Regulation of anxiety and initiation of sexual behavior by CREB in the nucleus accumbens

Abstract

Sexual deficits and other behavioral disturbances such as anxiety-like behaviors can be observed in animals that have undergone social isolation, especially in species having important social interactions. Using a model of protracted social isolation in adult rats, we observed increased anxiety-like behavior and deficits in both the latency to initiate sexual behavior and the latency to ejaculate. We show, using transgenic cAMP response element (CRE)-LacZ reporter mice, that protracted social isolation also reduces CRE-dependent transcription within the nucleus accumbens. This decrease in CRE-dependent transcription can be mimicked in nonisolated animals by local viral gene transfer of a dominant negative mutant of CRE-binding protein (CREB). We previously showed that this manipulation increases anxiety-like behavior. We show here that it also impairs initiation of sexual behavior in nonisolated animals, a deficit that can be corrected by anxiolytic drug treatment. This local reduction in CREB activity, however, has no influence on ejaculation parameters. Reciprocally, we used the viral transgenic approach to overexpress CREB in the nucleus accumbens of isolated animals. We show that this local increase in CREB activity completely rescued the anxiety phenotype of the isolated animals, as well as their deficit in initiating sexual behavior, but failed to rescue the deficit in ejaculation. Our data suggest a role for the nucleus accumbens in anxiety responses and in specific aspects of sexual behavior. The results also provide insight into the molecular mechanisms by which social interactions affect brain plasticity and behavior.

Fig. 1.

CREB and sexual behavior in double-housed males. (A) Schematic of bilateral cannula placement for nucleus accumbens shell injections. (B) CREB immunohistochemistry in nucleus accumbens shell after HSV-LacZ (Left) or HSV-CREB (Right) injections. (C) The initiation of sexual behavior with a receptive female, as measured by latency to the first mount, is impaired after microinjection of HSV-mCREB, but not HSV-CREB, into the nucleus accumbens shell (F_3,44 = 2.83, P < 0.05; ^*, P < 0.04 against other three groups). The initiation of subsequent copulatory series is not affected by CREB manipulation. (D) Similar results are observed when considering the latency to intromission (F_3,44 = 3.92, P < 0.015; ^*, P < 0.015 against other three groups). (E) The delay between the first intromission and the first ejaculation is not affected by CREB manipulation. (F) The total number of mounts necessary to reach ejaculation is not affected by CREB manipulation.

Fig. 2.

Diazepam reversal of HSV-mCREB phenotype. (A) The increased delay to initiate sexual behavior, observed after HSV-mCREB microinjection, is corrected by diazepam (0.75 mg/kg, s.c.) (^*, P < 0.045 against other groups; n = 6-8 per group). (B) The initiation of the second copulatory series is not affected by the diazepam injection.

Fig. 3.

Influence of social isolation on CRE-dependent transcription. (A) CRE-LacZ mice isolated for 10-12 wk show decreased density of β-gal-positive nuclei within the shell of the nucleus accumbens (^*, P < 0.013). GH, group-housed controls; SH, adult animals isolated for 10-12 wk. (B) Schematic of a brain coronal section at the level of the nucleus accumbens. (Inset) Field view in C and D. (C) β-gal immunostaining in the nucleus accumbens of a group-housed CRE-LacZ mouse. (D) β-gal immunostaining in the nucleus accumbens of an adult CRE-LacZ mouse isolated for 10-12 wk, which shows a selective reduction in CRE activity in the shell region.

Fig. 4.

Influence of social isolation on sexual behavior and anxiety. (A) Isolated animals spend less time in the open arms of an elevated plus-maze (^*, P < 0.015). DH, double-housed controls; SH, animals isolated for 10-12 wk. (B-D) Both latency to initiate sexual behavior (^*, P < 0.035) (B and C) and latency to ejaculate after intromission (^*, P < 0.025) (D) are delayed in isolated animals. (E) In isolated animals, the deficit in the time spent in open arms of an elevated plus-maze is corrected by local microinjection of HSV-CREB. (F-H) In isolated animals, the microinjection of HSV-CREB into the nucleus accumbens shell corrects the increased latency to initiate sexual behavior (^*, P < 0.04) (F and G), but it does not correct the increased latency to ejaculate (H).