A panel of polymorphic EST-derived microsatellite loci for the bay scallop (Argopecten irradians)

Abstract

The constant progress in aquaculture research has provoked an increasing interest in molecular tools for population analysis, marker-assisted early selection, controlled crosses and fertilization analysis (Weising et al., 1998; Roberts et al., 2005). Microsatellite markers have proved to be one of the best choices. Microsatellites, or simple sequence repeats (SSRs), are small tandemly repeated sequences that are abundantly scattered throughout a genome (Toth et al., 2000). The advantages of microsatellites include co-dominance, high reproducibility and hypervariability (Jarne & Lagoda, 1996). Consequently, the use of microsatellite markers has expanded considerably over the past decade, both in the number of studies and the number of organisms, primarily due to their facility and power for population genetic analyses (Xu et al., 2001; Senanan et al., 2004). The bay scallop, Argopecten irradians, widely distributed along the Atlantic and Gulf coasts of the United States, is an hermaphroditic bivalve mollusc with a single, large adductor muscle (Roberts et al., 2005). In China, the bay scallop became one of the main cultured bivalves after being transplanted from USA in the early 1980s. Recently, 17 EST-SSRs have been developed (Roberts et al., 2005; Zhan et al., 2005). In this study, we describe the mining of microsatellites via bioinformatic analysis of the bay scallop ESTs, and nine new polymorphic EST-SSRs are also reported. The EST database (dbEST) of bay scallop in NCBI was mined for microsatellites with a homemade PC program, Repeat Reporter 1.5, with the parameters: seven repetitions for di-, five repetitions for tri-, four repetitions for tetraand three repetitions for pentaand hexanucleotides. The EST sequences containing SSRs were clustered using BioEdit program (http://www.mbio.ncsu.edu/BioEdit/bioedit.html). After cluster analysis, the unique SSR-containing sequences recently released to GenBank by Song et al. were identified. The sequences with the longest perfect repetitions and flanking regions were selected for PCR primer design by software Primer Premier 5.0 (www.PremierBiosoft.com/faq.html). PCR amplification and electrophoresis were performed as described by Zhan et al. (2005). The allelic diversity was assessed for 30 specimens sampled from the Huangdao hatchery, Shandong Province. The observed heterozygosity (Ho) was obtained using observed heterozygotes divided by total samples and the expected heterozygosity (He) was estimated using the formula He 1⁄4 12 P pi , where the pi is the frequency of the ith allele. The Markov chain method was employed to estimate the probability of significant deviation from Hardy-Weinberg equilibrium (HWE) using GENEPOP online version (http://wbiomed.curtin.edu. au/genepop/). Pairwise tests for linkage disequilibrium (LD) were performed using the program Fstat (http://www2.unil. ch/popgen/softwares/fstat.htm). Sequential Bonferroni corrections were applied for all multiple tests. A total of 7057 EST sequences were downloaded and used for the data mining analysis. A complete search of 2–6 nucleotide-motif microsatellites resulted in the identification of 193 microsatellite-containing ESTs, accounting for 2.73% of all the bay scallop ESTs. Of 193 SSR-containing ESTs, 71 sequences fell into 19 contigs, which represented genes sequenced more than once. Cluster analysis showed that TA/AT repeat type had the largest number of clusters (7), and two trinucleotide types, CAG and GAT, had formed two big clusters that contained 16 and 12 sequences, respectively. The remaining 122 ESTs were singletons, whose transcripts were sequenced only once. This study identified a total of 141 unique microsatellite-containing ESTs and the repeat number of the independent microsatellites ranged from 3 to 51. The bay scallop EST-SSRs contained a variety of simple sequence types. Diand trinucleotide repeats were the most abundant, which accounted for 69.95% of all SSR-containing ESTs, followed by 10.36%, 15.03% and 4.66%, respectively, for tetra-, pentaand hexanucleotide repeats. A total of 19 (9.84%) ESTs contained more than one microsatellite. Of the dinucleotide repeats, TA/AT was the most abundant, accounting for 78.48% of all dinucleotide repeats detected in the bay scallop ESTs. The second most abundant dinucleotide repeat type was GA/TC, accounting for 15.19%, while CA/TG repeats occurred only rarely (6.33%). No GC/CG repeats were detected in this study. Thirteen repeat types were included in the 70 trinucleotide SSR-ESTs, and the distribution of different types was uneven. The repeat types of ATG and AGC were abundant (27.14%) in all the independent trinucleotide microsatellites found. In contrast to diand trinucleotide repeat types, the result indicated that the tetra-, pentaand hexanucleotide types are distributed more evenly. Ten SSR-containing sequences were selected for the design of PCR primers (Table 1). When the PCR reactions were carried out with genomic DNA, all the 10 loci could be amplified and 9 revealed length polymorphisms. The average number of alleles of polymorphic loci was 6.4, ranging from 3 to 11, and the average Ho was 0.4259, ranging from 0.0667 to 0.8000. The locus AIMS023 had the lowest Ho (0.0667), while the locus AIMS024 had the greatest number of alleles (11) and Ho (0.8000) (Table 1). Among the nine polymorphic loci, the locus AIMS023 had the lowest He value (0.0961), and the other eight loci had the He values ranging from 0.5394 to 0.8253. The results of Markov chain tests showed that all the nine loci significantly deviated from HWE. Four loci (AIMS020, AIMS021, AIMS025 and AIMS026) deviated from HWE due to heterozygosity deficiency. The remaining five cases showed no excess or deficit of heterozygotes. Deviations from HWE expectations are common in bivalves and may have various causes (Gaffney & Scott, 1984; Li et al., 2003; Roberts et al., 2005), and it is beyond the scope of this paper to expand on this issue. Pairwise tests indicated that there was no LD between any loci. Presently, the application of microsatellite marker systems has been limited due to the costs involved in isolating, cloning, sequencing and characterizing microsatellite loci from the Correspondence: Z. Bao; e-mail: zmbao@ouc.edu.cn