Cathepsin B Contributes to Autophagy-related 7 (Atg7)-induced Nod-like Receptor 3 (NLRP3)-dependent Proinflammatory Response and Aggravates Lipotoxicity in Rat Insulinoma Cell Line

Background: Previous studies show that autophagy deficiency leads to inflammasome activation. Results: Excessive autophagy activation induces a proinflammatory response. Conclusion: Our findings provide new insights into the release mechanisms of proinflammatory cytokines regulated by autophagy. Significance: That autophagy may play a deleterious role in pathogenesis in T2D is a novel finding. Impairment of glucose-stimulated insulin secretion caused by the lipotoxicity of palmitate was found in β-cells. Recent studies have indicated that defects in autophagy contribute to pathogenesis in type 2 diabetes. Here, we report that autophagy-related 7 (Atg7) induced excessive autophagic activation in INS-1(823/13) cells exposed to saturated fatty acids. Atg7-induced cathepsin B (CTSB) overexpression resulted in an unexpected significant increase in proinflammatory chemokine and cytokine production levels of IL-1β, monocyte chemotactic protein-1, IL-6, and TNF-α. Inhibition of receptor-interacting protein did not affect the inflammatory response, ruling out involvement of necrosis. CTSB siRNA suppressed the inflammatory response but did not affect apoptosis significantly, suggesting that CTSB was a molecular linker between autophagy and the proinflammatory response. Blocking caspase-3 suppressed apoptosis but did not affect the inflammatory response, suggesting that CTSB induced inflammatory effects independently of apoptosis. Silencing of Nod-like receptor 3 (NLRP3) completely abolished both IL-1β secretion and the down-regulation effects of Atg7-induced CTSB overexpression on glucose-stimulated insulin secretion impairment, thus identifying the NLRP3 inflammasome as an autophagy-responsive element in the pancreatic INS-1(823/13) cell line. Combined together, our results indicate that CTSB contributed to the Atg7-induced NLRP3-dependent proinflammatory response, resulting in aggravation of lipotoxicity, independently of apoptosis in the pancreatic INS-1(823/13) cell line.


Impairment of glucose-stimulated insulin secretion caused by the lipotoxicity of palmitate was found in ␤-cells.
Recent studies have indicated that defects in autophagy contribute to pathogenesis in type 2 diabetes. Here, we report that autophagy-related 7 (Atg7) induced excessive autophagic activation in INS-1(823/13) cells exposed to saturated fatty acids. Atg7-induced cathepsin B (CTSB) overexpression resulted in an unexpected significant increase in proinflammatory chemokine and cytokine production levels of IL-1␤, monocyte chemotactic protein-1, IL-6, and TNF-␣. Inhibition of receptor-interacting protein did not affect the inflammatory response, ruling out involvement of necrosis. CTSB siRNA suppressed the inflammatory response but did not affect apoptosis significantly, suggesting that CTSB was a molecular linker between autophagy and the proinflammatory response. Blocking caspase-3 suppressed apoptosis but did not affect the inflammatory response, suggesting that CTSB induced inflammatory effects independently of apoptosis. Silencing of Nod-like receptor 3 (NLRP3) completely abolished both IL-1␤ secretion and the down-regulation effects of Atg7-induced CTSB overexpression on glucosestimulated insulin secretion impairment, thus identifying the NLRP3 inflammasome as an autophagy-responsive element in the pancreatic INS-1(823/13) cell line. Combined together, our results indicate that CTSB contributed to the Atg7-induced NLRP3-dependent proinflammatory response, resulting in aggravation of lipotoxicity, independently of apoptosis in the pancreatic INS-1(823/13) cell line.
The prevalence of type 2 diabetes (T2D) 3 is growing rapidly worldwide, and insulin resistance has been implicated as a major risk factor for the disease. Impaired glucose-stimulated insulin secretion (GSIS) is an essential feature of ␤-cell dysfunction, which is widely believed to be secondary to prolonged exposure to high glucose and lipid levels termed "glucolipotoxicity." Palmitate, because of its lipotoxicity, was found to impair the GSIS in ␤-cells (1). Several lines of evidence appear to prove that diabetes is a low grade inflammatory disease with insulin resistance. The low grade inflammatory state of obesity is characterized by increased infiltration of immune cells and an increase in proinflammatory M1 state macrophages in the metabolic tissue, which in turn has been shown to negatively affect insulin sensitivity (2). It has been shown that elevated levels of IL-1␤, IL-6, MCP-1, and C-reactive protein are predictive of T2D (3). The association of cytokines with T2D is also demonstrated by an increased risk of developing the disease for individuals with elevated blood levels of IL-1␤, IL-6, and MCP-1 (4). In addition, the IL-1 receptor antagonist effectively counters the effects of IL-1 in diet-induced obesity (5), and T2D patient therapy with IL-1 receptor antagonist significantly improves the proinsulin/insulin ratio and markers of systemic inflammation (6). Thus, understanding the mechanisms involved in the regulation of diabetic inflammation may help to identify novel therapeutic targets with important clinical applications.
Autophagy is an intracellular, lysosome-mediated catabolic mechanism that is responsible for the bulk degradation and recycling of damaged or dysfunctional cytoplasmic components and intracellular organelles (7). Cathepsin B (CTSB) is a lysosomal cysteine protease primarily involved in the degradation or processing of lysosomal proteins. In certain conditions, CTSB has been shown to be involved in additional cellular processes including cell invasion, vesicle trafficking, inflammasome formation, autophagy-based IL-1␤ secretion (8), and cell death. Induction of autophagy in diabetic and in C57BL/6 mice fed high fat diets revealed active formation of autophagosomes and a higher level of conversion of LC3I to LC3II with low grade systemic inflammation (9). INS-1 cells treated with very high concentrations of palmitic acid (PA) demonstrate increased transcription of CTSB and progressive ␤-cell apoptosis characterized by cellular hypertrophy followed by depletion of insulin immunoreactivity (10,11). Recently, studies have clearly indicated that defects in autophagy contribute to the inflammatory response in T2D. For instance, Atg16L1 is an essential component of the autophagic machinery responsible for regulation of endotoxin-induced IL-1␤ production (12). Defective hypothalamic autophagy directs the central pathogenesis of obesity by induction inflammation and subsequently leads to obesity and insulin resistance in mice fed a high fat diet (13). Depletion of the autophagic proteins LC3B and beclin 1 enhances the activation of caspase-1 and secretion of IL-1␤ and IL-18 (14). However, several recent studies suggest that autophagy also has a deleterious role at the cellular level. For instance, it has been shown that RNAi knockdown of essential autophagy genes inhibits type II programmed cell death in a variety of cell types under different conditions (15). Excessive levels of autophagy could contribute to cardiomyocyte damage and subsequent death (16). At the organismal level, excessive autophagy activation was found to have deleterious effects in runx1 mutant mice, which exhibited severe wasting of enervated myofibers (17). Although these studies have shown that excessive autophagy activation under different pathological conditions plays different negative roles (18), there is currently no information concerning possible involvement of excessive autophagy activation in the pathogenesis of T2D. In addition, there is increasing evidence that autophagy is activated in failing pancreatic cells, a condition associated with insulin resistance, but the mechanism of the dual role of autophagy in T2D has not been explored. In this study, we show that Atg7 induces excessive autophagy activation in the INS-1(823/13) cell line exposed to PA and that Atg7-induced CTSB overexpression contributes to an NLRP3-dependent proinflammatory response and subsequently impairs GSIS independently of apoptosis. Understanding how excessive autophagy activation increases the risk for diabetic inflammation may provide insight into the development of T2D and have important clinical applications.

EXPERIMENTAL PROCEDURES
Palmitic Acid Solutions-Palmitic (C16:0) acid was purchased from Sigma. Palmitic acid solutions were prepared according to Martino et al. (19). Briefly, PA was dissolved at 70°C in 0.1 M NaOH to obtain a 100 mM stock solution. A 5% (w/v) solution of PA-free bovine serum albumin (BSA) was pre-pared in serum-free RPMI 1640 medium. Then a 5 mM PA, BSA mixture was prepared by suitable combination of the two above mentioned solutions. Finally, the PA stock solutions were diluted in RPMI 1640 medium supplemented with 1% fetal bovine serum (FBS) to obtain 0, 0.25, 0.5, and 1 mM final concentrations at a fixed concentration of 0.5% BSA. INS-1(823/ 13) cells were transfected with construct using Lipofectamine Plus (Invitrogen).
Glucose Quantitative Real Time PCR-Total RNA was extracted using TRIzol reagent (Invitrogen), and cDNA was synthesized using the iScript TM cDNA synthesis kit (Bio-Rad) according to the manufacturer's instruction. The sequences of probes and primers are summarized in Table 1. TaqMan PCR was performed using the ABI Prism 7700 Sequence Detection System as instructed by the manufacturer (Applied Biosystems). The level of mRNA was normalized to that of 18 S mRNA. Confocal Microscopy-INS-1(823/13) cells grown in a chamber slide were fixed with paraformaldehyde and permeabilized with 0.5% Triton X-100 for 4 min at room temperature. After washing out primary antibodies, the plates were washed in PBS and incubated with anti-LC3 antibodies diluted in PBA (PBS ϩ 1% (w/v) BSA). Plates were then further washed in PBS and incubated with anti-rabbit Alexa Fluor 633 (Invitrogen). The samples were then further washed in PBS and mounted using Vectashield mounting medium (Abacus ALS, Australia). Images were captured using a Zeiss Meta-510 LSM microscope (excitation, 633 nm; emission, 647 nm). LC3-labeled puncta were counted in individual cells (four to five per field, 20 fields per experiment, three experiments total), and values were averaged. Each value represents a total of 200 cells counted. For the Annexin V/propidium iodide (PI) double staining assay, INS-1(823/13) cells (4 ϫ 10 5 ) were cultured in a 6-well plate. After cells were treated with test agents and harvested, the cells were dyed using a FITC-Annexin V/PI kit. The stained cells were centrifuged at 1,000 rpm for 5 min, and the cell pellets were dropped on sterile coverslips and then examined by fluorescence microscopy.
Flow Cytometric Assay of Apoptosis-5 ϫ 10 5 INS-1(823/13) cells were cultured with or without PA and later incubated for a minimum of 5 h in culture medium at 37°C with or without additional drug or plasmid treatment. Cells were then harvested and fixed with cold 70% ethanol, stored overnight at Ϫ20°C, and stained with PI staining buffer (1 mg/ml RNase A, 0.1% Triton X-100, 50 g/ml PI in PBS) to determine the apoptotic/necrotic cell population (the sub-G 1 /G 0 cell fraction). To evaluate the early stages of apoptosis, treated cells were stained with Annexin V-FITC (Caltag) in Annexin V staining buffer for 15 min at room temperature and counterstained with 50 g/ml PI in Hanks' balanced salt solution (Stem Cell Technologies) with no phenol red and analyzed with a FACSCalibur flow cytometer (BD Biosciences).
Statistics-All values are expressed as mean Ϯ S.D. Differences between means were analyzed using either one-way or two-way analysis of variance followed by Newman-Keuls post hoc testing for pairwise comparison using SPSS. The null hypothesis was rejected when the p value was Ͻ0.05.

PA Induced Autophagy Activation in INS-1(823/13) Cells-
The saturated fatty acid PA was reported to be autophagic to ␤-cells (19). We first sought to determine which PA concentration caused proper levels of both autophagy (Fig. 1) and apoptosis (Fig. 2) 1A). We found the maximum effect with 0.5 mM PA, which significantly enhanced INS-1(823/13) cell autophagy activity after 12-h incubation (p Ͻ 0.01; Fig. 1, A and B). These results were further confirmed by quantification of LC3II/actin in the 0.5 mM PA group compared with the 0 mM PA group (p Ͻ 0.01; Fig. 1B). Notably, the level of LC3II/actin was decreased in the 1 mM PA group in contrast to the 0.5 mM PA group in agreement with previous reports (Fig. 1, A and B) (21,22). Thus, we chose 0.5 mM PA as the proper PA concentration for autophagic activation. We observed numerous and large puncta in the cytoplasm of INS-1(823/13) cells by LC3II staining in the 0.5 mM PA group (Fig. 1D) compared with the control group, which had no detectable dots (Fig. 1C), and the proportion of INS-1(823/13) cells containing LC3II-stained dots/cell was counted. The number of puncta/cell significantly increased at 12 h after 0.5 mM PA treatment as compared with cells that were not exposed to PA (p Ͻ 0.01; Fig. 1E).
Exposure of INS-1(823/13) Pancreatic Cells to 0.5 mM PA Induced Proper Cell Apoptosis-PA has been shown to induce apoptosis in pancreatic ␤-cell lines as well as in primary cultured ␤-cells (23). To investigate which of 0, 0.25, 0.5, and 1.0 mM PA induced appropriate apoptosis in INS-1(823/13) cells, we examined the effect of PA treatment on apoptotic cell death in INS-1(823/13) cells by morphological and FACS analyses (Fig. 2). Annexin V/PI staining indicated that 0.5 mM PA could induce proper apoptosis consistent with a previous report (Fig.  2, C, G, and K) (24). Little if any apoptosis was observed in control cells grown in the presence of a concentration lower than 0.5 mM PA (Fig. 2, A, E, and I, 0 mM PA; Fig. 2, B, F, and J, 0.25 mM PA). Excessive loss of cells through apoptosis was induced by 1 mM PA concentration (Fig. 2, D, H, and I). Treatment of the cells with 0.5 mM PA resulted in a great increase in the amount of early apoptotic cells as analyzed by FACS (Fig.  2M) (0, 0.25, 0.5, and 1 mM PA from left to right, respectively). Although early apoptotic cells showed no differences between 0.5 mM PA and 1 mM PA concentrations, the proportion of late apoptotic cells increased significantly with the latter concentra-tion of PA (p Ͻ 0.01; Fig. 2N). Thus, we chose 0.5 mM PA for the subsequent experiments.
Atg7-induced Excessive Autophagy Activation Resulted in Enhanced Proinflammatory Cytokine Response-To study the effect of Atg7-induced autophagy activation in INS-1(823/13) exposed to 0.5 mM PA, we transfected cells with Atg7-inserted pCMV6-AC-GFP plasmid and a Jab-1-inserted pCMV6-AC-GFP unrelated plasmid as a control. Cells were incubated for 5 h before experiments. The ability of 0.5 mM PA to induce autophagy activation in INS-1(823/13) cells was revealed by increased expression of Atg7 and CTSB and proteolytic conversion of LC3I into the LC3II isoform (Fig. 3A), consistent with Fig. 1A. Significantly higher Atg7-induced CTSB expression and ratio of conversion of LC3I to LC3II were observed in the  0.5 mM PA ϩ Atg7 group compared with the 0.5 mM PA group, and the relative quantification of LC3II/actin and CTSB/actin is shown in Fig. 3B (p Ͻ 0.05) (25). To determine whether the Atg7-induced excessive autophagy activation was an important contributor to the increased expression of proinflammatory genes and that this process was involved in necrosis, we detected the mRNA and protein expression levels of these genes in INS-1(823/13) cells. Surprisingly, as shown in Fig. 3C, Atg7-induced autophagy activation significantly increased mRNA expression of TNF-␣, IL-1␤, IL-6, and MCP-1 in the 0.5 mM PA ϩ Atg7 group in contrast to the 0.5 mM PA group or the 0.5 mM MPA ϩ Jab-1 group, respectively (Fig. 3C) (p Ͻ 0.01). The protein level was higher in the 0.5 mM PA ϩ Atg7 group with respect to the other two groups (Fig. 3C). PA-induced autophagy results in necroptosis in endothelial cells, and it may initiate inflammation (26). To test whether necrosis was involved in Atg7 overexpression in INS-1(823/13) cells, leading to the initiation of inflammation, cells were pretreated with 10 mM 3-methyladenine, a class III phosphatidylinositol 3-kinase autophagy inhibitor, or 30 M necrostatin-1, a receptorinteracting protein inhibitor of necroptosis. Atg7-induced autophagy overexpression was characterized by increased expression of Atg7 and CTSB and an increased conversion ratio of LC3II/LC3I. 3-Methyladenine reduced the LC3II/LC3I ratio and CTSB expression and blocked the mRNA expression of TNF-␣, IL-1␤, IL-6, and MCP-1, whereas necrostatin-1 failed to exhibit any effect on Atg7 overexpression-induced autophagy activity or inflammatory response (Fig. 3D), thus further confirming that the mode of Atg7-induced inflammatory response did not involve necrosis.
Blocking CTSB Expression Suppressed Inflammatory Response but Did Not Affect Apoptosis Markedly-To study whether the induction of proinflammatory cytokine expression in INS-1(823/13) cells was caused by CTSB overexpression, cells were transfected with 1 M siRNA against CTSB for 5 h prior to additional treatment. As expected, we found significantly higher expression of CTSB and a higher conversion ratio of LC3II/LC3I in Atg7-inserted pCMV6-AC-GFP-transfected INS-1(823/13) cells in the 0.5 mM PA ϩ Atg7 group and the 0.5 mM PA ϩ Atg7 ϩ siCTSB group compared with control (0.5 mM PA) (p Ͻ 0.05; Fig. 4A). siCTSB completely inhibited CTSB protein expression in the 0.5 mM PA ϩ Atg7 ϩ siCTSB group ( Fig. 4A) (27). These results were further confirmed by quantification of Atg7/actin, LC3II/actin, and CTSB/actin (p Ͻ 0.05; Fig. 4A, right panel). To validate whether the role of CTSB in proinflammatory induction is related to cell apoptosis, we analyzed the above four groups of cells by FACS. Fig. 4B demonstrates that CTSB was not involved in PA-induced apoptosis as determined by FACS analysis (p ϭ 0.31; Ͼ0.05). However, inhibition of CTSB in the 0.5 mM PA ϩ Atg7 ϩ siCTSB group significantly reduced mRNA expression of TNF-␣, IL-1␤, IL-6, and MCP-1 compared with the 0.5 mM PA ϩ Atg7 group as determined by real time PCR (p Ͻ 0.01; Fig. 4C), suggesting that expression of these genes strongly depends on CTSB protein synthesis in INS-1(823/13) cells. These findings confirmed the pivotal role of Atg7-induced overexpression of CTSB, which contributed to the proinflammatory response in INS-1(823/13) cells.
Blocking Caspase-3 Suppressed Apoptosis but Did Not Affect Inflammatory Response-To examine whether the effect of Atg7-induced CTSB overexpression on the proinflammatory cytokine response required caspase-3, INS-1(823/13) cells cultured with 0.5 mM PA were pretreated with a 50 M concentra-tion of the caspase-3 inhibitor Z-DEVD-fmk (Santa Cruz Biotechnology) before additional treatments. As expected, Z-DEVD-fmk completely inhibited caspase-3 expression in INS-1(823/13) cells, and the protein expression level of CTSB and ratio of LC3II/LC3I was markedly higher in the Atg7 ϩ 0.5 mM PA group and the Atg7 ϩ 0.5 mM PA ϩ Z-DEVD-fmk group compared with the 0.5 mM PA group (Fig. 5A). No differences in the CTSB expression and ratio of LC3II/LC3I were observed between the Atg7 ϩ 0.5 mM PA group and the Atg7 ϩ 0.5 mM PA ϩ Z-DEVD-fmk group (Fig. 5A). These results were further confirmed by quantification of LC3II/actin, cleaved caspase-3/actin, and CTSB/actin (p Ͻ 0.05; Fig. 5A, right  panel). Apoptosis was markedly reduced in cells treated with Z-DEVD-fmk (p Ͻ 0.01; Fig. 5B). To determine whether the induction of the inflammatory response was dependent upon caspase-3 activation, we detected mRNA expression of TNF-␣, IL-1␤, IL-6, and MCP-1 in INS-1(823/13) cells exposed to PA after pretreatment with Z-DEVD-fmk. The mRNA expression of TNF-␣, IL-1␤, IL-6, and MCP-1 was not affected in the group treated with Z-DEVD-fmk compared with the control (Fig. 5C) (28). Taken together, these results indicate that blocking  6). B, cell apoptosis was determined by FACS in the 0.5 mM PA group, Atg7 ϩ 0.5 mM PA group, and Atg7 ϩ 0.5 mM PA ϩ siCTSB group. There were no significant differences among them, although siRNA caused a lower trend of apoptosis. C, relative levels of mRNA of the proinflammatory cytokines were determined by real time PCR. Data are expressed as the mean Ϯ S.D. (error bars) (**, p Ͻ 0.01) (n ϭ 6). caspase-3 suppressed apoptosis but did not affect the proinflammatory response.
NLRP3 Knockdown Blocked Atg7-induced Inflammatory Response-Caspase-1 itself is synthesized as a zymogen, procaspase-1, that undergoes autocatalytic processing, resulting in two subunits that form the active caspase-1. Activation of caspase-1 occurs within the NLRP3 inflammasome following its assembly, which is essential for the cleavage of pro-IL-1␤ into its mature, biologically active form. To define the role of NLRP3 inflammasome components in the activation of caspase-1 by excessive autophagic activation, NLRP3 in INS-1(823/13) cells was depleted by siRNA. Compared with nontarget siRNA control, protein levels of NLRP3 knockdown cells were totally inhibited (Fig. 6A). NLRP3 and non-target control knockdown INS-1(823/13) cells were then challenged with additional treatments, and caspase-1 activation was revealed by the appearance of caspase-1 and IL-1␤ in Western blots. When compared with non-target control cells of the 0.5 mM PA ϩ Atg7 ϩ siControl group, depletion of NLRP3 caused a distinct reduction of IL-1␤ in the 0.5 mM PA ϩ Atg7 ϩ siNLRP3 group (Fig. 6B). Concomitantly, a distinct reduction of caspase-1 levels was also observed (Fig. 6C). Taken together, the results show In the presence of 16.7 mM glucose, the insulin release in INS-1(823/13) cells exposed to Atg7 ϩ 0.5 mM PA was markedly reduced compared with the control group (0 mM PA) (p Ͻ 0.01; Fig. 7A and B). There was similar insulin release between the 0.5 mM PA group and the 0.5 mM PA ϩ Atg7 ϩ siNLRP3 group (Fig.  7, A and B), but these two groups showed a significantly lower insulin release compared with the control group. 0.5 mM PA elevated insulin secretion at low glucose (basal insulin release; 3 mM glucose) (p Ͻ 0.05; Fig. 7B), whereas GSIS at high glucose (16.7 mM glucose) was markedly decreased (p Ͻ 0.05; Fig. 7B) in both the 0.5 mM PA group and the 0.5 mM PA ϩ Atg7 ϩ siNLRP3 group. In the Atg7-and 0.5 mM PA-treated group (0.5 mM PA ϩ Atg7), the GSIS in INS-1(823/13) cells was severely decreased compared with the control (p Ͻ 0.01; Fig. 7B). These results suggest that Atg7-induced excessive autophagy activity was deleterious, impairing GSIS in INS-1(823/13) ␤-cells exposed to PA, and blockade of the proinflammatory response attenuated this effect. To further determine the toxic effects of Atg7-induced excessive autophagy activation, we next investigated the total insulin content in INS-1(823/13) cells (Fig. 7C). INS-1(823/13) cells were treated for 24 h with 0.5 mM PA, 0.5 mM PA ϩ Atg7, and 0.5 mM PA ϩ Atg7 ϩ siNLRP3, which reduced the total insulin content by 38.3, 34.5, and 36.2%, respectively, compared with the control group (0 mM PA) (p Ͻ 0.05), suggesting that INS-1(823/13) cells may be damaged by the toxic effects of Atg7-induced excessive autophagy activation.  Taken together, these data reveal that excessive autophagy activity induced caspase-1 activation and cytokine secretion through an NLRP3-dependent pathway.  6). B, the insulin secretion in INS-1(823/13) cells in response to 3.0 and 16.7 mM glucose stimulation was examined (*, p Ͻ 0.05; **, p Ͻ 0.01) (n ϭ 6). C, total insulin content in INS-1(823/13) cells was extracted by the acid/ethanol method and detected using an RIA kit. Data are expressed as the mean Ϯ S.D. (error bars) (*, p Ͻ 0.05) (n ϭ 6). cells exposed to PA. 3) CTSB-induced MCP-1, TNF-␣, IL-1␤, and IL-6 expression was independent of apoptosis; expression of inflammatory markers upon inhibition of necrosis was not changed, ruling out the involvement of necrosis. 4) siCTSB suppressed the inflammatory response but did not affect apoptosis. 5) Caspase-3 inhibitor Z-DEVD-fmk suppressed apoptosis but did not affect the inflammatory response (30). 6) siNLRP3 blocked the Atg7-induced inflammatory response. 7) Atg7-induced excessive autophagy activation aggravated the resultant GSIS impairment, and siNLRP3 attenuated this effect. Thus, this study has identified for the first time a molecular target of CTSB that contributed to the Atg7-induced NLRP3-dependent proinflammatory response and subsequently led to GSIS aggravation independent of apoptosis.

DISCUSSION
The results in the present study may have an important implication in the cell biology of CTSB. CTSB, a lysosomal cysteine protease, has been suggested to be involved in autophagyinduced inflammasome formation (31,32). We report here that the increased level of CTSB induced by Atg7 transfection acted as a stimulus, which was demonstrated to trigger NLRP3 activation and promote caspase-1-dependent proinflammatory cytokine secretion and may enhance pathogenesis in type 2 diabetes (33). Further evidence from other researchers has indicated that release of CTSB after lysosomal damage can activate NLRP3, consistent with our results (27). In macrophages, excessive autophagy activation enhances lysosomal destabilization and is also associated with NLRP3 activation induced by cholesterol crystals, which are probably involved in the inflammation that promotes atherosclerosis. Inhibition of CTSB suppresses its release from destabilized lysosomes, attenuating the inflammatory response in this report (35) in agreement with our study.
Recent reports reveal a complex interplay between inflammation and autophagic pathways. Blockade of autophagy by genetic ablation of the autophagy regulator Atg16L1 or Atg7 enables LPS-dependent inflammasome activation, suggesting that autophagy normally counters inflammasome activation by LPS (12). It has been suggested that autophagosomes may target inflammasomes for degradation, and possible target reactive oxygen species may indirectly inhibit inflammasome activity (36). However, our results suggest a cross-talk between the toxic effect of excessive autophagy activation and the expression of the proinflammatory cytokines. Our observations are partly congruent with a recent study demonstrating in macrophages that PA could activate NLRP3, whereas there was no response to the unsaturated oleate (37). The two seemingly contradictory functions of autophagy suggest that the complex roles of autophagy can act as either a friend or foe of T2D, depending on its level of activation or the different stages of disease. At the beginning, autophagy provides a protective role of survival advantage to ␤-cells where they suffer from metabolic stress. After induction, excessive autophagy activation instead causes an elevation in the production of CTSB, which can stimulate intracellular inflammasomes, and then activates NLRP3 and subsequently inflammatory caspases (38), which stimulate procytokine secretion or possibly cause autophagic cell death resulting in GSIS impairment. The results reported herein are also consistent with recent studies demonstrating that autophagy might play a deleterious role in acute pancreatitis via the maturation of autophagosomes, which result in CTSB-mediated trypsinogen activation (39).
Previous reports have shown that autophagy is regulated by immunologically relevant cytokines with activation by TNF-␣ and MCP-1, major Th1 cytokines, and inhibition by IL-4 and IL-13, major Th2 cytokines (40). Thus, autophagy is an effector of Th1/Th2 polarization where either Th1 cytokines are protective or Th2 cytokines are permissive when it comes to controlling intracellular pathogens (34). Proinflammatory cytokines usually observed in T2D ultimately disrupt the balance of autophagy in the affected organ. Our findings provide new insights into the release mechanisms of proinflammatory cytokines regulated by autophagy, showing that autophagy also has negative effects on secretion of the proinflammatory Th1 cytokines integrated with aggravation of lipotoxicity. With respect to this possibility, it would be interesting to examine the complex interplay between autophagy and the NLRP3-dependent proinflammatory Th1 response during T2D progression in islets of diabetic patients and animal models.