GPER Expression Prevents Estrogen-induced Urinary Retention in Obese Mice

Abstract

Long-term administration of exogenous estrogen is known to cause urinary retention and marked, often fatal, bladder distention in both male and female mice. Estrogen-treated mice have increased bladder pressure and decreased urine flow, suggesting that urinary retention in estrogen-treated mice is due to infravesicular obstruction to urine outflow. Thus, the condition is commonly referred to as bladder outlet obstruction (BOO). Obesity can also lead to urinary retention. As the effects of estrogen are mediated by multiple receptors, including estrogen receptors ERα and ERβ and the G protein-coupled estrogen receptor (GPER), we sought to determine whether GPER plays a role in estrogen-induced BOO, particularly in the context of obesity. Wild type and GPER knockout (KO) mice fed a high-fat diet were ovariectomized or left ovary-intact (sham surgery) and supplemented with slow-release estrogen or vehicle-only pellets. Supplementing both GPER KO and wild type obese mice with estrogen for 8 weeks resulted in weight loss, splenic enlargement, and thymic atrophy, as expected. However, estrogen-treated obese GPER KO mice developed abdominal distension, debilitation, and ulceration of the skin surrounding the urogenital opening. At necropsy, these mice had prominently distended bladders and hydronephrosis. In contrast, estrogen-treated wild type mice fed a high-fat diet only rarely displayed these signs. Our results suggest that, under conditions of obesity, estrogen induces BOO as a result of ERα-driven pathways and that GPER expression is protective against BOO.

Introduction

Chronic administration of exogenous estrogen as subcutaneous injections in vegetable oil or in slow-release pellets has long been known to cause urinary retention and marked, often fatal, bladder distention in both male and female mice (1–4). Cystometric studies show that estrogen-treated mice have increased bladder pressure, increased intervoid interval, prolonged micturition time, and decreased urine flow rate compared to untreated mice (5–7). These findings strongly suggest that urinary retention in estrogen-treated mice is due to infravesicular obstruction to urine outflow, a condition commonly referred to as bladder outlet obstruction (BOO) (8).

Because estrogen-induced urinary retention often occurs in the absence of any physical obstruction of the urethra, the underlying cause of restricted urine outflow is believed to be an increase in urethral tone (6,9,10). Urethral tone is controlled by nitrergic parasympathetic nerves. Nitric oxide produced by the neuronal nitric oxide synthase (nNOS) in these nerves stimulates NO-sensitive guanylyl cyclase in urethral smooth muscle, resulting in urethral relaxation (11,12). nNOS knockout (KO) mice and mice treated with pharmacologic inhibitors of nNOS develop bladder distention similar to that seen with estrogen administration (11,13). In estrogen-treated mice, there is a marked decrease in nNOS expression in the urethra, which is ultimately responsible for a significant increase in urethral tone (6). Mice made obese with high-fat diets may also develop urinary retention, which is associated with down-regulation of soluble guanylyl cyclase (14,15). Thus, both estrogen and a high-fat diet appear to increase urethral tone via a similar NO-activated guanylyl cyclase pathway.

ERα, rather than ERβ, appears to mediate the urinary retention observed in estrogen-treated mice. Urinary retention occurs in wild type male mice treated with testosterone plus estrogen; however, similarly treated ERα KO mice do not develop enlarged bladders, whereas similarly treated ERβ KO mice phenocopy wild type mice (16,17). Moreover, treatment with an ERα antagonist, but not an ERβ antagonist, partially blocks the effects of testosterone and estrogen administration (16). Bisphenols A, F and S, recognized as xenoestrogenic compounds, also induce lower urinary tract dysfunction in male mice (18).

The 7-transmembrane-spanning G protein-coupled estrogen receptor GPER has been implicated in multiple aspects of normal physiology and pathology (19–23), particularly related to the actions of estrogen, but the possible role(s) of GPER in modulating urethral tone have not been determined. GPER deficiency has been associated with obesity, insulin resistance and glucose intolerance, atherosclerosis, and a proinflammatory state (24,25), whereas selective GPER stimulation promotes weight loss, improves glucose tolerance, and reduces inflammation and blood pressure (26–31). GPER deficiency or antagonism can also lead to reduced superoxide production, resulting in reduced age-related fibrosis and hypertension (32,33).

We examined the effects of continuous exogenous estrogen supplementation on wild type and GPER KO mice fed a high-fat diet. GPER KO mice developed much more severe urinary retention than wild type mice fed the same. This suggests that GPER enhances urethral relaxation, an activity diametrically opposed to that of ERα, to prevent chronic urinary obstruction.

Materials and methods

Ethics statement

All procedures were carried out in accordance with the National Institutes of Health Guide for the Humane Care and Use of Laboratory Animals and approved by the University of New Mexico Institutional Animal Care and Use Committee as previously described (24).

Mouse cohorts and treatments

C57BL/6 mice were obtained from Harlan Laboratories (Indianapolis, IN); GPER-deficient (GPER KO) mice were provided by Dr. Jan S. Rosenbaum (Proctor & Gamble, Cincinnati, OH) (34) and subsequently backcrossed for 10 generations onto the C57BL/6 background. Mice were housed at the Animal Resource Facility at the University of New Mexico Health Sciences Center. Mice were maintained under a controlled temperature of 22–23°C with a 12-hour light, 12-hour dark cycle and fed a soy protein-free chow (2920X, Envigo, Indianapolis, IN) ad libitum, which provides 16% of calories as fat. Female mice were ovariectomized at 6–8 weeks of age. Immediately after ovariectomy, all mice were placed on a high-fat diet that provided 60% of calories from anhydrous milk fat (TD.09766, Envigo, Indianapolis, IN). Twelve weeks after surgery, 1.7 mg 60-day slow-release estradiol (estrogen) or vehicle pellets (Innovative Research of America, Sarasota, FL) were implanted subcutaneously in the neck area. Eight weeks after pellet implantation, mice were sacrificed with pentobarbital (59 mg/mL stock) at 0.1 mL/25 g mouse. Additional control groups included 1) wild type and GPER KO ovary intact female mice, not treated with estrogen pellets and 2) wild type and GPER KO female mice that were ovariectomized, and treated with vehicle pellets.

Necropsy and histopathology

At necropsy (performed on 5–6 mice per group), the volume of urine remaining in the bladder was determined. Body, spleen, thymus, and kidney weights were recorded. Bladder and spleen were fixed in neutral buffered formalin for routine paraffin embedding and H&E staining. Necropsy and histopathology were performed by a board-certified veterinary pathologist (DFK).

Water consumption

Water consumption was determined by placing mice in a computer-controlled indirect calorimetry system (Promethion^®, Sable Systems). Prior to data collection, mice were acclimatized in the chambers for 24 hours, after which data were collected for a subsequent 48 hours.

Statistics

Statistical analyses were carried out by one-way ANOVA followed by Tukey’s multiple comparisons test or, in the case of uterus weights, the Kruskal-Wallis test or unpaired t test (Graphpad Prism). A p-value of less than 0.05 was considered statistically significant.

Results

Estrogen treatment significantly reduced the body weight of wild type and GPER KO mice fed a high-fat diet and also produced significant thymic atrophy (by absolute weight, Figs 1A and B; as well as by % body weight, S1 Fig), consistent with previous studies describing the effects of exogenous estrogen (31,34–37). Spleens of estrogen-treated GPER KO mice were substantially enlarged, whereas spleens of wild type mice, although also enlarged, were significantly smaller than those of estrogen-treated GPER KO mice (Fig 1C; S1 Fig). Histology revealed that splenic enlargement was due to marked extramedullary hematopoiesis, characterized by a striking increase in red pulp (S2 Fig). Ovariectomy reduced the average weight of the uterus by 59% in ovariectomized GPER KO mice (intact: 0.0637 +/−0.0039 g vs OVX+Veh:0.0258+/−0.0025 g; p<0.0001, unpaired t test) and by 47% in ovariectomized wild type mice (intact: 0.1128 +/−0.0093 g vs OVX+Veh:0.0602+/−0.0185 g; p<0.05, unpaired t test) (Fig 1D). Estrogen treatment resulted in a substantial increase in uterine weight in both wild type and GPER KO mice (Fig 1D; S1 Fig), consistent with the established role of estrogen in uterine imbibition through ERα and the lack of a major role for GPER in uterine function (38–40).

Figure 1.

Effects of chronic estrogen supplementation on body and organ weights. A. Body weight. B. Thymus weight. C. Spleen weight. D. Uterus weight. Mouse cohorts included ovary-intact wild type (WT) and GPER KO (KO) mice, ovariectomized wild type (WT-OVX) and GPER KO (KO-OVX) mice, and estrogen pellet-supplemented wild type (WT-OVX-E2) and GPER KO (KO-OVX-E2) mice. Data shown are mean ± SEM, with significant p values indicated. n= 5–6 for all cohorts. A-C, one-way ANOVA followed by Tukey’s multiple comparisons test; D, Kruskall-Wallis test followed by Dunn’s multiple comparisons test. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Ovariectomized GPER KO mice fed the high-fat diet began to show evidence of abdominal distension 4–5 weeks following estrogen treatment. At the time of sacrifice, the mice were hunched and had an unsteady gait. All had moist skin ulcerations more than 1 cm in diameter surrounding the urogenital opening. Inguinal and paraaortic lymph nodes were markedly enlarged. These mice had prominently distended bladders. Bladders contained 1.5–2.5 mL of urine (Fig. 2A). Kidney weights were increased (Fig. 2B; S1 Fig). The statistically significant increase in kidney weight in ovariectomized GPER KO mice compared to intact GPER KO mice or to ovariectomized GPER KO mice was due to hydronephrosis, which was evident on the cut surface of the kidneys in all obese estrogen-treated GPER KO mice. In contrast, estrogen-treated wild type mice had a normal outward appearance at the time of sacrifice, with no perineal ulceration and a normal gait and activity level. Only one mouse had an enlarged bladder and hydronephrosis. Estrogen-treated wild type mice had somewhat increased kidney weights (Fig. 2B), although differences in kidney weights in intact wild type versus ovariectomized or ovariectomized estrogen-treated mice were not statistically significant. Histologic examination of the bladders displayed markedly thinned bladder walls in estrogen-treated GPER KO mice, compared to the much thicker bladder walls of mice that did not receive estrogen (Fig. 3). Thinning of the bladder walls was a direct consequence of stretching due to urine retention. Both the bladder urothelium and smooth muscle wall are highly adapted to changes in urine volume. Lastly, wild type and GPER KO mice that did not receive estrogen, whether ovariectomized or not, exhibited no thymus atrophy (Fig. 1B), no splenic enlargement (Fig. 1C), and no bladder distension (Fig. 2A).

Figure 2.

Effect of chronic estrogen supplementation on urine volume. A. Urine volume from each of the six cohorts. Data shown are mean ± SEM, with significant p values indicated. n= 5–6 for all cohorts. B. Kidney weights from each of the six cohorts. Data shown are mean ± SEM, with significant p values indicated. n= 5–6 for all cohorts. C. Water intake by estrogen-supplemented mice. Water intake was determined over 48 hours for wild type and GPER KO mice that were either ovary intact (Intact), ovariectomized (OVX) or ovariectomized and supplemented with estrogen pellets (OVX+E2). No significant differences in water intake were observed. Data shown are mean ± SEM. n = 4 for all cohorts. One-way ANOVA followed by Tukey’s multiple comparisons test. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Figure 3.

Representative H&E-stained sections of bladder walls from GPER KO mice fed high fat diets that were ovariectomized (A) or ovariectomized and supplemented with estrogen pellets (B). Note the markedly thinned bladder wall in the estrogen-treated GPER KO mouse compared to the much thicker bladder wall in the mouse that did not receive estrogen. In the bladder of the mouse that did not receive estrogen, the urothelium (U) is multilayered and the muscle wall (M) is contracted. In the bladder of the estrogen-treated mouse, both the urothelium (arrowheads) and the smooth muscle (M) of the bladder wall are markedly reduced in thickness. Scale bars represent 100 μm.

As estrogen treatment has been shown to increase water consumption (41), we measured water intake to determine whether GPER deficiency enhanced water consumption sufficiently to account for the difference in urine volume between GPER KO and wild type mice. As shown in Fig 4, there was a modest, but not statistically significant, increase in water consumption in estrogen-treated wild type and GPER KO mice. Thus, increased water intake in estrogen-treated GPER KO mice did not appear to account for differences in urine volume in the bladder.

Discussion

The type and dose of estrogen administered, and the strain, sex, and age of the treated mice determine the incidence and severity of estrogen-induced bladder outlet obstruction (1–4). In one study, ovariectomized C57BL/6J mice given estrogen by subcutaneous implantation of 90-day release 0.72 mg or 1.7 mg pellets every 90 days died by 8 months of age; necropsy and histopathology revealed urinary retention, hydronephrosis, and proliferative cystitis (42). In the present study, wild type C57BL/6J mice received only a single 1.7 mg 60-day slow-release subcutaneous estradiol pellet and were sacrificed after only 2 months; thus, it was not surprising that few of these wild type mice showed evidence of obstructive uropathy, even though fed a high-fat diet known to enhance urinary retention (14,15). Somewhat increased kidney weights in estrogen-treated wild type mice on high-fat diet and evidence of BOO in one of these mice suggest that some urinary retention was occurring within the timeframe of these studies. In the present study, it appeared that both GPER deficiency and E2 supplementation were required for the accelerated BOO onset observed.

Estrogen-induced obstructive uropathy is believed to be mediated via ERα (16,17), and ERα signaling is largely intact in GPER KO mice, as indicated by their unaltered reproductive capacity (34). Our study suggests that, in the context of BOO, the effects of estrogen mediated through ERα are enhanced or dominant in the absence of estrogen signaling through GPER. Thus, GPER activation likely functions to reduce urethral tone and enhance urine voiding. Multiple studies have demonstrated that GPER activation results in smooth muscle relaxation. GPER activity mediates vascular smooth muscle cell relaxation both in vitro/ex vivo (43–46) and in vivo (30,47), in part through activation of endothelial nitric oxide synthase (eNOS) (46). GPER activation in myenteric neurons results in decreased contraction of colonic smooth muscle, and this GPER effect appears to be mediated by enhanced NO release from myenteric nerves (48). Endothelial NO production is also involved in GPER-mediated vasodilation in mesenteric arteries (49). Thus, GPER is a likely modulator of urethral tone via activation of the nitric oxide pathway, the same pathway suppressed by activation of ERα. This is not the only instance in which estrogen signaling through GPER versus ERα has been observed to have divergent effects. For example, while signaling through GPER in vascular endothelium suppresses tumor necrosis factor-induced upregulation of proinflammatory proteins, signaling through classical ERs suppresses this anti-inflammatory effect (50).

Although GPER deficiency markedly altered the effects of estrogen on urinary retention in the present study, the effects of estrogen treatment on body weight and thymic size were not different in GPER KO mice versus wild type mice. The reduction in body weight seen in estrogen-treated wild type mice fed a high-fat diet was consistent with other reports (31,36,37). Moreover, estrogen treatment of wild type mice is also known to result in thymic atrophy, as confirmed in the present study (35,51). Our findings however differed from a previous report that estrogen-induced thymic atrophy (following 8 days of treatment) was suppressed by 30% in ovary-intact GPER KO mice fed normal chow compared to that in wild type mice (34).

Splenomegaly due to extramedullary hematopoiesis has been reported to occur in mice treated with estrogen (52). In the present study, we confirmed this finding; however, spleens in estrogen-treated GPER KO mice were even larger than those in estrogen-treated wild type mice. This additional splenic enlargement may be due to the inflammatory response to perineal ulceration observed in GPER KO mice. Collectively however these findings suggest that estrogen effects on body weight, thymus atrophy, and splenic enlargement can be mediated through ERα in the absence of GPER.

Conclusions

In conclusion, clinical extrapolation of our results suggests that GPER-selective antagonism (39,40) could be beneficial in individuals with lower urinary tract symptoms of incontinence, increased voiding frequency and urgency, nocturia and incomplete bladder emptying, whereas GPER-selective agonists (53) could promote urination in individuals with hesitancy, weak stream or straining. In contrast, ER-selective agonists would be expected to mimic the effects of GPER deficiency observed here (38). Such studies represent important future directions of this research.

Highlights.

• Prolonged estrogen treatment of mice leads to infravesicular obstruction to urine outflow, commonly referred to as bladder outlet obstruction (BOO).

• This study explored the contribution of the G protein-coupled estrogen receptor in estrogen-induced BOO.

• Estrogen-treated GPER knockout mice developed abdominal distension, debilitation, and urogenital skin ulceration with prominently distended bladders and hydronephrosis.

• In contrast, estrogen-treated wild type mice only rarely displayed these outcomes in the same time frame.