Female genomic response to mate information

Abstract

Females should be choosier than males about prospective mates because of the high costs of inappropriate mating decisions. Both theoretical and empirical studies have identified factors likely to influence female mate choices. However, male-male social interactions also can affect mating decisions, because information about a potential mate can trigger changes in female reproductive physiology. We asked how social information about a preferred male influenced neural activity in females, using immediate early gene (IEG) expression as a proxy for brain activity. A gravid female cichlid fish (Astatotilapia burtoni) chose between two socially equivalent males and then saw fights between these two males in which her preferred male either won or lost. We measured IEG expression levels in several brain nuclei including those in the vertebrate social behavior network (SBN), a collection of brain nuclei known to be important in social behavior. When the female saw her preferred male win a fight, SBN nuclei associated with reproduction were activated, but when she saw her preferred male lose a fight, the lateral septum, a nucleus associated with anxiety, was activated instead. Thus social information alone, independent of actual social interactions, activates specific brain regions that differ significantly depending on what the female sees. In female brains, reproductive centers are activated when she chooses a winner, and anxiety-like response centers are activated when she chooses a loser. These experiments assessing the role of mate-choice information on the brain using a paradigm of successive presentations of mate information suggest ways to understand the consequences of social information on animals using IEG expression.

Fig. 1.

Aquaria used for both the female choice part of the experiment (Upper) and for exposing females to their preferred male (P) either winning or losing a fight to their nonpreferred male (NP) (Lower). Dashed lines within the figures represent clear barriers. Solid lines represent both clear and opaque barriers. In these tanks, clear and opaque barriers were used to expose or occlude parts of the tank.

Fig. 2.

Mean (+SE) of relative gene expression of (A) c-fos and (B) egr-1 for each of the six nodes of the social behavior network, plotted as a function of whether females saw their preferred males win (filled bars) or lose (open bars) a fight. Asterisks above pairs of mean values indicate significant differences (t tests, corrected for multiple comparisons). Between panels A and B is a schematic sagittal section of the A. burtoni brain showing the approximate locations of the brain regions sampled. Cce, cerebellum; R, Raphe nucleus. As described in the text, all these brain regions are interconnected, sensitive to steroid hormones, and show high densities of androgen receptors (12, 13).

Fig. 3.

Average principal component 1 (PC1) scores of each female as a function of principle component 2 (PC2) scores for egr-1 (Upper) and c-fos (Lower). Females who saw their preferred males win a fight formed a significant cluster distinct from females who saw their preferred males lose a fight. This clustering is based exclusively on PC1 loadings. PC1 loadings suggest a continuum of brain activation from anxiety-like (large positive loading from the LS) to reproductive (large positive loadings from the POA, Vm, and AH), whereas PC2 could suggest a stress axis with large positive loadings from the LS, Dm, and PAG.