Influence of ERβ selective agonism during the neonatal period on the sexual differentiation of the rat hypothalamic-pituitary-gonadal (HPG) axis

Abstract

Background: It is well established that sexual differentiation of the rodent hypothalamic-pituitary-gonadal (HPG) axis is principally orchestrated by estrogen during the perinatal period. Here we sought to better characterize the mechanistic role the beta form of the estrogen receptor (ERβ) plays in this process.

Methods: To achieve this, we exposed neonatal female rats to three doses (0.5, 1 and 2 mg/kg) of the ERβ selective agonist diarylpropionitrile (DPN) using estradiol benzoate (EB) as a positive control. Measures included day of vaginal opening, estrous cycle quality, GnRH and Fos co-localization following ovariectomy and hormone priming, circulating luteinizing hormone (LH) levels and quantification of hypothalamic kisspeptin immunoreactivity. A second set of females was then neonatally exposed to DPN, the ERα agonist propyl-pyrazole-triol (PPT), DPN+PPT, or EB to compare the impact of ERα and ERβ selective agonism on kisspeptin gene expression in pre- and post-pubescent females.

Results: All three DPN doses significantly advanced the day of vaginal opening and induced premature anestrus. GnRH and Fos co-labeling, a marker of GnRH activation, following ovariectomy and hormone priming was reduced by approximately half at all doses; the magnitude of which was not as large as with EB or what we have previously observed with the ERα agonist PPT. LH levels were also correspondingly lower, compared to control females. No impact of DPN was observed on the density of kisspeptin immunoreactive (-ir) fibers or cell bodies in the arcuate (ARC) nucleus, and kisspeptin-ir was only significantly reduced by the middle (1 mg/kg) DPN dose in the preoptic region. The second experiment revealed that EB, PPT and the combination of DPN+PPT significantly abrogated preoptic Kiss1 expression at both ages but ARC expression was only reduced by EB.

Conclusion: Our results indicate that selective agonism of ERβ is not sufficient to completely achieve male-typical HPG organization observed with EB or an ERα agonist.

Figure 1

Effects on reproductive development. (A) Day of vaginal opening (DOV), a hallmark of pubertal onset in the female rat, was significantly advanced by EB and all doses of DPN compared to the vehicle treated controls. (B) There was no overall effect of treatment on bodyweight near the time of vaginal opening. (C) By 10 weeks of age, nearly all of the females neonatally administered EB or the 2 mg/kg dose of DPN (HIGH DPN) were acyclic while all of the vehicle control animals (Oil) were cycling normally. By 16 weeks of age, 75% of the MID DPN treated females and 85% of the LOW DPN treated females were acyclic compared to only one of the control females. (Means ± S.E.M., *P ≤ 0.001).

Figure 2

GnRH activation and LH release following hormone priming. (A) Immunolabeling of GnRH (green) and FOS (red) in the region surrounding the OVLT following hormone priming (Experiment 1). Representative double labeled cells are indicated by the blue arrows and single labeled GnRH neurons are indicated by the white arrows. Double labeled cells were first visualized at low magnification and then verified at higher magnification (inset). (B) The percentage of double labeled neurons was significantly reduced by neonatal DPN administration but the magnitude of the effect was substantially greater in the EB treated group. (C) LH levels were below the limit of detection (indicated by the red line, 1 ng/ml) in all females neonatally exposed to EB, and three of the five animals in the HIGH DPN group. LH levels were significantly lower in the EB and HIGH DPN groups compared to the vehicle treated control group (OIL, P ≤ 0.05). A trend for lower LH levels in the MID DPN group was also observed (P ≤ 0.06). The overall pattern of serum LH levels was largely reflective of GnRH and FOS co-labeling. (3v = 3^rd ventricle, Means ± S.E.M., *P ≤ 0.05, **P ≤ 0.001, † < 0.07).

Figure 3

Kisspeptin immunolabeling in the anterior hypothalamus. Fluorescence photomicrographs showing Kiss immunoreactive (-ir) labeling in the AVPV of the hormone primed females (Experiment 1). (A) Kiss-ir was localized to extended lengths of fibers within and surrounding the AVPV. (B) Kiss-ir fibers (blue arrows) were readily visible and quantifiable at high magnification with the confocal microscope. The density of Kiss immunostaining was significantly lower in the EB and MID DPN (1 mg/kg) treated groups compared to the control animals. (3v = 3^rd ventricle, ox = optic chiasm, (A) Scale bar = 50 μm, (B) Scale bar = 25 μm, Means ± S.E.M., *P ≤ 0.05, **P ≤ 0.005).

Figure 4

Kisspeptin immunolabeling in the mediobasal hypothalamus. Fluorescence photomicrographs showing Kiss immunoreactive (-ir) labeling in the ARC of the hormone primed females (Experiment 1). Kiss immunolabeling was considerably denser (A) than in the AVPV (see Figure 3) and localized to both fibers, (B blue arrows) as well as a small number of cell bodies that were often difficult to discern from the heavy fiber labeling (B, white arrows). (C) The density of Kiss-ir labeling was significantly lower in the females neonatally treated with EB but not DPN at any dose examined. (D) The number of Kiss-ir cell bodies did not statistically differ between groups. (3v = 3^rd ventricle, (A) Scale bar = 50 μm, (B) Scale bar = 25 μm, Means ± S.E.M., **P ≤ 0.001).

Figure 5

Hypothalamic kisspeptin mRNA expression in peripubertal females. (A) Representative autoradiographs depicting Kiss1 mRNA signal in the AVPV of periputertal females (Experiment 2). (B) Expression was significantly higher in the control (OIL) group compared to the PPT, DPN+PPT and EB groups at both PND24 and 33. (C) Within the ARC, expression was only significantly lower in the EB group compared to the OIL controls at both ages examined. Expression was not significantly altered in any of the other groups. (Scale bar = 500 μm, **P ≤ 0.005).