Estrogen modulates learning in female rats by acting directly at distinct memory systems

Abstract

Physiologically high levels of circulating estradiol enhance the use of place learning and impair the use of response learning to find food on a land maze. These two types of learning are impaired by lesions of distinct neuronal structures, i.e. the hippocampus and striatum, respectively. Moreover, it has been shown in male rats that compromising hippocampal function can promote the use of response learning, while compromising striatal function can promote place learning. These findings suggest an ongoing competition between the hippocampus and striatum during cognition, such that intact functioning of one structure somehow obstructs the relative participation of the other. The goal of this study was to determine if estrogen's opposing effects on place and response learning in female rats are due to direct actions, either independent or interacting, at the hippocampus and striatum. We infused 0.5 microM 17beta-estradiol 3-sulfate sodium or vehicle bilaterally into the dorsal hippocampus or dorsolateral striatum of ovariectomized young adult female rats, 48, 24 and 2 h before training. Rats were tested on one of three appetitive tasks in a Y-maze: place learning, response learning, or response learning with reduced visual cues (cue-poor condition). Intrahippocampal estradiol infusions enhanced place learning, reversing a cannula-induced impairment, whereas intrastriatal infusions had no effects on place learning. Estradiol infusions into neither structure significantly affected response learning when extramaze cues were visible. However, in the response task, cue-poor condition, intrastriatal but not intrahippocampal infusions impaired learning. These data demonstrate that estrogen modulates place and response learning at the hippocampus and striatum respectively, most likely through independent actions at these two structures.

Figure 1

Infusion sites aimed at the A) dorsal hippocampus or overlying cortex and B) dorsolateral striatum or overlying cortex. Circle shading represents the task used during training, black = place, gray = response, white = response, cue-poor.

Figure 2

Learning in the place task following intrahippocampal (A) or intrastriatal (B) infusions. A) E2-S into the hippocampus enhanced place learning relative to aCSF, reversing a cannula-related impairment. B) E2-S infusions into the dorsal striatum had no significant effects on place learning. Dark bars = unimplanted control (UC); white bar = aCSF; light gray bar = E2-S. Values reflect median trials to criterion with interquartile ranges. + = p < .005 vs UC; * = p < .01 vs aCSF

Figure 3

Learning in the response task following intrahippocampal (A) or intrastriatal (B) infusions. A) E2-S into the hippocampus was ineffective in modulating response learning. B) E2-S infusions into the dorsal striatum also had no significant effects on response learning when visual cues were present. Dark bars = unimplanted controls (UC); white bar = aCSF; light gray bar = E2-S. Values reflect median trials to criterion with interquartile ranges.

Figure 4

Learning in the response, cue-poor task following intrahippocampal (A) or intrastriatal (B) infusions. A) E2-S into the hippocampus had no significant effect on response learning when visual cues were obstructed. Cannula implantation and aCSF infusions into the hippocampus impaired learning relative to unimplanted controls (UC). B) E2-S infusions into the dorsal striatum significantly impaired response learning without visual cues. Dark bars = UC; white bar = aCSF; light gray bar = E2-S. Values reflect median trials to criterion with interquartile ranges. + = p < .05 vs UC; * = p < .02 vs aCSF.