Rehin Mitigates Chronic Kidney Injury by Decreasing the Release of Perforin and IFN-γ from Immune Cells Via Inhibiting the STING/TBK1/IRF3 Pathway Activation

Shenkang suppository (SKS), a Chinese medicinal preparation rich in various natural ingredients, has not been reported in any studies related to brosis. Our experiments validated the anti-brotic and anti-inammatory effects of Rhein (Rh), which is a major component of SKS, and explored its potential immune mechanisms. Tissue and serum specimens from chronic kidney disease (CKD) patients and normal subjects were collected in 30 cases each, and the expression differences of perforin and IFN-γ were analyzed by ELISA. Further, the CKD mice model constructed with folic acid (FA) was used to validate these differences by WB and qRT-PCR to explore the potential nephroprotective mechanism of Rh. Besides, in vitro experiments were conducted to identify the release sources of perforin and IFN-γ. ELISA showed that perforin and IFN-γ were upregulated in CKD patients, and this phenomenon was also corroborated in CKD mice. WB and qRT-PCR data showed that Rh reversed perforin and IFN-γ upregulation, inammatory factor recruitment, and extracellular matrix (ECM) protein upregulation. Results from in vitro experiments demonstrate that the upregulation of perforin and IFN-γ originates from the stress response of CD4 + T lymphocytes (CD4 + cells), CD8 + T lymphocytes (CD8 + cells) and natural killer cells (NK cells), which can be suppressed by Rh. More importantly, the activated STING/TBK1/IRF3 pathway in CKD was also inhibited by Rh. Our data suggest that Rh possesses anti-brotic and nephroprotective effects, which mechanistically are associated with decreased release of perforin and IFN-γ from immune cells, which may be achieved by suppressing the STING/TBK1/IRF3 pathway. Conceptualization, H.X. and J.W.; Methodology and Investigation, X.H., J.W., Y.L., and X.C.; Formal analysis and software, Y.L., X.C. and J.X.; data curation, X.C.; Writing—original draft preparation, X.H. and J.W.; Writing—review and editing, H.X. and J.W., J.H. and C.Z.; Funding acquisition, J.H. and C.Z. All


Introduction
As a global public health issue, CKD poses a tremendous challenge to human welfare and health, and the prevalence of CKD among adults in China has reached 10.8% [1]. The common pathological features of CKD include renal interstitial brosis (RIF), glomerulosclerosis, and mall vascular sclerosis, and the former has received particularly close attention in the eld of CKD because inhibition of RIF is more conducive to parenchymal recovery. The pathological mechanisms of RIF are closely related to multifarious factors, such as in ammatory response, apoptosis, oxidative stress, and imbalance of growth factors and cytokines [2][3]. Therefore, treatment strategies for CKD often revolve around these targets, but effective therapies are still lacking.
Immune cells are widely diffused in the organism and participate in antigen presentation, immune response and regulation, and in the pathogenesis of several immune-related in ammatory diseases. The roles of immune cells in kidney diseases are particularly sophisticated and, in some cases, even contradictory [4][5]. NK cells were classi ed into CD56 bright NK cells and CD56 dim NK cells upon the difference in expression of CD56, both of which can be switched under certain conditions [6]. NK cells are involved in the interstitial lesions of immune nephritis [7], and CD56 bright NK cells in the interstitium can exacerbate CKD by secreting IFN-γ through signal transduction to obtain pro-in ammatory effects [8][9][10]. The released IFN-γ also mediates early injury and inhibits tissue remodeling and brotic processes [11].
Moreover, NK cells can release cytotoxic particles like perforins through extravasation and employ the caspase pathway to inspire apoptosis in target cells [12][13]. CD4 + , CD8 + can also release IFN-γ and perforin [14] and closely correlate with cardiovascular adverse events in CKD patients [15]. Previous studies have focused more on the e cacy of immune cells in kidney transplantation to observe the effects of toxicity and injury. However, effective immune monitoring is still lacking. In this context, we conceived this study and explored the mechanism of immune cells rescue in chronic kidney injury and prevention of renal brosis.
The stimulator of interferon genes (STING) is an important protein that links upstream DNA sensors and downstream factors in the innate immune signaling pathway [16], and its activation can trigger IFN immunity [17][18]. The second messenger cyclic guanosine monophosphate-adenosine monophosphate (cGAMP) directly activates STING and its key synthase cyclic GMP-AMP synthase (cGAS), which subsequently recruits STING to TANK-binding kinase 1 (TBK1) and activates downstream signals, such as activation of IFN regulatory factor 3 (IRF3) which then promotes IFN gene expression [19][20]. Notably, it is unclear regarding the involvement of the STING/TBK1/IRF3 pathway in the progression of renal brosis.
SKS has been used in China for over ten years as a representative class of agents for the treatment of chronic renal failure. Its curative mechanisms include in ammation and antioxidation and also the reparation of cellular injuries [21][22][23][24]. SKS primarily comprises extracts of 4 herbs used in Chinese medicines, namely rhubarb (Rheum O cinale), salvia (Salvia miltiorrhiza), sa ower (Carthamus tinctorius L) and astragalus (Astragalus membranaceous). Research has not been conducted to analyze the single components of SKS and to screen the therapeutic bene ts of the corresponding components.
Rh, one of the main active components of rhubarb, has been recently reported for the treatment of renal brosis by mechanisms involving traditional targets such as oxidative stress and apoptosis [22,25]. Emodin, Chrysophanol were 0.2mg/g and Rh 0.7mg/g, respectively. All the data were analyzed by Chemstation software. The results of HPLC analysis and the molecular formula of Rh are shown in Fig. 1.

Acquisition of clinical specimens
Thirty patients with stage III CKD ( brosis group) who were hospitalized and underwent renal biopsy at the Department of Nephrology, The First Hospital of China Medical University from January 2020 to December 2020 were included in the case group, including three males and three females with a median age of 64 years (range: 47-80 years). Another thirty normal cases of renal specimens from the patients with post-traumatic nephrectomy (the cases with primary kidney diseases were excluded) were selected as the normal control group (normal group), including four males and two females, with a median age of 54 years (range: 37-69 years). Clinical specimens were acquired and mounted in OCT compounds (Tissue-Tek, Torrance, CA, USA) and stored at -80℃ for subsequent analysis.

Animal experimental design
A total of 24 C57BL/6 male mice (age, eight weeks; weight, 22± 2g) were purchased from the animal laboratory of China Medical University. All the mice were kept in a speci c pathogen-free facility and placed under standard laboratory conditions. The light/dark cycle was 12 hours, the temperature was 22± 2℃, and the relative humidity was 55%-60%. After one week of acclimatization, the mice were divided into NC, FA and FA+Rh groups using the random number table method (n=8 per group). CKD mice model induced by FA was established in the FA and FA+ Rh group on the rst day. The FA group was modeled by a single intraperitoneal administration of FA at a concentration of 100 mg/kg, which was prepared with 0.9% sodium bicarbonate injection. While in the FA+Rh group, after the intraperitoneal administration of FA, we gave continuous gavage administration of Rh at a concentration of 120 mg/kg once daily for 14 days. NC and FA groups were given equal amounts of vehicles by the same means. After two weeks, all mice were sacri ced. The left kidney was frozen to isolate RNA or protein, and the right kidney was perfused with PBS and xed in 10% paraformaldehyde buffer for histological examination. The schematic diagram of the animal experiment is shown in Fig. 2.

Biochemical assays
The plasma was collected by tail clamping method with heparin tube to determine the serum creatinine (SCr) and blood urea nitrogen (BUN) of mice on the 14th day after modeling. The samples were analyzed by kits purchased from Nanjing Jiancheng Institute of Biological Engineering (C011-2-1, C013-2-1) according to the method described by the manufacturer.

Renal histopathology
Both kidney tissues from mice and humans were xed in 10% paraformaldehyde buffer for 24 h, followed by gradient alcohol dehydration, para n embedding, and sectioning (4 µm), according to manufacturer's instructions for Hematoxylin and eosin (H&E) and Periodic acid Schiff (PAS) and Masson's Trichrome (Masson) staining. The corresponding kits were purchased from Beijing Solabao (G1120, G1340, G1281). Renal injury including tubular and glomerular deformation and tubular epithelial cell detachment and assessed by semi-quantitative analysis with the following criteria: 0: no abnormality; 1: impairment less than 10%; 2: impairment less than 25%; 3: impairment less than 50%; 4: impairment less than 75%; 5: impairment greater than 75%. Semi-quantitative analysis of brotic area was assessed by Masson staining, i.e., calculating the percentage of blue staining over the entire view.

Immunohistochemistry (IHC)
Para n sections with tissue thickness of 4um were sequentially dewaxed, hydrated, and subsequently placed in xylene and graded alcohol, followed by antigen repair, DAB staining, hematoxylin re-staining, and nally sealed. The immunohistochemistry kit was purchased from MXB Biotechnologies (KIT-9710, DAB-0031), and the dilution ratios of the primary antibody were as follows: α-SMA (1:200, Abcam, ab5694), COL IV (1:200, Abcam, ab182744). Sections were photographed under a 400x light microscope (Olympus, Japan). The integrated optical density (IOD) of α-SMA and COL was measured using Image-Pro Plus 6.0, and the mean density was calculated by IOD/area.

Quantitative real-time PCR analysis (qRT-PCR)
Total RNA was extracted from the kidney by Trizol and enzyme RNA Extraction Kit (Takara, RR820A, China). Plentiful (1ug) RNA was reverse-transcribed by reverse transcription Kit (Takara, RR047A, China). Quantitative real-time PCR was performed subsequently with gene-speci c primers (Biotechnology) and a 7500abi biological system machine. The absolute mRNA number was calculated by comparison with the threshold. Primers for the examined genes are given in Table 1.

Statistical analysis
All statistical analyses were performed using GraphPad Prism 7.0 (GraphPad Inc, La Jolla, CA, USA). For the measurement data, continuous variables were expressed as mean ± standard deviation (x ± s). Nonnormally distributed data were described as median (range). Categorical data were expressed as percentages. Differences in categorical variables among groups were determined by the Chi-square test, with the Yates' correction or the Fisher exact probability test as appropriate. Differences in continuous variables were determined by analysis of variance (unpaired Student's t-test) or non-parametric tests, as appropriate. Multiple comparisons were performed with two-way analysis of variance followed by Student's t-test with post hoc Bonferroni's correction. Comparison between the time points was performed by one-way ANOVA followed by pairwise comparisons with P-value adjustment. For all tests, a P-value (two-tailed) of < 0.05 was recognized as a statistically signi cant statement if the study did not report any data.

CKD patients exhibit features of renal brosis and perforin induction
The pathological staining of the biopsy specimens is shown in Fig. 3A. Generally, In the normal group, the kidney microstructure was normal. No anomalies were observed. Epithelial cells were not dilated or deformed, and there was no interstitial in ammation. In the brosis group, the glomerular morphology was changed, with abnormalities such as tubular dilatation and deformation, epithelial cell detachment, marked interstitial edema, diffuse in ltration of in ammatory cells and excessive collagen deposition. This is consistent with the IHC results of α-SMA (Fig. 3A, D), indicating that these patients all have typical brotic pathology. In addition, we focused on the expression of perforin and IFN-γ in both groups. As shown by ELISA (Fig. 3E, F), perforin secretion was signi cantly increased in the brosis group compared to the normal group. This suggests that there may be a link between renal brosis and perforin and IFN-γ induction.

Rh alleviates renal dysfunction and brosis in FA-induced CKD mice model
Generally, in our experiments, FA successfully caused deterioration of renal function in mice, as evidenced by a signi cant increase in SCr and BUN (Fig. 4A, B). Whereas renal function was preserved after Rh treatment, suggesting that Rh plays a nephroprotective role. H&E and PAS staining (Fig. 4C-E) showed dilated tubules and interstitium, atrophy of tubular epithelium, and increased in ammatory cell in ltration in the FA group. The FA+Rh group showed distinctly fewer histological abnormalities than the FA group but still had a small proportion of aberrant glomeruli and proximal tubular epithelial cell hyperplasia with microprotein tubular patterns visible in the tubular lumen and occasional in ammatory cell in ltration in the interstitium. Masson staining (Fig. 4C, E) showed that collagen bers were most abundant in the FA group, while Rh mitigated this trend. Furthermore, WB (Fig. 5A-C), IHC (Fig. 5D-F) and qRT-PCR (Fig. 5G, H) data demonstrated that Rh attenuates the FA-mediated ECM deposition. To summarize, these data suggest that we successfully established the FA-induced CKD mice model and proved that Rh could preserve renal function was associated with reduced ECM deposition.

Rh suppresses perforin induction in FA-induced CKD mice model
As displayed in Fig. 6A, B, the WB results showed that perforin was expressed at a signi cantly higher level in the FA group than in the FA+Rh and NC groups. The qRT-PCR (Fig. 6C) results were consistent with the WB scenario. These results suggested that high expression of perforin also occurred in CKD mice as in brosis patients. Animal experiments revealed that Rh inhibited FA-mediated induction of perforin.
We also examined the mRNA expression levels of some in ammatory factors, as shown in Fig. 6D-G, the expression of IFN-γ, TNF-α, IL-6 and IL-8 was signi cantly upregulated in the FA group, indicating that FAinduced CKD mice were accompanied by in ammatory disorders, whereas Rh inhibits the upregulation of the aforementioned in ammatory factors. Our results suggest that Rh exerts renoprotective and antibrotic effects on the renal by inhibiting the expression of IFN-γ and other in ammatory factors, as well as perforin.
Rh represses the release of perforin and IFN-γ from CD4 + , CD8 + and NK cells IL-2 is secreted by T lymphocytes and is responsible for stimulating the growth and differentiation of T cells, B cells, and NK cells [27]. Therefore, we performed in vitro experiments using IL-2 treatment of immune cells to determine the source of perforin and IFN-γ release. As shown by qRT-PCR results in Fig. 7A-F, IL-2 enhanced perforin and IFN-γ release from CD4 + , CD8 + and NK cells. But this trend was reversed by Rh. These results demonstrated that the perforin and IFN-γ were derived largely from the CD4 + , CD8 + and NK cells as a response to stress out.

Rh inhibits the immune cells-mediated release of perforin and IFN-γ from immune cells via inhibiting the STING/TBK1/IRF3 pathway activation
Previous studies have shown that nonspeci c agonists of STING can induce IFN-γ release from NK cells [25,[28][29], which suggests that STING signaling is involved in the release of cellular particles by immune cells. Therefore, we observed the effects of Rh treatment on the STING/TBK1/IRF3 pathway in the mice model and compared the WB results, as presented in Fig. 8A-D. We found that STING, p-TBK1 and p-IRF3 proteins were most pronounced in the FA group, while Rh inhibited this trend, suggesting that Rh inhibits the release of perforin and IFN-γ from immune cells perhaps by inhibiting the STING/TBK1/IRF3 immune pathway activation.

Discussion
Renal brosis is a complex and prolonged course characterized by in ltration of in ammatory cells and damage to the renal parenchyma and interstitium, gradually leading to the formation of scar tissue [30]. The aberrant tissue repair allows excessive accumulation of ECM, including collagen (types I, III, and IV), bronectin, and proteoglycans. Meanwhile, activated broblasts are transformed into myo broblasts, and α-SMA is induced to generate constriction tension [31]. It seems di cult to change the irreversible fate of brosis simply by inhibiting the ECM sedimentation.
In recent years, NK cells have attracted increasing attention in the eld of brosis. In 2017, elevated IFN-γ levels of CD56 dim NK cells and CD56 bright NK cells in specimens from patients with renal brosis were found for the rst time to be associated with decreased renal function [8]. In response to cytokine stimulation, CD56 + cells carry cytotoxicity and proliferate in the renal tubule interstitial matrix, which in turn receives activation signals and exerts anti-in ammatory (IFN-γ) effects in CKD, which indicates that this is a pathogenic phenotype. In addition, the activated receptor CD335 and the differentiation marker CD117 were selectively co-expressed on CD56 bright NK cells, and this co-expression is thought to be a source of pro-in ammatory cytokines as well as an intermediate factor in the development of brosis and deterioration of renal function [9,[32][33][34]. Zhang et al. found that NK cells can directly damage renal tubular epithelial cells in vitro and that perforin is also involved in their cytotoxic acquisition and deeply involved in brosis during their evolution [35][36]. Similarly, the accumulation of NK cells in renal tubules leads to irreversible renal brosis [37][38][39]. This is consistent with our data suggesting a correlation between renal brosis and increased perforin and IFN-γ.
It has been indicated that cGAMP and STING can activate NK cells [13,[40][41], and are correlated with CD4 + and CD8 + recruitment [42][43]. In our study, we found that STING/p-TBK1/p-IRF3 pathway protein expression was signi cantly higher in the FA group than in the NC and FA+ Rh groups, as was brotic pathology and ECM protein expression, which indicates that the brotic environment exerts an activating effect on STING and its downstream molecules and that this may be responsible for the recruitment of NK cells, CD8 + and CD4 + cells and the release of perforin and IFN-γ. A groundbreaking study demonstrated that the STING signaling is involved in the progression of renal brosis, but they focused on the upstream cGAS of STING, while the downstream of STING failed to be explored [44]. Together with their ndings, we can draw the conclusion that CKD is accompanied by STING signaling activation, but the exact level of activation may be related to the modeling method and CKD staging. Furthermore, STING activation was also observed in the patients with liver brosis and animal models, while liver brosis and in ammatory responses were alleviated after STING downregulation [45], which is consistent with our experimental data, indicating that there exists a trigger for STING signaling in the brotic environment, and the speci c molecule remains to be further screened. Given the role of STING in regulating multiple in ammatory and immune responses, an increasing number of studies suggest that STING is a promising new target for the treatment of liver, lung, and cardiac brosis [46][47][48]. The experiment in vitro showed that knockdown of STING could reduce TGF-β-promoted ECM deposition, and conversely, STING can accelerate TGF-β-induced brosis. Our study revealed in greater depth the activation of STING and its downstream in renal brosis patients and animal models, which may be responsible for the release of perforin and IFN-γ by immune cells. Therefore, new therapeutic strategies for renal brosis and chronic kidney injury could revolve around the inhibition of immune signaling pathways, including STING and its upstream and downstream, which may be of greater therapeutic signi cance in immune-related nephritis, such as lupus nephritis.
Previous studies have shown that Rh can signi cantly inhibit the compensatory hypertrophy and hardening of the glomerulus and the in ammatory response during the renal brosis progress. Meanwhile, it can reduce the excretion of urine protein and effectively improve the permeability of the glomerular ltration membrane and attenuate the loss of nephron [49][50]. Our study also demonstrated the renoprotective effects of Rh. Further investigation of the regulatory mechanisms of how Rh inhibits the release of cellular granules and in ammatory factors from immune cells may better predict its renoprotective outcome, which would allow us to more effectively utilize immunomodulatory therapeutic strategies to prevent kidney injury. Further investigation of the regulatory network and potential functions of immune cells in RIF requires the development of speci c therapeutic agents capable of acting on the STING/TBK1/IRF3 signaling pathway for patients with brotic kidney disease.

Conclusion
Our data suggest that Rh may suppress perforin and IFN-γ released from misguided immune cells in a pronounced pro-in ammatory response through inhibiting the STING/TBK1/IRF3 pathway activation and thereby attenuate renal brosis to rescue renal function. More importantly, we discovered for the rst time the activation of STING downstream molecules in CKD mice model, which may help provide new strategies for clinical treatment of CKD. Moreover, our study revealed a new mechanism by which SKS exerts its renal protective effect, which would facilitate a more precise dissection of its pharmacological effects.

Con icts of Interest
The author reports no con icts of interest in this work.

Author information
Xin Hunag and Jingyu Wang have contributed equally to this work.

Author Contributions
Conceptualization, H.X. and J.W.; Methodology and Investigation, X.H., J.W., Y.L., and X.C.; Formal analysis and software, Y.L., X.C. and J.X.; data curation, X.C.; Writing-original draft preparation, X.H. and J.W.; Writing-review and editing, H.X. and J.W., J.H. and C.Z.; Funding acquisition, J.H. and C.Z. All authors have read and agreed to the published version of the manuscript.

Data Availability
All the data in this study are available upon request through the correspondence author.

Ethical approval
The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the ethics committee of the First A liated Hospital of China Medical University and implemented after approval (Ethical Lot Number: AF-SOP-07-1,1-01). Informed consent was obtained from all subjects involved in the study, and written informed consent was obtained from the patients to publish this paper.   The schematic diagram of the animal experiment.      Rh represses the release of perforin and IFN-γ from CD4 + , CD8 + and NK cells. (A-F) Perforin, IFN-γ, TNF-α, IL-6 and IL-8 was determined by qRT-PCR and normalized to GAPDH. *P<0.05, vs immune cell groups; # P<0.05, vs immune cell+IL-2 groups.