The Effect of Vertical Sleeve Gastrectomy on a Rat Model of Polycystic Ovarian Syndrome

Abstract

Polycystic ovarian syndrome (PCOS) is the most common endocrine disorder of women of reproductive age. Although some of the primary symptoms of PCOS are reproductive abnormalities, including hyperandrogenism, menstrual dysfunction, and hirsutism, other metabolic disturbances are also common, including obesity and insulin resistance. Women with PCOS who have undergone weight-loss bariatric surgery have reported surprising postoperative benefits beyond weight loss, including resolution of menstrual dysfunction and improvement of hirsutism. Here, we use a chronic dihydrotestosterone (DHT) exposure model of PCOS in female rats and investigate the efficacy of a specific type of bariatric surgery, namely vertical sleeve gastrectomy (VSG), to resolve the reproductive and metabolic disturbances induced by DHT treatment. We find that VSG causes loss of body weight and body fat in DHT-treated rats but does not improve glucose tolerance or restore estrous cyclicity. Although human PCOS patients have shown decreased androgen levels after bariatric surgery, the chronic nature of DHT administration in this rat model both before and after VSG renders this effect impossible in this case. Therefore, the lack of improvement in glucose tolerance and estrous cyclicity may implicate a direct effect of androgen knockdown as a mechanism for the improvements seen in human PCOS patients after bariatric surgery. In addition, the dissociation of body weight loss without improved glucose tolerance suggests that glucose intolerance may be a body weight-independent phenomenon in women with PCOS.

Polycystic ovarian syndrome (PCOS) is the most common endocrine disorder of women of reproductive age, affecting 5–10% of this population (1). The cardinal manifestations of PCOS are hyperandrogenism and polycystic ovary morphology and also include menstrual dysfunction and hirsutism. In addition, women with PCOS have greatly increased risk for obesity, type 2 diabetes, and associated metabolic comorbidities (2–6).

In light of the increased prevalence of metabolic disturbances in women with PCOS, some women with PCOS have elected to undergo weight-reduction bariatric surgery. In the general population, bariatric surgery causes not only weight loss, but many procedures have been shown to improve glucose tolerance and other metabolic parameters as well (7–9). Relatively few studies have investigated the outcome of bariatric surgery in women with PCOS, but results have indicated that the benefits of surgery are above and beyond the expected metabolic improvements of weight loss and improved glucose tolerance. Two studies that investigated women with PCOS who underwent Roux-en-Y gastric bypass (RYGB) or biliopancreatic diversion showed 100% remission of menstrual dysfunction and almost uniform improvement of hirsutism after surgery (10, 11).

The term bariatric surgery encompasses several kinds of procedures that modify the digestive system in various ways to induce weight loss. RYGB and biliopancreatic both decrease the effective size of the stomach and reroute the flow of nutrients through the intestine. In contrast, vertical sleeve gastrectomy (VSG) is a comparatively simple procedure, modifying only the stomach without manipulating the intestine. VSG involves removing approximately 80% of the stomach along the greater curvature, creating a gastric “sleeve” connecting the esophagus and pylorus. Like RYGB, this procedure induces loss of weight, fat mass, and improves glucose tolerance in humans and in rodent models (7–9, 12–14).

Animal models of PCOS (15–18) allow investigation of this disease in a way that is not possible in humans and avoid many of the confounds of the human population, including heterogeneity of symptoms and medications. We investigated the effect VSG in rodents with experimentally induced symptoms of PCOS. We chose a model of chronic exposure to dihydrotestosterone (DHT) via a sc-implanted pellet, starting at 3 wk of age. This model was chosen because it has been shown to induce both reproductive and metabolic disturbances associated with PCOS (17, 19). Given that we have observed a variety of metabolic impacts of VSG that are beyond weight loss (12, 14, 20), we hypothesized that VSG would normalize both the reproductive and metabolic disturbances associated with this model.

Materials and Methods

Animals

All procedures for animal use were approved by the University of Cincinnati Institutional Animal Care and Use Committee. Nine pregnant female Wistar rats (Harlan Laboratories, Indianapolis, IN) were obtained on the 14th d of pregnancy. Litters were culled at postnatal d 2, such that each litter consisted of eight pups, four to six of which were females to be used in the study. Male pups were not used. DHT pellet implantation and weaning occurred at postnatal d 21, at which time the female pups were singly housed with ad libitum access to standard laboratory chow (Harlan-Teklad, Indianapolis, IN) and water at all times. Food hoppers placed in each cage were weighed daily or weekly to measure food intake. At 9 wk of age, DHT-treated rats were divided into two body weight- and fat mass-matched groups (VSG and sham) before gastric surgery. Animals were killed after an additional 6 wk. Experimental groups were control-sham (n = 14), DHT-sham (n = 12), and DHT-VSG (n = 8). Pellet and surgery type were counterbalanced across litters.

Pellet implantation

At postnatal d 21, anesthetized (isoflurane) female pups were implanted sc with a 90-d continuous-release pellet containing 7.5 mg of 5α-DHT or placebo (Innovative Research of America, Sarasota, FL) to deliver the appropriate stimulus until the rats were killed at 15 wk. This paradigm and dose were chosen based on their ability to induce both ovarian and metabolic characteristics of PCOS (17) and to mimic the hyperandrogenic state in women with PCOS (21, 22).

Gastric surgery

VSG or sham surgery was conducted at 9 wk of age using isofluorane anesthesia, as previously described (12). Briefly, the lateral 80% of the stomach was excised, leaving a tubular gastric remnant in continuity with the esophagus superiorly and the pylorus and duodenum inferiorly. The sham procedure involved analogous isolation of the stomach followed by manually applying pressure with blunt forceps along a vertical line between the esophageal sphincter and the pylorus.

Estrous cycle

The stage of the estrous cycle was determined during five to six consecutive days at 6 and 12 wk. Cells were obtained by vaginal lavage and were stained with DipQuick staining kit (Jorgensen Laboratories, Inc., Loveland, CO) for the determination of the estrous cycle phase (23).

Body composition and body length

Magnetic resonance imaging was performed on all rats at 8, 10, and 14 wk to determine fat and lean body composition, using an Echo magnetic resonance imaging whole-body composition analyzer (EchoMedical Systems, Houston, TX). Each rat was tested in duplicate, and readings were accepted if they differed by less than 10%. Fat-free mass was calculated by subtracting fat mass from total body weight. Body length was determined at sacrifice by measuring naso-anal distance.

Glucose tolerance tests (GTT)

GTT were performed at 8 and 14 wk of age. Rats were fasted for 18 h, beginning 2 h into the dark phase of the light-dark cycle. After a baseline blood sample was taken (0 min), 50% D-glucose (Phoenix Pharmaceutical, St. Joseph, MO) was injected ip. Blood glucose was measured at 0, 15, 30, 45, 60, and 120 min after glucose administration on duplicate samples using Accu-chek glucometers and test strips (Roche, Indianapolis, IN). All blood samples were obtained from the tip of the tail vein of freely moving rats. Because lean mass accounts for the majority of glucose uptake during a GTT, each rat received a dose of glucose equal to 1.75 g/kg of fat-free mass, as determined from body composition parameters measured the week before the GTT. Rats were excluded from the data analysis if they did not exhibit a rise in blood glucose of greater than 20 mg/dl in the first 15 min after injection or if they exhibited diarrhea, because such symptoms indicate that the glucose injection did not hit the ip cavity.

Insulin and leptin

Insulin and leptin were measured from plasma taken at killing, which occurred after a 16-h fast. Blood was cold centrifuged, and plasma was stored at −80 C until insulin and leptin were assessed using ELISA (CrystalChem, Inc., Downers Grove, IL).

Statistics

Data were analyzed using Student's t test or the appropriate ANOVA. Where appropriate, Tukey's post hoc comparisons were used to determine pair-wise differences between groups. P < 0.05 was considered significant for each of these analyses. All data were analyzed using GraphPad Prism (GraphPad Software, Inc., San Diego, CA).

Results

Chronic DHT treatment causes increased body weight, food intake, and fat mass and disrupts estrous cyclicity

DHT-treated females had significantly greater body weight beginning at 5 wk of age (Fig. 1A) and greater food intake beginning at 33 d (P < 0.01) (Fig. 1B). At 8 wk, 5 d before gastric surgery, DHT-treated rats had significantly greater fat mass (P < 0.01) and lean mass (P < 0.001) than the rats that received a control pellet (Fig. 1C). When expressed as a percentage of body weight, fat mass and fat-free mass were not different between groups (Supplemental Fig. 1, A and B, published on The Endocrine Society's Journals Online web site at http://endo.endojournals.org). At 8 wk of age, there were no significant differences in glucose tolerance between groups (Fig. 1D), although by 14 wk, significant glucose intolerance had developed in the DHT-treated group compared with the control group (Fig. 2F). Baseline glucose levels did not differ significantly between groups (control, 78.92 ± 2.340 sem; DHT, 82.81 ± 1.431 sem). Compared with control rats, DHT-treated rats displayed a disrupted estrous cycle (Fig. 1E), whose vaginal cytology indicated more days in diestrous 1, and fewer days in diestrous 2, proestrous, and estrous phases (P < 0.01).

Fig. 1.

DHT-treated females had greater body weight (A), food intake (B), fat mass, and fat-free mass (C) than control females. An ip GTT revealed no differences in glucose tolerance at 8 wk of age (D). Estrous cyclicity (E) was disrupted in DHT-treated females, such that most rats did not exhibit vaginal cytology characteristic of proestrus or estrus during the time period in which they were examined. Data shown are mean ± sem; *, P < 0.05 compared with control.

Fig. 2.

Control or DHT-treated rats received sham or VSG surgery and were monitored postoperatively for body weight (A), food intake (B), fat mass (C), fat-free mass (D), and body length (E). Glucose tolerance was assessed 7 wk after surgery by an ip GTT (F). Fasting leptin (G) and insulin (H) were determined in plasma taken at killing, 7 wk after surgery. Data shown are mean ± sem; *, P < 0.05 for control-sham vs. DHT-sham; #, P < 0.05 for control-sham vs. DHT-VSG; †, P < 0.05 for DHT-sham vs. DHT-VSG.

VSG causes loss of body weight and fat mass without improving glucose tolerance in DHT-treated females

After surgery, DHT-VSG rats lost significantly more weight than the DHT-sham rats starting 1 wk after surgery and maintained a lower body weight through the termination of the experiment (P < 0.05). Control rats maintained a body weight lower than the DHT-sham group (P < 0.05 at all time points) and lower than or not different than the DHT-VSG group (P < 0.05 at d 0, no difference 1–5 wk; P < 0.05, wk 6 and 7) (Fig. 2A). Food intake was measured starting 4 d after surgery. DHT-sham rats displayed greater daily food intake than control-sham rats, which was significant beginning 20 d after surgery. DHT-VSG animals exhibited a transient reduction in food intake relative to the other two groups (vs. control-sham P < 0.05 at 5 and 7 d, vs. DHT-sham from 5 to 34 d) (Fig. 2B) consistent with what we observed after VSG in other experiments (12, 14, 20).

Body composition was measured at 10 and 35 d after surgery. DHT-sham rats maintained a higher fat mass than control-sham, which was significant at 35 d (P < 0.05). DHT-VSG females had lower fat mass than DHT-sham and control-sham at both postoperative time points (P < 0.05) (Fig. 2C). DHT-sham animals also maintained higher fat-free mass than control-sham animals (P < 0.001 at both time points), whereas DHT-VSG exhibited an intermediate phenotype (Fig. 2D). When fat and fat-free mass are expressed as a percentage of body weight, there are no differences between control-sham and control-DHT, but DHT-VSG rats have a lower percentage of body fat and a higher percentage of fat-free mass compared with both control groups (Supplemental Fig. 1, C and D). Body length, as assessed by naso-anal distance, was significantly greater in DHT-sham rats than control-sham or DHT-VSG rats (P < 0.05) (Fig. 2E).

Five weeks after surgery, DHT-sham rats had impaired glucose tolerance compared with control-sham (P < 0.05). Interestingly, DHT-VSG females showed no improvement in glucose tolerance compared with DHT-sham, despite their decreased body weight and fat mass (Fig. 2F). VSG was similarly ineffective at restoring normal estrous cyclicity, because both DHT-sham and DHT-VSG rats maintained a disrupted estrous cycle compared with control-sham females and did not differ significantly from each other (data not shown). Fasting insulin was significantly higher in DHT-sham rats than in control-sham rats and was normalized by VSG surgery (P < 0.05). These results are intriguing in light of the lack of improvement in glucose tolerance in the DHT-VSG group (Fig. 2G). Leptin levels paralleled fat mass and were elevated in DHT-sham rats compared with control-sham rats and were normalized by VSG surgery (P < 0.05) (Fig. 2H).

Discussion

The current results indicate that although VSG can reduce food intake, body weight, and body fat in a model of PCOS, it cannot restore normal glucose tolerance or normal cyclicity. These results are surprising in several ways. First, previous human work has shown that RYGB can produce metabolic and reproductive improvements in women diagnosed with PCOS (10, 11). There are several possible explanations for this discrepancy. First, this may be a difference between RYGB and VSG. Specifically, the rerouting of nutrients (which occurs in RYGB but not VSG) may cause differential endocrine or neuroendocrine changes that are more effective at relieving the effects of elevated androgens than VSG, which includes only a gastric manipulation. Although this explanation is appealing, it does not account for recent findings showing comparable metabolic impacts of RYGB and VSG on both lipids (20, 24) and insulin sensitivity (7, 9, 14, 24). Furthermore, the known effect of the insulin-sensitizing drug metformin, which improves menstrual function in women with PCOS (25–27), may indicate that improved insulin sensitivity is the relevant feature of bariatric surgery (achieved by both RYGB and VSG), which drives endocrine improvements in women with PCOS.

The second possible explanation is a discrepancy between the model of PCOS used here compared with human women with PCOS. Although this could be a species difference, it may also relate to the exogenous source of androgens used to induce PCOS symptoms in the female rats used in these experiments. Although this model has been widely used (17, 19, 28, 29), the continuous delivery of DHT before and after surgery does not allow for therapies that would otherwise lower androgen production to be effective. In the single study that measured androgen levels of women with PCOS who underwent RYGB, lower androgens were reported after RYGB (11). This raises the possibility that lowered androgens are a critical mediator of the effect of RYGB. Hence, failure of VSG to improve glucose tolerance or restore estrous cyclicity in the present experiments may be the result of VSG being unable to alter the levels of the exogenously administered androgens.

The second surprising aspect of the current data is the dissociation of body weight and glucose tolerance effects of VSG. Our data show that, even in the context of marked weight loss, glucose tolerance did not improve in DHT-treated rats, indicating that PCOS-related glucose intolerance may be a body weight-independent phenomenon. Further, our data indicate that although weight loss can occur despite unchanged hyperandrogenemia, this is not true for glucose regulation. This is a significant, because it implies that androgen-induced glucose intolerance is not amenable to improvement from either weight loss or the weight loss-independent effects of VSG on glucose regulation.

Treatment strategies for women with PCOS include targeted therapies for menstrual function/fertility, androgen-related symptoms (hirsutism and acne), or other metabolic endpoints, such as insulin sensitivity. However, few treatments are effective at simultaneously targeting all of these symptoms. One exception is weight loss, which may decrease androgen levels, normalize menstrual function, and improve glucose tolerance (30). Metformin has similar effects, although it does not substantially improve hirsutism (26) and is itself associated with slight weight loss (31, 32). Although data are limited, bariatric surgery (RYGB and biliopancreatic diversion) also treats fertility-related, androgen-related, and metabolic endpoints in women with PCOS. Studies using the rat model of DHT-induced PCOS employed in the present experiments have shown benefits of exercise to improve insulin sensitivity and reduce adiposity but did not evaluate estrous cyclicity (33). The same authors showed that electroacupunture also improves insulin sensitivity, restores estrous cyclicity, but does not reduce adiposity (19, 28, 33). Therefore, although the maintained levels of high androgens induced by this model may hinder certain types of therapies, others have nonetheless demonstrated significant benefits.

Although a variety of treatment options exists for women with PCOS, comprehensive strategies that treat the syndrome as a whole are largely lacking due to limited understanding of its etiology and the interplay of the various components of PCOS, including weight loss, androgen levels, and glucose homeostasis. The mechanistic discrepancy between the actions of androgens and VSG surgery implies that different treatment strategies will be necessary to treat the metabolic effects of androgen excess when the hyperandrogenemia cannot be addressed directly.