Y-chromosomes can constrain adaptive evolution via epistatic interactions with other chromosomes

Abstract

Background: Variation in the non-coding regions of Y-chromosomes have been shown to influence gene regulation throughout the genome in some systems; a phenomenon termed Y-linked regulatory variation (YRV). This type of sex-specific genetic variance could have important implications for the evolution of male and female traits. If YRV contributes to the additive genetic variation of an autosomally coded trait shared between the sexes (e.g. body size), then selection could facilitate sexually dimorphic evolution via the Y-chromosome. In contrast, if YRV is entirely non-additive (i.e. interacts epistatically with other chromosomes), then Y-chromosomes could constrain trait evolution in both sexes whenever they are selected for the same trait value. The ability for this phenomenon to influence such fundamental evolutionary dynamics remains unexplored.

Results: Here we address the evolutionary contribution of Y-linked variance by selecting for improved male geotaxis in populations possessing multiple Y-chromosomes (i.e. possessed Y-linked additive and/or epistatic variation) or a single Y-chromosome variant (i.e. possessed no Y-linked variation). We found that males from populations possessing Y-linked variation did not significantly respond to selection; however, males from populations with no Y-linked variation did respond. These patterns suggest the presence of a large quantity of Y-linked epistatic variance in the multi-Y population that dramatically slowed its response.

Conclusions: Our results imply that YRV is unlikely to facilitate the evolution of sexually dimorphic traits (at least for the trait examined here), but can interfere with the rate of trait evolution in both males and females. This result could have real biological implications as it suggests that YRV can affect how quickly a population responds to new selective pressures (e.g. invasive species, novel pathogens, or climate change). Considering that YRV influences hundreds of genes and is likely typical of other independently-evolved hemizygous chromosomes, YRV-like phenomena may represent common and significant costs to hemizygous sex determination.

Keywords: Drosophila melanogaster; Epistasis; Evolutionary constraint; Selection; Y-linked regulatory variation.

Fig. 1

Response to selection. a Males from the Y1 replicate populations (circles) responded to selection by increasing negative geotaxis, while males from the Y25 replicate populations (squares) showed no significant increase. Filled symbols represent geotaxis score prior to selection and empty symbols represent geotaxis after selection. b Females from the Y1 and Y25 replicate populations showed variation in their response to male selection on the genome. Significant differences between generation 1 and 16 within a population are marked by an asterisk and based on t-tests adjusted for multiple comparisons (k = 6). Error bars represent standard errors. See methods for LS means calculation

Fig. 2

Overall change in geotaxis score after 15 generations of selection. Y1 male geotaxis score (filled circle) increased significantly after selection, while Y25 male geotaxis score (filled square) did not. No overall change in Y1 female (open circle) or Y25 female (open square) was detected. Sex-specific significant differences between Y1 and Y25 geotaxis scores are marked by an asterisk and based on t-tests adjusted for multiple comparisons (k = 2). Data points represent the grand means of each sex-specific treatment (n = 3) and error bars represent standard errors. See methods for LS grand means calculation

Fig. 3

Creation of experimental populations. In the YRV parental generation (P1), 1 male from each isofemale line (1–25) was mated to 10 virgin females from the outbred population. In each subsequent filial generation (F1 - F10), 10 males from the previous line cross were mated to 10 virgin females from the outbred population. By the 10th filial generation, all lines were expected to be 99.9% similar, with the exception of the Y-chromosome. To establish each replicate population, 10 males from each line (n = 250 total) were placed in a large cage with 250 virgin females from the outbred population. The same approach was taken with the no-YRV population; however, all males shared the same Y-chromosome. YRV and No-YRV populations were created in parallel. This design maintained similar genetic variation between YRV and no-YRV populations through ample gene flow with outbred base population. All Y-chromosomes started at equal frequencies in the YRV population

Fig. 4

Geotaxis maze. Seven mazes were constructed in identical fashion, using 36 PVC cross junctions, 13 PVC elbow joints, 49 1 ml pipette tips embedded in hot glue, and numerous pieces of 0.5 in. diameter vinyl tubing to connect the cross junctions. Hot glue was also used to block unwanted horizontal movement (designated by diagonal lines) and pipette tips created one-way passage into an adjacent cross junction. Vials filled with food were placed at the terminal exits to attract flies. This maze provided flies with a total of 6 up or down choices