Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Insights into the strategy of micro-environmental adaptation: Transcriptomic analysis of two alvinocaridid shrimps at a hydrothermal vent

  • Fang-Chao Zhu,

    Roles Conceptualization, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing

    Affiliations Institute of Deep-sea Science and Engineering, Chinese Academy of Sciences, Sanya, Hainan, China, College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing, China

  • Jin Sun,

    Roles Conceptualization, Methodology, Resources, Writing – review & editing

    Affiliation Department of Ocean Science, The Hong Kong University of Science and Technology, Hong Kong, China

  • Guo-Yong Yan,

    Roles Investigation, Resources

    Affiliation Institute of Deep-sea Science and Engineering, Chinese Academy of Sciences, Sanya, Hainan, China

  • Jiao-Mei Huang,

    Roles Methodology

    Affiliations Institute of Deep-sea Science and Engineering, Chinese Academy of Sciences, Sanya, Hainan, China, College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing, China

  • Chong Chen,

    Roles Funding acquisition, Resources, Writing – review & editing

    Affiliation Japan Agency for Marine-Earth Science and Technology (JAMSTEC), Yokosuka, Kanagawa, Japan

  • Li-Sheng He

    Roles Conceptualization, Funding acquisition, Methodology, Writing – review & editing

    Affiliation Institute of Deep-sea Science and Engineering, Chinese Academy of Sciences, Sanya, Hainan, China


Diffusing fluid at a deep-sea hydrothermal vent creates rapid, acute physico-chemical gradients that correlate strongly with the distribution of the vent fauna. Two alvinocaridid shrimps, Alvinocaris longirostris and Shinkaicaris leurokolos occupy distinct microhabitats around these vents and exhibit different thermal preferences. S. leurokolos inhabits the central area closer to the active chimney, while A. longirostris inhabits the peripheral area. In this study, we screened candidate genes that might be involved in niche separation and microhabitat adaptation through comparative transcriptomics. The results showed that among the top 20% of overexpressed genes, gene families related to protein synthesis and structural components were much more abundant in S. leurokolos compared to A. longirostris. Moreover, 15 out of 25 genes involved in cellular carbohydrate metabolism were related to trehalose biosynthesis, versus 1 out of 5 in A. longirostris. Trehalose, a non-reducing disaccharide, is a multifunctional molecule and has been proven to act as a protectant responsible for thermotolerance in Saccharomyces cerevisiae. Putative positively selected genes involved in chitin metabolism and the immune system (lectin, serine protease and antimicrobial peptide) were enriched in S. leurokolos. In particular, one collagen and two serine proteases were found to have experienced strong positive selection. In addition, sulfotransferase-related genes were both overexpressed and positively selected in S. leurokolos. Finally, genes related to structural proteins, immune proteins and protectants were overexpressed or positively selected. These characteristics could represent adaptations of S. leurokolos to its microhabitat, which need to be confirmed by more evidence, such as data from large samples and different development stages of these alvinocaridid shrimps.


Deep-sea hydrothermal vents are highly dynamic and unstable, both temporally and spatially. Fluids emitted from these vents, at temperatures ranging from approximately 20°C to as high as 407°C [1], mix directly with ambient seawater (~2°C) and, thus, create steep thermal and chemical gradients. The vent-associated fauna exhibits clear zonation patterns that are consistent with the physico-chemical gradients [2]. Among the factors that affect the species distribution around hydrothermal vents, the temperature and sulphides always play predominant roles [3, 4].

In the hydrothermal fields of the Okinawa Trough, Shinkaicaris leurokolos (Alvinocarididae, Rimicaridinae) and Alvinocaris longirostris (Alvinocarididae, Alvinocaridinae) usually co-exist sympatrically but occupy distinct microbiotopes according to in situ observations [5]. For example, at the Iheya North Knoll in the middle Okinawa Trough, the fauna directly influenced by vent activity can be divided into four zones based on thermal conditions. Among the endemic crustaceans, S. leurokolos inhabits the central zone (defined as zone 2, 0.2–0.8 m from vent) together with the squat lobster Shinkaia crosnieri, while A. longirostris mainly inhabits the peripheral zone (zone 4, >2.5 m from the vent), far away from the active chimney, as do Bathymodiolus platifrons mussels. The area within a 0.2 m radius of the vent is considered zone 1, and the transitional area (0.8–2.5 m away from vent) between zone 2 and zone 4 is defined as zone 3 [6]. Shinkaicaris leurokolos exhibits a similar microhabitat preference to Rimicaris exoculata (Alvinocarididae, Rimicaridinae) [7]. Adult R. exoculata prefers to inhabit areas with temperatures in the range of 10–25°C, and swarms of this species may tolerate occasional heat shocks that exceed its maximum critical temperature (33–38.5±2°C) [8]. For A. longirostris, the ambient temperature is approximately 3–4°C in both hydrothermal vents and cold seeps [9]. An experiment showed that a higher optimal temperature (10–20°C) is required for S. leurokolos to reach the maximum hatching rate of its embryos than for A. longirostris (10°C) under atmospheric pressure [10]. In addition, the morphological trends of these species suitable for different vent microhabitats have been revealed. S. leurokolos and R. exoculata, which occur in the vicinity of vent fluids, have evolved a degenerate rostrum and reduced external spines, both of which reduce the impact of strong turbulent fluid flows; they also have a dorsal organ that is used for detecting dim light emitted from the vents inside their carapaces; however, A. longirostris does not have dorsal organs, and its rostrum and spines are well developed [11].

Studies have been performed to investigate the mechanisms of environmental adaptation in the vent fauna in comparison with their shallow-water relatives. The expression levels of metal-binding proteins (metallothioneins) and the activities of antioxidant enzymes (such as superoxide dismutase, catalase, and glutathione peroxidase) show significant differences between vent and coastal shrimps. These genes are thought to be associated with heavy metal detoxification [12, 13]; the expression of heat shock proteins increases in R. exoculata, the crab Chaceon affinis, and the annelid Paralvinella grasslei, following an acute heat stimulus in the laboratory [1416]. In recent years, large-scale gene profiles of vent-endemic invertebrates such as shrimp (Rimicaris sp.), mussel (Bathymodiolus platifrons) and tubeworms (Branchipolynoe pettiboneae, Lepidonotopodium sp.), have been analysed by next-generation sequencing [9, 1719]. Consequently, a group of genes involved in sulphur metabolism, immune defence, antioxidation and detoxification have been successfully identified as being associated with environmental adaptation. However, in addition to the dramatic changes between deep-sea and shallow-sea regions, physico-chemical characteristics also vary significantly at a finer scale around vents. Zonation may induce variable physiological and biochemical adaptations, even for the same species from different microhabitats in a single hydrothermal field [20]. Thus far, the strategies for coping with fine-scale environmental fluctuations within the deep-sea vent fauna are still unknown.

In this study, we assembled the transcriptomes of A. longirostris and S. leurokolos, compared highly expressed genes and identified positively selected genes, providing preliminary clues about the genetic basis of the microhabitat adaptation of hydrothermal alvinocaridid shrimps.

Materials and methods

Ethical statement

This study does not involve endangered or protected species. Sample collection was conducted in the Japanese exclusive economic zone by a Japanese government research vessel. No specific permission was required for the sampled location.

Sample collection and sequencing

Specimens of A. longirostris and S. leurokolos samples were collected from the Sakai hydrothermal vent field (27˚31.4749' N, 126˚59.021' E; depth = 1,550 m) in the middle Okinawa Trough by the JAMSTEC ROV KAIKO Mk-Ⅳ during R/V KAIREI cruise KR15-17 in November 2015 (PI: Hiroyuki Yamamoto) [21]. After being brought on board, the specimens were immediately preserved in RNAlater stabilization solution (Invitrogen, USA) at 4°C overnight, and then transferred to -80°C for long-term storage. Two specimens of each species were used for analysis: one for transcriptome sequencing and the other for absolute quantitative real-time PCR (qPCR). Total RNA was extracted from the dissected cephalothorax and pleon using TRIzol reagent (Invitrogen, USA). The quality and quantity of the RNA were examined by agarose gel electrophoresis and with a Qubit 2.0 Fluorometer (Invitrogen, USA). Then, cDNA libraries were constructed and sequenced on the Illumina HiSeq 4000 platform at Novogene (Beijing, China).

De novo transcriptome assembly

The quality of 150 bp paired-end reads was assessed by FastQC v0.10.1 ( Contaminated adapters and poor-quality bases were trimmed using Trimmomatic-0.36 in paired-end mode [22]. Bases at both ends of the reads were cut off if the quality score was less than 5. Then, the reads that would be cut if the average quality dropped below 15 were scanned in a 4-base-wide sliding window. Finally, reads of less than 36 bases were removed. The Trinity v2.3.2 software package was utilized to assemble clean reads into putative transcripts with the minimum k-mer coverage set to 2 and the other parameters set to default [23]. The completeness of each transcriptome assembly was evaluated by using BUSCO v3.0.2 and Arthropda OrthoDB9 [24]. To remove redundant isoforms, only the longest transcript of each gene set was selected as a unigene.

Phylogenetic analysis

Mitochondrial cytochrome c oxidase subunit I (COI) and 16S rRNA genes were separately used for phylogenetic analysis. The full-length COI and 16S rRNA genes of ten alvinocaridid species were downloaded from the NCBI database, and the pandalid shrimp Heterocarpus ensifer (Pandalidae) was used as an outgroup. The downloaded genes were searched against the unigenes using the Blastn program (Blast+ v2.5.0) to retrieve the assembled COI and 16S sequences. Multiple sequence alignment was performed using the MAFFT v7.294b program [25], and the aligned sequences were subsequently trimmed using the trimAl v1.4 tool [26]. Total lengths of 1,534 bp (COI) and 1,303 bp (16S) were reserved for the construction of maximum-likelihood phylogenetic trees. The TIM2+F+I+G4 model and TIM3+F+G4 model were selected for the COI and 16S rRNA sequences using ModelFinder, respectively. The phylogenetic trees were inferred by using IQ-TREE version 1.6.12 with 1,000 ultrafast bootstraps [27].

Annotation of protein-coding genes

TransDecoder v3.0.1 was used to predict candidate open reading frames (ORFs) from unigenes with homology to known proteins via Blast or pfam searches ( All predicted protein sequences were searched against the NCBI non-redundant (nr, downloaded in 08/03/2017) protein database via Blastp alignment (Blast+ v2.5.0) with an e-value cutoff of 1e-05. Conserved protein domains were identified by searching the Pfam 30.0 database using InterProScan v5.22 [28]. Gene Ontology (GO) annotation was implemented with Blast2Go Basic v5.2.5 [29]. KEGG pathway annotation was carried out with the online tool KEGG Automatic Annotation Server (

Comparison of highly expressed genes

Gene expression levels measured as transcripts per million (TPM) values were calculated with RSEM 1.3.0 [30]. The top 20% of highly expressed proteins were extracted and then classified into particular groups based on the annotated GO terms by using the online tool WEGO 2.0 [31]. The percentages of each gene group were compared between A. longirostris and S. leurokolos. Pearson’s Chi-square test was applied for 2×2 matrixes if all the expected gene numbers were greater than 5. A p-value < 0.05 indicated a significant difference.

Positive selection analysis

Orthogroups of pairwise species were predicted using InParanoid 4.1 with default parameters [32]. The coding sequences of Daphnia pulex were obtained from Ensembl Genomes and served as an outgroup [33]. Only orthogroups with single-copy genes (one to one orthologue pairs) were retained for positive selection analysis. For each single-copy orthogroup, protein-coding sequence alignment was implemented with ParaAT v2.0, in which the multiple sequence alignment program was specified as MAFFT, and both aligned codons with gaps and mismatched codons were removed [34]. The ratio of the number of nonsynonymous substitutions per nonsynonymous site (Ka) to the number of synonymous substitutions per synonymous site (Ks) was calculated using KaKs Calculator 2.0 with the model-averaging method [35]. Multiple testing correction was performed via false discovery rate (FDR) estimation. Orthologous pairs with an FDR>0.05, Ks<0.01, Ks>1, or Ka>1 were discarded [36]. Ka/Ks>1 indicated strong positive selection. Ka/Ks>0.5 was also used as a less conservative cut-off that had proven to be useful for identifying positively selected genes (PSGs) [37]. The functions of candidate PSGs with Ka/Ks>0.5 were enriched using TBtools under the threshold of an adjusted p-value < 0.05 [38].

Quantitative real-time PCR

Approximately 2 μg of total RNA was used for cDNA synthesis by using High Capacity cDNA Reverse Transcription Kits (Applied Biosystems, USA), and contaminating DNA was removed by using the TURBO DNA-free Kit (Ambion, USA). The cDNA products were diluted 10-fold and used as templates. Primer pairs were designed with the NCBI on-line tool Primer-BLAST (S1 Table). The target gene fragment was amplified using PrimeSTAR HS DNA Polymerase (Takara, Japan) and cloned into the pMD18-T vector (Takara, Japan). Then, the recombinant plasmid was transformed into DH5α competent cells and positive clones were sent to BGI for Sanger sequencing. The quantification results for the transcriptomes were validated by absolute qPCR using TB Green Premix Ex Taq Ⅱ (Takara, Japan) and the StepOnePlus Real-Time PCR system (Applied Biosystems, USA). The recombinant plasmid was extracted using a TIANprep Mini Plasmid Kit (TIANGEN, China). A standard curve was generated with serial 10-fold dilutions of the recombinant plasmid. The real-time PCR mixture (20 μl) contained 10 μl of TB Green Premix Ex Taq II (2×), each of forward and reverse primers at 0.4 μM, 0.4 μl of ROX reference dye, and 2 μl of diluted cDNA. The amplification program was as followes: 95°C for 30 s, followed by 40 cycles of 95°C for 5 s and 60°C for 30 s. All samples were tested in three technical replicates. Putative homologous genes of trehalose-6-phosphate synthases (TPSs) were also confirmed by PCR. The sequenced data were searched against predicted TPSs in transcriptomes with Blastx and aligned with Clustal Omega [39].


Transcriptome assembly and annotation

A total of 33,324,253 and 39,266,315 pairs of raw reads were generated for A. longirostris and S. leurokolos, respectively. After quality filtering, 82.98% and 82.41% the raw reads were retained for de novo transcriptome assembly, which generated 158,408 trinity transcripts for A. longirostris and 173,354 for S. leurokolos. Accordingly, the assembled transcripts had average lengths of 757.75 and 677.35 bp, with N50 lengths of 1,460 and 1,198 bp for A. longirostris and S. leurokolos (Table 1). When aligned with 1,066 benchmarking universal single-copy orthologues (BUSCOs) from arthropods, 90.7% complete BUSCOs were found to be present in the transcriptome of A. longirostris and 89.6% in S. leurokolos (S2 Table).

Table 1. Information on the de novo transcriptome assembly.

Excluding redundant isoforms, 129,409 and 143,754 unigenes were retained for A. longirostris and S. leurokolos, respectively. Then, 28,782 and 34,390 ORFs with a minimum length of 300 bp were predicted for A. longirostris and S. leurokolos, respectively. By database searching, 20,730 (for A. longirostris) and 23,720 (for S. leurokolos) predicted proteins were annotated in at least one database. In particular, 20,581 and 23,548 sequences returned significant hits in the nr database (Table 2). The top-hits species distribution showed that most of the predicted protein sequences were similar to proteins from the amphipod Hyalella azteca, indicating that no obvious contamination was present in both of the assembled transcriptomes (S1 Fig).

Phylogenetic analysis of shrimp

Both the 16S rRNA and COI nucleotide sequences of A. longirostris used in this study shared 100% identity with an individual collected from the Hatoma Knoll in the southern Okinawa Trough (accession number of the mitochondrial genome: AB821296), while the 16S rRNA and COI nucleotide sequences of S. leurokolos shared 99.85% and 99.74% identity with a reported sample from the middle Okinawa Trough (accession number of the mitochondrial genome: MF627741). Their phylogenetic relationships with other alvinocaridid shrimps were reconstructed based on the 16S rRNA and COI genes, respectively. The two trees displayed similar topologic structures. S. leurokolos clustered with the Opaepele, Manuscaris and Rimicaris genera, and they formed the Rimicaridinae subfamily clade. Then, this clade was separated from the Nautilocaris and Alvinocaris genera (Fig 1).

Fig 1.

Maximum-likelihood phylogenetic tree of alvinocaridid shrimps based on COI (A) and 16S rRNA (B) genes. The sequences in red are from this study. Statistical supports is indicated as bootstrap values, and the values of less than 50 are omitted. Subfamilies are masked by different colours: Alvinocaridinae (blue), Mirocaridinae (yellow), Rimicaridinae (green). Heterocarpus ensifer (family: Pandalidae) serves as the outgroup. Accession numbers are labeled in parentheses. The tree scale bar represents the number of expected substitutions per site.

Gene family-based comparison

After quantification, 5,761 and 6,879 genes ranked in the top 20% of highly expressed genes in A. longirostris and S. leurokolos, respectively. Among these genes, 4,144 and 4,982 successfully returned GO annotations (S3 Table). At GO level 2, the highly expressed genes showed a similar distribution for A. longirostris and S. leurokolos, most of which were concentrated in binding, metabolic process, cellular process and catalytic activity (S2 Fig).

The percentages of each gene group among the top 20% of highly expressed genes were compared between A. longirostris and S. leurokolos. Notably, there was a higher percentage of genes involved in cellular carbohydrate metabolism in S. leurokolos compared to A.longirostris (Table 3). There were 23 genes involved in this category for S. leurokolos compared to 5 genes for A.longirostris. Moreover, 15 out of 23 genes were annotated as TPSs, versus only 1 out of 5 genes in A. longirostris. The genes involved in structural components, including the extracellular matrix, integral component of membrane and cell cortex groups, were also more abundant in S. leurokolos than in A. longirostris. A higher percentage of genes involved in ribosome and translation was also observed in S. leurokolos. In addition, two groups of genes related to sulphate metabolism presented a higher proportion in S. leurokolos. One group was related to sulfuric ester hydrolases, mainly including arylsulfatase A, arylsulfatase B and N-acetylgalactosamine-6-sulfatase. The other group was related to sulfotransferases and was composed of carbohydrate sulfotransferases 9 and 11, and sulfotransferase 1C4. In contrast, two gene groups were over-represented in A. longirostris: NAD+ ADP-ribosyltransferase activity (poly (ADP-ribose) polymerase (PARP)) and cysteine-type peptidase (mainly cathepsin and ubiquitin carboxyl-terminal hydrolase) (Table 3).

Table 3. GO families with significant differences between A. longirostris and S. leurokolos.

To validate the RNA-seq results, six orthogroups with single-copy genes were randomly selected for absolute quantification analysis, and the variation tendencies of the gene copy numbers were consistent with the TPM values (Fig 2). Furthermore, 7 out of 15 putative TPS homologous genes from S. leurokolos were successfully amplified from the cDNA libraries, and the PCR products shared 100% amino acid identity with the corresponding TPSs assembled from the transcriptomes. The TPS segments, ranging from 100 to 325 amino acids, were aligned with TPSs from Penaeus chinensis and Callinectes sapidus. S43407_c0_g2 and S55571_c2_g2 were aligned to the glycosyltransferase family 20 domain (PF00982) of TPS from Penaeus chinensis (residues 7–483), while the other five unigenes were highly similar (with 59.90~81.14% sequence identity in amino acid level) to the trehalose-phosphatase domain (PF02358, residues 520–745) (S3 Fig). The presence of a TPS sequence from A. longirostris was also confirmed.

Fig 2. Validation of gene expression by absolute qPCR.

Six single-copy orthologues for each species were randomly selected and quantified in the cephalothorax of S. leurokolos (A), abdomen of S. leurokolos (B), cephalothorax of A. longirostris (C) and abdomen of A. longirostris (D). The histograms show the gene copy number per μl (mean ± SD) with three technical replicates quantified by qPCR. The line charts show the TPM value quantified with RSEM software. Gene names are indicated on the x-axis.

Positively selected genes

In total, 12,544 orthogroups were identified for A. longirostris and S. leurokolos. A total of 11,002 pairs were comprised of single-copy genes and used for positive selection analysis. After filtering the data with an FDR>0.05, Ks<0.01, Ks>1 and Ka>1, 9,114 pairs of single-copy orthogroups were finally retained. Among these orthogroups, 402 pairs of orthologous genes exhibited a Ka/Ks value greater than 0.5 (S4 Fig), and 20 pairs of orthologues presented a Ka/Ks value greater than 1. However, only four PSGs were successfully annotated in the nr database: CUB-serine protease, trypsin, collagen alpha-1(IV) chain and NAD-specific glutamate dehydrogenase (S4 Table).

These genes with Ka/Ks values > 0.5 were further analysed for functional enrichment (S4 Table). As a result, five genes possessing sulfotransferase activity and eight genes participating in chitin metabolic processes were found to be significantly enriched among the moderate PSGs. Another eight enriched PSGs exhibited carbohydrate binding activities, and six of them were lectins. The next enriched group belonged to endopeptidase, including nine serine proteinases. The antimicrobial peptides (AMPs) were enriched in the groups of molecular function regulators and extracellular regions (Table 4).

Table 4. Enriched positively selected genes in S. leurokolos.


How organisms adapt to deep-sea environments has always been an interesting topic. The subject of adaptation to the microenvironment in special areas such as hydrothermal vent fields is easy to be ignore but important. In this paper, two vent-endemic alvinocaridid shrimps were used as an example to illustrate the possible genes and pathways involved in microenvironmental adaptation. The protein synthesis rate has a significant impact on thermal acclimation, although the relationship between them is complex [40, 41]. Proteins are usually vulnerable to elevated temperatures because they maintain their function within only a narrow range of temperatures. It has been demonstrated that in vitro high temperatures inhibit mRNA translation by suppressing Met-tRNA synthetase activity [42]. Decreased protein turnover reduces metabolic sensitivity to environmental change [43]. Therefore, active protein synthesis may be a compensation mechanism to balance the protein turnover. In this study, genes associated with ribosomes and translation were highly expressed, indicating more active protein synthesis in S. leurokolos compared to A. longirostris. In the group of cellular carbohydrate metabolism of S. leurokolos, more than 60% of genes were annotated as TPSs, which were key enzymes for trehalose biosynthesis. Trehalose, a non-reducing disaccharide, is a multifunctional molecule that plays important roles in sugar metabolism, stress recovery, chitin synthesis and other biological processes [44]. Trehalose is also a protectant responsible for thermotolerance, as demonstrated in Saccharomyces cerevisiae [45]. Functionally, it acts as a chemical co-chaperone to delay protein degradation and aggregation, possibly due to the preferential formation of the peptide-trehalose hydrogen bond [46, 47]. The presence of TPS homologs was validated by PCR, and we inferred that there were at least four different TPS genes in S. leurokolos, according to sequence alignment. The TPSs existing in deep-sea invertebrates have been poorly investigated, and the functions of different TPSs within the same species are still unclear. However, as a primary enzyme in trehalose synthesis, an increase in trehalose might help S. leurokolos to cope with temperature variation and other stresses.

Basic structural proteins such as extracellular matrix, integral component of membrane and cell cortex proteins displayed distinct expression patterns between S. leurokolos and A. longirostris. Among these proteins, the basement-membrane collagen alpha-1(IV) chain protein was found to be under particularly strong positive selection. The deep-sea polychaetous annelids Alvinella pompejana and Riftia pachyptila present a similar living pattern to the shrimps investigated in this study: the former inhabits the surface of chimney walls and tolerates temperatures up to 60–65°C; the latter inhabits regions with a relatively lower temperature (approximately 37°C). The thermal tolerance of A. pompejana is mainly due to interstitial collagen because of its increased proline content and hydroxylation [48]. However, in another kind of fibrillar collagen from R. pachyptila, glycosylated threonine but not 4-hydroxyproline contributes to triple helix stability [49]. All of the potential collagens in the transcriptomes of A. longirostris and S. leurokolos were identified by conserved domain searches, and the compositions of their amino acids were calculated. The results showed that the percentages of proline (12.34% in A. longirostri versus 13.80% in S. leurokolos) and threonine (5.38% versus 4.46%) in collagens were significantly different between the species (S5 Table). We infer that the thermostability of collagen from S. leurokolos differs from that of A. longirostris. Another structure-related group that was enriched in the putative PSGs of S. leurokolos was the chitin metabolic process category. Peritrophin-44 and mucin are components of the peritrophic membrane, a non-cellular structure secreted from the midgut epithelium of invertebrates. Chitinase is generally found in tissues (such as the peritrophic membrane) that either require the remodelling of chitinous structures or degradation of digested chitin [50]. In decapod crustaceans, the peritrophic membrane commonly provides an intestinal barrier that protects against mechanical and chemical damage and prevents pathogen infection [51].

Genes participating in innate immunity of invertebrates were enriched among the positively selected genes, including lectin, caspase, serine proteinase and AMP. As important pattern recognition receptors (PRRs), the crustacean lectins recognize glycans on the cell surface of invading pathogens and activate a range of immune responses [52]. However, PRRs are also required to promote the normal colonization of gut microbiota [53]. A C-type lectin from R. exoculata recognizes and agglutinates Escherichia coli in vitro without the inhibition of bacterial growth [54]. Commonly, AMPs directly kill pathogens by disrupting their cell membranes. However, an AMP known as coleoptericin-A from weevil selectively targets endosymbionts within bacteriocytes and controls their growth through the inhibition of cell division [55]. More importantly, it has been reported that caspases regulate endosymbiont density in deep-sea Bathymodiolus mussels through the mechanism of gill cell apoptosis [56]. The serine proteinases not only regulate antimicrobial peptide synthesis and prophenoloxidase activation but also mediate apoptosis-like cell death [57, 58]. Thus, the lectins, caspases, serine proteinases and AMPs as well as other innate immune molecules are also possibly involved in the management of symbiont populations. The dominant chemosynthetic bacteria associated with A. longirostris and S. leurokolos are assumed to be different because of the divergence of their carbon fixation pathways [59]. In addition, the size of the host symbiotic bacterial population varies according to the supply of free H2S/HS- in the environment [60]. By analogy with R. exoculata, S. leurokolos probably has more abundant symbionts in its gill chambers. Therefore, the identified immune molecules may contribute to the differences between A. longirostris and S. leurokolos in terms of distinguishing different symbiotic bacteria and regulating their densities to address environmental fluctuations.

Extremely high genetic diversity of S. leurokolos was revealed in the Okinawa Trough, but A. longirostris showed low genetic diversity [61]. The 16S rRNA nucleotide sequences obtained in our study were nearly identical (>99%) with those of previously reported A. longirostris and S. leurokolos samples, as were those of the COI genes [62]. However, even within the same species, variation in environmental acclimation exists between populations and phylogenetic lineages [63]. Therefore, the genes screened in our study still need to be further confirmed based on a dataset including replicate specimens for each species.

In conclusion, genes related to protein synthesis, structural components and trehalose biosynthesis might be involved in thermal acclimation, and a group of immune proteins might be involved in symbiosis preservation. The differences observed between the two species for these genes provide clues about the discrepancy in microhabitats between A. longirostris and S. leurokolos.

Supporting information

S1 Fig. Top-hit species classification of predicted proteins with nr annotation.

A indicates A. longirostris and B indicates S. leurokolos.


S2 Fig. Gene ontology distribution.

The top 20% of highly expressed genes were analysed. The X-axis shows the GO terms in level 2; the y-axis shows the percentages of genes (number of a particular gene divided by total gene number) on the left and the number of genes on the right.


S3 Fig. Multiple alignment of validated TPSs.

NCBI accession number ACD74843.1 indicates TPS from Penaeus chinensis, and ACI12944.1 indicates TPS from Callinectes sapidus.


S4 Fig. Distribution of Ka and Ks values.

Dots between the y-axis and the grey line represent orthologous pairs with a Ka/Ks ratio>1, dots between the x-axis and the red line represent orthologous pairs with a Ka/Ks ratio<0.5, and dots between the red and grey lines represent a 1>Ka/Ks ratio>0.5.


S2 Table. Completeness estimation of transcriptome assemblies.


S3 Table. List of the top 20% of highly expressed genes.


S4 Table. Orthologous genes displaying evidence of positive selection.


S5 Table. The percentages of amino acids in collagens from A. longirostris and S. leurokolos.



We would like to thank the crew and scientists on-board R/V KAIREI cruise KR15-17 as well as the operation team of ROV KAIKO Mk-IV. The cruise chief scientist Dr. Hiroyuki Yamamoto (JAMSTEC) is gratefully acknowledged for leading the cruise to success and for bridging collaborations.


  1. 1. Pedersen RB, Rapp HT, Thorseth IH, Lilley MD, Barriga FJ, Baumberger T, et al. Discovery of a black smoker vent field and vent fauna at the Arctic Mid-Ocean Ridge. Nat Commun. 2010; 1: 126. pmid:21119639
  2. 2. Kelly N, Metaxas A, Butterfield D. Spatial and temporal patterns of colonization by deep-sea hydrothermal vent invertebrates on the Juan de Fuca Ridge, NE Pacific. Aquat Biol. 2007; 1: 1–16.
  3. 3. Cuvelier D, Sarradin PM, Sarrazin J, Colaço A, Copley JT, Desbruyères D, et al. Hydrothermal faunal assemblages and habitat characterisation at the Eiffel Tower edifice (Lucky Strike, Mid-Atlantic Ridge). Mar Ecol. 2011; 32(2): 243–255.
  4. 4. Husson B, Sarradin PM, Zeppilli D, Sarrazin J. Picturing thermal niches and biomass of hydrothermal vent species. Deep Sea Res Part 2 Top Stud Oceanogr. 2017; 137: 6–25.
  5. 5. Komai T, Segonzac M. A revision of the genus Alvinocaris Williams and Chace (Crustacea: Decapoda: Caridea: Alvinocarididae), with descriptions of a new genus and a new species of Alvinocaris. J Nat Hist. 2005; 39(15): 1111–1175.
  6. 6. Watanabe H, Kojima S. Vent fauna in the Okinawa Trough. In: Ishibashi J, Okino K, Sunamura M, editors. Subseafloor biosphere linked to hydrothermal systems. Tokyo: Springer; 2015. p. 449–459.
  7. 7. Gebruk AV, Southward EC, Kennedy H, Southward AJ. Food sources, behaviour, and distribution of hydrothermal vent shrimps at the Mid-Atlantic Ridge. J Mar Biol Assoc U.K. 2000; 80(3): 485–499.
  8. 8. Cottin D, Shillito B, Chertemps T, Thatje S, Léger N, Ravaux J. Comparison of heat-shock responses between the hydrothermal vent shrimp Rimicaris exoculata and the related coastal shrimp Palaemonetes varians. J Exp Mar Bio Ecol. 2010; 393(1–2): 9–16.
  9. 9. Hui M, Cheng J, Sha Z. Adaptation to the deep-sea hydrothermal vents and cold seeps: insights from the transcriptomes of Alvinocaris longirostris in both environments. Deep Sea Res Part 1 Oceanogr Res Pap. 2018; 135: 23–33.
  10. 10. Watanabe H, Yahagi T, Nagai Y, Seo M, Kojima S, Ishibashi J, et al. Different thermal preferences for brooding and larval dispersal of two neighboring shrimps in deep‐sea hydrothermal vent fields. Mar Ecol. 2016; 37(6): 1282–1289.
  11. 11. Vereshchaka AL, Kulagin DN, Lunina AA. Phylogeny and new classification of hydrothermal vent and seep shrimps of the family Alvinocarididae (Decapoda). PLoS One. 2015; 10(7): e0129975. pmid:26161742
  12. 12. Gonzalez‐Rey M, Serafim A, Company R, Bebianno MJ. Adaptation to metal toxicity: a comparison of hydrothermal vent and coastal shrimps. Mar Ecol. 2010; 28(1): 100–107.
  13. 13. Gonzalez‐Rey M, Serafim A, Company R, Gomes T, Bebianno MJ. Detoxification mechanisms in shrimp: comparative approach between hydrothermal vent fields and estuarine environments. Mar Environ Res. 2008; 66(1): 35–37. pmid:18405963
  14. 14. Cottin D, Ravaux J, Léger N, Halary S, Toullec JY, Sarradin PM, et al. Thermal biology of the deep-sea vent annelid Paralvinella grasslei: in vivo studies. J Exp Biol. 2008; 211: 2196–2204. pmid:18587113
  15. 15. Cottin D, Shillito B, Chertemps T, Tanguy A, Léger N, Ravaux J. Identification of differentially expressed genes in the hydrothermal vent shrimp Rimicaris exoculata exposed to heat stress. Mar Genomics. 2010; 3(2): 71–78. pmid:21798199
  16. 16. Mestre NC, Cottin D, Bettencourt R, Colaço A, Correia SP, Shillito B, et al. Is the deep-sea crab Chaceon affinis able to induce a thermal stress response? Comp Biochem Physiol A Mol Integr Physiol. 2015; 181: 54–61. pmid:25434602
  17. 17. Zhang J, Sun Q, Luan Z, Lian C, Sun L. Comparative transcriptome analysis of Rimicaris sp. reveals novel molecular features associated with survival in deep-sea hydrothermal vent. Sci Rep. 2017; 7: 2000. pmid:28515421
  18. 18. Wong YH, Sun J, He LS, Chen LG, Qiu JW, Qian PY. High-throughput transcriptome sequencing of the cold seep mussel Bathymodiolus platifrons. Sci Rep. 2015; 5: 16597. pmid:26593439
  19. 19. Zhang Y, Sun J, Chen C, Watanabe HK, Feng D, Zhang Y, et al. Adaptation and evolution of deep-sea scale worms (Annelida: Polynoidae): insights from transcriptome comparison with a shallow-water species. Sci Rep. 2017; 7: 46205. pmid:28397791
  20. 20. Fisher CR, Childress JJ, Arp AJ, Brooks JM, Distel D, Favuzzi JA, et al. Microhabitat variation in the hydrothermal vent mussel, Bathymodiolus thermophilus, at the Rose Garden vent on the Galapagos Rift. Deep Sea Res A. 1988; 35(10–11): 1769–1791.
  21. 21. Nakamura K, Kawagucci S, Kitada K, Kumagai H, Takai K, Okino K. Water column imaging with multibeam echo-sounding in the mid-Okinawa Trough: implications for distribution of deep-sea hydrothermal vent sites and the cause of acoustic water column anomaly. Geochem J. 2015; 49(6): 579–596.
  22. 22. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014; 30(15): 2114–2120. pmid:24695404
  23. 23. Haas BJ, Papanicolaou A, Yassour M, Grabherr M, Blood PD, Bowden J, et al. De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis. Nat Protoc. 2013; 8: 1494–1512. pmid:23845962
  24. 24. Simão FA, Waterhouse RM, Ioannidis P, Kriventseva EV, Zdobnov EM. BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics. 2015; 31(19): 3210–3212. pmid:26059717
  25. 25. Katoh K, Standley DM. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol Biol Evol. 2013; 30(4): 772–780. pmid:23329690
  26. 26. Capella-Gutiérrez S, Silla-Martínez JM, Gabaldón T. TrimAl: a tool for automated alignment trimming in large-scale phylogenetic analyses. Bioinformatics. 2009; 25(15): 1972–1973. pmid:19505945
  27. 27. Nguyen LT, Schmidt HA, von Haeseler A, Minh BQ. IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol Biol Evol. 2015; 32(1): 268–274. pmid:25371430
  28. 28. Quevillon E, Silventoinen V, Pillai S, Harte N, Mulder N, Apweiler R, et al. InterProScan: protein domains identifier. Nucleic Acids Res. 2005; 33(suppl_2): W116–W120.
  29. 29. 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(18): 3674–3676. pmid:16081474
  30. 30. Li B, Dewey CN. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics. 2011; 12: 323. pmid:21816040
  31. 31. Ye J, Fang L, Zheng H, Zhang Y, Chen J, Zhang Z, et al. WEGO: a web tool for plotting GO annotations. Nucleic Acids Res. 2006; 34(suppl_2): W293–W297.
  32. 32. Remm M, Storm CE, Sonnhammer EL. Automatic clustering of orthologs and in-paralogs from pairwise species comparisons. J Mol Biol. 2001; 314(5): 1041–1052. pmid:11743721
  33. 33. Kersey PJ, Allen JE, Allot A, Barba M, Boddu S, Bolt BJ, et al. Ensembl Genomes 2018: an integrated omics infrastructure for non-vertebrate species. Nucleic Acids Res. 2018; 46(D1): D802–D808. pmid:29092050
  34. 34. Zhang Z, Xiao J, Wu J, Zhang H, Liu G, Wang X, et al. ParaAT: a parallel tool for constructing multiple protein-coding DNA alignments. Biochem Biophys Res Commun. 2012; 419(4): 779–781. pmid:22390928
  35. 35. Wang D, Zhang Y, Zhang Z, Zhu J, Yu J. KaKs_Calculator 2.0: a toolkit incorporating gamma-series methods and sliding window strategies. Genomics Proteomics Bioinformatics. 2010; 8(1): 77–80. pmid:20451164
  36. 36. Chen LY, Zhao SY, Wang QF, Moody ML. Transcriptome sequencing of three Ranunculus species (Ranunculaceae) reveals candidate genes in adaptation from terrestrial to aquatic habitats. Sci Rep. 2015; 5: 10098. pmid:25993393
  37. 37. Swanson WJ, Wang A, Wolfner MF, Aquadro CF. Evolutionary expressed sequence tag analysis of Drosophila female reproductive tracts identifies genes subjected to positive selection. Genetics. 2004; 168(3): 1457–1465. pmid:15579698
  38. 38. Chen C, Chen H, He Y, Xia R. TBtools, a toolkit for biologists integrating various biological data handling tools with a user-friendly interface. bioRxiv. 2018; p.289660.
  39. 39. Madeira F, Park Y, Lee J, Buso N, Gur T, Madhusoodanan N, et al. The EMBL-EBI search and sequence analysis tools APIs in 2019. Nucleic Acids Res. 2019; 47(W1): W636–W641. pmid:30976793
  40. 40. Rastrick SP, Whiteley NM. Influence of natural thermal gradients on whole animal rates of protein synthesis in marine gammarid amphipods. PloS One. 2013; 8(3): e60050. pmid:23544122
  41. 41. Ravaux J, Gaill F, Bris NL, Sarradin PM, Jollivet D, Shillito B. Heat-shock response and temperature resistance in the deep-sea vent shrimp Rimicaris exoculata. J Exp Biol. 2003; 206: 2345–2354. pmid:12796451
  42. 42. Hutchison JS, Moldave K. The effect of elevated temperature on protein synthesis in cell-free extracts of cultured Chinese hamster ovary cells. Biochem Biophys Res Commun. 1981; 99(2): 722–728. pmid:7236297
  43. 43. Hawkins AJ, Day AJ. Metabolic interrelations underlying the physiological and evolutionary advantages of genetic diversity. Integr Comp Biol. 1999; 39(2): 401–411.
  44. 44. Tang B, Wang S, Wang SG, Wang HJ, Zhang JY, Cui SY. Invertebrate trehalose-6-phosphate synthase gene: genetic architecture, biochemistry, physiological function, and potential applications. Front Physiol. 2018; 9: 30. pmid:29445344
  45. 45. Virgilio CD, Hottiger T, Dominguez J, Boller T, Wiemken A. The role of trehalose synthesis for the acquisition of thermotolerance in yeast. I. Genetic evidence that trehalose is a thermoprotectant. FEBS J. 1994; 219(1–2): 179–186.
  46. 46. Paul S, Paul S. Molecular insights into the role of aqueous trehalose solution on temperature induced protein denaturation. J Phys Chem B. 2015; 119(4): 1598–1610. pmid:25558880
  47. 47. Bailly X, Vinogradov S. The sulfide binding function of annelid hemoglobins: relic of an old biosystem? J Inorg Biochem. 2005; 99(1): 142–150. pmid:15598498
  48. 48. Pradillon F, Gaill F. Adaptation to deep-sea hydrothermal vents: some molecular and developmental aspects. Journal of Marine Science and Technology. 2007; 15(15_S): 37–53.
  49. 49. Mann K, Mechling DE, Bächinger HP, Eckerskorn C, Gaill F, Timpl R. Glycosylated threonine but not 4-hydroxyproline dominates the triple helix stabilizing positions in the sequence of a hydrothermal vent worm cuticle collagen. J Mol Biol. 1996; 261(2): 255–266. pmid:8757292
  50. 50. Merzendorfer H, Zimoch L. Chitin metabolism in insects: structure, function and regulation of chitin synthases and chitinases. J Exp Biol. 2003; 206: 4393–4412. pmid:14610026
  51. 51. Dias RO, Cardoso C, Pimentel AC, Damasceno TF, Ferreira C, Terra WR. The roles of mucus‐forming mucins, peritrophins and peritrophins with mucin domains in the insect midgut. Insect Mol Biol. 2018; 27(1): 46–60. pmid:28833767
  52. 52. Sánchez-Salgado JL, Pereyra MA, Agundis C, Vivanco-Rojas O, Sierra-Castillo C, Alpuche-Osorno JJ, et al. Participation of lectins in crustacean immune system. Aquac Res. 2017; 48(8): 4001–4011.
  53. 53. Chu H, Mazmanian SK. Innate immune recognition of the microbiota promotes host-microbial symbiosis. Nat Immunol. 2013; 14: 668–675. pmid:23778794
  54. 54. Liu XL, Ye S, Cheng CY, Li HW, Lu B, Yang WJ, et al. Identification and characterization of a symbiotic agglutination-related C-type lectin from the hydrothermal vent shrimp Rimicaris exoculata. Fish Shellfish Immun. 2019; 92: 1–10.
  55. 55. Login FH, Balmand S, Vallier A, Vincent-Monégat C, Vigneron A, Weiss-Gayet M, et al. Antimicrobial peptides keep insect endosymbionts under control. 2011; Science. 334: 362–365. pmid:22021855
  56. 56. Piquet B, Shillito B, Lallier FH, Duperron S, Andersen AC. High rates of apoptosis visualized in the symbiont-bearing gills of deep-sea Bathymodiolus mussels. Plos One. 2019; 14(2): e0211499. pmid:30716127
  57. 57. Cerenius L, Söderhäll K. The prophenoloxidase-activating system in invertebrates. Immunol Rev. 2004; 198(1): 116–126.
  58. 58. Egger L, Schneider J, Rhême C, Tapernoux M, Häcki J, Borner C. Serine proteases mediate apoptosis-like cell death and phagocytosis under caspase-inhibiting conditions. Cell Death Differ. 2003; 10: 1188–1203. pmid:14502242
  59. 59. Wang X. Nutritional sources analysis and the heavy-metal enrichment of the macrofauna from deep-sea chemotrophic ecosystem. Ph.D. Thesis, Institute of Oceanology, Chinese Academy of Sciences. 2018.
  60. 60. Luther GW III, Rozan TF, Taillefert M, Nuzzio DB, Meo CD, Shank TM, et al. Chemical speciation drives hydrothermal vent ecology. Nature. 2001; 410: 813–816. pmid:11298448
  61. 61. Yahagi T, Watanabe H, Ishibashi J, Kojima S. Genetic population structure of four hydrothermal vent shrimp species (Alvinocarididae) in the Okinawa Trough, Northwest Pacific. Mar Ecol Prog Ser. 2015; 529: 159–169.
  62. 62. Sun S, Hui M, Wang M, Sha Z. The complete mitochondrial genome of the alvinocaridid shrimp Shinkaicaris leurokolos (Decapoda, Caridea): Insight into the mitochondrial genetic basis of deep-sea hydrothermal vent adaptation in the shrimp. Comp Biochem Phys D. 2018; 25: 42–52.
  63. 63. Seebacher F, Holmes S, Roosen NJ, Nouvian M, Wilson RS, Ward AJW. Capacity for thermal acclimation differs between populations and phylogenetic lineages within a species. Func Ecol. 2012; 26: 1418–1428.