Metformin ameliorates insulitis in STZ-induced diabetic mice

Animals

Male C57BL/6 mice (6∼8 w) were purchased from Sino-British SIPPR/BK Lab Animal Ltd. (Shanghai, China). Mice were maintained in a well-ventilated humidified room (12 h light-dark cycle, 23 ± 2 °C), received water ad libitum, and fed standard chow at the Laboratory Animal Science Center of the Second Military Medical University. All the animal experiments were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Protocols were approved by the Institutional Animal Care and Use Committee of the Second Military Medical University (SYXK-2012-0003).

STZ model of diabetes and experimental designs

C57BL/6 mice were given STZ (Amresco, Solon, Ohio, USA) for a consecutive 5-day schedule (at 8∼12 a.m. every day) according to published methods and our previous protocol (Szkudelski, 2001; Yu et al., 2016). Briefly, the compound was dissolved in a citrate buffer (pH 4.5), and injected intraperitoneally (60 mg/kg/d) within 15 min of dissolution. The control group received citrate buffer solution without STZ correspondingly. Three weeks post STZ stimulation, animals with random blood glucose value ≥300 mg/dL were defined as STZ-induced diabetic mice. The mice were then divided into three groups: (1) Control; (2) STZ; and (3) metformin treated (STZ + Met), and housed in groups. The dose of metformin administered to mice in this study was calculated according to clinically relevant human dose based on body surface area. Metformin (250 mg/kg/d; Sigma-Aldrich, St. Louis, MO, USA) dissolved in vehicle solution (5% sodium carboxymethylcellulose, CMC-Na; Sangon Biotech, Shanghai, China) was administered by gavage for 2 weeks (at 2∼4 p.m. every day). Mice in the Control and STZ groups were treated with vehicle correspondingly. On day 34, mice were used for ex vivo studies (Fig. 1A)

Blood glucose and body weight determinations

Blood samples were collected from the mice tail vein for the determination of blood glucose (at 2∼4 p.m.) according to previously described methods (Li et al., 2016; Yu et al., 2016) using a monitoring system (Major Biosystem, Taipei, Taiwan, China). Changes in the body weight of each group were determined after metformin administration on day 34 at 2∼4 p.m. Also, the daily amount of food and water intake was measured.

Histological preparation and assessments of pancreatic islets in H&E and immunofluorescence stained sections

C57BL/6 mice were euthanized by cervical dislocation at the end of metformin treatment. Then the pancreas were excised, separated immediately from the fat, cleaned carefully, and blotted dry. The collected pancreata were divided into two parts along the pancreatic duct, either fixed in 10% buffered paraformaldehyde and embedded in paraffin for histological processing, or processed for insulin and cytokines analysis. Thin sections (5 µm) of these embedded samples were cut and stained with hematoxylin and eosin (H&E) for histopathological evaluations. Images were taken at 200× magnifications under an inverted phase-contrast light microscope (Leica Microsystems, Wetzlar, Germany).

For immunohistochemical analysis, insulin in the pancreas islets was detected using a mouse monoclonal antibody (diluted 1:50; BD Bioscience, San Jose, CA, USA) followed by immunofluorescence staining with Cy3-labeled goat anti-mouse IgG antibody (Jackson ImmunoResearch, West Groove, PA, USA). The nuclei were labeled using DAPI (diluted 1:250; Molecular Probes/Invitrogen Life Technologies, Eugene, OR, USA) staining. Images were obtained with a microscope (Leica Microsystems, Wetzlar, Germany) at a magnification of ×200 using identical acquisition settings for each section. The relative intensities of fluorescence, two sections per mouse, were determined using Image-Pro Plus software (Media Cybernetics, Silver Spring, MD, USA).

Insulitis score

The severity of insulitis was semiquantitatively measured according to previously reported method (Citro et al., 2015) using a scale of 0–4, where 0 referred to no sign of pathological changes in the pancreas islets, 1 referred to peri-insulitis (sign of pathological infiltration changes observed only restricted to the surrounding of islets), 2 referred to mild evidence of insulitis (less than 25% of the islets showed sign of pathological infiltration changes), 3 referred to severe evidence of insulitis (between 25 and 75% of the islets showed sign of pathological infiltration changes), and 4 corresponded to apparent evidence of destructive insulitis (more than 75% of the islets showed sign of pathological infiltration changes). The analysis of islet infiltration was performed by two independent observers who were blinded to the experimental conditions.

Insulin assay

After all of the final measurements had been obtained, mice were anesthetized with ketamine (100 mg/kg, i.p.), and blood samples used for serum insulin measurement were obtained by excising the eyeballs. Those collected pancreatic tissues were homogenized and centrifuged at 3, 200× g for 30 min. The supernatant was collected and stored at −80 °C until further use.

The determination of insulin levels in both serum and pancreatic tissue samples were performed by ELISA according to the manufacturer’s instructions. Kit for estimation of insulin was obtained from the Shibayagi Co., Ltd. (Ishihara, Shibukawa, Gunma, Japan). Briefly, 10 µl of the collected serum and tissue specimens were incubated in wells coated with biotin conjugated monoclonal anti insulin for 2 h at room temperature (RT). After incubation, washing buffer was added to rinse wells, followed by addition of HRP conjugated streptavidin to incubate for 30 min at RT. After washing, the remaining HRP conjugated streptavidin were reacted with substrate chromogen reagent for another 30 min at RT. Reaction was then stopped by addition of acidic solution, and the readings were made on a microplate reader (Bio-Rad, Hercules, CA, USA) at 450 nm. All samples, blanks and standards were analyzed in duplicate.

Inflammatory cytokines determination

Pancreatic tissue samples were analyzed for proinflammatory cytokines (IL-1β and TNF-α) using ELISA kits (Anogen, Mississauga, Ontario, Canada) according to the manufacturer’s protocols. Estimation of IL-1β and TNF-α were performed using monoclonal anti-mouse IL-1β and TNF-α as primary antibodies. All readings were made on an ELISA Plate Reader (Bio-Rad, Hercules, CA, USA). All samples, standards and blanks were analyzed in duplicate.

Statistical analyses

Data were expressed as means ± SEM and analyzed by one-way ANOVA followed by Tukey’s test using GraphPad Prism Software (Version 5.01). Statistical significance was set at P<0.05.

Metformin reduced blood glucose level in STZ-induced diabetic mice

In Figs. 1B and 1C and Figs. 2A–2B, the plasma glucose levels and body weight changes are shown after injection of STZ or treatment with metformin. Animal glucose levels in the STZ group were significantly elevated. Metformin treatment significantly reduced blood glucose in STZ-induced diabetic mice compared with STZ group (P < 0.05). There was no significant difference in the body weight between metformin-treated and SZT group, though the body weight was significantly decreased after injection of STZ (P < 0.001).

In addition, improvement of aberrant food and water intake behavior was also observed following metformin treatment (Figs. 2C and 2D). The animals in STZ group evidenced significant increase in both water and food intake after receiving STZ stimulation (Figs. 2C and 2D). The aberrant water intake behavior was significantly improved after metformin treatment compared with vehicle-administered STZ-induced diabetic mice (Fig. 2C), and the increased food intake tended to be lower, but it was not statistically significant (Fig. 2D).

Metformin increased pancreatic and serum insulin levels in STZ-induced diabetic mice

To determine the recovery of β-cell function, insulin levels in pancreatic tissues and sera were detected at the end of the study. Significant differences were observed when compared mice stimulated with STZ to mice from the control group (Figs. 3 and 4). However, metformin administration significantly increased insulin secretion of STZ-induced diabetic mice: the insulin levels in either pancreas (P < 0.01; Fig. 4B) or serum (P < 0.001; Fig. 4C) were significantly improved in mice treated with metformin when compared to those in the STZ group. Also shown in Fig. 4B is ELISA analysis of insulin production in pancreas from different groups. Thus, the application of metformin may provide a significant degree of β-cell protection in STZ-induced diabetes model.

Metformin attenuated insulitis in STZ-induced diabetic mice

The administration of metformin has been found to protect against diabetes-related complications in mice/rats (Sartoretto et al., 2005; Yu et al., 2016), when STZ is administered as diabetogenic doses. Figure 5 shows the action of metformin to alleviate the toxic activity of STZ in C57BL/6 mice. In the presence of diabetogenic STZ, the islet integrity became disrupted, with uneven boundaries. However, administration of metformin in STZ-induced diabetic mice preserved islets integrity, with regularly shaped islets observed under microscope (Fig. 5A).

The severity of insulitis was also scored in experimental diabetic mice 2 weeks after treatment to determine the pancreas islet leukocytic infiltration degree. A significant difference was found between the STZ-induced diabetic mice and non-STZ stimulated control mice (Fig. 5). However, the insulitis score in STZ-induced diabetic mice was greatly lowered by metformin administration (Fig. 5B).

Metformin reduced IL-1β and TNF-α level of pancreatic tissues in STZ-induced diabetic mice

Proinflammatory cytokines, such as IL-1β and TNF-α, in pancreas reflects the severity of pancreas islets injury. The administration of metformin has been reported to possess potential anti-inflammatory effects (Saisho, 2015). Therefore, we give this agent to each of the STZ-induced diabetic mice. To determine inflammatory cytokines in the pancreas, tissue IL-1β and TNF-α level were measured at the end of the study. Significant increase of the two cytokines was detected in STZ-induced diabetic mice, compared to those from control group (Fig. 6). Treatment with metformin reduced inflammatory cytokines production in the pancreas of experimental diabetic mice: IL-1β level in mice treated with metformin was similar to that in mice from control group (Fig. 6A); though not statically significant reduced, TNF-α level tended to be lower when compared to STZ-induced diabetic mice (Fig. 6B).