Chromosomal transmission bias in laboratory hybrids between wild strains of the two European subspecies of house mice.

Laboratory crosses between wild strains of the two European house mouse subspecies Mus musculus domesticus (2n = 34) and M. m. musculus (2n = 40) were performed to analyze the selective processes involved in the non-introgression of centromeric regions of Robertsonian (Rb) fusions in the Danish hybrid zone. The chromosomal analysis of 226 backcross progeny from 22 reciprocal crosses showed that the segregation of the three Rb fusions present did not significantly differ from Mendelian expectations. However, a significant negative correlation was found between Rb transmission rates and the average litter sizes of the F1 pairs. Among the different models of selection discussed, the most likely one supported the existence of two opposing selective factors resulting in an overall compensation of chromosomal types in the backcross progeny. A two-phase selective process involving embryo competition was postulated with non-Rb carriers being favored during pre-implantation but disadvantaged after implantation. Such balanced selective pressures acting on musculus non-Rb centromeres are compatible with the steep slope and off-centered position of the chromosomal cline observed in the Danish hybrid zone. These results suggested that these selective factors may be more related to centromere origin (musculus or domesticus) than to centromere structure (Rb or non-Rb).