Temperature dependent sex ratios in selected lines and crosses with a YY-male in Nile tilapia ( Oreochromis niloticus)

Abstract

Sex determination in Nile tilapia is very labile. It was shown that the sex ratio in temperature treated groups (36 °C, 10 days) of Oreochromis niloticus can be selected for as a quantitative trait. The standardized test procedure involved the artificial reproduction of breeders with nine days of egg incubation at 28 °C and subsequent testing of fry at an elevated temperature of 36 °C for 10 days. Family selection was carried out upon the male percentages in treatment groups (high-line > 80%, low-line < 60%). Selected candidates were taken from corresponding full sib control groups, which were kept at 28 °C. After two generations of divergent selection for the male proportion in temperature treated groups, male percentages of 90% and 54% were obtained in the high- and low-line, respectively. Herein, the data for a third generation of selection are presented. Twelve families in the high-line and 20 in the low-line were tested. Average male percentages in the treatment groups of 92.7% and 50.4% were obtained. After three generations of selection the cumulated realized heritabilities for the temperature dependent sex ratio were 0.63 in the high- and 0.84 in the low-line, still indicating good potential for selection. Two families in the high-line exhibited only males, whereas in the low-line two families showed a feminising effect in the treatment groups. Generally elevated temperatures (> 34 °C) are considered to be masculinising in Nile tilapia, feminising effects so far have only been reported in genetic all-male progenies (YY, XY). In the third generation of selection, it could be shown that elevated temperatures might take over a feminising role through selection in genetically mixed sex (XX/XY) progenies. Consequently both, temperature-masculinising and -feminising genes, exist in the same population and can be accumulated through selection. However, the load of temperature-masculinising genes of dams from the high- or low-line had no effect on mean sex ratios in temperature treated groups when a YY-male was used as a sire. From an evolutionary point of view, the prevalence of both temperature-masculinising and -feminising effects might allow the sex ratio to equilibrate at the population level if random mating occurs. This in turn might counteract the evolution of sex chromosomes.