CHARACTERIZATION OF SHADOO AND DPPX: TWO … · exist in prion biology including the cellular role of PrPC, the potential involvement of auxiliary factors in prion replication, and

CHARACTERIZATION OF SHADOO AND DPPX:

TWO PROTEINS OF POTENTIAL RELEVANCE

TO PRION BIOLOGY

by

Joel Christopher Watts

A thesis submitted in conformity with the requirements

for the degree of Doctor of Philosophy

Graduate Department of Laboratory Medicine and Pathobiology

University of Toronto

© Copyright by Joel Watts 2008

Joel Christopher Watts Characterization of Shadoo and DPPX: Two Proteins of Potential Relevance to Prion Biology Doctor of Philosophy Department of Laboratory Medicine and Pathobiology University of Toronto 2008

ABSTRACT

Prion diseases are fatal neurodegenerative disorders of humans and animals. The prion

hypothesis states that PrPSc, a misfolded conformational isoform of the cellular prion

protein (PrPC), is the sole component of the infectious particle. Many open questions

exist in prion biology including the cellular role of PrPC, the potential involvement of

auxiliary factors in prion replication, and the mechanism of PrPSc-induced toxicity in

prion disease. The identification of novel prion-like proteins and authentic in vivo prion

protein-interacting proteins would certainly assist the process of demystifying these

unsolved mysteries. Accordingly, two newly-identified proteins with potential relevance

to prion protein biology, Shadoo and DPPX, were selected for biochemical and functional

characterization. Shadoo, a hypothetical prion-like protein, is revealed as being a

glycoprotein which possesses many overlapping properties with PrPC including neuronal

expression, C1-like endoproteolytic processing, and the ability to protect against

apoptotic stimuli in cerebellar neurons. Shadoo loosely resembles the disordered N-

terminal domain of PrPC and consistent with this notion, Shadoo appears to lack a well-

defined structure. Remarkably, Shadoo levels in the brains of mice with clinical prion

disease are significantly decreased suggesting that Shadoo may be inherently linked to

prion replication or prion disease pathogenesis. These experiments define Shadoo as the

third member of the prion protein family and, because of its functional similarities to

ii

PrPC, Shadoo may be a useful tool for deciphering the in vivo function of PrPC. DPPX, a

neuronal type II transmembrane protein, is demonstrated to be the first protein capable of

interacting with all three members of the prion protein family (PrPC, Doppel, and

Shadoo) in vivo. Complex formation between prion proteins and DPPX appears to be

mediated by multiple binding sites. When coupled with high levels of DPPX expression

in cerebellar granular neurons, DPPX is a strong candidate for mediating phenotypic

interactions between prion proteins in cerebellar cells. Thus, Shadoo and DPPX comprise

two new entry points for studying prion proteins. Further investigation of the roles of

Shadoo and DPPX in both the cell biology of prion proteins and prion disease may yield

important clues to these enigmatic topics.

iii

ACKNOWLEDGEMENTS

I would like to thank Dr. David Westaway for giving me the opportunity to

conduct prion research in his laboratory and agreeing to be my PhD thesis supervisor.

Throughout the past four and a half years he has provided me with an enormous amount

of guidance and assistance and I can’t thank him enough for everything he’s done for me.

By allowing me a substantial amount of independence in my research David has prepared

me well for a career in science and I certainly attribute any future success to him.

I would also like to thank Dr. Gerold Schmitt-Ulms for agreeing to be my PhD

co-supervisor. Gerold has given me an incredible amount of technical advice and career

assistance and has been especially helpful following David’s move to Alberta. I also

would like to thank my other two PhD thesis advisory committee members, Dr. JoAnne

McLaurin and Dr. Rod Bremner. I appreciate the guidance you have provided me

throughout the duration of my PhD studies.

I feel very privileged to have been able to work in a laboratory with such

wonderful people. I’d like to thank all the past and present members of the Westaway lab

for providing such a friendly working environment, for all the help and technical

assistance, and for being such great friends. I would like to extend a huge thank you to

Bettina Drisaldi, Erwan Paitel, Peter Mastrangelo, Patrick Horne, Bob Strome, Jing

Yang, Vivian Ng, and Joannis Sekoulidis for everything they’ve done for me in the past

four years. I’d also like say a special thank you to Michael Quejada, an extremely

talented fourth year project student who helped with a lot of the DPPX studies.

I certainly could not have completed my thesis without the help and guidance of

various members of the CRND. In particular, I’d like to thank Dr. JoAnne McLaurin and

Dr. Janice Robertson for their willingness to help me with whatever problem I happened

to be experiencing. My project also would not have been possible without the assistance

of Dr. Paul Fraser and Dr. Howard Mount and I thank them for their large contributions. I

would also like to acknowledge the vast amount of help that has been given to me by my

colleagues at the CRND. In particular, I would like to express my gratitude to Monika

Duthie, Rosemary Ahrens, Ling Wu, Dwayne Ashman, Teresa Sanelli, Kelly Markham,

Yu Bai, and Jenny Griffin.

iv

I also must thank my fellow PhD graduate students and friends at the CRND:

Kevin DaSilva, David Gelinas, Veronique Dorval, Daniela Fenili, and Jessie McLean.

Your friendship has meant a lot to me and has helped me get through days when you

want to throw your experiments out the window.

I would also like to acknowledge the support of the Natural Sciences and

Engineering Research Council of Canada for helping to fund my PhD research and the

Canadian Institutes of Health Research for funding research in the Westaway laboratory.

I would also like to thank the people at PrioNet Canada for their assistance.

Finally, I would like to say thank you to my family and in particular my wife

Kathy for putting up with the long hours and weekends in the lab and all the boring talk

about biology. I certainly would not have made it this far without their support.

v

TABLE OF CONTENTS ABSTRACT ........................................................................................................................ ii

ACKNOWLEDGEMENTS ............................................................................................... iv

TABLE OF CONTENTS ................................................................................................... vi

LIST OF TABLES ........................................................................................................... viii

LIST OF FIGURES ......................................................................................................... viii

LIST OF ABBREVIATIONS ........................................................................................... xii

Chapter 1 ............................................................................................................................. 1

Introduction ......................................................................................................................... 1

1.1 An Overview of Basic Concepts in Prion Biology ................................................... 2

1.2 Prion Genes and Proteins ........................................................................................ 22

1.3 Cellular Functions of Prion Proteins ....................................................................... 33

1.4 Prion Protein Ligands ............................................................................................. 38

1.5 Functional Interactions Between Members of the Prion Protein Family ................ 44

Chapter 2 ........................................................................................................................... 59

Rationale, Hypotheses, and Objectives ............................................................................. 59

2.1. Rationale ................................................................................................................ 60

2.2. Hypotheses ............................................................................................................. 61

2.3. Objectives .............................................................................................................. 62

Chapter 3 ........................................................................................................................... 63

Characterization of Shadoo, a Putative Prion-Like Protein .............................................. 63

3.1 Abstract ................................................................................................................... 64

3.2 Introduction ............................................................................................................. 65

3.3 Materials and Methods ............................................................................................ 67

3.4 Results ..................................................................................................................... 75

3.5 Discussion ............................................................................................................. 107

vi

Chapter 4 ......................................................................................................................... 117

Characterization of Interactions Between Members of the Prion Protein Family and the

Type II Transmembrane Protein DPPX .......................................................................... 117

4.1 Abstract ................................................................................................................. 118

4.2 Introduction ........................................................................................................... 119

4.3 Materials and Methods .......................................................................................... 123

4.4 Results ................................................................................................................... 131

4.5 Discussion ............................................................................................................. 178

Chapter 5 ......................................................................................................................... 190

Conclusions and Future Directions ................................................................................. 190

References ....................................................................................................................... 204

vii

LIST OF TABLES

CHAPTER 1 Table 1.1. Commonly-used methods for detecting TSEs and PrPSc ..................................19

Table 1.2. Proteins or molecules which have been reported to interact with PrPC or

PrPSc ...................................................................................................................................39

Table 1.3. Characteristics of strains of Prnp0/0 Mice .........................................................46

Table 1.4. Summary of transgenic mice expressing Doppel or ΔPrP ................................48

CHAPTER 4

Table 4.1. N-terminal splicing isoforms of murine DPPX and DPP10 ...........................137

LIST OF FIGURES

CHAPTER 1

Figure 1.1. Prion disease nomenclature and modes of acquisition ......................................4

Figure 1.2. The neuropathological hallmarks of prion disease ..........................................11

Figure 1.3. Proposed models of prion replication ..............................................................14

Figure 1.4. Schematic representation of the PMCA procedure .........................................21

Figure 1.5. Schematic structural representation of the mouse genes encoding the PrP,

Doppel, and Shadoo proteins .............................................................................................23

Figure 1.6. Schematic representation of the domain architecture of the prion protein

family members .................................................................................................................27

Figure 1.7. High resolution structures of prion proteins ....................................................28

Figure 1.8. Alignment of PrP and Sho protein sequences from a variety of species .........31

Figure 1.9 Proposed models for Doppel neurotoxicity and PrPC neuroprotection in

cerebellar cells ...................................................................................................................53

Figure 1.10. The LPrP model of functional interactions between prion proteins in

cerebellar neurons of transgenic mice ................................................................................57

CHAPTER 3

Figure 3.1. Domain structure of PrP, ΔPrP, Doppel, and Shadoo .....................................76

viii

Figure 3.2. Construction of Shadoo polyclonal antibodies, epitope mapping, and analysis

of specificity.......................................................................................................................78

Figure 3.3. Biochemical characterization of murine Sho expressed in N2a cells ..............79

Figure 3.4. Analysis of mouse Shadoo in tissue preparations ...........................................82

Figure 3.5. ‘C1-like’ endoproteolytic processing of Shadoo in N2a cells and mouse

brains ..................................................................................................................................83

Figure 3.6. Expression of Sprn mRNA and Shadoo protein in the hippocampus .............86

Figure 3.7. Expression of Sprn mRNA and Shadoo protein in the cerebellum .................87

Figure 3.8. Neuronal expression of Shadoo in the cerebral cortex, thalamus, and

medulla ...............................................................................................................................89

Figure 3.9. Reciprocal and overlapping expression of Sho and PrPC in the CNS .............92

Figure 3.10. Expression of Shadoo in the spinal cord and retina ......................................94

Figure 3.11. Neuroprotective activity and Sho expression in CGN cells ..........................96

Figure 3.12. No change in Shadoo expression or distribution in PrP knockout brains .....99

Figure 3.13. Reduced Sho levels in clinically ill prion-infected mice .............................100

Figure 3.14. Shadoo levels are decreased in transiently transfected prion-infected cells

compared to uninfected cells ...........................................................................................102

Figure 3.15. Biochemical properties of Shadoo in infected and uninfected tissues and

cells ..................................................................................................................................104

Figure 3.16. Analysis of Shadoo stability in transfected N2a and ScN2a cells ...............106

Figure 3.17. Effects of over-expression and knockdown of Shadoo in ScN2a cells .......108

CHAPTER 4

Figure 4.1. Alignment of murine DPPX-S and DPP10-1 amino acid sequences ............133

Figure 4.2. Construction of DPPX and DPP10 polyclonal antibodies and analysis of their

specificities ......................................................................................................................134

Figure 4.3. Cloning of murine DPPX isoforms and expression in N2a cells ..................136

Figure 4.4. Cloning and characterization of a novel DPPX splice variant (“DPPX-

E_SV1”) ...........................................................................................................................139

Figure 4.5. Analysis of DPPX and DPP10 expression in the wild-type mouse brain .....141

Figure 4.6. PrPC levels have no effect on DPPX expression or endoproteolysis .............145

ix

Figure 4.7. DPPX exists as a dimer in vivo ......................................................................146

Figure 4.8. DPPX forms high molecular weight complexes with all three members of the

mammalian prion protein family as assessed in tissue culture cells ................................149

Figure 4.9. Analysis and confirmation of complexes containing prion proteins and

DPPX ...............................................................................................................................151

Figure 4.10. PrPC/DPPX complexes are present at the cell surface and are composed of

adjacent molecules displayed on the same cell ................................................................153

Figure 4.11. Mapping of DPPX complex determinants in PrPC demonstrates that the C-

terminal α-helical domain is required for complex formation and implies the existence of

a second binding site ........................................................................................................155

Figure 4.12. Mapping of DPPX complex determinants in Doppel demonstrates that the

helix B/B’ region is necessary for complex formation ....................................................157

Figure 4.13. Mapping of DPPX complex determinants in Shadoo demonstrates that the

N-terminal domain and the N-glycosylation site contribute to complex formation ........159

Figure 4.14. Membrane anchorage but not the cytoplasmic domain of DPPX is required

for complex formation with prion proteins ......................................................................161

Figure 4.15. Two distinct sites in DPPX mediate complex formation with prion

proteins .............................................................................................................................162

Figure 4.16. Competition experiments suggest that Dpl and PrPC share a common binding

site on DPPX ....................................................................................................................164

Figure 4.17. DPPX is expressed at high levels in the granule cell layer of the

cerebellum ........................................................................................................................166

Figure 4.18. PrPC/DPPX complexes are present in CGNs and DPPX is essential for the in

vitro survival of Prnp0/0 CGNs ........................................................................................169

Figure 4.19. No change in PrPC or Sho levels in mice genetically deficient for DPPX

expression ........................................................................................................................171

Figure 4.20. Biochemical characterization of prion proteins in wild-type and DPP6df5J/Rw

mouse brains ....................................................................................................................173

Figure 4.21. No change in prion disease incubation time in mice hemizygous for the

DPP6 df5J deletion allele ................................................................................................175

x

Figure 4.22. Analysis of DPPX and DPP10 expression in non-inoculated and RML prion-

inoculated mouse brains ...................................................................................................177

Figure 4.23. Over-expression and knockdown of DPPX and DPP10 in prion-infected N2a

cells ..................................................................................................................................179

CHAPTER 5

Figure 5.1. Model to explain to phenotypic interactions between prion proteins in

cerebellar neurons ............................................................................................................197

Figure 5.2. DPPX is expressed in motor neurons of the spinal cord ...............................202

xi

LIST OF ABBREVIATIONS

ADAM A disintegrin and metalloprotease ALS Amyotrophic lateral sclerosis ANOVA Analysis of variance APP Amyloid precursor protein BACE Beta amyloid cleaving enzyme BASE Bovine amyloidotic spongiform encephalopathy BCA Bicinchoninic acid BSA Bovine serum albumin BSE Bovine spongiform encephalopathy CDI Conformation-dependent immunoassay CGN Cerebellar granular neuron CHX Cycloheximide CJD Creutzfeldt-Jakob disease CNS Central nervous system CPEB Cytoplasmic polyadenylation element binding protein CSF Cerebrospinal fluid CWD Chronic wasting disease DDM n-dodecyl-β-D-maltoside DIG Digoxigenin DMEM Dulbecco’s modified Eagle’s medium DNA Deoxyribonucleic acid Dpl Doppel DPP Dipeptidyl peptidase ECL Enhanced chemiluminescence EDTA Ethylenediaminetetraacetic acid ELISA Enzyme-linked immunosorbent assay ELISPOT Enzyme-linked immunosorbent spot ER Endoplasmic reticulum FBS Fetal bovine serum FFI Fatal familial insomnia GABA Gamma-aminobutyric acid gCJD Genetic Creutzfeldt-Jakob disease GFAP Glial fibrillary acidic protein GFP Green fluorescent protein GPI Glycosylphosphatidylinositol GSS Gerstmann-Sträussler-Scheinker disease HA Hemagglutinin HBSS Hank’s balanced salt solution HEK Human embryonic kidney HRP Horseradish peroxidase iCJD Iatrogenic Creutzfeldt-Jakob disease IHC Immunohistochemistry ISH In situ hybridization KLH Keyhole limpet hemocyanin LDS Lithium dodecyl sulphate LINGO-1 LRR and Ig domain-containing, Nogo receptor-interacting protein MBM Meat and bone meal

xii

xiii

MEM Modified Eagle’s medium MES 2-(N-morpholino)ethanesulfonic acid MOPS 3-(N-morpholino)propanesulfonic acid mRNA Messenger ribonucleic acid MTT 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide N2a Neuro 2a NBT/BCIP Nitro blue tetrazolium chloride/5-bromo-4-chloro-3-indolyl phosphate N-CAM Neural cell adhesion molecule NFH Neurofilament H NMR Nuclear magnetic resonance NSE Neuron specific enolase ORF Open reading frame PBS Phosphate-buffered saline PCR Polymerase chain reaction PDI Protein disulfide isomerase PI-PLC Phosphatidylinositol-specific phospholipase C PK Proteinase K PMCA Protein misfolding cyclic amplification PMSF Phenylmethanesulfonylfluoride PNGaseF Peptide N-glycosidase F PrP Prion protein PrPC Cellular prion protein PrPres Protease-resistant prion protein PrPSc Infectious prion protein isoform RACK1 Receptor for activated C kinase1 RIPA Radio immunoprecipitation assay RML Rocky mountain laboratory RNA Ribonucleic acid RNAi RNA interference RT-PCR Reverse transcriptase polymerase chain reaction SCA Scrapie cell assay sCJD Sporadic Creutzfeldt-Jakob disease SDS Sodium dodecyl sulphate SDS-PAGE Sodium dodecyl sulphate polyacrylamide gel electrophoresis sFI Sporadic fatal insomnia Sho Shadoo shRNA Short hairpin RNA siRNA Small interfering RNA STI1 Stress-inducible protein 1 SOD Superoxide dismutase SSC Sodium chloride sodium citrate SV1 Splice variant 1 TACE Tumour necrosis factor-alpha converting enzyme TBS Tris-buffered saline TcTPC Time-controlled transcardiac perfusion crosslinking Tg Transgenic TMP21 Transmembrane protein of molecular mass 21 kDa tPa Tissue plasminogen activator TSE Transmissible spongiform encephalopathy USA United States of America vCJD Variant Creutzfeldt-Jakob disease

Chapter 1

Introduction

Prion Genes, Proteins, and Ligands:

Implications for Cellular Functions and Prion Disease

Portions of this section have been published in two review articles:

Watts, J.C., Balachandran, A., and Westaway, D. “The expanding universe of prion

diseases.” (2006) PLoS Pathogens, 2(3) e26.

Watts, J.C. and Westaway, D. “The prion protein family: Diversity, rivalry, and

dysfunction.” (2007) Biochimica et Biophysica Acta, 1772(6): 654-672.

1

1.1 An Overview of Basic Concepts in Prion Biology

Classes of Infectious Agents and Transmissible Spongiform Encephalopathies

The vast majority of infectious diseases of humans and animals are caused either

by parasites, fungi, bacteria, or viruses. A common component to all these classes of

disease-causing agents is the presence of an informational molecule such as DNA or

RNA which is capable of encoding structural or enzymatic constituents of the agent.

Although components of the host’s cellular machinery may be required for replication of

the agent (as is the case with certain viruses), maintenance of the nucleic acid genome is

essential for the persistence and spread of the infectious agent.

An emerging class of diseases is the transmissible spongiform encephalopathies

(TSEs) which are neurodegenerative disorders affecting certain mammalian species.

These diseases are atypical in that 1) there can be a long latency period between infection

and progression to clinical disease and 2) the diseases can manifest with sporadic, genetic,

or infectious etiologies. Originally thought to be caused by a ‘slow virus’ [1-3], the

infectious agents in these diseases exhibit numerous properties which are inconsistent

with the existence of a nucleic acid genome. Firstly, the TSE agent is largely resistant to

treatments which destroy nucleic acids such as ionizing radiation and nucleases but are

sensitive to reagents which destroy proteins [4-6]. Secondly, no nucleic acid of

substantial length has ever been found in purified fractions of the infectious agent

implying the absence of a nucleic acid genome [7-10]. Thirdly, unlike viruses, TSEs do

not elicit an immune response [11, 12]. Finally, although certain viral diseases are

compatible with a genetic etiology (i.e. the integration of retroviruses into the host

genome or the modulation of virus receptors by heritable mutations) spontaneous viral

diseases cannot be easily rationalized. These results led to the idea that TSEs are caused

by infectious self-replicating proteins [6, 13].

Purification of the infectious agent revealed that a single protein required for

transmission of the disease was present in the brain of infected animals (in this case,

hamsters infected with an isolate originally obtained from a sheep TSE) but absent in

brains of healthy animals [14]. This protein had a molecular weight of 27-30 kDa and

levels of the protein correlated well with the titre of the infectious agent. The

accumulation of evidence linking a protein to the TSE agent led Stanley Prusiner to

2

introduce the concept of a prion and the protein-only hypothesis to explain TSEs in 1982

for which he was later awarded the 1997 Nobel Prize in Medicine.

Prions, Nomenclature and the Protein-Only Hypothesis

Prions (proteinaceous infectious particles) are defined as small proteinaceous

infectious agents which are resistant to most procedures which modify nucleic acids [6].

The protein-only hypothesis states that a single protein termed PrP (the prion protein) is

the sole component of the TSE infectious agent. Prions are formed by a post-translational

conformational remodeling event in which a cellular protein, PrPC, is converted into the

disease-associated infectious form, PrPSc or PrPD. PrPSc can be differentiated from PrPC

by an increased resistance to digestion with proteases such as proteinase K (PK), poor

solubility in non-ionic detergents, an increased β-sheet content, and its propensity for

forming higher order structures such as amyloid fibrils [15-17]. Protease resistant PrP

formed by the digestion of PrPSc with PK is referred to as PrPres (or PrP27-30) to

distinguish it from full-length PrPSc (Figure 1.1A). The use of the term PrPSc implies that

the protein is infectious whereas PrPres does not necessarily imply infectivity.

The protein-only hypothesis is unique in offering explanations for all three

etiologies of prion diseases (Figure 1.1B). In infectious disease, exogenous PrPSc enters

the host, recruits copies of host-encoded PrPC and then templates their conversion

resulting in prion replication and disease pathogenesis. In genetic TSE disease, missense

or insertional mutations within the gene encoding PrP (the Prnp gene) can lead to an

increased propensity for misfolding of PrPC into PrPSc and thus the initiation of the

disease. Finally, a simple explanation for sporadic disease is that the ‘initiating event’

corresponds to the rare stochastic refolding of PrPC into PrPSc.

Although opponents of the protein-only hypothesis still exist [18], several lines of

evidence argue persuasively that it is the most plausible explanation for the accepted

properties of prion disease. Transgenic mice expressing high levels of a mutant mouse

PrP allele with the P101L mutation (akin to the P102L mutation in the human PRNP gene

which causes genetic prion disease) develop a neurodegenerative disease reminiscent of

scrapie which can be transmitted to asymptomatic mice expressing the same mutant PrP

at low levels [19, 20]. Secondly, PrPres perpetuated in vitro by serial dilution and rounds

3

Figure 1.1. Prion disease nomenclature and modes of acquisition. A: In prion disease, the host-encoded cellular prion protein (PrPC) undergoes a conformational transition to a disease-associated conformer termed PrPSc or PrPD. The approximate region of PrPC known to be conformationally altered in PrPSc is shown in red, although the exact boundaries are unknown. Detection of prion disease is commonly achieved in vitro by treatment with proteinase K which cleaves PrPSc near residue 90 to generate the protease-resistant fragment termed PrPres or PrP27-30. B: Prion diseases are unique in that they have a trimodal epidemiological manifestation. Disease can either be spontaneous, inherited via germ line mutations in the Prnp gene, or infectious whereby PrPSc is introduced into the host from an exogenous source.

4

of protein misfolding cyclic amplification (PMCA) is infectious to wild type hamsters

[21]. Finally, in one instance, ‘synthetic prions’ generated from recombinant PrP

expressed in E. coli have been generated. Truncated PrP (encompassing mouse residues

89-230) refolded into β-rich structures causes prion disease when inoculated into

transgenic mice expressing high levels of the identical truncated PrP allele [22]. However,

it has been argued that this phenomenon may be better characterized as the acceleration

of a cryptic disease present in the transgenic mice [23]. The irrefutable final proof of the

protein-only hypothesis would be the creation of synthetic prions, produced from full

length recombinant PrP, that are infectious to wild-type mice.

Non-Mammalian Prions

Following the discovery of mammalian prions, protein-based modes of

inheritance were also described in yeast and filamentous fungi [24]. To date, four non-

mammalian prions have been identified: [URE3] which is the prion form of Ure2p,

[PSI+] which is the prion form of Sup35p, [PIN+] which is the prion form of Rnq1p [25]

and [Het-s] which is the prion form of the HET-s protein [26]. These prions are inherited

in a non-Mendelian fashion and can be transferred from cell to cell by cytoplasmic

mixing. Yeast prions have distinct prion-forming domains that are predicted to be

unstructured and are unusually rich in asparagine and glutamine residues. Unlike

mammalian prions for which no physiological function has been ascribed to either PrPC

or PrPSc, the protein precursors to yeast prions are involved in the regulation of a variety

of biological processes. For instance, Ure2p is involved in nitrogen metabolism with the

wild-type protein inhibiting the uptake of ureidosuccinate whereas the prion form

([URE3]) is inactive [24]. The Sup35p protein is involved in the termination of

translation. In [PSI+] cells, read-through of nonsense (STOP) codons occurs more

frequently because Sup35p is inactive in the prion form. It has been proposed that [PSI+]

may be a genetic mechanism for revealing hidden genetic variation and promoting the

survival of cells in restrictive environments [27, 28], although others have argued that

yeast prions should be classified as a disease of yeast [29]. Studies on yeast prions have

revealed properties which can potentially be extrapolated to mammalian prions (such as

the conformational basis of prion strains). However, there are certainly fundamental

5

differences between mammalian and yeast prions which preclude a direct comparison.

For instance, yeast prions reside exclusively in the cytoplasm whereas mammalian prions

transit the secretory pathway and are present on the cell surface.

Another interesting example of potential prion-like behaviour occurs in the CPEB

protein of Aplysia californica. The CPEB family of proteins is involved in the regulation

of mRNA translation and the neuronal version has an extended domain reminiscent of the

prion-forming domains of yeast prions. Expressed in yeast, Aplysia CPEB can obtain

prion-like properties with the prion state binding more mRNA than the wild-type state

[30]. The authors have proposed that this prion-like mechanism may be involved in the

regulation of neuronal memory storage.

Prion Diseases of Animals

Scrapie

Scrapie is a prion disease of sheep and goats that has been recognized in Europe

since the mid 18th century [31] and there is speculation, based on the composition of

Chinese characters, that scrapie may even have existed in ancient China [32]. There is no

evidence that scrapie is transmissible to humans. Unlike prion diseases of humans,

scrapie is known to be horizontally transmissible (i.e. the infection of a disease-free

animal can occur by placing the animal in the presence of an infected animal). The

mechanism of horizontal transmission is unclear, though at one stage attributed to

contaminated placentae [33, 34]. Recent work has illustrated the novel idea that chronic

inflammation can modify the organ tropism of prions [35], and that infectivity can be

detected in the urine of mice with chronic inflammatory kidney disease [36]. In addition,

PrPSc has been found in the kidneys of scrapie-infected sheep suggesting that some PrPSc

may also be present in the urine [37]. Furthermore, deposits of PrPSc have been found in

the mammary glands of sheep with coincident scrapie infection and mastitis [38]. These

results have led to the suggestion that inflammation may play a role in the horizontal

spread of prions.

Bovine Spongiform Encephalopathy

Bovine spongiform encephalopathy (BSE or mad cow disease) is a disease of

cattle that was first recognized in 1987 [39]. Evidence suggests that BSE spreads by the

6

feeding of cows with meat and bone meal (MBM) containing animal-derived components

(i.e. recycled brain-derived material from cattle). Contamination of the MBM supply with

prions appears to be the origin of the BSE epidemic which occurred in the United

Kingdom [40]. Unlike scrapie, BSE appears to be transmissible to humans in the form of

variant CJD (vCJD). However, despite the presence of a large number of infected cattle

(on the order of one million), comparatively few people have died from vCJD. As of

January 2008, 163 people have succumbed to vCJD in the United Kingdom

(http://www.cjd.ed.ac.uk/figures.htm). The BSE and vCJD epidemics appear to be on the

decline in large part due to regulations forbidding the use of animal products in MBM

[41]. In Canada, eleven cases of BSE have been identified which have resulted in severe

economic consequences for the agricultural industry, particularly in Alberta.

Chronic Wasting Disease

Chronic Wasting Disease (CWD) is a prion disease of mule and white-tailed deer

as well as elk [42]. It also appears that moose are susceptible to CWD [43]. The most

striking aspect of CWD is that it seems to be extremely horizontally transmissible, and

perhaps is even more communicable than scrapie [44, 45]. CWD can attain spectacular

attack rates in affected populations (approximately 30% in some local populations of deer

and essentially 100% in captive research facilities [46]), and is undergoing a swift

geographical spread eastward across the USA (a migration that may be determined in

greater part by the commercial trucking of deer) and includes outbreaks in farmed and

feral animals in Saskatchewan and Alberta. There are reported and anecdotal instances

where introduction of one affected animal led to disease in several other members of a

farmed herd [47]. Also, physical proximity by penning is claimed to increase disease

incidence. An oral route of exposure is inferred from the presence of infectivity in

lymphoid tissue of the alimentary canal [48, 49], yet the problem remains as to which

bodily fluid or secretion contains infectivity. Possibilities include blood and saliva as

both have been determined to harbor infectious prions as demonstrated by bioassay [50].

There is no evidence for transmission of CWD to humans thus far [51], although the

eating of venison from animals with the disease should be avoided since prions are found

within the muscle tissue of CWD-infected animals [52].

7

Other Prion Diseases of Animals

The spectrum of prion diseases appears to be increasing and several new prion

diseases of animals have been described recently [53]. Abnormal presentation of prion

disease in two cattle in Italy was sufficiently distinct from BSE as to suggest a distinct

neurological syndrome denoted bovine amyloidotic spongiform encephalopathy (BASE)

[54]. BASE is transmissible to mice confirming that it is a veritable TSE and strikingly,

its properties in mice were indistinguishable from BSE suggesting that BASE may

represent the original source of BSE [55]. Another disease, Nor98, sometimes referred to

as atypical scrapie, was first detected in Norwegian sheep [56]. There is no evidence of

lateral or horizontal transmission, with cases (one per flock) observed in geographically

dispersed locations. Furthermore, Nor98 possesses distinct genetic, biochemical, and

histopathological signatures from scrapie. Nor98 is transmissible to mice, confirming its

classification as a prion disease [57]. The origins of BASE and Nor98 are unclear,

although it is conceivable that both represent distinct sporadic prion diseases of animals.

Prion Diseases of Humans

Kuru

Kuru is a prion disease of the Fore people of Papua New Guinea. The

transmissibility of the disease was confirmed by transmission to chimpanzees [58]. Kuru

is spread orally via ritualistic cannibalism of the brains of deceased individuals [58, 59].

After the cessation of cannibalism, the disease has largely died out. The recent deaths of

individuals infected with Kuru in the 1950’s have led to speculation that prion diseases

can have incubation periods in excess of 50 years [60].

Creutzfeldt-Jakob disease

Creutzfeldt-Jakob disease (CJD) is a neurodegenerative disease with an incidence

of approximately one case per million that was first described in the early 20th century.

CJD can occur via all three prion disease etiologies. Sporadic CJD (sCJD) is the most

common form accounting for approximately 85% of all cases. The origin of sCJD is

unclear although it is thought to stem from the spontaneous formation of PrPSc and its

subsequent propagation. Genetic CJD (gCJD, previously referred to as familial CJD or

fCJD) is caused by autosomal dominant mutations in the PRNP gene (the gene encoding

8

PrP). Over ten independent mutations in PRNP have been identified which cause gCJD

[61]. Because the mutations are not clustered within a specific region of the PrP amino

acid sequence, ascertaining the causal link between mutation and disease has been

difficult. It was thought that all mutations destabilize the structure of PrPC, increasing its

propensity for misfolding to PrPSc, but the cumulative data has revealed exceptions to this

rule [62, 63]. Infectious or iatrogenic CJD (iCJD) occurs when PrPSc is accidentally

introduced into the body. Historical examples include the use of prion-contaminated

neurosurgical instruments [64], treatment with contaminated human growth hormone [65],

and dura matter grafts [66].

Variant CJD

Variant CJD is an infectious prion disease that was first recognized in 1996 [67]

and a vast amount of evidence suggests that vCJD results from exposure to BSE-derived

prions [68-70]. In human PrP, codon 129 is polymorphic with either methionine (M) or

valine (V) residues possible at this position. All clinically confirmed cases of vCJD have

been of the PRNP codon 129 M/M genotype suggesting that infections may be

proceeding at a slower pace in M/V and V/V individuals exposed to BSE prions, and that

the possibility exists that these individuals currently have subclinical prion disease. In

agreement with these ideas, transgenic mice expressing just the human 129V form of

PrPC exhibit a significant barrier to infection with either BSE or vCJD prions, and those

that do become infected have distinct neuropathological characteristics and propagate a

different type of PrPSc from vCJD [71]. One particularly worrying aspect about vCJD is

that it appears to be transmissible by blood transfusions, including to individuals with an

M/V codon 129 genotype [72-74].

Gerstmann-Sträussler-Scheinker disease

Gerstmann-Sträussler-Scheinker disease (GSS) is a genetic prion disease that like

gCJD can result from a variety of mutations located throughout the PRNP gene [61]. One

interesting mutation that causes GSS is the Y145STOP mutation in human PrP which

generates a truncated PrP molecule that lacks the majority of the α-helical domain and

accumulates in amyloid deposits in the parenchyma and blood vessels [75, 76]. In

contrast to classical CJD, smaller PK-resistant PrP fragments (7-11 kDa) can be found in

9

certain types of GSS resulting from C-termini positioned in the vicinity of residue 150 of

human PrP [77, 78].

Fatal Familial Insomnia and Sporadic Fatal Insomnia

Fatal Familial Insomnia (FFI) is a genetic prion disease caused by a PRNP D178N

mutation in cis to a methionine polymorphism at codon 129 [61]. In contrast, patients

with the D178N mutation in cis to valine at codon 129 develop gCJD. FFI is

characterized by a reduction in total sleep time with a preferential involvement of the

thalamus in disease pathology. More recent work has described a sporadic version of the

same syndrome. To date, eight cases of sporadic fatal insomnia (sFI) have been reported

[79-82], all of which resemble typical fatal familial insomnia and are homozygous for the

methionine codon 129 polymorphism but lack any mutation in the PRNP gene. sFI has

successfully been transmitted to mice and the resulting neuropathology and

electrophoretic signature of PrPSc are indistinguishable from those obtained with mice

inoculated with fatal familial insomnia isolates [80].

Neuropathology and Pathogenesis of Prion Diseases

There are four principle neuropathological findings for prion disease (Figure 1.2):

spongiform change (vacuolar degeneration of brain parenchyma), death of neurons,

astrocytic gliosis (in itself a non-specific reactive response to CNS damage but of unusual

intensity in prion disease [83]), and the presence of extracellular amyloid plaques in some

but not all varieties of prion disease. The majority of prion diseases (including scrapie,

mouse-adapted scrapie, BSE, and CJD) are characterized by large amounts of spongiform

degeneration and the accumulation of PK-resistant PrP, but with little or no PrP amyloid

plaque formation [83]. Amyloid plaques are frequently found in vCJD brains and are

often described as ‘florid plaques’ due to decoration of the perimeter of the plaque with

regions of spongiform change. In contrast, plaques are only found in 5-10% of CJD cases

[83]. Regions of pathology within the brain are usually specific to and diagnostic of a

given prion disease or prion strain. For instance, in FFI the thalamus is preferentially

targeted whereas in D178N CJD the cerebral cortex is the most severely damaged area

[61].

10

A B

DC

Figure 1.2. The neuropathological hallmarks of prion disease. A: Spongiform (vacuolar) degeneration in the grey matter of the brain of a mouse infected with the RML strain of prions. Vacuolation of the brain (black arrows) is observed in most prion diseases and causes the degeneration of neuronal processes and eventually results in the death of neurons. B: Activation of astrocytes (reactive astrocytic gliosis: white arrows) in the brain of a patient with CJD as observed by immunostaining for glial fibrillary acidic protein (GFAP). Image taken from DeArmond et al. [83] C: Deposits of PrP amyloid as observed by immunohistochemistry in the brain of an elk infected with CWD. Image taken from Watts et al. [53] D: ‘Florid’ plaque (black arrow) consisting of a central PrP amyloid deposit surrounded by a halo of spongiform change in the brain of patient with vCJD. Image taken from Will et al. [67]

11

The mechanism(s) by which PrPSc causes disease are unclear. One study which

used Prnp0/0 mice that had been grafted with Prnp+/+ tissue demonstrated that only the

tissue which expressed PrPC was damaged despite the large amounts of PrPSc in Prnp0/0

tissue [84]. This suggests that the presence of PrPC is required for prion pathology and

that PrPSc is not inherently neurotoxic. In contrast, in mice expressing PrPC exclusively in

astrocytes, pathology in prion-inoculated mice was observed in tissue surrounding

astrocytes (i.e. in tissue lacking PrPC) whereas astrocytes remained undamaged [85].

Recent studies on transgenic mice expressing PrP lacking its GPI anchor (i.e. secreted

PrP) have shown that these mice produce large amounts of PrPSc but fail to develop

clinical prion disease following prion challenge [86]. This suggests that membrane

anchorage of PrP is essential for prion pathology, perhaps implying the existence of a

transmembrane protein which mediates PrPSc toxicity [87]. There is also considerable

debate as to whether the accumulation of cytoplasmic PrP may play a role in prion

toxicity. Transgenic mice expressing cytoplasmic PrP exhibit a neurodegenerative

phenotype (albeit one that differs from classical prion disease) and accumulation of PrP

in the cytoplasm results in the generation of a PrPSc-like molecule [88, 89]. However,

other investigators have failed to find any toxicity associated with cytoplasmic PrP [90-

92].

Prion Disease Therapeutics

Prion diseases of both humans and animals are invariably fatal. At the present

time there is no treatment or vaccine for this class of diseases. Prion vaccines are unlikely

to be effective because of immune tolerance to the host-encoded PrPC protein although it

may be possible to circumvent this tolerance by various means [93, 94]. Numerous

compounds or reagents which prevent the formation of PrPSc or accelerate its clearance in

vitro (typically prion-infected N2a or GT1 cells) have been identified [95-99]. However,

none of these have been successful to date in a clinical setting [100]. Antibodies to PrP

prevent the formation of PrPSc in vitro, likely by preventing PrPC from interacting with

PrPSc [101, 102]. The in vivo efficacy of anti-PrP antibodies is questionable because they

may have a hard time crossing the blood-brain-barrier. Nonetheless, peripheral prion

inoculation followed by infusion of antibody at the same site was effective at delaying

12

prion disease [103]. Additionally, transgenic expression of an anti-PrP antibody was

effective at preventing disease [104]. Interestingly, transgenic expression of a soluble

dimeric prion protein (via fusion to the Fcγ region of immunoglobulin) is also

prophylactic for prion disease [105]. No effective means for treating prion disease

following the onset of clinical disease have been uncovered thus far. Nonetheless, it has

recently been shown that depletion of PrPC in mice following prion inoculation can

reverse behavioural deficits and neuropathology [106] providing hope that prion disease

pathology can also be reversed in humans.

Prion Replication

The exact mechanism which governs prion replication is unknown. However, it is

clear that PrPC is absolutely required for both prion replication and disease progression as

Prnp0/0 mice do not develop disease following inoculation with infectious prions and do

not propagate prions in their brains [107, 108]. There are two models which have been

proposed to explain the mechanism of prion replication (Figure 1.3). The first, termed

template-directed refolding, postulates that exogenous PrPSc interacts with host PrPC and

templates its conversion into an additional copy of PrPSc. These PrPSc molecules would

then serve as additional templates for converting more molecules of PrPC allowing the

infectious PrPSc form to propagate. In this model, spontaneous conversion of PrPC to

PrPSc is not favoured, likely due to a large energy barrier between the two conformational

isoforms. The second model, termed the seeded nucleation model, assumes that PrPC and

PrPSc are both normally present in a reversible equilibrium within the brain with the

balance shifted strongly towards PrPC. In order for a seed to form, several molecules of

PrPSc need to come together. Once the seed is formed, PrPSc becomes stabilized and

recruitment of additional PrPSc molecules occurs at a much faster rate. This model

postulates that small amounts of PrPSc are present in a healthy brain. In potential

agreement with this, PrPres can sometimes be amplified from control healthy brain

samples using the protein misfolding cyclic amplification (PMCA) procedure [109],

however this could be due to an induction of a PrPres-like structure by thermal and kinetic

energy from the repeated sonication cycles needed for amplification events in PMCA.

13

Figure 1.3. Proposed models of prion replication. A: In the template-directed refolding model, PrPSc is not normally present in the brain and its spontaneous formation is impeded by a large energy barrier between PrPC and PrPSc. Exogenous PrPSc recruits host PrPC and templates its conversion to an additional copy of PrPSc. These two copies of PrPSc would then interact with further copies of PrPC allowing prion replication to progress. Amyloid formation is a byproduct of prion replication and does not figure explicitly in the mechanism. B: In the seeded nucleation model, an equilibrium exists in the brain between PrPC and PrPSc (although the balance is shifted greatly towards PrPC). The formation of a PrPSc seed occurs slowly and is favoured by the introduction of exogenous PrPSc. Once the seed has formed, recruitment of additional PrPSc occurs rapidly allowing the formation of larger amyloids. Fragmentation of amyloid into smaller pieces generates new seeds and allows prion replication to progress. In this model, the postulated minute amounts of monomeric PrPSc present in the brain would be harmless and the infectious agent would necessarily consist of multimeric PrPSc. Figure adapted from Aguzzi and Polymenidou [87].

14

There has been considerable debate surrounding whether or not other proteins or

molecules are required for prion replication in vivo. A hypothetical prion-converting

factor termed Protein X was postulated based on the properties of transgenic mice

expressing human or chimeric mouse-human PrP proteins [110]. Using an indirect

approach, the binding site for Protein X on PrP was deciphered despite the fact that

Protein X remains unidentified to date [111]. The definition of Protein X requires that it

bind to both PrPC and PrPSc and that Protein X knockout mice are refractory to prion

disease. However, despite enormous research efforts, no plausible candidates for Protein

X have surfaced. Other proteins such as the 37-kDa/67-kDa laminin receptor have been

reported to be required for prion replication [112], however this has not been confirmed

in vivo. In vitro, numerous studies suggest that polyanionic molecules such as RNA or

heparan sulfate proteoglycans can increase the rate of PrPres formation [113-116]. In fact,

the minimal components for infectious prion formation in vitro appear to be purified PrPC

and co-purified lipid molecules, purified PrPSc, and a polyanionic molecule [117]. It is

not known whether a polyanion is required for prion replication in vivo.

For determinants within PrP, it is known that the N-terminus is dispensable for

prion replication. Transgenic expression of truncated mouse PrP alleles in Prnp0/0 mice

has revealed that deletions in mouse PrP up to residue 93 do not abrogate prion

replication [118-120]. In contrast, mice with PrP deletions extending to residue 106 or

further are not susceptible to disease [121]. Miniprions have been formed from a

compound deletion in mouse PrP where residues 23-88 and 141-176 have been removed

(commonly referred to as PrP106) [122]. In vitro, residues 112-119 within the

hydrophobic tract have also been shown to be required for PrPSc formation [123]. In

summary, it appears that the central region of PrP (the region following the octarepeats

and preceding the start of the α-helical domain) and portions of the α-helical domain are

required for prion replication (Figure 1.1A). Recent data has demonstrated that a

significant portion (residues 160-220) of the human PrP C-terminal domain is refolded in

amyloid fibrils prepared from recombinant prion protein suggesting that the

conformational remodeling of PrP in prion disease may be more extensive than originally

believed [124].

15

Prion Transmission and the Species Barrier

When prions from one species are introduced into a second species, there is often

a significant species barrier effect which limits the transmission of prions [125]. Species

barrier effects include prolonged incubation time and inefficient transmission (i.e. only a

small proportion of inoculated animals get the disease). The species barrier has important

implications for human health as it appears that there is a significant species barrier

between humans and CWD or scrapie prions. In contrast, BSE prions appear to be rather

promiscuous and can enter a number of diverse species including humans in the form of

vCJD [67]. At the molecular level, the species barrier is likely dictated by the amino acid

sequence of PrP. For instance, infection of mice expressing both hamster and mouse PrP

with hamster prions leads to formation of hamster PrPSc and conversely, inoculation of

mouse prions leads to the formation of mouse PrPSc [126]. Interestingly, the bank vole

appears to be susceptible to prions from a variety of human prion diseases despite a low

degree of sequence homology between human and vole PrP [127]. This suggests that

factors other than the PrP amino acid sequence may influence the species barrier.

Within a given species, the amino acid of PrP can also profoundly influence prion

disease susceptibility and incubation times. For example, two polymorphisms in the

mouse Prnp gene (encoding the L108F and T189V polymorphisms in mouse PrP) control

prion incubation time [128-131]. A second example is the sheep gene in which

polymorphisms at ovine PrP residues 136, 154, and 171 have a strong effect on

susceptibility to scrapie with the VRQ allele being associated with increased

susceptibility [132]. Thirdly, no cases of sporadic CJD have been found in individuals

with the lysine variant of the codon 219 E/K polymorphism in human PrP [133]. Finally,

the codon 129 M/V polymorphism in the human PRNP gene controls both prion disease

phenotype (i.e. determining FFI versus gCJD phenotypes when in cis to a D178N

mutation) and susceptibility to disease (no cases of vCJD have been observed in non-

methionine homozygotes [134]).

Prion Strains

Prion strains are distinct prion isolates which display unique biological properties

and result in characteristic phenotypes when propagated in a given species. Strains can be

16

identified and differentiated on the basis of clinical manifestation (such as the

“scratching” and “drowsy” strains of scrapie when given to goats) [135], differential

incubation times [136], pathological lesion profiles (i.e. different strains target different

neuroanatomic areas of the brain) [137], the ratio of glycoforms (unglycosylated,

monoglycosylated and di-glycosylated) within protease-digested preparations of PrP [68],

the size of the PK-resistant PrP fragment [138-140], differential reactivity to luminescent

conjugated polymers [141], and other biochemical properties such as conformational

stability [142]. The existence of prion strains comprises one of the main challenges to the

protein-only hypothesis and opponents argue that such diversity cannot be encoded in the

absence of a nucleic acid [143]. Indeed, multiple prion strains can exist for a given PrP

amino acid sequence suggesting that strain variety is encoded by a different mechanism.

Accumulating evidence suggests that strain-specific properties are encoded by the

conformation of PrP. Subtle changes in conformation could lead to the phenotypic

differences observed such as differential sizes of PK-resistant fragments and

neuroanatomic target areas [140]. This hypothesis has not been definitively proven for

mammalian prions to date, in large part due to the lack of high resolution structural data

for PrPSc. Studies on various prion strains using the conformation-dependent

immunoassay suggest that individual strains have differential availability of antibody

epitopes implying that each strain has a different PrP conformation [144]. Also, small

changes in amino acid sequence of a mutant PrP23-144 molecule lead to distinct

ultrastructural properties of prion fibrils in vitro [145, 146]. Furthermore, prion strains

which are more susceptible to chemical denaturation have the shortest incubation times

and vice versa suggesting that the conformational stability of a given strain governs it

replication rate [147]. However, the strongest evidence for the conformational encoding

of prion strains has come from studies on yeast prions. Amyloids of the Sup35 protein

prepared at different temperatures lead to unique conformations which are stably

propagated. When these variants are introduced into yeast, different strains of [PSI+] are

obtained suggesting that conformation of Sup35 amyloid governs strain-specific

properties [148, 149]. Furthermore, the structural differences in Sup35 strain variants

have been probed by hydrogen/deuterium exchange revealing that differences in the

length of the amyloid β-sheet core dictate strain structure and biological properties [150].

17

Nonetheless, the issue of prion strains will remain an open question in prion biology until

parallel high resolution structures exist for different strains of mammalian prions.

Detection of Prions and Prion Bioassays

Several techniques are used for post mortem diagnosis of prion disease [151] (Table 1.1).

Brains of infected individuals usually exhibit pronounced spongiform change (areas of

vacuolation), neuronal degeneration and death, astrogliosis, and occasionally, the

accumulation of amyloid plaques containing aberrant PrP. Spongiform change can be

observed using standard histological procedures while abnormal PrP deposits can be

viewed following pretreatment with formic acid and hydrated or hydrolytic autoclaving

[152], to reduce the immunoreactivity of PrPC, prior to staining with a PrP-specific

antibody. Other diagnostic tests rely on the detection of PrPres as a surrogate marker for

PrPSc. Following PK treatment, PrPres can be detected using either a Western blot or an

ELISA, with these strategies forming the basis for two of the most widely-used

commercial tests for BSE. A distinct approach, one that circumvents protease digestion,

is the conformation-dependent immunoassay (CDI) [144, 153, 154]. This method takes

advantage of the differential availability of sequestered antibody epitopes between PrPC

and PrPSc. An antibody is used which recognizes a central epitope with differential

accessibility between PrPC (available) and PrPSc (not accessible until thermal or chemical

denaturation). Ratios are calculated between signals obtained by ELISA for the native

and denatured forms of the test samples, which are then used to ascertain the presence of

PrPSc. In this technique there is the potential to detect protease-sensitive forms of PrPSc

which may outnumber their insoluble and more protease-resistant counterparts.

Conversely, antibodies have now been described which may react with determinants

unique to PrPSc [155-158].

In contrast to ‘static diagnostics’ which biochemically detect aberrant PrP, ‘active

diagnostics’ amplify PrPSc or infectivity in vivo or in vitro before a detection step or

biological read-out. The prion bioassay in mice is the most commonly used method for

assaying infectivity in vivo and is considered to be the ‘gold standard’ for the detection of

prions. Following intracerebral inoculation of the test sample, mice typically succumb to

prion disease following an incubation period of approximately 150 days (which is

18

Table 1.1. Commonly-used methods for detecting TSEs and PrPSc Method Principle Notes Reference(s)

Histology/ Immunohistochemistry

Fixed tissue is examined for the hallmarks of TSEs or processed

by hydrated or hydrolytic autoclaving followed by

immunodetection of aberrant PrP deposits

‘Gold standard’ for confirming TSE diagnosis; neuroanatomical lesions are characteristic of individual

prion strains

Reviewed in [151] and [159]

Western blot/ ELISA

Homogenate is digested with PK to remove PrPC followed by

immunodetection of PrPSc using either Western blot or ELISA

formats

Basis for many commercially-available tests for TSEs; rapid

method

Reviewed in [151] and [159]

Conformation- Dependent

Immunoassay (CDI)

An antibody which recognizes a central epitope with differential

accessibility between PrPC (available) and PrPSc (not

accessible until denaturation) is used in a sandwich ELISA format. Ratios calculated

between signals obtained for the native and denatured forms of the test sample signify the presence

or absence of PrPSc

No protease digestion is required;

capable of detecting soluble and more protease-sensitive forms of PrPSc; rapid method

[144, 153, 154]

Bioassay (wild-type mice)

Test samples are injected intracerebrally into mice and

disease progression monitored

Subject to species barrier effects (inefficient

transmission or extended incubation times);

expensive, time-consuming, and labour-intensive

[160]

Bioassay (transgenic indicator

mice)

Transgenic mice expressing the PrP gene of interest (on a mouse

Prnp0/0 background) are inoculated as above

Eliminates species barrier effects resulting in shorter

incubation times

[161-164]

Scrapie Cell Assay Highly prion-susceptible N2a cells are exposed to the test

sample, split three times, filtered onto an ELISPOT plate, digested

with PK, and then stained for PrPSc. Positive cells are counted

using a specialized computer setup.

Can quantitate levels of infectivity;

ten times faster than mouse bioassay and over two orders of magnitude less expensive

[165]

Protein Misfolding Cyclic Amplification

(PMCA)

An excess of PrPC and repeated cycles of amplification and

sonication are used to amplify any PrPSc present in the test

sample.

Can detect PrPSc in pre-clinical (asymptomatic)

infected animals; can detect PrPSc in the blood of

experimentally-infected hamsters; amplified PrPres is

infectious to hamsters

[21, 166-168]

19

dependent on the strain of inbred mouse and prion strain utilized). One drawback of

bioassays is that unless mouse or rodent-adapted prions are being tested, these bioassays

may be subject to species-barrier effects leading to prolonged incubation times or

inefficient transmission of disease. Attempts to circumvent the species barrier have

resulted in the development of transgenic indicator mice, mice which express PrPC of the

same amino acid sequence as the PrPSc in the test inoculum [110, 161-164]. Ex vivo

mouse prion bioassays which utilize either neurospheres [169] or cerebellar slices [170]

have also recently been described and can be performed much more rapidly than

conventional bioassays.

Another ‘active’ technique which is ten times faster than conventional bioassays,

and approximately two orders of magnitude cheaper is the Scrapie Cell Assay (SCA)

[165]. This cell culture-based method utilizes sub-lines of mouse N2a neuroblastoma

cells that have been selected for enriched susceptibility to prions and measures the ability

of a test sample to generate PrPSc-positive cells. One drawback of this technique is that

attempts to use mouse-adapted prion strains other than the Rocky Mountain Lab (RML)

isolate (such as the murine Me7 and 22A isolates) were unsuccessful [165]. This assay

has now been expanded to incorporate multiple cell lines (the cell panel assay) and can be

used to discriminate between the RML, 22L, Me7, and 301C strains of prions [171].

A third ‘active’ diagnostic technique denoted PMCA (Protein Misfolding Cyclic

Amplification) was developed with the goal of producing large quantities of PrPres in

vitro [166] (Figure 1.4). Notably, PMCA-generated PrPres is as infectious as brain PrPSc

[21, 172]. The method is loosely similar in a conceptual sense to nucleic acid PCR. Fresh

brain homogenate from non-infected animals is used as a source of PrPC, and brain

homogenate from scrapie-infected animals as a source of PrPSc [173]. In this analogy,

PrPSc is akin to the rare nucleic acid target sequence of PCR and PrPC is akin to the

cocktail of oligonucleotide primers and mononucleotides that allow de novo nucleic acid

synthesis. Small amounts of infected material are diluted into normal brain homogenate

and in vitro conversion – presumably a form of templated protein refolding – is allowed

to proceed at 37ºC. A key ingredient is a subsequent sonication step, formally analogous

to thermal denaturation of complementary DNA strands in a PCR reaction. Here,

mechanical energy is used to break up newly formed PrP aggregates into smaller

20

A B

Figure 1.4. Schematic representation of the PMCA procedure. A: Brain homogenate from an infected animal (containing PrPSc) is diluted into homogenate from a healthy animal (containing PrPC) and incubated at 37ºC. During this stage, some molecules of PrPC are converted to PrPres by conformational rearrangement and are added to the growing PrPres unit. Sonication breaks up PrPres into smaller units generating new seeds for conversion. These steps are repeated in a cyclic fashion in order to amplify the amount of PrPres present in the initial sample. The stoichiometry and directionality of amplification depicted are for illustrative purposes only and are not meant to represent intrinsic properties of the system. B: Serial PMCA is used to detect minute quantities of PrPSc in the test sample. Following one round of PMCA cycles, the amplified material is diluted into fresh brain homogenate and additional cycles of PMCA performed. This can be repeated in order to perform multiple rounds of PMCA cycles.

21

structures, with the latter providing new seeds for PrPres formation in reiterations of the 2-

step procedure. In this way repeated cycles of conversion and sonication are performed in

order to amplify any PrPSc present in the starting sample. PMCA is a very powerful and

sensitive technique and has been used to detect prions in the blood of pre- and post-

symptomatic experimentally-inoculated animals [168, 174]. A recently-published

technique which builds on PMCA (termed rPrP-PMCA) uses recombinant PrP as the

source of PrPC and can detect as little as 50 attograms of PrPSc [175].

1.2 Prion Genes and Proteins

The Prion Supergene Family

PRNP

The genes encoding the human, hamster, and mouse prion proteins were first cloned in

1986 [176-179]. The Prnp gene is located on chromosome 2 in mice and PRNP is on

chromosome 20 in humans. Both genes consist of three exons with the entire open

reading frame located within exon 3 (Figure 1.5). Polymorphisms in both the human and

mouse prion genes have been described with the codon 129 polymorphism in human

PRNP being a critical determinant of prion disease susceptibility. The effects of Prnp

polymorphisms on documented PrPC functions, such as neuroprotection against Doppel

toxicity, have not yet been investigated. Several mutations in PRNP are known to cause

gCJD including D178N (which causes either gCJD or fatal familial insomnia, depending

on the residue in cis at codon 129), E200K (which causes gCJD), and P102L (which

causes GSS) [61, 180]. Interestingly, no genetic prion disease-causing mutations in Prnp

have been described in the published literature for non-human species. One possible

exception is a bovine PrP E211K polymorphism found in a single BSE case with atypical

PrPres from the USA. Notably, this polymorphism is equivalent to the E200K variant of

human PrP, which causes gCJD. Genetic ablation of Prnp in mice causes no spectacular

phenotypes other than conferring resistance to prion disease [107]. The predominant area

of Prnp expression is in post-mitotic neuronal cells within the CNS [181], although extra-

neuronal expression at lower levels is well-documented in a number of other regions

including lymphoid tissue, on lymphocytes, and in muscle [182-185].

22

Figure 1.5. Schematic structural representation of the mouse genes encoding the PrP, Doppel, and Shadoo proteins. The Prn locus on chromosome 2 contains the Prnp and Prnd genes encoding the PrP and Doppel proteins, respectively. The Sprn gene encoding the hypothetical Shadoo protein is found on chromosome 7. For all three genes, the open reading frame (ORF) is contained within a single exon. In ataxic strains of Prnp0/0 mice, the splice acceptor site of Prnp exon 3 is deleted resulting in the production of chimeric Prnp/Prnd mRNA’s. The exonic structure of the most abundant chimeric mRNA generated is shown.

23

PRND

The discovery of the gene encoding the Doppel protein came about in a rather indirect

manner. While the earliest strains of Prnp0/0 mice exhibited no phenotypic abnormalities

[186, 187], other strains exhibited late-onset ataxia characterized by loss of Purkinje cells

in the cerebellum [188, 189]. Initially, these results were interpreted in favour of a role

for PrPC in the long-term survival of Purkinje cells. However, upon closer examination, it

became apparent that the phenotypic effect resulted from an artifact of the genetic

engineering used in ataxia-prone Prnp0/0 mice. A previously unappreciated gene lying 16

kb downstream of the murine Prnp locus was discovered as a result of an extensive

sequencing effort [190]. The gene, dubbed Prnd, contained an open reading frame

encoding the Doppel (Dpl) protein. Strains of Prnp0/0 mice in which the splice acceptor

site of exon 3 of the Prnp gene has been removed results in the production of chimeric

Prnp/Prnd mRNA’s (essentially putting the Prnd gene under the control of the Prnp

promoter) and permitting the Doppel protein (whose post-embryonic expression is

principally confined to the testis) to be expressed in the CNS (Figure 1.5). Bovine Doppel

expression has also been observed in circulating lymphoid cells, in B-cells, neutrophils,

and in follicular dendritic cells within lymphoid follicles [191, 192]. Several

polymorphisms have been noted in the human PRND gene (both within the coding

sequence and in the untranslated regions) and there is debate as to whether or not these

confer risk for sporadic CJD [193-195]. The functional consequences of PRND

polymorphisms have not yet been fully explored, although they are known not to alter

Doppel expression or trafficking in tissue culture cells [196].

SPRN

The third member of the prion gene family was sighted in a search of publicly-available

databases for nucleotide sequences with similarity to the Prnp sequence. One such hit

was for a cDNA predicted to encode a short protein which has similarity to the alanine-

and valine-rich central hydrophobic region of PrP. The protein was coined Shadoo

(supposedly the Japanese word for ‘shadow’) and is commonly referred to as Sho, and

the gene was termed Sprn (for ‘shadow’ of the prion protein) [197]. Sprn seems to be

widely conserved in nature, being present in the genomes of lower organisms such as

zebrafish all the way up to rodents and primates. Sprn is not part of the Prn genomic

24

locus (containing Prnp and Prnd) and instead can be found on chromosomes 7 and 10 in

mice and humans, respectively. Like Prnp and Prnd, the entire open reading frame of

Sprn is contained within a single exon (Figure 1.5). Analysis of expression patterns imply

that Sprn expression is restricted to the brain, suggesting that unlike Doppel, Shadoo may

be pertinent to prion protein-associated CNS phenomena. Mammalian genomes have a

single Sprn gene, whereas two zebrafish genes have been noted (the Sprna and Sprnb

genes encoding the Sho-1 and Sho-2 proteins, respectively). Bioinformatic analysis of

Sprn and Prnp sequences from diverse organisms has suggested that the archaic prion

protein gene may have been related to Sprn, with the Prnp and Prnd genes evolving later

[198]. Polymorphisms in the human SPRN gene are rare but not unlike ovine PRNP,

ovine SPRN has a number of missense variants clustered within the central region of the

protein (Nathalie Daude, submitted).

PRNT

A hypothetical fourth prion gene has been described in the form of PRNT, a sequence

found in the same genomic cluster as human PRNP and PRND (situated approximately 3

kb downstream of PRND) [199]. PRNT exists as three distinct splicing isoforms with the

corresponding predicted amino acid sequences lacking any distinctive homology to either

PrPC or Doppel. Further work has shown that PRNT is not present in rodents [198] and

that its existence may be restricted to primate species [200]. Expression appears to be

limited to the testis, the major site of Doppel protein expression [199]. No evidence has

been tendered concerning the existence of the predicted 94 residue protein encoded by

the open reading frame and it is currently unclear as to whether PRNT even comprises a

bona fide gene in humans.

Prion Proteins: Domain Architecture and Structure

PrPC

PrPC is synthesized with N- and C-terminal signal sequences, the former targeting

it to the endoplasmic reticulum (although not with perfect efficiency which may lead to

the generation of small amounts of cytoplasmic PrP [201]) and the secretory pathway and

the latter directing removal of a C-terminal signal peptide (followed by the addition of a

GPI anchor, which tethers PrPC to the outer leaflet of the plasma membrane) [202]. As is

25

common with GPI-anchored proteins, PrPC is found in cholesterol-rich lipid raft domains

within the membrane [203]. PrPC contains two consensus sequences for N-linked

glycosylation and un-, mono-, and di-glycosylated versions of PrPC are simultaneously

present in the cell. The structures of PrPC from various species have been determined and

are remarkably similar: the N-terminus is unstructured in solution whereas the C-terminal

domain consists of a three α-helix bundle with two short β-strands (Figure 1.6, Figure

1.7A) [204-207]. Subtle differences exist amongst structures from various species

including elk PrPC which possesses a well-defined loop between β-sheet 2 and α-helix 2,

a region which is disordered in other species [208]. A single disulfide bond is present in

the PrPC C-terminal domain linking together cysteine residues present in helices B and C.

Another feature of the cellular prion protein is an N-terminal octarepeat region consisting

of five repeats of the sequence PHGGGWGQ in the case of human PrP. The octarepeat

region is notable for several reasons. Firstly, the histidine residues are capable of binding

copper with four copper-binding sites located within the octarepeat region [209, 210] and

copper is known to induce endocytosis of PrPC [211, 212]. Secondly, expansions of the

octarepeat domain (with up to fourteen total repeats) are known to cause genetic prion

disease classified as either gCJD or GSS [213] and expression of this mutant protein in

transgenic mice causes a non-transmissible neurological syndrome [214]. The second

notable region is the hydrophobic tract prior to the beginning of the structured domain.

This sequence is well conserved amongst PrPC sequences from a variety of species.

Mutations within this sequence can increase the formation of CtmPrP (a transmembrane

topological variant of PrP in which the C-terminus is lumenal and the N-terminus is

cytoplasmic) which has been linked to neurodegeneration [215, 216]. Furthermore, as N-

terminal deletions in PrPC invade this domain, the truncated proteins become increasingly

toxic to transgenic mice [217]. The extreme N-terminus of PrP (the basically-charged

region beginning at residue 23 of mouse PrP) is important for its proper trafficking in the

cell [218, 219]. Upon suramin-induced misfolding of PrPC, constructs lacking this region

remain on the cell surface, indicating a defect in the trafficking and recycling of

misfolded proteins [220].

26

Figure 1.6. Schematic representation of the domain architecture of the prion protein family members. Both Doppel and PrPC have structured C-terminal domains consisting of 3 α-helices and 2 short β-strands as well as basically-charged N-terminal regions. Disulfide bridges are indicated above the proteins (-S-S-) and N-glycosylation sites (CHO) are denoted below the proteins. Repetitive regions are found in both PrPC and Shadoo with the former possessing octarepeats capable of binding copper and the latter possessing tetrarepeats rich in arginine and glycine residues. A well-conserved hydrophobic tract is also illustrated in PrPC and Shadoo. The locations of residues defining the N- and C-termini of ‘ΔPrP’ deletions (Δ32-121 and Δ32-134) are also noted for PrPC.

27

PrPC Doppel

NC

N

C

A

B PrPSc Model PrPSc Trimer

Figure 1.7. High resolution structures of prion proteins. A: NMR structures of murine PrP (PDB #1AG2: Riek et al. [204]) and Doppel (PDB #1I17: Mo et al. [221]) C-terminal domains. In both proteins, α-helices A, B, and C are coloured blue, red, and green, respectively. The two short β-strands are coloured magenta. The N- and C-termini of the structures are shown although it should be noted that PrPC has an additional large flexible N-terminal domain that is not resolved by NMR. The structures of PrPC and Doppel are similar, both consisting of a three α-helix bundle. However, helix B in Doppel is separated into two regions, generating a kinked structure. Images were created using MBT PDB ProteinWorkshop 1.50. B: Proposed model for the structure of PrPSc. A portion of the unstructured N-terminal domain and the first α-helical domain of PrPC are converted into a parallel left-handed β-helix in PrPSc (left) which then self-associates to form trimeric structures (right). PrPSc images taken from Govaerts et al. [222]

28

PrPSc

High resolution structural analysis of PrPSc by NMR spectroscopy or X-ray

crystallography has been hindered by the aggregated nature and general insolubility of

purified PrPSc and PrPres. At the secondary structure level it is known that PrPSc and PrPres

have an increased β-sheet content compared to PrPC [17]. Studies using antibodies have

shown that certain epitopes which are exposed in native PrPC are masked in native PrPSc

and this idea forms the basis for the CDI assay for prions [144]. In particular, residues 90

to 120 of hamster PrP are inaccessible in PrPSc suggesting that regions within the N-

terminal unstructured domain of PrPC are conformationally altered in PrPSc [223]. In

contrast, a recent study using hydrogen/deuterium backbone amide exchange found that

the β-sheet core of amyloid fibrils formed from recombinant human PrP90-231 is

composed of residues from α-helices 2 and 3 [224].

Upon treatment with proteinase K, PrPSc spontaneously forms amyloid fibrils

termed prion rods [16]. Electron microscopy on two-dimensional crystals obtained from

PrPres has provided low resolution structural data and implies the presence of parallel β-

helices [225]. Further studies on these crystals has led to a model for PrPSc in which

individual units are composed of a trimer of PrPSc containing parallel left-handed β-

helices [222, 226] (Figure 1.7B). A distinct model for the structure of PrPSc has been

obtained by molecular dynamics simulations [227]. Simulations of hamster PrP with the

D147N mutation at low pH lead to a β-sheet enriched structure which readily forms

hexameric protofibrils. Currently there is no consensus as to which model, if either, is

correct and to date there is no high resolution structural information available for PrPSc.

Doppel

The tertiary structure of Doppel is very similar to the C-terminal domain of PrPC,

even though their C-terminal domains only share approximately 25% sequence identity

and 55% homology at the amino acid level. The structures of human and mouse

recombinant Doppel have been solved by NMR and are very similar, both consisting of a

globular fold with the same secondary structural elements as PrPC [221, 228]. Notable

differences to PrPC include a disruption in helix B separating it into B and B’ regions and

generating a ‘kinked’ helix, a different orientation of the two β-strands, and an additional

outer disulfide bond (Figure 1.6, Figure 1.7A). Like PrPC, Doppel migrates through the

29

secretory pathway where its N-terminal signal sequence is removed, its C-terminal signal

sequence replaced with a GPI anchor, and it is modified by the addition of N-linked

sugars at asparagine residues 98 and 110 (human gene residue numbering) and possibly

O-linked carbohydrate moieties at threonine residue 43, although the latter may be

restricted to human Doppel as Thr43 is not conserved in mouse [196, 229]. Like PrPC,

Doppel localizes to lipid rafts [230]. The extreme N-terminus of Doppel is flexibly

disordered and positively charged like that of PrPC and likely contributes to its proper

trafficking. Fusion of the Doppel N-terminus to PrP lacking its N-terminus and

octarepeats restores wild-type trafficking properties [220]. In short, Doppel resembles the

C-terminal domain of PrPC but lacks the N-terminal hydrophobic and octarepeat regions

(Figure 1.6).

Shadoo

Murine Shadoo is predicted to be a 98 residue protein with N- and C-terminal

signal sequences mirroring those in PrPC and Doppel (Figure 1.6) [197]. N-glycosylation

is also conserved, but is only predicted to occur at a single site. The overall architecture

of Shadoo is reminiscent of the N-terminal half of PrPC and thus stands in stark contrast

to Doppel, which shares homology to the PrPC C-terminal domain. Although Shadoo

lacks the octarepeat sequences found in PrPC and lacks histidine residues capable of

coordinating copper ions, it does have a series of N-terminal charged tetrarepeats rich in

glycine, serine, alanine and arginine. The bulk of the homology between Shadoo and

PrPC is found within the hydrophobic tract (Figure 1.8). Notably, this is one of the most

conserved regions amongst PrP sequences from diverse organisms, perhaps implying

functional homology between Shadoo and PrPC. Shadoo appears to lack any discernible

elements of secondary structure (at least by bioinformatic analyses), in concordance with

its homology to the unstructured N-terminus of PrPC. The Shadoo region analagous to the

α-helical domain present in PrPC and Doppel is shorter, and is not long enough to

accommodate three α-helices. Additionally, Shadoo is devoid of cysteine residues,

preventing the formation of stabilizing disulfide bridges. Thus, the C-terminal domain of

Shadoo is unique within the prion protein family and is likely to be unstructured. Thus far,

the only characterization the hypothetical Shadoo protein has come from a study which

demonstrates that an epitope-tagged cDNA encoding the zebrafish Sho-2 protein

30

Human_PrP KKRPKPG-GWNTGGSRYPGQGSPGGNRYPPQGGGGWGQPHGGGWGQPHGGGWGQPH--------GGGWGQPHGGG-WGQ-GGGTHSQWNKP---SKPKTNMKHMAGAAAAG--AVVGGLGChimpanzee_PrP KKRPKPG-GWNTGGSRYPGQGSPGGNRYPPQGGGGWGQPHGGGWGQPHGGGWGQPH--------GGGWGQPHGGG-WGQ-GGGTHSQWNKP---SKPKTNMKHMAGAAAAG--AVVGGLGCow_PrP KKRPKPGGGWNTGGSRYPGQGSPGGNRYPPQGGGGWGQPHGGGWGQPHGGGWGQPHGGGWGQPHGGGWGQPHGGGGWGQ-GG-THGQWNKP---SKPKTNMKHVAGAAAAG--AVVGGLGSheep_PrP KKRPKPGGGWNTGGSRYPGQGSPGGNRYPPQGGGGWGQPHGGGWGQPHGGGWGQPH--------GGGWGQPHGGGGWGQ-GG-SHSQWNKP---SKPKTNMKHVAGAAAAG--AVVGGLGMouse_PrP KKRPKPG-GWNTGGSRYPGQGSPGGNRYPPQ-GGTWGQPHGGGWGQPHGGSWGQPH--------GGSWGQPHGGG-WGQ-GGGTHNQWNKP---SKPKTNLKHVAGAAAAG--AVVGGLGRat_PrP KKRPKPG-GWNTGGSRYPGQGSPGGNRYPPQSGGTWGQPHGGGWGQPHGGGWGQPH--------GGGWGQPHGGG-WSQ-GGGTHNQWNKP---SKPKTNLKHVAGAAAAG--AVVGGLGRabbit_PrP KKRPKPGGGWNTGGSRYPGQSSPGGNRYPPQ-GGGWGQPHGGGWGQPHGGGWGQPH--------GGGWGQPHGGG-WGQ-GG-THNQWGKP---SKPKTSMKHVAGAAAAG--AVVGGLGHuman_Sho -------------------K-------------------------------------GGR----GGARGSARGGV-RGGARGASRVRVRPAQRYGAPGSSLRVAAAGAAAG--AAAGAAAChimpanzee_Sho -------------------K-------------------------------------GGR----GGARGSARGGV-RGGARGASRVRVRPAQRYGAPGSSLRVAAAGAAAG--AAAGAAACow_Sho -------------------K-------------------------------------GGR----GGARGSARG------GRGAARVRVRPAPRY---AGSSMRVAAGAAAG--AAAGAAASheep_Sho -------------------K-------------------------------------GGR----GGARGSARG------GRGAARVRVRPAPRY---AGSSVRAAAGAAAGAAAAAGVAAMouse_Sho -------------------K-------------------------------------GGR----GGARGSARG-V-RGGARGASRVRVRPAPRY---GSSLRVAAAGAAAG--AAAGVAARat_Sho -------------------K-------------------------------------GGR----GGARGSARG-V-RGGARGASRVRVRPAPRY---SSSLRVAAAGAAAG--AAAGVAARabbit_Sho -------------------K-------------------------------------GGR----GGARGSARGGI-RGGARGTSRVRVRPAPRY---GSSPRVAAAGAAAG--AAAGAAA : **. *..:* * :: : . . *..**** *..* . Human_PrP GYMLGSAMSRPIIHFGSDYEDRYYRENMHRYPNQVYYRPMDEYSNQNNFVHDCVNITIKQHTVTTTTKGENFTETDVKMMERVVEQMCITQYERESQAYYQRG-S Chimpanzee_PrP GYMLGSAMSRPIIHFGSDYEDRYYRENMHRYPNQVYYRPMDQYSSQNNFVHDCVNITIKQHTVTTTTKGENFTETDVKMMERVVEQMCITQYERESQAYYQRG-S Cow_PrP GYMLGSAMSRPLIHFGSDYEDRYYRENMHRYPNQVYYRPVDQYSNQNNFVHDCVNITVKEHTVTTTTKGENFTETDIKMMERVVEQMCITQYQRESQAYYQRGAS Sheep_PrP GYMLGSAMSRPLIHFGNDYEDRYYRENMYRYPNQVYYRPVDRYSNQNNFVHDCVNITVKQHTVTTTTKGENFTETDIKIMERVVEQMCITQYQRESQAYYQRGAS Mouse_PrP GYMLGSAMSRPMIHFGNDWEDRYYRENMYRYPNQVYYRPVDQYSNQNNFVHDCVNITIKQHTVTTTTKGENFTETDVKMMERVVEQMCVTQYQKESQAYYDGRRS Rat_PrP GYMLGSAMSRPMLHFGNDWEDRYYRENMYRYPNQVYYRPVDQYSNQNNFVHDCVNITIKQHTVTTTTKGENFTETDVKMMERVVEQMCVTQYQKESQAYYDGRRS Rabbit_PrP GYMLGSAMSRPLIHFGNDYEDRYYRENMYRYPNQVYYRPVDQYSNQNSFVHDCVNITVKQHTVTTTTKGENFTETDIKIMERVVEQMCITQYQQESQAAYQRAAG Human_Sho GLAAGSGWRRAAGPGERGLEDE-----------------------------------------EDGVPGGNGTGPGI-------------------YSYRAWTSG Chimpanzee_Sho GLAAGSGWRRAAGPGERGLEDE-----------------------------------------EDGVPGGNGTGPGI-------------------YSYRAWTSG Cow_Sho GLAAGSSWRRAAGPAELGPEDA-----------------------------------------EDGAPGSNGTGRGV-------------------YSYWAWTSG Sheep_Sho GLAAGSSWRRAAGPAELGLEDA-----------------------------------------EDGAPGSNGTGRGV-------------------YSYWAWTSG Mouse_Sho GLATGSGWRRTSGPGELGLEDD-----------------------------------------ENGAMGGNGTDRGV-------------------YSYWAWTSG Rat_Sho GLATGSGWRRTSGPGELGLEDD-----------------------------------------ENGAMGGNGTDRGV-------------------YSYWAWTSG Rabbit_Sho GLAAGPGWRRAAGPGERG-PDE-----------------------------------------EDLASGGNGT--GV-------------------YSYWTWTSG * *.. *. . * . * * * .: : .

Figure 1.8. Alignment of PrP and Sho protein sequences from a variety of species. A multiple sequence alignment of PrP and Sho protein sequences (with N- and C-terminal signal peptides removed) from a variety of mammalian species performed using the T-COFFEE algorithm is shown. The principle region of similarity between PrP and Sho sequences falls within the alanine/glycine/valine-rich hydrophobic tract (highlighted in yellow).

31

generates an N-glycosylated, GPI-anchored cell-surface protein when expressed in

mammalian cells [231].

Endoproteolysis and Membrane Topology of PrPC

A proportion of mature PrPC undergoes endoproteolysis at the cell surface.

Human PrPC is cleaved between residues 110 and 111 to generate a C-terminal fragment

termed C1 and an N-terminal fragment (N1) which is presumably released into the

parenchyma of the brain [232]. C1 cleavage is thought to be mediated by the

metalloproteases ADAM10 and ADAM17 (also known as TACE) as inhibition of these

proteases in N2a cells results in decreased levels of the N1 fragment in conditioned

medium [233]. ADAM9 may also be indirectly involved in the C1 cleavage event via

regulation of ADAM10 activity [234]. A second cleavage event in PrPC occurs in the

vicinity of residue 88 to generate a C-terminal fragment termed C2 which is similar in

molecular weight to PrPres [232]. C2 cleavage appears to be mediated by calpains, which

are Ca2+-activated cysteine proteases [235]. It is not known what role PrP endoproteolysis

plays in regulating its in vivo physiology. However, C1 cleavage does occur within the

middle of residues 106-126 of human PrP, a fragment which has been shown to be toxic

to neurons [236]. Additionally, the C1 and C2 fragments have differential effects on p53-

dependent staurosporine-induced caspase-3 activation in HEK293 cells [237].

During the biogenesis of PrPC in the endoplasmic reticulum, several different

topological variants are generated. The predominant form of PrPC is fully translocated

across the membrane so that the entire protein is on the lumenal side of the membrane. In

contrast, two topological variants exist in which incomplete translocation occurs [238-

240]. In CtmPrP, the C-terminus of the molecule is lumenal and the N-terminus is

cytoplasmic. In NtmPrP, the N-terminus is lumenal and the C-terminus is cytoplasmic. In

both variants, the hydrophobic tract of PrP (consisting approximately of residues 112 to

131 of human PrP) spans the membrane. It is thought that the CtmPrP variant may be

relevant to prion disease as transgenic mice expressing PrP alleles with mutations that

increase the amount of CtmPrP have a neurodegenerative phenotype [215]. Additionally,

the GSS-causing A117V mutation in PRNP also increases the relative amount of CtmPrP.

32

Because Shadoo possesses a hydrophobic tract similar to PrP, it is possible that CtmSho

and NtmSho topological variants also exist.

1.3 Cellular Functions of Prion Proteins

Proposed Functions for PrPC

Although the lack of pronounced phenotypes in Prnp0/0 mice has hindered

deciphering the function of PrPC [186, 241], several more subtle phenotypes have

provided clues towards its physiological role. Unfortunately, repetition of some of these

phenotypes has proven challenging suggesting the possibility of confounding variables

such as differences in genetic background. The first phenotype reported in Prnp0/0 mice

(mixed C57BL6/129/Sv background) was a defect in synaptic function in the form of

impaired GABA receptor-mediated fast inhibition and long term potentiation in

hippocampal CA1 pyramidal neurons [242]. These synaptic defects were later shown to

be rescued by introduction of a wild-type human PrP over-expressing transgene but not a

transgene with only moderate PrP expression [243]. In contrast, another group was

unable to observe any defects in synaptic transmission in the CA1 region of the

hippocampus of Prnp0/0 mice with either a mixed C57BL6/129/Sv or an FVB background

[244] and no impairment in Purkinje cell synaptic transmission was found in Prnp0/0 mice

(mixed C57BL6/129/Sv background) [245]. Other changes which have been observed in

the hippocampus of Prnp0/0 mice (mixed C57BL6/129/Sv background) include disrupted

Ca2+-activated K+ currents and reorganization of mossy fibres [246, 247].

Electrophysiological defects (reductions in after hyperpolarization potentials in CA1

neurons) have also been observed in mice (FVB background) where the Prnp gene has

been ablated post-natally [248]. Another phenotype which has been noted in Prnp0/0 mice

on the mixed C57BL6/129/Sv background is an alteration in circadian activity rhythms

and sleep patterns [249]. Notably, this is true for two independent knockout strains (with

different gene-targeting strategies) and suggests that the behavioural alterations in the

human prion disease fatal familial insomnia (and perhaps sleep-wake disturbances in

sporadic CJD [250]) may be due in part to a loss of PrPC function. Surprisingly, this

phenomenon remains relatively unexplored although one group has found that PrP

mRNA levels in the rat forebrain are regulated in a circadian manner [251].

33

A recent publication has documented a learning impairment in Prnp0/0 mice with

defects in hippocampal-dependent spatial learning and reduction in long-term

potentiation in the dentate gyrus [252]. These defects could be rescued by introduction of

a wild-type PrP transgene under the control of the neuron-specific enolase promoter and

were independent of the genetic background of the mice. Impairments to long- and short-

term memory retention have also been observed in 9 month old (but not 3 month old)

Prnp0/0 mice [253, 254]. In contrast, other studies have failed to find any overt changes to

learning or behaviour in Prnp0/0 mice [254, 255]. Thus, the role of PrPC in learning and

behaviour in mice remains to be clarified.

In part because PrPC is capable of binding copper in vivo [210], it has been

proposed that PrPC may function as a superoxide dismutase (SOD). Cu2+-bound SOD is

involved in the detoxification of damaging superoxide radicals by converting them to

oxygen and hydrogen peroxide. Decreased Cu/Zn SOD activity has been observed in

brain homogenates prepared from Prnp0/0 mice [256] or Prnp+/+ brain homogenates

which have been immuno-depleted of PrPC [257], and may be caused by an ability of

PrPC to regulate the incorporation of Cu2+ into SOD [258]. Alternatively, PrPC may

possess some intrinsic SOD activity [259], although another group has failed to find any

SOD-1 activity in recombinant PrP [260]. The ability of PrPC to regulate SOD activity

remains contentious as multiple groups have reported that changes in Prnp gene dosage

do not affect SOD1 or SOD2 activity levels in vivo [261, 262].

Two recent papers have begun to unravel subtle roles for PrPC in the development

of the nervous and hematopoietic systems. One study has focused on the effect of PrPC

expression levels on the developing nervous system [263]. The authors find that PrPC

increases neuronal differentiation in the dentate gyrus in vitro (i.e. the lowest amount of

differentiation is observed in Prnp0/0 mice) and increases cellular proliferation in the

subventricular zone in vivo. However, adult mice, regardless of PrP genotype, have the

same number of mature neurons suggesting a minor role, if any, of PrPC in the

development of the nervous system. In the second study the authors show that

hematopoietic stem cells isolated from the bone marrow of Prnp0/0 mice are impaired in

their ability to self-renew following reconstitution of irradiated mice and that this defect

can be repaired by introduction of PrPC into cells by retroviral infection [264]. These

34

results point to a role for PrPC in the long-term maintenance of hematopoietic stem cells

but fail to enlighten as to why the predominant site of PrPC expression is in the adult CNS.

A recently-published study has proposed that PrPC is involved in the regulation of

cleavage of the amyloid precursor protein (APP) by β-secretase (BACE1) [265]. BACE1

is responsible for one of the two proteolytic cleavage events that liberates the β-amyloid

(Aβ) peptide from APP. Increased levels of Aβ are associated with Alzheimer’s disease

pathology. The authors find that over-expression of PrPC leads to a decrease in the

BACE1-mediated cleavage of APP and conversely that β-secretase cleavage and Aβ

secretion increases with down-regulation of PrP. These results suggest that there may be

interplay between disease mechanisms in prion disease and Alzheimer’s disease. In

support of these ideas, PrPC has been found to reside in spatial proximity to APP within

the cell membrane [266, 267].

Of the several proposed cellular functions for PrPC, none have been described in

molecular terms. One broadly accepted observation is a role for PrPC in neuroprotection

against pro-apoptotic stimuli. For example, PrPC is capable of protecting cerebellar cells

from Doppel-induced neurodegeneration and engagement of PrPC through the use of a

PrPC-binding peptide is capable of preventing anisomycin-induced death in rodent retinal

explants [268]. Additionally, PrPC over-expression in human neurons appears to

counteract the toxic actions of Bax expression [269]. The protective activity of PrPC is

dependent on the presence of the octarepeat region but interestingly is not lost when PrPC

lacking its GPI anchor (i.e. secreted PrPC) is used. This phenomenon also has a parallel in

yeast—mammalian PrPC specifically engineered to enter the yeast secretory pathway is

capable of blocking Bax toxicity [270]. In this paradigm, however, the octarepeats are not

absolutely required and a deletion of residues 23-31 blunts protective ability (a similar

deletion in PrPC also abrogates its ability to protect against Doppel toxicity and is likely

due to a deletion-induced defect in PrP-trafficking [271]). In both paradigms, cell-surface

localization of PrPC (or at least proper transport through the secretory pathway) is

required for protection. A molecular interpretation of this activity remains unclear,

although a signaling event from the extracellular milieu may be involved since cell-

surface PrPC is capable of blocking a cytotoxic event occurring in the cytoplasm. Another

example of the neuroprotective activity of PrPC is found in its influence on stroke biology.

35

PrPC expression is up-regulated following ischemic brain damage, in both humans and

mice [272, 273]. Furthermore, PrPC deficiency in mice increases infarct size following

cerebral artery occlusion and increases caspase-3 activation [274], and PrPC over-

expression improves neurological behaviour and reduces infarct volume in a rat stroke

model [275]. Paradoxically, pro-apoptotic activities of PrPC have also been observed.

Antibody-mediated crosslinking of PrPC in vivo causes death of both hippocampal and

cerebellar cells although this pro-apoptotic activity may be limited to dimerized PrPC

[276]. Additionally, it has been shown that PrPC over-expression in HEK293 cells can

enhance staurosporine-induced toxicity and activation of caspase-3 [277] and that over-

expression in TSM1 cells can sensitize neurons to apoptotic stimuli via regulation of p53-

dependent caspase-3 activation [278]. A further paradox concerns the central region of

PrP which is toxic (pro-apoptotic) when made as an aggregation-prone peptide [236], yet

when deleted from the centre of wild-type PrP creates a toxic gain-of-function transgene

[279, 280].

Cellular Function of Doppel

In contrast to PrPC, more definitive clues concerning the function of Doppel have

been uncovered. Although Doppel is present at low levels in the brain during embryonic

development and neonatal life, expression in adults is most evident in the testis (and to a

lesser extent in the heart) [190, 281]. Doppel immunoreactivity is present in both somatic

and germ cells of the male reproductive tract including ejaculated spermatozoa, seminal

plasma, seminiferous tubules, Sertoli cells, and the acrosome, with possible species-

specific variations in levels of expression and localization [196, 282, 283]. Therefore, it is

not surprising that Doppel knockout mice suffer from a reproductive defect. Males from

two published lines of Doppel knockout (Prnd0/0) mice are sterile, suggesting that Doppel

plays a role in the normal functioning of the male reproductive system [284, 285]. Sperm

from both lines appear to be incapable of undergoing a proper acrosome reaction (the

enzyme-releasing reaction which occurs at the tip of a sperm which permits fusing of an

egg and sperm), either preventing or impeding fertilization in vitro. A mechanism-based

interpretation of these events has not yet been described. It seems unlikely that Doppel

and PrPC have functional overlap since no reproductive defects have been observed in

Prnp0/0 mice. Perhaps the CNS-specific role(s) of PrPC are dependent on the domains not

36

present in Doppel (i.e. the octarepeats and the hydrophobic tract). Interestingly, a

truncated Doppel-like form of PrPC is the most prominent PrP species in the testes [196]

suggesting that truncated PrPC and Doppel may have overlaps in function within the testis.

In contrast to Prnd0/0 mice, transgenic mice (accidentally) over-expressing ovine Doppel

under the control of its endogenous promoter did not exhibit any alterations in

reproductive behaviour [286].

No defects in the development or maintenance of the central nervous system or

other tissues were observed in either Prnd0/0 or Prnp0/0/Prnd0/0 double knockout mice

implying that like PrPC, Doppel is dispensable in the brain and that the lack of a

phenotype in Prnp0/0 mice is not a result of functional compensation by Doppel [285].

Two studies have shown that, like PrPC, Doppel is capable of binding copper [287,

288]. This likely involves histidine residue 131 of the mouse sequence that is within the

αB/B’-loop-αC region which stands in marked contrast to the octarepeat copper binding

sites in PrPC [287]. The physiological implications of copper binding may be different for

the two proteins as PrPC can bind more equivalents of copper and therefore may be more

suited as a copper scavenger. Additionally, in contrast to PrPC, no internalization of

Doppel is observed following copper binding [288]. One study has reported that

detergent-extracted Doppel from mouse testis does not bind to an immobilized metal

affinity column loaded with copper leading to the suggestion that Doppel is incapable of

binding copper or that it is perhaps already metallated [230].

The necessity of PrPC expression for the occurrence of prion disease and the

propagation of prions has been well-documented [107, 108]. In contrast, several lines of

evidence suggest that Doppel is neither required for prion pathogenesis nor is capable of

modulating the disease process. Transgenic mice expressing Doppel in the CNS on a

wild-type PrP background have unaltered disease incubation times and pathological

lesion profiles upon challenge with infectious prions [289]. Secondly, no differences in

either disease incubation time, vacuolation profiles, patterns of neuronal loss, or PrP

deposition were observed between Doppel over-expressing and Doppel non-expressing

Prnp0/0 mice crossed with mice expressing PrP when inoculated with prions [290]. In

humans, Doppel expression is unchanged in patients with sporadic CJD and no changes

in murine Doppel expression between uninfected and infected N2a cells have been found

37

[281]. Furthermore, expression of Doppel in infected cells does not modify toxicity, and

there are as yet no indications of abnormally folded forms of the protein. One hypothesis

as to why Doppel is unable to adopt a β-sheet enriched structure is that its second

disulfide bridge imparts added stability, making the transition thermodynamically

unfavourable. Whyte et al looked into this possibility by creating a Doppel mutant with a

single disulfide bridge, analogous to the structure of PrPC [291]. This mutant was indeed

less stable, but the stability of wild-type Doppel was less than wild-type PrPC, suggesting

that thermodynamic instability alone is not sufficient to explain the inability of Doppel to

undergo a conformational transition. Furthermore, no evidence of an α to β transition

(even if β-rich recombinant PrP was present) was observed for either wild-type or single

disulfide-mutant Doppel [291]. In a yeast two-hybrid setting, Doppel is incapable of

interacting with itself [292], possibly indicating an inherent inability to undergo

dimerization or multimerization steps that are associated with conformational conversion

and templated refolding. However, this conclusion is tempered by the caveat that fusion

proteins expressed in the reducing environment of the yeast cytosol do not acquire the

disulfide linkages and glycosyl modifications found in the mature mammalian proteins.

When added to the observation that Doppel does not accumulate to appreciable amounts

in the brain of adults, these results suggest that Doppel is not significantly involved in

prion disease.

1.4 Prion Protein Ligands

PrPC-Interacting Proteins

A common route to inferring the function of a protein is to uncover its binding

partners and then tap into the known pathways and functional properties of the identified

proteins. There is no shortage of identified candidate PrPC-interacting proteins (a list of

over thirty identified PrPC ligands is given in Table 1.2)—the main problem has been the

inability of different labs to uncover the same ligands and to prove functional relevance

of the binding proteins using in vivo assays. It seems likely that some of the putative

binding partners only bind to PrPC in either in vitro systems or as a result of protein over-

expression since differences in compartmentalization should occlude binding of the two

proteins within a cell under normal conditions (i.e. Bcl-2 and Grb-2, both of which are

38

Table 1.2. Proteins or molecules which have been reported to interact with PrPC or PrPSc

PrPC-Interacting Protein

Sub-Cellular Localization Method of Discovery

Binding Epitope on

PrPC Reference(s)

GFAP Cytoplasmic (Cytoskeleton)

Ligand blots Unknown [293]

Bcl-2 The cytoplasmic face of organelles

Yeast 2-hybrid Residues 72-254

[294]

Grb2 Cytoplasmic Yeast 2-hybrid N-terminal and C-terminal

binding sites

[295]

Synapsin-1b Synapse Yeast 2-hybrid N-terminal and C-terminal

binding sites

[295]

Pint1 Unknown Yeast 2-hybrid Residues 90-231

[295]

Glycosaminoglycans (i.e. heparin)

Cell membrane, extracellular

matrix

Heparin-agarose pull-down on PI-PLC-treated

cells

Residues 23-52, 53-93, and 110-128

[296, 297]

Caveolin Caveolae (cell membrane)

Co-purification in N2a cells

Unknown [298]

Dystroglycan Cell membrane (transmembrane)

Co-immunoprecipitation

Unknown [299]

Synaptophysin Synaptic vesicles (transmembrane)


Unknown [299]

Neuronal nitric oxide synthase

(nNOS)

Peripheral membrane protein


Unknown [299]

ApoE* Secreted Co-immunoprecipitation/

pull-downs with recombinant proteins

Residues 23-90

[300]

Plasminogen Secreted Immobilized serum proteins probed with

PrPC- or PrPres-containing material

Binds to both PrPC and PrPres

(PrP27-30)

[301, 302]

ER Chaperones (calnexin,

calreticulin, PDI*, BiP*, grp94)

Endoplasmic reticulum

Two sets of immunoprecipitations on radio-labeled cells

Unknown [303]

39

Table 1.2 (Continued)



Binding Epitope on

PrPC Reference(s)

Hsp60 Mitochondria Yeast 2-hybrid Residues 180-210

[304]

NRAGE Cytoplasm and peripheral

membrane protein


[305]

TREK-1 Cell membrane (transmembrane)

Bacterial 2-hybrid Residues 128-230

[306]

Rdj2 Cytoplasmic face of membranes

In vitro pull-downs with GST fusions

Unknown [307]

Tetraspanin-7 Cell membrane (transmembrane)


[308]

Selectins Cell membrane (transmembrane)

PrP-Ig fusions and co-immunoprecipitation

Unknown [309]

Vitronectin Secreted (extracellular

matrix)

Overlay assays and co-immunoprecipitation

Residues 105-119

[310]

Tubulin Microtubules Chemical crosslinking Unknown [311]

αB-crystallin Cytoplasm Yeast 2-hybrid Unknown [312]

ZAP-70 Cytoplasm Co-immunoprecipitation

Unknown [313]

Fyn Cytoplasmic face of membranes


Unknown [313]

Casein Kinase 2 α/α’ subunits

Cytoplasmic Far-Western blots and plasmon resonance using recombinant

proteins

Residues 105-242

[314]

Nrf2 Cytoplasm/ nucleus

Alkaline phosphatase-PrPC fusion used to

screen a mouse brain cDNA library

Unknown [315]

APLP1 Cell membrane (transmembrane)

Alkaline phosphatase-PrPC fusion used to

screen a mouse brain cDNA library

Unknown [315]

40

Table 1.2 (Continued)



Binding Epitope on

PrPC Reference(s)

β-secretase (BACE1)

Cell membrane (transmembrane)


N-terminus (residues 23-28)

[265]

RNA aptamers N/A GST-PrPC fusion and library of RNA

sequences

N-terminus (residues 23-52)

[316]

Laminin Secreted GST-PrPC and laminin co-purification

Unknown [317]

α2/β2 Na+/K+-ATPase*

Cell membrane (transmembrane)


Unknown [318]

37-kDa/67-kDa laminin receptor

Cell membrane Yeast 2-hybrid Residues 144-179 (direct)

and residues 53-93 (HSPG-

dependent)

[319, 320]

Stress-inducible protein 1 (STI1)

Cytoplasm Complementary hydropathy

Residues 113-128

[321]

N-CAM* Cell membrane Mild formaldehyde crosslinking in N2a

cells

Residues 141-176

[322]

‘Protein X’ (hypothetical

protein)

Unknown Inferred from prion inoculations in transgenic mice

expressing chimeric human/mouse PrPC.

Discontinuous epitope

comprising residues

167/171 and 214/218

[110, 111]

*PrPC-interacting proteins which were also found in an in vivo study employing time-controlled transcardiac perfusion crosslinking of wild-type mice brains [266].

41

cytoplasmic proteins) [294, 295]. That being said, there has been considerable debate

surrounding the toxicity of PrP, either targeted directly to the cytoplasm or for molecules

which have undergone retrotranslocation from the ER into the cytoplasm. Some labs have

found that cytoplasmic PrP is toxic [88, 323] whereas others have failed to confirm this

finding [90, 91, 324]. A recent paper has suggested that under conditions of proteasomal

inhibition, cytoplasmic PrP co-aggregates with Bcl-2, a process which correlates with

toxicity [323]. Thus Bcl-2 may be an authentic ligand for cytosolic PrP. Other ligands

may be veritable PrPC-interacting proteins but have not been demonstrated to modulate

either the biochemistry of PrPC or disease pathogenesis involving PrPSc. Two putative

ligands which have received much attention are the 37-kDa/67-kDa laminin receptor

[319] and the neural cell adhesion molecule (N-CAM) [322]. The laminin receptor has

been reported to act as a cellular receptor for PrPC and for PrPSc and a requisite for PrPSc

propagation in infected neuronal cell lines [112, 325, 326]. However, the in vivo

relevance of the laminin receptor to prion biology and scrapie pathogenesis remains

obscure but the existence of laminin receptor ‘knock-down’ mice [327] may help to

clarify this issue. One possible role of the laminin receptor is as an intestinal receptor for

ingested prions [328]. Indeed, blocking the laminin receptor with antibodies prevented

the uptake of bovine prions by enterocytes [329]. The interaction between PrPC and N-

CAM was uncovered in N2a neuroblastoma cells without protein overexpression using

mild formaldehyde crosslinking, a technique which can lock together protein complexes

which may only be transiently associated [322]. This interaction has been confirmed

using in vivo crosslinking in the brains of wild-type mice [266] and recently, the N-CAM/

PrPC interaction has been shown to be physiologically relevant (although not to disease

pathogenesis since N-CAM knockout mice have unaltered prion incubation times [322]).

Santuccione et al. have shown that PrPC can recruit N-CAM to lipid rafts where it

participates in the activation of Fyn kinase [330]. Interestingly, signal transduction

through PrPC and involving Fyn kinase has previously been documented [331], although

genetic ablation of Fyn in mice has no significant effect on prion disease pathogenesis

[332]. Furthermore, interactions between PrPC and N-CAM are involved in neurite

outgrowth. Thus, N-CAM may represent one of the more physiologically plausible PrPC-

interacting proteins identified to date. Another noteworthy putative PrPC-interacting

42

protein is stress-inducible protein 1 (STI1) [321]. Recombinant STI1 and a peptide

comprising the PrP-binding site on STI1 are both capable of decreasing cell death in

retinal explants following treatment with the protein synthesis inhibitor anisomycin.

STI1/PrPC-dependent neuritogenesis and neuroprotection were dependent on mitogen-

activated protein kinase and protein kinase A pathways, respectively [333]. Another

study has found that the STI1-PrPC interaction is involved in regulating superoxide

dismutase activity [334]. One final interacting protein of note is plasminogen, a zymogen

precursor to a serine protease present in serum. Plasminogen was first reported to interact

specifically with disease-associated forms of PrP (PrPres and PrPSc) [301] although

subsequent studies have shown that it is also capable of interacting with PrPC [335, 336].

This interaction is likely mediated by the kringle domains present in plasminogen [337].

Binding of PrPC to plasminogen stimulates both tPa-mediated activation of plasminogen

[302, 338] and cleavage of PrPC [339]. In vivo studies have shown that plasminogen

deficiency has no major effect on disease progression or survival in prion-inoculated

mice [340] and does not alter the endoproteolytic processing of PrPC [341].

In an attempt to accurately define the PrPC interactome in the brain, Schmitt-Ulms

and co-workers developed a novel technique (termed time-controlled transcardiac

perfusion crosslinking (tcTPC)), which uses mild formaldehyde crosslinking to lock

together protein complexes followed by immunoaffinity purification of the target protein

and the identification of binding partners by mass spectrometry [266]. The validity of this

technique was verified by isolating other known members of the γ-secretase complex

(which is of relatively low abundance) using one component of the complex as the

subject for immunoaffinity chromatography. Further applications of this technique have

uncovered two novel proteins, TMP21 and LINGO-1, which have an impact upon

Alzheimer’s disease-related endpoints in functional assays, demonstrating the utility of

this system [267, 342]. Using this same approach for PrPC, several candidate interacting

proteins were uncovered, including various splicing isoforms of N-CAM and the amyloid

precursor protein (APP). Some of the identified proteins may have been isolated solely

because of their proximity to PrPC in the membrane and their abundance, and may not

actually interact with PrPC in a phenotypically significant manner. However, further

43

application of this technology may ultimately reveal proteins which provide insight into

PrPC (and PrPSc) physiology.

Doppel-Interacting Proteins

Only one ligand for Doppel has surfaced thus far (although there are contradictory

reports concerning the ability of Doppel to bind to the 37-kDa/67-kDa laminin receptor

[292, 343]). Using a yeast 2-hybrid approach, the receptor for activated C-kinase RACK1

was identified as a Doppel-interacting protein but the functional significance of this

interaction remains to be deciphered [344]. Unlike PrPC, Doppel is unable to bind to

either Grb-2 or GFAP in a two-hybrid assay [345]. In addition, a large-scale yeast two-

hybrid screen failed to find a Doppel-interacting protein from a human ORF cDNA

library [346]. In an interesting report, fusions of either Doppel or PrPC to the Fc portion

of IgG bind to the granule cell layer of the cerebellum, suggesting the existence of a

cerebellar protein which may mediate Doppel toxicity [347]. The discovery of a common

ligand for PrPC and Doppel would be immediately useful for dissecting the pathways

involved in genotypic interactions between Prnp and Prnd.

1.5 Functional Interactions Between Members of the Prion Protein Family

Doppel and PrP in Knockout and Transgenic Mice

Other than its aforementioned role in the male reproductive system, the majority

of interest in studying the Doppel protein has stemmed from its neurotoxic properties

when present in the brains of Prnp0/0 mice [189, 190, 289]. This effect is either

completely blocked or significantly reduced upon co-expression of PrPC, implying either

a direct interaction between the two proteins or the existence of shared binding partners.

Understanding the biology of this paradigm is important for other areas in prion research

since 1) The ability of PrPC to block the pro-apoptotic activity of Doppel hints at its

cellular function and detailed studies using this paradigm may help to elucidate the

enigmatic function of PrPC and 2) The pro-apoptotic activity of Doppel may partly mimic

the action of PrPSc and thus may be useful for deciphering the molecular events

underlying prion disease pathogenesis.

44

Doppel in Prnp0/0 and Transgenic Mice

The discovery of the Prnd gene and the Doppel protein was hastened by a need to

explain phenotypic discrepancies between various strains of Prnp0/0 mice. A summary of

the properties of all strains of Prnp0/0 mice (ZrchI, NPU, Ngsk, Rcm0, ZrchII, and Rikn) is

presented in Table 1.3. These data have been discussed at length [188-190, 348] leading

to the conclusion that ectopic Doppel expression rather than loss of PrPC is the underlying

cause of phenotypic variation (presence or absence of a cerebellar ataxia syndrome).

Indeed, an increase in Doppel protein expression can be seen in the brains of Rcm0

Prnp0/0 mice [229] and immunohistochemical analyses of the brains of Ngsk and Rcm0

Prnp0/0 mice have demonstrated Doppel expression in the cerebellum [289]. An elegant

proof of this concept was achieved by meiotic deletion of the Prnd gene in ZrchII Prnp0/0

mice which are ataxic and have Purkinje cell loss [349]. No pathology was observed in

these mice, confirming that the phenotypic defects in ataxic strains of Prnp0/0 mice result

from expression of Doppel and not the loss of PrPC function or the cis-activation of genes

adjacent to the Prn gene complex. Interestingly, cerebellar Doppel in Ngsk Prnp0/0 mice

was associated with a distinct molecular signature on Western blots but further tests using

independent antibodies and other strains of mice are required to validate this observation

[350].

ZrchII Prnp0/0 mice have an ataxic phenotype with loss of Purkinje cells and the

age of onset is inversely related to the level of ectopic Doppel expression. Hybrid

ZrchI/ZrchII mice (with only a single copy of the Doppel-producing ZrchII allele)

develop disease at a later age whereas mice with two copies of the ZrchII allele and also

expressing a cosmid containing the Prnd and Prnp genes, but in which the PrP open

reading frame has been replaced by the tetracycline transactivator, express higher levels

of Doppel and develop disease at an earlier age [189]. The ataxic phenotype and Purkinje

cell degeneration could be rescued in the progeny of crosses with tga20 mice (over-

expressing PrPC), even though the transgene promoter elements in tga20 mice do not

drive PrPC expression in Purkinje cells. These data prompt two possible interpretations:

1) PrPC is able to counteract in trans a Doppel-dependent molecular event occurring in

Purkinje cells (ie by virtue of PrPC expression in CGNs) or 2) The critical event occurs in

CGNs expressing Doppel and the consequent dysfunction (but not necessarily

45

Table 1.3. Characteristics of strains of Prnp0/0 Mice

Strain Knockout Strategy Deletion of Prnp

Exon 3 Splice Acceptor Site?

Late-Onset

Ataxia?

Up-Regulation of Doppel in the

Brain?

Reference(s)

ZrchI Replacement of PrP residues 4-187 with a

neo cassette

No No No [186]

NPU Insertion of a neo cassette following residue 93 of PrP

No No No [187]

Ngsk Replacement of a region containing 0.9

kb of Prnp intron 2, the entire ORF, and 0.45

kb of the 3’ UTR with a neo cassette

Yes Yes Yes [188]

Rcm0 Replacement of a region containing 0.9

kb of Prnp intron 2, the entire ORF, and 0.45

kb of the 3’ UTR with an HPRT cassette

Yes Yes Yes [190]

ZrchII Replacement of 0.27 kb of intron 2 to 0.6 kb

following exon 3 with a loxP site

Yes Yes Yes [189]

Rikn Replacement of a portion of intron 3, the

entire ORF, and a portion of the 3’ UTR

with a neo cassette

Yes Yes Yes (in cell lines derived

from the hippocampus)

[348]

46

degeneration) causes Purkinje cell loss via an indirect mechanism (e.g., by loss of a

hypothetical trophic signal emanating from CGNs). In this scenario, PrPC in CGNs

expressed from the tga20 transgene array would directly block Doppel toxicity.

Transgenic mice expressing Doppel under control of the hamster Prnp promoter

superimposed on a ZrchI Prnp0/0 genetic background (non-ataxic, non-Doppel over-

expressing) develop ataxia and have profound loss of cerebellar granular neurons (CGNs)

and Purkinje cells (for a summary of Doppel and ΔPrP transgenic mice, see Table 1.4)

providing additional support for the causal relationship between Doppel expression and

the ataxic phenotype [289]. Other areas of the brain remain apparently healthy at the time

that cerebellar pathology develops and despite the presence of high levels of Doppel

expression, arguing that a CGN- or Purkinje cell-specific co-factor or Doppel-binding

protein hastens the neurodegenerative phenotype. In agreement with the aforementioned

studies on ZrchII mice, age of onset of ataxia was inversely correlated with Doppel

protein levels. Phenotypic rescue could be achieved by crossing the mice to mice over-

expressing hamster PrPC (Tg(SHaPrP+/+)7/Prnp0/0-ZrchI) although one line expressing

Doppel at a high level still exhibited some neurodegeneration in the cerebellar cortex, in

agreement with other studies in which high levels of Doppel expression were unable to be

counteracted by PrPC expression [351].

Transgenic mice expressing Doppel specifically targeted to Purkinje neurons

results in ataxia and Purkinje cell loss, followed later by granule cell reduction and gliosis

(which the authors attribute to a distinct mechanism stemming from Purkinje cell death)

confirming that Doppel expression is toxic to Purkinje cells [351]. Only mice expressing

lower levels of Doppel could be rescued from the ataxic phenotype by co-expression of

PrPC (and even in this case, some late onset Purkinje cell loss was apparent). Mice

expressing high levels of Doppel on either a wild-type Prnp background or a tga20

transgenic (mouse PrPC over-expressed from the Prnp ‘minigene’ construct) background

exhibited severe ataxia and cellular loss. These results imply that PrPC is unable to

counteract the deleterious effects of high levels of Doppel expression. A separate report

has confirmed that Purkinje cell-specific expression of Doppel (as well as neuron-specific

expression controlled by the neuron-specific enolase (NSE) promoter) is sufficient to

confer ataxia and Purkinje cell degeneration [352]. Again, age of onset was inversely

47

Table 1.4. Summary of transgenic mice expressing Doppel or ΔPrP Transgenic

Line Promoter/Transgene

Construct Transgene Expression Phenotype(s) Rescue by

PrPC? Reference

PrPΔ32-93 (PrPΔC)

Murine Prnp ‘minigene’ construct

Neuronal (no Purkinje

cells)

-- N/A [217]

PrPΔ32-106 (PrPΔD)



cells)

-- N/A [217]

PrPΔ32-121 (PrPΔE)



cells)

Ataxia with CGN

degeneration

Yes [217]

PrPΔ32-134 (PrPΔF)



cells)

Ataxia with CGN

degeneration

Yes [217]

L7- PrPΔ32-134

L7 promoter Purkinje cells Ataxia with Purkinje cell

loss

Yes (some Purkinje cell

loss apparent)

[353]

PrPΔ105-125

(PrPΔCR)



cells)

Rapid illness with CGN loss

Yes [280]

PrPΔ94-134 (PrPΔCD)



cells)

Rapid illness without CGN

loss

Yes [279]

Tg(Dpl) Syrian hamster Prnp promoter

Neuronal Ataxia with CGN

degeneration and Purkinje

cell loss

Yes (limited degeneration

in high-expresser

lines)

[289]

L7-Dpl L7 promoter Purkinje cells Ataxia with Purkinje cell

loss

Yes (low-expressing lines only)

[351]

Tg-N(Dpl) NSE promoter Neuron-specific

Ataxia with Purkinje cell

loss

Yes (Prnp+/+ only)

[352]

Tg-P(Dpl) PCP-2 promoter Purkinje cells Ataxia with Purkinje cell

loss

Yes (Prnp+/+ only)

[352]

48

related to levels of Doppel. A single copy of wild-type PrP was able to delay, but not

prevent the ataxic phenotype. In this system, no ataxia was observed in any Doppel-

expressing line on a wild-type Prnp+/+ background.

PrP Mutants with Deletions in the N-terminal Unstructured Region Expressed in a

Prnp0/0 Genetic Background

The phenotypic defects seen in mice ectopically expressing Doppel in the brain

are reminiscent of an earlier-described artificially-induced disease (‘Shmerling

syndrome’) in transgenic mice expressing N-terminally truncated PrP molecules [217].

Mice expressing PrPΔ32-121 or PrPΔ32-134 (collectively referred to as ΔPrP) but not

PrPΔ32-93 or PrPΔ32-106 develop ataxia characterized by degeneration of cerebellar

granule cells. As is the case for Doppel, this phenotype is only apparent when the mutant

PrP’s are expressed on a Prnp0/0 background demonstrating that the process can be

abrogated by wild type PrPC. Similar to Doppel, Purkinje-cell specific expression of ΔPrP

is also sufficient to induce ataxia and Purkinje cell loss in a manner that is dependent on

the absence or presence of wild-type PrP [353]. Two recent studies, which utilize

transgenic mice expressing more restricted deletions on a Prnp0/0 background, have

shown that deletions in PrP’s central domain (in the form of either Δ94-134 [279] or

Δ105-125 [280]) are sufficient to cause ataxia and cerebellar granule cell degeneration.

Remarkably, toxicity is greatly enhanced in these mice compared to mice expressing the

longer Δ32-121 and Δ32-134 deletions, pointing to the central domain of PrP (and in

particular the hydrophobic tract) as the major determinant of this phenotype. The

structural similarities between Doppel and ΔPrP certainly suggest the economical

explanation that the two diseases are caused by a single mechanism [221]. One problem

in positing a single pathogenic pathway, however, has been the differences in target cell

populations in the cerebellum (i.e. Purkinje cells vs. cerebellar granular cells vs. both

types of cell) amongst the various strains of mice. Although trivial explanations are

possible (i.e. differences in expression owing to the use of different transgenic cassettes,

different genetic backgrounds), alternate explanations could involve trans interactions

between Purkinje neurons and CGNs with respect to Doppel or ΔPrP and a loss of trophic

factors. However, other possibilities include binding of Doppel/ΔPrP to different splicing

isoforms within a hypothetical protein ligand family.

49

Transgenic expression of either ΔPrP or Doppel also results in a late-onset white

matter pathology characterized by axon and myelin degeneration implying a shared

mechanism of white matter toxicity between the two structurally similar proteins [354].

White matter pathology is also observed in transgenic mice expressing PrP alleles with

central domain deletions [279, 280]. Expression of PrPC under control of an

oligodendrocyte-specific promoter rescued the axon-myelin degeneration and enhanced

survival but did not abrogate the cerebellar pathology. In contrast, neuronal-specific PrPC

expression conferred partial resistance to the ΔPrP-induced cerebellar pathology and

increased survival but had no effect on the axon-myelin degeneration. These experiments

point to a role for PrPC in the maintenance of myelin integrity and is supported by the

observation that demyelination occurs in the spinal cord and peripheral nerves of aged

ZrchI Prnp0/0 mice [355].

Cellular Models and Prnp/Prnd Interactions

Some investigators have sought to create cellular models of Doppel/PrPC phenotypic

interactions. In one strategy, primary CGNs—a target of Doppel/ΔPrP toxicity in vivo in

transgenic mice—are cultured from Prnp0/0 mice and transfected with constructs

encoding Doppel and PrP alleles of interest. CGNs transfected with Doppel exhibit a

significant increase in apoptotic events and levels are returned to baseline upon co-

transfection with wild-type PrPC [271]. In agreement with the observations in transgenic

mice, no increase in apoptosis is observed when CGNs isolated from wild-type (Prnp+/+)

mice are utilized. This model system has appeal because it utilizes differentiated neurons

and not immortalized cells, although it suffers from the typical inability to obtain high

transfection rates in primary cultures. A second strategy involves serum deprivation of a

hippocampal cell line derived from Doppel-expressing Rikn Prnp0/0 mice [356].

Following serum deprivation, Prnp0/0 cells die via apoptosis (and have shorter neurite

extensions) whereas Prnp+/+ cells or Prnp0/0 cells transfected with PrPC remain healthy.

These cells express Prnp/Prnd chimeric mRNA’s and low amounts of Doppel protein

[357]. Serum deprivation-induced toxicity is moderately enhanced in cells expressing

additional amounts of Doppel and toxicity is blocked in the presence of wild-type PrPC.

However, there is some doubt as to whether serum-induced apoptosis in Prnp0/0 cells is

50

additive in a pathologic sense to Doppel over-expression, since serum deprivation of cells

prepared from a strain of Doppel non-expressing Prnp0/0 mice (ZrchI) also induces

apoptosis [357]. A third model in a recently-published study has provided evidence that

N2a neuroblastoma cells (a popular cell line for studying prion replication) can be

sensitized to Doppel toxicity following depletion of endogenous PrPC using RNA

interference [358]. Qin et al. have also shown that recombinant Doppel (rDpl) is capable

of inducing caspase-3 activation when added to either N2a cells or primary rat adult

reactive astrocytes following transfection with PrP RNAi [358]. Addition of rDpl to

cultures of Prnp0/0 CGNs also causes apoptosis, but this effect is also apparent in Prnp+/+

wild-type mice casting doubt on its relevance to in vivo phenotypes [271].

Protein Structural Determinants and Prnp/Prnd Interactions—Determinants in PrP and

Doppel

In an attempt to characterize the domains in Doppel and PrPC responsible for

eliciting the neurotoxic or neuroprotective activity, respectively, several investigators

have utilized a mutagenesis approach. The rescue of Doppel-induced cerebellar

degeneration by PrPC requires an intact N-terminal domain of PrPC. Transgenic mice

expressing PrPΔ23-88 on the ataxic Ngsk Prnp0/0 background develop ataxia and Purkinje

cell loss indistinguishable from non-transgenic littermates suggesting that the octarepeat

region is required for neuroprotection [359]. In agreement with these results, PrP alleles

lacking either the charged N-terminal region (Δ23-28) or the octarepeats (Δ51-90) also

failed to rescue Doppel-induced apoptosis in primary cultures of Prnp0/0 cerebellar

granule cells [271]. In contrast, rescue of ΔPrP toxicity could be achieved by co-

expression of PrPΔ32-93 in transgenic mice [360], perhaps pointing to the existence of

subtle differences in the mechanisms of Doppel and ΔPrP toxicity or the necessity of

residues 23-31 of PrPC for neuroprotection. Fusion of PrPC residues 1-124 but not

residues 1-95 to the N-terminus of Doppel protected cells from Doppel toxicity during

serum deprivation implicating residues 96-124 in the neuroprotective function of PrPC

[361] and this region contains a portion of the highly conserved hydrophobic domain. In

agreement with this data, PrP lacking residues 95-132 was incapable of inhibiting serum

deprivation-induced apoptosis whereas a 124-146 deletion retained this ability

51

reinforcing the importance of the region between residues 95-123 for neuroprotection

[357]. A role for copper binding in PrPC neuroprotection is implied from analyses of PrP

alleles lacking the copper-binding sites in the octarepeats, which proved to be non-

protective [271], although this has not yet been established at the level of physical

PrPC/Cu2+ interactions in vivo, and also because one or two other copper binding sites

still exist in this construct. Collectively, these results show that various regions of the N-

terminus of PrPC are important for maintaining its neuroprotective activity. Several of

these regions (residues 23-28 and the octarepeats) have been implicated in the proper

trafficking of PrPC [212, 218] suggesting that their deletion may prevent PrPC from

localizing with a required co-factor.

Mutagenesis experiments have shown that the helix B/B’ region of Doppel

(residues 101-125) are required for Doppel-induced toxicity in cultured Prnp0/0 neurons

[271]. Notably, this is a region of structural deviation from PrPC, suggesting that altered

binding to an essential co-factor might alter the experimental endpoint.

Cellular Mechanisms in Prnp/Prnd Interactions

The downstream events initiated by CNS expression of Doppel or ΔPrP and

culminating in loss of cerebellar cells remain enigmatic. Any proposed mechanism must

take into account the ability of PrPC to rescue the Doppel-induced toxicity. Several

scenarios have been proposed (Figure 1.9) including 1) Direct competition for a shared

ligand or co-factor (such as copper) required by Doppel to activate an apoptotic pathway

2) Direct competition for a shared ligand required by Doppel to subvert a cellular

pathway required for cell viability 3) Triggering of opposing neurotoxic and

neuroprotective pathways by Doppel and PrPC, respectively 4) Direct binding of PrPC to

Doppel prevents it from eliciting its toxic effect 5) Increased free radicals arising from

Doppel expression cause oxidative damage to membranes or proteins with PrPC

antagonizing this effect and 6) PrPC interfering with the formation of ER or plasma

membrane multimeric pores composed of Doppel. The potential existence of a common

neurotoxic mechanism used by both Doppel and mutant forms of PrP in genetic prion

disease has been proposed based on the fact that several mutations in PRNP which cause

gCJD (such as E200K) result in amino acid shifts within the α-helical domain to residues

52

Figure 1.9: Proposed models for Doppel neurotoxicity and PrPC neuroprotection in cerebellar cells. A: Doppel is capable of activating a cell death-associated pathway but requires a co-factor (a protein or an inorganic molecule) to do so. PrPC efficiently competes with Doppel for binding to the co-factor and prevents activation of the death pathway. B: PrPC binds to Doppel which either prevents activation of the death-associated pathway directly or blocks access of a necessary co-factor for death signaling. C: Doppel is involved in the inhibition of a pathway essential for cell viability (as opposed to activating a pro-death pathway). PrPC would efficiently compete with Doppel for binding to a co-factor necessary for Doppel to subvert the cell viability pathway. D: Doppel forms multimers which assemble into a pore-like structure leading to increased cell permeability and activation of cell death. PrPC interferes with pore formation and thus blocks the toxicity. E: Doppel expression in the cerebellum leads to the production of free radicals and oxidative damage. PrPC is able to counteract this process and is therefore neuroprotective. Figure taken from Westaway et al. [362]

53

conserved in Doppel [363]. Thus, these substitutions may result in a PrP protein capable

of mimicking Doppel, leading to neurotoxic events.

Although the Prnp and Prnd genes “interact” in CNS toxicity assays, this is not to

say that there are physical interactions between their cognate protein products. Indeed,

given that neither PrPC nor Doppel have been suggested to exist as dimers in vivo,

physical interactions are not necessarily expected even though some investigators have

persevered with this assumption. Accordingly, there is little evidence of a direct

interaction between Doppel and PrPC [292, 364]. Nonetheless, binding of recombinant

Doppel to recombinant PrPC was observed in one study and was inhibited by pre-

incubation of Doppel with a PrP106-126 peptide [365] and co-purification of Doppel and

bands that appear to be endogenous mono- or un-glycosylated PrPC was achieved in vitro

in N2a cells by using a TAP-tagged Doppel construct as bait [358, 366]. Doppel and PrPC

co-localize on the surface of transfected human and mouse neuroblastoma cells implying

proximity of the two proteins in the plasma membrane [358], whereas Doppel and PrPC

may not inhabit the same membrane microdomains in testis [230]—perhaps a neuron-

specific co-factor is involved in the interaction. Again, however, co-localization as

determined by light microscopy certainly does not confirm a physical interaction. An

indirect interaction between Doppel and PrPC is implied by studies in which antibodies to

either PrPC or Doppel induce co-internalization of both proteins [366]. Despite these

observations, co-immunoprecipitation of wild type Doppel and PrPC (a simple point of

reference for confirming stable protein-protein interactions) has never been achieved,

although it is possible to invoke a caveat that the proper detergent conditions were not

implemented. Additionally, in a yeast 2-hybrid setting Doppel fails to interact with PrPC

[292].

With regards to possible downstream pathways initiated by Doppel, cultures of

Prnp0/0 cerebellar granule neurons die via an apoptotic pathway upon exposure to

transfected Doppel as demonstrated by activation of caspase-3 [271]. In addition, PrP-

depleted N2a cells expressing Doppel also exhibit caspase-3 and caspase-10 (but not

caspase-9) activation [358]. These studies clearly demonstrate that Doppel somehow

engages a pro-apoptotic pathway (interestingly, nanomolar concentrations of purified

PrPSc added to tissue culture cells also induces apoptosis in a caspase-12-dependent

54

manner [367]). Caspase-3 and TUNEL-positive cerebellar granule neurons are also

observed in the brains of mice expressing PrPΔ105-125 [280], perhaps suggesting a

common route of toxicity for both Doppel and N-terminally deleted forms of PrP.

Crossing mice expressing a toxic PrPΔ32-134 allele with mice lacking Bax (a pro-

apoptotic protein which is an important regulator of neuronal death in the CNS) has

shown that although Bax deletion can delay clinical illness and suppress cerebellar

granule neuron loss in young mice, it does not reduce white matter pathology or age of

death [368]. Similarly, Bax deletion has no effect on the progression of prion disease in

inoculated mice [369]. These experiments suggest that apoptotic proteins other than Bax

are likely involved in mediating Doppel/ΔPrP toxicity. In support of the apoptosis

mechanism, studies in Chinese Hamster Ovary cells have pointed to perturbations in

subplasma membrane, ER, and mitochondrial calcium pools upon expression of either

PrPC or Doppel [370]. In general, PrPC and Doppel elicit opposing effects associated with

protective and apoptotic calcium signaling, respectively, whereas co-expression results in

a counter-balancing neutral effect. There is some evidence that Doppel causes oxidative

damage to the brain and cultured cells from Prnp0/0 mice. Rcm0 Prnp0/0 mice were

reported to have increased expression of heme oxygenase 1 and nitric oxide synthase, and

increased levels of nitrite, protein oxidation, nitrotyrosine, and lipid peroxidation, all

pointing to an increase in oxidative stress in Doppel-expressing mice [371]. In

comparison, no changes in either nitrotyrosine or in modification of carbonyl groups

(both markers of oxidative damage) were found in other transgenic mice expressing

Doppel and succumbing to a neurodegenerative syndrome [287].

A recent report has demonstrated that PrPC and Doppel are sorted differently in

cultured Madin-Darby canine kidney cells [364]. Doppel is typically found on the apical

surface whereas PrPC is found on the basolateral side. Interestingly, co-expression of

Doppel and PrPC directs PrPC to the apical side, putting PrPC into the vicinity of Doppel

and possibly providing a mechanism by which PrPC can abrogate Doppel toxicity. Also,

the hydrophobic tract of PrPC is necessary for basolateral sorting and can confer

basolateral sorting to Doppel in a dominant manner. Of further interest is the observation

that a pathogenic PrP mutation (the AV3 hydrophobic mutation which induces the CtmPrP

topology and causes neurodegeneration [215]) causes PrPC to be sorted apically.

55

A Molecular Model for Prnp/Prnd Interactions

One model has been proposed which attempts to explain interactions between

Doppel and PrPC in terms of an unidentified prion protein receptor termed LPrP or Tr. In

this model, PrPC binds to LPrP and elicits an unknown “positive” signaling event favoring

cell survival (Figure 1.10) [217, 279, 280, 372]. In the absence of PrPC, Doppel (or ΔPrP)

also binds to LPrP but lacks the effector domain (which is present on PrPC) to initiate

positive signaling and instead either functions as a dominant negative by blocking

signaling or initiates ‘improper signaling’ through LPrP. PrPC has a higher affinity for LPrP

than either Doppel or ΔPrP explaining its ability to block initiation of the toxic phenotype.

No candidates for LPrP have come forward, but the following characteristics are predicted

based on the current model: 1) Expression in the CNS with particular localization in the

Purkinje cells or cerebellar granule cells 2) Two binding sites for PrPC: one not present in

Doppel which acts as an effector domain and a second which is shared between the two

proteins and 3) A cytoplasmic domain (or the ability to interact with another protein with

cytoplasmic sequences) which can initiate downstream signaling. It seems plausible that

altered binding to LPrP could also explain the toxicity of CtmPrP. Because the hydrophobic

tract of CtmPrP is embedded in the membrane, the remaining cell surface-exposed portion

of PrP may structurally resemble ΔPrP (in particular PrPΔ32-134), which could bind LPrP

in a manner reminiscent of ΔPrP and induce a similar phenotype. In support of this idea,

transgenic mice expressing a PrP mutant which generates increased amounts of CtmPrP

develops a neurological illness characterized by a progressive loss of cerebellar granule

neurons [373]. However, toxicity is dependent on the presence of wild-type PrPC, which

is unlike the case of ΔPrP in which PrPC must be absent for toxicity to occur. An

intriguing question is whether or not LPrP (or a similar protein) may play a role in prion

disease. Indeed, the presence of a transmembrane protein which transduces neurotoxic

signals emanating from PrPSc has been postulated based on studies of transgenic mice

expressing PrP lacking the GPI anchor [87]. Following prion inoculation, these mice have

abundant PrPSc-containing plaques and protease-resistant PrP but never develop clinical

prion disease [86] suggesting that PrPSc binding to a component of the cell membrane is

essential for eliciting neurodegeneration.

56

Figure 1.10. The LPrP model of functional interactions between prion proteins in cerebellar neurons of transgenic mice. A: In wild-type (Prnp+/+) mice, PrPC binds to a hypothetical ligand (LPrP) and initiates an as-yet-unidentified signaling event favouring cell survival. This binding event could occur either in cis or trans configurations. Two binding sites for LPrP on PrPC are implied—a C-terminal anchoring site and an N-terminal effector site which enables signaling. B: In Prnp0/0 mice, a hypothetical PrPC-like protein, π, binds to LPrP and initiates the signaling event in the absence of PrPC. π shares in common with PrPC the N-terminal effector domain and can therefore elicit signaling activity through LPrP. C: When Doppel or ΔPrP bind to LPrP in Prnp0/0 mice, ‘improper’ signaling is initiated resulting in cellular death since both proteins lack the N-terminal effector domain required for pro-survival signaling. D: When Doppel or ΔPrP are expressed in Prnp+/+ mice, PrPC possesses higher affinity for LPrP and thus prevents or reduces Doppel/ΔPrP binding.

57

Another facet of the above model is a hypothetical PrPC-like protein termed ‘π’

which was invoked to explain the lack of a phenotype in Prnp0/0 mice. π also binds to LPrP

and can initiate the positive signaling event favoring cell survival. No candidate proteins

which fit the definition of the π molecule have been identified to date.

Summary

Numerous unanswered questions exist in prion biology. In particular, the function

of PrPC remains enigmatic, the necessity of auxiliary proteins for prion replication in vivo

is unclear, and the mechanism by which PrPSc causes neurotoxicity is unresolved. Thus, it

is necessary to uncover new proteins of relevance to prion biology and disease, including

proteins with similarity to PrPC as well as authentic prion protein-interacting proteins.

The identification and functional characterization of such proteins may provide new

insights into the role of prion proteins in health and disease.

58

Chapter 2

Rationale, Hypotheses, and Objectives

59

2.1. Rationale

Despite considerable research efforts, the cellular function of PrPC remains

enigmatic. However, PrPC is quite well-conserved so it is reasonable to assume that it

possesses some beneficial property since it seems highly unlikely that it would have been

conserved in nature for the sole purpose of conferring susceptibility to prion diseases

[372]. It remains possible that the lack of an overt phenotype in Prnp0/0 mice is due to the

existence of a compensatory protein. This hypothetical protein has been named π but no

candidates have surfaced to date [217, 372]. With respect to this notion, the identification

of a putative third prion gene (Sprn) which is expressed within the CNS opens up new

avenues of potential research [197]. Although Doppel has a remarkably similar three-

dimensional structure to PrPC [221], it clearly is not π since 1) Doppel is not expressed in

the post-embryonic brain [190] and 2) Prnp0/0/Prnd0/0 double knockout mice do not have

any additional phenotypes compared to either single knockout [285]. The similarity

between Shadoo and PrPC is largely confined to the hydrophobic tract, a region of PrPC

which is emerging as a key determinant in controlling its neuroprotective behaviour [279,

280]. Thus, it is reasonable to assume that the Shadoo protein may possess similar

biochemical and functional properties to PrPC and thus is a reasonable candidate for the π

molecule. Consequently, a careful examination of the hypothetical Shadoo protein in the

brain is warranted. In addition, the hydrophobic tract of PrP likely corresponds to the

epicenter of misfolding to disease-associated forms such as PrPSc [223]. Because Shadoo

also contains this domain, it is worthwhile to investigate the properties of Shadoo within

the context of prion disease and to assess whether or not Shadoo is capable of modulating

the folding of PrP or vice versa.

The understanding of both the cellular function of PrPC and the mechanisms

which govern prion replication and pathogenesis has been hindered by the inability to

find proteins which interact with PrPC in vivo and in a biologically relevant manner. A

large number of candidate PrPC-interacting proteins have been described, but none have

been informative about PrP’s role in normal and disease settings [374]. A recent study

has defined the proteins which reside in the vicinity of PrPC in the brains of wild type

mice [266]. A careful assessment of the candidate PrPC-interacting proteins within this

list is likely to reveal proteins which are important for understanding prion biology. One

60

protein in particular, DPPX, seems especially worthy of further investigation for the

following reasons: 1) It is not expressed at high levels in the mouse brain and therefore is

unlikely to have co-purified with PrPC simply due to its abundance 2) DPPX does not

contain the Fibronectin Type III and C2 domains found in other candidate PrPC-

interactors on the list 3) Excellent tryptic peptide coverage was obtained in the TcTPC

experiments (13 unique peptides were identified) and 4) DPPX has a documented role in

the trafficking of neuronal membrane proteins [375].

DPPX was first identified in 1992 as a new member of the dipeptidyl

aminopeptidase family of membrane-bound serine proteases which is expressed in the

brain with several different N-terminal splicing isoforms [376]. Little is known about the

function of DPPX, although it is known to modulate the properties of neuronal A-type

potassium channels by increasing the trafficking of Kv4.2 to the cell surface [375]. In

addition, DPPX is unlikely to function as a canonical serine protease since the catalytic

serine residue has been mutated [377]. Because DPPX may be the most intriguing of the

strong candidate in vivo PrPC-interacting proteins [266], an explorative study of the

effects of DPPX on well-defined PrP properties including PrPC neuroprotective activity,

prion replication, and prion pathogenesis is warranted.

2.2. Hypotheses

1. It is hypothesized that the notional Sprn open reading frame encodes an actual

protein (Shadoo) that is present in the brains of healthy animals and, because of

its sequence similarity to PrP, has similar biochemical and functional properties to

PrPC.

2. It is hypothesized that DPPX interacts with PrPC in vivo. Furthermore, it is

hypothesized that DPPX is capable of modulating the cellular activities of PrPC

and/or is involved in regulating pathophysiological aspects of PrP biology such as

prion replication or prion pathogenesis.

61

2.3. Objectives

1. Clone the Sprn gene and raise antibodies to the putative Shadoo open reading

frame in order to characterize Shadoo biochemically and to evaluate Shadoo

protein expression in mouse brain and in tissue culture cells.

2. Evaluate potential functional redundancies between Shadoo and PrPC by assessing

the ability of Shadoo to modulate Doppel and ΔPrP-induced neurotoxicity.

3. Assess any potential involvement of Shadoo in prion replication.

4. Clone the DPP6 gene and raise peptide antisera to its translated product, DPPX,

in order to evaluate protein expression in mouse brain and in tissue culture cells.

5. Validate the proposed interaction between PrPC and DPPX.

6. Map the binding sites which govern PrPC/DPPX complex formation on both PrPC

and DPPX.

7. Assess whether DPPX is capable of modulating genetic interactions observed

between Doppel/ΔPrP and PrPC in cerebellar granular neurons.

8. Determine if DPPX is involved in either the cellular function of PrPC or the

pathophysiology of PrPSc including prion replication and prion pathogenesis.

62

Chapter 3

Characterization of Shadoo, a Putative Prion-Like

Protein

Portions of this section have been published in the following article:

Watts, J.C., Drisaldi, B., Ng, V., Yang, J., Strome, B., Horne, P., Sy, M.-S., Yoong, L.,

Young, R., Mastrangelo, P., Bergeron, C., Fraser, P.E., Carlson, G.A., Mount, H.T.J.,

Schmitt-Ulms, G., and Westaway, D. “The CNS glycoprotein Shadoo has PrPC–like

protective properties and displays reduced levels in prion infections”, (2007) EMBO

Journal 26(17): 4038-50.

63

3.1 Abstract The prion protein family currently consists of two members: PrPC which is a

neuronal glycoprotein of unknown function and is the precursor to prion disease-

associated isoforms such as PrPSc, and Doppel, a protein with a similar three-dimensional

structure which is expressed in the testis and is required for the proper functioning of the

male reproductive system. The lack of a major phenotype in Prnp0/0 mice has led to

speculation that a hypothetical second CNS prion protein termed π compensates for the

loss of PrPC. One candidate for π is Shadoo (Sho), the protein product of the hypothetical

Sprn gene and which bears some resemblance to the middle portion of PrPC. Antibodies

generated against the Sho open reading frame reveal that Sho is an N-glycosylated, GPI-

anchor protein which is present in the brains of wild-type mice. High levels of Sho

expression are observed in hippocampal and Purkinje neurons, as well as in other neurons

in the brain and motor neurons of the spinal cord. Similar to PrPC, Shadoo undergoes

endoproteolytic processing to liberate an N-terminal peptide. Biochemical

characterization of Shadoo protein expressed in N2a cells suggests that it is an unstable

protein denoted by a short half-life and a high sensitivity to protease digestion. When

tested in a functional assay, Sho exhibits PrPC-like neuroprotective properties in that it

protects cultured cerebellar granule neurons from the toxicity associated with expressing

either Doppel or N-terminally truncated PrP suggesting that Sho and PrPC may be

functionally redundant. Remarkably, levels of Sho are decreased by approximately 90%

in the brains of terminally ill prion-infected animals and transfected Sho levels are

diminished in prion-infected cells compared to non-infected cells. Cumulatively, these

results define Shadoo as the third member of the prion protein family and the second

CNS prion protein. Biochemical and functional similarities to PrPC make Sho a strong

candidate for the π molecule and further studies of Sho within the context of prion

disease may yield insights into the mechanisms of prion replication or pathogenesis.

64

3.2 Introduction Prions are proteinaceous infectious particles and are the causative agents of

neurodegenerative diseases which include bovine spongiform encephalopathy (BSE) in

cattle, scrapie in sheep, chronic wasting disease in mule deer and elk, and Creutzfeldt-

Jakob Disease (CJD) in humans. The infectious agent is believed to consist of improperly

folded forms of a host-encoded protein, the cellular prion protein (PrPC). Conversion of

PrPC into the prion disease-associated isoform, PrPSc, is thought to proceed by a template-

directed refolding mechanism and is believed to be the primary pathogenic event,

although the mechanisms by which PrPSc causes neurodegeneration are poorly

understood. PrPC is absolutely required for disease progression as PrP knock-out

(Prnp0/0) mice do not succumb to disease and do not propagate infectivity following

intracerebral challenge with infectious prions [107, 108]. The minimum components for

templated prion replication in vitro have been defined as PrPC and co-purified lipids,

PrPSc, and a polyanionic molecule [117]. The requirement of auxiliary molecules such as

polyanions [114, 116] or other hypothetical protein co-factors [110] for prion replication

in vivo has not been confirmed.

The mammalian prion protein family currently consists of two proteins: the

cellular prion protein, PrPC, which is expressed at high levels in the central nervous

system, and Doppel (Dpl), a molecule with a similar C-terminal domain whose postnatal

expression is normally confined to the testis [221, 229]. Whereas a role for Doppel in the

proper functioning of the male reproductive system has been confirmed in two lines of

Doppel knock-out mice [284, 285], the function of PrPC, a well-conserved neuronal

glycoprotein, has remained enigmatic, in part because phenotypic alterations in Prnp0/0

mice have been subtle or disputed [186, 242, 244, 256, 262]. One emerging area of

consensus concerns a protective effect of PrPC against neuronal insults [268, 356, 365,

378]. For instance, PrPC can protect against Bax toxicity in mammalian cells and in yeast

[269, 270]. In addition, PrPC is up-regulated following ischemic brain damage, in both

humans and mice [272, 273]. PrPC deficiency in mice increases infarct size following

cerebral artery occlusion and increases caspase 3 activation [274], and PrPC

overexpression improves neurological behavior and reduces infarct volume in a rat stroke

model [275]. Strong evidence for a neuroprotective activity for PrPC against apoptosis in

65

vivo has come from studies of transgenic mice expressing internally deleted forms of PrP

or wild-type Doppel within the CNS. The presence of Doppel in the brain of Prnp0/0 mice

leads to a neurodegenerative syndrome characterized by a profound apoptotic loss of

cerebellar cells [188, 189, 289, 351]. A similar phenotype is observed when N-terminally

truncated versions of PrPC (PrPΔ32-121 or PrPΔ32-134, collectively referred to as ΔPrP)

are expressed in the brain [217, 353]. Remarkably, both syndromes are abrogated by the

co-expression of wild-type PrPC. Recently, it has been shown that a smaller deletion

restricted to the well-conserved central domain of PrP is sufficient to elicit a highly toxic

phenotype in Prnp0/0 mice [279, 280]. The above studies have led to a model in which

Doppel and ΔPrP initiate aberrant signaling through a hypothetical prion ligand termed

LPrP, a process which is blocked by PrPC binding [121, 372]. Assuming that the

interaction between PrPC and LPrP represents an essential physiological event, the authors

also proposed the existence of a PrPC-like protein termed π, which binds to LPrP and is

capable of compensating for the absence of PrPC in Prnp0/0 mice [217]. To this date, no

candidates for π (or LPrP) have been put forward.

Recently, an open reading frame was discovered which, when translated, exhibits

homology to the central hydrophobic domain in PrPC. This gene, denoted Sprn (“shadow

of the prion protein”), is present from zebrafish to humans and is predicted to encode a

short protein, Shadoo (Sho) [197]. Mammals possess a single Sprn gene, whereas

multiple Sprn genes (termed Sprna and Sprnb) have been found in zebrafish and other

species of fish [198]. Sprn is located on chromosome 7 in mice, away from the Prn gene

complex (which contains the Prnp and Prnd genes) on chromosome 2. Phylogenetic

analysis indicates that all modern prion and prion-like genes may have evolved from an

ancestral Shadoo gene [198]. Outside of the cloning of the Shadoo gene from various

mammalian species, transcriptional profiling, and expression of tagged zebrafish Sho

molecules in cultured cells [231, 379, 380], little work has been done on Shadoo. Sprn

mRNA appears to be confined to the CNS [197, 380], but no evidence towards the

existence of Shadoo protein in the brain of a wild-type animal (or spliced Sprn mRNA’s)

has been tendered.

Based on the phenotypic interaction between PrPC and Dpl or ΔPrP, an assay for

PrPC activity in primary cultures of cerebellar granule cells has been developed [271].

66

Here, cerebellar granule neurons (CGNs) cultured from Prnp0/0 mice are transfected with

plasmids encoding Dpl or PrP alleles of interest and individual apoptotic events scored.

This assay recapitulates the phenotypes produced by multiple PrP alleles in transgenic

mice, including neurotoxicity of both Dpl and ΔPrP, and neuroprotective activity of PrPC

against the toxicity elicited by either Dpl or ΔPrP [271]. In conjunction with biochemical

and histological analyses, the CGN assay has been used to explore the properties of the

notional Sho protein. Shadoo is revealed as being a GPI-anchored neuronal glycoprotein

present in the CNS from early post-natal life. Biochemical characterization of Sho

suggests that it is much less stable than PrPC, likely indicating a lack of well-defined

elements of secondary structure. In addition, not only is Sho PrPC−like in its ability to

counteract Dpl or ΔPrP toxicity in the CGN assay, but it is also strikingly reduced in

prion infections and in multiple lines of prion-infected cells.

3.3 Materials and Methods

Bioinformatics and Statistics

Protein alignments were created using the T-COFFEE algorithm

(http://www.ch.embnet.org/software/TCoffee.html). Transfected CGN datasets were

analyzed by one-way ANOVA and Tukey pair-wise comparisons with significance set at

p<0.05 using GraphPad Prism software (version 5, GraphPad Inc.). Statistical analysis on

protein levels obtained by densitometry was performed using unpaired t tests.

Cloning of Shadoo and Plasmid Generation

The Shadoo open reading frame was amplified from mouse genomic DNA using

Platinum Taq polymerase (Invitrogen) in the presence of 5% DMSO and then inserted

between the HindIII and XbaI sites of either pcDNA3 or pBUD.CE4.GFP [271]

(Invitrogen). Shadoo deletion mutants and FLAG-tagged constructs were generated using

pairs of mutagenic oligonucleotides and the QuikChange (Stratagene) site-directed

mutagenesis procedure with Pfu Turbo DNA polymerase. The Thy-1 plasmid (Thy-1.2

isoform) was generated by amplification of the Thy-1 open reading frame from the

MGC:62652 cDNA clone by PCR and then insertion into either pcDNA3 or

67

pBUD.CE4.GFP. The identity of all constructs was verified by DNA sequencing. All

plasmids were prepared using endotoxin-free plasmid maxi-prep kits (Qiagen).

Shadoo Polyclonal Antibody Production

The peptide CRRTSGPGELGLEDDE (Shadoo residues 86-100 with an

additional N-terminal cysteine) was conjugated to maleimide-activated KLH (Pierce) and

injected into New Zealand White rabbits. Polyclonal antibodies were precipitated from

serum using ammonium sulfate and then affinity purified using the immunogenic Sho(86-

100) peptide conjugated to a SulfoLink column (Pierce) to generate the 04SH-1 and

06SH-3 anti-Sho antibodies. Alternatively, serum from the 06SH-3 rabbit was purified

over a column consisting of purified full-length recombinant Sho [381] conjugated to an

AminoLink Plus Coupling Gel column (Pierce) to generate the 06SH-3a anti-Sho

antibody.

Cell Culture, Transfection, and Lysis

N2a cells were cultured in DMEM medium containing 10% FBS and 0.2×

penicillin/streptomycin (Gibco) and maintained in a humidified incubator with 5% CO2.

ScN2a, ScGT1-trk, and GT1-trk cells were cultured in OptiMEM medium (Gibco)

containing 10% FBS, 1× GlutaMax (Gibco), and 0.2× Penicillin/Streptomycin and

maintained as above. SMB and SMB-PS cells were cultured in Medium 199 (Gibco)

containing 10% FBS and 0.2× penicillin/streptomycin and maintained as above. Cells

were transfected with Lipofectamine-2000 (Invitrogen) according to the manufacturer’s

instructions using a 1 µg DNA: 1 µL Lipofectamine-2000 ratio. For siRNA treatment of

ScN2a cells, cells were transfected with ON-TARGETplus siRNA (Dharmacon) at a final

concentration of 100 nM in OptiMEM using Lipofectamine-2000 as per the

manufacturer’s instructions. Cells were incubated with the transfection mixture for 48hrs

and then in serum-containing medium for an additional 24 hrs. For the creation of stable

cell lines, cells were selected and maintained in 1 mg/mL and 0.2 mg/mL G418 (Gibco),

respectively. For cell lysis, cells were washed twice with PBS and then lysed with RIPA

lysis buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.5% sodium deoxycholate and 1%

NP-40) containing Complete Mini Protease Inhibitor Cocktail tablets (Roche). Lysates

68

were incubated on ice and then cleared by centrifugation at 20,000× g for 10 min at 4ºC.

Protein concentrations were determined using the BCA assay (Pierce).

PI-PLC and PNGaseF Treatments

For PI-PLC treatment, Shadoo-transfected N2a cells were processed 24 hr post-

transfection by washing three times with PBS, and then treated with PI-PLC (Invitrogen)

diluted in PBS for 40 min at 4ºC. Cells were then washed twice and lysed as above.

PNGaseF digestions (New England Biolabs) were performed by boiling protein in

1× denaturing buffer (0.5% SDS, 0.5% β-mercaptoethanol) for 10 min, followed by the

addition of NP-40 and G7 reaction buffers to 1× and 50 units of PNGaseF per 50 µg

protein. Digestions were allowed to proceed at 37ºC for 4 hours.

Western Blotting

For Western blotting, 20 to 50 µg total protein was prepared in sample buffer,

boiled, and then separated on either 4-12% NuPAGE gels with the MES buffer system

(Invitrogen) or by conventional SDS-PAGE using 14% polyacrylamide gels. Proteins

were transferred to either nitrocellulose or PVDF and blocked with either 5% non-fat

skim milk or 2% BSA in TBS containing 0.05% Tween-20. Blots were then incubated

overnight with primary antibodies at 4ºC or at room temperature in the presence of 0.05%

sodium azide. Following three washes with TBS containing 0.05% Tween-20 (TBST),

blots were incubated with HRP-conjugated secondary antibody (BioRad) and developed

using Western Lightning ECL (Perkin-Elmer). The following primary antibodies were

used: anti-Sho 04SH-1, anti-Sho 06rSH-1, anti-Sho 06SH-3, anti-Sho 06SH-3a, anti-

FLAG M2 (Sigma), anti-PrP D18 (InPro Inc.), anti-PrP D13 (InPro Inc.), anti-PrP

monoclonal antibodies 8H4 and 7A12 (generous gifts from Man-Sun Sy), anti-APP C-

terminal rabbit polyclonal antibody, anti-neomycin phosphotransferase II (Millipore),

anti-Thy-1 R194 (a generous gift from Roger Morris), anti-actin 20-33 (Sigma), anti-

synaptophysin SY38 (Chemicon), and anti-calbindin D-28K (Chemicon). The embryo

blot was supplied from Zyagen (San Diego, CA), probed with 04SH-1 and then stripped

using 0.2 M Glycine pH 2.2, 1% Tween-20, 0.1% SDS followed by re-probing with D13

antibody.

69

Cerebellar Granular Neuron Cultures and Transfections

CGN cultures were obtained essentially as previously described [271]. Briefly,

cerebella from either wild-type or Prnp0/0 mice (ZrchI strain, both on a C57/B6

background) were dissected from 7 day old pups in HBSS. Cerebellar tissue was

disrupted by incubation with Trypsin-EDTA for 15 min at 37ºC and then the medium

replaced with MEM (Sigma #4655) containing 10% heat-inactivated FBS, 0.1×

penicillin-streptomycin, and 25mM KCl (K25+S medium) with the addition of soybean

trypsin inhibitor to a concentration of 0.25 mg/mL. Cerebella were triturated by pipetting

up and down with a fire-polished Pasteur pipet and following sedimentation of un-

digested material, cells were spun at 1300 rpm in an IEC Centra-EC4R centrifuge for 5

min at room temperature. Cells were resuspended in K25+S medium (without inhibitor),

filtered through a cell strainer, and then plated on 12-well tissue culture dishes (Costar)

that had been coated overnight with 0.1 mg/mL poly-L-lysine. Cells were incubated at

37ºC for 4 days prior to transfection. Individual wells of cells were transfected with 2 µg

DNA and 3 µL Lipofectamine-2000 in MEM medium without serum for 1 hr before

replacing with conditioned K25+S medium. For co-transfections, a 1:3 ratio of toxic

plasmid to test plasmid was used. The pBUD.GFP.Dpl and pBUD.GFP.PrPΔ32-121

plasmids have been previously described [271]. 24 hr post-transfection, cells were fixed

with 4% paraformaldehyde and nuclei stained with Hoechst 33342 (5 µg/mL in PBS).

Individual apoptotic events were scored by scoring nuclear morphology in GFP-positive

transfectants or by staining for activated caspase-3 using a cleaved caspase-3 (Asp175)

antibody (Cell Signaling) and an Alexafluor594-conjugated secondary antibody

(Invitrogen).

Preparation of Mouse Brain Homogenates and Membrane Fractions

Mice were perfused with saline, half brains were extracted, and then brains were

either homogenized directly or snap frozen and stored at -80ºC for future use. Brains or

other organs were homogenized in nine volumes of 0.32M sucrose (10% homogenates)

containing Complete Mini Protease Inhibitor Cocktail tablets (Roche). For preparation of

RML-infected brain homogenates, brains were homogenized in PBS without protease

inhibitors. For preparation of crude membranes, homogenates were spun at 700× g for 10

70

min at 4ºC, pellets washed with 1 volume of homogenization buffer, spun again as above,

and then supernatants from the two spins pooled. The supernatant was spun at 100,000×

g for 1 hr at 4ºC and pellets resuspended in 50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1

mM EDTA and solubilized with 1% Triton X-100 on ice for 1 hr. A final spin (100,000×

g, 1 hr, 4ºC) was performed and the supernatant stored at -80ºC.

Immunocytochemistry

N2a cells 24 hours post-transfection were washed with PBS, fixed with 4%

paraformaldehyde, washed with PBS, blocked with 2% goat serum, and then incubated

with primary antibody overnight at 4ºC. Following PBS washes, cells were incubated

with Alexafluor488- or Alexafluor594-conjugated secondary antibodies (Invitrogen) for

two hours and then washed three times with PBS. For immunocytochemistry on CGNs,

cells were fixed 24 hours post-transfection in ice-cold methanol for 5 min at -20ºC, and

then processed and stained. Fluorescent images were obtained using either a Zeiss

Axiovert microscope or a Leica DMI6000 B inverted microscope in conjunction with

Volocity software (Improvision).

Immunohistochemistry

For tissue analysis, mice were perfused with saline, brains bisected in the mid-

sagittal plane, and fixed in either 10% neutral buffered formalin or methacarn fixative

(60% methanol, 30% chloroform, and 10% glacial acetic acid). Formalin-fixed brains

were fixed for a minimum of 24 hours at room temperature and methacarn-fixed brains

were fixed either at room temperature for 3-4 hours or overnight at 4ºC prior to

immersion in 70% ethanol. Spinal cord and eye samples were fixed in methacarn fixative.

Brains or other organs were then processed to paraffin wax. Sections (6 µm) were cut,

dried (63ºC for 1 hour), de-paraffinized with xylene, re-hydrated through a graded series

of ethanol, and then rinsed in TBS pH 7.2. Formalin-fixed sections were subjected to

heat-induced epitope retrieval by microwaving in citrate buffer (10 mM, pH 6) for 30 min

and then cooling at room temperature for 30 min prior to antibody application. Primary

antibodies were diluted in Antibody Dilution Buffer (DAKO Cytomation) and applied to

the tissue sections for an overnight incubation at 4ºC in a humidified chamber. For

71

peptide blocking of antibody reactivity, peptide was added in 4-fold mass excess to the

diluted antibody and rotated overnight at 4°C prior to application to tissue sections. The

sections were then rinsed in TBS and processed in EnVision Labeled Polymer (DAKO

Cytomation), rinsed with water, visualized with DAB (3,3'-diaminobenzidine) and

counterstained with Harris' hematoxylin (Sigma). Images were captured on a Leica

DM6000 B microscope using a Micropublisher 3.3RTV camera (Q Imaging Inc.) in

conjunction with OpenLab software (Improvision). For fluorescent double-labeling

experiments, the secondaries used were Alexafluor488- and Alexafluor594-conjugated

antibodies (Invitrogen). Some fluorescent images were de-convoluted using the iterative

restoration algorithm in Volocity.

The following antibodies were used for immunohistochemistry: anti-Sho 06rSH-1,

anti-Sho 04SH-1, anti-Sho 06SH-3, anti-Sho 06SH-3a, anti-PrP 7A12, anti-PrP 3F4

(Signet), and anti-Neurofilament H SMI-32 (Sternberger Monoclonals, Inc.).

In situ Hybridizations

The template for probe generation was pcDNA3.Shadoo, a plasmid containing the

entire open reading frame of murine Shadoo inserted between the HindIII and XbaI sites

of pcDNA3. The plasmid was linearized by digestion with either HindIII or BglII in order

to generate anti-sense and sense probes, respectively. Digoxigenin (DIG)-labeled RNA

probes were created using the DIG RNA labeling kit (Roche). The labeling reaction

included 1 µg of purified linearized plasmid and either SP6 or T7 RNA polymerase for

anti-sense and sense probes, respectively. Template DNA was removed by digestion with

DNaseI for 15 min at 37ºC. Labeled probes were purified over Sephadex G50 spin

columns (GE Health Sciences) and probe concentrations were estimated by performing

serial dilutions of the probes on nylon membranes and comparing them to a known

standard (Roche).

Sections (6 µm) of formalin-fixed, paraffin-embedded mouse brains were cut

using an RNase-free blade, mounted on slides, and dried overnight at 63ºC. Paraffin was

removed by incubation in xylene followed by rehydration through a graded series of

ethanol. Sections were post-fixed in 10% formalin for 20 min, rinsed three times with

TBS, and then incubated in 200 mM HCl for 15 min to denature proteins. Following

72

three rinses with TBS, sections were placed in 0.5% acetic anhydride (in 0.1M Tris-HCl,

pH 8) for 10 min. Slides were rinsed three times with TBS and then incubated with 20

µg/mL proteinase K (Invitrogen) in TBS containing 2 mM CaCl2 for 20 min at 37ºC.

Sections were rinsed three times with TBS and then incubated in TBS at 4ºC for 5 min to

stop the digestion. The sections were then dehydrated through a graded series of ethanol,

incubated in chloroform for 20 min, rehydrated through decreasing concentrations of

ethanol, and then incubated in 2× SSC for 5 min. Sections were pre-hybridized with

hybridization buffer (2× SSC, 10% dextran sulfate, 0.01% sheared salmon sperm DNA,

0.02% SDS, 50% formamide) for 1 hr at 56ºC. For hybridization, DIG-labeled RNA

probes were diluted 1/200 to 1/400 in hybridization buffer and 40 µL of probe was

applied per slide. The slides were cover-slipped, heated at 95ºC for 5 min, and then

hybridized overnight at 56ºC. Cover-slips were removed by incubating the slides in 2×

SSC for 15 min. Non-hybridized probe was removed by digestion with 20 µg/mL

RNaseA (Fermentas) in 0.5M NaCl, 10 mM Tris-HCl, pH 8.0 for 30 min at room

temperature. Sections were rinsed in 2× SSC and then incubated in 50% formamide/1×

SSC for 1 hr at 56ºC to remove unbound probe. Sections were washed twice with 1× SSC

for 15 min and then rinsed with TBS. Slides were blocked in blocking buffer (1×

Blocking Reagent (Roche) diluted in 0.1M maleic acid, 0.15M NaCl, pH 7.5) for 30 min

at room temperature. Anti-DIG antibody conjugated to alkaline phosphatase (Roche) was

diluted 1/500 in blocking buffer and then applied to the sections for 1 hr at room

temperature. Following two 15 min rinses with TBS, sections were incubated with

detection buffer (0.1M NaCl, 0.1M Tris-HCl, pH 9.5) for 15 min. The alkaline phosphate

substrate NBT/BCIP was added and colour development was allowed to proceed

overnight at room temperature in the dark. Images were captured on a Leica DM6000 B

microscope using a Micropublisher 3.3RTV camera (Q Imaging Inc.) in conjunction with

OpenLab software (Improvision).

Detergent Insolubility Assays

One tenth volume of 10× detergent was added to either brain homogenate or cell

lysate for a final concentration of 0.5% Triton X-100 and 0.5% sodium deoxycholate.

Samples were mixed, incubated briefly on ice, and cell debris removed by spinning at

73

1000× g for 5 min at 4ºC. Samples were then spun at 120,000× g for 40 min at 4ºC using

a TLA-55 rotor and a Beckman TL-100 ultracentrifuge. Supernatant (S2) and pellet (P2)

samples were prepared in 1× loading buffer, boiled, and then analyzed by Western

blotting.

Detection of PrPres and Proteinase K Digestions

For detection of PrPres in ScN2a cells, cells were lysed with RIPA buffer in the

absence of protease inhibitors. Lysates were adjusted to 1 mg/mL and 200 µg total

protein was digested with 20 µg/mL proteinase K (a PK:protein ratio of 1:50) for 30 min

at 37ºC. Digestion was terminated by the addition of PMSF to a final concentration of 2

mM and incubation on ice for 15 min. PrPres was precipitated by the addition of sodium

phosphotungstic acid (4% w/v stock in 170 mM MgCl2, pH 7.4) to a final concentration

of 0.3% and incubation at 37ºC for 30 min. Pellets were recovered by spinning at

38,000× g for 40 min at 4ºC and then were resuspended in 1× LDS sample buffer and

boiled for 10 min. PrPres levels were then analyzed by Western blotting. For PK titration

experiments, PK was added to lysates or homogenates to various final concentrations,

incubated for 30 min at 37ºC, and digestions terminated by the addition of loading buffer

to 1× and subsequent boiling.

Cycloheximide Treatment of Cultured Cells

N2a or ScN2a cells were transfected with Shadoo plasmid for 6 hours and then

incubated overnight in medium containing serum. The next morning, the medium was

replaced with medium containing 30 µg/mL cycloheximide (Sigma) to inhibit protein

synthesis and incubated at 37ºC for various amounts of time. Cells were then lysed in

RIPA buffer and processed normally. Residual Shadoo levels were calculated by Western

blotting and densitometry (Scion Image) by comparison to a standard curve created from

serial dilutions of untreated (0 hr) Shadoo-transfected cells. Curve fitting and half-life

estimation was performed using GraphPad Prism software.

74

Inoculation of Mice with Prions

Mice for inoculations (C3H/B6 hybrids) were inoculated intracerebrally with 30

µL of 0.1% RML-infected brain homogenate diluted in PBS containing BSA (50 mg/mL),

penicillin (0.5 U/mL), and streptomycin (0.5 µg/mL). Age matched non-inoculated mice

were used as negative controls. Clinically ill mice were sacrificed and 10% brain

homogenates in PBS were made. Proteinase K digestions (50 µg/mL) were performed at

37ºC for 1 hr. Shadoo levels in brains were calculated by densitometry (Scion Image) and

comparison to a standard curve created from serial dilutions of non-infected brain

homogenate. TgCRND8 mice [382] on a C3H/C57BL6 outbred background were

sacrificed at clinical illness (~8 months) and 10% brain homogenates performed in 0.32M

sucrose containing protease inhibitors.

3.4 Results Prion protein domain architecture

The first two identified members of the prion protein family, PrP and Doppel,

possess similar C-terminal domains which consist of three α-helices and two short β-

strands. Indeed, the three-dimensional structures of these two proteins are very similar

[204, 221]. In contrast, the hypothetical Shadoo protein loosely resembles the flexibly

disordered N-terminal domain of PrPC (Figure 3.1). Like PrP, Shadoo has a series of N-

terminal positively-charged repeats. However, these are in the form of tetrarepeats in

Shadoo compared to the copper-binding octarepeats found in PrP. The series of five

tetrarepeats are rich in Arg, Gly, and Ala residues with the consensus sequence of GARG.

Bioinformatic analysis predicts that, like PrP and Doppel, Sho is modified by N-

glycosylation and the addition of a GPI anchor [197]. An alignment of human and mouse

PrP and Sho sequences demonstrates that the largest area of conservation between PrP

and Shadoo is found within the hydrophobic tract (Figure 3.1). Both PrP and Sho

hydrophobic tracts contain a palindromic sequence (AGAAAAGA in PrP,

AAAGAAAGAAA in Sho). An accumulating set of data suggests that the hydrophobic

tract in PrP comprises its bioactive site [123, 279, 280, 381] and considering the striking

similarity to Sho within this region, a thorough investigation of Shadoo structure and

function is warranted.

75

Mouse PrP G----GGWGQGGGTHNQWNKPSK----PKTNLKHVAGAAAAGAVVGGLGGYMLGSAMSRPMIHF Human PrP G----GGWGQGGGTHSQWNKPSK----PKTNMKHMAGAAAAGAVVGGLGGYMLGSAMSRPIIHF Mouse Sho GSARG-VRGGARGASRVRVRPAPRY---GSSLRVAAAGAAAGAAAGVAAGLATGSGWRRTSGPG Human Sho GSARGGVRGGARGASRVRVRPAQRYGAPGSSLRVAAAGAAAGAAAGAAAGLAAGSGWRRAAGPG * * . *: :*: :.:: *..*****..* .* **. *.

Δ62-77 Figure 3.1. Domain structure of PrP, ΔPrP, Doppel, and Shadoo. α-helices (A, B, C) are boxed. Unlike Doppel which is similar to the C-terminal α-helical domain of PrPC, Shadoo loosely resembles the N-terminal unstructured region of PrPC. Shadoo possesses an N-terminal Arg/Gly-rich series of tetrarepeats, a hydrophobic tract with strong similarity to PrPC (red), a single consensus site for N-glycosylation (CHO), and is predicted to be attached to the cell membrane by a GPI anchor. An alignment of PrP and Shadoo sequences with the T-COFFEE algorithm reveals that the principle region of homology between the two proteins is located within the Ala/Gly/Val-rich hydrophobic tract. Two residues that correspond to the N-termini of human PrP C1 endoproteolytic fragment are underlined. A Shadoo allele with a precise hydrophobic tract deletion (Δ62-77) used in this study is bracketed.

76

Generation of polyclonal antibodies against the putative Sho open reading frame

In order to begin to assess the existence and properties of the notional Shadoo

protein, polyclonal antibodies were first raised against a peptide epitope comprising

residues 86-100 of the predicted Sho protein. Affinity-purified antisera (using

immobilized Sho86-100 peptide) obtained from two rabbits (04SH-1 and 06SH-3) had

numerous cross-reactive bands present in Western blots of mouse brain homogenates

(Figure 3.2C). However, upon purification of 06SH-3 serum over a column with

immobilized full-length recombinant Sho [381], a much cleaner antibody (termed 06SH-

3a) was obtained. A distinct form of polyclonal antibody (06rSH-1) was raised in rabbits

against full-length recombinant Sho produced in E. coli [381]. This antibody has very

minimal cross-reactive species (Figure 3.2C) and recognizes an epitope contained within

residues 30 to 61 of Sho as demonstrated by the failure of 06rSH-1 to recognize a

ShoΔ30-61 deletion mutant (Figure 3.2B). As expected, the 06SH-3 antibody failed to

recognize a ShoΔ78-100 deletion mutant since this deletion interval contains the

immunogenic peptide epitope.

Cloning and expression of murine Shadoo in N2a cells

The hypothetical murine Sprn open reading frame was amplified from mouse

genomic DNA (since the Sho coding sequence is contained within a single exon) and

inserted into the pcDNA3 mammalian expression vector. N2a neuroblastoma cells were

transfected with the Shadoo expression construct and then analyzed by staining of non-

permeabilized cells with the 04SH-1 anti-Sho antibody. Transfection with an empty

vector plasmid reveals only moderate background staining (Figure 3.3A), likely reflecting

the cross-reactive species recognized by this antibody (Figure 3.2C). In contrast, staining

of cells transfected with a Sho-expressing plasmid reveals prominent labeling along the

periphery of cells. Because cells were not permeabilized, this indicates that Sho protein is

present on the cell surface of transfected cells. Similarly, a Sho deletion mutant

containing a precise deletion of the hydrophobic tract (ShoΔ62-77) is also present on the

cell surface and is expressed at similar levels to wild-type Sho.

Biochemical analysis of the Sho protein sequence predicts that Shadoo is tethered

to the cell membrane by a GPI anchor and is modified post-translationally by the addition

77

28

14

6Antibody: 06rSH-1 06SH-3

FLAG-ShoConstruct:

A

B

C- + - + PNGaseF

06SH-3 06rSH-1

148986450

36

16

- +

04SH-1Antibody:

Mature ShoDe-glycosylated Sho

- +

06SH-3a Figure 3.2. Construction of Shadoo polyclonal antibodies, epitope mapping, and analysis of specificity. A: Schematic representation of the domain structure of the murine Shadoo protein. The location of an N-terminal FLAG tag (inserted following residue 25) and the mapped epitopes of the anti-Sho antibodies are shown. B: Epitope mapping of anti-Sho antibodies. Lysates from N2a cells expressing N-terminally FLAG-tagged wild-type Sho or Sho deletion mutants were analyzed by Western blotting with the indicated antibodies. As expected, 06SH-3 (which was raised against a Sho86-100 peptide) does not recognize the Δ78-100 mutant whereas 06rSH-1 (which was raised against recombinant Sho) fails to recognize the Δ30-61 mutant illustrating its specificity for the N-terminus of Shadoo. C: Brain homogenate from a wild-type mouse with or without PNGaseF treatment was analyzed by Western blotting with the indicated purified anti-Sho polyclonal antibodies. Only faint cross-reactive species are observed with the 06SH-3a and 06rSH-1 antibodies.

78

Empty vector wt Sho ShoΔ62-77A

28

1714

0 0.2 0.5

PI-PLC (U)

Shadoo

B

1714

6

- - + + PNGaseF

C

28

17

14

6

- + - + PNGaseF

FLAG-ShoConstruct:

Un-glycosylated Shadoo

ShoC1 fragment

D

ShoShoC1 fragment

Figure 3.3. Biochemical characterization of murine Sho expressed in N2a cells. A: Cell surface expression of wild-type Sho and a mutant Sho allele lacking the hydrophobic tract in non-permeabilized transfected N2a cells as demonstrated by immunocytochemistry. Scale bar = 50 μm. B: Diminution of Sho signal in the cell lysates of Sho-transfected N2a cells following pre-treatment with increasing concentrations of PI-PLC. C: Western blot showing expression of a wild-type Sho allele in N2a cells with or without PNGaseF treatment. A lysate from cells transfected with empty vector is included to show antibody specificity. D: Western blot of lysates from N2a cells transfected with either wild-type or N107Q FLAG-tagged Sho constructs. The N107Q mutant is insensitive to PNGaseF and displays the same electrophoretic mobility as de-glycosylated wild-type Shadoo confirming the location of the N-glycosylation event at Asn107.

79

of N-glycans [197]. In order to confirm these inferences, biochemical analyses were

performed on lysates of N2a cells transfected with Sho. Treatment of transfected cells

with increasing concentrations of the GPI anchor-cleaving enzyme PI-PLC led to

decreasing amounts of Sho protein in the cell lysate fraction (Figure 3.2B), implying that

Shadoo is being released into the medium by PI-PLC. This indicates that Shadoo, like the

other two prion proteins, is anchored to the cell membrane by a GPI anchor. To assess N-

glycosylation, transfected cell lysates were treated with the enzyme PNGaseF to remove

Asn-linked sugars. Following treatment, Sho signal on a Western blot decreased in size

indicating that Shadoo is modified by N-glycans (Figure 3.2C). The lack of signal in

lysates from empty vector-transfected cells confirms the specificity of the 04SH-1

antibody. In all blot analyses using the 04SH-1 antibody, Sho signal increased following

treatment with PNGaseF. This may reflect a slight hindrance of antibody recognition by

N-glycosylation since the N-glycosylation site is close to the antibody epitope (Figure

3.2A). To confirm the location of the N-glycosylation event, an N107Q mutant was

generated. This mutant migrates at the same molecular weight as de-glycosylated wild-

type Sho and is insensitive to PNGaseF (Figure 3.2D). These results affirm the notion

that Sho is modified by N-glycosylation at a single site (Asn107).

Cumulatively, these results argue that the Shadoo open reading frame is capable

of forming a stable protein that is modified by a single N-glycosylation event and is

present at the cell surface by virtue of a GPI anchor. The size of de-glycosylated Shadoo

(~9 kDa) is in good agreement with its predicted molecular weight suggesting that full-

length Sho protein is being generated in transfected N2a cells.

Analysis of Shadoo expression in mouse brains

Although experiments using the cloned Shadoo open reading frame in N2a cells

suggest that Shadoo is indeed a real protein, they do not prove the existence of Shadoo

protein in vivo. For instance, it is possible that Sprn is simply a pseudogene that is not

actually expressed. To address this issue, α-Sho antibodies were used to examine mouse

brain homogenates for the presence of Shadoo protein. Two distinct anti-Sho antibodies

recognizing different epitopes (04SH-1 and 06rSH-1) reveal identical bands

corresponding to Shadoo protein in both neonatal (P1 and P2) and adult mouse brains

80

(Figure 3.4B). These bands are PNGaseF-sensitive and migrate at the predicted molecular

weight for authentic Shadoo protein confirming for the first time the existence of Sho in

the mouse CNS. Sho expression begins to appear at embryonic day 16 whereas PrP is

present throughout embryonic development (Figure 3.4A). Sho signal was also found in

membrane-enriched preparations from mouse brains (Figure 3.4C) corroborating its

membrane anchorage. As expected, signal for PrPC was also found in the membrane

fraction.

Endoproteolytic processing and secretion of Shadoo

In PrPC, a well-documented endoproteolytic cleavage event occurs just N-terminal

to the hydrophobic tract to generate a membrane-bound C1 fragment [232, 233, 383, 384].

To investigate whether Shadoo undergoes a similar endoproteolytic processing event,

N2a cells which stably express Shadoo with an N-terminal FLAG tag (inserted following

residue 25—Figure 3.2A) were created. Lysates from these cells revealed two prominent

Shadoo bands following treatment with PNGaseF and probing with a C-terminal anti-Sho

antibody (Figure 3.5A). This phenomenon was also observed with transiently-transfected

wild-type Sho (Figure 3.3C). Since cell lysates were prepared in the presence of protease

inhibitors, it seems likely that an endoproteolytic processing event analogous to the C1

processing of PrPC figures in the biogenesis of Sho. Consequently, the cleaved Sho

fragment was termed ShoC1. Based on the size of the ShoC1 fragment (~6 kDa), it is

likely that, similar to PrPC, the C1 cleavage event occurs near the start of the hydrophobic

tract. An anti-FLAG antibody failed to detect the ShoC1 fragment in cell lysates

confirming that the cleavage event releases an N-terminal peptide. Accordingly, a faint

band corresponding to the ShoN1 fragment was detected in the conditioned media but not

the cell lysate of FLAG-Sho-expressing cells (Figure 3.5B). Similarly, the PrP N1

fragment can be found in the conditioned medium of HEK293 cells over-expressing wild-

type PrP [233, 384]. A faint band with a molecular weight in between that of full-length

and C1-truncated Sho is also observed following PNGaseF treatment in cell lysates

(Figure 3.5A). It is possible that this band is analogous to the C2 cleavage event that

occurs in PrP in the vicinity of residue 90 [232, 235]. Strong signal corresponding to full-

length Sho was also observed in the conditioned medium of FLAG-Sho-expressing cells

81

A

20

25

37

10 11 12 13 14 15 16 17Embryonic day

Shadoo

PrP

C

28

1714

Mouse Brain Membranes

- + PNGaseF

Shadoo

28

1714

38PrP

C1

B

49

38

28PrP

Shadoo22

16

PNGaseF- -+ ++-P1 P2 Adult

2216

Shadooα-Sho(86-100)peptide

α-rSho

α-PrP

Figure 3.4. Analysis of mouse Shadoo in tissue preparations. A: Expression of Sho in whole mouse embryos as assessed by Western blotting. Sho is expressed beginning at embryonic day 16 whereas PrP is expressed throughout development. B: Postnatal expression of Sho in neonatal (P1 or P2) and adult mouse brains with or without PNGaseF treatment assessed by Western blotting using 14% Tris-Glycine (α-rSho) or 4-12% NuPAGE (α-Sho(86-100) peptide and α-PrP) gels. Full-length species before and after enzymatic treatment are bracketed. C: Sho is present in a membrane-enriched fraction prepared from mouse brains. Membrane preparations with or without PNGaseF treatment were analyzed by Western blotting. Sho was detected using either α-Sho peptide polyclonal 04SH-1 (A-C) or α-recombinant Sho polyclonal 06rSH-1 (B), while PrP was probed with single chain antibodies D13 (A-B) or D18 (C).

82

A B

17

14

6ShoC1Fragment

06rSH-1 06SH-3aAb:

- + - +Mouse Brains

PNGaseF

C

20

15

10

3.5α-FLAG

M2α-Sho

06rSH-1α-Sho

06SH-3

D

- + -PNGaseF:6 ShoN1

Fragmentα-FLAG

M2

28

17

14

- + -

α-Sho 06SH-3a

- + -

α-FLAG M2

PNGaseF

ShoC1Fragment6

Figure 3.5. ‘C1-like’ endoproteolytic processing of Shadoo in N2a cells and mouse brains. A: Endoproteolysis and shedding of Shadoo protein in N2a cells stably expressing FLAG-tagged Sho. Western blots of cell lysates or conditioned media with or without PNGaseF treatment probed with the indicated antibodies. The ShoC1 fragment is not detected by the anti-FLAG M2 antibody confirming that the cleavage event is N-terminal. Shadoo protein can also be found in the conditioned media confirming that a proportion of Shadoo, like other GPI-anchored proteins, is shed from the cell membrane. B: A band corresponding to the ‘ShoN1’ fragment is detected with the M2 antibody in the conditioned medium but not the cell lysates of cells stably expressing FLAG-tagged Sho. C: Endoproteolysis of mutant Shadoo proteins. Western blots of PNGaseF-treated cell lysates prepared from N2a cells transfected with the indicated FLAG-tagged Shadoo mutants and probed with the indicated antibodies. Based on the size of the ShoC1 fragment, cleavage likely occurs near the beginning of the hydrophobic tract. D: Shadoo endoproteolysis in vivo. C1-like endoproteolysis of Shadoo is also observed by Western blotting of detergent-extracted brain homogenate prepared from a wild type mouse.

83

(Figure 3.5A) indicating that, like many GPI-anchored proteins including PrPC [385], a

proportion of Sho is secreted from the cell.

In order to further map the ShoC1 cleavage site, the endoproteolytic processing of

various FLAG-tagged Shadoo deletion mutants was investigated. As expected the anti-

FLAG antibody which recognizes the N-terminus failed to detect the ShoC1 fragment for

any of the mutants (Figure 3.5C). Similarly, the anti-Sho antibody 06rSH-1 (which

recognizes residues 30-61) also failed to detect cleaved Sho. This reinforces the concept

that the cleavage event occurs near the beginning of the hydrophobic tract at residue 62.

The C-terminal anti-Sho antibody 06SH-3 was able to detect the ShoC1 fragment in the

Δ30-61 and Δ62-77 mutants. In support of the idea that the cleavage event occurs in the

vicinity of residue 62, deletion of residues 30-61 was unfavourable for endoproteolytic

processing but did not totally abrogate cleavage (Figure 3.5C). Unexpectedly, deletion of

residues 62-77 (residues which are C-terminal to the predicted cleavage site) did not alter

the size of the ShoC1 fragment as would be expected if cleavage was occurring in a site-

specific manner. This suggests that cleavage may be based primarily on the length of the

protein and not on the presence of a specific cleavage site. Unfortunately, the 06SH-3

antibody does not recognize the Δ78-100 construct which would be useful for testing this

inference. Deletion of Shadoo residues 101-120 completely abrogates cleavage (Figure

3.5C) suggesting that, since this region is clearly C-terminal to the cleavage site in wild-

type Sho, it may function as a docking site for the unidentified protease.

To confirm that C1-like processing of Sho also occurs in vivo, wild-type mouse

brain homogenates were analyzed for the presence of the ShoC1 fragment. A band

corresponding to the ShoC1 fragment was detected in brain homogenates with the C-

terminal 06SH-3a antibody but not the N-terminal 06rSH-1 antibody (Figure 3.5D)

confirming that endoproteolytic processing figures in the biogenesis of Shadoo both in

tissue culture and in brains.

Localization of Sprn and Sho expression in mouse brains

In order to localize Sprn mRNA and Sho protein expression within the CNS, both

in situ hybridization (ISH) and immunohistochemistry (IHC) were performed. For ISH,

digoxigenin (DIG)-labeled anti-sense Sho riboprobes were generated and used to

84

interrogate wild-type mouse brain sections. Strong Sprn mRNA signal was obtained in

hippocampal neurons including the CA1-CA3 pyramidal neurons and the dentate gyrus

(Figure 3.6B). No signal was obtained when a sense-strand riboprobe was utilized (Figure

3.6A). Positive staining for Sho protein was also observed in the hippocampus by IHC

using the 04SH-1 antibody on methacarn-fixed brains (Figure 3.6D) that was absent

when the antibody was pre-incubated with its immunogenic peptide (Figure 3.6C). In

order to compare PrPC and Sho expression profiles, parallel analyses were also performed

using anti-PrP antibodies. Staining for mouse PrP (MoPrP) in the hippocampus of wild-

type mice revealed widespread staining (Figure 3.6F) whereas no staining was observed

in Prnp0/0 mice (Figure 3.6E). Similar results were obtained for PrP in Tg(SHaPrP)7

transgenic mice which over-express hamster PrP (HaPrP) from a cosmid transgene [126,

386]. For these mice, the HaPrP-specific 3F4 antibody was used and revealed strong

widespread staining in the hippocampus (Figure 3.6H) that was absent in wild-type mice

(Figure 3.6G). These results argue that Sho is expressed in hippocampal neurons and has

a more restricted pattern of expression than PrPC within the hippocampus.

Another principle region of Sho expression is the cerebellum. Strong signal for

Sprn mRNA (Figure 3.7B) and Sho protein (Figure 3.7D) in Purkinje cells was obtained

by ISH and IHC, respectively. Positive staining disappeared when the appropriate

negative controls were performed (Figure 3.7A, C). No staining of the granule cell layer

or the cerebellar white matter was observed with the 04SH-1 (Figure 3.7D) or 06rSH-1

(data not shown) antibodies. For PrPC, strong staining was observed throughout the

cerebellum for both MoPrP in wild-type mice or HaPrP in Tg(SHaPrP)7 mice (Figures

3.7F and 3.7H, respectively) that was absent in the negative control sections (Figure 3.7E,

G). Notably, little to no staining for PrP was observed in the Purkinje cells. Similar to the

hippocampus, Sho has a much more restricted pattern expression than PrPC within the

cerebellum.

Although the most prominent Sho signals were obtained in the hippocampus and

cerebellum, signal for both Sprn mRNA and Sho protein was also observed in other

regions of the brain including the cerebral cortex, the thalamus, and the medulla as

demonstrated by ISH and IHC (Figure 3.8). In order to determine what type(s) of cell in

the CNS expresses Shadoo, double immunofluorescent labeling on methacarn-fixed wild-

85

α-Sho mRNA

α-Sho

α-HaPrP

α-MoPrP

A

HG

E F

DC

B

Wild-type

Wild-type

Wild-type

Wild-type

Wild-type

Wild-type Tg(SHaPrP)

Prnp0/0

Negative Controls Figure 3.6. Expression of Sprn mRNA and Shadoo protein in the hippocampus. Wild-type C57/B6 mice are presented in all sections with the exception of B6 congenic Prnp0/0 (panel E) and Tg(SHaPrP)7 mice (panel H). Panels A, C, E, and G in the left-hand column comprise negative controls for data presented in the right-hand columns. A-B: Detection of Sprn mRNA in hippocampal neurons by in situ hybridization with either a Sho sense-strand (A) or anti-sense (B) RNA probe. Sections are not counter-stained and blue staining from NBT/BCIP substrate represents hybridization to Sprn mRNA. C-D: Detection of Sho protein in hippocampal neurons by immunohistochemistry with the anti-Sho antibody 04SH-1 (D). Pre-incubation of the antibody with its immunogenic peptide (Sho86-100) abolishes staining (C). E-F: Detection of mouse PrPC in hippocampal neurons by immunohistochemistry with the anti-PrP antibody 7A12 (F). No staining is observed with this antibody in Prnp0/0 mice (E). G-H: Detection of over-expressed hamster PrPC in hippocampal neurons of Tg(SHaPrP)7 mice by immunohistochemistry with the anti-PrP antibody 3F4 (H). The specificity of this antibody for transgenically-expressed hamster PrPC is demonstrated by the absence of staining in wild-type mice (G). Note the widespread expression pattern of PrPC compared to the more restricted expression pattern of Sho in the hippocampus. Scale bar = 100 µm.

86

α-Sho

α-Sho mRNA

α-MoPrP

α-HaPrP G

E

C

B

H

F

D

A

Wild-type

Wild-type

Wild-type

Wild-type

Wild-type

Wild-type Tg(SHaPrP)

Prnp0/0

Negative Controls Figure 3.7. Expression of Sprn mRNA and Shadoo protein in the cerebellum. Wild-type C57/B6 mice are presented in all sections with the exception of B6 congenic Prnp0/0 (panel E) and Tg(SHaPrP)7 mice (panel H). Panels A, C, E, and G in the left-hand column comprise negative controls for data presented in the right-hand columns. A-B: Detection of Sprn mRNA in Purkinje cells by in situ hybridization with either a Sho sense-strand (A) or anti-sense (B) RNA probe. Sections are not counter-stained and blue staining from NBT/BCIP substrate represents hybridization to Sprn mRNA. C-D: Detection of Sho protein in Purkinje cells by immunohistochemistry with the anti-Sho antibody 04SH-1 (D). Pre-incubation of the antibody with its immunogenic peptide (Sho86-100) abolishes staining (C). E-F: Detection of mouse PrPC in the cerebellum by immunohistochemistry with the anti-PrP antibody 7A12 (F). No staining is observed with this antibody in Prnp0/0 mice (E). G-H: Detection of over-expressed hamster PrPC in the cerebellum of Tg(SHaPrP)7 mice by immunohistochemistry with the anti-PrP antibody 3F4 (H). The specificity of this antibody for transgenically-expressed hamster PrPC is demonstrated by the absence of staining in wild-type mice (G). Note the expression of Sho in the Purkinje cell layer of the cerebellum (B, D: white arrows) compared to the lack or relative paucity (F, H: black arrows) of PrPC staining in these cells. Scale bar = 100 µm.

87

type mouse brain tissue was conducted. The vast majority of Shadoo signal in the

cerebral cortex, thalamus, and medulla was coincident with signal for the neuronal

marker neurofilament H (NFH) indicating that neurons are the primary site of Sho

expression in the CNS (Figure 3.8E-G, L-N, and S-U). No evidence for Sho expression in

the other types of CNS cells (namely astrocytes, oligodendrocytes, and microglia) has

been obtained in either ISH or IHC studies.

To more precisely compare the expression profiles of PrPC and Sho within the

hippocampus and cerebellum, higher magnifications of stained methacarn-fixed mouse

brain sections were analyzed. In the hippocampal CA1 pyramidal neurons, strong Shadoo

signal is observed both in the cell body and the apical dendritic processes (Figure 3.9A-

C). Staining of the dendritic processes with the 06rSH-1 antibody was less pronounced

(Figure 3.9C) although further experiments using more concentrated antibody and

formalin-fixed tissue have confirmed that 06rSH-1 reactivity is present in these processes

(data not shown). In contrast, PrPC is under-represented in hippocampal CA1 neuron cell

bodies and, by virtue of a ‘negative staining’ effect, is revealed as being largely absent in

the apical dendritic processes in either wild-type (Figure 3.9D) or Tg(SHaPrP)7 mice

(Figure 3.9E). This suggests that PrPC and Sho exhibit a partial interlocking pattern of

expression in the hippocampus. This concept is further supported by double

immunofluorescent labeling in which no co-localization of PrPC and Sho is observed in

hippocampal CA1 neuronal cell bodies or dendritic processes (Figure 3.9F). Similar

analyses were also performed in the cerebellum. Strong Shadoo signal is observed in both

the cell bodies and dendritic arborizations (which project into the molecular layer of the

cerebellum) of Purkinje cells (Figure 3.9G-I). Staining of Purkinje cells and their

processes in both methacarn and formalin-fixed brains has been observed with all anti-

Sho antibodies tested to date. In contrast, Purkinje cells and their processes do not stain

positive for PrPC in wild-type mice (Figure 3.9J) and there is a relative paucity of staining

for PrPC in Purkinje cell bodies in mice over-expressing HaPrP (Figure 3.9K). These

results suggest that, similar to the hippocampus, PrPC and Sho have a reciprocal pattern

of expression in the cerebellum. To confirm this notion, double immunofluorescent

labeling was performed and demonstrates a lack of co-localization between PrPC and Sho

in the cerebellum (Figure 3.9L).

88

Thalamus

H

J

I

K

α-Sho mRNA

α-Sho

Shadoo NFH Merge

L M N

Medulla

O

Q

P

R

α-Sho mRNA

α-Sho

Shadoo NFH Merge

S T U

Cerebral Cortex

A B

C D

α-Sho mRNA

α-Sho

Shadoo NFH Merge

E F G

89

Figure 3.8. Neuronal expression of Shadoo in the cerebral cortex, thalamus, and medulla. Panels A-G represent the cerebral cortex, panels H-N the thalamus, and panels O-U the medulla. Wild-type C57/B6 mice are presented in all sections. Panels A, C, H, J, O, and Q in the left-most column comprise negative controls for data presented in the adjacent column. In situ hybridization: Panels A-B, H-I, and O-P, hybridizations with a Shadoo sense-strand (A, H, O) or anti-sense (B, I, P) RNA probe. Sections are not counter-stained and blue staining from NBT/BCIP substrate represents hybridization to Sprn mRNA. Immunohistochemistry: panels C-D, J-K, and Q-R. Anti-Sho antibody 06rSH-1 was used with (C, J, Q) or without (D, K, R) pre-incubation with recombinant Sho. Positive staining is denoted by brown signal. Consistent results were obtained when the anti-Sho antibody 04SH-1 was used (data not shown). Fluorescent immunohistochemical double labeling: panels E-G, L-N, and S-U. Sections were simultaneously labeled with anti-Shadoo antibody 06rSH-1 (green) and anti-neurofilament H (NFH) antibody SMI-32 (red). Shadoo (E, L, S), NFH (F, M, T) and merged (G, N, U) fluorescent images are shown. The merged images confirm that Shadoo signal is associated with NFH-positive cells (i.e. neurons). Scale bar = 20 µm (panels A-G, L-U) or 50 µm (panels H-K).

90

Although the above results suggest that PrPC and Sho have a non-overlapping

pattern of expression in the hippocampus and cerebellum, this is clearly not the case

throughout all areas of the brain. For instance, co-localization of Shadoo and PrPC is

observed in neuronal cell bodies within the cerebral cortex (Figure 3.9M-O) and PrPC is

also present in the medulla and thalamus which are areas of strong Sho expression

(Figure 3.8) confirming that Sho and PrPC expression can overlap in various regions of

the brain.

The overlapping profile of Sho ISH and IHC within the CNS is persuasive

evidence that veritable Sprn mRNA/Sho protein is being detected. However, as

evidenced by Western blotting of mouse brain homogenates, both the 04SH-1 and 06SH-

3 antibodies recognize numerous cross-reactive species (Figure 3.2C) and it is highly

likely that some of the signal obtained by IHC (such as the staining of the molecular layer

of the cerebellum with the 04SH-1 antibody in Figure 3.7D) represents staining of cross-

reactive proteins. To address this concern, Sho IHC experiments were repeated using the

06rSH-1 and 06SH-3a antibodies which are largely devoid of cross-reactive species

(Figure 3.2C). Consistent results to those presented above were obtained with both

antibodies using either methacarn-fixed brains or formalin-fixed brains (following heat-

induced epitope retrieval), namely strong expression of Sho in CA1 hippocampal neurons,

Purkinje cells, neurons in the cerebral cortex, etc. (data not shown). Furthermore, no

signal was observed in IHC experiments using 06rSH-1 antibody following pre-

incubation with recombinant Sho. These experiments reinforce the validity of the IHC

experiments using the 04SH-1 or 06SH-3 antibodies presented in Figures 3.6, 3.7, and 3.9.

A further series of control experiments will be feasible upon completion of Sprn

knockout mice (Nathalie Daude and David Westaway, in progress), to verify these

inferences.

Areas of Sho expression outside of the brain

Gene expression studies using RT-PCR suggest that Shadoo expression is

restricted to the brain [197] with perhaps lower levels of expression in the intestine and

testis [380]. Although the highest levels of PrPC are found in the CNS, PrPC expression

can also be found in tissues including the retina [387], spleen [182], and skeletal muscle

91

G

A

α-Shadoopeptide04SH-1

K

ED

JI

α-recomb.Shadoo06rSH-1

α-MoPrP α-HaPrP

CB

α-Shadoopeptide06SH-3

H L

F

α-PrPα-Shadoo

PrPC Shadoo MergeM N O

Figure 3.9. Reciprocal and overlapping expression of Sho and PrPC in the CNS. A–F: CA1 hippocampal neurons of adult mice probed as indicated. The signal for Sho immunohistochemistry in apical dendrites of wild-type mice (A–C) has an equivalent in a ‘negative image’ (white brackets) in the molecular layer of the neuropil imaged for either mouse PrPC in wild-type mice (D) or hamster PrPC in Tg(SHaPrP)7 mice (E). Apical dendrites were less intensely stained with the anti-Sho 06rSH-1 antibody (C, black arrowhead) G-L: Analogous analyses are presented for Purkinje cells. Somatodendritic localization of Sho in Purkinje cell bodies and dendritic arborizations (G–I) is contrasted by a reciprocal ‘negative image’ in molecular layer of the neuropil imaged for either mouse PrPC in wt mice (J) or hamster PrPC in Tg(SHaPrP)7 mice (K). Note the complete absence of PrPC staining in cell bodies (J, black arrowheads). Double immunofluorescent labeling of mouse brains confirms that Shadoo and PrPC do not co-localize in either the hippocampus (F) or cerebellum (L). M–O: The cerebral cortex probed simultaneously with anti-Sho (06rSH-1, green) and anti-PrP (8H4, red) antibodies. Overlapping PrPC and Sho expression is observed in neurons of the cerebral cortex, including colocalization in cell bodies (N, white arrowheads). Scale bar, 25 μm (A–F), 10 μm (G–L) or 20 μm (M–O).

92

[183, 388] as well as on immune cells [184, 185, 389]. Consequently, the presence of

Shadoo protein in areas outside the brain was assessed by Western blotting and

immunohistochemistry.

Western blotting using the 06rSH-1 anti-Sho antibody reveals that Shadoo protein

can also be found in homogenates prepared from the spinal cord and eyes of a wild-type

mouse (Figure 3.10A). However, levels of Shadoo in the spinal cord and eye are clearly

much lower than those present in the brain. Other tissues that were tested for Sho

expression were the thymus, heart, kidney, lung, spleen, liver, skeletal muscle, large

intestine, small intestine, and the testis. No Shadoo protein in any of these tissues was

found by Western blotting, even following long exposures (Figure 3.10B). One possible

exception is the liver, in which a faint band at approximately the correct molecular

weight for Shadoo was identified. Further studies are needed to validate the existence of

Sho protein in the liver. It remains possible that Shadoo protein is indeed present in some

of the above tissues, although at levels which fall beyond the limits of detection for the

currently-available anti-Sho antibodies.

In order to localize Shadoo expression in the spinal cord and eye, IHC on

methacarn-fixed tissue was performed. Strong Shadoo staining was observed in motor

neurons of the spinal cord with both the 06rSH-1 and 04SH-1 anti-Sho antibodies (Figure

3.10C). This result is in agreement with the neuronal pattern of Shadoo expression in the

brain. Shadoo staining was also obtained in the retina in which strong Sho protein signal

was found within the inner membrane stacks of the photoreceptor cells in both human

and mouse retinas (Figure 3.10D). This result confirms that, like PrPC, Shadoo is present

in the retina.

Functional studies on Shadoo in cerebellar granule neurons

PrPC is known to be neuroprotective against the cerebellar toxicity of either

transgenically-expressed Doppel or ΔPrP [217, 289]. Based on this phenomenon, a

cellular assay which utilizes primary cerebellar granule neurons (CGNs) cultured from

Prnp0/0 mice was developed in order to be able to rapidly analyze neurotoxicity and

neuroprotective determinants in Doppel and PrPC, respectively [271]. In this assay,

increased levels of apoptosis are observed following transfection with either Doppel or

93

06rSH-1 04SH-1

Spinal Cord

06rSH-1

Retina

Mouse

Human

C D

22

BrainSpinalcord Eye

Sho

Ab: 06rSH-1

A B

22

Figure 3.10. Expression of Shadoo in the spinal cord and retina. A: Western blot of homogenates prepared from the indicated tissues of a wild type mouse probed with an anti-Sho antibody. In addition to the brain, Shadoo protein signal is also observed in the spinal cord and the eye. B: Long exposure of a Western blot of homogenates prepared from the indicated non-neuronal tissues of a wild type mouse probed with an anti-Sho antibody. Shadoo is not expressed in any of the tissues tested with the possible exception of the liver. C: Immunohistochemistry using the indicated anti-Sho antibodies on methacarn-fixed spinal cords from a wild type mouse. Prominent staining of Shadoo protein in motor neurons is observed. Scale bar = 160 µm (top panels) or 40 µm (lower panels). D: Immunohistochemistry on methacarn-fixed mouse and human retina sections using the 06rSH-1 anti-Sho antibody. Strong Shadoo protein signal is observed in the inner membrane stacks of the photoreceptor cells in both mouse and human retinas (arrows). Scale bar = 40 µm.

94

PrPΔ32-121 plasmids compared to empty vector controls. As is the case in vivo, co-

transfection of a wild-type PrP plasmid is effective at blocking this toxicity. Because

Shadoo has similar biochemical properties to PrPC and shares homology to PrP’s

hydrophobic domain, a region implicated in PrP neuroprotection [279, 280, 361, 381,

390], the ability of Shadoo to engender a PrPC-like neuroprotective activity was tested

using the CGN assay.

As expected, an approximate two-fold increase in the number of transfected cells

(as defined by GFP expression) undergoing apoptosis (as defined by cell staining with an

activated caspase-3 antibody) was obtained in cells transfected with a Doppel plasmid

compared to cells transfected with empty vector (Figure 3.11A). Remarkably, co-

transfection of Doppel with a wild-type Shadoo plasmid restored toxicity to baseline

levels (Figure 3.11A). This suggests that Shadoo possesses a PrPC-like neuroprotective

activity against Doppel. Because the hydrophobic domain of PrP has been implicated in

its neuroprotective function, a Shadoo allele bearing a precise deletion of the

hydrophobic tract (residues 62-77) was also tested for neuroprotective activity. Co-

transfections of Doppel with ShoΔ62-77 were not significantly different from Doppel

single transfections indicating that the hydrophobic tract of Shadoo is required for its

neuroprotective behaviour (Figure 3.11A). The deletion of residues 62-77 in Shadoo does

not alter its subcellular localization (Figure 3.3A) or its expression level (Figure 3.2B)

arguing that its impaired function is not simply due to a trivial deficiency in expression or

to improper targeting. The ability of Shadoo to protect against a toxic PrPΔ32-121 allele

was also assessed. As predicted, baseline toxicity levels increased approximately two-

fold upon transfection with PrPΔ32-121 plasmid (Figure 3.11B). In contrast, co-

transfection with a wild-type Shadoo plasmid completely abrogated ΔPrP-induced

toxicity (Figure 3.11B). Co-transfection of ShoΔ62-77 with PrPΔ32-121 did not reduce

toxicity levels confirming that Shadoo’s hydrophobic tract is necessary for

neuroprotection. Cumulatively, these results suggest that Shadoo possesses a PrPC-like

neuroprotective activity and implies possible functional redundancy between Shadoo and

PrPC.

Both Shadoo and PrPC are neuronal GPI-anchored proteins. In order to confirm

that neuroprotection in the CGN assay is not simply conferred by over-expression of a

95

A B

E

22

16

- + - + PNGaseF

MouseBrain

Prnp0/0

CGN’s

Mature Sho

De-glycosylated Sho

36

FShadoo PrPΔ32-121 Merge

C D

******

***

***n.s.

***

n.s.

******

***

Figure 3.11. Neuroprotective activity and Sho expression in CGN cells. A-D: Toxicity assays in Prnp0/0 CGN cells [271]. Results of co-transfections of wild-type Sho, internally-deleted Sho (ShoΔ62-77) and a pBUD.GFP control with either a toxic Doppel (A) or PrPΔ32-121 plasmid (B). Determinations (% cells ± s.e.m undergoing apoptosis) reflect the results of two or more triplicate transfections of independent batches of Prnp0/0 CGNs. Like PrPC, wild-type Sho had potent neuroprotective activity against Doppel (p<0.001, n=10) and PrPΔ32-121 (p<0.001, n=6) whereas ShoΔ62-77 co-transfections were not significantly different from either Doppel or PrPΔ32-121 single transfections. In contrast, Thy-1 was not protective against Doppel (C) and the ShoΔ62-77 allele was not toxic on its own (D). E: Western blot analysis using 06rSH-1 detects Sho in mouse brain homogenate but not in a normalized loading of lysate derived from mouse Prnp0/0 CGNs. F: Immunocytochemistry on non-permeabilized CGNs co-transfected with non-fluorescent Sho and PrPΔ32-121 plasmids (3:1 ratio). Sho was detected with 06rSH-1 and PrPΔ32-121 with 8H4. Both Sho and PrPΔ32-121 are expressed in a single cell and co-localization is observed in the cell body. Scale bar = 15 µm.

96

GPI-anchored protein, co-transfections of Doppel and Thy-1 were performed. Thy-1 is a

well-studied GPI-anchored protein that is expressed on neurons and involved in processes

such as T cell activation [391]. No significant differences in toxicity were obtained

between Doppel single transfections and Doppel/Thy-1 co-transfections (Figure 3.11C).

This indicates that over-expression of a GPI-anchored protein is not sufficient to confer

neuroprotection against Doppel and suggests that the observed neuroprotective activities

of Shadoo and PrPC are specific.

Deletion of the hydrophobic tract in PrP is sufficient to activate the protein to a

pro-apoptotic state [279, 280, 381] and the addition of the PrP hydrophobic tract to the N-

terminus of Doppel is sufficient to prevent Doppel-induced toxicity [361]. Therefore, it

was tested whether deletion of Shadoo’s hydrophobic tract is also sufficient to confer

spontaneous neurotoxic behaviour. No significant differences in toxicity were observed

between empty vector controls and either wild-type Shadoo or ShoΔ62-77 (Figure 3.11D).

This result implies that deletion of Shadoo’s hydrophobic tract does not activate the

molecule to a pro-apoptotic state and implies that the C-terminal α-helical domain

present in both Doppel and PrPC is necessary for spontaneous toxicity.

The ability of Doppel and ΔPrP to elicit toxicity in transfected CGNs implies that

there is no compensatory endogenous neuroprotective activity in these cells. Indeed, no

toxicity is observed when wild-type CGNs (expressing PrPC) are transfected with Doppel

[271]. To ensure that no endogenous Shadoo (which would theoretically prevent toxicity)

is present in Prnp0/0 CGNs, a Western blot on lysates obtained from cultured CGNs was

performed. While Shadoo signal was readily detected in homogenates from whole mouse

brain, no Shadoo signal was observed in the CGN lysate (Figure 3.11E). This result also

comprises an additional negative control for the immunohistochemical analysis of Sho in

the brain.

Co-localization of Shadoo and PrPΔ32-121 in transfected CGNs was observed by

immunofluorescence (Figure 3.11F). Although co-localization does not necessarily imply

a physical interaction between Sho and PrPΔ32-121, it suggests that they inhabit similar

membrane microdomains within the cell. This argues that Sho and PrPΔ32-121 (and

likely Doppel) may be capable of tapping into similar biochemical pathways and that

97

perturbations within this pathway may potentially explain Doppel and ΔPrP-induced

neurotoxicity.

Shadoo in Prnp0/0 mice

Since Shadoo and PrPC share functional properties, it was assessed whether or not

their expression is cross-regulated. No changes in Shadoo protein levels as determined by

Western blotting were observed between wild-type and Prnp0/0 mice (Figure 3.12A).

Furthermore, no change in the distribution of Shadoo expression was found within the

brains of Prnp0/0 mice (Figure 3.12B-G). Thus, Shadoo is not up-regulated in response to

PrPC deficiency. However, because Shadoo and PrPC have functional redundancy, it

remains possible that the lack of a strong phenotype in Prnp0/0 mice is due to a

compensatory activity elicited by endogenous levels of Sho. This may be particularly true

in regions of the brain with overlapping PrPC/Sho expression such as the cerebral cortex

(Figure 3.9O).

Shadoo in prion-infected mice

In order to assess the biochemical properties of Shadoo in prion-infected mice,

wild-type mice (on a mixed C3H/C57B6 background) were inoculated intracerebrally

with the RML strain of mouse-adapted scrapie prions. Inoculated mice were allowed to

progress to clinical illness (approximately 170 days post-inoculation) and then sacrificed.

Half-brains were homogenized and Shadoo levels assessed by Western blotting.

Unexpectedly, Shadoo levels were strongly decreased in prion-infected brains compared

to non-inoculated control brains (Figure 3.13A). This result was observed with two

independent anti-Sho antibodies confirming its validity. When normalized to the decrease

in signal for the GPI-anchored protein Thy-1 (which likely reflects the neuronal loss that

occurs in prion disease), Shadoo levels were decreased to 12.1 ± 2.8% the levels

observed in non-inoculated brains (Figure 3.13B). To examine the specificity of this

phenomenon, levels of other neuronal proteins in prion-infected brains were examined.

No changes in either NSE (neuron-specific enolase, a general marker for neurons) or

calbindin (a marker for Purkinje cells) levels were observed (Figure 3.13C). This argues

that the down-regulation of Shadoo protein cannot be explained solely on the basis of

98

Shadoo

PrP

Actin

Prnp+/+ Prnp0/0A Prnp+/+ Prnp0/0

B

D E

F G

C

Figure 3.12. No change in Shadoo expression or distribution in PrP knockout brains. A: Western blot of wild-type (Prnp+/+) or PrP knockout (Prnp0/0) mouse brain homogenates probed with antibodies to Shadoo (06rSH-1), PrP (D18), and actin. There is no apparent change in Shadoo levels between the two genotypes. B-G: Shadoo immunohistochemistry on Prnp+/+ or Prnp0/0 methacarn-fixed mouse brain sections. Panels B-C represent the whole brain, panels D-E the hippocampus, and panels F-G the cerebellum. Panels B-C were stained with 06rSH-1 and panels D-G with 04SH-1. No obvious change in Shadoo distribution is observed between the two genotypes. Scale bar = 500 µm (panels B-C), 100 µm (panels D-E), or 50 µm (panels F-G).

99

Shadoo(06rSH-1)

Actin

PrPres

Non-inoculated RML-inoculated

Thy-1

Shadoo(06SH-3a)

Uninfected Infected0.0

0.2

0.4

0.6

0.8

1.0

1.2 ShadooThy-1

Rel

ativ

e Le

vel

******

***

A B

CRML-infectedNon-infected

NSE

Calbindin

Synaptophysin

TgCRND8Non Tg

Shadoo

APP

Actin

D

Figure 3.13. Reduced Sho levels in clinically ill prion-infected mice. A: Western blot of homogenates prepared from the brains of non-inoculated or clinically ill (average of 172 days post-inoculation) RML prion-inoculated mice (C3H/C57B6 background). Sho protein levels are notably reduced in prion-infected brains. Levels of the GPI-anchored protein Thy-1 are shown for comparison purposes. B: Quantitation of Sho (06rSH-1) and Thy-1 blot signals in panel A by densitometry. Sho levels in prion infected brains (normalized against Thy-1 signal) are reduced to 12.1 ± 2.8% (p < 0.001) the levels observed in non-inoculated mice. C: Expression of neuronal markers in prion-infected and control mouse brains as assessed by Western blotting. No change in neuron-specific enolase (NSE) or calbindin levels, and only a moderate decrease in synaptophysin levels are observed in prion-infected brains. D: No change in Sho levels are observed in brain homogenates prepared from clinically ill (8 month old) transgenic mice (TgCRND8) expressing a familial Alzheimer’s disease-associated variant of the amyloid precursor protein and control non-transgenic littermates.

100

neuron loss. A moderate decrease in synaptophysin levels was observed in prion-infected

mice (Figure 3.13C), likely indicative of a small amount of neuronal loss and loss of

synaptic density, but did not parallel the large decrease in Shadoo levels. These results

suggest that Shadoo levels are specifically diminished during clinical prion disease in

mice. In order to test whether Shadoo down-regulation is specific to prion disease, or is

instead merely reflective of a neurodegenerative disease, Shadoo levels were investigated

in a mouse model of familial Alzheimer’s disease (TgCRND8 mice over-expressing a

mutant allele of the β-amyloid precursor protein transgene [382]). Using mice in the

clinical phase of the disease (marked by overt amyloid deposits, normalized Aβ levels

equaling those of sporadic Alzheimer’s disease, and profound cognitive impairment), no

difference was apparent between TgCRND8 and age-matched non-Tg control mice with

respect to Shadoo levels (Figure 3.13D). Thus, depletion of Sho is absent in a non-

infectious CNS amyloidosis and appears to be specific for prion disease.

Transfected Shadoo levels in uninfected and prion-infected tissue culture cells

In an attempt to characterize the mechanism by which Sho levels are diminished

in prion infections, it was investigated whether or not the same phenomenon also occurs

in prion-infected tissue culture cells. For this purpose, levels of Shadoo protein 24 hours

post-transfection were compared between prion-infected cells and their uninfected

counterparts. Transient transfection assays were used since 1) no endogenous Shadoo can

be found in N2a cells using the existing anti-Sho antibodies (Figure 3.3C) and 2) there

are large differences between individual clones of ScN2a cells with respect to PrPres

levels [165] which could lead to misleading results following selection of stably-

transfected clones. Three different pairs of infected/uninfected cells were utilized: N2a

neuroblastoma cells and prion-infected ScN2a cells [392], GT1 hypothalamic neuronal

cells and prion-infected ScGT1 cells [393], and prion-infected SMB cells (mesodermal

cells derived from a prion-infected mouse brain) and cured SMB-PS cells [394].

Following transfection of N2a and ScN2a cells with Shadoo, a marked difference in

Shadoo protein levels was apparent between the two cell lines (Figure 3.14A). Sho

accumulated to considerably lower levels in ScN2a cells than in uninfected N2a cells.

This was not caused by a large difference in transfection efficiency—the transfection

101

α-Sho

α-NPTII

- + + +- -- + + +- -Infected:

Emptyvector

EmptyvectorShadoo ShadooTransfect:

N2a/ScN2a Cells GT1/ScGT1 CellsA B

α-Sho

α-NPTII

Infected:

Transfect:

DC***

- + + +- -

Emptyvector Shadoo

SMB-PS/SMB Cells

Infected:

Transfect:

α-Sho

α-NPTII

Figure 3.14. Shadoo levels are decreased in transiently transfected prion-infected cells compared to uninfected cells. A: Western blot of lysates from N2a (non-infected) or ScN2a (prion-infected) cells transiently transfected with empty vector or Shadoo plasmids and probed with an anti-Sho (06rSH-1) antibody. An antibody to neomycin phosphotransferase II (α-NPTII) was used as a control for transfection efficiency. B: Western blot of lysates from GT1 (non-infected) or ScGT1 (prion-infected) cells transiently transfected with empty vector or Shadoo plasmids and probed with an anti-Sho (06rSH-1) antibody. C: Western blot of lysates from SMB-PS (non-infected) or SMB (prion-infected) cells transiently transfected with empty vector or Shadoo plasmids and probed with an anti-Sho (06rSH-1) antibody. D: Quantitation of Shadoo levels in transiently transfected N2a cells and ScN2a cells (n = 6) by densitometry and normalized to NPTII levels. Levels of transfected Shadoo in ScN2a are approximately 50% lower than levels in N2a cells. ***P < 0.001.

102

efficiencies of N2a and ScN2a cells were similar as judged by immunoblotting for the

neomycin phosphotransferase II protein also encoded by the pcDNA3 vector. When

quantified and normalized for transfection efficiency, ScN2a cells accumulated

approximately 50% lower levels of transfected Shadoo (Figure 3.14D) and this difference

was highly statistically significant (p < 0.001). Similar results were obtained in paired

GT1 and ScGT1 transfections as levels of Shadoo obtained in ScGT1 cells were much

lower than those obtained in uninfected GT1 cells (Figure 3.14B). Although there was a

slight difference in transfection efficiency between the two cell lines, the difference was

not sufficiently large to account for the difference in Shadoo levels. Lower Shadoo levels

were also obtained in prion-infected SMB cells compared to cured SMB-PS cells (Figure

3.14C). However, the difference in transfection efficiency between uninfected and

infected cells was greater in this case and may partially explain the reduced Sho levels for

this pair of cells. In summary, it appears as though the in vivo phenomenon of Shadoo

down-regulation in prion-infected CNS tissue is also manifest in prion-infected tissue

culture cells.

Biochemical characterization of Sho in prion-infected and uninfected cells and tissue

There are several biochemical differences between PrPC in uninfected tissue and

PrPSc in prion-infected tissue. For instance, PrPSc is more detergent-insoluble and

displays a greater resistance to proteinase K digestion. Thus, it was considered whether or

not the biochemical properties of Shadoo are altered between uninfected and prion-

infected tissue. Following detergent extraction with Triton X-100/deoxycholate, the

solubility of Shadoo in brain homogenates from either non-inoculated or RML prion-

inoculated mice was assessed by ultracentrifugation. In non-inoculated brains, Shadoo

was essentially absent from the pellet fraction indicating that the protein is largely soluble

in non-ionic detergents (Figure 3.15A). No obvious change in Shadoo solubility was

observed in prion-infected brains although the significantly lower levels of Shadoo

protein may obscure any subtle change. A similar scenario was observed in N2a and

ScN2a cells. The vast majority of Shadoo was in the soluble fraction in both N2a and

ScN2a cells (Figure 3.15B). Collectively, these results suggest that Shadoo is a soluble

protein and that its solubility is not altered in prion-infected tissue. The sensitivity of

103

S2 S2 S2 S2P2 P2 P2 P2Non-infected brains RML-infected brains

Sho

A B

Sho

S2 S2P2 P2

N2aCells

ScN2aCells

C

20

15

20

15

N2a

ScN2a

Sho

Sho

*

*

0 0.1 0.5 1 2 5 10 20 μg/mLPK:

20

20

30

30

N2a

ScN2a

PrP

PrP

D

Sho

*20

15

0 0.1 0.5 1 2 5 10 20 μg/mLPK:Uninfected Mouse Brain

Figure 3.15. Biochemical properties of Shadoo in infected and uninfected tissues and cells. A: Analysis of Shadoo detergent solubility in uninfected and RML prion-infected brains. Brain homogenates were extracted with detergent and the supernatant (S2) and pellet (P2) fractions following ultracentrifugation analyzed by Western blotting with the 06rSH-1 anti-Sho antibody. Shadoo is largely soluble in both uninfected and infected brains. B: Analysis of Shadoo detergent solubility in transfected uninfected N2a cells and prion-infected ScN2a cells. No difference in Shadoo solubility is observed between infected and uninfected cells. C: Analysis of the sensitivity of Shadoo and PrP to proteinase K digestion in transfected uninfected N2a cells and prion-infected ScN2a cells by Western blotting. Shadoo is readily digested by low concentrations of PK whereas higher concentrations of PK are required to completely digest PrPC. No differences in Shadoo PK sensitivity are observed between N2a and ScN2a cells. D: Analysis of the sensitivity of Shadoo from mouse brain to PK digestion. Shadoo is completely digested by PK at a concentration of 0.5 µg/mL. A cross-reactive band in present in blots in panels C-D is denoted by an asterisk (*).

104

Shadoo to PK digestion was also investigated. In both N2a and ScN2a cells transfected

with Shadoo, Shadoo was extremely sensitive to PK digestion and was completely

digested by PK at a concentration of 0.5 µg/mL (Figure 3.15C). In contrast, higher

concentrations of PK were required to completely digest PrPC in N2a cells

(approximately 5 µg/mL) and, as expected, a fraction of PrP was PK-resistant up to a

concentration of 20 µg/mL in prion-infected ScN2a cells. Shadoo was also found to be

highly sensitive to PK digestion in wild-type mouse brain homogenates (Figure 3.15D).

These results suggest that Shadoo is highly sensitive to PK digestion, which likely

reflects an intrinsic flexible structure without many defined elements of secondary

structure. No evidence was obtained for increased protease resistance of Shadoo in prion-

infected tissue (data not shown).

The stability of Shadoo in N2a and ScN2a cells was investigated by treating

transfected cells with cycloheximide (CHX), an inhibitor of protein translation, and

measuring residual Shadoo levels at defined time points. Following CHX treatment,

Shadoo levels rapidly decreased in both N2a and ScN2a cells (Figure 3.16A) and the

majority of Shadoo was gone following six hours of CHX treatment. In contrast, PrP

levels persisted much longer following treatment—most of the protein was still present

following 12 hours of CHX treatment in N2a cells. The half-life of Shadoo in N2a cells

was estimated to be 2.2 hours by performing densitometry on blots of CHX-treated

lysates (Figure 3.16B). Clearly, the half-life of Shadoo in N2a cells is much shorter than

that of PrPC. There was no large difference in the stability of Shadoo in N2a and ScN2a

cells as judged by the rate of signal decay (Figure 3.16A). However, there appears to be

increased amounts of a residual Shadoo band in ScN2a cells. This result requires

substantiation and repetition in a distinct cell line. These experiments define Shadoo as a

short-lived flexible protein and these characteristics may be important to tailoring its

biological activities in vivo.

Effects of over- and under-expression of Shadoo on PrPres levels in ScN2a cells

In order to assess the effects of Shadoo on prion replication, Shadoo over-

expression and knockdown experiments were performed in ScN2a cells. For knockdown

105

N2a

N2a

ScN2a

ScN2a

Sho*

Sho*

PrP

PrP

0 2 4 6 8 10 12 hrCHX:

A

B

Figure 3.16. Analysis of Shadoo stability in transfected N2a and ScN2a cells. A: Sho-transfected N2a or ScN2a cells were treated with cycloheximide (CHX) for the indicated amounts of time and residual Sho and PrP levels were analyzed by Western blotting using the anti-Sho and anti-PrP antibodies 06rSH-1 and 7A12, respectively. Compared to PrP, Shadoo has a much shorter half-life. B: Measurement of Shadoo half-life in transfected N2a cells. Following treatment with CHX for the indicated amount of time, residual Sho levels were quantified by densitometry and Shadoo half-life calculated to be 2.2 hr.

106

experiments, pooled siRNA consisting of a pool of four independent siRNAs directed

against various parts of the Sprn gene was used. Endogenous levels of Shadoo in

N2a/ScN2a are too low to be detected with the current Shadoo antibodies. Thus, the

efficacy of the Sprn siRNA was tested by performing a co-transfection with a Shadoo-

expressing plasmid in N2a cells. The siRNA was highly effective at reducing levels of

transfected Shadoo (Figure 3.17A). When ScN2a cells were transfected with the Sprn

siRNA, no changes in PrPres levels were observed whereas siRNA directed at Prnp was

effective at reducing PrPres levels (Figure 3.17B). Similarly, over-expression of Shadoo in

ScN2a cells failed to modulate levels of PrPres whereas over-expression of PrP induced an

increase in PrPres (Figure 3.17C). Cumulatively, these results suggest that Shadoo levels

have no overt effect on prion replication as monitored by PrPres levels in ScN2a cells.

However, the transfection efficiency of ScN2a cells is not ideal implying that sufficient

over-expression or knockdown of Shadoo to modulate endogenous PrPres levels may not

be achievable. Unfortunately, the use of stably-transfected ScN2a cells was not a viable

approach to this problem since clonal differences in subcloned cell lines with respect to

PrPres levels and susceptibility to prions [171] preclude comparisons between PrPres levels

in stable cell lines expressing or not expressing Shadoo. Therefore, further in vivo

experiments using Shadoo over-expressing transgenic mice and Shadoo knockout mice

are necessary to either confirm or refute the above results.

3.5 Discussion The work presented in this chapter represents the first biochemical and functional

characterization of the putative prion-like protein Shadoo. Although Sprn genes from a

variety of mammalian species have previously been identified and cloned, progress thus

far has been restricted to RT-PCR experiments to identify the tissue specificity of Sprn

expression as well as phylogenetic analysis [197, 198, 379, 380]. Crucially, no evidence

supporting the existence of authentic Shadoo protein in vivo has been previously

published. Numerous lines of experimental evidence contained within this body of

research support the notion that Shadoo is indeed a real protein and is present in the

brains of wild-type mice. Firstly, two distinct antibodies raised against the predicted Sho

amino acid sequence and which recognize separate epitopes in the protein both recognize

107

20

15

10

A B

40

3020

40

3020

PrP

PrPres

C

40

3020

40

3020

PrP

PrPres

Sho302015

Figure 3.17. Effects of over-expression and knockdown of Shadoo in ScN2a cells. A: Western blot of lysates from N2a cells transfected with Shadoo plasmid or Shadoo plasmid plus Sprn siRNA. The Sprn siRNA is effective at knocking down Shadoo expression. B: Knockdown of Shadoo has no effect on PrPres levels in ScN2a cells. ScN2a cells were transfected with the indicated siRNAs, lysates collected 72 hours post-transfection, and PrPres levels analyzed by Western blotting. Although Prnp siRNA was effective at decreasing PrPres levels, no effects of Sprn siRNA on PrPres levels were observed. C: Over-expression of Shadoo in ScN2a cells does not alter PrPres levels. ScN2a cells were transfected with the indicated plasmids, lysates collected 72 hours post-transfection, and PrPres levels analyzed by Western blotting. Expression of Shadoo had no effect on PrPres levels whereas over-expression of PrP led to increased amounts of PrPres.

108

PNGaseF-sensitive bands of identical molecular weight in brain homogenates prepared

from wild-type mice (Figure 3.4). Because the identified bands migrate at the predicted

molecular weight of the Shadoo protein, this result highly suggests that authentic Shadoo

protein exists in the brain. Secondly, anti-Sho antibodies and Sprn anti-sense riboprobes

both generated signals in tissue sections from mouse brains. Importantly, both reagents

defined an identical restricted pattern of expression in the brain arguing that 1) actual

Shadoo protein or Sprn mRNA was identified and 2) there was not a general problem of

widespread cross-reactivity. Thirdly, transfection of an expression plasmid containing the

Sho open-reading frame into N2a cells generated a protein recognized by various anti-

Sho antibodies which possessed the predicted biochemical properties of the Shadoo

protein (Figure 3.3). These experiments argue persuasively that the Sprn gene codes for a

protein which is produced in the brains of wild-type mice and is a normal component of

the adult CNS.

Originally, Shadoo was identified on the basis of similarity to the hydrophobic

domain of PrP [197]. Consequently, Shadoo has been referred to as the third member of

the prion protein family [374], joining the well-characterized members PrP and Doppel.

Is the designation of Shadoo as a prion protein justified? Besides the sequence similarity

to PrP’s hydrophobic tract and the presence of N-terminal positively-charged repeats,

Shadoo possesses other biochemical signatures of the prion protein family. For instance,

Shadoo is post-translationally modified by the addition of N-glycans (although only at a

single site compared to the two N-glycosylation sites present in PrP and Doppel), is

processed by the removal of N- and C-terminal signal sequences, and undergoes the

addition of a GPI anchor which tethers the protein to the cell surface. However, although

these biochemical characteristics are consistent with the other two prion proteins, many

other GPI-anchored proteins (such as Thy-1) will also possess these qualities. In support

of the hypothesis that Shadoo is a bona fide member of the prion protein family, Shadoo

was found to undergo endoproteolysis both in the brains of mice and in transfected tissue

culture cells (Figure 3.5). This cleavage event is reminiscent of the C1 cleavage of PrPC

[232, 383] in that it takes place just N-terminal to the commencement of the hydrophobic

tract. Endoproteolytic cleavage of Doppel has also been observed (Janaky

Coomaraswamy, unpublished observations) suggesting that this processing event may be

109

a common feature of all three prion protein family members. The protease responsible for

Shadoo C1 cleavage remains to be identified. Treatment of N2a cells which stably

express FLAG-tagged Shadoo (Figure 3.5) with various protease inhibitors may prove

useful for this purpose. Thus, the biochemical properties of Shadoo and its sequence

similarity to PrP support its inclusion in the prion protein family (despite its lack of the

α-helical C-terminal domain present in PrPC and Doppel [204, 206, 207, 221, 228]).

Perhaps the strongest evidence for a similarity to PrPC is demonstrated by the ability of

Shadoo to abrogate the neuronal toxicity of both Doppel and ΔPrP (Figure 3.11).

Importantly, the neuroprotective activity of both Shadoo and PrPC is absolutely

dependent on the hydrophobic tract [381], the principle region of conservation between

the two proteins. Therefore, sequence similarity between the two proteins is also

reflective of functional similarities. In summary, both biochemical and functional

similarities to PrPC support the designation of Shadoo as a prion protein and defines

Shadoo as the third member of the prion protein family and the second CNS prion protein.

Unlike PrPC and Doppel, Shadoo is not predicted to contain well-defined elements

of secondary structure. Consistent with this notion, circular dichroism spectra of

recombinant Shadoo failed to reveal peaks characteristic of either α-helices or β-strands

[381]. This suggests that Shadoo is an intrinsically disordered protein that has a very

flexible structure. In support of this idea, Shadoo was found to be highly sensitive to

proteinase K digestion (Figure 3.15) implying a flexible structure that is easily accessible

to the protease. In contrast, PrPC is, relatively speaking, more resistant to proteinase K

digestion and a portion of the protein (the structured α-helical C-terminal domain)

persists at higher protease concentrations. Furthermore, treatment of N2a cells expressing

Shadoo with cycloheximide revealed that Shadoo has a much shorter half-life in tissue

culture cells than PrPC (i.e. Shadoo has a fast turnover rate). One possible explanation for

this is that Shadoo’s lack of structure allows it to be degraded by endogenous proteases

more easily than other proteins such as PrPC that have a more defined structure. The

instability and structural flexibility of Sho is perhaps not surprising given the similarity of

Shadoo to the N-terminus of PrP (Figure 3.1). This region of PrPC is known to be flexibly

disordered and therefore does not appear in any high resolution structure of PrP to date

[204, 206, 207, 395]. Although this region of PrP is not essential for prion replication and

110

pathogenesis [118, 119], it has been implicated in other physiological activities of PrPC

such as neuroprotection [271, 361, 390, 396], copper binding [209, 210], and PrP

trafficking [218]. Likewise, Shadoo has a positively-charged N-terminus which may

contribute to its trafficking and clearly possesses similar neuroprotective properties to

PrPC (Figure 3.11). However, it is unlikely that Shadoo is capable of binding copper since

it lacks any histidine residues in its primary structure. Instead, the tetrarepeat sequences

in the Shadoo N-terminus may serve as sites for protein-protein interactions or

interactions with extracellular metabolites.

A comprehensive analysis of the expression pattern of Shadoo within the brain

was performed using both immunohistochemistry and in situ hybridization. The strongest

Shadoo signals are observed in neurons throughout the brain with particularly strong

signals in the CA1 to CA3 neurons of the hippocampus and the Purkinje cells of the

cerebellum (Figures 3.6-3.9). No staining was observed in other CNS cells such as

astrocytes or oligodendrocytes. Thus, Shadoo may be useful as a specific marker of

neurons. In contrast to Shadoo, PrPC has a much more widespread pattern of expression

in the brain [397] and its expression is not limited to neurons [398]. Although Shadoo and

PrPC were co-expressed in neurons in certain regions of the brain such as the cerebral

cortex, a reciprocal expression pattern was found in the hippocampus and cerebellum. In

the hippocampus, Sho was found in the cell bodies of CA1 to CA3 neurons and in their

apical dendritic processes whereas these areas were largely devoid of PrPC expression.

Similarly, Sho was found in Purkinje cell bodies and dendritic arborizations whereas

PrPC was not found to substantially accumulate in these cells, a result that has been

repeatedly observed [279, 399]. This interlocking pattern of Sho/PrPC expression in the

hippocampus and cerebellum may imply that Shadoo has evolved to ‘fill-in’ a PrPC-like

function to the few regions of the brain that are naturally deficient in PrPC expression.

Consistent with the idea, PrPC and Sho display similar neuroprotective properties when

measured against Doppel or ΔPrP toxicity (Figure 3.11).

As mentioned previously, PrP knockout mice lack a pronounced phenotypic

defect when Prnp is ablated either pre- or post-natally [186, 248]. Possible explanations

for the absence of a pronounced phenotype include 1) PrPC plays a role in a process that

is not relevant when mice are observed under resting conditions (i.e. the mice may have

111

to be stressed in some specific way to elicit a phenotype) and 2) PrPC plays a role in an

essential physiological process but its absence is compensated for by the presence of a

functionally redundant protein. With regard to the latter possibility, certain investigators

have postulated that a hypothetical PrPC-like molecule termed π exists and compensates

for the lack of PrPC in Prnp0/0 mice [217, 372]. To this date, no candidate molecules or

proteins for π have been put forward. However, Shadoo seems a likely candidate for the

notional π protein for the following reasons. Firstly, Shadoo exhibits many biochemical

similarities to PrPC including N-glycosylation, addition of a GPI anchor, and C1

endoproteolysis and possesses homology to PrP’s hydrophobic tract, a strong candidate

for the active site of PrPC (Figures 3.1, 3.3-3.5). Secondly, Shadoo is expressed in

neurons and is present in many of the same regions of the brain as PrPC (Figures 3.6-3.9).

Thirdly, Shadoo is not found in cerebellar granular neurons (Figure 3.11) as an

endogenous π-like activity in these cells would prevent the CGN toxicity assay from

working. Fourthly, and perhaps most importantly, Shadoo exhibits functional redundancy

with PrPC in the CGN assay in that both proteins are capable of preventing the neuronal

toxicity elicited by expression of either Doppel or ΔPrP (Figure 3.11). Thus, Shadoo

possesses all of the characteristics which define the hypothetical π protein. Whether or

not Shadoo compensates for the lack of PrPC in Prnp0/0 mice remains to be determined

and future studies using Sprn0/0 and Prnp0/0/Sprn0/0 mice should be useful for this purpose.

It should be noted that Shadoo protein levels are not increased in Prnp0/0 mice (Figure

3.12) which argues that the protein products of the Prnp and Sprn genes are not cross-

regulated. Therefore, if Sho compensates for the lack of PrPC in Prnp0/0 mice, the

mechanism does not involve up-regulation of steady-state levels of Shadoo mRNA or

protein.

One possible complication in drawing parallels between Shadoo and π is that

Shadoo is clearly expressed in Purkinje cells (Figure 3.9). However, degeneration of

Purkinje cells occurs in mice with targeted expression of either Doppel or ΔPrP in

Purkinje cells [351-353], in mice transgenically expressing Doppel under the control of

the Prnp promoter [289], and in Prnp0/0 mice with ectopic expression of Doppel [188,

189]. If Shadoo is π and is present in Purkinje cells, then why do Purkinje cells

degenerate in these mouse lines? Firstly, the lack of Shadoo neuroprotection could be due

112

to a simple issue of stoichiometry. Over-expression of Doppel/ΔPrP in Purkinje cells may

overwhelm any endogenous π-like activity of Shadoo resulting in incomplete

neuroprotection. In support of this idea, incomplete rescue of Doppel-induced

neurodegeneration by PrPC is observed in lines of mice expressing high levels of Doppel

[289]. This idea can be tested by generating Sprn0/0 mice and crossing them with mice

expressing Doppel/ΔPrP in Purkinje cells to see if neurodegeneration is more pronounced

in mice lacking Shadoo expression. A second possibility is that Purkinje cell death

proceeds by a distinct mechanism from the death of cerebellar granular neurons. In this

scenario, Purkinje cell death occurs via a secondary mechanism (i.e. loss of essential

trophic signals from degenerating CGNs) following either the degeneration of CGNs or

the dysregulation of CGN homeostasis resulting from Doppel/ΔPrP expression. The

presence of Shadoo in Purkinje cells would have no effect on Purkinje cell survival in

this hypothetical mechanism. In support of this idea it has been recently shown that the

death of Purkinje cells in Doppel transgenic mice likely proceeds via a non-apoptotic

mechanism [400] whereas the death of CGNs clearly occurs via apoptosis. These two

mechanisms are not necessarily mutually exclusive and the actual mechanism may

include facets of both possibilities.

Although the exact mechanism of Doppel/ΔPrP toxicity and PrPC/Sho

neuroprotection remains enigmatic, a model has been proposed which explains these

phenomena in terms of physiological signaling events. In this model, PrPC normally binds

to a hypothetical cell-surface receptor (termed LPrP) and elicits an unknown ‘positive’

signaling event [372]. The exact nature of this signaling event is unknown, although this

concept is not without precedence as signaling through PrPC has previously been

observed [331]. Two binding sites are thought to exist between PrPC and LPrP, one in the

α-helical domain and one in the flexibly disordered N-terminus. Binding of PrPC to LPrP

promotes positive signaling whereas binding of a protein without the complete set of

interaction sites (such as Doppel or ΔPrP) causes transduction of a signal which leads to

neurodegeneration. When both PrPC and Doppel/ΔPrP are present, PrPC out-competes

Doppel/ΔPrP for binding and prevents the initiation of the neurodegeneration signal. How

does Shadoo fit in with this model? Presumably, Shadoo also binds to LPrP and thus

prevents Doppel/ΔPrP binding by either direct competition or by steric hindrance.

113

However, Shadoo does not possess the C-terminal α-helical domain of PrPC/Doppel and

therefore must bind to LPrP by virtue of its flexible N-terminal domain which bears

resemblance to the N-terminal domain of PrPC (Figure 3.1). Importantly, this is the

domain of PrPC that is associated with a protective signal in this model and is consistent

with the observed neuroprotective activity of Shadoo. The hydrophobic domains of both

Shadoo and PrPC are essential for neuroprotection and are therefore likely to figure in this

model [381]. Whether the hydrophobic tract serves directly as the second binding site to

LPrP or functions as an effector or modulator domain remains to be established. Clearly,

the identification of LPrP is required in order to draw further inferences from this model.

As is the case for PrPC, the exact role of Shadoo in the brain is unknown and will

be the subject of extensive future investigations. Based on the functional redundancy

between PrPC and Shadoo as measured in the CGN assay, it seems reasonable to

speculate that Shadoo plays a neuroprotective role in the brain. It will be worthwhile to

investigate whether or not Shadoo, like PrPC, is also protective against toxic stimuli such

as Bax expression [269, 270] and stroke-induced ischemia [273-275]. Presumably, the

neuroprotective role of Shadoo is tailored to neuron-specific requirements, with the

possible exception of the photoreceptor cell layer in the retina, as Shadoo expression is

primarily observed in neurons in the brain and spinal cord (Figures 3.8, 3.10). An exact

description of Shadoo’s in vivo function will require the creation of Sprn0/0 mice. These

mice are currently being generated (David Westaway, personal communication).

An unexpected result that has emerged from studies on Shadoo is that Shadoo

protein is notably down-regulated in both prion-infected brains and prion-infected tissue

culture cells (Figures 3.13-14). This suggests that Shadoo is either directly or indirectly

involved in prion disease. The mechanism of Shadoo down-regulation remains to be

determined but it is unlikely to result from a decrease in transcription as Sprn mRNA

levels as assessed by quantitative RT-PCR are unaltered throughout the course of prion

disease in mice (Inyoul Lee, personal communication). Thus, it seems likely that the

down-regulation occurs via a protein-based mechanism. Biochemical properties of

Shadoo such as solubility, PK resistance, and half-life are not notably altered in either

mice with clinical prion disease or in prion-infected tissue culture cells (Figures 3.15-16)

arguing against changes in Shadoo stability causing an increased rate of turnover in

114

diseased tissue. Therefore, it is possible that Shadoo is more closely-linked to prion

replication and that its down-regulation results from physiological interplay with PrPSc or

other conformationally-altered forms of PrP. In support of this idea, the decrease in

Shadoo protein levels tracks very well chronologically with the appearance of PrPSc in a

time course of prion disease, being an early event apparent months before the onset of

neurological symptoms in the infected mice (George Carlson, personal communication).

As mentioned previously, the well-conserved hydrophobic tract comprises the principle

region of similarity between PrPC and Shadoo. Notably, this region corresponds to the

epicenter of protein misfolding in PrP to disease-associated isoforms such as PrPSc [101,

223, 401]. It is possible that Shadoo and PrPSc may interact by virtue of the similarity of

their hydrophobic tracts causing a conformational alteration in Shadoo and its

degradation. Additionally, the domain structure of Shadoo resembles the domain

architecture of the mutant PrP allele Y145Stop which causes genetic prion disease in the

form of GSS and is known to be amyloidogenic [75, 76]. Recombinant wild-type Shadoo

can also adopt amyloid structures, as assessed by thioflavin binding and electron

microscopy, upon incubation at neutral pH (Vivian Ng, unpublished observations)

suggesting that it is thermodynamically possible for Shadoo to adopt β-rich structures.

Because PrPC and Sho have only a partial overlapping expression pattern in the brain,

contact between Sho and PrPSc may be increased by virtue of Shadoo shedding (Figure

3.5) into the parenchyma. Further experiments are required to define the relationship

between Shadoo and PrPSc, to clarify the mechanism of Shadoo down-regulation in prion

disease, and to assess any direct effect of Shadoo on prion replication.

Over and above enquiries into the mechanism of prion replication, the marked

down-regulation of Shadoo in mouse brains with clinical prion disease suggests that

Shadoo may have practical use as a biomarker of prion infections. Currently, the only

definitive biomarker of prion disease is PrPres and unambiguous confirmation of disease

can only be accomplished post-mortem. Thus, it will be worthwhile to assess whether

Shadoo is also down-regulated in other naturally-occurring prion diseases such as

sporadic CJD, BSE, scrapie, and CWD. Down-regulation of Shadoo appears to be

specific to prion disease as no decrease in Sho protein level was observed in a mouse

model of Alzheimer’s disease (Figure 3.13). It will be important to test the levels of

115

Shadoo in human cases of Alzheimer’s disease as well as in other neurodegenerative

conditions such as Parkinson’s disease and ALS, both in human cases and in mouse

models, in order to further assess the specificity of this phenomenon. If Shadoo is found

to be present in the blood, urine, or CSF, quantitation of Shadoo levels by ELISA may

serve as a useful diagnostic marker of prion disease.

Finally, it should be considered whether or not Shadoo plays a role in prion

disease pathogenesis. Studies of prion-infected rodents suggest that target areas exist in

the brain which give rise to the clinical phenotypes of prion disease [402]. However, the

exact mechanisms by which disease-associated isoforms of PrP cause dysfunction in the

brain are unknown. It has been suggested that the presence of PrPC is essential for PrPSc

toxicity as grafting PrPC-expressing tissue into Prnp0/0 mice and performing prion

inoculations leads to pathology only in the PrPC-expressing tissue despite the large

quantities of PrPSc in the surrounding areas [84]. Paradoxically, the opposite result,

namely pathology in PrP null tissue, has also been observed in prion-inoculated mice

expressing PrP specifically in astrocytes [85]. Perhaps some aspects of prion disease

pathology arise due to the interference of a Shadoo-associated pathway in trans by

conformationally-altered forms of PrP. For instance, the dendritic spines of CA1

hippocampal neurons constitute an early clinical target in mouse prion disease [403-405].

However, these structures do not express PrPC but do express Shadoo (Figure 3.9)

suggesting that morphological alterations in these structures are proceeding by a PrPC-

independent mechanism. It is also worthwhile to consider whether Shadoo down-

regulation in prion disease may have implications for clinical phenotypes. Because

Shadoo possesses a neuroprotective activity (Figure 3.11), the dramatic reduction in the

levels of this protein may be responsible for some aspects of the clinical presentation of

prion disease. Further investigations on Shadoo within the context of prion disease

biology are likely to provide insights into the many unresolved questions surrounding

prion replication and pathology.

116

Chapter 4

Characterization of Interactions Between Members of

the Prion Protein Family and the Type II

Transmembrane Protein DPPX

117

4.1 Abstract The function of the cellular isoform of the prion protein, PrPC, remains enigmatic

and a molecular description of its involvement in cellular processes has been hindered by

the inability to uncover interacting proteins authenticated in vivo. PrPC is known to

possess a neuroprotective activity in a number of experimental paradigms. In particular,

PrPC is capable of abrogating the cerebellar toxicity elicited by the expression of either

Doppel or N-terminally truncated PrP in the CNS. Although the mechanism of this

phenomenon is unclear, a model has been created which postulates the existence of a

prion ligand termed LPrP which is involved in transducing neurotoxic and neuroprotective

signals from prion proteins. In this study, the type II transmembrane protein DPPX has

been identified as the first common interacting protein for all three members of the prion

protein family. DPPX forms high molecular weight cell surface complexes with PrPC,

Doppel, and Shadoo in tissue culture cells and in cultured primary neurons. At least two

distinct sites in prion proteins are required for complex formation with DPPX: an N-

terminal site present in Sho and in PrPC and a C-terminal site in the α-helical domain

represented in PrPC and Doppel. Consistent with a two-site hypothesis, an intact β-

propeller ectodomain in DPPX is required for complex formation with PrPC or Doppel,

whereas the DPPX juxtamembrane region is necessary for binding to Sho. Remarkably,

the determinants for DPPX complex formation are in close agreement with determinants

for neurotoxicity and neuroprotection positioned by deletion mapping of Doppel and

PrPC. Furthermore, DPPX is expressed at high levels in the cerebellar granule neurons,

cells which are susceptible to Doppel toxicity in vivo and in vitro. Thus, DPPX is a

plausible candidate for the hypothetical LPrP protein which controls Doppel neurotoxicity

and PrPC-mediated neuroprotection in cerebellar cells. DPPX may provide insight into

the physiological functions of PrPC and Shadoo, and, since intronic variations in the

DPP6 gene increase the risk for sporadic ALS, it may be prudent to evaluate DPPX in

non-transmissible neurodegenerative disorders as well as in prion disease.

118

4.2 Introduction Prions are infectious proteins which cause neurodegenerative disorders of both

humans and animals. The prion hypothesis states that the conformational remodeling of a

normal host protein, PrPC, into a β-sheet-enriched isoform termed PrPSc is the central

event in prion diseases such as bovine spongiform encephalopathy (mad cow disease),

scrapie, and Creutzfeldt-Jakob disease [6, 406]. Unlike other neurodegenerative diseases

in which amyloidogenic proteins play a role such as Alzheimer’s disease and Parkinson’s

disease, misfolded prion protein is infectious and is thought to replicate by templating the

conversion of PrPC into additional copies of the disease-associated PrPSc isoform.

Although the requirements for in vivo prion replication are unclear, it is known that

seeded prion replication in vitro requires PrPC, PrPSc, lipids, and a polyanionic molecule

such as RNA [117]. It has been hypothesized that an additional factor, termed ‘Protein X’

is necessary for in vivo prion replication [110, 111]. However, despite considerable effort,

no candidates for Protein X have surfaced to date. Clearly, PrPC is required for both prion

replication and propagation since mice lacking the gene encoding the prion protein are

resistant to prion disease [107, 108].

There are many unanswered questions in prion disease biology. For instance, the

function of the cellular version of the prion protein, PrPC, has remained enigmatic despite

the existence of numerous strains of Prnp0/0 knockout mice [186, 187]. These mice have

only subtle or disputed phenotypes which has hindered the assignment of a cellular role

to PrPC [242, 244, 246, 252, 256, 262, 263]. PrPC is a GPI-anchored neuronal

glycoprotein which is expressed in most areas of the brain. It is possible that the absence

of an obvious phenotype in these mice is due to the existence of a PrPC-like molecule in

the brain which functionally compensates for the lack of PrPC [217, 372]. Alternatively,

unmasking of a cryptic phenotype in Prnp0/0 mice may require subjecting the mice to a

specific stress. Secondly, very little is known about how PrPSc causes degeneration of

neuronal tissue. Prion diseases are characterized by profound changes in the brain such as

the death of neurons, spongiform change (partially caused by neuronal vacuolation),

infiltration of astrocytes, and in some instances, the presence of amyloid plaques

containing aggregated PrPSc. Although one elegant tissue grafting experiment

demonstrated the requirement of PrPC-expressing tissue for prion-induced

119

neuropathology [84], a second experiment found that pathology can occur in tissue which

lacks PrPC [85]. Studies using transgenic mice which express PrPC lacking its GPI anchor

(and thus is secreted from the cell) have shown that although prion replication is

unimpeded in these mice, no clinical symptoms arise following challenge with infectious

prions [86]. This has led to the suggestion that cell surface-anchored PrPSc is required to

induce prion neurotoxicity, possibly by virtue of interacting with a hypothetical prion

receptor [87]. Thirdly, the requirement of additional factors for prion replication in vivo

remains unclear. Finally, studies using transgenic mice have shown that expression of

either Doppel, a PrPC paralog normally present in the testis, or N-terminally truncated

PrP (ΔPrP) in the CNS leads to a neurodegenerative phenotype characterized by the loss

of Purkinje and/or cerebellar granular neurons [188, 190, 217, 279, 280, 289].

Remarkably, this phenomenon only occurs in Prnp0/0 mice—co-expression of PrPC either

completely or partially abrogates the Doppel/ΔPrP-induced neurotoxicity. This has led to

a model in which neurotoxic/neuroprotective signals emanating from prion proteins are

controlled by a hypothetical prion ligand termed LPrP [217, 279, 372] or Tr [280]. No

candidates for LPrP have yet been identified.

The numerous unsolved mysteries in prion disease biology would certainly

benefit from the identification and authentication of in vivo PrPC- and PrPSc-interacting

proteins. The identification of PrP-interacting proteins has been a contentious issue in

prion research. Although there is no shortage of identified candidate PrP-binding proteins

(see Table 1.2), problems have arisen with respect to the lack of repetition of identified

binding partners amongst various labs, the paucity of identified proteins which interact

with PrPC in vivo in a biologically relevant manner, and the inability to find PrP-

interacting proteins which have an effect on prion replication. For instance, two of the

more characterized PrP-interacting proteins, the 37 kDa/67 kDa laminin receptor and

stress inducible protein 1 (STI1) are unlikely to have a large effect on prion replication

and may only offer small contributions to understanding the in vivo neurobiology of PrPC

[112, 319, 321, 325, 326, 329, 333, 334]. In contrast, two other PrPC-interacting proteins,

the neural cell adhesion molecule (N-CAM) and the α2/β2-Na+/K+-ATPase co-operate

with PrPC in regulating well-defined biological processes. PrPC promotes the localization

of N-CAM to lipid rafts and neurite outgrowth [322, 330] whereas PrPC and the α2/β2-

120

Na+/K+-ATPase are involved in regulating glutamate-dependent lactate transport in

astrocytes [318]. Notably, these PrPC-interacting proteins were also identified in an in

vivo screen for candidate PrPC-interacting proteins using a novel protein-protein

interaction tool in mice [266]. Time-controlled transcardiac perfusion crosslinking

(tcTPC) utilizes dilute formaldehyde to form covalent crosslinks between individual

members of protein complexes within the brain or other organs. Following capture of

target protein complexes by immunoprecipitation, associated proteins are identified by

tandem mass spectrometry. Advantages of this approach over other protein-protein

interaction strategies include the ability to analyze protein complexes in their native

setting and at endogenous levels of expression, amenability to membrane protein

complexes, and the ability to perform stringent washes due to the covalent crosslinks.

Accordingly, this technique has been successfully used to uncover biologically relevant

binding partners for both the amyloid precursor protein and presenilin-1 [267, 342].

When PrPC-containing complexes were isolated following tcTPC, several proteins in

proximity to PrPC were identified with confidence [266]. One of these, DPPX, was

selected for further investigation.

DPPX (also known as DPP6 or dipeptidyl aminopeptidase-like protein 6) is a type

II transmembrane protein encoded by the DPP6 gene located on chromosome 7 in

humans and chromosome 5 in mice. DPPX is expressed as multiple splice variants with

the divergent sequences occurring in the N-terminal cytoplasmic domain [376]. The large

extracellular domain of DPPX classifies it as a member of the dipeptidyl aminopeptidase

family of membrane-anchored serine proteases which includes DPP4 (also known as

CD26), a protease which is responsible for cleaving N-terminal dipeptides of the form

Xaa-Pro/Ala (where Xaa represents any amino acid) from proteins [407]. DPP4 cleaves a

variety of proteins including chemokines, inflammatory modulators, and neuropeptides

which results in modulation of their biological properties. For instance, DPP4 cleavage of

GLP-1, GLP-2, or GIP-1, proteins involved in stimulating insulin secretion, results in

their inactivation by enhancing their subsequent degradation [408]. Thus, inhibitors of

DPP4 represent potential treatments for type II diabetes [409, 410]. However, DPPX has

an aspartate residue at the position normally occupied by the catalytic serine residue

within the consensus sequence for serine proteases [377] and does not exhibit any

121

proteolytic activity against canonical DPP4 substrates, even when the aspartate residue is

mutated to a serine [376, 411]. Thus, DPPX may have evolved to perform a function

distinct from proteolysis. Accordingly, DPPX is known to modulate the properties of

neuronal A-type potassium channels and is required for the successful in vitro

reconstitution of nascent channel properties [375]. In this system, DPPX likely functions

by promoting the trafficking of Kv4.2, the pore-forming channel subunit, to the cell

surface. DPPX expression appears to be primarily restricted to the CNS with a

widespread pattern of localization within individual regions of the brain [375, 376, 411,

412] and the various splice variants of DPPX are known to have unique patterns of

expression [413]. The structure of the DPPX ectodomain has been solved and consists of

two principal domains: an α/β-hydrolase domain and a β-propeller domain [414].

Recently, the DPP6 gene has been linked to increased susceptibility to amyotrophic

lateral sclerosis (ALS) suggesting that DPPX may play a role in neurodegeneration [415,

416]. DPPX is closely related to DPP10 (also known as DPPY), another member of the

dipeptidyl aminopeptidase family which is encoded by the DPP10 gene on chromosome

2 in humans and chromosome 1 in mice [417]. Like DPPX, DPP10 exists as multiple N-

terminal splicing isoforms, lacks the catalytic serine residue, and can modulate the

properties of A-type potassium channels [418-422]. Interestingly, the DPP10 gene has

been linked to asthma [423].

Analysis of DPPX as a putative PrPC-interacting protein demonstrates that DPPX,

but not DPP10, forms high molecular weight cell-surface complexes with PrPC, Doppel,

and Shadoo in N2a neuroblastoma cells and in cultured primary cerebellar granule

neurons. Thus, DPPX represents the first protein identified which interacts with all three

prion proteins. Mapping studies of binding determinants in both prion proteins and DPPX

argue for a two-site binding mechanism and the identified binding sites are in good

agreement with mapped toxicity and neuroprotection determinants in Doppel and PrPC,

respectively. In view of a high level of expression of DPPX in the granule cell layer of

the cerebellum, the data imply that DPPX is a plausible candidate for LPrP. Thus, DPPX

may be important to both the normal and disease-associated cell biology of prion proteins.

122

4.3 Materials and Methods

Bioinformatics and Statistics

Protein alignments were created using the T-COFFEE algorithm (http://tcoffee.vital-

it.ch/cgi-bin/Tcoffee/tcoffee_cgi/index.cgi). Transfected CGN datasets and MTT assays

were analyzed by one-way ANOVA and Tukey pair-wise comparisons with significance

set at p<0.05 using GraphPad Prism software (version 5, GraphPad Inc.). Three

dimensional protein structure pictures were generated using MBT PDB ProteinWorkshop

1.50 with structure files downloaded from the RCSB Protein Data Bank

(http://www.rcsb.org).

mRNA Purification, Cloning of DPPX Isoforms, and Plasmid Generation

Mouse brain total RNA was isolated from half brains using the acid guanidinium

thiocynatate-phenol-chloroform extraction protocol of Chomczynski and Sacchi [424]

and mRNA purified using the Oligotex mRNA Mini Kit (Qiagen). cDNA synthesis was

performed in the presence of SUPERase-In RNase inhibitor (Ambion) using 0.5 µg

mRNA, Omniscript reverse transcriptase (Qiagen), and an oligo-dT primer. PCR

amplification of DPPX and DPP10 isoforms was conducted using Pfu Turbo DNA

polymerase (Invitrogen) and a nested PCR strategy. DPPX-M was cloned by

amplification of an N-terminal portion of the coding sequence and then ligation into the

complete sequence. The coding sequences for all DPPX and DPP10 isoforms were

inserted between the BamHI and XbaI sites of the pcDNA3 mammalian expression vector

(Invitrogen).

FLAG-tagged Shadoo and Shadoo deletion mutants as well as 3F4-tagged PrP

and PrP deletion mutants were generated using pairs of mutagenic oligonucleotides and

the QuikChange (Stratagene) site-directed mutagenesis procedure with Pfu Turbo DNA

polymerase. Doppel deletion mutants in pBUD.CE4.GFP have been previously described

[271]. HA-tagged DPPX-S, DPPXΔCyto, and DPPX deletion mutants were generated by

standard PCR-based techniques. The secreted DPPX ectodomain construct was generated

by fusing the DPPX ectodomain to the PrP N-terminal signal sequence using an

introduced BsrGI site.

123

Vector-based shRNA directed against the DPP6 gene was prepared by annealing

complimentary oligonucleotides using a step cooling protocol and then insertion into the

pSUPER vector (Oligoengine) as per the manufacturer’s instructions.

The Thy-1 plasmid (Thy-1.2 isoform) was generated by amplification of the Thy-

1 open reading frame from the MGC:62652 cDNA clone by PCR and then insertion into

pcDNA3. The identity of all constructs was verified by DNA sequencing and all plasmids

were prepared using endotoxin-free plasmid maxi-prep kits (Qiagen).

RT-PCR

First strand cDNA was generated from mouse brain mRNA as above and then

pre-amplified with primers flanking the entire DPPX-E open reading frame. Internal

primers on either side of the SV1 junction were then used for a second round of PCR

amplification using Platinum Taq polymerase (Invitrogen) and the following protocol: 1

cycle of 95ºC for 5 min, 40 cycles of 95ºC for 30 s, 58ºC for 30 s, and 72ºC for 1 min,

and finally, 1 cycle of 72ºC for 7 min. PCR products were then analyzed by

electrophoresis using an 8% polyacrylamide gel.

DPPX and DPP10 Polyclonal Antibody Production

To generate peptide antisera directed against murine DPPX, peptides were

synthesized (each containing an additional N-terminal cysteine residue for KLH

conjugation), conjugated to maleimide-activated KLH (Pierce) and then injected into

New Zealand White rabbits. 04DX-2 was raised against DPPX-S residues 59-78

(EDTSLSQKKKVTVEDLFSED), 03J2 was raised against DPPX-S residues 507-522

(DKRRMFDLEANEEVQK), and 03K1 was raised against DPPX-S residues 788-803

(QDKLPTATAKEEEEED). Similarly, the 06D10-2 polyclonal antibody recognizing

murine DPP10 was generated using a peptide comprising DPP10-1 residues 784-800

(CLKEEVSVLPQEPEEDE) where the N-terminal cysteine residue is part of the natural

DPP10 sequence. Polyclonal antibodies were precipitated from serum using ammonium

sulfate and then affinity purified using the immunogenic DPPX or DPP10 peptide

conjugated to a SulfoLink column (Pierce).

124

Cell Culture, Transfection, and Lysis

N2a and HEK293 cells were cultured in DMEM medium containing 10% FBS

and 0.2× penicillin/streptomycin (Gibco) and maintained in a humidified incubator with

5% CO2. ScN2a cells were cultured in OptiMEM medium (Gibco) containing 10% FBS,

1× GlutaMax (Gibco), and 0.2× Penicillin/Streptomycin and maintained as above. Cells

were transfected with Lipofectamine-2000 (Invitrogen) according to the manufacturer’s

instructions using a 1 µg DNA: 1 µL Lipofectamine-2000 ratio. For competition

experiments, N2a cells were transfected with 2 µg DPPX-S plasmid and increasing

amounts (0 to 2 µg) of either Doppel or PrPΔ32-121 plasmid. Plasmid amounts were

equalized by the addition of empty vector (pcDNA3). For siRNA treatment of ScN2a

cells, cells were transfected with ON-TARGETplus siRNA (Dharmacon) at a final

concentration of 100 nM in OptiMEM using Lipofectamine-2000 as per the

manufacturer’s instructions. Cells were incubated with the transfection mixture for 48hrs

and then in serum-containing medium for an additional 24 hrs. For the creation of stable

cell lines, cells were selected and maintained in 1 mg/mL and 0.2 mg/mL G418 (Gibco),

respectively. For cell lysis, cells were washed twice with PBS and then lysed with RIPA

lysis buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.5% sodium deoxycholate and 1%

NP-40) containing Complete Mini Protease Inhibitor Cocktail tablets (Roche). Lysates

were incubated on ice and then cleared by centrifugation at 20,000× g for 10 min at 4ºC.

Protein concentrations were determined using the BCA assay (Pierce).

Mild Formaldehyde Crosslinking of Cultured Cells

Cells (either N2a or HEK293 cells 24 hours post-transfection or CGN cultures 4

days post-plating) were washed twice with PBS and then incubated with 2%

formaldehyde (diluted into PBS from a 37% w/v stock solution) for 15 min at room

temperature. The crosslinking reaction was quenched by the addition of 125 mM glycine

in PBS for 10 min. Crosslinked cells were then washed twice with PBS and lysed with

ice cold RIPA lysis buffer containing protease inhibitors for 10 min at 4ºC. Lysates were

cleared by centrifugation at 20,000× g for 10 min at 4ºC. Prior to Western blotting,

protein in crosslinked lysates was precipitated with acetone (5 volumes with overnight

incubation at -20ºC) and the pellets resuspended directly in 1× loading buffer containing

125

β-mercaptoethanol. For crosslink reversal, samples were incubated for 30 min at 95ºC

using a PCR machine.

Biotinylation of Transfected N2a Cells

Transfected N2a cells were washed twice with PBS and then incubated with cell-

impermeant Sulfo-NHS-SS-Biotin (Pierce) at a concentration of 1 mg/mL in PBS for 1 hr

at 4ºC. Quenching solution (Pierce) was then added per the manufacturer’s instructions

and the cells incubated at 4ºC for an additional 20 min. Cells were then washed two times

with PBS and then crosslinked with 2% formaldehyde as described above. Following

lysis, lysates were incubated with neutravidin beads (Pierce) for 1 hr at room temperature

to capture biotinylated proteins. Proteins were eluted by incubation with loading buffer

containing 50 mM dithiothreitol for 1 hr at room temperature. Beads were then removed

by spin column filtration and crosslinked complexes analyzed by Western blotting.

PNGaseF Treatments

PNGaseF digestions (New England Biolabs) were performed by boiling protein in

1× denaturing buffer (0.5% SDS, 0.5% β-mercaptoethanol) for 10 min, followed by the

addition of NP-40 and G7 reaction buffers to 1× and 50 units of PNGaseF per 50 µg

protein. Digestions were allowed to proceed at 37ºC for 4 hours and then terminated by

the addition of SDS-PAGE loading buffer.

Immunoprecipitations

For immunoprecipitations from mouse brains, brain homogenates were first

solubilized on ice with PBS containing 1% DDM (n-dodecyl-β-D-maltoside) and then

cleared by centrifugation at 20,000× g for 10 min. Cell lysates or detergent-extracted

mouse brain homogenates (typically 0.5-1 mg) were incubated with antibody (5 µg) for 4

hours at 4ºC with end-over-end rotation. In some instances, lysates or homogenates were

pre-cleared with protein-G-Sepharose beads for 2 hours prior to the addition of antibody.

Antibody-target complexes were captured with 30 µL of pre-washed protein-G-

Sepharose beads (Sigma) by rotation overnight at 4ºC. Beads were washed 3 times with

RIPA lysis buffer or solubilization buffer and proteins eluted by boiling in 1× loading

126

buffer containing reducing agent. Beads were then removed by spin column filtration and

proteins analyzed by Western blotting.

Western Blotting

For Western blotting, 20 to 50 µg total protein was prepared in sample buffer

containing reducing agent, boiled, and then separated on either 4-12% NuPAGE gels with

the MES or MOPS buffer systems (Invitrogen) or by conventional SDS-PAGE using

freshly-poured polyacrylamide gels of various percentages. Crosslinked protein samples

were not boiled prior to gel loading. Proteins were transferred to either nitrocellulose or

PVDF membranes and then blocked with either 5% non-fat skim milk or 2% BSA in

TBS containing 0.05% Tween-20. Blots were incubated overnight with primary

antibodies at 4ºC or at room temperature in the presence of 0.05% sodium azide.

Following three washes with TBS containing 0.05% Tween-20 (TBST), blots were

incubated with HRP-conjugated secondary antibody (BioRad) and developed using

Western Lightning ECL (Perkin-Elmer). The following primary antibodies were used:

anti-DPPX 04DX-2, anti-DPPX 03J2, anti-DPPX 03K1, anti-DPP10 06D10-2, anti-Sho

06rSH-1, anti-Sho 06SH-3a, anti-FLAG M2 (Sigma), anti-HA HA-7 (Sigma), anti-Dpl

E6977 (a generous gift from Stanley Prusiner), anti-Dpl 03A2 [271], anti-PrP antibodies

D18, D13, and R1 (InPro Inc.), anti-PrP antibodies 8H4 and 7A12 (generous gifts from

Man-Sun Sy), anti-Thy-1 R194 (a generous gift from Roger Morris), and anti-actin 20-33

(Sigma).

Blue Native PAGE

Mouse brain homogenates were solubilized on ice with 1% digitonin and samples

prepared in 1× NativePAGE loading buffer. Protein complexes were separated on 4-16%

NativePAGE gels (Invitrogen) in the presence of Coomassie Brilliant Blue G-250 as per

the manufacturer’s instructions. Following transfer to PVDF membranes and rinsing with

methanol, proteins were fixed with 8% acetic acid. Membranes were then processed

using standard Western blotting procedures as described above.

127

Cerebellar Granular Neuron Cultures and Transfections

CGN cultures were obtained essentially as previously described [271]. Briefly,

cerebella from either wild-type or Prnp0/0 mice (ZrchI strain, both on a C57/B6

background) were dissected from 7 day old pups in HBSS. Cerebellar tissue was

disrupted by incubation with Trypsin-EDTA for 15 min at 37ºC and then the medium

replaced with MEM (Sigma #4655) containing 10% heat-inactivated FBS, 0.1×

penicillin-streptomycin, and 25mM KCl (K25+S medium) with the addition of soybean

trypsin inhibitor to a concentration of 0.25 mg/mL. Cerebella were triturated by pipetting

up and down with a fire-polished Pasteur pipet and following sedimentation of un-

digested material, cells were spun at 1300 rpm in an IEC Centra-EC4R centrifuge for 5

min at room temperature. Cells were resuspended in K25+S medium (without inhibitor),

filtered through a cell strainer, and then plated on 12-well tissue culture dishes (Costar)

that had been coated overnight with 0.1 mg/mL poly-L-lysine. Cells were incubated at

37ºC for 4 days prior to transfection. Individual wells of cells were transfected with 2 µg

DNA and 3 µL Lipofectamine-2000 in MEM medium without serum for 1 hr before

replacing with conditioned K25+S medium. For DPPX shRNA transfections, pSUPER

plasmids were co-transfected with the fluorescent pBUD.CE4.GFP plasmid at a 3:1 ratio.

The pBUD.GFP.Dpl plasmid has been previously described [271]. 24 hr post-transfection,

cells were fixed with 4% paraformaldehyde and nuclei stained with Hoechst 33342 (5

µg/mL in PBS). Individual apoptotic events were assessed by scoring nuclear

morphology in GFP-positive transfectants.

Antibody Treatment of CGNs and MTT Assays

Antibodies (diluted in PBS) were added directly to Prnp0/0 CGN cultures

following 4 days of culture to a final concentration of 10 µg/mL. Cells were incubated

with the antibodies (anti-DPPX 03K1, anti-DPP10 06D10-2, or vehicle) for 72 hrs.

Toxicity analysis was performed using the MTT assay. MTT (3-(4,5-Dimethylthiazol-2-

yl)-2,5-diphenyltetrazolium bromide) was added to the cells at a final concentration of

0.5 mg/mL and incubated at 37ºC for 4 hrs. Cells were then solubilized by the addition of

an equal volume of 10% SDS/0.01 M HCl and incubated overnight at 37ºC. Absorbances

were then read at a wavelength of 562 nm.

128

http://en.wikipedia.org/wiki/Di-

http://en.wikipedia.org/wiki/Methyl

http://en.wikipedia.org/wiki/Thiazole

http://en.wikipedia.org/wiki/Phenyl

Mouse Lines and Preparation of Mouse Brain Homogenates

Prnp0/0 mice (ZrchI strain) and Tg7 mice were maintained on a C57/B6

background. DPP6df5J/Rw mice [425] were a generous gift from John Schimenti and were

maintained on a C3H background. Rw+/- and df5J+/- mice were generated by crossing

DPP6df5J/Rw mice with wild-type C57/B6 mice. df5J+/- mice were identified by real time

PCR and Rw+/- were identified visually by the presence of a white posterior.

Mice were perfused with saline, half brains were extracted, and then brains were

either homogenized directly or snap frozen and stored at -80ºC for future use. Brains

were homogenized in nine volumes of 0.32M sucrose (10% homogenates) containing

Complete Mini Protease Inhibitor Cocktail tablets (Roche). For preparation of RML-

infected brain homogenates, brains were homogenized in PBS without protease inhibitors.

Time-Controlled Transcardiac Perfusion Crosslinking

Mice (either C57/B6 or FVB strains) were subjected to the tcTPC procedure as

described by Schmitt-Ulms and co-workers [266]. Following in vivo crosslinking, brains

were extracted and then homogenized in 0.32 M sucrose containing protease inhibitors.

Protein complexes were then analyzed by Western blotting.

Immunofluorescence

N2a cells 24 hours post-transfection were washed with PBS, fixed with ice cold

methanol (5 min), washed with PBS, blocked with 2% goat serum, and then incubated

with primary antibody overnight at 4ºC. Following PBS washes, cells were incubated

with Alexafluor488-conjugated secondary antibodies (Invitrogen) for two hours and then

washed three times with PBS. For immunocytochemistry on CGNs, cells were fixed after

5 days in culture with 4% paraformaldehyde, and then processed and stained. Fluorescent

images were obtained using either a Zeiss Axiovert microscope or a Leica DM6000 B

microscope in conjunction with OpenLab software (Improvision).

Immunohistochemistry

For tissue analysis, mice were perfused with saline, brains bisected in the mid-

sagittal plane, and fixed in methacarn fixative (60% methanol, 30% chloroform, and 10%

129

glacial acetic acid). Brains were fixed either at room temperature for 3-4 hours or

overnight at 4ºC prior to immersion in 70% ethanol. Brains were then processed to

paraffin wax. Sections (6 µm) were cut, dried (63ºC for 1 hour), de-paraffinized with

xylene, re-hydrated through a graded series of ethanol, and then rinsed in TBS pH 7.2.

Primary antibodies were diluted in Antibody Dilution Buffer (DAKO Cytomation) and

applied to the tissue sections for an overnight incubation at 4ºC in a humidified chamber.

For peptide blocking of antibody reactivity, peptide was added in 4-fold mass excess to

the diluted antibody and rotated overnight at 4°C prior to application to tissue sections.

The sections were then rinsed in TBS and processed in EnVision Labelled Polymer

(DAKO Cytomation), rinsed with water, visualized with DAB (3,3'-diaminobenzidine)

and counterstained with Harris' hematoxylin (Sigma). Images were captured on a Leica

DM6000 B microscope using a Micropublisher 3.3RTV camera (Q Imaging Inc.) in

conjunction with OpenLab software (Improvision). For fluorescent double labeling

experiments, the secondaries used were Alexafluor488- and Alexafluor594-conjugated

antibodies (Invitrogen). The following primary antibodies were used for

immunohistochemistry: anti-DPPX 04DX-2, anti-DPPX 03J2, anti-DPPX 03K1, and

anti-DPP10 06D10-2.

Detergent Insolubility Assays

One tenth volume of 10× detergent was added to brain homogenate for a final

concentration of 0.5% Triton X-100 and 0.5% sodium deoxycholate. Samples were

mixed, incubated briefly on ice, and cell debris removed by spinning at 1000× g for 5 min

at 4ºC. Samples were then spun at 120,000× g for 40 min at 4ºC using a TLA-55 rotor

and a Beckman TL-100 ultracentrifuge. Supernatant (S2) and pellet (P2) samples were

prepared in 1× loading buffer, boiled, and then analyzed by Western blotting.

Detection of PrPres and Proteinase K Digestions

For detection of PrPres in ScN2a cells, cells were lysed with RIPA buffer in the

absence of protease inhibitors. Lysates were adjusted to 1 mg/mL and 200 µg total

protein was digested with 20 µg/mL proteinase K (a PK:protein ratio of 1:50) for 30 min

at 37ºC. Digestion was terminated by the addition of PMSF to a final concentration of 2

130

mM and incubation on ice for 15 min. PrPres was precipitated by the addition of sodium

phosphotungstic acid (4% w/v stock in 170 mM MgCl2, pH 7.4) to a final concentration

of 0.3% and incubation at 37ºC for 30 min. Pellets were recovered by spinning at

38,000× g for 40 min at 4ºC and then were resuspended in 1× LDS sample buffer and

boiled for 10 min. PrPres levels were then analyzed by Western blotting. For PK titration

experiments, PK was added to brain homogenates to various final concentrations,

incubated for 30 min at 37ºC, and digestions terminated by the addition of loading buffer

to 1× and subsequent boiling.

Inoculation of Mice with Prions

Mice for inoculations (C3H/B6 hybrids) were inoculated intracerebrally with 30

µL of 0.1% RML-infected brain homogenate diluted in PBS containing BSA (50 mg/mL),

penicillin (0.5 U/mL), and streptomycin (0.5 µg/mL). Age matched non-inoculated mice

inoculated were used as negative controls. Clinically ill mice were sacrificed and 10%

brain homogenates in PBS were made.

4.4 Results Identification of DPPX as a Putative PrPC-Interacting Protein

An in vivo screen using tcTPC for proteins residing in proximity to PrPC at the

cell surface resulted in the identification of DPPX as a strong candidate PrPC-interacting

protein [266]. Other top candidates such as N-CAM and ApoE were not considered for

further investigation because they are known not to be involved in prion replication or

prion disease [322, 426]. Of the identified candidate interacting proteins DPPX was

selected for further investigation based on the following reasons: 1) DPPX is not known

to be present at high concentrations in the brain and therefore is unlikely to have co-

purified with PrPC simply due to its abundance and consistent with this idea, DPPX has

not been identified in pull-downs of other target proteins (Gerold Schmitt-Ulms, personal

communication) 2) DPPX lacks the fibronectin type III and C2 domains found in many of

the other candidate PrPC-interacting proteins on the list 3) Excellent tryptic peptide

fragment coverage was obtained (13 unique peptides) and 4) DPPX has a documented

role in the trafficking of neuronal membrane proteins [375].

131

DPPX is a member of a large family of membrane bound and secreted serine

proteases. However, DPPX is unlikely to possess any proteolytic activity since it lacks

the catalytic serine residue present in other catalytically active members of the family. A

related family member, DPP10, also lacks the catalytic serine residue and an alignment of

the two proteins reveals that they share approximately 50% identity and 63% similarity

(Figure 4.1). Despite the absence of the catalytic serine residue (which has been mutated

to an aspartate residue in DPPX and a glycine residue in DPP10), the remaining two

residues of the serine protease catalytic triad remain intact as do the two glutamate

residues necessary for substrate binding. Both DPPX and DPP10 are type II

transmembrane proteins which contain a short cytoplasmic N-terminal domain, a single

transmembrane domain, and a large extracellular C-terminal domain (Figure 4.2A). The

structure of the human DPPX ectodomain has been solved and reveals two principle

domains: an 8-bladed β-propeller domain and an α/β-hydrolase domain [414].

Generation of Polyclonal Antibodies Which Recognize Murine DPPX and DPP10

In order to validate and characterize potential interactions between DPPX and

PrPC, a series of polyclonal antibodies was raised against distinct epitopes in DPPX

(Figures 4.1, 4.2A). Peptide epitopes were selected based on their position in the protein

and their potential for immunogenicity. Three polyclonal antisera directed against murine

DPPX were raised in rabbits and affinity-purified: 04DX-2 which recognizes a

juxtamembrane region in the DPPX ectodomain, 03J2 which recognizes an epitope near

the interface between the β-propeller and α/β-hydrolase domains, and 03K1 which

recognizes the C-terminus of DPPX. A polyclonal antibody, 06D10-2, which recognizes

the C-terminus of DPP10 was also generated.

The specificity of the purified anti-DPPX antibodies was tested by Western

blotting using brain homogenates prepared from either wild-type or DPPX-deficient

(DPP6df5J/Rw) mice (Figure 4.2B). Any band present in both homogenates clearly

132

Mouse_DPPX-S MTT---AKEPS--ASGKSVQQQDQELVGSNPPQRNWKGIAIALLVILVICSLIVTSVILL 55 Mouse_DPP10-1 MTAMKQEQQPTPGARATQSQPADQELGSNSPPQRNWKGIAIALLVILVVCSLITMSVILL 60 **: ::*: * ... * **** ...******************:****. ***** Mouse_DPPX-S TPAEDTSLSQKKKVTVEDLFSEDFKIHDPEAKWISNKEFIYRERKGSVILRNVETNNSTV 115 Mouse_DPP10-1 TPDELTN-SSETRLSLEELLGKGFGLHNPEPRWINDTVVVYKTNNGHVMKLNTESNASTL 119 ** * *. *.:.::::*:*:.:.* :*:**.:**.:. .:*: .:* *: *.*:* **: Mouse_DPPX-S LIEGKKIESLRAIRYEISPDKEYVLFSYNVEPVYQHSHTGYYVLSKIPHGDPQSLDPPEV 175 Mouse_DPP10-1 LLDNSTFVTFKASRHSLSPDLKYVLLAYDVKQIFHYSFTASYLIYNIHTGEVWELNPPEV 179 *::...: :::* *:.:*** :***::*:*: ::::*.*. *:: :* *: .*:**** Mouse_DPPX-S SNAKLQYAGWGPKGQQLIFIFENNIYYCAHVGKQAIRVVSTGKEGVIYNGLSDWLYEEEI 235 Mouse_DPP10-1 EDSVLQYAAWGVQGQQLIYIFENNIYYQPDIKSSSLRLTSSGKEGIIFNGIADWLYEEEL 239 .:: ****.** :*****:******** ..: ..::*:.*:****:*:**::*******: Mouse_DPPX-S LKSHIAHWWSPDGTRLAYATINDSRVPLMELPTYTGSVYPTVKPYHYPKAGSENPSISLH 295 Mouse_DPP10-1 LHSHIAHWWSPDGERLAFLMINDSLVPNMIIPRFTGALYPKAKQYPYPKAGQANPSVKLY 299 *:*********** ***: **** ** * :* :**::**..* * *****. ***:.*: Mouse_DPPX-S VIGLNGPTHDLEMMPPDDPRMREYYITMVKWATSTKVAVTWLNRAQNVSILTLCDATTGV 355 Mouse_DPP10-1 VVNLYGPTHTLELMPPDIFKSREYYITMVKWVSNTRTVVRWLNRPQNISILTLCESTTGA 359 *:.* **** **:**** : **********.:.*:..* ****.**:******::***. Mouse_DPPX-S CTKKHEDESEAWLHRQNEEPVFSKDGRKFFFVRAIPQGGRGKFYHITVSSSQPNSSNDNI 415 Mouse_DPP10-1 CSRKYEMTSDTWLSKQNEEPVFSRDGSKFFMTVPVKQGGRGEFHHIAMFLVQSKSEQITV 419 *::*:* *::** :********:** ***:. .: *****:*:**:: *.:*.: .: Mouse_DPPX-S QSITSGDWDVTKILSYDEKRNKIYFLSTEDLPRRRHLYSANTVDDFNRQCLSCDLV-ENC 474 Mouse_DPP10-1 RHLTSGNWEVIRILAYDETTQKIYFLSTESSPQGRQLYSASTEGLLNRDCISCNFMKEDC 479 : :***:*:* :**:***. :********. *: *:****.* . :**:*:**::: *:* Mouse_DPPX-S TYVSASFSHNMDFFLLKCEGPGVPTVTVHNTTDKRRMFDLEANEEVQKAINDRQMPKIEY 534 Mouse_DPP10-1 TYFDASFSPMNQHFLLFCEGPKVPVVSLHITDNPSRYFLLENNSVMKETIQKKKLAKRET 539 **..**** :.*** **** **.*::* * : * * ** *. ::::*:.:::.* * Mouse_DPPX-S RKIEVEDYSLPMQILKPATFTDTAHYPLLLVVDGTPGSQSVTERFEVTWETVLVSSHGAV 594 Mouse_DPP10-1 RILHIDDYELPLQLSFPKDFMEKNQYALLLIMDEEPGGQMVTDKFHVDWDSVLIDTDNVI 599 * :.::**.**:*: * * :. :*.***::* **.* **::*.* *::**:.:...: Mouse_DPPX-S VVKCDGRGSGFQGTKLLQEVRRRLGFLEEKDQMEAVRTMLKEQYIDKTRVAVFGKDYGGY 654 Mouse_DPP10-1 VARFDGRGSGFQGLKVLQEIHRRIGSVEAKDQVAAVKYLLKQPYIDSKRLSIFGKGYGGY 659 *.: ********* *:***::**:* :* ***: **: :**: ***..*:::***.**** Mouse_DPPX-S LSTYILPAKGENQGQTFTCGSALSPITDFKLYASAFSERYLGLHGLDNRAYEMTKLAHRV 714 Mouse_DPP10-1 IASMIL----KSDEKFFKCGAVVAPISDMKLYASAFSERYLGMPSKEESTYQASSVLHNI 715 ::: ** :.: : *.**:.::**:*:*************: . :: :*: :.: *.: Mouse_DPPX-S SALEDQQFLIIHATADEKIHFQHTAELITQLIKGKANYSLQIYPDESHYFHSVALKQHLS 774 Mouse_DPP10-1 HGLKEENLLIIHGTADTKVHFQHSAELIKHLIKAGVNYTLQVYPDEGYHIS-DKSKHHFY 774 .*:::::****.*** *:****:****.:***. .**:**:****.::: *:*: Mouse_DPPX-S RSIIGFFVECFRVQDKLPTATAKEEEEED 803 Mouse_DPP10-1 STILRFFSDCLKEEVSV---LPQEPEEDE 800 :*: ** :*:: : .: .:* **::

Figure 4.1. Alignment of murine DPPX-S and DPP10-1 amino acid sequences. Mouse DPPX-S and DPP10-1 protein sequences were aligned using the T-COFFEE algorithm and display approximately 50% identity. An arrow denotes the point at which N-terminal splicing isoforms of DPPX and DPP10 diverge and the single transmembrane domain is shown in yellow. The consensus sequence for serine proteases is boxed and note that for both DPPX and DPP10 the catalytic serine residue has been mutated (red) whereas the remaining residues of the catalytic triad remain intact (green). DPPX/DPP10 polyclonal antibody epitopes are highlighted as follows: 04DX-2 (magenta), 03J2 (turquoise), 03K1 (dark grey), and 06D10-2 (light grey).

133

A

B

250

98

64

50

36

α-DPPX Ab: 04DX-2 03J2 03K1

C

250

98

64

50

36

α-DPP10 Ab: 06D10-2*

*

*

*

*

*

***

Figure 4.2. Construction of DPPX and DPP10 polyclonal antibodies and analysis of their specificities. A: Schematic representation of DPPX and DPP10 proteins with the approximate locations of antibody epitopes shown above the protein. 03K1 and 06D10-2 are C-terminal antibodies directed against DPPX and DPP10, respectively. 03J2 recognizes an epitope in the middle of the ectodomain of DPPX and 04DX-2 recognizes an extracellular ‘juxtamembrane’ epitope in DPPX. All antibodies recognize the different N-terminal splicing isoforms of the two proteins. B: Analysis of the specificity of anti-DPPX antibodies by Western blotting. Brain homogenates from wild-type or DPPX-deficient (DPP6df5J/RW) mice were probed with the indicated anti-DPPX antibodies. Bands appearing in both strains of mice represent non-specific cross-reactivity (asterisks). Both the 03J2 and 03K1 antibodies are very clean with only minor cross-reactive bands in addition to signals at the size anticipated for authentic DPPX (arrows), while 04DX-2 has three non-specific bands of slightly greater intensity. C: Analysis of the specificity of the α-DPP10 antibody by Western blotting. Lysates from N2a cells transfected with empty vector or DPP10-1 were probed with the 06D10-2 anti-DPP10 antibody. In N2a cells, only minor cross-reactive bands are present (asterisks) in addition to authentic transfected DPP10 (arrow).

134

represents non-specific antibody cross-reactivity. Analysis of the 03J2 and 03K1

antibodies reveals that both are highly specific for DPPX (i.e. only faint cross-reactive

bands are present). Stronger cross-reactive bands are present in 04DX-2 blots, although

the authentic DPPX signal is stronger than the non-specific signal. All three antibodies

recognize a strong band at an approximate molecular weight of 100 kDa which

corresponds to the predicted molecular weight of mature glycosylated DPPX. Therefore,

all three anti-DPPX antibodies recognize their intended target. The specificity of the anti-

DPP10 antibody 06D10-2 was also evaluated by Western blotting. However, DPP10

knockout mice do not exist so lysates from empty vector- and DPP10-transfected N2a

cells were used to analyze specificity. The DPP10 antibody recognizes a band specific to

the transfected cells at the predicted molecular weight for glycosylated DPP10 and only

faint cross-reactive signals are present with this antibody (Figure 4.2C). Therefore, the

06D10-2 antibody is highly specific for DPP10.

Cloning of Various Splicing Isoforms of DPPX and DPP10

DPPX was first identified as a dipeptidyl aminopeptidase-like protein with

multiple N-terminal splice variants [376]. Two main splice variants for rat DPPX were

initially identified: DPPX-S (Short) and DPPX-L (Long). A distinct Embryonic splice

variant was later identified in mice and called DPPX-E [413, 427]. Other N-terminal

splice variants of the rat protein (including DPPX-K and DPPX-D) have also been

documented [413]. Similarly, there are multiple N-terminal splice variants of DPP10 with

4 distinct variants of the human protein (DPP10a, b, c and d) and 3 splice variants of the

mouse and rat proteins (DPP10a, c and d) [418, 421].

For the purpose of characterizing interactions between PrPC and DPPX, three

splice variants of DPPX were cloned from mRNA isolated from an adult wild-type

mouse brain (Figure 4.3A). Each variant possesses a distinct N-terminal sequence which

is fused to a short sequence common to all variants that completes the cytoplasmic

domain (Table 3.1). Sequences corresponding to the open reading frames of murine

DPPX-S and DPPX-E were inserted into the pcDNA3 mammalian expression vector. All

attempts to clone either full-length murine DPPX-L or the DPPX-L-specific N-terminal

region failed. Instead, an isoform with a cytoplasmic domain of intermediate length

135

Figure 4.3. Cloning of murine DPPX isoforms and expression in N2a cells. A: Schematic representation of DPPX N-terminal splicing isoforms. DPPX-S (short), DPPX-E (embryonic), and DPPX-M (medium) isoforms have divergent short cytoplasmic domains. B: Expression of DPPX isoforms in N2a cells. N2a cells were transfected with the indicated DPPX isoforms and lysates analyzed by Western blotting with the 03J2 anti-DPPX antibody. All three isoforms express at equivalent levels and at the predicted molecular weight. Expression of DPPX has no effect on the levels of PrPC in N2a cells as determined by Western blotting with the 7A12 anti-PrP antibody. C: DPPX isoforms are correctly localized at the cell surface in transfected N2a cells. Non-permeabilized cells transfected with the indicated isoforms of DPPX were analyzed by immunofluorescence using the indicated anti-DPPX antibodies. Strong signal along the periphery of cells indicates that DPPX is correctly targeted to the cell membrane. Scale bar = 50 µm.

136

Table 4.1. N-terminal splicing isoforms of murine DPPX and DPP10 Splice

Variant Unique N-terminal Sequence Remaining Cytoplasmic Sequence

Length(a.a.)

DPPX-S MTTAKEPSASGKSVQQQDQ ELVGSNPPQRNWKG 803

DPPX-E MNQTAGASNNVRCPPGKGHK ELVGSNPPQRNWKG 804

DPPX-M1 MASLYQRFTGGAGGRPRFQYQARSDCDEED

ELVGSNPPQRNWKG 814

DPP10-1 MTAMKQEQQPTPGARATQSQPADQ ELGSNSPPQRNWKG 800

DPP10-2 MNQTASVSHHIKCQPSKTIK ELGSNSPPQRNWKG 796 1 In DPPX-L, additional sequence is present between the two underlined residues of DPPX-M.

137

between DPPX-S and DPPX-L termed DPPX-M (Medium) was isolated. This isoform

contains portions from the beginning and end of the DPPX-L sequence but is missing

approximately 45 residues present within the middle of the DPPX-L N-terminus. The

exact origin of this splice variant is unclear since the removed DNA sequence does not

conform to the consensus sequence for mRNA splicing. Two distinct N-terminal splicing

isoforms of DPP10 were also cloned from mouse brain mRNA: DPP10-1 (which

corresponds to DPP10c) and DPP10-2 (which corresponds to DPP10a). The amino acid

sequences of all DPPX and DPP10 N-terminal splice variants used in this study are

presented in Table 4.1.

Expression of DPPX cDNAs in N2a Cells

Transfection of DPPX cDNAs into N2a neuroblastoma cells resulted in the

production of a protein recognized by an anti-DPPX antibody with an approximate

molecular weight of 100 kDa (Figure 4.3B). Expression levels of DPPX-S, DPPX-E, and

DPPX-M were approximately equal. Endogenous levels of DPPX in N2a cells are very

low and are not visible by Western blotting without using higher concentrations of

antibody or increasing the exposure time. Expression of DPPX isoforms in N2a cells had

no effect on the level of endogenous PrPC protein or its maturation. As assessed by

immunofluorescent labeling with two anti-DPPX antibodies, transfected DPPX isoforms

are correctly targeted to the plasma membrane (Figure 4.3C).

Isolation of a Novel Internal Splice Variant of DPPX-E

During the analysis of putative DPPX-E bacterial clones by restriction mapping, a

plasmid which contained a shorter-than-expected restriction fragment was isolated. When

sequenced, this clone lacked 57 nucleotides compared to the wild-type full-length

sequence. Upon analysis of the deleted DNA sequence it was noted that the 5’ end of the

deleted sequence conforms to the consensus splice donor sequence for mRNA splicing

(GT) and that the 3’ end corresponds to the junction between intron 22 and exon 23 of the

DPP6 gene (Figure 4.4A). Therefore, it appears that by virtue of a cryptic splice donor

site, 57 nucleotides of the DPP6 gene can be spliced out to generate a truncated protein

that lacks 19 amino acids within the α/β-hydrolase domain (Figure 4.4B). This splice

138

A

DPPX-E: TACATCCTCCCAGCCAAGGGAGAAAATCAAGGTCAGACTTTCACCTGCGGCTCTGCGCTCTCTCCAATAACAGACTTCAAACTCTATGCCTCTGCATTTTCTGAGAGGTACCTT ||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||DPPX-E_SV1: TACATCCTCCCAGCCAAGGGAGAAAATCAAG.........................................................CCTCTGCATTTTCTGAGAGGTACCTT

Intron 22/Exon 23Boundary

Cryptic SpliceDonor Site

B

C D

97

64

α-DPPX

DPPX-E DPPX-E_SV1

E F

Size (bp)300

200

100

263 bp (wt)206 bp (SV1)Deleted

Region

DW423

DW424

97

64

191

Monomers

Dimers

Oligomers/Aggregates

Figure 4.4. Cloning and characterization of a novel DPPX splice variant (“DPPX-E_SV1”). A: The DPPX-E_SV1 variant is generated by the utilization of a cryptic splice donor site and the standard exon 23 splice acceptor site during mRNA splicing. B: Schematic representation of the DPPX-E_SV1 protein. The splicing out of 57 nucleotides generates an in-frame deletion of 19 amino acids (residues 669 to 687) which fall within the α/β hydrolase domain. C: RT-PCR analysis of SV1 expression in mouse brain using the primers DW423 and DW424. The SV1 variant is expressed at much lower levels than full-length DPPX. D: Expression of SV1 in transfected N2a cells reveals that DPPX-E_SV1 is expressed at similar levels to wild-type DPPX. DPPX isoforms were detected with the 03J2 anti-DPPX antibody. E: Immunofluorescence on transfected N2a cells using the 03J2 anti-DPPX antibody demonstrates that some SV1 is expressed at the cell surface whereas some of the protein is aggregated (white arrows). Scale bar = 50 µm. F: SV1 forms oligomers/aggregates in transfected N2a cells. N2a cells were transfected as indicated, crosslinked with 2% formaldehyde, lysed and then analyzed by Western blotting with the 03J2 antibody. Unlike DPPX-S and DPPX-E which predominantly exist as dimers, SV1 forms high molecular weight oligomers and/or aggregates.

139

variant has not been previously documented. Consequently, the protein product was

named DPPX-E_SV1 (Splice Variant 1) or SV1 for short. Analysis of the relative

abundance of SV1 compared to full-length DPPX-E was performed using RT-PCR with

primers that flank the deleted region. SV1 is present at a much lower level in the brain

than full-length DPPX-E (Figure 4.4C). When transfected into N2a cells, SV1 expresses

at similar levels to DPPX-E although it exists as a doublet band (Figure 4.4D). The origin

of this doublet is unclear although it may be that the SV1 deletion alters the N-

glycosylation profile of the protein. When the subcellular localization of SV1 was

analyzed by immunofluorescent staining using an anti-DPPX antibody the observed

staining pattern was different than that of DPPX-E. Although some SV1 is clearly present

along the periphery of the cell indicating partial localization at the plasma membrane,

there are distinct ‘clumps’ of SV1 staining at the poles of cells (Figure 4.4E). This could

indicate that a portion of SV1 aggregates or possibly that it gets stuck in a post-

endoplasmic reticulum compartment within the cell. Further studies are required to

clarify this observation. When transfected N2a cells are treated with formaldehyde to

crosslink protein complexes, a substantial portion of SV1 is found in high molecular

weight smears on a Western blot which suggests that SV1 oligomers or aggregates may

be present in the cell (Figure 4.4F). In contrast, DPPX-S and DPPX-E exist primarily as

dimers following crosslinking. Further studies on SV1 are required to fully understand its

biophysical properties and any potential functional differences from full-length DPPX

isoforms.

Analysis of DPPX and DPP10 Expression in the Mouse Brain

Anti-DPPX antibodies were used to analyze the expression of DPPX in mouse

brains. On Western blots of brain homogenates prepared from wild-type mice, three

lower molecular weight bands in addition to full-length DPPX are visible with the 03J2

and 03K1 antibodies (Figure 4.5A). These bands are sensitive to PNGaseF digestion

indicating that they are N-glycosylated. In contrast, these additional bands are not

recognized by the 04DX-2 antibody which recognizes the juxtamembrane region of

DPPX (the additional bands present in the 04DX-2 blot are non-specific—see Figure 4.2).

Thus, these additional bands likely represent C-terminal fragments of DPPX that have

140

A

DPPX Ab: 04DX-203J203K1

- + PNGaseF97

64

51

- + - +B

03J2

03K1

Wild-typeWild-type

(+ peptide)

C D

FE

11080

60

- + PNGaseF

α-DPP10

Wild-type

06D10-2

G

11080

11080

3020

3020

15

α-DPPX

α-DPP10

α-PrP

α-Dpl

H

141

Figure 4.5. Analysis of DPPX and DPP10 expression in the wild-type mouse brain. A: Western blots of wild-type mouse brain homogenate with or without PNGaseF treatment probed with the indicated anti-DPPX antibodies. B: Western blot of wild-type mouse brain homogenate with or without PNGaseF treatment probed with the anti-DPP10 antibody 06D10-2. C-G: Detection of DPPX and DPP10 in mouse brains by immunohistochemistry. Methacarn-fixed wild-type mouse brains were probed with the indicated anti-DPPX antibodies with (D, F) or without (C, E) blocking peptide or a DPP10 antibody (G) without blocking peptide. DPPX expression is widespread in the brain and staining is abrogated in the presence of the immunogenic peptide. DPP10 is also expressed throughout the brain. Scale bar = 500 µm. H: Tissue expression profile of DPPX and DPP10 in a wild-type mouse. Homogenates from the indicated tissues were analyzed by Western blotting with antibodies to the indicated proteins. DPPX is exclusively present in the brain, spinal cord, and eye, whereas DPP10 has a more widespread expression profile.

142

been released from the cell membrane by endoproteolytic processing. Consistent with this

hypothesis, a secreted form of DPP4 generated by post-translational endoproteolytic

cleavage of the full-length protein is well-documented and is present in the serum of

humans [428]. A corresponding series of truncated membrane-associated DPPX stubs

was not observed with the 04DX-2 antibody suggesting that these products are unstable

and are rapidly removed from the brain. Based on the size of the three C-terminal

fragments, they likely represent secreted versions of DPPX which lack various amounts

of the β-propeller domain. Interestingly, analysis of DPP10 expression in mouse brain

homogenates using the 06D10-1 antibody (which recognizes the same region of DPP10

that 03K1 recognizes in DPPX) failed to demonstrate the existence of cleaved DPP10

products (Figure 4.5B). Thus, endoproteolytic processing of DPPX may be a feature

unique to this protein that is involved in tailoring its biological activity.

Immunohistochemical labeling of methacarn-fixed mouse brains with anti-DPPX

antibodies demonstrates that DPPX has a widespread pattern of expression within the

brain (Figure 4.5C, E). Staining is observed in numerous neuroanatomic regions

including the cerebellum, hippocampus, cerebral cortex, olfactory bulb, and the brainstem.

All observed signal represents authentic DPPX staining for the following reasons: 1) An

identical pattern of staining is observed with two distinct anti-DPPX antibodies, both of

which are largely devoid of any cross-reactive species on a Western blot (Figures 4.2,

4.5C, E), 2) Staining with both antibodies is completely abrogated when the antibodies

are pre-incubated with their immunogenic peptides (Figure 4.5D, F), and 3) The vast

majority of signal obtained with either anti-DPPX antibody is absent when brains from

DPPX-deficient mice (DPP6df5J/Rw) are probed (data not shown). Like DPPX, DPP10 is

also expressed throughout the brain (Figure 4.5G). Thus, DPPX protein is present in most

areas of the mouse brain and the existence of secreted versions of DPPX may further

augment its distribution pattern.

The tissue expression patterns of DPPX and DPP10 were next investigated by

Western blotting using tissue homogenates prepared from a wild-type mouse. Signal for

DPPX protein was only observed in homogenates prepared from either the brain, spinal

cord, or eye (Figure 4.5H). While DPP10 protein was also observed in these three tissues,

it was also present (albeit at lower levels) in the spleen, kidney, lung, and liver implying

143

that DPP10 has a more widespread tissue distribution than DPPX. The high levels of

DPPX found in the brain and spinal cord mirror the high levels of PrPC found in these

tissues. In contrast, Doppel protein was only observed in the testis and, to a lesser extent,

the heart which is in agreement with published studies on the expression pattern of Prnd

mRNA [190].

Effect of PrP Levels on DPPX Expression and Endoproteolytic Processing

In order to assess whether the level of PrPC protein in the brain has any effect on

DPPX levels or processing, brain homogenates from mice expressing various amounts of

PrPC were analyzed by Western blotting with anti-DPPX antibodies. The following

mouse lines were tested: wild-type C57/B6 mice (1× relative PrP expression level), the

ZrchI strain of Prnp0/0 mice (no PrP expression) [186], and Tg(SHaPrP)7 mice which

over-express hamster PrP (Tg7+/+ homozygotes or Tg7+/- heterozygotes which express

PrP at approximately 8× and 4× wild-type levels, respectively) [126, 386, 388]. No effect

of PrPC expression level on either DPPX levels or endoproteolytic cleavage was observed

using two distinct anti-DPPX antibodies (Figure 4.6). Thus, the stability and processing

of DPPX is not modulated by the amount of PrPC protein present in the brain.

DPPX Exists as a Homodimer in Tissue Culture Cells and in the Brain

The soluble human DPPX ectodomain generated in insect cells using the

baculovirus system crystallized as a dimer (Figure 4.7A) [414]. Similarly, the crystal

structure of the human DPP4 ectodomain also exhibits a dimeric stoichiometry [429]. In

order to confirm that DPPX exists as a dimer in vivo (i.e. on the cell surface),

crosslinking assays were performed using either live cells or whole mouse brains. When

N2a cells were transfected with plasmids encoding either DPPX-S or DPPX-E, a band at

an approximate molecular weight of 95 kDa was obtained (Figure 4.7B). When

transfected cells were crosslinked with 2% formaldehyde prior to cell lysis in order to

lock together protein complexes, the vast majority of the DPPX signal shifted to an

approximate molecular weight of 190 kDa (Figure 4.7B). This indicates that transfected

DPPX exists as a dimer on the surface of N2a cells. The assignment of this high

molecular weight band as a DPPX dimer was made based on the following two reasons:

144

148

98

64

50

148

98

64

50

α-DPPX 03J2

α-DPPX 03K1

Figure 4.6. PrPC levels have no effect on DPPX expression or endoproteolysis. Brain homogenates from mice (all on the C57/B6 background) with various levels of PrPC expression were analyzed by Western blotting using the indicated anti-DPPX polyclonal antibodies. Levels of DPPX and its associated endoproteolytic products were not modulated by the level of PrPC protein in the brain.

145

Figure 4.7. DPPX exists as a dimer in vivo. A: Crystal structure of the secreted human DPPX ectodomain (PDB #1XFD) generated using the baculovirus expression system [414]. Recombinantly-produced DPPX ectodomain crystallized as a dimer. B: Transfected membrane-anchored DPPX exists as a dimer in N2a cells. N2a cells were transfected with plasmids encoding the indicated proteins, cross-linked as indicated with 2% formaldehyde, lysed, and then analyzed by Western blotting with the 03J2 anti-DPPX antibody. Crosslinked dimeric DPPX is the main species observed. C: DPPX exists as a dimer in mouse brains. Brains from mice of the indicated strain were crosslinked in vivo using the tcTPC procedure and analyzed by Western blotting. Although the efficiency of crosslinking is lower, dimeric DPPX is still visible. D: Analysis of DPP10 dimer formation in transfected N2a cells. N2a cells were transfected with plasmids encoding the indicated proteins, cross-linked as indicated with 2% formaldehyde, lysed, and then analyzed by Western blotting with the 06D10-2 anti-DPP10 antibody. The efficiency of crosslinked DPP10 dimer formation is lower than that of DPPX.

146

1) the size of the cross-linked band is approximately double the size of the monomeric

protein and 2) the efficiency of formation of the high molecular weight band is so high

that it is unlikely to represent a monomer of DPPX forming complex with a separate

protein. Dimeric DPPX was also observed in mouse brains which had been crosslinked in

vivo with formaldehyde using the tcTPC procedure [266] (Figure 4.7C). In this situation

crosslinking efficiency was lower, likely due to the mild conditions used and the unequal

penetration of the perfused formaldehyde into all regions of the brain. The ability of

DPP10 to form dimers in vivo was also tested using DPP10-transfected N2a cells.

However, only a small fraction of the DPP10 signal shifted to the dimeric molecular

weight upon crosslinking (Figure 4.7D). The reason for this is not clear, although it could

either reflect an inherent inability of DPP10 to form homodimers or the lack of suitable

residues in DPP10 for the efficient formation of formaldehyde-induced covalent

crosslinks. Epitope-tagged DPP10 has been demonstrated to interact with itself in

transfected HEK293 cells using co-immunoprecipitation assays [420] suggesting that the

latter possibility may be true, although the efficiency of dimer formation cannot be

determined using this technique. Further experiments are required to clarify this issue.

Interactions Between DPPX and Prion Proteins in N2a Cells

Mild formaldehyde crosslinking of transfected N2a cells was used to analyze

complex formation between members of the prion protein family and isoforms of either

DPPX or DPP10. Formaldehyde forms covalent crosslinks between two accessible amino

acid residues (primarily lysine, arginine, and tyrosine residues [430, 431]) that are

sufficiently proximal to each other (around 2-3 Å). This technique has numerous

advantages over other protein-protein interaction tools. Firstly, crosslinking is performed

on live cells prior to cell lysis meaning that all observed complexes represent interactions

that cannot simply be rationalized as being formed upon cell lysis. Secondly, protein

complexes are analyzed in their native environments (the plasma membrane, for instance).

Thirdly, the crosslinking moiety is short and is therefore likely to favour crosslinking of

proteins which are interacting with each other over proteins which reside in loose

proximity to each other. Fourthly, unlike most protein-protein interaction techniques,

formaldehyde crosslinking is amenable to the study of protein complexes containing

147

membrane proteins. Fifthly, formaldehyde crosslinking has been widely used to study

protein-DNA interactions using chromatin immunoprecipitation assays. Sixthly,

crosslinking with formaldehyde is not known to introduce rearrangement of molecular

architecture when assessed by electron microscopy. Finally, this technique has previously

been used to identify N-CAM as a PrPC-interacting protein [322]. Notably, N-CAM is

one of the few identified candidate PrPC-interacting proteins which may have

physiological relevance to PrPC biology [330].

N2a neuroblastoma cells were transfected with various isoforms of DPPX and

DPP10, crosslinked with formaldehyde, and then PrPC-containing protein complexes

analyzed by Western blotting. Transfection of either DPPX-S or DPPX-E led to the

formation of a high molecular weight complex that contains PrPC (Figure 4.8A). In

contrast, no complex formation was observed when isoforms of DPP10 were transfected.

The fact that both DPPX-S and DPPX-E formed complexes with PrPC suggests that the

binding determinant for complex formation exists in the DPPX ectodomain (since the

DPPX isoforms differ in their cytoplasmic domains) and is consistent with localization of

PrPC on the cell surface. Identical results were also obtained by PrP and DPPX co-

transfections in HEK293 cells (data not shown) implying that complex formation

between PrPC and DPPX is not cell type-specific. The exact size of the crosslinked

complex is difficult to assess since the crosslinked proteins do not form a linear chain and

likely retain some aspects of structure due to intramolecular crosslinks. However,

PrPC/DPPX complexes migrate more slowly on a Western blot than DPPX dimers (data

not shown) arguing that the complex consists of a homodimer of DPPX and either one or

two molecules of PrPC. Although a dimeric DPPX structure implies the capability to bind

two copies of PrPC, this has not been confirmed experimentally.

The three dimensional structures of the C-terminal α-helical domains of PrPC and

Doppel are very similar [221]. Therefore, the ability of Doppel to form complexes with

DPPX was also tested. Remarkably, transfected Doppel also formed high molecular

weight complexes with DPPX-S and DPPX-E but not DPP10 (Figure 4.8B). N-terminally

truncated PrP (PrPΔ32-121 or ΔPrP) has a similar domain structure to Doppel and both

proteins cause a similar neurodegenerative disease when expressed in the brain [217,

289]. Accordingly, the ability of ΔPrP to interact with DPPX isoforms was also assessed.

148

Figure 4.8. DPPX forms high molecular weight complexes with all three members of the mammalian prion protein family as assessed in tissue culture cells. A: N2a cells were transfected with plasmids encoding the indicated proteins, crosslinked with formaldehyde, and PrPC-containing complexes assessed by Western blotting. Endogenous PrPC forms complexes with DPPX-S and DPPX-E but not DPP10. B: N2a cells were co-transfected with Doppel and the indicated plasmids and then crosslinked with formaldehyde. Doppel also forms complexes with DPPX-S and DPPX-E but not DPP10. C: HEK293 cells were co-transfected with PrPΔ32-121 and the indicated plasmids and then crosslinked with formaldehyde. ΔPrP is capable of forming complexes with DPPX-S and DPPX-E. D: N2a cells were co-transfected with Shadoo and the indicated plasmids and then crosslinked with formaldehyde. Like PrPC and Doppel, Shadoo forms complexes with DPPX-S and DPPX-E but not DPP10. E: N2a cells were co-transfected with Thy-1 and the indicated plasmids and then crosslinked with formaldehyde. No high molecular weight complexes are observed between Thy-1 and either DPPX or DPP10.

149

For this purpose, HEK293 cells were used since endogenous PrPC in N2a cells would

interfere with the experimental analysis (as there is no ΔPrP-specific antibody). ΔPrP was

also able to form complexes with both DPPX-S and DPPX-E (Figure 4.8C). The ability

of the third member of the prion protein family, Shadoo, to form complexes with DPPX

was tested as well. Surprisingly, Shadoo was also able to form complexes with DPPX

isoforms but not DPP10 isoforms in transfected N2a cells (Figure 4.8D). PrPC, Doppel,

and Shadoo are all anchored to the cell surface by means of a GPI anchor [202, 229, 381].

Therefore, another GPI-anchored protein, Thy-1, was tested for complex formation to see

if DPPX forms complexes non-specifically with many GPI-anchored proteins. However,

no complex formation between Thy-1 and either DPPX or DPP10 was observed in

transfected N2a cells (Figure 4.8E). This argues that complex formation between DPPX

and member of the prion protein family is a specific event.

In order to confirm that DPPX is present in the high molecular weight PrP-

containing complex, co-immunoprecipitations were performed. This experiment was

necessary because it is theoretically possible that transfection of DPPX results in the up-

regulation of another protein which subsequently forms a high molecular weight complex

with PrPC of similar size to a predicted PrPC/DPPX complex. Following

immunopurification of PrP-containing complexes and blotting with an anti-DPPX

antibody, a band was observed that is identical in size to the one observed on the PrP blot

(Figure 4.9A). Thus, the crosslinked high molecular complex contains both PrPC and

DPPX. Following crosslink reversal and comparison to non-crosslinked DPPX it was

determined that full-length glycosylated DPPX is present in the complex.

Co-immunoprecipitation experiments were also performed on non-crosslinked

brain homogenates prepared from wild-type mice. For this purpose, homogenates were

solubilized with the non-ionic detergent DDM prior to immunoprecipitation.

Immunoprecipitation with the anti-DPPX antibody 03K1 led to efficient precipitation of

DPPX from mouse brain homogenates (Figure 4.9B). PrPC co-precipitated with DPPX

suggesting that DPPX and PrPC form a complex in vivo. Although a small amount of

PrPC was obtained in a mock immunoprecipitation (which is probably due to a non-

specific interaction between a fraction of PrPC and the protein G-Sepharose beads), the

signal for PrPC was clearly enriched in the anti-DPPX immunoprecipitation. To further

150

A

IP:Reversal:

Crosslink:

DPPX

191

97

α-DPPX

+ + + -

- - + -NMS PrP PrP

PrP/DPPXComplex

-

B

C D

E

26016011060

30

40

2015

α-PrP

IP with α-HA:Reversal:

HA-DPPX-S: - + - +

- - + +- ++-

26016011060

30

40

2015

α-Dpl


HA-DPPX-S: - + - +

- - + +- ++-

26016011060

30

40

2015

α-Sho


HA-DPPX-S: - + - +

- - + +- ++-

Sho

DplPrP

α-DPPX

α-PrP

α-PrP

α-DPPX

wt wt Prnp0/0

Figure 4.9. Analysis and confirmation of complexes containing prion proteins and DPPX. A: N2a cells were transfected with DPPX-S, crosslinked with 2% formaldehyde, lysed, and then immunoprecipitations performed with either normal mouse serum (NMS) or the anti-PrP antibody 7A12. Following pull-down of PrPC-containing complexes, a band reactive to DPPX antibodies is observed confirming the identity of the band as a complex between PrPC and DPPX. After cross-link reversal, full-length glycosylated DPPX is obtained. B: PrPC and DPPX can be co-immunoprecipitated without cross-linking. Detergent-extracted wild-type mouse brain homogenates were subjected to immunoprecipitation with either anti-DPPX (03K1) or anti-PrP (8B4) antibodies. DPPX and PrPC reciprocally co-immunoprecipitate at endogenous levels of protein expression. C-E: Analysis of prion proteins in high molecular weight complexes with DPPX. N2a cells were co-transfected with HA-tagged DPPX-S and either PrP (C), Dpl (D), or Shadoo (E) plasmids. Following crosslinking, high molecular weight complexes are observed (arrowheads). Prion protein/DPPX complexes were isolated by immunoprecipitation using an anti-HA antibody and then crosslinks were reversed. The prion proteins found within complexes with DPPX are full-length and glycosylated.

151

confirm this result, reciprocal co-immunoprecipitations were also performed.

Immunoprecipitation with an anti-PrP antibody (8B4) resulted in efficient precipitation of

PrPC and co-precipitation of DPPX (Figure 4.9B). No DPPX was observed when the pull-

down was performed using homogenate from a Prnp0/0 mouse brain suggesting that

DPPX does not bind non-specifically to the 8B4 antibody or the protein G beads. These

experiments suggest that PrPC and DPPX interact at physiological levels of protein

expression in a wild-type animal and that the interaction occurs in the absence of a

crosslinking agent.

PrPC exists on the cell surface as a mixture of un-, mono-, and di-glycosylated

forms and is subject to endoproteolytic cleavage. Both Doppel and Shadoo exist

primarily in fully-glycosylated forms but both are also susceptible to endoproteolytic

processing. Therefore, the exact species of prion protein present in the high molecular

weight complex with DPPX was analyzed by co-immunoprecipitation and Western

blotting. Following pull-down of DPPX-containing complexes and blotting for PrP, all

three glycosylated variants of PrP were obtained (Figure 4.9C). However, DPPX appears

to have a slight preference for di-glycosylated PrPC. Comparison to input PrPC signals

confirms that full-length PrPC is the predominant species present in the high molecular

weight complexes with DPPX. Similar analyses were also performed for Doppel and

Shadoo. As is the case for PrPC, full-length fully-glycosylated Doppel (Figure 4.9D) and

Shadoo (Figure 4.9E) are preferentially present in complexes with DPPX.

In order to determine whether prion protein/DPPX complexes are present on the

cell surface, biotinylation assays were performed. Transfected cells were treated with a

cell impermeant biotin moiety followed by formaldehyde crosslinking and isolation of

biotinylated complexes by neutravidin pull-down. High molecular weight complexes

between DPPX and PrPC, Doppel, or Shadoo were observed in the biotinylated fraction

confirming that the complexes exist on the cell surface (Figure 4.10A). It was next

investigated whether PrPC/DPPX complexes result from interactions occurring in the

same cell (cis) or between neighbouring cells (trans). HEK293 cells were either co-

transfected with PrP and DPPX-S (to assess cis interactions) or singly transfected with

either PrP or DPPX-S and then mixed (to assess trans interactions) prior to performing

the crosslinking assay. A strong band was obtained for the co-transfected cells whereas

152

191

97

64

51

39

28

19

14

- -+ +DPPX-S:cis trans

α−PrP

A B

191

97

64

51

39

28

1914

α−PrP α−Dpl α−Sho

Figure 4.10. PrPC/DPPX complexes are present at the cell surface and are composed of adjacent molecules displayed on the same cell. A: Prion protein/DPPX complexes are located at the cell surface. N2a cells were transiently transfected with the indicated plasmids and then 24 hours later cells were biotinylated, crosslinked with 2% formaldehyde and lysed. Biotinylated complexes were isolated with neutravidin and then analyzed by Western blotting with the indicated antibodies. PrPC/DPPX, Dpl/DPPX, and Sho/DPPX complexes (black arrow) were all found in the biotinylated fraction indicating that they are present on the cell surface. B: PrPC/DPPX complexes result from interactions within the same cell. HEK293 cells were either co-transfected with PrP with or without DPPX-S (in “cis” with respect to cellular disposition) or singly transfected with either PrP or DPPX-S (in “trans” with respect to cellular disposition). 24 hours post-transfection, cells were trypsinized and either re-plated (cis) or PrP and DPPX-S singly transfected cells were mixed (trans). After an additional 24 hour incubation, cells were crosslinked with 2% formaldehyde, lysed, and PrPC-containing complexes analyzed by Western blotting. PrPC/DPPX complexes (black arrow) are only formed when PrPC and DPPX are present in the same cell (i.e. co-transfected).

153

only a faint band was obtained with the mixed singly-transfected cells indicated that

PrPC/DPPX complexes result primarily from interactions within the same cell (Figure

4.10B).

Mapping of DPPX Complex Determinants in PrPC

In order to map determinants in PrPC which govern complex formation with

DPPX a series of PrP deletion mutants (Figure 4.11A) was used in conjunction with the

crosslinking assay. HEK293 cells were used exclusively for PrP mapping experiments

due to the extremely low levels of endogenous wild-type PrPC in these cells. For mapping

of determinants in the α-helical C-terminal domain of PrPC, a PrPΔ23-88 backbone was

used due to the previous characterization of similar mutants [122] and the decrease in

endocytosis (and concomitant increase in cell surface levels) obtained by deleting the N-

terminus of PrP ([212] and data not shown). Whereas PrPΔ23-88 retained the ability to

form complexes with DPPX, PrP deletion mutants encompassing either helix A, B, or C

failed to form distinguishable complexes with DPPX-S (Figure 4.11B). This suggests that

an intact α-helical domain in PrPC is essential for complex formation and is consistent

with the ability of DPPX to bind to PrPΔ32-121 (Figure 4.8C). It should be noted that the

PrP mutants with helical deletions appear to express at lower levels than the parental

mutant. Although all three PrP helical deletion mutants were expressed, one possible

caveat exists in that the correct targeting of these mutants to the cell surface has not been

confirmed. Deletions in the α-helical domain of PrP alleles with intact N-termini have

been documented to accumulate intracellularly and cause a neuronal storage-like disease

[120]. Nonetheless, one of the mutants used here (PrPΔ23-88/Δ141-176, also known as

PrP106) is known to be capable of forming miniprions and therefore is likely to exist at

the cell surface [122]. To search for additional complex formation determinants in PrPC,

various N-terminal deletion mutants of PrP were used (Figure 4.11A). As deletions

progressively invaded the hydrophobic tract of PrP, complex formation with DPPX-S

became noticeably weaker (Figure 4.11C). This suggests that additional N-terminal

DPPX complex determinants are present in PrPC and that the hydrophobic tract is

important for modulating complex formation. Slight differences in expression levels

between the individual mutants hinders precise evaluation of complex formation

154

B C

A

191

97

64

51

39

281914

α-PrP C-terminus (R1)

- + - + - + - +DPPX-S:

α-PrP D13

- + - + - + - +

191

97

64

51

39

281914

DPPX-S:

Figure 4.11. Mapping of DPPX complex determinants in PrPC demonstrates that the C-terminal α-helical domain is required for complex formation and implies the existence of a second binding site. A: Schematic representation of murine PrP domain architecture showing the deletion mutants used for mapping DPPX complex determinants. B-C: Mapping of DPPX complex determinants in PrPC. HEK293 cells were transfected with the indicated PrP deletion mutants in the presence or absence of DPPX-S, cross-linked with 2% formaldehyde, and PrP/DPPX complexes (arrows) assessed by Western blotting using the indicated anti-PrP antibodies. No complex formation was observed for any PrP mutants with deletions within the α-helical domain suggesting that the α-helical domain is necessary for binding to DPPX. As deletions progressively invade the hydrophobic tract, complex formation with DPPX weakens implying the existence of a second DPPX binding site within this region.

155

efficiency. Nonetheless, there is a clear difference in the complex formation efficiency

between the PrPΔ32-105 and PrPΔ32-121 mutants and these two mutants express at

similar levels. Thus, complex formation with DPPX requires at least two distinct sites in

PrPC. Both PrPΔ32-121 and PrPΔ32-134 exhibited lower binding to DPPX compared to

PrPΔ32-105 and interestingly, the former two molecules are toxic to cerebellar cells in

Prnp0/0 mice whereas the latter mutant is not [217].

Mapping of DPPX Complex Determinants in Doppel

Regions in Doppel responsible for complex formation with DPPX were

determined by co-transfection of N2a cells with DPPX-S and Doppel deletion mutants

using the crosslinking assay. The deletion mutants used (Figure 4.12A) were Δ29-49

(which deletes the N-terminus of Doppel), Δ50-90 (which deletes helix A and the two

short β-strands), and Δ126-149 ( which deletes helix C). Both wild-type Doppel and

DplΔ29-49 formed complexes with DPPX as detected with the E6977 antibody which

was raised against recombinant Doppel (Figure 4.12B). It should be noted that the

efficiency of DPPX complex formation for the Δ29-49 mutant was weaker, implying that,

although not necessary for complex formation, this region may contribute to binding

affinity or contain residues that are more amenable to crosslinking. Doppel mutants with

deletions encompassing either Doppel helix A or helix C retained the ability to form

complexes with DPPX as detected by the 03A2 antibody which recognizes the N-

terminus of Doppel. Collectively, these results suggest that helix B of Doppel and the

preceding loop (residues 91-125) are necessary for complex formation with DPPX.

Consistently, a Doppel deletion mutant which lacks both helices A and C (Δ50-90/Δ126-

149) retained the ability to form complexes with DPPX (Figure 4.12B). The region in

Doppel required for complex formation with DPPX corresponds to one of the structural

features that distinguishes Doppel from PrPC. The helix B region in Doppel is separated

into two parts (helix B/B’) generating a kinked structure (Figure 4.12C). Notably, this is

also the region in Doppel which is both necessary and sufficient for Doppel-induced

toxicity in cultured Prnp0/0 CGNs [271].

156

03A2

Dpl/DPPXComplex191

97

64

51

39281914

E6977

DPPX-S:

Doppel Ab: 03A2

- + - + - + - + - +

B

A

C

N

C Figure 4.12. Mapping of DPPX complex determinants in Doppel demonstrates that the helix B/B’ region is necessary for complex formation. A: Schematic representation of murine Doppel domain structure showing the deletion mutants used for mapping DPPX complex determinants. B: Mapping of DPPX complex determinants in Doppel. N2a cells were transfected with the indicated Doppel mutants in the presence or absence of DPPX-S, crosslinked with 2% formaldehyde, and Dpl/DPPX complexes assessed by Western blotting using the indicated anti-Dpl antibodies. Both Δ29-49 and Δ50-90/Δ126-149 Doppel mutants retain the ability to form complexes with DPPX suggesting that residues 91-125 are required for complex formation. Weaker complexes were observed for some mutants (such as Δ29-49) suggesting that other regions of Doppel may be capable of modulating complex formation. C: The DPPX complex determinant (red) in the murine Doppel structure (PDB #1I17) encompasses the helix B/B’ region and the preceding loop.

157

Mapping of DPPX Complex Determinants in Shadoo

Determinants in Shadoo were mapped using deletion mutants of FLAG-tagged

Shadoo (Figure 4.13A). FLAG-tagged Shadoo was used in order to facilitate the

immunodetection of the deletion mutants. Shadoo deletion mutants lacking either the

hydrophobic tract (residues 62-77) or a C-terminal region (residues 78-100) were wild-

type Shadoo-like in their ability to form complexes with DPPX (Figure 4.13B). In

contrast, a Shadoo deletion mutant which lacks the N-terminus (residues 30-61 which

contain the positively-charged tetrarepeats and the region preceding the hydrophobic

tract) exhibited markedly reduced complex formation with DPPX. This suggests that the

N-terminus of Shadoo is important for complex formation. Interestingly, higher levels of

monomeric non-crosslinked protein were observed with this mutant although it expresses

at a similar level to wild-type Shadoo in non-crosslinked lysates (see Figure 4.2).

Therefore, deletion of the N-terminus of Shadoo may also diminish its ability to interact

with other proteins. Consistently, high molecular smears were also reduced with this

Shadoo deletion mutant following crosslinking. An alternate interpretation of this

phenomenon is that deletion of residues 30-61 in Shadoo removes all the residues that are

amenable to crosslinking. Further experiments are required in order to investigate this

possibility. The necessity of the glycosylation site in Shadoo for complex formation with

DPPX was also tested. Surprisingly, removal of the N-glycosylation site (by using an

N107Q mutant) abolished complex formation (Figure 4.13C). This suggests that both the

N-terminus of Shadoo and the Asn-linked sugar contribute to DPPX complex formation.

Mapping of Complex Determinants in DPPX

DPPX-S is a large protein of 803 amino acids with a complex domain structure in

which N- and C-terminal portions of the ectodomain contribute to the α/β-hydrolase

domain and dimer formation is governed by residues in both the α/β-hydrolase domain

and the β-propeller domain [414]. The first two DPPX mutants tested for complex

formation with prion proteins were a mutant lacking the entire cytoplasmic domain

(including the cytoplasmic sequence which is common to all DPPX N-terminal splicing

isoforms), termed DPPXΔCyto, and a secreted version of the DPPX ectodomain (residues

56-803) generated by fusing the ectodomain sequence to the PrP signal sequence (Figure

158

26016011080605040

30

201510

DPPX-S: - + - +

α-FLAG M2

26016011080605040

30

201510

DPPX-S: - + - + - + - +

α-FLAG M2

B C

A

Figure 4.13. Mapping of DPPX complex determinants in Shadoo demonstrates that the N-terminal domain and the N-glycosylation site contribute to complex formation. A: Schematic representation of murine Shadoo domain architecture showing the deletion mutants used for mapping DPPX complex determinants. The N107Q mutant abrogates the N-glycosylation event (CHO) at residue 107. B-C: Mapping of DPPX complex determinants in Shadoo. N2a cells were transfected with the indicated FLAG-tagged Shadoo mutants in the presence or absence of DPPX-S, crosslinked with 2% formaldehyde, and Sho/DPPX complexes (arrows) assessed by Western blotting using an anti-FLAG antibody. The Δ30-61 mutant exhibits markedly reduced complex formation suggesting that residues 30-61 of Shadoo which contain the basic tetrarepeats are necessary for binding to DPPX. An N107Q mutant which lacks the N-glycosylation site also fails to form complexes with DPPX.

159

4.14A). The DPPXΔCyto mutant retained the ability to form complexes with PrP in N2a

cells suggesting that the N-terminus of DPPX is not essential for complex formation

(Figure 4.14B). This is not surprising since PrPC exists on the cell-surface. When the

secreted DPPX ectodomain construct was tested, it failed to form complexes with

endogenous PrPC in N2a cells (Figure 4.14C). This implies either that the transmembrane

domain of DPPX is necessary for complex formation or that anchorage of DPPX to the

cell membrane is required. Similarly, Shadoo was capable of forming complexes with

DPPXΔCyto but not secreted DPPX (Figure 4.14D). Both DPPXΔCyto and secreted

DPPX efficiently formed dimers (Figure 4.14E) and secreted DPPX was enriched in the

conditioned medium fraction compared to wild-type DPPX (Figure 4.14F) confirming

that the mutant DPPX constructs behave as designed. Cumulatively, these results argue

that complex formation between DPPX and prion proteins requires extracellular DPPX

epitopes and that DPPX must be anchored to the plasma membrane.

In order to map the region(s) in the DPPX ectodomain responsible for complex

formation with prion proteins, a series of C-terminally truncated HA-tagged DPPX

deletion mutants were tested using the crosslinking assay (Figure 4.15A). In this instance,

a slight modification to the crosslinking assay was utilized: following crosslinking,

DPPX-containing complexes were immunoprecipitated using an anti-HA antibody and

crosslinks reversed prior to the detection of PrP, Doppel, or Shadoo by Western blotting.

PrP was capable of forming complexes with DPPXΔ516-803 and smaller C-terminal

DPPX deletion mutants but did not bind to a Δ394-803 mutant (Figure 4.15B). This

suggests that residues 394-515 of DPPX are important for complex formation with PrPC.

Interestingly, this sequence of residues lies within the 8-bladed β-propeller domain.

Doppel exhibited a similar pattern of complex formation with the series of DPPX

deletion mutants although low levels of binding were also observed for the Δ81-803 and

Δ394-803 mutants. This implies that both PrPC and Doppel bind primarily to the same

region of DPPX, namely the β-propeller domain. In stark contrast, Shadoo bound

strongly to all DPPX mutants tested suggesting that residues N-terminal to residue 81 in

DPPX govern complex formation with Shadoo. This region corresponds to the

juxtamembrane region of DPPX. All DPPX deletion mutants expressed at comparable

levels in non-crosslinked transfected N2a cells (Figure 4.15C) and were

160

19197

64

51

39

281914

A

B C

26016011080

60

5040

302015

D

80

260160110

60

5040

302015

α-PrP α-PrP α-Sho

E F

98

98

α-DPPX

Cell lysate

Medium

26016011080

α-DPPX: 03K1 03J2

Dimers

Monomers

Figure 4.14. Membrane anchorage but not the cytoplasmic domain of DPPX is required for complex formation with prion proteins. A: Schematic representation of DPPX mutants. DPPXΔCyto lacks all but three residues of the cytoplasmic domain whereas secDPPX directs the secretion of the DPPX ectodomain into the lumen/extracellular space. B-C: N2a cells were transfected with the indicated plasmids, crosslinked, lysed, and PrPC/DPPX complexes (arrows) analyzed by Western blotting. The cytoplasmic domain of DPPX is not required for complex formation with PrPC whereas the secreted DPPX ectodomain fails to bind to PrPC. D: N2a cells were co-transfected with Shadoo and the indicated plasmids, crosslinked, lysed, and Sho/DPPX complexes (arrow) analyzed by Western blotting. DPPXΔCyto forms complexes with Sho whereas secDPPX does not. E: N2a cells were transfected with the indicated plasmids, crosslinked, lysed, and DPPX analyzed by Western blotting. Both DPPXΔCyto and secDPPX are capable of forming dimers. F: N2a cells were transfected with the indicated plasmids, lysed, and DPPX protein in the cell lysate and conditioned medium analyzed by Western blotting. Whereas only minimal amounts of wild-type DPPX-S are present in the conditioned medium, large amounts of secDPPX are found in this fraction. Accordingly, a smaller proportion of secDPPX is found in the cell lysate.

161

B

40

30

20

40

30

20

30

20

15

PrP

Dpl

Sho

C

98

62

49

38

28

1714

α-HA

D

A

Figure 4.15. Two distinct sites in DPPX mediate complex formation with prion proteins. A: Schematic representation of HA-tagged DPPX deletion mutants. B: N2a cells were co-transfected with the indicated HA-tagged DPPX mutants and either PrP (top), Dpl (middle), or Sho (bottom). Transfected cells were crosslinked, lysed, and DPPX complexes immunoprecipitated with an anti-HA antibody. Prion protein/DPPX complexes were detected by Western blotting following crosslink reversal with antibodies to either PrP (D18), Dpl (03A2), or Sho (06rSH-1). Whereas both Dpl and PrPC bind most strongly to DPPX mutants containing residues 394-515, Sho binds strongly to all deletion mutants. C: Expression of DPPX deletion mutants in transfected N2a cells as assessed by Western blotting. D: The DPPX protein structure with mapped PrPC/Dpl (red) and Sho (green) epitopes. PrPC and Dpl bind to the β-propeller domain whereas Shadoo binds to the juxtamembrane region.

162

immunoprecipitated equally by the anti-HA antibody (data not shown). It should be noted

that the slight decrease in expression of the Δ666-803 mutant relative to the Δ516-803

and Δ787-803 mutants likely explains the smaller amounts of co-precipitated PrP and

Doppel obtained for the Δ666-803 mutant. Therefore, it appears that two distinct epitopes

in DPPX control complex formation with prion proteins (Figure 4.15D). This notion is in

agreement with the existence of multiple binding DPPX binding determinants in prion

proteins.

Competition for DPPX Complex Formation Between PrPC and Doppel

In order to test whether or not PrPC and Doppel bind to the same site on DPPX,

competition experiments were performed in transfected N2a cells. Increasing amounts of

transfected Doppel led to an increase in Doppel/DPPX complexes and a concomitant

diminution in endogenous PrPC/DPPX complex levels (Figure 4.16A). This suggests that

over-expression of Doppel can remove PrPC from DPPX and that the two proteins share a

common binding site on DPPX (i.e. the β-propeller domain). Although an alternative

explanation is possible (i.e. the non-specific swamping of DPPX binding sites by over-

expressed protein), the fact that both PrP and Doppel require an intact α-helical domain

for complex formation with DPPX argues that a shared binding site exists within this

domain. In contrast, when increasing amounts of PrPΔ32-121 were transfected, the levels

of endogenous wild-type PrPC/DPPX complexes did not decrease substantially (Figure

4.16B). This result implies that endogenous wild-type PrPC binds much more strongly to

DPPX than transfected PrPΔ32-121. This result is further suggestive of a second binding

site for DPPX in PrPC (somewhere N-terminal to residue 121), reinforcing the epitope

mapping data presented in Figure 4.11.

Expression of DPPX in the Granule Cells of the Cerebellum

The granule cells of the cerebellum are known to be permissive to Doppel or

ΔPrP-induced toxicity both in vivo and in vitro in a manner that is dependent on the

absence of PrPC [217, 271, 289]. Because DPPX binds to PrPC, Doppel, ΔPrP, and

Shadoo, it is a reasonable candidate for controlling prion protein neurotoxicity and/or

neuroprotection in CGNs. Accordingly, the expression of DPPX in CGNs was

163

Dpl:PrP/DPPXComplex191

97

64

51

39

28

1914

α-PrP

A

Dpl/DPPXComplex

Dpl:

α-Dpl

191

97

64

51

39

28

1914

191

97

64

51

39

28

19

PrPΔ32-121:

wt PrP/DPPXComplex

B

α-PrP (wt) Figure 4.16. Competition experiments suggest that Dpl and PrPC share a common binding site on DPPX. A: N2a cells were co-transfected with DPPX-S and increasing amounts of Doppel plasmid, crosslinked, lysed, and PrPC (left) and Dpl (right) complexes analyzed by Western blotting. As the amount of Dpl/DPPX complex increases the levels of PrPC/DPPX complex decrease suggesting that the two proteins compete for a single binding site on DPPX. B: N2a cells were co-transfected with DPPX-S and increasing amounts of PrPΔ32-121 plasmid, crosslinked, lysed, and wild-type PrPC complexes analyzed by Western blotting with the D13 antibody (which doesn’t recognize transfected PrPΔ32-121). Very little competition for DPPX complex formation is observed between wild-type PrPC and PrPΔ32-121 suggesting that wild-type PrPC binds more strongly to DPPX, perhaps by virtue of a second binding site N-terminal to residue 121.

164

investigated. For tissue analysis, methacarn-fixed tissue was used exclusively as a signal

was not readily obtained on formalin-fixed tissue with anti-DPPX antibodies.

Immunohistochemistry on a methacarn-fixed wild-type mouse cerebellum using an anti-

DPPX antibody demonstrates strong labeling of the granule cell layer (Figure 4.17A).

Interestingly, the distribution of DPPX in the cerebellar lobules was not uniform (i.e.

DPPX expression is distributed asymmetrically between the different lobules). A recent

study has found that Purkinje cell death in Doppel transgenic mice is asymmetric among

the different cerebellar lobules [400] and it should be noted that there is good agreement

between the pattern of Purkinje cell loss and the pattern of DPPX expression in the

granule cell layer. Granule cell labeling was present with three independent anti-DPPX

antibodies (Figure 4.17B) validating the existence of DPPX in these cells. Consistent

with previous reports [375, 412], Purkinje cell labeling for DPPX was also observed with

all three antibodies, although it was more variable than labeling of the granule cell layer.

Therefore, DPPX is expressed in the two cerebellar cell types (CGNs and Purkinje cells)

that are pertinent to Doppel/ΔPrP-induced toxicity. Not surprisingly, DPPX is also

present on the surface of cultured wild-type CGNs as demonstrated by

immunofluorescence using an anti-DPPX antibody (Figure 4.17C). The presence of

DPPX in CGNs was also confirmed by analyzing cell lysates from either wild-type

(Prnp+/+) or Prnp0/0 cultured neurons. Strong DPPX signal that was sensitive to PNGaseF

digestion was observed with two distinct DPPX antibodies confirming that N-

glycosylated DPPX is expressed at high levels in these cells (Figure 4.17D). The

presence or absence of PrPC had no effect on the levels of DPPX in CGNs. Co-

localization of PrPC and DPPX was observed in the granule cell layer of a wild-type

mouse cerebellum as assessed by double immunofluorescent labeling (Figure 4.17E). In

summary, DPPX is present at high levels in the granule cell layer of the cerebellum and is

expressed in the correct cerebellar cell types for it to control prion protein neurotoxicity

and neuroprotection.

DPPX Forms Complexes with PrPC and Doppel in Cultured CGNs

The existence of complexes between PrPC and DPPX in vivo at endogenous levels

of protein expression is already implied by the identification of DPPX as an interactor of

165

04DX-2 03J2 03K1

PrP DPPX Merge

A

B

C

E

D03K1

16011080

16011080

40

302015

α-DPPX 03J2

α-DPPX 03K1

α-PrP 7A12

- -+ + PNGaseF

Prnp+/+

CGNsPrnp0/0

CGNs

166

Figure 4.17. DPPX is expressed at high levels in the granule cell layer of the cerebellum. A: Immunohistochemical labeling of the cerebellar granule cell layer from a wild-type mouse using the anti-DPPX antibody 03K1. Note the asymmetric distribution of DPPX between the lobules of the cerebellum. B: Immunohistochemical labeling of the granule cell layer from a wild-type mouse with the indicated anti-DPPX antibodies. Purkinje cell staining is also observed but is more variable. C: Immunofluorescent labeling of cultured CGNs from a wild-type mouse with the anti-DPPX antibody 03K1. D: Western blots of lysates prepared from cultured CGNs from either wild-type or Prnp0/0 mice with or without PNGaseF treatment. High levels of DPPX are observed in the lysates. E: Co-localization of PrPC and DPPX within the granule cell layer of the cerebellum as assessed by double immunofluorescent labeling with PrP (7A12) and DPPX (03J2) antibodies. Scale bar = 200 µm (A), 100 µm (B, E), or 10 µm (C).

167

PrPC by tcTPC using wild-type mice [266]. Nonetheless, the presence of PrPC/DPPX

complexes specifically in CGNs was tested using the crosslinking assay in conjunction

with immunoprecipitation. Cultured wild-type (Prnp+/+) CGNs were crosslinked with

formaldehyde and then PrPC-containing complexes isolated by immunoprecipitation.

DPPX was readily detected in a PrPC-containing high molecular weight complex that is

similar in molecular weight to those obtained in transfected N2a cells (Figure 4.18A).

This confirms that PrPC and DPPX interact in CGNs at endogenous levels of protein

expression. To assess whether Doppel forms complexes with DPPX in CGNs, CGNs

prepared from transgenic mice with inducible Doppel expression were analyzed. Because

an antibody which effectively immunoprecipitates Doppel is not available, the standard

Western blot crosslinking assay was used. Following crosslinking, a crosslinked Doppel-

containing band was obtained that is consistent in size with a Doppel/DPPX complex

(Figure 4.18B). Although the identity of this band has not been confirmed in a more

direct way, it seems reasonable to assume that it represents a complex between Doppel

and DPPX due to the high levels of expression of the two proteins in these cells and the

known molecular weight of Doppel/DPPX complexes from studies on transfected N2a

cells.

Functional Analysis of DPPX in CGNs

In order to test whether or not DPPX is involved in modulating Doppel-induced

toxicity in CGNs, the ideal experiment is to remove DPPX expression and observe

whether or not Doppel retains its toxicity. To implement this experiment using the CGN

Doppel toxicity assay [271], RNA interference constructs for the DPP6 gene were

developed. Short hairpin RNA (shRNA) constructs were made using the pSUPER vector-

based RNAi system [432]. Co-transfection of DPP6 shRNA with a DPPX expression

plasmid in N2a cells resulted in efficient suppression of DPPX protein expression (Figure

4.18C). In contrast, a plasmid containing a scrambled DPP6 shRNA sequence failed to

knockdown transfected DPPX protein levels. When these RNA interference constructs

were tested in the CGN assay, it was observed that knockdown of DPPX is toxic to

CGNs (Figure 4.18D). This result has been validated by an independent investigator

(Bettina Drisaldi, unpublished observations). In contrast, no toxicity was obtained when

168

191

97

64

51

39

28

19

Doppel/DPPXComplex

Doppel

Transgene: - + +BA

191

97

IP:

PrPC/DPPXComplex

Prnp+/+ CGNs

α-DPPX

Mock PrP

C D

E

DPPX-S

Actin

*****

n.s.

Figure 4.18. PrPC/DPPX complexes are present in CGNs and DPPX is essential for the in vitro survival of Prnp0/0 CGNs. A: PrPC/DPPX complexes can be isolated from wild-type CGNs. Prnp+/+ CGNs were crosslinked with 2% formaldehyde, lysed, and then PrP complexes immunoprecipitated with 7A12. A Western blot using the 03J2 anti-DPPX antibody demonstrates that DPPX is found in a complex with PrPC. B: CGNs isolated from transgenic mice expressing Doppel were crosslinked with 2% formaldehyde and Doppel complexes analyzed by Western blotting with the E6977 anti-Dpl antibody. A band with the correct size for a Dpl/DPPX complex is observed. C: Development of shRNA directed against DPP6. N2a cells were co-transfected with DPPX-S and the indicated plasmids and then DPPX levels in lysates analyzed by Western blotting with the 03J2 anti-DPPX antibody. The DPPX RNAi plasmid (but not a scrambled sequence) is effective at knocking down DPPX protein levels. D: Toxicity assays in Prnp0/0 CGNs [271] demonstrate that knockdown of DPPX is toxic to the cells (p < 0.01 compared to the pSUPER empty vector) whereas a scrambled RNAi plasmid has no effect. Data is presented as the percentage of transfected cells ± s.e.m. undergoing apoptosis (n = 3). There is no significant difference (n.s.) between the DPPX RNAi plasmid and a toxic Doppel plasmid. E: Treatment of Prnp0/0 CGNs with either DPPX (03K1) or DPP10 (06D10-2) antibodies (10 µg/mL) for 72 hours has no effect on cell viability as assessed by the MTT assay.

169

the scrambled shRNA was transfected. The lack of a significant difference in cell toxicity

between DPP6 shRNA and a toxic Doppel control allele prevents the testing of the

hypothesis that DPPX is required for Doppel toxicity by using a Doppel/DPP6 shRNA

co-transfection. However, this result does demonstrate that perturbation of DPPX

homeostasis in CGNs can lead to apoptosis, a result that is perhaps not surprising since

potassium levels are critical to the in vitro survival of CGNs [433] and DPPX has been

shown to be involved in the proper functioning of neuronal potassium channels [375]. In

an attempt to find other ways of testing the importance of DPPX to Doppel-induced

neurotoxicity, Prnp0/0 CGNs were treated with anti-DPPX antibodies since binding of

antibodies to DPPX may sterically interfere with prion protein binding to DPPX. No

toxicity of DPPX (or DPP10) antibodies at a concentration of 10 µg/mL (as judged by the

MTT assay) was observed (Figure 4.18E). Therefore, treatment of CGNs with anti-DPPX

antibodies following transfection with a toxic Doppel plasmid may be a feasible means of

testing the hypothesis that DPPX is involved in Doppel-mediated CGN toxicity.

Characterization of DPPX-Deficient Mice

The murine DPP6 gene resides on chromosome 5 and the rump white mutant

mouse strain is a product of the Rw allele which has been mapped to chromosome 5 in

mice. Homozygosity for the Rw allele results in an embryonic lethal phenotype and the

Rw allele has been determined to result from an inversion on the distal arm of

chromosome 5 [434, 435]. The distal breakpoint of the inversion (Figure 4.19A) is

known to cause dysregulation of Kit expression resulting in a phenotype characterized by

white colouration of the posterior and ventral portion of the abdomen [436, 437]. The

proximal breakpoint has been mapped to the middle of the DPP6 gene and is predicted to

truncate the DPPX protein after residue 516 (DPPX-S numbering) [427]. However, Rw

mRNA is likely to be unstable as no evidence for the existence of a unique Rw mRNA

species was obtained [427]. This suggests that the Rw allele prevents the production of

DPPX protein. Mutant mice have also been generated which possess a series of

overlapping deletions of regions near the Rw locus on chromosome 5 [425]. One of these

deletions (the df5J deletion) completely encompasses the DPP6 gene (Figure 4.19A).

df5J homozygous mice also die embryonically, likely due to the existence of other genes

170

DPP6 Kit

a b

ab

wt

Rw

df5J

Chromosome 5

A BDPP6df5J/Rw

C D

1048

720480

242α-DPPX α-DPP10

α-DPPX 03J2

α-DPPX 03K1

α-DPPX 04DX-2

α-DPP10

α-PrP

α-Sho

α-actin

Figure 4.19. No change in PrPC or Sho levels in mice genetically deficient for DPPX expression. A: Schematic representation of wild-type, Rw, and df5J DPP6 alleles on chromosome 5 in mice. In the Rump White (Rw) allele, an inversion occurs with the distal breakpoint resulting in dysregulated expression of the Kit gene and the proximal breakpoint resulting in truncation of the DPP6 gene. In the df5J allele, a chromosomal deletion completely spans the DPP6 gene. B: Compound heterozygote mice bearing both the Rw and df5J alleles (DPP6df5J/Rw) have the same phenotype as Rw mice, namely white colouring of the posterior which is due to dysregulation of Kit expression. C: Western blot analysis of protein expression in young (37 days) and old (526 days) DPP6df5J/Rw mice. No DPPX expression is observed in these mice and no differences in expression of DPP10, PrPC, or Shadoo are observed between wild-type and DPP6df5J/Rw mice. D: Analysis of DPPX and DPP10 complexes in wild-type and DPP6df5J/Rw mice by Blue Native PAGE. Although levels of DPP10 are unaltered in brain homogenates from DPP6df5J/Rw mice (C), DPP10 participation in high molecular weight complexes is increased in DPPX deficient mice (arrow).

171

within the df5J deletion interval that are haplosufficient but are essential for viability.

However, Rw/df5J compound heterozygotes are viable and are referred to as DPP6df5J/Rw

mice. These mice have the rump white phenotype (Figure 4.19B) and theoretically do not

produce any full-length DPPX protein. Thus, DPP6df5J/Rw mice may be a useful tool for

studying prion protein biology in a DPPX-deficient system.

As predicted, no DPPX protein was observed in brain homogenates from either

young or aged DPP6df5J/Rw mice as determined by immunoblotting with three distinct

anti-DPPX antibodies (Figure 4.19C). No evidence for a truncated DPPX species

generated by the Rw allele was obtained, even with the 04DX-2 antibody which

recognizes a juxtamembrane epitope in DPPX and theoretically would react with the

protein product of the Rw allele. No changes in either PrPC or Shadoo protein levels were

observed between wild-type and DPP6df5J/Rw mice suggesting that DPPX levels have no

effect on the steady state levels of prion proteins. A diminution in Shadoo signal between

young and aged DPP6df5J/Rw mice was observed, but this phenomenon was also present in

wild-type mice suggesting that Shadoo levels decrease slightly with aging. Interestingly,

no changes in DPP10 protein levels were observed in DPP6df5J/Rw mice suggesting that

the two proteins are not cross-regulated. However, Blue Native PAGE analysis on brain

homogenates from wild-type and DPP6df5J/Rw mice demonstrates that an increase in a

DPP10-containing high molecular weight complex is present in DPP6df5J/Rw mice (Figure

4.19D). This argues that in the absence of DPPX, DPP10 is recruited into protein

complexes that may preferentially contain DPPX in wild-type mice. One candidate

complex is the neuronal A-type potassium channel assembly since both DPPX and

DPP10 bind to members of this complex and are involved in their proper functioning

[375, 419]. Thus, the lack of an overt phenotype in DPP6df5J/Rw mice may result from the

functional compensation of DPPX by DPP10.

Biochemical Properties of Prion Proteins in DPP6df5J/Rw Mice

The endoproteolytic processing of PrPC and Shadoo was compared in wild-type

and DPP6df5J/Rw mice. No differences in the production of the PrP C1 and C2 fragments

or the ShoC1 fragment were observed between wild-type and DPP6df5J/Rw mice (Figure

4.20A). This suggests that DPPX does not contribute to the endoproteolytic processing of

172

A B

40

30

20

15

15

20

10

PrP

Sho

ShoC1

C1

- +-+ PNGaseF

PrP

Sho

S2 P2 S2 P2 S2 P2S2 P2

C

1024

20

66146242

480

720

α-PrP

D

E

110

40

302040

3020

40

3020

DPPX(0 μg/mL PK)

PrP(0 μg/mL PK)

PrP(10 μg/mL PK)

PrP(50 μg/mL PK)

α-Sho

Wild-type

DPP6df5J/Rw

0 0.5 2 5 20 μg/mLPK:

Figure 4.20. Biochemical characterization of prion proteins in wild-type and DPP6df5J/Rw mouse brains. A: Brain homogenates from the indicated mice were analyzed with or without PNGaseF treatment by Western blotting. No change in the endoproteolytic processing of either PrPC or Sho is observed between wild-type and DPPX-deficient mice. B: Brain homogenates from the indicated mice were solubilized in detergent and the supernatant (S2) and pellet (P2) fractions analyzed following ultracentrifugation by Western blotting. No change in either PrPC or Sho solubility is observed between wild-type and DPP6df5J/Rw mice. C: PrPC-containing complexes in brain homogenates from the indicated mice were analyzed by Blue Native PAGE and subsequent Western blotting. PrPC-containing complexes appear to be similar in both wild-type and DPP6df5J/Rw mice. D: Brain homogenates from the indicated mice were treated with various concentrations of PK and then analyzed by Western blotting. There is no evidence for increased amounts of protease-resistant PrP in DPP6df5J/Rw mice. E: The PK resistance profiles of Shadoo are identical in wild-type and DPP6df5J/Rw mice as assessed by Western blotting with the 06rSH-1 antibody.

173

prion proteins. PrPC and Shadoo detergent solubility was investigated in the two strains of

mice using ultracentrifugation. No consistent differences were observed in the amount of

PrPC or Shadoo present in the pellet fractions and the majority of protein is present in the

soluble fraction in both lines of mice (Figure 4.20B). Thus, DPPX is unlikely to regulate

the formation of detergent insoluble structural isoforms of PrP or Shadoo. Next, PrPC-

containing complexes were evaluated using blue native PAGE. No obvious differences

between wild-type and DPP6df5J/Rw mice with respect to PrPC-containing complexes were

observed between the two strains of mice (Figure 4.20C). This implies that DPPX does

not regulate the stoichiometry of PrPC. It should be noted that the inability to see a band

corresponding to a PrPC/DPPX complex by blue native PAGE analysis likely reflects the

instability or transient nature of the complex, a low level of the complex under resting

conditions, or the unsuitability of the detergent used. No consistent differences in the

amount of protease resistant PrP present after digestion with low (10 µg/mL) or high (50

µg/mL) concentrations of proteinase K were observed between wild-type, DPP6df5J/+,

DPP6Rw/+, and DPP6df5J/Rw mice (Figure 4.20). This suggests that the absence of DPPX

does not promote the formation of PrPSc-like protease resistant forms of PrP. Similarly,

no increase in the relative protease resistance of Shadoo was observed between wild-type

and DPP6df5J/Rw mice (Figure 4.20E). Thus, the absence of DPPX does not appreciably

alter the biochemical properties of either PrPC or Shadoo.

Evaluation of the Role of DPPX in Prion Disease in Mice

In order to test whether or not DPPX has any relevance to prion disease or prion

replication, DPP6df5J/+ mice were inoculated with the RML strain of murine-adapted

scrapie prions. DPP6df5J/+ mice have lower levels of DPPX protein due to the presence of

the df5J deletion allele (Figure 4.21A). No statistically significant differences in survival

were observed between wild-type and DPP6df5J/+ mice following inoculation (Figure

4.21B) with the median survival times being 165 and 170 days, respectively. Additionally,

no obvious differences in either the level or biochemical profile of PrPres between the two

lines of mice were observed (Figure 4.21C). Thus, a small decrease in DPPX protein

levels does not appear to affect prion disease or prion replication. However, the decrease

in DPPX protein level in DPP6df5J/+ mice is minimal so the potential involvement of

174

A

B C

148

98

64

5050

36

2216

α-DPPX

α-PrP

493828

1714

Figure 4.21. No change in prion disease incubation time in mice hemizygous for the DPP6 df5J deletion allele. A: Western blot of uninoculated wild-type (DPP6+/+) or df5J hemizygous (DPP6df5J/+) mouse brain homogenates. A slight decrease in DPPX expression is apparent in the DPP6df5J/+ mice. No change in PrPC expression is observed between the two mouse lines. B: Kaplan-Meier survival curve of mice inoculated with the RML strain of prions. The median survival was 165 days for wild-type mice and 170 days for DPP6df5J/+ mice although there was no significant difference between the survival curves (P > 0.05) by the Log-rank Test. C: Western blot of PrPres from clinically ill RML prion-inoculated mice. There is no apparent difference between the PrPres profiles of wild-type and DPP6df5J/+ mice.

175

DPPX in prion disease cannot be fully evaluated using these mice. Unfortunately,

breeding problems prevented the inoculation of DPP6df5J/Rw mice (which completely lack

DPPX expression) with prions.

No changes in levels of full-length DPPX protein were observed between non-

inoculated and RML prion-inoculated wild-type mice (Figure 4.22). Similarly, the levels

of full-length DPP10 were unaltered. With regard to the endoproteolytic processing of

DPPX, no obvious differences between infected and non-infected mice were obtained in

blots using the 03J2 and 03K1 antibodies. Small changes in the banding pattern between

infected and non-infected mice were observed with the 04DX-2 antibody. However, these

bands have not been further characterized or validated. A unique band reactive to a

DPP10 antibody was observed at around 65 kDa in prion-inoculated mice (Figure 4.22).

Interestingly, this band corresponds in size to one of the endoproteolytic products of

DPPX. As demonstrated above, DPP10 is not normally subject to endoproteolytic

processing in a wild-type mouse brain. Whether DPP10 endoproteolysis is specific to

prion-diseased brains remains to be determined and further experimental validation and

characterization of the observed band is necessary.

Evaluation of the Role of DPPX in Prion Replication in Tissue Culture Cells

Numerous strategies were initially employed in an attempt to analyze the effect(s)

of over-expression of DPPX on prion replication (as monitored by PrPres levels) in prion-

infected ScN2a and SMB cells. First, individual clones of ScN2a cells stably-transfected

with DPPX isoforms were isolated. However, the results were highly variable and

inconsistent, likely reflecting the inherent differences in susceptibility to prion replication

between individual N2a/ScN2a subclones [165, 171]. In an attempt to circumvent this

problem, bulk-selected stably-transfected ScN2a cells expressing DPPX isoforms were

generated. However, the variation between ScN2a subclones was not completely

removed using this strategy as inconsistent results upon repetition were obtained

rendering the experiments uninterpretable. Similarly, the de novo infection of bulk-

selected stably-transfected DPPX-expressing N2a cells with RML prions failed to yield

repeatable results (data not shown). Thus, clonal variability in N2a and ScN2a cells with

respect to prion susceptibility precludes the utilization of stably-transfected clones or

176

16011080

6050

16011080

6050

16011080

6050

160110806050

α-DPPX 03J2

α-DPPX 03K1

α-DPPX 04DX-2

α-DPP10 06D10-2

Non-inoculated

RML-inoculated

Figure 4.22. Analysis of DPPX and DPP10 expression in non-inoculated and RML prion-inoculated mouse brains. Western blots of homogenates prepared from the brains of non-inoculated or clinically ill (average of 172 days post-inoculation) RML prion-inoculated mice (C3H/C57BL6 background) probed with the indicated antibodies. There is no change in the levels of either DPPX or DPP10 in the brains of prion-infected mice. Changes in proteolytic product bands for DPP10 may reflect an alteration in DPP10 metabolism in prion-infected brains.

177

populations of cells. Similar inconsistent results were obtained with stably-transfected

SMB cells. Consequently, experiments with prion-infected tissue culture cells were

limited to transient transfections.

Transient over-expression of DPPX or DPP10 isoforms in ScN2a cells had no

overt effect on PrPres levels despite robust expression of DPPX and DPP10 in the

transfected cells (Figure 4.23A). Next, knockdown of DPP6 in ScN2a was performed

using pooled siRNA consisting of four individual RNAi sequences. Effective knockdown

of endogenous DPPX in ScN2a cells was achieved 72 hours post-transfection (Figure

4.23B). A slight decrease in PrPres levels was obtained in DPP6 siRNA-transfected cells

compared to Mock siRNA- and DPP10 siRNA-transfected cells. However, the observed

difference was small and further confirmation of this result is necessary. Knockdown of

DPPX or DPP10 did not have a noticeable effect on total PrP levels in ScN2a cells. As a

control, Prnp siRNA was effective at knocking down PrP expression and lowering PrPres

levels. Therefore, the role of DPPX in prion replication remains ambiguous and needs to

be further assessed.

4.5 Discussion The work presented in this chapter represents the identification and

characterization of the first protein which is capable of binding to all three members of

the prion protein family. Moreover, DPPX is the first identified binding partner for

Shadoo and only the second identified binding partner for Doppel [344]. Unlike many

other identified candidate PrPC-interacting proteins, DPPX interacts with PrPC in vivo and

at the cell surface where numerous important events in prion protein biology are thought

to take place. Thus, DPPX is a plausible candidate for modulating the physiological roles

of PrPC and Shadoo in healthy brains as well as prion disease pathobiology.

DPPX was initially identified in an in vivo screen for PrPC-interacting proteins

using tcTPC [266]. The validity of this technique for uncovering meaningful protein-

protein interactions has been previously demonstrated by the identification of TMP21 as

a presenilin 1-interacting protein and LINGO-1 as an APP-interacting protein [267, 342].

Furthermore, tcTPC has been successfully used to isolate the other members of the γ-

secretase complex using Aph-1 as the bait [266]. Therefore, tcTPC is a useful method for

178

A

DPP10

DPPX

PrPTotal

PrPres

B

DPPX

PrPTotal

PrPres

Figure 4.23. Over-expression and knockdown of DPPX and DPP10 in prion-infected N2a cells. A: DPPX and DPP10 isoforms were transiently transfected as indicated into ScN2a cells and incubated for 72 hours, followed by analysis of PrPres levels in cell lysates by Western blotting. Neither expression of DPPX or DPP10 had any discernible effect on PrPres levels. B: ScN2a cells were transfected with the indicated siRNAs and lysates collected 72 hours later. Whereas knockdown of Prnp resulted in a large decrease in PrPres levels, knockdown of DPP6 did not have any large effect on PrPres levels. DPPX was detected with either 03K1 (A) or 03J2 (B), DPP10 with 06D10-2, and PrP with 7A12.

179

uncovering novel protein-protein interactions with physiological relevance since

crosslinking is performed on intact wild-type mouse brains with endogenous levels of

protein expression. The identification of DPPX as a hypothetical PrPC-interacting protein

was unequivocal because it was based on 13 unique DPPX peptides [266]. Notably,

DPP10, a protein which shares approximately 50% sequence identity with DPPX [417]

(Figure 4.1), was not found in this screen implying that the interaction between PrPC and

DPPX is specific. It will be of interest to perform tcTPC analysis using Shadoo as the bait

protein to see if DPPX is authenticated as an in vivo Sho-interacting protein.

Numerous lines of evidence suggest that DPPX is a legitimate in vivo interacting

partner of PrPC, Doppel, and Shadoo. Firstly, crosslinking of transfected cells results in

the formation of high molecular weight complexes between the members of the prion

protein family and DPPX (Figure 4.8). This analysis is performed on intact live cells

which removes the possibility that complex formation occurs spuriously upon the release

of proteins from the membrane by cell lysis. Secondly, complexes between DPPX and

PrPC were observed in primary cultures of CGNs isolated from wild-type mice with

endogenous levels of protein expression (Figure 4.18). Thus, protein over-expression is

not sufficient to explain the interaction between PrPC and DPPX. Consistent with this

notion, over-expression of DPP10 did not lead to the formation of complexes. Thirdly,

reciprocal co-immunoprecipitation of DPPX and PrPC from wild-type mouse brains was

achieved in the absence of a crosslinking agent (Figure 4.9B). This result confirms that

complex formation is not an artifact of formaldehyde crosslinking, that PrPC and DPPX

form a complex containing physical contacts, and that complexes are not trivially

obtained due to the proximity of PrPC and DPPX in the plasma membrane. Finally,

complex formation does not simply result from the over-expression of proteins in the

secretory pathway as complexes between DPPX and members of the prion protein are

located on the cell surface (Figure 4.10A) and contain mature glycosylated proteins

(Figure 4.9). Therefore, the interaction of DPPX with members of the prion protein

family occurs in vivo at endogenous levels of protein expression and cannot be

rationalized as an experimental artifact owing to protein over-expression or spurious

interactions.

180

The precise stoichiometry of prion protein/DPPX complexes remains to be

determined. It is tempting to speculate that due to the existence of DPPX as a dimer

(Figure 4.7) the complex contains a homodimer of DPPX and two prion protein

molecules. Consistent with this notion, PrPC/DPPX complexes are larger than DPPX

dimers as determined by Western blotting suggesting that the complex contains a DPPX

dimer and one or two copies of PrPC. However, a precise estimation of the molecular

weight of the complex by SDS-PAGE is impossible due to the non-linearity of the

separation method and the presence of crosslink-induced residual structure. Resolution of

this issue may be possible by using PrP alleles with different epitope tags. For instance,

pull-down of 3F4-tagged PrP-containing complexes and immunoblotting for FLAG-

tagged PrP complexes (or vice versa) may demonstrate the co-existence of 3F4- and

FLAG-tagged PrP in a single complex with DPPX. This would imply that the DPPX

dimer simultaneously binds at least two copies of PrPC. Alternatively it may be possible

to purify PrPC/DPPX complexes from cells or reconstitute them in vitro using

recombinant proteins and then use biochemical methods such as analytical

ultracentrifugation or gel filtration chromatography to determine their precise molecular

weight.

Although the in vivo function of PrPC and Shadoo remains enigmatic [241, 372,

374], various candidate physiological roles for PrPC have been proposed. Is it possible

that DPPX is involved in the cellular function(s) of PrPC and/or Shadoo? DPPX is

expressed throughout the brain (Figure 4.5) and its expression pattern coincides well with

those of PrPC and Shadoo (see Figures 3.6-3.8). Clearly, any interaction between PrPC

and DPPX is not essential under resting conditions as Prnp0/0 mice have only subtle or

disputed phenotypes [186, 187, 248]. In support of this notion, complexes between PrPC

and DPPX are likely weak or transient in nature (as demonstrated by the requirement of a

crosslinking agent to ‘lock’ the complex together) and are unlikely to exist constitutively.

The functional relevance of Sho/DPPX complexes may become apparent upon the study

of potential phenotypic defects in Sprn0/0 mice. PrPC/DPPX or Sho/DPPX complexes may

exhibit a specialized function that is only pertinent to specific situations. For instance,

PrPC/DPPX complexes may play an important role in the brain under stressed conditions.

Paradoxically, PrPC has been shown to exhibit both pro- and anti-apoptotic properties in

181

various experimental systems [268, 269, 271, 272, 276-278, 356] and Shadoo is PrPC-like

in its ability to protect against toxic stimuli (see Figure 3.11). Because DPPX is a ligand

for both PrPC and Shadoo, it is a strong candidate for controlling the neuroprotective

properties of these proteins. The neuroprotective and neurotoxic properties of prion

proteins presumably require access to intracellular apoptosis pathways. However, neither

PrPC nor Shadoo have intracellular domains. DPPX contains a cytoplasmic domain which

contains multiple consensus sequences for either Protein kinase C or Casein kinase II

phosphorylation [413] which may provide a link between prion proteins and intracellular

signal transduction pathways. DPPX may also be relevant to the observed signal

transduction through PrPC involving Fyn kinase [331]. Although Doppel/DPPX

complexes may be relevant to neurodegeneration in mice ectopically or transgenically

expressing Doppel in the brain, their relevance to the physiology of wild-type mice

remains to be investigated. Doppel is not normally expressed in the adult brain (the

principal site of DPPX expression [413, 427]) and is instead present primarily in the

testes [190, 229] where it is required for the proper functioning of the male reproductive

system [284, 285]. However, a molecular description of the role of Doppel in the testes

has not emerged. DPP6 mRNA has been found in the testes, albeit at levels below those

present in the brain [376, 427]. It is therefore possible that the binding of Doppel to

DPPX (or a related protein) in the testes may be involved in reproductive biology.

Consistent with this notion, problems were experienced in the breeding of DPP6df5J/Rw

mice which lack DPPX expression (Figure 4.19) suggesting that the lack of DPPX may

hinder reproductive physiology.

Although DPPX is a member of the dipeptidyl aminopeptidase family of serine

proteases it is unlikely to possess any proteolytic activity since it lacks the catalytic serine

residue required for proteolysis [377, 414]. However, this cannot be the sole reason for its

lack of proteolytic activity since restoration of the catalytic serine residue by mutagenesis

failed to generate any catalytic activity when measured using traditional dipeptidyl

peptidase substrates [411]. Therefore, it seems likely that DPPX has evolved a unique

function that is independent of proteolysis. This phenomenon has also been documented

in other proteins. For example, the DJ-1 protein which has been implicated in Parkinson’s

disease [438] is structurally similar to a bacterial cysteine protease but does not exhibit

182

any protease activity [439]. Interestingly, DPP6 has been identified as a gene that has

undergone accelerated evolution in the primate lineage suggesting that it may have

evolved a new function specific to more complex nervous systems [440]. Like DPP4, the

ectodomain of DPPX contains a large 8-bladed β-propeller domain [414]. β-propeller

domains are known to serve as scaffolds for protein-protein interactions [441]. Thus, the

function of DPPX may be exerted through the binding of various proteins (such as PrPC)

to the β-propeller domain. Production of DPP6 mRNA in the hippocampus is known to

be stimulated by kainic acid-induced synaptic activity [412] suggesting that DPPX may

play a role in synaptic transmission. Interestingly, one group has reported synaptic

defects in Prnp0/0 mice leading to the suggestion that PrPC is necessary for normal

synaptic function [242, 243] whereas others have failed to confirm this finding [244].

Further studies may uncover whether or not PrPC/DPPX complexes are involved in

synaptic transmission.

The most well-documented role for DPPX surrounds its ability to modulate the

properties of neuronal A-type potassium channels [375]. In this paradigm, DPPX is

thought to function by assisting the transport of Kv4.2 (the membrane-spanning channel

subunit) to the cell surface. Like DPPX, DPP10 also modulates A-type potassium

channels by altering the trafficking of Kv4.2 [422]. All identified N-terminal splice

variants of DPPX exert similar effects on the electrophysiological properties of potassium

channels and in agreement with this result, the cytoplasmic domain of DPPX is not

necessary for modulating channel properties [413]. Furthermore, a DPPX construct with

just the cytoplasmic and transmembrane domains is wild-type DPPX-like in its ability to

bind Kv4.2 and to modulate potassium channels [422]. Thus, the transmembrane domain

of DPPX seems to be important to its role in potassium channel biology. Similarly, the N-

terminus and transmembrane domain of DPP10 have been shown to be sufficient for

binding to Kv4.3 and channel modulation [420]. The modulation of neuronal potassium

channels by DPPX is clearly not essential for viability since DPP6df5J/Rw mice survive to

old age. The electrophysiological properties of these mice have not been investigated but

may reveal subtle alterations to potassium channels. It is also possible that the lack of an

overt phenotype in these mice is due to the recruitment of the functionally-related DPP10

protein into potassium channel complexes normally occupied by DPPX (Figure 4.19D).

183

Are PrPC or Shadoo involved in potassium channel biology via their interaction

with DPPX? PrPC binds to the β-propeller domain of DPPX whereas Shadoo likely binds

to an extracellular juxtamembrane epitope (Figure 4.15). These regions of DPPX are non-

essential for the modulation of neuronal A-type potassium channels. Nonetheless, it

remains possible that PrPC and/or Shadoo may contribute to the proper functioning of

potassium channels. Two studies have noted that calcium-activated potassium currents

are disrupted in Prnp0/0 mice [246, 442]. Additionally, PrPC has been shown to bind to

TREK-1, a two-pore potassium channel protein [306], although this interaction was

uncovered using over-expressed proteins in the bacterial cytoplasm casting doubt on its

relevance. More work is needed to decipher the potential link DPPX provides between

prion proteins and potassium channels.

One well-characterized aspect of prion protein biology in which an involvement

of DPPX seems especially plausible is the phenotypic interactions between Doppel/ΔPrP

and PrPC that occur in certain strains of transgenic mice. In the absence of PrPC,

expression of Doppel or ΔPrP in the CNS results in an ataxic neurodegenerative

phenotype characterized by the loss of cerebellar cells (Purkinje neurons and/or CGNs)

[188, 190, 217, 289]. This has led to a model in which Doppel/ΔPrP-mediated

neurotoxicity and PrPC-mediated neuroprotection occurs via a hypothetical prion protein

ligand termed LPrP [217, 279, 280, 372]. In the most recent version of the model, PrPC

binding to an LPrP dimer results in the production of a positive trophic signal that confers

a neuroprotective phenotype [279]. The lack of a phenotype in Prnp0/0 mice is explained

by the existence of a residual amount of constitutive autoactivation of LPrP in the absence

of PrPC, although functional compensation by a PrPC-like molecule or the non-essential

nature of signals from PrPC/LPrP complexes are also plausible explanations. The toxicity

of Doppel/ΔPrP is postulated to result from the binding of either protein to LPrP. However,

since Doppel and ΔPrP lack a second neuroprotection-associated LPrP binding

determinant (that is present in the N-terminus of PrPC), a neurotoxic signal is transduced

instead. Alternatively, if LPrP can undergo autoactivation to produce a baseline level of

neuroprotective signal, the binding of Doppel/ΔPrP may be dominant negative in nature

and completely block the generation of any signal resulting in cell death.

184

DPPX is a strong candidate for LPrP for a number of reasons. Firstly, DPPX is

present at high levels in the granule cell layer of the cerebellum, in Purkinje cells, and in

cultured Prnp0/0 CGNs (Figure 4.17) and these are the cell types that are permissive to

Doppel/ΔPrP-mediated toxicity [217, 271, 289, 351, 353]. Thus, DPPX is in the correct

location for mediating phenotypic interactions between Doppel/ΔPrP and PrPC and

indeed co-localizes and forms complexes with PrPC in CGNs (Figure 4.17D, Figure

4.18A). Secondly, the asymmetric distribution of DPPX between the individual lobules of

the cerebellum (Figure 4.17A) corresponds well with the asymmetric cell death of

Purkinje neurons between the different cerebellar lobules in Doppel transgenic mice

[400]. Thirdly, DPPX exists as a dimer in vivo (Figure 4.7), as is the case for the

hypothetical LPrP molecule [279]. Fourthly, DPPX is capable of forming cell surface

complexes with all members of the prion protein family (Figure 4.8): PrPC and Shadoo

which are associated with neuroprotection as well as Doppel and ΔPrP which are

associated with neurotoxicity. When coupled with the idea that Shadoo and the

hypothetical PrPC-like and LPrP-binding protein π are synonymous (see Chapter 3), DPPX

fulfills all the prion protein binding characteristics of LPrP. Fifthly, as postulated in the

LPrP model, complex formation between prion proteins and DPPX are governed by

multiple interaction sites. A common binding site exists in the α-helical domain found in

PrPC and in Doppel which binds to the β-propeller domain of DPPX and an N-terminal

binding site present in Shadoo and likely in PrPC which binds to a juxtamembrane epitope

in DPPX (Figures 4.11-4.13, 4.15). In the model, the α-helical binding site would be

associated with the transduction of a neurotoxic signal (as is the case with Doppel and

ΔPrP) unless the N-terminal binding site which is associated with the transduction of a

neuroprotective signal is simultaneously present (as in PrPC). Shadoo also possesses the

neuroprotective binding determinant and is therefore functionally similar to PrPC. Sixthly,

the mapped DPPX binding sites in Doppel and PrPC correspond well to experimentally-

deciphered neurotoxicity and neuroprotection determinants in these proteins, respectively

[271, 381]. Seventhly, Doppel and PrPC compete for complex formation with DPPX

(Figure 4.16) providing a mechanism by which PrPC expression may abrogate Doppel-

mediated toxicity. Finally, knockdown of DPPX protein levels in Prnp0/0 CGNs results in

cell death (Figure 4.18) suggesting that perturbations in DPPX homeostasis can lead to

185

apoptosis. Thus, an involvement of DPPX in Doppel/ΔPrP-mediated apoptosis of CGNs

is plausible.

While various lines of experimental evidence suggest that DPPX is a strong

candidate for LPrP, the functional involvement of DPPX in Doppel/ΔPrP-mediated

neurotoxicity and PrPC-mediated neuroprotection has not been confirmed. Unfortunately,

due to the inherent toxicity of knocking down the DPP6 gene in cultured Prnp0/0 CGNs, a

Doppel/DPP6 siRNA co-transfection would not be informative with respect to the

potential requirement of DPPX for Doppel toxicity. A possible solution to this problem is

the construction of a dominant negative DPPX molecule that would bind to Doppel but

be incapable of transducing the neurotoxic signal. A DPPX allele lacking its cytoplasmic

domain (Figure 4.14) may be useful for this purpose. Alternatively, peptides

corresponding to the Doppel binding site on DPPX may be useful for preventing the

interaction between transfected Doppel and endogenous DPPX in CGNs permitting a

functional assessment of the interaction. However, the most convincing experiment

would be to generate a DPP6 knockout mouse on a Prnp0/0 background and then cross

with transgenic mice expressing either PrPΔ32-134 [217] or Doppel [289, 352]. Analysis

of clinical disease and neurodegeneration in these mice would provide an accurate

characterization of the role of DPPX in phenotypic interactions between prion proteins

and may conclusively identify DPPX as LPrP.

If DPPX and LPrP are determined to be synonymous, it will be of interest to

investigate the mechanism by which DPPX mediates prion protein neurotoxicity and

neuroprotection. DPPX is involved in the maturation of neuronal A-type potassium

channels [375] and potassium is a critical regulator of the survival of CGNs in vitro. For

instance, lowering the concentration of potassium from 25 mM to 5 mM causes apoptosis

of cultured CGNs [433, 443]. Furthermore, neuronal apoptosis can be induced by

potassium efflux [444, 445]. Thus, it is possible that Doppel toxicity involves the

perturbation of potassium channels. The downstream transducers of Doppel/ΔPrP toxicity

are not clear although an involvement of activated caspase-3 seems probable [271, 280].

A direct link between DPPX and caspase-3 has not been uncovered but the presence of

cytoplasmic DPPX sequence provides potential access to cytoplasmic signaling pathways.

To this end, the various N-terminal splice variants of DPPX may be involved in the

186

differential regulation of cytoplasmic signaling pathways as they are known not to affect

the DPPX-mediated regulation of potassium channels [413]. Consistently, the

intracellular domain of DPPX-S contains consensus sequences for phosphorylation by

Protein kinase C and Casein kinase II whereas DPPX-E lacks the proper motifs for

phosphorylation by these kinases [413]. Interestingly, Protein kinase C has been linked to

both pro- and anti-apoptotic signaling [446]. It will be important to determine at the

molecular level how the binding of Doppel and PrPC to DPPX can modulate its biological

properties.

Because Doppel and PrPC compete for complex formation with DPPX (Figure

4.16), it seems likely that both proteins bind to the same site on DPPX, specifically the β-

propeller domain (Figure 4.15). The individual blades of the 8-bladed β-propeller domain

form a funnel in which the narrow end directs access to the catalytic site [414, 429]. The

binding of proteins to β-propeller domains typically occurs via interactions with the loops

between individual β-strands. Therefore, these loops are likely to constitute the binding

sites for the prion protein α-helical domain and probably form direct contacts with the

helix B region of PrPC or the helix B/B’ region of Doppel. This interaction may serve as

an anchor so that the flexible N-terminus of PrPC is free to bind to another epitope on

DPPX. In support of this idea, Shadoo, which loosely resembles the N-terminus of PrPC,

binds to a juxtamembrane region on DPPX. The binding of PrPC or Shadoo to this second

site may generate the production of a neuroprotective signal. In contrast, the binding of

ΔPrP or Doppel to DPPX may result in the production of a toxicity signal due to the

absence of the second interaction site. The exact nature of the second binding site

remains to be determined. The N-terminus of Shadoo serves as the binding site for DPPX

and contains a series of basically-charged tetrarepeats. Although PrPC also contains N-

terminal repeats in the form of octarepeats, they are not similar to Shadoo at the amino

acid level. In contrast to Shadoo, the N-terminus of PrPC is clearly not required for

complex formation with DPPX (Figure 4.11). However, it remains possible that it is

involved in some aspect of complex formation. Similarly, the hydrophobic tract of PrP

appears to be involved in complex formation with DPPX (Figure 4.11) whereas it has no

effect on the ability of Shadoo to bind to DPPX (Figure 4.13). Numerous lines of

evidence point to the importance of the hydrophobic tract of PrPC in determining the

187

balance between neuroprotection and neurotoxicity [279, 280, 381]. Therefore, it seems

reasonable to assume that if DPPX is LPrP, the hydrophobic tract plays an important role

either in governing binding of prion proteins to DPPX or as an effector domain.

Interestingly, one report has proposed that the neurotoxicity of the PrP106-126 peptide

[236] can also be rationalized by the LPrP model [280]. In mice expressing wild-type PrPC,

PrP106-126 would compete with the hydrophobic tract in PrPC for the second site

binding event. This may mimic the PrPΔ105-125 allele which is highly toxic to Prnp0/0

neurons. Notably, this region of PrPC also contributes to complex formation with DPPX.

In the LPrP model, the parsimonious explanation for the ability of PrPC to counteract

Doppel/ΔPrP toxicity is direct competition between the two proteins for LPrP binding. The

fact that Doppel and PrPC can compete for complex formation with DPPX (Figure 4.16)

is consistent with this notion. If it is found that DPPX can simultaneously bind multiple

copies of prion proteins then the binding stoichiometry may also need to be considered.

The role, if any, that DPPX plays in prion disease and prion replication remains to

be determined. Preliminary evidence using cell culture models suggests that the over-

expression of DPPX isoforms has no effect on prion replication in ScN2a cells (Figure

4.23). Furthermore, prion disease incubation time is unaltered in mice that transgenically

over-express DPPX (George Carlson, personal communication). However, it is not

known to what degree these mice over-express DPPX. Hemizygous mice with a single

copy of either the df5J (Figure 4.21) or Rw allele (George Carlson, personal

communication) have prion disease incubation times indistinguishable from wild-type

mice. However, this may not be surprising since these mice have only very moderate

decreases in DPPX protein levels (Figure 4.20D, Figure 4.21A). A comprehensive

assessment of the role of DPPX in prion disease and prion replication requires the

creation of DPP60/0 mice as DPP6df5J/Rw mice are not ideal for analysis since multiple

genes are removed in the df5J deletion interval [425]. Challenge of DPP60/0 mice with

prions and the subsequent monitoring of disease incubation time, PrPres level and

biochemical profile, and neuropathology will help to unmask the role of DPPX in prion

disease.

Theoretically, DPPX seems like an ideal candidate for involvement in either prion

replication or prion neurotoxicity. DPPX and PrPC expression profiles within the brain

188

are in good agreement (Figure 4.5) suggesting that DPPX is ideally positioned to affect

some aspect of prion disease. Furthermore, the dimeric stoichiometry of DPPX (Figure

4.7) implies that DPPX may be capable of binding multiple copies of the prion protein.

Thus, DPPX could hypothetically serve as a surface for the template-directed refolding of

prion proteins by bringing together PrPC and PrPSc. In this scenario, the β-structure-rich

environment of the β-propeller domain in DPPX to which PrPC (and potentially PrPSc)

binds may aid the conformational transition from PrPC to PrPSc. This would make DPPX

a strong candidate for Protein X, a hypothetical protein which is involved in the

formation of PrPSc [110, 111]. However, a potential role for DPPX in prion toxicity

should also be considered. The absence of clinical prion disease and prion

neuropathology in mice expressing PrP lacking its GPI anchor following prion challenge

[86] implies the existence of a cell surface transmembrane protein that is involved in

potentiating PrPSc-mediated neurodegeneration [87]. DPPX may constitute such a

molecule due to its potential link to toxicity emanating from Doppel/ΔPrP expression.

Because the hydrophobic tract of PrPC is conformationally altered in PrPSc [223], it may

exhibit altered binding to DPPX at the second site. Within the framework of the LPrP

hypothesis, this may cause structural mimicry of ΔPrP and result in the generation of a

toxic signal from DPPX.

Finally, an investigation of DPPX within the context of other neurodegenerative

diseases is also warranted. Recently, two published reports have linked an intronic

variation in the DPP6 gene to increased susceptibility to sporadic ALS [415, 416].

Therefore, it is possible that DPPX plays a common role in various neurodegenerative

diseases that is likely related to neuronal toxicity. The creation of authentic DPP60/0 mice

and crossing them with mouse models of other neurodegenerative diseases such as the

TgCRND8 model of Alzheimer’s disease [382] or mice expressing SOD1 containing

ALS-linked mutations [447-449] may be informative for this purpose.

189

Chapter 5

Conclusions and Future Directions

190

Identification of Shadoo as the Third Member of the Prion Protein Family

Prior to the commencement of the work presented herein, the prion protein family

consisted of two members: PrPC and Doppel. A putative third member, the hypothetical

Shadoo protein encoded by the Sprn gene, was identified in 2003 [197] but no evidence

to support the existence of the predicted protein product in vivo had been provided. The

work presented in Chapter 3 represents the first biochemical and functional

characterization of the hypothetical Shadoo protein product from mice. Various

antibodies raised against the predicted Shadoo protein sequence recognize a PNGaseF-

sensitive band of the correct size on a Western blot of brain homogenate prepared from a

wild-type mouse. This result provides the first evidence for the existence of authentic

Shadoo protein in the brain and confirms that transcripts from the Sprn gene are capable

of encoding a stable protein in vivo. The identification of Shadoo as a neuronal protein

was reinforced by the demonstration that signal for both Sprn mRNA and Shadoo protein

can be detected in neurons of the mouse brain. Thus, Shadoo is an active component of

the adult mouse CNS and Sprn is not simply a pseudogene lying dormant in the genome.

Shadoo exhibits many biochemical properties that are reminiscent of PrPC. Firstly,

Shadoo is post-translationally modified by the addition of N-glycans and is anchored to

the cell surface via the addition of a GPI anchor at its C-terminus. Secondly, Shadoo

undergoes physiological endoproteolytic processing in the vicinity of its hydrophobic

tract to release an N-terminal peptide from the cell surface. This is strikingly similar to

the C1 cleavage of PrPC. Thirdly, the principal region of amino acid sequence

conservation between PrPC and Shadoo is found within the alanine/glycine-rich

hydrophobic tract shared between the two proteins. Numerous lines of evidence suggest

that this region corresponds to a site of bioactivity in PrPC [279, 280, 361, 381]

suggesting that Shadoo may share functional properties with PrPC. Indeed, Shadoo is

PrPC-like in its ability to abrogate the toxicity associated with the expression of either

Doppel or ΔPrP in CGNs cultured from Prnp0/0 mice and the hydrophobic tract appears to

be important for conferring neuroprotective activity in both proteins. The remarkable

similarities between Shadoo and PrPC suggest that the labeling of Shadoo as the third

member of the prion protein family is warranted.

191

Probing the Cellular Function of Shadoo and Potential Similarities to PrPC

The similar behaviour of Shadoo and PrPC in an ex vivo functional assay provides

the motivation for testing the potential functional redundancy of the two proteins in vivo

in the mouse brain. The lack of spectacular phenotypes in Prnp0/0 mice [186, 187] has

hindered the elucidation of the function of PrPC. One possible explanation is that a PrPC-

like protein exists in the brain which functionally compensates for the absence of PrPC

[217]. Shadoo is a strong candidate for such a protein due to its PrPC-like neuroprotective

properties. Furthermore, Shadoo is present in Purkinje cells and the apical dendrites of

hippocampal CA1 neurons which are two of the few areas in the brain in which PrPC is

notably absent. Thus, Shadoo may have evolved to fill-in a PrPC-like function to

neuroanatomic regions of the brain that are naturally-deficient in PrPC expression. To test

the hypothesis that Shadoo compensates for the loss of PrPC in Prnp0/0 mice, mice lacking

the Sprn gene (Sprn0/0 mice) need to be created. Because the gene structures of Prnp and

Sprn are comparable, a strategy similar to those used to generate Prnp0/0 mice could be

employed. For instance, the entire Shadoo open-reading frame present in exon 2 of Sprn

could be replaced with a neomycin resistance cassette. This would ensure that no residual

truncated Shadoo protein sequence is produced. It will be interesting to assess the

phenotypic deficit(s), if any, in Sprn0/0 mice. If PrPC and Shadoo are truly functionally

redundant, then it can be predicted that, like Prnp0/0 mice, Sprn0/0 mice will have no

phenotype due to the functional compensation of Shadoo by PrPC. If this is the case,

analysis of Prnp0/0/Sprn0/0 double knockout mice may prove more informative as the

genetic ablation of both genes may completely remove an essential function, presumably

related to neuroprotection, resulting in the potential unmasking of the cryptic function of

PrPC. It seems plausible that some of the subtle phenotypes in Prnp0/0 mice (such as

deficits in neuronal differentiation and proliferation [263]) may be accentuated upon the

removal of Sprn.

Because of the functional similarities between Shadoo and PrPC, it will be of

interest to probe the Shadoo interactome in the mouse brain. For this purpose, tcTPC

[266] seems ideally suited due to its simplicity and its capacity for uncovering

physiologically important protein-protein interactions [267]. Two of the polyclonal anti-

Shadoo antibodies used in this study (06rSH-1 and 06SH-3a) would be appropriate for

192

this purpose due to their specificity and their ability to be used for immunoprecipitations.

A comparison between the Shadoo interactome and the published PrPC interactome [266]

may help to decipher overlapping and divergent functions of the two proteins in the brain

via the analysis of shared and unique binding partners. Since DPPX was identified as a

common binding partner for PrPC and Shadoo in this study, it would not be surprising to

obtain this protein in the list of candidate in vivo Shadoo-interacting proteins. It will also

be of interest to characterize the binding epitopes of the identified proteins on Shadoo.

The N-terminus of Shadoo clearly plays a role in complex formation with DPPX and

appears to govern binding to multiple other proteins in N2a cells. Thus, this domain may

serve as a critical protein-protein interaction domain and may be important to the

function of Shadoo. A comparative proteomic analysis of the interactomes of PrPC,

Doppel, and Shadoo is also currently underway using N2a cells stably-transfected with

the individual FLAG-tagged prion proteins. The use of an epitope tag in this case allows

the same isolation technique to be used for all three proteins and therefore permits a

meaningful direct comparison between the three interactomes. It is hoped that this

experiment will provide insight into the divergent and overlapping functions of PrPC,

Doppel, and Shadoo through the identification of unique and shared binding partners,

respectively.

The high sensitivity of Shadoo to proteinase K digestion and its short half-life in

N2a cells suggests that Shadoo is highly flexible and natively unstructured. Consistent

with this notion, the circular dichroism spectrum of recombinant Shadoo lacks well-

defined peaks implying a paucity of secondary structural elements [381]. Interestingly,

other proteins with relevance to neurodegenerative diseases such as α-synuclein and tau

are also intrinsically disordered proteins [450, 451]. It seems reasonable to assume that

the intrinsic flexibility of the Shadoo protein has important implications for its function.

Consequently, the structural characterization of Shadoo would aid the understanding of

its biological roles. For this purpose, traditional x-ray crystallography or NMR

spectroscopy analysis is likely to be uninformative due to the inherent flexibility of

Shadoo. However, NMR techniques such as relaxation dispersion analysis may prove

useful for probing specific aspects of Shadoo structure. Additionally, a more defined

Shadoo structure may be induced upon binding to an interaction partner such as DPPX,

193

potentially allowing for structural characterization using traditional NMR or x-ray

crystallography approaches.

Potential Roles of Shadoo in Prion Disease

Perhaps the most important result obtained with regards to Shadoo is the fact that

Shadoo protein levels are significantly diminished in the brains of clinically-ill prion-

infected mice. When coupled with the neuroprotective activity exhibited by Shadoo

against apoptotic stimuli, it is tempting to speculate that some of the neuropathological

and clinical aspects of prion disease may be accentuated due to the loss of Shadoo protein

and a concomitant decrease in Shadoo-associated neuroprotective activity. In order to test

this inference, it will be necessary to first demonstrate that Shadoo protein levels are

decreased at the clinical stage of other prion diseases, in particular non-induced prion

diseases such as sporadic or genetic CJD, scrapie, BSE, and CWD. This will determine

whether Shadoo down-regulation is a general feature of all prion diseases or is specific to

mouse-adapted scrapie in experimentally-infected rodents. One possible system for

testing the role of Shadoo in clinical prion disease is mice which express PrP lacking its

GPI anchor [86]. In these mice, clinical prion disease and prion replication are

dissociated—prion replication is uncompromised (i.e. high prion titres are present in

infected brains) yet the mice fail to develop the clinical symptoms of prion disease. By

assessing the levels of Shadoo in these mice following prion challenge, it should be

possible to ascertain any link between Shadoo levels, prion replication, and clinical prion

disease.

The mechanism of Shadoo down-regulation remains enigmatic and will need to

be investigated further. The fact that Shadoo levels are also reduced in transfected prion-

infected cells should provide a simple model system for probing the biochemical basis of

this phenomenon. The decrease in Shadoo protein levels corresponds well with the

appearance of PrPres in prion-infected Prnp+/0 mice hundreds of days before the onset of

neurological disease symptoms (George Carlson, personal communication). This

suggests that the disappearance of Shadoo is inherently linked with prion replication.

Thus, it will be interesting to determine the effect of Shadoo protein levels on prion

replication. For this purpose, the aforementioned Sprn0/0 mice will certainly be useful.

194

Inoculation of these mice with prions will clearly demonstrate whether or not Shadoo is

absolutely necessary for prion replication. Additionally, prion replication in mice over-

expressing Shadoo in the brain should also be tested. Mice over-expressing Shadoo

(TgSho) could be created using the hamster prion promoter-containing Cos.Fse.Tet

cosmid vector in order to ensure that Shadoo is expressed in the same tissues as PrPC.

Challenge of TgSho mice with prions may provide useful clues towards the mechanism

of prion replication as well as prion disease neuropathology. The fact that Shadoo has a

highly flexible structure suggests that it may be an ideal candidate for involvement in

enciphering prion strain-specific properties. The inoculation of both Sprn0/0 and TgSho

mice with various prion strains should help to assess any putative relationship between

Shadoo and prion strains.

TgSho mice will also provide a method for confirming that Shadoo can counteract

Doppel/ΔPrP toxicity in vivo. TgSho mice on a Prnp0/0 background can be crossed with

mice expressing ΔPrP or Doppel [217, 289] or even PrP alleles lacking the hydrophobic

tract [279, 280] in order to test the in vivo neuroprotective activity of Shadoo. Because

the hamster Prnp promoter will direct Shadoo expression in the granule cell layer of the

cerebellum, a direct evaluation of the protective properties of Shadoo can be made.

TgSho mice may also be useful for assaying the protective properties of Shadoo in other

neuronal insult paradigms such as in stroke models [275].

Identification of DPPX as a Plausible Candidate for LPrP

The identification of Shadoo as a strong candidate for the hypothetical π protein

gives credence to the original LPrP model which was developed to explain phenotypic

interactions between prion proteins in cerebellar neurons [217]. The other major

component of this model is LPrP, a putative prion protein ligand capable of binding to

PrPC, Doppel, ΔPrP, and π and which is responsible for controlling neurotoxic and

neuroprotective signals emanating from prion proteins. No candidates for LPrP have

surfaced, frustrating the characterization of Doppel/ΔPrP-induced neurotoxicity and

PrPC-mediated neuroprotection. However, the type II transmembrane protein DPPX has

now been identified as a plausible candidate for LPrP based upon patterns of expression

and binding properties. Thus, DPPX is the first protein identified capable of interacting

195

with all three members of the prion protein family. DPPX forms high molecular cell

surface complexes with PrPC, Doppel, ΔPrP, and Shadoo in N2a and HEK293 cells and is

expressed at high levels in the granule cells of the cerebellum. Complexes between prion

proteins and DPPX are governed by multiple binding determinants. At least two DPPX-

binding sites exist in prion proteins: a C-terminal binding site located in the α-helical

domain and present in PrPC, Doppel, and ΔPrP, and an N-terminal binding site present in

Shadoo and PrPC. In DPPX, the β-propeller domain is essential for complex formation

with PrPC and Doppel whereas Shadoo binds to a juxtamembrane epitope in DPPX.

When coupled to the fact that knockdown of DPPX in cultured CGNs leads to apoptosis,

DPPX becomes a leading candidate for LPrP.

A proposed refinement to the original LPrP model is outlined in Figure 5.1. In

wild-type mice, PrPC interacts with a DPPX dimer to produce a ‘signal’ that is associated

with cell survival. The exact nature of this ‘signal’ is unclear, but it is presumably linked

to the observed neuroprotective properties of PrPC in numerous experimental paradigms

[268, 269, 275, 356]. Whether or not the binding of PrPC to DPPX results in the

engagement of an actual signal transduction pathway or rather is associated with a non-

signaling event which promotes cell survival remains to be determined. Two binding

interactions regulate the formation of complexes between PrPC and DPPX. The α-helical

domain of PrPC binds to the β-propeller domain of DPPX and this interaction serves to

anchor the two proteins together. The second interaction which involves an epitope in the

N-terminus of PrPC and the juxtamembrane region of DPPX is responsible for

transduction of the neuroprotective signal. There are two hypotheses for the nature of the

interaction between PrPC and DPPX. The trophic hypothesis assumes that the interaction

between PrPC and DPPX is beneficial (but not necessarily essential) for neuronal survival

under resting conditions. The lack of an overt phenotype in Prnp0/0 mice can then be

explained by the compensatory action of Shadoo. Although Shadoo does not bind the

DPPX β-propeller domain, it is capable of binding to a juxtamembrane epitope on DPPX

via its flexible N-terminal domain. Similar to PrPC, this binding event is associated with

the production of a neuroprotective ‘signal’ and thus Shadoo is capable of functionally

compensating for the absence of PrPC via its interaction with DPPX. In neurons lacking

Shadoo and PrPC (such as the CGNs of Prnp0/0 mice) it is possible that DPPX can

196

Figure 5.1. Model to explain to phenotypic interactions between prion proteins in cerebellar neurons. A: In wild-type mice, PrPC forms a complex with DPPX which results in the production of a signal associated with cell survival. PrPC forms two contacts with DPPX: one involving the C-terminal α-helical domain of PrPC and the other involving the flexible N-terminal domain (red). The interaction between the PrPC N-terminal domain and DPPX is associated with the generation of the neuroprotective signal. B: In Prnp0/0 mice, the lack of a phenotype can be attributed to the functional compensation of PrPC by Shadoo. An interaction between the Shadoo N-terminal domain (green) and DPPX results in the production of the cell survival signal. C: In Prnp0/0 mice expressing either Doppel or ΔPrP, DPPX forms a complex with Doppel/ΔPrP which involves the prion protein α-helical C-terminal domain. However, the second DPPX interaction site is absent in Doppel/ΔPrP resulting in the generation of an improper cell death-associated signal.

197

produce a small amount of the cell survival-associated ‘signal’ via autoactivation [279].

The second hypothesis (the neuroprotection hypothesis) assumes that the interaction

between DPPX and PrPC or Shadoo is non-essential under resting conditions. Instead,

complexes between PrPC and/or Sho and DPPX are important for protecting against pro-

apoptotic neuronal insults (such as the expression of Doppel or ΔPrP). In mice lacking

PrPC but expressing Doppel or ΔPrP in cerebellar neurons, Doppel or ΔPrP bind to DPPX

via their C-terminal α-helical domains but are unable to stimulate the production of the

neuroprotective ‘signal’ due to the absence of the N-terminal domain. This results in the

generation of an improper signal by DPPX which leads to cell death via an unidentified

mechanism. Presumably, the ability of PrPC to counteract this toxicity is due to direct

competition for DPPX binding between PrPC and Doppel/ΔPrP with PrPC possessing a

higher affinity for DPPX.

DPPX and Prion Protein Signaling

The identification of DPPX as a conserved interacting partner for the prion

protein family opens up multiple avenues of further research. First and foremost, it must

be demonstrated conclusively that DPPX can modulate the neurotoxic properties of

Doppel/ΔPrP and the neuroprotective properties of PrPC/Shadoo in a functional assay as

this is paramount to confirming the identification of DPPX as LPrP. One possible way of

testing this using the CGN assay is to use the DPPXΔCyto allele as a hypothetical

dominant negative. This mutant should be able to heterodimerize with wild-type

endogenous DPPX in CGNs resulting in a loss of function if the hypothesis that the

cytoplasmic domain of DPPX is critical for transducing neurotoxic signals from Doppel

is correct. Since DPPXΔCyto retains the ability to bind Doppel, it may prevent

transfected Doppel from initiating the neurotoxic signaling pathway. These studies are

currently in progress. Alternatively, it may be possible to use a recently-developed

Doppel toxicity model which involves N2a cells [358]. In this system, co-transfection of

Doppel and Prnp siRNA leads to apoptosis of N2a cells. Therefore, it may be feasible to

perform Prnp/DPP6 siRNA co-transfections of N2a cell stably expressing wild-type

Doppel. However, this assay needs to be validated before such studies can commence.

198

Even if DPPX and LPrP are confirmed to be synonymous, the downstream events

that are responsible for determining the neuroprotection or neurotoxic endpoints remain

to be determined. As a start, it will be useful to identify and characterize DPPX-

interacting proteins. To date, the only known interacting partners of DPPX are Kv4.2 (a

pore subunit of A-type potassium channels), the prion proteins identified in this study,

and DPP10 [375, 420]. However, when crosslinking analysis is performed on DPPX-

transfected N2a cells, high molecular weight smears are obtained on a Western blot with

DPPX antibodies. This suggests that DPPX is capable of binding to numerous mouse

proteins, perhaps due to the presence of its β-propeller domain. TcTPC could be used to

characterize the DPPX interactome in the mouse brain in conjunction with the 03K1 anti-

DPPX antibody which works well for immunoprecipitations. A parallel approach could

also be taken on N2a cells stably transfected with various DPPX N-terminal splicing

isoforms. Since splice variant-specific DPPX antibodies do not yet exist, this type of

approach may help to identify isoform-specific interactors of DPPX. This may be

particularly important for identifying pathways downstream of DPPX since the splice

variants differ in their cytoplasmic domains.

The strongest evidence for DPPX involvement in prion protein signaling may

derive from future studies on DPP60/0 mice. Although DPP6df5J/Rw mice do not express

full-length DPPX, they are not an ideal system for studying the relevance of DPPX to

prion protein biology since multiple genes are deleted in the df5J allele and difficulties

were experienced in breeding these mice. Consequently, it is essential that true DPP60/0

mice be generated. Since DPP6df5J/Rw mice have no overt phenotypes, it seems reasonable

to predict that DPP60/0 mice will also be viable and phenotypically normal. Nonetheless,

DPP60/0 mice are being constructed as conditional knockouts (David Westaway, personal

communication). These mice can then be crossed with Prnp0/0 mice to generate mice

lacking both PrPC and DPPX. The double knockout mice can then be crossed with mice

expressing Doppel, PrPΔ32-134, or PrPΔ105-125 in the cerebellum. If the identification

of DPPX as LPrP is valid then these mice should not develop a neurodegenerative disease.

199

DPPX and Prion Disease Biology

The creation of DPP60/0 mice will also facilitate a definitive assessment of the

role of DPPX in prion replication and disease. As discussed in Chapter 4, DPPX seems

like an ideal candidate for involvement in prion replication due to its dimeric structure

and the binding of PrPC to its β-propeller domain. However, no concrete evidence has

been obtained to date that suggests DPPX is involved in any aspect of prion disease.

Nonetheless, the gold standard for testing a protein’s role in prion disease is to inoculate

knockout mice with different strains of prions via a variety of routes and then evaluate

the clinical progression of prion disease, neuropathology, PrPres biochemistry, and prion

titres in the brains of inoculated animals. This strategy has previously been used to show

that two other PrPC-interacting proteins, N-CAM and ApoE, have no effect on prion

disease in mice, at least in the context of intracerebral inoculation with RML scrapie

prions [322, 426]. A careful assessment of DPP60/0 mice inoculated with prions should

provide conclusive evidence as to whether or not DPPX is relevant to prion disease

biology. In parallel to studies involving DPPX knockout mice, prion inoculations of

transgenic mice over-expressing DPPX should also be considered. Over-expression of

DPPX may either hinder or promote prion replication resulting in alterations to the

clinical progression of prion disease in these mice. If, for instance, DPPX over-

expression is found to enhance prion replication, these mice could then be crossed with

mice over-expressing wild-type PrP [388] in order to create transgenic animals which

rapidly succumb to prion disease. Such mice would greatly improve time-consuming

prion bioassays. It may also be of interest to generate transgenic mice which over-express

the DPPX-E_SV1 splice variant. In transfected cells, this isoform appears to be prone to

forming oligomers and higher order structures and therefore may be well-suited to

concentrating numerous PrP molecules together in a small space. This may facilitate

interactions between PrPC and PrPSc and influence the efficiency of prion replication.

Another possible method for forming a link between DPPX and prion disease is to

examine the DPPX binding efficiency of various PrP alleles that are either associated

with genetic prion disease or are known to modulate prion disease incubation time. For

instance, two polymorphisms in the mouse Prnp gene are known to control prion disease

incubation time. The L108F and T189V dimorphisms (an allele commonly referred to as

200

PrP-B) cause a significant lengthening of prion disease incubation time in mice

expressing this allele [128, 131]. Therefore, it will be interesting to evaluate complex

formation between DPPX and mouse PrPC containing either or both of the L108F and

T189V variations. Notably, the T189V polymorphism is located within helix B of PrPC, a

suspected binding site for DPPX and the L108F polymorphism is located near the

hydrophobic tract, a region that is capable of modulating PrP complex formation with

DPPX. An altered binding efficiency may provide further evidence that DPPX is

involved in prion disease. Secondly, it is known that the M129V polymorphism in human

PrP (which is equivalent to M128V in mouse PrP) has a profound influence on prion

disease and is known to alter the structure of recombinant PrP [71, 163, 452, 453]. This

polymorphism also falls within the vicinity of the hydrophobic tract and therefore its

effect on the efficiency of PrP complex formation with DPPX should be tested. Similarly,

PrP alleles containing prion-disease associated mutations should also be assayed for their

ability to form complexes with DPPX. In this case, specific attention should be paid to

mutations that occur within or near the DPPX binding sites identified in this study. Two

possible mutations to consider are the D178N mutation which causes either FFI or gCJD

and is located in helix B of PrP and the A117V mutation which causes GSS and is

located in the middle of the PrP hydrophobic tract [61].

DPPX and Other Neurodegenerative Diseases

Although caution must be exercised when interpreting whole genome association

studies [454], the recently identified association between an intronic variant in the DPP6

gene and susceptibility to sporadic ALS in at least two distinct populations has provided

the motivation for examining DPPX in the context of other neurodegenerative diseases

[415, 416]. To begin to examine the plausibility of the proposed link between DPPX and

ALS, DPPX expression was examined in the spinal cords of mice. Prominent labeling of

motor neurons was observed with three different anti-DPPX antibodies in a spinal cord

isolated from a wild-type mouse (Figure 5.2A). Furthermore, DPPX signal was readily

detected in a spinal cord homogenate from a wild-type mouse by Western blotting

(Figure 5.2B). Therefore, DPPX is expressed in motor neurons and thus is a plausible

modulator of ALS pathology. Further studies are clearly required to begin to dissect any

201

A04DX-2 03J2 03K1

B

α-DPPX 03J2

α-DPPX 03K1

α-DPP10 06D10-2

α-actin

Figure 5.2. DPPX is expressed in motor neurons of the spinal cord. A: Immunohistochemistry on methacarn-fixed wild-type mouse spinal cords using the indicated anti-DPPX antibodies. Strong signal for DPPX protein is present in motor neurons with all three antibodies. Scale bar = 40 µm. B: Western blots of wild-type mouse brain or spinal cord homogenates probed with the indicated antibodies. Signal for both DPPX and DPP10 protein is present in the spinal cord.

202

link between DPPX and motor neuron pathology. Thus far, only intronic variations in the

DPP6 gene have been identified and it is unclear how they increase susceptibility to

sporadic ALS. Perhaps the variations result in a destabilization of DPP6 mRNA and a

concomitant decrease in DPPX protein levels. This hypothesis can easily be tested using

ALS and healthy spinal cord tissue in conjunction with screening for the identified

genetic variant. Secondly, DPPX protein levels can be examined in mouse models of

ALS which express mutant SOD1 proteins [447-449]. It will also be of interest to

examine spinal cord tissue in DPP60/0 mice. The lack of an obvious motor phenotype in

DPP6df5J/Rw mice suggests that any motor neuron pathology stemming from the absence

of DPPX is mild. However, spinal cords of DPP6df5J/Rw mice have not yet been checked

for the presence of pathology. Due to the putative association between DPPX and ALS,

links between DPPX and other neurodegenerative diseases should be considered. For

instance, many of the crucial biochemical events in Alzheimer’s disease such as the γ-

secretase-mediated release of Aβ peptides from APP occur at the cell surface. It seems

likely that toxic Aβ42 molecules would have access to DPPX and may be capable of

interfering with DPPX physiology. Again, DPP60/0 mice would be useful for testing this

hypothesis. These mice could be crossed with the TgCRND8 mouse model of

Alzheimer’s disease in order to assess the importance of DPPX to pathological and

behavioural deficits in this model [382, 455]. In addition, the aforementioned DPPX

interactome studies may be useful for revealing DPPX-interacting proteins that are

components of a hypothetical neurodegeneration pathway common to various diseases.

In conclusion, two new proteins have been identified, Shadoo and DPPX, which

are likely to shed light on the enigmatic function of PrPC and demystify certain aspects of

prion disease. Further study of Shadoo and its biological similarities to PrPC may help to

clarify the role of PrPC in the brain and may provide insight into prion replication.

Similarly, a careful examination of DPPX may provide a link between cell surface prion

protein biology and downstream neuroprotective and neurotoxic events. Thus, it is

possible that the identification of Shadoo and DPPX as important contributors to prion

protein biology may serve to unite two currently divergent fields of prion research,

namely prion cell biology and prion disease pathobiology.

203

References

[1] B. Sigurdsson, Rida, a chronic encephalitis of sheep with general remarks on infections which develop slowly and some of their special characteristics, Br Vet J 110 (1954) 341-354.

[2] W.J. Hadlow, Scrapie and kuru, Lancet 2 (1959) 289-290.

[3] D.C. Gajdusek, Unconventional viruses and the origin and disappearance of kuru, Science 197 (1977) 943-960.

[4] T. Alper, D.A. Haig, M.C. Clarke, The exceptionally small size of the scrapie agent, Biochem Biophys Res Commun 22 (1966) 278-284.

[5] T. Alper, W.A. Cramp, D.A. Haig, M.C. Clarke, Does the agent of scrapie replicate without nucleic acid?, Nature 214 (1967) 764-766.

[6] S.B. Prusiner, Novel proteinaceous infectious particles cause scrapie, Science 216 (1982) 136-144.

[7] N. Meyer, V. Rosenbaum, B. Schmidt, K. Gilles, C. Mirenda, D. Groth, S.B. Prusiner, D. Riesner, Search for a putative scrapie genome in purified prion fractions reveals a paucity of nucleic acids, J Gen Virol 72 ( Pt 1) (1991) 37-49.

[8] K. Kellings, N. Meyer, C. Mirenda, S.B. Prusiner, D. Riesner, Further analysis of nucleic acids in purified scrapie prion preparations by improved return refocusing gel electrophoresis, J Gen Virol 73 ( Pt 4) (1992) 1025-1029.

[9] K. Kellings, N. Meyer, C. Mirenda, S.B. Prusiner, D. Riesner, Analysis of nucleic acids in purified scrapie prion preparations, Archives of virology 7 (1993) 215-225.

[10] J.G. Safar, K. Kellings, A. Serban, D. Groth, J.E. Cleaver, S.B. Prusiner, D. Riesner, Search for a prion-specific nucleic acid, J Virol 79 (2005) 10796-10806.

[11] M.C. Clarke, D.A. Haig, Attempts to demonstrate neutralising antibodies in the sera of scrapie-affected animals, Vet Rec 78 (1966) 647-649.

[12] M. Worthington, Interferon system in mice infected with the scrapie agent, Infection and immunity 6 (1972) 643-645.

[13] J.S. Griffith, Self-replication and scrapie, Nature 215 (1967) 1043-1044.

[14] D.C. Bolton, M.P. McKinley, S.B. Prusiner, Identification of a protein that purifies with the scrapie prion, Science 218 (1982) 1309-1311.

[15] M.P. McKinley, D.C. Bolton, S.B. Prusiner, A protease-resistant protein is a structural component of the scrapie prion, Cell 35 (1983) 57-62.

204

[16] S.B. Prusiner, M.P. McKinley, K.A. Bowman, D.C. Bolton, P.E. Bendheim, D.F. Groth, G.G. Glenner, Scrapie prions aggregate to form amyloid-like birefringent rods, Cell 35 (1983) 349-358.

[17] K.M. Pan, M. Baldwin, J. Nguyen, M. Gasset, A. Serban, D. Groth, I. Mehlhorn, Z. Huang, R.J. Fletterick, F.E. Cohen, et al., Conversion of alpha-helices into beta-sheets features in the formation of the scrapie prion proteins, Proc Natl Acad Sci U S A 90 (1993) 10962-10966.

[18] L. Manuelidis, A 25 nm virion is the likely cause of transmissible spongiform encephalopathies, Journal of cellular biochemistry 100 (2007) 897-915.

[19] K.K. Hsiao, D. Groth, M. Scott, S.L. Yang, H. Serban, D. Rapp, D. Foster, M. Torchia, S.J. Dearmond, S.B. Prusiner, Serial transmission in rodents of neurodegeneration from transgenic mice expressing mutant prion protein, Proc Natl Acad Sci U S A 91 (1994) 9126-9130.

[20] K.K. Hsiao, M. Scott, D. Foster, D.F. Groth, S.J. DeArmond, S.B. Prusiner, Spontaneous neurodegeneration in transgenic mice with mutant prion protein, Science 250 (1990) 1587-1590.

[21] J. Castilla, P. Saa, C. Hetz, C. Soto, In vitro generation of infectious scrapie prions, Cell 121 (2005) 195-206.

[22] G. Legname, I.V. Baskakov, H.O. Nguyen, D. Riesner, F.E. Cohen, S.J. DeArmond, S.B. Prusiner, Synthetic mammalian prions, Science 305 (2004) 673-676.

[23] K.E. Nazor, F. Kuhn, T. Seward, M. Green, D. Zwald, M. Purro, J. Schmid, K. Biffiger, A.M. Power, B. Oesch, A.J. Raeber, G.C. Telling, Immunodetection of disease-associated mutant PrP, which accelerates disease in GSS transgenic mice, Embo J 24 (2005) 2472-2480.

[24] R.B. Wickner, [URE3] as an altered URE2 protein: evidence for a prion analog in Saccharomyces cerevisiae, Science 264 (1994) 566-569.

[25] I.L. Derkatch, M.E. Bradley, J.Y. Hong, S.W. Liebman, Prions affect the appearance of other prions: the story of [PIN(+)], Cell 106 (2001) 171-182.

[26] V. Coustou, C. Deleu, S. Saupe, J. Begueret, The protein product of the het-s heterokaryon incompatibility gene of the fungus Podospora anserina behaves as a prion analog, Proc Natl Acad Sci U S A 94 (1997) 9773-9778.

[27] H.L. True, I. Berlin, S.L. Lindquist, Epigenetic regulation of translation reveals hidden genetic variation to produce complex traits, Nature 431 (2004) 184-187.

[28] H.L. True, S.L. Lindquist, A yeast prion provides a mechanism for genetic variation and phenotypic diversity, Nature 407 (2000) 477-483.

205

[29] T. Nakayashiki, C.P. Kurtzman, H.K. Edskes, R.B. Wickner, Yeast prions [URE3] and [PSI+] are diseases, Proc Natl Acad Sci U S A 102 (2005) 10575-10580.

[30] K. Si, S. Lindquist, E.R. Kandel, A neuronal isoform of the aplysia CPEB has prion-like properties, Cell 115 (2003) 879-891.

[31] S.B. Prusiner, E. Williams, J.L. Laplanche, M. Shinagawa, Scrapie, Chronic Wasting Disease, and Transmissible Mink Encephalopathy, in: S.B. Prusiner (Ed.), Prion Biology and Diseases, 2nd Edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 2004, pp. 545-594.

[32] R.B. Wickner, Scrapie in ancient China?, Science 309 (2005) 874.

[33] I.H. Pattison, M.N. Hoare, J.N. Jebbett, W.A. Watson, Spread of scrapie to sheep and goats by oral dosing with foetal membranes from scrapie-affected sheep, Vet Rec 90 (1972) 465-468.

[34] T. Onodera, T. Ikeda, Y. Muramatsu, M. Shinagawa, Isolation of scrapie agent from the placenta of sheep with natural scrapie in Japan, Microbiol Immunol 37 (1993) 311-316.

[35] M. Heikenwalder, N. Zeller, H. Seeger, M. Prinz, P.C. Klohn, P. Schwarz, N.H. Ruddle, C. Weissmann, A. Aguzzi, Chronic lymphocytic inflammation specifies the organ tropism of prions, Science 307 (2005) 1107-1110.

[36] H. Seeger, M. Heikenwalder, N. Zeller, J. Kranich, P. Schwarz, A. Gaspert, B. Seifert, G. Miele, A. Aguzzi, Coincident scrapie infection and nephritis lead to urinary prion excretion, Science 310 (2005) 324-326.

[37] C. Ligios, G.M. Cancedda, I. Margalith, C. Santucciu, L. Madau, C. Maestrale, M. Basagni, M. Saba, M. Heikenwalder, Intraepithelial and interstitial deposition of pathological prion protein in kidneys of scrapie-affected sheep, PLoS ONE 2 (2007) e859.

[38] C. Ligios, C.J. Sigurdson, C. Santucciu, G. Carcassola, G. Manco, M. Basagni, C. Maestrale, M.G. Cancedda, L. Madau, A. Aguzzi, PrPSc in mammary glands of sheep affected by scrapie and mastitis, Nat Med 11 (2005) 1137-1138.

[39] G.A. Wells, A.C. Scott, C.T. Johnson, R.F. Gunning, R.D. Hancock, M. Jeffrey, M. Dawson, R. Bradley, A novel progressive spongiform encephalopathy in cattle, Vet Rec 121 (1987) 419-420.

[40] J.W. Wilesmith, J.B. Ryan, M.J. Atkinson, Bovine spongiform encephalopathy: epidemiological studies on the origin, Vet Rec 128 (1991) 199-203.

206

[41] N.J. Andrews, C.P. Farrington, H.J. Ward, S.N. Cousens, P.G. Smith, A.M. Molesworth, R.S. Knight, J.W. Ironside, R.G. Will, Deaths from variant Creutzfeldt-Jakob disease in the UK, Lancet 361 (2003) 751-752.

[42] C.J. Sigurdson, A. Aguzzi, Chronic wasting disease, Biochim Biophys Acta 1772 (2007) 610-618.

[43] L.A. Baeten, B.E. Powers, J.E. Jewell, T.R. Spraker, M.W. Miller, A natural case of chronic wasting disease in a free-ranging moose (Alces alces shirasi), J Wildl Dis 43 (2007) 309-314.

[44] M.W. Miller, M.A. Wild, E.S. Williams, Epidemiology of chronic wasting disease in captive Rocky Mountain elk, J Wildl Dis 34 (1998) 532-538.

[45] M.W. Miller, E.S. Williams, C.W. McCarty, T.R. Spraker, T.J. Kreeger, C.T. Larsen, E.T. Thorne, Epizootiology of chronic wasting disease in free-ranging cervids in Colorado and Wyoming, J Wildl Dis 36 (2000) 676-690.

[46] E.S. Williams, Chronic wasting disease, Vet Pathol 42 (2005) 530-549.

[47] S. Kahn, C. Dube, L. Bates, A. Balachandran, Chronic wasting disease in Canada: Part 1, Can Vet J 45 (2004) 397-404.

[48] C.J. Sigurdson, E.S. Williams, M.W. Miller, T.R. Spraker, K.I. O'Rourke, E.A. Hoover, Oral transmission and early lymphoid tropism of chronic wasting disease PrPres in mule deer fawns (Odocoileus hemionus), J Gen Virol 80 ( Pt 10) (1999) 2757-2764.

[49] K.I. O'Rourke, D. Zhuang, A. Lyda, G. Gomez, E.S. Williams, W. Tuo, M.W. Miller, Abundant PrP(CWD) in tonsil from mule deer with preclinical chronic wasting disease, J Vet Diagn Invest 15 (2003) 320-323.

[50] C.K. Mathiason, J.G. Powers, S.J. Dahmes, D.A. Osborn, K.V. Miller, R.J. Warren, G.L. Mason, S.A. Hays, J. Hayes-Klug, D.M. Seelig, M.A. Wild, L.L. Wolfe, T.R. Spraker, M.W. Miller, C.J. Sigurdson, G.C. Telling, E.A. Hoover, Infectious prions in the saliva and blood of deer with chronic wasting disease, Science 314 (2006) 133-136.

[51] Q. Kong, S. Huang, W. Zou, D. Vanegas, M. Wang, D. Wu, J. Yuan, M. Zheng, H. Bai, H. Deng, K. Chen, A.L. Jenny, K. O'Rourke, E.D. Belay, L.B. Schonberger, R.B. Petersen, M.S. Sy, S.G. Chen, P. Gambetti, Chronic wasting disease of elk: transmissibility to humans examined by transgenic mouse models, J Neurosci 25 (2005) 7944-7949.

[52] R.C. Angers, S.R. Browning, T.S. Seward, C.J. Sigurdson, M.W. Miller, E.A. Hoover, G.C. Telling, Prions in skeletal muscles of deer with chronic wasting disease, Science 311 (2006) 1117.

207

[53] J.C. Watts, A. Balachandran, D. Westaway, The expanding universe of prion diseases, PLoS Pathog 2 (2006) e26.

[54] C. Casalone, G. Zanusso, P. Acutis, S. Ferrari, L. Capucci, F. Tagliavini, S. Monaco, M. Caramelli, Identification of a second bovine amyloidotic spongiform encephalopathy: molecular similarities with sporadic Creutzfeldt-Jakob disease, Proc Natl Acad Sci U S A 101 (2004) 3065-3070.

[55] R. Capobianco, C. Casalone, S. Suardi, M. Mangieri, C. Miccolo, L. Limido, M. Catania, G. Rossi, G. Di Fede, G. Giaccone, M.G. Bruzzone, L. Minati, C. Corona, P. Acutis, D. Gelmetti, G. Lombardi, M.H. Groschup, A. Buschmann, G. Zanusso, S. Monaco, M. Caramelli, F. Tagliavini, Conversion of the BASE prion strain into the BSE strain: the origin of BSE?, PLoS Pathog 3 (2007) e31.

[56] S.L. Benestad, P. Sarradin, B. Thu, J. Schonheit, M.A. Tranulis, B. Bratberg, Cases of scrapie with unusual features in Norway and designation of a new type, Nor98, Vet Rec 153 (2003) 202-208.

[57] A. Le Dur, V. Beringue, O. Andreoletti, F. Reine, T.L. Lai, T. Baron, B. Bratberg, J.L. Vilotte, P. Sarradin, S.L. Benestad, H. Laude, A newly identified type of scrapie agent can naturally infect sheep with resistant PrP genotypes, Proc Natl Acad Sci U S A 102 (2005) 16031-16036.

[58] D.C. Gajdusek, C.J. Gibbs, M. Alpers, Experimental transmission of a Kuru-like syndrome to chimpanzees, Nature 209 (1966) 794-796.

[59] D.C. Gajdusek, C.J. Gibbs, Jr., M. Alpers, Transmission and passage of experimenal "kuru" to chimpanzees, Science 155 (1967) 212-214.

[60] J. Collinge, J. Whitfield, E. McKintosh, J. Beck, S. Mead, D.J. Thomas, M.P. Alpers, Kuru in the 21st century--an acquired human prion disease with very long incubation periods, Lancet 367 (2006) 2068-2074.

[61] Q. Kong, W.K. Surewicz, R.B. Petersen, W. Zou, S.G. Chen, P. Gambetti, P. Parchi, S. Capellari, L. Goldfarb, P. Montagna, E. Lugaresi, P. Piccardo, B. Ghetti, Inherited Prion Diseases, in: S.B. Prusiner (Ed.), Prion Biology and Diseases Second Edition, Cold Spring Harbor Laboratory Press, New York, N.Y., 2004, pp. 673-775.

[62] W. Swietnicki, R.B. Petersen, P. Gambetti, W.K. Surewicz, Familial mutations and the thermodynamic stability of the recombinant human prion protein, J Biol Chem 273 (1998) 31048-31052.

[63] S. Liemann, R. Glockshuber, Influence of amino acid substitutions related to inherited human prion diseases on the thermodynamic stability of the cellular prion protein, Biochemistry 38 (1999) 3258-3267.

208

[64] R.G. Will, M.P. Alpers, D. Dormont, L.B. Schonberger, Infectious and Sporadic Prion Diseases, in: S.B. Prusiner (Ed.), Prion Biology and Diseases Second Edition, Cold Spring Harbor Laboratory Press, New York, N.Y., 2004, pp. 629-671.

[65] T.K. Koch, B.O. Berg, S.J. De Armond, R.F. Gravina, Creutzfeldt-Jakob disease in a young adult with idiopathic hypopituitarism. Possible relation to the administration of cadaveric human growth hormone, N Engl J Med 313 (1985) 731-733.

[66] P. Brown, M. Preece, J.P. Brandel, T. Sato, L. McShane, I. Zerr, A. Fletcher, R.G. Will, M. Pocchiari, N.R. Cashman, J.H. d'Aignaux, L. Cervenakova, J. Fradkin, L.B. Schonberger, S.J. Collins, Iatrogenic Creutzfeldt-Jakob disease at the millennium, Neurology 55 (2000) 1075-1081.

[67] R.G. Will, J.W. Ironside, M. Zeidler, S.N. Cousens, K. Estibeiro, A. Alperovitch, S. Poser, M. Pocchiari, A. Hofman, P.G. Smith, A new variant of Creutzfeldt-Jakob disease in the UK, Lancet 347 (1996) 921-925.

[68] J. Collinge, K.C. Sidle, J. Meads, J. Ironside, A.F. Hill, Molecular analysis of prion strain variation and the aetiology of 'new variant' CJD, Nature 383 (1996) 685-690.

[69] M.E. Bruce, R.G. Will, J.W. Ironside, I. McConnell, D. Drummond, A. Suttie, L. McCardle, A. Chree, J. Hope, C. Birkett, S. Cousens, H. Fraser, C.J. Bostock, Transmissions to mice indicate that 'new variant' CJD is caused by the BSE agent, Nature 389 (1997) 498-501.

[70] M.R. Scott, R. Will, J. Ironside, H.O. Nguyen, P. Tremblay, S.J. DeArmond, S.B. Prusiner, Compelling transgenetic evidence for transmission of bovine spongiform encephalopathy prions to humans, Proc Natl Acad Sci U S A 96 (1999) 15137-15142.

[71] J.D. Wadsworth, E.A. Asante, M. Desbruslais, J.M. Linehan, S. Joiner, I. Gowland, J. Welch, L. Stone, S.E. Lloyd, A.F. Hill, S. Brandner, J. Collinge, Human prion protein with valine 129 prevents expression of variant CJD phenotype, Science 306 (2004) 1793-1796.

[72] C.A. Llewelyn, P.E. Hewitt, R.S. Knight, K. Amar, S. Cousens, J. Mackenzie, R.G. Will, Possible transmission of variant Creutzfeldt-Jakob disease by blood transfusion, Lancet 363 (2004) 417-421.

[73] A.H. Peden, M.W. Head, D.L. Ritchie, J.E. Bell, J.W. Ironside, Preclinical vCJD after blood transfusion in a PRNP codon 129 heterozygous patient, Lancet 364 (2004) 527-529.

[74] S.J. Wroe, S. Pal, D. Siddique, H. Hyare, R. Macfarlane, S. Joiner, J.M. Linehan, S. Brandner, J.D. Wadsworth, P. Hewitt, J. Collinge, Clinical presentation and

209

pre-mortem diagnosis of variant Creutzfeldt-Jakob disease associated with blood transfusion: a case report, Lancet 368 (2006) 2061-2067.

[75] B. Ghetti, P. Piccardo, M.G. Spillantini, Y. Ichimiya, M. Porro, F. Perini, T. Kitamoto, J. Tateishi, C. Seiler, B. Frangione, O. Bugiani, G. Giaccone, F. Prelli, M. Goedert, S.R. Dlouhy, F. Tagliavini, Vascular variant of prion protein cerebral amyloidosis with tau-positive neurofibrillary tangles: the phenotype of the stop codon 145 mutation in PRNP, Proc Natl Acad Sci U S A 93 (1996) 744-748.

[76] T. Kitamoto, R. Iizuka, J. Tateishi, An amber mutation of prion protein in Gerstmann-Straussler syndrome with mutant PrP plaques, Biochem Biophys Res Commun 192 (1993) 525-531.

[77] F. Tagliavini, P.M. Lievens, C. Tranchant, J.M. Warter, M. Mohr, G. Giaccone, F. Perini, G. Rossi, M. Salmona, P. Piccardo, B. Ghetti, R.C. Beavis, O. Bugiani, B. Frangione, F. Prelli, A 7-kDa prion protein (PrP) fragment, an integral component of the PrP region required for infectivity, is the major amyloid protein in Gerstmann-Straussler-Scheinker disease A117V, J Biol Chem 276 (2001) 6009-6015.

[78] F. Tagliavini, F. Prelli, J. Ghiso, O. Bugiani, D. Serban, S.B. Prusiner, M.R. Farlow, B. Ghetti, B. Frangione, Amyloid protein of Gerstmann-Straussler-Scheinker disease (Indiana kindred) is an 11 kd fragment of prion protein with an N-terminal glycine at codon 58, Embo J 10 (1991) 513-519.

[79] P. Parchi, S. Capellari, S. Chin, H.B. Schwarz, N.P. Schecter, J.D. Butts, P. Hudkins, D.K. Burns, J.M. Powers, P. Gambetti, A subtype of sporadic prion disease mimicking fatal familial insomnia, Neurology 52 (1999) 1757-1763.

[80] J.A. Mastrianni, R. Nixon, R. Layzer, G.C. Telling, D. Han, S.J. DeArmond, S.B. Prusiner, Prion protein conformation in a patient with sporadic fatal insomnia, N Engl J Med 340 (1999) 1630-1638.

[81] F. Scaravilli, R.J. Cordery, H. Kretzschmar, P. Gambetti, B. Brink, V. Fritz, J. Temlett, C. Kaplan, D. Fish, S.F. An, W.J. Schulz-Schaeffer, M.N. Rossor, Sporadic fatal insomnia: a case study, Ann Neurol 48 (2000) 665-668.

[82] Y.S. Piao, A. Kakita, H. Watanabe, T. Kitamoto, H. Takahashi, Sporadic fatal insomnia with spongiform degeneration in the thalamus and widespread PrPSc deposits in the brain, Neuropathology 25 (2005) 144-149.

[83] S.J. DeArmond, J.W. Ironside, E. Bouzamondo-Bernstein, D. Peretz, J.R. Fraser, Neuropathology of Prion Diseases, in: S.B. Prusiner (Ed.), Prion Biology and Diseases, 2nd Edition, Cold Spring Harbour Laboratory Press, Cold Spring Harbour, New York, 2004, pp. 777-856.

210

[84] S. Brandner, S. Isenmann, A. Raeber, M. Fischer, A. Sailer, Y. Kobayashi, S. Marino, C. Weissmann, A. Aguzzi, Normal host prion protein necessary for scrapie-induced neurotoxicity, Nature 379 (1996) 339-343.

[85] M. Jeffrey, C.M. Goodsir, R.E. Race, B. Chesebro, Scrapie-specific neuronal lesions are independent of neuronal PrP expression, Ann Neurol 55 (2004) 781-792.

[86] B. Chesebro, M. Trifilo, R. Race, K. Meade-White, C. Teng, R. LaCasse, L. Raymond, C. Favara, G. Baron, S. Priola, B. Caughey, E. Masliah, M. Oldstone, Anchorless prion protein results in infectious amyloid disease without clinical scrapie, Science 308 (2005) 1435-1439.

[87] A. Aguzzi, Cell biology. Prion toxicity: all sail and no anchor, Science 308 (2005) 1420-1421.

[88] J. Ma, R. Wollmann, S. Lindquist, Neurotoxicity and neurodegeneration when PrP accumulates in the cytosol, Science 298 (2002) 1781-1785.

[89] J. Ma, S. Lindquist, Conversion of PrP to a self-perpetuating PrPSc-like conformation in the cytosol, Science 298 (2002) 1785-1788.

[90] X. Roucou, Q. Guo, Y. Zhang, C.G. Goodyer, A.C. LeBlanc, Cytosolic prion protein is not toxic and protects against Bax-mediated cell death in human primary neurons, J Biol Chem 278 (2003) 40877-40881.

[91] L. Fioriti, S. Dossena, L.R. Stewart, R.S. Stewart, D.A. Harris, G. Forloni, R. Chiesa, Cytosolic prion protein (PrP) is not toxic in N2a cells and primary neurons expressing pathogenic PrP mutations, J Biol Chem 280 (2005) 11320-11328.

[92] C. Crozet, J. Vezilier, V. Delfieu, T. Nishimura, T. Onodera, D. Casanova, S. Lehmann, F. Beranger, The truncated 23-230 form of the prion protein localizes to the nuclei of inducible cell lines independently of its nuclear localization signals and is not cytotoxic, Mol Cell Neurosci 32 (2006) 315-323.

[93] N. Fernandez-Borges, A. Brun, J.L. Whitton, B. Parra, F. Diaz-San Segundo, F.J. Salguero, J.M. Torres, F. Rodriguez, DNA vaccination can break immunological tolerance to PrP in wild-type mice and attenuates prion disease after intracerebral challenge, J Virol 80 (2006) 9970-9976.

[94] G. Kaiser-Schulz, A. Heit, L. Quintanilla-Martinez, F. Hammerschmidt, S. Hess, L. Jennen, H. Rezaei, H. Wagner, H.M. Schatzl, Polylactide-coglycolide microspheres co-encapsulating recombinant tandem prion protein with CpG-oligonucleotide break self-tolerance to prion protein in wild-type mice and induce CD4 and CD8 T cell responses, J Immunol 179 (2007) 2797-2807.

211

[95] C. Korth, B.C. May, F.E. Cohen, S.B. Prusiner, Acridine and phenothiazine derivatives as pharmacotherapeutics for prion disease, Proc Natl Acad Sci U S A 98 (2001) 9836-9841.

[96] A. Mange, N. Nishida, O. Milhavet, H.E. McMahon, D. Casanova, S. Lehmann, Amphotericin B inhibits the generation of the scrapie isoform of the prion protein in infected cultures, J Virol 74 (2000) 3135-3140.

[97] S. Supattapone, H.O. Nguyen, F.E. Cohen, S.B. Prusiner, M.R. Scott, Elimination of prions by branched polyamines and implications for therapeutics, Proc Natl Acad Sci U S A 96 (1999) 14529-14534.

[98] B. Caughey, R.E. Race, Potent inhibition of scrapie-associated PrP accumulation by congo red, J Neurochem 59 (1992) 768-771.

[99] B. Caughey, G.J. Raymond, Sulfated polyanion inhibition of scrapie-associated PrP accumulation in cultured cells, J Virol 67 (1993) 643-650.

[100] M. Nakajima, T. Yamada, T. Kusuhara, H. Furukawa, M. Takahashi, A. Yamauchi, Y. Kataoka, Results of quinacrine administration to patients with Creutzfeldt-Jakob disease, Dement Geriatr Cogn Disord 17 (2004) 158-163.

[101] D. Peretz, R.A. Williamson, K. Kaneko, J. Vergara, E. Leclerc, G. Schmitt-Ulms, I.R. Mehlhorn, G. Legname, M.R. Wormald, P.M. Rudd, R.A. Dwek, D.R. Burton, S.B. Prusiner, Antibodies inhibit prion propagation and clear cell cultures of prion infectivity, Nature 412 (2001) 739-743.

[102] C. Feraudet, N. Morel, S. Simon, H. Volland, Y. Frobert, C. Creminon, D. Vilette, S. Lehmann, J. Grassi, Screening of 145 anti-PrP monoclonal antibodies for their capacity to inhibit PrPSc replication in infected cells, J Biol Chem 280 (2005) 11247-11258.

[103] A.R. White, P. Enever, M. Tayebi, R. Mushens, J. Linehan, S. Brandner, D. Anstee, J. Collinge, S. Hawke, Monoclonal antibodies inhibit prion replication and delay the development of prion disease, Nature 422 (2003) 80-83.

[104] F.L. Heppner, C. Musahl, I. Arrighi, M.A. Klein, T. Rulicke, B. Oesch, R.M. Zinkernagel, U. Kalinke, A. Aguzzi, Prevention of scrapie pathogenesis by transgenic expression of anti-prion protein antibodies, Science 294 (2001) 178-182.

[105] P. Meier, N. Genoud, M. Prinz, M. Maissen, T. Rulicke, A. Zurbriggen, A.J. Raeber, A. Aguzzi, Soluble dimeric prion protein binds PrP(Sc) in vivo and antagonizes prion disease, Cell 113 (2003) 49-60.

[106] G.R. Mallucci, M.D. White, M. Farmer, A. Dickinson, H. Khatun, A.D. Powell, S. Brandner, J.G. Jefferys, J. Collinge, Targeting cellular prion protein reverses early

212

cognitive deficits and neurophysiological dysfunction in prion-infected mice, Neuron 53 (2007) 325-335.

[107] H. Bueler, A. Aguzzi, A. Sailer, R.A. Greiner, P. Autenried, M. Aguet, C. Weissmann, Mice devoid of PrP are resistant to scrapie, Cell 73 (1993) 1339-1347.

[108] A. Sailer, H. Bueler, M. Fischer, A. Aguzzi, C. Weissmann, No propagation of prions in mice devoid of PrP, Cell 77 (1994) 967-968.

[109] P. Saa, J. Castilla, C. Soto, Ultra-efficient replication of infectious prions by automated protein misfolding cyclic amplification, J Biol Chem 281 (2006) 35245-35252.

[110] G.C. Telling, M. Scott, J. Mastrianni, R. Gabizon, M. Torchia, F.E. Cohen, S.J. DeArmond, S.B. Prusiner, Prion propagation in mice expressing human and chimeric PrP transgenes implicates the interaction of cellular PrP with another protein, Cell 83 (1995) 79-90.

[111] K. Kaneko, L. Zulianello, M. Scott, C.M. Cooper, A.C. Wallace, T.L. James, F.E. Cohen, S.B. Prusiner, Evidence for protein X binding to a discontinuous epitope on the cellular prion protein during scrapie prion propagation, Proc Natl Acad Sci U S A 94 (1997) 10069-10074.

[112] C. Leucht, S. Simoneau, C. Rey, K. Vana, R. Rieger, C.I. Lasmezas, S. Weiss, The 37 kDa/67 kDa laminin receptor is required for PrP(Sc) propagation in scrapie-infected neuronal cells, EMBO Rep 4 (2003) 290-295.

[113] N.R. Deleault, J.C. Geoghegan, K. Nishina, R. Kascsak, R.A. Williamson, S. Supattapone, Protease-resistant prion protein amplification reconstituted with partially purified substrates and synthetic polyanions, J Biol Chem 280 (2005) 26873-26879.

[114] N.R. Deleault, R.W. Lucassen, S. Supattapone, RNA molecules stimulate prion protein conversion, Nature 425 (2003) 717-720.

[115] S. Supattapone, Prion protein conversion in vitro, J Mol Med 82 (2004) 348-356.

[116] C. Wong, L.W. Xiong, M. Horiuchi, L. Raymond, K. Wehrly, B. Chesebro, B. Caughey, Sulfated glycans and elevated temperature stimulate PrP(Sc)-dependent cell-free formation of protease-resistant prion protein, Embo J 20 (2001) 377-386.

[117] N.R. Deleault, B.T. Harris, J.R. Rees, S. Supattapone, From the Cover: Formation of native prions from minimal components in vitro, Proc Natl Acad Sci U S A 104 (2007) 9741-9746.

213

[118] S. Supattapone, T. Muramoto, G. Legname, I. Mehlhorn, F.E. Cohen, S.J. DeArmond, S.B. Prusiner, M.R. Scott, Identification of two prion protein regions that modify scrapie incubation time, J Virol 75 (2001) 1408-1413.

[119] E. Flechsig, D. Shmerling, I. Hegyi, A.J. Raeber, M. Fischer, A. Cozzio, C. von Mering, A. Aguzzi, C. Weissmann, Prion protein devoid of the octapeptide repeat region restores susceptibility to scrapie in PrP knockout mice, Neuron 27 (2000) 399-408.

[120] T. Muramoto, S.J. DeArmond, M. Scott, G.C. Telling, F.E. Cohen, S.B. Prusiner, Heritable disorder resembling neuronal storage disease in mice expressing prion protein with deletion of an alpha-helix, Nat Med 3 (1997) 750-755.

[121] E. Flechsig, J.C. Manson, R. Barron, A. Aguzzi, C. Weissmann, Knockouts, Knockins, Transgenics, and Transplants in Prion Research, in: S.B. Prusiner (Ed.), Prion Biology and Diseases, 2nd Edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 2004, pp. 373-434.

[122] S. Supattapone, P. Bosque, T. Muramoto, H. Wille, C. Aagaard, D. Peretz, H.O. Nguyen, C. Heinrich, M. Torchia, J. Safar, F.E. Cohen, S.J. DeArmond, S.B. Prusiner, M. Scott, Prion protein of 106 residues creates an artifical transmission barrier for prion replication in transgenic mice, Cell 96 (1999) 869-878.

[123] E.M. Norstrom, J.A. Mastrianni, The AGAAAAGA palindrome in PrP is required to generate a productive PrPSc-PrPC complex that leads to prion propagation, J Biol Chem 280 (2005) 27236-27243.

[124] N.J. Cobb, F.D. Sonnichsen, H. McHaourab, W.K. Surewicz, Molecular architecture of human prion protein amyloid: a parallel, in-register beta-structure, Proc Natl Acad Sci U S A 104 (2007) 18946-18951.

[125] I.H. Pattison, Experiments with scrapie with special reference to the nature of the agent and the pathology of the disease, in: D.C. Gajdusek (Ed.), Slow, latent and temperate virus infections, U.S. Government Printing Office, Washington, D.C., 1965, pp. 249-257.

[126] S.B. Prusiner, M. Scott, D. Foster, K.M. Pan, D. Groth, C. Mirenda, M. Torchia, S.L. Yang, D. Serban, G.A. Carlson, et al., Transgenetic studies implicate interactions between homologous PrP isoforms in scrapie prion replication, Cell 63 (1990) 673-686.

[127] R. Nonno, M.A. Di Bari, F. Cardone, G. Vaccari, P. Fazzi, G. Dell'Omo, C. Cartoni, L. Ingrosso, A. Boyle, R. Galeno, M. Sbriccoli, H.P. Lipp, M. Bruce, M. Pocchiari, U. Agrimi, Efficient Transmission and Characterization of Creutzfeld-Jakob Disease Strains in Bank Voles, PLoS Pathog 2 (2006) e12.

214

[128] D. Westaway, P.A. Goodman, C.A. Mirenda, M.P. McKinley, G.A. Carlson, S.B. Prusiner, Distinct prion proteins in short and long scrapie incubation period mice, Cell 51 (1987) 651-662.

[129] D. Westaway, C.A. Mirenda, D. Foster, Y. Zebarjadian, M. Scott, M. Torchia, S.L. Yang, H. Serban, S.J. DeArmond, C. Ebeling, et al., Paradoxical shortening of scrapie incubation times by expression of prion protein transgenes derived from long incubation period mice, Neuron 7 (1991) 59-68.

[130] R.M. Barron, H. Baybutt, N.L. Tuzi, J. McCormack, D. King, R.C. Moore, D.W. Melton, J.C. Manson, Polymorphisms at codons 108 and 189 in murine PrP play distinct roles in the control of scrapie incubation time, J Gen Virol 86 (2005) 859-868.

[131] R.C. Moore, J. Hope, P.A. McBride, I. McConnell, J. Selfridge, D.W. Melton, J.C. Manson, Mice with gene targetted prion protein alterations show that Prnp, Sinc and Prni are congruent, Nat Genet 18 (1998) 118-125.

[132] D. Westaway, V. Zuliani, C.M. Cooper, M. Da Costa, S. Neuman, A.L. Jenny, L. Detwiler, S.B. Prusiner, Homozygosity for prion protein alleles encoding glutamine-171 renders sheep susceptible to natural scrapie, Genes Dev 8 (1994) 959-969.

[133] S. Shibuya, J. Higuchi, R.W. Shin, J. Tateishi, T. Kitamoto, Codon 219 Lys allele of PRNP is not found in sporadic Creutzfeldt-Jakob disease, Ann Neurol 43 (1998) 826-828.

[134] P.G. Smith, S.N. Cousens, J.N. d' Huillard Aignaux, H.J. Ward, R.G. Will, The epidemiology of variant Creutzfeldt-Jakob disease, Curr Top Microbiol Immunol 284 (2004) 161-191.

[135] I.H. Pattison, G.C. Millson, Scrapie produced experimentally in goats with special reference to the clinical syndrome, Journal of comparative pathology 71 (1961) 101-109.

[136] A.G. Dickinson, V.M. Meikle, H. Fraser, Identification of a gene which controls the incubation period of some strains of scrapie agent in mice, Journal of comparative pathology 78 (1968) 293-299.

[137] S.J. DeArmond, S.L. Yang, A. Lee, R. Bowler, A. Taraboulos, D. Groth, S.B. Prusiner, Three scrapie prion isolates exhibit different accumulation patterns of the prion protein scrapie isoform, Proc Natl Acad Sci U S A 90 (1993) 6449-6453.

[138] R.A. Bessen, R.F. Marsh, Distinct PrP properties suggest the molecular basis of strain variation in transmissible mink encephalopathy, J Virol 68 (1994) 7859-7868.

215

[139] R.A. Bessen, D.A. Kocisko, G.J. Raymond, S. Nandan, P.T. Lansbury, B. Caughey, Non-genetic propagation of strain-specific properties of scrapie prion protein, Nature 375 (1995) 698-700.

[140] G.C. Telling, P. Parchi, S.J. DeArmond, P. Cortelli, P. Montagna, R. Gabizon, J. Mastrianni, E. Lugaresi, P. Gambetti, S.B. Prusiner, Evidence for the conformation of the pathologic isoform of the prion protein enciphering and propagating prion diversity, Science 274 (1996) 2079-2082.

[141] C.J. Sigurdson, K.P. Nilsson, S. Hornemann, G. Manco, M. Polymenidou, P. Schwarz, M. Leclerc, P. Hammarstrom, K. Wuthrich, A. Aguzzi, Prion strain discrimination using luminescent conjugated polymers, Nature methods 4 (2007) 1023-1030.

[142] G. Legname, H.O. Nguyen, I.V. Baskakov, F.E. Cohen, S.J. Dearmond, S.B. Prusiner, Strain-specified characteristics of mouse synthetic prions, Proc Natl Acad Sci U S A 102 (2005) 2168-2173.

[143] M.E. Bruce, A.G. Dickinson, Biological evidence that scrapie agent has an independent genome, J Gen Virol 68 ( Pt 1) (1987) 79-89.

[144] J. Safar, H. Wille, V. Itri, D. Groth, H. Serban, M. Torchia, F.E. Cohen, S.B. Prusiner, Eight prion strains have PrP(Sc) molecules with different conformations, Nat Med 4 (1998) 1157-1165.

[145] D.L. Vanik, K.A. Surewicz, W.K. Surewicz, Molecular basis of barriers for interspecies transmissibility of mammalian prions, Mol Cell 14 (2004) 139-145.

[146] E.M. Jones, W.K. Surewicz, Fibril conformation as the basis of species- and strain-dependent seeding specificity of mammalian prion amyloids, Cell 121 (2005) 63-72.

[147] G. Legname, H.O. Nguyen, D. Peretz, F.E. Cohen, S.J. DeArmond, S.B. Prusiner, Continuum of prion protein structures enciphers a multitude of prion isolate-specified phenotypes, Proc Natl Acad Sci U S A 103 (2006) 19105-19110.

[148] M. Tanaka, P. Chien, N. Naber, R. Cooke, J.S. Weissman, Conformational variations in an infectious protein determine prion strain differences, Nature 428 (2004) 323-328.

[149] C.Y. King, R. Diaz-Avalos, Protein-only transmission of three yeast prion strains, Nature 428 (2004) 319-323.

[150] B.H. Toyama, M.J. Kelly, J.D. Gross, J.S. Weissman, The structural basis of yeast prion strain variants, Nature 449 (2007) 233-237.

[151] E. Kubler, B. Oesch, A.J. Raeber, Diagnosis of prion diseases, Br Med Bull 66 (2003) 267-279.

216

[152] M. Haritani, Y.I. Spencer, G.A. Wells, Hydrated autoclave pretreatment enhancement of prion protein immunoreactivity in formalin-fixed bovine spongiform encephalopathy-affected brain, Acta Neuropathol (Berl) 87 (1994) 86-90.

[153] J.G. Safar, M. Scott, J. Monaghan, C. Deering, S. Didorenko, J. Vergara, H. Ball, G. Legname, E. Leclerc, L. Solforosi, H. Serban, D. Groth, D.R. Burton, S.B. Prusiner, R.A. Williamson, Measuring prions causing bovine spongiform encephalopathy or chronic wasting disease by immunoassays and transgenic mice, Nat Biotechnol 20 (2002) 1147-1150.

[154] J.G. Safar, M.D. Geschwind, C. Deering, S. Didorenko, M. Sattavat, H. Sanchez, A. Serban, M. Vey, H. Baron, K. Giles, B.L. Miller, S.J. Dearmond, S.B. Prusiner, Diagnosis of human prion disease, Proc Natl Acad Sci U S A 102 (2005) 3501-3506.

[155] C. Korth, B. Stierli, P. Streit, M. Moser, O. Schaller, R. Fischer, W. Schulz-Schaeffer, H. Kretzschmar, A. Raeber, U. Braun, F. Ehrensperger, S. Hornemann, R. Glockshuber, R. Riek, M. Billeter, K. Wuthrich, B. Oesch, Prion (PrPSc)-specific epitope defined by a monoclonal antibody, Nature 390 (1997) 74-77.

[156] E. Paramithiotis, M. Pinard, T. Lawton, S. LaBoissiere, V.L. Leathers, W.Q. Zou, L.A. Estey, J. Lamontagne, M.T. Lehto, L.H. Kondejewski, G.P. Francoeur, M. Papadopoulos, A. Haghighat, S.J. Spatz, M. Head, R. Will, J. Ironside, K. O'Rourke, Q. Tonelli, H.C. Ledebur, A. Chakrabartty, N.R. Cashman, A prion protein epitope selective for the pathologically misfolded conformation, Nat Med 9 (2003) 893-899.

[157] G. Moroncini, N. Kanu, L. Solforosi, G. Abalos, G.C. Telling, M. Head, J. Ironside, J.P. Brockes, D.R. Burton, R.A. Williamson, Motif-grafted antibodies containing the replicative interface of cellular PrP are specific for PrPSc, Proc Natl Acad Sci U S A 101 (2004) 10404-10409.

[158] V. Curin Serbec, M. Bresjanac, M. Popovic, K. Pretnar Hartman, V. Galvani, R. Rupreht, M. Cernilec, T. Vranac, I. Hafner, R. Jerala, Monoclonal antibody against a peptide of human prion protein discriminates between Creutzfeldt-Jacob's disease-affected and normal brain tissue, J Biol Chem 279 (2004) 3694-3698.

[159] D. Gavier-Widen, M.J. Stack, T. Baron, A. Balachandran, M. Simmons, Diagnosis of transmissible spongiform encephalopathies in animals: a review, J Vet Diagn Invest 17 (2005) 509-527.

[160] R.L. Chandler, Encephalopathy in mice produced by inoculation with scrapie brain material, Lancet 1 (1961) 1378-1379.

[161] M.R. Scott, J. Safar, G. Telling, O. Nguyen, D. Groth, M. Torchia, R. Koehler, P. Tremblay, D. Walther, F.E. Cohen, S.J. DeArmond, S.B. Prusiner, Identification

217

of a prion protein epitope modulating transmission of bovine spongiform encephalopathy prions to transgenic mice, Proc Natl Acad Sci U S A 94 (1997) 14279-14284.

[162] C. Crozet, F. Flamant, A. Bencsik, D. Aubert, J. Samarut, T. Baron, Efficient transmission of two different sheep scrapie isolates in transgenic mice expressing the ovine PrP gene, J Virol 75 (2001) 5328-5334.

[163] E.A. Asante, J.M. Linehan, M. Desbruslais, S. Joiner, I. Gowland, A.L. Wood, J. Welch, A.F. Hill, S.E. Lloyd, J.D. Wadsworth, J. Collinge, BSE prions propagate as either variant CJD-like or sporadic CJD-like prion strains in transgenic mice expressing human prion protein, Embo J 21 (2002) 6358-6366.

[164] S.R. Browning, G.L. Mason, T. Seward, M. Green, G.A. Eliason, C. Mathiason, M.W. Miller, E.S. Williams, E. Hoover, G.C. Telling, Transmission of prions from mule deer and elk with chronic wasting disease to transgenic mice expressing cervid PrP, J Virol 78 (2004) 13345-13350.

[165] P.C. Klohn, L. Stoltze, E. Flechsig, M. Enari, C. Weissmann, A quantitative, highly sensitive cell-based infectivity assay for mouse scrapie prions, Proc Natl Acad Sci U S A 100 (2003) 11666-11671.

[166] G.P. Saborio, B. Permanne, C. Soto, Sensitive detection of pathological prion protein by cyclic amplification of protein misfolding, Nature 411 (2001) 810-813.

[167] C. Soto, L. Anderes, S. Suardi, F. Cardone, J. Castilla, M.J. Frossard, S. Peano, P. Saa, L. Limido, M. Carbonatto, J. Ironside, J.M. Torres, M. Pocchiari, F. Tagliavini, Pre-symptomatic detection of prions by cyclic amplification of protein misfolding, FEBS Lett 579 (2005) 638-642.

[168] J. Castilla, P. Saa, C. Soto, Detection of prions in blood, Nat Med 11 (2005) 982-985.

[169] R.K. Giri, R. Young, R. Pitstick, S.J. DeArmond, S.B. Prusiner, G.A. Carlson, Prion infection of mouse neurospheres, Proc Natl Acad Sci U S A 103 (2006) 3875-3880.

[170] J. Falsig, C. Julius, I. Margalith, P. Schwarz, F.L. Heppner, A. Aguzzi, A versatile prion replication assay in organotypic brain slices, Nat Neurosci (2007).

[171] S.P. Mahal, C.A. Baker, C.A. Demczyk, E.W. Smith, C. Julius, C. Weissmann, Prion strain discrimination in cell culture: The cell panel assay, Proc Natl Acad Sci U S A (2007).

[172] P. Weber, A. Giese, N. Piening, G. Mitteregger, A. Thomzig, M. Beekes, H.A. Kretzschmar, Cell-free formation of misfolded prion protein with authentic prion infectivity, Proc Natl Acad Sci U S A 103 (2006) 15818-15823.

218

[173] P. Saa, J. Castilla, C. Soto, Cyclic amplification of protein misfolding and aggregation, Methods Mol Biol 299 (2005) 53-65.

[174] P. Saa, J. Castilla, C. Soto, Presymptomatic detection of prions in blood, Science 313 (2006) 92-94.

[175] R. Atarashi, R.A. Moore, V.L. Sim, A.G. Hughson, D.W. Dorward, H.A. Onwubiko, S.A. Priola, B. Caughey, Ultrasensitive detection of scrapie prion protein using seeded conversion of recombinant prion protein, Nature methods 4 (2007) 645-650.

[176] C. Locht, B. Chesebro, R. Race, J.M. Keith, Molecular cloning and complete sequence of prion protein cDNA from mouse brain infected with the scrapie agent, Proc Natl Acad Sci U S A 83 (1986) 6372-6376.

[177] K. Basler, B. Oesch, M. Scott, D. Westaway, M. Walchli, D.F. Groth, M.P. McKinley, S.B. Prusiner, C. Weissmann, Scrapie and cellular PrP isoforms are encoded by the same chromosomal gene, Cell 46 (1986) 417-428.

[178] Y.C. Liao, R.V. Lebo, G.A. Clawson, E.A. Smuckler, Human prion protein cDNA: molecular cloning, chromosomal mapping, and biological implications, Science 233 (1986) 364-367.

[179] H.A. Kretzschmar, L.E. Stowring, D. Westaway, W.H. Stubblebine, S.B. Prusiner, S.J. Dearmond, Molecular cloning of a human prion protein cDNA, DNA 5 (1986) 315-324.

[180] S.B. Prusiner, M.R. Scott, Genetics of prions, Annu Rev Genet 31 (1997) 139-175.

[181] P. Tremblay, E. Bouzamondo-Bernstein, C. Heinrich, S.B. Prusiner, S.J. DeArmond, Developmental expression of PrP in the post-implantation embryo, Brain Res 1139 (2007) 60-67.

[182] M. Horiuchi, N. Yamazaki, T. Ikeda, N. Ishiguro, M. Shinagawa, A cellular form of prion protein (PrPC) exists in many non-neuronal tissues of sheep, J Gen Virol 76 ( Pt 10) (1995) 2583-2587.

[183] S. Huang, J. Liang, M. Zheng, X. Li, M. Wang, P. Wang, D. Vanegas, D. Wu, B. Chakraborty, A.P. Hays, K. Chen, S.G. Chen, S. Booth, M. Cohen, P. Gambetti, Q. Kong, Inducible overexpression of wild-type prion protein in the muscles leads to a primary myopathy in transgenic mice, Proc Natl Acad Sci U S A 104 (2007) 6800-6805.

[184] T. Liu, R. Li, B.S. Wong, D. Liu, T. Pan, R.B. Petersen, P. Gambetti, M.S. Sy, Normal cellular prion protein is preferentially expressed on subpopulations of murine hemopoietic cells, J Immunol 166 (2001) 3733-3742.

219

[185] N.R. Cashman, R. Loertscher, J. Nalbantoglu, I. Shaw, R.J. Kascsak, D.C. Bolton, P.E. Bendheim, Cellular isoform of the scrapie agent protein participates in lymphocyte activation, Cell 61 (1990) 185-192.

[186] H. Bueler, M. Fischer, Y. Lang, H. Bluethmann, H.P. Lipp, S.J. DeArmond, S.B. Prusiner, M. Aguet, C. Weissmann, Normal development and behaviour of mice lacking the neuronal cell-surface PrP protein, Nature 356 (1992) 577-582.

[187] J.C. Manson, A.R. Clarke, M.L. Hooper, L. Aitchison, I. McConnell, J. Hope, 129/Ola mice carrying a null mutation in PrP that abolishes mRNA production are developmentally normal, Mol Neurobiol 8 (1994) 121-127.

[188] S. Sakaguchi, S. Katamine, N. Nishida, R. Moriuchi, K. Shigematsu, T. Sugimoto, A. Nakatani, Y. Kataoka, T. Houtani, S. Shirabe, H. Okada, S. Hasegawa, T. Miyamoto, T. Noda, Loss of cerebellar Purkinje cells in aged mice homozygous for a disrupted PrP gene, Nature 380 (1996) 528-531.

[189] D. Rossi, A. Cozzio, E. Flechsig, M.A. Klein, T. Rulicke, A. Aguzzi, C. Weissmann, Onset of ataxia and Purkinje cell loss in PrP null mice inversely correlated with Dpl level in brain, Embo J 20 (2001) 694-702.

[190] R.C. Moore, I.Y. Lee, G.L. Silverman, P.M. Harrison, R. Strome, C. Heinrich, A. Karunaratne, S.H. Pasternak, M.A. Chishti, Y. Liang, P. Mastrangelo, K. Wang, A.F. Smit, S. Katamine, G.A. Carlson, F.E. Cohen, S.B. Prusiner, D.W. Melton, P. Tremblay, L.E. Hood, D. Westaway, Ataxia in prion protein (PrP)-deficient mice is associated with upregulation of the novel PrP-like protein doppel, J Mol Biol 292 (1999) 797-817.

[191] S. Paltrinieri, S. Comazzi, V. Spagnolo, M. Rondena, W. Ponti, F. Ceciliani, Bovine Doppel (Dpl) and prion protein (PrP) expression on lymphoid tissue and circulating leukocytes, J Histochem Cytochem 52 (2004) 1639-1645.

[192] S. Paltrinieri, V. Spagnolo, A. Giordano, D. Gelmetti, S. Comazzi, Bovine prion (PrP) and Doppel (Dpl) proteins expression after in vitro leukocyte activation or Dpl/PrP blocking, J Cell Physiol 208 (2006) 446-450.

[193] K. Peoc'h, C. Guerin, J.P. Brandel, J.M. Launay, J.L. Laplanche, First report of polymorphisms in the prion-like protein gene (PRND): implications for human prion diseases, Neurosci Lett 286 (2000) 144-148.

[194] S. Mead, J. Beck, A. Dickinson, E.M. Fisher, J. Collinge, Examination of the human prion protein-like gene doppel for genetic susceptibility to sporadic and variant Creutzfeldt-Jakob disease, Neurosci Lett 290 (2000) 117-120.

[195] B.H. Jeong, N.H. Kim, E.K. Choi, C. Lee, Y.H. Song, J.I. Kim, R.I. Carp, Y.S. Kim, Polymorphism at 3' UTR +28 of the prion-like protein gene is associated with sporadic Creutzfeldt-Jakob disease, Eur J Hum Genet 13 (2005) 1094-1097.

220

[196] K. Peoc'h, C. Serres, Y. Frobert, C. Martin, S. Lehmann, S. Chasseigneaux, V. Sazdovitch, J. Grassi, P. Jouannet, J.M. Launay, J.L. Laplanche, The human "prion-like" protein Doppel is expressed in both Sertoli cells and spermatozoa, J Biol Chem 277 (2002) 43071-43078.

[197] M. Premzl, L. Sangiorgio, B. Strumbo, J.A. Marshall Graves, T. Simonic, J.E. Gready, Shadoo, a new protein highly conserved from fish to mammals and with similarity to prion protein, Gene 314 (2003) 89-102.

[198] M. Premzl, J.E. Gready, L.S. Jermiin, T. Simonic, J.A. Marshall Graves, Evolution of vertebrate genes related to prion and Shadoo proteins--clues from comparative genomic analysis, Mol Biol Evol 21 (2004) 2210-2231.

[199] E. Makrinou, J. Collinge, M. Antoniou, Genomic characterization of the human prion protein (PrP) gene locus, Mamm Genome 13 (2002) 696-703.

[200] S.H. Choi, I.C. Kim, D.S. Kim, D.W. Kim, S.H. Chae, H.H. Choi, I. Choi, J.S. Yeo, M.N. Song, H.S. Park, Comparative genomic organization of the human and bovine PRNP locus, Genomics 87 (2006) 598-607.

[201] N.S. Rane, J.L. Yonkovich, R.S. Hegde, Protection from cytosolic prion protein toxicity by modulation of protein translocation, Embo J 23 (2004) 4550-4559.

[202] N. Stahl, D.R. Borchelt, K. Hsiao, S.B. Prusiner, Scrapie prion protein contains a phosphatidylinositol glycolipid, Cell 51 (1987) 229-240.

[203] N. Naslavsky, R. Stein, A. Yanai, G. Friedlander, A. Taraboulos, Characterization of detergent-insoluble complexes containing the cellular prion protein and its scrapie isoform, J Biol Chem 272 (1997) 6324-6331.

[204] R. Riek, S. Hornemann, G. Wider, M. Billeter, R. Glockshuber, K. Wuthrich, NMR structure of the mouse prion protein domain PrP(121-321), Nature 382 (1996) 180-182.

[205] D.G. Donne, J.H. Viles, D. Groth, I. Mehlhorn, T.L. James, F.E. Cohen, S.B. Prusiner, P.E. Wright, H.J. Dyson, Structure of the recombinant full-length hamster prion protein PrP(29-231): the N terminus is highly flexible, Proc Natl Acad Sci U S A 94 (1997) 13452-13457.

[206] F. Lopez Garcia, R. Zahn, R. Riek, K. Wuthrich, NMR structure of the bovine prion protein, Proc Natl Acad Sci U S A 97 (2000) 8334-8339.

[207] R. Zahn, A. Liu, T. Luhrs, R. Riek, C. von Schroetter, F. Lopez Garcia, M. Billeter, L. Calzolai, G. Wider, K. Wuthrich, NMR solution structure of the human prion protein, Proc Natl Acad Sci U S A 97 (2000) 145-150.

221

[208] A.D. Gossert, S. Bonjour, D.A. Lysek, F. Fiorito, K. Wuthrich, Prion protein NMR structures of elk and of mouse/elk hybrids, Proc Natl Acad Sci U S A 102 (2005) 646-650.

[209] M.P. Hornshaw, J.R. McDermott, J.M. Candy, Copper binding to the N-terminal tandem repeat regions of mammalian and avian prion protein, Biochem Biophys Res Commun 207 (1995) 621-629.

[210] D.R. Brown, K. Qin, J.W. Herms, A. Madlung, J. Manson, R. Strome, P.E. Fraser, T. Kruck, A. von Bohlen, W. Schulz-Schaeffer, A. Giese, D. Westaway, H. Kretzschmar, The cellular prion protein binds copper in vivo, Nature 390 (1997) 684-687.

[211] P.C. Pauly, D.A. Harris, Copper stimulates endocytosis of the prion protein, J Biol Chem 273 (1998) 33107-33110.

[212] W.S. Perera, N.M. Hooper, Ablation of the metal ion-induced endocytosis of the prion protein by disease-associated mutation of the octarepeat region, Curr Biol 11 (2001) 519-523.

[213] S. Krasemann, I. Zerr, T. Weber, S. Poser, H. Kretzschmar, G. Hunsmann, W. Bodemer, Prion disease associated with a novel nine octapeptide repeat insertion in the PRNP gene, Brain Res Mol Brain Res 34 (1995) 173-176.

[214] R. Chiesa, P. Piccardo, B. Ghetti, D.A. Harris, Neurological illness in transgenic mice expressing a prion protein with an insertional mutation, Neuron 21 (1998) 1339-1351.

[215] R.S. Hegde, J.A. Mastrianni, M.R. Scott, K.A. DeFea, P. Tremblay, M. Torchia, S.J. DeArmond, S.B. Prusiner, V.R. Lingappa, A transmembrane form of the prion protein in neurodegenerative disease, Science 279 (1998) 827-834.

[216] R.S. Hegde, P. Tremblay, D. Groth, S.J. DeArmond, S.B. Prusiner, V.R. Lingappa, Transmissible and genetic prion diseases share a common pathway of neurodegeneration, Nature 402 (1999) 822-826.

[217] D. Shmerling, I. Hegyi, M. Fischer, T. Blattler, S. Brandner, J. Gotz, T. Rulicke, E. Flechsig, A. Cozzio, C. von Mering, C. Hangartner, A. Aguzzi, C. Weissmann, Expression of amino-terminally truncated PrP in the mouse leading to ataxia and specific cerebellar lesions, Cell 93 (1998) 203-214.

[218] C. Sunyach, A. Jen, J. Deng, K.T. Fitzgerald, Y. Frobert, J. Grassi, M.W. McCaffrey, R. Morris, The mechanism of internalization of glycosylphosphatidylinositol-anchored prion protein, Embo J 22 (2003) 3591-3601.

222

[219] N.S. Hachiya, K. Watanabe, M. Yamada, Y. Sakasegawa, K. Kaneko, Anterograde and retrograde intracellular trafficking of fluorescent cellular prion protein, Biochem Biophys Res Commun 315 (2004) 802-807.

[220] S. Gilch, M. Nunziante, A. Ertmer, F. Wopfner, L. Laszlo, H.M. Schatzl, Recognition of lumenal prion protein aggregates by post-ER quality control mechanisms is mediated by the preoctarepeat region of PrP, Traffic 5 (2004) 300-313.

[221] H. Mo, R.C. Moore, F.E. Cohen, D. Westaway, S.B. Prusiner, P.E. Wright, H.J. Dyson, Two different neurodegenerative diseases caused by proteins with similar structures, Proc Natl Acad Sci U S A 98 (2001) 2352-2357.

[222] C. Govaerts, H. Wille, S.B. Prusiner, F.E. Cohen, Evidence for assembly of prions with left-handed beta-helices into trimers, Proc Natl Acad Sci U S A 101 (2004) 8342-8347.

[223] D. Peretz, R.A. Williamson, Y. Matsunaga, H. Serban, C. Pinilla, R.B. Bastidas, R. Rozenshteyn, T.L. James, R.A. Houghten, F.E. Cohen, S.B. Prusiner, D.R. Burton, A conformational transition at the N terminus of the prion protein features in formation of the scrapie isoform, J Mol Biol 273 (1997) 614-622.

[224] X. Lu, P.L. Wintrode, W.K. Surewicz, Beta-sheet core of human prion protein amyloid fibrils as determined by hydrogen/deuterium exchange, Proc Natl Acad Sci U S A 104 (2007) 1510-1515.

[225] H. Wille, M.D. Michelitsch, V. Guenebaut, S. Supattapone, A. Serban, F.E. Cohen, D.A. Agard, S.B. Prusiner, Structural studies of the scrapie prion protein by electron crystallography, Proc Natl Acad Sci U S A 99 (2002) 3563-3568.

[226] H. Wille, C. Govaerts, A. Borovinskiy, D. Latawiec, K.H. Downing, F.E. Cohen, S.B. Prusiner, Electron crystallography of the scrapie prion protein complexed with heavy metals, Arch Biochem Biophys (2007).

[227] M.L. DeMarco, V. Daggett, From conversion to aggregation: protofibril formation of the prion protein, Proc Natl Acad Sci U S A 101 (2004) 2293-2298.

[228] T. Luhrs, R. Riek, P. Guntert, K. Wuthrich, NMR structure of the human doppel protein, J Mol Biol 326 (2003) 1549-1557.

[229] G.L. Silverman, K. Qin, R.C. Moore, Y. Yang, P. Mastrangelo, P. Tremblay, S.B. Prusiner, F.E. Cohen, D. Westaway, Doppel is an N-glycosylated, glycosylphosphatidylinositol-anchored protein. Expression in testis and ectopic production in the brains of Prnp(0/0) mice predisposed to Purkinje cell loss, J Biol Chem 275 (2000) 26834-26841.

[230] Y. Shaked, N. Hijazi, R. Gabizon, Doppel and PrP(C) do not share the same membrane microenvironment, FEBS Lett 530 (2002) 85-88.

223

[231] M. Miesbauer, T. Bamme, C. Riemer, B. Oidtmann, K.F. Winklhofer, M. Baier, J. Tatzelt, Prion protein-related proteins from zebrafish are complex glycosylated and contain a glycosylphosphatidylinositol anchor, Biochem Biophys Res Commun 341 (2006) 218-224.

[232] S.G. Chen, D.B. Teplow, P. Parchi, J.K. Teller, P. Gambetti, L. Autilio-Gambetti, Truncated forms of the human prion protein in normal brain and in prion diseases, J Biol Chem 270 (1995) 19173-19180.

[233] B. Vincent, E. Paitel, P. Saftig, Y. Frobert, D. Hartmann, B. De Strooper, J. Grassi, E. Lopez-Perez, F. Checler, The disintegrins ADAM10 and TACE contribute to the constitutive and phorbol ester-regulated normal cleavage of the cellular prion protein, J Biol Chem 276 (2001) 37743-37746.

[234] M.A. Cisse, C. Sunyach, S. Lefranc-Jullien, R. Postina, B. Vincent, F. Checler, The disintegrin ADAM9 indirectly contributes to the physiological processing of cellular prion by modulating ADAM10 activity, J Biol Chem 280 (2005) 40624-40631.

[235] R. Yadavalli, R.P. Guttmann, T. Seward, A.P. Centers, R.A. Williamson, G.C. Telling, Calpain-dependent endoproteolytic cleavage of PrPSc modulates scrapie prion propagation, J Biol Chem 279 (2004) 21948-21956.

[236] G. Forloni, N. Angeretti, R. Chiesa, E. Monzani, M. Salmona, O. Bugiani, F. Tagliavini, Neurotoxicity of a prion protein fragment, Nature 362 (1993) 543-546.

[237] C. Sunyach, M.A. Cisse, C.A. da Costa, B. Vincent, F. Checler, The C-terminal products of cellular prion protein processing, C1 and C2, exert distinct influence on p53-dependent staurosporine-induced caspase-3 activation, J Biol Chem 282 (2007) 1956-1963.

[238] B. Hay, R.A. Barry, I. Lieberburg, S.B. Prusiner, V.R. Lingappa, Biogenesis and transmembrane orientation of the cellular isoform of the scrapie prion protein [published errratum appears in Mol Cell Biol 1987 May;7(5):2035], Mol Cell Biol 7 (1987) 914-920.

[239] C.D. Lopez, C.S. Yost, S.B. Prusiner, R.M. Myers, V.R. Lingappa, Unusual topogenic sequence directs prion protein biogenesis, Science 248 (1990) 226-229.

[240] C.S. Yost, C.D. Lopez, S.B. Prusiner, R.M. Myers, V.R. Lingappa, Non-hydrophobic extracytoplasmic determinant of stop transfer in the prion protein, Nature 343 (1990) 669-672.

[241] L. Westergard, H.M. Christensen, D.A. Harris, The cellular prion protein (PrP(C)): its physiological function and role in disease, Biochim Biophys Acta 1772 (2007) 629-644.

224

[242] J. Collinge, M.A. Whittington, K.C. Sidle, C.J. Smith, M.S. Palmer, A.R. Clarke, J.G. Jefferys, Prion protein is necessary for normal synaptic function, Nature 370 (1994) 295-297.

[243] M.A. Whittington, K.C. Sidle, I. Gowland, J. Meads, A.F. Hill, M.S. Palmer, J.G. Jefferys, J. Collinge, Rescue of neurophysiological phenotype seen in PrP null mice by transgene encoding human prion protein, Nat Genet 9 (1995) 197-201.

[244] P.M. Lledo, P. Tremblay, S.J. DeArmond, S.B. Prusiner, R.A. Nicoll, Mice deficient for prion protein exhibit normal neuronal excitability and synaptic transmission in the hippocampus, Proc Natl Acad Sci U S A 93 (1996) 2403-2407.

[245] J.W. Herms, H.A. Kretzchmar, S. Titz, B.U. Keller, Patch-clamp analysis of synaptic transmission to cerebellar purkinje cells of prion protein knockout mice, Eur J Neurosci 7 (1995) 2508-2512.

[246] S.B. Colling, J. Collinge, J.G. Jefferys, Hippocampal slices from prion protein null mice: disrupted Ca(2+)-activated K+ currents, Neurosci Lett 209 (1996) 49-52.

[247] S.B. Colling, M. Khana, J. Collinge, J.G. Jefferys, Mossy fibre reorganization in the hippocampus of prion protein null mice, Brain Res 755 (1997) 28-35.

[248] G.R. Mallucci, S. Ratte, E.A. Asante, J. Linehan, I. Gowland, J.G. Jefferys, J. Collinge, Post-natal knockout of prion protein alters hippocampal CA1 properties, but does not result in neurodegeneration, Embo J 21 (2002) 202-210.

[249] I. Tobler, S.E. Gaus, T. Deboer, P. Achermann, M. Fischer, T. Rulicke, M. Moser, B. Oesch, P.A. McBride, J.C. Manson, Altered circadian activity rhythms and sleep in mice devoid of prion protein, Nature 380 (1996) 639-642.

[250] H.P. Landolt, M. Glatzel, T. Blattler, P. Achermann, C. Roth, J. Mathis, J. Weis, I. Tobler, A. Aguzzi, C.L. Bassetti, Sleep-wake disturbances in sporadic Creutzfeldt-Jakob disease, Neurology 66 (2006) 1418-1424.

[251] F.R. Cagampang, S.A. Whatley, A.L. Mitchell, J.F. Powell, I.C. Campbell, C.W. Coen, Circadian regulation of prion protein messenger RNA in the rat forebrain: a widespread and synchronous rhythm, Neuroscience 91 (1999) 1201-1204.

[252] J.R. Criado, M. Sanchez-Alavez, B. Conti, J.L. Giacchino, D.N. Wills, S.J. Henriksen, R. Race, J.C. Manson, B. Chesebro, M.B. Oldstone, Mice devoid of prion protein have cognitive deficits that are rescued by reconstitution of PrP in neurons, Neurobiol Dis 19 (2005) 255-265.

[253] A.S. Coitinho, R. Roesler, V.R. Martins, R.R. Brentani, I. Izquierdo, Cellular prion protein ablation impairs behavior as a function of age, Neuroreport 14 (2003) 1375-1379.

225

[254] R. Roesler, R. Walz, J. Quevedo, F. de-Paris, S.M. Zanata, E. Graner, I. Izquierdo, V.R. Martins, R.R. Brentani, Normal inhibitory avoidance learning and anxiety, but increased locomotor activity in mice devoid of PrP(C), Brain Res Mol Brain Res 71 (1999) 349-353.

[255] H.P. Lipp, M. Stagliar-Bozicevic, M. Fischer, D.P. Wolfer, A 2-year longitudinal study of swimming navigation in mice devoid of the prion protein: no evidence for neurological anomalies or spatial learning impairments, Behav Brain Res 95 (1998) 47-54.

[256] D.R. Brown, W.J. Schulz-Schaeffer, B. Schmidt, H.A. Kretzschmar, Prion protein-deficient cells show altered response to oxidative stress due to decreased SOD-1 activity, Exp Neurol 146 (1997) 104-112.

[257] B.S. Wong, T. Pan, T. Liu, R. Li, P. Gambetti, M.S. Sy, Differential contribution of superoxide dismutase activity by prion protein in vivo, Biochem Biophys Res Commun 273 (2000) 136-139.

[258] D.R. Brown, A. Besinger, Prion protein expression and superoxide dismutase activity, Biochem J 334 ( Pt 2) (1998) 423-429.

[259] D.R. Brown, B.S. Wong, F. Hafiz, C. Clive, S.J. Haswell, I.M. Jones, Normal prion protein has an activity like that of superoxide dismutase, Biochem J 344 Pt 1 (1999) 1-5.

[260] S. Jones, M. Batchelor, D. Bhelt, A.R. Clarke, J. Collinge, G.S. Jackson, Recombinant prion protein does not possess SOD-1 activity, Biochem J 392 (2005) 309-312.

[261] D.J. Waggoner, B. Drisaldi, T.B. Bartnikas, R.L. Casareno, J.R. Prohaska, J.D. Gitlin, D.A. Harris, Brain copper content and cuproenzyme activity do not vary with prion protein expression level, J Biol Chem 275 (2000) 7455-7458.

[262] G. Hutter, F.L. Heppner, A. Aguzzi, No superoxide dismutase activity of cellular prion protein in vivo, Biol Chem 384 (2003) 1279-1285.

[263] A.D. Steele, J.G. Emsley, P.H. Ozdinler, S. Lindquist, J.D. Macklis, Prion protein (PrPc) positively regulates neural precursor proliferation during developmental and adult mammalian neurogenesis, Proc Natl Acad Sci U S A 103 (2006) 3416-3421.

[264] C.C. Zhang, A.D. Steele, S. Lindquist, H.F. Lodish, Prion protein is expressed on long-term repopulating hematopoietic stem cells and is important for their self-renewal, Proc Natl Acad Sci U S A 103 (2006) 2184-2189.

[265] E.T. Parkin, N.T. Watt, I. Hussain, E.A. Eckman, C.B. Eckman, J.C. Manson, H.N. Baybutt, A.J. Turner, N.M. Hooper, Cellular prion protein regulates beta-

226

secretase cleavage of the Alzheimer's amyloid precursor protein, Proc Natl Acad Sci U S A 104 (2007) 11062-11067.

[266] G. Schmitt-Ulms, K. Hansen, J. Liu, C. Cowdrey, J. Yang, S.J. DeArmond, F.E. Cohen, S.B. Prusiner, M.A. Baldwin, Time-controlled transcardiac perfusion cross-linking for the study of protein interactions in complex tissues, Nat Biotechnol 22 (2004) 724-731.

[267] Y. Bai, K. Markham, F. Chen, R. Weerasekera, J. Watts, P. Horne, Y. Wakutani, P.M. Mathews, P.E. Fraser, D. Westaway, P. St George-Hyslop, G. Schmitt-Ulms, The in vivo brain interactome of the amyloid precursor protein, Mol Cell Proteomics (2007).

[268] L.B. Chiarini, A.R. Freitas, S.M. Zanata, R.R. Brentani, V.R. Martins, R. Linden, Cellular prion protein transduces neuroprotective signals, Embo J 21 (2002) 3317-3326.

[269] Y. Bounhar, Y. Zhang, C.G. Goodyer, A. LeBlanc, Prion protein protects human neurons against Bax-mediated apoptosis, J Biol Chem 276 (2001) 39145-39149.

[270] A. Li, D.A. Harris, Mammalian prion protein suppresses Bax-induced cell death in yeast, J Biol Chem 280 (2005) 17430-17434.

[271] B. Drisaldi, J. Coomaraswamy, P. Mastrangelo, B. Strome, J. Yang, J.C. Watts, M.A. Chishti, M. Marvi, O. Windl, R. Ahrens, F. Major, M.S. Sy, H. Kretzschmar, P.E. Fraser, H.T. Mount, D. Westaway, Genetic mapping of activity determinants within cellular prion proteins: N-terminal modules in PrPC offset pro-apoptotic activity of the Doppel helix B/B' region, J Biol Chem 279 (2004) 55443-55454.

[272] N.F. McLennan, P.M. Brennan, A. McNeill, I. Davies, A. Fotheringham, K.A. Rennison, D. Ritchie, F. Brannan, M.W. Head, J.W. Ironside, A. Williams, J.E. Bell, Prion protein accumulation and neuroprotection in hypoxic brain damage, Am J Pathol 165 (2004) 227-235.

[273] J. Weise, O. Crome, R. Sandau, W. Schulz-Schaeffer, M. Bahr, I. Zerr, Upregulation of cellular prion protein (PrPc) after focal cerebral ischemia and influence of lesion severity, Neurosci Lett 372 (2004) 146-150.

[274] A. Spudich, R. Frigg, E. Kilic, U. Kilic, B. Oesch, A. Raeber, C.L. Bassetti, D.M. Hermann, Aggravation of ischemic brain injury by prion protein deficiency: role of ERK-1/-2 and STAT-1, Neurobiol Dis 20 (2005) 442-449.

[275] W.C. Shyu, S.Z. Lin, M.F. Chiang, D.C. Ding, K.W. Li, S.F. Chen, H.I. Yang, H. Li, Overexpression of PrPC by adenovirus-mediated gene targeting reduces ischemic injury in a stroke rat model, J Neurosci 25 (2005) 8967-8977.

[276] L. Solforosi, J.R. Criado, D.B. McGavern, S. Wirz, M. Sanchez-Alavez, S. Sugama, L.A. DeGiorgio, B.T. Volpe, E. Wiseman, G. Abalos, E. Masliah, D.

227

Gilden, M.B. Oldstone, B. Conti, R.A. Williamson, Cross-linking cellular prion protein triggers neuronal apoptosis in vivo, Science 303 (2004) 1514-1516.

[277] E. Paitel, C. Alves da Costa, D. Vilette, J. Grassi, F. Checler, Overexpression of PrPc triggers caspase 3 activation: potentiation by proteasome inhibitors and blockade by anti-PrP antibodies, J Neurochem 83 (2002) 1208-1214.

[278] E. Paitel, R. Fahraeus, F. Checler, Cellular prion protein sensitizes neurons to apoptotic stimuli through Mdm2-regulated and p53-dependent caspase 3-like activation, J Biol Chem 278 (2003) 10061-10066.

[279] F. Baumann, M. Tolnay, C. Brabeck, J. Pahnke, U. Kloz, H.H. Niemann, M. Heikenwalder, T. Rulicke, A. Burkle, A. Aguzzi, Lethal recessive myelin toxicity of prion protein lacking its central domain, Embo J 26 (2007) 538-547.

[280] A. Li, H.M. Christensen, L.R. Stewart, K.A. Roth, R. Chiesa, D.A. Harris, Neonatal lethality in transgenic mice expressing prion protein with a deletion of residues 105-125, Embo J 26 (2007) 548-558.

[281] K. Peoc'h, H. Volland, A. De Gassart, P. Beaudry, V. Sazdovitch, M.C. Sorgato, C. Creminon, J.L. Laplanche, S. Lehmann, Prion-like protein Doppel expression is not modified in scrapie-infected cells and in the brains of patients with Creutzfeldt-Jakob disease, FEBS Lett 536 (2003) 61-65.

[282] M. Rondena, F. Ceciliani, S. Comazzi, V. Pocacqua, C. Bazzocchi, C. Luvoni, S. Chigioni, S. Paltrinieri, Identification of bovine doppel protein in testis, ovary and ejaculated spermatozoa, Theriogenology 63 (2005) 1195-1206.

[283] C. Serres, K. Peoc'h, A.M. Courtot, C. Lesaffre, P. Jouannet, J.L. Laplanche, Spatio-developmental distribution of the prion-like protein doppel in Mammalian testis: a comparative analysis focusing on its presence in the acrosome of spermatids, Biol Reprod 74 (2006) 816-823.

[284] A. Behrens, N. Genoud, H. Naumann, T. Rulicke, F. Janett, F.L. Heppner, B. Ledermann, A. Aguzzi, Absence of the prion protein homologue Doppel causes male sterility, Embo J 21 (2002) 3652-3658.

[285] D. Paisley, S. Banks, J. Selfridge, N.F. McLennan, A.M. Ritchie, C. McEwan, D.S. Irvine, P.T. Saunders, J.C. Manson, D.W. Melton, Male infertility and DNA damage in Doppel knockout and prion protein/Doppel double-knockout mice, Am J Pathol 164 (2004) 2279-2288.

[286] R. Essalmani, S. Taourit, N. Besnard, J.L. Vilotte, Sequence determination and expression of the ovine doppel-encoding gene in transgenic mice, Gene 285 (2002) 287-290.

228

[287] K. Qin, J. Coomaraswamy, P. Mastrangelo, Y. Yang, S. Lugowski, C. Petromilli, S.B. Prusiner, P.E. Fraser, J.M. Goldberg, A. Chakrabartty, D. Westaway, The PrP-like protein Doppel binds copper, J Biol Chem 278 (2003) 8888-8896.

[288] G.M. Cereghetti, A. Negro, E. Vinck, M.L. Massimino, M.C. Sorgato, S. Van Doorslaer, Copper(II) binding to the human Doppel protein may mark its functional diversity from the prion protein, J Biol Chem 279 (2004) 36497-36503.

[289] R.C. Moore, P. Mastrangelo, E. Bouzamondo, C. Heinrich, G. Legname, S.B. Prusiner, L. Hood, D. Westaway, S.J. DeArmond, P. Tremblay, Doppel-induced cerebellar degeneration in transgenic mice, Proc Natl Acad Sci U S A 98 (2001) 15288-15293.

[290] N.L. Tuzi, E. Gall, D. Melton, J.C. Manson, Expression of doppel in the CNS of mice does not modulate transmissible spongiform encephalopathy disease, J Gen Virol 83 (2002) 705-711.

[291] S.M. Whyte, I.D. Sylvester, S.R. Martin, A.C. Gill, F. Wopfner, H.M. Schatzl, G.G. Dodson, P.M. Bayley, Stability and conformational properties of doppel, a prion-like protein, and its single-disulphide mutant, Biochem J 373 (2003) 485-494.

[292] C. Hundt, S. Weiss, The prion-like protein Doppel fails to interact with itself, the prion protein and the 37 kDa/67 kDa laminin receptor in the yeast two-hybrid system, Biochim Biophys Acta 1689 (2004) 1-5.

[293] B. Oesch, D.B. Teplow, N. Stahl, D. Serban, L.E. Hood, S.B. Prusiner, Identification of cellular proteins binding to the scrapie prion protein, Biochemistry 29 (1990) 5848-5855.

[294] C. Kurschner, J.I. Morgan, The cellular prion protein (PrP) selectively binds to Bcl-2 in the yeast two-hybrid system, Brain Res Mol Brain Res 30 (1995) 165-168.

[295] C. Spielhaupter, H.M. Schatzl, PrPC directly interacts with proteins involved in signaling pathways, J Biol Chem 276 (2001) 44604-44612.

[296] B. Caughey, K. Brown, G.J. Raymond, G.E. Katzenstein, W. Thresher, Binding of the protease-sensitive form of PrP (prion protein) to sulfated glycosaminoglycan and congo red [corrected], J Virol 68 (1994) 2135-2141.

[297] R.G. Warner, C. Hundt, S. Weiss, J.E. Turnbull, Identification of the heparan sulfate binding sites in the cellular prion protein, J Biol Chem 277 (2002) 18421-18430.

[298] J.H. Harmey, D. Doyle, V. Brown, M.S. Rogers, The cellular isoform of the prion protein, PrPc, is associated with caveolae in mouse neuroblastoma (N2a) cells, Biochem Biophys Res Commun 210 (1995) 753-759.

229

[299] G.I. Keshet, O. Bar-Peled, D. Yaffe, U. Nudel, R. Gabizon, The cellular prion protein colocalizes with the dystroglycan complex in the brain, J Neurochem 75 (2000) 1889-1897.

[300] C. Gao, Y.J. Lei, J. Han, Q. Shi, L. Chen, Y. Guo, Y.J. Gao, J.M. Chen, H.Y. Jiang, W. Zhou, X.P. Dong, Recombinant Neural Protein PrP Can Bind with Both Recombinant and Native Apolipoprotein E In Vitro, Acta Biochim Biophys Sin (Shanghai) 38 (2006) 593-601.

[301] M.B. Fischer, C. Roeckl, P. Parizek, H.P. Schwarz, A. Aguzzi, Binding of disease-associated prion protein to plasminogen, Nature 408 (2000) 479-483.

[302] V. Ellis, M. Daniels, R. Misra, D.R. Brown, Plasminogen activation is stimulated by prion protein and regulated in a copper-dependent manner, Biochemistry 41 (2002) 6891-6896.

[303] S. Capellari, S.I. Zaidi, C.B. Urig, G. Perry, M.A. Smith, R.B. Petersen, Prion protein glycosylation is sensitive to redox change, J Biol Chem 274 (1999) 34846-34850.

[304] F. Edenhofer, R. Rieger, M. Famulok, W. Wendler, S. Weiss, E.L. Winnacker, Prion protein PrPc interacts with molecular chaperones of the Hsp60 family, J Virol 70 (1996) 4724-4728.

[305] B.T. Bragason, A. Palsdottir, Interaction of PrP with NRAGE, a protein involved in neuronal apoptosis, Mol Cell Neurosci 29 (2005) 232-244.

[306] A. Azzalin, V. Ferrara, A. Arias, S. Cerri, D. Avella, M.B. Pisu, R. Nano, G. Bernocchi, L. Ferretti, S. Comincini, Interaction between the cellular prion (PrPC) and the 2P domain K+ channel TREK-1 protein, Biochem Biophys Res Commun 346 (2006) 108-115.

[307] K.E. Beck, J.G. Kay, J.E. Braun, Rdj2, a J protein family member, interacts with cellular prion PrP(C), Biochem Biophys Res Commun 346 (2006) 866-871.

[308] M. Guo, T. Huang, Y. Cui, B. Pan, A. Shen, Y. Sun, Y. Yi, Y. Wang, G. Xiao, G. Sun, PrP(C) interacts with tetraspanin-7 through bovine PrP(154-182) containing alpha-helix 1, Biochem Biophys Res Commun (2007).

[309] C. Li, P. Wong, T. Pan, F. Xiao, S. Yin, B. Chang, S.C. Kang, J. Ironside, M.S. Sy, Normal cellular prion protein is a ligand of selectins: binding requires Le(X) but is inhibited by sLe(X), Biochem J 406 (2007) 333-341.

[310] G.N. Hajj, M.H. Lopes, A.F. Mercadante, S.S. Veiga, R.B. da Silveira, T.G. Santos, K.C. Ribeiro, M.A. Juliano, S.G. Jacchieri, S.M. Zanata, V.R. Martins, Cellular prion protein interaction with vitronectin supports axonal growth and is compensated by integrins, Journal of cell science 120 (2007) 1915-1926.

230

[311] K. Nieznanski, H. Nieznanska, K.J. Skowronek, K.M. Osiecka, D. Stepkowski, Direct interaction between prion protein and tubulin, Biochem Biophys Res Commun 334 (2005) 403-411.

[312] G. Sun, M. Guo, A. Shen, F. Mei, X. Peng, R. Gong, D. Guo, J. Wu, P. Tien, G. Xiao, Bovine PrPC directly interacts with alphaB-crystalline, FEBS Lett 579 (2005) 5419-5424.

[313] V. Mattei, T. Garofalo, R. Misasi, A. Circella, V. Manganelli, G. Lucania, A. Pavan, M. Sorice, Prion protein is a component of the multimolecular signaling complex involved in T cell activation, FEBS Lett 560 (2004) 14-18.

[314] F. Meggio, A. Negro, S. Sarno, M. Ruzzene, A. Bertoli, M.C. Sorgato, L.A. Pinna, Bovine prion protein as a modulator of protein kinase CK2, Biochem J 352 Pt 1 (2000) 191-196.

[315] F. Yehiely, P. Bamborough, M. Da Costa, B.J. Perry, G. Thinakaran, F.E. Cohen, G.A. Carlson, S.B. Prusiner, Identification of candidate proteins binding to prion protein, Neurobiol Dis 3 (1997) 339-355.

[316] S. Weiss, D. Proske, M. Neumann, M.H. Groschup, H.A. Kretzschmar, M. Famulok, E.L. Winnacker, RNA aptamers specifically interact with the prion protein PrP, J Virol 71 (1997) 8790-8797.

[317] E. Graner, A.F. Mercadante, S.M. Zanata, O.V. Forlenza, A.L. Cabral, S.S. Veiga, M.A. Juliano, R. Roesler, R. Walz, A. Minetti, I. Izquierdo, V.R. Martins, R.R. Brentani, Cellular prion protein binds laminin and mediates neuritogenesis, Brain Res Mol Brain Res 76 (2000) 85-92.

[318] R. Kleene, G. Loers, J. Langer, Y. Frobert, F. Buck, M. Schachner, Prion protein regulates glutamate-dependent lactate transport of astrocytes, J Neurosci 27 (2007) 12331-12340.

[319] R. Rieger, F. Edenhofer, C.I. Lasmezas, S. Weiss, The human 37-kDa laminin receptor precursor interacts with the prion protein in eukaryotic cells, Nat Med 3 (1997) 1383-1388.

[320] C. Hundt, J.M. Peyrin, S. Haik, S. Gauczynski, C. Leucht, R. Rieger, M.L. Riley, J.P. Deslys, D. Dormont, C.I. Lasmezas, S. Weiss, Identification of interaction domains of the prion protein with its 37-kDa/67-kDa laminin receptor, Embo J 20 (2001) 5876-5886.

[321] S.M. Zanata, M.H. Lopes, A.F. Mercadante, G.N. Hajj, L.B. Chiarini, R. Nomizo, A.R. Freitas, A.L. Cabral, K.S. Lee, M.A. Juliano, E. de Oliveira, S.G. Jachieri, A. Burlingame, L. Huang, R. Linden, R.R. Brentani, V.R. Martins, Stress-inducible protein 1 is a cell surface ligand for cellular prion that triggers neuroprotection, Embo J 21 (2002) 3307-3316.

231

[322] G. Schmitt-Ulms, G. Legname, M.A. Baldwin, H.L. Ball, N. Bradon, P.J. Bosque, K.L. Crossin, G.M. Edelman, S.J. DeArmond, F.E. Cohen, S.B. Prusiner, Binding of neural cell adhesion molecules (N-CAMs) to the cellular prion protein, J Mol Biol 314 (2001) 1209-1225.

[323] A.S. Rambold, M. Miesbauer, D. Rapaport, T. Bartke, M. Baier, K.F. Winklhofer, J. Tatzelt, Association of Bcl-2 with misfolded prion protein is linked to the toxic potential of cytosolic PrP, Mol Biol Cell 17 (2006) 3356-3368.

[324] M. Kristiansen, M.J. Messenger, P.C. Klohn, S. Brandner, J.D. Wadsworth, J. Collinge, S.J. Tabrizi, Disease-related prion protein forms aggresomes in neuronal cells leading to caspase activation and apoptosis, J Biol Chem 280 (2005) 38851-38861.

[325] S. Gauczynski, D. Nikles, S. El-Gogo, D. Papy-Garcia, C. Rey, S. Alban, D. Barritault, C.I. Lasmezas, S. Weiss, The 37-kDa/67-kDa Laminin Receptor Acts as a Receptor for Infectious Prions and Is Inhibited by Polysulfated Glycanes, J Infect Dis 194 (2006) 702-709.

[326] S. Gauczynski, J.M. Peyrin, S. Haik, C. Leucht, C. Hundt, R. Rieger, S. Krasemann, J.P. Deslys, D. Dormont, C.I. Lasmezas, S. Weiss, The 37-kDa/67-kDa laminin receptor acts as the cell-surface receptor for the cellular prion protein, Embo J 20 (2001) 5863-5875.

[327] C. Leucht, K. Vana, I. Renner-Muller, D. Dormont, C.I. Lasmezas, E. Wolf, S. Weiss, Knock-down of the 37-kDa/67-kDa laminin receptor in mouse brain by transgenic expression of specific antisense LRP RNA, Transgenic Res 13 (2004) 81-85.

[328] A.N. Shmakov, J. Bode, P.J. Kilshaw, S. Ghosh, Diverse patterns of expression of the 67-kD laminin receptor in human small intestinal mucosa: potential binding sites for prion proteins?, J Pathol 191 (2000) 318-322.

[329] E. Morel, T. Andrieu, F. Casagrande, S. Gauczynski, S. Weiss, J. Grassi, M. Rousset, D. Dormont, J. Chambaz, Bovine prion is endocytosed by human enterocytes via the 37 kDa/67 kDa laminin receptor, Am J Pathol 167 (2005) 1033-1042.

[330] A. Santuccione, V. Sytnyk, I. Leshchyns'ka, M. Schachner, Prion protein recruits its neuronal receptor NCAM to lipid rafts to activate p59fyn and to enhance neurite outgrowth, J Cell Biol 169 (2005) 341-354.

[331] S. Mouillet-Richard, M. Ermonval, C. Chebassier, J.L. Laplanche, S. Lehmann, J.M. Launay, O. Kellermann, Signal transduction through prion protein, Science 289 (2000) 1925-1928.

232

[332] A. Schwarz, M. Burwinkel, C. Riemer, J. Schultz, M. Baier, Unchanged Scrapie Pathology in Brain Tissue of Tyrosine Kinase Fyn-Deficient Mice, Neurodegener Dis 1 (2004) 266-268.

[333] M.H. Lopes, G.N. Hajj, A.G. Muras, G.L. Mancini, R.M. Castro, K.C. Ribeiro, R.R. Brentani, R. Linden, V.R. Martins, Interaction of cellular prion and stress-inducible protein 1 promotes neuritogenesis and neuroprotection by distinct signaling pathways, J Neurosci 25 (2005) 11330-11339.

[334] A. Sakudo, D.C. Lee, S. Li, T. Nakamura, Y. Matsumoto, K. Saeki, S. Itohara, K. Ikuta, T. Onodera, PrP cooperates with STI1 to regulate SOD activity in PrP-deficient neuronal cell line, Biochem Biophys Res Commun 328 (2005) 14-19.

[335] Y. Shaked, R. Engelstein, R. Gabizon, The binding of prion proteins to serum components is affected by detergent extraction conditions, J Neurochem 82 (2002) 1-5.

[336] M. Cuccioloni, M. Amici, A.M. Eleuteri, M. Biagetti, S. Barocci, M. Angeletti, Binding of recombinant PrPc to human plasminogen: kinetic and thermodynamic study using a resonant mirror biosensor, Proteins 58 (2005) 728-734.

[337] C. Ryou, S.B. Prusiner, G. Legname, Cooperative binding of dominant-negative prion protein to kringle domains, J Mol Biol 329 (2003) 323-333.

[338] G. Epple, W.D. Schleuning, G. Kettelgerdes, E. Kottgen, R. Gessner, M. Praus, Prion protein stimulates tissue-type plasminogen activator-mediated plasmin generation via a lysine-binding site on kringle 2, J Thromb Haemost 2 (2004) 962-968.

[339] J.A. Kornblatt, S. Marchal, H. Rezaei, M.J. Kornblatt, C. Balny, R. Lange, M.P. Debey, G. Hui Bon Hoa, M.C. Marden, J. Grosclaude, The fate of the prion protein in the prion/plasminogen complex, Biochem Biophys Res Commun 305 (2003) 518-522.

[340] M. Salmona, R. Capobianco, L. Colombo, A. De Luigi, G. Rossi, M. Mangieri, G. Giaccone, E. Quaglio, R. Chiesa, M.B. Donati, F. Tagliavini, G. Forloni, Role of plasminogen in propagation of scrapie, J Virol 79 (2005) 11225-11230.

[341] K. Barnewitz, M. Maringer, G. Mitteregger, A. Giese, U. Bertsch, H.A. Kretzschmar, Unaltered prion protein cleavage in plasminogen-deficient mice, Neuroreport 17 (2006) 527-530.

[342] F. Chen, H. Hasegawa, G. Schmitt-Ulms, T. Kawarai, C. Bohm, T. Katayama, Y. Gu, N. Sanjo, M. Glista, E. Rogaeva, Y. Wakutani, R. Pardossi-Piquard, X. Ruan, A. Tandon, F. Checler, P. Marambaud, K. Hansen, D. Westaway, P. St George-Hyslop, P. Fraser, TMP21 is a presenilin complex component that modulates gamma-secretase but not epsilon-secretase activity, Nature 440 (2006) 1208-1212.

233

[343] S.M. Yin, M.S. Sy, H.Y. Yang, P. Tien, Interaction of Doppel with the full-length laminin receptor precursor protein, Arch Biochem Biophys 428 (2004) 165-169.

[344] A. Azzalin, I. Del Vecchio, L. Ferretti, S. Comincini, The prion-like protein Doppel (Dpl) interacts with the human receptor for activated C-kinase 1 (RACK1) protein, Anticancer Res 26 (2006) 4539-4547.

[345] A. Azzalin, I. Del Vecchio, L.R. Chiarelli, G. Valentini, S. Comincini, L. Ferretti, Absence of interaction between doppel and GFAP, Grb2, PrPc proteins in human tumor astrocytic cells, Anticancer Res 25 (2005) 4369-4374.

[346] J. Lim, T. Hao, C. Shaw, A.J. Patel, G. Szabo, J.F. Rual, C.J. Fisk, N. Li, A. Smolyar, D.E. Hill, A.L. Barabasi, M. Vidal, H.Y. Zoghbi, A protein-protein interaction network for human inherited ataxias and disorders of Purkinje cell degeneration, Cell 125 (2006) 801-814.

[347] G. Legname, P. Nelken, Z. Guan, Z.F. Kanyo, S.J. DeArmond, S.B. Prusiner, Prion and doppel proteins bind to granule cells of the cerebellum, Proc Natl Acad Sci U S A 99 (2002) 16285-16290.

[348] T. Yokoyama, K.M. Kimura, Y. Ushiki, S. Yamada, A. Morooka, T. Nakashiba, T. Sassa, S. Itohara, In vivo conversion of cellular prion protein to pathogenic isoforms, as monitored by conformation-specific antibodies, J Biol Chem 276 (2001) 11265-11271.

[349] N. Genoud, A. Behrens, G. Miele, D. Robay, F.L. Heppner, S. Freigang, A. Aguzzi, Disruption of Doppel prevents neurodegeneration in mice with extensive Prnp deletions, Proc Natl Acad Sci U S A 101 (2004) 4198-4203.

[350] R. Al Bersaoui, I. Robert, Y. Lutz, F. Blanc, G. Sommermeyer-Leroux, H. Shibaguchi, D. Aunis, J.P. Fuchs, Purkinje-cell degeneration in prion protein-deficient mice is associated with a cerebellum-specific Doppel protein species signature, FEBS Lett 579 (2005) 2715-2721.

[351] L. Anderson, D. Rossi, J. Linehan, S. Brandner, C. Weissmann, Transgene-driven expression of the Doppel protein in Purkinje cells causes Purkinje cell degeneration and motor impairment, Proc Natl Acad Sci U S A 101 (2004) 3644-3649.

[352] N. Yamaguchi, S. Sakaguchi, K. Shigematsu, N. Okimura, S. Katamine, Doppel-induced Purkinje cell death is stoichiometrically abrogated by prion protein, Biochem Biophys Res Commun 319 (2004) 1247-1252.

[353] E. Flechsig, I. Hegyi, R. Leimeroth, A. Zuniga, D. Rossi, A. Cozzio, P. Schwarz, T. Rulicke, J. Gotz, A. Aguzzi, C. Weissmann, Expression of truncated PrP targeted to Purkinje cells of PrP knockout mice causes Purkinje cell death and ataxia, Embo J 22 (2003) 3095-3101.

234

[354] I. Radovanovic, N. Braun, O.T. Giger, K. Mertz, G. Miele, M. Prinz, B. Navarro, A. Aguzzi, Truncated prion protein and Doppel are myelinotoxic in the absence of oligodendrocytic PrPC, J Neurosci 25 (2005) 4879-4888.

[355] N. Nishida, P. Tremblay, T. Sugimoto, K. Shigematsu, S. Shirabe, C. Petromilli, S.P. Erpel, R. Nakaoke, R. Atarashi, T. Houtani, M. Torchia, S. Sakaguchi, S.J. DeArmond, S.B. Prusiner, S. Katamine, A mouse prion protein transgene rescues mice deficient for the prion protein gene from purkinje cell degeneration and demyelination, Lab Invest 79 (1999) 689-697.

[356] C. Kuwahara, A.M. Takeuchi, T. Nishimura, K. Haraguchi, A. Kubosaki, Y. Matsumoto, K. Saeki, T. Yokoyama, S. Itohara, T. Onodera, Prions prevent neuronal cell-line death, Nature 400 (1999) 225-226.

[357] A. Sakudo, D.C. Lee, I. Nakamura, Y. Taniuchi, K. Saeki, Y. Matsumoto, S. Itohara, K. Ikuta, T. Onodera, Cell-autonomous PrP-Doppel interaction regulates apoptosis in PrP gene-deficient neuronal cells, Biochem Biophys Res Commun 333 (2005) 448-454.

[358] K. Qin, L. Zhao, Y. Tang, S. Bhatta, J.M. Simard, R.Y. Zhao, Doppel-induced apoptosis and counteraction by cellular prion protein in neuroblastoma and astrocytes, Neuroscience 141 (2006) 1375-1388.

[359] R. Atarashi, N. Nishida, K. Shigematsu, S. Goto, T. Kondo, S. Sakaguchi, S. Katamine, Deletion of N-terminal residues 23-88 from prion protein (PrP) abrogates the potential to rescue PrP-deficient mice from PrP-like protein/doppel-induced Neurodegeneration, J Biol Chem 278 (2003) 28944-28949.

[360] E. Flechsig, C. Weissmann, The role of PrP in health and disease, Curr Mol Med 4 (2004) 337-353.

[361] D.C. Lee, A. Sakudo, C.K. Kim, T. Nishimura, K. Saeki, Y. Matsumoto, T. Yokoyama, S.G. Chen, S. Itohara, T. Onodera, Fusion of Doppel to octapeptide repeat and N-terminal half of hydrophobic region of prion protein confers resistance to serum deprivation, Microbiol Immunol 50 (2006) 203-209.

[362] D. Westaway, L.E. Hood, S.B. Prusiner, Doppel, a New PrP-like Mammalian Protein, in: S.B. Prusiner (Ed.), Prion Biology and Diseases Second Edition, Cold Spring Harbor Laboratory Press, New York, N.Y., 2004, pp. 283-304.

[363] P. Mastrangelo, L. Serpell, T. Dafforn, A. Lesk, P. Fraser, D. Westaway, A cluster of familial Creutzfeldt-Jakob disease mutations recapitulate conserved residues in Doppel: a case of molecular mimicry?, FEBS Lett 532 (2002) 21-26.

[364] A. Uelhoff, J. Tatzelt, A. Aguzzi, K.F. Winklhofer, C. Haass, A pathogenic PrP mutation and doppel interfere with polarized sorting of the prion protein, J Biol Chem 280 (2005) 5137-5140.

235

[365] T. Cui, A. Holme, J. Sassoon, D.R. Brown, Analysis of doppel protein toxicity, Mol Cell Neurosci 23 (2003) 144-155.

[366] M.L. Massimino, C. Ballarin, A. Bertoli, S. Casonato, S. Genovesi, A. Negro, M.C. Sorgato, Human Doppel and prion protein share common membrane microdomains and internalization pathways, Int J Biochem Cell Biol 36 (2004) 2016-2031.

[367] C. Hetz, M. Russelakis-Carneiro, K. Maundrell, J. Castilla, C. Soto, Caspase-12 and endoplasmic reticulum stress mediate neurotoxicity of pathological prion protein, Embo J 22 (2003) 5435-5445.

[368] A. Li, S.J. Barmada, K.A. Roth, D.A. Harris, N-terminally deleted forms of the prion protein activate both Bax-dependent and Bax-independent neurotoxic pathways, J Neurosci 27 (2007) 852-859.

[369] M. Coulpier, S. Messiaen, R. Hamel, M. Fernandez de Marco, T. Lilin, M. Eloit, Bax deletion does not protect neurons from BSE-induced death, Neurobiol Dis 23 (2006) 603-611.

[370] M. Brini, M. Miuzzo, N. Pierobon, A. Negro, M.C. Sorgato, The prion protein and its paralogue Doppel affect calcium signaling in Chinese hamster ovary cells, Mol Biol Cell 16 (2005) 2799-2808.

[371] B.S. Wong, T. Liu, D. Paisley, R. Li, T. Pan, S.G. Chen, G. Perry, R.B. Petersen, M.A. Smith, D.W. Melton, P. Gambetti, D.R. Brown, M.S. Sy, Induction of HO-1 and NOS in doppel-expressing mice devoid of PrP: implications for doppel function, Mol Cell Neurosci 17 (2001) 768-775.

[372] A. Aguzzi, M. Polymenidou, Mammalian prion biology: one century of evolving concepts, Cell 116 (2004) 313-327.

[373] R.S. Stewart, P. Piccardo, B. Ghetti, D.A. Harris, Neurodegenerative illness in transgenic mice expressing a transmembrane form of the prion protein, J Neurosci 25 (2005) 3469-3477.

[374] J.C. Watts, D. Westaway, The prion protein family: Diversity, rivalry, and dysfunction, Biochim Biophys Acta 1772 (2007) 654-672.

[375] M.S. Nadal, A. Ozaita, Y. Amarillo, E. Vega-Saenz de Miera, Y. Ma, W. Mo, E.M. Goldberg, Y. Misumi, Y. Ikehara, T.A. Neubert, B. Rudy, The CD26-related dipeptidyl aminopeptidase-like protein DPPX is a critical component of neuronal A-type K+ channels, Neuron 37 (2003) 449-461.

[376] K. Wada, N. Yokotani, C. Hunter, K. Doi, R.J. Wenthold, S. Shimasaki, Differential expression of two distinct forms of mRNA encoding members of a dipeptidyl aminopeptidase family, Proc Natl Acad Sci U S A 89 (1992) 197-201.

236

[377] N. Yokotani, K. Doi, R.J. Wenthold, K. Wada, Non-conservation of a catalytic residue in a dipeptidyl aminopeptidase IV-related protein encoded by a gene on human chromosome 7, Human molecular genetics 2 (1993) 1037-1039.

[378] V. Novitskaya, O.V. Bocharova, I. Bronstein, I.V. Baskakov, Amyloid fibrils of mammalian prion protein are highly toxic to cultured cells and primary neurons, J Biol Chem 281 (2006) 13828-13836.

[379] C. Uboldi, M. Paulis, E. Guidi, A. Bertoni, G.P. Meo, A. Perucatti, L. Iannuzzi, E. Raimondi, R.M. Brunner, A. Eggen, L. Ferretti, Cloning of the bovine prion-like Shadoo (SPRN) gene by comparative analysis of the predicted genomic locus, Mamm Genome 17 (2006) 1130-1139.

[380] E. Lampo, M. Van Poucke, K. Hugot, H. Hayes, A. Van Zeveren, L.J. Peelman, Characterization of the genomic region containing the Shadow of Prion Protein (SPRN) gene in sheep, BMC genomics 8 (2007) 138.

[381] J.C. Watts, B. Drisaldi, V. Ng, J. Yang, B. Strome, P. Horne, M.S. Sy, L. Yoong, R. Young, P. Mastrangelo, C. Bergeron, P.E. Fraser, G.A. Carlson, H.T. Mount, G. Schmitt-Ulms, D. Westaway, The CNS glycoprotein Shadoo has PrP(C)-like protective properties and displays reduced levels in prion infections, Embo J 26 (2007) 4038-4050.

[382] M.A. Chishti, D.S. Yang, C. Janus, A.L. Phinney, P. Horne, J. Pearson, R. Strome, N. Zuker, J. Loukides, J. French, S. Turner, G. Lozza, M. Grilli, S. Kunicki, C. Morissette, J. Paquette, F. Gervais, C. Bergeron, P.E. Fraser, G.A. Carlson, P.S. George-Hyslop, D. Westaway, Early-onset amyloid deposition and cognitive deficits in transgenic mice expressing a double mutant form of amyloid precursor protein 695, J Biol Chem 276 (2001) 21562-21570.

[383] K.M. Pan, N. Stahl, S.B. Prusiner, Purification and properties of the cellular prion protein from Syrian hamster brain, Protein Sci 1 (1992) 1343-1352.

[384] B. Vincent, E. Paitel, Y. Frobert, S. Lehmann, J. Grassi, F. Checler, Phorbol ester-regulated cleavage of normal prion protein in HEK293 human cells and murine neurons, J Biol Chem 275 (2000) 35612-35616.

[385] E.T. Parkin, N.T. Watt, A.J. Turner, N.M. Hooper, Dual mechanisms for shedding of the cellular prion protein, J Biol Chem 279 (2004) 11170-11178.

[386] M. Scott, D. Foster, C. Mirenda, D. Serban, F. Coufal, M. Walchli, M. Torchia, D. Groth, G. Carlson, S.J. DeArmond, et al., Transgenic mice expressing hamster prion protein produce species-specific scrapie infectivity and amyloid plaques, Cell 59 (1989) 847-857.

[387] M.A. Chishti, R. Strome, G.A. Carlson, D. Westaway, Syrian hamster prion protein (PrP(C)) is expressed in photoreceptor cells of the adult retina, Neurosci Lett 234 (1997) 11-14.

237

[388] D. Westaway, S.J. DeArmond, J. Cayetano-Canlas, D. Groth, D. Foster, S.L. Yang, M. Torchia, G.A. Carlson, S.B. Prusiner, Degeneration of skeletal muscle, peripheral nerves, and the central nervous system in transgenic mice overexpressing wild-type prion proteins, Cell 76 (1994) 117-129.

[389] J. Burthem, B. Urban, A. Pain, D.J. Roberts, The normal cellular prion protein is strongly expressed by myeloid dendritic cells, Blood 98 (2001) 3733-3738.

[390] A. Sakudo, D.C. Lee, T. Nishimura, S. Li, S. Tsuji, T. Nakamura, Y. Matsumoto, K. Saeki, S. Itohara, K. Ikuta, T. Onodera, Octapeptide repeat region and N-terminal half of hydrophobic region of prion protein (PrP) mediate PrP-dependent activation of superoxide dismutase, Biochem Biophys Res Commun 326 (2005) 600-606.

[391] S.M. Haeryfar, D.W. Hoskin, Thy-1: more than a mouse pan-T cell marker, J Immunol 173 (2004) 3581-3588.

[392] D.A. Butler, M.R. Scott, J.M. Bockman, D.R. Borchelt, A. Taraboulos, K.K. Hsiao, D.T. Kingsbury, S.B. Prusiner, Scrapie-infected murine neuroblastoma cells produce protease-resistant prion proteins, J Virol 62 (1988) 1558-1564.

[393] H.M. Schatzl, L. Laszlo, D.M. Holtzman, J. Tatzelt, S.J. DeArmond, R.I. Weiner, W.C. Mobley, S.B. Prusiner, A hypothalamic neuronal cell line persistently infected with scrapie prions exhibits apoptosis, J Virol 71 (1997) 8821-8831.

[394] C.R. Birkett, R.M. Hennion, D.A. Bembridge, M.C. Clarke, A. Chree, M.E. Bruce, C.J. Bostock, Scrapie strains maintain biological phenotypes on propagation in a cell line in culture, Embo J 20 (2001) 3351-3358.

[395] K.J. Knaus, M. Morillas, W. Swietnicki, M. Malone, W.K. Surewicz, V.C. Yee, Crystal structure of the human prion protein reveals a mechanism for oligomerization, Nat Struct Biol 8 (2001) 770-774.

[396] A. Li, P. Piccardo, S.J. Barmada, B. Ghetti, D.A. Harris, Prion protein with an octapeptide insertion has impaired neuroprotective activity in transgenic mice, Embo J 26 (2007) 2777-2785.

[397] A. Taraboulos, K. Jendroska, D. Serban, S.L. Yang, S.J. DeArmond, S.B. Prusiner, Regional mapping of prion proteins in brain, Proc Natl Acad Sci U S A 89 (1992) 7620-7624.

[398] M. Moser, R.J. Colello, U. Pott, B. Oesch, Developmental expression of the prion protein gene in glial cells, Neuron 14 (1995) 509-517.

[399] T. Liu, T. Zwingman, R. Li, T. Pan, B.S. Wong, R.B. Petersen, P. Gambetti, K. Herrup, M.S. Sy, Differential expression of cellular prion protein in mouse brain as detected with multiple anti-PrP monoclonal antibodies, Brain Res 896 (2001) 118-129.

238

[400] J. Dong, A. Li, N. Yamaguchi, S. Sakaguchi, D.A. Harris, Doppel induces degeneration of cerebellar Purkinje cells independently of Bax, Am J Pathol 171 (2007) 599-607.

[401] V. Beringue, D. Vilette, G. Mallinson, F. Archer, M. Kaisar, M. Tayebi, G.S. Jackson, A.R. Clarke, H. Laude, J. Collinge, S. Hawke, PrPSc binding antibodies are potent inhibitors of prion replication in cell lines, J Biol Chem 279 (2004) 39671-39676.

[402] R.H. Kimberlin, C.A. Walker, Pathogenesis of scrapie (strain 263K) in hamsters infected intracerebrally, intraperitoneally or intraocularly, J Gen Virol 67 ( Pt 2) (1986) 255-263.

[403] R.N. Hogan, J.R. Baringer, S.B. Prusiner, Scrapie infection diminishes spines and increases varicosities of dendrites in hamsters: a quantitative Golgi analysis, Journal of neuropathology and experimental neurology 46 (1987) 461-473.

[404] A.R. Johnston, C. Black, J. Fraser, N. MacLeod, Scrapie infection alters the membrane and synaptic properties of mouse hippocampal CA1 pyramidal neurones, J Physiol 500 ( Pt 1) (1997) 1-15.

[405] D. Brown, P. Belichenko, J. Sales, M. Jeffrey, J.R. Fraser, Early loss of dendritic spines in murine scrapie revealed by confocal analysis, Neuroreport 12 (2001) 179-183.

[406] S.B. Prusiner, Prions, Proc Natl Acad Sci U S A 95 (1998) 13363-13383.

[407] E. Boonacker, C.J. Van Noorden, The multifunctional or moonlighting protein CD26/DPPIV, European journal of cell biology 82 (2003) 53-73.

[408] R. Mentlein, B. Gallwitz, W.E. Schmidt, Dipeptidyl-peptidase IV hydrolyses gastric inhibitory polypeptide, glucagon-like peptide-1(7-36)amide, peptide histidine methionine and is responsible for their degradation in human serum, European journal of biochemistry / FEBS 214 (1993) 829-835.

[409] D.J. Drucker, Therapeutic potential of dipeptidyl peptidase IV inhibitors for the treatment of type 2 diabetes, Expert opinion on investigational drugs 12 (2003) 87-100.

[410] D. Marguet, L. Baggio, T. Kobayashi, A.M. Bernard, M. Pierres, P.F. Nielsen, U. Ribel, T. Watanabe, D.J. Drucker, N. Wagtmann, Enhanced insulin secretion and improved glucose tolerance in mice lacking CD26, Proc Natl Acad Sci U S A 97 (2000) 6874-6879.

[411] Y. Kin, Y. Misumi, Y. Ikehara, Biosynthesis and characterization of the brain-specific membrane protein DPPX, a dipeptidyl peptidase IV-related protein, Journal of biochemistry 129 (2001) 289-295.

239

[412] L. de Lecea, E. Soriano, J.R. Criado, S.C. Steffensen, S.J. Henriksen, J.G. Sutcliffe, Transcripts encoding a neural membrane CD26 peptidase-like protein are stimulated by synaptic activity, Brain Res Mol Brain Res 25 (1994) 286-296.

[413] M.S. Nadal, Y. Amarillo, E. Vega-Saenz de Miera, B. Rudy, Differential characterization of three alternative spliced isoforms of DPPX, Brain Res 1094 (2006) 1-12.

[414] P. Strop, A.J. Bankovich, K.C. Hansen, K.C. Garcia, A.T. Brunger, Structure of a human A-type potassium channel interacting protein DPPX, a member of the dipeptidyl aminopeptidase family, J Mol Biol 343 (2004) 1055-1065.

[415] M.A. van Es, P.W. van Vught, H.M. Blauw, L. Franke, C.G. Saris, L. Van den Bosch, S.W. de Jong, V. de Jong, F. Baas, R. van't Slot, R. Lemmens, H.J. Schelhaas, A. Birve, K. Sleegers, C. Van Broeckhoven, J.C. Schymick, B.J. Traynor, J.H. Wokke, C. Wijmenga, W. Robberecht, P.M. Andersen, J.H. Veldink, R.A. Ophoff, L.H. van den Berg, Genetic variation in DPP6 is associated with susceptibility to amyotrophic lateral sclerosis, Nat Genet 40 (2008) 29-31.

[416] S. Cronin, S. Berger, J. Ding, J.C. Schymick, N. Washecka, D.G. Hernandez, M.J. Greenway, D.G. Bradley, B.J. Traynor, O. Hardiman, A genome-wide association study of sporadic ALS in a homogenous Irish population, Human molecular genetics (2007).

[417] S.Y. Qi, P.J. Riviere, J. Trojnar, J.L. Junien, K.O. Akinsanya, Cloning and characterization of dipeptidyl peptidase 10, a new member of an emerging subgroup of serine proteases, Biochem J 373 (2003) 179-189.

[418] H.H. Jerng, A.D. Lauver, P.J. Pfaffinger, DPP10 splice variants are localized in distinct neuronal populations and act to differentially regulate the inactivation properties of Kv4-based ion channels, Mol Cell Neurosci 35 (2007) 604-624.

[419] H.H. Jerng, Y. Qian, P.J. Pfaffinger, Modulation of Kv4.2 channel expression and gating by dipeptidyl peptidase 10 (DPP10), Biophysical journal 87 (2004) 2380-2396.

[420] X. Ren, Y. Hayashi, N. Yoshimura, K. Takimoto, Transmembrane interaction mediates complex formation between peptidase homologues and Kv4 channels, Mol Cell Neurosci 29 (2005) 320-332.

[421] K. Takimoto, Y. Hayashi, X. Ren, N. Yoshimura, Species and tissue differences in the expression of DPPY splicing variants, Biochem Biophys Res Commun 348 (2006) 1094-1100.

[422] E. Zagha, A. Ozaita, S.Y. Chang, M.S. Nadal, U. Lin, M.J. Saganich, T. McCormack, K.O. Akinsanya, S.Y. Qi, B. Rudy, DPP10 modulates Kv4-mediated A-type potassium channels, J Biol Chem 280 (2005) 18853-18861.

240

[423] M. Allen, A. Heinzmann, E. Noguchi, G. Abecasis, J. Broxholme, C.P. Ponting, S. Bhattacharyya, J. Tinsley, Y. Zhang, R. Holt, E.Y. Jones, N. Lench, A. Carey, H. Jones, N.J. Dickens, C. Dimon, R. Nicholls, C. Baker, L. Xue, E. Townsend, M. Kabesch, S.K. Weiland, D. Carr, E. von Mutius, I.M. Adcock, P.J. Barnes, G.M. Lathrop, M. Edwards, M.F. Moffatt, W.O. Cookson, Positional cloning of a novel gene influencing asthma from chromosome 2q14, Nat Genet 35 (2003) 258-263.

[424] P. Chomczynski, N. Sacchi, Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction, Analytical biochemistry 162 (1987) 156-159.

[425] J.C. Schimenti, B.J. Libby, R.A. Bergstrom, L.A. Wilson, D. Naf, L.M. Tarantino, A. Alavizadeh, A. Lengeling, M. Bucan, Interdigitated deletion complexes on mouse chromosome 5 induced by irradiation of embryonic stem cells, Genome research 10 (2000) 1043-1050.

[426] J. Tatzelt, N. Maeda, M. Pekny, S.L. Yang, C. Betsholtz, C. Eliasson, J. Cayetano, A.P. Camerino, S.J. DeArmond, S.B. Prusiner, Scrapie in mice deficient in apolipoprotein E or glial fibrillary acidic protein, Neurology 47 (1996) 449-453.

[427] R.B. Hough, A. Lengeling, V. Bedian, C. Lo, M. Bucan, Rump white inversion in the mouse disrupts dipeptidyl aminopeptidase-like protein 6 and causes dysregulation of Kit expression, Proc Natl Acad Sci U S A 95 (1998) 13800-13805.

[428] J.S. Duke-Cohan, C. Morimoto, J.A. Rocker, S.F. Schlossman, A novel form of dipeptidylpeptidase IV found in human serum. Isolation, characterization, and comparison with T lymphocyte membrane dipeptidylpeptidase IV (CD26), J Biol Chem 270 (1995) 14107-14114.

[429] H.B. Rasmussen, S. Branner, F.C. Wiberg, N. Wagtmann, Crystal structure of human dipeptidyl peptidase IV/CD26 in complex with a substrate analog, Nat Struct Biol 10 (2003) 19-25.

[430] B. Metz, G.F. Kersten, G.J. Baart, A. de Jong, H. Meiring, J. ten Hove, M.J. van Steenbergen, W.E. Hennink, D.J. Crommelin, W. Jiskoot, Identification of formaldehyde-induced modifications in proteins: reactions with insulin, Bioconjugate chemistry 17 (2006) 815-822.

[431] B. Metz, G.F. Kersten, P. Hoogerhout, H.F. Brugghe, H.A. Timmermans, A. de Jong, H. Meiring, J. ten Hove, W.E. Hennink, D.J. Crommelin, W. Jiskoot, Identification of formaldehyde-induced modifications in proteins: reactions with model peptides, J Biol Chem 279 (2004) 6235-6243.

[432] T.R. Brummelkamp, R. Bernards, R. Agami, A system for stable expression of short interfering RNAs in mammalian cells, Science 296 (2002) 550-553.

241

[433] S.R. D'Mello, C. Galli, T. Ciotti, P. Calissano, Induction of apoptosis in cerebellar granule neurons by low potassium: inhibition of death by insulin-like growth factor I and cAMP, Proc Natl Acad Sci U S A 90 (1993) 10989-10993.

[434] M. Bucan, D.L. Nagle, R.B. Hough, V.M. Chapman, C.W. Lo, Lethality of Rw/Rw mouse embryos during early postimplantation development, Dev Biol 168 (1995) 307-318.

[435] D.A. Stephenson, K.H. Lee, D.L. Nagle, C.H. Yen, A. Morrow, D. Miller, V.M. Chapman, M. Bucan, Mouse rump-white mutation associated with an inversion of chromosome 5, Mamm Genome 5 (1994) 342-348.

[436] A.G. Searle, G.M. Truslove, A gene triplet in the mouse, Genetical research 15 (1970) 227-235.

[437] D.L. Nagle, P. Martin-DeLeon, R.B. Hough, M. Bucan, Structural analysis of chromosomal rearrangements associated with the developmental mutations Ph, W19H, and Rw on mouse chromosome 5, Proc Natl Acad Sci U S A 91 (1994) 7237-7241.

[438] V. Bonifati, P. Rizzu, M.J. van Baren, O. Schaap, G.J. Breedveld, E. Krieger, M.C. Dekker, F. Squitieri, P. Ibanez, M. Joosse, J.W. van Dongen, N. Vanacore, J.C. van Swieten, A. Brice, G. Meco, C.M. van Duijn, B.A. Oostra, P. Heutink, Mutations in the DJ-1 gene associated with autosomal recessive early-onset parkinsonism, Science 299 (2003) 256-259.

[439] M.A. Wilson, J.L. Collins, Y. Hod, D. Ringe, G.A. Petsko, The 1.1-A resolution crystal structure of DJ-1, the protein mutated in autosomal recessive early onset Parkinson's disease, Proc Natl Acad Sci U S A 100 (2003) 9256-9261.

[440] S. Dorus, E.J. Vallender, P.D. Evans, J.R. Anderson, S.L. Gilbert, M. Mahowald, G.J. Wyckoff, C.M. Malcom, B.T. Lahn, Accelerated evolution of nervous system genes in the origin of Homo sapiens, Cell 119 (2004) 1027-1040.

[441] M.D. Gorrell, First bite, Nat Struct Biol 10 (2003) 3-5.

[442] J.W. Herms, T. Tings, S. Dunker, H.A. Kretzschmar, Prion protein affects Ca2+-activated K+ currents in cerebellar purkinje cells, Neurobiol Dis 8 (2001) 324-330.

[443] T.M. Miller, E.M. Johnson, Jr., Metabolic and genetic analyses of apoptosis in potassium/serum-deprived rat cerebellar granule cells, J Neurosci 16 (1996) 7487-7495.

[444] S.P. Yu, C.H. Yeh, S.L. Sensi, B.J. Gwag, L.M. Canzoniero, Z.S. Farhangrazi, H.S. Ying, M. Tian, L.L. Dugan, D.W. Choi, Mediation of neuronal apoptosis by enhancement of outward potassium current, Science 278 (1997) 114-117.

242

243

[445] S.P. Yu, C. Yeh, U. Strasser, M. Tian, D.W. Choi, NMDA receptor-mediated K+ efflux and neuronal apoptosis, Science 284 (1999) 336-339.

[446] H.J. Mackay, C.J. Twelves, Targeting the protein kinase C family: are we there yet?, Nature reviews 7 (2007) 554-562.

[447] M.E. Gurney, H. Pu, A.Y. Chiu, M.C. Dal Canto, C.Y. Polchow, D.D. Alexander, J. Caliendo, A. Hentati, Y.W. Kwon, H.X. Deng, et al., Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation, Science 264 (1994) 1772-1775.

[448] P.C. Wong, C.A. Pardo, D.R. Borchelt, M.K. Lee, N.G. Copeland, N.A. Jenkins, S.S. Sisodia, D.W. Cleveland, D.L. Price, An adverse property of a familial ALS-linked SOD1 mutation causes motor neuron disease characterized by vacuolar degeneration of mitochondria, Neuron 14 (1995) 1105-1116.

[449] L.I. Bruijn, M.W. Becher, M.K. Lee, K.L. Anderson, N.A. Jenkins, N.G. Copeland, S.S. Sisodia, J.D. Rothstein, D.R. Borchelt, D.L. Price, D.W. Cleveland, ALS-linked SOD1 mutant G85R mediates damage to astrocytes and promotes rapidly progressive disease with SOD1-containing inclusions, Neuron 18 (1997) 327-338.

[450] K.A. Conway, J.D. Harper, P.T. Lansbury, Accelerated in vitro fibril formation by a mutant alpha-synuclein linked to early-onset Parkinson disease, Nat Med 4 (1998) 1318-1320.

[451] O. Schweers, E. Schonbrunn-Hanebeck, A. Marx, E. Mandelkow, Structural studies of tau protein and Alzheimer paired helical filaments show no evidence for beta-structure, J Biol Chem 269 (1994) 24290-24297.

[452] M.T. Bishop, P. Hart, L. Aitchison, H.N. Baybutt, C. Plinston, V. Thomson, N.L. Tuzi, M.W. Head, J.W. Ironside, R.G. Will, J.C. Manson, Predicting susceptibility and incubation time of human-to-human transmission of vCJD, Lancet neurology 5 (2006) 393-398.

[453] N. Pham, S. Yin, S. Yu, P. Wong, S.C. Kang, C. Li, M.S. Sy, Normal cellular prion protein with a methionine at position 129 has a more exposed helix 1 and is more prone to aggregate, Biochem Biophys Res Commun (2008).

[454] K. Garber, Genetics. The elusive ALS genes, Science 319 (2008) 20.

[455] C. Janus, J. Pearson, J. McLaurin, P.M. Mathews, Y. Jiang, S.D. Schmidt, M.A. Chishti, P. Horne, D. Heslin, J. French, H.T. Mount, R.A. Nixon, M. Mercken, C. Bergeron, P.E. Fraser, P. St George-Hyslop, D. Westaway, A beta peptide immunization reduces behavioural impairment and plaques in a model of Alzheimer's disease, Nature 408 (2000) 979-982.

Documents

CHARACTERIZATION OF SHADOO AND DPPX: TWO … · exist in prion biology including the cellular role of PrPC, the potential involvement of auxiliary factors in prion replication, and