Fig. 3 in Zavreliella shidai Cao & Tang, 2017, sp. n.

© 2017 Academia Sinica, Taiwan Open Access Population Structure of the Blue Swimmer Crab Portunus pelagicus in Coastal Areas of Malaysia Inferred from Microsatellites Chuan Jian Chai1, Yuzine Bin Esa1,2,*, Muhammad Fadhil Syukri Ismail1,2, and Mohd. Salleh Kamarudin1 1Department of Aquaculture, Faculty of Agriculture, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia. E-mail: [email protected] (Chai); [email protected] (Kamarudin) 2Institute of Biosciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia. E-mail: [email protected] (Received 13 May 2017; Accepted 5 September 2017; Published 16 October 2017; Communicated by Benny K.K. Chan) Chuan Jian Chai, Yuzine Bin Esa, Muhammad Fadhil Syukri Ismail, and Mohd. Salleh Kamarudin (2017) Portunus pelagicus, distributed throughout the Indo-West Pacific region, is one of the large and edible species of blue swimmer crabs. Increasing demand for the frozen and canned crabmeat industry worldwide has now relied mainly on P. pelagicus which in turn generates splendid income for the fisherman communities. In the present study, the population genetic structure of P. pelagicus was examined using six pairs of microsatellite loci. A total of 87 crab samples were collected from five different coastal areas of Malaysia. Genomic DNA was extracted from each sample for polymerase chain reaction (PCR) amplification and fragment analysis. Four out of six microsatellite primers revealed polymorphic loci in P. pelagicus sampled. The number of alleles per locus in P. pelagicus ranged from 14 to 34. Microsatellites analyses indicated low levels of genetic differentiation among the P. pelagicus populations. The average observed heterozygosity (HO = 0.48) obtained was lower than the standard heterozygosity found in most marine populations (HO = 0.79). The high FIS values (mean FIS = 0.4756) and low FST values (mean FST = 0.0413) also suggested the existence of inbreeding among different populations of P. pelagicus. In conclusion, this study was able to shed light on the population structure of P. pelagicus in coastal areas of Malaysia. Key words: Portunus pelagicus, Blue swimmer crabs, Population genetic structure, Microsatellites. *Correspondence: E-mail: [email protected] BACKGROUND In Malaysia, the population genetic structure of the blue swimmer crabs, Portunus pelagicus has not been well-studied unlike countries such as Australia and Thailand (Yap et al. 2002; Klinbunga et al. 2007). However, the increasing demands of P. pelagicus in the fisheries industry of Malaysia currently have led to a growing interest on the broodstock of this particular crab species. The knowledge on the genetic differentiation of P. pelagicus is no doubt useful for the effective management of this edible crab species with wide distribution range and long planktonic larval stages (Klinbunga et al. 2007). Microsatellites, also termed as short tandem repeats (STRs) or simple sequence repeats (SSRs) are tandemly repeating motifs of DNA (1-6 bases long) which are widely distributed throughout the nuclear genomes of eukaryotes (Putman and Carbone 2014; Senan et al. 2014). With high levels of allelic polymorphism and codominant inheritance, they have become the mainstay of population genetics, conservation management, parentage identification and fingerprinting (Nolan et al. 2000; Putman and Carbone 2014). Lately, Yap et al. (2002) identified eight microsatellites in P. pelagicus (seven dinucleotides and one tetranucleotide). All eight microsatellite loci were polymorphic when inspected against Zoological Studies 56: 26 (2017) doi:10.6620/ZS.2017.56-26 1 © 2017 Academia Sinica, Taiwan the genomic DNA of P. pelagicus collected throughout Australia. In general, the mean observed heterozygosity (HO) was not significantly different from the expected heterozygosity (HE). To date, Sezmis (2004) investigated the population genetic structure of P. pelagicus from 16 diverse assemblages in Australia via six microsatellite loci. Large genetic distances between pairs of geographic samples indirectly reflected strong intraspecific genetic differentiation of P. pelagicus. Recent studies have verified that microsatellites are more variable and informative than dominant markers like RAPD and AFLP (He et al. 2003; Senan et al. 2014). Bryars and Adams (1999) on the other hand reported that P. pelagicus shows relatively high polymorphisms when tested with allozymes. Nevertheless, microsatellite markers which exhibit higher levels of polymorphism than the allozyme loci are ideal for this research. Thus, the present study examined the population structure of P. pelagicus in coastal areas of Malaysia using microsatellites. The detailed information on the genetic diversity of P. pelagicus populations is necessary for the breeding programs of this exploited taxon. MATERIALS AND METHODS Sampling Location and Sample Collection A total of 87 Portunus pelagicus samples were collected from selected sites on both east coast (South China Sea) and west coast (Strait of Malacca) of Peninsular Malaysia including Perak (Pantai Remis and Kuala Sepetang), Johor (Pendas), Negeri Sembilan (Port Dickson), and Terengganu (Besut) and Sarawak (Bako) of Borneo (Fig. 1). Samples were identified using taxonomic keys provided in Ng (1998) and Lai et al. (2010). The crab samples (muscle tissue from the chelipid manus or whole crab) were preserved in 95% ethanol. These samples were then stored at -20°C until further analyses. DNA Extraction, Polymerase Chain Reaction Amplification (PCR), Agarose Gel Electrophoresis, DNA Screening and Fragment Analysis Total genomic DNA extraction was performed using the DNeasy® Blood and Tissue Kit (QIAGEN). Alternatively, cetyl-trimethylammonium bromide (CTAB) method in the presence of proteinase K Fig. 1. Sampling locations of Portunus pelagicus in coastal areas of Malaysia. ●Pendas, Johor ●Port Dickson, Negeri Sembilan ●Besut, Terengganu ●Pantai Remis, Perak ●Kuala Sepetang, Perak Bako, Sarawak 100°E 105°E 110°E 110°E 0° 5°N ● N page 2 of 12Zoological Studies 56: 26 (2017) © 2017 Academia Sinica, Taiwan was modified and applied (Grewe et al. 1993). Microsatellite amplifications were carried out with six pairs of microsatellite loci (Table 1) developed by Yap et al. (2002) and Xu and Liu (2011). Approximately 1.0 µl of DNA template was amplified in a reaction mixture containing 7.5 µl of 5X MyTaqTM Red Mix (Bioline, USA) and 1.0 µl of each primer. The reaction mixture was then adjusted to a final volume of 15 µl with ddH2O. The thermal cycling parameters included initial denaturation at 94°C for 2 minutes followed by 25 cycles of denaturation at 94°C for 20 seconds, primer annealing at Ta°C (Table 1) for 20 seconds, extension at 72°C for 40 seconds and a final extension at 72°C for 3 minutes. The completed amplification process hold at a routine 10°C. All amplifications were carried out with negative controls to check for contamination throughout the experiment. The PCR products of the microsatellites were viewed under 2% high resolution agarose gel. A total of 1.2 grams of HR agarose powder (HydraGene, USA) was mixed with an exact amount of 60 ml of 1X TBE buffer (Promega, USA) to prepare the gel. BenchTop 50bp DNA ladder was used as a standard DNA size marker. PCR products with multiple fluorescence bands indicated the presence of DNA polymorphisms. The samples were then subjected to microsatellite screening for estimation of the expected size of PCR products. Colourless MyTaq Red Mix and labelled primers with appropriate fluorescent dyes (FAM and HEX) were employed for PCR amplification at this stage. Also, loading dye was added into the PCR products for agarose gel electrophoresis. All the PCR products were wrapped with aluminium foils and sent for fragment analysis through Applied Biosystems Genetic Analyzer. Fragment sizes were interpreted according to the 500-ROX DNA size standard using GeneMapper® version 5.0 (Chatterji and Pachter 2006). The results of fragment analysis obtained were then applied in the statistical analyses of microsatellites. Statistical Analyses The CONVERT 1.31 software (Glaubitz 2004) was used to translate the genotypic data into required formats for microsatellite analyses. These included the GENEPOP, ARLEQUIN and STRUCTURE formats. Allelic frequencies of the microsatellite loci found among the Portunus pelagicus populations were also computed through the CONVERT software. Besides, MicroChecker 2.2.3 software (Van Oosterhout et al. 2004) was utilised to check for genotyping errors, particularly due to null alleles and allele dropouts. In addition, the exact tests for both HardyWeinberg Equilibrium (HWE) and linkage disequilibrium among pairs of loci with 10,000 permutations were conducted via GENEPOP version 4.4 (Rousset 2008). The inbreeding coefficient (FIS), allelic richness (AR), observed (HO) and expected heterozygosity (HE) on the other hand were calculated through FSTAT version 2.9.3.2 (Goudet 1995). Furthermore, ARLEQUIN version 3.0 (Excoffier et al. 2005) was used to estimate the significance of spatial variation in genetic diversity of P. pelagicus populations implemented in AMOVA. The fixation index (FST) was also measured using the ARLEQUIN software Table 1. Primer sequences of six microsatellite loci Locus Sequences Repeat motif Ta (°C) Size range (bp) pPp2 F: GTGACCAGTAGGCGACCGAG R: ACGACTGCTTGTACGACCTTCA (CA)16 59 69-141 pPp5 F: GCTACGACAGTCCAATAACAACGT R: GATAGACCGACCTCACCTCAAAA (AG)35 57 87-151 pPp9 F: GACTTGAGCGATGCTGAAAG R: ATGGATAGATGGAATGCAAAAT (TG)19 52 133-187 pPp10 F: CCTGTATTGTCATGTGTTTGATTTT R: CTACGACCAACTTTACCGCC (TG)34 52 91-155 Ptri1 F: ACGCGTCTGGTAGTCATC R: TGTTCCCAAAGTTAGCAG (TGC)12 57 367-454 Ptri2 F: CAATGGCGGGTATGGTA R: TAAATGAAGGAAGCTAAAGACAAA (TC)28 52 257-359 (Source: Yap et al. 2002; Xu and Liu 2011). page 3 of 12Zoological Studies 56: 26 (2017) © 2017 Academia Sinica, Taiwan to examine the genetic differentiation among P. pelagicus populations (Weir and Cockerham 1984). The STRUCTURE version 2.3.4 (Pritchard et al. 2000) was used to allocate each individual to their genetic groups (K). This was done by admixing the ancestry within individuals. Ten independent runs were normally achieved with a burn-in period of 10,000 iterations and 10,000 replications. The best value of K was selected through the ad hoc statistic recommended by Evanno et al. (2005) and the log posterior probability of the data for a given K, In Pr (X|K) proposed by Pritchard et al. (2000). Moreover, assignment tests of each individuals were performed via GeneClass version 2.0 (Piry et al. 2004). The probability of an individual belonged to a specific population was identified. The Bayesian approach by Rannala and Mountain (1997) was then employed to determine the likelihood of inherited population with 10,000 simulations and threshold value of 0.05. Finally, the BOTTLENECK version 1.2.02 (Piry et al. 1999) was implemented to test for the occurrence of current bottlenecks in each population. The two phase mutation (TPM) model (Di Rienzo et al. 1994) was introduced with different percentages (60-80%) of the stepwise mutation model (SMM) and a variance of 12 in 5,000 replications. All bottleneck analyses were derived from the Wilcoxon sign rank test (Maudet et al. 2002). RESULTS Microsatellite Genotyping Of the six microsatellite primer pairs selected, only five primer pairs were successfully amplified. Of these five microsatellite loci, one (Ptri1) was monomorphic while the others were polymorphic (pPp2, pPp9, pPp10, Ptri2) and yielded PCR bands (Fig. 2) consistently. Ptri1 and Ptri2 were chosen for the cross-species amplification of Portunus pelagicus as these two primers were initially designed for P. tritubeculatus populations in China. The sizes of all the microsatellite loci used ranged from 69-454 bp. Most of the microsatellite primers contained dinucleotide repeat units except for Ptri1 with trinucleotide repeats. However, both pPp5 and Ptri1 loci were excluded from the statistical analyses to avoid scoring errors. The pPp5 primer had no PCR product at temperature higher or lower than the annealing temperature. Microsatellite Variations The allele frequencies of four microsatellite loci obtained from five different populations of Portunus pelgicus across the coastal areas of Malaysia were depicted in table 2. The highest and lowest number of alleles were acquired by loci pPp2 and Ptri2 with an amount of 34 alleles and 14 alleles respectively. On the other hand, the mean Fig. 2. Gel image of Portunus pelagicus samples obtained using microsatellite primers (Ptri2). S36-S37: Sample 36-Sample 37; S39-S51: Sample 39-Sample 51; S53: Sample 53. page 4 of 12 Zoological Studies 56: 26 (2017) © 2017 Academia Sinica, Taiwan Table 2. Allele frequencies of five Portunus pelagicus populations through four pairs of microsatellite loci Locus Allele Size Perak Johor Negeri Sembilan Terengganu Sarawak Overall pPp2 174 0.0000 0.0625 0.0000 0.0000 0.0000 0.0172 276 0.0455 0.0000 0.0000 0.0000 0.0385 0.0115 3 78 0.0000 0.0208 0.0000 0.0000 0.1538 0.0287 4 80 0.0000 0.0000 0.0000 0.0000 0.3077 0.0460 5 82 0.0000 0.0000 0.0357 0.0000 0.0000 0.0057 6 84 0.0000 0.0417 0.0000 0.0000 0.0000 0.0115 786 0.0455 0.0833 0.1429 0.0800 0.0000 0.0747 8 88 0.0000 0.0000 0.0357 0.0000 0.0000 0.0057 9 90 0.0000 0.0000 0.0000 0.1000 0.0000 0.0287 10 92 0.0000 0.0000 0.0000 0.0200 0.0000 0.0057 11 94 0.0000 0.0208 0.0000 0.0400 0.0769 0.0287 12 96 0.0909 0.0625 0.0000 0.0000 0.0000 0.0287 13 98 0.0909 0.0208 0.1429 0.0000 0.0385 0.0460 14 100 0.1364 0.0208 0.0357 0.1600 0.1538 0.0977 15 102 0.0000 0.0417 0.0000 0.0000 0.0000 0.0115 16 104 0.0000 0.0417 0.0000 0.0400 0.0000 0.0230 17 106 0.3182 0.0625 0.0000 0.1400 0.0000 0.0977 18 108 0.0455 0.0417 0.0000 0.0000 0.0000 0.0172 19 110 0.1818 0.0000 0.0000 0.1200 0.0000 0.0575 20 112 0.0000 0.0417 0.0000 0.0000 0.0000 0.0115 21 114 0.0455 0.1042 0.0000 0.0200 0.0000 0.0402 22 116 0.0000 0.0417 0.0000 0.0000 0.0000 0.0115 23 118 0.0000 0.0208 0.1429 0.0000 0.0769 0.0402 24 120 0.0000 0.1875 0.0000 0.0000 0.0000 0.0517 25 122 0.0000 0.0000 0.0000 0.0400 0.0000 0.0115 26 124 0.0000 0.0625 0.1071 0.0400 0.0000 0.0460 27 126 0.0000 0.0000 0.0714 0.1200 0.0000 0.0460 28 128 0.0000 0.0000 0.1429 0.0000 0.0000 0.0230 29 130 0.0000 0.0000 0.0714 0.0000 0.0000 0.0115 30 132 0.0000 0.0208 0.0357 0.0000 0.0000 0.0115 31 134 0.0000 0.0000 0.0000 0.0200 0.0385 0.0115 32 136 0.0000 0.0000 0.0357 0.0200 0.0385 0.0172 33 138 0.0000 0.0000 0.0000 0.0200 0.0000 0.0057 34 140 0.0000 0.0000 0.0000 0.0200 0.0769 0.0172 pPp9 1134 0.0000 0.2917 0.0357 0.0400 0.0769 0.1092 2136 0.0000 0.1250 0.0000 0.0800 0.1923 0.0862 3 138 0.2727 0.0000 0.1071 0.1000 0.0000 0.0805 4 140 0.0000 0.0208 0.0714 0.0000 0.0000 0.0172 5 142 0.1364 0.0417 0.2857 0.2800 0.1538 0.1782 6 144 0.0909 0.0000 0.0000 0.0000 0.0000 0.0115 7146 0.0455 0.0000 0.0357 0.0000 0.0000 0.0115 8 148 0.0455 0.0000 0.0357 0.0000 0.0000 0.0115 9150 0.0000 0.0417 0.1429 0.0400 0.1923 0.0747 10 152 0.0000 0.0417 0.0000 0.0000 0.0000 0.0115 11 154 0.0000 0.0417 0.0000 0.0000 0.0000 0.0115 12 156 0.2727 0.0625 0.0000 0.0800 0.0385 0.0805 13 158 0.0000 0.0208 0.0000 0.0000 0.0000 0.0057 14 160 0.0000 0.0000 0.1071 0.0000 0.1154 0.0345 15 162 0.0909 0.0417 0.0357 0.0800 0.0769 0.0632 16 164 0.0000 0.0208 0.0000 0.0400 0.0000 0.0172 17 166 0.0000 0.0208 0.0000 0.0400 0.0000 0.0172 18 168 0.0000 0.0833 0.0000 0.0800 0.0000 0.0460 19 170 0.0000 0.0833 0.0357 0.0600 0.0769 0.0575 page 5 of 12Zoological Studies 56: 26 (2017) © 2017 Academia Sinica, Taiwan Locus Allele Size Perak Johor Negeri Sembilan Terengganu Sarawak Overall 20 172 0.0000 0.0417 0.0357 0.0000 0.0000 0.0172 21 174 0.0000 0.0208 0.0714 0.0400 0.0769 0.0402 22 176 0.0000 0.0000 0.0000 0.0400 0.0000 0.0115 23 178 0.0455 0.0000 0.0000 0.0000 0.0000 0.0057 pPp10 1 92 0.0000 0.0000 0.0000 0.0200 0.0000 0.0057 294 0.0000 0.0000 0.0000 0.0200 0.0000 0.0057 3 96 0.0000 0.0000 0.0000 0.0800 0.0000 0.0230 4 98 0.0000 0.0208 0.0000 0.0000 0.0000 0.0057 5100 0.0000 0.0833 0.0714 0.0000 0.0000 0.0345 6102 0.0000 0.0417 0.0000 0.0000 0.0000 0.0115 7104 0.0000 0.0417 0.0714 0.1200 0.0000 0.0575 8 106 0.0000 0.0000 0.0357 0.0000 0.0000 0.0057 9108 0.0000 0.0417 0.0000 0.0000 0.0000 0.0115 10 110 0.0000 0.0625 0.0357 0.0000 0.0000 0.0230 11 112 0.0000 0.0417 0.0000 0.0200 0.0000 0.0172 12 114 0.0000 0.0417 0.0000 0.0000 0.0000 0.0115 13 116 0.0000 0.0000 0.0000 0.0200 0.0385 0.0115 14 118 0.0000 0.0000 0.1429 0.0400 0.0000 0.0345 15 120 0.0909 0.0000 0.0714 0.0400 0.0000 0.0345 16 122 0.0455 0.0208 0.1071 0.1400 0.2692 0.1092 17 124 0.0000 0.0000 0.0714 0.0400 0.0000 0.0230 18 126 0.0455 0.0833 0.0000 0.0000 0.0000 0.0287 19 128 0.3636 0.1250 0.0714 0.0000 0.0000 0.0920 20 130 0.0000 0.1042 0.0000 0.0000 0.0385 0.0345 21 132 0.0000 0.0417 0.0000 0.0000 0.0769 0.0230 22 134 0.2273 0.0625 0.0357 0.1000 0.0769 0.0920 23 136 0.0000 0.0417 0.1071 0.0000 0.0769 0.0402 24 138 0.0000 0.0625 0.1429 0.0400 0.0385 0.0575 25 140 0.0000 0.0417 0.0000 0.1800 0.0769 0.0747 26 142 0.0455 0.0000 0.0000 0.0000 0.0000 0.0057 27 144 0.0000 0.0000 0.0357 0.0200 0.0000 0.0115 28 146 0.0909 0.0000 0.0000 0.0600 0.0000 0.0287 29 148 0.0000 0.0417 0.0000 0.0000 0.0000 0.0115 30 150 0.0000 0.0000 0.0000 0.0400 0.0385 0.0172 31 152 0.0000 0.0000 0.0000 0.0000 0.0385 0.0057 32 154 0.0909 0.0000 0.0000 0.0200 0.2308 0.0517 Ptri2 1 258 0.3182 0.1250 0.2143 0.3200 0.0769 0.2126 2260 0.0000 0.2917 0.0000 0.0200 0.0769 0.0977 3 262 0.0455 0.1667 0.2857 0.1000 0.1923 0.1552 4 264 0.0000 0.1250 0.0357 0.2800 0.2692 0.1609 5 266 0.0909 0.0000 0.1071 0.0000 0.0769 0.0402 6 268 0.0000 0.0000 0.0714 0.0000 0.0385 0.0172 7 270 0.2273 0.0417 0.0000 0.0000 0.0000 0.0402 8272 0.0455 0.0000 0.0000 0.0000 0.0000 0.0057 9274 0.1818 0.0417 0.2143 0.1000 0.0000 0.0977 10 276 0.0000 0.1250 0.0714 0.0200 0.1154 0.0690 11 278 0.0000 0.0417 0.0000 0.1200 0.0769 0.0575 12 280 0.0000 0.0417 0.0000 0.0200 0.0000 0.0172 13 282 0.0000 0.0000 0.0000 0.0200 0.0000 0.0057 14 284 0.0909 0.0000 0.0000 0.0000 0.0769 0.0230 Number of samples 11 24 14 25 13 87 Table 2. (continued) page 6 of 12Zoological Studies 56: 26 (2017) © 2017 Academia Sinica, Taiwan allelic frequencies ranged from 0.0057 to 0.2126. P. pelagicus of the Terengganu and Negeri Sembilan populations displayed the highest and the lowest percentage of allele discrepancies, each with 12 and seven distinct allele sizes. Hardy-Weinberg Equilibrium and Linkage Disequilibrium among Portunus pelagicus Populations All four microsatellite loci assigned to characterise the HWE and linkage disequilibrium of Portunus pelagicus populations showed significant genetic differentiations (Table 3). The allele richness ranged from 6.643 (locus Ptri2 of the Terengganu population) to 13.662 (locus pPp10 of the Johor population). In addition, the observed heterozygosity (HO) values over all loci extended from 0.182 (loci pPp10 and Ptri2 of the Perak population) to 1.000 (locus pPp2 of the Perak population). On the contrary, the highest and the lowest expected heterozygosity (HE) was gleaned from locus pPp10 of the Johor population (0.951) and locus Ptri2 of the Terengganu population (0.799) respectively. In all P. pelagicus populations, the inbreeding coefficient (FIS) values were significantly different from zero (p < 0.05). However, a negative FIS value was detected for locus pPp2 of the Perak population, indicating a loss of heterozygosity in this particular population. The informative contents illustrated that all four pairs of microsatellite loci applied were highly polymorphic. There was no linkage disequilibrium found within the microsatellite loci tested through the sequential Bonferroni correction (data not shown). Nevertheless, significant deviations from the HWE expectations in P. pelagicus populations except for locus Ptri2 of the Negeri Sembilan population reflected the occurrence of heterozygote deficiency. This phenomenon in turn Table 3. Genetic diversity of Portunus pelagicus populations at four microsatellite loci Locus Perak Johor Negeri Sembilan Terengganu Sarawak N11 24 14 25 13 pPp2 Nα9 19 12 16 10 AR9.000 13.303 10.840 11.205 9.329 HO1.000 0.750 0.357 0.606 0.462 HE0.861 0.938 0.923 0.919 0.868 FIS -0.170 0.204 0.622 0.352 0.478 P0.002 0.000 0.000 0.000 0.000 pPp9 Nα8 16 12 13 9 AR8.000 11.240 10.622 10.579 8.771 HO0.272 0.375 0.643 0.360 0.385 HE0.848 0.887 0.889 0.890 0.898 FIS 0.689 0.583 0.284 0.601 0.582 P0.000 0.000 0.000 0.000 0.000 pPp10 Nα8 18 13 17 11 AR8.000 13.662 11.931 11.756 10.157 HO0.182 0.542 0.500 0.560 0.462 HE0.848 0.951 0.939 0.922 0.878 FIS 0.787 0.436 0.477 0.397 0.484 P0.000 0.000 0.000 0.000 0.000 Ptri2 Nα7 9 7 9 9 AR7.000 7.787 6.700 6.643 8.752 HO0.182 0.208 0.786 0.160 0.769 HE0.831 0.851 0.833 0.799 0.880 FIS 0.789 0.759 0.059 0.803 0.130 P0.000 0.000 0.137 0.000 0.003 N: Sample size; Nα: Number of allele; AR: Allele richness; HO: Observed heterozygosity; HE: Expected heterozygosity; FIS: Inbreeding coefficient; P: p-value (p < 0.05). page 7 of 12Zoological Studies 56: 26 (2017) © 2017 Academia Sinica, Taiwan was typically caused by the events of inbreeding and the presence of null alleles. Genetic Differentiation among Portunus pelagicus Populations Hierarchical results of AMOVA revealed that the individuals of Portunus pelagicus had a major contribution on the genetic variations of this species, with approximately 50% of total variance (Table 4). In contrast, only 4.13% of variance was accounted for the inter-population differentiations. Besides, the pairwise FST values unveiled significant genetic variations among the P. pelagicus populations (Table 5). The Sarawak population exhibited the highest degree of differentiations with the other populations, ranging from 0.04239 to 0.10237 (p < 0.05). The FST demonstrated between the Terengganu and Negeri Sembilan populations was relatively lower at 0.03569, signifying sufficient gene flow between these two populations. Moreover, assignment tests of P. pelagicus disclosed that nearly all individuals were correctly assigned to their original populations (Table 6). The assignment scores ranged from 19.599133.195, relating both the Perak and Johor populations. The Bayesian structure analysis on the other hand suggested that the most suitable K identified for P. pelagicus was K = 4 (Ln P(D) = -4265.5; Var[Ln P(D)] = 5096.5). The clustering roughly corresponded to geographic locations, with majority of individuals from Perak assigned to Cluster 1, Johor and Negeri Sembilan to Cluster 2, Terengganu to Cluster 3 and Sarawak to Cluster 4 (Fig. 3). Lastly, the bottleneck analyses elucidated that the Sarawak population involved in current population reduction (Table 7). The decrease Table 4. Hierarchical analysis of molecular variance (AMOVA) in Portunus pelagicus Source of variation Sum of squares Variance components Percentage of variation Among populations Among individuals within populations Within individuals 21.069 217.500 82.000 0.07735 0.85495 0.94253 4.13 45.60 50.27 Table 5. Estimation of FST among Portunus pelagicus populations via four microsatellite loci Populations Perak Johor Negeri Sembilan Terengganu Sarawak Perak - Johor 0.07921 - Negeri Sembilan 0.06698 0.05137 - Terengganu 0.05406 0.05358 0.03569* - Sarawak 0.10237 0.05199 0.04808* 0.04239* - *p < 0.05. Table 6. Assignment tests of Portunus pelagicus based on four microsatellite loci frequencies Assigned population CA (%) Origin Perak Johor Negeri Sembilan Terengganu Sarawak Perak 100 19.599 76.998 68.729 68.432 77.209 Johor 100 133.195 42.384 124.708 128.956 121.687 Negeri Sembilan 100 89.446 89.228 27.875 84.894 83.774 Terengganu 100 113.494 117.821 109.239 37.891 105.700 Sarawak 100 89.364 77.645 75.213 72.793 24.665 CA: Correct assignment. page 8 of 12Zoological Studies 56: 26 (2017) © 2017 Academia Sinica, Taiwan in population size of P. pelagicus from Sarawak was further evidenced by the infinite allele model (IAM) and the two phase model (TPM). However, the stepwise mutation model (SMM) strongly implied the absence of bottleneck incidents in all populations. In accordance, mode shift allele frequency distributions were perceived in all five P. pelagicus populations. DISCUSSION Genetic Differentiation of Portunus pelagicus Four out of six microsatellite loci including pPp2, pPp9, pPp10 and Ptri2 showed favourable polymorphic results for microsatellite analyses. Of the four microsatellite loci applied, Ptri2 acted as a pair of primers for cross-species amplification of Portunus pelagicus. The positive polymorphic bands acquired were comparable to the results established in other cross-amplification studies of blue swimmer crabs. Yap et al. (2002) tested the microsatellite loci developed for P. pelagicus on an unidentified species, Portunus sp. in the northern Australia and recognized considerable levels of polymorphisms. Similar outcomes were also achieved by Xu and Liu (2011) where crossspecies amplification of P. trituberculatus primers was carried out in two other portunid species, P. sanguinolentus and P. pelagicus. Undeniably, the flanking regions of microsatellite sequences within related taxa were highly conservative (Scribner et al. 1996). Crossspecies amplification was further enhanced when homologous loci in one species were successfully amplified in another species via a single primer pairs developed (Zardoya et al. 1996). Thus, the cross-amplification between P. trituberculatus and P. pelagicus loci in the current study was plausible. However, the cross-amplification of P. pelagicus using Ptri2 exhibited lower annealing temperature than the one reported by Xu and Liu (2011), from 57°C to 52°C. General assumption that cross-species amplification inclines to have lower annealing temperature than the amplification of the primitive species was practically supported (Zane et al. 2002; Esa et al. 2011). Besides, moderate allele frequencies were derived from this study through 87 samples and four microsatellite loci, ranging from 34 to 14 alleles. The allele frequencies obtained were further implemented in the exact probability tests. These tests are not biased by low allele frequencies, signifying low mutation rates and hence are suitable for microsatellite analyses (Raymond and Rousset 1995; Chakraborty et al. 1997). The mean observed heterozygosity (HO) at Table 7. Current bottleneck evidences within populations of Portunus pelagicus I.A.M. T.P.M. S.M.M. Mode shift 60% 70% 80% Perak 0.562500 0.562500 0.843750 0.906250 0.906250 Y Johor 0.437500 0.906250 0.906250 0.906250 0.968750 Y Negeri Sembilan 0.093750 0.156250 0.156250 0.562500 0.906250 Y Terengganu 0.156250 0.562500 0.562500 0.843750 0.906250 Y Sarawak 0.031250* 0.031250* 0.031250* 0.031250* 0.562500 Y I.A.M.: Infinite allele model; T.P.M.: Two phase model; S.M.M.: Stepwise mutation model; Y: Yes; N: No. *p < 0.05. Fig. 3. Bar plot of Portunus pelagicus populations. 1: Perak; 2: Johor; 3: Negeri Sembilan; 4: Terengganu; 5: Sarawak. page 9 of 12Zoological Studies 56: 26 (2017)