Methylglyoxal Impairs Insulin Secretion of Pancreatic β-Cells through Increased Production of ROS and Mitochondrial Dysfunction Mediated by Upregulation of UCP2 and MAPKs

Methylglyoxal (MG) is a highly reactive glucose metabolic intermediate and a major precursor of advanced glycation end products. MG level is elevated in hyperglycemic disorders such as diabetes mellitus. Substantial evidence has shown that MG is involved in the pathogenesis of diabetes and diabetic complications. We investigated the impact of MG on insulin secretion by MIN6 and INS-1 cells and the potential mechanisms of this effect. Our study demonstrates that MG impaired insulin secretion by MIN6 or ISN-1 cells in a dose-dependent manner. It increased reactive oxygen species (ROS) production and apoptosis rate in MIN6 or ISN-1 cells and inhibited mitochondrial membrane potential (MMP) and ATP production. Furthermore, the expression of UCP2, JNK, and P38 as well as the phosphorylation JNK and P38 was increased by MG. These effects of MG were attenuated by MG scavenger N-acetyl cysteine. Collectively, these data indicate that MG impairs insulin secretion of pancreatic β-cells through increasing ROS production. High levels of ROS can damage β-cells directly via JNK/P38 upregulation and through activation of UCP2 resulting in reduced MMP and ATP production, leading to β-cell dysfunction and impairment of insulin production.


Introduction
Diabetes mellitus is a chronic and progressive metabolic condition characterized by hyperglycemia. According to statistics from International Diabetes Federation, as of 2013 an estimated 382 million people worldwide had diabetes, and an estimated 592 million are expected to have diabetes by 2035 [1,2]. About 90% of the cases are type 2 diabetes, which features -cell failure and chronic insulin resistance [3]. While type 1 diabetes is caused by loss of insulin secretion due to destruction of pancreatic -cells [4], progressive -cell failure, including disruption of -cell function and reduction of -cell mass, is the central component of the onset and progression of type 2 diabetes [5]. The mass reduction and dysfunction of -cells have been shown in many studies to lead ultimately to insulin deficiency in patients with type 2 diabetes [3,6]. Collectively, pancreatic -cell dysfunction and reduced insulin secretion play an important role in the pathogenesis of both type 1 and type 2 diabetes.
Methylglyoxal (MG) is a highly reactive intermediate metabolite that normally is produced only in small amounts from basal carbohydrate, lipid, and protein metabolism [7,8]. MG production is highly increased in hyperglycemic disorders such as diabetes mellitus [9][10][11]. It is believed to be a major precursor of advanced glycation end products that are involved in the pathogenesis of diabetes and diabetic 2 Journal of Diabetes Research complications [8]. Several studies have proved that MG can induce damage to tissues, including vascular endothelial and smooth muscle tissue and cells, in diabetes [12,13]. However, data on the direct effects of MG on -cells and insulin secretion are scarce.
Uncoupling protein (UCP) is a mitochondrial inner membrane protein that can decrease metabolic efficiency by dissipating the proton gradient in the mitochondrion from ATP synthesis [14]. UCP2, which was first described in 1997, is expressed in multiple tissues, including the pancreas [15]. Earlier studies demonstrated that UCP2 can decrease formation of reactive oxygen species (ROS) [16,17] and regulate free fatty acid metabolism and transport [14,18]. Increased UCP2 expression under oxidative stress may be a negative modulator of insulin secretion. Recent studies have shown that increased expression of UCP2 can suppress glucose-stimulated insulin secretion [19].
Oxidative stress, which may induce -cell apoptosis and decrease -cell mass, is involved in the pathological process of diabetes [20,21]. Pancreatic -cells are susceptible to ROS, and the action of ROS on these cells is considered a potential mechanism of glucose toxicity in diabetes [20]. ROS can activate the c-Jun N-terminal kinase (JNK) and the P38 mitogen-activated protein kinase (MAPK), both of which may induce mitochondrial dysfunction and then cell apoptosis [22,23]. We and others have shown that increased MG production increases ROS production [24,25].
Although the pathological effects of MG in diabetes and diabetic complications are known, there is little evidence showing whether MG has an impact on insulin secretion by -cells. In the present study, we investigated whether MG impairs insulin secretion by -cells and the potential mechanisms involved. Furthermore, we examined whether N-acetyl cysteine (NAC), an MG scavenger, could prevent the deleterious effects of MG.

Determination of ROS Production and Mitochondrial
Membrane Potential. MIN6 or INS-1 cells were seeded in Corning 24-well tissue culture plates at 2 × 10 5 per well and allowed to attach overnight. Medium containing various concentrations of MG (0.05 mM or 0.1 mM) or H 2 O 2 (0.2 mM) with or without NAC (0.6 mM) was applied and the cells were incubated for 3 h. After that, cells were treated with ROS Detection Solution (1 : 1000 dilution; Reactive Oxygen Species Assay Kit) for 20 min at 37 ∘ C and then washed according to the manufacturer's protocol. The ROS levels were determined by the fluorescence intensity of dichlorodihydrofluorescein diacetate (DCF) with a FACSCalibur Flow Cytometer and fluorescence microscope (Nikon, Tokyo, Japan).
Mitochondrial membrane potential (MMP) was assayed by using JC1 (from the Mitochondrial Membrane Potential Assay Kit with JC1) according to the manufacturer's protocol. The fluorescence was determined with a FACSCalibur Flow Cytometer. Cells in the bottom right gate have low MMP, and the rate of these cells that is low MMP rate can reflect the level of decreased MMP.

Measurement of ATP Production. MIN6 cells were plated
in Corning 6-well tissue culture plates at 1 × 10 6 per well and allowed to attach overnight. The medium was then replaced with medium containing various concentrations of MG (0.05 mM or 0.1 mM) with or without NAC (0.6 mM) and incubated for 3 h. Cells were collected and washed twice with PBS. The sediment was resuspended with ATP extracting solution (100 mM Tris, 4 mM EDTA, pH adjusted to 7.75 with glacial acetic acid) after centrifugation (500 ×g, 4 min). This cell resuspension was heated at 100 ∘ C for 90 s and subjected to centrifugation (10000 ×g, 1 min). The supernatants were extracted for detection of ATP production with the ATP Determination Kit according to the manufacturer's protocol; ATP activity was determined by an Automatic Microplate Reader.
2.6. Determination of Ucp2 mRNA. Total RNA from MIN6 cells treated with MG (0.05 mM or 0.1 mM) with or without NAC (0.6 mM) was extracted by using Trizol (Invitrogen) and reverse-transcribed into cDNA with the PrimeScript RT reagent Kit (Takara Biotechnology, Dalian, China) according to the manufacturer's instructions. Real-time quantitative polymerase chain reaction (PCR) was performed with the QuantiTect SYBR Green PCR Kit (Qiagen, Hilden, Germany). Reactions were performed with the 7500 Standard program on a 7500 Fast Real-Time PCR System (Applied Biosystems, Grand Island, NY). Cycling parameters were as follows: 95 ∘ C 5 min followed by 40 cycles of 95 ∘ C 10 s + 60 ∘ C 30 s. The relative expression of Ucp2 mRNA was normalized by using the 2 −ΔΔCt -method relative to -actin. The specific primers were as follows: forward 5 -GTCGGAGATACCAGAGCACT-3 , reverse 5 -GTGACCTGCGCTGTGGTACT-3 . The primers foractin were forward 5 -GAGACCTTCAACACCCCAGC-3 , reverse 5 -CCACAGGATTCCATACCCAA-3 . All primers were synthesized by Sangon Biotech (Shanghai, China).

Statistical
Analyses. Data were analyzed by analysis of variance (ANOVA) and two-tailed Student's -test and results are expressed as mean ± standard deviation. All experiments were repeated independently at least three times. Statistical analysis was performed by SPSS 17.0 software (Chicago, IL). A value <0.05 was considered statistically significant.

MG Increased ROS Production and Reduced MMP in -
Cells. Incubation of cultured MIN6 cells with MG (0.05 mM or 0.1 mM) for 3 h significantly increased the fluorescence intensity of DCF, an indicator of ROS production, as shown by fluorescence microscopy and flow cytometry (Figures  3(a) and 3(b)). The increased fluorescence intensity of DCF was attenuated by coincubation with NAC (Figures 3(a) and  3(b)). Similarly, ROS production of INS-1 was increased after incubation with MG (0.05 mM or 0.1 mM) for 3 h, which was reversed by coincubation with NAC (Figure 3(c)). The rate of low MMP was increased by exposure to 0.05 or 0.1 mM MG for 3 h, indicating that MG decreased MMP in MIN6 cells; MMP was restored by coculturing with NAC (Figures 3(d) and 3(e)).

MG Increased UCP2 mRNA and Protein
Expression in MIN6 Cells. The Ucp2 mRNA level was markedly increased in MIN6 cells incubated with 0.05 or 0.1 mM MG for 3 h compared with the control group (1.46 ± 0.12 or 1.63 ± 0.18, resp., versus 1.00 ± 0.00 fold change over baseline, < 0.01; Figure 5(a)). The effects of MG on Ucp2 mRNA expression were attenuated by coincubation of NAC (1.20 ± 0.06 versus 1.63 ± 0.18, < 0.01; Figure 5(a)). Similar effects were observed in UCP2 protein levels between the group of cells treated with 0.05 or 0.1 mM MG and the controls (1.44 ± 0.18 or 2.13 ± 0.31, resp., versus 1.00 ± 0.00 fold change over baseline,

Discussion
Despite the recognized pathological role of MG in diabetes and diabetic complications, little previous study has, to our knowledge, investigated the possible impact of MG on insulin secretion or the mechanism underlying any such effect. In this study, we provide evidence that MG impaired insulin secretion in cultured MIN6 and INS-1 cells. The effects of MG on MIN6 and INS-1 cells are dose dependent and can be prevented by NAC, an antioxidant and MG scavenger [12,26]. After demonstrating that MG suppressed insulin secretion by MIN6 and INS-1 cells, we investigated the potential mechanisms underlying this effect. We found that MG increases ROS production and apoptosis in these cells and that this increase of ROS production and apoptosis was attenuated by NAC. These findings suggest that MG may aggravate oxidative stress and apoptosis in pancreatic -cells.
Previous studies have demonstrated that MG can induce ROS production and lead to oxidative stress in various cells [25,27,28]. Oxidative stress can induce apoptosis and increase chemical modification of proteins and lead to tissue damage. Oxidative stress and resultant tissue damage are hallmarks of chronic disease and cell death, and diabetes is no exception [29]. Furthermore, excessive ROS levels can damage -cells and induce -cell dysfunction [30]. Thus, increased ROS production and apoptosis by MG are responsible for the MGinduced impairment of -cells and insulin secretion. Mitochondrial dysfunction is a central contributor to -cell failure. UCP2 is a member of a family of proteins that are located in the mitochondrial inner membrane and act as proton channels to uncouple mitochondrial oxidative phosphorylation [31]. UCP2 is involved in various physiological and pathological processes including insulin secretion [14]. Increased ROS can induce UCP2 upregulation [14]. UCP2 facilitates proton leak to reduce the MMP and thus attenuates ATP synthesis. Reduced ATP production results in the dysfunction of -cells [31]. It has been reported that UCP2 negatively regulates ATP production and insulin secretion [14]. Indeed, optimal ATP synthesis is necessary for efficient activation of the triggering pathway of insulin secretion [32]. We have shown that MG increased UCP2 expression and reduced MMP and ATP production. Our findings indicate that ROS activate UCP2, which results in proton leak across the mitochondrial inner membrane and reduced -cell ATP production, leading to impairment of insulin secretion.
JNK and P38 MAPKs are members of the complex superfamily of MAP serine/threonine protein kinases, which are activated by phosphorylation. JNK and P38 MAPKs are stress-activated kinases and are responsive to stress-inducing  stimuli such as ROS [30,33]. It has been reported that oxidative stress induces damage to -cells by activating stresssensing pathways via JNK and P38 [23,34]. When isolated rat islets were exposed to oxidative stress, the JNK, P38, and protein kinase C pathways were activated. Activation of the JNK pathway can induce -cell apoptosis and decrease pancreatic and duodenal homeobox factor-1 (PDX-1) activity and subsequent suppression of insulin gene transcription and resultant inhibition of insulin secretion in the diabetic state [34]. In the present study, we demonstrate that not only MG increased apoptosis in MIN6 and INS-1 cells but it also increased the expression and activation of JNK and P38. The effects of MG on MAPK upregulation were earlier than the biological effects and continued to increase as the time was prolonged. We speculate, therefore, that increasing the expression of JNK/P38 is another mechanism by which MG can cause cell damage and impair insulin secretion. Since H 2 O 2 can induce ROS production, MIN6 cells were incubated with H 2 O 2 to verify the effects of ROS in MIN6 cells. Results showed that apoptotic rate and MAPK protein expression were increased. These further confirmed that the effects of MG in pancreatic -cells might be related to ROS production. Studies have shown that the effects of MG can be attenuated by NAC [12,35]. Our findings are consistent with the previous studies. Our findings may explain the previous observation that chronic infusion of MG reduces insulin secretion and the development of type 2 diabetes in SD rats and expand the underlying mechanisms of MGinduced impaired insulin secretion [36,37].
In conclusion, the data presented here demonstrate that MG can damage insulin secretion of pancreaticcells through increased production of ROS and apoptosis, mitochondrial dysfunction, and upregulation of UCP2 and MAPKs. Based on our findings, we postulate that MG induces increased production of ROS, which may trigger the upregulation of both UCP2 and JNK/P38. Increased JNK and P38 can directly damage -cells causing cell dysfunction. UCP2 can reduce MMP and ATP synthesis, resulting incell damage and impairment of insulin secretion (Figure 9). These findings reveal a novel pathological role and potential mechanisms of MG in the pathogenesis of diabetes. These findings also provide evidence that MG could be a new therapeutic target in diabetes management and that improving -cell function should be given attention in the treatment of diabetes.

Disclaimer
The authors alone are responsible for the content and writing of this paper.