TLR4/NF-κB Signaling Induces GSDMD-Related Pyroptosis in Tubular Cells in Diabetic Kidney Disease

Gasdermin D (GSDMD) has been proven to be a key protein in the activation of pyroptosis. Pyroptosis of renal tubular epithelial cells contributes to the progression of tubular injury in kidney diseases. However, it remains elusive whether and how GSDMD is involved in the regulation of diabetic kidney disease (DKD). In this study, we found that tubular injury is accompanied by the up-regulation of Toll-like receptor 4 (TLR4) and GSDMD in patients with diabetic kidney disease. In addition, we discovered that the expressions of cleaved Caspase-1, active N-terminal fragments of GSDMD (GSDMD-NT), IL-18, and the secretion of IL-1β also increased in the kidneys of db/db mice. These changes were partially ameliorated following intraperitoneal injection of TAK-242, an inhibitor of TLR4. Similar results were observed in human tubular cells (HK-2) subjected to high-glucose (HG) conditions and treated with TAK-242 or parthenolide (inhibitor of NF-κB) by Western blot, Enzyme-linked immunosorbent assay (ELISA), and flow cytometry. These results indicated that TLR4/NF-κB signaling could induce GSDMD-mediated pyroptosis in tubular cells in DKD.


INTRODUCTION
Recently, tubular injury induced by high glucose has been taken into consideration to be an independent risk factor for the progression of diabetic kidney disease (DKD) (1,2). Pyroptosis is a programmed cell death mediated by the activation of caspase family, such as caspase-1, in the pathogenesis of disorders characterized by excessive cell death and inflammation. This form of cell death is accompanied by water influx, cellular swelling, osmotic lysis, activation of NLRP3 inflammasome, and the release of pro-inflammatory cytokines, such as IL-1β and IL-18 (3,4). Researchers have found that pyroptosis could be induced by hyperglycemia in diabetic cardiomyopathy (5). In addition, several reports have proved that pyroptosis of renal tubular epithelial cells is an indispensable process for contrast-induced acute kidney injury and renal ischemia-reperfusion injury (6,7), indicating the contribution of pyroptosis in the progression of tubular injury in kidney diseases. However, few studies have focused on the role of pyroptosis in tubular cells of diabetic kidney disease.
Toll-like receptor 4 (TLR4) is one of TLRs family which plays a fundamental role in the activation of innate and adaptive immune responses (8,9). It has been reported to be involved in the pathogenesis of acute and chronic renal disorders such as acute kidney injury (AKI), renal fibrosis, and DKD (10,11). TLR4 typically signals through its downstream partner MyD88 to activate the nuclear factor-κB (NF-κB) pathway, leading to ROS and cytokine production. In HG-stimulated podocytes, TLR4 was found to activate NF-κB, increasing the release of proinflammatory cytokines and chemokines (12,13). All of these findings indicate that TLR4/NF-κB signaling plays a crucial part in the progression of DKD. TLR4 was also proved to induce Caspase-11 expression and Caspase-1 activation, resulting in cellular pyroptosis and increased release of pro-inflammatory cytokines (14). Recently, gasdermin D (GSDMD), which belongs to a family of gasdermins, mainly expressed in the skin and mucosa, has been proven to be a key protein in the activation of pyroptosis. It could be cleaved by activated inflammatory caspases such as Caspase-1. The active N-terminal fragment oligomerizes to form membrane pores which cause leakage of the pro-inflammatory cytokines. But the role of GSDMD in the TLR4-induced pyroptosis in DKD patients is not clear.
Our aim in this study is to explore the effect of the TLR4/NF-κB pathway on the pyroptosis of renal tubular epithelial cells in DKD, and to investigate the role of GSDMD in this TLR4/NF-κBmediated pathway.

Morphological Analysis of Kidney
Thirty patients with DKD and 13 non-DKD controls were recruited for this study. We chose patients without diabetes (non-DKD) who clinically manifest as mild hematuria or proteinuria patients, have kidney biopsies diagnosed as minor glomerular lesions, and do not incorporate with other glomerular diseases such as IgA nephropathy or membranous nephropathy as controls. Kidney biopsy tissues were stained with periodic acidsilver methenamine (PASM). A semiquantitative scoring system was used to evaluate the tubulointerstitial lesion index and tubular damage according to guidelines in 2010 by Tervaet et al. (1,15). The percentage area of tubular injury was estimated by a pathologist in a blinded manner according to the following criteria: tubular dilation, detachment of brush border, tubular atrophy, and interstitial fibrosis in every high-power field across whole tissue sample (magnification, ×200). All patients did not use adrenal cortical hormones or immunosuppression. All of the DKD patients use insulin combined with metformin to control blood glucose. The institutional review board and the administrators of the Department of Nephrology in Second Xiangya Hospital approved the protocol for this study. Written informed consent was obtained from all the participants.

Animal Experimental Design
A total of 5 adult male db/m mice and 10 adult male db/db mice at 16 weeks of age were divided into three groups of 5 animals each. Male db/m mice served as a control (db/m group). Db/db mice received an intraperitoneal injection with the vehicle alone (db/db group), or TLR4 inhibitor for 7 consecutive days (3 mg/kg) (dbdb + TAK242 group). The mice were sacrificed at 17 weeks following administration. The kidney tissue of the mice was processed as described previously (16). The Institutional Animal Experimentation Ethics Committee as described above approved the experimental animal protocols.

Immunohistochemistry (IHC)
Renal tissue sections for immunostaining were deparaffinized, rehydrated, and antigen retrieval in a microwave oven. Immunohistochemistry was performed using anti-TLR4 (1:100), GSDMD (1:100, Proteintech), GSDMD-NT (1:800, abcam) and IL-18 (1:100, abcam) antibody as the primary antibodies, followed by a secondary antibody. Then slides were visualized by using a DAB detection kit according to the manufacturer's instructions, and the tissue specimens were examined by light microscopy. The TUNEL staining was used to detect cell death conditions following the manufacturer's instructions.

Data Analysis
Statistical analysis was carried out using the SPSS 20 software. Quantitative Data are expressed as means ± SEM (standard error of the mean). Statistical differences among groups were analyzed by One-way ANOVA, and two-tailed P-values were reported. Spearman's correlation was used to evaluate the correlations among each histopathological and clinical finding. P-values < 0.05 were considered statistically significant.  Frontiers in Endocrinology | www.frontiersin.org

Clinical Characteristics and Kidney Morphological Changes in Patients With DKD
Patients with DKD exhibited higher blood glucose and 24 h urine protein levels compared with non-DKD patients, as shown in Table 1. PASM staining showed focal tubular atrophy and interstitial fibrosis in DKD patients (arrow) compared with those in non-DKD patients (Figure 1Aa,b). TLR4 and GSDMD expression was examined in paraffin-embedded sections of human diabetic kidney tissues by immunohistochemical staining. Both enhanced TLR4 (Figure 1Ac,d) and GSDMD (Figure 1Ae,f) expressions were observed predominantly in proximal tubules of DKD patients.
To further determine the relationship between GSDMD expression and diabetic nephropathy and tubular injury, correlation analyses were made. Results showed that GSDMD expression was positively correlated with TLR4 expression (r = 0.988, P = 0.000), proteinuria (r = 0.423, P = 0.01), tubular interstitial fibrosis and tubular atrophy (IFTA) scores (r = 0.510, P = 0.007), and negatively correlated with eGFR (r = −0.388, P = 0.012). These findings revealed that TLR4 and GSDMD may be involved in the development of DKD.

Inhibition of TLR4 Protects Tubular Cell Injury by Suppressing GSDMD-Related Pyroptosis in db/db Mice
To explore whether in vivo inhibition of the TLR4 influences tubular injury and GSDMD-mediated pyroptosis under diabetic conditions, we examined GSDMD-NT, caspase-1 p20, and IL-18 expression in renal tissues by western blot or immunohistochemistry staining, and the level of IL-1β in kidney cortex by ELISA.
As shown in Figure 2Aa-c, increased detachment of renal tubular brush border, tubular atrophy, and interstitial fibrosis in db/db mice was seen when compared with the control group, which were ameliorated by the injection of TLR4 inhibitor TAK-242. IHC staining of GSDMD (Figure 2Adf) and IL-18 (Figure 2Ag-i) showed increased expression predominately located in renal tubular cells in db/db mice, which decreased after TAK-242 treatment. Increased cell death was seen in tubular cells of diabetic mouse kidneys by TUNEL staining (Figure 2Aj-l), while the effect was alleviated by TLR4 inhibitor treatment.
There was no significant difference in body weight and blood glucose levels between db/db mice group and db/db+TAK-242 group (Figures 2B,C). As seen in Figure 2D, urinary excretion of β-NAG, the marker of tubular injury, was consistently increased in diabetic mice. Treatment with TAK-242 significantly lowered the elevated urinary β-NAG level. The same changes also happened in tubular injury score (Figure 2E), indicating improvement of tubular function after TLR4 inhibition.
We further detected the protein levels of Caspase-1 p20 and GSDMD-NT in renal cortex tissue. Both of which were increased in db/db mice and obviously decreased after TAK-242 injection (Figures 2F,G). IL-1β level in kidney homogenates ( Figure 2H) was upregulated in db/db mice and decreased after TAK-242 treatment. These findings revealed that TLR4 inhibition distinctly suppressed GSDMD-related pyroptosis and alleviated renal injury in db/db mice.

TLR4/ NF-κB and GSDMD-NT Expressions Were Increased in HK-2 Cells Under HG Ambience
To dissect the effect of high glucose on tubular TLR4 and GSDMD-NT expression in vitro, we exposed cultured HK-2 cells to D-glucose (15-45 mM) for 24 h and showed an upregulation of TLR4 ( Figure 3A) and GSDMD-NT expression in a dosedependent manner, whereas exposure to an equivalent dose of mannitol (45 mM) had no effect on GSDMD-NT expression ( Figure 3B).

Inhibition of TLR4/NF-κB Signaling Decreased the Expression of Caspase-1 and GSDMD-NT in HK-2 Cells Under HG Ambience
Due to the similar change of TLR4 and GSDMD-NT in HK-2 cells under different concentration of high glucose, we next examined the effect of TLR4 or NF-κB inhibition on pyroptosisrelated protein. The levels of p-p65, p65, Caspase-1, Caspase-1 p20, GSDMD, and GSDMD-NT were determined by western blot, and ELISA was performed to detect the level of IL-1β in supernatant of HK-2 cells.
The results demonstrated that the expression levels of p-p65, Caspase-1 p20, and GSDMD-NT were increased in HGstimulated HK-2 cells than that in control group. However, TLR4 or NF-κB inhibition led to a significant decrease of p-p65, Caspase-1 p20, and GSDMD-NT in HG-stimulated HK-2 cells (Figure 4), indicating that NF-κB was the downstream signal molecules of TLR4 under HG ambience and the activation of pyroptosis in HK-2 cells was markedly inhibited by TLR4/NF-κB signaling inhibition.

Inhibition of TLR4/NF-κB Signaling Alleviated Pyroptosis in HK-2 Cells Under HG Ambience
Pyroptosis is a caspase-1-dependent regulated cell death. To verify the necrotic nature of high glucose induced pyroptosis, caspase-1 activation in HK-2 cells were detected by labeling the cells with PI and Alexa Fluor 488 labeled caspase-1 FLICA. Double positive cells were considered to be pyroptotic cells. As shown in Figure 5, high glucose induced an increase in HK-2 cells pyroptosis compared to the control. However, TLR4 or NF-κB blockers treatment alleviated these changes, suggesting an essential role of TLR4/NF-κB pathway in pyroptosis in HGstimulated HK-2 cells.

DISCUSSION
In this study, we identified that TLR4/NF-κB signaling could activate GSDMD-related pyroptosis in high glucose ambience. We have proved that GSDMD, which is the key factor in the process of pyroptosis, had a close correlation with tubular injury. In addition, TLR4/NF-κB signaling might involve in the mechanism of GSDMD-related tubular pyroptosis in high glucose ambience during the progression of DKD.
Researches have been proved that TLR4-mediated pathway play an essential role in the progression of DKD (11,16). In addition, several evidences have been shown that TLRs activation can trigger inflammasome formation and cell death pyroptosis, which is dependent on caspase-1 (17)(18)(19)(20). These findings promote us to detect the expression of TLR4 in DKD patients and analyze the correlation between them in tubular cells. We detected the increased expression of TLR4 in tubular cells was related with increased tubular injury (Figure 1). In vivo and in vitro, high-glucose induced increased expression of TLR4, cleaved caspase-1, GSDMD and secretion of IL-1β, which could partly reversed by TLR4 and NF-κB inhibitors. In addition, the inhibition of TLR4 could reduce the activation of NF-κB, and inhibitor of NF-κB (parthenolide) could also ameliorate the pyroptosis level of HK-2 cells under the high glucose condition. These results indicated that TLR4/NF-κB signaling modulates pyroptosis in tubular cells in DKD.
Recently, GSDMD has been identified to be one of the key proteins in caspase-1-mediated pyroptosis and controlling IL-1β release in mouse bone marrow macrophages (21,22). Activated caspase-1 cleaves GSDMD to generate an N-terminal cleavage product (GSDMD-NT) that triggers pyroptosis. But there is no research focusing on whether GSDMD also contributes to the development of pyroptosis in DKD. In this study, we have detected strong expression of GSDMD in the tubulars of DKD patients. Diabetic mice model and HK-2 cells experiments show that the inhibition of TLR4/NF-κB signaling could reverse the increased expression of GSDMD-NT under the high glucose ambience, accompanied with the inhibition of pyroptosis and the release of IL-1β. Thus, TLR4/NF-κB signaling was proved to be involved in GSDMD-related pyroptosis in DKD.

CONCLUSION
TLR4/NF-κB signaling could activate GSDMD-related pyroptosis in tubular cells in vitro and in vivo, hoping to provide a better understanding and more effective therapeutic approach for the prevention and treatment of DKD.

DATA AVAILABILITY
The raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher.

ETHICS STATEMENT
Kidney biopsy sections from patients diagnosed with DN (n = 30) and non-diabetic patients diagnosed as minor glomerular lesions (N-DKD, n = 13) were obtained from the second Xiangya Hospital of Central South University following the Institutional Review Board of the Second Xiangya Hospital. Written informed consent was obtained from all the participants. Animal experiments were approved by the institutional committees of the Animal Research Committee and Animal Ethics Committee of the Second Xiangya Hospital.

AUTHOR CONTRIBUTIONS
YW revised the data, edited the manuscript, and completed submission. XZ conceived and supervised the study. SY reviewed the manuscript and helped interpreted the histology data. SW performed most experiments and drafted the manuscript. XL guided the cell experiments. ZQ and CW helped in animal experiments. JL and HL helped designed the experiments. FL provided guidance to the study. All authors read and approved the manuscript.