Hypoglycaemic effects of glimepiride in sulfonylurea receptor 1 deficient rat

Abstract

Background and Purpose

Sulfonylureas (SUs) have been suggested to have an insulin‐independent blood glucose‐decreasing activity due to an extrapancreatic effect. However, a lack of adequate in vivo evidence makes this statement controversial. Here, we aimed to evaluate whether glimepiride has extrapancreatic blood glucose‐lowering activity in vivo.

Experimental Approach

Sulfonylurea receptor 1 deficient (SUR1 ^−/−) rats were created by means of transcription activator‐like effector nucleases (TALEN)‐mediated gene targeting technology. Type 2 diabetic models were established by feeding a high‐fat diet and administering a low‐dose of streptozotocin. These rats were then randomly divided into four groups: glimepiride, gliclazide, metformin and saline. All rats were treated for 2 weeks.

Key Results

Glimepiride decreased blood glucose levels and increased insulin sensitivity without elevating insulin levels. Gliclazide showed similar effects as glimepiride. Both agents were weaker than metformin. Further mechanistic investigations revealed that glimepiride increased hepatic glycogen synthesis and decreased gluconeogenesis, which were accompanied by the activation of Akt in the liver. Moreover, glimepiride increased both total and membrane glucose transporter 4 (GLUT4) levels in muscle and fat, which might be attributed to insulin receptor‐independent IRS1/Akt activation.

Conclusion and Implications

Glimepiride possesses an extrapancreatic blood glucose‐lowering effect in vivo, which might be attributed to its direct effect on insulin‐sensitive tissues. Therefore, the combination of glimepiride with multiple insulin injections should not be excluded per se.

Abbreviations

SU

sulfonylurea

SUR

sulfonylurea receptor

TALEN

transcription activator‐like effector nucleases

GIRs

glucose infusion rates

GSK3

glycogen synthase kinase 3

GS

glycogen synthase

IPGTT

intraperitoneal glucose tolerance test

IPITT

intraperitoneal insulin tolerance test

InsR

insulin receptor

IRS

insulin receptor substrate

GLUT4

glucose transporter 4

Introduction

Sulfonylureas (SUs) are generally accepted to reduce blood glucose by stimulating the release of insulin from pancreatic beta cells (Broichhagen et al., 2014). However, there is also evidence suggesting that SUs can lower blood glucose by an insulin‐independent mechanism. Several studies have suggested that SUs can decrease blood glucose by improving insulin resistance, which is known as the extrapancreatic glucose‐lowering activity of SUs (Faber et al., 1990; Muller et al., 1995; Tsiani et al., 1995; Gribble and Ashcroft, 2000). The extrapancreatic glucose‐lowering effects of SUs have been demonstrated in vitro ; however, its action in vivo has not yet been determined. Thus, the existence of in vivo extrapancreatic glucose‐lowering effects of SUs has sparked widespread controversies. SU‐induced blood glucose control without a significant elevation of mean plasma insulin was interpreted as a drug‐induced postprandial kick of insulin release resulting in lower mean blood glucose values and thereby also reduced basal insulin levels (Bodansky et al., 1982; Hosker et al., 1985). The improvement in insulin resistance has also been attributed to an alleviation of glucotoxicity derived from a SU‐induced increase in insulin secretion (Muller et al., 1994; Muller et al., 1995). Moreover, researchers have attempted to confirm the extrapancreatic glucose‐lowering effects of SUs by inhibiting SU‐induced insulin secretion, using somatostatins or KK‐Ay mice (Muller et al., 1995; Tsiani et al., 1995; Overkamp et al., 2002). However, in view of the fact that the insulin secretion induced by SUs is not completely blocked in these models, the results of the above‐mentioned studies are not enough to confirm the existence of the extrapancreatic blood glucose‐decreasing effects of SUs. Thus, further in vivo studies are urgently needed to clarify the occurrence of these effects of SUs.

Glimepiride, a widely used SU, harbours the typical SU moiety and shares the wide spectrum of activities attributed to conventional SUs. In pancreatic beta cells, the glimepiride‐binding affinity is lower than that of other SUs, leading to a lower effect on insulin secretion. However, the effect of glimepiride on glycaemic homeostasis is similar to that observed for other SUs, which indicates that glimepiride possesses extrapancreatic blood glucose‐improving activity (Draeger et al., 1996; Muller, 2000).

Sulfonylurea receptor 1 (SUR1; ABCC8), the main functional receptor of SUs on islet beta cells, stimulates insulin secretion when combined with SUs (Thule and Umpierrez, 2014). In this study, an SUR1 gene (Abcc8) knockout (SUR1 ^−/−) rat model was used to determine the extrapancreatic glucose‐lowering activity of glimepiride, in which glimepiride could not stimulate insulin secretion by combining with SUR1, thus excluding its intrinsic ability to decrease glucose levels. We found that glimepiride could still reduce blood glucose without significantly elevating plasma insulin levels in this model. The hyperinsulinaemic‐euglycaemic clamp test further indicated that insulin resistance was ameliorated in glimepiride‐treated diabetic SUR1 ^−/− rats. To gain insight into the underlying mechanisms of these processes, key mediators of glucose metabolism in the liver, skeletal muscle and fat were further evaluated in the present study.

Methods

The effects of metformin, a popular insulin sensitizer (Sofer et al., 2011) and also the first choice of agents for treating type 2 diabetes (T2DM) (Nathan et al., 2006), and gliclazide, another SU that only binds to SUR1 (Renstrom et al., 2002), were included for comparison in this study.

Animals and treatments

Animal studies are reported in compliance with the ARRIVE guidelines (Kilkenny et al., 2010) and were approved by the Ethics Committee of Qianfoshan Hospital, Shandong University.

Establishment of SUR1^−/− rats

SUR1 plays a major role in SU‐induced insulin secretion; thus, the extrapancreatic hypoglycaemic effects of glimepiride were investigated in SUR1 ^−/− rats in the present study. The SUR1 gene ID (25559) was searched for in the Entrez Gene database (RRID: SCR_002473, URL: http://www.ncbi.nlm.nih.gov/gene). The SUR1 ^−/− rats (Abcc8, the approved symbol for the SUR1 gene was completely ablated) were established by use of transcription activator‐like effector nucleases (TALEN)‐mediated gene targeting technology (Cyagen Biosciences Inc., Guangzhou, China) in the genetic background of Sprague Dawley (SD) rats (RGD Cat# 734476, RRID: RGD_734476), a species that is commonly used in the genome editing field (Wei et al., 2015; Chen et al., 2017). Genomic DNA was isolated from the tail tips of SUR1 ^−/− rats. The primers TALEN‐F (TGCCGAGGTGTACGTGTGTGTGA) and TALEN‐R (CAGATGCAAAGGATTCCTGGGTTG) were used to amplify the region containing the SUR1 target site from the rats. The PCR products that included TALEN‐target sites were purified using the Gel Extraction Kit (Tiangen, Beijing, China) and sequenced in Biosune biotechnology (Shanghai, China) CO.LTD, and gene mutation was analysed in each rat to confirm deletion sites. Rats assessed in the present experiment were 5th generation or more. SUR1 protein expression was examined by Western blotting to further corroborate the successful deletion of the SUR1 gene.

Induction of diabetes

All rats studied were male with a weight of 120–140 g at the beginning of the experiments. The animals were housed in plastic cages, corncob bedding and maintained in a temperature‐ and humidity‐controlled room (24 ± 2°C and 60% humidity) with 12 h light–dark cycles. Diabetes was induced by feeding a high‐fat diet (HFD) and administration of a low‐dose of streptozotocin (STZ, 27.5 mg·kg⁻¹) to the SUR1 ^−/− rats; this method has been in use for several years (Reed et al., 2000; Ti et al., 2011; Geng et al., 2016). Rats were fed a standard irradiated HFD (34.5% fat, 17.5% protein and 48% carbohydrate; Beijing Keaoxieli FEEDS Co., Ltd, China) for 4 weeks. An i.p. glucose tolerance test (IPGTT) and i.p. insulin tolerance test (IPITT) were performed before and after HFD. Diabetes was induced by a single i.p. injection of low dose STZ (27.5 mg·kg⁻¹ i.p. in 0.1 mol·L⁻¹ citrate buffer, pH 4.5; Sigma, St. Louis, MO) to rats with insulin resistance. Blood samples were obtained from the tail vein to test blood glucose, triglyceride (TG) and total cholesterol (TC) levels. One week after STZ administration, rats with a fasting blood glucose >11.1 mmol·L⁻¹ in two consecutive analyses were regarded as diabetic rat models.

Experimental grouping

In view of the influence of SUs–SUR1 interaction‐induced insulin secretion on the results, the following studies were mainly performed in diabetic SUR1 ^−/− rats. Thirty‐six diabetic SUR1 ^−/− rats were randomly divided into four groups: glimepiride (0.5 mg·kg⁻¹·day⁻¹, Sanofi, France; n = 9), gliclazide (10 mg·kg⁻¹·day⁻¹, Servier, France; n = 9), metformin (212.5 mg·kg⁻¹·day⁻¹, Bristol‐Myers Squibb, Shanghai; n = 9) and control (normal saline, n = 9). All drugs and saline were administered by gastric gavage. Body weight, fasting blood glucose, TC and TG were measured each week. After 2 weeks of treatment, IPGTT, IPITT and hyperinsulinaemic‐euglycaemic clamps were performed to test glucose/insulin tolerance and insulin sensitivity. At the end of the experiment, rats were fasted overnight and deeply anaesthetized with an overdose of isoflurane (i.p., Cat# YZ‐1349003, Beijing Solarbio Science & Technology Co., Ltd., Beijing, China). Tissues were immediately dissected, weighed and stored until further analysis.

IPGTT and IPITT

Glucose tolerance was assessed by IPGTT after rats had been fasted for 12 h. A bolus of glucose (1 g·kg⁻¹, i.p.) was injected, and blood samples were collected sequentially from the tail vein at 0, 15, 30, 60 and 120 min. Plasma glucose was measured with a One‐Touch Glucometer (Ascensia Breeze, Bayer, Germany). To evaluate insulin tolerance, IPITT was performed after rats had been fasted for 4 h. A bolus of insulin (1 U·kg⁻¹ i.p.) was administered. Considering that glycaemia is a dynamic phenomenon, mean AUC was calculated for glucose to assess the overall effect and not just at one point of time.

Hyperinsulinaemic‐euglycaemic clamp experiments

Hyperinsulinaemic‐euglycaemic clamps were performed in six rats randomly selected from each group. One week before the clamps, rats were anaesthetized with isoflurane (1.7% end‐tidal concentration) in 40% oxygen, and mechanically ventilated. Assess the depth of anesthesia by lack of reflex to toe pinch. Catheters were implanted in the left femoral artery and tunnelled to the back of the neck, which the rats were unable to reach. Following an overnight fast, a 2 h hyperinsulinaemic‐euglycaemic clamp was performed in conscious rats. A venous indwelling needle was inserted into the tail vein, and three‐way stopcocks were used so that glucose and insulin could be infused simultaneously into the tail veil. Insulin (10 mU·kg⁻¹·min⁻¹) was continuously infused to raise plasma insulin within a physiological range, and an infusion of 20% glucose was started at variable rates throughout the 2 h experiment to maintain the plasma glucose at the basal concentration. An arterial catheter was used for blood sampling into heparin‐containing centrifuge tubes, and the blood samples were centrifuged and stored. At the end of the clamp, rats were killed, and the organs were collected for multiple biochemical analyses.

Hepatic glycogen analysis

Livers were fixed in 4% paraformaldehyde. Then, the tissues were embedded in paraffin and sectioned for 4 μm thick samples. Periodic acid‐Schiff (PAS) staining was used to detect hepatic glycogen deposits. The PAS staining kit were purchased by Jiancheng Bioengineering Institute (Nanjing, China), which was used according to the manufacturer's instructions.

Western blot analysis

For Western blot analysis, tissues were removed and homogenized with commercial RIPA (Beyotime, China) added with protease inhibitor (PMSF, 1:100) and phosphatase inhibitor (1:100) (Beijing Solarbio Science & Technology Co., Ltd.). The protein content was determined by use of a BCA kit (Beyotime). Equal amounts of protein, 50 ug per lane, were subjected to SDS‐PAGE and transferred to PVDF membranes (Millipore Corporation, MA, USA). Blots were blocked for 2 h in 5% nonfat dry milk‐TBS‐0.0% Tween 20 or 5% BSA and then washed. Primary antibodies against glycogen synthase kinase‐3 (GSK3α/β, Ser^21/9) (1:1000, Cat#5676, RRID: AB_10547140), p‐GSK3α/β (Ser^21/9) (1:1000, Cat#9331, RRID: AB_329830), glycogen synthase (GS, 1:1000, Cat#3893, RRID: AB_2279563), p‐GS (Ser⁶⁴¹) (1:1000, Cat#3891, RRID: AB_2116390), Akt (1:1000, Cat#4691, RRID: AB_915783), p‐Akt (1:1000, Cat#4060, RRID: AB_2315049), AMPK (1:1000, Cat#5831, RRID: AB_10622186), p‐AMPK (1:1000, Cat#2535s, RRID: AB_331250), insulin receptor (InsR, 1:1000, Cat#3025, RRID: AB_2280448), phospho‐IGF1 receptor β (Tyr^1135/1136)/insulin receptor β (Tyr^1150/1151) (1:1000, Cat#3024, RRID: AB_331253), insulin receptor substrate (IRS1) (1:1000, Cat#2382, RRID: AB_330333) from Cell Signaling Technology (MN, USA); p‐IRS1 (pTyr⁶¹²) (1:1000, Cat#I2658, RRID: AB_260161) from Sigma‐Aldrich; glucose 6‐phosphatase (G6pase) (1:500, ab83690, RRID: AB_1860503), glucose transporter 4 (GLUT4) (1:1000, ab65267, RRID: AB_1140009), PPAR‐γ (1:1000, ab191407, RRID: AB_2751013) (Abcam Ltd Cambridge, UK) and PEPCK (1:1000, 16754‐1‐AP, RRID: AB_2160031), β‐actin (1:5000, 60008‐1‐Ig, RRID: AB_2289225) (Proteintech Group, Wuhan, China) were purchased. Membranes were incubated with specific antibodies overnight at 4°C followed by an HRP‐conjugated secondary anti‐rabbit antibody (1:10000; #7074, RRID: AB_2099233, Cell Signaling Technology Co., Ltd) for 2 h. Following this, the membranes were washed, and the luminescent signal was detected using the FluorChem E enhanced chemiluminescent system (ProteinSimple, San Jose, CA, USA). Membrane proteins were obtained using a membrane protein extraction kit (PROTMEM, Sigma) to detect the subcellular localization of GLUT4 protein. The values of protein band densities were normalized to those of the control group. Average values of Western blot band density were normalized to the internal control β‐actin and to the data obtained from the control groups and expressed as fold change relative to the control group, so as to control for unwanted sources of variation.

Immunofluorescence staining

Muscle and fat tissues were processed for immunofluorescence staining to evaluate GLUT4 expression. The free‐floating sections of the samples were incubated with anti‐GLUT4 antibody (1:500, ab65267, RRID: AB_1140009, Abcam Ltd Cambridge). Immunoreactivity products were visualized by incubation with appropriate Alexa Fluor dye‐conjugated antibodies (1:200; Cat# A‐21202, RRID: AB_141607, Invitrogen, Carlsbad, CA). All steps were performed in 0.125 to 0.5% Triton X‐100 in PBS. The sections were mounted with Mowiol 4‐88 (Hoechst, Frankfurt, Germany) supplemented with 2.5% anti‐bleaching agent (Sigma‐Aldrich). After being stained, the sections were photographed with an OLYMPUS FSX100 imaging system (Olympus, Tokyo, Japan).

Randomization and blinding

Diabetic SUR1 ^−/− rats were randomly assigned to the experimental groups and treatment conditions therein using a random number generator (https://www.random.org). The specimens of experiments were coded using a random number generator (https://www.random.org). Both biochemical detection and data statistics were carried out in a blind fashion.

Data and statistical analysis

The data and statistical analysis comply with the recommendations on experimental design and analysis in pharmacology (Curtis et al., 2015; Curtis et al., 2018). Data are presented as mean ± SEM. Comparisons between two groups were performed by Student t‐test, while one‐way ANOVA followed by Tukey's post hoc test using SPSS statistics (version 20.0, RRID: SCR_002865, IBM, NY, USA) were used for multi‐group comparison. Post hoc tests were only carried out if F achieved P < 0.05 and there was no significant variance inhomogeneity. Significance was defined as P < 0.05. The graphs were made using Graphpad Prism 6.0.

Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Harding et al., 2018), and are permanently archived in the Concise Guide to PHARMACOLOGY 2017/18 (Alexander et al., 2017a,b,c,d).

Results

Identification of SUR1 ^−/− rats and generation of type 2 diabetes model

SUR1 ^−/− rats: SUR1 genomic structure and the TALEN binding sites were shown in Supporting Information Figure S1A. The genotype of the rat was determined by gene sequencing. The SUR1 ^−/− rats demonstrated the lack of a 16‐base pair segment. The knockout was confirmed by Western blots. Results showed that SUR1 was not expressed in SUR1 ^−/− rats, which indicated that SUR1 ^−/− rats were successfully established (Supporting Information Figure S1B and C).

Generation of type 2 diabetes model based on SUR1 ^−/− rats: After 4 weeks of HFD, insulin resistance was confirmed by IPGTT and IPITT. By IPGTT, the levels of blood glucose in the diabetic group were significantly higher at week 4 than at baseline at all of the time points tested (Supporting Information Figure S2A). Also, AUC for glucose level was higher at week 4 than baseline (P < 0.05) (Supporting Information Figure S2C). Similarly, IPITT revealed impaired insulin sensitivity (Supporting Information Figure S2B and D). Thus, rats showed insulin resistance after 4 weeks HFD. Then, after low‐dose STZ injection (Supporting information, Figure S3), body weight (BW) and serum levels of FBG, TG and TC were increased significantly in diabetic SUR1 ^−/− rats (Supporting Information Table S1, P < 0.05). A total of 36 of the SUR1 ^−/− rats developed insulin resistance, moderate hyperglycaemia and hyperlipidaemia, which were in accordance with the features of type 2 diabetes and included in our following study. Rats without these characteristics were excluded from this study. Among them, two rats (one in the gliclazide group and one in the control group) died before the end of the experiment.

Glimepiride had no significant effect on body weight and plasma lipid

As shown in Figure 1. Before treatment, BW, TG and TC levels were consistent in the four groups. After a 2 week treatment, there were no significant changes in BW, TG and TC in each group, and no significant differences were found between treatment groups and control group (P > 0.05) (Figure 1A, C and D).

Figure 1.

The changes in the metabolic characteristics of T2DM SUR1 ^−/− rats in the four groups before and after treatment. (A–D) Changes in body mass, blood glucose, TC and TG in each group. IPGTT and IPITT were performed and their AUC were calculated (E, G, I and K) before (0 day) and (F, H, J and L) after (14 days) treatment. Data are expressed as mean ± SEM (n = 9 for glimepiride group, n = 8 for gliclazide group, n = 9 for metformin group and n = 8 for control group), *P < 0.05 versus control.

Glimepiride decreased the blood glucose level without increasing the level of insulin

After 2 weeks of treatment, the fasting glucose levels were significantly decreased in the glimepiride (15.70 ± 1.27 vs. 20.71 ± 1.58 mmol·L⁻¹, P < 0.05), gliclazide (16.07 ± 1.47 vs. 20.71 ± 1.58 mmol·L⁻¹, P < 0.05) and metformin (12.56 ± 1.87 vs. 20.71 ± 1.58 mmol·L⁻¹, P < 0.05) group compared with that of the control group (Figure 1B). Gliclazide and glimepiride showed a similar fasting glucose reduction but not as remarkable as that of metformin. There were no significant differences in insulin levels among these groups before and after treatment (both P > 0.05, Supporting Information Figure S4). The above results indicated that glimepiride and gliclazide could still decrease the blood glucose level in T2DM SUR1 ^−/− rats without increasing plasma insulin levels.

Glimepiride alleviated insulin resistance and increased insulin sensitivity in T2DM SUR1^−/− rats

IPGTT and IPITT

IPGTT and ITT assays were respectively used to evaluate the glucose tolerance and insulin tolerance. There was no significant difference among the groups before treatment (P > 0.05, Figure 1E, G, I and K). After 2 weeks of treatment, compared with the control group, the AUCs of IPGTT in the glimepiride (P < 0.05) and metformin (P < 0.05) groups were significantly decreased, while no significant difference was observed between the gliclazide group and control group (P > 0.05, Figure 1F and J). Additionally, IPITT showed that the efficiency of insulin in T2DM SUR1 ^−/− rats treated with metformin, glimepiride and gliclazide was significantly enhanced, compared to that in the control group (P < 0.05) with the following ranking in the improvement of insulin tolerance: metformin > glimepiride ≈ gliclazide > control (Figure 1H and L).

Hyperinsulinaemic‐euglycaemic clamp

To directly examine the quantitative effect of glimepiride on insulin sensitivity, we next subjected rats to hyperinsulinaemic‐euglycaemic clamping. Glucose concentrations were clamped to the same level to ensure that there were no significant deviations in plasma glucose over the course of the experiment. Whole body glucose disposal, reflecting insulin sensitivity in peripheral tissues, was calculated as the sum of glucose infusion rates (GIRs) in the peripheral vein. As shown in Figure 2, compared with the control group, GIR in T2DM SUR1 ^−/− rats treated with glimepiride was significantly increased, a similar effect was found in the gliclazide group. These results indicate that glimepiride did possess extrapancreatic blood glucose‐lowering effect, and gliclazide showed a similar effect as glimepiride in improving insulin sensitivity in T2DM SUR1 ^−/− rats, but both the improvement was inferior to that of metformin.

Figure 2.

Whole body insulin sensitivity in vivo in the four groups. (A) Blood glucose and (B) GIR during the 2 h insulin clamp (10 mU·min⁻¹·kg⁻¹). Data are presented as mean ± SEM (n = 9 for glimepiride group, n = 8 for gliclazide group, n = 9 for metformin group and n = 8 for control group), *P < 0.05 versus control; ^# P < 0.05 versus gliclazide; and ^& P < 0.05 versus glimepiride.

Glimepiride increased hepatic glycogen synthesis and prevented gluconeogenesis

Decreased glycogen synthesis and increased hepatic gluconeogenesis, which are major manifestations of hepatic insulin resistance (Yan et al., 2018), lead to increased blood glucose levels. The effect of glimepiride on hepatic glycogen was identified by PAS staining, which stains stored glycogen as purple particles. As shown in Figure 3A, the level of hepatic glycogen was significantly higher in glimepiride‐treated rats than that of control group. A similar effect was also found in gliclazide‐treated rats, and an increased effect was observed in livers of rats treated with metformin, as indicated by obviously enhanced glycogen staining. In the liver, glycogen synthase kinase 3 (GSK3) mediates the activation of liver glycogen synthase (GS) and plays a key role in hepatic glycogen synthesis (Beurel et al., 2015). Thus, we further investigated the activation of hepatic GSK3 and GS. Our result showed that, compared with control group, the phosphorylation levels of GSK3 (p‐GSK3) were obviously increased and that of GS (p‐GS) were significantly decreased in glimepiride, gliclazide and metformin groups, especially in the metformin group, which was consistent with above results of PAS staining and explained the changes in glycogen content to some extent (Figure 3B, E and F).

Figure 3.

Effects of glimepiride on hepatic glycogen synthesis and gluconeogenesis. (A) PAS staining of liver sections from (a) glimepiride (n = 9), (b) gliclazide (n = 8), (c) metformin (n = 9) and (d) control (n = 8). (B–H) Western blot and analysis of key mediators in hepatic glucose metabolism. Data are expressed as mean ± SEM. *P < 0.05 versus control; ^# P < 0.05 versus gliclazide; and ^& P < 0.05 versus glimepiride.

The levels of PEPCK and glucose‐6‐phosphase (G‐6pase), key enzymes in hepatic gluconeogenesis (Wang et al., 2014), were decreased in livers of T2DM SUR1 ^−/− rats treated with glimepiride. Similar effects were found in the gliclazide group, which were weaker than that of the metformin group (Figure 3B, C and D).

Glimepiride promoted activation of Akt in the livers of T2DM SUR1^−/− rats

Akt, the key molecule in insulin resistance modulation, is upstream of GSK3 and GS, whose activation facilitates hepatic glycogen synthesis (Wang et al., 2014; Liu et al., 2015). The p‐Akt and total Akt in each group were evaluated by Western blotting. Our results showed that the ratio of p‐Akt to total Akt was significantly increased in the glimepiride‐treated SUR1 ^−/− rats, gliclazide‐treated rats and metformin‐treated rats compared with that of control group (Figure 3G). These results are in agreement with the inhibition of GSK3, activation of GS in liver and up‐regulation of hepatic glycogen storage.

AMPK is a potential target for glucose metabolism and plays a key role in the treatment of T2DM (Yan et al., 2018). Hepatic gluconeogenesis was increased in T2DM and AMPK controls hepatic glucose homeostasis mainly through the inhibition of gluconeogenesis. Our results showed that the level of hepatic AMPK phosphorylation at Thr¹⁷² was significantly increased in the liver of T2DM SUR1 ^−/− rats treated with metformin, while not increased in the glimepiride or gliclazide group, compared with the control (Figure 3H).

Glimepiride elevated the expression of total and membrane GLUT4 in muscle and fat tissues of T2DM SUR1^−/− rats

Peripheral tissues, mainly muscle and fat, whose glucose uptake is mainly mediated by GLUT4 (Waldhart et al., 2017). To further elucidate the mechanisms of extrapancreatic blood glucose‐lowering effects of glimepiride, the levels of total GLUT4 in muscle and fat tissues of SUR1 ^−/− rats in each group were examined. As shown in Figure 4A–C, the expression of GLUT4 protein in muscle was increased in the glimepiride and gliclazide groups compared with that of control respectively. The increase was highest in the glimepiride group among the three therapeutic groups.

Figure 4.

Effects of glimepiride on glucose transport in muscle from T2DM SUR1 ^−/− rats. (A) GLUT4 was stained with FITC, revealing the presence of GLUT4 in (a) glimepiride (n = 9), (b) gliclazide (n = 8), (c) metformin (n = 9) and (d) control (n = 8). (B–E) Western blot confirmed the total and membrane GLUT4 protein levels in muscle of rats in the four groups. (F–K) Changes in GLUT4‐related mediators and PPAR‐γ in muscle of rats from these four groups. Data are presented as mean ± SEM. (C, E, H, I and K) *P < 0.05 versus control; ^# P < 0.05 versus gliclazide; and ^& P < 0.05 versus metformin. (J) ^# P < 0.05 metformin versus gliclazide; and ^& P < 0.05 metformin versus glimepiride.

The expressions of total GLUT4 in fat tissues were consistent with that in muscle to some extent. The expressions of total GLUT4 protein were significantly increased in the glimepiride and gliclazide groups, compared with that of the control respectively (Figure 5A–C). Among the three therapeutic drugs, glimepiride showed a more significant up‐regulation in GLUT4 expression than metformin, whereas there was no significant difference compared with that of gliclazide group.

Figure 5.

Effects of glimepiride on glucose transport in fat from T2DM SUR1 ^−/− rats. (A) GLUT4 was stained with FITC, revealing the presence of GLUT4 in (a) glimepiride (n = 9), (b) gliclazide (n = 8), (c) metformin (n = 9) and (d) control (n = 8). (B–E) Western blot confirmed the total and membrane GLUT4 protein levels in fat of rats from the four groups. (F–K) Changes in GLUT4‐related mediators and PPAR‐γ in fat of rats from these four groups. Data are presented as mean ± SEM. (C, E, H, I and K) *P < 0.05 versus control; ^# P < 0.05 versus gliclazide; and ^& P < 0.05 versus metformin. (J) ^# P < 0.05 metformin versus gliclazide; and ^& P < 0.05 metformin versus glimepiride.

GLUT4 is the predominant glucose transporter in muscle and fat, and it mediates cellular glucose uptake when it localizes to the cell membrane (Jaldin‐Fincati et al., 2017). Membrane fractions were isolated from muscle and fat tissues, and the expressions of plasma membrane (PM) GLUT4 of these tissues in each group were also examined. Results showed that the expressions of PM GLUT4 protein in muscle tissues were increased in the glimepiride, gliclazide and metformin groups, compared to that of control group respectively (Figure 4D and E). Comparisons among the three therapeutic groups indicated that glimepiride was superior to gliclazide and metformin in enhancing the expression of PM GLUT4.

Furthermore, compared with the control group, the PM GLUT4 protein in fat tissues were increased in the glimepiride and gliclazide groups respectively (Figure 5D and E). Among the three therapeutic groups, the increase of PM GLUT4 protein in glimepiride group was higher than that of metformin group (P < 0.05), but no significant difference was found when compared with that of gliclazide group (P > 0.05).

Glimepiride increased IRS1 tyrosine phosphorylation and promoted the activation of Akt in muscle and fat tissues

The changes in InsR, IRS1 and Akt in muscles are shown in Figure 4F–I. The levels of p‐IRS1 and p‐Akt were significantly higher in muscles of glimepiride‐treated rats than that of the other three groups. In addition, the expression of p‐Akt was also significantly elevated in the gliclazide group, compared with that of control group. However, compared with the control group, the levels of phospho‐InsR (p‐InsR) in muscle were not increased by glimepiride treatment. Similarly, no alteration was found in the gliclazide and metformin groups, and no significant differences were found among these therapeutic groups (Figure 4G).

The changes in Akt, InsR and IRS1 in fat tissues are shown in Figure 5F–I. Compared with the control, both glimepiride and gliclazide significantly increased the levels of p‐Akt and p‐IRS1, but not the level of p‐InsR. Glimepiride and gliclazide were more effective than metformin at increasing the levels of p‐Akt and p‐IRS1. No significant differences were found in the levels of p‐InsR in fat tissues among the four groups (Figure 5G).

We next investigated the level of p‐AMPK and found that p‐AMPK protein was increased significantly in muscles of rats treated with metformin. Glimepiride and gliclazide had no effect on the activation of AMPK compared to the control group (Figure 4J).

As shown in Figure 5J, the changes in AMPK in fat were similar to that in muscle. Compared with the control group, the expression of p‐AMPK in fat was significantly increased in the metformin‐treated group, but not in the glimepiride or gliclazide group.

Glimepiride increased the expression of PPAR‐γ in muscle and fat tissues

PPAR‐γ, a member of the nuclear receptor superfamily, plays a crucial role in the regulation of glucose metabolism and insulin resistance. Glimepiride has been reported to induce PPAR‐γ expression, thereby improving insulin resistance (Inukai et al., 2005). Thus, the expression of PPAR‐γ was also determined by Western blotting. Results showed that, compared with the control, glimepiride significantly elevated the level of PPAR‐γ in muscles of T2DM SUR1 ^−/− rats. The expression of PPAR‐γ also tended to be increased in the gliclazide group (Figure 4K). Among the three therapeutic groups, the level of PPAR‐γ in the glimepiride group was elevated to a higher extent than that of the gliclazide and metformin group.

The change in PPAR‐γ in fat was similar to that in muscle, except that the increase in the glimepiride group was much higher compared with that in the control group. Compared with control, PPAR‐γ was also significantly increased in the gliclazide but not the metformin group (Figure 5K).

Discussion

SUs have been introduced into the therapy of non‐insulin‐dependent diabetes mellitus with great success. The hypoglycaemic potency of SUs has been attributed primarily to stimulation of insulin secretion by binding to their main functional receptor, SUR1, a major subunit of the ATP‐sensitive K⁺ channel in pancreatic beta cells (Powell et al., 2018). Of note, during glimepiride treatment, extrapancreatic glucose‐lowering mechanisms may operate, which have been suggested in several studies (Muller and Wied, 1993; Mori et al., 2008; Thule and Umpierrez, 2014). However, a lack of sufficient in vivo evidence makes this postulate controversial. In the present study, the data obtained from SUR1 ^−/− rats confirmed the extrapancreatic glucose‐lowering effect of glimepiride in vivo. We observed a decrease in blood glucose as well as an increase in insulin sensitivity in glimepiride‐treated T2DM SUR1 ^−/− rats, which might be attributed to the inhibition of gluconeogenesis, promotion of glycogen synthesis in liver and the increase in GLUT4 and PPAR‐γ in muscle and fat. Our findings make up for the deficiencies in previous studies on the extrapancreatic glucose‐lowering effects of glimepiride.

Insulin resistance mainly occurs in liver, muscle and adipose tissues, which contribute to the impairment of blood glucose metabolism and cause hyperglycaemia (Nooron et al., 2017). Thus, in the present study, key mediators of glucose metabolism in the liver, skeletal muscle and fat of glimepiride‐treated T2DM SUR1 ^−/− rats were examined to elucidate its underlying mechanisms in these insulin target organs.

The liver plays a central role in the maintenance of glucose homeostasis by balancing glycogen synthesis and gluconeogenesis. Hepatic insulin resistance can lead to reduced glycogen synthesis and elevated gluconeogenesis in the liver and then causes hyperglycaemia (Petersen et al., 2017). Our results showed that glimepiride increased the phosphorylation of GSK3 and decreased that of GS. As the rate‐limiting step for glycogen synthesis, GS activation is manifested as a decrease in its phosphorylation through inhibition of GSK3 by increasing its phosphorylation and then finally results in increased glycogen synthesis (Montori‐Grau et al., 2013; Yan et al., 2018). In the present study, the glimepiride‐induced increase in hepatic glycogen storage was accompanied by the inhibition of GSK3 and the activation of GS. Moreover, the expressions of hepatic PEPCK and G6pase, which catalyse the committed steps of gluconeogenesis (Liu et al., 2017), were also inhibited. The above findings suggest that glimepiride could directly improve hepatic insulin resistance, which might be attributed to the increase in hepatic glycogen synthesis and a decrease in gluconeogenesis.

In insulin resistance, hepatic gluconeogenesis and glycogen synthesis are less responsive to insulin despite adequate circulating levels of glucose (Ochiai et al., 2011). Akt, one of the major indicators in the insulin signalling pathway, is considered as a key regulator in the process of gluconeogenesis and glycogen synthesis. Activation of Akt inhibits GSK3 and subsequently activates GS to lead to increased glycogen synthesis in the liver (Miralem et al., 2016).Furthermore, Akt activation also inhibits hepatic PEPCK and G6pase expressions to restrain the gluconeogenesis (Liu et al., 2017). Our further investigation showed that Akt was activated in glimepiride‐treated rats, which suggests that Akt might play a vital part in the above effects of glimepiride on hepatic glycogen metabolism.

Intriguingly, gliclazide showed similar effects to glimepiride in the liver. Additionally, the protective effects of metformin on hepatic glucose metabolism were obviously stronger than that of glimepiride and gliclazide, and this could be attributed to the activation of AMPK, which are consistent with previous studies (Duca et al., 2015). In contrast, no AMPK activation was observed in either the glimepiride or gliclazide group, suggesting that the AMPK pathway was not involved in their effects on hepatic glucose metabolism.

Apart from the imbalance in hepatic glucose metabolism, the glucose transport and uptake in peripheral tissues (muscle and fat) are also insufficient during conditions of insulin resistance (Leto and Saltiel, 2012; Park et al., 2015). Thus, the effects of glimepiride on muscle and fat were next explored in T2DM SUR1 ^−/− rats. Cellular mechanisms of glucose uptake have revealed that the entry of glucose into cells is facilitated by glucose transporter proteins (GLUTs). Among them, GLUT4 is the predominant glucose transporter in muscle and fat tissues (Zorzano et al., 2005; Ng et al., 2008). Both the elevation of GLUT4 protein content and its normal translocation to the cell membrane contribute to the improvement in insulin sensitivity in peripheral insulin target organs (Bogan, 2012; Leto and Saltiel, 2012; Jaldin‐Fincati et al., 2017). In our study, glimepiride and gliclazide, especially the former, significantly increased the expressions of total and plasma membrane (PM) GLUT4 in muscle and fat, which might facilitate glucose uptake and utilization and importantly contribute to their extrapancreatic glucose‐lowering activity. In contrast, metformin showed limited effects on GLUT4 in muscle and fat in T2DM SUR1 ^−/− rats, which were mainly consistent with previous studies (Kristensen et al., 2014). It is meaningful to clarify the potential mechanisms by which glimepiride up‐regulates GLUT4 expression and translocation in muscle and fat. Functional membrane translocation of glucose transporters in insulin‐responsive tissues is primarily regulated by the insulin signalling pathway (Mackenzie and Elliott, 2014). Upon insulin binding to the InsR, the receptor catalyses the tyrosine phosphorylation of the IRS proteins, resulting in the activation of Akt, a key enzyme downstream of p‐IRS, which then accelerates the translocation of GLUT4 and promotes the glucose uptake in muscle and fat (Saltiel, 2016). Different from regular insulin signalling activation, we discovered that both the tyrosine phosphorylation of IRS1 and the activation of Akt were enhanced in glimepiride‐treated T2DM SUR1 ^−/− rats, nevertheless, no activation of InsR was observed. This means glimepiride has a direct effect on the post‐binding pathways of insulin and activates IRS/Akt independently of InsR. Based on the literature (Mori et al., 2008), glimepiride can intercalate the lipid bilayer of cell membranes by acting as a lipophilic molecule, which provides possibilities that glimepiride can directly modulate the insulin signalling cascade in an InsR‐independent manner, so as to stimulate the expression and translocation of GLUT4 in fat and muscle.

Similar but weaker effects were observed in gliclazide‐treated rats. Gliclazide has been reported to promote glucose transport by directly activating a series of enzymes, which seems to initiate the tyrosine phosphorylation of IRS1 and its association with PI3K/Akt (Rodriguez et al., 2004). Furthermore, it has been shown that the activity of insulin receptors in skeletal muscle is unaffected by gliclazide treatment (Palmer and Brogden, 1993). Identical results were found in our study that glicalzide failed to activate InsR in skeletal muscle and fat, which suggests that it possesses a post‐receptor effect on GLUT4. Additionally, as a positive control, metformin elevated the expression of PM GLUT4 in muscle, rather than total GLUT4, which might be associated with the activation of AMPK (Kristensen et al., 2014). But in adipocytes, metformin activated AMPK without increasing PM GLUT4, which was consistent with previous studies (Salt et al., 2000; Ciaraldi et al., 2002).

In addition to conventional insulin signalling, PPAR‐γ also has an important role in regulating insulin sensitivity and glucose transport (Park et al., 2016). PPAR‐γ knockout mice were reported to exhibit aggravated insulin resistance due to an abnormal insulin sensitivity in muscle and adipose tissues (He et al., 2003; Hevener et al., 2003). Our study indicated that glimepiride and gliclazide, especially the former, increased the expressions of PPAR‐γ in muscle and fat, which demonstrated that their protective effects on insulin sensitivity could also be achieved by augmentation of PPAR‐γ. Apart from regulating insulin sensitivity, PPAR‐γ also has positive effects in other pathological processes, such as acting as an anti‐inflammatory and anti‐apoptosis agent (Ko et al., 2016; Massaro et al., 2016), which might also play a part in other unrecognized extrapancreatic effects of glimepiride and gliclazide and contribute to the beneficial effects induced by these SUs.

Moreover, the ability of gliclazide to improve insulin sensitivity in T2DM SUR1 ^−/− rats was equivalent to that of glimepiride, but its effects on key factors that are responsible for glucose uptake were weaker, which suggests that other molecular signals or pathways might be involved in this gliclazide‐induced improvement of insulin sensitivity. As the only SU with an azabicyclo‐octyl ring structure (Palmer and Brogden, 1993), gliclazide can also scavenge ROS, inhibit NADPH oxidase and protect various tissues from injury induced by a hyperglycaemic environment (Onozato et al., 2004; Sena et al., 2009; Chen et al., 2011), which may lead to its additional beneficial effects on insulin sensitivity. This needs to be clarified in future studies.

In summary, this is the first demonstration that glimepiride possesses extrapancreatic glucose‐decreasing effects in the diabetic SUR1 ^−/− rat model. For several decades, SUs were thought to decrease blood glucose mainly by stimulating the secretion of insulin and were merely used to treat T2DM in patients with sufficient beta cell function. The combinination of SUs with twice daily premixed insulin or intensive insulin therapy is not, at present, recommended for treating T2DM. Our results indicate that SUs (glimepiride and gliclazide) can also decrease blood glucose by improving peripheral tissue insulin sensitivity, which suggests that the combination of these SUs with the above insulin therapy should not be excluded per se. Moreover, modifications of SUs to elevate their extrapancreatic effect while maintaining their secretagogue capability would help to generate a more optimized intervention therapy for diabetes in the future.

Author contributions

L.L. and J.D. conceived and designed the study. Z.Z., X.S. and T.X. performed the experiments. X.Z. interpreted the results and R.Z. wrote the manuscript. X.Z., R.Z. and C.X. edited the figures in the manuscript. All authors read and approved the final version of the manuscript. L.L. and J.D. contributed equally to this work.

Conflict of interest

The authors declare no conflicts of interest.

Declaration of transparency and scientific rigour

This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research recommended by funding agencies, publishers and other organisations engaged with supporting research.

Supporting information

Figure S1 The construction and identification of SUR1 ^−/− rats. (A) SUR1 genomic structure and the TALEN binding sites. (B‐C) Western blot and sequence analysis confirmed the deletion of SUR1.

Figure S2 Diabetic rats exhibited insulin resistance. (A‐D) SUR1 ^−/− rats exhibited insulin resistance after 4‐week high‐fat diet. (A‐B) IPGTT and IPITT were performed at baseline and at week 4 in the DM group. AUC for (C) IPGTT and (D) IPITT. Data are present as mean ± SEM. *P < 0.05 vs. baseline.

Figure S3 The comparisons of beta cells in pancreas between (A) HFD fed SUR1 ^−/− rats and (B) HFD + STZ treated SUR1 ^−/− rats by immunofluorescence. Data are present as mean ± SEM. HFD: high‐fat diet.

Figure S4 The comparisons of plasma insulin levels in each group before (0 day) and after treatment (14 days). Data are present as mean ± SEM (n = 9 for glimepiride group, n = 8 for gliclazide group, n = 9 for metformin group and n = 8 for control group).

Table S1 Clinical parameters of rats before and after STZ injection.

Zhou, X. , Zhang, R. , Zou, Z. , Shen, X. , Xie, T. , Xu, C. , Dong, J. , and Liao, L. (2019) Hypoglycaemic effects of glimepiride in sulfonylurea receptor 1 deficient rat. British Journal of Pharmacology, 176: 478–490. 10.1111/bph.14553.