Heritability of fat accumulation in the house fly and its implication for the selection of nutritionally tailored phenotypes

Fats make up a large fraction of the larval biomass in insects and are of increasing relevance for industrial purposes. The quantity of fats accumulated during the larval period can vary greatly among individuals. To selectively breed favourable fat accumulation phenotypes for commercial purpose, a genetic component underlying these differences would be required. In this study we determined the heritability of larval fat accumulation in the common housefly (Musca domestica L.). A nested paternal half-sib breeding design resulted in a total of 47 full-sib families, which produced in total 633 larvae, each phenotyped for their dry weight and fat content (absolute and relative). Adult selection was strictly standardized to reduce the variation in fat content induced by differences in development time during the immature stages: sires and dams were separated according to their pupation day and emergence day, and subsequently pooled into randomized groups. Two animal models were built to estimate the heritability of larval fat accumulation: (1) a Markov Chain Monte Carlo linear mixed model (MCMCglmm) where three sets of priors were used, and (2) one based on linear mixed model using restricted maximum likelihood (ASReml). The results showed a moderate heritability for larval absolute fat content for both MCMCglmm (between 0.30 and 0.38, depending on the set of prior used) and ASReml (0.37 ± 0.11). Slightly higher estimates were obtained for the relative fat content with both the MCMCglmm (between 0.34-0.48) and ASReml (0.47 ± 0.10). In contrast, heritability estimates for larval dry weight were low and could be obtained only with the MCMCglmm model (between 0.11-0.22). This work provides valuable insights into the quantitative genetics of larval fat characteristics, a trait of high relevance for the emerging sector producing insects for feed and food.