Gastric bypass prevents diabetes in genetically modified mice and chemically induced diabetic mice

Obese subjects have increase probabilities of developing type 2 diabetes (T2D). In this study, we sought to determine whether gastric bypass prevents the progression of prediabetes to overt diabetes in genetically modified mice and chemically induced diabetic mice. Roux-en-Y gastric bypass (RYGB) was performed in C57BL/KsJ-db/db null (BKS-db/db,) mice, high-fat diet (HFD)-fed NONcNZO10/LtJ (NZO) mice, C57BL/6 db/db null (B6-db/db) mice and streptozotocin (STZ)-induced diabetic mice. Food consumption, body weight, fat mass, fast blood glucose level, circulating insulin and adiponectin and glucose tolerance test were analyzed. The liver and pancreatic tissues were subjected to H&E and immunohistochemistry staining and islet cells to flow cytometry for apoptotic analysis. RYGB resulted in sustained normoglycemia and improved glucose tolerance in young prediabetic BKS-db/db mice (at the age of 6 weeks with hyperglycemia and normal insulinemia) and HFD-fed NZO and B6-db/db mice. Remarkably, RYGB improved liver steatosis, preserved the pancreatic β-cells and reduced β-cell apoptosis with increases in circulating insulin and adiponectin in young prediabetic BKS-db/db mice. However, RYGB neither reversed hyperglycemia in adult diabetic BKS-db/db mice (12 weeks old) nor attenuated hyperglycemia in STZ-induced diabetic mice. These results demonstrate that gastric bypass improves hyperglycemia in genetically modified prediabetic mice; however, it should be performed prior to β-cells exhaustion.


Introduction
Obesity is associated with insulin resistance and type 2 diabetes (T2D) and is considered a public health crisis, responsible for an exponential increase in health care costs [1]. In response to insulin resistance, pancreatic β-cells compensate by increasing insulin secretion characterized by hyperglycemia and hyperinsulinemia or normal insulinemia [2]. This metabolic setting can be maintained for long periods of time [3][4][5][6]. Eventually, in some individuals who are genetically predisposed to develop T2D, pancreatic β-cells are unable to maintain the increased db mice (6 weeks old with blood glucoses of 300-350mg/dl and normal insulinemia), adult BKS-db/db mice (� 12 weeks old with blood glucoses of � 400 mg/dl and reduced plasma insulin levels) and adult B6-db/db mice (� 12 weeks old with blood glucoses of � 300 mg/dl) were used. All BKS-db/db and B6-db/db mice were fed chow for the duration of the experiment. NZO mice were fed with chow up to 6 weeks of age, after which gastric bypass and sham surgeries were performed, followed by a HFD until the end of the experiment. To test whether RYGB reverses chemically induced diabetes, MIP-luc mice were made diabetic (blood glucoses were � 300 mg/dl for more than two consecutive days) by a single intravenous injection of STZ (170 mg/kg, Sigma-Aldrich, St. Louis, MO). The mouse models and diet feeding are described in Table 1 and the S1 File. RYGB was performed with a success rate of �80% as described in our previous reports and the S1 File [16,17,30], and RYGB with a small gastric pouch was used in this study. Blood glucose was tested before and after surgery using a glucose meter (SureStep, LifeScan, Inc., Milpitas, CA). In the sham procedure, the stomach, duodenum and jejunum were mobilized, and the esophagus, stomach and jejunum were clamped without incision followed by the closure of the abdomen. This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. At the endpoint (12 weeks post-surgery), mice with RYGB or sham surgery were sacrificed under the anesthesia of isoflurane, and the livers were collected for H&E staining. All surgery was performed under isoflurane, and all efforts were made to minimize suffering. The protocol (ACUP number: 72357) was approved by the University of Chicago Institutional Animal Care and Use Committee.

Whole body composition
Body mass was measured using the mq10 NMR analyzer (Bruker Optics Inc., Billerica, MA) following a 2-hr fast.

Intraperitoneal glucose tolerance tests (IPGTT)
Mice were fasted for 4 hrs prior to the IPGTT. Blood was sampled from the tail vein at 0, 30, 60, 90 and 120 min after an intraperitoneal injection of 20% dextrose dosed at 2g/kg body weight. Blood glucose was measured using a blood glucose meter (SureStep, Lifescan, Inc.). The area under the cure (AUC) was calculated using the trapezoidal rule [31]. Any result exceeding the maximum reading of meter (showing HIGH) was recorded as 600 mg/dL.

Plasma insulin and adiponectin analyses
Circulating insulin and adiponectin were analyzed by the MilliPlex map kit using the Luminex 200 System analyzer (Millipore, Co., Billerica, MA) following company's instruction.

Flow cytometry for annexin V positive cell analysis
Single islet cells were collected by collagenase digestion and Ficoll gradient procedure [28,32]

Bioluminescence imaging (BLI)
BLI in MIP-luc mice allows for direct visualization of β-cell viability under isoflurane anesthesia [29]. Luciferin (Roche Diagnostics, Indianapolis, IN) was injected intravenously at a dose of 50mg/kg. Mice were housed inside a light-tight box and imaged with an ICCD camera (Hamamatsu C2400-32, Hopkinton, MA). Light emission through the ventral body was detected as photon counts over a standardized area of the organs of interest using the ARGUS-50 software for image processing [29,33].

Statistics
Statistical significance was analyzed by ANOVA test (StatView 4.5, Abacus Concepts, Berkeley, CA). P-value of < 0.05 was considered significant. Results were presented as the mean ± SEM. The level of significance was set at a probability of p < 0.05.

Results
Gastric bypass improves hyperglycemia and glucose tolerance in young prediabetic BKS-db/db mice RYGB or sham surgery was performed in young prediabetic BKS-db/db mice showing hyperglycemia and normal insulinemia. Diet intake was analyzed in untreated (BKS-db/db Control), sham surgery treated (BKS-db/db-SHAM) and RYGB-treated (BKS-db/db-RYGB) mice. Fig  1a shows that chow consumption in BKS-db/db-Control mice was 6.2 ± 0.5 g/day at 6 weeks of age and increased to 7.2 ± 0.5 g/day at 18 weeks of age. Chow intake in BKS-db/db-RYGB mice was reduced immediately post-surgery, and BKS-db/db-SHAM mice revealed the similar change to RYGB mice in the first 2 weeks post-surgery. However, diet consumption was increased 4 weeks after surgery in BKS-db/db-SHAM mice and reduced through the end point  Fig 1b). Fat mass in BKS-db/db mice was enhanced compared to wild-type (WT) C57BL/6 LEAN mice and significantly reduced by RYGB (BKS-db/db-RYGB vs. BKS-db/db-Control, p < 0.05 from the 1 st week to 12 weeks post-surgery, Fig 1c).
To test whether RYGB improved β-cell apoptosis in BKS-db/db mice, islet cells were collected at 4 weeks post-surgery for Annexin V analysis by flow cytometry [28,32]. Increased apoptotic islet cells were found in control BKS-db/db mice (CONTROL, Fig 3d) compared to lean C57BL/6 mice (LEAN). Both SHAM-PF and RYGB reduced Annex V-and PI-double positive cells (Annex V-PI, p <0.001); however, when compared to SHAM-PF, RYGB further reduced apoptotic cells (p < 0.01, Fig 3d). The results show that RYGB improves β-cell viability and prevents β-cell apoptosis. These changes were also reflected by the observed changes in blood glucose and plasma insulin levels (Fig 3a).

Gastric bypass improves β-cell mass and prevents diabetes in HFD-fed NZO mice
NZO male mice were fed chow in the first 6 weeks of age followed by HFD feeding to the end point (12 weeks). HFD accelerated body weight gain compared to chow (Fig 4a). Four of eight chow-fed and all HFD-fed NZO males developed hyperglycemia (fasting blood glucoses of > 300 mg/dl) after 6 weeks on HFD. SHAM-PF delayed but did not completely prevent hyperglycemia in 6 of 8 mice. RYGB prevented hyperglycemia in 100% of HFD-fed NZO mice https://doi.org/10.1371/journal.pone.0258942.g004 (Fig 4b), suggesting that HFD accelerates the development of overt diabetes, which can be prevented by RYGB, and RYGB, but not SHAM-PF, preserves insulin + cells, contributing to improved hyperglycemia in HFD-fed NZO mice.

Gastric bypass fails to improve glucose tolerance in genetic and chemically induced diabetic mice
To determine whether RYGB reverses diabetes when β-cells were severely damaged or exhausted, RYGB surgeries were performed in adult diabetic BKS-db/db mice (� 12 weeks of age) with blood glucoses of � 400 mg/dl. In contrast to young prediabetic BKS-db/db mice, RYGB did not improve hyperglycemia in adult diabetic BKS-db/db mice (Fig 5c).
STZ induced diabetes (blood glucoses of � 300 mg/dl) in 75-80% of MIP-luc mice after 3-7 days of injection (referred to as day 0 in Fig 5d and 5e). Fig 5d indicates that RYGB failed to reverse hyperglycemia, and diabetic mice remained hyperglycemic through 4 weeks post-surgery. Bioluminescence imaging (Fig 5e) reveals reduced luciferase activity in STZ-induced diabetic mice, indicating decrease of β-cell activity, which cannot be improved by RYGB.

Discussion
In prediabetes metabolic states such as obesity and glucose intolerance, β-cells biosynthesize and secrete more insulin in response to increased blood glucose level. With chronic hyperglycemia, there is a perpetually high demand for insulin, and β-cells eventually become exhausted, resulting in insufficient insulin synthesis and secretion causing overt diabetes [36]. The current study shows that RYGB preserves β-cell mass and prevents or slows the progression toward the insulin-deficient state, thereby preventing the development of overt diabetes in genetically modified prediabetic mice. However, RYGB neither reverses hyperglycemia nor improves βcell viability in genetically modified diabetic mice and chemically induced diabetic mice. Gastric bypass must be performed before β-cells are completely exhausted, and it is ineffective in reversing the overt diabetic state once β-cells have already been severely damaged.
Although weight loss is associated with decreased insulin resistance, recent clinical results suggest that gastric bypass normalizes insulin sensitivity well before a significant reduction in body weight [37,38]. We have recently shown that RYGB induces sustained weight loss, fat mass reduction and improves insulin resistance in HFD-induced obese mice [16,17]. However, in BKS-db/db mice, body weight and fat mass are restored to pre-surgery levels from the second week post-surgery, and there is no statistical difference with SHAM-PF mice. However, while RYGB mice maintain normoglycemia post-surgery, SHAM-PF mice maintain hyperglycemia through the end point with a severe loss of β-cell mass, indicating that RYGB prevents the progression to overt diabetes and that this remarkable effect is independent of weight loss.
Consistent with a previous report that plasma insulin is decreased by 40% and 80% at 8 and 12 weeks of age, respectively [39], our findings show that plasma insulin in adult diabetic BKS- db/db mice (12 weeks of age) is reduced to 25% of that in young prediabetic BKS-db/db mice (6 weeks of age). At the meantime, plasma adiponectin is also reduced to 50%. Adiponectin has been shown to induce ERK and Akt phosphorylation, stimulate insulin secretion and protect β-cells against apoptosis [40]. Gastric bypass preserves adiponectin expression, contributing to the prevention of β-cell exhaustion, and maintains normal insulinemia and normoglycemia when RYGB is performed in prediabetic BKS-db/db mice.
Both B6-db/db and BKS-db/db mice reveal increased blood glucose; however, B6-db/db mice glucose levels peak then decrease with aging due to β-cell hypertrophy and do not develop the full phenotype of T2D [15]. Gastric bypass improved glucose tolerance throughout the 12-week period in B6-db/db mice. T2D is characterized by a combination of genetic and epigenetic factors, and it is increasingly attributed to environmental factors. NONcNZO10/LtJ (NZO) is a new polygenic strain developed to more realistically model human obesity induced T2D. Consistent with previous reports [25,26], our results show that NZO male mice develop hyperglycemia and exhibited moderate to severe liver steatosis and islet atrophy in 40% of mice. HFD (60% fat)-feeding results in hyperglycemia in 100% of NZO male mice. Insulin + cells are reduced in HFD-induced diabetic NZO mice due to β-cell apoptosis, as we found in the BKS-db/db model. RYGB preserves β-cell mass, maintains normoglycemia and prevents HFD-induced diabetes, suggesting that genetic-and HFD-mediated diabetes can be prevented by gastric bypass.
STZ targets insulin-producing β-cells in the pancreas and results in severe damage of βcells in mice, mimicking T1D phenotypes. Recent studies have suggested that gastric bypass yields metabolic benefits as well as leads to better cardiovascular outcomes for the treatment of T1D [41,42]. However, our results show that gastric bypass neither rescues damaged β-cells nor improves hyperglycemia in STZ-induced diabetic mice. We understand that one major limitation of our findings relates to the different proliferation of β-cells, i.e. many mitogens, growth factors and nutrients have been shown to induce expansion in rodent models. However, human β-cells show poor responsiveness to these stimuli [43]. The definition of prediabetes in the clinic is that blood glucose level (100 to 125 mg/dL) is higher than it should be (in normal condition) but not high enough to diagnose diabetes (� 126 mg/dL) [44]. Consequently, the gastric bypass related results in mice must be carefully interpreted concerning human implications.
In conclusion, gastric bypass results in sustained normoglycemia and improved glucose tolerance associated with preservation of β-cells in prediabetic BKS-db/db mice. These effects are independent of weight loss. Gastric bypass also prevents the development of overt diabetes in HFD-fed NZO mice. In adult diabetic BKS-db/db mice, however, gastric bypass neither reverses diabetes nor improves β-cell viability. Similarly, gastric bypass did not reverse STZinduced overt diabetes. Our findings suggest that RYGB improves hyperglycemia and must be performed before β-cells are severely damaged.