Genetic Relationships among Species of Meretrix (Mollusca: Veneridae) in the Western Pacific Ocean1 Ayako Yashiki Yamakawa,2,3,6 Masashi Yamaguchi,4,5 and Hideyuki Imai4 Abstract: We compared allozymes at 12 loci in 12 populations of six species of Meretrix: M. lusoria ( Japan, Korea, and Taiwan), M. petechialis (China and Ko- rea), M. ovum (Thailand and Mozambique), M. lyrata (China), M. lamarckii ( Ja- pan), and Meretrix sp. A (Okinawa, Japan). Our allozyme results were generally consistent with the major groupings currently recognized within the genus based on morphological characters. However, we found two cryptic or un- described species: Meretrix sp. A from Okinawa and M. cf. lusoria from Taiwan. The shell characters of Meretrix sp. A were similar to those of M. lamarckii, but the species was genetically distinct (Nei’s genetic distance D > 0.845) from all other species examined. The Taiwanese Meretrix population was morphologi- cally indistinguishable from Japanese M. lusoria, although the genetic distance between the Taiwanese and Japanese populations showed a high degree of ge- netic differentiation (D > 0.386). Meretrix lusoria seedlings were introduced into Taiwan from Japan in the 1920s, and Japanese M. lusoria was previously thought to be established as a cultured stock. However, our results suggest that the Taiwanese population may represent a sibling or cryptic species of M. lusoria. Asianhardclams, genus Meretrix (Vener- (Yoosukh and Matsukuma 2001). These idae), are commercially important bivalves clams inhabit the tidal flats, estuaries, and in East and Southeast Asia and East Africa sandy beaches of the Indian Ocean, including East Africa and Southeast Asia, and the west- ern Pacific along the Chinese coast, Korean 1 Financial support was provided from the 21st Peninsula, and Japanese Archipelago. The Century Center of Excellence (COE) Program of the University of the Ryukyus; the TaKaRa Harmonist genus comprises nine recognized species: M. Fund by the Takara Shuzo. Co., Ltd.; the Sasakawa Sci- meretrix (Linnaeus, 1758), M. casta (Chem- entific Research Grant by The Japan Science Society; the nitz, 1782), M. lusoria (Ro¨ding, 1798), M. pe- Mikimoto Fund for Marine Ecology by the Mikimoto & techialis (Lamarck, 1818), M. ovum (Hanley, Co., Ltd.; and a research grant by the Interdisciplinary 1845), M. planisulcata (Sowerby, 1851), M. Research Institute of Environmental Sciences. Travel funding was provided from a graduate student research lyrata (Sowerby, 1851), M. lamarckii Gray, grant by the University of the Ryukyus Foundation. 1853, and M. attenuata Dunker, 1862 (OBIS Manuscript accepted 4 September 2007. Indo-Pacific Molluscan Database 2006). Most 2 Graduate School of Engineering and Science, Uni- are important fishery resources. People have versity of the Ryukyus, Nishihara, Okinawa, 903-0213, Japan. eaten Asian hard clams since ancient times, 3 Department of Regional Economics and Environ- and Meretrix shells are one of the most abun- mental Policy, Okinawa International University, 2-6-1 dant mollusks found in shell middens in Japan Ginowan, Ginowan, Okinawa, 901-2701, Japan. and the Middle East (Kanamaru 1932, Char- 4 Faculty of Science, University of the Ryukyus, pentier et al. 2004). Nishihara, Okinawa, 903-0213, Japan. 5 Current address: 9658-2 Shiomi, Hyuga, Miyazaki Because of the economic importance of 883-0033, Japan. Meretrix, previous work has focused mainly 6 Corresponding author (e-mail: a.yamakawa@okiu on its aquaculture (Yoshida 1941, Wu and .ac.jp). Liu 1992, Tuan and Phung 1998), organotin compounds (Midorikawa et al. 2004, Harino Pacific Science (2008), vol. 62, no. 3:385–394 et al. 2006), and shellfish poisoning (Nguyen : 2008 by University of Hawai‘i Press et al. 2006). Seedlings of a few Meretrix spe- All rights reserved cies are mass-produced, and nearly all hard 385 386 PACIFIC SCIENCE . July 2008 clams in Taiwanese markets are cultured (Wu Meretrix species using electrophoretically de- and Liu 1989). tectable allozyme variation. Past taxonomic studies of Meretrix consid- ered only shell morphology (Fischer-Piette materials and methods and Fischer 1940–1941, Yoosukh and Matsu- kuma 2001), although shell shape and color Samples patterns often show marked intraspecific vari- A total of 12 local samples of spp. ability (Hamai 1934, 1935, Kosuge 2006). Meretrix was collected in East and Southeast Asia and Systematic descriptions of Meretrix species East Africa in 2005. These clam samples are often confusing, and the specific name were identified based on shell morphology M. meretrix has apparently been used for var- and coloration. Individuals were sampled ious species (Yoosukh and Matsukuma 2001). from the following locations: M. lusoria from Moreover, the shell form and color of M. Mutsu Bay (32 individuals examined) and lusoria and M. petechialis are very similar, lead- Ariake Sea (30), Japan (MluJ-M and MluJ-A, ing to many erroneous identifications and respectively); cf. from Korea notations in shell books, reports, and refer- M. lusoria (32) and Taiwan (30) (McfK and McfT, ences. respectively); from northern Mitochondrial DNA (mtDNA) has been M. petechialis China (30) and Korea (31) (MpC and MpK, widely studied in several groups of animals, respectively); from Mozambique mainly for taxonomic and phylogenetic M. ovum (32) and Thailand (26) (MoM and MoT, re- purposes. The advantages of using mtDNA spectively); from southern China include its simple maternal inheritance, ab- M. lyrata (25) (MlyC); from Miyagi (32) sence of recombination, and high substitution M. lamarckii and Shimane (32), Japan (MlaJ-M and MlaJ- rates (Wolstenholme 1992). However, a very S, respectively); and sp. A from Oki- high level of gender-associated mtDNA het- Meretrix nawa, Japan (23) (MspJ-O). The foot muscle eroplasmy has been detected in a few bivalve and hepatopancreas were dissected from fresh mollusks of the families Mytilidae, Unioidae, specimens and immediately frozen at À40 and Veneridae (Hoeh et al. 1996, Liu et al. C. 1996, Beagley et al. 1997, Passamonti and Scali 2001). Thus, the use of mtDNA to esti- Allozyme Electrophoresis mate evolutionary relationships in bivalves requires great care. Furthermore, the number Horizontal starch-gel electrophoresis was of loci available for mtDNA analysis is lim- carried out based on the methods of Harris ited. In contrast, allozyme analysis can in- and Hopkinson (1976) and May et al. (1979). clude a number of loci coded in nuclear Tissue fragments were homogenized with DNA simultaneously. Allozyme analysis re- sterilized distilled water. Hydrolyzed starch mains an effective molecular tool with which gels (12.5%) were run at constant voltage. A to investigate phylogenetic relationships and total of 12 loci from 12 enzymes that showed has been much used in phylogenetic and pop- adequate activity and resolution were rou- ulation genetic studies of mollusks (Benzie tinely examined (Table 1). Terminology and and Williams 1998, Marins and Levy 1999, notation of the allozymes were based on rec- Rı´os et al. 2002, Inness-Campbell et al. 2003, ommendations by Shaklee et al. (1990). Two Va¨ino¨la¨ 2003, Martı´nez et al. 2005, Zaslav- buffer systems were assessed: the Tris-citrate skaya 2006). pH 7.0 buffer system (CT-7; gel:buffer dilu- Despite the economic importance of tion 1:10) was run at 250 V for 5 hr for six Meretrix, few phylogenetic and population enzymes, and the citrate-N-(3-aminopropyl) genetic studies of the genus have been morphoine pH 6.0 buffer system (CAPM-6; undertaken. Molecular techniques can aid gel:buffer dilution 1:4) was run at 250 V for in understanding the taxonomy and rela- 9 hr for six enzymes. All zymograms were vi- tionships within this genus. Therefore, we in- sualized using enzyme-specific stains follow- vestigated the genetic relationships among ing recipes in Numachi (1989). Alleles at Genetic Relationships among Pacific Meretrix . Yamakawa et al. 387 TABLE 1 Resolved Enzymes, Most Effective Buffer Systems, and Tissues Used for Allozyme Analyses Enzyme Abbreviation E.C.a Buffer Tissueb Aspartate aminotranseferase AAT 2.6.1.1 CAPM-6 F Aconitase hydrogenase ACO 4.2.1.3 CAPM-6 H Catalase CAT 1.11.1.6 CT-7 H Glyceraldehyde-3-phosphate dehydrogenase GAPDH 1.2.1.12 CAPM-6 F Glucose-6-phosphate isomerase GPI 5.3.1.9 CT-7 F Isocitrate dehydrogenase (NADPþ) IDH 1.1.1.42 CT-7 H Malate dehydrogenase MDH 1.1.1.37 CAPM-6 F Malic enzyme (NADPþ) ME 1.1.1.40 CAPM-6 F Peptidase using leucyl-glycyl-glycine substrate PEP-lgg 3.4.11–13 CT-7 H 6-Phosphogluconate dehydrogenase 6PGD 1.1.1.44 CT-7 H Phosphoglucomutase PGM 2.7.5.1 CT-7 F Superoxidase dismutase SOD 1.15.1.1 CAPM-6 H a Enzyme commission number. b F, foot; H, hepatopancreas. each locus were labeled alphabetically in diagnostic alleles were observed: allele F and order of the relative electrophoretic mobility G at the ACO locus in M. ovum, allele A at of the allozymes. the GPI locus in M. lyrata, the combination of allele A and C at the MDH-2 locus and al- lele B and C at the SOD-2 locus in M. la- Data Analysis marckii, and the combination of allele A at Genotype and allele frequencies for the 12 the IDH locus and allele C at the MDH-2 lo- loci were determined for all local samples. cus in Meretrix sp. A (Table 2). However, no Genotype frequencies at polymorphic loci diagnostic alleles were observed for M. luso- were examined for agreement with the ex- ria, M. cf. lusoria, or M. petechialis. pectations of Hardy-Weinberg equilibrium Unbiased genetic identities (I ) (Nei 1978) using a chi-square test. The mean biased esti- between taxa varied from 0.162 between mate of expected heterozygosity (He), mean MluJ-A and MoT to 0.850 between McfK observed heterozygosity (Ho), and percentage and MpK (Table 3).