Genetic parameters and genotype by environment interactions for harvest traits in the Pacific oyster (Crassostrea gigas)

A key aspect of selective breeding programs is whether the traits of interest respond similarly in different environments. The magnitude of the genotype by environment (G x E) interaction and heritability are responsible for this aspect. Commercial hatcheries in northern China supply batches of seed to oyster farmers each year, which are cultivated in a wide range of growing sites. However, the extent to which G x E interactions affect the ability to identify quality oyster lines has not been measured. Therefore, the objective of this study was to estimate the genetic parameters and G x E interactions for harvest traits in C. gigas cultured in northern China. Oysters from 40 mostly full-sib families (the offspring of 36 sires and 40 dams) were tested at three major commercial cultivation sites: Rushan, Hungdao and Rongchceng. We recorded shell height (SH), shell length (SL), shell width (SW), individual weight (IW), survival rate (SR), wet meat weight (WMW), meat ratio (MR) and shell shape (SS) at harvest. Compared with control line, selected line exhibited high superiority in survival rate at three sites. Heritability estimates for the harvest traits were low to high across different sites (0.55-0.75 for SH, 0.25-0.49 for SL, 0.20-0.33 for SW, 0.37-0.71 for IW, 0.10-0.12 for SR, 0.40-0.52 for WMW, 0.14-0.34 for MR and 0.18-0.30 for SS). Genetic correlations among growth traits were medium to high and positive, ranging between 0.52 +/- 0.15 (between SH and SW at Huangdao) to 0.90 +/- 0.04 (between SH and IW at Rongcheng). Nevertheless, we found significantly negative genetic correlations between growth traits and shell shape, an index that takes into account growth traits and shell shape should be developed to improve both simultaneously. Since it is not possible to disentangle this result from the correlation that mathematically determined alone, this result requires further study. Notably, genetic correlations between growth traits and survival were low, ranging from - 0.31 +/- 0.17 (between SL and SS at Huangdao) to 0.14 +/- 0.19 (between SW and SS at Rongcheng). G x E interactions were estimated as genetic correlations between the same trait measured on different oysters at three sites. Moderate G x E interactions were found for growth traits (0.45-0.87), meat yield traits (0.40-0.73) and shell shape (0.22-0.69), and weak G x E interaction was found for survival at harvest (0.80-0.86). This study reveals the presence of weak to high re-ranking of genotypes for harvest traits in C. gigas cultured in different commercial sites, suggesting that G x E interactions should be considered in oyster breeding program servicing multiple sites.