IDENTIFICATION OF QPX-RESISTANCE MARKERS BY GENOME-WIDE CANDIDATE-GENE ASSOCIATION STUDY IN THE HARD CLAM  

The hard clam Mercenaria mercenaria supports important aquaculture industries along the Atlantic Coast of the United States. Since the 1990's, farmed clams in several Atlantic states have suffered serious mortalities due to a disease caused by a protistan parasite Quahog Parasite Unknown (QPX). The identification of disease-resistance markers may facilitate marker-assisted selection and improve selection efficiency as breeding decisions are often made in years of low disease pressure. Disease-resistance markers can be identified by several approaches including QTL mapping, family and population-based association studies using random genetic markers or candidate genes. Genome-wide association study (GWAS) is a powerful approach although the required high-density and high-throughput genotyping platforms are expensive to develop. Genome-wide candidate-gene association studies (GCAS) may provide a cost-effective alternative by narrowing the target regions. In GCAS, all candidate genes in the genome that are related to a particular trait, as indicated by function or transcriptional profile, are collected and targeted for genotyping by next-generation sequencing. Variations in these candidate genes are more likely associated with phenotypes of the trait than random genetic variations. In this study, we tested a limited GCAS for the identification of QPX-resistance markers in the hard clam.

We obtained a set of 373 candidate genes that are involved in immune response from hard clam transcriptomes. Primers were designed with the Ion AmpliSeq^TM Designer to amplify all candidate genes in a single-tube multiplex PCR. Hard clams were collected from three stocks before deployment in Massachusetts and New Jersey. The deployed clams were sampled again after field mortalities (47 - 93%) caused primarily by QPX. DNAs from 80 - 100 clams were pooled in equal amounts producing 3 before and 6 after-mortality DNA pools. The pooled DNAs were used as templates for amplification of 373 candidate genes with the AmpliSeq^TM primer panel. Amplified products were sequenced to about 1000x per gene with the Ion Torrent PGM 400 bp module. Of the 373 genes targeted, 98 genes were successfully amplified and sequenced in all 9 samples. Single-nucleotide polymorphisms (SNPs) and indels were identified and analyzed for post-mortality frequency shifts. Variations in 12 genes showed significant allele frequency shifts (>15%) including genes encoding metalloproteinase, interferon-induced proteins, complement factor b and several proteins related to apoptosis such as tumor necrosis factor (TNF) receptor-associated factor, TNF ligand superfamily member 6, apoptosis-stimulating of p53 protein, programmed cell death protein 2. Some loci showed consistent allele frequency shifts in all three stocks and at both sites, while others showed divergent selection at the two sites in all stocks, indicating allele-specific local adaptation. The markers identified in this study are likely associated with QPX resistance and useful for marker-assisted selection.