1
2
Genomic resource development for shellfish of conservation concern
3
4
5
6
Authors:
7
Emma B. Timmins-Schiffman, Carolyn S. Friedman, Dave C. Metzger, Samuel J. White, Steven
8
B. Roberts*
9
10
University of Washington, School of Aquatic and Fishery Sciences
11
Box 355020
12
Seattle, WA 98195
13
14
Keywords: high-throughput sequencing, pinto abalone, Olympia oyster, transcriptome, gene
15
expression
16
17
*Corresponding author
18
University of Washington
19
Box 355020
20
Seattle, WA 98195
21
Fax: (206) 685-7471
22
Email: sr320@uw.edu
23
24
Running Title: Genomics of shellfish of conservation concern
25
Abstract
26
Effective conservation of threatened species depends on the ability to assess organism
27
physiology and population demography. In order to develop genomic resources to better
28
understand the dynamics of two ecologically vulnerable species in the Pacific Northwest of the
29
United States, larval transcriptomes were sequenced for the pinto abalone Haliotis
30
kamtschatkana kamtschatkana and the Olympia oyster Ostrea lurida. Based on comparative
31
species analysis the Ostrea lurida transcriptome (41,136 contigs) is relatively complete. These
32
transcriptomes represent the first significant contribution to genomic resources for both species.
33
Genes are described based on biological function with a particular attention to those associated
34
with temperature change, oxidative stress, and immune function. In addition, transcriptome
35
derived genetic markers are provided. Together, these resources provide valuable tools for
36
future studies aimed at conservation of Haliotis kamtschatkana, Ostrea lurida and related
37
species.
38
39
40
Introduction
Effective conservation efforts are dependent on understanding how organisms respond
41
to environmental conditions (Wikelski & Cooke 2006). One means to assess physiological
42
responses is to evaluate changes at the molecular level. Often, valuable insight can be gained
43
from assessing gene expression variation that would not be readily evident from ecological
44
observations. For instance, in Crassostrea virginica larvae, specific gene expression patterns as
45
determined through qPCR were found to signal an early mortality event (Genard et al. 2012).
46
Quantitative PCR has also been used to characterize the effects of ocean acidification on the
47
purple sea urchin Strongylocentrotus purpuratus (Stumpp et al. 2011), Mediterranean sea
48
urchin Paracentrotus lividus (Martin et al. 2011), red abalone Haliotis rufescens (Zippay and
49
Hofmann 2010), and red sea urchin Strongylocentrotus franciscanus (O’Donnell et al. 2009).
50
These studies revealed that ocean acidification alters metabolism and calcification (Stumpp et
51
al. 2011; Martin et al. 2011) and limits an effective heat shock response (O’Donnell et al. 2009).
52
Transcriptomic data can be used to increase our understanding not only of an
53
organismal response, but can also aid in characterizing population structure and dynamics.
54
Understanding population genetic structure can further the knowledge needed to increase the
55
success of conservation efforts (e.g. restoration, relocation). This includes providing necessary
56
information to maintain effective genetic diversity (Reed & Frankham 2003) and, in some cases,
57
ensuring native cohorts are not diminished (Frankham 2002). High-throughput sequencing of
58
transcriptomes can reveal a breadth of markers (e.g. single nucleotide polymorphisms or SNPs)
59
that are informative not only for population demographics, but also for understanding ecological
60
and evolutionary mechanisms. For example high-throughput sequencing of transcriptomes has
61
been used to identify SNPs in the non-model Glanville fritillary butterfly (Melitaea cinxia; Vera et
62
al. 2008), species in the fish genus Xiphophorus (Shen et al. 2012), and oilseed rape (Brassica
63
napus; Trick et al. 2009). The SNPs discovered in B. napus were successfully tested for use in
64
linkage mapping (Trick et al. 2009). While not necessarily required a priori, sufficient genomic
65
resources can greatly increase a researcher’s ability to plan, implement, and monitor success of
66
conservation efforts.
67
The overall objective of the current effort was to develop genomic resources for two
68
shellfish species of conservation concern in the coastal region of northwest North America: the
69
pinto abalone, Haliotis kamtschatkana kamtschatkana (hereafter referred to as H.
70
kamtschatkana), and Olympia oysters, Ostrea lurida. The pinto (or northern) abalone is an IUCN
71
listed species of concern (NOAA 2007) distributed from Alaska through Point Conception,
72
California. Harvest of pinto abalone has been outlawed along its range since the 1990s, except
73
in Alaska where recreational and subsistence harvest are still permitted. Within the past two
74
decades, significant declines have been reported in pinto abalone populations in the San Juan
75
Archipelago (Washington, USA), with evidence of decreased recruitment and potential for
76
inhibited capacity to reproduce based on low population density (Rothaus et al. 2008, Bouma et
77
al. 2012). Greater than expected mortality of mature H. kamtschatkana has been observed in
78
the southeast Queen Charlotte Islands, BC, suggesting that poaching is still a valid concern for
79
this threatened species (Hankewich et al. 2008). In the southern end of their range, H.
80
kamtschatkana are threatened by warming ocean temperatures, poaching, and predation by
81
sea otters (Rogers-Bennett 2007). Marine reserves have been shown to promote both
82
population growth and reproductive output in pinto abalone (Wallace 1999).
83
Olympia oysters are the only native oyster species on the United States west coast and
84
are another species of conservation concern with a coastwide distribution. In the state of
85
Washington O. lurida are listed as a candidate Species of Concern. Overexploitation of O.
86
lurida in the late 1800s through the early 1900s led to significant reductions of natural
87
populations (White et al. 2009). Water pollution and decreased availability of suitable habitat
88
compounded the effects of overharvest and led to further declines in O. lurida populations along
89
the coast (White et al. 2009; Groth & Rumrill 2009; Gillespie 2009). Olympia oyster populations
90
are still found along the majority of their historical distribution, indicating a possibility for
91
rebuilding the species, however, densities are much lower than they once were (Polson &
92
Zacherl 2009). O. lurida restoration efforts have been increasing in recent years (McGraw
93
2009) and include efforts such as habitat remediation and hatchery supplementation of natural
94
populations (e.g. Dinnel et al. 2009).
95
96
This study describes the larval transcriptomes of the pinto (northern) abalone, Haliotis
kamtschatkana, and the Olympia oyster, Ostrea lurida, with the objective of developing genomic
97
resources for further conservation and management purposes. Along with descriptions of the
98
full transcriptomes and identification of genes associated with environmental stress response,
99
we also identify putative genetic markers for both species.
100
101
Materials & Methods
102
103
104
Sampling and Library Construction
Olympia oyster larvae were spawned at the Taylor Shellfish Hatchery in Quilcene, WA.
105
Larvae were transferred to the University of Washington six hours post spawning. Larvae (12
106
larvae/ml) were evenly distributed to six 4.5-L larval chambers. Larvae were sampled from all
107
chambers by filtering them onto a 35 µm screen and flash freezing in liquid N2 on days one, two,
108
and three post-fertilization. Two RNA-seq libraries were constructed from mRNA and
109
sequenced at the University of Washington High Throughput Genomics Unit (HTGU) on the
110
Illumina Hi-Seq 2000 platform (Illumina, San Diego, CA, USA). Both libraries were made of
111
pooled mRNA from 3 chambers across all sampling time points. Each library was run on a
112
single lane.
113
Pinto abalone larvae were spawned at the Mukilteo Research Station in Mukilteo, WA
114
and transported to the University of Washington at 3 days post-fertilization. Larvae (two
115
larvae/ml) were held in eight 4.0-L chambers and on day 4 post-fertilization, all larvae were
116
sampled for sequencing by filtering them onto a 70 µm screen and flash freezing them in liquid
117
N2. Similar to the Olympia oyster samples, two RNA-Seq libraries were constructed from pooled
118
mRNA samples (from four chambers each). Pinto abalone libraries were run together on one
119
lane of the Illumina Hi-Seq 2000 (Illumina) at the University of Washington HTGU.
120
121
Sequence Analysis
122
Sequence analysis was performed with CLC Genomics Workbench version 5.0 (CLC
123
bio, Katrinebjerg, Denmark). Quality trimming was performed using the following parameters:
124
quality score 0.05 (Phred; Ewing & Green, 1998; Ewing et al., 1998), number of ambiguous
125
nucleotides <2 on ends, and reads shorter than 25 bp were removed. De novo assembly was
126
performed on the quality trimmed reads using the following parameters: mismatch cost = 2,
127
deletion cost = 3, similarity fraction = 0.9, insertion cost = 3, length fraction = 0.8 and minimum
128
contig size of 200 bp. Sequences were annotated by comparing contiguous sequences to the
129
UniProtKB/Swiss-Prot database (http://uniprot.org) using the BLASTx algorithm (Altschul et al.
130
1997) with a 1.0E-5 e-value threshold. Genes were classified according to their biological
131
processes as determined by their Gene Ontology (GO) and classified into their parent
132
categories (GO Slim).
133
To assess the completeness of the transcriptomes, contigs from both species were
134
assessed to determine if they contained orthologs to proteins found in all Metazoa. Specifically,
135
OrthoDB (Waterhouse et al. 2011, http://cegg.unige.ch/orthodb6) was used obtain a suite of
136
proteins from Lottia gigantea (the giant owl limpet) found as single copy, which have orthologs
137
in all other metazoans in OrthoDB. Sequence comparisons (tBLASTn; Altschul et al. 1997)
138
were performed to find matching contigs in H. kamtschatkana and O. lurida transcriptomes. An
139
e-value threshold of 1E-10 was used. In order to determine what proportion of full-length coding
140
regions were obtained, sequences were translated and aligned to corresponding L. gigantea
141
proteins (Geneious Pro version 5.3.6; Kearse et al. 2012). Alignment parameters are as follows:
142
BLOSUM 62 cost matrix, gap open penalty of 12, gap extension penalty of 3, global alignment
143
with free end gaps, and 2 refinement iterations. Percent coverage of the L. gigantea protein by
144
the O. lurida or H. kamtschatkana contig was then calculated.
145
Using a more global approach to assess the larval transcriptomes of O. lurida and H.
146
kamtschatkana, all contigs were compared using the BLASTn algorithm (Altschul et al. 1997) to
147
publicly available transcriptomes of other shellfish, as well as to zebrafish. Specific datasets
148
used include transcriptomes of Pacific oyster, Crassostrea gigas (GigasDatabase version 8;
149
Fleury et al., 2009); pearl oyster, Pinctada fucata (Predicted Transcripts from Genome
150
Assembly ver 1.0; http://marinegenomics.oist.jp/genomes/download?project_id=0; Takeuchi et
151
al. 2012); South African abalone, Haliotis midae transcriptome (Franchini et al. 2011); Manila
152
clam, Ruditapes philippinarum (RuphiBase (http://compgen.bio.unipd.it/ruphibase/); and
153
zebrafish, Danio rerio (NCBI RefSeq mRNA, ftp://ftp.ncbi.nih.gov/refseq/D_rerio/).
154
In order to investigate the functions of contigs that did not have significant matches in
155
the transcriptomes of the species listed above, gene enrichment analysis was performed. Gene
156
ontology information associated with enriched unique gene set was identified using the
157
Database for Annotation, Visualization, and Integrated Discovery (DAVID) v. 6.7 (Huang et al.
158
2009a and 2009b, http://david.abcc.ncifcrf.gov/). The background gene list was made from the
159
UniProtKB/Swiss-Prot database accession numbers that corresponded to the entire
160
transcriptome of either H. kamtschatkana or O. lurida.
161
Phylogenetic analysis of select stress response genes was conducted with sequences
162
from the following species: O. lurida, H. kamtschatkana, H. midae, P. fucata, C. gigas, R.
163
philippinarum, and D. rerio. Target genes and corresponding sequence accession numbers for
164
each species are provided in Table 1. Sequences were aligned in Geneious Pro version 5.3.6
165
(Kearse et al. 2012) and trimmed to the length of the shortest sequence. Parameters for
166
alignment included cost matrix 65% similarity (5.0/-4.0), gap open penalty of 12, and gap
167
extension penalty of 3. Phylogenetic trees were made using Geneious tree builder with the
168
genetic distance model HKY, neighbor-joining tree build method, Danio rerio as the outgroup,
169
100x bootstrap, and 50% support threshold.
170
171
172
Single Nucleotide Polymorphism (SNP) Identification
Single Nucleotide Polymorphism (SNP) detection was carried out on O. lurida and H.
173
kamtschatkana assemblies using the following parameters: window length = 11, maximum gap
174
and mismatch count = 2, minimum central base quality = 20, minimum coverage = 10,
175
maximum coverage =1000, and minimum variant frequency = 35% (Genomics Workbench v. 5;
176
CLC bio).
177
178
Results
179
Larval Transcriptomes
180
Two larval transcriptome libraries were constructed and sequenced for Olympia oyster
181
(O. lurida) and pinto abalone (H. kamtschatkana). Given that quality and yield from
182
complementary libraries were consistent, all sequencing reads were combined within each
183
species. Following quality trimming, 387,720,843 sequencing reads remained from the
184
combined Olympia oyster (O. lurida) larval libraries. For pinto abalone (H. kamtschatkana),
185
94,777,799 quality reads remained. Raw sequence data is available in the NCBI Short Read
186
Archive (Accession Number SRA057107).
187
De novo assembly of sequencing reads from each species generated 41,136 and 8,641
188
contigs for O. lurida and H. kamtschatkana, respectively (Supp_1_Ostrea_lurida transcriptome
189
and Supp_2_Haliotis_kamtschatkana_transcriptome). The average contig lengths were 611 bp
190
for O. lurida and 401 bp for H. kamtschatkana (Table 2).
191
192
193
Transcriptome Annotation
A total of 15,381 (37%) O. lurida contigs were annotated based on the UniProt-
194
SwissProt database (Supp_3_Ostrea_lurida_SPIDs) with associated gene ontology information
195
available for 8,030 contigs (Supp_4_Ostrea_lurida_GO and Figure 1). A total of 1,277 (15%)
196
H. kamtschatkana contigs were annotated based on the UniProt-SwissProt database
197
(Supp_5_Haliotis_kamtschatkana_SPIDs) with associated gene ontology information available
198
for 574 contigs (Supp_6_Haliotis_kamtschatkana_GO and Figure 2).
199
Of the 30 orthologus proteins found across Metazoa, and found as a single copy in L.
200
gigantea, 28 O. lurida contigs were identified with sequence similarity matches. A majority of the
201
O. lurida contigs (24) had e-values <1E-50. The average coverage of O. lurida contigs based on
202
the corresponding L. gigantea protein was 53.7%. In contrast, only 2 proteins found across
203
Metazoa were identified in the H. kamtschatkana transcriptome.
204
205
206
Comparative Transcriptome Analysis
Of the 41,136 contigs from the O. lurida larval library, transcriptomes from the C. gigas
207
and P. fucata oysters had the highest similarity to O. lurida, 21,748 and 11,101 matching
208
contigs, respectively (Figure 3). The D. rerio transcriptome had the next highest number of
209
matches with 6,015, followed in order by H. midae, R. philippinarum, and H. kamtschatkana
210
(Figure 3). A summary of matches to O. lurida contigs in each dataset is provided with
211
respective bit scores provided for each pairwise blast comparison
212
(Supp_7_Ostrea_lurida_bitscores).
213
Of the 8,641 contigs from the H. kamtschatkana library, the H. midae database had the
214
most matches with 3,090 BLAST hits (Figure 4). Comparison with the other transcriptomic
215
datasets all produced fewer than 1,000 matches (Figure 4). A summary of matches to H.
216
kamtschatkana contigs in each dataset is provided with respective bit scores for each pairwise
217
blast comparison (Supp_8_Haliotis_kamtschatkana_bitscores).
218
There were 1,315 genes without matches to other transcriptomes in the O. lurida
219
transcriptome (3.2% of the all the contigs) and 243 with no matches to other transcriptome in
220
the H. kamtschatkana transcriptome (2.8% of all contigs). The O. lurida unique contigs were
221
enriched in the GO Slim categories of developmental processes, cell organization and
222
biogenesis, cell adhesion, transport, stress response, signal transduction, and protein
223
metabolism (Table 3). The H. kamtschatkana unique contigs were enriched in the GO Slim
224
terms of stress response and RNA metabolism (Table 3).
225
Analysis using select stress response genes resulted in phylogenetic tree clustering that
226
was consistent with taxonomic relationships. Specifically, H. kamtschatkana and H. midae
227
clustered together with high confidence (100%) except for the gene hsp82 where H. midae and
228
R. philippinarum clustered together with a bootstrap confidence of 50% but within the same
229
phyletic group as H. kamtschatkana (bootstrap confidence of 93%) (Figure 5). O. lurida
230
consistently clustered with C. gigas with high bootstrap support (v-type proton ATPase 85%,
231
transmembrane protein 85 99%, hsp82 66%, heat shock 70 cognate 67%, GABA 99%) (data
232
not shown). However, O. lurida hsp70 clustered more closely with the corresponding gene in P.
233
fucata (bootstrap confidence 68%).
234
235
236
SNP Characterization
In the O. lurida transcriptome, 59,221 putative SNPs were identified within 19,354
237
contigs (Supp_9_Ostrea_lurida_SNPs). A subset of these putative SNPs (27,818) were within
238
annotated genes, corresponding to 6,328 protein descriptions. In the H. kamtschatkana
239
transcriptome, 6,596 putative SNPs were identified within 3,378 contigs
240
(Supp_10_Haliotis_kamtschatkana_SNPs). A subset of these putative SNPs (1,038) were
241
within annotated genes, corresponding to 442 protein descriptions. The functional distribution of
242
SNPs among contigs (as determined by GO and GO Slim terms) was not significantly different
243
from the annotation of the entire transcriptome for both O. lurida and H. kamtschatkana.
244
245
246
Discussion
This research effort has dramatically increased the amount of genomic data available in
247
O. lurida and H. kamtschatkana. At the time the project was undertaken there were on a few
248
hundred publicly available sequences for the species combined. Here we provide 8,641 H.
249
kamtschatkana and 41,136 O. lurida contig sequences representing the larval transcriptomes.
250
The O. lurida contigs represent a more complete transcriptome as compared to the H.
251
kamtschatkana dataset. This is not unexpected as approximately 3 times more reads were
252
sequenced for O. lurida compared to H. kamtschatkana. Evidence to support the relative
253
completeness of the O. lurida transcriptome includes the fact that the number of contigs from de
254
novo assembly is similar to what has been reported in other species. Specific examples include
255
Ciona intestinalis species B (27,426 contigs; Cahais et al. 2012), Lepus granatensis (38,540
256
contigs; Cahais et al. 2012), and Radix balthica (>20,000; Feldmeyer et al. 2011). Furthermore,
257
using OrthoDB, we found evidence that over 90% of the proteins found across Metazoa were
258
present in the O. lurida transcriptome. Together these data suggest that 14 gigabases of ultra
259
short read (~36bp) high throughput transcriptome sequencing can produce a robust
260
transcriptome. It would be expected that more sequences and longer read length would
261
increase the number of full length sequences and facilitate identification of transcripts expressed
262
at a minimal level. While not complete in nature, both transcriptomes characterized here
263
provide a wealth of information, evidenced in part by the number of sequences that did not have
264
a significant match in cross-species comparisons.
265
In addition to characterizing the transcripts based on sequence homology, a suite of
266
transcriptome derived SNPs are reported. Recently there have been several examples where
267
transcriptome derived SNPs have provided unique insight into population dynamics. For
268
example, high-throughput sequencing of the Pacific herring, Clupea pallasii, transcriptome
269
yielded SNPs that were used to discriminate population structure between ocean basins
270
(Roberts et al. 2012). Interestingly, one locus (Cpa_APOB) was a virus response gene and
271
differentiation across populations at this locus suggested a disease pressure could contribute to
272
the delayed recovery of Pacific herring (Roberts et al 2012). In Pacific oysters, Crassostrea
273
gigas, transcriptome derived SNPs have been used to create a linkage map and find
274
quantitative trait loci linked to summer mortality and viral infection (Sauvage et al. 2010).
275
Another example includes the application of SNPs derived from the Glanville fritillary butterfly,
276
Melitaea cinxia, transcriptome (Vera et al. 2008) that were used in a separate large-scale study
277
to characterize metapopulation structure (Wheat et al. 2011). It should be pointed out that while
278
there are advantages to using transcriptome derived SNPs, there are also drawbacks.
279
Foremost, these markers are characterized at the genomic DNA level and the absence of
280
introns in the transcriptome can contribute to marker “drop-out” during validation (Seeb et al
281
2011). While it is beyond the scope of this effort to validate and use these markers to
282
characterize population structure, the annotated SNPs provided for H. kamtschatkana and O.
283
lurida will be a valuable resource future efforts to examine restoration and conservation activity.
284
Through phylogenetic analysis and comparative species analysis, both O. lurida and H.
285
kamtschatkana contigs have the overall greatest transcriptome similarity with closely related
286
species, C. gigas and H. midae, respectively. There was not a consistent direct relationship
287
between phylogenetic relatedness and the number of shared sequences. This is likely related to
288
the incomplete nature of some of the transcriptome datasets used for comparison. For example,
289
in both species the fully sequenced transcriptome of Danio rerio had a greater number of
290
sequence matches than some invertebrates. Gene specific phylogenetic analysis generally
291
demonstrated clustering along taxonomic lines, though there were some exceptions. Similarly,
292
this is likely due to the incomplete nature of transcripts from some species, which limited the
293
robustness of the comparison.
294
Even though most genes were shared across multiple, related species, both O. lurida
295
and H. kamtschatkana had significant subsets of expressed genes that did not have matches in
296
other species based on sequence similarity. For both species, these contigs were enriched for
297
genes involved in the environmental stress response. The sequence dissimilarity across taxa is
298
likely associated with the species and environmental specific interactions. In other words it
299
would be expected that the constitutively expressed, housekeeping genes would be conserved,
300
while genes involved in transient processes (e.g. immune function) could diverge based on
301
environmental pressure (e.g. disease, temperature). Another explanation for the relatively large
302
number of unique sequences is that a majority of the other datasets were not specifically larval
303
transcriptomes, but were derived from adult tissue. This is supported by the fact that the
304
enriched biological processes wtih the largest number of genes in O. lurida was developmental
305
processes. It is also possible that both species may possess some genes that have no
306
homologs across taxa, or potential “orphan” genes (Tautz & Domazet-Loso 2011). As additional
307
sequence information is available for related species it is likely there will be greater attention
308
directed toward these whole genome level evolutionary questions.
309
310
One important aspect of conservation is understanding how an organism responds to
environmental change. As part of this sequencing effort we identified a suite of genes that are
311
involved in the stress response that could be used in future studies to assess the physiological
312
state and fitness of an organism. One group of genes that fits this description is molecular
313
chaperones. O. lurida and H. kamtschatkana larvae express genes homologous to high
314
molecular weight molecular chaperones, including heat shock protein 70 (hsp70)
315
(Ostrea_lur_contig14040, Haliotis_kam_contig8247), hsp83 (Ostrea_lur_contig402,
316
Haliotis_kam_contig7441), and hsp90 (Ostrea_lur_contig7674, Haliotis_kam_contig1648).
317
Chaperones in the HSP70 family are responsible for the proper folding of nascent polypeptides
318
as well as providing a response to environmental stress (reviewed in Hendrick & Hartl 1993;
319
Kawabe & Yokoyama 2011; Chapman et al. 2011; Genard et al. 2012). Different isoforms of
320
HSP70 are expressed either constitutively, in response to stress, or in the endoplasmic
321
reticulum or mitochondria (reviewed in Parcellier et al. 2003). Heat shock protein 70 and its
322
transcription factor, heat shock transcription factor 1, are important in the physiological response
323
of Crassostrea gigas to both emersion and hypoxia (Kawabe & Yokoyama 2011). Increased
324
expression of hsp70 transcript can also act as a biomarker, signaling the onset of a mortality
325
event in C. virginica larvae up to a week before it occurs (Genard et al. 2012) as well as
326
indicating adult C. virginica exposure to high temperature or low pH in natural populations
327
(Chapman et al. 2011). Expression of HSP70 also increases in the gills of zebra mussel,
328
Dreissena polymorpha, by 24 hours post-exposure to copper and mercury (Navarro et al. 2011).
329
The HSP90 family, including HSP90 and HSP83, bind to various receptors and can be
330
associated with signaling proteins (Hendrick & Hartl 1993; Parcellier et al. 2003). The
331
expression of HSP90 can also be stress-induced and can stimulate and inhibit ubiquitylation
332
(Parcellier et al. 2003). The mRNA of hsp90 is up-regulated in response to changes in dietary
333
selenium in the Pacific abalone, Haliotis discus hannai Ino (Zhang W et al. 2011), dietary zinc in
334
H. discus hanai (Wu et al. 2011), and salinity and bacteria challenges in Crassostrea
335
hongkongensis (Fu et al. 2011). Global climate change is expected to significantly change the
336
temperature regimes and other environmental factors in the coastal ocean (IPCC 2007). Using
337
general stress biomarkers such as heat shock proteins could be useful in monitoring resilience
338
and adaptive potential in native invertebrate populations.
339
A second set of genes identified in O. lurida and H. kamtschatkana larval transcriptomes
340
included genes associated with oxidative stress and antioxidant activity. Invertebrate exposure
341
to pathogens and other common environmental stresses frequently results in increased
342
expression of genes and proteins associated with oxidative stress. Oxidative stress is caused
343
by proliferation of reactive oxygen species (ROS), which can promote apoptosis in cells (Wang
344
et al. 1998). ROS oxidize proteins, which can accelerate cellular aging and lead to disease
345
(Berlett & Stadtman 1997). The degradation of these damaged oxidized proteins is affected by
346
the concentration of antioxidants (Berlett & Stadtman 1997), such as the enzymes identified in
347
O. lurida and H. kamtschatkana. A number of proteins scavenge reactive oxygen species to
348
minimize cellular damage, which makes them reliable biomarkers of a variety of stresses.
349
Antioxidant genes identified in O. lurida and H. kamtschatkana included peroxiredoxin-6
350
(Ostrea_lur_contig1185, Haliotis_kam_contig6404), superoxide dismutase
351
(Ostrea_lur_contig1691, Haliotis_kam_contig8349), catalase (Ostrea_lur_contig24174),
352
cytochrome P450 (Ostrea_lur_contig24499, Haliotis_kam_contig5816), and glutathione-s-
353
transferase (Ostrea_lur_contig27247, Haliotis_kam_contig5954). Peroxiredoxin-6 (prx6) is
354
expressed at a higher level in C. virginica larvae one week before they experience a mortality
355
event, when compared to those that do not undergo mortality (Genard et al. 2012). The
356
transcript of prx6 is also expressed at higher levels in natural populations of C. gigas in the
357
winter in contaminated estuaries (David et al. 2007). Superoxide dismutase (sod) and catalase
358
(cat) increase in expression in H. discus hannai exposed to elevated dietary zinc (Wu et al.
359
2011). Catalase is an important component of C. hongkongensis’s immune response during a
360
bacterial challenge and has been shown to protect cells from H2O2-induced apoptosis (Zhang Y
361
et al. 2011). Both CAT and SOD activities were up-regulated in channel catfish, Ictalurus
362
punctatus, after exposure to contaminated sediment (Di Guilio et al. 1993). Cytochrome P450 is
363
up-regulated in the mussel Mytilus galloprovincialis upon exposure to and bioaccumulation of
364
the contaminant PCB (Livingstone et al. 1997). Glutathione-s-transferase (gst) has been found
365
to be a good predictor of C. virginica exposure to copper and iron (Chapman et al. 2011). GST
366
activity is also up-regulated in channel catfish that were exposed to contaminated sediment that
367
promoted tumor growth in aquatic organisms (Di Giulio et al. 1993). All of these antioxidant
368
enzymes are reliable indicators of physiological stress caused by a harmful environment and
369
can be applied as biomarkers to monitor populations of O. lurida and H. kamtschatkana.
370
Genes that code for proteins that accomplish a wide variety of roles in the innate
371
immune response were also identified. These include cathepsins (Ostrea_lur_contig19125,
372
Haliotis_kam_contig4888), tumor necrosis factor (TNF) receptor (Ostrea_lur_contig22192), Toll-
373
like receptor (Ostrea_lur_contig811), and mitogen activated protein kinase (MAPK)
374
(Ostrea_lur_contig10231). Some of these genes, like TNF receptor and MAPK, act upstream of
375
pathways that affect the oxidative stress response (Wang et al., 1998). MAPKs are conserved
376
across eukaryotes and are instrumental in activating a variety of enzymes in response to
377
environmental change (reviewed in Johnson & Lapadat 2002). A subfamily of MAPKs, p38
378
enzymes, have proven to be an important component of invertebrate response to a variety of
379
environmental stressors, such as oxidative stress, hypoxia, osmostic stress, and immune
380
challenge (Travers et al. 2009; Gaitanaki et al. 2004; Canesi et al. 2002). TNF receptor is
381
associated with several immune response pathways. TNFα induces phagocytic activity and
382
affects phosphorylation of the MAPK pathway in Mytilus galloprovincialis hemocytes (Betti et al.
383
2006). A TNFα homolog was cloned and sequenced in the disk abalone, Haliotis discus discus,
384
and its transcript increased expression in the abalone gill tissue upon exposure to both bacteria
385
and a virus (DeZoysa et al. 2009). Similarly, Toll-like receptors also play an integral role in
386
signaling pathways during an immune response (reviewed in Barton & Medzhitov 2003).
387
Homologs of toll-like receptors have been implicated in bacterial immune response in the
388
Zhikong scallop, Chlamys farreri (Qiu et al. 2007), and in C. gigas (Zhang L et al. 2011).
389
Cathepsins represent a wide class of enzymes, the roles of which include immune function,
390
repair, and apoptosis (reviewed in Conus & Simon 2010). Gene expression of cathepsins has
391
been shown to be an indicator of environmental stress in invertebrates. A cathepsin-B
392
precursor transcript was expressed more highly in grass shrimp, Palaemonetes pugio, exposed
393
to the pesticides fibpronil and endosulfan when compared to controls (Griffitt et al. 2006). The
394
expression of cathepsin-L mRNA also increased upon C. gigas exposure to thermal stress
395
(Meisterzheim et al. 2007). While disease mortality is has not considered a major culprit in the
396
decreased population levels of O. lurida or H. kamtschatkana, these gene sequences will be
397
important resource to examine immune function and resilience. Additionally these gene
398
sequences could be used in isolating and characterizing homologs in closely related species.
399
For example, the pathogen Candidatus Xenohaliotis californiensis has been identified as a
400
major contributor to the decline of other Haliotis species such as the black abalone, H.
401
cracherodii (Friedman & Finley 2003). Gene sequences from H. kamtschatkana could easily be
402
used to study differential disease resistance in population, which could directly aid in restoration
403
activity.
404
In summary, we have characterized genomic resources for two species of conservation
405
concern that, prior to this effort, had limited molecular information available. These data will
406
provide essential information that could be used for future physiological (i.e. gene expression)
407
and genetic based research carried out for planning, implementation, and monitoring of
408
conservation efforts.
409
410
References
411
Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ (1997) Gapped
412
BLAST and PSI-BLAST: a new generation of protein database search programs. Nucl. Acids
413
Res., 25, 3389-3402.
414
415
Barton GM, Medzhitov R (2003) Toll-Like Receptor Signaling Pathways. Science, 300, 1524-
416
1525.
417
418
Berlett BS, Stadtman ER (1997) Protein Oxidation in Aging, Disease, and Oxidative Stress.
419
JBC, 272, 20313-20316.
420
421
Betti M, Ciacci C, Lorusso LC, Cononico B, Falcioni T, Gallo G, Canesi L (2006) Effects of
422
tumor necrosis factor α (TNFα) on Mytilus haemocytes: role of stress-activated mitogen-
423
activated protein kinases (MAPKs). Biology of the Cell, 98, 233-244.
424
425
Bouma JV, Rothaus DP, Straus KM, Vadopalas B, Friedman CS (2012) Low Juvenile Pinto
426
Abalone Haliotis kamtschatkana kamtschatkana Abundance in the San Juan Archipelago,
427
Washington State. Transactions of the American Fisheries Society, 141, 76-83.
428
429
Cahais V, Gayral P, Tsagkogeorga G, Melo-Ferreira J, Ballenghien M, Weinert L, Chiari Y,
430
Belkhir K, Ranwez V, Galtier N (2012) Reference-free transcriptome assembly in non-model
431
animals from next-generation sequencing data.
432
433
Canesi L, Betti M, Ciacci C, Scarpato A, Cietterio B, Pruzzo C, Gallo G (2002) Signaling
434
pathways involved in the physiological response of mussel hemocytes to bacterial challenge:
435
the role of stress-activated p38 MAP kinases. Developmental & Comparative Immunology, 26,
436
325-334.
437
438
Chapman RW, Mancia A, Beal M, Veloso A, Rathburn C, Blair A, Holland AF, Warr GW,
439
Didinato G, Sokolova IM, Writh EF, Duffy E, Sanger D (2011) The transcriptomic responses of
440
the eastern oyster, Crassostrea virginica, to environmental conditions. Molecular Ecology, 20,
441
1431-1449.
442
443
Conus S, Simon H-U (2010) Cathepsins and their involvement in immune responses. Swiss
444
Med Wkly, 140, w13042.
445
446
David E, Tanguy A, Moraga D (2007) Peroxiredoxin 6: A new physiological and genetic indicator
447
of multiple environmental stress response in Pacific oyster Crassostrea gigas. Aquatic
448
Toxicology, 84, 389-398.
449
450
DeZoysa M, Jung S, Lee J (2009) First molluscan TNF- α homologue of the TNF superfamily in
451
disk abalone: Molecular characterization and expression analysis. Fish & Shellfish Immunology,
452
26, 625-631.
453
454
Di Guilio RT, Habig C, Gallagher EP (1993) Effects of Black Rock Harbor sediments on indices
455
of biotransformation, oxidative stress, and DNA integrity in channel catfish. Aquatic Toxicology,
456
26, 1-22.
457
458
Dinnel, PA, Peabody B, Peter-Contesse T (2009) Rebuilding Olympia oysters, Ostrea lurida
459
Carpenter 1864, in Fidalgo Bay, Washington.Journal of Shellfish Research, 28, 79-85.
460
461
Ewing B, Green P (1998) Base-Calling of Automated Sequencer Traces Using Phred. II. Error
462
Probabilities. Genome Res., 8, 186-194.
463
464
Ewing B, Hillier L, Wendl MC, Green P (1998) Base-Calling of Automated Sequencer Traces
465
Using Phred. I. Accuracy Assessment. Genome Res., 8, 175-185.
466
467
Feldmeyer B, Wheat CW, Krezdorn N, Rotter B, Pfenninger M (2011) Short read Illumina data
468
for the de novo assembly of a non-model snail species transcriptome (Radix balthica,
469
Basommatophora, Pulmonata), and a comparison of assembler performance. BMC Genomics,
470
12, doi: 10.1186/1471-2164-12-317.
471
472
Fleury E, Huvet A, Lelong C, Lorgeril J, Voulo V, Gueguen Y, Bachère E, Tanguy A, Moraga D,
473
Fabioux C, Lindeque P, Shaw J, Reinhardt R, prunet P, Davey G, Lapègue S, Sauvage C,
474
Corporeau C, Moal J, Gavory F, Wincker P, Moreews F, Klopp C, Mathieu M, Boudry P, Favrel
475
P (2009) Generation and analysis of 29,745 unique Expressed Sequence Tags from the Pacific
476
oyster (Crassostrea gigas) assembled into a publicly accessible database, the GigasDatabase.
477
BMC Genomics, 10, doi:10.1186/1471-2164-10-341.
478
479
Franchini P, van der Merwe M, Roodt-Wilding R (2011) Transcriptome characterization of the
480
South African abalone Haliotis midae using sequencing-by-synthesis. BMC Research Notes, 4,
481
doi:10.1186/1756-0500-4-59.
482
483
Frankham R (2002) Relationship of Genetic Variation to Population Size in Wildlife.
484
Conservation Biology, 10, 1500-1508.
485
486
Friedman CS, Finley CA (2003) Anthropogenic introduction of the etiological agent of withering
487
syndrome into northern California abalone populations via conservation efforts. Canadian
488
Journal of Fisheries and Aquatic Sciences, 60, 1424-1431.
489
490
Fu D, Chen J, Zhang Y, Yu Z (2011) Cloning and expression of a heat shock protein (HSP) 90
491
gene in the haemocytes of Crassostrea hongkongensis under osmotic stress and bacterial
492
challenge. Fish & Shellfish Immunology, 31, 118-125.
493
494
Gaitanaki C, Kefaloyianni E, Marmari A, Beis I (2004) Various stressors rapidly activate the p38-
495
MAPK signaling pathway in Mytilus galloprovincialis (Lam.). Molecular and Cellular
496
Biochemistry, 260, 119-127.
497
498
Genard B, Moraga D, Pernet F, David E, Boudry P, Tremblay R (2012) Expression of candidate
499
genes related to metabolism, immunity and cellular stress during massive mortality in the
500
American oyster Crassostrea virginica larvae in relation to biochemical and physiological
501
parameters. Gene, 499, 70-75.
502
503
Gillespie GE (2009) Status of the Olympia Oyster, Ostrea lurida Carpenter 1864, in British
504
Columbia, Canada. Journal of Shellfish Research, 28, 59-68.
505
506
Griffitt RJ, Chandler GT, Greig TW, Quattro JM (2006) Cathepsin B and Glutathione Peroxidase
507
Show Differing Transcriptional Responses in the Grass Shrimp, Palaemonetes pugio Following
508
Exposure to Three Xenobiotics. Environ. Sci. & Technol., 40, 3640-3645.
509
510
Groth S, Rumrill S (2009) History of Olympia Oysters (Ostrea lurida Carpenter 1864) in Oregon
511
Estuaries, and a Description of Recovering Populations in Coos Bay. Journal of Shellfish
512
Research, 28, 51-58.
513
514
Hankewich S, Lessard J, Grebeldinger E (2008) Resurvey of Northern Abalone, Haliotis
515
kamtschatkana, Populations in Southeast Queen Charlotte Islands, British Columbia, May 2007.
516
Can. Manuscr. Rep. Fish. Aquat. Sci., 2839, vii + 39 pp.
517
518
Hendrick JP, Hartl F (1993) Molecular Chaperone Functions of Heat-shock Proteins. Annu. Rev.
519
Biochem., 62, 349-384.
520
521
Huang DW, Sherman BT, Lempicki RA (2009a) Systematic and integrative analysis of large
522
gene lists using DAVID Bioinformatics Resources. Nature Protoc., 4, 44-57.
523
524
Huang DW, Sherman BT, Lempicki RA (2009b) Bioinformatics enrichment tools: paths toward
525
the comprehensive functional analysis of large gene lists. Nucleic Acids Res., 37, 1-13.
526
527
Johnson GL, Lapadat R (2002) Mitogen-Activated Protein Kinase Pathways Mediated by ERK,
528
JNK, and p38 Protein Kinases. Science, 298, 1911-1912.
529
530
IPCC (Intergovernmental Panel on Climate Change) (2007) Contribution of Working Groups I, II,
531
and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change.
532
Core Writing Team, Pachauri RK, Reisinger A (Eds) IPCC, Geneva, Switzerland, pp104.
533
534
Kawabe S, Yokoyama Y (2011) Novel isoforms of heat shock transcription factor 1 are induced
535
by hypoxia in the Pacific oyster Crassostrea gigas. Journal of Experimental Zoology Part A:
536
Ecological Genetics and Physiology, 315A, 394-407.
537
538
Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A,
539
Markowtiz S, Duran C, Thierer T, Ashton B, Meintjes P, Drummond A (2012) Geneious Basic:
540
An integrated and extendable desktop software platform for the organization and analysis of
541
sequence data. Bioinformatics, 28, 1647-1649.
542
543
Livingstone DR, Nasci C, Solé M, Da Ros L, O’Hara SCM, Peters LD, Fossato V, Wootton AN,
544
Godfarb PS (1997) Apparent induction of a cytochrom P450 with immunochemical similarities to
545
CYP1A in digestive gland of the common mussel (Mytilus galloprovincialis L.) with exposure to
546
2,2’,3,4,4’,5’-hexachlorobiphenyl and Arochlor 1254. Aquatic Toxicology, 38, 205-224.
547
548
Martin S, Richier S, Pedrotti M, Dupont S, Castejon C, Gerakis Y, Kerros M, Oberha¨nsli F,
549
Teyssié J, Jeffree R, Gattuso J (2011) Early development and molecular plasticity in the
550
Mediterranean sea urchin Paracentrotus lividus exposed to CO2-driven acidification. J Exp.
551
Biol., 214, 1357-1368.
552
553
McGraw K (2009) The Olympia Oyster, Ostrea lurida Carpenter 1864 Along the West Coast of
554
North America. Journal of Shellfish Research, 28, 5-10.
555
556
Navarro A, Faria M, Barata C, Piña B (2011) Transcriptional response of stress genes to metal
557
exposure in zebra mussel larvae and adults. Environmental Pollution, 159, 100-107.
558
559
NOAA (National Oceanographic and Atmospheric Administration) (2007) Species of Concern
560
NOAA National Marine Fisheries Service, Pinto abalone Haliotis kamtschatkana. NOAA, Silver
561
Spring, MD.
562
563
O’Donnell MJ, Hammond LM, Hofmann GE (2009) Predicted impact of ocean acidification on a
564
marine invertebrate: elevated CO2 alters response to thermal stress in sea urchin larvae. Mar.
565
Biol., 156, 439-446.
566
567
Parcellier A, Burbuxani S, Schmitt E, Solary E, Garrido C (2003) Heat shock proteins, cellular
568
chaperones that modulate mitochondrial cell death pathways. Biochemical and Biophysical
569
Research Communications, 304, 505-512.
570
571
Polson MP, Zacherl DC (2009) Geographic Distribution and Intertidal Population Status for the
572
Olympia Oyster, Ostrea lurida Carpenter 1864, from Alaska to Baja. Journal of Shellfish
573
Research, 28, 68-77.
574
575
Qiu L, Song L, Xu W, Ni D, Yu Y (2007) Molecular cloning and expression of a Toll receptor
576
gene homologue from Zhikong Scallpo, Chlamys farreri. Fish & Shellfish Immunology, 22, 451-
577
466.
578
579
Reed DH, Frankham R (2003) Correlation between Fitness and Genetic Diversity. Conservation
580
Biology, 17, 230-237.
581
582
Roberts SB, Hauser L, Seeb LW, Seeb JE (2012) Development of Genomic Resources for
583
Pacific Herring through Targeted Transcriptome Pyrosequencing. PLoS One, 7, e30908.
584
doi:10.1371/journal.pone.0030908.
585
586
Rogers-Bennett L (2007) Is climate change contributing to range reductions and localized
587
extinctions in northern (Haliotis kamtschatkana) and flat (Haliotis walallensis) abalones? Bulletin
588
of Marine Science, 81, 283-296.
589
590
Rothaus DP, Vadopalas B, Friedman CS (2008) Precipitous declines in pinto abalone (Haliotis
591
kamtschatkana kamtschatkana) abundance in the San Juan Archipelago, Washington, USA,
592
despite statewide fishery closure. Canadian Journal of Fisheries and Aquatic Sciences, 65,
593
2703-2711.
594
595
Sauvage C, Boudry P, De Koning D-J, Haley CS, Heurtebise S, Lapègue S (2010) QTL for
596
resistance to summer mortality and OsHV-1 load in the Pacific oyster (Crassostrea gigas).
597
Animal Genetics, 41, 390-399.
598
599
Seeb JE, Pascal CE, Grau ED, Seeb LW, Templin WD, Harkins T, Roberts SB (2011)
600
Transcriptome sequencing and high-resolution melt analysis advance single nucleotide
601
polymorphism discovery in duplicated salmonids. Molecular Ecology Resources, 11, 335-348.
602
603
Shen Y, Catchen J, Garcia T, Amores A, Beldorth I, Wagner J, Zhang Z, Postlethwait J, Warren
604
W, Schartl M, Walter RB (2012) Identification of transcriptome SNPs between Xiphophorus lines
605
and species for assessing allele specific gene expression within F1 interspecies hybrids.
606
Comparative Biochemistry and Physiology Part C: Toxicology & Pharmacology, 1, 102-108.
607
608
Stumpp M, Dupont S, Thorndyke MC, Melzner F (2011) CO2 induced seawater acidification
609
impacts sea urchin larval development II: Gene expression patterns in pluteus larvae.
610
Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology, 160, 320-
611
330.
612
613
Takeuchi T, Kawashima T, Koyanagi R, Fyoja F, Tanaka M, Ikuta T, Shoguchi E, Fujiwara M,
614
Shinzato C, Hisata K, Fujie M, Usami T, Nagai K, Maeyama K, Okamoto K, Aoki H, Ishikawa T,
615
Masaoka T, Fuijwara A, Endo K, Endo H, Hagasawa H, Kinoshita S, Asakawa S, Watabe S,
616
Satoh N (2012) Draft Genome of the Pearl Oyster Pinctada fucata: A Platform for
617
Understanding Bivalve Biology. DNA Research, 19, 117-130.
618
619
Tautz D, Domazet-Loso T (2011) The evolutionary origin of orphan genes. Nature Reviews, 12,
620
692-702.
621
622
Travers M, Le Bouffant R, Friedman CS, Buzin F, Cougard B, Huchette S, Koken M, Paillard C
623
(2009) Pathogenic Vibrio harveyi, in contrast to non-pathogenic strains, intervenes with the p38
624
MAPK pathway to avoid an abalone haemoctye immune response. Journal of Cellular
625
Biochemistry, 106, 152-160.
626
627
Trick M, Long Y, Meng J, Bancroft I (2009) Single nucleotide polymorphis (SNP) discovery in
628
the polyploid Brassica napus using Solexa transcriptome sequencing. Plant Biotechnology
629
Journal, 7, 334-346.
630
631
Vera JC, Wheat C, Fescemyer HW, Frilander MJ, Crawford CL, Hanski I, Marden JH (2008)
632
Rapid transcriptome characterization for a nonmodel organism using 454 pyrosequencing.
633
Molecular Ecology, 17, 1636-1647.
634
635
Wallace SS (1999) Evaluating the Effects of Three Forms of Marine Reserve on Northern
636
Abalone Populations in British Columbia, Canada. Conservation Biology, 13, 882-887.
637
638
Wang X, Martindale JL, Liu Y, Holbrook NJ (1998) The cellular response to oxidative stress:
639
influences of mitogen-activated protein kinase signaling pathways on cell survival. Biochem. J.,
640
15, 291-300.
641
642
Wheat CW, Fescemyer HW, Kvist J, Tas E, Vera JC, Frilander MJ, Hanski I, Marden JH (2011)
643
Functional genomics of life history variation in a butterfly metapopulation. Molecular Ecology,
644
20, 1813-1828.
645
646
White J, Ruesink JL, Trimble AC (2009) The Nearly Forgotten Oyster: Ostrea lurida Carpenter
647
1864 (Olympia Oyster) History and Management in Washington State. Journal of Shellfish
648
Research, 28, 43-49.
649
650
Waterhouse RM, Adobnov EM, Tegenfeldt F, Li J, Kriventseva EV (2011) OrthoDB: the
651
hierarchical catalog of eukaryotic orthologs in 2011. Nucl. Acids Res., 39,
652
doi:10.1093/nar/gkq930.
653
654
Wikelski M, Cooke SJ (2006) Conservation physiology. Trends in Ecology & Evolution, 21, 28-
655
46.
656
657
Wu C, Zhang W, Mai K, Xu W, Zhong X (2011) Effects of dietary zinc on gene expression of
658
antioxidant enzymes and heat shock proteins in hepatopancreas of abalone Haliotis discus
659
hannai. Comparative Biochemistry and Physiology Part C: Toxicology & Pharmacology, 154, 1-
660
6.
661
662
Zhang L, Li L, Zhang G (2011) A Crassostrea gigas Toll-like receptor and comparative analysis
663
of TLR pathway in invertebrates. Fish & Shellfish Immunology, 30, 653-660.
664
665
Zhang W, Wu C, Mai K, Chen Q, Xu W (2011) Molecular cloning, characterization and
666
expression analysis of heat shock protein 90 from Pacific abalone, Haliotis discus hannai Ino in
667
response to dietary selenium. Fish & Shellfish Immunology, 30, 280-286.
668
669
Zhang Y, Fu D, Yu F, Liu Q, Yu Z (2011) Two catalase homologs are involved in host protection
670
against bacterial infection and oxidative stress in Crassostrea hongkongensis. Fish & Shellfish
671
Immunology, 31, 894-903.
672
673
Zippay ML, Hofmann GE (2010) Effect of pH on Gene Expression and Thermal Tolerance of
674
Early Life History Stages of Red Abalone (Haliotis rufescens). Journal of Shellfish Research, 29,
675
429-439.
676
677
Data Accessibility
678
mRNA Sequences: NCBI SRA SRA057107
679
Final RNA sequence assembly uploaded as online supplemental material
680
SNP data: uploaded as online supplemental material
681
682
Acknowledgements
683
The authors would like to thank Taylor Shellfish Farms for supplying the Olympia oyster larvae
684
and Puget Sound Restoration Fund for supplying the pinto abalone larvae. We would also like
685
to thank Liza Ray who helped to execute the pinto abalone part of this experiment. Lisa
686
Crosson, Samantha Brombacker, and Elene Dorfmeier helped with larval care and sampling.
687
We are grateful to Josh Bouma who provided comments during the writing process. This work
688
was funded in part by the Washington Sea Grant project grant number NA10OAR4170057
689
(awarded to Dr. Carolyn Friedman). This work was also partially funded from Washington Sea
690
Grant, University of Washington, pursuant to National Oceanographic and Atmospheric
691
Administration Award No. NA10OAR4170057 Project R/LME/N—3 (awarded to Dr. Steven
692
Roberts). The views expressed herein are those of the authors and do not necessarily reflect
693
the views of NOAA or any of its sub-agencies.
694
695
Figure Legends
696
Figure 1. Functional classification of O. lurida contigs based on GO Slim terms associated with
697
Biological Processes. Other biological processes and other metabolic processes are not
698
shown.
699
700
Figure 2. Functional classification of H. kamtschatkana contigs based on GO Slim terms
701
associated with Biological Processes. Other biological processes and other metabolic
702
processes are not shown.
703
704
Figure 3. Percent coverage of O. lurida transcriptome in comparison with the transcriptomes of
705
closely related species. The C. gigas database has the most coverage of the O. lurida
706
transcriptome (52.8%) and the H. kamtschatkana database covers the least (1.9%). There are
707
a total of 82,312 contigs in the C. gigas transcriptome database, 72,597 in the P. fucata
708
database, 28,304 in the D. rerio database, 22,761 in the H. midae database, 32,606 in the R.
709
philippinarum database, and 8,641 in the H. kamtschatkana database.
710
711
Figure 4. Percent coverage of H. kamtschatkana transcriptome in comparison with the
712
transcriptomes of closely related species. The H. midae database has the most coverage of the
713
O. lurida transcriptome (35.8%) and the R. philippinarum database covers the least (6.0%).
714
There are a total of 22,761 contigs in the H. midae transcriptome database, 82,312 contigs in
715
the C. gigas database, 41,136 contigs in the O. lurida database, 28,304 contigs in the D. rerio
716
database, 72,597 contigs in the P. fucata database, and 32,606 contigs in the R. philippinarum
717
database.
718
719
Figure 5. Phylogenetic trees of hsp82 (panel A), hsp70 (panel B), transmembrane protein 85
720
(panel C), and cathepsin-L (panel D). Each tree includes sequences for all species (O. lurida,
721
H. kamtschatkana, C. gigas, H. midae, P. fucata, R. philippinarum, and D. rerio) except when
722
the sequence for a particular species did not fit in the alignment. Bootstrap consensus support
723
is shown at the nodes.
724
725
Tables & Figures
726
Table 1. Contig identification and accession numbers for the sequences used to build
727
phylogenetic trees. Numbers are specific to the database from which the contigs were taken
728
(see Materials & Methods section).
729
730
731
Table 2. Sequencing statistic summaries for high-throughput sequencing of O. lurida and H.
732
kamtschatkana transcriptomes.
733
734
735
Table 3. Summary of unique genes and their functional categories for O. lurida and H.
736
kamtschatkana. For each species, the total number of contigs corresponding to unique genes
737
in enriched biological processes are given. Total number of contigs (for enriched and non-
738
enriched biological processes) are in bold.
739
740
741
742
743
744
745
O.#lurida H.#kamtschatkana
Number$of$contigs$in$category
Number'Unique'Genes
1315
243
RNA$metabolism
.
5
Stress$response
65
3
Cell$adhesion
156
.
Cell$organization$and$biogenesis
241
.
Developmental$processes
328
.
Protein$metabolism
268
.
Signal$transduction
114
.
Transport
39
.
746
747
748
749
750
751
752
753
754
755
756
Figure 1
757
Figure 2
DNA(metabolism(
protein(metabolism(
developmental(processes(
death(
cell(organiza0on(
and(biogenesis(
cell(cycle(and(
prolifera0on(
cell(adhesion(
RNA(metabolism(
signal(transduc0on(
transport(
stress(response(
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
Figure 3
773
774
775
776
Figure 4
777
Figure 5
778
779
780
781
Supplementary Data Legends
782
Supp_1_Ostrea_lurida_transcriptome. Assembled contigs of O. lurida transcriptome
783
sequencing.
784
785
Supp_2_Haliotis kamtschatkana_transcriptome. Assembled contigs of H. kamtschatkana
786
sequencing.
787
788
Supp_3_Ostrea_lurida_SPIDs. BLASTx results for O. lurida contig search against the
789
UniProtKB/Swiss-Prot database. BLAST e-values and gene descriptions are also given.
790
791
Supp_4_Ostrea_lurida_GO. Gene Ontology annotations of O. lurida contigs. GO annotations
792
are made based on associations with a Swiss-Prot ID.
793
794
Supp_5_Haliotis_kamtschatkana_SPIDs. BLASTx results for H. kamtschatkana contig search
795
against the UniProtKB/Swiss-Prot database. BLAST e-values and gene descriptions are also
796
given.
797
798
Supp_6_Haliotis_kamtschatkana_GO. Gene Ontology annotations of H. kamtschatkana contigs.
799
GO annotations are made based on associations with a Swiss-Prot ID.
800
801
Supp_7_Ostrea_lurida_bitscores. Bit scores for BLASTn results of O. lurida contigs against
802
species-specific databases of other closely related species.
803
804
Supp_8_Haliotis_kamtschatkana_bitscores. Bit scores for BLASTn results of H. kamtschatkana
805
contigs against species-specific databases of other closely related species.
806
807
Supp_9_Ostrea_lurida_SNPs. SNP information for putative SNPs identified in the O. lurida
808
transcriptome. Contig numbers are listed in the leftmost column, followed by SNP location and
809
allele. Annotations of the contigs, as determined through a BLASTx against the
810
UniProtKB/Swiss-Prot database, are given along with the e-value for the BLAST result.
811
812
Supp_10_Haliotis_kamtschatkana_SNPs. SNP information for putative SNPs identified in the H.
813
kamtschatkana transcriptome. Contig numbers are listed in the leftmost column, followed by
814
SNP location and allele. Annotations of the contigs, as determined through a BLASTx against
815
the UniProtKB/Swiss-Prot database, are given along with the e-value for the BLAST result.
816
817
Author Contributions Box
818
ETS performed the analyses and contributed to writing of the manuscript. CSF and SBR were
819
responsible for experimental design and provided funding for the research. SBR helped to
820
write the manuscript as well. DCM and SJW performed the lab work for the O. lurida and H.
821
kamtschatkana parts of the experiment, respectively. All co-authors contributed comments and
822
editing during the manuscript writing process.
823
824