GPR30 is necessary for estradiol-induced desensitization of 5-HT1A receptor signaling in the paraventricular nucleus of the rat hypothalamus

Abstract

Estrogen therapy used in combination with selective serotonin reuptake inhibitor (SSRI) treatment improves SSRI efficacy for the treatment of mood disorders. Desensitization of serotonin 1A (5-HT(1A)) receptors, which takes one to two weeks to develop in animals, is necessary for SSRI therapeutic efficacy. Estradiol modifies 5-HT(1A) receptor signaling and induces a partial desensitization in the paraventricular nucleus (PVN) of the rat within two days, but the mechanisms underlying this effect are currently unknown. The purpose of this study was to identify the estrogen receptor necessary for estradiol-induced 5-HT(1A) receptor desensitization. We previously showed that estrogen receptor β is not necessary for 5-HT(1A) receptor desensitization and that selective activation of estrogen receptor GPR30 mimics the effects of estradiol in rat PVN. Here, we used a recombinant adenovirus containing GPR30 siRNAs to decrease GPR30 expression in the PVN. Reduction of GPR30 prevented estradiol-induced desensitization of 5-HT(1A) receptor as measured by hormonal responses to the selective 5-HT(1A) receptor agonist, (+)8-OH-DPAT. To determine the possible mechanisms underlying these effects, we investigated protein and mRNA levels of 5-HT(1A) receptor signaling components including 5-HT(1A) receptor, Gαz, and RGSz1. We found that two days of estradiol increased protein and mRNA expression of RGSz1, and decreased 5-HT(1A) receptor protein but increased 5-HT(1A) mRNA; GPR30 knockdown prevented the estradiol-induced changes in 5-HT(1A) receptor protein in the PVN. Taken together, these data demonstrate that GPR30 is necessary for estradiol-induced changes in the 5-HT(1A) receptor signaling pathway and desensitization of 5-HT(1A) receptor signaling.

Figure 1

EB treatment induces 5-HT_1A desensitization. Plasma OT (A) and ACTH (B) levels in response to saline or (+)8-OH-DPAT challenge in rats treated with oil or EB. Data are presented as mean ±SEM (n=6–8/group). *Significantly different from saline-challenged animals with same treatment, p<0.0001; #significantly different from oil/(+)8-OH-DPAT-treated animals, p<0.005 by Student-Newman-Keuls post hoc test.

Figure 2

EB-induced changes in 5-HT_1A signaling pathway. A) Left, characterization of RGSz1 antibody. RGSz1 protein at approximately 27kD was detected in HEK293 cells transfected with a vector expressing RGSz1 protein but not empty-vector transfected cells. In rat PVN, higher molecular weight proteins were also detected most likely due to post-translational modifications. Lane 1, HEK293 cells transfected with RGSz1. Lane 2, HEK293 cells transfected with empty vector (pcDNA). Lane 3, PVN tissue lysate. Right, Western blot of PVN protein after oil or EB treatment with antibodies against GPR30, 5-HT_1A, Gαz, and RGSz1, with β-actin as a loading control. B) Quantitation of protein levels after EB pretreatment. Bands were analyzed densitometrically (integrated optical density, IOD). Each band was normalized to β-actin and expressed as percent of control (oil). Data are expressed as mean ± SEM (n=6), *p<0.05, **p<0.001 by Student-Newman-Keuls post hoc test. C) qPCR of mRNA isolated from PVN tissue, normalized to control (oil). ΔCt was calculated as the target gene – TBP mRNA for each sample; ΔΔCt was calculated as ΔCt for the experimental condition − ΔCt for the control condition, for each target gene. Changes in mRNA levels are expressed as mean 2^−ΔΔCt ± SEM (n=4–6), *p<0.05 by Student-Newman-Keuls post hoc test.

Figure 3

Generation and evaluation of recombinant adenoviruses containing GPR30-siRNAs. A) Selection of GPR30-siRNAs: SOS-GPR30-siRNA-HUS (left panels), SOS-GPR30-mis-siRNA- HUS (right panels), and SOS-GPR30-HUS (control) (right panels) constructs were transfected into HEK293 cells. The number and brightness of GFP-containing cells were observed three days after the transfection. B) Diagram of PVN injection site. F = fornix. Vertical line represents the 3rd ventricle. PVN is within the dashed triangle. C) An example of RFP expression in the PVN five days after unilateral injection of GPR30-siRNA-Ads. 3V = 3rd ventricle. D) Examples of Western blot for GPR30 protein from PVN tissue with unilateral injection of saline, control-Ad, GPR30-mis-Ads, and GPR30-siRNA-Ads five days after infection. E) Quantitation of GPR30 knockdown. Bands were measured densitometrically and normalized to β-actin. The percent inhibition of GPR30-siRNA-Ads was calculated by comparing the ratio of GPR30:β-actin between the injected side (I) and the contralateral side (C) of each animal (% inhibition = (1−I/C) × 100%). Data are presented as mean percent inhibition. ± SEM (n=4–5). *p<0.05 by Student-Newman-Keuls post hoc test.

Figure 4

Confirmation of GPR30 knockdown. A) Examples of Western blots of GPR30 and ERβ with β-actin as loading control in GPR30-siRNA-Ad- and control-Ad-injected PVN. B) Quantitation of GPR30 knockdown and ERβ protein levels in the PVN of GPR30-siRNA-Ad- and control-Ad-injected rats. Bands were measured densitometrically and normalized to β-actin. Data are expressed as mean percent of control-Ad ± SEM (n=4–7), *p<0.01 by Student-Newman-Keuls post hoc test.

Figure 5

Effects of GPR30-siRNA-Ads on plasma OT (A) and ACTH (B) responses to (+)8-OH-DPAT in EB-treated rats. Data are expressed as mean ± SEM, (n=7–10). *Indicates significantly different from saline-challenged animals with same treatment, p<0.0001, #indicates significantly different from control-Ad/oil-treated rats with same challenge, p<0.05 by Student-Newman-Keuls post hoc test.

Figure 6

GPR30-siRNA-Ads prevented EB-induced reductions of 5-HT_1A receptor protein in the PVN. A) Representative Western blots of 5-HT_1A receptor and Gαz protein levels with β-actin loading control in the PVN of rats injected with control-Ad or GPR30-siRNA-Ad followed by treatment with oil or EB. B) Quantitation of Western blots. Bands were measured densitometrically and normalized to β-actin. Data are expressed as mean percent of control-Ad/oil, ± SEM (n=7–12). *p<0.001 by Student-Newman-Keuls post hoc test.