Scientific Notes Polymorphism in Litopenaeus vannamei genes and cross-species amplification in other shrimp species

The objective of this work was to assess polymorphisms within Litopenaeus vannamei candidate genes related to growth performance, and a possible association between weight genotypes and phenotypes in two shrimp closed lines. Three EST-SSR (serine-arginine, 60S ribosomal, and troponin) and one ESTSNP (crustacyanin) were selected and evaluated in L. vannamei and seven other marine shrimp species, besides two freshwater species. Genotypes had no effect on total body weight. Although no association was observed in the sampled population, these markers showed amplification for most of the evaluated species and, therefore, may be useful in further genetic studies on shrimp species.

Molecular markers such as simple sequence repeats (SSR) and single nucleotide polymorphisms (SNP) within genes have showed their wide application in aquaculture in the last decade (Yu & Li, 2008).The association between genotypes and productive traits can be highlighted for several aquaculture species, including the description of a SSR region linked to resistance to White Spot Syndrome Virus (WSSV) disease, in Penaeus monodon (Mukherjee & Mandal, 2009), and the description of SNPs associated with survival and disease resistance in Litopenaeus vannamei (Marti et al., 2010).Besides, molecular markers in coding regions have high-transferability rates among closely related species.Several studies have described the potential use of SSR and SNP linked to expressed sequence tags (EST) in shrimp species with still poorly known genomes (Freitas et al., 2007;Santos et al., 2012).
The objective of this work was to evaluate polymorphisms in L. vannamei shrimp in three microsatellite loci and one SNP, within candidate genes related to the growth performance.All molecular markers were into the protein main chain.A possible association between the genotypes and high growth performance rates in two L. vannamei families was also tested for high-growth performance rates.In Three genes containing SSR -Serine-Arginine (Ser-Arg), 60S ribosomal (60Srib), and Troponin (Trop) -, and one containing one SNP marker -Crustacyanin (Crust) -were obtained from the ShEST Project, available at TSA database, BioProject PRJNA77889, under JR494104, JR494145, and JR494393 accession numbers, respectively.Sets of specific primer pairs were designed for each loci using the Primer 3 software (Rozen & Skaletsky, 1999), considering the flanking regions and the respective annealing temperatures for the SSR and SNP, as follows: serinearginine (F: GATATGGCAGGCTTCGTGAT, and R: TACCTGTCAAGCCAAGCT, at 58°C); 60S ribosomal (F: CAGCTACTGACATCAACTTC, and R: CTCCTTCTTGGCTTCCTCCT, at 58°C); troponin (F: AGCCTGTGTACGAGGAGGAA, and R: AAAATAACGCGACGACATCC, at 64°C), and crustacyanin (F: ATCGAAATGAGCGGCTACAA, and R: TGTGTGTCAAGCTAG, at 55°C).
Polymerase chain reactions (PCRs) were performed for DNA samples of 200 specific pathogen-free shrimps, (SPF) (Genearch Ltda., Rio do Fogo, RN, Brazil).The specimens belonged to families challenged for highsurvival and growth rates.Pleopods of 100 animals (50 males and 50 females) from two families (F1 and F3) were sampled and stored in alcohol 100% at -20 o C. Whole body weight of each animal was measured with analytical balance, and sex was identified by visualization of the external morphology.The samples of each family were firstly separated into two groups: the heavier group of 50 animals (25 females and 25 males) from F1 and F3 families, with the greatest weight gain rates; and the lighter group of 50 animals (25 females and 25 males), from those same families, with the lowest weight gain rates.The heaviest and lightest animals from each group were then separated into four subgroups.
Biological samples for the nine shrimp species were obtained from the Tissue Bank, at the Laboratory of Molecular Biodiversity and Conservation (University of São Carlos, São Carlos, SP, Brazil), and used in cross-amplification tests.DNA extraction was performed using the phenol protocol proposed by Sambrook et al. (1989).PCR was carried out in 10 µL reaction mixtures containing 50 ηg µL -1 DNA, 0.2 mmol L -1 dNTPs, 1 x PCR buffer [200 mmol L -1 Tris-HCl, pH 8.4, and 500 mmol L -1 KCl (Invitrogen Life Technologies)], 1.5 mmol L -1 MgCl 2 , 1 U of Taq DNA polymerase, 8 pmol µL -1 of reverse primer, 2 pmol µL -1 of the M13-tailed forward primer, and 8 pmol µL -1 of the fluorescently-labeled M13 primer for SSR loci (Schuelke, 2000).The M13 primer was not used for the locus with SNP marker.The thermal cycler was programmed for an initial denaturation of 5 min at 95°C, followed by 35 cycles of 30 s at 94°C, 30 s at the annealing temperature, 30 s at 72°C, and a final extension of 20 min at 72°C.
The annealing temperatures were established using gradient tests that were performed in a Mastercycler gradient thermal cycler (Life Technologies).Alleles were genotyped in a Illumina automated sequencer MiSeq and scored using the Geneious software (Kearse et al., 2012).Gene product was identified using blastN and blastX against nrNCBI (National Center for Biotechnology Information), SwissProt (Protein Sequence Database), InterPro (Protein Sequence Analysis and classification), and GO (Gene Ontology Consortium) databases.The PCR products obtained from the cross-species amplifications were also sequenced, and blastN (nrNCBI) searches were performed in order to confirm the described protein and its respective species.Allele frequencies were estimated using Genepop version 3.4 (Raymond & Rousset, 1995).The genotype frequencies and analysis of variance were calculated using GLM (general linear model) procedure with SAS (SAS Institute Inc., Cary, NC, USA) version 9.3.The effects from family, sex, and genotypes, for each locus, were also considered.Sex was treated as a fixed effect, and family and genotypes as random effects.The animal whole body weight was considered a dependent variable, and data significance was evaluated at 5% probability.
For the 60Srib protein, the frequencies obtained for the 333 bp and 239 bp alleles were 73 and 27%, respectively.The most frequent genotypes were: 333/333 (66%) in F3 family, in the animals with weight means (WM) equal to 15.3 g; 239/333 (62%) in F1 (WM = 14.6 g); and 239/239 (71%) also in F1 (WM 13.6 g).Considering the SNP C153T observed in the Crust gene, the frequency for the C allele was 94% and, for the T allele was 6%.The CC (WM = 15.0 g) and CT (WM = 13.9 g) genotypes were equally distributed between F1 and F3 families.
Although the weight means differed among the animals, significant differences were not observed.The SSR into Trop and Ser-Arg genes were both removed from the association analysis, as the first marker exhibited monomorphism, and the second one showed frequency of 98% for the 594/594 genotype.
The effects and significance of family and group were tested in a nested model comparison (test A in the present study), considering sex and genotype effects individually for each locus.Only families and groups significantly affected the total variance for weight (p<0.001).The determination coefficient (R 2 ) was close to 83% for all loci, and family and group showed high-mean square (MS) values, which highlights the great effect of family and group on weight variation for the assumed model (Table 1).
As described earlier, no significant values for genotype means were observed in the previous analysis.Therefore, the two groups were subdivided into four subgroups according to family, in order to verify whether the weight variance could be partially assigned to the genotypes (test B).The four subgroups were considered as follows: subgroups 1 and 2, higher and lower weights, respectively, for animals from the F1 family; and subgroups 3 and 4, higher and lower weights, respectively, for animals from the F3 family.The results for test B were the same as those for test A, with no association observed between genotypes and phenotype weights (Table 1).Some possible interactions between two loci, or among the loci, and the other effects were also tested, but, once more, no association could be observed.Considering that weight is a quantitative trait usually influenced by several genes of small effects each, it is expected that a greater number of loci would explain the variance observed on weight (Jung et al., 2013).Additionally, this variance may be assigned to several factors, such as environmental variance in phenotype expression or bias in the sampling (Visscher et al., 2008).Hence, further studies are recommended to perform association analyses for these and other genes, in different populations with larger sampling.
Regarding the cross-species amplification reactions, the best rates were found for the SSR into the Ser-Arg and 60Srib genes (89% each).This can be justified by the fact that their encoded proteins participate in major metabolic steps, such as splicing control and repairs in translation (Ge et al., 1991;Zhai et al., 2008), tending to be highly conserved among species.Patterns for all SSR loci were observed for P. monodon, F. indicus, F. brasiliensis and F. paulensis amplification (Table 2).Thus, it is possible that these   (Ma et al., 2011).Overall, the markers that did not show polymorphisms in the association analyses herein may be potentially useful for species with a small amount or even with no molecular markers described yet, such as most of the shrimp mentioned in this study.Therefore, the markers identified in this study, including those with no allele variants, such as Ser-Arg and Trop, may be applied in further population genetic approaches for L. vannamei and other closely related species.

Table 1 .
Analysis of variance of the effects (1) of genes, families, and lineages on shrimp weight (Litopenaeus vannamei).