STING Promotes Renal Fibrosis in Diabetic Kidney Disease via ID1-Dependent Epithelial-Mesenchymal Transition
Article information
Abstract
Background
Tubulointerstitial fibrosis (TIF) due to epithelial-mesenchymal transition (EMT) is an inseparable feature of diabetic renal fibrosis. Although stimulator of interferon genes (STING) has been shown to play a potential role in regulating EMT, whether and how it modulates EMT in diabetic kidney disease (DKD) mice remain unclear. Here, we investigated the role and underlying mechanisms of STING-mediated EMT in TIF in DKD.
Methods
STING expression was detected in human renal biopsy tissues and serum samples from patients with DKD. Mouse models with genetic STING deletion or pharmacological by a STING inhibitor (C176) were established to further investigate the functions of STING in vivo. The in vitro roles of STING were analyzed in human renal tubular epithelial (HK2) cells with STING overexpression or knockdown. RNA sequencing was used to explore the underlying mechanisms.
Results
STING was upregulated in the kidneys and serum of patients with DKD and was negatively correlated with kidney function. STING deletion or pharmacological inhibition with C176 ameliorated pathological lesions, renal dysfunction and fibrosis in mouse models of DKD. STING deficiency alleviated renal fibrosis in DKD mice by inhibiting EMT. Mechanistically, RNA sequencing identified, inhibitor of differentiation 1 (ID1) as a downstream molecule of STING. Inhibition of ID1 in the setting of STING overexpression suppressed EMT and renal fibrosis.
Conclusion
Our study provides evidence that STING deficiency relieves renal fibrosis by inhibiting ID1-mediated EMT and that inhibition of STING and ID1 has therapeutic potential for patients with DKD.
Highlights
• STING expression is increased in the kidneys of individuals with DKD.
• STING acts as an important regulator of EMT in the kidneys of DKD mice.
• STING deficiency alleviates renal fibrosis by inhibiting ID1-mediated EMT.
INTRODUCTION
Diabetic kidney disease (DKD) is one of the most serious microvascular complications of type 2 diabetes mellitus (T2DM) and the leading cause of end-stage renal disease worldwide [1]. Among people worldwide who require renal replacement therapy, DKD accounts for a substantial proportion of cases, at 20% to 40%, and leads to more than 950,000 deaths globally every year [2,3]. Moreover, with the increasing prevalence of diabetes worldwide, this proportion is expected to steadily increase. Therefore, it is of great significance to investigate its pathogenic mechanisms. Studies have confirmed that multiple pathophysiological abnormalities contribute to the progression of DKD, such as inflammation, oxidative stress, autophagy, and tubulointerstitial fibrosis (TIF) [4]. Accumulating evidence has demonstrated that TIF substantially contributes to and is closely correlated with kidney dysfunction [5,6]. However, the molecular mechanisms responsible for TIF in DKD are not fully understood, and current strategies focused on controlling blood glucose and blood pressure are insufficient to prevent disease progression [7]. Therefore, novel antifibrotic approaches are urgently needed to retard the progression of TIF and kidney function decline in patients with DKD.
TIF is characterized by the accumulation of extracellular matrix (ECM), including collagen I (Col-I), collagen IV (Col-IV), and fibronectin (Fn) [8,9]. In response to chronic stimuli, tubular epithelial cells, one of the major cell types involved in TIF, undergo epithelial-mesenchymal transition (EMT) [10,11]. EMT of tubular epithelial cells is characterized by the loss of epithelial markers (e.g., E-cadherin and zonula occludens-1) and the acquisition of mesenchymal markers (e.g., vimentin, N-cadherin, Col-IV, and Fn) [12–14]. Therefore, EMT of tubular epithelial cells plays a crucial role in the pathological process of TIF in DKD, and interventions to slow or even reverse renal fibrosis in DKD by specifically targeting EMT of tubular epithelial cells urgently need to be developed. However, the mechanism underlying EMT of tubular epithelial cells remains largely unclear.
Stimulator of interferon genes (STING) is a transmembrane protein that predominantly resides in the endoplasmic reticulum membrane [15]. When pathological factors occur, STING undergoes extensive conformational changes, forming activated STING units that relay signals to downstream signaling molecules such as TANK-binding kinase 1 (TBK1) and interferon regulatory factor 3 (IRF3), to regulate immune responses and inflammation [15,16]. STING deletion skewed the pro-inflammatory state to an anti-inflammatory state in microglia by inhibiting autophagy, thereby alleviating ischemia-induced infarction and neuronal injury [17]. In addition to inflammatory diseases, recent studies have shown that STING plays an important role in a variety of diabetes-related complications. One study showed that STING promotes diabetic sarcopenia [18]. Our latest research revealed that STING is involved in diabetic aortic endothelial injury [19] and cardiomyocyte fibrosis in diabetic cardiomyopathy [20]. Interestingly, STING also plays a crucial role in DKD [21,22] and kidney fibrosis [23]. In radiation-induced lung damage, STING also promotes lung fibrosis by triggering EMT [24]. However, it remains unclear whether and how STING is involved in the initiation of kidney fibrosis in DKD by promoting EMT of tubular epithelial cells.
In this study, we found that STING expression was increased in both patients with DKD and DKD mice. Furthermore, STING deficiency suppressed renal tubular EMT in DKD-induced renal fibrosis in animal models and in human renal tubular epithelial (HK2) cells. Additionally, our results revealed that STING in tubular cells is required for renal fibrosis by promoting the expression of inhibitor of differentiation 1 (ID1), which inhibits E-cadherin expression and promotes vimentin expression, leading to EMT and fibrosis of tubular epithelial cells. Hence, STING may be considered a potential intervention target to prevent DKD progression.
METHODS
Human kidney biopsy and serum sample studies
Serum samples from patients with DKD were collected from hospitalized patients, and serum samples from normal controls and patients with T2DM were collected from physical examination centers. Patient characteristics are listed in Supplementary Table 1. DKD kidney samples were obtained from adult patients undergoing resection for kidney diseases. Non-DKD control kidney specimens were acquired from healthy kidney poles of individuals who underwent cancer nephrectomy. Patients provided written consent for their samples to be collected. The study protocol was approved by the Ethics Committee of the Affiliated Hospital of Southwest Medical University (clinical trial registration no. ChiCTR2100048381) and was conducted in accordance with the 1975 Declaration of Helsinki.
Mice
The STING knockout (STING−/−) mouse strain on a C57BL/6 background was generated using clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 technology (Cyagen Biosciences, Guangzhou, China). The guide RNA sequences were AGTATTCTCCTGGTACTTCGTGGG (gRNA1) and CTGAGGCCCCCAACCATTGAAGG (gRNA2). Mouse tail DNA was used for polymerase chain reaction (PCR) genotyping. Four- to 5-week-old db/m and db/db male mice were purchased from Tengxin Biotechnology Co., Ltd. (Chongqing, China). All mice were housed in a specific pathogen-free facility with a 12-hour light/dark cycle and standard fat diet (SFD) or high fat diet (HFD) feeding at a humidity of 50%±5% and a temperature of 20°C to 22°C. The experimental protocols for all mice were approved by the Animal Ethics Committee of Southwest Medical University (No. 20220225–014).
HFD/streptozotocin-induced T2DM mouse model
Male mice were fed 60% HFD (HFKbio, Beijing, China) for 12 weeks from 7 weeks of age. Eighteen-week-old HFD-fed mice were fasted for 12 hours and then injected intraperitoneally with streptozotocin (STZ) for 5 consecutive days (50 mg/kg/day). STZ powder was dissolved in 0.1 mol/L sodium citrate buffer and freshly prepared as a 10 mg/mL STZ solution. Age-matched male mice were injected with sodium citrate buffer as a control. The criterion for successful construction of the T2DM model was blood glucose ≥16.7 mmol/L for 3 consecutive days after the last STZ injection. Mice were sacrificed after another 12 weeks of HFD or SFD feeding, and the urinary microalbumin/creatinine was measured before sacrifice.
Pharmacological inhibition of STING in db/db mice
All db/db mice were continuously fed a 60% high fat diet (HFKbio) to stabilize the diabetic model. db/m and db/db mice were intraperitoneally injected with 750 nmol C176 (Selleck, Houston, TX, USA; HY-112906) per mouse daily in 200 μL corn oil (Selleck) for 8 weeks, and control mice were given 200 μL corn oil. Eight weeks later, the urinary microalbumin/creatinine ratio was measured before mice were euthanized.
Measurements of lipid and renal function
Urine was collected in sterile tubes, centrifuged to collect the supernatant, and stored at −80°C. Urinary creatinine (#C011-2-1) and urinary albumin (#H127-1-2) in the same samples were assessed using commercially available kits (Nanjing Jianjian Bioengineering Institute, Nanjing, China) according to the manufacturer’s instructions. Blood creatinine (#C011-2-1), blood total cholesterol (TC; #A111-1-1), and blood triglyceride (TG; #A110-1-1) were measured using commercially available kits (all products purchased from Nanjing Jianjian Bioengineering Institute) according to the manufacturer’s instructions.
Measurement of malondialdehyde
Malondialdehyde (MDA) levels in mouse kidneys were measured using commercial kits (Beyotime, Shanghai, China; Cat# S0131S). Tissue lysates were analyzed in a 96-well plate following the kit protocols. Absorbance was read at 532 nm using a microplate reader, with concentrations determined from standard curves.
Alanine aminotransferase/aspartate aminotransferase enzyme activity analysis
The activities of alanine aminotransferase (ALT; #C009-2-1) and aspartate aminotransferase (AST; #C010-2-1) were determined using commercial assay kits (Nanjing Jiancheng Bioengineering Institute) based on the Reitman-Frankel microplate method. Absorbance was measured at 505 nm (ALT) and 510 nm (AST), with enzyme activities calculated from standard curves and expressed as U/L.
Histological st2aining
Hematoxylin-eosin (HE), Masson, and Sirius red staining were used to observe kidney injury, fibrosis, and collagen deposition in mice. The stained sections were observed under a microscope and photographed. Staining intensity was quantified using ImageJ (National Institutes of Health, Bethesda, MD, USA).
Immunohistochemistry and immunofluorescence staining
Immunohistochemical staining of paraformaldehyde-fixed, paraffin-embedded 4-μm kidney sections was performed with antibodies against STING (1:100, Cell Signaling Technology, Danvers, MA, USA; #13647) or ID1 (1:50, Santa Cruz Biotechnology, Dallas, TX, USA; sc-133104), according to the manufacturer’s instructions (Absin, Shanghai, China; #ZM317D01). Paraffin-embedded mouse kidney sections were co-immunostained for E-cadherin (1:200, Cell Signaling Technology; #14472), viemntin (1:200, Cell Signaling Technology; #5741), and Col-IV (1:200, Abcam, Cambridge, UK; ab6586). Quantitative analysis was performed using ImageJ software.
Cell culture and intervention
HK2 cells were purchased from the National Model and Specialty Experimental Cell Resource Bank/Committee for Preservation of Typical Cultures of the Chinese Academy of Sciences cell bank and were cultured with 10% fetal bovine serum and Dulbecco’s Modified Eagle Medium (DMEM)/F12 in an incubator at 37°C, 5% CO2, and 90% humidity. The diabetes model was constructed by co-intervention of HK2 cells with high glucose (HG, 30 mmol/L) and palmitic acid (150 μmol/L).
Plasmid and siRNA intervention
In accordance with the kit instructions, HK2 cells were firstly passaged to an appropriate cell density. When the cell density reached approximately 50%, the cells were transfected with human-targeted STING or ID1 siRNA (RiboBio, Guangzhou, China) using transfection reagent. After HK2 cells grew to the appropriate density, they were transfected with transfection reagents targeting human-derived STING or ID1 plasmids (SynthBio, Beijing, China). All controls were simultaneously treated with isotonic solvent and control siRNA or empty plasmid.
Western blot
Cell lysate was prepared at a ratio of radioimmunoprecipitation assay (RIPA) lysis buffer to protease inhibitor mix of 100:1. An appropriate amount of freshly prepared lysate was added to cells or tissues. Protein blots were developed with an enhanced chemiluminescence system and imaged with an Odyssey FC imaging system (Li-COR Bioscience, Lincoln, NE, USA). Gray scale values were measured and statistically analyzed using Image J. Cytoplasmic proteins were normalized to tubulin or β-actin. The following primary antibodies were used: STING (1:1,000, Cell Signaling Technology; #13647), ID1 (1:200, Santa Cruz; sc-133104), E-cadherin (1:1,000, Cell Signaling Technology; #14472), vimentin (1:1,000, Cell Signaling Technology; #5741S), Smad2/3 (1:200, Santa Cruz; sc-133098), pSmad2/3 (1:200, Santa Cruz; sc-11769), and transforming growth factor β (TGFβ; 1:2,000, Proteintech, Rosemont, IL, USA; #81746).
Quantitative real-time polymerase chain reaction
Primers for quantitative real-time PCR were designed using the Primer BLAST tool from the National Center for Biotechnology Information. All primers were synthesized by BioWorks Bio (Uppsala, Sweden). PCR was performed using SYBR Green Master Mix (QIAGEN, Hilden, Germany) and the Applied Biosystems 7500 Real-Time PCR System (Waltham, MA, USA). Gene expression was normalized to the housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH), and the fold change in expression relative to the control was calculated using the 2−ΔΔCT method. Primer sequences are shown in Supplementary Table 2.
Statistical analysis
Continuous data are expressed as the mean±standard deviation. All statistical analyses were performed using GraphPad Prism version 8.0 software (La Jolla, CA, USA). Data were analyzed using the nonparametric Mann-Whitney test for comparisons between two groups. Statistical differences among multiple groups were analyzed using analysis of variance (ANOVA) with Tukey’s post hoc analysis, with P<0.05 considered significant.
RESULTS
STING is upregulated in kidneys from patients or animals with DKD and in high glucose and palmitic acid-cultured HK2 cells
To investigate whether STING is involved in the pathogenesis of DKD, we first searched available databases and found that STING was significantly upregulated in the kidneys of patients with DKD (Fig. 1A). We then collected renal specimens from patients with or without DKD, and immunohistochemical staining showed increased STING expression, mainly localized in renal tubular cells (Fig. 1B). We also determined STING expression in HFD/STZ-induced and db/db mice. The same changes of STING were observed in the renal tubular cells of DKD mice (Fig. 1C–E). The mRNA level of STING was also elevated in the kidneys of DKD mice (Fig. 1F). Because HG intervention alone had no effect on STING in HK2 cells (Supplementary Fig. 1), and in addition to hyperglycemia, dyslipidemia is also a major driver DKD development [25], we next established a glucolipotoxicity cell model in HK2 cells treated with high glucose and palmitic acid (HGPA). Consistent with the in vivo results, the protein level of STING was significantly increased in HK2 cells treated with HGPA (Fig. 1G and H).
Stimulator of interferon genes (STING) is upregulated in kidneys from patients or animals with diabetic kidney disease and in high glucose and palmitic acid (HGPA)-cultured human renal tubular epithelial (HK2) cells. The mRNA levels of STING in kidneys from healthy controls and patients diagnosed with type 2 diabetic nephropathy (T2DN) based on the GSE166239 dataset (A). Representative images of immunohistochemical staining of STING in kidneys from patients with T2DN (B), diabetic kidney disease mice (C), and db/db mice (D). Semiquantitative analysis of STING in panels C and D (E). Quantitative real-time polymerase chain reaction analysis of STING in high fat diet (HFD)/streptozotocin (STZ) mice and db/db mice (F). Western blot analysis of STING protein levels in HK2 cells treated with HG (30 mM) and palmitic acid (PA, 150 μM) (G, H). Serum STING levels in healthy control, type 2 diabetes mellitus (T2DM), and diabetic kidney disease (DKD) patient groups (I). Pearson correlation analysis between serum STING and urea (J), serum creatinine (K), urinary microalbumin/urinary creatinine ratio (UACR) (L), and estimated glomerular filtration rate (eGFR) (M). CPM, counts per million; NC, normal control. aP<0.05, bP<0.01, cP<0.001, and dP<0.0001 between indicated groups by analysis of variance (ANOVA) with Tukey’s post hoc analysis.
To further explore the correlation between serum STING levels and renal function indexes in DKD, 45 serum samples were collected (Supplementary Table 1). Serum STING was markedly elevated in patients with T2DM and DKD (Fig. 1I). Pearson correlation analysis suggested that serum STING concentration was positively correlated with urea, serum creatinine, and urinary microalbumin/urinary creatinine ratio (UACR), and negatively correlated with estimated glomerular filtration rate (eGFR) (Fig. 1J–M). These results demonstrated that STING may participate in the pathogenesis of DKD.
Genetic deletion of STING ameliorates the pathological manifestations, inflammation, oxidative stress and renal fibrosis of DKD mice
To explore the role of endogenous STING in the development of renal fibrosis in DKD mice, STING knockout (STING−/−) mice were constructed (Supplementary Fig. 2A). STING was absent in genetically deleted mice compared with wild-type (WT) mice (Supplementary Fig. 2B–E), indicating successful STING deletion. The DKD mouse model was induced using HFD/STZ in both WT and STING−/− mice (Fig. 2A). As shown, there were no differences in body weight or blood glucose between HFD/STZ WT mice and STING−/− mice (Supplementary Fig. 3). STING deletion ameliorated kidney dysfunction in HFD/STZ-induced DKD mice compared with WT mice, as evidenced by reduced UACR and blood creatinine levels (Fig. 2B, left two panels). In addition, both blood TC and TG levels were significantly lower in STING−/− mice (Fig. 2B, right two panels).
Genetic deletion of stimulator of interferon genes (STING) attenuates pathological manifestations, inflammation, oxidative stress and renal fibrosis in high fat diet (HFD)/streptozotocin (STZ)-induced diabetic kidney disease (DKD) mice. Flowchart of establishing type 2 diabetes mellitus (T2DM) model in wild-type (WT) and STING knockout mice (A). Urinary microalbumin/urinary creatinine ratio (UACR), blood creatinine levels, serum total cholesterol (TC), and triglycerides (TG) in each group of mice (B). Representative images of hematoxylin-eosin (HE), Masson, and Sirius red staining at 200× magnification (scale bars: 100 μm) (C). Pathological scoring of renal tubular and glomerular injury was performed on randomly selected fields from each sample (n=3 mice per group) (D). Quantitative analysis was performed on Masson’s trichrome and Sirius red staining to assess areas of fibrosis and collagen deposition (E). mRNA levels of transforming growth factor β (TGFβ), interleukin 6 (IL-6), and tumor necrosis factor-α (TNFα) and tissue content of malondialdehyde (MDA) in each group (F). SFD, standard fat diet. aP<0.05, bP<0.01, cP<0.001, and dP<0.0001 between indicated groups by analysis of variance (ANOVA) with Tukey’s post hoc analysis.
As shown in Fig. 2C and D, HE staining demonstrated that, compared with HFD/STZ WT mice, renal pathological damage was relieved in HFD/STZ STING−/− mice. Decreased collagen and ECM accumulation were observed in both the glomeruli and renal interstitium of STING deficient mice, as shown by Masson and Sirius red staining (Fig. 2C and E). To explore the effects of STING on inflammation and oxidative stress, inflammation-related genes and MDA, an oxidative stress marker, were assessed in each group of mice. Compared with HFD/STZ WT mice, the expression of inflammation-related genes (Fig. 2F, first three panels) and MDA levels (Fig. 2F, last panel) were significantly down-regulated in HFD/STZ STING−/− mice. These findings suggest that STING is sufficient to drive the progression of renal tubular fibrosis in DKD.
Pharmacological inhibition of STING with C176 alleviates pathological manifestations, inflammation, oxidative stress and renal fibrosis in db/db mice
To explore the therapeutic effect of STING inhibition on the progression of renal fibrosis, db/db and db/m mice were intraperitoneally injected with the selective covalent STING inhibitor (C176) once daily for 8 or 12 weeks (Fig. 3A). As shown in Supplementary Fig. 4, long-term treatment with C176 showed no off-target effects or liver damage. C176 treatment remarkably improved kidney function, accompanied by lower UACR and blood creatinine levels in db/db mice (Fig. 3B). Consistent with this observation, HE, Masson, and Sirius red staining showed that kidneys had severe histopathological changes in db/db mice, and C176 treatment significantly attenuated renal pathological damage (Fig. 3C and D) and interstitial fibrosis (Fig. 3C and E). In addition, STING inhibition with C176 reduced inflammation (Fig. 3F, first three panels) and oxidative stress (Fig. 3F, last panel). These findings further confirmed the importance of STING in renal fibrosis and showed that pharmacological intervention with C176 had a beneficial effect on DKD-induced renal fibrosis.
Pharmacological inhibition of stimulator of interferon genes (STING) with C176 alleviates pathological manifestations, inflammation, oxidative stress and renal fibrosis in db/db mice. Flow chart of pharmacological inhibition of STING with C176 in db/db mice (A). Urinary microalbumin/urinary creatinine ratio (UACR) and blood creatinine levels in each group of mice (B). Representative images of hematoxylin-eosin (HE), Masson, and Sirius red staining at 200× magnification (scale bars: 100 μm) (C). Pathological scoring of renal tubular and glomerular injury was performed on randomly selected fields from each sample (n=3 mice per group) (D). Quantitative analysis was performed on Masson’s trichrome and Sirius red staining to assess areas of fibrosis and collagen deposition (E). mRNA levels of transforming growth factor β (TGFβ), interleukin 6 (IL-6), and tumor necrosis factor-α (TNFα) and tissue content of malondialdehyde (MDA) in each group (F). aP<0.05, bP<0.01, cP<0.001, and dP<0.0001 between indicated groups by analysis of variance (ANOVA) with Tukey’s post hoc analysis or nonparametric Mann-Whitney test.
STING deficiency attenuates renal fibrosis by inhibiting EMT in tubular epithelial cells
EMT of tubular epithelial cells plays a critical role in the pathogenesis of renal fibrosis caused by DKD [26]. We investigated the hypothesis that STING deficiency attenuates the fibrosis process in DKD by inhibiting the EMT in tubular epithelial cells. Therefore, the levels of EMT-related factors were detected in DKD and db/db mice. Immunofluorescence results showed that STING deficiency or inhibition induced upregulation of the epithelial cell marker (E-cadherin) and downregulation of mesenchymal markers (vimentin and vimentin/E-cadherin ratio) (Fig. 4A and B, Supplementary Fig. 5). In addition, a significant reduction in Col-IV was also observed in DKD mice with STING deficiency (Fig. 4A and B, Supplementary Fig. 5). The same trends in the protein levels of E-cadherin and vimentin and in the vimentin/E-cadherin ratio were also observed in DKD mice with STING deletion (Fig. 4C).
Inhibition of stimulator of interferon genes (STING) ameliorates renal tubulointerstitial fibrosis by inhibiting epithelial-mesenchymal transition (EMT). Representative images of immunofluorescence (IF) staining of EMT-associated factors in kidney tissues of high fat diet (HFD)/streptozotocin (STZ)-induced mice. E-cadherin is an epithelial cell marker, and vimentin and collagen IV are mesenchymal cell markers. Scale bars: 50 μm, n=3 mice per group (A). Semiquantitative analysis of vimentin and collagen IV intensity per E-cadherin-positive area (B). Western blot analysis of E-cadherin, vimentin, and STING (left) and densitometric quantification of the vimentin/E-cadherin ratio and STING in kidneys of each group of mice (right) (C). Western blot analysis of STING and E-cadherin in control and high glucose and palmitic acid (HGPA)-induced human renal tubular epithelial (HK2) cells with STING deficiency (D). SFD, standard fat diet; WT, wild-type; DAPI, 4′,6-diamidino-2-phenylindole. aP<0.05, bP<0.01, cP<0.001, and dP<0.0001 between indicated groups by analysis of variance (ANOVA) with Tukey’s post hoc analysis.
To further demonstrate that STING mediates TIF by promoting EMT of tubular epithelial cells, siRNA was used to decrease STING expression in HGPA-cultured HK2 cells. The constructed STING siRNA effectively downregulated STING expression in HK2 cells (Supplementary Fig. 6A and C). Similarly, upregulated E-cadherin expression was observed in the STING knockdown group (Fig. 4D). Collectively, these results indicate that STING deficiency alleviates renal fibrosis by inhibiting EMT in tubular epithelial cells.
STING promotes EMT via upregulation of ID1
To identify potential downstream molecules regulated by STING, kidney tissues from HFD/STZ WT and HFD/STZ STING−/− mice were subjected to RNA sequencing. A total of 340 differentially expressed genes were identified. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis showed that these genes were involved in the TGFβ signaling pathway (Fig. 5A). TGFβ signaling is a major driver of fibrosis in multiple organs. Phosphorylation of Smad2/3 was upregulated in kidney tissues of DKD mice and downregulated after STING deletion, indicating that the TGFβ signaling pathway is related to STING (Fig. 5B). Then, we validated several genes in the TGFβ signaling pathway and found that the protein and mRNA levels of ID1 were upregulated in renal tissues of DKD mice and downregulated after STING deletion (Fig. 5B, Supplementary Fig. 7). We further examined the expression of ID1 by immunohistochemistry, and the results also showed that STING knockout or inhibition downregulated ID1 expression (Fig. 5C and D). Moreover, immunofluorescence results showed that STING and ID1 were co-localized and expressed in renal tubular epithelial cells (Fig. 5E, Supplementary Fig. 8). Collectively, these results suggest a possible regulatory relationship between STING and ID1.
RNA sequencing results reveal a correlation between inhibitor of differentiation 1 (ID1) and stimulator of interferon genes (STING). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of RNA sequencing data in high fat diet (HFD)/streptozotocin (STZ) wild-type (WT) mice and HFD/STZ-induced STING knockout mice (A). Western blot analysis of transforming growth factor β (TGFβ), pSmad2/3, Smad2/3, and ID1 in kidneys of control and HFD/STZ-induced mice (B). Representative images of immunohistochemical staining of ID1 in HFD/STZ-induced mice (C) and db/db mice (D); semiquantitative analysis of the ID1-positive area is shown on the right. Scale bars: 100 μm, n=3 mice per group. Representative images of immunofluorescence staining of STING and ID1 in kidney tissues of HFD/STZ-induced mice; semiquantitative analysis of the STING- and ID1-positive areas is shown on the right. Scale bars: 25 μm (E). GnRH, gonadotropin-releasing hormone; SFD, standard fat diet; DAPI, 4′,6-diamidino-2-phenylindole; IOD, integrated optical density. aP<0.05, bP<0.01, cP<0.001, and dP<0.0001 between indicated groups by analysis of variance (ANOVA) with Tukey’s post hoc analysis.
Studies have shown that ID1 is a positive regulator of EMT [27]. To determine whether STING is involved in renal fibrosis via ID1-mediated EMT in tubular epithelial cells, ID1 was knocked down or overexpressed in the setting of STING overexpression or knockdown under HGPA treatment in HK2 cells. The overexpression and knockdown efficiencies of STING and ID1 were examined (Supplementary Fig. 6). As shown in Fig. 6A, co-treatment with STING further reduced the expression of E-cadherin inhibited by HGPA, while further increasing the expression of vimentin and the vimentin/E-cadherin ratio induced by HGPA in HK2 cells. However, the detrimental effects of STING on renal tubules were reversed after ID1 protein reduction. More importantly, STING knockdown promoted HGPA-inhibited E-cadherin re-expression and decreased the expression of HGPA-induced vimentin and the vimentin/E-cadherin ratio, but this protective effect was disrupted after ID1 overexpression (Fig. 6B). Furthermore, the immunofluorescence results revealed identical alterations (Fig. 6C). These findings indicate that STING could promote EMT through upregulation of ID1.
Stimulator of interferon genes (STING) promotes epithelial-mesenchymal transition (EMT) via upregulation of inhibitor of differentiation 1 (ID1). Western blot analysis was performed to detect the expression of E-cadherin, vimentin, ID1, and STING in control and high glucose and palmitic acid (HGPA)-induced human renal tubular epithelial (HK2) cells with STING overexpression, with or without ID1 knockdown. Densitometric quantification of the vimentin/E-cadherin ratio, ID1, and STING is shown in the figure (A). Western blot analysis was used to assess E-cadherin, vimentin, ID1, and STING expression in control and HGPA-treated HK2 cells with STING deficiency, with or without ID1 overexpression. Densitometric data for the vimentin/E-cadherin ratio, ID1 and STING are presented in the figure (B). Representative images of immunofluorescence staining of EMT-associated factor E-cadherin and vimentin in control, HGPA-induced HK2 cells with STING deficiency/overexpression or ID1 deficiency/overexpression; semiquantitative analysis of the vimentin/E-cadherin ratio-positive area is shown on the right. Scale bars: 10 μm (C). DAPI, 4′,6-diamidino-2-phenylindole. aP<0.05, bP<0.01, cP<0.001, and dP<0.0001 between indicated groups by analysis of variance (ANOVA) with Tukey’s post hoc analysis.
DISCUSSION
This study identified an important role for STING in DKD-induced kidney dysfunction and renal fibrosis based on the results of genetic, in vivo, in vitro, and pharmacological experiments. The main findings of the present research include the following: (1) STING expression was significantly increased in kidney specimens from both patients with DKD and DKD mice, and increased serum STING concentrations in ptients with DKD were positively correlated with urea, serum creatinine, and UACR, and negatively correlated with eGFR; (2) STING was characterized as a promoter of renal fibrosis and STING deletion or inhibition ameliorated DKD-induced renal fibrosis by suppressing EMT; and (3) mechanistically, STING was identified as an important regulator of ID1, which subsequently promoted renal fibrosis via dysregulation of EMT-associated proteins in DKD. Collectively, these findings provide evidence that STING is a potential target for the treatment of renal fibrosis in DKD.
Previous studies have established the critical involvement of STING in DKD and renal fibrosis [21,22]. For instance, Zang et al. [22] focused on podocytes in DKD mice and explained the damaging effects of STING on podocytes through the classical mtDNA-cyclic GMP-AMP synthase (cGAS)-STING-TBK1-P65-inflammatory cytokine pathway, which worsens renal function and promotes DKD progression. Parallel research identified STING-mediated ferroptosis as a pathogenic mechanism in DKD, demonstrating that STING deficiency protected against diabetic renal injury by alleviating ferroptosis through stabilization of ferroportin 1 in DKD mice and mouse renal tubular epithelial cells, providing a potential therapeutic target for DKD [21]. In alignment with previous findings, our data corroborate that inhibition of STING ameliorates autophagy, inflammation, oxidative stress, and renal dysfunction, mitigates histological kidney injury, and reduces renal fibrosis. Notably, we reveal for the first time that this renal protective effect is linked to the suppression of EMT in both DKD mice and HGPA-stimulated HK2 cells, uncovering a previously unrecognized role of STING in DKD pathogenesis. However, the mechanism by which STING regulates EMT remains unclear.
Therefore, we subjected kidneys from STING knockout and WT DKD mice to RNA sequencing to explore the possible mechanism by which STING regulates EMT. Through KEGG analysis of differentially expressed genes, we found that the TGFβ signaling pathway was involved. TGFβ promotes fibrosis by activating Smad-driven EMT [28]. In our study, TGFβ and p-Smad2/3 expression were obviously increased in diabetic renal tissues, while STING deletion significantly decreased TGFβ and p-Smad2/3 expression in diabetic renal tissues. We then continued to verify genes involved in the TGFβ signaling pathway and focused on ID1. Immunohistochemical staining showed that STING and ID1 were mainly highly expressed in the renal tubules of diabetic mice, suggesting that both are important for their effects on renal tubular function. Previous studies have found that ID1 is involved in the development of chronic kidney disease by inducing EMT and inflammation, and immunohistochemical staining revealed that ID1 is mainly expressed in renal tubules [29,30]. In vivo, genetic ablation of the ID1 gene reduces peritubular inflammation in renal tubules after ureteral obstruction [30]. We found that STING and ID1 were upregulated in DKD and that ID1 expression varied with STING expression, suggesting that ID1 is a downstream molecule of STING. Moreover, STING deficiency partly inhibited the effects of ID1 overexpression on EMT, while inhibition of ID1 partially limited the activating effects of STING overexpression on EMT. These results confirm that STING regulates EMT by affecting ID1 expression. Our study elucidates a new mechanism of STING regulation of EMT through the STING-ID1 axis and enriches the regulatory network of STING in DKD. In addition, a previous study has shown that Smad3 mediates TGFβ-dependent transcriptional activation of ID1 [31]. Therefore, we propose that STING activates the TGFβ-Smad3 axis, which promotes the transcriptional expression of ID1, thereby inducing EMT in DKD mice.
In conclusion, we investigated the mechanism of STING involvement in DKD and demonstrated that STING activates EMT phenotypic transformation in DKD renal tubular epithelial cells through the ID1 signaling pathway. These observations also suggest that amelioration of ID1-induced EMT may be an effective therapeutic modality to alleviate fibrosis in DKD. However, the signaling networks involved in fibrosis are intricate and many unknown aspects still need to be addressed in future studies. Importantly, given that previous studies have explored the role of STING in DKD with respect to podocytes, a diabetes model using renal tubular epithelial cell-specific STING knockout mice should be constructed to investigate the intrinsic and specific effects of STING inhibition on renal tubular epithelial cells in DKD. In addition, other DKD models such as Akita or overexpression of calmodulin in pancreatic β-cells (OVE26) mice, should be used to improve generalizability. These are the limitations of the current study. Although more studies are needed, our preclinical studies suggest that STING and ID1 may be promising therapeutic targets for the treatment of DKD.
SUPPLEMENTARY MATERIALS
Supplementary materials related to this article can be found online at https://doi.org/10.4093/dmj.2024.0645.
Clinical characteristics
Quantitative real-time polymerase chain reaction primers
High glucose intervention alone has no effect on stimulator of interferon genes (STING) expression. (A) Western blot analysis of STING protein levels in human renal tubular epithelial (HK2) cells treated with high glucose (HG) (40 mM). (B) The densitometry data of STING. NC, normol control; NS, not significant.
Identification of stimulator of interferon genes (STING) knockout mice. (A) DNA sequencing results of STING knockout positive mice. (B) Agarose gel electrophoresis of STING knockout mice. (C, D) Western blot analysis of STING expression levels in kidneys of wild-type (WT) and STING−/− mice. (E) Quantitative real-time polymerase chain reaction showing STING mRNA levels in kidneys of WT and STING−/− mice. SFD, standard fat diet. aP<0.001 and bP<0.0001 between indicated groups by nonparametric Mann-Whitney test.
Characterization of an experimental mouse model of high fat diet (HFD)/streptozotocin (STZ) induced diabetes. Random weight (A) and blood glucose (B) measured once of each 2 weeks in indicated groups of mice, starting before STZ injection and continuing until 12 weeks after injection. SFD, standard fat diet; WT, wild-type; STING, stimulator of interferon genes.
Treatment with C176 showed no off-target and liver damage. (A) Western blot analysis of stimulator of interferon genes (STING) expression levels in kidneys of db/db and db/db+C176 mice. (B, C) The activities of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in each group of mice. SFD, standard fat diet; WT, wild-type; NS, not significant. aP<0.05, bP<0.01, and cP<0.001 between indicated groups by analysis of variance (ANOVA) with Tukey’s post hoc analysis or nonparametric Mann-Whitney test.
Representative images of immunofluorescence staining of epithelial-mesenchymal transition-associated markers in kidney tissues of db/db mice. (A) E-cadherin as an epithelial cell marker, vimentin and collagen IV as mesenchymal cell markers. (B) Semiquantitative analysis of the vimentin and collagen IV intensity per E-cadherin positive area. Scale bars: 50 μm, n=3 mice per group. DAPI, 4ʹ,6-diamidino-2-phenylindole. aP<0.001 and bP<0.0001 between indicated groups by analysis of variance (ANOVA) with Tukey’s post hoc analysis.
Efficiency test of stimulator of interferon genes (STING) and inhibitor of differentiation 1 (ID1) gene knockdown and overexpression. Representative Western blots showing the protein level of STING and ID1 in human renal tubular epithelial (HK2) cells transfected with vector, siSTING (A, C) and siID1 (E, G). Representative western blots showing the protein level of STING and ID1 in HK2 cells transfected with vector, flag-STING (B, D) and oeID1 (F, H). si, small interfering; oe, overexpression. aP<0.05 and bP<0.001 between indicated groups by nonparametric Mann-Whitney test.
The genes associated with the transforming growth factor β signaling pathway analyzed by Kyoto Encyclopedia of Genes and Genomes (KEGG). Quantitative real-time polymerase chain reaction analysis of inhibitor of differentiation 1 (ID1) (A), inhibin subunit beta B (INHBB) (B), insulin like growth factor binding protein 3 (IGFBP3) (C), and ID3 (D) in kidney of control and high fat diet/streptozotocin-induced mice (C). NS, not significant. aP<0.05 and bP<0.01 between indicated groups by analysis of variance (ANOVA) with Tukey’s post hoc analysis.
Stimulator of interferon genes (STING) co-locates and associates with inhibitor of differentiation 1 (ID1) in db/db mice. Representative images of immunofluorescence staining of STING and ID1 in kidney tissues of db/db mice. Scale bars: 25 μm, n=3 mice per group. DAPI, 4ʹ,6-diamidino-2-phenylindole.
Notes
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
AUTHOR CONTRIBUTIONS
Conception or design: H.W., X.M., Y.H., X.F., Y.X.
Acquisition, analysis, or interpretation of data: all authors.
Drafting the work or revising: H.W., X.M., Y.H., X.F., Y.X.
Final approval of the manuscript: all authors.
FUNDING
This study was supported by funds from the National Natural Science Foundation of China (U22A20286 and 82401001), Sichuan Science and Technology Program (2024NSFSC1616 and 2022YFS0617), Luzhou Science and Technology Projects (2022-JYJ-162), the Scientific research project of Southwest Medical University (2024LCYXZX02) and Open Project Program of Metabolic Vascular Diseases Key Laboratory of Sichuan Province (2023MVDKL-G4).
ACKNOWLEDGMENTS
None