ox-LDL induces autophagy-mediated apoptosis by suppressing secretagogin-regulated autophagic flux in pancreatic β-cells

Abstract

Lipotoxicity has been shown to induce the loss of functional β-cell mass and lead to type 2 diabetes, but the mechanism remains unknown. In this study, we aim to explore the role of secretagogin (SCGN) in lipotoxicity-induced β-cell injury. Our results indicate that ox-LDL treatment leads to autophagic cell death, as evidenced by decreased cell viability, aggravated cell apoptosis, and the accumulation of the p62 protein in MIN6 cells. LysoTracker Red staining, TEM and mRFP-GFP-LC3 assays demonstrate that autophagic flux is blocked in ox-LDL-treated MIN6 cells. Intriguingly, SCGN is significantly decreased in MIN6 cells under lipotoxic conditions. Additionally, siRNA-guided SCGN knockdown blocks autophagic flux triggered by rapamycin, while SCGN restoration alleviates autophagic flux retardation and mitigates cell apoptosis. The physical interaction between SCGN and SNAP29 is validated by bioinformatics analysis, coimmunoprecipitation assay and SCGN knockdown test. Downregulation of SCGN expression reduces the interaction of these two proteins. Taken together, our results indicate that ox-LDL treatment induces apoptotic β-cell death by blocking autophagic flux dependent on SCGN downregulation. SCGN administration prevents lipotoxic β-cell injury and may be a potential therapeutic strategy to promote β-cell expansion in type 2 diabetes.

Introduction

Chronic exposure to hyperlipidemia results in lipotoxicity in pancreatic β-cells, which contributes to progressive deterioration in β-cell function and mass and ultimately aggravation of type 2 diabetes (T2D) [ 1, 2] . Evidence indicates that a mild concentration and short-term lipid overload can activate the lysosomal degradative pathway of autophagy to degrade lipid droplets in autolysosomes and relieve β-cells from lipotoxic injury [ 3‒ 5] , while sustained high-level lipotoxic exposure can repress the autophagic flux process by blocking autophagosome-lysosome fusion to generate autolysosomes, resulting in β-cell dysfunction and apoptotic death [ 6, 7] . However, the detailed mechanism by which excessive lipid accumulation impairs β-cell autophagic flux remains largely uncharacterized.

The mass and functions of pancreatic β-cells are maintained by a variety of regulatory factors. Secretagogin (SCGN), a novel EF-hand calcium (Ca ²⁺) sensor protein, is predominantly enriched in pancreatic islets and plays crucial roles in preserving β-cell development and functions [ 8‒ 10] . Emerging evidence indicates that SCGN levels in both plasma and pancreatic islets are decreased in mice fed with a high-fat diet (HFD) and in T2D patients [ 11, 12] . The level of SCGN is positively correlated with islet density, islet area and the level of plasma insulin in HFD-fed mice [12], and purified recombinant mouse SCGN protein administration significantly increased β-cell proliferation and insulin secretion [13]. Consistent with these findings, our previous study showed that oxidized low density lipoprotein (ox-LDL) treatment blunted insulin release, exacerbated cell apoptosis, and downregulated SCGN expression, while recombinant plasmid-mediated SCGN restoration ameliorated insulin release and apoptotic cell death [14]. This evidence suggests that SCGN protects pancreatic β-cells against lipotoxic injury. However, whether SCGN modulates lipotoxicity-induced autophagy in β-cells has not yet been established.

Autophagy is a physiologically conserved process for maintaining homeostatic cellular functions through lysosomal degradation of misfolded proteins and damaged organelles [15], which is critical for the proper functioning of β-cells [ 4, 16] . Autophagy is a highly dynamic and complex process that is coordinated by a cascade of regulatory factors [17]. Soluble nethylmaleimide-sensitive-factor attachment receptor (SNARE) complexes, primarily composed of three proteins, syntaxin 17 (STX17), synaptosome associated protein 29 (SNAP29) and vesicle-associated membrane protein 8 (VAMP8), are critical for facilitating autophagosome-lysosome fusion [18]. Usually, STX17 and VAMP8 are anchored in the respective membranes of autophagosomes and lysosomes by their C-terminal domains, whereas SNAP29 is flexible in linking the two anchors; therefore, SNAP29 is reported to be an important target for the regulation of autophagosome-lysosome fusion [ 19‒ 21] . As a member of the SNAP protein subfamily, SNAP29 shares structure and functional homology with SNAP25, while SNAP25 has been shown to interact with SCGN to regulate SNARE complex assembly and to facilitate insulin granule release [ 9, 22, 23] . However, whether and how SCGN is involved in regulating autophagic flux in β-cells under lipotoxicity remains to be determined.

In the present study, we found that ox-LDL treatment triggers autophagy, exacerbates apoptosis, and downregulates SCGN in MIN6 cells. SCGN downregulation blocked the fusion of autophagosomes with lysosomes, while SCGN expression restoration antagonized ox-LDL-induced autophagic flux retardation and cell apoptosis. And the interaction between SCGN and SNAP29 in ox-LDL-treated MIN6 cells was also proven. SCGN might be a key regulator involved in SNARE complex-mediated autophagosome-lysosome fusion.

Materials and Methods

Antibodies, siRNAs and plasmids

The primary antibodies against LC3I/II (12741), SQSTM1/p62 (23214), Caspase3 (9662) and GAPDH (5174) were purchased from Cell Signaling Technology (Beverly, USA), while the primary antibodies against SNAP29 (ab181151) and SCGN (ab137017) were from Abcam (Cambridge, UK). The goat anti-rabbit IgG secondary antibody (ab205718) was from Abcam. SCGN-specific siRNAs, SNAP29-specific siRNAs and their control siRNAs were from RiboBio (Guangzhou, China). A pcDNA3.1 vector containing the full-length cDNA sequence of mouse Scgn (NM_145399) was constructed by Gene Chem (Shanghai, China).

Cell culture and treatment

MIN6 cells were obtained from Cancer Research Institute of Central South University (Changsha, China) and maintained in DMEM (Sigma, Oakville, Canada) containing 10% fetal bovine serum (FBS; Gibco, Carlsbad, USA), 100 IU/mL penicillin, 50 μg/mL streptomycin, 2 mM glutamine and 16.7 mM glucose in a 37°C humidified incubator with 5% CO ₂. Cells in the exponential growth phase were prepared for experimental treatment. First, MIN6 cells were incubated with ox-LDL (YB-002; Yiyuanbiotech, Guanzhou, China) at different concentrations (0, 10, 20, 40 μg/mL) for 24 h or different time intervals (0, 6, 12, 24, 48, 72 h) at a concentration of 40 μg/mL to determine the optimal treatment time and concentration, and subsequent treatments were based on the optimal concentration and treatment time.

SCGN and SNAP29 expression manipulation

For SCGN overexpression, MIN6 cells were transfected with the pcDNA3.1-Scgn vector (rSCGN) using Lipofectamine ^TM 3000 transfection reagent (Invitrogen, Carlsbad, USA), and a cell line stably expressing SCGN was generated and identified. Conversely, knockdown of SCGN by siRNA reduced SCGN expression in MIN6 cells. The specific transfection processes and the detailed siRNA sequences were described in our previous study [14]. For SNAP29 knockdown, MIN6 cells stably expressing rSCGN were transfected with SNAP29-targeting siRNAs (SNAP29-siRNA1, 5′-GAAACATATCAACAGCATT-3′; SNAP29-siRNA2, 5′-CGAACTTATCACCAGAAGA-3′; and SNAP29-siRNA3, 5′-TGACAACCAAAGTGGACAA-3′) or scrambled siRNA (siN0000002-1-5; RiboBio) as the control before treatment with ox-LDL (40 μg/mL). The efficiency of gene overexpression and knockdown was validated by western blot analysis.

Cell viability assay

An MTT assay was used to examine the viability of MIN6 cells after treatment with ox-LDL. Briefly, MIN6 cells were seeded in 96-well plates at a density of 5000 cells per well and incubated for 24 h and then treated with increasing concentrations of ox-LDL (0, 10, 20, 40 μg/mL) for an additional 24 h. Next, the cells were incubated with 20 μL MTT solution (5 mg/mL; TianGen, Beijing, China) for 4 h under regular culture conditions. After the supernatant was removed, the crystal sediments were dissolved in 150 μL of DMSO, followed by measuring the optical density (OD) of each well with a microplate reader (ELx800; BioTek, Vermont, USA) at 570 nm. The cell viability was calculated and represented as a relative value compared to the control group.

Flow cytometric analysis of cell apoptosis

Apoptotic cells were quantified by flow cytometry using Annexin V and propidium iodide (PI) staining according to the protocol of the detection kit (Invitrogen, Carlsbad, USA). Briefly, MIN6 cells were seeded in 6-well plates overnight and then incubated with increasing concentrations of ox-LDL for 24 h. After trypsinization, the cells were collected into 1.5-mL centrifuge tubes and washed twice with ice-cold PBS. Then, the cells were fixed in an ice-cold 80% ethanol solution and stained with Annexin V-FITC and PI. The stained cells were finally counted by using a BD FACSDiva™ flow cytometer (BD Biosciences, Franklin Lakes, USA).

Western blot analysis

For protein expression analysis, cellular proteins were extracted with Mammalian Protein Extraction Buffer (M-PER; Thermo Fisher Scientific, Waltham, USA), and then, the protein concentrations were examined with a BCA protein assay kit (Beyotime, Shanghai, China). Protein lysates were separated by 10% SDS-polyacrylamide gel electrophoresis and subsequently transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, Beijing, China). After being blocked with 5% fat-free milk (Abbott, Shanghai, China) for 2 h, the PVDF membranes were incubated with primary antibodies against LC3I/II (1:1000), p62 (1:1000), Caspase3 (1:1000), SCGN (1:1000), SNAP29 (1:1000) or GAPDH (1:5000). After overnight incubation at 4°C, the membranes were washed 3 times with 1× Tris-buffered saline with 0.1% Tween (TBST) and subsequently incubated with horseradish peroxidase-conjugated secondary antibodies (1:5000) at room temperature for 2 h. The immunoreactive bands were visualized using efficient chemiluminescence reagent (CWBIO, Beijing, China) and further analysed using ImageJ software (National Institutes of Health, Bethesda, USA).

LysoTracker Red staining

To demonstrate that the fusion of autophagosomes with lysosomes was blocked in MIN6 cells treated with ox-LDL, a pH-sensitive LysoTracker Red kit (TianGen), which can monitor the changes in the intralysosomal pH, was employed to detect autophagy-associated lysosomal activity. Briefly, LysoTracker Red showed red fluorescence in a pH-dependent manner in lysosomes, and the increase in fluorescence intensity indicated a decrease in pH. In this experiment, cells were seeded in 6-well plates and cultured overnight and then treated with ox-LDL, rapamycin (Rap, an autophagy inducer; Sigma-Aldrich, St Louis, USA) or bafilomycin A1 (Baf A1, an inhibitor of the lysosomal proton pump V-ATPase), restraining lumenal acidification and lysosomal enzyme activation (Sigma-Aldrich) for 24 h. Next, the culture medium was removed, and the cells were labelled with LysoTracker Red staining solution (50 nM) at 37°C for 30 min. Finally, the staining solution was removed and replaced by fresh cell culture medium, and the fluorescence signal of LysoTracker Red was subsequently observed with a fluorescence microscope (Axioscope 5; Zeiss, Oberkochen, Germany).

Tandem fluorescent-tagged LC3 assay

The effect of ox-LDL or SCGN on the autophagic flux in MIN6 cells was further morphologically traced by transfection with double fluorescent adenovirus vector (mRFP-eGFP-LC3; Gene Chem). In brief, MIN6 cells were infected with mRFP-eGFP-LC3 adenovirus at an MOI of 20 for 8 h, followed by a further 24 h of incubation with refreshed culture medium. Then, the cells were treated with ox-LDL (40 μg/mL), Rap (5 μM), Rap with SCGN siRNA (Rap+si-SCGN), Rap with scrambled siRNA (Rap+Scr-siRNA) or ox-LDL with rSCGN (ox-LDL+rSCGN). After incubation for 24 h, the fluorescence signal was observed at room temperature under a laser-scanning confocal microscope (FV3000; Olympus, Tokyo, Japan). The GFP signal is immediately quenched by the acidic environment of the lysosome lumen. Following the induction of autophagy, yellow spots (merging of the GFP signal and RFP signal) are visible in autophagosomes; however, following the fusion of autophagosomes with lysosomes, GFP fluorescence is quenched, leaving only the red fluorescent signal in autolysosomes.

To demonstrate whether SCGN regulates autophagic flux by interacting with SNAP29, siRNA-mediated SNAP29 knockdown was carried out, and changes in autophagic flux were observed. Briefly, a MIN6 cell line stably expressing rSCGN was generated, and the cells in the exponential growth phase were infected with mRFP-eGFP-LC3 adenovirus and then treated with ox-LDL and SNAP29 siRNA (rSCGN+ox-LDL+si-SNAP29) or with ox-LDL and scrambled siRNA (rSCGN+ox-LDL+Scr-siRNA). Eventually, the fluorescence signal in each group was observed using a laser-scanning confocal microscope (Eclipse Ti; Nikon, Tokyo, Japan).

Transmission electron microscopy (TEM)

Autophagosomes and autolysosomes in MIN6 cells were observed by transmission electron microscopy. Using a cell scraper, MIN6 cells were removed from the bottom of the Petri dishes and then collected and fixed with 2.5% glutaraldehyde in 0.1 M phosphate buffer for 2 h, followed by washing with 0.1 M cacodylate buffer 3 times and postfixed with 1% osmium acid at 4°C for 2 h. Subsequently, the samples were washed with a series of gradient concentrations of ethanol solution (50%, 70%, 80%, 90%, 95% and 100%) and 100% acetone for 10 min per step. After dehydration, the cells were embedded in 1:1 propylene oxide/Epon resin at 37°C overnight and were then polymerized in a 60°C oven for 48 h. Next, ultrathin sections were made using an ultramicrotome and contrast stained with saturated solutions of uranyl acetate. The stained sections were finally observed under a transmission electron microscope (TVEM5; Toshiba, Tokyo, Japan).

Coimmunoprecipitation analysis

Cells were washed with ice-cold PBS and lysed with radio-immunoprecipitation assay (RIPA) buffer (Beyotime). Lysates were incubated with anti-SNAP29 (1:50), anti-SCGN (1:50) or IgG (1:200) antibodies at 4°C overnight, followed by incubation with 50 μL of protein G plus-agarose beads (Santa Cruz Biotechnology, Santa Cruz, USA) at 4°C for 24 h. Next, the beads were washed three times with ice-cold PBS and then resuspended with 5× SDS loading buffer for 9 min. Immunoprecipitated protein was eluted from the beads and analyzed by western blot analysis.

In silico protein binding site prediction

The binding capacity and specific binding sites between SCGN and SNAP29 were predicted using SWISS-MODEL ( https://swissmodel.expasy.org/) and Discovery Studio 4.5 (BIOVIA, Boston, USA) online software.

Statistical analysis

Data are shown as the mean±standard deviation and were calculated by SPSS 18.0 software. Independent experiments were performed at least three times. Differences between the groups were analysed by one-way analysis of variance (ANOVA) followed by Student–Newman–Keuls test. A P value less than 0.05 was considered statistically significant.

Results

Ox-LDL treatment activates autophagy, exacerbates apoptosis, and downregulates SCGN in MIN6 cells

To explore the effect of lipotoxic injury on pancreatic β-cells, we employed insulinoma MIN6 cells as a model system and treated the cells with ox-LDL to mimic lipotoxic injury in vitro. We found that ox-LDL treatment induced the conversion of microtubule-associated protein LC3-I to LC3-II in a time- and dose-dependent manner, while the level of p62 protein was increased aberrantly. A plateau level was found after treatment with ox-LDL (40 μg/mL) for 24 h ( Figure 1A,B), which indicated that autophagy was activated in response to ox-LDL treatment, but autophagic flux was blocked with a high dosage (≥ 20 μg/mL) and sustained ox-LDL exposure (≥24 h). Next, we observed the effect of ox-LDL treatment on cell apoptosis and viability. The results of western blot analysis, flow cytometric analysis and MTT assay also showed that a high dosage (≥ 20 μg/mL) of ox-LDL exposure for more than 24 h increased Caspase-3 protein level, aggravated cell apoptosis, and suppressed cell viability ( Figure 1C‒E and Supplementary Figure S1). Similar to a previous study [14], ox-LDL dose-dependent SCGN downregulation was confirmed ( Figure 1F). Based on the abovementioned results, MIN6 cells treated with 40 μg/mL ox-LDL for 24 h were selected for subsequent experiments.

Figure 1 .

ox-LDL treatment activates autophagy, promotes cell apoptosis, and downregulates SCGN expression in MIN6 cells

(A,B) The levels of autophagic proteins in MIN6 cells treated with increasing concentrations (0, 10, 20, 40 μg/mL) of ox-LDL for 24 h or for various durations (0, 6, 12, 24, 48, 72 h) at a concentration of 40 μg/mL. (C) The expression of caspase 3 in MIN6 cells treated with increasing concentrations of ox-LDL was measured by western blot analysis. (D,E) The ratios of apoptotic cells and the cell viability in MIN6 cells in response to different concentrations of ox-LDL treatment for 24 h were detected by flow cytometry and MTT assay, respectively. (F) The level of SCGN was measured by western blot analysis in MIN6 cells exposed to increasing concentrations of ox-LDL for 24 h. Levels of LC3 II, p62, caspase 3 and SCGN were quantified and normalized to GAPDH. * P<0.05, ** P<0.01 vs the control group (0 μg/mL or 0 h).

ox-LDL treatment blocks autophagic flux in MIN6 cells

p62 is localized to the autophagosome by interacting with LC3 directly and is degraded by the autophagosome-lysosome system in an acidic pH environment. Blocking autophagic flux or neutralizing the acidic pH in the lumen of lysosomal vesicles leads to the frustration of p62 degradation, resulting in the accumulation of p62 protein. To determine whether the accumulation of p62 is due to autophagic flux blockade or lysosomal deacidification in ox-LDL-treated MIN6 cells, we first monitored the pH of the lysosomal lumen by using the LysoTracker Red assay. The results showed that the fluorescence intensity in cells treated with ox-LDL and Rap was obviously increased, while it was quenched in the cells treated with bafilomycin ( Figure 2A,B), which indicated that the acidic environment of the lysosome was not impaired in ox-LDL-treated MIN6 cells. Subsequently, we observed the autophagic flux process in MIN6 cells by using a tandem fluorescent-tagged LC3 (mRFP-EGFP-LC3) assay, and we found that the number of yellow puncta but not red puncta was increased significantly in cells treated with ox-LDL ( Figure 2C), indicating that the fusion of autophagosomes with lysosomes was blocked. Furthermore, we visualized a remarkable increase in autophagosomes in ox-LDL-treated MIN6 cells by TEM ( Figure 2D). These results demonstrate that ox-LDL treatment triggers autophagy in MIN6 cells, while it blocks autophagic flux by impairing the fusion of autophagosomes with lysosomes but not impairing lysosome acidification.

Figure 2 .

ox-LDL treatment blocks autophagic flux in MIN6 cells

(A,B) The pH of the lysosomal lumen in cells treated with ox-LDL, rapamycin (Rap) or bafilomycin A1 (Baf A1) was monitored by a pH-sensitive LysoTracker Red assay (scale bar= 25 μm). (C) The fusion of autophagosomes with lysosomes was traced by mRFP-eGFP-LC3 assay in MIN6 cells exposed to 40 μg/mL ox-LDL for 24 h. The intensity of autophagic flux was assessed by the number of yellow and red puncta (scale bar= 5 μm). (D) Autolysosomes (indicated by red arrows) and autophagosomes (indicated by yellow arrows) in MIN6 cells treated with ox-LDL (40 μg/mL) for 24 h were examined by TEM (scale bar= 1 μm). ** P<0.01 vs the control group.

SCGN downregulation exacerbates autophagic flux blockade

To elucidate whether ox-LDL-induced SCGN downregulation is involved in the inhibition of autophagic flux, siRNA-guided SCGN silencing was performed to mimic ox-LDL-induced SCGN downregulation. Three candidate siRNAs were screened, and the most efficient siRNA was identified as described previously [14]. Next, MIN6 cells were treated with Rap alone, Rap in combination with SCGN siRNA or scrambled siRNA for 24 h, and then, the protein levels of LC3, p62 and SCGN were measured. Compared with the cells treated with Rap or Rap in combination with scrambled siRNA, Rap in combination with SCGN siRNA treatment decreased SCGN expression and significantly elevated LC3II and p62 levels ( Figure 3A). Meanwhile, the results of the mRFP-EGFP-LC3 assay showed that the number of yellow puncta was increased significantly in cells treated with Rap in combination with SCGN siRNA ( Figure 3B), which indicated that SCGN downregulation blocked the autophagic flux process in MIN6 cells. Meanwhile, the effect of SCGN on autophagic flux was further verified by TEM. We found that the number of autolysosomes was obviously increased in MIN6 cells treated with Rap and Rap along with scrambled siRNA, whereas it was significantly increased in the presence of Rap and SCGN siRNA ( Figure 3C). These results demonstrate the role of SCGN in facilitating the fusion of autophagosomes with lysosomes, acting as a pivotal autophagic flux regulator in β-cells.

Figure 3 .

SCGN downregulation exacerbates autophagic flux blockade in rapamycin-treated MIN6 cells

(A) The levels of SCGN, LC3 and p62 in MIN6 cells treated with Rap alone, Rap in combination with SCGN siRNA or with scrambled SCGN siRNA were measured by western blot analysis and normalized to GAPDH. (B) Autophagic flux in MIN6 cells treated with Rap or Rap in combination with SCGN siRNA or with scrambled SCGN siRNA was detected by confocal laser microscopy after transfection with the mRFP-eGFP-LC3 plasmids (scale bar= 5 μm). (C) Autolysosomes (indicated by red arrows) and autophagosomes (indicated by yellow arrows) in each group of cells were observed by TEM (scale bar= 1 μm). * P<0.05, ** P<0.01 vs the control group.

SCGN restoration alleviates ox-LDL-induced β-cell autophagic flux blockade and apoptosis

To further confirm that ox-LDL-induced SCGN downregulation facilitates autophagic flux blockade and associated apoptosis in β-cells, a recombinant pcDNA3.1-Scgn vector (rSCGN)-mediated SCGN restoration experiment was performed in MIN6 cells to assess its function. Briefly, MIN6 cells were treated with ox-LDL or ox-LDL in combination with the rSCGN vector or the control vector. The results showed that rSCGN treatment increased SCGN expression and effectively restored ox-LDL-induced SCGN downregulation ( Figure 4A and Supplementary Figure S2). Pretreatment of MIN6 cells with rSCGN alleviated ox-LDL-induced p62 and Caspase3 elevation and partly antagonized cell apoptosis ( Figure 4A,B and Supplementary Figure S3). Consistent with the change in p62 protein, both mRFP-EGFP-LC3 and TEM assays verified that rSCGN treatment alleviated ox-LDL-induced autophagic flux blockade, reflecting a significant decrease in autophagosomes and an increase in autolysosomes in cells upon SCGN restoration ( Figure 4C,D).

Figure 4 .

Recombinant SCGN treatment alleviates ox-LDL-induced autophagic flux blockade and apoptosis in β-cells

(A) The levels of SCGN, LC3 and p62 in MIN6 cells treated with ox-LDL or ox-LDL in combination with rSCGN or with the control vector were measured by western blot analysis and normalized to GAPDH. (B) The apoptotic ratios of MIN6 cells treated with ox-LDL, ox-LDL in combination with rSCGN or with the control vector were determined by flow cytometric analysis. (C) The autophagic flux in MIN6 cells treated with ox-LDL, ox-LDL in combination with rSCGN or with the control vector was traced by mRFP-eGFP-LC3 assay and detected by laser scanning confocal fluorescence microscopy. Recombinant SCGN treatment significantly decreased the number of autophagosomes (yellow puncta) and markedly increased the number of autolysosomes (free red puncta) in ox-LDL-treated MIN6 cells (scale bar= 5 μm). (D) Autolysosomes (indicated by red arrows) and autophagosomes (indicated by yellow arrows) in each group of cells were observed by TEM (scale bar= 1 μm). * P<0.05, ** P<0.01 vs the control group; ## P<0.01 vs the ox-LDL treatment group.

SCGN regulates ox-LDL-induced autophagic flux by interacting with SNAP29

Since rSCGN treatment facilitates autophagic flux in ox-LDL-treated MIN6 cells, further analysis of the relationship between SCGN and the SNARE complex, a crucial contributor in facilitating autophagosome-lysosome fusion, was conducted by using in silico protein–protein interaction analysis tools, including SWISS-MODEL and Discovery Studio 4.5. In this molecular analysis, we predicted the potential capability of SCGN to bind with the SNARE protein SNAP29, with several amino acids within the amino terminal domain (NT) of SCGN (PKDKITDERVQQI, amino acids 44-56) shown to be critical for binding ( Figure 5A,B). Meanwhile, no potential capability of direct binding was found between SCGN and the other two members of the SNARE complex (data not shown). Furthermore, an interaction between SCGN and SNAP29 was confirmed using a Co-IP assay in MIN6 cells exposed to either normal or ox-LDL-supplemented (40 μg/mL) cell culture medium ( Figure 5C). To further confirm the interaction between the two proteins, a siRNA-guided SNAP29 silencing experiment was conducted. The knockdown efficiency of three candidate siRNAs (SNAP29-siRNA1, SNAP29-siRNA2, and SNAP29-siRNA3) was identified, and SNAP29-siRNA1 was identified as the optimal siRNA used for further tests ( Supplementary Figure S4A). Consistently, the results of autophagic flux-related protein detection, cell apoptosis test, fluorescent-tagged LC3 assay and TEM analysis showed that SNAP29 silencing antagonized rSCGN-mediated autophagic flux restoration and promoted cell apoptosis in ox-LDL-treated MIN6 cells ( Figure 5D‒F and Supplementary Figure S4B). These results confirmed the physical interaction between SCGN and SNAP29 and revealed the pivotal role of SCGN in regulating autophagic flux in ox-LDL-treated β-cells.

Figure 5 .

Prediction and verification of the interaction between SCGN and SNAP29

(A) The three-dimensional structures of SCGN and SNAP29 were modelled, and the interaction between the two proteins was predicted. (B) Potential SNAP29 binding sites in the SCGN protein were predicted by Discovery Studio 4.5 software, and the amino acids (PKDKITDERVQQI, aa 44-56) within the N-terminal domain were predicted to be critical for binding. (C) Interaction of SCGN and SNAP29 was validated by coimmunoprecipitation in MIN6 cells treated with normal medium or medium containing 40 μg/mL ox-LDL. IgG was employed as a negative control. (D) The levels of SNAP29, LC3 and p62 in MIN6 cells treated with ox-LDL in combination with rSCGN and si-SNAP29 or with Scr-siRNA were measured by western blot analysis and normalized to GAPDH. (E) The autophagic flux of MIN6 cells in each group was traced by mRFP-eGFP-LC3 assay (scale bar= 5 μm). (G) Autolysosomes (indicated by red arrows) and autophagosomes (indicated by yellow arrows) in each group of cells were observed by TEM (scale bar= 1 μm). ** P<0.01 vs the Scr-siRNA treated group.

Discussion

Lipotoxicity-induced β-cell injury is a major risk factor contributing to the pathogenesis of type 2 diabetes [ 24, 25] . Autophagy functions to maintain β-cell homeostasis in response to lipotoxic β-cell injury to some extent, while sustained lipotoxic injury may block the autophagic flux process, resulting in apoptotic cell death [ 6, 26] . Therefore, impaired autophagic flux is involved in lipotoxicity-induced β-cell death. The autophagic flux process is regulated at multiple stages, including functional autophagosome and lysosome formation followed by autophagosome-lysosome fusion to generate autolysosomes that degrade misfolded proteins in the cargo [ 27, 28] . This process is precisely regulated by a set of regulatory proteins; however, the mechanism by which autophagic flux is blocked in response to lipotoxicity in β-cells remains unclear.

In the present study, a novel SCGN-mediated autophagy flux regulatory pathway involved in lipotoxic β-cell injury was proposed and investigated. As shown in Figure 6, sustained ox-LDL exposure induced the conversion of soluble LC3-I to LC3-II and promoted the formation of autophagosomes in MIN6 cells, while the level of p62 protein accumulated in the cytoplasm, indicating that the p62-mediated autophagic degradation of misfolded proteins in autolysosomes was retarded. Ox-LDL treatment led to a decrease in SCGN, which hampered the fusion of autophagosomes with lysosomes by reducing the interaction with SNAP29, a pivotal regulator linking STX17 and VAMP8 to form autophagosomal-specific SNARE complexes [29]; therefore, the fusion of autophagosomes with lysosomes was hampered, and apoptotic cell death was initiated.

Figure 6 .

Schematic diagram showing SCGN-regulated autophagic flux in ox-LDL-treated pancreatic β-cells

ox-LDL treatment actives autophagy and induces SCGN downregulation in MIN6 cells. The downregulation of SCGN hampers the fusion of autophagosomes with lysosomes by reducing the interaction with SNAP29, a pivotal regulator linking STX17 and VAMP8 to compose autophagosomal-specific SNARE complexes, leading to apoptotic cell death.

As a calcium sensor protein, SCGN is markedly enriched in the islets and plays crucial roles in the maintenance of β-cell identity and function [8]. SCGN modulates exocytosis in multiple endocrine cells and tissues by interacting with SNAP-25 and facilitating SNARE complex assembly [ 10, 30] , resulting in the fusion of vesicles with the plasma membrane [ 22, 31] . Herein, we report the first evidence that SCGN is involved in regulating autophagic flux in β-cells through binding to SNAP29, a member of the SNAP-25 family, and then facilitates autophagosomal-specific SNARE complex-mediated autophagosome-lysosome fusion. This work broadens our understanding of the role of SCGN in intracellular organelle membrane fusion.

In addition to its role in regulating membrane fusion and facilitating insulin vesicle secretion [30], SCGN also exerts a growth factor-like function in β-cells, as evidenced by β-cell mass loss upon SCGN knockout in mice [12] and β-cell mass recovery in streptozotocin-treated mice upon SCGN administration [13]. Our data on the effect of SCGN on cell survival under lipotoxic conditions further enriched the evidence of SCGN in the modulation of cell proliferation. Intriguingly, recent research revealed that circulatory SCGN can internalize into specialized non-SCGN-expressing cells, such as myoblasts and hepatocytes, through clathrin-mediated endocytosis. Moreover, exogenous SCGN treatment favors growth-related protein synthesis and increases cell viability in these specific target cells under stress [32], which likely can, to some extent, substantiate our findings of SCGN on cell stress responses and lipotoxic β-cell injury.

However, our findings have only provided preliminary evidence based on in vitro tests. Further in vivo experiments carried out in animal models of high-fat diet-fed mice would mimic the lipotoxic injury on β-cells more effectively and provide greater insight. Future studies are also needed to elucidate the structural interactions between SCGN and SNAP29 and to reveal the influence of SCGN on the binding of SNAP29 with STX17 and VAMP8. Moreover, the thorough molecular mechanisms of SCGN in modulating cell proliferation in lipotoxic or other cellular stress responses demand further exploration.

In summary, we found that ox-LDL treatment activates autophagy and decreases SCGN expression in pancreatic β-cells, whereas SCGN is a modulator that promotes the fusion of autophagosomes with lysosomes by facilitating autophagosomal-specific SNARE assembly. The decrease in SCGN hampered autophagic flux and led to apoptotic cell death, while SCGN restoration rescued lipotoxicity-induced β-cell injury. These findings provide novel insight into the mechanism of lipotoxic β-cell injury and prove that SCGN is a pivotal regulator in protective autophagy. The study suggests that functional SCGN supplementation is a promising therapeutic strategy for protecting β-cells from lipotoxic injury and maintaining β-cell mass.

Supplementary Data

Supplementary data is available at Acta Biochimica et Biphysica Sinica online.

COMPETING INTERESTS

The authors declare that they have no conflict of interest.

Funding Statement

This work was supported by the grants from the National Natural Science Foundation of China (No. 82170852), the Hunan Provincial Clinical Medical Technology Innovation Project (No. 2018SK51607), the Outstanding Youth Foundation Supported by the Education Commission of Hunan Province (No. 18B286), and the Scientific Research Project of the Health Commission of Hunan Province (No. B2019125).