SMURF1 activates the cGAS/STING/IFN-1 signal axis by mediating YY1 ubiquitination to accelerate the progression of lupus nephritis

Abstract

Aggravated endoplasmic reticulum stress (ERS) and apoptosis in podocytes play an important role in lupus nephritis (LN) progression, but its mechanism is still unclear. Herein, the role of SMURF1 in regulating podocytes apoptosis and ERS during LN progression were investigated. MRL/lpr mice was used as LN model in vivo. HE staining was performed to analyze histopathological changes. Mouse podocytes (MPC5 cells) were treated with serum IgG from LN patients (LN-IgG) to construct LN model in vitro. CCK8 assay was adopted to determine the viability. Cell apoptosis was measured using flow cytometry and TUNEL staining. The interactions between SMURF1, YY1 and cGAS were analyzed using ChIP and/or dual-luciferase reporter gene and/or Co-IP assays. YY1 ubiquitination was analyzed by ubiquitination analysis. Our results found that SMURF1, cGAS and STING mRNA levels were markedly increased in serum samples of LN patients, while YY1 was downregulated. YY1 upregulation reduced LN-IgG-induced ERS and apoptosis in podocytes. Moreover, SMURF1 upregulation reduced YY1 protein stability and expression by ubiquitinating YY1 in podocytes. Rescue studies revealed that YY1 knockdown abrogated the inhibition of SMURF1 downregulation on LN-IgG-induced ERS and apoptosis in podocytes. It was also turned out that YY1 alleviated podocytes injury in LN by transcriptional inhibition cGAS/STING/IFN-1 signal axis. Finally, SMURF1 knockdown inhibited LN progression in vivo. In short, SMURF1 upregulation activated the cGAS/STING/IFN-1 signal axis by regulating YY1 ubiquitination to facilitate apoptosis in podocytes during LN progression.

Keywords:

• Lupus nephritis
• endoplasmic reticulum stress
• SMURF1
• YY1
• cGAS/STING/IFN-1 signal axis

1. Introduction

Systemic lupus erythematosus (SLE) is a connective tissue disease invading multiple systems [1], which often occurs in young women [2]. SLE often injures the kidney, resulting in lupus nephritis (LN) [3,4]. LN is a major risk factor for overall incidence rate and mortality of SLE [2]. Although anti-inflammatory and immunosuppressive therapies represented by glucocorticoids and immunosuppressants have certain therapeutic effects on LN, there are still too many LN patients who develop into end-stage renal disease (ESRD) [2]. It’s suggested that understanding LN pathological mechanism is of great help to develop new therapeutic strategies for LN.

Smad ubiquitination regulatory factor 1 (SMURF1) is the key enzyme that recognizes the ubiquitinated protein substrate and selectively regulate the ubiquitinated degradation process of effector molecules such as Smads [5,6]. SMURF1 is involved in multiple kidney diseases. As proof, SMURF1 could aggravate renal fibrosis in diabetes nephropathy and obstructive nephropathy by promoting glomerular mesangial cell apoptosis [7,8]. Additionally, SMURF1 was markedly upregulated in the process of kidney injury, and its inhibition could prevent kidney injury to chronic kidney disease transition [9]. Nevertheless, the role of SMURF1 in LN remains unclear, which deserves further research. Yin Yang 1 (YY1), as a nuclear transcription factor with dual transcriptional activity [10], is a key player in kidney diseases. Gao et al. demonstrated that YY1 expression was markedly reduced in mesangial cells of advanced diabetes nephropathy, and YY1 knockdown aggravated glomerulosclerosis [11]. Notably, YY1 was downregulated in kidney tissues of LN mice [12]. Herein, we observed that SMURF1 might suppress YY1 expression by regulating YY1 ubiquitination by using bioinformatics software analysis. However, the interaction between SMURF1 and YY1 in LN is largely unknown.

Cyclic GMP-AMP synthase (cGAS), as the main cytosolic DNA sensor in human immortal foot cells, is implicated in SLE [13]. As revealed by An et al. cGAS was markedly upregulated in SLE clinical samples [14]. In LN, the extracellular DNA accumulation of necrotic cells activates the cGAS signal pathway and causes the subsequent interferon (IFN)-I secretion, thus promoting the inflammatory response [15]. Mechanically, the activation of cGAS stimulates the adapter protein stimulator of interferon genes (STING) to trigger IFN signal transduction [16]. Notably, it was previously described that cGAS-STING pathway activation exacerbated LN by facilitating endoplasmic reticulum stress (ERS) and inflammation [15]. In the current research, by using JASPAR (https://jaspar.genereg.net/) prediction, it was observed that YY1 potentially bound with cGAS promoter, suggesting that YY1 might regulate LN development by acting on the cGAS/STING/IFN1 signal axis.

Based on the above evidence, it’s speculated that SMURF1 reduced the expression of YY1 by regulating YY1 ubiquitination, thus activating the cGAS/STING/IFN-1 signal axis, and ultimately promoting ERS and inflammation to accelerate LN development. Our research provides a theoretical basis for developing new therapeutic strategies for LN.

2. Materials and methods

2.1. Clinical sample collection

Serum samples were collected from 22 diagnosed LN patients hospitalized. Serum samples from 22 healthy people were collected as control samples. All LN patients were confirmed by renal biopsy and did not have other kidney diseases. Disease activity was assessed using the Mexican version of Systemic Lupus Erythematosus Disease Activity Index (Mex-SLEDAI) [17] and the Systemic Lupus International Collaborating Clinics (SLICC) [18]. According to the Mex-SLEDAI score, the disease activity was stratified as follows: inactive (0–2), mild-moderate (3–5), and severe (≥6) activity. Table S1 reports demographic, disease activity clinical and laboratory features of enrolled patients. The study was authorized by the Ethics Committee of Hainan women and Children’s medical center before enrollment of patients. All participants signed informed consent.

2.2. Purification of serum IgG

Serum IgG was separated and purified using protein affinity chromatography column elution, and the concentration was measured using the human IgG concentration detection kit (Elabscience, TX, USA).

2.3. Cell culture and treatment

Mouse podocyte cells (MPC5 cells), obtained from ATCC (VA, USA), were cultured in RPMI-1640 (Gibco, MD, USA) containing 10% FBS (Gibco) at 37 °C with 5% CO₂. To establish a LN cell model, MPC5 cells were induced with LN serum IgG (1000 μg/mL) for 6 h.

2.4. Cell transfection

YY1 overexpressing lentivirus (OE-YY1), SMURF1 overexpressing lentivirus (OE-SMURF1), YY1 knockdowning lentivirus (sh-YY1), SMURF1 knockdowning lentivirus (sh-SMURF1), cGAS knockdowning lentivirus (sh-cGAS) and their negative controls were obtained from GenePharma (Shanghai, China). Furthermore, in order to construct the overexpressing plasmid of ATF4, TCF4 and KLF5, the subclonal sequences of ATF4, TCF4 and KLF5 were inserted into the pcDNA3.1 ((Invitrogen, US). MPC5 cells were cultured in 12-well plates at 50% confluency for 24 h. Cells were subsequently cultured in FBS-free DMEM/F-12 containing lentivirus (2 × 10⁸ transducing units/mL, 10 µL) for 6 h.

2.5. Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was extracted with TRIzol (ThermoFisher Scientific, MA, USA). The cDNA was synthesized using the Reverse Transcription Kit (Toyobo, Tokyo, Japan) and subjected to qRT-PCR assay with SYBR (ThermoFisher Scientific). GAPDH was used as the reference gene. The data was analyzed using 2^−ΔΔCT method. The primer sequences were listed in Table 1.

Table 1. Primers used in the study.

	Forward primer (5′-3′)	Reverse primer (5′-3′)
SMURF1	GACATAGGTTCCTGCTGGACAC	CGCCCATCAAGGGCCCCAAG
cGAS	AGGAAGCCCTGCTGTAACACTTCT	AGCCAGCCTTGAATAGGTAGTCCT
STING	GGCGTCTGTATCCTGGAGT	TAGACAATGAGGCGGCAGTTAT
YY1	ATGAGAAAGCATCTGCACACC	AGCCTTCGAATGTGCACTGAAA
NEDD4L	GTCCGGCTGTTCCGTACTC	AGGCCATAGTAGGGGTTAAACAT
BMI1	AGCAATGACTGTGATGCACTTGAG	CTCCAGCATTCGTCAGTCCATCCC
GAPDH	GGAGCGAGATCCCTCCAAAAT	GGCTGTTGTCATACTTCTCATGG

2.6. Western blot

The proteins were isolated with RIPA. Equal amounts of protein samples were separated by SDS/PAGE and further transferred to a PVDF membrane (Millipore, MA, USA). Then, membranes were incubated overnight with antibodies against YY1 (ab109237), Bax (ab32503), B-cell lymphoma-2 (Bcl-2) (ab182858), glucose-regulated protein 78 (GRP78) (ab21685), C/EBP-homologous protein (CHOP) (ab11419), activating transcription factor 6 (ATF6) (ab227830), PKR-like ER kinase (PERK) (ab229912), SMURF1 (ab300408), cGAS (ab224144), STING (ab239074), IFN-1 (ab171081) and GAPDH (ab8245), then hybridized with secondary antibody (Abcam, 1:10000, ab7090) for 60 min. Blots were visualized by an ECL detection kit (Beyotime). All antibodies were purchased from Abcam and were tested in appropriate dilutions according to the instructions.

2.7. Cell apoptosis assay

MPC5 cells were re-suspended in the binding buffer (500 μL, Beyotime, Shanghai, China), stained with Annexin V-FITC (10 μL) and PI (5 μL) for 10 min and subsequently analyzed by flow cytometry (BD, NJ, USA).

2.8. Cell counting kit-8 (CCK-8) assay

MPC5 cells were cultured in 24-well plates (2 × 10⁴ cells/well) for 24 h and incubated with CCK-8 solution (10 μL, Sangon) at 37 °C for 3 h. Absorbance at 450 nm was subsequently analyzed.

2.9. Chromatin immunoprecipitation (ChIP) assay

Cells were fixed and quenched. DNA was fragmented by sonication. Then cell lysate was incubated with anti-YY1 (Abcam, 1:50, ab109237) or anti-IgG (Abcam, 1:100, ab211493) at 4 °C overnight. DNA was enriched and subjected to qPCR analysis.

2.10. Dual-luciferase reporter gene assay

cGAS promoter fragments containing YY1/ATF4/TCF4/KLF5 WT/MUT binding site were amplified by PCR and inserted into the pGL3 reporter plasmid (GenePharma). Cells were co-transfected with cGAS-WT or cGAS-MUT and vector or sh-YY1/oe-ATF4/oe-TCF4/oe-KLF5, and the luciferase activity was subsequently examined.

2.11. Coimmunoprecipitation (Co-IP)

Cells were lysed using the lysis buffer, and the lysates were incubated with IgG (Abcam, 1:50, ab172730) and SMURF1 (Abcam, 1:50, ab57573) antibodies at 4 °C overnight. The lysates were subsequently incubated with Protein G agarose (Millipore) for 3 h and subjected to Western blot analysis.

2.12. YY1 ubiquitination analysis

After treatment, MPC5 cells were lysed in 1% SDS buffer and boiled (10 min), the lysates were incubated with anti-YY1 or anti-IgG antibody and protein A/G IP magnetic beads for 12 h. YY1 ubiquitination was analyzed by Western blot with anti-Ub antibody.

2.13. Immunofluorescence

Cells were fixed and then sealed for 30 min, followed by incubation overnight with antibodies against GRP78 (Abcam, 1:100, ab21685). Cells were incubated with the secondary antibody (Abcam, 1:3000, ab7090) for 1 h, and the nucleus was stained with DAPI (Sangon). Cells were subsequently sealed and observed under a fluorescence microscope (Olympus, Tokyo, Japan).

2.14. Cycloheximide (CHX) chase assay

After transfection with vector or OE-SMURF1 for 24 h, MPC5 cells were subjected to CHX (100 μg/mL, Sigma-Aldrich, MO, USA) for 0, 4, 8 and 12 h and lysed in a lysis buffer containing 2% SDS. The lysates were subjected to Western blot analysis.

2.15. Animal experiments

C57BL/6J female mice (8-week-old, n = 8) and MRL/lpr female mice (8-week-old, n = 24) were obtained from SJA LABORATORY Animal Co, Ltd. (Hunan, China) and randomly divided into four groups, each with eight mice [1]: Sham group: C57BL/6J mice [2]; LN group: MRL/lpr mice [3]; LN + sh-NC group: MRL/lpr mice injected with lentivirus empty vector [4]; LN + sh-SMURF1 group: MRL/lpr mice injected with knockdown SMURF1 lentivirus. The concentration of lentivirus used was 1 × 10⁸ vg/mL, and the effects of lentivirus in all mice lasted for one month after in vivo injection. Then, mice were euthanized with carbon dioxide (CO₂), and the kidney tissues were collected for further experiments. All procedures involving animals were performed strictly in accordance to the guide for the Care and Use of Laboratory Animals, under the approval by Hainan women and Children’s medical center.

2.16. Hematoxylin-eosin (HE) staining

The paraffin sections of kidney tissue were prepared (4 μm) and dehydrated using alcohol. The sections were subsequently stained with HE (Sigma-Aldrich) and observed under a microscope (Olympus).

2.17. Terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) staining

The kidney tissues were fixed and permeabilized followed by wash with PBS. The staining was performed using the TUNEL staining kit (Beyotime), and counterstaining was carried out with DAB.

2.18. Immunohistochemistry (IHC)

The kidney tissue sections (4 μm in thickness) were prepared. After deparaffinization and antigen retrieval (Dako, CA, USA), sections were then blocked and incubated with antibody against GRP78 (Abcam, 1:200, ab21685) overnight followed by incubation with an appropriate secondary antibody (Abcam, 1:500, ab150077) for 1 h. The sections were stained with DAB and then counterstained with hematoxylin, dehydrated and mounted. The images were taken using a microscope (Olympus).

2.19. Urine protein level measurement

Urine protein level was measured by the urine protein detection kit (Biorab, Beijing, China) following the manufacturers’ instructions.

2.20. Online database analyses

JASPAR (http://jaspar.genereg.net/) is an open-access database containing manually curated, non-redundant transcription factor (TF) binding profiles for TFs across six taxonomic groups [19]. JASPAR was used to predict TFs that might affect cGAS. UbiBrowser 2.0 (http://ubibrowser.ncpsb.org.cn) is a comprehensive resource for proteome-wide known and predicted ubiquitin ligase/deubiquitinase-substrate interactions in eukaryotic species [20]. UbiBrowser 2.0 was used to predict the possible ubiquitinases for regulating YY1.

2.21. Statistical analysis

All the data from three independent experiments were analyzed by SPSS 19.0 and expressed as means ± SD. Student’s t-test or one-way ANOVA was used to determine the differences among groups. The correlation between, SMURF1, YY1 and cGAS was analyzed using Pearson correlation analysis. The p values less than 0.05 were considered significant.

3 Results

3.1. SMURF1, cGAS and STING were markedly upregulated in the serum of LN patients, while YY1 was downregulated

As shown in Figure 1A–D, SMURF1, cGAS and STING mRNA levels were markedly increased in the serum of LN patients in comparison to those in the serum of healthy people, while YY1 was markedly downregulated. In addition, SMURF1 expression was negatively correlated with YY1 expression in clinical samples, and YY1 expression was negatively correlated with cGAS expression (Figure 1E–F). Collectively, SMURF1, YY1, cGAS and STING dysregulation might be associated to LN progression.

Figure 1. SMURF1, cGAS and STING were markedly upregulated in the serum of LN patients, while YY1 was downregulated. (A–D) SMURF1, YY1, cGAS and STING mRNA levels in the serum of LN patients (n = 22) and healthy people (n = 22) were determined by qRT-PCR. (E-F) The correlation between, SMURF1, YY1 and cGAS was analyzed using Pearson correlation analysis. Data were expressed as mean ± SD. ***p < 0.001.

3.2. YY1 overexpression alleviated LN-IgG-induced ERS and apoptosis in podocytes

To investigate the role of YY1 in LN, podocytes were treated with serum IgG from LN patients (LN-IgG) and YY1 overexpression was induced. It was firstly indicated that YY1 in podocytes was markedly upregulated following OE-YY1 transfection (Figure 2A), suggesting the transfection was successful. As illustrated in Figure 2B–C, LN-IgG treatment reduced YY1 expression level in podocytes, whereas these changes were reversed by OE-YY1 transfection. In addition, podocytes viability was reduced by LN-IgG stimulation, which was abolished by YY1 overexpression (Figure 2D). Podocyte apoptosis and ERS play essential roles in LN progression [21]. Herein, it was observed that LN-IgG treatment markedly promoted podocyte apoptosis, which was reversed by YY1 overexpression (Figure 2E). Meanwhile, LN-IgG stimulation upregulated Bax level and downregulated Bcl-2 level in podocytes, while these changes in apoptosis marker protein expression induced by LN-IgG were eliminated following YY1 overexpression (Figure 2F). In addition, the levels of ERS-related proteins (GRP78, CHOP, ATF6 and PERK) in podocytes were markedly increased by LN-IgG, which were abrogated by YY1 overexpression (Figure 2G–H). Taken together, YY1 overexpression could reduce ERS and apoptosis in podocytes during LN development.

Figure 2. YY1 overexpression alleviated LN-IgG-induced ERS and apoptosis in podocytes. (A) YY1 expression in MPC5 cells after vector or OE-YY1 transfection was examined by qRT-PCR. MPC5 cells were treated with serum IgG from LN patients (LN-IgG) and YY1 overexpression was induced. (B-C) YY1 level in cells was determined using qRT-PCR and western blot. (D) CCK8 assay was performed to determine cell viability. (E) Cell apoptosis was analyzed by flow cytometry. (F) Bax and Bcl-2 protein levels were measured by western blot. (G) Immunofluorescence was employed to analyze GRP78 level in cells. (H) GRP78, CHOP, ATF6 and PERK protein levels in cells were assessed using western blot. Data were expressed as mean ± SD. All our data were obtained from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

3.3. SMURF1 inhibited YY1 expression by regulating YY1 protein ubiquitination

Subsequently, ubibrowser V2 was used to predict the possible ubiquitination enzymes (SMURF1, NEDD4L, BMI1, etc.) regulating YY1. Preliminary experiments showed that the expressions of BMI1 and NEDD4L were decreased and the expression of SMURF1 was increased in the IgG-induced podocytes (Figure S1). SMURF1, as an E3 ubiquitin ligase, functions in regulating the ubiquitination level of targets [22]. Our results illustrated that OE-SMURF1 transfection resulted in increased SMURF1 protein level and decreased YY1 level in podocytes (Figure 3A). The protein-binding relationship between SMURF1 and YY1 was confirmed by Co-IP assay (Figure 3B). It was also turned out that protease inhibitor (MG132) treatment abrogated SMURF1 overexpression mediated inhibition on YY1 expression in podocytes (Figure 3C). The following evidences showed that YY1 degradation rate in OE-SMURF1 cells was dramatically accelerated compared with OE-NC cells (Figure 3D). In addition, the YY1 ubiquitination level was significantly increased after SMURF1 overexpression (Figure 3E). Collectively, SMURF1 inhibited YY1 expression in podocytes by regulating YY1 ubiquitination modification.

Figure 3. SMURF1 inhibited YY1 expression by regulating YY1 protein ubiquitination. (A) YY1 and SMURF1 expressions in MPC5 cells after vector or OE-SMURF1 transfection were examined by western blot. (B) The binding relationship between SMURF1 and YY1 was verified using Co-IP assay. (C) YY1 protein level in SMURF1-overexpressing podocytes after MG132 treatment was analyzed by western blot. (D) The degradation of YY1 in SMURF1-overexpressing podocytes after CHX treatment was analyzed by western blot. (E) YY1 ubiquitination was analyzed by ubiquitination analysis. Data were expressed as mean ± SD. All our data were obtained from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

3.4. SMURF1 knockdown inhibited LN-IgG-induced ERS and apoptosis in podocytes by mediating ubiquitination of YY1

To investigate the interaction between SMURF1 and YY1 in regulating ERS and apoptosis in podocytes during LN development, both SMURF1 knockdown and YY1 knockdown were induced in LN-IgG-treated podocytes. As shown in Figure 4A, SMURF1 expression in podocytes was markedly reduced by sh-SMURF1 transfection, and the expression of YY1 was reduced by sh-YY1 transfection. Meanwhile, it was indicated that SMURF1 knockdown increased YY1 expression in LN-IgG-treated podocytes, while this effect of sh-SMURF1 was eliminated by YY1 knockdown (Figure 4B–C). It was also showed that SMURF1 knockdown prevented LN-IgG-induced decrease in podocyte viability, whereas this effect of sh-SMURF1 was abolished after YY1 silencing (Figure 4D). Our results also revealed LN-IgG-induced increase in podocyte apoptosis was ameliorated by SMURF1 knockdown, whereas this regulatory effect of SMURF1 knockdown was abolished by YY1 knockdown (Figure 4E). Consistently, SMURF1 knockdown reduced Bax level and increased Bcl-2 level in LN-IgG-treated podocytes, while YY1 knockdown eliminated all these effects of SMURF1 knockdown (Figure 4F). Moreover, the protein levels of ERS-related proteins (GRP78, CHOP, ATF6 and PERK) in LN-IgG-treated podocytes were markedly reduced by sh-SMURF1 transfection, while this sh-SMURF1’s effect was reversed by YY1 knockdown (Figure 4G–H). In total, SMURF1 knockdown inhibited ERS and apoptosis in podocytes during LN development by targeting YY1.

Figure 4. SMURF1 knockdown inhibited LN-IgG-induced ERS and apoptosis in podocytes by mediating ubiquitination of YY1. (A) SMURF1 or YY1 expression in MPC5 cells after sh-NC or sh-SMURF1/sh-YY1 transfection was examined by qRT-PCR. Both SMURF1 knockdown and YY1 knockdown were induced in LN-IgG-treated podocytes. (B-C) YY1 expression level in cells was assessed using qRT-PCR and western blot. (D) Cell viability was measured using CCK8 assay. (E) Cell apoptosis was detected using flow cytometry. (F) Bax and Bcl-2 protein levels in cells were examined using western blot. (G) GRP78 level in cells was analyzed using immunofluorescence. (H) GRP78, CHOP, ATF6 and PERK protein levels in cells were determined using western blot. Data were expressed as mean ± SD. All our data were obtained from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

3.5. YY1 alleviated podocyte injury by transcriptional inhibition cGAS/STING/IFN-1 signal axis

The cGAS/STING/IFN-1 signal axis is closely related to LN progression [15]. In addition, by using JASPAR prediction, it was found that cGAS had potential binding sites to multiple transcription factors (KLF5, TCF4 and ATF4). However, we subsequently found through dual luciferase reporter gene assay that transcription factor KLF5, TCF4 and ATF4 had no significant effect on the transcriptional activity of cGAS (Figure S2A–C). Therefore, the role of YY1/cGAS in LN was subsequently investigated. Herein, by using JASPAR prediction, it was found that YY1 had potential binding sites to cGAS promoter region (Figure 5A). It was subsequently displayed that YY1 directly bound with cGAS (Figure 5B–C). It was also turned out that OE-YY1 transfection led to a reduction in cGAS, STING and IFN-1 expression in podocytes (Figure 5D). As demonstrated in Figure 5E, sh-cGAS transfection resulted in decrease in cGAS mRNA level in podocytes, suggesting the transfection was successful. In addition, cGAS, STING and IFN-1 protein levels in podocytes were markedly increased by LN-IgG treatment, while cGAS knockdown ameliorated LN-IgG-induced increase in cGAS, STING and IFN-1 levels, and this effect of cGAS knockdown was abrogated after YY1 knockdown (Figure 5F). Functional experiments subsequently demonstrated that LN-IgG-induced decrease in podocyte viability was prevented by cGAS knockdown, which was abolished after YY1 silencing (Figure 5G). Additionally, cGAS knockdown ameliorated LN-IgG-induced increase in podocyte apoptosis, whereas this regulatory effect of cGAS knockdown was abolished by YY1 knockdown (Figure 5H). Meanwhile, it was observed that cGAS knockdown reduced Bax level and increased Bcl-2 level in LN-IgG-treated podocytes, whereas YY1 knockdown eliminated the regulatory effect of sh-cGAS on Bax and Bcl-2 expressions (Figure 5I). Moreover, the protein levels of ERS-related proteins (GRP78, CHOP, ATF6 and PERK) in LN-IgG-treated podocytes were markedly reduced by cGAS knockdown, while this cGAS knockdown’s effect was reversed by YY1 knockdown (Figure 5J–K). All these results suggested that YY1 alleviated ERS and apoptosis in podocytes during LN development by inactivating cGAS/STING/IFN-1 signal axis.

Figure 5. YY1 alleviated podocytes injury by transcriptional inhibition cGAS/STING/IFN-1 signal axis. (A) The potential binding sites between YY1 and cGAS promoter region was predicted by JASPAR. (B-C) ChIP and dual-luciferase reporter gene assays were performed to analyze the interaction between YY1 and cGAS. (D) cGAS, STING and IFN-1 protein levels in podocytes following vector or OE-YY1 transfection were assessed by western blot. (E) cGAS mRNA level in MPC5 cells after sh-NC or sh-cGAS transfection was examined by qRT-PCR. Both cGAS knockdown and YY1 knockdown were induced in LN-IgG-treated podocytes. (F) cGAS, STING and IFN-1 protein levels in podocytes were detected by western blot. (G) CCK8 assay was employed to analyze cell viability. (H) Cell apoptosis was examined using flow cytometry. (I) Bax and Bcl-2 protein levels in cells were assessed by western blot. (J) GRP78 level in cells was analyzed using immunofluorescence. (K) GRP78, CHOP, ATF6 and PERK protein levels in cells were examined by western blot. Data were expressed as mean ± SD. All our data were obtained from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

3.6. SMURF1 knockdown inhibited cGAS/STING/IFN-1 signal axis by targeting YY1 to suppress LN progression in mice

To further study the molecular mechanism of SMURF1 in LN progression in vivo, sh-SMURF1 lentivirus were injected into LN mice. The level of urinary protein of MRL/lpr mice was gradually increased at 20 weeks compared with C57BL/6 mice (Figure 6A). HE staining results revealed that typical glomerular hyperplasia, crescent body, glomerulosclerosis and interstitial inflammation were observed in the renal interstitium of MRL/lpr mice; mice in LN + sh-SMURF1 group still showed slight glomerular hyperplasia and slight interstitial inflammation, but these severe kidney injuries were significantly reduced (Figure 6B). It was also turned out that SMURF1 was markedly upregulated and YY1 was downregulated in kidneys of MRL/lpr mice, while these changes were all eliminated by SMURF1 knockdown (Figure 6C). In addition, cGAS, STING and IFN-1 protein levels were markedly increased in kidney tissues of MRL/lpr mice, which were abrogated by SMURF1 knockdown (Figure 6D). It was also turned out that GRP78 protein level was markedly increased in kidney tissues of MRL/lpr mice, whereas SMURF1 knockdown reversed this change (Figure 6E). Moreover, cell apoptosis was markedly increased in the kidney of LN group, while this change was abrogated following SMURF1 knockdown (Figure 6F). In summary, SMURF1 knockdown inactivated cGAS/STING/IFN-1 signal axis by targeting YY1 to inhibit LN progression in vivo.

Figure 6. SMURF1 knockdown inhibited cGAS/STING/IFN-1 signal axis by targeting YY1 to suppress LN progression in mice. sh-SMURF1 lentivirus were injected into LN mice (MRL/lpr mice). (A) The level of urinary protein was shown by urinary Albumin/Creatinine. (B) The representative pictures of HE staining of kidney tissues. (C) SMURF1 and YY1 mRNA levels in kidney tissues were detected using qRT-PCR. (D) Western blot was conducted to analyze cGAS, STING and IFN-1 protein levels in kidney tissues. (E) GRP78 protein level in kidney tissues was determined using IHC. (F) Cell apoptosis was examined using TUNEL staining. Data were expressed as mean ± SD. n = 8. *p < 0.05, **p < 0.01, ***p < 0.001.

4 Discussion

LN is a kind of immune complex glomerulonephritis, which is the most important disabling and fatal factor of SLE [23]. Podocytes are epithelial cells in the visceral layer of renal sacs located outside the basement membrane of the glomerulus, which participate in the ultrafiltration of blood. Podocyte injury is a key feature of LN progression [21]. ERS and apoptosis are key players in LN-induced podocyte injury, but its mechanism is still unclear. The primary novel findings in the current research are that SMURF1 activated the cGAS/STING/IFN-1 signal axis by mediating YY1 ubiquitination to promote ERS and apoptosis in podocytes during LN progression.

Ubiquitination is closely related to LN pathogenesis [24,25]. Ubiquitination modification is regulated by ubiquitination regulatory enzymes [26]. SMURF1 is an E3 ubiquitin ligase, which recognizes the ubiquitination protein substrate and selectively regulates the ubiquitination degradation process of various effector molecules including Smad1, RhoA, MEKK2, and RUNX [5,6]. Li et al. revealed that SMURF1 inhibition could prevent acute kidney injury to CDK transition [9]. In addition, it was previously reported SMURF1 inhibition ameliorated aristolochic acid-induced tubular necrosis and interstitial fibrosis [27]. Herein, our results revealed that SMURF1 knockdown alleviated LN-IgG-induced ERS and apoptosis in podocytes in vitro and improved renal lesions, renal cell apoptosis and ERS in LN mice. Collectively, SMURF1 upregulation aggravated LN progression by promoting ERS and apoptosis in podocytes.

As widely reported, SMURF1, an E3 ubiquitin ligase, achieves its role in diseases by regulating the ubiquitination modification of target proteins [28,29]. SMURF1 upregulation promoted renal fibrosis in diabetic kidney disease by regulating the ubiquitination of Smad3 [30]. YY1 is a relatively conserved evolutionary zinc finger protein which functions as a transcription factor [31]. The role of YY1 in kidney diseases has been widely reported. As proof, Du et al. demonstrated that YY1 knockdown exacerbated renal fibrosis in diabetic nephropathy [32]. In addition, YY1 was markedly upregulated in rapamycin-treated kidney angiomyolipoma, and its upregulation inhibited renal fibrosis [33]. Notably, YY1 was markedly downregulated in the kidney tissues of LN mice, and its overexpression reduced proinflammatory factor production, cell apoptosis and Th17/Treg cells ratio in LN mice and TNF-α-treated Jurkat cells [12]. Herein, our results showed that YY1 was significantly downregulated in LN, and its overexpression alleviated LN-IgG-induced ERS and apoptosis in podocytes. Encouragingly, our results revealed that SMURF1 reduced YY1 protein stability and expression by ubiquitinating YY1 protein to aggravate LN-IgG-induced ERS and apoptosis in podocytes.

Transcription factor YY1 is involved in the development of disease by transcriptional regulation of the downstream targets [34]. For instance, YY1 accelerated tubulointerstitial fibrosis in diabetic nephropathy by inactivating transcription PGC-1α [35]. cGAS is the main DNA sensor in human immortal podocytes, which can recognize the DNA that should not appear in the cytoplasm and synthesize cyclic GMP-AMP to activate the ER protein STING, which then transduces signals into the nucleus through the STING pathway to regulate gene transcription and initiate immune response [36]. cGAS-STING signal transduction plays an important regulatory role in immunology by inducing cytokine secretion (mainly IFN-1) [37]. In LN, the release of HMGB1 contributes to the endocytosis of extracellular DNA accumulation and cGAS/STING/IFN-1 signal pathway activation, which is a key pro-inflammatory pathway in LN progression [15]. In addition, it has been widely illustrated that cGAS/STING/IFN-1 signal axis activation induces ERS in multiple types of cells, such as alveolar epithelial cells [38] and macrophages [39]. Nevertheless, the role of cGAS/STING/IFN-1 signal axis in regulating ERS in podocytes during LN progression remains unknown. In the current research, cGAS and STING were significantly upregulated in LN. Our subsequently experiments revealed that YY1 inhibited ERS and apoptosis in podocytes during LN development by transcriptional inhibition cGAS/STING/IFN-1 signal axis. Taken together, the cGAS/STING/IFN-1 signal axis acted as the downstream pathway of YY1 in regulating ERS and apoptosis in podocytes during LN progression.

Taken together, our research provided evidence that SMURF1 activated the cGAS/STING/IFN-1 signal axis by mediating YY1 ubiquitination, thereby promoting ERS and apoptosis in podocytes to facilitate LN progression, which systematically elucidated the mechanism of SMURF1 in LN development and suggested that targeting SMURF1 might be a potential therapeutic approach for LN.

Authors’ contributions

Xiaoyan Li: Conceptualization; Methodology; Validation; Writing – Original Draft; Sisi Tao: Formal analysis; Investigation; Zhiquan Xu: Resources; Data Curation; Yi Ren: Visualization; Wei Xiang: Supervision; Xiaojie He: Writing – Review and Editing; Project administration; Funding acquisition.

Ethics approval and consent to participate

The study was authorized by the Ethics Committee of Hainan women and Children’s medical center before enrollment of patients. All participants signed informed consent. All procedures involving animals were performed strictly in accordance to the guide for the Care and Use of Laboratory Animals, under the approval by Hainan women and Children’s medical center. Approval number: [Ethical Review No. 118 of 2022].

Competing interests

The authors declare that there is no conflict of interest.

Consent for publication

The informed consent was obtained from study participants.

Abbreviations
(SLE)	=	Systemic lupus erythematosus
(LN)	=	Lupus nephritis
(CKD)	=	Chronic kidney disease
(ESRD)	=	End-stage renal disease
(SMURF1)	=	Smad ubiquitination regulatory factor 1
(YY1)	=	Yin Yang 1
(cGAS)	=	Cyclic GMP-AMP synthase
(STING)	=	Stimulator of interferon genes
(IFN)	=	Interferon
(TGFβ-1)	=	Transforming growth factor β-1
(ERS)	=	Endoplasmic reticulum stress
(CCK-8)	=	Cell counting kit-8
(IHC)	=	Immunohistochemistry
(HE)	=	Hematoxylin-eosin
(Co-IP)	=	Coimmunoprecipitation
(CHX)	=	Cycloheximide
(ChIP)	=	Chromatin immunoprecipitation
(qRT-PCR)	=	Quantitative real-time polymerase chain reaction
(SD)	=	Standard deviation
(ANOVA)	=	Analysis of variance
(EMT)	=	Epithelial-mesenchymal transition

Availability of data and materials

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The authors express their gratitude to for his help for directing our article. This work was supported by grants from 2022JJ30064, 2022JJ70048, Natural Science Foundation of Hunan Province; 2021SK53201, Hunan Province clinical medical technology innovation guide project; S2022JJKWLH0255, Hainan Province science and health joint project; 821RC1130, Natural Science Foundation of Hainan Province; ZDYF2021SHFZ088, Key R&D Plan of Hainan Province