Rapid effects of estradiol on aggression depend on genotype in a species with an estrogen receptor polymorphism

Abstract

The white-throated sparrow (Zonotrichia albicollis) represents a powerful model in behavioral neuroendocrinology because it occurs in two plumage morphs that differ with respect to steroid-dependent social behaviors. Birds of the white-striped (WS) morph engage in more territorial aggression than do birds of the tan-striped (TS) morph, and the TS birds engage in more parenting behavior. This behavioral polymorphism is caused by a chromosomal inversion that has captured many genes, including estrogen receptor alpha (ERα). In this study, we tested the hypothesis that morph differences in aggression might be explained by differential sensitivity to estradiol (E2). We administered E2 non-invasively to non-breeding white-throated sparrows and quantified aggression toward a conspecific 10 min later. E2 administration rapidly increased aggression in WS birds but not TS birds, consistent with our hypothesis that differential sensitivity to E2 may at least partially explain morph differences in aggression. To query the site of E2 action in the brain, we administered E2 and quantified Egr-1 expression in brain regions in which expression of ERα is known to differ between the morphs. E2 treatment decreased Egr-1 immunoreactivity in nucleus taeniae of the amygdala, but this effect did not depend on morph. Overall, our results support a role for differential effects of E2 on aggression in the two morphs, but more research will be needed to determine the neuroanatomical site of action.

A behavioral polymorphism caused by a single gene inside a supergene

“…2B). 11 The animals treated with scrambled, control oligonucleotides exhibited the expected morph difference, 11 in that a bolus dose of E2 rapidly facilitated aggression in WS but not TS birds (morph ´ treatment interaction, p < 0.001 for both behaviors; Table S3,4). In contrast, knocking down ESR1 blocked E2-facilitated aggression in WS birds (treatment ´ antisense interaction, p < 0.035 for both behaviors) ( Fig.…”

“…3B,E). Thus, the degree to which a bird engaged in territorial aggression, which was markedly higher in the WS than TS birds, 9,11,16 was predicted not only by overall ESR1 expression in TnA but also specifically by the relative expression of the ZAL2 m allele, which is present only in WS birds.…”

“…Whereas in our previous studies we measured aggression in freeliving populations, 9,16,17 in the present study we used in a lab-based assay. 11,18 To confirm that ESR1 expression in TnA explains aggression as operationalized in this lab-based assay, we tested for a correlation between aggressive behavior and the level of ESR1 expression in TnA. Even when our analysis was limited to the control animals and "misses", in other words animals not receiving knockdown in TnA, ESR1 expression predicted aggression independently of morph (p < 0.01 for both behaviors) ( Fig.…”

“…Data were analyzed as described by Merritt et al 11 using generalized linear models (GLM) with a Gaussian distribution. The best-fit model was selected on the basis of AIC values.…”

“…9 Third, exogenous estradiol facilitates aggression in WS but not TS birds. 10,11 Although these observations suggest that ESR1 contributes to behavioral differences between the morphs, they provide only correlational evidence of such. To causally connect genotype to phenotype, we need evidence that morph differences in ESR1 expression are (1) directly attributable to cis-regulatory variation in ESR1 and (2) responsible for morph differences in behavior.…”