INTRODUCTION
Alviniconcha spp. and Ifremeria nautilei Bouchet & Warén (family Provannidae) are large snails endemic to deep-sea hydrothermal vent fields in the Indo-Pacific Ocean (Sasaki et al., Reference Sasaki, Warén, Kano, Okutani, Fujikura and Kiel2010). Alviniconcha consists of a single described species Alviniconcha hessleri Okutani & Ohta and at least 4 more genetically distinct lineages (Kojima et al., Reference Kojima, Segawa, Fujiwara, Fujikura, Ohta and Hashimoto2001, Reference Kojima, Fujikura, Okutani and Hashimoto2004; Suzuki et al., Reference Suzuki, Kojima, Sasaki, Suzuki, Utsumi, Watanabe, Urakawa, Tsuchida, Nunoura, Hirayama, Takai, Nealson and Horikoshi2006a). Although Ifremeria consists of a single species, I. nautilei, its population in the Manus Basin of Papua New Guinea is genetically distinct from those in the North Fiji and Lau basins of the south-west Pacific Ocean (Kojima et al., Reference Kojima, Segawa, Fujiwara, Hashimoto and Ohta2000; Suzuki et al., Reference Suzuki, Kojima, Watanabe, Suzuki, Tsuchida, Nunoura, Hirayama, Takai, Nealson and Horikoshi2006b). Johnson et al. (Reference Johnson, Warén, Lee, Kano, Kaim, Davis, Strong and Vrijenhoek2010) analysed the phylogenetic relationships among provannids and related species by comparing the nucleotide sequences of 3 mitochondrial and 3 nuclear genes. Their findings showed that Provannidae, which includes the genera Provanna, Desbruyeresia, Alviniconcha and Ifremeria, is paraphyletic and is derived from a monophyletic group that includes 2 other genera, Abyssochrysos and Rubyspira. This analysis further revealed a sister relationship between Alviniconcha and Ifremeria.
Recently, sulphide deposits associated with deep-sea hydrothermal vents have attracted public attention as a mineral resource, and full-scale commercial mining is now planned in the hydrothermal vent fields of the south-west Pacific where Alviniconcha and Ifremeria snails dominate (Van Dover, Reference Van Dover2011). The genetic diversity of many other hydrothermal vent-endemic species, which are expected to provide genetic resources, is also likely to be affected by commercial mining (Arrieta et al., Reference Arrieta, Arnaud-Haond and Duarte2010). To minimize the potential impact of mining activities on the genetic diversity of these species, it is necessary to establish protected area(s) based on genetic connectivity among local populations (Van Dover, Reference Van Dover2011). Unfortunately, such baseline studies are seldom conducted by trained molecular ecologists and are often lacking in time and budget. Mitochondrial DNA markers such as the cytochrome c oxidase I (COI) gene fragment are therefore adequate for this purpose because these nucleotide sequences can be determined relatively easily by using direct-sequencing techniques without the need for molecular cloning. However, hydrothermal vent fields are unstable and relatively ephemeral environments that may change drastically within relatively short time-frames (Vrijenhoek, Reference Vrijenhoek2010). The present hydrothermal activities of back-arc basins in the south-west Pacific are estimated to have started between 4 and 8 million years ago (Tamaki & Honza, Reference Tamaki and Honza1991) and have been subjected to large changes such as explosions, reductions in activity, and chemical alterations in hydrothermal fluid (Nojiri et al., Reference Nojiri, Ishibashi, Kawai, Otsuki and Sakai1989; Gamo et al., Reference Gamo, Sakai, Ishibashi, Nakayama, Isshiki, Matsuura, Shitashima, Takeuchi and Ohta1993; Ishibashi et al., Reference Ishibashi, Luption, Yamaguchi, Querellou, Nunoura and Takai2006). On the other hand, the general evolutionary rate of the protostomia COI gene is 0.01 per million years (Wilke, Reference Wilke2003), which suggests that mutations are expected to occur at approximately 5 sites in 500-(base pairs)bp COI sequences, which can be analysed by a single round of sequencing, in a million years. Thus, due to the relatively low mutation rate, this gene may not be a suitable choice for determining the population history of vent animals, which is closely associated with hydrothermal environments. Alternative mitochondrial markers with higher evolutionary rates should therefore be sought for this purpose. Although large non-coding regions such as those found in the control regions of vertebrates have not been found in the mitochondrial genomes of gastropods, pseudogenes formed by the rearrangement of the mitochondrial genome are good candidates for such DNA markers (Rawlings et al., Reference Rawlings, MacInnis, Bieler, Boore and Collins2010).
Although the gene order in metazoan mitochondrial genomes is generally conserved (Boore & Brown, Reference Boore and Brown1998), changes have nonetheless frequently occurred in some animal groups such as Platyhelminthes, Brachiopod and Gastropod (Le et al., Reference Le, Blair, Agatsuma, Humair, Campbell, Iwagami, Littlewood, Peacock, Johnston, Bartley, Rollinson, Herniou, Zarlenga and McManus2000; Endo et al., Reference Endo, Noguchi, Ueshima and Jacobs2005; Grande et al., Reference Grande, Templado and Zardoya2008). In the current study, we have found evince of genome rearrangement and the resultant non-coding regions in the mitochondrial genomes of Alviniconcha and Ifremeria gastropods and showed the potential utility of these regions as DNA markers for phylogeographical studies.
MATERIALS AND METHODS
Table 1 summarizes the specimens used in this study: a Provanna glabra Okutani, Tsuchida & Fujikura specimen was collected from a seep area off Hatsushima Island, Sagami Bay during dive 306 of the remotely operating vessel (ROV) ‘Hyper-Dolphin’ of the Japan Agency for Marine–Earth Science and Technology (JAMSTEC); a P. aff. glabra 1 specimen was collected from the North Iheya Knoll, Okinawa Trough, during dive 392 of the ROV ‘Hyper-Dolphin’ of JAMSTEC; a P. aff. glabra 2 specimen was collected from the Hatoma Knoll, Okinawa Trough, during dive 1352 of the submersible ‘Shinkai2000’ of JAMSTEC; a Desbruyeresia cancellata Warén & Bouchet specimen was collected from the Lau Basin during dive 846 of the submersible ‘Shinkai6500’ of JAMSTEC; a Desbruyeresia sp. specimen, collected from the Bayonnaise Knoll, Izu-Ogasawara Arc during dive 1153 of the submersible ‘Shinkai6500’ of JAMSTEC, was provided by the Japan Oil, Gas and Metals National Corporation (JOGMEC); a specimen of Abyssochrysos melanioides Tomlin was obtained from the Nankai Trough, off Kochi, during the KT-11-12 cruise of the research vessel ‘Tansei-maru’ of the Atmosphere and Ocean Research Institute (AORI) of the University of Tokyo; and an Abyssochrysos sp. (yellow-type) specimen was obtained from the East China Sea (near Amami Oshima Island) during the N275 cruise of the training ship ‘Nagasaki-maru’ of Nagasaki University. Voucher specimens are stored at JAMSTEC and AORI.
List of specimens used in the present study, along with the primers used for polymerase chain reaction (PCR).
Either mitochondrial DNA (mtDNA) or total genomic DNA was extracted from Alviniconcha spp. and I. nautilei in previous studies (Kojima et al., Reference Kojima, Segawa, Fujiwara, Hashimoto and Ohta2000, Reference Kojima, Segawa, Fujiwara, Fujikura, Ohta and Hashimoto2001, Reference Kojima, Fujikura, Okutani and Hashimoto2004; Suzuki et al., Reference Suzuki, Kojima, Sasaki, Suzuki, Utsumi, Watanabe, Urakawa, Tsuchida, Nunoura, Hirayama, Takai, Nealson and Horikoshi2006a, Reference Suzuki, Kojima, Watanabe, Suzuki, Tsuchida, Nunoura, Hirayama, Takai, Nealson and Horikoshib). Total genomic DNA was extracted from the foot region of 7 other specimens by using a DNeasy Tissue Extraction Kit (Qiagen, Valencia, CA).
GeneReleaser (BioVenture Inc., Murfreesboro, TN) was used to sequester the products of cell lysis that might have inhibited polymerase activity. Fragments (~6500 bp) of mtDNA were amplified by polymerase chain reaction (PCR) with the extracted DNA as a template, Takara Ex Taq Hot Start Version (Takara Bio Inc., Shiga, Japan), and the primers LCO1490 (Folmer et al., Reference Folmer, Black, Hoeh, Lutz and Vrijenhoek1994) and 16S-IV (Kojima, Reference Kojima2010). The PCR conditions were as follows: incubation at 94°C for 60 seconds, followed by 40 cycles of incubation at 92°C, 50°C, and 72°C for 40 seconds, 5 minutes, and 120 seconds, respectively. Shorter mtDNA fragments (~2000–3000 bp) were amplified by PCR with the extracted DNA as a template, Takara Ex Taq Hot Start Version (Takara Bio Inc.), and the primers listed in Table 1. The PCR conditions for this amplification were as follows: incubation at 94°C for 60 seconds, followed by 40 cycles of incubation at 92°C, 42–50°C and 72°C for 40 seconds, 1 minute, and 90 seconds, respectively.
Each PCR product was incubated with EXOSAP-IT (United States Biochemical, Cleveland, OH) to digest the unused primers and nucleotides. Each treated PCR product was used in cycle sequencing reactions performed using a BigDye Terminator Cycle Sequencing Kit (Version 3.0; Applied Biosystems Inc., Foster City, CA). The PCR products were sequenced using an automated DNA sequencer (ABI 3100; Applied Biosystems Inc., Foster City, CA). The nucleotide sequences of both ends of the gene fragments were determined by cycle sequencing with the same primers that were used for PCR. Long fragments were comprehensively sequenced by primer walking based on the determined sequences. The nucleotide sequences of the primers used in the current study are listed in Table 2.
Primers used in the present study.
Gene annotation was performed using FASTA (Pearson & Lipman, Reference Pearson and Lipman1988) comparisons against published gastropod mtDNA sequences. Protein-coding genes were identified by inference with the open reading frames, comparing deduced amino acid sequences with published gastropod sequences, and by delimiting the start and stop codons. Genes for tRNAs were identified on the basis of their corresponding anticodons and typical cloverleaf structures, using the search program tRNAscan-SE Search 1.21 (Lowe & Eddy, Reference Lowe and Eddy1997) or by performing a manual search. Secondary structures of tRNA genes were manually reconstructed. Because the exact boundaries of the rRNA genes could not be determined, it was tentatively assumed that no intergenic regions were present between rRNA genes and the adjacent genes. The nucleotide sequences determined in the current study were deposited in the DDBJ/EMBL/GenBank databases under the accession numbers AB749232–749271.
For phylogenetic analysis, nucleotide sequences of only 7 tRNA genes were used because it was difficult to align sequences of the non-coding regions between these genes. Littorina saxatilis (Willing et al., Reference Willing, Mill and Grahame1999) was used as an outgroup due to the suggested close relationship between Littorinidae and Provannidae (Ponder et al., Reference Ponder, Colgan, Healy, Nützel, Simone, Strong, Ponder and Lindberg2008). Sequences were aligned using ProAlign ver. 0.5 (Löytynoja & Milinkovitch, Reference Löytynoja and Milinkovitch2003), and only sites with posterior probability scores greater than 50% were used for phylogenetic analysis. A phylogenetic tree was constructed using the Bayesian method with the program MrBayes ver. 3.1.2 (Ronquist & Huelsenbeck, Reference Ronquist and Huelsenbeck2003). The general time-reversible model was used with invariant site–frequency and gamma-shape parameters estimated from the data (GTR + I + G). All the data were treated as a single partition. Two parallel runs were made for 3,000,000 generations (with a sample frequency of 1000) by using the default value of 4 Markov chains. The first 1500 trees for each run were discarded to ensure that the 4 chains reached stationarity. The consensus tree and posterior probabilities were computed from the remaining 3000 trees (1500 trees × 2 runs).
A phylogenetic network was constructed on the basis of differences in the nucleotide sequences by the median-joining method by using Network version 4.6.1.0 (Bandelt et al., Reference Bandelt, Forster and Röhl1999). The genetic diversity of populations was estimated on the basis of both haplotype (gene) diversity (h), which is the probability that 2 randomly chosen haplotypes are different (Nei, Reference Nei1987), and nucleotide diversity (π), which is the probability that 2 randomly chosen homologous nucleotides are different (Tajima, Reference Tajima1983; Nei, Reference Nei1987), by using the Arlequin version 3.5.1.2 software package (Excoffier & Lischer, Reference Excoffier and Lischer2010).
RESULTS
We first determined the nucleotide sequences of mtDNA fragments that were flanked by genes for COI and the large subunit of rRNA (16S rRNA) in Alviniconcha hessleri, obtained from the Mariana Trough, and in Alviniconcha sp. 1 and I. nautile, obtained from the Manus Basin (Table 1). Each of these fragments contained genes for cytochrome c oxidase subunits I and II, ATPase subunits 6 and 8, large and small subunits of rRNA, and 9 tRNAs. Comparison of the gene order of these species with that of typical caenogastropods (Cunha et al., Reference Cunha, Grande and Zardoya2009) showed that in all 3 species, the tRNATrp gene had translocated from between the tRNACys and tRNAGln genes to between the tRNAMet and tRNATyr genes (Figure 1).
Partial mitochondrial genomes of provannid gastropods and related species. The numbers denote the lengths (base pairs) of the non-coding regions between neighbouring genes; the numbers in parentheses are the number of overlapping base pairs between neighbouring genes.
To identify the time point at which this translocation occurred, we amplified shorter mtDNA fragments and determined the nucleotide sequences of fragments containing 7 tRNA genes in the following genetically distinct representatives of the 2 genera: Alviniconcha sp. 2, Alviniconcha sp. 3, Alviniconcha sp. 4, I. nautilei from the North Fiji Basin, and I. nautilei from the Lau Basin, as well as 7 related species. These related species were P. glabra, P. aff. glabra 1, P. aff. glabra 2, D. cancellata, Desbruyeresia sp., A. melanioides and Abyssochrysos sp (Figure 1). All the taxa of the genera Alviniconcha and Ifremeria showed translocation of the tRNATrp gene, whereas sequences obtained from the specimens of the other genera showed the gene order typical of caenogastropods (Cunha et al., Reference Cunha, Grande and Zardoya2009).
Relatively long non-coding regions were observed on both sides of the tRNATrp gene and at the original position of this gene between the tRNACys and tRNAGln genes in all the Alviniconcha and Ifremeria gastropods (Figure 1). These non-coding regions were generally shorter in Alviniconcha spp. than in Ifremeria gastropods (Figure 1).
Phylogenetic relationships among provannids and related species were analysed by the Bayesian method (Figure 2). Each of the 5 genera formed a monophyletic group supported by the highest posterior possibility (100%). Alviniconcha and Ifremeria formed a robust clade (100%), which comprised a moderately supported branch with Desbruyeresia and Abyssochrysos (87%). Although the phylogenetic relationships among the Alviniconcha + Ifremeria clade, Desbruyeresia, and Abyssochrysos could not be determined with any certainty (52%), they are identical to those based on the nucleotide sequences of 3 mitochondrial and 3 nuclear genes (Johnson et al., Reference Johnson, Warén, Lee, Kano, Kaim, Davis, Strong and Vrijenhoek2010).
Bayesian phylogenetic tree of provannid gastropods and related species based on the nucleotide sequences of 7 mitochondrial tRNA genes. Posterior probabilities greater than 70% are shown above or below the branches.
Furthermore, the region between the tRNAMet and 12S rDNA genes (631–632 bp) was sequenced in 40 A. hessleri specimens, for which partial nucleotide sequences (696 bp) of the mitochondrial COI gene have been previously determined (Kojima et al., Reference Kojima, Segawa, Fujiwara, Fujikura, Ohta and Hashimoto2001). Genetic diversity was estimated (Table 3) and a phylogenetic network was constructed (Figure 3) on the basis of the nucleotide sequences of each of the COI genes (Kojima et al., Reference Kojima, Segawa, Fujiwara, Fujikura, Ohta and Hashimoto2001) and the region between the tRNAMet and 12S rRNA genes. Although 2 genetically distinct groups were recognized in the haplotype network based on the latter data set, such a genetic structure was not recognized in the haplotype network based on the COI dataset (Figure 3).
Median-joining networks of the haplotypes of Alviniconcha hessleri based on: (A) nucleotide sequences of the gene for cytochrome c oxidase subunits I; (B) the region between the tRNAMET and 12S rRNA genes. The haplotypes that were not detected in the sample are indicated by closed circles. The areas of the open circles are proportional to the frequency of the occurrence of the haplotypes.
Genetic diversity of an Alviniconcha hessleri population based on nucleotide sequences of the gene for cytochrome c oxidase subunit I (COI) and the region between the tRNAMet and 12S rRNA genes (tRNAs–12SrDNA).
DISCUSSION
Gene order changes occur relatively frequently in gastropods (Grande et al., Reference Grande, Templado and Zardoya2008). However, in caenogastropods, the largest gastropod group, rearrangement of the mitochondrial genome has only been reported in the families: Conidae; Turridae; Potamididae; Batillariidae; and Vermetidae (Cunha et al., Reference Cunha, Grande and Zardoya2009; Kojima, Reference Kojima2010; Rawlings et al., Reference Rawlings, MacInnis, Bieler, Boore and Collins2010). To our knowledge, this is the first report of mitochondrial genome rearrangement in the caenogastropod family Provannidae.
Because the genera Alviniconcha and Ifremeria form a monophyletic group (Johnson et al., Reference Johnson, Warén, Lee, Kano, Kaim, Davis, Strong and Vrijenhoek2010; Figure 2), the translocation of the tRNATrp gene is thought to have occurred within the common ancestor of these 2 genera. Although a specimen of the genus Rubyspira was not available for the current study, its gene order is expected to be the same as that of typical caenogastropods because this genus has been shown to form a monophyletic group with the genera Desbruyeresia and Abyssochrysos (Johnson et al., Reference Johnson, Warén, Lee, Kano, Kaim, Davis, Strong and Vrijenhoek2010).
Relatively long non-coding regions were observed on both sides of the tRNATrp gene and at the original position of this gene in all of the Alviniconcha and Ifremeria gastropod specimens (Figure 1). Such regions are thought to be remnants of duplicated genes that became pseudogenes (Boore & Brown, Reference Boore and Brown1998), although it was difficult to align sequences of the non-coding regions and tRNA genes.
The current study shows that mitochondrial non-coding regions provide much more information about population genetic structures of Alviniconcha hessleri than dose the COI gene (Table 3). Indeed, genetic subdivision of the population of A. hessleri is only detectable by using the non-coding regions (Figure 3).
An accurate evaluation of the spatial distribution of the genetic diversity and the extent of genetic connectivity between local populations is crucial in order to prevent the loss of genetic diversity of species endemic to hydrothermal vents as a consequence of sulphide-deposits mining. For this purpose, DNA markers with high evolutionary rates which can be used relatively easily by non-specialist researchers are particularly useful. Based on these criteria, the new DNA marker presented in this study is expected to enable more detailed population structure analyses of Alviniconcha and Ifremeria snails, which are representatives of the vent fauna in the south-west Pacific where full-scale commercial mining has been planned.
ACKNOWLEDGEMENTS
We thank Drs Tadashi Maruyama, Katsunori Fujikura, Ken Takai and Takuro Nunoura and the captains, officers and crews of the research vessels ‘Yokosuka’ and ‘Natsushima’ of the Japan Agency for Marine–Earth Science and Technology (JAMSTEC), as well as the operation teams of the submersibles ‘Shinkai2000’ and ‘Shinkai6500’ and the ROV ‘Hyper-Dolphin’ of JAMSTEC for their support in obtaining Provanna and Desbruyeresia specimens. The Japan Oil, Gas and Metals National Corporation (JOGMEC) provided a specimen from the Bayonnaise Knoll. We especially thank Dr Jun Hashimoto and the crew of the training ship ‘Nagasaki-maru’ of Nagasaki University and the crew of the research vessel ‘Tansei-maru’ of the Atmosphere and Ocean Research Institute (AORI), University of Tokyo, for their help in obtaining Abyssochrysos specimens. This work was conducted in the scientific research innovative areas ‘trans-crustal advection & in-situ bio-geochemical processes of global sub-seafloor aquifer (TAIGA)’, supported by the Ministry of Education, Culture, Sports, Science and Technology (grant number 20109004).
