Genetic differentiation among three color variants of Japanese sea cucumber Stichopus japonicus

Blackwell Science, LtdOxford, UK FISFisheries Science0919-92682003 Blackwell Science Asia Pty Ltd 694August 2003 690 Genetic differences in Stichopus japonicus M Kan-No and A Kijima 10.1046/j.0919-9268.2003.00690.x Original Article806812BEES SGML FISHERIES SCIENCE 2003; 69: 806 812 Genetic differentiation among three color variants of Japanese sea cucumber Stichopus japonicus Manami KAN-NO AND Akihiro KIJIMA* Education and Research Center of Marine Bio-resources (ERCOMB), Graduate School of Agricultural Science, Tohoku University, Onagawa, Miyagi 986-2242, Japan ABSTRACT: In order to evaluate genetic differentiation among three color variants (red, green and black) of Japanese sea cucumber, Stichopus japonicus, 11 isozyme loci were used as genetic markers for 448 individuals collected from four localities around Japan. Significant differences of allele frequencies were observed between all pairs of color types sympatrically. Average genetic distances at seven loci between the three sympatric color types were 0.0173 between the red and the other color types, and 0.0015 between the green and the black types. A dendrogram drawn from genetic distances among the three color types across all four localities showed two distinct clusters comprising all the red types and all the green and the black types, respectively. These results indicated that the red type showed definite genetic differentiation from the other color types. KEY WORDS: color variation, genetic difference, Holothuroidea, isozyme, sea cucumber, Stichopus japonicus. INTRODUCTION The Japanese sea cucumber Stichopus japonicus, inhabiting the shallow shelf of coastal sea around Japan (from Hokkaido to Kyushu), China, the Korean Peninsula and Far Eastern Russia is one of the most important seafoods in those countries. 1 Color variation of the species has been found almost all around Japan; 1 in general, they are classified into three groups: red, green and black, according to their visual color. Color variation is one of this species most important traits, affecting the taste and the price of the products in Japan. Choe 1 researched the morphological, physiological and ecological differences among these color types, and reported many differences (i.e. habitat, spawning season, the morphology of ossicle, polian vesicle) between the red and the green types. Nishimura 2 reported that the red type inhabits the gravel bed offshore, while the green and the black types inhabit the sand-muddy bottom inshore. Kanno and Kijima 3 evaluated the three color types and indicated that the red type could be clearly distinguished from the other color types; however, the taxonomical position of the three color types has not yet been resolved. *Corresponding author: Tel: 81-0225-53-2436. Fax: 81-0225-53-2303. Email: a-kijima@mail.cc.tohoku.ac.jp Received 21 June 2002. Accepted 14 March 2003. Genetic markers, such as isozymes and/or DNA polymorphisms, are considered to be powerful tools for detecting taxonomical differentiation. On the genetic analysis of Holothroidea, Uthicke et al. 4 surveyed isozyme variations at five polymorphic loci in Holothuria atra, and Arndt and Smith 5 examined the population structure of two species of sea cucumber based on mitochondrial DNA sequence analysis. Only a few genetic studies have been reported on sea cucumber species and, until recently, none on the Japanese sea cucumber S. japonicus. Kanno and Kijima 6 identified 10 isozyme loci controls for nine enzymes and reported the exceedingly high genetic variability in S. japonicus. These isozymes are expected to be useful genetic markers and provide a means of determining the genetic differences between these color types of the species. The aims of the present study were to clarify the genetic differences, to estimate a degree of genetic differentiation and to discuss the reproductive isolation among the three color variants of the Japanese sea cucumber, by using electrophoretically detectable isozyme markers. MATERIALS AND METHODS Specimens of S. japonicus were sampled from four localities around Japan (Fig. 1). Data on the specimens are shown in Table 1. The color types of

Genetic differences in Stichopus japonicus FISHERIES SCIENCE 807 Fig. 1 Locality of sampling of Stichopus japonicus. Table 1 Sample lot Sampling data of Japanese sea cucumber Stichopus japonicus examined Sampling site Date of collection No. samples Average body wall weight (g)* Color type Red Black 99Hokkaido Funka Bay April 1999 41 50.6 ± 28.1 0 41 0 01Aomori Mutsu Bay January 2001 87 87.4 ± 41.1 0 53 34 99Miyagi Oshika Peninsula Apr October 1999 84 61.9 ± 45.5 0 84 0 01Miyagi Oshika Peninsula January 2001 90 81.6 ± 44.0 24 37 29 01Oita Saeki Bay January 2001 146 214.5 ± 111.4 53 53 40 *Mean ± SD. specimens were determined by ventral color based on the standard established by Choe and Oshima. 7 Typical individuals of each color type are shown in Fig. 2. The ventral marking of the red type is characteristically magenta or cranberry red; of the green type, anything from dark bluish green to yellowish brown or dark brown; and of the black type, inky or intense black. The specimens collected were transported to our laboratory (ERCOMB) and kept in an aquarium until tested. The body color of all specimens did not change during the 1 2 years they were maintained. Detection of isozymes was carried out according to a modification of the methods given in Fujio and Ikeda. 8 Approximately 200 mg intestine from each specimen was minced in a 1.5-mL microtube and frozen at - 20 C until the electrophoretic run. A total of nine enzymes were assayed: aspartate aminotransferase (AAT, E.C.2.6.1.1), glyceraldehyde-3- phosphate dehydrogenase (G3PDH, E.C.1.1.1.12), glucose-phosphate isomerase (GPI, E.C.5.3.1.9), isocitrate dehydrogenase (IDHP, E.C.1.1.1.42), malate dehydrogenase (MDH, E.C.1.1.1.37), mannose phosphate isomerase (MPI, E.C.5.3.1.8), phosphoglucomutase (PGM, E.C.2.7.5.1), 6- phosphogluconate dehydrogenase (PGDH, E.C.1.1.1.44) and superoxide dismutase (SOD, E.C.1.15.1.1). Full details of electrophoresis buffer systems and designation of the alleles are given in Kanno and Kijima. 6 Hardy Weinberg exact tests and population differentiation tests among the color types were performed by GENEPOP version 3.1d. 9,10 Probability values for these tests were estimated using a

808 FISHERIES SCIENCE M Kan-No and A Kijima Fig. 2 Typical individuals of three color types of Stichopus japonicus. Markov chain method (dememorization number = 1000; 100 batches; and 1000 iterations per batch). Nei s genetic distances 11 were calculated between every pair of color types of all lots and the dendrogram was drawn by using UPGMA. 12 RESULTS A total of 11 loci (10 loci were identified by Kanno and Kijima 6 and an additional locus) were estimated; AAT*, G3PDH*, GPI*, IDHP-1*, IDHP-2*, MDH-1*, MDH-2*, MPI*, PGM-1*, PGDH* and SOD*. Because of low enzyme activity, we could not detect the locus IDHP-1* in 99 Hokkaido or 99 Miyagi, nor could we detect the loci AAT*, MPI* and PGM-1* in 01Oita. Therefore, statistical analyses within regions were carried out using all available loci, and those across regions were carried out using seven loci that were detected in all regions. Allele frequencies for each locality and each color type were calculated as shown in Table 2. Major alleles of each locus were generally identical in all localities and in all color types, but allele frequencies fluctuated among them. At AAT*, GPI*, PGDH* and PGM- 1*, obvious differences of allele frequencies were observed between the red and the other color types. Hardy Weinberg exact tests were carried out for two localities comprising three color types. Significant deviation was observed in 01Miyagi

Genetic differences in Stichopus japonicus FISHERIES SCIENCE 809 Table 2 Locus Allele frequencies of the specimens of Japanese sea cucumber Stichopus japonicus examined Allele 99- Hokkaido 01Aomori 99-01Miyagi 01Oita Miyagi Total Black Total Red Black Total Red Black AAT* *120 0.006 *110 0.021 0.035 0.006 0.013 0.032 *100 0.890 0.829 0.814 0.852 0.845 0.769 0.684 0.694 0.926 ND ND ND ND *90 0.110 0.007 0.012 0.131 0.090 0.079 0.113 0.056 *80 0.143 0.140 0.148 0.012 0.128 0.237 0.161 0.019 G3PDH* *110 0.012 0.006 0.009 0.006 0.021 *105 0.006 *100 0.976 0.988 0.991 0.985 0.964 0.977 0.979 0.971 0.983 1.000 1.000 1.000 1.000 *95 0.012 0.006 0.015 0.030 0.017 0.029 0.017 GPI* *120 0.012 0.009 0.015 0.006 0.034 0.125 0.028 0.059 0.020 *100 0.988 0.953 0.943 0.970 0.952 0.920 0.813 0.929 1.000 0.905 0.804 0.931 1.000 *85 0.011 0.042 0.039 0.078 0.029 *70 0.012 0.035 0.047 0.015 0.042 0.011 0.042 0.071 0.028 0.059 0.020 IDHP-1* *100 ND 0.971 0.972 0.969 ND 0.972 1.000 0.926 1.000 0.989 1.000 0.980 0.986 *90 0.029 0.028 0.031 0.028 0.074 0.011 0.020 0.014 IDHP-2* *120 0.012 0.006 0.011 0.042 *110 0.280 0.297 0.292 0.303 0.298 0.293 0.271 0.324 0.286 0.352 0.284 0.396 0.375 *100 0.707 0.657 0.660 0.652 0.625 0.667 0.688 0.632 0.679 0.629 0.705 0.583 0.597 *90 0.047 0.047 0.045 0.071 0.029 0.044 0.036 0.020 0.011 0.021 0.028 MDH-1* *110 0.006 0.015 0.006 *100 0.976 0.977 0.972 0.985 0.976 0.966 0.938 0.985 0.983 0.993 0.981 1.000 1.000 *90 0.006 0.009 0.006 0.023 0.021 0.015 0.017 *80 0.012 0.019 0.012 0.011 0.042 0.007 0.019 *70 0.024 MDH-2* *110 0.012 *100 0.988 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.989 0.990 0.981 1.000 *90 0.011 0.010 0.019 MPI* *110 0.037 0.012 0.019 0.018 0.045 0.021 0.029 0.086 *100 0.683 0.849 0.915 0.742 0.833 0.858 0.938 0.794 0.862 ND ND ND ND *90 0.268 0.140 0.066 0.258 0.149 0.097 0.042 0.176 0.052 *80 0.012 PGDH* *100 0.415 0.477 0.481 0.470 0.470 0.506 0.646 0.485 0.431 0.538 0.767 0.380 0.473 *90 0.073 0.064 0.066 0.061 0.036 0.068 0.083 0.059 0.052 0.068 0.067 0.070 0.068 *80 0.512 0.459 0.453 0.470 0.488 0.426 0.271 0.456 0.517 0.394 0.167 0.550 0.459 *70 0.006 PGM-1* *115 0.035 0.038 0.030 *110 0.037 0.163 0.160 0.167 0.131 0.137 0.354 0.044 0.040 *105 0.366 0.273 0.302 0.227 0.190 0.268 0.396 0.221 0.220 *100 0.232 0.308 0.264 0.379 0.274 0.196 0.125 0.265 0.180 ND ND ND ND *95 0.122 0.134 0.123 0.152 0.149 0.202 0.063 0.279 0.240 *85 0.159 0.070 0.085 0.045 0.190 0.143 0.063 0.176 0.180 *80 0.085 0.017 0.028 0.060 0.048 0.015 0.120 *75 0.006 0.006 SOD* *120 0.006 0.015 *100 0.951 1.000 1.000 1.000 0.958 0.977 0.979 0.971 0.983 1.000 1.000 1.000 1.000 *70 0.049 0.036 0.017 0.021 0.015 0.017 *60 0.006 ND, no data. (c 2 = 31.5, P = 0.049, at 11 loci), but not in 01Oita (c 2 = 11.0, P = 0.528, at eight loci). Population differentiation tests were then performed on color variants of S. japonicus in the different localities (Table 3). The first tests for allele frequencies at the available loci were performed on 01Miyagi samples. Among the three sympatric color types, significant differences (P < 0.05) were observed at

810 FISHERIES SCIENCE M Kan-No and A Kijima Table 3 Genetic differentiation for all pairs of sympatric color types of Japanese sea cucumber Stichopus japonicus (by GENEPOP) 01Oita P value 01Miyagi P value Locus Total Red Red Black Black Locus Total Red Red Black Black AAT* 0.012** 0.817 0.003*** 0.008*** AAT* ND ND ND ND G3PDH* 0.552 0.298 0.700 1.000 G3PDH* GPI* 0.000*** 0.001*** 0.001*** 0.039** GPI* 0.001*** 0.068 0.000*** 0.138 IDHP-1* 0.091 0.207 0.129 IDHP-1* 0.508 0.230 0.402 1.000 IDHP-2* 0.648 0.367 0.371 0.972 IDHP-2* 0.428 0.218 0.336 0.954 MDH-1* 0.149 0.137 0.236 1.000 MDH-1* 0.336 0.498 0.509 MDH-2* MDH-2* 0.784 1.000 1.000 0.511 MPI* 0.029** 0.057 0.400 0.039** MPI* ND ND ND ND PGDH* 0.131 0.133 0.029** 0.828 PGDH* 0.000*** 0.000*** 0.000*** 0.500 PGM-1* 0.000*** 0.000*** 0.000*** 0.215 PGM-1* ND ND ND ND SOD* 1.000 1.000 1.000 1.000 SOD* Significant at **P < 0.05, ***P < 0.01. ND, no data. four loci (AAT*, GPI*, MPI* and PGM-1*), suggesting the existence of genetic heterogeneity among the three sympatric color types. In the pair analyses, significant differences were observed at two loci (GPI* and PGM-1*) between the red and the green types, four loci (AAT*, GPI*, PGDH* and PGM-1*) between the red and the black types, and three loci (AAT*, GPI* and MPI*) between the green and black types. The same tests on the three color types were performed at eight available loci in 01Oita samples (Table 3). Significant differences (P < 0.05) were observed at two loci (GPI* and PGDH*) among the three color types, at one locus (PGDH*) between the red and the green types, at two loci (GPI* and PGDH*) between the red and the black types, but at no locus between the green and the black types. The tests on 01Aomori samples performed at 11 available loci showed significant differences at one locus (MPI*, P = 0.001) between the green and the black types (data not shown). Nei s genetic distances (D) between the color types in our four localities were calculated from the allele frequencies based on the seven loci (Table 4). As the results, the minimum D-value, 0.0003 was detected between the green and the black types in the 01Aomori sample, and the maximum D-value, 0.0304, was discovered between the red and the green types in the 01Oita sample (Table 4). Average D-values within each color type were 0.0032 within the red type, 0.0020 within the green type and 0.0011 within the black type, while the averages between types were 0.0175 between the red and the green types, 0.0171 between the red and the black types, and 0.0015 between the green and the black types. Clearly, larger D-values were observed between the red and the other color types. The dendrogram drawn among all sampling lots showed two distinct clusters, comprising all the red types and all the other types, respectively (Fig. 3). However, the green and the black types were not divided clearly. DISCUSSION The major alleles of each locus were generally identical through the color types, and any peculiar allele to one color type was not observed at all loci surveyed. However, significance of allele frequencies was clearly observed between all pairs of color types. This indicates that the gene flow between sympatric color types was restricted to some degree, and three color types do not constitute a single population. Choe and Oshima 7 reported some distinct morphological and ecological differences between the

Genetic differences in Stichopus japonicus FISHERIES SCIENCE 811 Fig. 3 Genetic relationship of color type lots with seven loci in Stichopus japonicus. Table 4 Nei s genetic distance between all pairs of color type lots with seven loci in Japanese sea cucumber Stichopus japonicus 01Oita- 01Oita- Red 99Miyagi- 01Miyagi- Black 01Miyagi- 01Miyagi- Red 01Aomori- Black 01Aomori- green and the black types; the shape of the polian vesicles, egg, ossicles, the spawning period and capable of regeneration etc. From these observations, Choe 1 concluded that these differences between the red and the green types suggested far more than a mere variation or, more precisely, to constitute fair grounds for establishing a new species. Nei s genetic distance (D) between the red and the other color types was 0.0096 0.0304 (0.0173 in average), based on seven loci. Nei 13 noted that the D-value is approximately 1.0 between species, 0.1 between subspecies and 0.01 between local races in a variety of animals. According to the above criteria, the D-value between the red and the other types in the present study could only be considered as a local race level difference. However, these differences were observed even in the same locality and, in addition, the red types from distant two localities (Miyagi and Oita) clustered together. This indicated that the red type is reproductively isolated from the other color types. Indeed, experimental D-values between closely related sympatric invertebrate species or subspecies; 0.178 between two sympatric types of shrimp Palaemon paucidens 14,15 and 0.156 between two sympatric forms of grapsid crab Hemigrapsus penicillatus 16 indicating the existence of reproductive isolation between the types. Regarding the green and black types, genetic differences were detected within regions, but D- values were consistently low; 0.0005 0.0038 (0.0015 average). This indicated that the green and the black types are very closely related genetically although the gene flow between them is not limitless. Kanno and Kijima 3 evaluated the color traits of S. japonicus qualitatively (composition of pig- 99Hokkaido- Lot 01Aomori- 0.0019 01Aomori-Black 0.0015 0.0003 01Miyagi-Red 0.0147 0.0087 0.0104 01Miyagi- 0.0024 0.0005 0.0006 0.0096 01Miyagi-Black 0.0005 0.0012 0.0006 0.0145 0.0017 99Miyagi- 0.0018 0.0007 0.0005 0.0118 0.0005 0.0009 01Oita-Red 0.0259 0.0170 0.0193 0.0032 0.0175 0.0251 0.0210 01Oita- 0.0038 0.0036 0.0028 0.0187 0.0031 0.0029 0.0031 0.0304 01Oita-Black 0.0030 0.0015 0.0009 0.0125 0.0013 0.0018 0.0015 0.0210 0.0020

812 FISHERIES SCIENCE M Kan-No and A Kijima ments of skin) and quantitatively (distribution of color values) and showed that the same pigments and continuous distribution of color were observed between the green and the black types, while the red type was clearly independent in both ways. The genetic features quantified in the present study corroborate their observations. In nature, red-type individuals tend to inhabit gravel beds offshore while the green and the black types inhabit sand-muddy beds inshore, although the three color types inhabit quite sympatrically in some area. 2 Reproductive isolation might have occurred in such microhabitat differentiation. However, the clear genetic independence of the red types could not be explained by differences of microhabitat only. Other reproductive isolation systems preventing the free mating between the red and the other types probably exist. To confirm such conclusion, it will be necessary to examine the possibility of mating between color types and to confirm the expression of pigments in offspring through mating experiments. ACKNOWLEDGMENTS The authors are grateful to Professor Hiroyuki Munehara, Hokkaido University, Dr Yoshinobu Kosaka, Aomori Prefectural Aquaculture Research Center and Mr Yoshihiro Yamamoto, Oita Institute of Marine and Fisheries Science for collecting the samples. This study was supported in part by the Grant-in-Aid for Scientific Research (No.13460080) from the Japanese Ministry of Education, Culture, Sports, Science and Technology. REFERENCES 1. Choe S. Biology of the Japanese Common Sea Cucumber Stichopus japonicus Selenka. Kaibundo, Tokyo. 1963. 2. Nishimura S. Guide to Seashore Animals of Japan with Color Pictures and Keys, Vol. 2. Hoikusha, Osaka. 1995. 3. Kanno M, Kijima A. Quantitative and qualitative evaluation on the color variation of the Japanese sea cucumber Stichopus japonicus. Suisanzoshoku 2002; 50: 63 69. 4. Uthicke S, Benzie JAH, Ballment E. Genetic structure of fissiparous populations of Holothuria (Halodeima) atra on the Great Barrier Reef. Mar. Biol. 1998; 132: 141 151. 5. Arndt A, Smith MJ. Genetic diversity and population structure in two species of sea cucumber: differing patterns according to mode of development. Mol. Ecol. 1998; 7: 1053 1064. 6. Kanno M, Kijima A. High genetic variability of isozymes in Japanese sea cucumber Stichopus japonicus. Fish Genet. Breed. Sci. 2002; 31: 7 12. 7. Choe S, Oshima Y. On the morphological and ecological differences between two commercial forms, and Red, of the Japanese common sea cucumber, Stichopus japonicus Selenka. Nippon Suisan Gakkaishi 1961; 27: 97 105. 8. Fujio Y, Ikeda M. Isozyme analysis and DNA analysis. In: Fujio Y (ed). A Mode of Existence and Conservation for Genetic Resources of Aquatic Animals. Oyster Research Institute, Sendai. 1999; 13 41. 9. Raymond M, Rousset F. An exact test for population differentiation. Evolution 1995; 49: 1280 1283. 10. Raymond M, Rousset F. Population genetics software for exact tests and ecumenicism. J. Heredity 1995; 86: 248 249. 11. Nei M. Genetic distance between population. Am. Nat. 1972; 114: 238 280. 12. Sneath PHA, Sokal R. Numerical Taxonomy. Freeman, San Francisco. 1973. 13. Nei M. Molecular Population Genetics and Evolution. North-Holland, Amsterdam. 1975. 14. Chow S, Fujio Y, Nomura T. Reproductive isolation and distinct population structures in two types of the freshwater shrimp Palaemon paucidens. Evolution 1988; 42: 804 813. 15. Fidhiany L, Kijima A, Fujio Y. Genetic divergence between two types in Palaemon paucidens. Tohoku J. Agric. Res. 1988; 39: 39 45. 16. Takano M, Ikeda M, Kijima A. Biochemical and morphological evidence of two sympatric forms, interpreted as sibling species, in the estuarine grapsid crab Hemigrapsus penicillatus (De Haan). Benthos Res. 1997; 52: 111 117.