Investigating genetic, gene expression and proteomic changes over temperature gradients in intertidal Nerita species i

INVESTIGATING GENETIC, GENE

EXPRESSION AND PROTEOMIC CHANGES

OVER TEMPERATURE GRADIENTS IN

INTERTIDAL NERITA SPECIES

Shorash Amin

Bachelor of Biomedical Science (1A Honours)

Submitted in fulfilment of the requirements for the degree of

Doctor of Philosophy

School of Biomedical Sciences

Faculty of Health

Queensland University of Technology

2018

ii Investigating genetic, gene expression and proteomic changes over temperature gradients in intertidal Nerita species

Keywords

De novo assembly; digital gene expression; genomics; heat shock protein; Ion torrent;

transcriptome; Nerita albicilla; Nerita melanotragus; molluscs; proteome; RNAseq;

thermal stress; Nerita melanotragus, Illumina.

Investigating genetic, gene expression and proteomic changes over temperature gradients in intertidal Nerita species iii

Abstract

A key area of research in physiological genomics is understanding the gene

expression and proteomic responses of specific species to abiotic change in their

habitat. In order to investigate these responses, an appropriate group of organisms is

required that is distributed across an environmental gradient. One such group of

organisms that meet this requirement are class Gastropoda, which are distributed

globally in a range of different environments. This highly speciose group are important

socially, economically and ecologically. Species from this taxonomic group form a

large component of intertidal zone fauna in many areas, globally.

The intertidal zone is amongst the harshest of environments on Earth, with

constant changes in temperature, pH, sea level and UV exposure. Furthermore, species

inhabiting these areas are periodically submerged due to the tidal cycle. The intertidal

zone can be further subdivided into the spray, upper, mid and lower intertidal sub

zones. Abiotic stresses also vary across these habitats as does the level of

submergence. As a consequence, species have adapted to sub zones within the

intertidal area resulting in a strong zonation of organisms among sub zones. It is

suggested that zonation of species is determined in a large part by their physiological

limits, with many intertidal species living near their upper thermal limits. This may

suggest that organisms within these areas are under extreme physiological stress.

Some intertidal organisms possess physiological defense mechanisms which

allow them to cope when faced with external stressors. This is of particular importance

for organisms inhabiting the intertidal zone as they are presented with continuous and

varying levels of stress. Although some species have adapted to these stressors through

standing genetic variation, others respond through gene expression or protein

abundance changes to these specific stressors. In order to gain further understanding

of how environmental stress affects intertidal grastropod species, studies are required

to determine how environmental stress influences the gene repertoire, gene expression

patterns and protein abundance in these species. Nerita melanotragus and N.

albicilla are widespread intertidal gastropod species, distributed across a number of

temporally and spatially fluctuating environmental gradients, including abrupt changes

in temperature over a tidal cycle. These species differ in their ecology, as N.

iv Investigating genetic, gene expression and proteomic changes over temperature gradients in intertidal Nerita species

melanotragus is a mid-littoral species, while N. albicilla is a low-littoral species.

Colonisation of different areas in the intertidal zone means that N. melanotragus (mid-

littoral) is likely to sustain longer periods of temperature stress than N. albicilla (low-

littoral), but N. albicilla is likely to experience more acute temperature stress when

exposed to high temperatures. Consequently, these two species present an interesting

case to examine differences in the gene repertoire, gene expression and protein

abundance patterns of intertidal gastropod species in response to the same temperature

stress conditions. Therefore the aim of this project was to investigate and compare the

physiological response of two closely related intertidal marine snails from the

genus Nerita to temperature stress using a combination of RNAseq and proteomic

experiments. To achieve this, I undertook three separate but inter-related experiments.

The first experiment was to optimize the assembly of next generation sequencing

(NGS) data and determine which assembler produced the best quality assemblies for

gastropod species. The sequencing, de novo assembly and annotation of transcriptome

datasets generated with next generation sequencing (NGS) has enabled biologists to

answer physiological questions in non-model species with unprecedented ease.

Reliable and accurate de novo assembly and annotation of transcriptomes, however, is

a critically important step for transcriptome assemblies generated from short read

sequences. Typical benchmarks for assembly and annotation reliability have been

performed with model species. To address the reliability and accuracy of de novo

transcriptome assembly in a non-model gastropod species, an RNAseq dataset was

generated for an intertidal gastropod mollusc species, Nerita melanotragus, and

compared the assembly produced by four different de novo transcriptome assemblers;

Velvet, Oases, Geneious and Trinity, for a number of quality metrics and redundancy.

Both the Trinity and Oases de novo assemblers produced the best assemblies based on

all quality metrics including fewer contigs, increased N50 and average contig length

and contigs of greater length. Overall, the BLAST and annotation success of the

assemblies was not high with only 15-19% of contigs assigned a putative function.

The second experiment examined differences in the expressed gene families of

N. melanotragus and N. albicilla under normal temperature conditions. Deep

transcriptome sequencing was undertaken for both species to determine differences

in the expressed gene compliment under natural living conditions. Samples were

collected from King’s Beach, Caloundra, Australia and RNA was extracted and

Investigating genetic, gene expression and proteomic changes over temperature gradients in intertidal Nerita species v

sequenced on an Illumina Hiseq 2500. A total of 6.2 Gbp and 7.4 Gbp of sequence

data was produced for N. melanotragus and N. albicilla respectively. Ortholog

analysis of the two Nerita species with Crassostrea gigas and Lottia gigantea

revealed a total of 6,457 orthologs in common. The two Nerita species shared a total

of 8,501 orthologs, with 3,618 unique orthologs in N. melanotragus and 2,280 in

Nerita albicilla. Gene set enrichment of the common genes between the Nerita

species revealed two over-represented terms (activation of protein kinase B and

positive regulation of guanylate cyclase activity), however, neither is directly related

to stress. Overall a larger number of stress transcripts were found to be expressed in

N. albicilla under normal conditions when compared to N. melanotragus.

The third experiment examined differences in the gene expression and protein

abundance patterns in response to the three temperature stress conditions in N.

melanotragus and N. albicilla. For the RNAseq component of this experiment, nine

individual samples from each of N. melanotragus and N. albicilla were randomly

allocated into three treatments (14 °C, 22 °C and 31 °C) with three replicates in each

treatment. Five temperature treatments (14 °C, 22 °C, 31 °C, 38 °C and 45 °C) were

used for the proteomic experiment with five replicate individuals in each treatment

for each species. Treatment animals were euthanized, RNA/protein extracted and

each individual was sequenced on an Illumina Hiseq 2500 or run on a TripleTOF

5600+ mass spectrometer (Sciex) coupled with an Ekspert nanoLC 400 system

(Eksigent). The two species had highly divergent patterns of gene expression and

protein abundance under specific treatment conditions. Few differentially expressed

genes/proteins (~22 and 109 respectively, ) were observed in Nerita albicilla, and

these were dominated by molecular chaperones. More differentially expressed genes

but fewer proteins (~131 and 80 respectively) were observed in N. melanotragus, but

no dominant class of genes was observed in either datatset. Little overlap existed

between differentially expressed transcripts and differentially abundant proteins in

either species.

The molecular data generated in these experiments is the largest produced for

the genus Nerita. Overall the data generated from these three experiments has

enabled us to determine that N. albicilla, the lower intertidal species iniates a thermal

stress response at much lower temperatures than N. melanotragus. This supports the

idea that low-littoral species undergo thermal stress at lower temperatures than mid-

vi

littoral species. Using a genome wide approach, this study has established that lower

intertidal species may be far more susceptible to future climate change due to a more

acute temperature stress response.

vii

Table of Contents

KEYWORDS ......................................................................................................................................... II ABSTRACT .......................................................................................................................................... III TABLE OF CONTENTS .................................................................................................................... VII LIST OF FIGURES .............................................................................................................................. IX LIST OF TABLES ................................................................................................................................. X LIST OF PUBLICATIONS ................................................................................................................. XI STATEMENT OF ORIGINAL AUTHORSHIP ................................................................................ XIII ACKNOWLEDGEMENTS ............................................................................................................... XIV

CHAPTER 1: LITERATURE REVIEW ............................................................ 15 1.1 BACKGROUND ......................................................................................................................... 15 1.2 CLIMATE CHANGE AND ENVIRONMENTAL STRESS ...................................................... 18 1.3 MARINE ANIMAL RESPONSE TO ENVIRONMENTAL STRESS ....................................... 20 1.4 THE CELLULAR STRESS RESPONSE ................................................................................... 22 1.5 MARINE GASTROPODS AS A SYSTEM TO INVESTIGATE TEMPERATURE STRESS . 23 1.6 PHYSIOLOGICAL INVESTIGATION INTO REPRESENTATIVE SPECIES FROM THE GENUS NERITA ................................................................................................................................... 24 1.7 RESEARCH QUESTION AND APPROACH ........................................................................... 26

1.7.1 Research objective 1 ....................................................................................................... 26 1.7.2 Research objective 2 ....................................................................................................... 26 1.7.3 Research objective 3 ....................................................................................................... 27

CHAPTER 2: ASSEMBLY AND ANNOTATION OF A NON-MODEL GASTROPOD (NERITA MELANOTRAGUS) TRANSCRIPTOME: A COMPARISON OF DE NOVO ASSEMBLERS ................................................... 29 2.1 BACKGROUND ......................................................................................................................... 29 2.2 METHODS .................................................................................................................................. 31

2.2.1 Sample acquisition and sequencing ................................................................................ 31 2.2.2 Assembly and annotation ................................................................................................ 32

2.3 RESULTS .................................................................................................................................... 34 2.3.1 Ion torrent sequencing and reads assembly .................................................................... 34 2.3.2 Functional annotation of contigs ..................................................................................... 35 2.3.3 PCR validation of contigs ............................................................................................... 41

2.4 DISCUSSION ............................................................................................................................. 42 2.5 CONCLUSION ........................................................................................................................... 44

CHAPTER 3: COMPARATIVE TRANSCRIPTOMICS AND CHARACTERIZATION OF GENES AND GENE FAMILIES INVOLVED IN ABIOTIC STRESS RESPONSE FROM TWO INTERTIDAL MARINE SNAILS 45 3.1 BACKGROUND ......................................................................................................................... 45 3.2 METHODS .................................................................................................................................. 46

3.2.1 RNA extraction, sequencing and read processing .......................................................... 46 3.2.2 Transcriptome quality and completeness ........................................................................ 48

viii

3.2.3 Identification of unique and common genes ................................................................... 49 3.2.4 Identification and comparative analysis of full length candidate genes .......................... 49

3.3 RESULTS .................................................................................................................................... 49 3.3.1 Sequencing results, reads assembly and functional annotation of contigs ...................... 49 3.3.2 Unique and common genes ............................................................................................. 51 3.3.3 Identification and comparative analysis of candidate genes ........................................... 52

3.4 DISCUSSION ............................................................................................................................. 54 3.4.1 Comparative copy number analysis of stress genes ........................................................ 55 3.4.2 Conclusion ...................................................................................................................... 56

CHAPTER 4: DIFFERENTIAL GENE EXPRESSION OF TWO INTERTIDAL SNAILS IN RESPONSE TO TEMPERATURE STRESS ......... 58 4.1 BACKGROUND ......................................................................................................................... 58 4.2 METHODS .................................................................................................................................. 60

4.2.1 Organism conditioning and treatment ............................................................................. 60 4.2.2 RNA extraction and sequencing ..................................................................................... 60 4.2.3 Assembly and annotation ................................................................................................ 61 4.2.4 Differential gene expression analysis ............................................................................. 61 4.2.5 Orthologous transcript identification .............................................................................. 61 4.2.6 Proteome analysis ........................................................................................................... 61

4.3 RESULTS .................................................................................................................................... 63 4.3.1 Sequencing, assembly and annotation ............................................................................ 63 4.3.2 Differential gene expression changes under temperature treatments .............................. 64 4.3.3 Frontloading of HSP Genes ............................................................................................ 75 4.3.4 Differential protein abundance under temperature stress ................................................ 75 4.3.5 Correlation of gene expression and protein abundance .................................................. 88

4.4 DISCUSSION ............................................................................................................................. 90 4.4.1 Transcriptome changes in response to heat stress ........................................................... 90 4.4.2 Proteome changes in response to heat stress ................................................................... 92 4.4.3 Conclusion ...................................................................................................................... 94

CHAPTER 5: GENERAL DISCUSSION ........................................................... 95 5.1 DE NOVO ASSEMBLERS IN MOLLUSC TRANSCRIPTOME ASSEMBLIES .................... 95 5.2 STRESS GENE EXPANSION IN LOW VS MID INTERTIDAL SPECIES ............................. 97 5.3 GENE EXPRESSION IN INTERTIDAL ZONES ...................................................................... 98 5.4 RESILIENCE UNDER FUTURE CLIMATES .......................................................................... 99 5.5 PROTEIN AND GENE CORRELATION ................................................................................ 100 5.6 CONCLUSION ......................................................................................................................... 101

REFERENCES ....................................................................................................... 102

APPENDIX A – POSTER PRESENTATIONS .................................................. 118

APPENDIX B – OUTCOMES FROM ASSOCIATED WORKS ..................... 122

ix

List of Figures

Figure 1. The intertidal zone ..................................................................................... 16

Figure 2. Nerita melanotragus and N. albicilla distribution ..................................... 18

Figure 3. Black nerite (Nerita melanotragus) ........................................................... 31

Figure 4. Nerita melanotragus transcriptome functional annotation based on Trinity Blast2GO analysis ........................................................................... 37

Figure 5. Nerita melanotragus transcriptome functional annotation based on Oases Blast2GO analysis ............................................................................ 38

Figure 6. GO category assignment ............................................................................ 39

Figure 7. GO category assignment ............................................................................ 40

Figure 8. Visualisation of PCR products ................................................................... 41

Figure 9. Orthologous genes between Nerita melanotragus, N. albicilla, Crasosstrea gigas and Lottia gigantea at 88% similarity cutoff ................ 52

Figure 10. GO Classification of stress related genes for N. melanotragus and N. albicilla ....................................................................................................... 53

Figure 11. WEGO plot for differentially expressed genes of N. melanotragus and N. albicilla ............................................................................................ 65

Figure 12. WEGO plot for differentially expressed proteins of N. melanotragus and N. albicilla ............................................................................................ 87

Figure 13. Scatterplot comparison of gene expression and protein abundance for BWN (Nerita albicilla) and BN (Nerita melanotragus) at 14 °C, 22 °C and 31 °C .......................................................................................... 88

x

List of Tables

Table 1. Primer sequences for β-actin and NADH dehydrogenase subunit 5 ........... 33

Table 2. Assembly quality metrics ............................................................................ 34

Table 3. Annotation results ...................................................................................... 35

Table 4. Synonymous to non-synonymous substitution calculations ....................... 54

Table 5. Mean expression values for N. melanotragus differentially expressed genes under different temperature treatments ............................................. 66

Table 6. Mean expression values for N. albicilla differentially expressed genes under different temperature treatments ....................................................... 73

Table 7 Mean baseline expression of eat shock protein in Nerita melanotragus and N. albicilla ............................................................................................ 75

Table 8. List of proteins that show differential expression profiles, the name of their top BLAST hit and their putative functions for N. melanotragus ...... 76

Table 9. List of proteins that show differential expression profiles, the name of their top BLAST hit and their putative functions for N. albicilla ............... 80

xi

List of Publications

PEER REVIEWED PUBLICATIONS

Amin S, Prentis PJ, Gilding EK, and Pavasovic A. 2014. Assembly and annotation of

a non-model gastropod (Nerita melanotragus) transcriptome: a comparison of

de novo assemblers. BMC Research Notes 7:488.

CONFERENCE POSTER PRESENTATIONS

Amin S, Prentis PJ, Gilding EK, Collet C, and A Pavasovic. Identifying genes

involved in physiological adaptation of Nerita melanotragus to temperature

stress using comparative transcriptome sequencing. International Marine

Biotechnology Conference, Brisbane Convention and Entertainment Centre,

Brisbane 2013.

Amin S, Prentis PJ, Gilding EK, Collet C, and A Pavasovic. Comparative

transcriptomics and characterisation of temperature stress genes in two

intertidal marine snails from the genus Nerita. Big Biology and Bioinformatics

Symposium, QUT Brisbane 2014.

Amin S, Prentis PJ, Gilding EK, Collet C, and A Pavasovic. Comparative analysis of

differentially expressed genes in two intertidal marine snails from the genus

Nerita under different temperatures. Lorne Genome Conference Mantra Lorne,

Victoria 2015.

Amin S, Prentis PJ, Gilding EK, Collet C, and A Pavasovic. De novo sequencing and

comparative analysis of temperature stress genes in two intertidal marine snails

from the genus Nerita. B3 Symposium, QUT Brisbane 2015.

xii

SEMINARS

Trinity De Novo Assembler. Systems Biology, QUT Brisbane 2014.

xiii

Statement of Original Authorship

The work contained in this thesis has not been previously submitted to meet

requirements for an award at this or any other higher education institution. To

the best of my knowledge and belief, the thesis contains no material previously

published or written by another person except where due reference is made.

Signature: QUT Verified Signature

Date: September 2018

xiv

Acknowledgements

This thesis is dedicated to my father, Nazeer and my mother, Dilvin, who have

supported and guided me throughout my life, I love you both. I would also like to thank

my three younger sisters for their support.

Firstly, I would like to thank my supervisory team, Dr. Ana Pavasovic, Dr. Peter

Prentis, Associate Professor Christopher Collet and Dr. Edward Gilding, for their

support throughout the years. It is due to their excellent supervision, guidance and

advice that I was able to complete this journey. I would like to take this moment to

express my utmost gratitude to Ana and Peter for their time devoted to weekly

meetings and their constant support and guidance. You have both helped me discover

my capabilities and guided me in using them, both as my supervisors in the university

but also as friends outside the campus.

Secondly, I would like to thank my colleagues from the ePGL group for creating

such a warm and welcoming atmosphere to work in. It has been a joy to have worked

with each and everyone of you.

Thirdly, I would like to thank QUT for providing me with the opportunity to

conduct my studies. In particular, the School of Biomedical Science and the Faculty

of Health. I would also like to thank QUT’s Molecular Genomics Research Facility

(MGRF) for providing the utilities to conduct my experiments.

Finally, I would like to thank a special person who entered my life and supported

me up until the final stages, my loving wife Shiva.

15

Chapter 1: Literature Review

1.1 BACKGROUND

Phylum Mollusca is one of the most morphologically diverse animal groups. It

is the second largest animal phylum following Arthropoda. Mollusc species comprise

a great proportion of marine biodiversity, making up over 23% of all currently

recognised marine organisms [Giribet et al., 2006; Fiedler et al., 2010]. This highly

speciose group is important both ecologically and economically; as a fisheries

commodity, a source of bioactive and medically important compounds, being used as

models in the study of neuroscience, and in the study of adaptation to environmental

extremes [Hold et al., 2013; Feng et al., 2009; Asta Lakshmi, 2011].

The better-known classes within this morphologically diverse phylum are

Cephalopoda (e.g. squid, cuttlefish and octopus), Bivalvia (e.g. oysters, mussels and

arkshells), Polyplacophora (e.g. chitons) and Gastropoda (e.g. snails, limpets and

nudibranchs) [Passamaneck et al., 2004]. Gastropods represent the most diverse class,

both in number (80% of all taxa) and life form (hard shell, soft shell, colouration and

life history) and have colonised many geographically and ecologically diverse regions,

including the tropical and polar biomes [Faulkner, 2011]. With such great diversity, it

is not surprising that gastropods have successfully colonised a wide range of terrestrial

and aquatic environments, including some of the harshest environments on earth such

as hot thermal vents, cold seeps and the intertidal zone [Khripounoff et al., 2017;

Sigwart et al., 2017]. Gastropod species are most common in marine environments,

and are particularly abundant in the rocky intertidal zone (Figure 1) [Miloslavich et

al., 2013]. The rocky intertidal zone is defined as “the part of the seafloor that lies

between the highest high tide and the lowest low tide” (Castro & Huber, 2003). It is

periodically submerged and exposed by daily tides (Barnwell, 1968). The intertidal

zone is a rapidly changing environment with great changes in both biotic and abiotic

conditions over a tidal cycle. Consequently, it is characterised as an extreme

environment, with great variation in air and water temperature, salinity, oxygen and

food availability, as well as constant pounding from wave energy [Gracey et al., 2008].

The rocky intertidal zone can be further subdivided into four distinct areas at

characteristic heights or distance from the shoreline. These areas are the spray zone,

16

upper intertidal zone, mid intertidal zone, and lower intertidal zone (Figure 1). This

subzonation results from changes in biological and physical characteristics associated

with tidal dynamics, wave exposure, temperature, salinity, substrate orientation and

composition, amongst others (Castro & Huber, 2003; Harly & Helmuth, 2003; Miura

et al., 2014). The upper intertidal zone is only immersed at the peak of high tide, and

the splash zone is almost never immersed, with wave splash the only source of

moisture (Reece & Campbell, 2011). The mid and lower intertidal zones are

submerged much of the time, but have high wave energy and strong currents

Temperature, food availability, and dissolved oxygen levels also vary greatly among

the subzones.

Figure 1. The intertidal zone, illustrating organism exposure to the tidal cycle. Adapted from https://www.oceanclassrooms.com/ms101_u7_c2_sd_1-.

The changes in environmental conditions among the zones also leads to large

changes in the composition, density and diversity of gastropod species across the areas

(Shotwell, 1950; Haven, 1970). This often leads to strong habitat partitioning of

gastropod species among these zones, which is a process referred to as zonation

(Underwood & Jernakoff, 1981). As a consequence, intertidal gastropods from

17

different areas will have marked differences in their exposure to changes in many

environmental variables including temperature, pH, salinity, dissolved oxygen and the

availability of food as tidal conditions change [Firth et al., 2011; Van Dyck et al.,

2015]. Fluctuating abiotic conditions are key regulators of metabolic activity in marine

gastropod species and are fundamental in controlling the rate of numerous biological

processes such as development rate, physiology and survival [O’Connor et al., 2007].

Intertidal gastropod species have a range of physiological adaptations to cope,

survive and thrive in these different areas. These adaptations include the extensive

duplication of heatshock protein 70 and inhibition of apoptosis proteins, as observed

in the Pacific oyster (Crassostrea gigas) [Wang et al., 2012], as well as the

frontloading of stress response genes in species from the upper intertidal zone that

encounter prolonged periods of thermal stress [Dong et al., 2008]. Another example

in a non-model organism is the RNA-seq experiment in the marine snail Chlorostoma

funebralis, revealing differing stress responses across populations of the same species

[Gleason and Burton, 2015]. Specifically, this experiment found differences in the

onset and expression of heat shock proteins between the two populations, with the

more resilient population having upregulation of genes in response to stress under

baseline conditions. This phenomenon better known as frontloading of expression is

suggested to have resulted from an evolutionary history of frequent heat exposure.

These examples show some of the mechanisms and physiological adaptations

proposed to allow gastropod species to withstand extreme environmental conditions in

the intertidal zone. While this recent research has started to illustrate how mollusc

species persist in stressful environments, physiological genomic research focusing on

physiology of gastropod species is limited.

Research into the physiological genomics of gastropod species has lagged

largely because the genomic, proteomic or expressed sequence tag (EST) resources

required to conduct this type of research have not yet developed for most species. In

fact, only a few marine invertebrate species have had significant genomic resources

developed for this purpose, including the California sea hare (Aplysia californica)

[Heyland et al., 2011], the striped venus clam (Chamelea gallina) (Coppe et al., 2012),

the owl limpet (Lottia gigantea) [Simakov et al., 2013] and the California two-spot

octopus (Octopus bimaculoides) [Albertin et al., 2015]. While genomic resources

developed for mass cultured invertebrates such as the Pacific oyster (C. gigas) [Wang

18

et al., 2012] may be applied to their non-model relatives, this does not provide a

comprehensive foundation to undertake physiological genomic studies in these

species. Two non-model marine gastropod species Nerita melanotragus and Nerita

albicilla (Figure 2 (A and B)) are both found in the intertidal zone and have large

geographic distributions in Australia and New Zealand (Figure 2 (C)) [Crandall et al.,

2008]. Currently, little is known about how the physiological genomics of N.

melanotragus and N. albicilla show specific genetic changes to allow them to persist

in intertidal environments. This paucity of knowledge is primarily due to a lack of

genomic resources for both species, which are required to measure gene expression

and proteomic changes in response to environmental variables. This genetic data will

enable for the first time, an insight into the molecular physiology of these species and

which genes and pathways are important for their survival and persistence in an

extreme and dynamic environment.

Figure 2. Nerita melanotragus and N. albicilla distribution. N. melanotragus (A) and N. albicilla (B) (Amin, S) distribution across Australia and New Zealand, including the collection site which is indicted with the arrow (C) (adapted from http://noavg.me/map-of-australia-and-nz/new-zealand-map-blank-political-with-cities-throughout-of-australia-and-nz/). N. melanotragus distribution is highlighted in blue, while N. albicilla distribution is represented in orange.

1.2 CLIMATE CHANGE AND ENVIRONMENTAL STRESS

Under many climate change scenarios, marine species are expected to experience

increased environmental stress in their habitats as a result of changes in environmental

19

conditions [Roessig et al., 2004; Harley et al., 2006]. For example, species that occur

in coastal ecosystems will experience possible increases in water temperatures and sea

level, reduced pH and decreased oxygen availability [Scavia et al., 2002]. Publications

such as IPCC (International Panel on Climate Change) provide extensive reporting on

predicted future climate change scenarios and based on this and a multitude of research

articles, ocean temperatures are expected to increase by at least 3 °C over the next

century with cold water habitats expected to be the most affected [Hungate et al., 2003;

Zeh et al., 2012; Albright and Mason, 2013]. Large increases in sea surface

temperatures will also affect a number of other biotic and abiotic variables in the ocean

including, primary productivity and oxygen saturation [Hinrichsen et al., 2011].

Changes in oxygen saturation and water temperature are likely to have negative

consequences on a range of stress sensitive species, including some species of

cnidarians, molluscs, arthropods and other marine invertebrates [Jakubowska and

Normant, 2015; Wijgerde et al., 2014].

Intertidal gastropod species are likely to be subjected to sea level increases,

higher temperatures and decreasing pH levels [Fabry et al., 2008; Pörtner and Peck,

2010; De Wit and Palumbi, 2013] associated with environmental warming increases

in sea surface temperatures as a consequence of climate change [Burrows et al., 2011;

Lima and Wethey, 2012]. Specifically, mollusc species are identified as one of the

most sensitive taxa (Helmuth et al., 2006; Byrne et al., 2013; Przeslawski et al., 2015).

These environmental changes will have a large impact on a range of biological

processes in these species [O’Connor et al., 2007], which may also result in the

contraction or expansion of species ranges [Southward et al., 1995]. Based on the

predicted changes in marine environments under climate change, it is also likely that

the extent of environmental stress experienced by many marine gastropod species will

increase dramatically in the near future. As a result, it is predicted that a broad range

of marine species will be affected in terms of growth, reproduction and survival

[Pespeni et al., 2013]. Consequently, it is important to better understand how marine

species currently cope with environmental stress, to better understand how species may

respond in the future.

20

1.3 MARINE ANIMAL RESPONSE TO ENVIRONMENTAL STRESS

Marine organisms occurring in the intertidal zone frequently experience

fluctuations in environmental abiotic factors such as water temperature, tidal

immersion, desiccation, oxygen levels, ultraviolet radiation and humidity [Menge and

Branch, 2001]. However, how marine species respond to changes in environmental

conditions is likely to be both varied and complex, even in closely related species.

Some species will likely be able to adapt to future climatic conditions from standing

genetic variation, while others may respond to climate variability through changes in

gene expression or even in extreme cases, go locally extinct if they are already close

to their upper thermal maxima. While many studies have attempted to determine how

marine species will respond to environmental variability e.g., Harley et al., 2006, for

the purposes of the current review, recent representative examples from key marine

taxa will be the focus.

The most well studied marine organisms are corals, a group characteristically

sensitive to temperature stress and ocean acidification [Alley et al., 2003]. In a recent

study, Barshis et al., (2013) examined genome wide patterns of gene expression in

thermally sensitive and thermally resilient populations of Acropora hyacinthus, under

projected future temperature conditions. The study focused on the molecular pathways

contributing to stress response differences between the two groups of corals.

Interestingly, resilient coral populations had higher baseline expression of 60 genes

that showed lower levels of up-regulation when compared to sensitive populations

under thermal stress. These 60 genes were enriched for those involved in stress

response such as heat shock proteins, antioxidant proteins and genes involved in

apoptosis regulation. This data suggests that higher baseline expression (referred to as

frontloading) of stress response genes may make marine species more resilient to

environmental stress. Overall, such data provides a strong indication that a species

response to environmental stress is not limited to a specific gene or gene family, but

rather a suite of genes involved in diverse physiological processes.

In contrast to the previous example, a study on multiple populations of the model

echinoderm species, purple sea urchin (Strongylocentrotus purpuratus), investigated

their response to climate change and environmental stress under realistic future CO2

levels [Pespeni et al., 2013]. The study found little evidence of morphological or

developmental changes to elevated CO2 levels, but did find genetic changes. At a

21

genetic level, substantial allelic change was observed in 42 functional classes of

proteins, with an excess of amino acid replacements detected in all populations. The

set of genes that showed an excess of amino acid replacements was enriched for

proteins involved in biomineralization, lipid metabolism, and ion homeostasis. This

result is intuitive as these gene classes are involved in exoskeleton formation and

maintenance, and pH regulation. These are all processes known to be affected by ocean

acidification. Results of this study found that all populations were able to rapidly adapt

to these conditions and were resilient to increased environmental stress associated with

ocean acidification. Standing genetic variation at physiologically important genes,

therefore, may allow select marine species to adapt to future climates in some cases.

Different developmental stages of biphasic organisms may respond to climate

change in different ways as well. An example of this comes from the widespread

sponge species, Rhopaloeides odorabile, a species whose adult form is highly sensitive

to temperatures above 32 °C, yet their larvae can withstand temperatures above 36 °C

[Webster et al., 2013]. This study examined gene expression changes that underpin

these contrasting thermal tolerances and found that heatshock proteins were

significantly up-regulated in adults at lower temperatures than in larval stages.

Importantly, this study also found that genes involved in signal transduction, protein

synthesis/degradation, oxidative stress and detoxification were all down regulated in

adults exposed to temperature stress. This data provides important baseline

information about the effects of temperature stress across different developmental

stages. Coupled with phenotypic data, this can be used to monitor the long-term impact

of climate change on sponge population dynamics.

There is now clear evidence that environmental variability associated with

climate change and climate variability induces significant stress on an organism’s

ability to persist in their habitat (Lathlean & Minchinton, 2012). This is often most

evident in exothermic species whose metabolic processes are highly dependent on

external temperatures [Van Dyck et al., 2015]. Therefore, identifying the genes or

pathways that are responsible for maintaining internal homeostasis within organisms

is important for understanding how marine invertebrates cope with changes within

their environments [Walsh and Jefferis, 2006]. Studying which molecular processes

are impacted by environmental stress is significant and provides a basis for

22

understanding how species respond to such stress and persist in their habitats [Lemos

et al., 2010].

1.4 THE CELLULAR STRESS RESPONSE

The cellular stress response (CSR) is a universal cellular defence mechanism

which is initiated in response to changes in the extracellular environment [Kültz,

2005]. Depending on the duration and severity of stress, cells either reinstate the

process of cellular homeostasis to the previous state or take on a distorted state in the

new and changed environment [Booth and Bilodeau-Bourgeois, 2009]. The response

includes control of the cell cycle, DNA and chromatin stabilization and repair, removal

of damaged proteins, protein chaperoning and repair, as well as various metabolism

functions [Kültz, 2003]. Consequently, different stressors and levels of stress can

generate different cellular responses, which can induce alternative mechanisms of cell

repair that allow the cell to re-establish normal function [Bakkenist and Kastan, 2004].

The CSR is comprised of a large suite of genes/proteins involved in multiple

cellular processes, with key proteins responsible for major functions being conserved

across phyla within a kingdom [Kültz, 2005]. In fact, over 300 stress related proteins

are considered to be highly conserved at the amino acid level among metazoans, with

44 having known functions in CSR [Kültz, 2005]. The proteins that have been

functionally categorised through gene ontologies have been shown to contribute to

different aspects of the cellular stress response. Among the genes/proteins that have

been functionally categorised are those involved in redox regulation, DNA damage

sensing/repair, molecular chaperones, protein degradation, fatty acid/lipid metabolism

and energy metabolism (Fulda et al., 2010).

The expression of several conserved genes involved in the CSR, have been

shown to be strongly induced under periods of stress [Beck et al., 2000], such as

HSP60, HSP70, peroxiredoxin and superoxide dismutase. Induction of expression for

these genes has been linked to changes in abiotic factors such as temperature, oxidative

stress and water contamination among others [Snyder et al., 2001; Cong et al., 2009;

Casellato et al., 2012]. Although some components of the CSR such as the heat shock

proteins (HSP’s) have been extensively studied and their expression patterns in

response to stress well established, the function and expression of many genes

involved in the CSR are yet to be examined in many taxa. Novel proteins have also

23

been found to play a role in CSR in some lineages, suggesting that more information

can be discovered by investigating stress response in non-model organisms [Kültz,

2005] such as marine molluscs. The stress response of numerous marine species has

been understudied and this is particularly true of diverse invertebrate phyla such as

cnidarians and molluscs.

1.5 MARINE GASTROPODS AS A SYSTEM TO INVESTIGATE TEMPERATURE STRESS

To investigate which genes are important components of stress response in

marine gastropods will require detailed studies of how environmental stress influences

patterns and levels of gene expression and protein abundance [Mohamed et al., 2014].

Currently, genome wide studies on stress response in marine mollusc species have

been largely restricted to the economically important species, such as the Pacific oyster

(Crassostrea gigas) and far fewer exist for gastropod species. Examinations of genome

wide patterns of gene expression and protein abundance in response to environmental

stress are only recently being investigated in intertidal gastropods [but see; Gleason

and Burton, 2015; Sun et al., 2012].

In the bivalve mollusc (C. gigas) Wang et al., (2012) sequenced multiple

transcriptomes for multiple individuals in both ambient and high temperature

treatments to investigate genome-wide responses to temperature stress. Transcript

abundance could be compared among thermally stressed and control groups using

patterns of gene expression derived from direct RNA sequencing. In individuals

subject to thermal stress, there was an almost 2,000 fold induction in multiple heat

shock protein 70 genes, indicating that these genes play an important role in thermal

stress in intertidal bivalve mollusc species. A number of other genes involved in

cellular homeostasis, including calreticulin and calnexin, were also up-regulated under

thermal stress. This indicates that genes involved in protein quality control and

refolding are critical to the thermal stress response in this species and may be one

reason that C. gigas can tolerate dramatic changes in temperature during tidal cycles

[Wang et al., 2012].

Sequencing of the C. gigas genome also revealed it contained an expanded

number of stress response genes when compared to humans and other terrestrial and

marine species. In fact, the oyster genome was found to contain 88 functional heat

shock protein 70 genes (HSP70) compared to 17 in the human genome and 39 in the

24

sea urchin genome [Wang et al., 2012]. Wang et al., (2012) also conducted

phylogenetic analyses and found that 71 of the HSP70 genes formed their own cluster,

indicating that the expansion may be specific to the oyster or bivalve lineage. The

oyster genome also contained 48 duplicate copies of inhibition of apoptosis proteins

compared to 8 in humans and 7 in sea urchins, which suggests that oysters may have

a powerful anti-apoptosis system [Reusch, 2014]. Several other stress related genes

were found to have undergone expansions in the oyster genome, including cytochrome

P450, multi-copper oxidase and extracellular superoxide dismutases. This study

highlights that although many aspects of stress response are conserved between

species, many lineage specific mechanisms may also play a large role in response to

environmental stress [Brierley and Kingsford, 2009].

While this landmark study has begun to elucidate the genomic and gene

expression response of intertidal mollusc species to environmental stress, there have

been few in depth characterisations of stress response at a genetic level in molluscs

beyond model taxa [but see Gleason and Burton, 2016]. It remains unknown whether

the findings from this study can be applied to most gastropod species or if stress

response varies widely among different species, particularly those from different areas

of the intertidal zone. As a consequence, we are unable to predict how marine

gastropod species from different intertidal areas will respond to increased

environmental stress associated with global climate change.

1.6 PHYSIOLOGICAL INVESTIGATION INTO REPRESENTATIVE SPECIES FROM THE GENUS NERITA

Understanding gene expression and proteomic responses of specific species to

temporal and spatial changes in their external environments is a key area of research

in physiological genomics. To conduct this research, it requires an appropriate group

of species that inhabit a temporally and spatially fluctuating environment. The genus

Nerita is a highly diverse and abundant group of organisms in the class gastropoda,

with an extensive pantropical distribution [Crandall et al., 2008]. Members of this

genus are abundant components of the intertidal fauna in a range of areas around the

world [Spencer et al., 2007]. Across their range, Nerita have exploited a variety of

environmental niches (different areas of the intertidal zone), and thrive in intertidal

habitats, which exhibit extreme fluctuations in temporal and spatial conditions

[Chapperon et al., 2013]. One common feature of Nerita species inferred from

25

biogeographic studies is that most species that occur in sympatry are ecologically

segregated into different areas of the intertidal zone [Crandall et al., 2008].

Nerita melanotragus and N. albicilla are widespread intertidal gastropod

species, which co-occur in south-east Queensland, and are distributed across a number

of temporally and spatially fluctuating environmental gradients, including a variety of

mean water temperatures (Waters et al., 2005; Postaire et al., 2014). These two species

have colonised different areas of the intertidal zones, with N. melanotragus in the mid

intertidal zone, while N. albicilla inhabits the low intertidal zone. Colonisation of

different areas of the intertidal zone means that N. melanotragus is more likely to

sustain long periods of temperature fluctuation, while N. albicilla is more frequently

subjected to immersion and more buffered from temperature fluctuation.

Consequently, these two species present an interesting case to examine their repertoire

of stress-related genes, patterns of differential gene expression and protein abundance

when placed in different thermal environments.

Understanding the genetic basis of physiological traits that are associated with

environmental variations has been limited by a lack of genomic resources for both

species. This study aims to conduct large scale sequencing of the N. melanotragus and

N. albicilla transcriptomes to generate genomic resources for functional genomic and

proteomic analyses. This resource will also allow us to determine if there is variation

in copy number of stress related genes between the two species. A large-scale gene

expression and proteomic study will then examine which genes and proteins are

differentially expressed/abundant in response to temperature changes. Despite the fact

that gastropod species are key indicator species for a suite of stressors, little research

has been undertaken on the physiological genomics of these species or how the

expression of these genes are affected by abiotic environmental variables [but see

Wang et al., 2008, Gleason and Burton, 2015]. The data generated from this research

will provide a significant contribution to the understanding of physiological

mechanisms that gastropod species use to cope with thermal stress, which has

applications in the development of indicator species that can be used to measure

thermal stress in marine environments.

26

1.7 RESEARCH QUESTION AND APPROACH

This project focused on investigating the molecular stress response of two key

intertidal gastropod species, Nerita melanotragus and a closely related warm water

species N. albicilla. To achieve this, genomic resources were developed, and this data

was interrogated to determine candidate genes involved in physiological adaptation

and stress response to abiotic environmental conditions on both temporal and spatial

scales. In addition, the project measured patterns of gene expression and protein

abundance of the key genes/proteins and pathways involved in temperature stress

response in these species, in a series of in vivo experiments providing a basis for our

understanding of physiological response to thermal stress in marine gastropod species.

In order to increase our understanding of stress response and adaptation in

marine invertebrates, this study addressed key knowledge gaps by focusing on two

representative gastropod species, N. melanotragus and N. albicilla. The study

consisted of the following objectives:

1.7.1 Research objective 1

Establishment of a reliable, efficient and accurate de novo assembler for Nerita

spp.

This objective corresponds to the first study of the project. The key objective of

this study is to critically compare the reliability and accuracy of current de novo

assemblers and to establish the most suitable, for species from the genus Nerita. Using

several quality metrics, four de novo assemblers (Geneious, Velvet, Oases and Trinity)

will be used to reconstruct a transcriptome assembly from curated raw read data. The

establishment of a reliable and accurate assembler will ensure the quality of

downstream analysis when investigating the gene repertoire, gene expression and

protein abundance patterns of the Nerita spp. under different temperatures.

1.7.2 Research objective 2

Development of genomic resources and comparative analysis of expressed stress

genes in Nerita melanotragus and N. albicilla.

This objective corresponds to the second study of the project. The key objective

of this study was to generate transcriptome assemblies from the two species from the

genus Nerita, to address the lack of genomic resources available for both species, and

27

gastropods in general. Nerita melanotragus is a temperate, cold water adapted species,

while N. albicilla is a tropical, warm water adapted species. Simultaneous sequencing

of the two species provided data for comparative genomic analysis to identify

candidate genes that were involved in temperature stress response, as well as an array

of other genes that regulate physiological processes. This comparative genomic

approach, in combination with annotation to specific gene ontologies allowed us to

determine the copy number of expressed genes important in thermal stress response in

these two intertidal gastropod species. Downstream application of this data

encompassed identification of potential stress response biomarkers and an extensive

set of candidate genes were used in a detailed investigation of the response of N.

melanotragus and N. albicilla towards temperature fluctuations.

1.7.3 Research objective 3

Comparative transcriptomic and proteomic characterization of genes/proteins

and gene families involved in the abiotic stress response of two intertidal marine snails

inhabiting different areas of the intertidal zones.

This objective will correspond to the third study of this project. The objectives

of this study are to identify genes involved in temperature stress response, i.e. those

that are differentially expressed in the two Nerita species under different temperature

conditions. This study will focus on gene expression and protein abundance at different

temperatures that correspond to the extreme range of water temperatures that these

species endure during periods of thermal stress. The information generated by this

study will help to better understand if species from lower intertidal zone experience

thermal stress at lower tempertures, compared to those from the mid-upper intertidal

zones.

29

Chapter 2: Assembly and Annotation of a Non-Model Gastropod (Nerita melanotragus) Transcriptome: a Comparison of De novo Assemblers

2.1 BACKGROUND

The phylum Mollusca is a highly abundant group of marine animals accounting

for over 23% of all marine species (Ponder and Lindberg, 1997; Hou et al., 2011), and

as such are a dominant taxa of many marine ecosystems. Some of these organisms are

also of significant economic importance as a source of bioactive compounds in

addition to being aquaculture and fisheries commodities. Molluscs also serve as

valuable models for behavioural neurobiology, respiration and feeding in animals

(Peterson, 2002; Herpin et al., 2004; Feng et al., 2009; Sadamoto et al., 2012).

Consequently, molluscs are very important both economically and ecologically.

However, genomic resources remain scarce for Mollusc species, with transcriptome

data available for only selected species such as Crassostrea gigas, Macoma balthica,

Aplysia californica and Lymnaea stagnalis (Fiedler et al., 2010; Feng et al., 2009 Pante

et al., 2012; Zhao et al., 2012). Despite the availability of these genomic resources,

this group of organisms remains relatively poorly studied at the genomic level.

Research into the genomics of gastropod molluscs has lagged, because genomic

resources are not developed for many species in this class of organisms. Next

generation sequencing platforms such as Illumina and Ion Torrent have recently been

used to rapidly characterise transcriptome sequences from a number of non-model

organisms (Chiara et al., 2013; Li et al., 2013; Hook et al., 2014; Schunter et al., 2014).

In this study, the Ion Torrent platform is employed, an efficient and cost effective

platform to sequence the Nerita melanotragus transcriptome, a non-model species

without a reference genome.

30

Precise and accurate de novo assembly and annotation of transcriptomes,

however, is a commonly overlooked but critically important step for assemblies

generated from short reads (~100-150 bp). Recently, many de novo assemblers have

been developed with specific algorithms for transcriptome assembled from Illumina

short reads., nonetheless their effectiveness for de novo assembly of Ion Torrent data

remains relatively unexplored as the Ion Torrent technology is newer and still gaining

acceptance in the research community. Accurate assembly of short reads into longer

contigs is important for the functional annotation of ESTs in non-model organisms. In

fact, one of the major challenges for genomic research in mollusc species is that many

genes remain unannotated. In this study, these issues are addressed by comparing the

performance of a number of short read de novo transcriptome assemblers, specifically

using Ion Torrent sequence data in Nerita melanotragus.

The black nerite (N. melanotragus) is a marine gastropod within the phylum

Mollusca. This species inhabits the intertidal zone and has a large geographic

distribution from central Queensland, Australia to southern New Zealand (Crandall et

al., 2008). Specifically, these species are found in rockpools where they are found to

occur in groups. This gastropod species is identified as a herbivore, grazing algae from

rocks. The environmental conditions that this species is exposed to change temporally

and spatially, on both micro and macro geographic scales. Thus, this organism is a

good candidate to explore the genetic and gene expression changes, which allow it to

persist in such a dynamic environment. Little is known about adaptation genetics and

plastic gene expression changes in N. melanotragus due to an explicit lack of genomic

resources. To address this issue, the first de novo assembly of the N. melanotragus

transcriptome is reported here. Specifically, this study focuses on addressing the

following aims: 1. to generate genomic resources for this species through whole

organism transcriptome sequencing; and 2. to assess the accuracy and precision of four

different short read de novo transcriptome assemblers using specific quality metrics.

31

2.2 METHODS

2.2.1 Sample acquisition and sequencing

Five Black nerite, N. melanotragus, (Figure 3), individuals were collected from

the rocky intertidal zone at Caloundra, Queensland, Australia (26°48′17“S

153°8′28”E). Ethics approval and collection permits/licenses were not required for

specimen collection. Individual animals were classified as N. melanotragus based on

operculum colour (Waters et al., 2005). A single individual was snap frozen in liquid

nitrogen (LN2) and stored at −80°C until RNA extraction. The frozen tissue sample

from the whole organism was homogenised in LN2 and total RNA was extracted using

a Trizol/Chloroform extraction protocol followed by a clean-up using an RNeasy

Minikit (Qiagen). RNA samples were treated using Turbo DNase (Ambion), according

to manufacturer’s protocol.

Figure 3. Black nerite (Nerita melanotragus). Black nerite displaying external morphology (A), and black nerite displaying tan/brown colouration of its operculum (B).

To check the quantity and integrity of the total RNA, the sample was run on a

Bioanalyzer 2100 RNA Nano chip (Agilent Technologies). Messenger RNA was

isolated from total RNA using the Dynabeads mRNA Purification Kit (Life

Technologies). A Bioanalyzer 2100 Pico chip (Agilent Technologies) was used to

determine the quality and quantity of isolated mRNA.

High quality mRNA (100–500 ng) was fragmented into 200–700 bp pieces using

RNase III (Life Technologies) and Agincourt beads were used to remove small RNA

fragments from two samples. The yield and size distribution of fragmented RNA was

determined on a Bioanalyzer 2100 using an RNA 6000 Pico chip (Agilent

Technologies). Library construction was conducted on a single sample as per the Ion

A B

32

Total RNA-Seq Kit (Life Technologies) for whole transcriptome libraries and cDNA

yield and size was determined using a Bioanalyzer 2100 high sensitivity DNA chip.

Template preparation for sequencing was conducted on a single sample

according to the OneTouch Ion™ Template Kit (Life Technologies). Ion Torrent

sequencing was conducted using the Ion PGM 200 Sequencing Kit (Life

Technologies) on an Ion Torrent Personal Genome Machine (PGM, Life

Technologies) using a 318-chip (Ion 318TM chip, Life Technologies).

2.2.2 Assembly and annotation

Raw sequencing reads were converted to FastQ files and assessed for quality

scores. Reads were accepted based on a quality threshold (Q > 20, ambiguous bases

less than 1%), and adapter sequences were removed prior to downstream analyses. To

critically assess the quality of this Ion Torrent data, a number of analytical approaches

described by Wheat and Vogel, 2011 were conducted, including the sequencing depth

and coverage for expressed genes from the publically available mitochondrial genome

of N. melanotragus using Geneious Pro (Version 5.6) (Drummond et al., 2010).

High quality reads were assembled into contiguous sequences (contigs) using

four different short read de novo assemblers, which included the following: 1)

Geneious Pro (Version 5.6) (Drummond et al., 2010); 2) Velvet short read assembler,

Version 1.2.08 (Zerbino and Birney, 2008); 3) Oases short read assembler, Version

0.2.08 (Schulz et al., 2012) and 4) Trinity short read assembler (Grabherr et al., 2011).

All assemblers except Geneious used the following assembly parameters: kmer hash

length = 25, coverage cut-off = 3x; minimum contig length = 100 bp. In the Geneious

software, kmer hash length and coverage cut-off could not be changed, so default

settings were used with a minimum contig length of 100 bp. The assembly created by

the four different assemblers were compared for three different parameters; the number

of contigs produced, the N50 statistic and the longest contig to determine which

assembler performed best.

To determine the redundancy of the assemblies produced by the four different

assemblers, assembled datasets were remapped to the mitochondrial reference gene set

from N. melanotragus (publically available from NCBI). All contigs produced by the

four different assemblers were remapped to this gene set and the overall number of

hits was calculated as a quality score.

33

Following contig generation, the transcriptome assemblies for Trinity and Oases

were referenced to the NR database at NCBI as BLASTx queries using the Blast2GO®

software suite (Conesa et al., 2005). In order to be used in downstream analyses,

BLASTx hits had to be below an E-value of 1 × 10−6. Annotation analyses were

performed at levels 2 and 3. The Blast2GO software suite was also used to predict the

functions of contigs with BLASTx hits and assign Gene Ontology (GO) terms to the

sequences. To determine which of the short read assemblers produced the best

assembly of our Ion Torrent data, BLAST and annotation success of these different

datasets was compared.

To validate the reliability and accuracy of the assembly and annotation, two

contigs (annotated as β (beta) - actin and NADH dehydrogenase subunit 5) were

randomly chosen for primer design and for PCR and Sanger sequencing. Primers were

designed using BatchPrimer3 (Version 1.0) using settings as per (Sexton et al., 2010).

Details of the primer sequences are provided in Table 1. PCR was performed according

to the MyTaq (Bioline) protocol with the following concentrations of reagents 1 × PCR

Buffer, 1 μM of each primer, 0.1 units of MyTaqTM DNA Polymerase (Bioline) and

20 ng of template genomic DNA (from same individual that was sequenced) in a total

volume of 25 μL. PCR conditions were as follows: 3 min at 94°C, followed by 30

cycles of 30 sec at 94°C, 30 sec at 52°C, 30 sec 72°C, 3 min at 72°C. Amplicons were

purified using the Isolate PCR Kit (Bioline) and cycle sequencing was carried out

using BigDye Terminator v3.1 Cycle Sequencing Kit (Life Technologies). After a

MgSO4 clean-up, the amplicons were run on an ABI 3500 Genetic Analyzer (Life

Technologies). Sequences were visualised and edited by eye using Geneious Pro

Version 5.6. These sequences were then used as BLASTn queries against the

nucleotide database at NCBI and were compared for differences against the original

sequences.

Table 1. Primer sequences for β-actin and NADH dehydrogenase subunit 5. Primers were designed to validate the reliability and accuracy of our assembly and annotation.

Primer name Primer sequence Fragment length (bp)

Beta-actin-Nm (F) GAAGCTGTGCTATGTTGTCCTC 450

Beta-actin-Nm (R) GATCTTGATCTTCATGGTGCTG 450

NADH (F) GGCGCATTAGCATCTCAAAT 414

NADH (R) GCTCCTGCAAGGGTAACTGA 414

34

2.3 RESULTS

2.3.1 Ion torrent sequencing and reads assembly

Transcriptome sequencing of mRNA from N. melanotragus on the Ion Torrent

PGM platform generated a total of 249.67 Mbp of sequence from 1,883,624 raw reads

(accession number SRR1054996). Mean length of reads was 133 bp, with the longest

read being 392 bp. Sequence reads that did not meet our strict quality criteria (Q < 20,

ambiguous bases > 1%) were excluded and 84.19 Mb of high quality data was retained

for downstream analysis, as low quality bases are likely to reduce the accuracy of

transcriptome assemblies.

Based on high quality reads, a total of 112, 762, 78, 306, 10, 886 and 3, 090

contigs were generated using the following four different assemblers Geneious,

Velvet, Trinity and Oases, respectively (Table 2). Overall, the Oases assembly

produced the longest contig at 1700 bp closely followed by Trinity at 1618 bp. The

longest contigs produced by Geneious and Velvet were over 700 bp shorter than the

other two de novo assemblers (Table 2). The length of the N50 statistic in the Geneious,

Velvet and Oases assemblies were noticeably shorter than that calculated for the

Trinity assembly (Table 2). Average contig length showed a similar trend, with the

Trinity assembly also having the longest average contig length.

Table 2. Assembly quality metrics. Assembly statistics for the transcriptomes produced by the four different short read de novo assemblers

Both Velvet and Geneious assemblies had a greater number of contigs remapped

to the mitochondrial expressed gene set with 450 and 420 hits respectively, compared

to 37 and 25 hits for Trinity and Oases, respectively. The coverage of the contigs

produced by both Trinity and Oases was greater than 95%, while the coverage

produced by the Geneious and Velvet contigs was less than 55% for both assemblies.

Assembly Statistic AssemblerOases Trinity Velvet Geneious

Number of contigs 3 090 10 886 78 306 112 762 Average contig length 175 293 111 140 Longest contig 1 700 1 618 458 711 N50 149 258 107 124

35

The Geneious and Velvet assemblies were found to be highly redundant and produced

more fragmented contigs. As the reliability of these two assemblies was not be

established, , they were disregarded from further analyses.

Remapping of high quality reads to the transcribed genes in N. melanotragus

mitochondrial genome resulted in an assembly with an average of approximately 374

× read depth and greater than 99.5% coverage. The sequencing depth was highest for

the 16S rRNA gene with >2000 × read depth. All genes had coverage of greater than

98% with the lowest coverage occurring in NADH dehydrogenase subunit 2, which

contained a 40 bp region with no coverage.

2.3.2 Functional annotation of contigs

Of the 10886 and 3090 contigs queried against the NR database only 2069 and

475 returned significant hits at greater than 1 × 10−6 stringency (Table 3). This meant

that approximately 19 and 15.4% of contigs could be assigned putative functions for

the Trinity and Oases assemblies, respectively.

Table 3. Annotation results. The number of contigs allocated to different annotation categories for the Trinity and Oases assemblies.

Annotation category Annotation result (number of sequences)

Trinity Oases

Without blast result 0 (0%) 0 (0%) Without blast hits 8823 (81%) 2615 (84.6%) With blast result 301 (2.7%) 66 (2.1%) With mapping result 177 (1.6%) 28 (0.9%) Annotated sequences 1585 (14.5%) 381 (12.3%) Total sequences 10886 3090

Despite the limited number of contigs assigned BLAST hits, the contigs

generated by both Trinity and Oases captured a broad range of different types of

transcripts, as indicated by the variety of Gene Ontology (GO) terms assigned. A total

functional annotation dataset is provided in Figures 4 and 5, while the results for the

top 20 GO terms for each category are presented in Figure 6. The GO category with

the highest number of terms assigned was molecular function, followed by cellular

component while biological process had the least contigs assigned terms. The most

commonly assigned GO terms in the molecular function GO category were the

housekeeping genes involved in ATP binding, protein binding and structural

constituent of ribosome for both assemblies. Oxidation-reduction process, translation

36

and translational elongation were the most commonly assigned terms for the biological

process GO category. The three most commonly assigned GO terms for cellular

component were cytosol, cytoplasm and nucleus, and cytoplasm, integral to membrane

and mitochondrion for the Trinity and Oases assembly, respectively. Over 65% of

BLAST hits were made up of different mollusc species as presented in Figure 7. The

Pacific oyster, C. gigas, which made up 27 and 34% of BLAST hits for the Oases and

Trinity assemblies, dominated top BLAST hits. Other molluscs including Haliotis

discus and H. diversicolor were also in the top four species that made up top BLAST

hits for both assemblies.

37

Figure 4. Nerita melanotragus transcriptome functional annotation based on Trinity Blast2GO analysis. Functional annotation results indicate the relative amount of each category of contigs with protein hits. The results are summarized as follows: Biological Process (BP), Molecular Function (MF) and Cellular Component (CC).

38

Figure 5. Nerita melanotragus transcriptome functional annotation based on Oases Blast2GO analysis. Functional annotation results indicate the relative amount of each category of contigs with protein hits. The results are summarized as follows: Biological Process (BP), Molecular Function (MF) and Cellular Component (CC).

39

Figure 6. GO category assignment. Comparative analysis and functional classification of the top 20 GO terms for the Trinity and Oases assembly.

40

Figure 7. GO category assignment. Comparative analysis and functional classification of the top 20 GO terms for the Trinity and Oases assembly.

41

2.3.3 PCR validation of contigs

The PCR primer pairs designed for beta-actin and NADH dehydrogenase subunit

5 amplified a single product of the correct size, images provided in the Figure 8. High

quality sequence was obtained for both amplicons using both forward and reverse

primers. The BAC and NAD sequences were assigned top nucleotide BLAST hits for

beta-actin from Aplysia californica and NADH dehydrogenase subunit 5 from N.

melanotragus, respectively. Protein BLAST confirmed this result with an E-value of

greater than 1 × 10−27. Alignment of the beta-actin and NADH dehydrogenase subunit

5 sequences to the contigs from which they were designed resulted in a perfect match

for beta-actin and the presence of a single one base pair indel for NADH

dehydrogenase subunit 5, in an adenosine homopolymer region. PCR validation gene

sequences are available under the accession number KM025036 (beta-actin) and

KM025037 (NADH dehydrogenase subunit 5).

Figure 8. Visualisation of PCR products. Agarose electrophoresis gel showing two candidate genes beta-actin (A) and NADH dehydrogenase (B) (Molecular marker Hyperladder IV).

500 bp300 bp200 bp

100 bp

42

2.4 DISCUSSION

The availability and throughput of next generation sequencing technologies has

enabled the rapid and efficient sequencing of transcriptomes for model and non-model

species. The majority of de novo transcriptome assemblies in non-model organisms

have in the past been produced using the long reads (300-600 bp) generated using

Roche 454 (Gibbons et al., 2009). With the recent developments in sequencing

technology, short read sequencers (90-400 bp), such as Illumina and Ion Torrent, are

starting to be more commonly used for the generation of large next generation

sequencing data sets, as the costs are much lower for the same output (Li et al., 2010).

Consequently, the use of short read sequencers to generate de novo transcriptome

assemblies for non-model organisms may lead to a more complete gene set for these

species at a lower cost. The reliability of de novo transcriptome assemblies generated

from short read sequencers, however, needs to be validated to ensure that assemblies

are accurate and won’t compromise the downstream applications of next generation

sequencing projects. This chapter focuses on comparing a number of de novo

assemblers to demonstrate that short read RNA-seq data generated by an Ion Torrent

PGMTM sequencing system can reliably and accurately be assembled for a non-model

organism.

Accurate de novo assembly of transcriptomes is crucial for next generation

sequencing projects in non-model organisms. Of particular importance is finding short

read assembly algorithms that produce accurate and reliable assemblies from the short

reads produced by Ion Torrent or Illumina sequencers. In this comparison of four

different short read assemblers using Ion Torrent data, it was found that Trinity and

Oases outperformed Velvet and Geneious in all performance metrics, including longer

N50 and average contig lengths, producing fewer and longer contigs and having less

redundant contigs. Overall, these results are similar to those obtained when comparing

Trinity or Oases against other short read assemblers in simulation studies and

empirically, with Illumina data (Martin and Wang, 2011). Even though Trinity and

Oases outperformed the other assemblers in all metrics, their respective assemblies

performed better for different quality metrics. For example, Trinity had a longer N50,

while Oases produced fewer contigs with less redundancy.

43

The de novo assemblies generated by both Trinity and Oases produced N50 and

average contig lengths similar to many past transcriptome sequencing studies (Li et

al., 2013; Hook et al., 2014; Schunter et al., 2014). The N50 and average contig size

of the Trinity assembly is also similar to that reported for the recently sequenced

transcriptome of the common pond snail, Radix balthica (Feldmeyer et al., 2011). In

contrast, the N50 and average contig size (>1200 bp) reported for a transcriptome

sequence of a different pond snail, Lymnaea stagnalis (Sadamoto et al., 2012) are 6x

larger than that of the Trinity assembly for this dataset. A few differences between this

transcriptome assembly and that for the L. stagnalis transcriptome assembly may

account for this difference. Firstly, their dataset had approximately 40x more 100 bp

Illumina sequences than in this study. Secondly, the L. stagnalis study was conducted

for a single tissue type, the central nervous system, while this study utilized the whole

animal. These two factors may explain much of the difference in N50 and average

contig length between the two studies. Therefore, it is highly likely that de novo

assemblies generated with a similar amount of Ion Torrent data could result in

assemblies with more comparable N50 and average contig lengths.

The blast and annotation success for both the Trinity and Oases assemblies was

quite low (15-19%). This level of annotation success is much lower than that often

reported in the literature even for non-model species (Li et al., 2010). The level of

annotation success achieved in this study, however, is in a similar range to that reported

for two recently sequenced gastropod transcriptomes using short read technologies (R.

balthica 17% and L. stagnalis 20.1%) (Feldmeyer et al., 2011; Sadamoto et al., 2012).

One of the reasons put forward to explain the low degree of annotation success is the

fact that few reference genome sequences exist for mollusc species (Sadamoto et al.,

2012). Furthermore, it is hypothesised that an improvement in annotation success of

gastropod species will require more representative gastropod reference genome

sequences and an increase in mollusc protein sequences in public databases.

This chapter describes an EST collection generated by Ion Torrent sequencing

and de novo assembly to characterize the transcriptome of a non-model gastropod

species, N. melanotragus. This marine gastropod is a common component of the

intertidal zone on rocky substrates and distributed from Mackay (Queensland,

Australia) to southern Tasmania (Australia) and New Zealand. Across this large

44

geographic distribution N. melanotragus spans a number of environmental gradients

such as clines in water temperature, oxygen concentration and substrate type (Waters

et al., 2005). The environment of this species also varies temporally on a micro

geographic scale between tidal cycles and consequently N. melanotragus is exposed

to large changes in a number of environmental factors including temperature, pH,

salinity and dissolved oxygen (Waters et al., 2005). Therefore generating genomic

datasets such as in this study is crucial as research efforts towards understanding

genetic and gene expression changes responsible for species adaptations.

2.5 CONCLUSION

The large number of contigs annotated and functionally characterized in this

study provides a first step towards a systems biology approach to physiological

genomics in gastropod species. By identifying a wide variety of genes from a number

of different GO classes, it is now possible to determine which genes are important for

adaptation across broad environmental changes and for stress response to micro

geographic environmental fluctuations. This is very important because we still know

remarkably little about the physiology and evolution of many marine organisms and

in particular the physiological basis of adaptation to both spatial and temporal

environmental variation in intertidal zone species (Chapperon and Seuront, 2011;

Place et al., 2012).

45

Chapter 3: Comparative transcriptomics and characterization of genes and gene families involved in abiotic stress response from two intertidal marine snails

3.1 BACKGROUND

While a core metazoan geneset is found in almost all animals, it is becoming

apparent that gene loss and gain within lineages is more common than originally

thought [Markov et al, 2015]. The evolution of new genes occurs primarily through

gene duplication and can lead to large increases in gene copy number in specific

lineages. In molluscs, for example, this has been observed in Crassostrea gigas, where

tyrosinase genes have been extensively duplicated (26 copies) when compared to

Homo sapiens (2 copies) and Caenorhabditis elegans (1 copy) [Wang et al., 2012].

While some lineage specific expansions appear random, many also seem to have an

adaptive basis. For example, an expansion of protocadherins has been observed in

Octopus bimaculoides (Albertin et al., 2015). Protocadherins are involved in neuronal

development and this correlates with an increase in nervous system complexity

observed in octopods. More research is required, however, to determine which genes

are expanded in specific mollusc lineages and whether they are likely to have an

adaptive basis.

Many lineage specific expansions in expressed genes have often been observed

in gene families with roles in stress response or immune function [Lespinet et al.,

2002] and this can have major implications for stress tolerance in particular species

[Wang et al., 2012]. In molluscs, multiple expansions of different heat shock proteins

(HSPs) have been observed, including the expansion of HSP70 in C. gigas [Wang et

al., 2012] and HSP90 in Aplysia californica (Pantzartzi et al., 2013), where they have

been associated with physiological adaptation of these organisms to thermal stress.

Similarly, immune related genes, such as toll-like receptors (TLR), appear to have

undergone extensive duplication in C. gigas and also in the echinoderm

46

Strongylocentrotus purpuratus [Wang et al., 2015]. These TLRs appear functionally

divergent and show differences in expression under bacterial or viral challenge. The

study of the expansion and contraction of gene families has largely been limited to

those mollusc species with significant genomic resources but for many mollusc

species, information on copy number within specific gene families is still lacking.

The genus Nerita is a highly diverse and abundant group of gastropod molluscs

with an extensive pantropical distribution [Frey and Vermeij, 2008]. Across their

range, Nerita species have exploited a variety of ecological niches and thrive in the

intertidal zone, where they are a common component of rocky intertidal communities.

Two common yet ecologically divergent Nerita species found in Southeast

Queensland are Nerita melanotragus and Nerita albicilla (N. melanotragus is a mid

littoral species, while N. albicilla is a low littoral species [Frey, 2010]). This means

that N. melanotragus is more likely to sustain far longer periods of thermal stress than

N. albicilla. Consequently, these two species present an interesting case to examine if

increases in gene copy number (lineage specific expansions) of expressed stress

response genes may underpin some of the physiological differences. Specifically, this

study aimed to test whether N. melanotragus had higher copy number of stress related

genes expressed under normal conditions when compared to N. albicilla, as N.

melanotragus is more likely to sustain far longer periods of thermal stress than N.

albicilla. To achieve this, examinations were made into lineage specific differences in

the copy number of stress related genes. Specifically, the two Nerita species’

transcriptome sequences were interrogated to investigate these differences.

Furthermore, investigations were made to examine if any of the stress genes showed

evidence of selection, which would indicate whether changes in gene copy number

may have an adaptive basis in these species.

3.2 METHODS

3.2.1 RNA extraction, sequencing and read processing

Two representative individuals of N. melanotragus and N. albicilla were

collected from King’s Beach, Caloundra, Australia (GPS coordinates: 153°8'14"E,

26°48'17"S). The collection site represents an overlap in the distribution of N.

melanotragus and N. albicilla (Figure S1). Two whole animals from each speies were

47

snap frozen in liquid nitrogen and stored at -80 °C. Prior to transcriptome sequencing,

species were verified based on sequences for the DNA barcode CO1 (cytochrome

oxidase 1) using the primers HCO-2198 and LCO-1490 [Folmer et al., 1994]. DNA

extraction was undertaken on two whole organism samples from each species using an

ISOLATE II Genomic DNA Kit (Bioline). PCR conditions for amplification of the

CO1 gene were as follows: 1 min at 95°C, 1 minute at 40°C, and 90 sec at 72°C for a

total of 35 cycles, followed by a final extension step at 72°C for seven minutes.

Amplicons were purified using ISOLATE II PCR and Gel Kit (Bioline), followed by

cycle sequencing using BigDye® Terminator v3.1. Sequences were resolved on an

ABI 3500 Genetic Analyzer (Life Technologies) and edited using Geneious [Kearse

et al., 2012]. Sequences were then referenced as BLASTn queries against the

nucleotide database at NCBI for species validation.

Figure S 1. Distribution of N. melanotragus (highlighted in blue) and N. albicilla (highlighted in orange). Collection point of specimens is indicated with the arrow.

Three samples from each species were homogenized in liquid nitrogen before

total RNA was extracted using a trizol/chloroform extraction protocol (Invitrogen),

followed by a silica membrane based column clean-up (Qiagen) as per Amin et al.,

2014. RNA integrity was determined using a Bioanalyser 2100 (Agilent). RNA

sequencing was performed on a single sample for each species using 91bp paired-end

chemistry on an Illumina HiSeq 2000 platform following the protocols of Prentis and

Pavasovic (2014).

48

3.2.2 Transcriptome quality and completeness

Sequence reads were converted to FastQ files and only reads that met a quality

criteria of Q > 20 and ambiguous bases < 0.5% were retained for further analysis.

These high quality reads were assembled into contigs using two software packages:

CLC Genomics Workbench [CLC Genomics Workbench

(https://www.qiagenbioinformatics.com/)] and Trinity short read de novo assembler

[Haas et al., 2013]. Once assembled, contigs were filtered for redundancy and

chimeras using CD-Hit [Huang et al., 2010]. To create the combined assembly,

sequences with 95% similarity or higher were clustered into a single contig while

unique genes were retained as singletons.

To determine the completeness of our assemblies, the number of full length and

partial ultra-conserved core eukaryotic gene set sequences (CEGs) in each assembly

using CEGMA were identified [Parra et al., 2007]. Open reading frames (ORFs) of

greater than 300 bp in length were batch extracted using TransDecoder [Haas et al.,

2013]. Final assemblies were then used as BLASTx queries against the NR database

using BLAST+ [Camacho et al., 2009]. BLAST hits were retained if they met a

stringency E-value of 10-5. BLAST2GO [Conesa et al., 2005] and Trinotate [Haas et

al., 2013] were used to assign gene ontologies (GOs). In addition, contigs were used

as BLASTx queries while extracted and translated ORFs were used as BLASTp

queries against the Swiss-Prot and TrEMBL databases. Protein family (PFAM)

domains were assigned using HMMER [Finn et al., 2011] searches while signalP

[Petersen et al., 2011] was used to predict signal peptides. A Trinity script was also

used to assess the number of full-length transcripts with 100 % coverage in both

assemblies. This was undertaken through BLASTx searches with a stringency E-value

of 10-20 which was used to determine the amount of predicted proteins from our

assemblies that had 100 % coverage for the top matching entries. Contigs matching

known parasitic species of gastropods (Chelonia mydas, Schistosoma japonicum and

S. mansoni) were removed based on top BLAST hit information. A total of 70 contigs

in N. melanotragus and 51 contigs in N. albicilla had top BLAST hits to parasites and

were removed from downstream analysis. WEGO [Ye et al., 2006] was used to

categorise and visualise GO terms for both species. Gene set enrichment analysis

(GSEA) was then performed to determine if transcripts from a particular GO class

were over-represented in either species, using GOSeq within the Trinotate suite.

49

3.2.3 Identification of unique and common genes

To identify common orthologs encoding proteins in our datasets, the predicted

peptide sets for N. melanotragus, N. albicilla, Lottia gigantea and Crassostrea gigas

were used. Peptide files for all four species were analysed using OrthoVenn [Wang et

al., 2015]. An E-value of 10-5 was used with an inflation value of 1.5. Gene set

enrichment analysis was then undertaken to find gene ontologies which are

significantly over-represented in the orthologs shared between the two Nerita species

as well as those unique to each Nerita species using a hypergeometric test with a

significance level of p < 0.05.

3.2.4 Identification and comparative analysis of full length candidate genes

In order to identify candidate genes associated with stress response, annotations

were exported from BLAST2GO. Gene ontology terms associated with stress were

obtained from the Gene Ontology Consortium [Ashburner et al., 2000] and used to

identify candidate genes (Table S2). Select candidate gene families (HSP90, HSP70,

HSP60, HSP20, catalase and nitric oxide synthase) were then investigated for

differences in copy number in the transcriptome sequences of both species.

In order to test whether specific candidate genes showed signatures of selection

the ratio of synonymous to non-synonymous substitutions was calculated, homologs

of interest were extracted from our datasets. Nucleotide sequences from each species

were entered into an automated KaKs online calculation tool using default paramters

(Zhang et al 2006). A KaKs ratio of greater than 1 indicates positive selection on the

gene, a value of less than 1 indicates purifying selection and values equal to 1 indicate

neutral evolution.

3.3 RESULTS

3.3.1 Sequencing results, reads assembly and functional annotation of contigs

CO1 barcoding confirmed the identity for each species (GenBank: KP981645,

KP981646) with 100 % identity to the N. melanotragus haplotype (EU732275) and N.

albicilla (EU253341), respectively.

Transcriptome sequencing of N. melanotragus and N. albicilla using Illumina

91bp paired end sequencing chemistry produced a total of 6.2 Gbp and 7.4 Gbp of

sequence from 68,678,334 and 61,367,256 raw reads, respectively. Greater than 99.99

50

% of reads were retained after quality trimming. Reads from N. melanotragus and N.

albicilla were assembled into 77,098 and 78,577 contigs, respectively, using CLC

Genomics Workbench and 127,068 and 124,992 contigs, respectively, using Trinity

software (see Table S1 for the complete assembly metrics). Merged assemblies

produced a total of 159,523 and 159,506 non redundant contigs for N. melanotragus

and N. albicilla, respectively. From these, 45,438 (28.48 %) and 44,887 (28.14 %)

contigs returned significant hits for known genes, within N. melanotragus and N.

albicilla, respectively. Over 75 % of top BLAST hits for both Nerita species were

assigned to molluscs (Figure S2).

Table S 1. Assembly quality metrics. Assembly statistics for the transcriptomes produced by the different short read de novo assemblers.

Assembly statistics N. melanotragus N. albicilla

CLC Genomics

Trinity Combined CLC Genomics

Trinity Combined

Number of contigs 77,098 127,068 159,523 78,577 124,992 159,506

Longest contig 14,653 14,718 14,718 15,413 15,538 15,538

N50 800 995 897 776 887 801

Figure S 2. BLAST top hit species distribution. The 20 species most commonly represented in BLAST hits for

N. albicilla and N. melanotragus assemblies.

A total of 203 (81.85 %) and 207 (83.55 %) full length proteins were identified

based on the 248 CEGs. In the overall dataset, however, 2566 (5.6 %) and 2296 (5.1

%) full length proteins were observed for N. melanotragus and N. albicilla,

51

respectively. The distribution of GO terms across the three broad categories biological

process, cellular component and molecular function are presented in Figure S3. Gene

set enrichment analysis revealed no significant enrichment of GO terms between the

two datasets.

Figure S 3. Gene ontology (GO) functional classification. Distribution of annotated contigs for N. melanotragus and N. albicilla across three main GO categories; including cellular component, molecular

function and biological process. The left Y axis indicates the percentage of transcripts assigned to a GO term, and the right Y axis represents the number of transcripts within the same category.

3.3.2 Unique and common genes

Overall, 6457 orthologs were found in common amongst the two Nerita species,

L. gigantea and C. gigas, while 3321, 10561 and 8134 were in common among three

species comparisons, two species comparisons or only found in individual species,

respectively (Figure 9). In total, the two Nerita species shared 8501 orthologs with

each other, and Nerita melanotragus had 3618 unique orthologs, while N. albicilla had

2280 unique orthologs. A total of 636 orthologs were shared among N. melanotragus,

C. gigas and L. gigantea while fewer orthologs were shared between C. gigas and N.

melanotragus (183) than L. gigantea and N. melanotragus (312). Crassostrea gigas

52

and L. gigantea share only 392 orthologs with N. albicilla, and when comparing two

species, C. gigas and L. gigantea share 155 and 259 orthologs with N. albicilla,

respectively.

Figure 9. Orthologous genes between Nerita melanotragus, N. albicilla, Crasosstrea gigas and Lottia gigantea at 88% similarity cutoff.

Geneset enrichment analysis found two terms over-represented for orthologs

shared between the two Nerita species (activation of protein kinase B activity and

positive regulation of guanylate cyclase activity), but neither are directly related to

stress. For orthologs unique to N. melanotragus only a single GO term was over-

represented (phosphoribosylglycinamide formyltransferase activity), while for N.

albicilla no GO terms were over-represented.

3.3.3 Identification and comparative analysis of candidate genes

The functional annotation of contigs allowed the identification of a number of

candidate genes associated with stress based on multiple GO terms. From these GO

terms, 447 and 604 genes with one or more of the stress GO terms were identified for

N. melanotragus and N. albicilla, respectively. In the specific “response to stress” GO

53

term N. melanotragus and N. albicilla had 99 and 262 genes, respectively. Similar

numbers of genes in the “response to heat” GO term were observed for N.

melanotragus (40) and N. albicilla (33) (Figure 10). When examining expansions in

specific gene families, heat shock protein 90 was expanded in N. albicilla, while Heat

shock protein 20 was expanded in N. melanotragus (Table 4). Nitric oxide synthase

and catalase showed similar number of gene copies across N. melanotragus and N.

albicilla. Other classic stress response genes such as heat shock protein 60, 70 and 90

were also expanded in N. albicilla. Analysis into the ratio of synonymous to non-

synonymous substitutions of these genes within the Nerita species showed patterns of

nucleotide variation consistent with the action of purifying selection (Table 5).

Figure 10. GO Classification of stress related genes for N. melanotragus and N. albicilla.

Table 4. Candidate gene counts. A list of candidate gene counts for Nerita melanotragus, N. albicilla, Crasosstrea gigas and Lottia gigantea.

Gene Count

Nerita melanotragus

Nerita albicilla

Crasosstrea gigas

Lottia gigantea

0 50 100 150 200 250 300

response to biotic stimulus

response to activity

response to abiotic stimulus

detection of stimulus

response to stress

response to external stimulus

response to endogenous stimulus

cellular response to stimulus

immune response

protein activation cascade

positive regulation of response to stimulus

negative regulation of response to stimulus

response to chemical

response to dietary excess

response to heat

Stress Gene GO Term Classification

N. albicilla

N. melanotragus

54

HSP90 (with isoforms)

10

13

3

3

HSP90 (without isoforms)

5

10

3

3

HSP20 (with isoforms)

58

43

17

10

HSP20 (without isoforms)

19

16

17

10 Catalase (with isoforms)

2

2

3

1

Catalase (without isoforms)

2

2

3

1

Nitric oxide synthase (with isoforms)

12

15

5

16 Nitric oxide synthase (without isoforms)

7

8

5

16

Table 4. Synonymous to non-synonymous substitution calculations.A list of synonymous to non-synonymous substitution calculations for select candidate genes.

Annotation Ka Ks Ka/Ks Cellular process

Heat shock protein 90 0.0036 0.0043 0.8381 Response to stress

Heat shock protein 70 0.0044 0.0049 0.9025 Response to stress

Heat shock protein 60 0.0047 0.0060 0.7785 Response to stress

Heat shock protein 20 0.0090 0.0132 0.6835 Response to stress

Catalase 0.0067 0.0082 0.8205 Response to stress

Nitric oxide synthase 0.0077 0.0087 0.8807 Response to stress

Globin 0.0092 0.0128 0.7150 Transport

3.4 DISCUSSION

This study expanded on the currently available genomic resources (Adamson et

al 2015; De Oliveira et al 2016; Ip et al 2016) for Class Gastropoda by describing two

whole organism transcriptomes of the ecologically divergent neritid species, N.

melanotragus and N. albicilla. This data provides the first largely complete

transcriptomes for the highly speciose Neritimorpha lineage of gastropods. Overall,

55

the N50 metrics generated in this study are similar to a recent transcriptome study of

the gastropod Lottia kogamogai (817 bp; De Oliveira et al 2016), but less than that

reported for the intertidal whelk Reishia clavigera (2236 bp; Ip et al 2016). A reason

for the observed difference in N50 between this study and that of Ip et al (2016) could

be the difference between contigs generated. This study generated ~ 150 000 contigs

for both the neritid species, while the study by Ip et al (2016) generated less than 40

000 total transcripts. This study and those of (Adamson et al 2015; De Oliveira et al

2016; Ip et al 2016) had similar BLAST and annotation success. This resource has

allowed us to examine comparative genomic analysis with other mollusc species and

whether lineage specific gene expansions in gene families involved in stress response

had occurred in either species. Changes in gene copy number and over representation

of specific GO classes in unique orthologs in both species may also support the

physiological basis of the ecological differentiation (which intertidal zones they

occupy) observed between the two neritid species.

Comparative analysis of ortholog distribution showed that many genes were

shared among the four mollusc species, but that a larger number were unique to the

two neritid species. This is not surprising as the other two species, C. gigas and L.

gigantea, are quite divergent from the two Nerita species. Furthermore, no evidence

of any stress gene ontologies over represented in genes unique to the Nerita species

were observed. An explanation for this may be that all the species used here are

intertidal and therefore have reasonably similar gene sets, but this remains to be

formally tested.

3.4.1 Comparative copy number analysis of stress genes

Overall, it was found that genes involved in response to stress were expanded in

the low intertidal species compared to the mid intertidal species. This finding did not

support the hypothesis that N. melanotragus should have higher stress gene copy

number as this species is more likely to sustain far longer periods of thermal stress

than N. albicilla. While at first glance this result may appear counter-intuitive, i.e., the

species that receives less exposure to higher temperatures has a larger number of stress

genes expressed under normal conditions. This observed pattern likely results from the

fact that lower intertidal species exhibit thermal stress more quickly than mid to upper

intertidal species at lower temperatures (Tomanek, 2002). Therefore, increased

56

numbers of stress response genes expressed in N. albicilla, may help this species to

cope with acute thermal stress encountered by low intertidal species when exposed at

low tides. Other molecular mechanisms, such as frontloaded expression of stress genes

[Barshis et al., 2013] may be an alternative explanation of why fewer stress response

genes are expressed in N. melanotragus.

Changes in the number of stress response genes within specific gene families

were also observed among N. melanotragus and N. albicilla. Of these gene families,

specific heat shock proteins (HSPs) gene families showed much of the variation. This

diverse family of proteins are known to assist in the repair, folding and translocation

of proteins damaged under various stresses, as well as the folding and translocation of

newly synthesized proteins [Fulda et al., 2010]. In N. albicilla, the largest increases in

HSPs expressed relative to N. melanotragus were those with molecular masses of 60,

70 and 90. These three classes of HSPs have been the most extensively investigated

[Sumizawa and Igisu, 2008; Jurgen et al., 2011; Rupik et al., 2011; Xu and Qin, 2012]

and show facultative expression in response to multiple stressors including thermal

stress [Boone and Vijayan, 2002]. This reinforces the previous finding that increases

in gene copy number of expressed stress related genes may occur in species that exhibit

acute thermal stress rather than those that experience a greater duration of exposure to

stress on a regular basis.

All of the stress response genes that were tested for evidence of selection

including those with changes in gene copy number were found to be under purifying

selection. This result is not surprising as many stress response genes are highly

conserved among metazoan taxa in general [Beck et al., 2000]. In fact, the vast

majority of stress response genes form part of the cellular stress response a universal

molecular mechanism largely conserved across all metazoan lineages (Kültz, 2005) to

maintain cellular homeostasis in organisms. Therefore, it is likely that increased copy

number in lower intertidal species is associated with increased gene expression under

stressful conditions.

3.4.2 Conclusion

The present study corresponds to initial attempts in providing genomic data for

two closely related gastropod species. This study has identified the difference in stress

57

related genes for species inhabiting low and mid intertidal zones under normal living

conditions. Furthermore, results from this study suggest lower intertidal species

express more stress related genes than those from the mid intertidal zone. In addition,

this study presents an ideal foundation for further investigation into stress related gene

expression in intertidal species.

58

Chapter 4: Differential Gene Expression of Two Intertidal Snails in Response to Temperature Stress

4.1 BACKGROUND

Examining how marine mollusc species respond at a genetic and proteomic level

to temporal and spatial changes in their abiotic environments is central to our

understanding of their physiology. Some marine molluscs undergo larval dispersal and

are largely sedentary in their post larval developmental stages [O'Connor et al., 2007].

As a consequence, these organisms need to possess mechanisms to survive and cope

with dynamic and changing environmental conditions. This area is between the low

tide and high tide marks, where organisms are exposed to dramatic fluctuations in a

number of environmental variables such as temperature, pH, salinity, dissolved oxygen

and the availability of food. In fact, recent research has determined that many intertidal

molluscs inhabit areas near the upper range of their thermal limits [Tomanek &

Somero, 1999; Tomanek & Zuzow, 2010] and are particularly vulnerable to thermal

stress. Physiological adaptations, such as the extensive duplication of heatshock

protein 70 and inhibition of apoptosis proteins in the Pacific oyster (Crassostrea gigas)

[Wang et al., 2012] and distinct changes in gene expression profiles or protein

abundance [Wang et al., 2014] have been proposed to allow mollusc species to

withstand the extreme environmental fluctuations present in the intertidal zone. While

this recent research has started to demonstrate how intertidal mollusc species persist

in such stressful environments, more research into the transcriptomic and proteomic

response of other mollusc species is required.

A consistent pattern that has been observed among closely related marine

molluscs is that mid to high intertidal species tend to have higher baseline expression

patterns and protein abundance for candidate stress response genes, where as low

intertidal species showed an induction of these genes under mild thermal stress. Such

a pattern has been observed in closely related marine snail species, Chlorostoma and

in Lottia where baseline levels heat shock protein 70 expression were higher in mid to

high intertidal species. Such a pattern has been found in a wide range of species that

59

display physiological resilience to thermal stress [e.g., Barshis et al, 2013]. In marine

molluscs, however, genome wide patterns of gene expression and protein abundance

have not been compared among closely related species that inhabit different areas of

the intertidal zone under thermal stress.

Nerita albicilla and Nerita melanotragus are widespread intertidal snails,

distributed across a number of temporally and spatially fluctuating environmental

gradients, including latitudinal clines in water temperature and abrupt changes in water

temperature over a tidal cycle. These two species differ in their ecology and geographic

ranges, as N. melanotragus is a mid intertidal species, which ranges from temperate to

subtropical climes, while N. albicilla is a low intertidal species found in tropical and

subtropical regions. While the distribution of these two species marginally overlaps in

subtropical regions, colonisation of different intertidal areas means that one species

(mid intertidal) is more likely to sustain long periods of temperature fluctuation, while

the other (low intertidal) is more frequently subjected to flooding and more buffered

from temperature fluctuation. Consequently, these two species present an interesting

case to examine patterns of differential gene expression and protein abundance when

placed in different thermal environments. Therefore, in this study, the aim was to

examine if N. albicilla (low intertidal species) had a more acute transcriptomic and

proteomic thermal stress response compared to N. melanotragus (mid intertidal

species) under the same temperature conditions.

60

4.2 METHODS

4.2.1 Organism conditioning and treatment

Nine individuals each of N. melanotragus (Figure 1a) and N. albicilla (Figure

1b) were collected from Coolum Beach (Queensland), GPS position: 26°31'42"S,

153°5'26"E. Only snails of 2cm circumference and greater were collected for this

experiment. Snails were allowed to acclimatise for 10 days in a glass tank filled with

seawater (pH: 6.7 salinity: 34%). Algae had developed in the storage tanks and were

readily available for consumption during acclimation. Following this, snails were

placed into three groups, each containing three individuals and placed into one of three

temperature conditions, 14 °C, 22 °C and 31 °C, for each respective species.

Temperatures were raised or lowered from 22 °C to the target temperatures over 1 hr

and organisms were completely submerged in 500 mL beakers containing seawater.

After three hours, organisms were euthanized and foot tissue extracted. Collected

tissue samples were immediately snap frozen in LN2 and stored at -80 °C for RNA

extractions.

4.2.2 RNA extraction and sequencing

Total RNA was extracted from foot tissue using a trizol/chloroform extraction

protocol (Invitrogen), followed by a column based cleanup (Qiagen Midi).

Specifically, each sample was homogenised using LN2, added to 1 ml of Trizol and

allowed to incubate at room temperature for five minutes. Following this, 0.2 ml of

chloroform was added and the mixture was centrifuged at maximum speed (25

200RCF) for 10 minutes. Supernatant was isolated and placed into a mixture of 2 ml

RLT buffer (Qiagen) and 2ml 70% ethanol combined. The samples were then placed

into an RNeasy column for a series of subsequent wash steps as per the RNeasy Midikit

(Qiagen) manufacturer’s instructions. Upon completeion, quality and quantity of RNA

samples were tested on a Bioanalyzer 2100.

Messenger RNA (mRNA) was isolated using the Dynabeads mRNA Purification

Kit (Life Technologies) as per manufacturer’s instructions. Following this, samples

were sent to Beijing Genomics Institute (BGI) for sequencing on an Illumina paired-

end sequencing platform.

61

4.2.3 Assembly and annotation

Raw sequence reads were received as paired-end FastQ files. Raw reads were

concatenated to create combined left and right read files for each respective species.

These combined files were assembled de novo using Trinity software, specifically

chosen for the availability of downstream pipelines (Haas, 2013). Assemblies were

conducted using two minimum contig cut-off lengths for comparison (200 bp & 300

bp) to limit redundancy. Default assembly parameters were used, with the exception

of a –trimmomatic command to enable filtering of raw reads prior to assembly.

Following this, transcriptomes were annotated by referencing them to the NR database

at NCBI as BLASTx queries using an E-value of 10-5. Gene ontology (GO) terms were

retained for mapping and used to create GO maps.

4.2.4 Differential gene expression analysis

The assembled transcriptomes were analysed for differentially expressed genes

using Trinity perl scripts, RSEM and edgeR as per Haas (2013). Specifically, raw reads

from each individual were mapped back to the assembled file for transcript abundance

estimation using RSEM. Following this, differentially expressed transcripts were

identified using edgeR, standard parameters were used with the biological replicates

option. In order to analyse differentially expressed transcripts and create heat maps,

differentially expressed matrix files were analysed using the analyze_diff_expr.pl perl

script and edgeR with a minimum abs fold change of two and p-value cut-off of 0.001.

4.2.5 Orthologous transcript identification

In order to identify orthologous transcripts between differentially expressed

transcripts from each temperature treatment for respective species, ProteinOrtho was

used. Files containing nucleotide sequences of differentially expressed genes were

translated to amino acid sequence using EMBOSS TranSeq. Files containing amino

acid sequences were then entered into Proteinortho for ortholog identification using a

hit similarity threshold of 88%.

4.2.6 Proteome analysis

A total of 25 samples each of N. melanotragus and N. albicilla were collected

from Coolum Beach (Queensland), GPS position: 26°31'42"S, 153°5'26"E. Organisms

size and acclimation conditions were as above. Snails were placed into groups of five

62

and assigned to one of five temperature settings, 14 °C, 22 °C, 31 °C, 38 °C and 45 °C

for each species. Temperatures were gradually raised or lowered from 22 °C to the

target temperatures over 1 hr and organisms were completely submerged in beakers

containing seawater. After 3 hours, organisms were euthanized and foot tissue was

extracted. Collected tissue samples were immediately snap frozen in LN2 and stored

at -80 °C. Following this, foot tissue was homogenized in LN2 and proteins extracted

as per Wisniewski et al., [2009] with the following modifications. Samples were

physically disrupted in extraction buffer containing protease (Thermo) and

phosphatase (Pierce) inhibitor cocktails (150 µL per 25 mg) using 18 gauge blunt

ended needles (Terumo) due to high sample viscosity. Protein concentration of each

sample was estimated using a Bradford assay kit according to the manufacturer’s

instructions (Pierce) and visualised using colloidal Coomassie following SDS-PAGE,

with pre-cast Novex NuPAGE 4-12% Bis-Tris 1 mm gradient gels run under

manufacturer’s recommendations (Life Technologies).

The supernatant of homogenised tissue was added to the filter of 3 kDa

ultracentrifugation tubes along with 200 µL of 8 M urea (Sigma, U5128) in 0.1 M

Tris/HCl, pH 8.5, 0.025 M DTT (UA buffer), allowed to stand at RT for 60 min and

then spun at 14,000 x g for 15 min. Cysteine alkylation was performed with 100 µL of

0.05 M iodoacetamide in UA buffer for 20 min in the dark prior to centrifugation at

14,000 x g for 10 min. Subsequent washing was performed with UA buffer (x3) and

0.05 M triethylammonium bicarbonate (TEAB) (x3) and centrifuged at 14,000 x g for

15 and 10 min, respectively. An aliquot of 40 µL trypsin (1:100 enzyme to protein

ratio) was added to the filter and incubated in a humidified chamber for 18 hours.

Peptides were collected into fresh tubes via 2 centrifugation steps at 14,000 x g for 10

min with 40 µL TEAB added in the latter. Peptides were acidified with formic acid to

a final concentration of 0.1%.

An aliquot from each temperature group was pooled with its respective replicates

and proteins were separated by isoelectric point across a 24 cm non-linear pH 3-10

immobilised pH gradient gel (GE) using an 3100 OFFGel fractionator (Agilent) as per

manufacturer’s instructions. Each fraction and whole samples were subsequently

digested using the filter-aided sample preparation technique [Wisniewski et al., 2009]

using 30 kDa NMW cut-off filters (Millipore). Peptides were de-salted using stage-

63

tips prepared in-house [Rappsilber et al., 2007]. Data dependent acquisition (DDA)

was performed on the fractions and a representative un-fractioned sample using a

TripleTOF 5600+ mass spectrometer (Sciex) coupled with an Ekspert nanoLC 400

system (Eksigent). Peptides were washed on a C18 trap column (ChromXP 150 µm x

10 cm 3 µm, Eksigent) and then separated using a 90 min gradient (Buffer A = 0.1%

formic acid, 2% acetonitrile; Buffer B = 80% acetonitrile in 0.1% formic acid; 2-40%

Buffer B in 60 min, 40-95% Buffer B over 10 min, 95% Buffer B for 10 min before

re-equilibration in Buffer A for 10 min) on a C18 resolving column (ChromXP 150

µm x 10 cm 3 µm, Eksigent) prior to electrospray ionisation. Mass spectrometry data

were acquired using a looping cycle of single MS1 scan between 350-1350 m/z

followed by up to 40 MS2 scans between 100-1250 m/z with a dynamic exclusion

window of 10 sec. Data independent acquisition (DIA) was performed on the whole

samples only using the same system and gradient above, except with 100 variable

SWATH windows.

Raw DDA data were queried in ProteinPilot (Sciex) employing the Paragon

algorithm (Sciex) against custom protein databases, which were generated using batch

extracted and translated open-reading frames from both assemblies with TransDecoder

[Haas et al., 2013]. Amino acid substitutions and modifications were included and a

False Discovery Rate (FDR) was calculated using a forward-reverse database search.

Identification of proteins was based on a 1% global FDR and the detection of ≥ two

distinct peptides per protein.

4.3 RESULTS

4.3.1 Sequencing, assembly and annotation

Transcriptome sequencing of mRNA from N. melanotragus and N. albicilla on

the Illumina sequencing platform produced an average of 15 million raw paired end

reads, per sample. Raw sequence reads were filtered and data failing to meet quality

criteria (Q < 20, N bases < 1%) was excluded, but on average greater than 97% of high

quality reads were retained per sample.

High quality reads from N. melanotragus and N. albicilla were assembled into

268,819 and 239,421 contigs respectively using Trinity software. The assembled

64

contigs from the respective species produced a longest contig of 14,653 bp and 15,413

bp. The N50 statistic for N. melanotragus and N. albicilla were similar at 1,159 and

1,160.

Of the 268,818 and 239,420 contigs referenced against the NR database at NCBI

using BLAST+ software, 41,753 (15.53%) and 39,637 (16.55%) received hits with

greater than the specified E-value stringency of 10-5, for Nerita melanotragus and N.

albicilla respectively. GO terms categorisation revealed highest abundance of genes

in cell, cell part and catalytic categories for both species (Figure s4).

Figure S 4. Gene ontology (GO) functional classification. Distribution of all annotated contigs for N. melanotragus and N. albicilla.

4.3.2 Differential gene expression changes under temperature treatments

A total of 131 and 22 genes were differentially expressed with a FDR cut-off of

0.001 for Nerita melanotragus and N. albicilla, respectively. Specifically, at 14 °C, 22

°C and 31 °C respectively, 61, 68 and 63 genes were up-regulated in Nerita

65

melanotragus and a total of 5, 7 and 17 genes were up-regulated in N. albicilla. A

comparison of the gene ontologies present in the differentially expressed genes for N.

melanotragus and N. albicilla revealed a broadly similar overall presence of GO terms.

Some GO categories were overabundant in one species and these were translation

regulator, death and viral reproduction in Nerita albicilla (Figure 11) and envelope,

virion, virion part, structural molecule, transcriptional regulator, transporter, biological

adhesion and rhythmic process in N. melanotragus.

Figure 11. WEGO plot for differentially expressed genes of N. melanotragus and N. albicilla.

BLASTx analysis of these genes revealed biological functions for 53 of 131

differentially expressed genes for Nerita melanotragus and 17 of 22 for N. albicilla.

Therefore, whilst most genes in N. albicilla were able to be assigned putative functions

(77.2%), only 40.5% of the genes from N. melanotragus returned BLASTX results

and/or domains results that made it possible to identify the possible function of the

predicted proteins. For a full list of all the genes that show differential expression

profiles, the name of their top BLAST hit and their putative functions please see Table

6 and 7.

66

Table 5. Mean expression values for N. melanotragus differentially expressed genes under different temperature treatments.

Contig description Probable function(s) 14°C 22°C 31°C

---NA--- N/A 1.498583 2.913832 -1.36515

---NA--- N/A -2.09959 2.588479 -0.48889

---NA--- N/A -1.00375 0.941993 0.061759

---NA--- N/A -0.75334 -0.75334 1.506684

tetratricopeptide repeat

protein 8

Probable role in signal transduction via

primary cilia, possibly related to

sight/smell senses or kidney chemical

sensing.

0.310041 1.227388 -1.53743

tetratricopeptide repeat

protein 8

Probable role in signal transduction via

primary cilia, possibly related to

sight/smell senses or kidney chemical

sensing.

1.648767 0.205366 -1.85413

---NA--- N/A -0.92598 1.851968 -0.92598

---NA--- N/A -2.95539 1.827476 1.127917

---NA--- N/A -0.53241 -1.79825 2.330661

sco-spondin precursor Probable involvement in neural axon

structuring.

1.049608 -0.83834 -0.21127

wd repeat-containing protein

87

Possible involvement in signal

transduction related to cell cycle and

apoptosis regulation.

1.209321 -1.00288 -0.20645

---NA--- N/A 0.152571 1.694429 -1.847

---NA--- N/A 1.910321 -1.26962 -0.6407

tar dna-binding protein 43-

like

Probable role in transcriptional repression

of certain genes, alternative splicing,

recruitment of mRNA into ribosomes,

and/or miRNA biogenesis. This particular

protein has a possible focus on neural cells

and/or relation to stress granule formation.

-0.72221 0.941615 -0.2194

transmembrane protein 164 Molecule signalling/transport 0.495679 -0.68357 0.187888

transmembrane protein 164 Molecule signalling/transport 1.551974 -1.23453 -0.31745

---NA--- N/A -0.19089 -1.00974 1.200629

---NA--- N/A 0.786183 0.514089 -1.30027

electroneutral sodium

bicarbonate exchanger 1-like

Probable role in pH regulation by

bicarbonate regulation, possible focus on

neurons specifically.

0.771589 -0.19295 -0.57863

67

zinc finger fyve domain-

containing protein 26

Probable role in regulating abscission

during cytokinesis, possible role in DNA

break repair.

-0.4681 1.504489 -1.03639

---NA--- N/A -0.86811 -0.01672 0.884827

peregrin-like protein dorsal/ventral axon guidance -2.08523 0.74095 1.344282

---NA--- N/A -1.50362 1.802018 -0.2984

c16orf52 homolog b Probable role in the movement of

organelles and other intracellular objects

and/or role in cilia/flagella motility.

1.704829 -1.73112 0.026295

c16orf52 homolog b Probable role in the movement of

organelles and other intracellular objects

and/or role in cilia/flagella motility.

-1.4657 0.969876 0.49582

dynein heavy chain

axonemal

Probable role in the uptake and/or removal

of substrates for cellular activity.

0.141787 -0.93798 0.796197

uncharacterized transporter -

like

Probable role in the uptake and/or removal

of substrates for cellular activity.

0.767624 -0.78974 0.022117

cyclin-dependent kinase-like

2

Probable role in modulating dynein

proteins to enable cilia/flagella motility.

-1.57109 1.30349 0.267599

coiled-coil domain-

containing protein c16orf93

homolog

Probable role in maintaining the structural

organisation of cilia.

-0.95818 0.594478 0.363703

enolase-like protein eno4 N/A -1.20317 -0.59012 1.793288

---NA--- N/A -0.91968 -0.91968 1.839367

---NA--- Probable role in nuclear import of

ribosomal proteins for ribosome synthesis,

histones for DNA formation, and/or

transcriptional regulators.

-1.22457 0.520842 0.703732

importin-7 Ran-dependent transport cycle 2.602112 -1.01921 -1.5829

---NA--- N/A 0.928927 -1.14942 0.22049

---NA--- N/A -2.79042 2.217037 0.573384

---NA--- Probable role in eventual formation of

nitric oxide and/or neurotransmitters such

as dopamine or serotonin.

0.397061 0.734933 -1.13199

gtp cyclohydrolase 1 hydrolysis of guanosine

triphosphate (GTP)

0.025557 -1.32752 1.301963

---NA--- N/A -0.12553 -0.95228 1.077806

---NA--- N/A -2.86374 5.305254 -2.44152

---NA--- N/A 1.358908 -2.38264 1.02373

68

---NA--- Probable role as a transcriptional regulator

for certain genes, possibly associated with

the cell cycle and apoptosis.

-1.17385 -0.28607 1.459914

thap domain-containing

protein 4

Probable role as an apoptotic regulator 1.40605 -1.93739 0.531339

---NA--- N/A -1.33766 1.649973 -0.31231

---NA--- N/A 0.741576 -1.41762 0.676049

hypothetical protein

LOTGIDRAFT_232099

N/A -2.25993 0.395171 1.864757

---NA--- N/A 0.972101 -0.66456 -0.30754

protein arrd- isoform b N/A -0.88837 0.860892 0.027482

protein arrd- isoform b N/A -0.96505 0.96305 0.001995

---NA--- N/A -1.15856 -2.49996 3.658517

---NA--- N/A 1.799665 -0.32581 -1.47386

---NA--- N/A -1.76184 0.641936 1.119899

---NA--- N/A 1.571831 0.068855 -1.64069

---NA--- N/A -1.17244 1.580824 -0.40838

kinesin-related protein 1 Cellular function support 0.961943 -0.88881 -0.07313

kat8 regulatory nsl complex

subunit 1

Probable role in transcriptional activation -2.41931 0.513858 1.905452

tripartite motif-containing

protein 2

Probable role in host defense transcription

activation

-0.51674 -2.73753 3.254271

---NA--- N/A 0.406376 0.935107 -1.34148

hypothetical protein

HPF30_0708

N/A -0.3718 -0.91216 1.28396

hypothetical protein

HPF32_0596

SLIT/ROBO transcriptional activator,

possible involvement in neuron axon

regulation, cell cycle regulation, and/or

actin cytoskeleton organisation. Exact

function is uncertain.

1.088905 -0.83779 -0.25112

slit-robo rho gtpase-

activating protein 1

Mediate cell communication -1.51207 1.755028 -0.24296

sarcoplasmic reticulum

histidine-rich calcium-

binding

Probable role in regulation of the septin

filaments of the cytoskeleton in

ciliogenesis and collective cell migration.

1.201298 -0.25639 -0.94491

wd repeat-containing and

planar cell polarity effector

protein fritz homolog

Probable role in DNA repair, signalling

pathways, and/or transcription and

translational regulation.

0.794859 2.095552 -2.89041

69

heterogeneous nuclear

ribonucleoprotein a1-like

Probable role in maintaining the structural

organisation of cilia.

-1.33905 0.97899 0.36006

intraflagellar transport

protein 88 homolog

Probable role in transcriptional regulation

via chromatin remodelling, possibly

neuron focused role.

-0.00894 -0.88358 0.892519

swi snf complex subunit

smarcc2 isoform x3

Probable transcription factor binding

domain

-1.1036 0.560214 0.543391

tpr repeat-containing protein

ddb_g0287407-like

Probable role in neural signalling and/or

signal transduction pathways involving G-

protein coupled receptors.

-0.54238 0.791001 -0.24862

calcium-activated potassium

channel slowpoke-like

Negative regulator of TGF-beta pathway,

possible role in cell cycle regulation,

motility, and/or apoptosis.

0.070542 0.974957 -1.0455

ski oncogene Negative regulator of TGF-beta pathway,

possible role in cell cycle regulation,

motility, and/or apoptosis.

0.617868 1.20275 -1.82062

ski oncogene Probable role in cell cycle regulation via

the degradation of G1 cyclins and Cdk

inhibitors.

0.618962 -0.1415 -0.47747

f-box wd repeat-containing

protein 7

Probable role in cell cycle regulation via

ubiquitin-mediated degradation.

-0.47076 1.631734 -1.16097

---NA--- Probable role in mitochondrial

homeostasis and apoptosis.

0.644007 -0.6113 -0.0327

leucine-rich repeat serine

threonine-protein kinase 1

Probable involvement in the anchoring of

proteins into the cellular membrane.

-0.62762 -0.62762 1.255248

fer3-like protein Probable function in cell development -1.56726 1.01173 0.555525

hypothetical protein

CGI_10007431

Probable role in the removal of phosphate

groups from nucleosides, possibly for

nutritional purposes.

1.509177 -0.53581 -0.97337

5 -nucleotidase Frizzled-type receptor protein involved in

Wnt signalling pathway, possibly involved

in ciliogenesis.

1.604721 -1.32599 -0.27873

frizzled homolog 7b Possible involvement in signal

transduction related to cell cycle and

apoptosis regulation.

-0.58024 -0.58455 1.164783

dmx-like protein 2-like Probable role in Notch signalling pathway 0.284347 0.750848 -1.0352

zinc finger protein Molecule binding -1.78057 2.601294 -0.82073

u6 snrna-associated sm-like

protein lsm6

Probable role in pre-mRNA splicing 0.737259 -0.93816 0.200898

70

pre-mrna-splicing regulator

female-lethal d

Rab GTPase pathway activator, probably

involved in resulting intracellular

transport of vesicles along the

cytoskeleton.

0.477878 0.221834 -0.69971

small g protein signaling

modulator 1

Possible involvement in ciliary

organisation and function, though

uncertainty exists regarding this protein’s

function.

-1.14507 1.281308 -0.13624

leucine-rich repeat-

containing protein 48-like

Probable involvement in the anchoring of

proteins into the cellular membrane.

1.061098 1.027505 -2.0886

cadherin-23 isoform x1 Probable involvement in the incorporation

of histones H3 and H4 into new DNA or

DNA under repair.

2.151927 -1.46405 -0.68787

probable histone-binding

protein caf1-like

Probable role in DNA binding 1.089292 -1.38529 0.296

---NA--- N/A -2.56225 2.658204 -0.09596

---NA--- N/A -0.86644 0.785913 0.080524

---NA--- Probably involvement in the breakdown of

fatty acids for energy.

-1.03298 1.227206 -0.19423

3-ketoacyl- mitochondrial Probable role in RNA binding -0.99933 0.572135 0.427198

leucine-rich repeat-

containing protein 9-like

Probable involvement in the anchoring of

proteins into the cellular membrane.

-0.27596 1.322998 -1.04704

caax prenyl protease 2-like Probable involvement in the anchoring of

proteins into the cellular membrane.

-0.5213 0.488286 1.268568

caax prenyl protease 2-like Probable involvement in the anchoring of

proteins into the cellular membrane.

0.977371 -0.89009 -0.08728

farnesyl pyrophosphate

synthase-like

Possibly involvement in transcriptional

regulation of the Wnt signalling pathway

(such as that involved in c151559_g1_i3).

0.840878 -0.61403 -0.22684

mediator of rna polymerase

ii transcription subunit 12-

like protein

Probable transcription coactivator -0.6969 -0.36525 1.062149

---NA--- Probable role in the uptake and/or removal

of substrates for cellular activity,

otherwise uncertain role.

-0.5158 -0.85558 1.371381

solute carrier family 45

member 3-like isoform x1

Transmembrane transportation -1.25549 1.007915 0.247571

71

---NA--- Probable stress response role, possible

involvement in actin cytoskeleton

regulation.

1.19628 -0.59814 -0.59814

serine threonine-protein

kinase osr1-like

Probable role in phosphorylation -1.22247 0.593551 0.628921

zinc finger protein 318 Molecule binding 0.048031 -0.76392 0.715891

e3 ubiquitin-protein ligase

dtx4

Probable component of nuclear pores,

structures involved in the transport of

proteins and RNA in and out of the

nucleus and retention of immature mRNa.

1.089058 0.217575 -1.30663

nuclear pore complex

protein nup153-like

Probable involvement in

absorbtion/reabsorbtion of sugars in the

intestines and/or kidneys.

-0.06958 1.389346 -1.31976

sodium glucose

cotransporter 4

Possible involvement in the attachment of

cells to one another and/or to the

extracellular matrix.

1.019557 -0.74725 -0.27231

notch gene homolog 3-like Possible involvement in signal

transduction related to cell cycle and

apoptosis regulation, possible involvement

in platelet formation.

-1.18616 0.281792 0.904363

wd repeat-containing protein

66

Rab GTPase pathway activator, probably

involved in resulting intracellular

transport of vesicles along the

cytoskeleton.

-0.52416 -0.52416 1.048328

tbc1 domain family member

24-like isoform 1

Probale role in GTPase signalling -0.50646 0.633157 -0.12669

deubiquitinating protein

vcip135

Protease involved in ubiquitin cleaving -1.506 1.474988 0.031013

---NA--- N/A -0.69151 1.383025 -0.69151

---NA--- N/A 0.019061 0.841447 -0.86051

flocculation protein flo11-

like

Possible role in cell-cell adhesion -1.16424 0.802226 0.362013

PREDICTED:

uncharacterized protein

LOC101845475

Probable involvement in the transport of

ornithine into the inner mitochondrial

matrix as part of the urea cycle.

-0.08926 -0.78372 0.872986

PREDICTED:

uncharacterized protein

LOC101845475

Probable involvement in the transport of

ornithine into the inner mitochondrial

matrix as part of the urea cycle.

0.309428 -0.96577 0.656337

72

mitochondrial ornithine

transporter

Probable role in cell transport 1.602121 -0.23944 -1.36268

---NA--- N/A 1.084683 1.038366 -2.12305

---NA--- Probable involvement in inhibition of

apoptosis.

0.930956 -0.65659 -0.27437

---NA--- N/A 0.549234 -1.07695 0.527719

---NA--- Probable involvement in recognition of

mRNA with nonsense mutations leading

to subsequent degradation.

0.318546 0.686917 -1.00546

regulator of nonsense

transcripts 2-like

Probable involvement in the transport of

materials across the cell membrane.

-1.30069 0.623101 0.677586

copine-8 Probable involvement in the proper

functioning of lysosomes.

1.607827 -0.65992 -0.9479

lysosome membrane protein

2-like

Probable role in cell regulation 0.560005 -0.81783 0.257822

receptor-type tyrosine-

protein phosphatase mu-like

Probable role in regulation of cellular

processes as signalling molecules

1.081427 -0.93349 -0.14794

receptor-type tyrosine-

protein phosphatase t

Probable role in regulation of cellular

processes as signalling molecules

0.35996 -1.43062 1.070665

ankyrin repeat and sam

domain-containing protein

1a isoform x7

Probable role in protein-protein mediation -1.60627 1.744439 -0.13817

---NA--- N/A -0.82104 1.135895 -0.31486

microprocessor complex

subunit dgcr8-like

Required for miRNA processing 0.851314 -1.79653 0.945214

---NA--- N/A 1.266738 -1.55975 0.293014

---NA--- N/A 1.509309 -0.44546 -1.06385

---NA--- Probable pro and/or anti-apoptosis

functionality.

1.650166 -0.36384 -1.28632

baculoviral iap repeat-

containing protein 7

Probable role in transferring of ubiquitin

to proteins for subsequent degradation.

-0.29789 -1.35421 1.652098

protein rmd5 homolog a-like Probable role in the recognition of

destabilising proteins, resulting in

ubiquitination and degradation.

-0.36677 1.48637 -1.1196

e3 ubiquitin-protein ligase

ubr4

Cellular cytoskeleton organising protein -1.14482 0.449571 0.695247

spectrin beta brain 4 Probable role in homeostasis 1.027521 0.789944 -0.04839

---NA--- N/A -0.98688 1.035267 -0.04839

73

Table 6. Mean expression values for N. albicilla differentially expressed genes under different temperature treatments.

Contig description Probable function(s) 14°C 22°C 31°C

coup transcription factor 2

isoform x3

Probable function in development 2.272 3.519

-2.896

heat shock protein Probable involvement in the preventation of

protein denaturation and aggregation.

-1.667 1.357 2.881

serine threonine-protein

phosphatase 2a 56 kda

regulatory subunit delta

isoform

Serine/threonine-protein phosphatase,

possible involvement in cell cycle regulation

and/or cell upkeep.

-1.902 1.295

2.511

cugbp elav-like family member

2-like

Possible involvement in the regulation of

alternative splicing, as well as in maintaining

stability and promoting translation of mRNA

transcripts.

1.4501 -2.889 1.439

78 kda glucose-regulated

protein

Possibly located in the ER and involved in

protein folding and the prevention of

damaged proteins leaving the ER.

-1.743 4.817 2.379

heat shock protein beta-1-like

isoform 1

Probable involvement in the prevention of

protein denaturation and aggregation.

-2.004 3.437 3.482

---NA--- ---NA--- -2.63636 5.730 4.178

mitogen-activated protein

kinase kinase kinase 9

Tyrosine protein kinase, possible involvement

in cell cycle regulation, cell upkeep, and/or

cell motility.

-2.009 2.504 1.515

heat shock protein 90 Probable involvement in protein folding and

refolding of damaged proteins, with possible

focus on protein kinases and steroid hormone

receptors.

-1.501 7.496 2.201

eukaryotic translation initiation

factor 4h

Possible involvement in the recruitment of

mRNA transcripts into ribosomes, generally

increases the rate of transcript translation.

0.293 3.369 -1.831

protein furry-like Probable regulation of the actin cytoskeleton

structure.

-1.253 3.760 -1.253

heat shock protein 70 Possibly located in the ER and involved in

protein folding and the prevention of

damaged proteins leaving the ER.

-3.445 4.047 5.017

74

heat shock protein 70 Possibly located in the ER and involved in

protein folding and the prevention of

damaged proteins leaving the ER.

-3.814 3.103 5.688

transmembrane and coiled-coil

domains protein 1

Probable functions in endoplasmic reticulum -1.637 0.991 2.283

leucine-rich repeats and

calponin homology domain

containing 3

Probable regulation of the actin cytoskeleton

structure.

2.344 -0.490 -1.854

hypothetical protein

LOTGIDRAFT_159013

---NA--- -0.837 -1.688 2.525

lim domain-binding protein 3 Potential role in neurone development -1.344 -0.564 1.908

nitric oxide synthase Probable production of nitric oxide, possible

involvement in neural signalling or

neuroendocrine stress response.

-0.623 -2.082 2.706

nitric oxide synthase Probable production of nitric oxide, possible

involvement in neural signalling or

neuroendocrine stress response.

0.933 4.725 -2.829

rho guanine nucleotide

exchange factor 17

Probable activator of RhoA GTPase, possible

downstream resulting stress fibre formation

and actin cytoskeleton regulation.

-2.257 1.905 0.352

fras1-related extracellular

matrix protein 2-like isoform

x2

Possible involvement in basement membrane

structure and function.

-2.008 -0.473 2.482

polyketide synthase Probable involvement in polyketide synthesis.

The polyketide’s function is unknown.

-2.187 0.547 1.640

Expression values of differentially expressed genes revealed differing patterns

for the two Nerita species particularly at the highest temperature treatment (31 °C).

The higher temperature treatment for Nerita melanotragus primarily reveals up-

regulation of transport, signalling and transcriptional regulatory genes and the down

regulation of genes involved in apoptosis and sensory functions. Of the differentially

expressed genes showing up-regulation at the 31 °C treatment, five are involved in

cellular transport, four are involved in transcription related functions, one in cell

signalling and eight with unknown functions. In contrast, majority of differentially

expressed genes in Nerita albicilla were molecular chaperones of different molecular

weight. In fact, of the 13 genes found to only be up-regulated at 31 °C, six were

75

molecular chaperones and three were transcripts encoding heats shock protein 70. Of

the genes up-regulated in the 31 °C treatment, none were found in common between

the species.

4.3.3 Frontloading of HSP Genes

Of the seven heat shock proteins found to be differentially expressed in Nerita

albicilla only six had clear orthologs in N. melanotragus. Two of the heat shock protein

70 transcripts had the same ortholog in N. melanotragus, which meant only five unique

comparisons could be made. Of the five heat shock protein genes in common between

the species, there was little evidence of frontloading in three (Table 8). The three heat

shock proteins to show front loading were both N. albicilla heat shock protein 70

transcripts with the same ortholog in N. melanotragus and heat shock protein Beta in

N. albicilla (Table 8).

Table 7 Mean baseline expression of eat shock protein in Nerita melanotragus and N. albicilla.

Gene Nerita melanotragus Nerita albicilla

heat shock protein 1.538 1.356

78 kda glucose-regulated

protein

5.739 4.817

heat shock protein beta-1-

like isoform 1

1.499 3.437

heat shock protein 90 7.083 7.496

heat shock protein 70 1.103 4.047

4.3.4 Differential protein abundance under temperature stress

A total of 80 and 109 proteins showed differential abundance in Nerita

melanotragus and N. albicilla respectively. BLASTp analysis of these genes revealed

biological functions for 75 of 80 proteins in Nerita melanotragus and 103 of 109 for

N. albicilla. Proteins assigned with GO terms were compared between species (Figure

S5). For a full list of all the proteins that show differential abundance profiles, the

name of their top BLAST hit and their putative functions please see Table 9 and 10.

76

Table 8. List of proteins that show differential expression profiles, the name of their top BLAST hit and their putative functions for N. melanotragus.

Protein Description Probable function(s)

---NA--- N/A

sarcoplasmic calcium-binding

proteins and iv-like

Intracellular storage and release of calcium

from the sacroplasmic reticulum.

macpf domain-containing protein

4

Probable roles in immunity.

micos complex subunit mic10 Associated with the formation and

maintainence of mitochondrial cristae.

peptidoglycan binding domain-

containing protein

Peptidoglycan binding function.

protein crumbs partial Probable role in epithelial integrity.

deleted in malignant brain tumors

1 protein

Probable role in immunity.

transcription factor btf3 homolog

4

Probable role in transcriptional initiation. This

particular protein forms a stable complex with

RNA polymerase.

---NA--- N/A

---NA--- N/A

protein 1 N/A

peptidase m20 domain-containing

protein 2

Probable role in the regulation of cellular

proteins.

choline acetyltransferase Involved in the synthesis pathway of the

neurotransmitter aceytlcholine

vwa domain-containing protein Proabable functions in blood.

---NA--- N/A

williams-beuren syndrome

chromosomal region 27

N/A

signal transduction protein Functions in cell signal transduction. These

particular proteins have possible roles in

stress response.

77

coiled-coil-helix-coiled-coil-helix

domain-containing protein 2

Probable functions in stress response.

calmodulin-like protein 5 isoform

x1

Involved in calcium binding. Probable role in

stress response.

deleted in malignant brain tumors

1

Probable role in immunity.

ef-hand domain-containing

protein d2

Calcium binding protein. Probable role in

calcium regulation.

perlucin-like isoform x1 N/A

transmembrane emp24 domain-

containing protein 7

Probable functions in plasma membrane and

vesicular protein trafficking.

protein pif-like N/A

---NA--- N/A

ribonucleoprotein ptb-binding 2 Probable role in nucleotide binding.

multifunctional methyltransferase

subunit trm112-like protein

Functions in methylation of protein and tRNA

species.

chromobox protein homolog 1-

like

Functions in inner nuclear membrane.

chloride intracellular channel

protein 5-like

Probable function as chloride transporter.

universal stress protein a-like

protein

Probable functions in stress response.

sentrin-specific protease 8 Probable role in endopeptidase activity.

macpf domain-containing protein

4

Probable role in immunity.

metal binding protein Possible role in storage and transport of

proteins.

sushi domain-containing protein 2 Probable role in immunity.

sarcoplasmic calcium-binding

proteins and vii-like

Probable role in movement.

regucalcin Probable role in calcium homeostasis.

78

inter-alpha-trypsin inhibitor heavy

chain h4 isoform x4

Probable role in response to inflammation.

transforming growth factor-beta-

induced protein ig-h3-like

Probable role in cell adhesion.

cathepsin l1-like Plays a role in intracellular protein

catabolism.

---NA--- N/A

ump-cmp kinase mitochondrial Possible involvement in mtDNA depletion.

thap domain-containing protein 4 Probable role as an apoptotic regulator

retinoid-inducible serine

carboxypeptidase

Possible involvement in vascular wall and

kidney homeostasis.

clathrin light chain b-like isoform

x2

Probable role in peptide binding.

methylcrotonoyl- carboxylase

beta mitochondrial-like

Possible function in acid catabolism.

collagen alpha-2 chain Involved in extracellular matrix structure.

death-associated protein 1 Probable function as a cell death mediator.

low quality protein: sco-spondin Probable involvement in neural axon

structuring.

nucleolysin tiar-like isoform x4 RNA binding protein.

ribonuclease uk114 Possible involvement in mRNA cleaving.

glycerol-3-phosphate

mitochondrial-like isoform x1

Probable role in calcium binding.

tyrosine aminotransferase-like Probable function in tyrosine breakdown.

adp-ribose mitochondrial Probable role in modulating mitochondrial

activity.

apextrin-like protein Possible function in bacterial recognition.

programmed cell death protein 6-

like isoform x2

Probable function in programmed cell death.

deleted in malignant brain tumors

1

Probable role in immunity.

79

PREDICTED: uncharacterized

protein LOC101862525

N/A

deleted in malignant brain tumors

1 protein

Probable role in immunity.

beta-arrestin-1-like isoform x1 Probable functions in regulating agonist-

mediated G-protein coupled receptor.

c-binding protein Possible roles in RNA binding.

cleavage and polyadenylation

specificity factor subunit 6

Probable role in mRNA processing.

universal stress protein sll1388-

like isoform x1

Probable functions in stress response.

sorting nexin-12 Probable roles in membrane association.

microtubule-associated protein

tau-like isoform x7

Possible promoter of microtuble assembly and

stability.

carnitine o-acetyltransferase Possible function as an acetyl-CoA

transporter.

integrin alpha-4 Possible involvement in ligand binding.

ras-related c3 botulinum toxin

substrate 1

Functions in enzyme binding.

coatomer subunit beta partial Probable function in protein transport.

carnosine synthase 1 Probable role as a catalyst.

fibrillin-2-like Probable function as a regulator of elastic

fiber assembly.

choline transporter-like protein 2 Probable role as a choline transporter.

fibrillin-2-like Probable function as a regulator of elastic

fiber assembly.

cytochrome p450 family Probable function in the metabolization of

toxic compounds. Possible role in stress

response.

aspartate--trna cytoplasmic Probable role as a catalyst.

vacuolar protein sorting-

associated protein 13c isoform x3

Probable function in mitochondrial function.

Possible involvement in protein transport.

80

PREDICTED: uncharacterized

protein LOC105325083

N/A

trichohyalin-like isoform x2 Possible role in the organisation of the cell

envelope.

calmodulin-like protein 5 Involved in calcium binding. Probable role in

stress response.

rhomboid-related protein 2-like Probable involvement in regulated

intramembrane proteolysis

Table 9. List of proteins that show differential expression profiles, the name of their top BLAST hit and their putative functions for N. albicilla.

Protein Description Probable function(s)

low-density lipoprotein

receptor-related protein 5-like

Probable role in skeletal homeostasis

---NA--- ---NA---

serine threonine-protein

phosphatase 2a 56 kda

regulatory subunit epsilon

isoform

Probable involvement in the anchoring of

proteins into the cellular membrane.

PREDICTED: uncharacterized

protein LOC106876016

---NA---

3-ketoacyl- mitochondrial Probable role in lipid metabolism.

coronin-2b-like isoform x2 Possible role in vesicular transportation.

tyrosine--trna cytoplasmic Possible role in RNA binding

---NA--- ---NA---

---NA--- ---NA---

lambda-crystallin homolog Probable role in fatty acid metabolism.

PREDICTED: uncharacterized

protein LOC106078900

---NA---

collagen alpha-1 chain-like Probable role in the strengthening and support of

cartilage, bone, tendon and skin

---NA--- ---NA---

81

calcium calmodulin-dependent

protein kinase type 1-like

Probable role In the regulation of transcription

activators.

---NA--- ---NA---

uv excision repair protein

rad23 homolog b

Probable role in global genome nucleotide

excision repair.

endoglucanase 4-like isoform

x2

Probable function in cellulose activity.

---NA--- ---NA---

nadh dehydrogenase Involved in mitochondrial electron transport.

Possible functions in response to stress.

catenin delta-1-like isoform x7 Probable role in cell adhesion. Binds to and

inhibits the transcriptional repressor ZBTB33.

legumain Probable function as a catalyst.

hypothetical protein

LOTGIDRAFT_232799

---NA---

regucalcin Probable role in calcium uptake and

homeostasis.

---NA--- ---NA---

cysteine-rich secretory protein

2-like

Possible role in ion channel regulation.

titin- partial Probable role in stabilizing filaments.

PREDICTED: raftlin-like May play a pivotal role in the formation and/or

maintenance of lipid rafts

l-galactose dehydrogenase-like Possible role in catalyzing the oxidation of L-

galactose to L-galactono-1,4-lactone in the

presence of NAD

hydroxyacid-oxoacid

mitochondrial

Probable role in molecular hydrogen transport

---NA--- ---NA---

---NA--- ---NA---

hypothetical protein

LOTGIDRAFT_229978

---NA---

82

tryptophan -dioxygenase-like Probable role in the protein synthesis.

---NA--- ---NA---

epidermal growth factor

receptor kinase substrate 8-like

protein 2 isoform x1

Probable function as an actin

cytoskeleton regulator.

---NA--- ---NA---

epimerase family protein

sdr39u1

---NA---

upf0047 protein ---NA---

---NA--- ---NA---

proteasome subunit alpha type-

7-like

Probable involvement in the proteolytic

degradation of most intracellular proteins.

caveolin-1-like Probable function as a scaffolding protein within

caveolar membranes.

uncharacterized oxidoreductase Possible functions in phosphogluconate

dehydrogenase (decarboxylating) activity.

eukaryotic translation initiation

factor 3 subunit f-like

Probable function in the initiation of protein

synthesis.

dnaj homolog subfamily c

member 10-like

Probable involvement in both the correct folding

of proteins and degradation of misfolded

proteins.\

fibril-forming collagen alpha

chain-like

Functions in extracellular matrix as a

key structural protein.

fibril-forming collagen alpha

chain-like

Functions in extracellular matrix as a

key structural protein.

heat shock protein 70 Possibly located in the ER and involved in

protein folding and the prevention of damaged

proteins leaving the ER.

deleted in malignant brain

tumors 1

Probable role in immunity.

dead-box helicase dbp80 Probable functions in mRNA transport.

glutaryl- mitochondrial-like Possible role as a catalyst.

83

heat shock protein 68-like

isoform x13

Probable involvement in the prevention of

protein denaturation and aggregation.

hypothetical protein

CAPTEDRAFT_174341

---NA---

sco-spondin-like isoform x1 Possible role in development.

ribonuclease h1 Probable role in the degradation of RNA in

RNA-DNA hybrids

oxysterol-binding protein 1-

like isoform x2

Probable function as a lipid transporter involved

in lipid countertransport between the Golgi

complex and membranes of the endoplasmic

reticulum.

PREDICTED: uncharacterized

protein LOC101860536

---NA---

cellulase Probable function as an enzyme involved in the

breakdown of cellulose.

ran gtpase-activating protein 1 Possible role as a GTPase activator for the

nuclear RAS-related regulatory protein Ran,

converting it to the putatively inactive GDP-

bound state.

heat shock protein partial GTPase activator for the nuclear Ras-related

regulatory protein Ran, converting it to the

putatively inactive GDP-bound state.

sorting nexin-4 Possible involvement in several stages of

intracellular trafficking.

stomatin-like protein

mitochondrial

Mitochondrial protein that probably regulates

the biogenesis and the activity of mitochondria.

sorting nexin-4-like Possible involvement in several stages of

intracellular trafficking.

integrin alpha-m-like Possible roles in various adhesive interactions of

monocytes,

rhomboid-related protein 2-like Possible involvement in regulated

intramembrane proteolysis and the subsequent

84

release of functional polypeptides from their

membrane anchors.

cathepsin l-like Probable role in cellular turnover.

tumor protein p63-regulated

gene 1-like protein

---NA---

peroxisomal sarcosine oxidase-

like

Possible role in receptor binding.

short-chain collagen c4-like

dipeptidyl peptidase 1 Probable function as both an exopeptidase and

endopeptidase

xaa-pro partial Probable function in collagen metabolism.

cytochrome b5 Probable function function as an electron carrier

for several membrane bound oxygenases.

Sites

hypothetical protein ---NA---

syntaxin-7-like isoform x1 Possible involvement in protein trafficking from

the plasma membrane to the early endosome

(EE) as well as in homotypic fusion of endocytic

organelles.

---NA--- ---NA---

PREDICTED: uncharacterized

protein LOC106868605

isoform X3

---NA---

cathepsin l1-like Probable role in cellular turnover.

semaphorin-2a isoform x2 Probable role in axon function.

calponin homology domain-

containing protein

ddb_g0272472-like

Possible function as cytoskeletal and signal

transduction proteins.

voltage-dependent calcium

channel subunit alpha-2 delta-

2-like isoform x3

Possible function as a calcium regulator.

85

cysteine--trna cytoplasmic

isoform x4

Possible role in ATP binding.

aldose 1-epimerase-like Probable functions in glycolysis

and gluconeogenesis.

---NA--- ---NA---

cathepsin b Probable role in cellular turnover.

synaptosomal-associated

protein 25-like isoform x3

Probable role in the molecular regulation of

neurotransmitter release.

aminopeptidase b-like Possible functions in peptide binding.

zinc metalloproteinase nas-13-

like

Possible role in metal ion binding.

gtpase imap family member 7-

like

Probable functions GTP binding.

trichohyalin- partial Probable role in the organization of the cell

envelope.

ef-hand domain-containing

protein 1-like

Possible role as a regulator of cell morphology.

caspase-2 isoform x1 Probable role in apoptosis.

calcium binding protein Probable functions in calcium binding.

lipoyltransferase mitochondrial Possible role in transferase activity.

receptor-type tyrosine-protein

phosphatase alpha-like

Probable function as a signalling molecule.

mucin-like protein Probable function in secretory proteins.

dsba-like thioredoxin domain

containing protein

Possible role as a secretory protein,

alpha-l-fucosidase-like Probable role in the degradation of glycans.

erlin-1 Possible Involvement in regulation of cellular

cholesterol homeostasis.

exoglucanase -like Probable function in degrading cellulose.

2-oxoisovalerate

dehydrogenase subunit

mitochondrial

Possible functions in glyoxylate metabolic

processes.

86

hydroxymethylglutaryl-

mitochondrial-like

Probable functions in ketone synthesis.

adipocyte plasma membrane-

associated protein

Possible role in adipocyte differentiation.

kremen protein 2-like Possible role as a dependance receptor.

ankyrin repeat domain-

containing protein 40-like

Probable role in DNA binding.

leukotriene a-4 hydrolase Probable function in arachidonic acid

metabolism.

neurotrypsin- partial Possible role in neuronal plasticity.

von willebrand factor d and egf

domain-containing

Possible functions in the extracellular region.

exoglucanase -like Probable function in degrading cellulose.

microtubule-actin cross-linking

factor 1-like isoform x5

Possible role in delivery of transport vesicles

containing GPI-linked proteins from the trans-

Golgi network.

Of the 80 differentially abundant proteins in Nerita melanotragus, 65 revealed

decreased abundances and 15 showed increased abundances at 31 °C. At 45 °C, 58

proteins showed decreased abundances and 22 increased abundance. Of these, a

number of calcium regulatory proteins showed decreased abundance while a signal

transduction protein showed increased abundance. In N. albicilla, 109 proteins showed

differential abundance, of which 89 showed decreased abundance and 20 increased

abundance at 31 °C. At 45 °C, 92 showed decreased abundance and 17 increased

abundance. The higher temperature treatments revealed increased abundances of

proteins involved in cytoskeleton function/regulation. Large differences in GO terms

were observed when comparing differentially abundant proteins from both species

with a large over-representation of many GO categories in N. albicilla (Figure 12).

From the 80 and 109 differentially abundant proteins in the respective species, a

number of stress related proteins were identified. In particular, these included two

universal stress proteins, two calmodulin proteins, a cytochrome p450 protein, and a

coiled coil domain containing protein for Nerita melanotragus. For Nerita albicilla,

87

these includ a regucalcin protein and three heat shock proteins (HSP70, HSP68-like

and an unannotated heat shock protein).

Figure 12. WEGO plot for differentially expressed proteins of N. melanotragus and N. albicilla.

A comparison of differentially expressed genes and differentially abundant

proteins revealed little overlap between each species for both datasets (Figure S5). A

comparison of differentially expressed genes and differentially abundant proteins in

Nerita melanotragus revealed an overlap of a spondin protein and a thap domain-

containing protein. Nerita albicilla revealed an overlap of 3 heat shock proteins, a

serine threonine-protein phosphatase and a eukaryotic translation initiation factor.

When comparing the differentially expressed genes and differentially abundant

proteins across the Nerita species, only four proteins showed overlap. These include

regucalcin, a deleted in malignant brain tumor protein, a rhomboid-related protein and

a cathepsin protein. No differentially expressed transcripts were found in common

between the species.

88

Figure S 5. Venn diagram highlighting overlap of differentially expressed genes and proteins for N. melanotragus and N. albicilla.

4.3.5 Correlation of gene expression and protein abundance

Comparison of gene expression and corresponding protein abundance for each

species revealed a small but significant correlation between the proteome and

transcriptome (Figure 13). The linear regression indicates that the relationship between

protein and transcript abundance is positive, but is weak.

Figure 13. Scatterplot comparison of gene expression and protein abundance for BWN (Nerita albicilla) and BN (Nerita melanotragus) at 14 °C, 22 °C and 31 °C.

89

90

4.4 DISCUSSION

This study provides a physiological understanding of two ecologically important

neritid species, N. melanotragus and N. albicilla. Specifically, the data generated here

has allowed the examination of the molecular response for two closely related species

that occur in different areas of the intertidal zone under thermal stress. A strong thermal

stress response was observed in the transcriptomic data for the low intertidal zone

species, N. albicilla, but not in the mid intertidal zone species, N. melanotragus. Some

evidence of a thermal stress response was observed in the quantitative proteomic data

for both species, but different proteins were involved in this process for both species.

Little concordance was observed between the quantitative transcriptomic and

proteomic datasets within either species. The transcriptomic dataset supports the

hypothesis that low intertidal species undergo thermal stress responses at lower

temperatures compared to those that live higher in the intertidal zone, but this pattern

is far weaker in the proteomic dataset.

4.4.1 Transcriptome changes in response to heat stress

Nerita albicilla (low intertidal zone) displayed an expression pattern consistent

with a thermal stress response at 31°C. This can be seen through the up-regulation of

a suite of heat shock proteins (HSP20, HSP Beta, HSP70 and HSP90) and other stress

response genes, such as Mitogen-activated protein kinase kinase kinase 9 and

polyketide synthase. The mitogen-activated protein kinase 9 gene (MAP3K9) is part

of a super-family of MAP kinases that play a key role in regulation of gene expression

and intracellular signal transduction in response to changes in the environment,

including heat stress [Schaeffer & Weber, 1999; Kyriakis et al., 1994]. In fact, the

over-expression of genes in the MAPK pathway has been previously observed in

molluscs experiencing thermal stress [Kefaloyianni et al., 2005]. The MAP3K9 gene

has previously been shown to have an adaptive response to heat stress [Kyriakis et al.,

1994], potentially as a protective measure against premature cell death (anti-apoptosis

signalling), but also as an inducer of cell death (pro-apoptosis) when necessary

[Dhanasekaran & Reddy, 2008]. Consequently, the over-expression of MAP3K9 at

higher temperatures in N. albicilla is likely also an adaptive response to stress.

Polyketide synthase is an enzyme (or enzyme complex) responsible for the

synthesis of polyketides [Hopwood & Sherman, 1990]. Polyketides are a class of

91

secondary metabolites generated from a wide array of complex compounds that can

have a range of properties, such as anti-fungal, anti-biotic and immuno-suppressive

properties in some species [Hopwood & Sherman, 1990]. While it is not clear from the

transcriptome results which polyketide metabolite was being produced through the

increased of polyketide synthase under higher temperatures in N. albicilla, it is

interesting that this secondary metabolite producing gene was up-regulated during a

period of environmental stress. It also remains unclear whether the increased

expression of polyketide synthase is an adaptive response to stress or a detrimental

result of stress in this species.

The strong induction of heat shock proteins has been observed in other intertidal

mollusc species exhibiting thermal stress [Wang et al., 2012; Gleason & Burton, 2015]

including the induction of HSP70 in Crassostrea gigas [Wang et al., 2012]. Heat shock

proteins (HSPs) contain a number of highly conserved superfamilies of proteins

(HSP20, HSP Beta, HSP70 and HSP90) originally found to be expressed during

periods of heat stress [Schlesinger, 1990], but have since been discovered to be

expressed under a range of stresses [Morimoto et al., 1996]. During periods of heat

stress (and other forms of stress), proteins may not fold correctly or may denature and

unfold, posing a threat to cellular functioning. Under these conditions, HSPs are

rapidly expressed at high levels and work to stabilise and re-fold damaged or denatured

proteins [Parsell & Lindquist, 1993]. In N. albicilla, it is likely that the induction of

HSPs provides an adaptive protective response under thermal stress.

Heat shock proteins are also present in cells under normal environmental

conditions and are essential for cellular metabolism, acting as molecular chaperones

and aiding in folding newly synthesised proteins [Morimoto, 1996]. Some recent

studies have correlated the high baseline expression of HSPs (front loading) with

thermal stress resilience in a number of intertidal and marine species [Barshis et al,

2013]. Front loading usually occurs in species that experience chronic thermal stress,

but here, it was found that the only front loaded HSPs occurred in N. albicilla, which

were opposite to the pattern hypothesised as N. melanotragus experiences greater

periods of exposure as a mid intertidal species. Currently, there is no explanation as to

why front loading in N. melanotragus was not observed, but it may be that this species

uses other processes to escape thermal stress [Chapperon and Seuront, 2011].

92

Nerita melanotragus showed little evidence of a transcriptional response to

thermal stress at 31°C. The only indication of stress response was that this species

exhibited down-regulation of a protein translation promoting gene, which has been

seen previously during thermal stress response [Ribeiro et al., 2012]. Apart from this

gene, there does not appear to be a clearly defined molecular response in N.

melanotragus that would be associated with thermal stress. In part, this is not

surprising as previous research has shown that mid intertidal species do not experience

a thermal stress response at the same temperatures as low intertidal species. Overall,

this may indicate that N. melanotragus may be more resilient to thermal stress than N.

albicilla, particularly at the temperatures compared in the transcriptome sequencing

experiment.

4.4.2 Proteome changes in response to heat stress

Overall, only limited evidence of a thermal stress response was observed in the

proteome of either Nerita species. Three proteins that have previously been implicated

in a thermal stress response in other intertidal molluscs were found to be differentially

abundant in N. albicilla (Liu & Chen, 2013; Snyder et al., 2001). These proteins were

all heat shock proteins which are known to play an important role in thermal stress

response in mollusc species from the low intertidal zone (see discussion in section

4.5.1 for an explanation of function and their role in thermal stress response). Besidex

these three proteins, very few other differentially abundant proteins from N. albicilla

have been implicated as having a direct role in stress response in other organisms. In

N. melanotragus, six differentially abundant proteins have previously been shown to

play a role in stress response. These six proteins included two universal stress proteins,

two calmodulin proteins, a cytochrome p450 protein and a coiled coil domain

containing protein. Of these, those with the best known association with stress

response are the universal stress proteins (Nachin et al., 2005). Universal stress

proteins show increased abundant under a range of stress conditions, including thermal

stress, and are associated with increased resistance to stress in a range of organisms

[Timmins-Schiffman et al., 2014; Tkaczuk et al., 2013; Nachin et al., 2005]. The

mechanism by which universal stress proteins act in N. melanotragus are still largely

unknown. Further study is required to determine if universal stress proteins provide

93

resilience to thermal stress in mid intertidal species when compared to low intertidal

species.

Calcium signalling is considered a crucial process in response to a number of

abiotic and biotic stressors [Wu & Jinn, 2012]. An increased abundance of calmodulin

proteins in response to stress has been mostly documented in plant species and they

play important roles in the activation of cell signalling machinery during stress [Gupta

et al., 2014]. Due to a lack of studies of calmodulin in mollusc species, it is unknown

whether these proteins play a similar role in these species. Coiled coil proteins have

been found to be involved in crucial interactions such as transcriptional control [Mason

& Arndt, 2004]. Similarly, there is a lack of documentation of coiled coil proteins in

mollusc species, making it difficult to understand their function. Although not directly

involved in thermal stress response, cytochrome p450 differential abundance has been

identified in relation to hydrocarbon exposure [Snyder et al., 2001]. These proteins are

primarily involved in the metabolism of chemical compounds in mollusc species

[Gagnaire et al., 2010], so may only have a limited role in the thermal stress response

of these species.

There was a lack of overlap between the transcriptome/transcriptome, and

proteome/proteome among species as well as the transcriptome/proteome datasets

within species. Large differences in comparisons among species was expected as

previous research has shown that low and mid intertidal mollusc species have largely

different gene expression and protein responses to heat stress (Somero, 2002; Stillman

and Somero, 1996). This study only reinforces the idea that mid and low intertidal

species have very different thermal responses to similar temperatures and that low

intertidal species experience thermal stress at lower temperatures.

What was more interesting though was the lack of concordance between the

transcriptome/proteome datasets within species, with only a few differentially

expressed genes/proteins in common in both species. The genes/proteins found to be

in common in N. albicilla were dominated by heat shock proteins (3/5), confirming

the importance of these proteins to thermal stress response in this species. The two

genes in common between the transcriptome and proteome dataset in N. melanotragus

are a spondin protein and a thap domain-containing protein which have not previously

been implicated in a thermal stress response in molluscs. Similar patterns of low or no

94

overlap between protein and transcriptome datasets have been observed in a range of

other organisms [Lu et al. 2007; Fagerberg et al. 2014] and an almost complete lack

of overlap was observed in the marine coral species, Seriatopora hystrix [Mayfield et

al., 2016]. In S. hystrix, only two differentially abundant proteins and differentially

expressed genes were found in common across a temperature stress experiment. The

very weak overlap observed in a marine species during a thermal stress response

experiment strongly supports the results reported here.

For the same temperature treatments within species (Figure 12) a significant but

weak positive relationship was observed between transcript and protein abundance.

This indicates translation of only few transcripts to protein controlled by mechanism

which regulates the translation of transcript to protein, such as those involved in

transcription, translaton and post-translational modification.

4.4.3 Conclusion

The large amount of transcriptomic and proteomic data generated from two

temperature stressed Nerita species provide a novel insight into the potential effects of

future climate change. This study has found that low intertidal species exhibit a more

pronounced thermal response at lower temperatures than those from the mid intertidal

zone. The preliminary investigations into proteomic data from these species provides

little evidence of a thermal response and indicates the need for further research in this

area. Furthermore, by combining proteomic and genomic approaches, this study has

identified the limitations of comparing these datasets.

95

Chapter 5: General Discussion

The importance of genomic resources in the study of physiology and evolution

is well recognised [Mochida and Shinozaki, 2010; Ramos et al, 2015; Shivhare and

Lata, 2017]. Traditional approaches such as Sanger sequencing have been superseded

by next generation sequencing technologies, allowing for more robust and efficient

studies of the gene repertoire and gene expression patterns in non-model species

[Morozova and Marra, 2008]. A fundamental requirement for large sequence data sets

is the reliable and accurate assembly, especially for organisms with no previous

genomic data. Such assemblers allow for downstream analyses into gene expansions

and contractions, investigations into gene expression patterns and can facilitate

proteomic studies, as well. Collectively, these studies allow for a better understanding

of the molecular response of non-model organisms to stress conditions, how species

adapt to extreme environments and potentially, how species may respond to future

climate change and variability.

5.1 DE NOVO ASSEMBLERS IN MOLLUSC TRANSCRIPTOME ASSEMBLIES

The demand for genomic resources for many non-model species has fuelled the

development of next generation sequencing technologies, which produce vast amounts

of sequencing data for lower costs compared to traditional sequencing approaches

[Schadt et al., 2010]. These sequencers, however, produce short fragmented sequences

which must be assembled before they are able to be used in downstream applications

[Everett et al., 2011]. As a result, large amounts of raw data are produced that must be

assembled to develop gene sets for down-stream analysis. These gene sets allow

insight into the genetic and functional information used by organisms. Currently, a

large number of assemblers exist, however, the accuracy and reliability of these

assemblers requires validation [Zerbino and Birney, 2008; Kearse et al., 2012; Luo et

al., 2012; Haas et al., 2013]. This is especially true for non-model organisms due to a

lack of a reference genome, so that they require a de novo assembly to be undertaken.

The use of several assembly statistics are used in order to determine the accuracy

and reliability of de novo assemblies. These include, N50, average contig length,

96 Error! No text of specified style in document.

longest contig length and the number of overall contigs which may indicate a large

number of redundant sequences. Assemblers such as SOAPdenovo, Velvet-Oases and

Trans-ABySS are widely available examples of current de novo assemblers; however

all are extensions of earlier developed assemblers. The development of Trinity

provided a novel method for the reconstruction of transcripts from raw read data [Haas,

2013]. When comparing the accuracy and reliability of these assemblers, Trinity and

Oases outperformed Geneious and Velvet in all metrics. When comparing Trinity and

Oases, however, Trinity had a longer N50 while Oases had less redundancy. Although

Oases showed less redundancy, software such as CD-HIT is available to address this

issue for Trinity assemblies [Huang, 2010]. Furthermore, Trinity has an active

developer community, providing excellent support and updates, constantly increasing

both efficiency and accuracy [Haas, 2013].

The use of Trinity [Haas, 2013] as a short read de novo assembler has been

extensively documented in a range of studies and taxa [Gutierrez-Gonzalez et al.,

2013; Tassone et al., 2016 Thunders and Cavanagh, 2017]. This assembler consists of

three consecutive modules and employs the use of de Bruijn graphs to reconstruct

transcripts. Furthermore, Trinity accounts for alternatively spliced isoforms from

either single end or paired end reads making it particularly effective in the study of

physiological genomics in non-model species. Specifically, this allows the potential

identification of expansions in gene families, which is crucial to the understanding of

the driving forces underlying adaptations to environmental extremes [Zhang, 2015;

Nock, 2016].

The use of k-mers (inconsistent throughout) in de novo assembly of datasets is a

key component of the assembly process. The goal of the k-mer is to enable the

determination of the number of occurances each fixed-length word of length k in a

DNA data set [Marçais and Kingsford, 2011]. The efficiency of k-mer counting plays

a cruicial role in de novo assemblies [Zhang, 2014]. Although k-mers facilitate

efficient reconstruction of datasets, they can also provide erroneous data if incorrectly

set. The Velvet/Oases package requires the user to preset a k-mer for each assembly.

Although it allows the combining of multiple k-mers, each set of raw reads is unique

in its size, error rate and composition. Consequently, this may cause an inaccurate

dataset with too few contigs, where many have been missed, or one with high numbers

of contigs and large amounts of chimeric contigs. In contrast, Trinity initially employs

97

the use of a third party package, Jellyfish to build a k-mer catalog [Haas, 2013]. As

each set of raw reads is unique, this method provides optimum k-mer selection.

The use of an efficient, accurate and reliable de novo assembler is extremely

important for studies in all fields. Although assemblers such as Velvet/Oases can

provide assemblies with less redundancy, Trinity creates contigs with a longer N50,

has a larger support system and employs an efficient k-mer optimisation method.

These factors make the use of Trinity highly suitable as a de novo assembler for the

study of evolution and physiology in non-model organisms.

5.2 STRESS GENE EXPANSION IN LOW VS MID INTERTIDAL SPECIES

The intertidal zone consists of a lower, middle, upper and spray subzones. Of

these, the lower subzones represent areas where the resident species are submerged in

water more frequently and for longer periods than those that occur in the upper

subzones. Consequently, the upper intertidal zones are where organisms experience

the the most prolonged exposure to stressors, due to tidal cycles and constant air and

sun exposure. Furthermore, organisms inhabiting these zones are faced with greater

maximum temperatures [Tomanek, 2002]. Such environmental pressure coupled with

gene duplication events may lead to the expansion of gene families involved in stress

response. Such expansions of stress response gene families have been observed in the

Pacific Oyster (Crassostrea gigas), where an extensive duplication of heat shock

protein 70 and inhibition of apoptosis gene families has occurred [Wang, 2012]. It is

suggested this expansion may enable this species to be able to persist in the harsh

environment of the intertidal zone. Studies on species from the genus Pinctada also

support the idea that interitidal species have an expanded repertoire of stress related

genes [Takeuchi, 2016]. Few studies, however, have examined if species that inhabit

different areas of the intertidal zone have different numbers of expressed genes under

normal conditions.

Differences in expressed gene copy number in species that inhabit different areas

of the intertidal zone could occur if reduced gene copy number is beneficial in species

that encounter less frequent stress conditions (lower intertidal) than those found in the

upper intertidal areas. An alternative scenario of increased of expressed copy number

in species from the lower intertidal zone could also be beneficial if the level of stress

is more acute in organisms inhabiting the lower areas. Our data supports the second

98 Error! No text of specified style in document.

idea as the mid intertidal zone species N. melanotragus had fewer stress related genes

expressed when compared to the low intertidal species N. albicilla. This may indicate

N. melanotragus is a more resilient species and has a higher stress tolerance than that

of Nerita albicilla which exhibits a more acute heat stress response.

5.3 GENE EXPRESSION IN INTERTIDAL ZONES

Investigating the physiological changes of species in response to stress is a

crucial step in the understanding how an organism copes in a harsh environment.

Organisms can employ gene expression changes as a defence mechanism to cope with

abiotic stressors such as temperature changes [Liu, 2017]. Of particular interest are

environments such as the intertidal zone, where organisms are presented with a suite

of stressors affecting virtually all physiological processes to some degree [Helmuth,

2006]. Furthermore, organisms inhabiting these areas often exhibit strong vertical

zonation [Chappuis, 2014]. Therefore, studies focusing on organisms inhabiting

intertidal zones present a great opportunity to increase our current understanding of

gene expression patterns under stress conditions.

The two study species used here N. albicilla and N. melanotragus occur in the

lower and mid intertidal zones, respectively. Distinct differences in expression patterns

were observed in these species and N. albicilla appeared to have a classic stress

response at a lower temperature than N. melanotragus based on gene expression data.

In fact, many of the genes expressed by N. albicilla are part of the cellular stress

response, a universal cellular defence mechanism which is initiated in response to

changes in the extracellular environment [Kültz, 2005]. This indicates that N. albicilla

displayed a more acute stress response when compared with the mid intertidal species

N. melanotragus. Overall this observation provides further evidence to suggest that N.

melanotragus is more resilient to temperature stress than the lower intertidal species.

A study by Madeira et al ., (2014) investigating the expression of heat shock

protein 70 in crabs from three different thermal niches revealed that species occupying

more stable niches present peaks of cellular stress responses at lower temperatures,

when compared with those that inhabit more variable environments [Madeira et al.,

2014]. As N. albicilla is submerged for far longer periods of time, it may be that lower

temperatures are also needed to illicit a thermal stress response in this species as well.

A lack of a thermal stress response in N. melanotragus may indicate that the

99

experimental temperatures used in this study are within the specie’s tolerable range.

This may also indicate that N. melanotragus is able to cope with greater temperatures

than N. albicilla. Overall, this pattern is consistent with previous studies which have

highlighted low intertidal species undergo a thermal stress response at lower

temperatures than mid intertidal species [Dong & Somero, 2009].

5.4 RESILIENCE UNDER FUTURE CLIMATES

Future climate change implications pose a host of issues for marine species.

Marine habitats are expected to become more stressful for inhabitants through changes

in abiotic stressors, such as increased water temperature. The implications of this

change on organisms are severe and in some cases may cause the local extinction of

populations. Our gene expression study in N. melanotragus and N. albicilla revealed

a much more pronounced stress response in N. albicilla as opposed to N. melanotragus

under current extreme temperature conditions. The implications suggest that N.

albicilla is less resilient to stress when compared to N. melanotragus and is also more

likely to be sensitive to future increases in temperature. In fact, thermally sensitive

species are more likely to be adversely affected under scenarios of increased

environmental temperatures. For example, a temperature stress experiment on

thermally sensitive and thermally resilient populations of Acropora hyacinthus

highlighted differences in stress gene expression patterns, and this was ultimately

linked to mortality in the thermally sensitive population in the more stressful

environment [Barshis et al., 2013]. Based on this experiment, there are implications

that many thermal sensitive species may potentially face local extinction if

environmental stress increases in their habitats.

The proteomic experiment which examined current and future extreme

temperatures, however, did not find clear cut results in terms of the stress response of

either species. Both species displayed some evidence of a thermal stress response, but

it was not strong in either species. In contrast, the transcriptomic experiment revealed

an expression pattern consisten with thermal stress in N. albicilla, however, in N.

melanotragus, there was no defined thermal stress response. Evidently, differences in

transcript/protein expression also correlated with little overlap between the two.

100 Error! No text of specified style in document.

5.5 PROTEIN AND GENE CORRELATION

Recent advances in proteomics have allowed the large scale discovery of

proteins from complex samples using protein digests [Yates et al., 2009]. Using

predicted peptide sets from transcriptome data enables the generation of a proteome

data set. These data sets are currently being interrogated in a range of taxa and in a

number of studies, and some have been used to test the thermal stress response of

particular species [Poirier et al., 2014; Xie et al., 2015; Khondee et al., 2016].

Furthermore, these proteome datasets can allow indepth investigations into any given

tissue of an organism under stress conditons [Kosová et al., 2014]. Proteomic and

transcriptomic data, however, often seem to only show weak overlap or correlation

with each other in many experiments.

Although it is expected that a large portion of mRNA is translated into protein,

studies are finding that this may not always be the case [Lu et al. 2007; Fagerberg et

al. 2014; Mayfield et al., 2016]. There are a number of regulatory mechanisms that

influence the abundance of mRNA transcripts before the production of proteins, other

regulatory mechanisms that protein translation [Mayfield, 2016]. These mechanisms

include those involved in transcription, translation, post-translation modification,

processing and transport, that are governed by different temporal profiles and

regulatory pathways [Mayfield, 2016]. For example, the up-regulation of genes due to

thermal intervention may not result in increased protein production, as not every

transcript is actively translated. Many forms of RNA are known to actively regulate

transcribed mRNA before it is translated into protein. This includes small interfering

RNAs (siRNAs) and microRNAs (miRNAs) which can reduce the abundance of

mRNA transcripts and reduce translation into proteins [Rossbach, 2010]. Post

transcriptional regulation of mRNA transcripts may be one factor that accounts for

some of the variability observed between proteomic and transcriptomic datasets, but

other factors such as protein localisation also play a role in this variability.

Protein localisation, including storage and secretion, may also contribute to the

low correlation between transcriptome and proteome datasets, as it introduces a

temporal-spatial element into the distribution of proteins within cells, tissues and the

organism. Moreover, protein localisation may impact on negative and positive

feedback loops within the cellular system, as biochemical thresholds often need to be

met and exceeded for feedback loops to inhibit the expression of genes again

101

increasing variability between proteomic and transcriptomic data in many

experiments.

Other forms of protein regulation can also influence protein abundance in cells

and tissues. For example, the ubiquitin-proteosome pathway ensures that ubiquinated

proteins are actively degraded, even those that have been recently translated. A

degraded protein is unlikely to be quantitatively measured if the subsequent peptides

lack the tryptic cleavage site. Consequently, this may contribute to a large number of

proteins not accounted for due to degradation.

Furthermore, factors such as protein and mRNA half-life may also impact correlations

[Wu et al., 2008]. Protein half-lives vary with the stability of the polypeptide,

involvement in complexes, active enzymes, temperature and chemical constituents of

the microenvironment (e.g. reactive oxygen species, hydrogen ions, acids and bases,

etc.).

Finally, the difference in number of genes sequenced and proteins assayed differs

immensely with 10 000s of genes sequenced versus 1 000s of proteins assayed. These

characteristics, provide a reasonable overview that can explain the disparity between

the transcriptome and proteome and suggest comparisons between proteomic and

transcriptomic datasets have not completely matured.

5.6 CONCLUSION

By identifying the most suitable de novo assembler, it can accurately and reliably

assemble datasets for mollusc species. This is important as all downstream analyses

rely on the quality of the dataset. Consequently, any errors in the initial reconstruction

of the dataset will reflect upon downstream analyses.

The comparative transcriptome data generated has provided genomic resources

for a phylum lacking thereof. Very little is known about the physiology of organisms

from the phylum mollusca, however, genomic resources produced have provided a

step forward in this field. Furthermore, the discovery that lower intertidal zone species

exhibit a more pronounced stress response under normal conditions paves way for

expression studies which may provide a further understanding of the biology of closely

related organisms inhabiting the intertidal zone.

102 Error! No text of specified style in document.

By stress inducing two closely related marine snail species from the genus

Nerita, a further investigation into their physiology was able to be discovered.

Although transcriptomic data revealed a higher number of stress genes in the low

intertidal zone inhabiting species, the temperature stress related expression data further

confirmed the finding that the lower intertidal species is less resilient to heat stress.

Although the proteomic data generated did not correlate with the transcriptomic data,

it provides a step towards investigating not only the genomics of these species, but

also the proteomics, in order to obtain a complete systems understanding of the

physiological adaptation of these organisms.

References

Adamson KJ, Wang T, Zhao M, Bell F, Kuballa AV, Storey KB, Cummins SF: Molecular insights into land snail neuropeptides through transcriptome and comparative gene analysis. BMC Genomics 2015, 16:308.

Albertin CB, Simakov O, Mitros Td, Wang ZY, Pungor JR, Edsinger-Gonzales E, Brenner S, Ragsdale CW, Rokhsar DS: The octopus genome and the evolution of cephalopod neural and morphological novelties. Nature 2015, 524:220-224.

103

Albright R, Mason B: Projected Near-Future Levels of Temperature and pCO2 Reduce Coral Fertilization Success. PLoS ONE 2013, 8:e56468.

Alley RB, Talley LD, Wallace JM, Marotzke J, Nordhaus WD, Overpeck JT: Abrupt Climate Change. Science 2003, 299:2005-2010.

Amin S, Prentis PJ, Gilding EK, Pavasovic A: Assembly and annotation of a non-model gastropod (Nerita melanotragus) transcriptome: a comparison of De novo assemblers. BMC Research Notes 2014, 7:488.

Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, Harris MA, Hill DP, Issel-Tarver L, Kasarskis A, Lewis S, Matese JC, Richardson JE, Ringwald M, Rubin GM, Sherlock G: Gene ontology: Tool for the unification of biology. Nature Genetics 2000, 25:25-29.

Asta Lakshmi S: Wonder molluscs and their utilities. International Journal of Pharmaceutical Sciences Review and Research 2011, 6:30-33.

Bakkenist CJ, Kastan MB: Initiating cellular stress responses. Cell 2004, 118:9-17.

Barnwell FH: The role of rhythmic systems in the adaptation of fiddler crabs to the intertidal zone. Integrative and Comparative Biology 1968, 8:569-583.

Barshis DJ, Ladner JT, Oliver TA, Seneca FO, Traylor-Knowles N, Palumbi SR: Genomic basis for coral resilience to climate change. Proceedings of the National Academy of Sciences of the United States of America 2013, 110:1387-1392.

Beck FX, Grünbein R, Lugmayr K, Neuhofer W: Heat shock proteins and the cellular response to osmotic stress. Cellular Physiology and Biochemistry 2000, 10: 303-306.

Boone AN, Vijayan MM: Constitutive heat shock protein 70 (HSC70) expression in rainbow trout hepatocytes: Effect of heat shock and heavy metal exposure. Comparative Biochemistry and Physiology - C Toxicology and Pharmacology 2002, 132:223-233.

Booth NJ, Bilodeau-Bourgeois AL: Proteomic analysis of head kidney tissue from high and low susceptibility families of channel catfish following challenge with Edwardsiella ictaluri. Fish and Shellfish Immunology 2009, 26:193–96.

Brierley AS, Kingsford MJ: Impacts of Climate Change on Marine Organisms and Ecosystems. Current Biology 2009,19:602-612.

104 Error! No text of specified style in document.

Burrows MT, Schoeman DS, Buckley LB, Moore P, Poloczanska ES, Brander KM, Brown C, Bruno JF, Duarte CM, Halpern BS, Holding J, Kappel CV, Kiessling W, O'Connor MI, Pandolfi JM, Parmesan C, Schwing FB, Sydeman WJ, Richardson AJ: The pace of shifting climate in marine and terrestrial ecosystems. Science 2011, 334:652–655.

Byrne M, Lamare M, Winter D, Dworjanyn S, Uthicke S: The stunting effect of a high CO2 ocean on calcification and development in sea urchin larvae, a synthesis from the tropics to the poles. Philosophical Transactions of the Royal Society B: Biological Sciences 2013, 368:Article number 20120439

Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, Madden TL: BLAST+: Architecture and applications. BMC Bioinformatics 2009, 10: Article number 421.

Casellato S, Masiero L, Ballarin L: Toxicity of fluoride to the freshwater mollusc Dreissena polymorpha: Effects on survival, histology, and antioxidant enzyme activity. Fluoride 2012, 45:35-46.

Castro P, Huber ME: Marine Biology Boston: McGraw-Hill 2003.

Chapperon C, Seuront L: Space-time variability in environmental thermal properties and snail thermoregulatory behaviour. Functional Ecology 2011, 25:1040-1050.

Chapperon C, Seuront L: Temporal shifts in motion behaviour and habitat use in an intertidal gastropod. Journal of the Marine Biological Association of the United Kingdom 2013, 93:1025-1034.

Chapperon C, Seuront L: Variability in the motion behaviour of intertidal gastropods: ecological and evolutionary perspectives. Journal of the Marine Biological Association of the United Kingdom 2011, 91:1717.

Chappuis E, Terradas M, Cefalì ME, Mariani S, Ballesteros E: Vertical zonation is the main distribution pattern of littoral assemblages on rocky shores at a regional scale. Estuarine, Coastal and Shelf Science 2014, 147:113-122.

Chiara M, Horner DS, Spada A: De novo assembly of the transcriptome of the non-model plant Streptocarpus rexii employing a novel heuristic to recover locus-specific transcript clusters. PLoS One 2013, 8:e80961.

Conesa A, Götz S, García-Gómez JM, Terol J, Talón M, Robles M, Blast2GO: A universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 2005, 21:3674–3676.

Cong M, Ni D, Song L, Wang L, Zhao J, Qiu L, Li L: Molecular cloning, characterization and mRNA expression of peroxiredoxin in Zhikong scallop Chlamys farreri. Molecular Biology Reports 2009, 36:1451-1459.

105

Coppe A, Bortoluzzi S, Murari G, Marino IAM, Zane L, Papetti C: Sequencing and Characterization of Striped Venus Transcriptome Expand Resources for Clam Fishery Genetics. PLoS ONE 2012, 7:44185.

Crandall ED, Frey MA, Grosberg RK, Barber PH: Contrasting demographic history and phylogeographical patterns in two Indo-Pacific gastropods. Molecular Ecology 2008, 17:611-626.

Davenport J, Macalister H: Environmental conditions and physiological tolerances of intertidal fauna in relation to shore zonation at Husvik, South Georgia. Journal of the Marine Biological Association of the United Kingdom 1996, 76:985-1002.

De Oliveira AL, Wollesen T, Kristof A, Scherholz M, Redl E, Todt C, Bleidorn C, Wanninger A: Comparative transcriptomics enlarges the toolkit of known developmental genes in mollusks. BMC Genomics 2016, 17:905.

De Wit P, Palumbi SR: Transcriptome-wide polymorphisms of red abalone (Haliotis rufescens) reveal patterns of gene flow and local adaptation. Molecular Ecology 2013, 22:2884-2897.

Derbali A, Jarboui O, Ghorbel M: Distribution, abundance and population structure of Pinctada radiata (Mollusca: Bivalvia) in southern Tunisian waters (Central Mediterranean). Cahiers de Biologie Marine 2011, 52:23-31.

Dhanasekaran DN, Reddy EP: JNK signaling in apoptosis. Oncogene 2008, 27:6245-6251.

Dong Y, Miller LP, Sanders JG, Somero GN: Heat-shock protein 70 (Hsp70) expression in four limpets of the genus Lottia: Interspecific variation in constitutive and inducible synthesis correlates with in situ exposure to heat stress. Biological Bulletin 2008, 215:173-181.

Dong Y, Somero GN: Temperature adaptation of cytosolic malate dehydrogenases of limpets (genus Lottia): Differences in stability and function due to minor changes in sequence correlate with biogeographic and vertical distributions. Journal of Experimental Biology 2009, 212:169-177.

Drummond AJ, Ashton B, Buxton S, et al: 2010. Geneious v5.6, Available from http://www.geneious.com.

Fabry VJ, Siebel BA, Feely RA, Orr JC: Impacts of ocean acidification on marine fauna and ecosystem processes. ICES Journal of Marine Science 2008, 65:414–432.

Fagerberg L, Hallstrom BM, Oksvold P, Kampf C, Djureinovic D, Odeberg J, Habuka M, Tahmasebpoor S, Danielsson A, Edlund K, Asplund A, Sjostedt E, Lundberg E, Szigyarto CA-K, Skogs M, Ottosson Takanen J, Berling H, Tegel H, Mulder J, Nilsson P, Schwenk JM, Lindskog C, Danielsson F, Mardinoglu A, Sivertsson A, Von Feilitzen

106 Error! No text of specified style in document.

K, Forsberg M, Zwahlen M, Olsson I, Navani S, Huss M, Nielsen J, Ponten F, Uhlen M: Analysis of the human tissue-specific expression by genome-wide integration of transcriptomics and antibody-based proteomics. Molecular and Cellular Proteomics 2014, 13:397-406.

Faulkner P: Late Holocene mollusc exploitation and changing near-shore environments: A case study from the coastal margin of Blue Mud Bay, northern Australia. Environmental Archaeology 2011, 16:137-150.

Feldmeyer B, Wheat CW, Krezdorn N, Rotter B, Pfenninger M: Short read illumina data for the de novo assembly of a non-model snail species transcriptome (Radix balthica, basommatophora, pulmonata), and a comparison of assembler performance. BMC Genomics 2011, 12:317.

Feng Z-P, Zhang Z, van Kesteren RE, Straub VA, van Nierop P, Jin K, Nejatbakhsh N, Goldberg JI, Spencer GE, Yeoman MS, Wildering W, Coorssen JR, Croll RP, Buck LT, Syed NI, Smit AB: Transcriptome analysis of the central nervous system of the mollusc Lymnaea stagnalis. BMC Genomics 2009, 10:1471.

Feng ZP, Zhang Z, van Kesteren RE, Straub VA, van Nierop P, Jin K, Nejatbakhsh N, Goldberg JI, Spencer GE, Yeoman MS, Wildering W, Coorssen JR, Croll RP, Buck LT, Syed NI, Smit AB: Transcriptome analysis of the central nervous system of the mollusc Lymnaea stagnalis. BMC Genomics 2009, 10:451.

Fiedler TJ, Hudder A, McKay SJ, Shivkumar S, Capo TR, Schmale MC, Walsh PJ: The transcriptome of the early life history stages of the California Sea Hare Aplysia californica. Comparative Biochemistry and Physiology Part D: Genomics and Proteomics 2010, 5: 165-170.

Finn RD, Clements J, Arndt W, Miller BL, Wheeler TJ, Schreiber F, Bateman A, Eddy SR: HMMER web server: 2015 Update. Nucleic Acids Research 2015, 43:W30-W38.

Firth LB, Knights AM, Bell SS: Air temperature and winter mortality: Implications for the persistence of the invasive mussel, Perna viridis in the intertidal zone of the south-eastern United States. Journal of Experimental Marine Biology and Ecology 2011, 400:250-256.

Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R: DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular marine biology and biotechnology 1994, 5:294-299.

Frey MA, Vermeij GJ: Molecular phylogenies and historical biogeography of a circumtropical group of gastropods (Genus: Nerita): Implications for regional diversity patterns in the marine tropics. Molecular Phylogenetics and Evolution 2008. 48:1067-1086.

107

Frey MA: The relative importance of geography and ecology in species diversification: evidence from a tropical marine intertidal snail (Nerita). Journal of Biogeography 2010, 37:1515–1528.

Fulda S, Gorman AM, Hori O, Samali A: Cellular stress responses: Cell survival and cell death. International Journal of Cell Biology 2010, Article number 214074.

Gagnaire B, Geffard O, Noury P, Garric J: In vivo indirect measurement of cytochrome P450-associated activities in freshwater gastropod molluscs. Environmental Toxicology 2010, 25:545-553.

Gibbons JG, Janson EM, Hittinger CT, Johnston M, Abbot P, Rokas A: Benchmarking next-generation transcriptome sequencing for functional and evolutionary genomics. Mol Biol Evol 2009, 26:2731–2744.

Giribet G, Okusu A, Lindgren AR, Huff SW, Schrödl M, Nishiguchi MK: Evidence for a clade composed of molluscs with serially repeated structures: Monoplacophorans are related to chitons. Proceedings of the National Academy of Sciences of the United States of America 2006, 103:7723-7728.

Gleason LU, Burton RS: Regional patterns of thermal stress and constitutive gene expression in the marine snail Chlorostoma funebralis in northern and southern California. Marine Ecology Progress Series 2016, 24:143-159.

Gleason LU, Burton RS: RNA-seq reveals regional differences in transcriptome response to heat stress in the marine snail Chlorostoma funebralis. Molecular Ecology 2015, 24:610-627.

Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, Adiconis X, Fan L, Raychowdhury R, Zeng Q, Chen Z, Mauceli E, Hacohen N, Gnirke A, Rhind N, Di Palma F, Birren BW, Nusbaum C, Lindblad-Toh K, Friedman N, Regev A: Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotechnol 2011, 29:644–652.

Gracey AY, Chaney ML, Boomhower JP, Tyburczy WR, Connor K, Somero GN: Rhythms of Gene Expression in a Fluctuating Intertidal Environment. Current Biology 2008, 18:1501-1507.

Gupta K, Jha B, Agarwal PK: A Dehydration-Responsive Element Binding (DREB) Transcription Factor from the Succulent Halophyte Salicornia brachiata Enhances Abiotic Stress Tolerance in Transgenic Tobacco. Marine Biotechnology 2014, 16:657-673.

Gutierrez-Gonzalez JJ, Tu ZJ, Garvin DF: Analysis and annotation of the hexaploid oat seed transcriptome. BMC Genomics 2013, 14:471.

Haas BJ, Papanicolaou A, Yassour M, Grabherr M, Blood PD, Bowden Jf, Couger MB, Eccles D, Li B, Lieber M, Macmanes MD, Ott M, Orvis J, Pochet N, Strozzi F,

108 Error! No text of specified style in document.

Weeks N, Westerman R, William T, Dewey CN, Henschel R, Leduc RD, Friedman N, Regev A: De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis. Nature Protocols 2013, 8:1494-1512.

Harley CDG, Helmuth BST: Local- and regional-scale effects of wave exposure, thermal stress, and absolute versus effective shore level on patterns of intertidal zonation. Limnology and Oceanography 2003, 48:1498-1508.

Harley CDG, Hughes AR, Hultgren KM, Miner BG, Sorte CJB: The impacts of climate change in coastal marine systems. Ecological Letters 2006, 9:228–241.

Haven SB: Niche differences in the intertidal limpets Acmaea scabra and A. digitalis (Gastropoda) in central California. Veliger 1970, 13:231-248.

Helmuth B, Mieszkowska N, Moore P, Hawkins SJ: Living on the edge of two changing worlds: Forecasting the responses of rocky intertidal ecosystems to climate change. Annual Review of Ecology, Evolution, and Systematics 2006, 37:373-404.

Herpin A, Badariotti F, Rodet F, Favrel P: Molecular characterization of a new leucine-rich repeat-containing G protein-coupled receptor from a bivalve mollusc: evolutionary implications. Biochim Biophys Acta Gene Struct Expr 2004, 1680:137–144.

Heyland A, Vue Z, Voolstra CR, Medina M, Moroz LL: Developmental transcriptome of Aplysia californica. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution 2011, 316B:113-134.

Hinrichsen H-H, Huwer B, Makarchouk A, Petereit C, Schaber M, Voss R: Climate-driven long-term trends in Baltic Sea oxygen concentrations and the potential consequences for eastern Baltic cod (Gadus morhua). ICES Journal of Marine Science 2011, 68:2019-2028.

Hold N, Murray LG, Hinz H, Neill SP, Lass S, Lo M, Kaiser MJ: Environmental drivers of small scale spatial variation in the reproductive schedule of a commercially important bivalve mollusk. Marine Environmental Research 2013, 92:144-153.

Hook SE, Twine NA, Simpson SL, Spadaro DA, Moncuquet P, Wilkins MR: 454 pyrosequencing-based analysis of gene expression profiles in the amphipod Melita plumulosa: transcriptome assembly and toxicant induced changes. Aquat Toxicol 2014, article in press.

Hopwood DA, Sherman DH: Molecular genetics of polyketides and its comparison to fatty acid biosynthesis. Annual Review of Genetics 1990, 24:37-66.

Hou R, Bao Z, Wang S, Su H, Li Y, Du H, Hu J, Wang S, Hu X: Transcriptome sequencing and De novo analysis for yesso scallop (Patinopecten yessoensis) using 454 GS FLX. PLoS One 2011, 6:21560.

109

Huang Y, Niu B, Gao Y, Fu L, Li W: CD-HIT Suite: a web server for clustering and comparing biological sequences. Bioinformatics 2010, 26:680–682.

Hungate BA, Dukes JS, Shaw MR, Luo Y, Field CB: Nitrogen and Climate Change. Science 2003, 302:1512-1513.

Ip JCH, Leung PTY, Ho KKY, Qiu JW, Leung KMY: De novo transcriptome assembly of the marine gastropod Reishia clavigera for supporting toxic mechanism studies. Aquatic Toxicology 2016, 178:39-48.

Jakubowska M, Normant M: Metabolic rate and activity of blue mussel Mytilus edulis trossulus under short-term exposure to carbon dioxide-induced water acidification and oxygen deficiency. Marine and Freshwater Behaviour and Physiology 2015, 48:25-39.

Jurgen F, Valerio M, Roberto R, Paolo SG, Marta M: 2-DE proteomic analysis of HSP70 in mollusc Chamelea gallina. Fish and Shellfish Immunology 2011, 30-739-743.

Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, Markowitz S, Duran C, Thierer T, Ashton B, Meintjes P, Drummond A: Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 2012, 28:1647–1649.

Kefaloyianni E, Gourgou E, Ferle V, Kotsakis E, Gaitanaki C, Beis I: Acute thermal stress and various heavy metals induce tissue-specific pro- or anti-apoptotic events via the p38-MAPK signal transduction pathway in Mytilus galloprovincialis (Lam.). Journal of Experimental Biology 2005, 208:4427-4436.

Khondee P, Srisomsap C, Chokchaichamnankit D, Svasti J, Simpson RJ, Kingtong S: Histopathological effect and stress response of mantle proteome following TBT exposure in the Hooded oyster Saccostrea cucullata. Enviornmental Pollution 2016, 218:855-862.

Khripounoff A, Caprais JC, Decker C, Le Bruchec J, Noel P, Husson B: Respiration of bivalves from three different deep-sea areas: Cold seeps, hydrothermal vents and organic carbon-rich sediments. Deep-Sea Research Part II: Topical Studies in Oceanography 2017, 142:233-243.

Kosová K, Vítámvás P, Prášil IT: Proteomics of stress responses in wheat and barley-search for potential protein markers of stress tolerance. Frontiers in Plant Science 2014, 11:711.

Kültz D: Evolution of the cellular stress proteome: From monophyletic origin to ubiquitous function. Journal of Experimental Biology 2003, 206:3119-3124.

110 Error! No text of specified style in document.

Kültz D: Molecular and evolutionary basis of the cellular stress response. Annual Review of Physiology 2005, 67:225–57.

Kyriakis JM, Banerjee P, Nikolakaki E, Dai T, Rubie EA, Ahmad MF, Avruch J, Woodgett JR: The stress-activated protein kinase subfamily of c-Jun kinases. Nature 1994, 369:156-160.

Lathlean JA, Minchinton TE: Manipulating thermal stress on rocky shores to predict patterns of recruitment of marine invertebrates under a changing climate. Marine Ecology Progress Series 2012, 467:121-136.

Lemos MF, Soares AM, Correia AC, Esteves AC: Proteins in ecotoxicology—How, why and why not? Proteomics 2010, 10:873–87.

Lespinet O, Wolf YI, Koonin EV, Aravind L: The Role of Lineage-Specific Gene Family Expansion in the Evolution of Eukaryotes. Genome Resources 2002, 12:1048–1059

Li P, Deng W-Q, Li T-H, Song B, Shen Y-H: Illumina-based de novo transcriptome sequencing and analysis of Amanita exitialis basidiocarps. Gene 2013, 532:63–71.

Li R, Fan W, Tian G, Zhu H, He L, Cai J, Huang Q, Cai Q, Li B, Bai Y, Zhang Z, Zhang Y, Wang W, Li J, Wei F, Li H, Jian M, Li J, Zhang Z, Nielsen R, Li D, Gu W, Yang Z, Xuan Z, Ryder OA, Leung FC-C, Zhou Y, Cao J, Sun X, Fu Y, et al: The sequence and de novo assembly of the giant panda genome. Nature 2010, 463:311–317.

Lima FP, Wethey DS: Three decades of high-resolution coastal sea surface temperatures reveal more than warming. Nature Communications 2012, 3:704.

Liu D, Chen Z: The expression and induction of heat shock proteins in molluscs. Protein and Peptide Letters 2013, 20:602-606.

Liu Z, Wang L, Zhou Z, Liu Y, Dong M, Wang W, Song X, Wang M, Gao Q, Song L: Transcriptomic analysis of oyster Crassostrea gigas larvae illustrates the response patterns regulated by catecholaminergic system upon acute heat and bacterial stress. Developmental and Comparative Immunology 2017, 73:52-60.

Lu P, Vogel C, Wang R, Yao X, Marcotte EM: Absolute protein expression profiling estimates the relative contributions of transcriptional and translational regulation. Nature Biotechnology 2007, 25:117-124.

Luo R, Liu B, Xie Y, Li Z, Huang W, Yuan J, He G, Chen Y, Pan Q, Liu Y, Tang J, Wu G, Zhang H, Shi Y, Liu Y, Yu C, Wang B, Lu Y, Han C, Cheung DW, Yiu S-M, Peng S, Xiaoqian Z, Liu G, Liao X, Li Y, Yang H, Wang J, Lam T-W, Wang J: SOAPdenovo2: An empirically improved memory-efficient short-read de novo assembler. GigaScience 2012, 1:18.

111

Madeira D, Narciso L, Cabral HN, Diniz MS, Vinagre C: Role of thermal niche in the cellular response to thermal stress: Lipid peroxidation and HSP70 expression in coastal crabs. Ecological Indicators 2014, 36:601-606.

Marçais G, Kingsford C: A fast, lock-free approach for efficient parallel counting of occurrences of k-mers. Bioinformatics 2011, 27:764–770.

Markov GV, Baskaran P, Sommer RJ: The Same or Not the Same: Lineage-Specific Gene Expansions and Homology Relationships in Multigene Families in Nematodes. Journal of Molecular Evolution 2015, 80:18–36.

Martin JA, Wang Z: Next-generation transcriptome assembly. Nat Rev Genet 2011, 12:671–682.

Mason JM, Arndt KM: Coiled coil domains: Stability, specificity, and biological implications. ChemBioChem 2004, 5:170-176.

Mayfield AB, Wang Y-B, Chen C-S, Chen S-H, Lin C-Y: Dual-compartmental transcriptomic + proteomic analysis of a marine endosymbiosis exposed to environmental change. Molecular Ecology 2016, 25:5944-5958.

Menge BA, Branch, GM: Rocky intertidal communities. Marine community ecology 2001, 1:221–251.

Miloslavich P, Cruz-Motta JJ, Klein E, Iken K, Weinberger V, Konar B, Trott T, Pohle G, Bigatti G, Benedetti-Cecchi L, Shirayama Y, Mead A, Palomo G, Ortiz M, Gobin J, Sardi A, Díaz JM, Knowlton A, Wong M, Peralta AC: Large-Scale Spatial Distribution Patterns of Gastropod Assemblages in Rocky Shores. PLoS ONE 2013, 8:e71396.

Miura O, Keawtawee T, Sato N, Onodera K-I.: Vertical zonation of endosymbiotic zooxanthellae within a population of the intertidal sea anemone, Anthopleura uchidai. Marine Biology 2014, 161:1745-1754.

Mochida K, Shinozaki K: Genomics and bioinformatics resources for crop improvement. Plant and Cell Physiology 2010, 51:497-523.

Mohamed B, Hajer A, Susanna S, Caterina O, Flavio M, Hamadi B, Aldo V: Transcriptomic responses to heat stress and nickel in the mussel Mytilus galloprovincialis. Aquatic toxicology 2014, 13:104-112.

Morimoto RI, Kroeger PE, Cotto JJ: The transcriptional regulation of heat shock genes: a plethora of heat shock factors and regulatory conditions. EXS 1996, 77:139-163.

Morozova O, Marra MA: Applications of next-generation sequencing technologies in functional genomics. Genomics 2008, 92:255-264.

112 Error! No text of specified style in document.

Nachin L, Nannmark U, Nyström T: Differential roles of the universal stress proteins of Escherichia coli in oxidative stress resistance, adhesion, and motility. Journal of Bacteriology 2005, 187:6265-6272.

Narum SR, Campbell NR, Meyer KA, Miller MR, Hardy RW: Thermal adaptation and acclimation of ectotherms from differing aquatic climates. Molecular Ecology 2013, 22:3090-3097.

Nock CJ, Baten A, Barkla BJ, Furtado A, Henry RJ, King GJ: Genome and transcriptome sequencing characterises the gene space of Macadamia integrifolia (Proteaceae). BMC Genomics 2016, 13:937.

O'Connor MI, Bruno JF, Gaines SD, Halpern BS, Lester SE, Kinlan BP, Weiss JM: Temperature control of larval dispersal and the implications for marine ecology, evolution, and conservation. Proceedings of the National Academy of Sciences of the United States of America 2007, 104:1266-1271.

Pante E, Rohfritsch A, Becquet V, Belkhir K, Bierne N, Garcia P: SNP detection from de novo transcriptome sequencing in the bivalve Macoma balthica: marker development for evolutionary studies. PLoS One 2012, 7:e52302.

Pantzartzi CN, Drosopoulou E, Scouras ZG: Assessment and Reconstruction of Novel HSP90 Genes: Duplications, Gains and Losses in Fungal and Animal Lineages. PLoS ONE 2013, 8:e73217.

Parra G, Bradnam K, Korf I: CEGMA: a pipeline to accurately annotate core genes in eukaryotic genomes. Bioinformatics 2007, 23:1061–1067.

Parsell DA, Lindquist S: The function of heat-shock proteins in stress tolerance: Degradation and reactivation of damaged proteins. Annual Review of Genetics 1993, 27:437-496.

Passamaneck YJ, Schander C, Halanych KM: Investigation of molluscan phylogeny using large-subunit and small-subunit nuclear rRNA sequences. Molecular Phylogenetics and Evolution 2004, 32:25-38.

Pespeni MH, Sanford E, Gaylord B, Hill TM, Hosfelt JD, Jaris HK, LaVigne M, Lenz, EA, Russell AD, Young MK, Palumbi SR: Evolutionary change during experimental ocean acidification. Proceedings of the National Academy of Sciences of the United States of America 2013, 110:6937-6942.

Petersen TN, Brunak S, Von Heijne G, Nielsen H: SignalP 4.0: Discriminating signal peptides from transmembrane regions. Nature Methods 2011, 8:785-786.

Peterson CH: Recruitment overfishing in a bivalve mollusc fishery: hard clams (Mercenaria mercenaria) in North Carolina. Can J Fish Aquat Sci 2002, 59:96–104.

113

Place SP, Menge BA, Hofmann GE: Transcriptome profiles link environmental variation and physiological response of Mytilus californianus between Pacific tides. Functional Ecology 2012, 26:144-155.

Poirier I, Kuhn L, Caplat C, Hammann P, Bertrand M: The effect of cold stress on the proteome of the marine bacterium Pseudomonas fluorescens BA3SM1 and its ability to cope with metal excess. Aquatic Toxicology 2014, 157:120-133.

Ponder WF, Lindberg DR: Towards a phylogeny of gastropod molluscs: an analysis using morphological characters. Zool J Linnean Soc 1997, 119:83–265.

Pörtner HO, Peck MA: Climate change effects on fishes and fisheries: towards a cause-and-effect understanding. Journal of Fish Biology 2010, 77:1745–1779.

Postaire B, Bruggemann JH, Magalon H, Faure B: Evolutionary dynamics in the southwest indian ocean marine biodiversity hotspot: A perspective from the rocky shore gastropod genus Nerita. PLoS ONE 2014, 9:Article number e95040.

Prentis PJ, Pavasovic A: The Anadara trapezia transcriptome: A resource for molluscan physiological genomics. Marine Genomics 2014, 18:113-115.

Przeslawski R, Byrne M, Mellin C: A review and meta-analysis of the effects of multiple abiotic stressors on marine embryos and larvae. Global Change Biology 2015, 21:2122-2140.

Ramos AA, Weydmann A, Cox CJ, Canário AVM, Serrão EA, Pearson GA: A transcriptome resource for the copepod Calanus glacialis across a range of culture temperatures. Marine Genomics 2015, 23:27-29.

Rappsilber J, Mann M, Ishihama, Y: Protocol for micro-purification, enrichment, pre-fractionation and storage of peptides for proteomics using StageTips. Nature Protocols 2007, 2:1896-1906.

Reece JB, Neil AC: Campbell Biology. Boston: Benjamin Cummings / Pearson, 2011.

Reusch TBH: Climate change in the oceans: evolutionary versus phenotypically plastic responses of marine animals and plants. Evolutionary Applications 2014, 7:104-122.

Ribeiro DA, Ferraz LFC, Vicentini R, Ottoboni LMM: Gene expression modulation by heat stress in Acidithiobacillus ferrooxidans LR. Antonie van Leeuwenhoek, International Journal of General and Molecular Microbiology 2012, 101:583-593.

Roessig JM, Woodley CM, Cech Jr JJ, Hansen LJ: Effects of global climate change on marine and estuarine fishes and fisheries. Reviews in Fish Biology and Fisheries 2004, 14:251-275.

114 Error! No text of specified style in document.

Rossbach M: Small non-coding rnas as novel therapeutics. Current Molecular Medicine 2010, 10:361-368.

Rupik W, Jasik K, Bembenek J, Widłak W: The expression patterns of heat shock genes and proteins and their role during vertebrate's development. Comparative Biochemistry and Physiology - A Molecular and Integrative Physiology 2011, 159:349-366.

Sadamoto H, Takahashi H, Okada T, Kenmoku H, Toyota M, Asakawa Y: De novo sequencing and transcriptome analysis of the central nervous system of mollusc Lymnaea stagnalis by deep RNA sequencing. PLoS One 2012, 7:42546.

Scavia D, Field JC, Boesch DF, Buddemeier RW, Burkett V, Cayan DR, Fogarty M, Harwell MA, Howarth RW, Mason C, Reed DJ, Royer TC, Sallenger AH, Titus JG: Climate change impacts on U.S. coastal and marine ecosystems. Estuaries 2002, 25:149-164.

Schadt EE, Linderman MD, Sorenson J, Lee L, Nolan GP: Computational solutions to large-scale data management and analysis. Nature Reviews Genetics 2010, 11:647-657.

Schaeffer HJ, Weber MJ: Mitogen-activated protein kinases: Specific messages from ubiquitous messengers. Molecular and Cellular Biology 1999, 19:2435-2444.

Schlesinger MJ: Heat shock proteins. Journal of Biological Chemistry 1990, 265:12111-12114.

Schulz MH, Zerbino DR, Vingron M, Birney E: Oases: robust de novo RNA-seq assembly across the dynamic range of expression levels. Bioinformatics 2012, 28:1086–1092.

Schunter C, Vollmer SV, Macpherson E, Pascual M: Transcriptome analyses and differential gene expression in a non-model fish species with alternative mating tactics. Acta Vet Scand 2014, 15:167.

Sexton GJ, Frere CH, Dieters MJ, Godwin ID, Prentis PJ: Development and characterization of microsatellite loci for Khaya senegalensis (Meliaceae). Am J Bot 2010, 97:e111–e113.

Shivhare R, Lata C: Exploration of genetic and genomic resources for abiotic and biotic stress tolerance in pearl millet. Frontiers in Plant Science 2017, 7:2069.

Shotwell J: Distribution of Volume and Relative Linear Measurement Changes in Acmaea, the Limpet. Ecology 1950, 31:51-61.

Sigwart JD, Chen C, Marsh L: Is mining the seabed bad for mollusks? Nautilus 2017, 131-43-49.

115

Simakov O, Marletaz F, Cho S-J, Edsinger-Gonzales E, Havlak P, Hellsten U, Kuo D-H, Larsson T, Lv J, Arendt D, Savage R, Osoegawa K, De Jong P, Grimwood J, Chapman JA, Shapiro H, Aerts A, Otillar RP, Terry AY, Boore JL, Grigoriev IV, Lindberg DR, Seaver EC, Weisblat DA, Putnam NH, Rokhsar DS: Insights into bilaterian evolution from three spiralian genomes. Nature 2013, 493:526-531.

Snyder MJ, Girvetz E, Mulder EP: Induction of marine mollusc stress proteins by chemical or physical stress. Archives of Environmental Contamination and Toxicology 2001, 41:22-29.

Somero GN: Thermal physiology and vertical zonation of intertidal animals: Optima, limits, and costs of living. Integrative and Comparative Biology 2002, 42:780-789.

Southward AJ, Hawkins SJ, Burrows MT: Seventy years′ observations of changes in distribution and abundance of Zooplankton and intertidal organisms in the western English Channel in relation to rising sea temperature. Journal of Thermal Biology 1995, 20:127-155.

Spencer HG, Waters JM, Eichhorst TE: Taxonomy and nomenclature of black nerites (Gastropoda, Neritimorpha: Nerita) from the South Pacific. Invertebrate Systematics 2007, 21:229-237.

Stillman JH, Somero GN: Adaptation to temperature stress and aerial exposure in congeneric species of intertidal porcelain crabs (genus Petrolisthes): Correlation of physiology, biochemistry and morphology with vertical distribution. Journal of Experimental Biology 1996, 199:1845-1855.

Stillman JH: Causes and consequences of thermal tolerance limits in rocky intertidal porcelain crabs, genus Petrolisthes. Integrative and Comparative Biology 2002, 42:790-796.

Sumizawa T, Igisu H: Release of heat shock proteins from human neuroblastoma cells exposed to acrylamide. Journal of Toxicological Sciences 2008, 33:117-122.

Sun J, Wang M, Wang H, Zhang H, Zhang X, Thiyagarajan V, Qian PY, Qiu JW: De novo assembly of the transcriptome of an invasive snail and its multiple ecological applications. Molecular Ecology Resources 2012, 12:1133-1144.

Takeuchi T, Koyanagi R, Gyoja F, Kanda M, Hisata K, Fujie M, Goto H, Yamasaki S, Nagai K, Morino Y, Miyamoto H, Endo K, Endo H, Nagasawa H, Kinoshita S, Asakawa S, Watabe S, Satoh N, Kawashima T: Bivalve-specific gene expansion in the pearl oyster genome: implications of adaptation to a sessile lifestyle. Zoological Letters 2016, 2:3.

116 Error! No text of specified style in document.

Tassone EE, Geib SM, Hall B, Fabrick JA, Brent CS, Hull JJ: De novo construction of an expanded transcriptome assembly for the western tarnished plant bug, Lygus hesperus. GigaScience 2016, 5:6.

Thunders M, Cavanagh J, Li Y: De novo transcriptome assembly, functional annotation and differential gene expression analysis of juvenile and adult E. fetida, a model oligochaete used in ecotoxicological studies. Biological Research 2017, 50:7.

Timmins-Schiffman E, Coffey WD, Hua W, Nunn BL, Dickinson GH, Roberts SB: Shotgun proteomics reveals physiological response to ocean acidification in Crassostrea gigas. BMC Genomics 2014, 15:951.

Tkaczuk KL, Shumilin AI, Chruszcz M, Evdokimova E, Savchenko A, Minor W: Structural and functional insight into the universal stress protein family. Evolutionary Applications 2013, 6:434-449.

Tomanek L, Somero GN: Evolutionary and acclimation-induced variation in the heat-shock responses of congeneric marine snails (genus Tegula) from different thermal habitats: Implications for limits of thermotolerance and biogeography. Journal of Experimental Biology 1999, 202:2925-2936.

Tomanek L, Zuzow MJ: The proteomic response of the mussel congeners Mytilus galloprovincialis and M. trossulus to acute heat stress: Implications for thermal tolerance limits and metabolic costs of thermal stress. Journal of Experimental Biology 2010, 213:3559-3574.

Tomanek, L: The heat-shock response: Its variation, regulation and ecological importance in intertidal gastropods (genus Tegula). Integrative and Comparative Biology 2002, 42:797-807.

Underwood AJ, Jernakoff P: Effects of interactions between algae and grazing gastropods on the structure of a low-shore intertidal algal community. Oecologia 1981, 48:221-233.

Van Dyck H, Bonte D, Puls R, Gotthard K, Maes D: The lost generation hypothesis: Could climate change drive ectotherms into a developmental trap?. Oikos 2015, 124:54-61.

Walsh CT, Jefferis R: Post-translational modifications in the context of therapeutic proteins. Nature Biotechnology 2006, 24:1241-1252.

Wang J, Zhang G, Fang X, Guo X, Li L, Luo R, Xu F, Yang P, Zhang L, Wang X, Qi H, Xiong Z, Que H, Xie Y, Holland PWH, Paps J, Zhu Y, Wu F, Chen Y, Wang J, Peng C, Meng J, Yang L, Liu J, Wen B, Zhang N, Huang Z, Zhu Q, Feng Y, Mount A, Hedgecock D, Xu Z, Liu Y, Domazet-Lošo T, Du Y, Sun X, Zhang S, Liu B, Cheng P, Jiang X, Li J, Fan D, Wang W, Fu W, Wang T, Wang B, Zhang J, Peng Z, Li Y, Li N, Wang J, Chen M, He Y, Tan F, Song X, Zheng Q, Huang R, Yang H, Du X, Chen L, Yang M, Gaffney PM, Wang S, Luo L, She Z, Ming Y, Huang W, Zhang S, Huang B, Zhang Y, Qu T, Ni P, Miao G, Wang J, Wang Q, Steinberg CEW, Wang H, Li N,

117

Qian L, Zhang G, Li Y, Yang H, Liu X, Wang J, Yin Y, Wang J: The oyster genome reveals stress adaptation and complexity of shell formation. Nature 2012, 490:49-54.

Wang Y-J, Zheng H-P, Zhang B, Liu H-L, Deng H-J, Deng L-H: Cloning and respond of a cold shock domain protein (CnCSDP) gene to cold stress in noble scallop Chlamys nobilis (Bivalve: Pectinidae). Molecular Biology Reports 2014, 41:7985-7994.

Wang Y, Coleman-Derr D, Chen G, Gu YQ: OrthoVenn: A web server for genome wide comparison and annotation of orthologous clusters across multiple species. Nucleic Acids Research 2015, 43:W78-W84.

Waters JM, King TM, O'Loughlin PM, Spencer HG: Phylogeographical disjunction in abundant high-dispersal littoral gastropods. Molecular Ecology 2005, 14:2789-2802.

Webster N, Pantile R, Botté E, Abdo D, Andreakis N, Whalan S: A complex life cycle in a warming planet: Gene expression in thermally stressed sponges. Molecular Ecology 2013, 22:1854-1868.

Wheat CW, Vogel H: Transcriptome sequencing goals, assembly, and assessment. Methods Mol Biol 2011, 772:129–144.

Wijgerde T, Silva CIF, Scherders V, Van Bleijswijk J, Osinga R: Coral calcification under daily oxygen saturation and pH dynamics reveals the important role of oxygen. Biology Open 2014, 3:489-493.

Wiśniewski JR, Zougman A, Nagaraj N, Mann M: Universal sample preparation method for proteome analysis. Nature Methods 2009, 6:359-362.

Wu G, Nie L, Zhang W: Integrative analyses of posttranscriptional regulation in the yeast Saccharomyces cerevisiae using transcriptomic and proteomic data. Current Microbiology 2008, 57:18-22.

Wu H-C, Jinn T-L: Oscillation regulation of Ca2+/calmodulin and heat-stress related genes in response to heat stress in rice (Oryza sativa L.). Plant Signaling and Behavior 2007, 12:1056-1057.

Xie J, Bai X, Lavoie M, Lu H, Fan X, Pan X, Fu Z, Qian H: Analysis of the Proteome of the Marine Diatom Phaeodactylum tricornutum Exposed to Aluminum Providing Insights into Aluminum Toxicity Mechanisms. Environmental Science and Technology 2015, 49:11182-11190.

Xu Q, Qin Y: Molecular cloning of heat shock protein 60 (PtHSP60) from Portunus trituberculatus and its expression response to salinity stress. Cell Stress and Chaperones 2012, 17:589-601.

118 Error! No text of specified style in document.

Yates JR, Ruse CI, Nakorchevsky A: Proteomics by mass spectrometry: Approaches, advances, and applications. Annual Review of Biomedical Engineering 2009, 11:49-79.

Ye J, Fang L, Zheng H, Zhang Y, Chen J, Zhang Z, Wang J, Li S, Li R, Bolund L, Wang J: WEGO: a web tool for plotting GO annotations. Nucleic Acids Research 2006, 34(Web Server):W293–W297.

Zeh JA, Bonilla MM, Su EJ, Padua MV, Anderson RV, Kaur D, Yang D-S, Zeh DW: Degrees of disruption: Projected temperature increase has catastrophic consequences for reproduction in a tropical ectotherm. Global Change Biology 2012, 18:1833-1842.

Zerbino DR, Birney E: Velvet: Algorithms for de novo short read assembly using de Bruijn graphs. Genome Research 2008, 18: 821-829.

Zhang Q, Pell J, Canino-Koning R, Howe AC, Brown CT: These are not the K-mers you are looking for: Efficient online K-mer counting using a probabilistic data structure. PLoS ONE 2014, 7:e101271.

Zhang Z, Li J, Zhao X-Q, Wang J, Wong GK-S, Yu J: KaKs_Calculator: Calculating Ka and Ks Through Model Selection and Model Averaging. Genomics, Proteomics and Bioinformatics 2006, 4:259-263.

Zhao X, Yu H, Kong L, Li Q: Transcriptomic responses to salinity stress in the pacific oyster Crassostrea gigas. PLoS One 2012, 7:e46244.

Appendix A – Poster Presentations

Poster Presentations

119

International Marine Biotechnology Conference – Convention Centre, Brisbane

(2013)

121

Big Biology and Bioinformatics Symposium – QUT, Brisbane (2014)

122 Error! No text of specified style in document.

Appendix B – Outcomes from associated works

OUTCOMES FROM ASSOCIATED WORKS

Pavasovic, A., Hair, C., Amin, S., Hurwood, D, and P. Prentis. Characterisation of

candidate nuclear genes for species delineation in the genus Cherax.

Conservation Genetic Resources 2:331-333.

Ali MY, Pavasovic A, Amin S, Mather PB, and Prentis PJ. 2015. Comparative analysis

of gill transcriptomes of two freshwater crayfish, Cherax cainii and C.

destructor. Marine Genomics 22:11-13.

Rahi ML, Amin S, Mather PB, Hurwood DA: Candidate genes that have facilitated

freshwater adaptation by palaemonid prawns in the genus Macrobrachium:

Identification and expression validation in a model species (M.

koombooloomba). PeerJ 2017, 2:e2977.