Renn Lab Thesis Project

2010 Reed Graduate


Gene expression profiles of female aggression in the Julidochromis cichlid


Differences between male and female behavior can be explained by differences in genetic makeup, sex hormones, and sexually dimorphic neuroanatomy. For many animals, the expression of sex-typical behavior is somewhat plastic and depends on the social environment. I use the African Julidochromis cichlid fish as a model system to determine the underlying physiological and molecular mechanisms that account fo
both differences in, and plasticity of, sex-typical behavioral phenotypes.


behavior results

Julidochromis is a genus of monogamous, biparental cichlids that show territorial and parental sex-typical behavioral roles. They are unique among biparental cichlids
because, in some Julidochromis species, the female is larger than her mate and provides the majority of territory defense for the pair. This reversal in behavior appears to be dependent on relative mate size, as experimental reversal of size ratios causes a sex-reversal in territorial behavior (figure 2). Thus, sex-roles in Julidochromis appear to be plastic and contingent on the social context of relative mate size. While this degree of role plasticity has not been reported in other monogamous cichlids, it very well may occur in other species where male and female body size ranges greatly overlap.


Figure 2. Correlations between testosterone and behavior (squares=Male / circles=Female).
* indicates signifiant correlation at p<0.05. (a) Time spent in nest for females (dashed line) and males (NS). (b) Territorial behavior females (dashed line) and males (NS).



Despite the lack of significant sex differences, testosterone levels were highly correlated with several behaviors from observations on the same day fish were sacrificed (the day after spawning). In females, testosterone was negatively correlated in time spent in nest (R2 = 0.98, p = 0.01) and positively correlated with approach neighbor (R2 = 0.92, p = 0.04) (Figure 3.10). Males showed no such correlation for time in nest or territorial approach, but there were correlations between testosterone and parental behaviors. For males, the log-transformed testosterone concentration was negatively correlated with mouth eggs (R2 = 0.99, p = 0.003) and fanning eggs (R2 = 0.85, p = 0.08) (Figure 2). There was no correlation between attack mate behavior and testosterone for either sex.These results also show that, although testosterone does not explain patterns of sex-typical behavior, testosterone is important for the regulation of behavior in J. transcriptus.

To examine the molecular changes that accompany this plastic change in sex-typical behavior, I used microarrays to compare the gene expression profiles of behavioral_dataJulidochromis transcriptus males and females in two different social contexts (female-larger vs. male-larger pairs) in which they would show di erent patterns of territorial aggression. Aggressive females (female-larger pairs) and aggressive males (male-larger pairs) were found to have very di erent gene expression pro les, but there was a core set of genes whose expression level correlated with aggressive behavior in both sexes.

Figure 3: Venn diagram showing the number of genes that showed similar expression patterns between J. transcriptus reverse pairing female-larger pairs (Aggressive females/Non-aggressive males) and J. transcriptus natural paring male-larger pairs (Aggressive males/Non-aggressive females) and


This core set of similarly expressed genes may represent the genes whose expres-
sion is essential for aggressive behavior regardless of sex in J. transcriptus, while the
genes that showed different expression patterns in males and females may represent
sex-specific mechanisms for aggression. I also compared the gene expression pro les for J. transcriptus female-larger pairs to an expression dataset from female-larger pairs of a closely related species, J. marlieri, to see if the molecular basis for sex-roles was conserved or diverged between the species. There was a set of genes expressed similarly in aggressive females of both species, and this was distinct from those expressed in aggressive males. The majority of genes showed species-speci c variation in expression patterns, suggesting that Julidochromis species have undergone changes in gene expression patterns on an evolutionary time scale.

Figure 4: Venn diagram showing the number of genes that showed similar expression patterns between J. transcriptus reverse pairing female-larger pairs (Aggressive females/ Non-aggressive males) and J. marlieri natural pairing female-larger pairs (Aggressive females/Non-aggressive males).





This research was supported in part by a grant from the Reed College Biology Undergraduate Research Project Program (B.U.R.P.)