation in DLBCL. Nature Communications, 5, doi:10.1038/ncomms6413.

PMID: 18032482: Rowlett, R.M., Chrestensen, C.A., Nyce, M., Harp, M.G., Pelo, J.W., Cominelli, F.,

Ernst, P.B., Pizarro, T.T., Sturgill, T.W. & Worthington, M.T. (2008) MNK Kinases Regulate Multiple

TLR Pathways and Innate Proinflammatory Cytokines in Macrophages. American Journal of

Physiology - Gastrointestinal and Liver Physiology, 294, G452-G459.

PMID: 22370634: Shi, Y., Frost, P., Hoang, B., Yang, Y., Fukunaga, R., Gera, J. & Lichtenstein,

A. (2013) MNK Kinases Facilitate c-Myc IRES Activity in Rapamycin-Treated Multiple Myeloma

Cells. Oncogene, 32, 190-197.

PMID: 19298529: Tsuda, M., Toyomitsu, E., Kometani, M., Tozaki-Saitoh, H. & Inoue,

K. (2009) Mechanisms Underlying Fibronectin-Induced Up-Regulation of P2X4R Expression in

Microglia: Distinct Roles of PI3K-Akt and MEK-ERK Signalling Pathways. Journal of Cellular and

Molecular Medicine, 13, 3251-3259.

PMID: 25115400: Wu, C. A., Huang, D.Y. & Lin, W.W. (2014) Beclin-1-Independent Autophagy

Positively Regulates Internal Ribosomal Entry Site-Dependent Translation of Hypoxia-Inducible

Factor 1α under Nutrient Deprivation. Oncotarget, 5, 7525-7539.

356

PMID: 25800647: Yu, M., Li, P., K C Basnet, S., Kumarasiri, M., Diab, S., Teo, T., Albrecht, H. &

Wang, S. (2015) Discovery of 4-(Dihydropyridinon-3-Yl)Amino-5-Methylthieno[2,3-d]Pyrimidine

Derivatives as Potent Mnk Inhibitors: Synthesis, Structure-Activity Relationship Analysis and

Biological Evaluation. European Journal of Medicinal Chemistry, 95, 116-126.

PMID: 18694961: Zhang, M., Fu, W., Prabhu, S., Moore, J.C., Ko, J., Kim, J.W., Druker, B.J., Trapp,

V., Fruehauf, J., Gram, H., Fan, H.Y. & Ong, S.T. (2008) Inhibition of Polysome Assembly Enhances

Imatinib Activity Against Chronic Myelogenous Leukemia and Overcomes Imatinib Resistance.

Molecular and Cellular Biology, 28, 6496-6509.

GSK2334470 Mock Dataset

PMID: 21087210: Najafov, A., Sommer, E.M., Axten, J.M., Deyoung, M.P. & Alessi, D.R. (2011)

Characterization of GSK2334470, a Novel and Highly Specific Inhibitor of PDK1. The Biochemical

Journal, 433, 357-369.

Ku0063794 Mock Dataset

PMID: 19402821: García-Martínez, J.M., Moran, J., Clarke, R.G., Gray, A., Cosulich, S.C., Chresta,

C.M. & Alessi, D.R. (2009) Ku-0063794 is a Specific Inhibitor of the Mammalian Target of Rapamycin


LY294002 Mock Dataset

PMID: 11145615: Adi, S., Wu, N.Y. & Rosenthal, S.M. (2001) Growth Factor-Stimulated

Phosphorylation of Akt and p70(S6K) is Differentially Inhibited by LY294002 and

Wortmannin. Endocrinology, 142, 498-501.

PMID: 14982927: Alam, H., Maizels, E.T., Park, Y., Ghaey, S., Feiger, Z.J., Chandel, N.S. & Hunzicker-

Dunn, M. (2004) Follicle-Stimulating Hormone Activation of Hypoxia-Inducible Factor-1 by the

Phosphatidylinositol 3-kinase/AKT/Ras Homolog Enriched in Brain (Rheb)/mammalian Target of

Rapamycin (mTOR) Pathway is Necessary for Induction of Select Protein Markers of Follicular

Differentiation. The Journal of Biological Chemistry, 279, 19431-19440.

PMID: 15467756: Asano, T., Yao, Y., Zhu, J., Li, D., Abbruzzese, J.L. & Reddy, S.A. (2004) The PI 3-

kinase/Akt Signaling Pathway is Activated due to Aberrant Pten Expression and Targets Transcription

Factors NF-κB and c-Myc in Pancreatic Cancer Cells. Oncogene, 23, 8571-8580.

357

PMID: 17850214: Bain, J., Plater, L., Elliott, M., Shpiro, N., Hastie, C.J., McLauchlan, H., Klevernic,

I., Arthur, J.S., Alessi, D.R. & Cohen, P. (2007) The Selectivity of Protein Kinase Inhibitors: A Further


PMID: 11231341: Bhandari, B.K., Feliers, D., Duraisamy, S., Stewart, J.L., Gingras, A.C., Abboud,

H.E., Choudhury, G.G., Sonenberg, N. & Kasinath, B.S. (2001) Insulin Regulation of Protein

Translation Repressor 4E-BP1, an eIF4E-Binding Protein, in Renal Epithelial Cells. Kidney

International, 59, 866-875.

PMID: 22787058: Brix, B., Mesters, J.R., Pellerin, L. & Jöhren, O. (2012) Endothelial Cell-Derived

Nitric Oxide Enhances Aerobic Glycolysis in Astrocytes Via HIF-1α-Mediated Target Gene

Activation. The Journal of Neuroscience, 32, 9727-9735.

PMID: 8895571: Brunn, G.J., Williams, J., Sabers, C., Wiederrecht, G., Lawrence Jr., J.C. &

Abraham, R.T. (1996) Direct Inhibition of the Signaling Functions of the Mammalian Target of

Rapamycin by the Phosphoinositide 3-Kinase Inhibitors, Wortmannin and LY294002. The EMBO

Journal, 15, 5256-5267.

PMID: 22546857: Chung, E. Y., Psathas, J.N., Yu, D., Li, Y., Weiss, M.J. & Thomas-Tikhonenko,

A. (2012) CD19 is a Major B Cell Receptor-Independent Activator of MYC-Driven B-

Lymphomagenesis. The Journal of Clinical Investigation, 122, 2257-2266.

PMID: 17826953: Freeley, M., Park, J., Yang, K.J., Wange, R.L., Volkov, Y., Kelleher, D. & Long,

A. (2007) Loss of PTEN Expression does Not Contribute to PDK-1 Activity and PKC Activation-Loop

Phosphorylation in Jurkat Leukaemic T Cells. Cellular Signalling, 19, 2444-2457.

PMID: 11367542: Funakoshi, M., Sonoda, Y., Tago, K., Tominaga, S. & Kasahara, T. (2001)

Differential Involvement of p38 Mitogen-Activated Protein Kinase and Phosphatidyl Inositol 3-Kinase

in the IL-1-Mediated NF-κB and AP-1 Activation. International Immunopharmacology, 1, 595-604.

PMID: 15677443: Gäbele, E., Reif, S., Tsukada, S., Bataller, R., Yata, Y., Morris, T., Schrum, L.W.,

Brenner, D.A. & Rippe, R.A. (2005) The Role of p70S6K in Hepatic Stellate Cell Collagen Gene

Expression and Cell Proliferation. The Journal of Biological Chemistry, 280, 13374-13382.

PMID: 12070140: Gao, N., Ding, M., Zheng, J.Z., Zhang, Z., Leonard, S.S., Liu, K.J., Shi, X. & Jiang,

B.H. (2002) Vanadate-Induced Expression of Hypoxia-Inducible Factor 1α and Vascular Endothelial

Growth factor through Phosphatidylinositol 3-kinase/Akt Pathway and Reactive Oxygen Species. The


PMID: 17302559: Gharbi, S. I., Zvelebil, M.J., Shuttleworth, S.J., Hancox, T., Saghir, N., Timms, J.F.

& Waterfield, M.D. (2007) Exploring the Specificity of the PI3K Family Inhibitor LY294002. The

Biochemical Journal, 404, 15-21.

PMID: 20194722: Gonsalves, C. S. & Kalra, V.K. (2010) Hypoxia-Mediated Expression of 5-

Lipoxygenase-Activating Protein Involves HIF-1α and NF-κB and microRNAs 135a and 199a-5p. The


358

PMID: 11457733: Jiang, B. H., Jiang, G., Zheng, J.Z., Lu, Z., Hunter, T. & Vogt,

P.K. (2001) Phosphatidylinositol 3-Kinase Signaling Controls Levels of Hypoxia-Inducible Factor 1. Cell

Growth & Differentiation, 12, 363-369.

PMID: 9287347: Kanda, S., Hodgkin, M.N., Woodfield, R.J., Wakelam, M.J., Thomas, G. & Claesson-

Welsh, L. (1997) Phosphatidylinositol 3'-Kinase-Independent p70 S6 Kinase Activation by Fibroblast

Growth Factor Receptor-1 is Important for Proliferation but Not Differentiation of Endothelial


PMID: 11040049: Kang, K.W., Ryu, J.H. & Kim, S.G. (2000) The Essential Role of Phosphatidylinositol

3-Kinase and of p38 Mitogen-Activated Protein Kinase Activation in the Antioxidant Response

Element-Mediated rGSTA2 Induction by Decreased Glutathione in H4IIE Hepatoma Cells. Molecular


PMID: 9748166: Le Good, J.A., Ziegler, W.H., Parekh, D.B., Alessi, D.R., Cohen, P. & Parker, P.J.

(1998) Protein Kinase C Isotypes Controlled by Phosphoinositide 3-Kinase Through the Protein Kinase

PDK1. Science, 281, 2042-2045.

PMID: 11801244: Llorens, F., Garcia, L., Itarte, E. & Gomez, N. (2002) Apigenin and LY294002

Prolong EGF-Stimulated ERK1/2 Activation in PC12 Cells but are Unable to Induce Full

Differentiation. FEBS Letters, 510, 149-153.

PMID: 20876744: Luyimbazi, D., Akcakanat, A., McAuliffe, P.F., Zhang, L., Singh, G., Gonzalez-

Angulo, A.M., Chen, H., Do, K.A., Zheng, Y., Hung, M.C., Mills, G.B. & Meric-Bernstam, F. (2010)

Rapamycin Regulates Stearoyl CoA Desaturase 1 Expression in Breast Cancer. Molecular Cancer


PMID: 16051672: Moeller, L. C., Dumitrescu, A.M. & Refetoff, S. (2005) Cytosolic Action of Thyroid

Hormone Leads to Induction of Hypoxia-Inducible Factor-1alpha and Glycolytic Genes. Molecular


PMID: 20958264: Moon, J. S., Jin, W.J., Kwak, J.H., Kim, H.J., Yun, M.J., Kim, J.W., Park, S.W. & Kim,

K.S. (2011) Androgen Stimulates Glycolysis for De Novo Lipid Synthesis by Increasing the Activities of

Hexokinase 2 and 6-Phosphofructo-2-kinase/fructose-2,6-Bisphosphatase 2 in Prostate Cancer

Cells. The Biochemical Journal, 433, 225-233.

PMID: 12242656: Pene, F., Claessens, Y.E., Muller, O., Viguié, F., Mayeux, P., Dreyfus, F., Lacombe,

C. & Bouscary, D. (2002) Role of the Phosphatidylinositol 3-kinase/Akt and mTOR/P70S6-Kinase

Pathways in the Proliferation and Apoptosis in Multiple Myeloma. Oncogene, 21, 6587-6597.

PMID: 16849522: Pore, N., Jiang, Z., Shu, H.K., Bernhard, E., Kao, G.D. & Maity, A. (2006) Akt1

Activation can Augment Hypoxia-Inducible Factor-1α expression by Increasing Protein Translation

through a Mammalian Target of Rapamycin-Independent Pathway. Molecular Cancer Research, 4,

471-479.

PMID: 14670836: Pi, X., Yan, C. & Berk, B.C. (2004) Big Mitogen-Activated Protein Kinase

(BMK1)/ERK5 Protects Endothelial Cells from Apoptosis. Circulation Research, 94, 362-369.

359

PMID: 12417714: Stolovich, M., Tang, H., Hornstein, E., Levy, G., Cohen, R., Bae, S.S., Birnbaum,

M.J. & Meyuhas, O. (2002) Transduction of Growth Or Mitogenic Signals into Translational

Activation of TOP mRNAs is Fully Reliant on the Phosphatidylinositol 3-Kinase-Mediated Pathway but

Requires neither S6K1 nor rpS6 Phosphorylation. Molecular and Cellular Biology, 22, 8101-8113.

PMID: 16103051: Sun, S.Y., Rosenberg, L.M., Wang, X., Zhou, Z., Yue, P., Fu, H. & Khuri, F.R. (2005)

Activation of Akt and eIF4E Survival Pathways by Rapamycin-Mediated Mammalian Target of

Rapamycin Inhibition. Cancer Research, 65, 7052-7058.

PMID: 15714461: Shi, Y. H., Wang, Y.X., Bingle, L., Gong, L.H., Heng, W.J., Li, Y. & Fang,

W.G. (2005) In Vitro Study of HIF-1 Activation and VEGF Release by bFGF in the T47D Breast Cancer

Cell Line Under Normoxic Conditions: Involvement of PI-3K/Akt and MEK1/ERK Pathways. The

Journal of Pathology, 205, 530-536.

PMID: 25636967: Sodi, V. L., Khaku, S., Krutilina, R., Schwab, L.P., Vocadlo, D.J., Seagroves, T.N. &

Reginato, M.J. (2015) mTOR/MYC Axis Regulates O-GlcNAc Transferase (OGT) Expression and O-

GlcNAcylation in Breast Cancer. Molecular Cancer Research, doi: 10.1158/1541-7786.MCR-14-0536.

PMID: 11713299: Tang, H., Hornstein, E., Stolovich, M., Levy, G., Livingstone, M., Templeton, D.,

Avruch, J. & Meyuhas, O. (2001) Amino Acid-Induced Translation of TOP mRNAs is Fully Dependent

on Phosphatidylinositol 3-Kinase-Mediated Signaling, is Partially Inhibited by Rapamycin, and is

Independent of S6K1 and rpS6 Phosphorylation. Molecular and Cellular Biology, 21, 8671-8683.

PMID: 12032158: Treins, C., Giorgetti-Peraldi, S., Murdaca, J., Semenza, G.L. & Van Obberghen,

E. (2002) Insulin Stimulates Hypoxia-Inducible Factor 1 through a Phosphatidylinositol 3-

kinase/target of Rapamycin-Dependent Signaling Pathway. The Journal of Biological Chemistry, 277,

27975-27981.

PMID: 10480882: Viñals, F., Chambard, J.C. & Pouysségur, J. (1999) P70 S6 Kinase-Mediated Protein

Synthesis is a Critical Step for Vascular Endothelial Cell Proliferation. The Journal of Biological

Chemistry, 274, 26776-26782.

PMID: 11755539: Viñals, F., López-Rovira, T., Rosa, J.L. & Ventura, F. (2002) Inhibition of PI3K/p70

S6K and p38 MAPK Cascades Increases Osteoblastic Differentiation Induced by BMP-2. FEBS

Letters, 510, 99-104.

PMID: 8106507: Vlahos, C.J., Matter, W.F., Hui, K.Y. & Brown, R.F. (1994) A specific Inhibitor of

Phosphatidylinositol 3-Kinase, 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002). The


PMID: 19346248: Wang, L., Lawrence Jr, J.C., Sturgill, T.W. & Harris, T.E. (2009) Mammalian Target

of Rapamycin Complex 1 (mTORC1) Activity is Associated with Phosphorylation of Raptor by mTOR.

The Journal of Biological Chemistry, 284, 14693-14697.

PMID: 16551362: Xie, Z., Yuan, H., Yin, Y., Zeng, X., Bai, R. & Glazer, R.I. (2006) 3-Phosphoinositide-

Dependent Protein Kinase-1 (PDK1) Promotes Invasion and Activation of Matrix Metalloproteinases.

BMC Cancer, 6, doi:10.1186/1471-2407-6-77.

360

PMID: 10749120: Zhong, H., Chiles, K., Feldser, D., Laughner, E., Hanrahan, C., Georgescu, M.M.,

Simons, J.W. & Semenza, G.L. (2000) Modulation of Hypoxia-Inducible Factor 1α expression by the

Epidermal Growth factor/phosphatidylinositol 3-kinase/PTEN/AKT/FRAP Pathway in Human Prostate

Cancer Cells: Implications for Tumor Angiogenesis and Therapeutics. Cancer Research, 60, 1541-

1545.

PMID: 11157888: Zubilewicz, A., Hecquet, C., Jeanny, J., Soubrane, G., Courtois, Y. & Mascarelli, F.

(2001) Proliferation of CECs Requires Dual Signaling Through Both MAPK/ERK and PI 3-K/Akt

Pathways. Investigative Ophthalmology & Visual Science, 42, 488-496.

PD184352 Mock Dataset

PMID: 12534346: Bain, J., McLauchlan, H., Elliott, M. & Cohen, P. (2003) The Specificities of Protein

Kinase Inhibitors: An Update. The Biochemical Journal, 371, 199-204.




PMID: 21071439: Carriere, A., Romeo, Y., Acosta-Jaquez, H.A., Moreau, J., Bonneil, E., Thibault, P.,

Fingar, D.C. & Roux, P.P. (2011) ERK1/2 Phosphorylate Raptor to Promote Ras-Dependent Activation

of mTOR Complex 1 (mTORC1). The Journal of Biological Chemistry, 286, 567-577.

PMID: 18199551: Carroll, V. A. & Ashcroft, M. (2008) Regulation of Angiogenic Factors by HDM2 in

Renal Cell Carcinoma. Cancer Research, 68, 545-552.

PMID: 11431348: Dai, Y., Yu, C., Singh, V., Tang, L., Wang, Z., McInistry, R., Dent, P. & Grant, S.

(2001) Pharmacological Inhibitors of the Mitogen-Activated Protein Kinase (MAPK) Kinase/MAPK

Cascade Interact Synergistically with UCN-01 to Induce Mitochondrial Dysfunction and Apoptosis in

Human Leukemia Cells. Cancer Research, 61, 5106-5115.

PMID: 10998351: Davies, S.P., Reddy, H., Caivano, M. & Cohen, P. (2000) Specificity and

Mechanism of Action of Some Commonly Used Protein Kinase Inhibitors. The Biochemical Journal,

351, 95-105.

PMID: 15286700: Domínguez-Cáceres, M. A., García-Martínez, J.M., Calcabrini, A., González, L.,

Porque, P.G., León, J. & Martín-Pérez, J. (2004) Prolactin Induces c-Myc Expression and Cell Survival

through Activation of Src/Akt Pathway in Lymphoid Cells. Oncogene, 23, 7378-7390.

PMID: 18347148: Hamed, H., Hawkins, W., Mitchell, C., Gilfor, D., Zhang, G., Pei, X.Y., Dai, Y.,

Hagan, M.P., Roberts, J.D., Yacoub, A., Grant, S. & Dent, P. (2008) Transient Exposure of Carcinoma

Cells to RAS/MEK Inhibitors and UCN-01 Causes Cell Death in Vitro and in Vivo. Molecular Cancer


PMID: 11799119: Herbert, T.P., Tee, A.R. & Proud, C.G. (2002) The Extracellular Signal-Regulated

Kinase Pathway Regulates the Phosphorylation of 4E-BP1 at Multiple Sites. The Journal of Biological

Chemistry, 277, 11591-11596.

361

PMID: 18358237: Li, L., Tatake, R.J., Natarajan, K., Taba, Y., Garin, G., Tai, C., Leung, E.,

Surapisitchat, J., Yoshizumi, M., Yan, C., Abe, J. & Berk, B.C. (2008) Fluid Shear Stress Inhibits TNF-

Mediated JNK Activation Via MEK5-BMK1 in Endothelial Cells. Biochemical and Biophysical Research

Communications, 370, 159-163.

PMID: 19147570: Mirzoeva, O.K., Das, D., Heiser, L.M., Bhattacharya, S., Siwak, D., Gendelman, R.,

Bayani, N., Wang, N.J., Neve, R.M., Guan, Y., Hu, Z., Knight, Z., Feiler, H.S., Gascard, P., Parvin, B.,

Spellman, P.T., Shokat, K.M., Wyrobek, A.J., Bissell, M.J., McCormick, F., Kuo, W.L., Mills, G.B.,

Gray, J.W. & Korn, W.M. (2009) Basal Subtype and MAPK/ERK Kinase (MEK)-Phosphoinositide 3-

Kinase Feedback Signaling Determine Susceptibility of Breast Cancer Cells to MEK Inhibition. Cancer

Research, 69, 565-572.

PMID: 11478941: Mody, N., Leitch, J., Armstrong, C., Dixon, J. & Cohen, P. (2001) Effects of MAP

Kinase Cascade Inhibitors on the MKK5/ERK5 Pathway. FEBS Letters, 502, 21-24.

PMID: 16849472: Monick, M.M., Powers, L.S., Gross, T.J., Flaherty, D.M., Barrett, C.W. &

Hunninghake, G.W. (2006) Active ERK Contributes to Protein Translation by Preventing JNK-

Dependent Inhibition of Protein Phosphatase 1. The Journal of Immunology, 177, 1636-1645.

PMID: 19376132: Moore, C.E., Xie, J., Gomez, E. & Herbert, T.P. (2009) Identification of cAMP-

Dependent Kinase as a Third in Vivo Ribosomal Protein S6 Kinase in Pancreatic β-Cells. Journal of


PMID: 12881422: Morton, S., Davis, R.J., McLaren, A. & Cohen, P. (2003) A Reinvestigation of the

Multisite Phosphorylation of the Transcription Factor c-Jun. The EMBO Journal, 22, 3876-3886.

PMID: 15292274: Naegele, S. & Morley, S.J. (2004) Molecular Cross-Talk Between MEK1/2 and

mTOR Signaling During Recovery of 293 Cells from Hypertonic Stress. The Journal of Biological

Chemistry, 279, 46023-46034.

PMID: 12069688: Squires, M.S., Nixon, P.M. & Cook, S.J. (2002) Cell-Cycle Arrest by PD184352

Requires Inhibition of Extracellular Signal-Regulated Kinases (ERK) 1/2 but not ERK5/BMK1. The


PMID: 22076433: Sugiyama, N., Kohno, M. & Yokoyama, T. (2012) Inhibition of the p38 MAPK

Pathway Ameliorates Renal Fibrosis in an NPHP2 Mouse Model. Nephrology Dialysis Transplantation,

27, 1351-1358.

PMID: 17213817: Sutton, K. M., Hayat, S., Chau, N.M., Cook, S., Pouyssegur, J., Ahmed, A.,

Perusinghe, N., Le Floch, R., Yang, J. & Ashcroft, M. (2007) Selective Inhibition of MEK1/2 Reveals a

Differential Requirement for ERK1/2 Signalling in the Regulation of HIF-1 in Response to Hypoxia and

IGF-1. Oncogene, 26, 3920-3929.

PMID: 11359773: Tanoue, T., Yamamoto, T., Maeda, R. & Nishida, E. (2001) A Novel MAPK

Phosphatase MKP-7 Acts Preferentially on JNK/SAPK and p38α and β MAPKs. The Journal of


362

PMID: 11431469: Wang, L., Gout, I. & Proud, C.G. (2001) Cross-Talk Between the ERK and p70 S6

Kinase (S6K) Signaling Pathways. MEK-Dependent Activation of S6K2 in Cardiomyocytes. The Journal


PMID: 9155017: Waskiewicz, A.J., Flynn, A., Proud, C.G. & Cooper, J.A. (1997) Mitogen-Activated

Protein Kinases Activate the Serine/Threonine Kinases Mnk1 and Mnk2. The EMBO Journal, 16,

1909-1920.

PMID: 18497825: Ying, Q. L., Wray, J., Nichols, J., Batlle-Morera, L., Doble, B., Woodgett, J., Cohen,

P. & Smith, A. (2008) The Ground State of Embryonic Stem Cell Self-Renewal. Nature, 453, 519-523.

PMID: 22668349: Yuen, H.F., Abramczyk, O., Montgomery, G., Chan, K.K., Huang, Y.H., Sasazuki,

T., Shirasawa, S., Gopesh, S., Chan, K.W., Fennell, D., Janne, P., el-Tanani, M. & Murray, J.T. (2012)

Impact of Oncogenic Driver Mutations on Feedback Between the PI3K and MEK Pathways in Cancer

Cells. Bioscience Reports, 32, 413-422.

PMID: 17906627: Zaru, R., Ronkina, N., Gaestel, M., Arthur, J.S. & Watts, C. (2007) The MAPK-

Activated Kinase Rsk Controls an Acute Toll-Like Receptor Signaling Response in Dendritic Cells and is

Activated Through Two Distinct Pathways. Nature Immunology, 8, 1227-1235.

PD98059 Mock Dataset

PMID: 7499206: Alessi, D.R., Cuenda, A., Cohen, P., Dudley, D.T. & Saltiel, A.R. (1995) PD 098059 is

a Specific Inhibitor of the Activation of Mitogen-Activated Protein Kinase Kinase in Vitro and in Vivo.

The Journal of Biological Chemistry, 270, 27489-27494.

PMID: 15467756: Asano, T., Yao, Y., Zhu, J., Li, D., Abbruzzese, J.L. & Reddy, S.A. (2004) The PI 3-

kinase/Akt Signaling Pathway is Activated due to Aberrant Pten Expression and Targets Transcription

Factors NF-κB and c-Myc in Pancreatic Cancer Cells. Oncogene, 23, 8571-8580.

PMID: 12481421: Bondar, V.M., Sweeney-Gotsch, B., Andreeff, M., Mills, G.B. & McConkey, D.J.

(2002) Inhibition of the Phosphatidylinositol 3'-Kinase-AKT Pathway Induces Apoptosis in Pancreatic

Carcinoma Cells in Vitro and in Vivo. Molecular Cancer Therapeutics, 1, 989-997.




PMID: 20212154: Cannell, I. G., Kong, Y.W., Johnston, S.J., Chen, M.L., Collins, H.M., Dobbyn, H.C.,

Elia, A., Kress, T.R., Dickens, M., Clemens, M.J., Heery, D.M., Gaestel, M., Eilers, M., Willis, A.E. &

Bushell, M. (2010) p38 MAPK/MK2-Mediated Induction of miR-34c Following DNA Damage Prevents

Myc-Dependent DNA Replication. The Proceedings of the National Academy of Sciences of the United


PMID: 18199551: Carroll, V. A. & Ashcroft, M. (2008) Regulation of Angiogenic Factors by HDM2 in

Renal Cell Carcinoma. Cancer Research, 68, 545-552.

363

PMID: 16365184: Chen, S., Qiong, Y. & Gardner, D.G. (2006) A Role for p38 Mitogen-Activated

Protein Kinase and c-Myc in Endothelin-Dependent Rat Aortic Smooth Muscle Cell

Proliferation. Hypertension, 47, 252-258.

PMID: 10037749: Cheng, M., Wang, D. & Roussel, M.F. (1999) Expression of c-Myc in Response to

Colony-Stimulating Factor-1 Requires Mitogen-Activated Protein Kinase Kinase-1. The Journal of


PMID: 19176818: Codeluppi, S., Svensson, C.I., Hefferan, M.P., Valencia, F., Silldorff, M.D., Oshiro,

M., Marsala, M. & Pasquale, E.B. (2009) The Rheb-mTOR Pathway is Upregulated in Reactive

Astrocytes of the Injured Spinal Cord. The Journal of Neuroscience, 29, 1093-1104.



351, 95-105.

PMID: 9395471: De Fea, K. & Roth, R.A. (1997) Modulation of Insulin Receptor Substrate-1 Tyrosine

Phosphorylation and Function by Mitogen-Activated Protein Kinase. The Journal of Biological

Chemistry, 272, 31400-31406.

PMID: 11043572: Engelman, J.A., Berg, A.H., Lewis, R.Y., Lisanti, M.P. & Scherer, P.E. (2000) Tumor

Necrosis Factor α-Mediated Insulin Resistance, but not Dedifferentiation, is Abrogated by MEK1/2

Inhibitors in 3T3-L1 Adipocytes. Molecular Endocrinology, 14, 1557-1569.

PMID: 9315666: Font de Mora, J., Porras, A., Ahn, N. & Santos, E. (1997) Mitogen-Activated Protein

Kinase Activation is Not Necessary for, but Antagonizes, 3T3-L1 Adipocytic Differentiation. Molecular

and Cellular Biology, 17, 6068-6075.

PMID: 12489846: Fukazawa, H., Noguchi, K., Murakami, Y. & Uehara, Y. (2002) Mitogen-Activated

Protein/Extracellular Signal-Regulated Kinase Kinase (MEK) Inhibitors Restore Anoikis Sensitivity in

Human Breast Cancer Cell Lines with a Constitutively Activated Extracellular-Regulated Kinase (ERK)

Pathway. Molecular Cancer Therapeutics, 1, 303-309.

PMID: 12149254: Fukuda, R., Hirota, K., Fan, F., Jung, Y.D., Ellis, L.M. & Semenza,

G.L. (2002) Insulin-Like Growth Factor 1 Induces Hypoxia-Inducible Factor 1-Mediated Vascular

Endothelial Growth Factor Expression, which is Dependent on MAP Kinase and Phosphatidylinositol

3-Kinase Signaling in Colon Cancer Cells. The Journal of Biological Chemistry, 277, 38205-38211.

PMID: 12070140: Gao, N., Ding, M., Zheng, J.Z., Zhang, Z., Leonard, S.S., Liu, K.J., Shi, X. & Jiang,

B.H. (2002) Vanadate-Induced Expression of Hypoxia-Inducible Factor 1α and Vascular Endothelial

Growth factor through Phosphatidylinositol 3-kinase/Akt Pathway and Reactive Oxygen Species. The





PMID: 11440977: Görlach, A., Diebold, I., Schini-Kerth, V.B., Berchner-Pfannschmidt, U., Roth, U.,

Brandes, R.P., Kietzmann, T. & Busse, R. (2001) Thrombin Activates the Hypoxia-Inducible Factor-1

364

Signaling Pathway in Vascular Smooth Muscle Cells: Role of the p22Phox-Containing NADPH

Oxidase. Circulation Research, 89, 47-54.

PMID: 11435459: Han, Z., Boyle, D.L., Chang, L., Bennett, B., Karin, M., Yang, L., Manning, A.M. &

Firestein, G.S. (2001) c-Jun N-Terminal Kinase is Required for Metalloproteinase Expression and Joint

Destruction in Inflammatory Arthritis. The Journal of Clinical Investigation, 108, 73-81.



Chemistry, 277, 11591-11596.

PMID: 21233202: Hutchinson, J.A., Shanware, N.P., Chang, H. & Tibbetts, R.S. (2011) Regulation of

Ribosomal Protein S6 Phosphorylation by Casein Kinase 1 and Protein Phosphatase 1. The Journal of


PMID: 10473620: Kamakura, S., Moriguchi, T. & Nishida, E. (1999) Activation of the Protein Kinase

ERK5/BMK1 by Receptor Tyrosine Kinases. Identification and Characterization of a Signaling Pathway

to the Nucleus. The Journal of Biological Chemistry, 274, 26563-26571.

PMID: 11359907: Laughner, E., Taghavi, P., Chiles, K., Mahon, P.C. & Semenza, G.L. (2001) HER2

(Neu) Signaling Increases the Rate of Hypoxia-Inducible Factor 1alpha (HIF-1α) Synthesis: Novel

Mechanism for HIF-1-Mediated Vascular Endothelial Growth Factor Expression. Molecular and


PMID: 23615277: Le, X., Pugach, E.K., Hettmer, S., Storer, N.Y., Liu, J., Wills, A.A., DiBiase, A., Chen,

E.Y., Ignatius, M.S., Poss, K.D., Wagers, A.J., Langenau, D.M. & Zon, L.I. (2013) A Novel Chemical

Screening Strategy in Zebrafish Identifies Common Pathways in Embryogenesis and

Rhabdomyosarcoma Development. Development, 140, 2354-2364.

PMID: 18358237: Li, L., Tatake, R.J., Natarajan, K., Taba, Y., Garin, G., Tai, C., Leung, E.,

Surapisitchat, J., Yoshizumi, M., Yan, C., Abe, J. & Berk, B.C. (2008) Fluid Shear Stress Inhibits TNF-

Mediated JNK Activation Via MEK5-BMK1 in Endothelial Cells. Biochemical and Biophysical Research

Communications, 370, 159-163.

PMID: 21619683: Ma, Q., Guin, S., Padhye, S.S., Zhou, Y.Q., Zhang, R.W. & Wang, M.H. (2011)

Ribosomal Protein S6 Kinase (RSK)-2 as a Central Effector Molecule in RON Receptor Tyrosine Kinase

Mediated Epithelial to Mesenchymal Transition Induced by Macrophage-Stimulating Protein.

Molecular Cancer, 10, doi: 10.1186/1476-4598-10-66.

PMID: 12704121: Martin, M., Michalek, S.M. & Katz, J. (2003) Role of Innate Immune Factors in the

Adjuvant Activity of Monophosphoryl Lipid A. Infection and Immunity, 71, 2498-2507.

PMID: 20109524: Mauvoisin, D., Prévost, M., Ducheix, S., Arnaud, M.P. & Mounier, C. (2010) Key

Role of the ERK1/2 MAPK Pathway in the Transcriptional Regulation of the Stearoyl-CoA Desaturase

(SCD1) Gene Expression in Response to Leptin. Molecular and Cellular Endocrinology, 319, 116-128.

PMID: 16051672: Moeller, L. C., Dumitrescu, A.M. & Refetoff, S. (2005) Cytosolic Action of Thyroid

Hormone Leads to Induction of Hypoxia-Inducible Factor-1alpha and Glycolytic Genes. Molecular


365

PMID: 9211946: Morley, S.J. & McKendrick, L. (1997) Involvement of Stress-Activated Protein

Kinase and p38/RK Mitogen-Activated Protein Kinase Signaling Pathways in the Enhanced

Phosphorylation of Initiation Factor 4E in NIH 3T3 Cells. The Journal of Biological Chemistry, 272,

17887-17893.

PMID: 19061498: Nakai, T., Mochida, J. & Sakai, D. (2008) Synergistic Role of c-Myc and ERK1/2 in

the Mitogenic Response to TGFβ-1 in Cultured Rat Nucleus Pulposus Cells. Arthritis Research &

Therapy, 10, doi:10.1186/ar2567.

PMID: 16342118: Opperman, L.A., Fernandez, C.R., So, S. & Rawlins, J.T. (2006) Erk1/2 Signaling is

Required for Tgf-β2-Induced Suture Closure. Developmental Dynamics, 235, 1292-1299.

PMID: 12444974: Pintus, G., Tadolini, B., Posadino, A.M., Sanna, B., Debidda, M., Bennardini, F.,

Sava, G. & Ventura, C. (2002) Inhibition of the MEK/ERK Signaling Pathway by the Novel

Antimetastatic Agent NAMI-A Down Regulates c-Myc Gene Expression and Endothelial Cell

Proliferation. European Journal of Biochemistry, 269, 5861-5870.

PMID: 20439490: Ramírez-Valle, F., Badura, M.L., Braunstein, S., Narasimhan, M. & Schneider, R.J.

(2010) Mitotic Raptor Promotes mTORC1 Activity, G(2)/M Cell Cycle Progression, and Internal

Ribosome Entry Site-Mediated mRNA Translation. Molecular and Cellular Biology, 30, 3151-3164.

PMID: 23671727: Saliani, N., Darabi, M., Yousefi, B., Baradaran, B., Khaniani, M.S., Darabi, M.,

Shaaker, M., Mehdizadeh, A., Naji, T. & Hashemi, M. (2013) PPARγ Agonist-Induced Alterations in

Delta6-Desaturase and Stearoyl-CoA Desaturase 1: Role of MEK/ERK1/2 Pathway. World Journal of

Hepatology, 5, 220-225.

PMID: 12663671: Sharma, G.D., He, J. & Bazan, H.E. (2003) p38 and ERK1/2 Coordinate Cellular

Migration and Proliferation in Epithelial Wound Healing: Evidence of Cross-Talk Activation Between

MAP Kinase Cascades. The Journal of Biological Chemistry, 278, 21989-21997.

PMID: 11154262: Scheper, G.C., Morrice, N.A., Kleijn, M. & Proud, C.G. (2001) The Mitogen-

Activated Protein Kinase Signal-Integrating Kinase Mnk2 is a Eukaryotic Initiation Factor 4E Kinase

with High Levels of Basal Activity in Mammalian Cells. Molecular and Cellular Biology, 21, 743-754.





PMID: 19768695: Sumbayev, V. V., Nicholas, S.A., Streatfield, C.L. & Gibbs, B.F. (2009) Involvement

of Hypoxia-Inducible Factor-1 HiF(1α) in IgE-Mediated Primary Human Basophil Responses. European


PMID: 11353829: Surapisitchat, J., Hoefen, R.J., Pi, X., Yoshizumi, M., Yan, C. & Berk, B.C. (2001)

Fluid Shear Stress Inhibits TNF-α Activation of JNK but not ERK1/2 or p38 in Human Umbilical Vein

Endothelial Cells: Inhibitory Crosstalk Among MAPK Family Members. The Proceedings of the


366

PMID: 17213817: Sutton, K. M., Hayat, S., Chau, N.M., Cook, S., Pouyssegur, J., Ahmed, A.,

Perusinghe, N., Le Floch, R., Yang, J. & Ashcroft, M. (2007) Selective Inhibition of MEK1/2 Reveals a

Differential Requirement for ERK1/2 Signalling in the Regulation of HIF-1 in Response to Hypoxia and

IGF-1. Oncogene, 26, 3920-3929.

PMID: 11431469: Wang, L., Gout, I. & Proud, C.G. (2001) Cross-Talk Between the ERK and p70 S6

Kinase (S6K) Signaling Pathways. MEK-Dependent Activation of S6K2 in Cardiomyocytes. The Journal


PMID: 9155017: Waskiewicz, A.J., Flynn, A., Proud, C.G. & Cooper, J.A. (1997) Mitogen-Activated

Protein Kinases Activate the Serine/Threonine Kinases Mnk1 and Mnk2. The EMBO Journal, 16,

1909-1920.

PMID: 21655705: Xiangming, X., Yun, Q., Guoliang, Z., Jianjiang, L. & Lisong, T. (2011) Mechanisms

of RON-Mediated Epithelial-Mesenchymal Transition in MDCK Cells through the MAPK Pathway.

Revista Brasileira de Pesquisas Médicas e Biológicas, 44, 634-641.

PMID: 11408609: Yu, C., Wang, S., Dent, P. & Grant, S. (2001) Sequence-Dependent Potentiation of

Paclitaxel-Mediated Apoptosis in Human Leukemia Cells by Inhibitors of the Mitogen-Activated

Protein Kinase Kinase/Mitogen-Activated Protein Kinase Pathway. Molecular Pharmacology, 60, 143-

154.

Rapamycin Mock Dataset

PMID: 23612979: Babcock, J. T., Nguyen, H.B., He, Y., Hendricks, J.W., Wek, R.C. & Quilliam,

L.A. (2013) Mammalian Target of Rapamycin Complex 1 (mTORC1) Enhances Bortezomib-Induced

Death in Tuberous Sclerosis Complex (TSC)-Null Cells by a c-MYC-Dependent Induction of the

Unfolded Protein Response. The Journal of Biological Chemistry, 288, 15687-15698.

PMID: 21324487: Benoit, M., Dormond-Meuwly, A., Demartines, N. & Dormond, O. (2011)

Targeting the JNK Signaling Pathway Potentiates the Antiproliferative Efficacy of Rapamycin in

LS174T Colon Cancer Cells. The Journal of Surgical Research, 167, e193-e198.




PMID: 19151764: Chaturvedi, D., Gao, X., Cohen, M.S., Taunton, J. & Patel, T.B. (2009) Rapamycin

Induces Transactivation of the EGFR and Increases Cell Survival. Oncogene, 28, 1187-1196.

PMID: 20512842: Chen, X.G., Liu, F., Song, X.F., Wang, Z.H., Dong, Z.Q., Hu, Z.Q., Lan, R.Z., Guan,

W., Zhou, T.G., Xu, X.M., Lei, H., Ye, Z.Q., Peng, E.J., Du, L.H. & Zhuang, Q.Y. (2010) Rapamycin

Regulates Akt and ERK Phosphorylation through mTORC1 and mTORC2 Signaling Pathways.

Molecular Carcinogenesis, 49, 603-610.

367

PMID: 21567203: Chen, B., Yuping, S. & Ni, J. (2012) Rapamycin Decreases Survivin Expression to

Induce NSCLC Cell Apoptosis Under Hypoxia through Inhibiting HIF-1α induction. Molecular Biology

Reports, 39, 185-191.

PMID: 15899889: Chiang, G.G. & Abraham, R.T. (2005) Phosphorylation of Mammalian Target of

Rapamycin (mTOR) at Ser-2448 is Mediated by p70S6 Kinase. The Journal of Biological

Chemistry, 280, 25485-25490.

PMID: 21114628: Chou, Y.Y., Gao, J.I., Chang, S.F., Chang, P.Y. & Lu, S.C. (2011) Rapamycin Inhibits

Lipopolysaccharide Induction of Granulocyte-Colony Stimulating Factor and Inducible Nitric Oxide

Synthase Expression in Macrophages by Reducing the Levels of Octamer-Binding Factor-2. The FEBS

Journal, 278, 85-96.

PMID: 21911485: Cloninger, C., Bernath, A., Bashir, T., Holmes, B., Artinian, N., Ruegg, T.,

Anderson, L., Masri, J., Lichtenstein, A. & Gera, J. (2011) Inhibition of SAPK2/p38 Enhances

Sensitivity to mTORC1 Inhibition by Blocking IRES-Mediated Translation Initiation in

Glioblastoma. Molecular Cancer Therapeutics, 10, 2244-2256.

PMID: 19244117: Copp, J., Manning, G. & Hunter, T. (2009) TORC-Specific Phosphorylation of

Mammalian Target of Rapamycin (mTOR): Phospho-Ser2481 is a Marker for Intact mTOR Signaling

Complex 2. Cancer Research,69, 1821-1827.

PMID: 9933636: Cuenda, A. & Cohen, P. (1999) Stress-Activated Protein Kinase-2/p38 and a

Rapamycin-Sensitive Pathway are Required for C2C12 Myogenesis. The Journal of Biological

Chemistry, 274, 4341-4346.

PMID: 20616057: Carayol, N., Vakana, E., Sassano, A., Kaur, S., Goussetis, D.J., Glaser, H., Druker,

B.J., Donato, N.J., Altman, J.K., Barr, S. & Platanias, L.C. (2010) Critical Roles for mTORC2- and

Rapamycin-Insensitive mTORC1-Complexes in Growth and Survival of BCR-ABL-Expressing Leukemic

Cells. The Proceedings of the National Academy of Sciences of the United States of America, 107,

12469-12474.

PMID: 18725988: Carracedo, A., Ma, L., Teruya-Feldstein, J., Rojo, F., Salmena, L., Alimonti, A.,

Egia, A., Sasaki, A.T., Thomas, G., Kozma, S.C., Papa, A., Nardella, C., Cantley, L.C., Baselga, J. &

Pandolfi, P.P. (2008) Inhibition of mTORC1 Leads to MAPK Pathway Activation through a PI3K-

Dependent Feedback Loop in Human Cancer. The Journal of Clinical Investigation, 118, 3065-3074.

PMID: 18781596: Dayan, F., Bilton, R.L., Laferrière, J., Trottier, E., Roux, D., Pouyssegur, J. &

Mazure, N.M. (2009) Activation of HIF-1α in Exponentially Growing Cells Via Hypoxic Stimulation is

Independent of the Akt/mTOR Pathway. Journal of Cellular Physiology, 218, 167-174.

PMID: 20670887: Düvel, K., Yecies, J.L., Menon, S., Raman, P., Lipovsky, A.I., Souza, A.L.,

Triantafellow, E., Ma, Q., Gorski, R., Cleaver, S., Vander Heiden, M.G., MacKeigan, J.P., Finan, P.M.,

Clish, C.B., Murphy, L.O. & Manning, B.D. (2010) Activation of a Metabolic Gene Regulatory

Network Downstream of mTOR Complex 1. Molecular Cell, 39, 171-183.

PMID: 12149254: Fukuda, R., Hirota, K., Fan, F., Jung, Y.D., Ellis, L.M. & Semenza,

G.L. (2002) Insulin-Like Growth Factor 1 Induces Hypoxia-Inducible Factor 1-Mediated Vascular

368

Endothelial Growth Factor Expression, which is Dependent on MAP Kinase and Phosphatidylinositol

3-Kinase Signaling in Colon Cancer Cells. The Journal of Biological Chemistry, 277, 38205-38211.

PMID: 19402821: García-Martínez, J.M., Moran, J., Clarke, R.G., Gray, A., Cosulich, S.C., Chresta,

C.M. & Alessi, D.R. (2009) Ku-0063794 is a Specific Inhibitor of the Mammalian Target of Rapamycin


PMID: 14576155: Gera, J. F., Mellinghoff, I.K., Shi, Y., Rettig, M.B., Tran, C., Hsu, J.H., Sawyers, C.L.

& Lichtenstein, A.K. (2004) AKT Activity Determines Sensitivity to Mammalian Target of Rapamycin

(mTOR) Inhibitors by Regulating Cyclin D1 and c-Myc Expression. The Journal of Biological

Chemistry, 279, 2737-2746.




PMID: 21406405: Grzmil, M., Morin Jr., P., Lino, M.M., Merlo, A., Frank, S., Wang, Y., Moncayo, G.

& Hemmings, B.A. (2011) MAP Kinase-Interacting Kinase 1 Regulates SMAD2-Dependent TGF-

β signaling Pathway in Human Glioblastoma. Cancer Research, 71, 2392-2402.

PMID: 20009104: Guo, D., Prins, R.M., Dang, J., Kuga, D., Iwanami, A., Soto, H., Lin, K.Y., Huang,

T.T., Akhavan, D., Hock, M.B., Zhu, S., Kofman, A.A., Bensinger, S.J., Yong, W.H., Vinters, H.V.,

Horvath, S., Watson, A.D., Kuhn, J.G., Robins, H.I., Mehta, M.P., Wen, P.Y., DeAngelis, L.M.,

Prados, M.D., Mellinghoff, I.K., Cloughesy, T.F. & Mischel, P.S. (2009) EGFR Signaling through an

Akt-SREBP-1-Dependent, Rapamycin-Resistant Pathway Sensitizes Glioblastomas to Antilipogenic

Therapy. Science Signaling, 2, doi: 10.1126/scisignal.2000446.

PMID: 15767555: Hahn, M., Li, W., Yu, C., Rahmani, M., Dent, P. & Grant, S. (2005) Rapamycin and

UCN-01 Synergistically Induce Apoptosis in Human Leukemia Cells through a Process that is

Regulated by the Raf-1/MEK/ERK, Akt, and JNK Signal Transduction Pathways. Molecular Cancer


PMID: 12766174: Helliwell, P.A., Rumsby, M.G. & Kellett, G.L. (2003) Intestinal Sugar Absorption is

Regulated by Phosphorylation and Turnover of Protein Kinase C βII mediated by Phosphatidylinositol

3-Kinase- and Mammalian Target of Rapamycin-Dependent Pathways. The Journal of Biological

Chemistry, 278, 28644-28650.



Chemistry, 277, 11591-11596.

PMID: 15030312: Hornberger, T.A., Stuppard, R., Conley, K.E., Fedele, M.J., Fiorotto, M.L., Chin,

E.R. & Esser, K.A. (2004) Mechanical Stimuli Regulate Rapamycin-Sensitive Signalling by a

Phosphoinositide 3-Kinase-, Protein Kinase B- and Growth Factor-Independent Mechanism. The


369

PMID: 12820963: Huang, S., Shu, L., Dilling, M.B., Easton, J., Harwood, F.C., Ichijo, H. & Houghton,

P.J. (2003) Sustained Activation of the JNK Cascade and Rapamycin-Induced Apoptosis are

Suppressed by p53/p21(Cip1). Molecular Cell, 11, 1491-1501.

PMID: 12242281: Hudson, C. C., Liu, M., Chiang, G.G., Otterness, D.M., Loomis, D.C., Kaper, F.,

Giaccia, A.J. & Abraham, R.T. (2002) Regulation of Hypoxia-Inducible Factor 1α expression and

Function by the Mammalian Target of Rapamycin. Molecular and Cellular Biology, 22, 7004-7014.

PMID: 9016789: Ishizuka, T., Sakata, N., Johnson, G.L., Gelfand, E.W. & Terada, N. (1997)

Rapamycin Potentiates Dexamethasone-Induced Apoptosis and Inhibits JNK Activity in

Lymphoblastoid Cells. Biochemical and Biophysical Research Communications, 230, 386-391.

PMID: 17698586: Jossin, Y. & Goffinet, A.M. (2007) Reelin Signals through Phosphatidylinositol 3-

Kinase and Akt to Control Cortical Development and through mTor to Regulate Dendritic

Growth. Molecular and Cellular Biology, 27, 7113-7124.

PMID: 17425689: Kobayashi, S., Kishimoto, T., Kamata, S., Otsuka, M., Miyazaki, M. & Ishikura,

H. (2007) Rapamycin, a Specific Inhibitor of the Mammalian Target of Rapamycin, Suppresses

Lymphangiogenesis and Lymphatic Metastasis. Cancer Science, 98, 726-733.

PMID: 21075852: Kroczynska, B., Joshi, S., Eklund, E.A., Verma, A., Kotenko, S.V., Fish, E.N. &

Platanias, L.C. (2011) Regulatory Effects of Ribosomal S6 Kinase 1 (RSK1) in IFNλ signaling. The


PMID: 17502379: Land, S. C. & Tee, A.R. (2007) Hypoxia-Inducible Factor 1α is Regulated by the

Mammalian Target of Rapamycin (mTOR) Via an mTOR Signaling Motif. The Journal of Biological

Chemistry, 282, 20534-20543.

PMID: 10993886: Law, B. K., Waltner-Law, M.E., Entingh, A.J., Chytil, A., Aakre, M.E., Nørgaard, P.

& Moses, H.L. (2000) Salicylate-Induced Growth Arrest is Associated with Inhibition of p70s6k and

Down-Regulation of c-Myc, Cyclin D1, Cyclin A, and Proliferating Cell Nuclear Antigen. The Journal of


PMID: 21508335: Li, J., Liu, J., Song, J., Wang, X., Weiss, H.L., Townsend Jr., C.M., Gao, T. & Evers,

B.M. (2011) mTORC1 Inhibition Increases Neurotensin Secretion and Gene Expression through

Activation of the MEK/ERK/c-Jun Pathway in the Human Endocrine Cell Line BON. American Journal

of Physiology - Cell Physiology, 301, C213-C226.

PMID: 20133650: Li, S., Brown, M.S. & Goldstein, J.L. (2010) Bifurcation of Insulin Signaling Pathway

in Rat Liver: mTORC1 Required for Stimulation of Lipogenesis, but Not Inhibition of

Gluconeogenesis. The Proceedings of the National Academy of Sciences of the United States of

America, 107, 3441-3446.

PMID: 20876744: Luyimbazi, D., Akcakanat, A., McAuliffe, P.F., Zhang, L., Singh, G., Gonzalez-

Angulo, A.M., Chen, H., Do, K.A., Zheng, Y., Hung, M.C., Mills, G.B. & Meric-Bernstam, F. (2010)

Rapamycin Regulates Stearoyl CoA Desaturase 1 Expression in Breast Cancer. Molecular Cancer


370

PMID: 24068000: Ma, K. L., Liu, J., Wang, C.X., Ni, J., Zhang, Y., Wu, Y., Lv, L.L., Ruan, X.Z. & Liu,

B.C. (2013) Activation of mTOR Modulates SREBP-2 to Induce Foam Cell Formation through

Increased Retinoblastoma Protein Phosphorylation. Cardiovascular Research, 100, 450-460.

PMID: 16377759: Mahfouz, M.M., Kim, S., Delauney, A.J. & Verma, D.P. (2006) Arabidopsis Target

of Rapamycin Interacts with Raptor, which Regulates the Activity of S6 Kinase in Response to

Osmotic Stress Signals. The Plant Cell, 18, 477-490.

PMID: 22900063: Medici, D. & Olsen, B.R. (2012) Rapamycin Inhibits Proliferation of Hemangioma

Endothelial Cells by Reducing HIF-1-Dependent Expression of VEGF. PLoS ONE, 7, doi:

10.1371/journal.pone.0042913.

PMID: 17494629: Mieulet, V., Roceri, M., Espeillac, C., Sotiropoulos, A., Ohanna, M., Oorschot, V.,

Klumperman, J., Sandri, M. & Pende, M. (2007) S6 Kinase Inactivation Impairs Growth and

Translational Target Phosphorylation in Muscle Cells Maintaining Proper Regulation of Protein

Turnover. American Journal of Physiology - Cell Physiology, 293, C712-C722.

PMID: 22343118: Novalic, Z., van der Wal, A.M., Leonhard, W.N., Koehl, G., Breuning, M.H.,

Geissler, E.K., de Heer, E. & Peters, D.J. (2012) Dose-Dependent Effects of Sirolimus on mTOR

Signaling and Polycystic Kidney Disease. Journal of the American Society of Nephrology, 23, 842-853.

PMID: 11567647: Oh, C.D., Kim, S.J., Ju, J.W., Song, W.K., Kim, J.H., Yoo, Y.J. & Chun, J.S. (2001)

Immunosuppressant Rapamycin Inhibits Protein Kinase C α and p38 Mitogen-Activated

protein kinase Leading to the Inhibition of Chondrogenesis. European Journal of


PMID: 16160607: Omura, T., Yoshiyama, M., Izumi, Y., Kim, S., Matsumoto, R., Enomoto, S.,

Kusuyama, T., Nishiya, D., Nakamura, Y., Akioka, K., Iwao, H., Takeuchi, K. & Yoshikawa, J. (2005)

Involvement of c-Jun NH2 Terminal Kinase and p38MAPK in Rapamycin-Mediated Inhibition of

Neointimal Formation in Rat Carotid Arteries. Journal of Cardiovascular Pharmacology, 46, 519-525.

PMID: 19458076: Paternot, S. & Roger, P.P. (2009) Combined Inhibition of MEK and Mammalian

Target of Rapamycin Abolishes Phosphorylation of Cyclin-Dependent Kinase 4 in Glioblastoma Cell

Lines and Prevents their Proliferation. Cancer Research, 69, 4577-4581.

PMID: 25796352: Pekkala, S., Wiklund, P., Hulmi, J.J., Pollanen, E., Marjomaki, V., Munukka, E.,

Pierre, P., Mouly, V., Mero, A., Alen, M. & Cheng, S. (2015) Cannabinoid Receptor 1 and Acute

Resistance Exercise - In Vivo and in Vitro Studies in Human Skeletal Muscle. Peptides, doi:

10.1016/j.peptides.2015.03.007.

PMID: 16849522: Pore, N., Jiang, Z., Shu, H.K., Bernhard, E., Kao, G.D. & Maity, A. (2006) Akt1

Activation can Augment Hypoxia-Inducible Factor-1α expression by Increasing Protein Translation

through a Mammalian Target of Rapamycin-Independent Pathway. Molecular Cancer Research, 4,

471-479.

PMID: 25833823: Rad, E., Dodd, K.M., Thomas, L.E., Upadhyaya, M. & Tee, A.R. (2015) STAT3 and

HIF1α Signalling Drives Oncogenic Cellular Phenotypes in Malignant Peripheral Nerve Sheath Tumors

(MPNSTs). Molecular Cancer Research, doi: 10.1158/1541-7786.MCR-14-0182.

371

PMID: 16763566: Shahbazian, D., Roux, P.P., Mieulet, V., Cohen, M.S., Raught, B., Taunton, J.,

Hershey, J.W., Blenis, J., Pende, M. & Sonenberg, N. (2006) The mTOR/PI3K and MAPK Pathways

Converge on eIF4B to Control its Phosphorylation and Activity. The EMBO Journal, 25, 2781-2791.

PMID: 18602894: Sharpe, L. J. & Brown, A.J. (2008) Rapamycin Down-Regulates LDL-Receptor

Expression Independently of SREBP-2. Biochemical and Biophysical Research communications, 373,

670-674.

PMID: 15634685: Shi, Y., Sharma, A., Wu, H., Lichtenstein, A. & Gera, J. (2005) Cyclin D1 and c-Myc

Internal Ribosome Entry Site (IRES)-Dependent Translation is Regulated by AKT Activity and

Enhanced by Rapamycin through a p38 MAPK- and ERK-Dependent Pathway. The Journal of


PMID: 23437362: Soares, H.P., Ni, Y., Kisfalvi, K., Sinnett-Smith, J. & Rozengurt, E. (2013) Different

Patterns of Akt and ERK Feedback Activation in Response to Rapamycin, Active-Site mTOR Inhibitors

and Metformin in Pancreatic Cancer Cells. PLoS ONE, 8, doi: 10.1371/journal.pone.0057289.

PMID: 25636967: Sodi, V. L., Khaku, S., Krutilina, R., Schwab, L.P., Vocadlo, D.J., Seagroves, T.N. &

Reginato, M.J. (2015) mTOR/MYC Axis Regulates O-GlcNAc Transferase (OGT) Expression and O-

GlcNAcylation in Breast Cancer. Molecular Cancer Research, doi: 10.1158/1541-7786.MCR-14-0536.

PMID: 23831578: Stead, R.L. & Proud, C.G. (2013) Rapamycin Enhances eIF4E Phosphorylation by

Activating MAP Kinase-Interacting Kinase 2a (Mnk2a). FEBS Letters, 587, 2623-2628.

PMID: 16103051: Sun, S.Y., Rosenberg, L.M., Wang, X., Zhou, Z., Yue, P., Fu, H. & Khuri, F.R. (2005)

Activation of Akt and eIF4E Survival Pathways by Rapamycin-Mediated Mammalian Target of

Rapamycin Inhibition. Cancer Research, 65, 7052-7058.

PMID: 20857497: Sunayama, J., Matsuda, K., Sato, A., Tachibana, K., Suzuki, K., Narita, Y., Shibui,

S., Sakurada, K., Kayama, T., Tomiyama, A. & Kitanaka, C. (2010) Crosstalk between the PI3K/mTOR

and MEK/ERK Pathways Involved in the Maintenance of Self-Renewal and Tumorigenicity of

Glioblastoma Stem-Like Cells. Stem Cells, 28, 1930-1939.

PMID: 15833867: Takeuchi, H., Kondo, Y., Fujiwara, K., Kanzawa, T., Aoki, H., Mills, G.B. & Kondo,

S. (2005) Synergistic Augmentation of Rapamycin-Induced Autophagy in Malignant Glioma Cells by

Phosphatidylinositol 3-kinase/protein Kinase B Inhibitors. Cancer Research, 65, 3336-3346.

PMID: 21075311: Tan, J., Lee, P.L., Li, Z., Jiang, X., Lim, Y.C., Hooi, S.C. & Yu, Q. (2010) B55β-

Associated PP2A Complex Controls PDK1-Directed Myc Signaling and Modulates Rapamycin

Sensitivity in Colorectal Cancer. Cancer Cell, 18, 459-471.

PMID: 12032158: Treins, C., Giorgetti-Peraldi, S., Murdaca, J., Semenza, G.L. & Van Obberghen,

E. (2002) Insulin Stimulates Hypoxia-Inducible Factor 1 through a Phosphatidylinositol 3-

kinase/target of Rapamycin-Dependent Signaling Pathway. The Journal of Biological Chemistry, 277,

27975-27981.

PMID: 19439614: Tyler, W. A., Gangoli, N., Gokina, P., Kim, H.A., Covey, M., Levison, S.W. & Wood,

T.L. (2009) Activation of the Mammalian Target of Rapamycin (mTOR) is Essential for

Oligodendrocyte Differentiation. The Journal of Neuroscience, 29, 6367-6378.

372

PMID: 18621930: Wall, M., Poortinga, G., Hannan, K.M., Pearson, R.B., Hannan, R.D. & McArthur,

G.A. (2008) Translational Control of c-MYC by Rapamycin Promotes Terminal Myeloid

Differentiation. Blood, 112, 2305-2317.

PMID: 17968710: Wang, Y., Zhao, Q., Ma, S., Yang, F., Gong, Y. & Ke, C. (2007) Sirolimus Inhibits

Human Pancreatic Carcinoma Cell Proliferation by a Mechanism Linked to the Targeting of

mTOR/HIF-1 alpha/VEGF Signaling. IUBMB Life, 59, 717-721.

PMID: 21876130: Wang, B. T., Ducker, G.S., Barczak, A.J., Barbeau, R., Erle, D.J. & Shokat,

K.M. (2011) The Mammalian Target of Rapamycin Regulates Cholesterol Biosynthetic Gene

Expression and Exhibits a Rapamycin-Resistant Transcriptional Profile. The Proceedings of the


PMID: 10749120: Zhong, H., Chiles, K., Feldser, D., Laughner, E., Hanrahan, C., Georgescu, M.M.,

Simons, J.W. & Semenza, G.L. (2000) Modulation of Hypoxia-Inducible Factor 1α expression by the

Epidermal Growth factor/phosphatidylinositol 3-kinase/PTEN/AKT/FRAP Pathway in Human Prostate

Cancer Cells: Implications for Tumor Angiogenesis and Therapeutics. Cancer Research, 60, 1541-

1545.

PMID: 10339425: Ziegler, W.H., Parekh, D.B., Le Good, J.A., Whelan, R.D., Kelly, J.J., Frech, M.,

Hemmings, B.A. & Parker, P.J. (1999) Rapamycin-Sensitive Phosphorylation of PKC on a Carboxy-

Terminal Site by an Atypical PKC Complex. Current Biology, 9, 522-529.

SB203580 Mock Dataset

PMID: 12080469: Alarcon-Vargas, D., Tansey, W.P. & Ronai, Z. (2002) Regulation of c-Myc Stability

by Selective Stress Conditions and by MEKK1 Requires Aa 127-189 of c-Myc. Oncogene, 21, 4384-

4391.

PMID: 18502741: Bae, H.K. & Pestka, J.J. (2008) Deoxynivalenol Induces p38 Interaction with the

Ribosome in Monocytes and Macrophages. Toxicological Sciences, 105, 59-66.




PMID: 24135379: Bansal, S., Berk, M., Alkhouri, N., Partrick, D.A., Fung, J.J. & Feldstein,

A. (2014) Stearoyl-CoA Desaturase Plays an Important Role in Proliferation and Chemoresistance in

Human Hepatocellular Carcinoma. The Journal of Surgical Research, 186, 29-38.

PMID: 10960075: Birkenkamp, K.U., Tuyt, L.M., Lummen, C., Wierenga, A.T., Kruijer, W. &

Vellenga, E. (2000) The p38 MAP Kinase Inhibitor SB203580 Enhances Nuclear Factor-Kappa B

Transcriptional Activity by a Non-Specific Effect Upon the ERK Pathway. British Journal of


373

PMID: 16111636: Buxadé, M., Parra, J.L., Rousseau, S., Shpiro, N., Marquez, R., Morrice, N., Bain,

J., Espel, E. & Proud, C.G. (2005) The Mnks are Novel Components in the Control of TNFα

Biosynthesis and Phosphorylate and Regulate hnRNP A1. Immunity, 23, 177-189.

PMID: 20212154: Cannell, I. G., Kong, Y.W., Johnston, S.J., Chen, M.L., Collins, H.M., Dobbyn, H.C.,

Elia, A., Kress, T.R., Dickens, M., Clemens, M.J., Heery, D.M., Gaestel, M., Eilers, M., Willis, A.E. &

Bushell, M. (2010) p38 MAPK/MK2-Mediated Induction of miR-34c Following DNA Damage Prevents

Myc-Dependent DNA Replication. The Proceedings of the National Academy of Sciences of the United


PMID: 16365184: Chen, S., Qiong, Y. & Gardner, D.G. (2006) A Role for p38 Mitogen-Activated

Protein Kinase and c-Myc in Endothelin-Dependent Rat Aortic Smooth Muscle Cell

Proliferation. Hypertension, 47, 252-258.

PMID: 9598985: Clerk, A. & Sugden, P.H. (1998) The p38-MAPK Inhibitor, SB203580, Inhibits

Cardiac Stress-Activated Protein Kinases/c-Jun N-Terminal Kinases (SAPKs/JNKs). FEBS Letters, 426,

93-96.



351, 95-105.

PMID: 12482858: Duyndam, M. C., Hulscher, S.T., van der Wall, E., Pinedo, H.M. & Boven,

E. (2003) Evidence for a Role of p38 Kinase in Hypoxia-Inducible Factor 1-Independent Induction of

Vascular Endothelial Growth Factor Expression by Sodium Arsenite. The Journal of Biological

Chemistry, 278, 6885-6895.

PMID: 9753474: Frantz, B., Klatt, T., Pang, M., Parsons, J., Rolando, A., Williams, H., Tocci, M.J.,

O'Keefe, S.J. & O'Neill, E.A. (1998) The Activation State of p38 Mitogen-Activated Protein Kinase

Determines the Efficiency of ATP Competition for Pyridinylimidazole Inhibitor Binding. Biochemistry,

37, 13846-13853.

PMID: 11367542: Funakoshi, M., Sonoda, Y., Tago, K., Tominaga, S. & Kasahara, T. (2001)

Differential Involvement of p38 Mitogen-Activated Protein Kinase and Phosphatidyl Inositol 3-Kinase

in the IL-1-Mediated NF-κB and AP-1 Activation. International Immunopharmacology, 1, 595-604.




PMID: 21903942: Janzen, C., Sen, S., Cuevas, J., Reddy, S.T. & Chaudhuri, G. (2011) Protein

Phosphatase 2A Promotes Endothelial Survival via Stabilization of Translational Inhibitor 4E-BP1

Following Exposure to Tumor Necrosis Factor-α. Arteriosclerosis, Thrombosis, and Vascular Biology,

31, 2586-2594.

PMID: 23315073: Katholnig, K., Kaltenecker, C.C., Hayakawa, H., Rosner, M., Lassnig, C., Zlabinger,

G.J., Gaestel, M., Müller, M., Hengstschlager, M., Hörl, W.H., Park, J.M., Säemann, M.D. &

374

Weichhart, T. (2013) p38α Senses Environmental Stress to Control Innate Immune Responses Via

Mechanistic Target of Rapamycin. The Journal of Immunology, 190, 1519-1527.

PMID: 15665819: Kim, J.S., Park, Z.Y., Yoo, Y.J., Yu, S.S. & Chun, J.S. (2005) p38 Kinase Mediates

Nitric Oxide-Induced Apoptosis of Chondrocytes through the Inhibition of Protein Kinase C ζ by

Blocking Autophosphorylation. Cell Death and Differentiation, 12, 201-212.

PMID: 10702313: Lali, F.V., Hunt, A.E., Turner, S.J. & Foxwell, B.M. (2000) The Pyridinyl Imidazole

Inhibitor SB203580 Blocks Phosphoinositide-Dependent Protein Kinase Activity, Protein Kinase B

Phosphorylation, and Retinoblastoma Hyperphosphorylation in Interleukin-2-Stimulated T Cells

Independently of p38 Mitogen-Activated Protein Kinase. The Journal of Biological Chemistry, 275,

7395-7402.

PMID: 12781867: Morley, S.J. & Naegele, S. (2003) Phosphorylation of Initiation Factor 4E is

Resistant to SB203580 in Cells Expressing a Drug-Resistant Mutant of Stress-Activated Protein Kinase

2a/p38. Cellular Signalling, 15, 741-749.

PMID: 23876806: Nagahama, M., Shibutani, M., Seike, S., Yonezaki, M., Takagishi, T., Oda, M.,

Kobayashi, K. & Sakurai, J. (2013) The p38 MAPK and JNK Pathways Protect Host Cells Against

Clostridium perfringens Beta-Toxin. Infection and Immunity, doi:10.1128/IAI.00579-13.

PMID: 10455142: Parrott, L.A. & Templeton, D.J. (1999) Osmotic Stress Inhibits p70/85 S6 Kinase

through Activation of a Protein Phosphatase. The Journal of Biological Chemistry, 274, 24731-24736.

PMID: 11042204: Rane, M.J., Coxon, P.Y., Powell, D.W., Webster, R., Klein, J.B., Pierce, W., Ping, P.

& McLeish, K.R. (2001) p38 Kinase-Dependent MAPKAPK-2 Activation Functions as 3-

Phosphoinositide-Dependent Kinase-2 for Akt in Human Neutrophils. The Journal of Biological

Chemistry, 276, 3517-3523.

PMID: 11297530: Ranganathan, A.C., Nelson, K.K., Rodriguez, A.M., Kim, K.H., Tower, G.B., Rutter,

J.L., Brinckerhoff, C.E., Huang, T.T., Epstein, C.J., Jeffrey, J.J. & Melendez, J.A. (2001) Manganese

Superoxide Dismutase Signals Matrix Metalloproteinase Expression via H2O2-Dependent ERK1/2

Activation. The Journal of Biological Chemistry, 276, 14264-14270.

PMID: 11154262: Scheper, G.C., Morrice, N.A., Kleijn, M. & Proud, C.G. (2001) The Mitogen-

Activated Protein Kinase Signal-Integrating Kinase Mnk2 is a Eukaryotic Initiation Factor 4E Kinase

with High Levels of Basal Activity in Mammalian Cells. Molecular and Cellular Biology, 21, 743-754.

PMID: 16763566: Shahbazian, D., Roux, P.P., Mieulet, V., Cohen, M.S., Raught, B., Taunton, J.,

Hershey, J.W., Blenis, J., Pende, M. & Sonenberg, N. (2006) The mTOR/PI3K and MAPK Pathways

Converge on eIF4B to Control its Phosphorylation and Activity. The EMBO Journal, 25, 2781-2791.



375



PMID: 11978547: Shemirani, B. & Crowe, D.L. (2002) Hypoxic Induction of HIF-1α and VEGF

Expression in Head and Neck Squamous Cell Carcinoma Lines is Mediated by Stress Activated Protein

Kinases. Oral Oncology, 38, 251-257.

PMID: 10648601: Stoneley, M., Chappell, S.A., Jopling, C.L., Dickens, M., MacFarlane, M. & Willis,

A.E. (2000) c-Myc Protein Synthesis is Initiated from the Internal Ribosome Entry Segment during

Apoptosis. Molecular and Cellular Biology, 20, 1162-1169.

PMID: 11563982: Subkhankulova, T., Mitchell, S.A. & Willis, A.E. (2001) Internal Ribosome Entry

Segment-Mediated Initiation of c-Myc Protein Synthesis Following Genotoxic Stress. The Biochemical

Journal, 359, 183-192.

PMID: 15075293: Walsh, D. & Mohr, I. (2004) Phosphorylation of eIF4E by Mnk-1 Enhances HSV-1

Translation and Replication in Quiescent Cells. Genes & Development, 18, 660-672.

PMID: 9545260: Wang, X., Flynn, A., Waskiewicz, A.J., Webb, B.L., Vries, R.G., Baines, I.A., Cooper,

J.A. & Proud, C.G. (1998) The Phosphorylation of Eukaryotic Initiation Factor eIF4E in Response to

Phorbol Esters, Cell Stresses, and Cytokines is Mediated by Distinct MAP Kinase Pathways. The


PMID: 9299480: Wang, X. & Proud, C.G. (1997) p70 S6 Kinase is Activated by Sodium Arsenite in

Adult Rat Cardiomyocytes: Roles for Phosphatidylinositol 3-Kinase and p38 MAP Kinase. Biochemical

and Biophysical Research Communications, 238, 207-212.




SP600125 Mock Dataset

PMID: 21525782: Akinyeke, T. O. & Stewart, L.V. (2011) Troglitazone Suppresses c-Myc Levels in

Human Prostate Cancer Cells Via a PPARγ-Independent Mechanism. Cancer Biology & Therapy, 11,

1046-1058.

PMID: 19305497: Bachar, E., Ariav, Y., Ketzinel-Gilad, M., Cerasi, E., Kaiser, N. & Leibowitz, G.

(2009) Glucose Amplifies Fatty Acid-Induced Endoplasmic Reticulum Stress in Pancreatic β-cells via

Activation of mTORC1. PLoS ONE, 4, doi:10.1371/journal.pone.0004954.




PMID: 12603281: Bevilaqua, L.R., Kerr, D.S., Medina, J.H., Izquierdo, I. & Cammarota, M. (2003)

Inhibition of Hippocampal Jun N-Terminal Kinase Enhances Short-Term Memory but Blocks Long-

376

Term Memory Formation and Retrieval of an Inhibitory Avoidance Task. The European Journal of

Neuroscience, 17, 897-902.

PMID: 20868664: Bhattacharya, I., Damjanovic, M., Dominguez, A.P. & Haas, E. (2011) Inhibition of

Activated ERK1/2 and JNKs Improves Vascular Function in Mouse Aortae in the Absence of Nitric

Oxide. European Journal of Pharmacology, 658, 22-27.

PMID: 20335523: Champelovier, P., El Atifi-Borel, M., Issartel, J.P., Boutonnat, J., Berger, F. &

Seigneurin, D. (2010) Inhibition of c-Jun N-Terminal Kinase by SP600125: A cDNA Microarray

Analysis. Cancer Genomics & Proteomics, 7, 87-92.

PMID: 19996270: Cui, J., Wang, Q., Wang, J., Lv, M., Zhu, N., Li, Y., Feng, J., Shen, B. & Zhang,

J. (2009) Basal c-Jun NH2-Terminal Protein Kinase Activity is Essential for Survival and Proliferation of

T-Cell Acute Lymphoblastic Leukemia Cells. Molecular Cancer Therapeutics, 8, 3214-3222.

PMID: 21320501: Fujishita, T., Aoki, M. & Taketo, M.M. (2011) JNK Signaling Promotes Intestinal

Tumorigenesis through Activation of mTOR Complex 1 in ApcΔ716 Mice. Gastroenterology, 140, 1556-

1563.e6.

PMID: 20359507: Gandin, V., Brina, D., Marchisio, P.C. & Biffo, S. (2010) JNK Inhibition Arrests

Cotranslational Degradation. Biochimica et Biophysica Acta, 1803, 826-831.

PMID: 16099428: Hiratani, K., Haruta, T., Tani, A., Kawahara, J., Usui, I. & Kobayashi, M. (2005)

Roles of mTOR and JNK in Serine Phosphorylation, Translocation, and Degradation of IRS-

1. Biochemical and Biophysical Research Communications, 335, 836-842.

PMID: 16449644: Huang, W., Batra, S., Korrapati, S., Mishra, V. & Mehta, K.D. (2006) Selective

Repression of Low-Density Lipoprotein Receptor Expression by SP600125: Coupling of Histone H3-

Ser10 Phosphorylation and Sp1 Occupancy. Molecular and Cellular Biology, 26, pp. 1307-1317.

PMID: 16176902: Huang, Q., Wu, L.J., Tashiro, S., Onodera, S., Li, L.H. & Ikejima, T. (2005)

Silymarin Augments Human Cervical Cancer HeLa Cell Apoptosis via P38/JNK MAPK Pathways in

Serum-Free Medium. Journal of Asian Natural Products Research, 7, 701-709.

PMID: 20643107: Iñesta-Vaquera, F.A., Campbell, D.G., Arthur, J.S. & Cuenda, A. (2010) ERK5

Pathway Regulates the Phosphorylation of Tumour Suppressor hDlg During Mitosis. Biochemical and

Biophysical Research Communications, 399, 84-90.




PMID: 15890690: Gururajan, M., Chui, R., Karuppannan, A.K., Ke, J., Jennings, C.D. & Bondada,

S. (2005) c-Jun N-Terminal Kinase (JNK) is required for Survival and Proliferation of B-Lymphoma

Cells. Blood, 106, 1382-1391.

PMID: 21263197: Ito, M., Kitamura, H., Kikuguchi, C., Hase, K., Ohno, H. & Ohara, O. (2011)

SP600125 Inhibits Cap-Dependent Translation Independently of the c-Jun N-Terminal Kinase

Pathway. Cell Structure and Function, 36, 27-33.

377

PMID: 20359850: Jiang, Y., Zhang, S.H., Han, G.Q. & Qin, C.Y. (2010) Interaction of Pdcd4 with

eIF4E Inhibits the Metastatic Potential of Hepatocellular Carcinoma. Biomedicine &

Pharmacotherapy, 64, 424-429.

PMID: 16061660: Kuntzen, C., Sonuc, N., De Toni, E.N., Opelz, C., Mucha, S.R., Gerbes, A.L. &

Eichhorst, S.T. (2005) Inhibition of c-Jun-N-Terminal-Kinase Sensitizes Tumor Cells to CD95-Induced

Apoptosis and Induces G2/M Cell Cycle Arrest. Cancer Research, 65, 6780-6788.

PMID: 22493283: Kwak, D., Choi, S., Jeong, H., Jang, J.H., Lee, Y., Jeon, H., Lee, M.N., Noh, J., Cho,

K., Yoo, J.S., Hwang, D., Suh, P.G. & Ryu, S.H. (2012) Osmotic Stress Regulates Mammalian Target of

Rapamycin (mTOR) Complex 1 via c-Jun N-Terminal Kinase (JNK)-Mediated Raptor Protein


PMID: 12869635: Lahti, A., Jalonen, U., Kankaanranta, H. & Moilanen, E. (2003) c-Jun NH2-

Terminal Kinase Inhibitor Anthra(1,9-cd)pyrazol-6(2H)-one Reduces Inducible Nitric-Oxide Synthase

Expression by Destabilizing mRNA in Activated Macrophages. Molecular Pharmacology, 64, 308-315.

PMID: 17108331: LaRosa, C. & Downs, S.M. (2007) Meiotic Induction by Heat Stress in Mouse

Oocytes: Involvement of AMP-Activated Protein Kinase and MAPK Family Members. Biology of

Reproduction, 76, 476-486.

PMID: 25486532: Li, C. L., Yang, J.G., Lin, D., Zhao, Y.S., Liu, S., Xing, S.N., Zhao, S., Chen, C.Q.,

Jiang, Z.M., Pu, F.F., Cao, J.P. & Ma, D.C. (2014) Phosphorylation of Ribosomal Protein S6 Kinase 1 at

Thr421/Ser424 and Dephosphorylation at Thr389 Regulates SP600125-Induced Polyploidization of

Megakaryocytic Cell Lines. PLoS ONE, 9, doi: 10.1371/journal.pone.0114389.

PMID: 21744278: Martinez, K., Kennedy, A. & McIntosh, M.K. (2011) JNK Inhibition by SP600125

Attenuates trans-10, cis-12 Conjugated Linoleic Acid-Mediated Regulation of Inflammatory and

Lipogenic Gene Expression. Lipids, 46, 885-892.

PMID: 15454338: Minutoli, L., Altavilla, D., Marini, H., Passaniti, M., Bitto, A., Seminara, P.,

Venuti, F.S., Famulari, C., Macri, A., Versaci, A. & Squadrito, F. (2004) Protective Effects of

SP600125 A New Inhibitor of c-Jun N-Terminal Kinase (JNK) and Extracellular-Regulated Kinase

(ERK1/2) in an Experimental Model of Cerulein-Induced Pancreatitis. Life Sciences, 75, 2853-2866.

PMID: 21558088: Miyazawa, K. (2011) Encountering Unpredicted Off-Target Effects of

Pharmacological Inhibitors. The Journal of Biochemistry, 150, 1-3.

PMID: 16849472: Monick, M.M., Powers, L.S., Gross, T.J., Flaherty, D.M., Barrett, C.W. &

Hunninghake, G.W. (2006) Active ERK Contributes to Protein Translation by Preventing JNK-

Dependent Inhibition of Protein Phosphatase 1. The Journal of Immunology, 177, 1636-1645.

PMID: 23876806: Nagahama, M., Shibutani, M., Seike, S., Yonezaki, M., Takagishi, T., Oda, M.,

Kobayashi, K. & Sakurai, J. (2013) The p38 MAPK and JNK Pathways Protect Host Cells

Against Clostridium perfringens Beta-Toxin. Infection and Immunity, 81, 3703-3708.

PMID: 19418558: Nagata, H., Hatano, E., Tada, M., Murata, M., Kitamura, K., Asechi, H., Narita, M.,

Yanagida, A., Tamaki, N., Yagi, S., Ikai, I., Matsuzaki, K. & Uemoto, S. (2009) Inhibition of c-Jun NH2-

378

Terminal Kinase Switches Smad3 Signaling from Oncogenesis to Tumor- Suppression in Rat

Hepatocellular Carcinoma. Hepatology, 49, 1944-1953.

PMID: 24104553: Pannem, R. R., Dorn, C., Ahlqvist, K., Bosserhoff, A.K., Hellerbrand, C. &

Massoumi, R. (2014) CYLD Controls c-MYC Expression through the JNK-Dependent Signaling

Pathway in Hepatocellular Carcinoma. Carcinogenesis, 35, 461-468.

PMID: 12697808: Qiao, L., Han, S.I., Fang, Y., Park, J.S., Gupta, S., Gilfor, D., Amorino, G., Valerie,

K., Sealy, L., Engelhardt, J.F., Grant, S., Hylemon, P.B. & Dent, P. (2003) Bile Acid Regulation of

C/EBPβ, CREB, and c-Jun Function, via the Extracellular Signal-Regulated Kinase and c-Jun NH2-

Terminal Kinase Pathways, Modulates the Apoptotic Response of Hepatocytes. Molecular and

Cellular Biology, 23, pp. 3052-3066.

PMID: 20637208: Saxena, N. K., Fu, P.P., Nagalingam, A., Wang, J., Handy, J., Cohen, C., Tighiouart,

M., Sharma, D. & Anania, F.A. (2010) Adiponectin Modulates C-Jun N-Terminal Kinase and

Mammalian Target of Rapamycin and Inhibits Hepatocellular Carcinoma. Gastroenterology, 139,

1762-1773.

PMID: 18359001: Stempin, C.C., Garrido, V.V., Dulgerian, L.R. & Cerbán, F.M. (2008) Cruzipain and

SP600125 Induce p38 Activation, Alter NO/Arginase Balance and Favor the Survival of Trypanosoma

cruzi in Macrophages. Acta Tropica, 106, 119-127.

PMID: 19106158: Tanemura, S., Momose, H., Shimizu, N., Kitagawa, D., Seo, J., Yamasaki, T.,

Nakagawa, K., Kajiho, H., Penninger, J.M., Katada, T. & Nishina, H. (2009) Blockage by SP600125 of

Fcε Receptor-Induced Degranulation and Cytokine Gene Expression in Mast Cells is Mediated

through Inhibition of Phosphatidylinositol 3-Kinase Signalling Pathway. The Journal of


PMID: 12878189: Vaishnav, D., Jambal, P., Reusch, J.E. & Pugazhenthi, S. (2003) SP600125, an

Inhibitor of c-Jun N-Terminal Kinase, Activates CREB by a p38 MAPK-Mediated Pathway. Biochemical

and Biophysical Research Communications, 307, 855-860.

PMID: 20159943: van der Westhuizen, E.T., Christopoulos, A., Sexton, P.M., Wade, J.D. &

Summers, R.J. (2010) H2 Relaxin is a Biased Ligand Relative to H3 Relaxin at the Relaxin Family

Peptide Receptor 3 (RXFP3). Molecular Pharmacology, 77, 759-772.

Torin-1 Mock Dataset

The Torin-1 dataset was not used in this project but is included as a tool to be used in future work

PMID: 24424027: Atkin, J., Halova, L., Ferguson, J., Hitchin, J.R., Lichawska-Cieslar, A., Jordan,

A.M., Pines, J., Wellbrock, C. & Petersen, J. (2014) Torin1-Mediated TOR Kinase Inhibition Reduces

Wee1 Levels and Advances Mitotic Commitment in Fission Yeast and HeLa Cells. Journal of Cell

Science, 127, 1346-1356.

379

PMID: 21307192: Clippinger, A.J., Maguire, T.G. & Alwine, J.C. (2011) The Changing Role of mTOR

Kinase in the Maintenance of Protein Synthesis during Human Cytomegalovirus Infection. Journal of

Virology, 85, 3930-3939.

PMID: 21576368: Ekim, B., Magnuson, B., Acosta-Jaquez, H.A., Keller, J.A., Feener, E.P. & Fingar,

D.C. (2011) MTOR Kinase Domain Phosphorylation Promotes mTORC1 Signaling, Cell Growth, and

Cell Cycle Progression. Molecular and Cellular Biology, 31, 2787-2801.

PMID: 23547259: Fournier, M.J., Coudert, L., Mellaoui, S., Adjibade, P., Gareau, C., Côte, M.F.,

Sonenberg, N., Gaudreault, R.C. & Mazroui, R. (2013) Inactivation of the mTORC1-Eukaryotic

Translation Initiation Factor 4E Pathway Alters Stress Granule Formation. Molecular and Cellular

Biology, 33, 2285-2301.

PMID: 20679199: Furic, L., Rong, L., Koumakpayi, I.H., Yoshida, K., Brueschke, A., Petroulakis, E.,

Robichaud, N., Pollak, M., Gaboury, L.A., Pandolfi, P.P., Saad, F. & Sonenberg, N. (2010) EIF4E

Phosphorylation Promotes Tumorigenesis and is Associated with Prostate Cancer Progression. The


PMID: 22015718: Ghosh, S., Lau, H., Simons, B.W., Powell, J.D., Meyers, D.J., De Marzo, A.M.,

Berman, D.M. & Lotan, T.L. (2011) PI3K/mTOR Signaling Regulates Prostatic Branching

Morphogenesis. Developmental Biology, 360, 329-342.

PMID: 23319332: Guo, Y., Du, J. & Kwiatkowski, D.J. (2013) Molecular Dissection of AKT Activation

in Lung Cancer Cell Lines. Molecular Cancer Research, 11, 282-293.

PMID: 21659604: Hsu, P.P., Kang, S.A., Rameseder, J., Zhang, Y., Ottina, K.A., Lim, D., Peterson,

T.R., Choi, Y., Gray, N.S., Yaffe, M.B., Marto, J.A. & Sabatini, D.M. (2011) The mTOR-Regulated

Phosphoproteome Reveals a Mechanism of mTORC1-Mediated Inhibition of Growth Factor

Signaling. Science, 332, 1317-1322.

PMID: 23888043: Kang, S.A., Pacold, M.E., Cervantes, C.L., Lim, D., Lou, H.J., Ottina, K., Gray, N.S.,

Turk, B.E., Yaffe, M.B. & Sabatini, D.M. (2013) mTORC1 Phosphorylation Sites Encode their

Sensitivity to Starvation and Rapamycin. Science, 341, doi: 10.1126/science.1236566.

PMID: 24043828: Khanna, N., Fang, Y., Yoon, M.S. & Chen, J. (2013) XPLN is an Endogenous

Inhibitor of mTORC2. The Proceedings of the National Academy of Sciences of the United States of

America, 110, 15979-15984.

PMID: 24311379: Kim, J.D., Kang, Y., Kim, J., Papangeli, I., Kang, H., Wu, J., Park, H., Nadelmann,

E., Rockson, S.G., Chun, H.J. & Jin, S.W. (2014) Essential Role of Apelin Signaling during Lymphatic

Development in Zebrafish. Arteriosclerosis, Thrombosis, and Vascular Biology, 34, 338-345.

PMID: 23142081: Kim, S.J., DeStefano, M.A., Oh, W.J., Wu, C.C., Vega-Cotto, N.M., Finlan, M., Liu,

D., Su, B. & Jacinto, E. (2012) mTOR Complex 2 Regulates Proper Turnover of Insulin Receptor

Substrate-1 Via the Ubiquitin Ligase Subunit Fbw8. Molecular Cell, 48, 875-887.

PMID: 21508335: Li, J., Liu, J., Song, J., Wang, X., Weiss, H.L., Townsend Jr., C.M., Gao, T. & Evers,

B.M. (2011) mTORC1 Inhibition Increases Neurotensin Secretion and Gene Expression through

380

Activation of the MEK/ERK/c-Jun Pathway in the Human Endocrine Cell Line BON. American Journal

of Physiology - Cell Physiology, 301, C213-C226.

PMID: 24375676: Li, X. & Gao, T. (2014) mTORC2 Phosphorylates Protein Kinase Cζ to Regulate its

Stability and Activity. EMBO Reports, 15, 191-198.

PMID: 22223645: Liu, Q., Kirubakaran, S., Hur, W., Niepel, M., Westover, K., Thoreen, C.C., Wang,

J., Ni, J., Patricelli, M.P., Vogel, K., Riddle, S., Waller, D.L., Traynor, R., Sanda, T., Zhao, Z., Kang,

S.A., Zhao, J., Look, A.T., Sorger, P.K., Sabatini, D.M. & Gray, N.S. (2012) Kinome-Wide Selectivity

Profiling of ATP-Competitive Mammalian Target of Rapamycin (mTOR) Inhibitors and

Characterization of their Binding Kinetics. The Journal of Biological Chemistry, 287, 9742-9752.

PMID: 23436801: Liu, Q., Xu, C., Kirubakaran, S., Zhang, X., Hur, W., Liu, Y., Kwiatkowski, N.P.,

Wang, J., Westover, K.D., Gao, P., Ercan, D., Niepel, M., Thoreen, C.C., Kang, S.A., Patricelli, M.P.,

Wang, Y., Tupper, T., Altabef, A., Kawamura, H., Held, K.D., Chou, D.M., Elledge, S.J., Janne, P.A.,

Wong, K.K., Sabatini, D.M. & Gray, N.S. (2013) Characterization of Torin2, an ATP-Competitive

Inhibitor of mTOR, ATM, and ATR. Cancer Research, 73, 2574-2586.

PMID: 23707523: Liu, Y., Vertommen, D., Rider, M.H. & Lai, Y.C. (2013) Mammalian Target of Rapamycin-Independent S6K1 and 4E-BP1 Phosphorylation during Contraction in Rat Skeletal Muscle. Cellular Signalling, 25, 1877-1886.

PMID: 23934889: Ma, N., Liu, Q., Zhang, L., Henske, E.P. & Ma, Y. (2013) TORC1 Signaling is Governed by Two Negative Regulators in Fission Yeast. Genetics, 195, 457-468.

PMID: 22006022: Majmundar, A.J., Skuli, N., Mesquita, R.C., Kim, M.N., Yodh, A.G., Nguyen-McCarty, M. & Simon, M.C. (2012) O2 Regulates Skeletal Muscle Progenitor Differentiation through Phosphatidylinositol 3-kinase/AKT Signaling. Molecular and Cellular Biology, 32, 36-49.

PMID: 20089925: Malagelada, C., Jin, Z.H., Jackson-Lewis, V., Przedborski, S. & Greene, L.A. (2010) Rapamycin Protects Against Neuron Death in in Vitro and in Vivo Models of Parkinson's Disease. The Journal of Neuroscience, 30, 1166-1175.

PMID: 20735411: Mansley, M.K. & Wilson, S.M. (2010) Dysregulation of Epithelial Na+ Absorption Induced by Inhibition of the Kinases TORC1 and TORC2. British Journal of Pharmacology, 161, 1778-1792.

PMID: 23401004: Martina, J.A. & Puertollano, R. (2013) Rag GTPases Mediate Amino Acid-Dependent Recruitment of TFEB and MITF to Lysosomes. The Journal of Cell Biology, 200, 475-491.

PMID: 23963679: Montané, M.H. & Menand, B. (2013) ATP-Competitive mTOR Kinase Inhibitors Delay Plant Growth by Triggering Early Differentiation of Meristematic Cells but no Developmental Patterning Change. Journal of Experimental Botany, 64, 4361-4374.

PMID: 21592956: Moore, S.F., Hunter, R.W. & Hers, I. (2011) mTORC2 Protein Complex-Mediated Akt (Protein Kinase B) Serine 473 Phosphorylation is Not Required for Akt1 Activity in Human Platelets. The Journal of Biological Chemistry, 286, 24553-24560.

381

PMID: 23415771: Oubrahim, H., Wong, A., Wilson, B.A. & Boon Chock, P. (2013) Pasteurella Multocida Toxin (PMT) Upregulates CTGF which Leads to mTORC1 Activation in Swiss 3T3 Cells. Cellular Signalling, 25, 1136-1148.

PMID: 23524850: Schepetilnikov, M., Dimitrova, M., Mancera-Martinez, E., Geldreich, A., Keller, M. & Ryabova, L.A. (2013) TOR and S6K1 Promote Translation Reinitiation of uORF-Containing mRNAs Via Phosphorylation of eIF3h. The EMBO Journal, 32, 1087-1102.

PMID: 21278786: Serra, V., Scaltriti, M., Prudkin, L., Eichhorn, P.J., Ibrahim, Y.H., Chandarlapaty, S., Markman, B., Rodriguez, O., Guzman, M., Rodriguez, S., Gili, M., Russillo, M., Parra, J.L., Singh, S., Arribas, J., Rosen, N. & Baselga, J. (2011) PI3K Inhibition Results in Enhanced HER Signaling and Acquired ERK Dependency in HER2-Overexpressing Breast Cancer. Oncogene, 30, 2547-2557.

PMID: 22343943: Settembre, C., Zoncu, R., Medina, D.L., Vetrini, F., Erdin, S., Erdin, S., Huynh, T., Ferron, M., Karsenty, G., Vellard, M.C., Facchinetti, V., Sabatini, D.M. & Ballabio, A. (2012) A Lysosome-to-Nucleus Signalling Mechanism Senses and Regulates the Lysosome Via mTOR and TFEB. The EMBO Journal, 31, 1095-1108.

PMID: 22733130: Shanmugasundaram, K., Block, K., Nayak, B.K., Livi, C.B., Venkatachalam, M.A. & Sudarshan, S. (2013) PI3K Regulation of the SKP-2/p27 Axis through mTORC2. Oncogene, 32, 2027-2036.

PMID: 21858812: Shimizu, S., Takehara, T., Hikita, H., Kodama, T., Tsunematsu, H., Miyagi, T., Hosui, A., Ishida, H., Tatsumi, T., Kanto, T., Hiramatsu, N., Fujita, N., Yoshimori, T. & Hayashi, N. (2012) Inhibition of Autophagy Potentiates the Antitumor Effect of the Multikinase Inhibitor Sorafenib in Hepatocellular Carcinoma. International Journal of Cancer, 131, 548-557.

PMID: 22065737: Shin, S., Wolgamott, L., Yu, Y., Blenis, J. & Yoon, S.O. (2011) Glycogen Synthase Kinase (GSK)-3 Promotes p70 Ribosomal Protein S6 Kinase (p70S6K) Activity and Cell Proliferation. The Proceedings of the National Academy of Sciences of the United States of America, 108, E1204-E1213.

PMID: 21841310: Shuda, M., Kwun, H.J., Feng, H., Chang, Y. & Moore, P.S. (2011) Human Merkel Cell Polyomavirus Small T Antigen is an Oncoprotein Targeting the 4E-BP1 Translation Regulator. The Journal of Clinical Investigation, 121, 3623-3634.

PMID: 22053105: Smrž, D., Kim, M.S., Zhang, S., Mock, B.A., Smržová, S., DuBois, W., Simakova, O., Maric, I., Wilson, T.M., Metcalfe, D.D. & Gilfillan, A.M. (2011) mTORC1 and mTORC2 Differentially Regulate Homeostasis of Neoplastic and Non-Neoplastic Human Mast Cells. Blood, 118, 6803-6813.

PMID: 22552098: Thoreen, C.C., Chantranupong, L., Keys, H.R., Wang, T., Gray, N.S. & Sabatini, D.M. (2012) A Unifying Model for mTORC1-Mediated Regulation of mRNA Translation. Nature, 485, 109-113.

PMID: 19150980: Thoreen, C.C., Kang, S.A., Chang, J.W., Liu, Q., Zhang, J., Gao, Y., Reichling, L.J., Sim, T., Sabatini, D.M. & Gray, N.S. (2009) An ATP-Competitive Mammalian Target of Rapamycin Inhibitor Reveals Rapamycin-Resistant Functions of mTORC1. The Journal of Biological Chemistry, 284, 8023-8032.

382

PMID: 24008870: Völkers, M., Konstandin, M.H., Doroudgar, S., Toko, H., Quijada, P., Din, S., Joyo, A., Ornelas, L., Samse, K., Thuerauf, D.J., Gude, N., Glembotski, C.C. & Sussman, M.A. (2013) Mechanistic Target of Rapamycin Complex 2 Protects the Heart from Ischemic Damage. Circulation, 128, 2132-2144.

PMID: 21908613: Warfel, N.A., Niederst, M. & Newton, A.C. (2011) Disruption of the Interface between the Pleckstrin Homology (PH) and Kinase Domains of Akt Protein is Sufficient for Hydrophobic Motif Site Phosphorylation in the Absence of mTORC2. The Journal of Biological Chemistry, 286, 39122-39129.

PMID: 23077579: You, J.S., Frey, J.W. & Hornberger, T.A. (2012) Mechanical Stimulation Induces mTOR Signaling Via an ERK-Independent Mechanism: Implications for a Direct Activation of mTOR by Phosphatidic Acid. PLoS ONE, 7, doi: 10.1371/journal.pone.0047258.

XMD8-92 Mock Dataset

PMID: 23428871: Erazo, T., Moreno, A., Ruiz-Babot, G., Rodriguez-Asiain, A., Morrice, N.A.,

Espadamala, J., Bayascas, J.R., Gomez, N. & Lizcano, J.M. (2013) Canonical and Kinase Activity-

Independent Mechanisms for Extracellular Signal-Regulated Kinase 5 (ERK5) Nuclear Translocation

Require Dissociation of Hsp90 from the ERK5-Cdc37 Complex. Molecular and Cellular Biology, 33,

1671-1686.

PMID: 24880079: Sureban, S. M., May, R., Weygant, N., Qu, D., Chandrakesan, P., Bannerman-

Menson, E., Ali, N., Pantazis, P., Westphalen, C.B., Wang, T.C. & Houchen, C.W. (2014) XMD8-92

Inhibits Pancreatic Tumor Xenograft Growth Via a DCLK1-Dependent Mechanism. Cancer

Letters, 351, 151-161.

PMID: 20832753: Yang, Q., Deng, X., Lu, B., Cameron, M., Fearns, C., Patricelli, M.P., Yates III, J.R.,

Gray, N.S. & Lee, J.D. (2010) Pharmacological Inhibition of BMK1 Suppresses Tumor Growth through

Promyelocytic Leukemia Protein. Cancer Cell, 18, 258-267.

383

Supplementary Information 1b - Benchmark Dataset References for the Binary Model with Lipopolysaccharide

(LPS Benchmark)

The references shown below were used in the construction of the benchmark dataset concerning the effects each inhibitor

has on Lipopolysaccharide stimulated cells.

Note: Those that are faded denote studies conducted with Macrophage cell lines. These are used in Section 2.2.2.1 –

Updated Benchmark Dataset as the basis of the benchmark dataset

CGP57380 & LPS Mock Dataset

PMID: 16431125: Andersson, K. & Sundler, R. (2006) Posttranscriptional Regulation of TNFα

Expression via Eukaryotic Initiation Factor 4E (eIF4E) Phosphorylation in Mouse Macrophages.

Cytokine, 33, 52-57.

PMID: 23401599: Fortin, C.F., Mayer, T.Z., Cloutier, A. & McDonald, P.P. (2013) Translational

Control of Human Neutrophil Responses by MNK1. Journal of Leukocyte Biology, 94, 693-703.

PMID: 20375303: Gais, P., Tiedje, C., Altmayr, F., Gaestel, M., Weighardt, H. & Holzmann, B. (2010)

TRIF Signaling Stimulates Translation of TNF-α mRNA via Prolonged Activation of MK2. The Journal of

Immunology, 184, 5842-5848.

PMID: 24803542: Giambelluca, M.S., Bertheau-Mailhot, G., Laflamme, C., Rollet-Labelle, E.,

Servant, M.J. & Pouliot, M. (2014) TNF-α Expression in Neutrophils and its Regulation by Glycogen

Synthase Kinase-3: A Potentiating Role for Lithium. FASEB Journal, doi: 10.1096/fj.14-251900 fj.14-

251900

PMID: 22955274: Chan, C.S., Ming-Lum, A., Golds, G.B., Lee, S.J., Anderson, R.J. & Mui, A.L. (2012)

Interleukin-10 Inhibits Lipopolysaccharide-Induced Tumor Necrosis Factor-α Translation through a

SHIP1-Dependent Pathway. The Journal of Biological Chemistry, 287, 38020-38027.

PMID: 18032482: Rowlett, R.M., Chrestensen, C.A., Nyce, M., Harp, M.G., Pelo, J.W., Cominelli, F.,

Ernst, P.B., Pizarro, T.T., Sturgill, T.W. & Worthington, M.T. (2008) MNK Kinases Regulate Multiple

384

TLR Pathways and Innate Proinflammatory Cytokines in Macrophages. American Journal of

Physiology - Gastrointestinal and Liver Physiology, 294, G452-G459.

PMID: 22610140: Xu, H., Zhu, J., Smith, S., Foldi, J., Zhao, B., Chung, A.Y., Outtz, H., Kitajewski, J.,

Shi, C., Weber, S., Saftig, P., Li, Y., Ozato, K., Blobel, C.P., Ivashkiv, L.B. & Hu, X. (2012) Notch-RBP-J

Signaling Regulates the Transcription Factor IRF8 to Promote Inflammatory Macrophage

Polarization. Nature Immunology, 13, 642-650.

LY294002 & LPS Mock Dataset

PMID: 15701712: An, H., Xu, H., Zhang, M., Zhou, J., Feng, T., Qian, C., Qi, R. & Cao, X. (2005) Src

Homology 2 Domain-Containing Inositol-5-Phosphatase 1 (SHIP1) Negatively Regulates TLR4-

Mediated LPS Response Primarily through a Phosphatase Activity- and PI-3K-Independent

Mechanism. Blood, 105, 4685-4692.

PMID: 10910920: Ardeshna, K.M., Pizzey, A.R., Devereux, S. & Khwaja, A. (2000) The PI3 Kinase,

p38 SAP Kinase, and NFκB Signal Transduction Pathways are Involved in the Survival and Maturation

of Lipopolysaccharide-Stimulated Human Monocyte-Derived Dendritic Cells. Blood, 96, 1039-1046.

PMID: 17405917: Dos Santos, S., Delattre, A.I., De Longueville, F., Bult, H. & Raes, M. (2007) Gene

Expression Profiling of LPS-Stimulated Murine Macrophages and Role of the NF-κB and PI3K/mTOR

Signaling Pathways. Annals of the New York Academy of Sciences, 1096, 70-77.

PMID: 17579042: Dye, J.R., Palvanov, A., Guo, B. & Rothstein, T.L. (2007) B Cell Receptor Cross-

Talk: Exposure to Lipopolysaccharide Induces an Alternate Pathway for B Cell Receptor-Induced ERK

Phosphorylation and NF-κB Activation. The Journal of Immunology, 179, 229-235.

PMID: 20163623: Gerilechaogetu, Hanieh, H., Abe, A. & Kondo, Y. (2009) Extracellular Signal-

Regulated Kinase (ERK) Activation in Chicken Heterophils Stimulated with Phorbol 12-Myristate 13-

Acetate (PMA), Formyl-Methionylleucyl-Phenylalanine (fMLP) and Lipopolysaccharide (LPS). Animal

Science Journal, 80, 577-584.

PMID: 12052830: Guha, M. & Mackman, N. (2002) The Phosphatidylinositol 3-Kinase-Akt Pathway

Limits Lipopolysaccharide Activation of Signaling Pathways and Expression of Inflammatory

Mediators in Human Monocytic Cells. The Journal of Biological Chemistry, 277, 32124-32132.

PMID: 16762341: Gutierrez-Venegas, G., Kawasaki-Cardenas, P., Arroyo-Cruz, S.R. & Maldonado-

Frias, S. (2006) Luteolin Inhibits Lipopolysaccharide Actions on Human Gingival Fibroblasts. European

Journal of Pharmacology, 541, 95-105.

PMID: 9195953: Herrera-Velit, P., Knutson, K.L. & Reiner, N.E. (1997) Phosphatidylinositol 3-

Kinase-Dependent Activation of Protein Kinase C-ζin Bacterial Lipopolysaccharide-Treated Human

Monocytes. The Journal of Biological Chemistry, 272, 16445-16452.

385

PMID: 9873050: Hmama, Z., Knutson, K.L., Herrera-Velit, P., Nandan, D. & Reiner, N.E. (1999)

Monocyte Adherence Induced by Lipopolysaccharide Involves CD14, LFA-1, and Cytohesin-1.

Regulation by Rho and Phosphatidylinositol 3-Kinase. The Journal of Biological Chemistry, 274, 1050-

1057.

PMID: 17367785: Jung, I.D., Lee, C.M., Jeong, Y.I., Lee, J.S., Park, W.S., Han, J. & Park, Y.M. (2007)

Differential Regulation of Indoleamine 2,3-Dioxygenase by Lipopolysaccharide and Interferon

Gamma in Murine Bone Marrow Derived Dendritic Cells. FEBS letters, 581, 1449-1456.

PMID: 15894110: Kim, Y.H., Choi, K.H., Park, J.W. & Kwon, T.K. (2005) LY294002 Inhibits LPS-

Induced NO Production through a Inhibition of NF-κB Activation: Independent Mechanism of

Phosphatidylinositol 3-Kinase. Immunology Letters, 99, 45-50.

PMID: 15153523: Kim, W.K., Hwang, S.Y., Oh, E.S., Piao, H.Z., Kim, K.W. & Han, I.O. (2004) TGF-β1

Represses Activation and Resultant Death of Microglia via Inhibition of Phosphatidylinositol 3-Kinase

Activity. The Journal of Immunology, 172, 7015-7023.

PMID: 11129660: Lee, J., Mira-Arbibe, L. & Ulevitch, R.J. (2000) TAK1 Regulates Multiple Protein

Kinase Cascades Activated by Bacterial Lipopolysaccharide. Journal of Leukocyte Biology, 68, 909-

915.

PMID: 12847238: Martin, M., Schifferle, R.E., Cuesta, N., Vogel, S.N., Katz, J. & Michalek, S.M.

(2003) Role of the Phosphatidylinositol 3 Kinase-Akt Pathway in the Regulation of IL-10 and IL-12 by

Porphyromonas gingivalis Lipopolysaccharide. The Journal of Immunology, 171, 717-725.

PMID: 12072439: Monick, M.M., Robeff, P.K., Butler, N.S., Flaherty, D.M., Carter, A.B., Peterson,

M.W. & Hunninghake, G.W. (2002) Phosphatidylinositol 3-Kinase Activity Negatively Regulates

Stability of Cyclooxygenase 2 mRNA. The Journal of Biological Chemistry, 277, 32992-33000.

PMID: 9862729: Salh, B., Wagey, R., Marotta, A., Tao, J.S. & Pelech, S. (1998) Activation of

Phosphatidylinositol 3-Kinase, Protein Kinase B, and p70 S6 Kinases in Lipopolysaccharide-Stimulated

Raw 264.7 Cells: Differential Effects of Rapamycin, Ly294002, and Wortmannin on Nitric Oxide

Production. The Journal of Immunology, 161, 6947-6954.

PMID: 20828737: Schaeffer, V., Arbabi, S., Garcia, I.A., Knoll, M.L., Cuschieri, J., Bulger, E.M. &

Maier, R.V. (2011) Role of the mTOR Pathway in LPS-Activated Monocytes: Influence of Hypertonic

Saline. The Journal of Surgical Research, 171, 769-776.

PMID: 10733102: Weinstein, S.L., Finn, A.J., Dave, S.H., Meng, F., Lowell, C.A., Sanghera, J.S. &

DeFranco, A.L. (2000) Phosphatidylinositol 3-Kinase and mTOR Mediate Lipopolysaccharide-

Stimulated Nitric Oxide Production in Macrophages via Interferon-β. Journal of Leukocyte Biology,

67, 405-414.

PMID: 14996708: Wong, F., Hull, C., Zhande, R., Law, J. & Karsan, A. (2004) Lipopolysaccharide

Initiates a TRAF6-Mediated Endothelial Survival Signal. Blood, 103, 4520-4526.

PMID: 22285490: Zhao, W., Qi, J., Wang, L., Zhang, M., Wang, P. & Gao, C. (2012) LY294002

Inhibits TLR3/4-Mediated IFN-β Production via Inhibition of IRF3 Activation with a PI3K-Independent

Mechanism. FEBS Letters, 586, 705-710.

386

PD184352 & LPS Mock Dataset

PMID: 23997213: Barrio, L., Saez de Guinoa, J. & Carrasco, Y.R. (2013) TLR4 Signaling Shapes B Cell

Dynamics via MyD88-Dependent Pathways and Rac GTPases. The Journal of Immunology, 191, 3867-

3875.

PMID: 24341623: Edgar, A.J., Trost, M., Watts, C. & Zaru, R. (2013) A Combination of SILAC and

Nucleotide Acyl Phosphate Labelling Reveals Unexpected Targets of the Rsk Inhibitor BI-D1870.

Bioscience Reports, doi:10.1042/BSR20130094

PMID: 21363926: Hsu, R.Y., Chan, C.H., Spicer, J.D., Rousseau, M.C., Giannias, B., Rousseau, S. &

Ferri, L.E. (2011) LPS-Induced TLR4 Signaling in Human Colorectal Cancer Cells Increases β1 Integrin-

Mediated Cell Adhesion and Liver Metastasis. Cancer Research, 71, 1989-1998.

PMID: 19667062: Kaiser, F., Cook, D., Papoutsopoulou, S., Rajsbaum, R., Wu, X., Yang, H.T., Grant,

S., Ricciardi-Castagnoli, P., Tsichlis, P.N., Ley, S.C. & O'Garra, A. (2009) TPL-2 Negatively Regulates

Interferon-β Production in Macrophages and Myeloid Dendritic Cells. The Journal of Experimental

Medicine, 206, 1863-1871.

PMID: 16848763: Morton, S., Yang, H.T., Moleleki, N., Campbell, D.G., Cohen, P. & Rousseau, S.

(2006) Phosphorylation of the ARE-Binding Protein DAZAP1 by ERK2 Induces its Dissociation from

DAZ. The Biochemical Journal, 399, 265-273.

PMID: 23979601: McGuire, V.A., Gray, A., Monk, C.E., Santos, S.G., Lee, K., Aubareda, A., Crowe,

J., Ronkina, N., Schwermann, J., Batty, I.H., Leslie, N.R., Dean, J.L., O'Keefe, S.J., Boothby, M.,

Gaestel, M. & Arthur, J.S. (2013) Cross Talk Between the Akt and p38α Pathways in Macrophages

Downstream of Toll-Like Receptor Signaling. Molecular and Cellular Biology, 33, 4152-4165.

PMID: 22733747: Risco, A., del Fresno, C., Mambol, A., Alsina-Beauchamp, D., MacKenzie, K.F.,

Yang, H.T., Barber, D.F., Morcelle, C., Arthur, J.S., Ley, S.C., Ardavin, C. & Cuenda, A. (2012) p38γ

and p38δ Kinases Regulate the Toll-Like Receptor 4 (TLR4)-Induced Cytokine Production by

Controlling ERK1/2 Protein Kinase Pathway Activation. The Proceedings of the National Academy of


PMID: 18187448: Rousseau, S., Papoutsopoulou, M., Symons, A., Cook, D., Lucocq, J.M., Prescott,

A.R., O'Garra, A., Ley, S.C. & Cohen, P. (2008) TPL2-Mediated Activation of ERK1 and ERK2 Regulates

the Processing of pre-TNFα in LPS-Stimulated Macrophages. Journal of Cell Science, 121, 149-154.

PMID: 22733995: Yang, H.T., Papoutsopoulou, S., Belich, M., Brender, C., Janzen, J., Gantke, T.,

Handley, M. & Ley, S.C. (2012) Coordinate Regulation of TPL-2 and NF-κB Signaling in Macrophages

by NF-κB1 p105. Molecular and Cellular Biology, 32, 3438-3451.

PMID: 21217011: Yang, H.T., Wang, Y., Zhao, X., Demissie, E., Papoutsopoulou, S., Mambole, A.,

O'Garra, A., Tomczak, M.F., Erdman, S.E., Fox, J.G., Ley, S.C. & Horwitz, B.H. (2011) NF-κB1 Inhibits

TLR-Induced IFN-β Production in Macrophages through TPL-2-Dependent ERK Activation. The Journal


387




PD98059 & LPS Mock Dataset

PMID: 10910920: Ardeshna, K.M., Pizzey, A.R., Devereux, S. & Khwaja, A. (2000) The PI3 Kinase,

p38 SAP Kinase, and NF-κB Signal Transduction Pathways are involved in the Survival and Maturation

of Lipopolysaccharide-Stimulated Human Monocyte-Derived Dendritic Cells. Blood, 96, 1039-1046.

PMID: 19823867: Baines, D.L., Albert, A.P., Hazell, M.J., Gambling, L., Woollhead, A.M. & Dockrell,

M.E. (2010) Lipopolysaccharide Modifies Amiloride-Sensitive Na+ Transport Processes Across Human

Airway Cells: Role of Mitogen-Activated Protein Kinases ERK 1/2 and 5. Pflügers Archiv, 459, 451-

463.

PMID: 9464988: Bhat, N.R., Zhang, P., Lee, J.C. & Hogan, E.L. (1998) Extracellular Signal-Regulated

Kinase and p38 Subgroups of Mitogen-Activated Protein Kinases Regulate Inducible Nitric Oxide

Synthase and Tumor Necrosis Factor-α Gene Expression in Endotoxin-Stimulated Primary Glial

Cultures. The Journal of Neuroscience, 18, 1633-1641.

PMID: 12556480: Castrillo, A., Través, P.G., Martín-Sanz, P., Parkinson, S., Parker, P.J. & Boscá, L.

(2003) Potentiation of Protein Kinase C ζ Activity by 15-Deoxy-Δ12,14-Prostaglandin J2 Induces an

Imbalance Between Mitogen-Activated Protein Kinases and NF-κB that Promotes Apoptosis in

Macrophages. Molecular and Cellular Biology, 23, 1196-1208.

PMID: 11159988: Chakravortty, D., Kato, Y., Sugiyama, T., Koide, N., Mu, M.M., Yoshida, T. &

Yokochi, T. (2001) Inhibition of p38 Mitogen-Activated Protein Kinase Augments Lipopolysaccharide-

Induced Cell Proliferation in CD14-Expressing Chinese Hamster Ovary Cells. Infection and Immunity,

69, 931-936.

PMID: 11163183: Dumitru, C.D., Ceci, J.D., Tsatsanis, C., Kontoyiannis, D., Stamatakis, K., Lin, J.H.,

Patriotis, C., Jenkins, N.A., Copeland, N.G., Kollias, G. & Tsichlis, P.N. (2000) TNF-α Induction by LPS

is Regulated Posttranscriptionally via a Tpl2/ERK-Dependent Pathway. Cell, 103, 1071-1083.

PMID: 12234923: Eliopoulos, A.G., Dumitru, C.D., Wang, C.C., Cho, J. Tsichlis, P.N. (2002) Induction

of COX-2 by LPS in Macrophages is Regulated by Tpl2-Dependent CREB Activation Signals. The EMBO

Journal, 21, 4831-4840.

PMID: 12829589: Franchi, L., Condo, I., Tomassini, B., Nicolo, C. & Testi, R. (2003) A Caspaselike

Activity is Triggered by LPS and is Required for Survival of Human Dendritic Cells. Blood, 102, 2910-

2915.

PMID: 16762341: Gutiérrez-Venegas, G., Kawasaki-Cárdenas, P., Arroyo-Cruz, S.R. & Maldonado-

Frías, S. (2006) Luteolin Inhibits Lipopolysaccharide Actions on Human Gingival Fibroblasts. European

Journal of Pharmacology, 541, 95-105.

388

PMID: 21436242: Kanoh, S., Tanabe, T. & Rubin, B.K. (2011) Dapsone Inhibits IL-8 Secretion from

Human Bronchial Epithelial Cells Stimulated with Lipopolysaccharide and Resolves Airway

Inflammation in the Ferret. Chest, 140, 980-990.

PMID: 15566575: Lee, H.S., Kim, H.J., Moon, C.S., Chong, Y.H. & Kang, J.L. (2004) Inhibition of c-Jun

NH2-Terminal Kinase or Extracellular Signal-Regulated Kinase Improves Lung Injury. Respiratory

Research, 5, doi:10.1186/1465-9921-5-23

PMID: 23960234: Madera-Salcedo, I.K., Cruz, S.L. & Gonzalez-Espinosa, C. (2013) Morphine

Prevents Lipopolysaccharide-Induced TNF Secretion in Mast Cells Blocking IκB Kinase Activation and

SNAP-23 Phosphorylation: Correlation with the Formation of a β-Arrestin/TRAF6 Complex. The


PMID: 12244175: Masuda, A., Yoshikai, Y., Aiba, K. & Matsuguchi, T. (2002) Th2 Cytokine

Production from Mast Cells is Directly Induced by Lipopolysaccharide and Distinctly Regulated by c-

Jun N-Terminal Kinase and p38 Pathways. The Journal of Immunology, 169, 3801-3810.

PMID: 10024537: Procyk, K.J., Kovarik, P., von Gabain, A. & Baccarini, M. (1999) Salmonella

typhimurium and Lipopolysaccharide Stimulate Extracellularly Regulated Kinase Activation in

Macrophages by a Mechanism Involving Phosphatidylinositol 3-Kinase and Phospholipase D as Novel

Intermediates. Infection and Immunity, 67, 1011-1017.

PMID: 15025824: Serikov, V.B., Choi, H., Schmiel, K., Skaggs, C., Fleming, N.W., Wu, R. &

Widdicombe, J.H. (2004) Endotoxin Induces Leukocyte Transmigration and Changes in Permeability

of the Airway Epithelium via Protein-Kinase C and Extracellular Regulated Kinase Activation. Journal

of Endotoxin Research, 10, 55-65.

PMID: 21338578: Svensson, C., Part, K., Kunnis-Beres, K., Kaldmae, M., Fernaeus, S.Z. & Land, T.

(2011) Pro-Survival Effects of JNK and p38 MAPK Pathways in LPS-Induced Activation of BV-2 cells.

Biochemical and Biophysical Research Communications, 406, 488-492.

PMID: 21775707: Tarabees, R., Hill, D., Rauch, C., Barrow, P.A. & Loughna, P.T. (2011) Endotoxin

Transiently Inhibits Protein Synthesis through Akt and MAPK Mediating Pathways in C2C12

Myotubes. American Journal of Physiology - Cell Physiology, 301, C895-902.

PMID: 24619416: Wang, Q., He, Y., Shen, Y., Zhang, Q., Chen, D., Zuo, C., Qin, J., Wang, H., Wang,

J. & Yu, Y. (2014) Vitamin D Inhibits COX-2 Expression and Inflammatory Response by Targeting

Thioesterase Superfamily Member 4. The Journal of Biological Chemistry, 289, 11681-11694.

PMID: 17109352: Weng, T.I., Chen, W.J., Wu, H.Y. & Liu, S.H. (2006) Uropathogenic Escherichia coli

Alters Muscle Contractions in Rat Urinary Bladder via a Nitric Oxide Synthase-Related Signaling

Pathway. The Journal of Infectious Diseases, 194, 1774-1782.

PMID: 12145106: Xagorari, A., Roussos, C. & Papapetropoulos, A. (2002) Inhibition of LPS-

Stimulated Pathways in Macrophages by the Flavonoid Luteolin. British Journal of Pharmacology,

136, 1058-1064.

389

SB203580 & LPS Mock Dataset

PMID: 16431125: Andersson, K. & Sundler, R. (2006) Posttranscriptional Regulation of TNFα

Expression via Eukaryotic Initiation Factor 4E (eIF4E) Phosphorylation in Mouse Macrophages.

Cytokine, 33, 52-57.

PMID: 11033356: Brook, M., Sully, G., Clark, A.R. & Saklatvala, J. (2000) Regulation of Tumour

Necrosis Factor α mRNA Stability by the Mitogen-Activated Protein Kinase p38 Signalling Cascade.

FEBS Letters, 483, 57-61.

PMID: 11159988: Chakravortty, D., Kato, Y., Sugiyama, T., Koide, N., Mu, M.M., Yoshida, T. &

Yokochi, T. (2001) Inhibition of p38 Mitogen-Activated Protein Kinase Augments Lipopolysaccharide-

Induced Cell Proliferation in CD14-Expressing Chinese Hamster Ovary Cells. Infection and Immunity,

69, 931-936.

PMID: 17916230: Chanteux, H., Guisset, A.C., Pilette, C. & Sibille, Y. (2007) LPS Induces IL-10

Production by Human Alveolar Macrophages via MAPKinases- and Sp1-Dependent Mechanisms.

Respiratory Research, 8, doi:10.1186/1465-9921-8-71

PMID: 19026643: Liu, X.Y., Seh, C.C. & Cheung, P.C. (2008) HSP90 is Required for TAK1 Stability but

not for its Activation in the Pro-Inflammatory Signaling Pathway. FEBS Letters, 582, 4023-4031.

PMID: 9753474: Frantz, B., Klatt, T., Pang, M., Parsons, J., Rolando, A., Williams, H., Tocci, M.J.,

O'Keefe, S.J. & O'Neill, E.A. (1998) The Activation State of p38 Mitogen-Activated Protein Kinase

Determines the Efficiency of ATP Competition for Pyridinylimidazole Inhibitor Binding. Biochemistry,

37, 13846-13853.

PMID: 23212905: Gupta, A., Cooper, Z.A., Tulapurkar, M.E., Potla, R., Maity, T., Hasday, J.D. &

Singh, I.S. (2013) Toll-Like Receptor Agonists and Febrile Range Hyperthermia Synergize to Induce

Heat Shock Protein 70 Expression and Extracellular Release. The Journal of Biological Chemistry, 288,

2756-2766.

PMID: 23928505: Han, H.J., Li, M., Son, J.K., Seo, C.S., Song, S.W., Kwak, S.H. & Bae, H.B. (2013)

Sauchinone, a Lignan from Saururus chinensis, Attenuates Neutrophil Pro-Inflammatory Activity and

Acute Lung Injury. International Immunopharmacology, 17, 471-477.

PMID: 23872471: Iliev, D.B., Hansen, T., Jørgensen, S.M., Krasnov, A. and Jørgensen, J.B. (2013)

CpG- and LPS-Activated MAPK Signaling in in Vitro Cultured Salmon (Salmo salar) Mononuclear

Phagocytes. Fish & Shellfish Immunology, 35, 1079-1085.



Weichhart, T. (2013) p38α Senses Environmental Stress to Control Innate Immune Responses via


390

PMID: 18783239: Ku, K.T., Huang, Y.L., Huang, Y.J. & Chiou, W.F. (2008) Miyabenol A Inhibits LPS-

Induced NO Production via IKK/IκB Inactivation in RAW 264.7 Macrophages: Possible Involvement of

the p38 and PI3K Pathways. Journal of Agricultural and Food Chemistry, 56, 8911-8918.

PMID: 12794156: Lai, W.C., Zhou, M., Shankavaram, U., Peng, G. & Wahl, L.M. (2003) Differential

Regulation of Lipopolysaccharide-Induced Monocyte Matrix Metalloproteinase (MMP)-1 and MMP-9

by p38 and Extracellular Signal-Regulated Kinase 1/2 Mitogen-Activated Protein Kinases. The Journal


PMID: 11129660: Lee, J., Mira-Arbibe, L. & Ulevitch, R.J. (2000) TAK1 Regulates Multiple Protein

Kinase Cascades Activated by Bacterial Lipopolysaccharide. Journal of Leukocyte Biology, 68, 909-

915.

PMID: 10952668: Luo, S.F., Wang, C.C., Chiu, C.T., Chien, C.S., Hsiao, L.D., Lin, C.H. & Yang, C.M.

(2000) Lipopolysaccharide Enhances Bradykinin-Induced Signal Transduction via Activation of

Ras/Raf/MEK/MAPK in Canine Tracheal Smooth Muscle Cells. British Journal of Pharmacology, 130,

1799-1808.

PMID: 23979601: McGuire, V.A., Gray, A., Monk, C.E., Santos, S.G., Lee, K., Aubareda, A., Crowe,

J., Ronkina, N., Schwermann, J., Batty, I.H., Leslie, N.R., Dean, J.L., O'Keefe, S.J., Boothby, M.,

Gaestel, M. & Arthur, J.S. (2013) Cross Talk Between the Akt and p38α Pathways in Macrophages

Downstream of Toll-Like Receptor Signaling. Molecular and Cellular Biology, 33, 4152-4165.

PMID: 21743791: Ponzetti, K., King, M., Gates, A., Anwer, M.S. & Webster, C.R. (2010) Cyclic AMP-

Guanine Exchange Factor Activation Inhibits JNK-Dependent Lipopolysaccharide-Induced Apoptosis

in Rat Hepatocytes. Hepatic Medicine: Evidence and Research, 2010, 1-11.

PMID: 11319880: Potter, M.W., Shah, S.A., Elbirt, K.K. & Callery, M.P. (2001) Endotoxin (LPS)

Stimulates 4E-BP1/PHAS-I Phosphorylation in Macrophages. The Journal of Surgical Research, 97, 54-

59.

PMID: 15979653: Shen, J., Sakaida, I., Uchida, K., Terai, S. & Okita, K. (2005) Leptin Enhances TNF-α

Production via p38 and JNK MAPK in LPS-Stimulated Kupffer Cells. Life Sciences, 77, 1502-1515.




PMID: 21775707: Tarabees, R., Hill, D., Rauch, C., Barrow, P.A. & Loughna, P.T. (2011) Endotoxin

Transiently Inhibits Protein Synthesis through Akt and MAPK Mediating Pathways in C2C12

Myotubes. American Journal of Physiology - Cell Physiology, 301, C895-902.

PMID: 11123340: van den Blink, B., Juffermans, N.P., ten Hove, T., Schultz, M.J., van Deventer,

S.J., van der Poll, T. & Peppelenbosch, M.P. (2001) p38 Mitogen-Activated Protein Kinase Inhibition

Increases Cytokine Release by Macrophages in Vitro and during Infection in Vivo. The Journal of


391

PMID: 12145106: Xagorari, A., Roussos, C. & Papapetropoulos, A. (2002) Inhibition of LPS-

Stimulated Pathways in Macrophages by the Flavonoid Luteolin. British Journal of Pharmacology,

136, 1058-1064.




SP600125 & LPS Mock Dataset

PMID: 14699155: Arndt, P.G., Suzuki, N., Avdi, N.J., Malcolm, K.C. & Worthen, G.S. (2004)

Lipopolysaccharide-Induced c-Jun NH2-Terminal Kinase Activation in Human Neutrophils: Role of

Phosphatidylinositol 3-Kinase and Syk-Mediated Pathways. The Journal of Biological Chemistry, 279,

10883-10891.

PMID: 12556480: Castrillo, A., Través, P.G., Martín-Sanz, P., Parkinson, S., Parker, P.J. & Boscá, L.

(2003) Potentiation of Protein Kinase C ζ Activity by 15-Deoxy-Δ12,14-Prostaglandin J2 Induces an

Imbalance Between Mitogen-Activated Protein Kinases and NF-κB that Promotes Apoptosis in

Macrophages. Molecular and Cellular Biology, 23, 1196-1208.

PMID: 12867362: Comalada, M., Xaus, J., Valledor, A.F., López-López, C., Pennington, D.J. &

Celada, A. (2003) PKCε is Involved in JNK Activation that Mediates LPS-Induced TNF-α, Which

Induces Apoptosis in Macrophages. American Journal of Physiology - Cell Physiology, 285, C1235-

C1245.

PMID: 16978663: Hirasawa, N., Torigoe, M., Ohgawara, R., Murakami, A. & Ohuchi, K. (2006)

Involvement of MAP Kinases in Lipopolysaccharide-Induced Histamine Production in RAW 264 Cells.

Life Sciences, 80, 36-42.

PMID: 16797002: Jang, S.I., Kim, H.J., Kim, Y.J., Jeong, S.I. & You, Y.O. (2006) Tanshinone IIA

Inhibits LPS-Induced NF-κB Activation in RAW 264.7 Cells: Possible Involvement of the NIK-IKK,

ERK1/2, p38 and JNK Pathways. European Journal of Pharmacology, 542, 1-7.

PMID: 12438403: Jarvis, B.W., Harris, T.H., Qureshi, N. & Splitter, G.A. (2002) Rough

Lipopolysaccharide from Brucella abortus and Escherichia coli Differentially Activates the same

Mitogen-Activated Protein Kinase Signaling Pathways for Tumor Necrosis Factor Alpha in RAW 264.7

Macrophage-Like Cells. Infection and Immunity, 70, 7165-7168.

PMID: 12869635: Lahti, A., Jalonen, U., Kankaanranta, H. & Moilanen, E. (2003) c-Jun NH2-

Terminal Kinase Inhibitor Anthra(1,9-cd)pyrazol-6(2H)-one Reduces Inducible Nitric-Oxide Synthase

Expression by Destabilizing mRNA in Activated Macrophages. Molecular Pharmacology, 64, 308-315.

392

PMID: 15566575: Lee, H.S., Kim, H.J., Moon, C.S., Chong, Y.H. & Kang, J.L. (2004) Inhibition of c-Jun

NH2-Terminal Kinase or Extracellular Signal-Regulated Kinase Improves Lung Injury. Respiratory

Research, 5, doi:10.1186/1465-9921-5-23

PMID: 19299915: Lee, Y.S., Lan Tran, H.T. & Van Ta, Q. (2009) Regulation of Expression of Matrix

Metalloproteinase-9 by JNK in Raw 264.7 Cells: Presence of Inhibitory Factor(s) Suppressing MMP-9

Induction in Serum and Conditioned Media. Experimental & Molecular Medicine, 41, 259-268.

PMID: 20196785: Liu, C.J., Lo, J.F., Kuo, C.H., Chu, C.H., Chen, L.M., Tsai, F.J., Tsai, C.H., Tzang, B.S.,

Kuo, W.W. & Huang, C.Y. (2009) Akt Mediates 17β-Estradiol and/or Estrogen Receptor-α Inhibition

of LPS-Induced Tumor Necresis Factor-α Expression and Myocardial Cell Apoptosis by Suppressing

the JNK1/2-NFκB Pathway. Journal of Cellular and Molecular Medicine, 13, 3655-3667.

PMID: 21743791: Ponzetti, K., King, M., Gates, A., Anwer, M.S. & Webster, C.R. (2010) Cyclic AMP-

Guanine Exchange Factor Activation Inhibits JNK-Dependent Lipopolysaccharide-Induced Apoptosis

in Rat Hepatocytes. Hepatic Medicine: Evidence and Research, 2010, 1-11.




PMID: 12847227: Utsugi, M., Dobashi, K., Ishizuka, T., Endou, K., Hamuro, J., Murata, Y.,

Nakazawa, T. & Mori, M. (2003) c-Jun N-Terminal Kinase Negatively Regulates Lipopolysaccharide-

Induced IL-12 Production in Human Macrophages: Role of Mitogen-Activated Protein Kinase in

Glutathione Redox Regulation of IL-12 Production. The Journal of Immunology, 171, 628-635.

PMID: 24619416: Wang, Q., He, Y., Shen, Y., Zhang, Q., Chen, D., Zuo, C., Qin, J., Wang, H., Wang,

J. & Yu, Y. (2014) Vitamin D Inhibits COX-2 Expression and Inflammatory Response by Targeting

Thioesterase Superfamily Member 4. The Journal of Biological Chemistry, 289, 11681-11694.

PMID: 22610140: Xu, H., Zhu, J., Smith, S., Foldi, J., Zhao, B., Chung, A.Y., Outtz, H., Kitajewski, J.,

Shi, C., Weber, S., Saftig, P., Li, Y., Ozato, K., Blobel, C.P., Ivashkiv, L.B. & Hu, X. (2012) Notch-RBP-J

Signaling Regulates the Transcription Factor IRF8 to Promote Inflammatory Macrophage

Polarization. Nature Immunology, 13, 642-650.

PMID: 18713996: Yoshizawa, T., Hammaker, D., Sweeney, S.E., Boyle, D.L. & Firestein, G.S. (2008)

Synoviocyte Innate Immune Responses: I. Differential Regulation of Interferon Responses and the

JNK Pathway by MAPK Kinases. The Journal of Immunology, 181, 3252-3258.

PMID: 18378223: Zhou, H.Y., Shin, E.M., Guo, L.Y., Youn, U.J., Bae, K., Kang, S.S., Zou, L.B. & Kim,

Y.S. (2008) Anti-Inflammatory Activity of 4-Methoxyhonokiol is a Function of the Inhibition of iNOS

and COX-2 Expression in RAW 264.7 Macrophages via NF-κB, JNK and p38 MAPK Inactivation.

European Journal of Pharmacology, 586, 340-349.

Rapamycin & LPS Mock Dataset

393

PMID: 18758466: Cao, W., Manicassamy, S., Tang, H., Kasturi, S.P., Pirani, A., Murthy, N. &

Pulendran, B. (2008) Toll-Like Receptor-Mediated Induction of Type I Interferon in Plasmacytoid

Dendritic Cells Requires the Rapamycin-Sensitive PI(3)K-mTOR-p70S6K Pathway. Nature


PMID: 19285553: Dos Santos Mendes, S., Candi, A., Vansteenbrugge, M., Pignon, M.R., Bult, H.,

Boudjeltia, K.Z., Munaut, C. & Raes, M. (2009) Microarray Analyses of the Effects of NF-κB or PI3K

Pathway Inhibitors on the LPS-Induced Gene Expression Profile in RAW264.7 Cells: Synergistic Effects

of Rapamycin on LPS-Induced MMP9-Overexpression. Cellular Signalling, 21, 1109-1122.

PMID: 20805416: Haidinger, M., Poglitsch, M., Geyeregger, R., Kasturi, S., Zeyda, M., Zlabinger,

G.J., Pulendran, B., Horl, W.H., Saemann, M.D. & Weichhart, T. (2010) A Versatile Role of

Mammalian Target of Rapamycin in Human Dendritic Cell Function and Differentiation. The Journal




Weichhart, T. (2013) p38α Senses Environmental Stress to Control Innate Immune Responses via


PMID: 21200439: Lee, P.S., Wilhelmson, A.S., Hubner, A.P., Reynolds, S.B., Gallacchi, D.A., chiou,

T.T. & Kwiatkowski, D.J. (2010) mTORC1-S6K Activation by Endotoxin Contributes to Cytokine Up-

Regulation and Early Lethality in Animals. PLoS One, 5, doi: 10.1371/journal.pone.0014399

PMID: 22675026: López-Pelaéz, M., Fumagalli, S., Sanz, C., Herrero, C., Guerra, S., Fernandez, M. &

Alemany, S. (2012) Cot/tpl2-MKK1/2-Erk1/2 Controls mTORC1-Mediated mRNA Translation in Toll-

Like Receptor-Activated Macrophages. Molecular Biology of the Cell, 23, 2982-2992.

PMID: 20335217: Turnquist, H.R., Cardinal, J., Macedo, C., Rosborough, B.R., Sumpter, T.L., Geller,

D.A., Metes, D. & Thomson, A.W. (2010) mTOR and GSK-3 Shape the CD4+ T-Cell Stimulatory and

Differentiation Capacity of Myeloid DCs after Exposure to LPS. Blood, 115, 4758-4769.

PMID: 18492954: Ohtani, M., Nagai, S., Kondo, S., Mizuno, S., Nakamura, K., Tanabe, M.,

Takeuchi, T., Matsuda, S. & Koyasu, S. (2008) Mammalian Target of Rapamycin and Glycogen

Synthase Kinase 3 Differentially Regulate Lipopolysaccharide-Induced Interleukin-12 Production in

Dendritic Cells. Blood, 112, 635-643.

PMID: 22412198: Pan, H., O'Brien, T.F., Zhang, P. & Zhong, X.P. (2012) The Role of Tuberous

Sclerosis Complex 1 in Regulating Innate Immunity. The Journal of Immunology, 188, 3658-3666.

PMID: 24126911: Rastogi, R., Jiang, Z., Ahmad, N., Rosati, R., Liu, Y., Beuret, L., Monks, R.,

Charron, J., Birnbaum, M.J. & Samavati, L. (2013) Rapamycin Induces Mitogen-Activated Protein

(MAP) Kinase Phosphatase-1 (MKP-1) Expression through Activation of Protein Kinase B and

Mitogen-Activated Protein Kinase Kinase Pathways. The Journal of Biological Chemistry, 288, 33966-

33977.

394

PMID: 9862729: Salh, B., Wagey, R., Marotta, A., Tao, J.S. & Pelech, S. (1998) Activation of

Phosphatidylinositol 3-Kinase, Protein Kinase B, and p70 S6 Kinases in Lipopolysaccharide-Stimulated

Raw 264.7 Cells: Differential Effects of Rapamycin, Ly294002, and Wortmannin on Nitric Oxide

Production. The Journal of Immunology, 161, 6947-6954.

PMID: 20828737: Schaeffer, V., Arbabi, S., Garcia, I.A., Knoll, M.L., Cuschieri, J., Bulger, E.M. &

Maier, R.V. (2011) Role of the mTOR Pathway in LPS-Activated Monocytes: Influence of Hypertonic

Saline. The Journal of Surgical Research, 171, 769-776.

PMID: 18924132: Schmitz, F., Heit, A., Dreher, S., Eisenächer, K., Mages, J., Haas, T., Krug, A.,

Janssen, K.P., Kirschning, C.J. & Wagner, H. (2008) Mammalian Target of Rapamycin (mTOR)

Orchestrates the Defense Program of Innate Immune Cells. European Journal of Immunology, 38,

2981-2992.

PMID: 21422248: Wang, H., Brown, J., Gu, Z., Garcia, C.A., Liang, R., Alard, P., Beurel, E., Jope, R.S.,

Greenway, T. & Martin, M. (2011) Convergence of the Mammalian Target of Rapamycin Complex 1-

and Glycogen Synthase Kinase 3-β-Signaling Pathways Regulates the Innate Inflammatory Response.

The Journal of Immunology, 186, 5217-5226.

PMID: 21368289: Weichhart, T., Haidinger, M., Katholnig, K., Kopecky, C., Poglitsch, M., Lassnig,

C., Rosner, M., Zlabinger, G.J., Hengstschlager, M., Muller, M., Horl, W.H. & Saemann, M.D. (2011)

Inhibition of mTOR Blocks the Anti-Inflammatory Effects of Glucocorticoids in Myeloid Immune

Cells. Blood, 117, 4273-4283.

PMID: 10733102: Weinstein, S.L., Finn, A.J., Dave, S.H., Meng, F., Lowell, C.A., Sanghera, J.S. &

DeFranco, A.L. (2000) Phosphatidylinositol 3-Kinase and mTOR Mediate Lipopolysaccharide-

Stimulated Nitric Oxide Production in Macrophages via Interferon-β. Journal of Leukocyte Biology,

67, 405-414.

5Z-7-Oxozeaenol & LPS Mock Dataset

PMID: 21138416: Clark, K., Peggie, M., Plater, L., Sorcek, R.J., Young, E.R., Madwed, J.B., Hough, J.,

McIver, E.G. & Cohen, P. (2011) Novel Cross-Talk within the IKK Family Controls Innate Immunity.

The Biochemical Journal, 434, 93-104.

PMID: 21949249: Clark, K., Takeuchi, O., Akira, S. & Cohen, P. (2011) The TRAF-Associated Protein

TANK Facilitates Cross-Talk Within the IκB Kinase Family During Toll-Like Receptor Signaling. The


PMID: 20200282: Ear, T., Fortin, C.F., Simard, F.A. & McDonald, P.P. (2010) Constitutive Association

of TGF-β-Activated Kinase 1 with the IκB Kinase Complex in the Nucleus and Cytoplasm of Human

Neutrophils and its Impact on Downstream Processes. The Journal of Immunology, 184, 3897-3906.

395

PMID: 21469098: Fortin, C.F., Cloutier, A., Ear, T., Sylvain-Prevost, S., Mayer, T.Z., Bouchelaghem,

R. & McDonald, P.P. (2011) A Class IA PI3K Controls Inflammatory Cytokine Production in Human

Neutrophils. European Journal of Immunology, 41, 1709-1719.

PMID: 23401599: Fortin, C.F., Mayer, T.Z., Cloutier, A. & McDonald, P.P. (2013) Translational

Control of Human Neutrophil Responses by MNK1. Journal of Leukocyte Biology, doi:

10.1189/jlb.0113012.

PMID: 10606001: Rawlins, P., Mander, T., Sadeghi, R., Hill, S., Gammon, G., Foxwell, B., Wrigley, S.

& Moore, M. (1999) Inhibition of Endotoxin-Induced TNF-α Production in Macrophages by 5Z-7-Oxo-

zeaenol and Other Fungal Resorcylic Acid Lactones. International Journal of Immunopharmacology,

21, 799-814.

PMID: 17348859: Windheim, M., Lang, C., Peggie, M., Plater, L.A. & Cohen, P. (2007) Molecular

Mechanisms Involved in the Regulation of Cytokine Production by Muramyl Dipeptide. The


396

Supplementary Information 2 –

Dataset Against Which the Binary

Model was Tested (Final Benchmark

Dataset)

397

BIX02189

ppMEK1 No Effect (PMID: 18834865)


ppERK1 Partial Inhibition with Cadmium Chloride (PMID: 22521884); No Effect (PMID: 18834865)

ppERK2 Partial Inhibition with Cadmium Chloride (PMID: 22521884); No Effect (PMID: 18834865)

RSK-Pi Decrease for RSK3 (PMID: 22997248) & RSK2 & RSK4 (PMID: 18834865)

rpS6-Pi Reduction Implied But Not Directly Shown

pMNK1 Partial Inhibition (PMID: 18834865)

pMNK2 No Effect (PMID: 18834865)

eIF4E-Pi Not Tested

pJNK1 No Effect (PMID: 18834865)

cJun-Pi Not Tested But Potential Indirect Inhibition (PMID: 20075332; PMID: 12574153)

ppp38a Partial Inhibition (PMID: 18834865)

pERK5 Inhibition Demonstrated (PMID: 18834865; PMID: 22254155; PMID: 24491810)

PI3K-Act Not Tested

PIP3 Not Tested

PDK1-Act No Effect (PMID: 18834865)

pPKCz Not Tested

ppAKT Partial Inhibition (PMID: 18834865)

ppS6K Partial Increase (PMID: 18834865)

eIF4B-Pi Not Tested

p4E-BP1 Not Tested

mTORC1-Act Not Tested

pTSC2 Not Tested

398

CGP57380

ppMEK1 Not Tested

ppMEK2 Not Tested

ppERK1 No Effect (PMID: 11463832) - Disputed (Partial Increase) (PMID: 18694961)

ppERK2 No Effect (PMID: 11463832) - Disputed (Partial Increase) (PMID: 18694961)

RSK-Pi Possible Inhibition (PMID: 18694961) - Maybe Specific to CML

rpS6-Pi Possible Inhibition (PMID: 18694961) - Disputed (PMID: 20664001)

pMNK1 Inhibition Demonstrated (PMID: 23269249)


eIF4E-Pi Inhibition Demonstrated (PMID: 23269249; PMID: 18081851)


cJun-Pi Not Tested

ppp38a No Effect (PMID: 11463832)

pERK5 Not Tested

PI3K-Act Not Tested

PIP3 Not Tested

PDK1-Act Not Tested

pPKCz Not Tested

ppAKT Not Tested

ppS6K No Effect (PMID: 20664001)

eIF4B-Pi Not Tested

p4E-BP1 Partial Inhibition (PMID: 18694961)

mTORC1-Act No Effect (PMID: 18694961)

pTSC2 Not Tested

399

GSK2334470

ppERK1 No Effect (PMID: 21087210)




pRSK1 Partial Inhibition (PMID: 21087210)

rpS6-Pi Inhibition Demonstrated (PMID: 21087210)



eIF4E-Pi Not Tested

pJNK1 No Effect (JNK1 & JNK3); Partial Activation (JNK2) (PMID: 21087210)

cJun-Pi Not Tested


pERK5 Not Tested

PI3K-Act No Effect (PMID: 21087210)

PIP3 No Effect (PMID: 21087210)

pPDK1 Inhibition Demonstrated (PMID: 21087210)

pPKCz Partial Inhibition (PMID: 21087210)

ppAKT No Effect (PMID: 21087210)

ppS6K1 No Effect (PMID: 21087210)

eIF4B-Pi Not Tested

p4E-BP1 Not Tested


pTSC2 Not Tested

400

LY294002

ppMEK1 Partial Inhibition (PMID: 17850214)


ppERK1 Partial Inhibition (PMID: 17850214; PMID: 11157888); Disputed (No Effect) (PMID: 10480882; PMID: 12242656)

ppERK2 Partial Inhibition (PMID: 17850214; PMID: 11157888); Disputed (No Effect) (PMID: 10480882; PMID: 12242656)

pRSK1 No Effect (PMID: 17850214)

rpS6-Pi Inhibition Demonstrated (PMID: 8895571; PMID: 11713299)


pMNK2 Not Tested

eIF4E-Pi Blocks Rapamycin Induced Increase (PMID: 16103051)

pJNK1 No Effect (JNK1 & JNK2) Partial Inhibition (JNK3) (PMID: 17850214)

cJun-Pi Not Tested

ppp38a No Effect (PMID: 11040049; PMID: 11367542; PMID: 17850214)

pERK5 No Effect Implied (PMID: 14670836)

PI3K-Act Inhibition Demonstrated (PMID: 8106507)

PIP3 Inhibition Implied (PMID: 8106507)

PDK1-Act Inhibition Demonstrated (PMID: 17850214; PMID: 9748166)

pPKCz Inhibition Demonstrated (PMID: 9748166) Disputed (Activation) (PMID: 17850214)

ppAKT Inhibition Demonstrated (PMID: 11801244; PMID: 16551362; PMID: 11755539)

ppS6K Inhibition Demonstrated (PMID: 15677443; PMID: 10480882; PMID: 11755539; PMID: 11145615; PMID: 12242656; PMID: 9287347)

eIF4B-Pi Not Tested

p4E-BP1 Inhibition Demonstrated (PMID: 11231341)

mTORC1-Act Inhibition Demonstrated (PMID: 8895571; PMID: 19346248)

pTSC2 Not Tested

401

PD184352

ppMEK1 Inhibition Demonstrated (PMID: 10998351; PMID: 17850214; PMID: 12534346)

ppMEK2 Inhibition Demonstrated (PMID: 10998351; PMID: 17850214; PMID: 12534346)

ppERK1 Partial Inhibition (PMID: 12069688; PMID: 17850214)

ppERK2 Partial Inhibition (PMID: 12069688; PMID: 17850214)

pRSK1 Not Tested

rpS6-Pi Partial Inhibition (PMID: 19376132)

pMNK1 Not Tested But Implied (PMID: 9155017)

pMNK2 Not Tested But Implied (PMID: 9155017)

eIF4E-Pi Not Tested But Implied (PMID: 9155017)

pJNK1 Increase Demonstrated (PMID: 18347148)

cJun-Pi Not Tested

ppp38a No Effect (PMID: 22076433) But Disputed (PMID: 11431348) - Activation Reported (PMID: 18347148)

pERK5 No Effect (PMID: 12069688; PMID: 18358237)

PI3K-Act Not Tested

PIP3 Not Tested

PDK1-Act Partial Inhibition (PMID: 10998351)

pPKCz Not Tested

ppAKT Partial Activation (PMID: 19147570; PMID: 22668349)

ppS6K Partial Inhibition (PMID: 11431469)

eIF4B-Pi Not Tested


mTORC1-Act Inhibition Demonstrated (PMID: 21071439)

pTSC2 Partial Inhibition (PMID: 15292274)

402

PD98059

ppMEK1 Partial Inhibition (PMID: 9395471; PMID: 9315666; PMID: 11043572; PMID: 10998351)

ppMEK2 Partial Inhibition (PMID: 9395471; PMID: 9315666; PMID: 11043572; PMID: 10998351)

ppERK1 Partial Inhibition (PMID: 7499206; PMID: 12704121; PMID: 16342118; PMID: 21655705; PMID: 10998351; PMID: 23615277)

ppERK2 Partial Inhibition (PMID: 7499206; PMID: 12704121; PMID: 16342118; PMID: 21655705; PMID: 10998351; PMID: 23615277)

pRSK1 Inhibition Demonstrated (PMID: 21619683)

rpS6-Pi Inhibition Demonstrated (PMID: 21233202) But Disputed (PMID: 19176818)



eIF4E-Pi Inhibition Demonstrated (PMID: 11154262; PMID: 9211946)

pJNK1 Activation Demonstrated (PMID: 11353829; PMID: 11408609)

cJun-Pi Partial Inhibition (PMID:11435459)

ppp38a No Effect (PMID: 12663671; PMID: 23615277)

pERK5 Inhibition Demonstrated (PMID: 10473620; PMID: 18358237)

PI3K-Act Not Tested

PIP3 Not Tested

PDK1-Act No Effect (PMID: 10998351)

pPKCz Not Tested

ppAKT No Effect (PMID: 12481421; PMID: 23615277)

ppS6K S6K2 Inhibited (PMID: 11431469) Degree Disputed (PMID: 12489846) - No Effect (PMID: 10998351)

eIF4B-Pi Not Tested


mTORC1-Act No Effect (PMID: 20439490)

pTSC2 Not Tested

403

Rapamycin

ppERK1 Inhibition at Higher Doses (Activation at Lower Doses (PMID: 19151764)) (PMID: 20512842 ); No Effect (PMID: 23437362; PMID: 16160607; PMID: 11567647; PMID: 15767555; PMID: 12820963; PMID: 22343118)

ppERK2 Inhibition at Higher Doses (Activation at Lower Doses (PMID: 19151764)) (PMID: 20512842 ); No Effect (PMID: 23437362; PMID: 16160607; PMID: 11567647; PMID: 15767555; PMID: 12820963; PMID: 22343118)



pRSK1 Activation at Low Doses (PMID: 19151764)

rpS6-Pi Inhibition Demonstrated (PMID: 23437362; PMID: 21075311; PMID: 16763566)

pMNK1 No Effect Indicated (PMID: 21406405)

pMNK2 Increase Demonstrated (PMID: 23831578)

eIF4E-Pi Increase Demonstrated (PMID: 23831578; PMID: 16103051)

pJNK1 Activation Demonstrated (PMID: 21324487; PMID: 15767555); Inhibition Demonstrated (PMID: 9016789; PMID: 16160607)

cJun-Pi Activation Demonstrated (PMID: 12820963)

ppp38a Inhibition Demonstrated (PMID: 16160607; PMID: 11567647); No Effect (PMID: 12820963; PMID: 9933636)

pERK5 Not Tested

PI3K-Act No Effect on the Expression (PMID: 12766174)

PIP3 Not Tested

404

Rapamycin

pPDK1 No Effect Indicated (PMID: 16377759) - Any Affect May be Substrate-Dependent (PMID: 21075311)

pPKCz Not Tested Directly (PMID: 10339425)

ppAKT Inhibition at Higher Doses (PMID: 20512842 ); Activation Possible at Lower Doses (PMID: 16103051; PMID: 23437362; PMID: 17698586; PMID: 21075311)

ppS6K1 Inhibition Demonstrated (PMID: 23437362; PMID: 21114628; PMID: 11567647; PMID: 15030312; PMID: 16763566; PMID: 15833867)

eIF4B-Pi Increase Demonstrated at Lower Dose (PMID: 21075852); Prolonged Exposure - Inhibition Demonstrated (PMID: 16763566)

p4E-BP1 Partial Inhibition Demonstrated (PMID: 19402821; PMID: 17425689; PMID: 11799119)

mTORC1-Act

Inhibition Demonstrated (PMID: 15899889); No Effect Possible (PMID: 16103051)

pTSC2 Not Tested

405

SB203580



ppERK1 Activation Demonstrated (PMID: 10960075; PMID: 11297530; PMID: 16111636) Disputed (PMID: 17850214; PMID: 10998351)

ppERK2 Activation Demonstrated (PMID: 10960075; PMID: 11297530; PMID: 16111636) Disputed (PMID: 17850214; PMID: 10998351)

pRSK1 Partial Inhibition Demonstrated (PMID: 17906627)

rpS6-Pi Partial Inhibition Demonstrated (PMID: 23315073)

pMNK1 Not Tested


eIF4E-Pi Partial Inhibition Demonstrated (PMID: 9545260; PMID: 12781867; PMID: 15075293)

pJNK1 Activation Demonstrated (PMID: 10960075) But Disputed (PMID: 9598985)

cJun-Pi Activation Demonstrated (PMID: 10960075) But Disputed (PMID: 9598985)

ppp38a Inhibition Demonstrated (PMID: 9753474; PMID: 18502741; PMID: 21903942)

pERK5 Not Tested


PIP3 No Effect (PMID: 11367542)

PDK1-Act Inhibition Demonstrated (PMID: 10702313)

pPKCz Activation Demonstrated (PMID: 15665819)

ppAKT Inhibition Demonstrated (PMID: 10702313; PMID: 11042204)

ppS6K Partial Inhibition Demonstrated (PMID: 10455142; PMID: 9299480)

eIF4B-Pi No Effect (PMID: 16763566)

p4E-BP1 Partial Inhibition Demonstrated (PMID: 23315073)


pTSC2 Increase Demonstrated (PMID: 23315073)

406

SP600125



ppERK1 Partial Inhibition - Disputed (PMID: 15454338; PMID: 20868664; PMID: 12603281; PMID: 12697808)

ppERK2 Partial Inhibition - Disputed (PMID: 15454338; PMID: 20868664; PMID: 12603281; PMID: 12697808)

pRSK1 Partial Increase (PMID: 16449644) Disputed (PMID: 12603281)




eIF4E-Pi Partial Inhibition (PMID: 20359850; PMID: 20359507)

pJNK1 Inhibition Demonstrated (PMID: 17850214)

cJun-Pi Inhibition Demonstrated (PMID: 20868664; PMID: 16449644; PMID: 16176902; PMID: 19305497)

ppp38a No Effect (PMID: 16449644; PMID: 12603281; PMID: 17850214; PMID: 18359001); Very Early Increase Demonstrated (PMID: 12878189; PMID: 18359001)

pERK5 Partial Inhibition (PMID: 20643107)

PI3K-Act Not Tested BUT Inhibition of δ Isoform of p110 Subunit - Subunit Specific (PMID: 19106158; PMID: 21558088)

PIP3 Not Tested


pPKCz No Effect (PMID: 17850214)


ppS6K Partial Inhibition (PMID: 17850214; PMID: 16099428)

eIF4B-Pi Partial Inhibition (PMID: 17850214; PMID: 21263197)

p4E-BP1 Partial Inhibition (PMID: 17850214 & PMID: 21263197)

mTORC1-Act Inhibition Demonstrated (PMID: 21320501; PMID: 22493283)

pTSC2 Not Tested

407

XMD8-92

ppMEK1 Not Tested

ppMEK2 Not Tested



pRSK1 Partial Inhibition (PMID: 20832753)

rpS6-Pi Not Tested

pMNK1 Not Tested

pMNK2 Not Tested

eIF4E-Pi Not Tested

pJNK1 Not Tested

cJun-Pi Not Tested

ppp38a Not Tested

pERK5 Inhibition Demonstrated (PMID: 20832753; PMID: 23428871)

PI3K-Act Not Tested

PIP3 Not Tested

PDK1-Act Not Tested

pPKCz Not Tested

ppAKT Not Tested

ppS6K Not Tested

eIF4B-Pi Not Tested

p4E-BP1 Not Tested


pTSC2 Not Tested

408

Note the details of the effects of Ku0063794 are included as Ku0063794 was used in the initial model

Ku0063794





pRSK1 No Effect (PMID: 19402821)


pMNK1 Partial Activation (PMID: 19402821)

MNK1/2-Pi Partial Inhibition (PMID: 19402821)

eIF4E-Pi Not Tested

pJNK No Effect (JNK2α2) Partial Inhibition (JNK1α1) (PMID: 19402821)

cJun-Pi Not Tested

ppp38a No Effect (α-isoform) (PMID: 19402821)

pERK5 Not Tested


PIP3 No Effect - Implied (PMID: 19402821)


pPKCz Partial Inhibition (PMID: 19402821)


ppS6K No Effect (PMID: 19402821)

eIF4B-Pi Not Tested

P4E-BP1 Inhibition Demonstrated (PMID: 19402821)


pTSC2 Not Tested

409

Torin-1, whilst not used in this project, is included as the list of effects is a tool that can be used for other works.

Torin-1

ppERK1 Activation Demonstrated (PMID: 21508335; PMID: 21278786) - No Effect (PMID: 22343943; PMID: 21592956)

ppERK2 Activation Demonstrated (PMID: 21508335; PMID: 21278786) - No Effect (PMID: 22343943; PMID: 21592956)

ppMEK1 Not Tested

ppMEK2 Not Tested

pRSK1 Inhibition Demonstrated (At Higher Concentrations) (PMID: 22223645)

rpS6-Pi Inhibition Demonstrated (PMID: 24424027; PMID: 22552098; PMID: 24008870; PMID: 21908613; PMID: 22065737; PMID: 23415771; PMID: 21576368; PMID: 23319332)



eIF4E-Pi Inhibition Demonstrated (PMID: 20679199)


cJun-Pi Activation Demonstrated (PMID: 21508335)


pERK5 Not Tested

PI3K-Act No Effect (PMID: 22015718; PMID: 23436801) - But Activation Maybe Possible (PMID: 22015718)

PIP3 Not Tested

410

Torin-1

pPDK1 Not Tested

pPKCz Inhibition Demonstrated (PMID: 24375676) - No Effect (PMID: 22223645)

ppAKT Inhibition Demonstrated (PMID: 21508335; PMID: 21659604; PMID: 22223645; PMID: 22733130; PMID: 20089925; PMID: 23142081; PMID: 24311379; PMID: 20735411; PMID: 22006022; PMID: 24043828)

ppS6K1 Inhibition Demonstrated (PMID: 22552098; PMID: 21508335; PMID: 21659604; PMID: 22343943; PMID: 22223645; PMID: 21858812; PMID: 23707523; PMID: 23524850; PMID: 23888043)

eIF4B-Pi Not Tested

p4E-BP1 Inhibition Demonstrated (PMID: 21508335; PMID: 21659604; PMID: 23547259; PMID: 23077579; PMID: 21858812; PMID: 23707523; PMID: 21841310; PMID: 21307192)

mTORC1-Act

Inhibition Demonstrated (PMID: 19150980; PMID: 23934889; PMID: 23401004; PMID: 22053105; PMID: 21576368; PMID: 23963679)

pTSC2 Not Tested

411


Dataset Against Which the Model

Integrated with Metabolism was

Tested (Metabolic Dataset)

Page 412

BIX02189

c-Myc Not Tested

HIF-1α Not Tested with the Inhibitor - ERK5 Maybe Important though (PMID: 16735500)

Scd1 Not Tested

PFK Not Tested

CGP57380

c-Myc No Effect (PMID: 22370634)

HIF-1α No Effect (PMID: 25115400)

Scd1 Not Tested

PFK Not Tested

GSK2334470

c-Myc Not Tested

HIF-1α Not Tested

Scd1 Not Tested

PFK Not Tested

LY294002

c-Myc Inhibition Demonstrated (PMID: 22546857; PMID: 15467756; PMID: 25636967)

HIF-1α Inhibition Demonstrated (PMID: 16849522; PMID: 12032158; PMID: 10749120; PMID: 14982927; PMID: 12070140; PMID: 11457733; PMID: 20194722; PMID: 15714461; PMID: 22787058)

Scd1 Inhibition Demonstrated (PMID: 20876744)

PFK Inhibition Demonstrated (PMID: 16051672) - No Effect on Level (PMID: 20958264) - Activity May be Inhibited (PMID: 17302559; PMID: 20958264)

Page 413

PD184352

c-Myc No Effect on mRNA Level (PMID: 15286700) - Inhibition Demonstrated (Inhibitor Not Used Alone) (PMID: 18497825)

HIF-1a Inhibition Demonstrated (PMID: 17213817) - No Effect (PMID: 18199551)

SCD1 Not Tested

PFK Not Tested

Rapamycin

c-Myc Inhibition Demonstrated (PMID: 14576155; PMID: 18621930; PMID: 23612979; PMID: 15634685; PMID: 10993886; PMID: 25636967)

HIF-1α Inhibition Demonstrated (PMID: 20670887; PMID: 21567203; PMID: 18781596; PMID: 17968710; PMID: 17502379; PMID: 12032158; PMID: 12242281; PMID: 10749120; PMID: 22900063) - No Effect (PMID: 16849522; PMID: 12149254; PMID: 20194722; PMID: 22787058)

Scd1 Inhibition Demonstrated (PMID: 20670887; PMID: 20876744)

PFK Inhibition Demonstrated (PMID: 20670887)

SB203580

c-Myc Inhibition Demonstrated (PMID: 20212154; PMID: 16365184; PMID: 10648601; PMID: 12080469; PMID: 11563982)

HIF-1a Inhibition Demonstrated (PMID: 20194722; PMID: 11978547; PMID: 12482858) - No Effect (PMID: 15714461)

SCD1 No Effect (with 5-Fluorouracil Pretreatment) (PMID: 24135379)

PFK Not Tested

SP600125

c-Myc Inhibition Demonstrated (PMID: 15890690; PMID: 20335523; PMID: 21525782; PMID: 25486532; PMID: 19418558; PMID: 24104553; PMID: 19996270)

HIF-1a No Effect (PMID: 20194722)

SCD1 Increase Demonstrated (in Trans-10, cis-12 conjugated linoleic acid Cotreatment) (PMID: 21744278)

PFK Not Tested

Page 414

XMD8-92

c-Myc Inhibition Demonstrated (PMID: 24880079)

HIF-1α Not Tested

Scd1 Not Tested

PFK Not Tested

Page 415


Ingredients List for Solutions

Produced In-House

Page 416

Running Buffer

14.4gm Glycine (Thermo Fisher Scientific Inc., Rockford, IL (USA))

3.0gm Tris-Base (Thermo Fisher Scientific Inc., Rockford, IL (USA))

5ml 20% SDS (Thermo Fisher Scientific Inc., Rockford, IL (USA))

Made up to 1l with MIlli-Q® Ultrapure (EMD Millipore Corp., Merck KGaA, Darmstadt (DE))

Transfer Buffer

14.4gm Glycine (Thermo Fisher Scientific Inc., Rockford, IL (USA))

3.0gm Tris-Base (Thermo Fisher Scientific Inc., Rockford, IL (USA))

200ml Methanol (Sigma-Aldrich Co. LLC, St. Louis, MO (USA))


1x PBS-Tween Solution

10 Dulbecco A/PBS Tablets (Oxoid Ltd., Thermo Fisher Scientific Inc., Rockford, IL (USA))

1ml Tween-20 (Sigma-Aldrich Co. LLC, St. Louis, MO (USA))


Documents

Computer Simulation of a Large-Scale Signalling …epubs.surrey.ac.uk/841058/1/Thesis Copy 10th March.pdfComputer Simulation of a Large-Scale Signalling Network Regulating Translation