Loss of PKBβ/Akt2 predisposes mice to ovarian cyst formation and increases the severity of polycystic ovary formation in vivo

Abstract

Ovarian cysts affect women of all ages and decrease fertility. In particular, polycystic ovarian syndrome (PCOS), in which multiple follicular cysts develop, affects 5-10% of women of reproductive age and can result in infertility. Current non-invasive treatments for PCOS can resolve cysts and restore fertility, but unresponsive patients must undergo severe ovarian wedge resection and resort to in vitro fertilization. PCOS is related to the deregulation of leutinizing hormone (LH) signaling at various levels of the hypothalamic-pituitary-ovarian axis and resultant hyperproduction of androgens. Because insulin resistance and compensatory hyperinsulinemia are observed in 50-70% of individuals with PCOS, deregulated insulin signaling in the ovary is considered an important factor in the disease. Here we report that aged mice specifically lacking the PKBβ (also known as Akt2) isoform that is crucial for insulin signaling develop increased testosterone levels and ovarian cysts, both of which are also observed in insulin-resistant PCOS patients. Young PKBβ knockout mice were used to model PCOS by treatment with LH and exhibited a cyst area that was threefold greater than in controls, but without hyperinsulinemia. Thus, loss of PKBβ might predispose mice to ovarian cysts independently of hyperactive insulin signaling. Targeted therapeutic augmentation of specific PKBβ signaling could therefore provide a new avenue for the treatment and management of ovarian cysts.

Fig. 1.

Specific loss of PKBβ in aged mice results in the development of severe ovarian cysts with an increase in the thecal-interstitial cell population. (A) 91- to 120-week-old WT (i) and PKBβ KO (ii) mice with distended abdomens. (B) Cystic ovaries isolated from WT (i) and PKBα KO (iii) mice show no atresia or small ovarian cyst formation, whereas PKBβ KO mice show severe ovarian cyst formation. (C) Increased stromal accumulation in 91-week-old (early; i, iii) and 120-week-old (late; ii, iv) aged PKBβ KO mice, shown by H&E staining (i, ii). This accumulation reflects increased thecal-interstitial hyperplasia, as indicated by positive vimentin immunohistochemical staining (iii, iv). 40× magnification.

Fig. 2.

Aged PKBβ KO mice show active LH signaling in ovarian cysts, which display increased steroidogenic signaling and lipid accumulation, resulting in increased serum testosterone. (A) WT and PKBβ KO mice exhibit circulating serum LH, with no significant difference in hormone levels. (B) PKBβ KO ovarian cysts display both active CREB (i–iii) and ERK (iv–vi) signaling, which are required for steroidogenesis. ERK is located at the cystic lumen (arrowheads), and increases with the severity of the cysts and age of the KO mice (ii, iii, v, vi) but is absent in WT mice (i, iv). Magnification: 100×. (C) PKBβ ovarian cysts (ii, iv) display increased lipid accumulation adjacent to the cystic lumen, a prerequisite for conversion to steroids; this is absent in WT mice (i, iii). Magnifications: 40× and 100×. (D) Consistent with increased active steroidogenesis, PKBβ KO mice show increased serum testosterone levels compared with WT mice.

Fig. 3.

Loss of PKBβ has no significant impact on normal ovarian steroidogenic signaling or reproductive function in young WT or PKBβ KO mice. (A) Steroidogenic signaling through CREB (i, ii) and ERK (iii, iv) in WT (i, iii) and PKBβ KO (ii, iv) mice. (B) Circulating serum hormone levels of testosterone (i) and estrodiol (ii) in WT and PKBβ KO animals. (C) PKBβ KO mice are fertile and produce litter sizes similar to those of WT animals. HT, heterozygous.

Fig. 4.

Induction of PCOS via tonic LH administration increases the severity of ovarian cysts observed in PKBβ KO ovaries, with cyst formation being associated with ERK activation and lipid accumulation in steroidogenically active ovaries. (A) PKBβ KO ovaries showed an approximately threefold increase in ovarian cyst area in LH-treated ovaries (vi, viii) compared with WT (v, vii), independent of administration of a GnRHAnt. Treatment of WT and PKBβ KO mice with vehicle (i, ii) or GnRHAnt (iii, iv) alone had no effect on cyst formation. Values are shown in the graph on the right. (B) Steroidogenic signaling was active and seen in ovaries from both WT (i, iii, v, vii) and PKBβ KO (ii, iv, vi, viii) mice treated with LH. ERK, however, was also strongly active in thecal cells, which accumulated and lined large follicles predominantly in PKBβ KO ovaries (arrowheads). (C) Increased lipid accumulation in ovaries treated with LH was also observed in PKBβ KO mice (ii, iv) in areas with active androgen steroidogenesis [indicated by active 3β-HSD staining (v–viii)] compared with WT (i, iii). All magnifications for IHC are 100×.

Fig. 5.

Model outlining the pro-androgenic contributions of PKBβ loss to LH-driven pathogenic cyst formation. In normal physiology, the negative feedback along the HPO axis ensures that steroidogenic signaling is tightly controlled. In pathogenic scenarios of ovarian cyst formation, such as the formation of simple cysts in aged mice or cyst formation in PCOS, a lack of PKBβ combined with increased LH and activated LHR signaling results in increased ERK1/2 activation, lipid accumulation and testosterone production, leading to granulosa cell death and cyst formation.