PLAGL1 overexpression exacerbates type 1 diabetes by inducing β‐cell apoptosis via oxidative stress‐dependent dual DNA damage and cGAS/STING pathway activation

ABSTRACT

Background

Type 1 diabetes mellitus (T1DM) arises from autoimmune destruction of pancreatic β‐cells. Pleomorphic adenoma gene‐like 1 (PLAGL1) overexpression has been linked to β‐cell apoptosis, but molecular mechanisms remain incompletely understood. This study explored whether PLAGL1 exacerbates T1DM by promoting oxidative stress‐induced DNA damage and activating the cGAS/STING inflammatory pathway.

Methods

The mouse β‐cell line NIT‐1 was transfected with PLAGL1 overexpression plasmids or specific siRNA. Mitochondrial and nuclear DNA damage was assessed through comet assays, 8‐OHdG ELISA, and Western blot analysis of key DNA repair proteins, including XRCC1, OGG1, and PARP1. Oxidative stress was evaluated by measuring superoxide dismutase (SOD) activity and the glutathione redox state (GSH/GSSG ratio), while apoptosis was examined via expression levels of BCL2, BAX, and cleaved Caspase‐3. To investigate pathway involvement, pharmacological inhibitors—RU.521 (targeting cGAS) and H‐151 (targeting STING)—were applied. In NOD mice, PLAGL1 overexpression was combined with cGAS/STING inhibition; glucose tolerance was subsequently evaluated, and pancreatic tissue was subjected to histopathological examination.

Results

Overexpression of PLAGL1 triggered substantial mitochondrial and nuclear DNA damage, which was accompanied by elevated oxidative stress and compromised DNA repair. Consequently, cytoplasmic DNA accumulated, leading to activation of the cGAS/STING pathway and subsequent β‐cell apoptosis and functional decline. Treatment with the cGAS inhibitor RU‐521 or the STING inhibitor H‐151 markedly attenuated apoptosis and restored insulin secretion in PLAGL1‐overexpressing NIT‐1 cells. In NOD mice, PLAGL1 overexpression accelerated diabetes progression, whereas inhibition of the cGAS/STING axis preserved β‐cell mass, improved glucose homeostasis, and sustained insulin output. Histological evaluation further confirmed that inhibition of this signaling pathway helped maintain normal islet architecture.

Conclusion

Our findings demonstrated that PLAGL1 exacerbates β‐cell loss in type 1 diabetes by driving oxidative DNA damage and activating the cGAS/STING signaling cascade. Therapeutic intervention targeting this axis may therefore represent a promising strategy to protect β‐cells and attenuate disease progression.

ABBREVIATIONS

2^(−delta delta Ct)

2^(−ΔΔCt)

8‐OHdG

8‐hydroxy‐2′‐deoxyguanosine

ANOVA

analysis of variance

ATM

ataxia‐telangiectasia mutated

AUC

area under the curve

BAX

BCL2‐associated X protein

BCA

bicinchoninic acid

BCL2

B‐cell lymphoma 2

BSA

bovine serum albumin

C‐Caspase‐3

Cleaved caspase‐3

cDNA

complementary DNA

cGAS/STING

cyclic GMP‐AMP synthase/stimulator of interferon genes

DAPI

4′,6‐diamidino‐2‐phenylindole

DDR

DNA damage response

ELISA

enzyme‐linked immunosorbent assay

GSH/GSSG

glutathione/glutathione disulfide

GSIS

glucose‐stimulated insulin secretion

H&E

hematoxylin and eosin

HRP

horseradish peroxidase

IHC

immunohistochemistry

Ins1

insulin 1

Ins2

insulin 2

IRF3

interferon regulatory factor 3

KD

knockdown

KO

knockout

MAFA

v‐maf musculoaponeurotic fibrosarcoma oncogene homolog A

NC

National Cancer Institute

NF‐κB

nuclear factor‐κB

NOD/LtJ mice

non‐obese diabetic/LtJ mice

NRF2

nuclear factor E2‐related factor 2

NS

not significant

OE

overexpression

OGG1

8‐oxoguanine DNA glycosylase 1

OGTT

oral glucose tolerance test

PARP1

poly (ADP‐ribose) polymerase 1

PBS

phosphate‐buffered saline

PDX1

pancreatic and duodenal homeobox 1

PLAGL1

pleomorphic adenoma gene‐like 1

POLG

polymerase γ

qRT‐PCR

quantitative real‐time polymerase chain reaction

RIPA

radioimmunoprecipitation assay

RNA

ribonucleic acid

ROS

reactive oxygen species

SDS‐PAGE

sodium dodecyl sulfate‐polyacrylamide gel electrophoresis

SOD

superoxide dismutase

T1DM

Type 1 diabetes mellitus

TUNEL

terminal deoxynucleotidyl transferase‐mediated dUTP nick‐end labeling

WB

Western blot

XRCC1

X‐ray repair cross‐complementing protein 1

γH2AX

phosphorylated histone H2AX

INTRODUCTION

Type 1 diabetes mellitus (T1DM) is an autoimmune disorder driven by T cell‐mediated destruction of pancreatic β‐cells, culminating in absolute insulin deficiency and hyperglycemia ¹ . Within the pathological cascade of T1DM, the DNA damage response (DDR) in pancreatic β‐cells has emerged as a critical mediator of disease progression. Sustained DDR activation not only promotes apoptotic cell death but may also induce cellular senescence and functional decline ² . In particular, unresolved DNA damage can trigger a senescent phenotype in β‐cells, and the subsequent accumulation of these senescent cells is thought to further aggravate the disease process ² . Supportively, in chemically induced DNA damage models, β‐cells display hallmark features of senescence, including persistent DDR signaling and irreversible growth arrest ² . The role of these senescent β‐cells in T1DM pathogenesis may be substantial, potentially modulated by genetic background variations that influence both the cellular response to DNA damage and insulin secretory capacity. Moreover, gene regulatory studies indicate that factors such as TCF19 overexpression can modulate the DDR in β‐cells ³ , while inhibition of the deubiquitinating enzyme USP1 has been shown to suppress DDR‐mediated apoptosis, thereby conferring protection to β‐cells ⁴ . In summary, DNA aggregation and damage in pancreatic β‐cells are of great significance in the pathogenesis of T1DM, and in‐depth studies of these mechanisms will help develop new therapeutic strategies to protect β‐cells from damage and apoptosis.

As a suppressor of cell growth, pleomorphic adenoma gene‐like 1 (PLAGL1) can induce cell apoptosis ⁵ . Studies have shown that overexpression of PLAGL1 may lead to apoptosis of pancreatic β‐cells, thereby triggering the occurrence of T1DM ⁶ . Additionally, PLAGL1 expression is closely associated with DNA damage in pancreatic β‐cells, which may lead to loss of cell function and progression of diabetes ² . Studies have found that PLAGL1 expression may regulate the function and survival of pancreatic β‐cells by influencing DNA methylation status ⁷ . These findings suggest that PLAGL1 may be involved in the pathological process of diabetes through multiple mechanisms. In summary, the role of PLAGL1 in T1DM is multifaceted, involving multiple biological processes such as gene expression regulation, DNA damage repair, and apoptosis. These studies provide a basis for further exploring the possibility of PLAGL1 as a therapeutic target for diabetes ⁸ , ⁹ .

In T1DM patients, DNA damage in pancreatic β‐cells is considered to be caused by oxidative stress and immune attack, and this damage is partially correlated with changes in PLAGL1 expression ¹⁰ . Oxidative stress plays a crucial role in the DDR process of pancreatic β‐cells ¹¹ . Oxidative stress results from redox balance disorders in the body, leading to excessive production of reactive oxygen species (ROS), thereby inducing cellular damage and DNA damage ¹² . Studies have shown that oxidative stress not only causes DNA damage in pancreatic β‐cells but also induces apoptosis and dysfunction ¹³ . Oxidative stress activates multiple intracellular signaling pathways that play important roles in β‐cell apoptosis ¹⁴ . Moreover, oxidative stress further exacerbates β‐cell damage and dysfunction by inducing endoplasmic reticulum stress and protein misfolding ¹⁵ , ¹⁶ . In T1DM, these defense mechanisms may be insufficient to counteract persistent oxidative stress ¹⁷ . In summary, oxidative stress promotes DNA damage and dysfunction in pancreatic β‐cells through multiple mechanisms, thereby accelerating T1DM progression. Understanding these mechanisms is of great significance for developing new therapeutic strategies to protect pancreatic β‐cells from oxidative stress damage.

DNA damage in pancreatic β‐cells may exacerbate pancreatic islet inflammation by activating the cGAS/STING pathway, promoting type I interferon signaling and NF‐κB‐mediated inflammatory responses ⁶ . This inflammatory state is closely associated with β‐cell DNA damage responses and may be a key event in the early stages of T1DM ¹ . While our previous work established that PLAGL1‐induced cytoplasmic DNA accumulation activates the cGAS/STING pathway to drive β‐cell apoptosis ¹⁵ , the initial trigger and the full spectrum of DNA damage induced by PLAGL1 remained elusive. Specifically, it was unclear how PLAGL1 overexpression leads to DNA damage in the first place and whether it affects both mitochondrial and nuclear genomes. Therefore, the present study was designed to test the novel hypothesis that PLAGL1 acts through inducing oxidative stress, which serves as the upstream driver to cause dual DNA damage (both mitochondrial and nuclear), ultimately culminating in cytoplasmic DNA accumulation and activation of the apoptotic cascade. This mechanistic insight represents a significant advance beyond our previous findings, as it identifies oxidative stress as a critical therapeutic target and provides a more comprehensive understanding of the DNA damage landscape in PLAGL1‐mediated β‐cell failure.

MATERIALS AND METHODS

Reagents

Mitochondrial DNA replicase polymerase γ (POLG) siRNA was purchased from Shanghai Genechem Co., Ltd. (Shanghai, China). PLAGL1 and ATM kinase overexpression plasmids were constructed by Shanghai GeneChem (Shanghai, China) using the pLVX‐IRES‐ZsGreen1 vector. The mitochondrial‐targeted antioxidant mitochondria‐targeted antioxidant (MitoQ) (Cat. No. HY‐100116A) and STING inhibitor H‐151 (Cat. No. HY‐112693) were both obtained from MedChemExpress (USA). N‐acetylcysteine (NAC) (Cat. No. A9165) and MTT assay kit (Cat. No. 11465007001) were purchased from Sigma‐Aldrich (USA). The cGAS inhibitor RU.521 (Cat. No. S9117) was acquired from Selleck (USA). Antibodies against γH2AX (Cat. No. ab81299), XRCC1 (Cat. No. ab235196), OGG1 (Cat. No. ab233214), PARP1 (Cat. No. ab32064), BCL2 (Cat. No. ab182858), BAX (Cat. No. ab32503), Caspase‐3 (Cat. No. ab32351), and Cleaved Caspase‐3 (Cat. No. ab32042) were all purchased from Abcam (UK). The 8‐OHdG ELISA kit (Cat. No. ab285254) was obtained from Abcam (UK). The Superoxide Dismutase (SOD) activity assay kit (Cat. No. ab65354) and GSH/GSSH detection kit (Cat. No. ab239709) were both purchased from Abcam (UK). The OxiSelect single‐cell gel electrophoresis kit (Cat. No. STA‐355‐5) was obtained from Cell Biolabs Inc (USA). The nuclear isolation kit (Cat. No. 78833) was purchased from Thermo Fisher (USA). The Annexin V‐FITC/PI apoptosis detection kit (Cat. No. 556547) was acquired from BD Biosciences (USA). Unless otherwise specified, all other reagents were purchased from Beyotime Biotechnology (Shanghai, China).

Cell treatment and transfection

The mouse pancreatic islet β‐cell line NIT‐1 was purchased from American Type Culture Collection (ATCC, Cat. No. CRL‐2055). The NIT‐1 cells were cultured in RPMI‐1640 medium containing 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin, under a humidified atmosphere of 5% CO₂ at 37°C. For transfection, 2 × 10⁵ NIT‐1 cells were seeded into 6‐well plates, and Lipofectamine 3000 was used to transfect PLAGL1 overexpression plasmid (2 μg/well) or PLAGL1 siRNA (50 nM). Cells were collected 48 h after transfection. For POLG knockdown and ATM overexpression, POLG siRNA (50 nM) or ATM overexpression plasmid (2 μg/well) was transfected into NIT‐1 cells, and mitochondrial DNA damage and repair protein expression were detected 72 h after transfection.

Mitochondrial function and oxidative stress detection

Mitochondria were isolated from NIT‐1 cells using a combination of differential and density gradient centrifugation. Briefly, NIT‐1 cells were collected and resuspended in lysis buffer and homogenized gently. The cell suspension was centrifuged at 1,000 × g for 10 min at 4°C, and the supernatant was further centrifuged at 12,000 × g for 15 min at 4°C to pellet the crude mitochondrial fraction. The crude mitochondrial extracts were resuspended in 1 mL of lysis buffer and layered onto a pre‐prepared OptiPrep gradient (15%–30%–45%), followed by ultracentrifugation at 100,000 × g for 2 h at 4°C. Mitochondria at the 30%–45% interface were collected, washed with PBS, and centrifuged at 12,000 × g for 15 min. SOD activity and GSH/GSSG ratio in the isolated mitochondria were determined using the SOD activity assay kit and GSH/GSSG detection kit, respectively.

DNA damage and repair analysis

Nuclear DNA fragmentation was assessed by single‐cell gel electrophoresis kit (Trevigen, USA, #4250‐050‐K). Tail Length, Tail DNA%, and Tail Moment were analyzed using CometScore software. Additionally, the expression of DNA repair proteins in mitochondria or nuclei of NIT cells was analyzed by Western blot. Briefly, total proteins from mitochondria, nuclei, or whole NIT‐1 cells were lysed with RIPA buffer at low temperature. Total protein concentration was determined using a BCA kit. After SDS‐PAGE electrophoresis and membrane transfer, membranes were incubated with primary antibodies overnight at 4°C, followed by horseradish peroxidase (HRP)‐conjugated secondary antibodies for 1 h at room temperature. Band intensity was quantified using Image Lab 6.0 software (Bio‐Rad), and relative expression levels were calculated using β‐actin as an internal reference.

Apoptosis detection

Apoptosis levels in NIT‐1 cells were detected by flow cytometry. Briefly, 1 × 10⁶NIT‐1 cells were trypsinized and washed three times with PBS. Cells were resuspended in 100 μL of Binding Buffer, and 5 μL of Annexin V‐FITC and 5 μL of PI (50 μg/mL) were added. Subsequently, 400 μL of Binding Buffer was added to each sample, and apoptosis was immediately quantified using a BD FACSCanto™ flow cytometer (BD Biosciences). Fluorescence was measured with excitation at 488 nm; FITC and PI emissions were collected through 530/30 and 670/30 nm bandpass filters, respectively.

Glucose‐stimulated insulin secretion (GSIS) assay

NIT‐1 cells were seeded in 24‐well plates at a density of 5 × 10⁵ cells per well. Prior to the assay, cells were pre‐incubated for 2 h at 37°C in Krebs‐Ringer buffer containing a low concentration of glucose (2.8 mM). Subsequently, the cells were stimulated for 1 h at 37°C with fresh Krebs‐Ringer buffer containing either a low (2.8 mM) or a high (25 mM) glucose concentration. The supernatants were then collected and centrifuged at 1,000 × g for 5 min at 4°C to remove any cellular debris. Insulin concentration in the supernatant was determined using a mouse insulin ELISA kit, strictly following the manufacturer's protocol. The glucose stimulation index (GSI) was calculated as the ratio of insulin secretion under high‐glucose conditions to that under low‐glucose conditions.

qRT‐PCR method

Total RNA was extracted from NIT‐1 cells using TRIzol reagent (A260/A280 ≥ 1.8). cDNA was synthesized using PrimeScript™ RT Master Mix with the following program: 37°C for 15 min → 85°C for 5 s → 4°C storage. cDNA was amplified using the following program: 95°C for 30 s → 40 cycles (95°C for 5 s, 60°C for 30 s) → melting curve analysis (65°C–95°C, 0.5°C/s). β‐actin was used as an internal reference, and relative gene expression levels were calculated using the 2^−ΔΔCt method. qRT‐PCR primers were as follows:

Ins1: F: GCTTCTTCTACACACCCATGT, R: AGTGCAGCACTGATCCACA;

Ins2: F: GCTTCTTCTACACACCCATGT, R: AGTGCAGCACTGATCCACA;

PDX1: F: CAGCCGAGAGTGGAAGAAAC, R: GTTGCAGTTCCTCCGCTTAC;

MAFA: F: CAGCCTCCCACTTCACAGAT, R: GCGGTAGTTGTGGTTGTTCC;

β‐actin: F: GGCTGTATTCCCCTCCATCG, R: CCAGTTGGTAACAATGCCATGT.

In vivo animal experiments

Female NOD/LtJ mice aged 12–16 weeks were purchased from Beijing HFK Bioscience Co. Ltd (Beijing, China). All mice were divided into four groups: empty vector control group (tail vein injection of empty vector containing 5 × 10⁸ pLVX‐IRES‐ZsGreen1), PLAGL1 overexpression group (tail vein injection of adenovirus containing 5 × 10⁸ pLVX‐PLAGL1‐IRES‐ZsGreen1 particles), cGAS inhibitor intervention group (intraperitoneal injection of 10 mg/kg RU.521), and STING inhibitor intervention group (intraperitoneal injection of 10 mg/kg H‐151). Interventions were performed three times weekly for 8 weeks. After the intervention, mice were fasted for 6 h, and fasting blood glucose levels were measured using a handheld blood glucose monitor. Insulin levels in NOD mice were determined using an insulin ELISA kit. Subsequently, blood glucose and insulin levels were measured at 0, 15, 30, 60, 90, and 120 min after oral administration of 2 g/kg glucose using the same method. After 8 weeks of continuous intervention, mice were euthanized. Pancreatic tissues were fixed in 4% paraformaldehyde and embedded in paraffin for sectioning (4 μm). For H&E staining, sections were deparaffinized and hydrated, stained with hematoxylin and eosin, and observed under an optical microscope (Nikon Eclipse E100) to assess pathological changes in pancreatic tissues. For immunohistochemistry (IHC), sections were stained with anti‐insulin antibody, and positive areas were quantified using NIS‐Elements BR 5.30 software (Nikon). For TUNEL detection, sections were stained according to the TUNEL apoptosis detection kit instructions. Images were captured under a fluorescence microscope (Olympus BX53) at excitation wavelengths of 488 nm (TUNEL‐FITC) and 358 nm (DAPI). The percentage of TUNEL‐positive cells in islet total cells was calculated using Image J software. Animal experiments were approved by the institutional Animal Care and Use Committee (IACUC) of the animal Experimental Ethics Committee of Shenglun Biotechnology Co., LTD (Suzhou, China) with approval NO. 2024072224.

Statistical analysis

Data were statistically analyzed using GraphPad Prism 6.0 (CA, USA) and presented as mean ± standard deviation. Two‐sided Student's t‐test or one‐way analysis of variance (ANOVA) followed by Tukey's post‐hoc test was used to analyze differences between two groups or among multiple groups, respectively. All data met the assumptions for parametric tests unless otherwise noted in the figure legends or results. *P < 0.05 was considered statistically significant.

RESULT

Overexpression of PLAGL1 induces mitochondrial DNA damage by promoting oxidative stress and impeding DNA repair

Building upon our previous finding that PLAGL1 overexpression triggers cytoplasmic DNA accumulation and subsequent apoptosis ¹⁵ , we sought to delineate the underlying mechanisms. Our data revealed a significant increase in cytoplasmic DNA levels upon PLAGL1 overexpression in NIT‐1 cells (P < 0.001, Figure 1a). This effect was substantially mitigated by siRNA‐mediated knockdown of POLG (P < 0.01), implicating mitochondrial DNA damage as a primary source of the cytoplasmic DNA. Additionally, PLAGL1 overexpression was associated with the activity of superoxide dismutase (SOD), the GSH/GSSH ratio in mitochondria of NIT‐1 cells, and cell viability of NIT‐1 cells (P < 0.001) (Figure 1b,c). Both the MitoQ and NAC significantly reversed these changes (all P < 0.05). Interestingly, compared with MitoQ, NAC improved NIT‐1 cell viability to a greater extent but alleviated mitochondrial oxidative stress to a lesser extent, indicating that nucleolar oxidative stress also participates in cytoplasmic DNA accumulation induced by PLAGL1 overexpression (Figure 1d). ELISA analysis showed that PLAGL1 overexpression significantly increased the expression of the mitochondrial DNA oxidative damage marker 8‐OHdG. Overexpression of ATM, a key DNA repair enzyme, significantly reversed the elevated 8‐OHdG levels (P < 0.05) (Figure 1e). Western blot analysis showed that compared with the empty vector group, the expression of DNA repair proteins such as XRCC1, OGG1, and PARP1 in mitochondria was significantly decreased in the PLAGL1 overexpression group. Conversely, ATM overexpression significantly upregulated the expression of these DNA repair proteins (Figure 1f–i). Taken together, PLAGL1 overexpression exacerbates mitochondrial oxidative stress, which in turn induces mitochondrial DNA damage and concurrently impairs the DNA repair machinery, ultimately leading to cytoplasmic DNA accumulation.

Figure 1.

PLAGL1 exacerbates oxidative stress and impairs DNA repair in NIT‐1 cells. (a) Cytoplasmic DNA accumulation, (b) SOD activity, and (c) relative GSH/GSSG ratio. (d) Cell viability of NIT‐1 cells. (e) ELISA analysis of 8‐OHdG level. (f) Representative Western blot image of γH2AX, XRCC1, OGG1, and PARP1. (g–i) Quantitative analysis of (g) XRCC1, (h) OGG1, and (i) PARP1 protein levels normalized to β‐Actin. Data are mean ± SD (n = 5 independent cell cultures); *, **, *** indicates P < 0.05, 0.01, 0.001, respectively vs. Vector group; ^#,^##,^### indicates P < 0.05, 0.01, 0.001, respectively vs. PLAGL1_OE group.

Mechanism of PLAGL1 overexpression on nuclear DNA damage

The mechanism by which PLAGL1 overexpression induces nuclear DNA damage in NIT‐1 cells was further analyzed. First, comet assays were used to observe nuclear DNA damage in NIT‐1 cells after PLAGL1 overexpression. As shown in Figure 2a–c, PLAGL1 overexpression significantly increased Tail length, Tail DNA, and Tail moment in NIT‐1 cells, indicating severe DNA strand breaks. Conversely, knocking down PLAGL1 significantly reversed these effects. Meanwhile, PLAGL1 overexpression significantly reduced cell viability, while knocking down PLAGL1 partially rescued the viability of NIT‐1 cells, indicating a direct link between DNA damage accumulation and impaired cell survival (Figure 2d). These results suggest that PLAGL1 overexpression exacerbates nuclear DNA damage in NIT‐1 cells, while its knockdown alleviates this phenotype. To assess DNA double‐strand breaks, we measured the overall protein level of the DNA damage marker γ‐H2AX by Western blot. The results indicated an increase in γ‐H2AX signal in the PLAGL1 overexpression group compared to the control (Figure 2e,f). This imbalance indicates that the DNA repair mechanism is impaired under PLAGL1 overexpression. Conversely, the DNA repair enzyme ATM partially reversed the expression of these DNA damage repair proteins (Figure 2e–i). Together, these results demonstrate that PLAGL1 overexpression induces nuclear DNA damage in NIT‐1 cells by increasing DNA strand breaks and inhibiting repair proteins. These mechanisms collectively contribute to the reduction in NIT‐1 cell viability, highlighting the dual role of PLAGL1 in promoting DNA damage accumulation and inhibiting repair processes.

Figure 2.

PLAGL1 Overexpression exacerbates DNA damage and impairs repair capacity in NIT‐1 cells. (a–c) Quantitative analysis of (a) Tail length, (b) Tail DNA percentage and (c) Tail moment in comet assay. (d) Cell viability of NIT‐1 cells. (e) Representative Western blot image of γH2AX, XRCC1, OGG1, and PARP1. (f–i) Quantitative analysis of (f) γH2AX, (g) XRCC1, (h) OGG1, and (i) PARP1 protein levels normalized to β‐Actin. Data are mean ± SD (n = 5 independent cell cultures); *, **, *** indicates P < 0.05, 0.01, 0.001, respectively vs. Vector group; ^##,^### indicates P < 0.01, 0.001, respectively vs. PLAGL1_OE group.

Overexpression of PLAGL1 promotes apoptosis of pancreatic islet β cells by activating the cGAS/STING pathway

Having established that PLAGL1 overexpression mediates cytoplasmic dsDNA accumulation and thereby triggers cGAS/STING pathway activation, we next investigated the functional consequence of this signaling on pancreatic β‐cell apoptosis. As illustrated in Figure 3a, PLAGL1 overexpression markedly reduced NIT‐1 cell viability by approximately 60% compared to the empty vector control (P < 0.001). This loss in viability was significantly rescued, with levels restored to approximately 70% of baseline, upon concurrent knockdown of either cGAS or STING (both P < 0.01). Expression of apoptosis‐related proteins was analyzed by Western blot, and results are shown in Figure 3b–e. PLAGL1 overexpression significantly downregulated the anti‐apoptotic protein BCL2 (P < 0.001), upregulated the pro‐apoptotic protein BAX (P < 0.001), and increased the C‐Caspase‐3/Caspase‐3 ratio (P < 0.001), indicating increased apoptosis in NIT‐1 cells. Knockdown of cGAS or STING significantly reversed these effects to baseline levels (all P < 0.001). Collectively, these results demonstrate that the activation of the cGAS/STING signaling pathway is a critical mechanistic link between PLAGL1‐driven genomic instability and the execution of β‐cell apoptosis, primarily through the initiation of the mitochondrial apoptotic pathway, as evidenced by BCL2/BAX imbalance and Caspase‐3 activation.

Figure 3.

Effects of PLAG1 overexpression and cGAS/STING modulation on NIT‐1 cell viability and apoptosis‐related proteins. (a) Cell viability of NIT‐1 cells. (b) Western blot image of BCL2, BAX, Caspase‐3, Cleaved‐Caspase‐3, and β‐Actin. (c–e) Quantitative analysis of (c) BCL2, (d) BAX protein level normalized to β‐Actin as well as (e) Cleaved‐Caspase‐3/Caspase‐3 ratio. Data are mean ± SD (n = 5 independent cell cultures); **, *** indicates P < 0.01, 0.001, respectively vs. Vector group; ^##,^### indicates P < 0.05, 0.01, 0.001, respectively vs. PLAGL1_OE group.

Overexpression of PLAGL1 impairs pancreatic islet β cell function by activating the cGAS/STING pathway

To further delineate the functional impact of the cGAS/STING pathway on β‐cell apoptosis and functional integrity in the context of PLAGL1 overexpression, we employed specific pharmacological inhibitors. As illustrated in Figure 4a, PLAGL1 overexpression markedly promoted apoptosis in NIT‐1 cells, resulting in a 2.8‐fold increase in the apoptosis rate compared to the vector control group (P < 0.001). This effect was substantially attenuated by treatment with the cGAS inhibitor RU.521 or the STING inhibitor H‐151, which reduced the apoptosis rate to 19.1% and 18.6%, respectively (both P < 0.01), indicating that pharmacological blockade of the cGAS/STING pathway effectively suppresses PLAGL1‐triggered apoptotic signaling. Fluorescence co‐localization analysis showed that PLAGL1 overexpression significantly reduced the overlap rate of insulin (red) and nucleus (DAPI, blue) signals compared to the vector group (P < 0.001) (Figure 4b). Treatment with RU.521 or H‐151 significantly restored the overlap rate (both P < 0.01).

Figure 4.

Effects of PLAGL1 overexpression and pharmacological interventions on β‐cell apoptosis and function. (a) Quantification of apoptosis (%) in NIT‐1 cells. (b) Fluorescence images of DAPI (blue), insulin (red), and merged signals in NIT‐1 cells. (c) Glucose‐stimulated insulin secretion (stimulation index) in NIT‐1 cells. (d–g) mRNA fold changes of (d) Ins1, (e) Ins2, (f) PDX1, and (g) MAFA. Data are mean ± SD (n = 5 independent cell cultures); *, **, *** indicates p < 0.05, 0.01, 0.001, respectively vs. Vector group; ^#,^## indicates p < 0.05, 0.01, respectively vs. PLAGL1_OE group.

Consistent with these findings, assessment of glucose‐stimulated insulin secretion (GSIS) demonstrated that the stimulation index was significantly impaired in PLAGL1‐overexpressing cells, declining from 1.63 (vector group) to 1.08 (P < 0.001). Inhibition of cGAS or STING restored the index to 1.36 and 1.31, respectively (both P < 0.01), confirming that pathway blockade contributes to the functional recovery of β‐cells (Figure 4c). Expression of downstream insulin genes and regulatory factors was further analyzed by qRT‐PCR. As shown in Figure 4d–g, PLAGL1 overexpression significantly inhibited the expression of insulin synthesis genes Ins1 and Ins2, as well as the core transcription factors PDX1 and MAFA for β‐cell differentiation (all P < 0.01) through a cGAS/STING‐dependent mechanism. Additionally, treatment with RU.521 and H‐151 upregulated the expression of the above genes, respectively (all P < 0.05). These results indicate that PLAGL1 overexpression drives both apoptosis and functional impairment in β‐cells—characterized by compromised insulin transcription, disrupted subcellular localization, and diminished secretory capacity—largely through activation of the cGAS/STING signaling axis. The partial rescue achieved via its pharmacological inhibition underscores the pivotal role of this pathway in mediating PLAGL1‐induced β‐cell damage and highlights its potential as a therapeutic target for intervention.

Overexpression of PLAGL1 exacerbates diabetic phenotypes in NOD mice by activating the cGAS/STING pathway

The regulatory role of the cGAS‐STING axis on diabetic phenotypes was further validated in PLAGL1‐overexpressing NOD mice. As shown in Figure 5a, there was no significant change in body weight among groups of NOD mice. Interestingly, compared with the vector group, PLAGL1‐overexpressing mice showed significantly higher fasting blood glucose (P < 0.01) and significantly lower fasting insulin levels (P < 0.001), exhibiting the typical characteristics of insulin deficiency and β‐cell failure (Figure 5b,c). After intervention with RU.521 or H‐151, fasting blood glucose in NOD mice decreased and fasting insulin levels partially recovered (both P < 0.05). In oral glucose tolerance tests (OGTT), the area under the 2‐h blood glucose curve (AUC) in the PLAGL1_OE group was nearly doubled compared to the vector group (P < 0.01), while the insulin AUC was approximately halved compared to the vector group (P < 0.001) (Figure 5d,e). Compared with the PLAGL1 overexpression group, treatment with RU.521 and H‐151 significantly reversed these changes (all P < 0.05). Histopathological results showed that H&E staining of islets in the PLAGL1_OE group revealed a loose structure and reduced islet area (Figure 5f). Additionally, immunohistochemistry (IHC) confirmed that the proportion of insulin‐positive areas in the PLAGL1_OE group (15.69 ± 3.69%) was significantly lower than that in the NC group (43.29 ± 5.09%). Treatment with RU.521 (32.59 ± 4.12%) and H‐151 (33.87 ± 3.97%) significantly reversed the above histopathological changes. These results indicate that pharmacological inhibition of the cGAS/STING pathway effectively improves glucose homeostasis and preserves β‐cell mass in PLAGL1‐overexpressing NOD mice. The consistency between these therapeutic benefits and the specific molecular pathway targeted in our study provides strong in vivo support for the pathophysiological relevance of the cGAS/STING activation mechanism, which was mechanistically delineated in our in vitro experiments.

Figure 5.

Effects of PLAGL1 overexpression and pharmacological interventions on metabolic and histological parameters in NOD mice. (a–c) Bar graphs comparing (a) body weight, (b) fasting blood glucose, and (c) Fasting insulin across experimental groups. (d, e) Dynamic (d) blood glucose and (e) insulin levels during 0–120 min monitoring. (f) Representative histological images of pancreatic tissue: H&E staining, immunohistochemistry (IHC) assay. Data are mean ± SD (n = 10 mice per group); *, **indicates p < 0.05, 0.01, respectively vs. Vector group; ^#,^##indicates p < 0.05, 0.01, respectively vs. PLAGL1_OE group.

DISCUSSION

The overexpression of PLAGL1 triggered a series of complex biological responses in pancreatic islet β cells, ultimately leading to cell apoptosis. Studies have shown that oxidative stress is a critical factor contributing to cellular DNA damage and apoptosis ¹⁸ , ¹⁹ . In this context, PLAGL1 may influence cellular survival and function by regulating the intracellular redox state ¹⁹ . In pancreatic islet β cells, PLAGL1 overexpression may also promote cell apoptosis by affecting mitochondrial function ²⁰ . Mitochondrial dysfunction is typically associated with increased intracellular reactive oxygen species (ROS), which in turn cause DNA damage and apoptosis ²⁰ . Therefore, PLAGL1 may indirectly induce cellular DNA damage by influencing mitochondrial function and oxidative stress levels. In this study, after PLAGL1 overexpression, the cytoplasmic DNA level in NIT‐1 cells significantly increased, accompanied by a significant reduction in antioxidant enzyme activity. Further knockdown of the mitochondrial DNA replicase POLG significantly reversed the cytoplasmic DNA level, suggesting that PLAGL1 overexpression can induce mitochondrial oxidative stress and promote DNA damage in NIT‐1 cells, consistent with previous research findings ¹⁹ . Additionally, PLAGL1 overexpression significantly increased the expression of 8‐OHdG, a marker of mitochondrial DNA oxidative damage, and decreased the expression of DNA repair proteins. After overexpressing ATM, a key DNA repair enzyme, these changes were significantly reversed. The key finding of this study is that PLAGL1 overexpression triggers severe oxidative stress, which acts as the primary upstream event leading to dual DNA damage and the subsequent activation of the cGAS/STING‐mediated apoptotic pathway in pancreatic β cells. This discovery provides a deeper mechanistic explanation for our previous report that PLAGL1 causes cytoplasmic DNA accumulation, by identifying the root cause and expanding the damage profile to include the nuclear genome.

PLAGL1 overexpression induces cytoplasmic DNA accumulation in cells, a process that activates the cGAS/STING signaling pathway, thereby further stimulating downstream IRF3 and NF‐κB pathways and enhancing type I interferon signaling and NF‐κB‐mediated inflammatory responses ⁶ . This mechanism indicates that PLAGL1 plays a crucial role in the DNA damage response, particularly in promoting DNA damage by inhibiting DNA damage repair pathways. We observed nuclear DNA damage in NIT‐1 cells after PLAGL1 overexpression through comet assays. Conversely, knockdown of PLAGL1 significantly reversed these effects. In other studies, inhibiting the activity of the DNA damage sensor protein PARP1 can suppress DNA repair ²¹ . The accumulation of PARP1 in the nucleolus accelerates DNA repair and cell survival ²¹ . Overexpression of OGG1, a DNA oxidative damage repair protein, can significantly inhibit the damage of oxidized low‐density lipoprotein to cells and improve DNA repair function ²² . In this study, PLAGL1 overexpression significantly upregulated the expression of γH2AX, a DNA damage marker in the nucleolus of NIT‐1 cells, while significantly downregulating the expression of key DNA repair proteins such as XRCC1, OGG1, and PARP1. Furthermore, DNA repair agonists significantly reversed the above changes. These results suggest that the mechanism by which PLAGL1 promotes DNA damage in pancreatic islet β cells by inhibiting DNA damage repair pathways is similar to the DNA damage response mechanisms identified in other studies, collectively revealing the critical role of DNA damage repair in cell survival and disease progression.

First, PLAGL1 overexpression leads to cytoplasmic DNA accumulation, which is believed to occur through activation of the cGAS/STING signaling pathway ⁶ . Activation of the cGAS/STING pathway further triggers downstream activation of the IRF3 and NF‐κB pathways, enhancing type I interferon signaling and NF‐κB‐mediated inflammatory responses, thereby exacerbating the apoptotic process ⁶ . This mechanism has been observed in multiple pathological conditions. For example, in myocardial ischemia–reperfusion injury, activation of the cGAS/STING pathway is closely associated with myocardial cell apoptosis and dysfunction ²³ . Additionally, in intestinal ischemia–reperfusion injury, activation of the cGAS/STING signaling pathway is considered a key driver of inflammation and apoptosis ²⁴ . In diabetic cardiomyopathy, activation of the cGAS/STING signaling pathway is associated with inflammation and apoptosis in myocardial cells, and BRG1 deficiency exacerbates this effect ²⁵ . These studies indicate that the cGAS/STING signaling pathway plays a critical role in the pathological processes of multiple diseases, influencing disease progression and therapeutic effects by regulating apoptosis and inflammation. Therefore, we further analyzed the effect of the cGAS/STING pathway on apoptosis in PLAGL1‐overexpressing pancreatic islet β cells. PLAGL1 overexpression significantly reduced the viability of pancreatic islet β cells, significantly upregulated the expression of pro‐apoptotic proteins, and downregulated the expression of anti‐apoptotic proteins. Interestingly, knockout of cGAS or STING significantly reversed these effects to baseline levels. This also confirms that PLAGL1 overexpression induces apoptosis in pancreatic islet β cells by activating the cGAS/STING signaling pathway and triggering the mitochondrial‐dependent apoptotic pathway (BCL2/BAX imbalance and Caspase‐3 activation).

Inhibiting the cGAS/STING pathway plays a crucial role in restoring the function of PLAGL1‐overexpressing β cells. Studies have shown that the cGAS/STING signaling pathway plays a key role in the occurrence and development of diabetes, particularly in regulating insulin deficiency and β cell dysfunction ²⁶ . Inhibiting cGAS or STING can alleviate β cell dysfunction caused by PLAGL1 overexpression, thereby improving glucose‐stimulated insulin secretion (GSIS) function ⁶ . Furthermore, research has found that STING has a unique role in regulating glucose homeostasis, particularly in the regulation of insulin sensitivity and insulin secretion ²⁷ . Inhibiting STING can improve insulin deficiency and glucose intolerance induced by a high‐fat diet while protecting the normal function of β cells ²⁷ . Consistent with previous findings, intervention with the cGAS inhibitor (RU.521) or STING inhibitor (H‐151) significantly reduced apoptosis in pancreatic islet β cells after PLAGL1 overexpression and improved insulin secretion function. Additionally, previous studies have shown that cGAS inhibition can also promote β cell proliferation through a STING‐independent but CEBPβ‐dependent mechanism, thereby improving insulin secretion and glucose tolerance ²⁸ . The independence of this mechanism indicates that the role of cGAS in β cells is not solely dependent on the STING signaling pathway but is achieved by regulating other downstream factors such as CEBPβ. Interestingly, in this study, the cGAS inhibitor slightly outperformed the STING inhibitor in improving insulin secretion capacity in PLAGL1‐overexpressing pancreatic islet β cells. Moreover, in diabetes treatment strategies, regulation of the cGAS/STING signaling pathway is considered a potential target. Inhibiting this pathway can reduce inflammation and apoptosis, thereby protecting β cell function ²⁹ . This protective effect is crucial for maintaining the expression of insulin synthesis genes, as their expression directly affects insulin synthesis and secretion ²⁷ . Consistent with previous findings, inhibiting cGAS or STING in a PLAGL1‐overexpressing pancreatic islet β cell model significantly upregulated the expression of downstream genes related to insulin synthesis and secretion. In summary, inhibiting cGAS or STING not only restores the function of PLAGL1‐overexpressing β cells but also improves insulin secretion and glucose metabolism through multiple mechanisms, providing new insights and potential therapeutic targets for diabetes treatment.

In diabetes research, activation of the cGAS/STING pathway is closely associated with insulin deficiency and β cell dysfunction. Targeting this pathway can reduce inflammation and improve insulin sensitivity and glucose metabolism ²⁶ . Studies have found that specific drugs can improve insulin deficiencyand hepatorenal injury in high‐fat diet and streptozotocin‐induced type 2 diabetes models by regulating the cGAS/STING pathway ³⁰ . Additionally, chronic activation of cGAS/STING may promote an inflammatory state, exacerbate insulin deficiency and lipid accumulation, and form a feedback loop of metabolic dysfunction ³¹ . Therefore, targeting the cGAS/STING pathway may become a new strategy for treating metabolic disorders. In this study, PLAGL1‐overexpressing NOD mice exhibited significantly increased fasting and postprandial blood glucose levels, along with significantly decreased fasting and postprandial insulin levels, indicating exacerbation of the diabetic process. Furthermore, after intervention with RU.521 or H‐151, the above diabetic phenotypes were significantly improved. Moreover, intervention with RU.521 or H‐151 further improved the pathological changes in pancreatic islet tissues. In conclusion, overexpressed PLAGL1 exacerbates the phenotype of T1DM through the cGAS/STING signaling pathway. Inhibiting this pathway can effectively improve diabetic phenotypes, reduce inflammation, and protect pancreatic islet function.

Our study reveals that PLAGL1 overexpression induces both mitochondrial and nuclear DNA damage, but the relative contributions and interplay between these two types of damage warrant further discussion. While mitochondrial DNA damage appears to be the primary source of cytoplasmic DNA accumulation due to its proximity to ROS production and lack of histones, nuclear DNA damage may exacerbate this process by generating larger fragments that persist in the cytoplasm. The dual damage likely creates a feed‐forward loop: mitochondrial dysfunction increases ROS, which causes nuclear DNA damage, impairing repair mechanisms and further promoting genomic instability. Interestingly, our data suggest that mitochondrial damage might initiate the apoptotic cascade through cGAS/STING activation, while nuclear damage could amplify the response by activating additional DDR pathways. Future studies using targeted interventions to specifically protect mitochondrial or nuclear DNA could help delineate their individual contributions to β‐cell loss in T1DM.

Our study proposes a model wherein PLAGL1‐triggered oxidative stress serves as the upstream initiating event for DNA damage. While this model is strongly supported by the concerted indirect evidence—including the disturbance of antioxidant capacity (SOD, GSH/GSSG) and, most importantly, the functional rescue achieved with antioxidants (MitoQ, NAC)—we acknowledge that a limitation of this work is the lack of direct measurement of reactive oxygen species using fluorescent probes such as DCFH‐DA or MitoSOX Red. The inclusion of such direct evidence in the future will be crucial to irrefutably cement the foundational step of this pathway. Nevertheless, the convergence of our current findings provides a compelling rationale for this mechanistic sequence. Moreover, we acknowledge that the evidence positioning oxidative stress as the primary trigger is primarily based on the rescue effects observed with antioxidants MitoQ and NAC. While this provides strong pharmacological evidence, more direct genetic evidence, such as manipulating key antioxidant pathways (e.g., nuclear factor E2‐related factor 2 (NRF2)) in combination with PLAGL1 overexpression, would further strengthen this conclusion and represents an important direction for our future research. Besides, our study relies on an overexpression model to establish the toxic potential of PLAGL1 in β‐cells. While this approach provides a powerful tool for mechanistic dissection, it remains to be conclusively demonstrated whether endogenous PLAGL1 expression is upregulated during the spontaneous development of T1D in models such as the NOD mouse. Determining the temporal expression pattern of PLAGL1 in pre‐diabetic and diabetic NOD mice is a crucial future direction that will definitively establish the pathophysiological relevance of our findings.

CONCLUSION

In conclusion, this study delineates a multi‐layered mechanism through which PLAGL1 overexpression precipitates apoptosis in pancreatic β‐cells. We demonstrate that PLAGL1 instigates mitochondrial dysfunction and oxidative stress, which not only induces oxidative damage to mitochondrial DNA but also suppresses the expression of key DNA repair proteins. The consequent accumulation of cytoplasmic DNA serves as a critical signal that activates the cGAS/STING pathway, thereby initiating mitochondrial‐dependent apoptosis, as evidenced by BCL2/BAX dysregulation and Caspase‐3 activation. Genetic ablation of cGAS or STING was shown to significantly restore β‐cell viability and insulin secretory function. Importantly, these findings were corroborated in vivo, where pharmacological inhibition of the cGAS/STING pathway ameliorated hyperglycemia and preserved islet architecture in NOD mice. Collectively, our work elucidates a coherent pathogenic cascade linking PLAGL1‐driven oxidative stress, DNA damage accumulation, and innate immune activation that ultimately leads to β‐cell loss. These insights offer novel mechanistic foundations and highlight the cGAS/STING axis as a potential therapeutic target for mitigating β‐cell decline in diabetes.

AUTHOR CONTRIBUTIONS

Cheng Li, Lingyan Qiao, Juan Ge, and Sicui Hu performed the experiments, analyzed the data and drafted the manuscript. Hongxiu Yang, Conghui Hu, and Tang Li contributed to the study design and critical revision. Cheng Li supervised and administrated the projects. All authors reviewed and approved the final version of the manuscript.

DISCLOSURE

The authors declare that they have no conflicts of interest.

Approval of the research protocol: This research was approved by the institutional Animal Care and Use Committee (IACUC) of Shenglun Biotechnology Co., LTD (Suzhou, China).

Informed Consent: N/A.

Approval date of Registry and the Registration No. of the study/trial: 2024072224.

Animal Studies: Animal experiments were approved by the Animal Experimental Ethics Committee of Shenglun Biotechnology Co., Ltd (Suzhou, China).

DATA AVAILABILITY STATEMENT

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.