INTRODUCTION

Many hydrothermal vents with associated chemoautosynthesis-based faunal communities have been discovered in the south Western Pacific back-arc basins, namely, the Mariana Trough (Craig et al., 1987; Hessler and Lonsdale, 1991), the Manus Basin (Both et al., 1986; Tufar, 1990; Auzende et al., 1997), the North Fiji Basin (KAIYO 87 Shipboard Party, 1988; Desbruyères et al., 1994) and the Lau Basin (NAUTILAU Group, 1990; Desbruyères et al., 1994). The chemoautosynthesis-based faunal communities are composed of many endemic groups, such as vestimentiferan tube worms, large bivalves of the genera Calyptogena and Bathymodiolus, and crabs of the genus Bythograea (Tunnicliffe et al., 1998). Within such communities, those in the south Western Pacific are unique with respect to species composition (Tunnicliffe and Fowler, 1996). In the vent areas in the south Western Pacific, the communities are dominated by large endemic gastropods (Both et al., 1986; Okutani and Ohta, 1988), which live in symbiosis with chemosynthetic bacteria (Stein et al., 1988; Endow and Ohta, 1989; Windoffer and Giere, 1997). Furthermore, bivalves of the genus Calyptogena, which form a dominant group in the communities in both the Eastern Pacific and the north Western Pacific, have not been found in the south Western Pacific, with the exception of an undescribed species collected in the Desmos Cauldron of the Manus Basin (Ohta et al., 1997) and the New Ireland Basin (Herzig et al., 1994).

The dominant gastropods in the vent areas in the south Western Pacific have been classified into two genera in the family Provannidae, namely Alviniconcha and Ifremeria. Although Alviniconcha gastropods have been reported from the Mariana Trough, the Manus Basin, the North Fiji Basin and the Lau Basin, Ifremeria gastropods have not been discovered in the hydrothermal vent fields in the Mariana Trough (Bouchet and Warén, 1991; Warén and Bouchet, 1993). The absence of Ifremeria gastropods in the Mariana Trough might be attributable to lower larval dispersal ability of Ifremeria than Alviniconcha. The morphology of the larval shell suggests that Alviniconcha gastropods undergo planktotrophic development (Warén and Bouchet, 1993). The high-dispersal mode of development is thought to have been resulted in a wide geographical distribution of Alviniconcha gastropods. The degree of genetic differentiation between conspecific populations is expected to be closely related to the dispersal ability of the species. As expected from the high dispersal ability of the genus Alviniconcha, the absence of genetic differentiation between populations in the North Fiji Basin and the Manus Basin of an undescribed species in the genus Alviniconcha was reported (Kojima et al., 1998).

The larval type of Ifremeria gastropods remains unknown because of corrosion of the protoconch of all available specimens (Warén and Bouchet, 1993). In addition, no analysis of the genetic differentiation between populations of Ifremeria gastropods has been reported. In the present study, we analyzed the genetic differentiation of populations of I. nautilei between the Manus Basin and the North Fiji Basin (Fig. 1) by examining nucleotide sequences of mitochondrial DNA in an effort to characterize the larval dispersal ability of Ifremeria.

Fig. 1

Location of sites at which samples of Ifremeria nautilei were collected: 1, Manus Basin; 2, North Fiji Basin. An open circle denotes the location of the Lau Basin.

MATERIALS AND METHODS

During dives of the submersible “Nautile” of Institut français de recherche pour l'exploitation de la mer (IFREMER) and of the submersibles “Shinkai 2000” and “Shinkai 6500” of the Japan Marine Science and Technology Center (JAMSTEC), a total of 52 specimens of Ifremeria nautilei was collected, as summarized in Table 1. One specimen of Alviniconcha hessleri, collected at the Forecast Vent site on the South Mariana Ridge (1470 m depth) during a dive of “Shinkai 6500” (No. D186), was used as the outgroup for phylogenetic analysis. Warén and Bouchet (1993) examined young specimens and the anatomy of A. hessleri and I. nautilei, and they stated that the genera Alviniconcha and Ifremeria are closely related and that these genera should be grouped into the family Provannidae. Beck (1991) also stated that Alviniconcha and Olgaconcha (= Ifremeria) are closely related to one another.

Table 1

List of samples.

Mitochondrial DNA (mtDNA) was extracted from the deep-frozen head-foot region of each individual by a modified version of the method described by Komm et al. (1982). In the case of some damaged samples, which had been collected in the North Fiji Basin in 1990 and stored at −20°C, total DNA was extracted by grinding of samples, digestion with sodium dodecyl sulfate (SDS) and proteinase K, and extraction with phenol and chloroform. In the case of samples collected in the Manus Basin in 1998, total DNA was isolated from samples without digestion with proteinase K.

A fragment (about 450 bp) of the mitochondrial gene for cytochrome oxidase I (COI) was amplified by the polymerase chain reaction (PCR) with “universal” primers for amplification of most genes for metazoan COI, COI-3 and COI-6 (Shimayama et al., 1990). The conditions for PCR were as follows: 94°C for 60 s; then 30 to 40 cycles at 92°C for 40 s, 40°C for 60 s, and 72°C for 90 s. Genereleaser™ (BioVenture Inc., Murfreesboro, TN, USA) was used to sequester products of cell lysis that might have inhibited the polymerase. The nucleotide sequences of fragments (306 bp) were determined by the dideoxynucleotide chain-termination method using a Sequenase™ PCR product sequencing kit (United States Biochemical Coop., Cleveland, OH, USA). A longer fragment (about 960 bp) of genes for COI was amplified by PCR using primers COI-B (Hasegawa et al., 1996) and COI-6. The conditions for PCR were as follows: 94°C for 60 s; then 30 to 40 cycles at 92°C for 40 s, 50°C for 60 s, and 72°C for 90 s. Nucleotide sequences within the upper region of this fragment were determined with a sequencer (DSQ-2000L; Shimazu Corp., Kyoto, Japan) using primers TW-2 (Kojima et al., 1997) and Gastro-3. The sequence of primer Gastro-3 was based on sequences from some specimens of I. nautilei that had been determined using primer COI-B (data not shown). Nucleotide sequences of primers used in the present study are summarized in Table 2. Amino acid sequences of COI were deduced by reference to the modified genetic code of molluscan mtDNA (Shimayama et al., 1990; Hoffmann et al., 1992).

Table 2

Primers used in the present study. Y, R, S and W denote T or C, A or G, G or C, and A or T, respectively. Positions refer to the corresponding amino acid residues encoded by the gene for mitochondrial cytochrome oxidase I from Drosophila yakuba.

The genetic distance between sequences was calculated by Kimura's two-parameter method (Kimura, 1980). Phylogenetic trees were constructed by the neighbour-joining method (Saitoh and Nei, 1987) with the program from the MEGA package, Version 1.0 (Kumar et al., 1993) and by the maximum parsimony method with the computer program Parsimony, which was provided by Dr. K. Tamura (Tokyo Metropolitan University). Number of net nucleotide substitution between populations (Nei, 1987) was estimated on the basis of numbers of nucleotide substitution between haplotypes calculated by Kimura's two-parameter method (Kimura, 1980). Genetic diversity of populations was estimated in terms of two indices: gene diversity (Nei, 1987) and nucleotide diversity (Tajima 1983; Nei, 1987). Estimations were made with the computer program Arlequin (Schneider et al., 1996). Gene diversity is the probability that two randomly chosen haplotypes are different. Nucleotide diversity is the probability that two randomly chosen homologous nucleotides are different.

RESULTS

Partial sequences (792 bp) of mitochondrial genes for COI were determined for 52 specimens of Ifremeria nautilei and one specimen of Alviniconcha hessleri (Figs. 2 and 3). Two intraspecific amino acid substitutions were detected in the sequences from the specimens of I. nautilei. Each amino acid substitution was detected in the case of only one specimen collected in the Manus Basin. Six amino acid substitutions were detected when we compared the dominant sequence from I. nautilei and the sequence from A. hessleri (Fig. 2).

Fig. 2

Nucleotide sequences of mitochondrial genes for cytochrome oxidase I from Ifremeria nautilei (I) and Alviniconcha hessleri (A). Asterisks indicate nucleotides that are identical to those in I. Numbers above the sequence from I. nautilei denote positions at which intraspecific variations were observed. Y, R, S and W denote T or C, A or G, G or C, and A or T, respectively. Underlining indicates sites at which amino acid substitutions were identified. Positions of primers COI-3 and TW-2 are shown. The nucleotide sequences will appear in the GSDB, DDBJ, EMBL and NCBI nucleotide sequence databases under accession numbers AB010097 (I) and AB010094 (A).

Fig. 3

Nucleotides at variable positions in the sequence of the gene for COI from Ifremeria nautilei. Numbers refer to the positions shown in Fig. 2. Asterisks indicate nucleotides that are identical to those in the uppermost sequences. M and F, as prefixes in names of haplotypes, denote haplotypes obtained from specimens from the Manus Basin and those from the North Fiji Basin, respectively.

Within the sequences of I. nautilei, 58 sites were found to be polymorphic and 42 haplotypes were obtained from 52 individuals (Fig. 3). Between the population of I. nautilei in the North Fiji Basin and that in the Manus Basin, five nucleotide substitutions were fixed (Fig. 3) and the two populations had no common haplotypes.

The phylogenetic relationships among the deduced haplotypes of I. nautilei were analyzed by the neighbour-joining (NJ) method and the maximum parsimony (MP) method. Individuals collected in both basins formed reciprocally monophyletic clusters on the NJ tree (Fig. 4). While monophyly of haplotypes from the North Fiji Basin was supported by a high bootstrap probability (88%), that of haplotypes from the Manus Basin was supported by a very low bootstrap probability (26%). On the MP tree (Fig. 5), haplotypes from the North Fiji Basin formed a monophyletic cluster supported by a high bootstrap probability (82%) and a bootstrap probability supporting the monophyly of haplotypes from the Manus Basin was low (45%).

Fig. 4

Phylogenetic tree for haplotypes of Ifremeria nautilei. The tree was constructed by the neighbour-joining method with Alviniconcha hessleri as the outgroup. Bootstrap values are shown below branches of clades that are supported by bootstrap values of more than 50%.

Fig. 5

A 50% majority rule consensus phylogenetic tree for haplotypes of Ifremeria nautilei. The tree was constructed by the maximum parsimony method with Alviniconcha hessleri as the outgroup. Bootstrap values are shown above branches of clades that are supported by bootstrap values of more than 50%.

The number of net nucleotide substitution between two populations was estimated to be 0.0082. The genetic diversity of the two populations is summarized in Table 3. While gene diversity values were almost the same in the Manus Basin and the North Fiji Basin, the genetic diversity in the Manus Basin, as estimated in terms of nucleotide diversity, was about twice that in the North Fiji Basin. These results mean that diversity of haplotypes was the same in both populations but the average extent of genetic divergence among individuals was much greater in the Manus Basin than in the North Fiji Basin.

Table 3

Genetic diversity of populations of Ifremeria nautilei, as estimated from two indices, with standard deviations.

DISCUSSION

In the present study, we analyzed the genetic differentiation between the Manus Basin and the North Fiji Basin of populations of Ifremeria nautilei, one of the dominant species in the hydrothermal areas in the south Western Pacific. We found that these populations had no common haplotypes and that five nucleotide substitutions were fixed between them (Figs. 2 and 3). Thus, they appeared to be geographically isolated from one another.

For 20 and 12 individuals of an undescribed species in the genus Alviniconcha in the Manus Basin and the North Fiji Basin, respectively, Kojima et al. (1998) determined nucleotide sequences of the same region of the mitochondrial gene for COI as the region analyzed for I. nautilei in the present study. Both the exact test of population differentiation (Raymond and Rousset, 1995) and the randomized chisquared test of independence (Roff and Bentzen, 1989) showed the absence of genetic differentiation between populations of Alviniconcha sp. in the Manus Basin and the North Fiji Basin (Kojima et al., 1998).

The undescribed species in the genus Alviniconcha and I. nautilei live close to the same hydrothermal vents in the both basins. Thus, the present results suggest that the larval dispersal ability of I. nautilei might be lower than that of Alviniconcha sp. While Alviniconcha gastropods were shown to undergo planktotrophic development, the larval type of Ifremeria gastropods remains unknown (Warén and Bouchet, 1993). To determine the larval type of Ifremeria gastropods, it is necessary to collect larvae or juveniles at early development stages for direct observations using a submersible with a sampling system such as a slurpgun. The analysis of DNA sequences should be very useful for identification of such samples (Dixon et al., 1995).

On the neighbour-joining tree (Fig. 4) and the maximum parsimony tree (Fig. 5), all haplotypes of specimens of I. nautilei obtained from the North Fiji Basin formed a monophyletic cluster that was supported by high bootstrap probabilities. The cluster from the Manus Basin was supported only by very low bootstrap probabilities and, therefore, the monophyly of individuals from the Manus Basin is open to question. They might be paraphyletic with exclusion of individuals from the North Fiji Basin.

The genetic diversity of the population of I. nautilei in the North Fiji Basin was much smaller than that of the population in the Manus Basin. The limited genetic divergence and the clear monophyly of haplotypes from the North Fiji Basin can be explained if we postulate that I. nautilei in the North Fiji Basin is derived from a relatively recent dispersion from another basin. If the population in the North Fiji Basin was founded by a few immigrants and if the time since their arrival has been insufficient for creation of high genetic diversity via the introduction of genetic mutations, the genetic diversity of this population would be expected to remain at a low level. Since the environments around hydrothermal vents are unstable, it is also likely that large fluctuations in population size have occurred very frequently. If the population of I. nautilei in the North Fiji Basin were to have recently experienced a severe decrease, such a decrease might also have reduced the genetic diversity of the population. As I. nautilei and Alviniconcha sp. inhabit the same vent areas in the both basins, the genetic diversity of the population of Alviniconcha sp. in the North Fiji Basin should be lower than that in the Manus Basin, if low genetic diversity of the population of I. nautilei in the North Fiji Basin is attributed to the bottleneck effect caused by the change of the environments in the North Fiji Basin. Between the population of Alviniconcha sp. in the Manus Basin and that in the North Fiji Basin, little difference in the genetic diversity was shown in terms of both gene diversity (0.86±0.06 for the Manus Basin and 0.85±0.10 for the North Fiji Basin) and nucleotide diversity (0.0035±0.0022 for the Manus Basin and 0.0030±0.0020 for the North Fiji Basin) (unpublished data). The decrease of population size in the past seems unlikely to explain the low genetic diversity of the population of I. nautilei in the North Fiji Basin.

The hydrothermal areas in the Manus Basin is the westernmost habitat of I. nautilei and the nearest habitat is situated in the North Fiji Basin. The St. Georges Undercurrent and the Equatorial Undercurrent flow from the Manus Basin to the North Fiji Basin (Butt and Lindstrom, 1994). If individuals of I. nautilei in the Manus Basin are paraphyletic, with exclusion of individuals in the North Fiji Basin, it seems likely that the population in the North Fiji Basin was founded by immigrants from the Manus Basin and that the two populations have not yet become reciprocally monophyletic, even though each population has been accumulating genetic differences. If the rate of sequence divergence of the COI gene of snapping shrimps, namely 0.014 per million years (Knowlton and Weigt, 1998) is used, we can estimate that the divergence between the population in the North Fiji Basin and that in the Manus Basin occurred 0.59 million years ago. Such a time frame suggests that the recruitment of I. nautilei to the North Fiji Basin was a relatively recent event.