The Combination of G-CSF/AMD3100 Mobilizes Bone Marrow-Derived Stem Cells to Rescue Mice from Cisplatin-Induced Acute Kidney Injury

Background: Several studies have conrmed that mobilizing bone marrow-derived stem cells (BMSCs) ameliorates renal function loss following cisplatin-induced acute kidney injury (AKI). The aim of this study was to explore whether the combination of G-CSF/AMD3100 exerts benecial effects with respect to renal function recovery in a mouse model of cisplatin-induced nephrotoxicity. Methods: C57BL/6J mice received intraperitoneal injections of G-CSF (200 μg/kg/d) for 5 consecutive days. On the day of the last injection, the mice received a single subcutaneous dose of AMD3100 (5 mg/kg) 1 hour before cisplatin 20 mg/kg injection. 96 hours after cisplatin injection, the mice were euthanized, blood and tissue samples were collected to assess renal function and tissue damage. Cell mobilization was assessed by ow cytometry. Results: Mice pretreated with G-CSF/AMD3100 exhibited longer survival and signicantly lower serum creatinine and BUN levels than mice treated with only G-CSF or saline, exhibited attenuated tissue injury and cell death and enhanced tissue repair and cell regeneration. C57BL/6J mice pretreated with G-CSF/AMD3100 exhibited higher numbers of stem cells in peripheral blood than mice treated with only G-CSF or saline. Furthermore, G-CSF/AMD3100 administration prevented increases in the expression of proinammatory factors, such as IL-6 and TNF-α, and increased the expression of the anti-inammatory factor IL-10. Conclusions: These results suggest that G-CSF/AMD3100 mobilizes bone marrow cells to improve renal function and prevent cisplatin-induced renal tubular injury and that the combination of G-CSF/AMD3100 is superior to G-CSF alone for preventing AKI. This combination may represent a new therapeutic option for the treatment of AKI and warrants further investigation.


Background
The two most common causes of acute kidney injury (AKI) are ischemia/reperfusion (I/R) injury and nephrotoxic agent exposure 1 . Cisplatin is a chemotherapeutic drug widely administered to treat various solid tumors. Nephrotoxicity is a severe side effect of cisplatin administration and often results in acute renal disease 2 . The rate of cisplatin-induced renal damage is 25% to 35%, and the curative effects and nephrotoxicity of cisplatin are dose-dependent. Cisplatin-induced toxic nephropathy is typically characterized by tubular necrosis and apoptosis and greatly limits the use of this agent in clinical settings. Hence, novel therapies for cisplatin-induced AKI that do not compromise its anti-tumor effects are needed.
Stem cell therapy is of great interest as a treatment for AKI. Numerous studies have demonstrated that stem cells can repair damaged tubular cells and attenuate cisplatin-induced AKI 3,4 . Bone marrowderived stem cells (BMSCs) can transdifferentiate into endothelial and epithelial cells 5 . Several studies have demonstrated that mobilizing BMSCs to treat AKI in animal models facilitates signi cant improvements in renal function and enables the repair of renal tissue structural damage 6,7 . Granulocyte colony-stimulating factor (G-CSF) mobilizes BMSCs to sites of renal injury, where these cells transdifferentiate into renal stem cells 6,8 . The CXCR4 antagonist plerixafor (AMD3100) exerts bene cial effects against I/R-induced AKI 9 and myocardial infarction 10 . However, several studies have reported that continuous AMD3100 administration accelerates renal functional decline and exerts adverse effects on renal tissue repair 11,12 .Theiss et al. demonstrated that high concentrations of AMD3100 mobilize BMSCs, whereas low concentrations of AMD3100 do not mobilize BMSCs and do not improve survival in the setting of myocardial infarction 13 . However, Zuk et al. demonstrated that a single dose of 5 mg/kg AMD3100 was less effective at mobilizing BMSCs than a single dose of 1 mg/kg AMD3100 in a rat model of renal I/R 14 . Other studies have demonstrated that the combination of G-CSF and AMD3100 increases BMSC mobilization compared with G-CSF alone 15,16 . Therefore, the usefulness and optimal dosage of AMD3100 remain controversial. In addition, few studies have examined the combination of G-CSF/AMD3100 as a therapy for cisplatin-induced AKI.
Based on the above findings, we speculated that combination therapy with G-CSF/AMD3100 might be more effective for recruiting BMSCs into renal tissue to repair and ameliorate cisplatin-induced AKI than either therapy alone. In addition, the mechanisms underlying the renoprotective effects of BMSCs require further elucidation. In the present study, we rst demonstrated that G-CSF/AMD3100 mobilizes BMSCs into the peripheral blood and rescues mice from cisplatin-induced AKI. Second, we demonstrated that mobilization of BMSCs by G-CSF/AMD3100 decreases tubular cell apoptosis and promotes proliferation. The protective effects of this combination are associated with reductions in the levels of putative biomarkers of renal injury and in ammation.

Ablation of BM stem cells in mice
To con rm BM (bone marrow) stem cell mobilization following G-CSF/AMD3100 administration, mice were irradiated with 2 separate doses of 4.5 Gy of whole-body γ-radiation at a 2-hour interval for bone marrow ablation (BMA). Mice were randomly assigned to the following groups: irradiation + saline group, irradiation + G-CSF/AMD3100 group, non-irradiation + saline group, and non-irradiation + G-CSF/AMD3100 group. Beginning 1 day after irradiation, G-CSF/AMD3100 was administered as described in the above-mentioned treatment protocol. Control mice that were not subjected to irradiation were injected with an equivalent dose of saline (n=6 per group). The mice were euthanized at 96 hours after treatment, and peripheral blood samples were collected for subsequent ow cytometry (FCM) analyses.

Peripheral blood ow cytometric analysis
Mice were anesthetized with chloraldurat, and peripheral blood samples were collected for FCM analyses.
Red blood cells were lysed using Lysing Buffer (Sigma). The remaining cells were labeled with APClabeled rat anti-mouse CD34, CD133 and CD44 antibodies and PE-labeled rat anti-mouse CXCL4 and c-kit antibodies (all from BD Pharmingen).
Measurement of blood urea nitrogen and serum creatinine levels Serum samples were obtained from mice 4 d after cisplatin injection. Blood urea nitrogen (BUN) and creatinine levels were subsequently measured using an autoanalyzer (Hitachi 7150 Auto-analyzer; Hitachi, Tokyo, Japan).

Tissue processing and histopathological scoring
After the mice were euthanized, kidney specimens were xed immediately in 10% buffered formalin for 24 hours at room temperature and then embedded in para n, and some kidney specimens were embedded in OCT compound for freezing, then were rapidly frozen in liquid nitrogen. Kidney tissues were cut into 3μm thick sections and were stained with PAS staining for immunohistochemical analysis. Tubulars injury was diagnosed based on the presence of tubular epithelial necrosis, cast formation, tubular dilatation and brush border loss. Renal injury severity was scored in a blinded fashion as described in a previous study 17 based on the percentage of tubule lesions in ten randomly selected, non-overlapping elds (magni cation, 200x) as follows: 0, 0%; 1, ≤10%; 2, 11-25%; 3, 26-45%; 4, 46-75%; and 5, 76-100%.
Immuno uorescence and immunohistochemical (IHC) studies Mice were euthanized, and their kidneys were perfused with precooled PBS via the left ventricle. Specimens were embedded in para n and were cut into 3µm thick sections and processed for immunostaining. IHC labeling was performed to identify Ki-67-positive cells, which were counted by a blinded investigator in 20 randomly selected cortical and outer medullary (OSOM) elds at a magni cation of 400×. In addition, kidney tissue sections were subjected to immuno uorescence staining for BrdU using mouse anti-BrdU monoclonal antibodies (Roche) and Dylight 594-conjugated secondary antibodies (Amyjet) to evaluate tubular epithelial cell proliferation. The number of BrdU-positive cells was determined by counting the numbers of positive nuclei in 20 randomly selected, non-overlapping cortical and OSOM elds at a magni cation of 400×.

Apoptosis assay
Apoptosis was evaluated using a terminal deoxynucleotidyl transferase dUTP nick end-labeling (TUNEL) assay kit (Roche, Indianapolis, IN, USA). Brie y, C57BL/6J mouse kidney sections were depara nized, rehydrated, digested with proteinase K and labeled with a TUNEL reaction mixture for 60 minutes at 37°C.
The TUNEL-positive cells corresponding to apoptotic tubular epithelial cells were counted in 20 randomly selected cortical elds at high-power magni cation (x400). All tissue sections were viewed and labeled by a blinded examiner Western blot analysis A 50µg quantity of total renal cell lysate was separated by SDS-PAGE and transferred onto polyvinylidene di uoride (PVDF) membranes (Millipore; Billerica, MA, USA). Western blotting was performed as described in our previous study 18 . The following primary antibodies were used: PCNA (Ruiyng, China), Bcl-2 (Santa Cruz Biotechnology), and Bax (Santa Cruz Biotechnology). Anti-mouse or goat HRP-conjugated secondary antibodies (Amyjet) were used to detect protein using an ECL Assay Kit (Bipec Biopharma). β-Tubulin or β-actin was used as an internal control. Band intensity was quanti ed using ImageJ software (1.44 P).

Quantitative RT-PCR
Total RNA was extracted from tissues with TRIzol reagent (TaKaRa) according to the manufacturer's instructions. RNA reverse transcription was conducted using a PrimeScript ™ RT Reagent Kit (TaKaRa).
PCR enzymes and master mixes (DBI Bioscience) were used for real-time PCR, along with primers speci c for mouse GAPDH, TNF-α, IL-6, IL-10, Kim1, and Ngal. Relative expression levels were normalized to GAPDH and calculated using the 2 −ΔΔCt method. The primer sequences were as follows: Results are presented as the mean ± SEM. For normally distributed data, differences within groups were assessed by ANOVA, and differences between groups were assessed using Tukey's post hoc test. Student's t-test was performed to analyze differences between two groups. For data that were not normally distributed, differences within groups were evaluated using the Kruskal-Wallis Htest, and differences between groups were evaluated using the Mann-Whitney U-test. All tests were two-tailed, and a P-value <0.05 was considered signi cant.

Results
Pretreatment with the combination of G-CSF/AMD3100 improves survival in cisplatin-treated mice BMSCs can repair injured renal tubules, and G-CSF can mobilize BMSCs into the peripheral blood.
Therefore, we examined the ability of pretreatment with G-CSF/AMD3100 to prolong survival in cisplatintreated mice. As shown in Figure 2A, compared with saline pretreatment, G-CSF and G-CSF/AMD3100 pretreatment improve survival rate in cisplatin-treated mice. Only 16.7% of mice survived up to 10 days following cisplatin treatment, whereas 58.3% of mice pretreated with G-CSF/AMD3100 survived up to 10 days after cisplatin injection (P<0.01). Furthermore, G-CSF/AMD3100 group had better survival rate than the G-CSF alone group (P<0.05). These results indicate that GCSF/AMD3100 protects against cisplatininduced renal toxicity.
The combination of G-CSF/AMD3100 ameliorates cisplatin-induced renal functional deterioration Cisplatin causes renal tubular damage and induces acute renal failure. We measured serum BUN and creatinine levels to assess the severity of renal dysfunction in each group. As shown in Figure 2B and C, after 4 days of cisplatin treatment, serum creatinine and BUN levels were signi cantly increased in saline-treated mice (P<0.05 vs. untreated with cisplatin). By contrast, mice that received G-CSF or G-CSF/AMD3100 injections exhibited signi cantly lower serum creatinine and BUN levels than mice that were treated with saline. Moreover, mice pretreated with G-CSF/AMD3100 exhibited lower BUN and creatinine levels than mice in G-CSF-treated (P<0.05). These results suggest that G-CSF and G-CSF/AMD3100 pretreatment improve renal function in cisplatin-treated mice and that the combination of G-CSF/AMD3100 improves renal function more effectively than G-CSF alone.
The combination of G-CSF/AMD3100 ameliorates renal tubule lesions To evaluate the effects of G-CSF/AMD3100-induced BMSC mobilization on cisplatin-induced renal impairment, we subjected the kidneys of cisplatin-treated mice to Periodic Acid-Schiff stain (PAS) staining ( Figure 3A-H). Histological scoring was based on the typical microscopic features of acute tubular damage, including extensive tubular necrosis, cast formation, tubular dilatation, and brush border loss. Cisplatin-treated mice exhibited more severe tissue injury than G-CSF and G-CSF/AMD3100-treated mice, and G-CSF/AMD3100-treated mice exhibited signi cantly lower histopathological scores than G-CSF-treated mice (P<0.05; Figure 3I). These results suggest that G-CSF/AMD3100 pretreatment protects against cisplatin-induced renal damage.
The combination of G-CSF/AMD3100 mobilizes BMSCs CD34 + and CD133 + cells are widely recognized as hematopoietic stem cell (HSC) and endothelial progenitor cell (EPC) biomarkers, respectively. In this study, BMSCs in peripheral blood samples from C57BL/6 male mice were analyzed by ow cytometry. The CD34 + , CD133 + staining results were depicted in Figure 4. As expected, G-CSF (*P<0.05 vs. cisplatin) and G-CSF/AMD3100 (**P<0.01 vs. cisplatin) treatment enhanced circulating CXCR4 + cell mobilization. In addition, the G-CSF/AMD3100 group exhibited a much larger CXCR4 + cell population than the G-CSF treated group (P<0.01) ( Figure 4A). The CXCR4 + CD34 + and CXCR4 + CD133 + cell populations were also harvested and identi ed by double-staining ( Figure 4B,C). G-CSF/AMD3100 treatment increased the numbers of CXCR4 + CD34 + and CXCR4 + CD133 + cells in the peripheral blood by 1.8-fold and 1.9-fold, respectively, compared to G-CSF treatment. The percentages of CXCR4 + CD34 + and CXCR4 + CD133 + were signi cantly increased in the G-CSF/AMD3100treated and G-CSF-treated groups compared with the cisplatin-treated group. We also measured the numbers of CXCR4 + CD44 + and c-Kit + peripheral blood mononuclear cells (Supplemental Figure 1). G-CSF/AMD3100 treatment increased the numbers of CXCR4 + CD44 + and c-Kit + cells in the peripheral blood compared with G-CSF or saline treatment. These ndings suggest that G-CSF/AMD3100 effectively mobilizes stem cells into the peripheral blood and facilitates tubule repair and regeneration.

Bone marrow ablation prevents BMSC mobilization by the combination of G-CSF/AMD3100
To further determine whether G-CSF/AMD3100 administration mobilizes BMSCs into the peripheral blood, we performed bone marrow ablation (BMA) via irradiation to prevent BMSC development and mobilization. As shown in Supplemental Figure 2, when the bone marrow was irradiated, the percentages of CXCR4 + CD34 + and CXCR4 + CD133 + cells were not different between the irradiation + saline group and irradiation + G-CSF/AMD3100 groups (P>0.05); however, these percentages were signi cantly higher in the non-irradiation + G-CSF/AMD3100 group than in the non-irradiation + saline and irradiation + G-CSF/AMD3100 groups (P<0.01). These results further proved that BMSCs were mobilized by G-CSF/AMD3100.
The combination of G-CSF/AMD3100 enhances tubular epithelial cell regeneration To determine whether G-CSF/AMD3100 pretreatment facilitates tubular epithelial cell proliferation and regeneration, we utilized Western blotting to determine whether the combination of G-CSF/AMD3100signi cantly increases PCNA protein expression compared with cisplatin treatment (P<0.05). G-CSF/AMD3100-treated mice exhibited higher levels of PCNA protein expression than G-CSFtreated mice ( Figure 5A and B). We also assessed the effects of G-CSF/AMD3100 treatment on tubular  Figure 5H and I). G-CSF/AMD3100 pretreatment resulted in a 4-fold increase in tubular cell proliferation compared with saline treatment (P<0.01) and a 1.5-fold increase in cell proliferation compared with G-CSF treatment ( Figure 5H-L). These results suggest that G-CSF/AMD3100 pretreatment promoted tubular epithelial cell regeneration to a greater extent than G-CSF pretreatment.
The combination of G-CSF/AMD3100 promotes the protective effect of BMSCs against cisplatin-induced injury To determine the relationship between cisplatin nephrotoxicity and apoptosis, we performed Western blotting to quantify Bcl-2 and Bax protein expression. Bcl-2 expression was notably lower in the cisplatintreated group than in the other three groups and was signi cantly lower in the G-CSF-treated group than in the G-CSF/AMD3100-treated group (P<0.05, Figure 6A and C). By contrast, Bax expression was increased in the cisplatin-treated group; however, G-CSF and G-CSF/AMD3100 administration decreased Bax expression, and a signi cant difference in Bax expression was noted between G-CSF-treated mice and G-CSF/AMD3100-treated mice based on band intensities (P<0.05, Figure 6A and B). Furthermore, to evaluate renal tubular epithelial cell (RTEC) damage, we examined the percentages of apoptotic RTECs via TUNEL assay. As expected, both G-CSF-treated mice and G-CSF/AMD3100-treated mice exhibited signi cantly decreased proportions of TUNEL + cells compared with cisplatin-treated mice. Interestingly, G-CSF/AMD3100-treated mice exhibited signi cantly fewer TUNEL + cells than G-CSF-treated mice ( Figure  6D-G). To con rm that the combination of G-CSF/AMD3100 attenuates renal epithelial injury, we measured the levels of kidney injury molecule-1 (Kim-1) and neutrophil gelatinase-associated lipocalin (Ngal) by qRT-PCR. Kim1 and Ngal mRNA expression levels were 2-fold and 1.9-fold lower, respectively, in G-CSF/AMD3100-treated mice than in cisplatin-treated mice. G-CSF/AMD3100 treatment also decreased Kim1 and Ngal mRNA expression to a greater extent than G-CSF administration alone, although these differences were not statistically signi cant (P=0.065, P=0.058, respectively) (Supplemental gure 3).
These results indicate that G-CSF/AMD3100 administration prevents cisplatin-induced RTEC apoptosis.

G-CSF/AMD3100-induced BMSC mobilization impact the immune response
Recent studies have demonstrated that MSC administration protects against I/R injury by signi cantly downregulating the expression of pro-in ammatory cytokines such as IL-1b, TNF-α, IFN-γ 19,20 . We examined the expression of the pro-in ammatory cytokine IL-6, which increases abruptly in the setting of cisplatin-induced nephrotoxicity. G-CSF/AMD3100 treatment signi cantly decreased IL-6 mRNA expression (P<0.01) to a greater extent than G-CSF treatment alone (P<0.05) ( Figure 7A). We observed similar changes in TNF-α levels, although there was no signi cant difference in TNF-α levels between G-CSF-treated mice and G-CSF/AMD3100-treated mice (P=0.065) ( Figure 7B). However, cisplatin-treated mice exhibited much higher mRNA expression levels of the anti-in ammatory cytokine IL-10 than control mice. Moreover, IL-10 mRNA expression levels were increased in G-CSF/AMD3100-treated mice compared with cisplatin-treated mice (P<0.01) ( Figure 7C). These data indicate that G-CSF/AMD3100 pretreatment ameliorates cisplatin-induced renal injury.

Discussion
Stem cells can facilitate renal repair and regeneration after AKI. Recent studies have focused speci cally on BMSCs mobilization to sites of renal injury to facilitate tissue repair and regeneration [21][22][23] . However, the repair e ciency of BMSCs is limited due to the low survival and migration e ciency of BMSCs; only a small proportion of intravenously injected BMSCs migrate successfully to sites of renal tissue damage 24 . Thus, the viability of this therapy in clinical settings is limited. In the present study, we explored the effects of the combination of G-CSF/AMD3100 in a mouse model of cisplatin-induced AKI and observed that G-CSF/AMD3100 pretreatment exerted marked renoprotective effects against cisplatin-induced nephrotoxicity.
G-CSF is considered the gold standard for mobilizing hematopoietic and mesenchymal lineage cells from the BMSC population into the circulation and promoting the proliferation and differentiation of mobilized BMSCs 25,26 . AMD3100 has been approved by the United States Food and Drug Administration and the EU for hematopoietic stem cell mobilization in lymphoma and multiple myeloma patients who have failed hematopoietic stem cell-mobilization therapy with G-CSF alone 27,28 . In the present study, G-CSF/AMD3100-treated mice exhibited higher numbers of BMSCs in the peripheral blood than G-CSFtreated mice. These ndings are consistent with those of previous studies demonstrating that the combination of G-CSF/AMD3100 mobilized BMSCs more effectively than G-CSF alone [29][30][31] . The role of the CXCR4/SDF-1 pathway in stem cell mobilization and progenitor cell tra cking has recently been studied, revealing a vital role of this chemokine receptor in regulating stem cell fate. CXCR4 is expressed on the surface of MSCs, and SDF-1 (stromal cell-derived factor-1α) is expressed on the surface of bone marrow stromal cells 32,33 . To increase BMSC mobilization into the peripheral blood, we used AMD3100 to disrupt the interactions between cell receptors and their ligands in speci c microenvironments (Supplemental gure 4). We observed that G-CSF/AMD3100-treated mice exhibited a higher proportion of CXCR4 + cells in the peripheral blood than mice in another groups, consistent with a previous study that demonstrated that combining G-CSF with AMD3100 dramatically improved HSC mobilization into the peripheral blood 15 . Several studies have demonstrated that acute AMD3100 administration enhances tissue repair 10,34 , whereas continuous AMD3100 administration has adverse effects on tissue regeneration 13,14 . AMD3100 exerts rapid cell-mobilization effects that peak within 1-3 hours of administration in mice 31,35 . Therefore, we elected to administer ADM3100 subcutaneously 1 hour before cisplatin injection. The combination of G-CSF/AMD3100 synergistically enhanced BMSC mobilization into the peripheral circulation.
Previous studies have demonstrated that G-CSF signi cantly ameliorates ischemia/reperfusion-and cisplatin-induced renal injury 22,36 . Wu et al. 37 demonstrated that subcutaneous administration of 6mg/kg AMD3100 prevents ischemic acute kidney injury. In the present study, we observed that the combination of G-CSF/AMD3100 ameliorated cisplatin-induced nephrotoxicity than G-CSF, ndings supported by our analyses of plasma creatinine and BUN levels and renal tissue morphology. The mechanism underlying stem cell-based therapy involves the transdifferentiation of stem cells into renal cells and subsequent release of factors that increase cell survival and proliferation, exert anti-in ammatory and anti-apoptotic effects and modulate the immune response 38 . Zuket al. 14 reported that AMD3100 administration ameliorates I/R-induced kidney injury characterized by leukocyte in ltration and pro-in ammatory chemokine/cytokine expression. In the present study, we failed to elucidate the mechanism by which BMSCs transdifferentiate into RTECs. However, we found that pretreatment of cisplatin-treated mice with G-CSF/AMD3100 or G-CSF could contribute to promote renal repair by promoting cell proliferation and decreasing apoptosis, as indicated by the increased protein expression levels of several cell proliferationrelated markers, such as PCNA, Ki-67 and BrdU. In addition, signi cantly decreased protein expression levels of Bax and notably enhanced protein expression levels of Bcl-2 were observed in cisplatin-treated mice pretreated with G-CSF/AMD3100 or G-CSF. Furthermore, these effects were greater in mice pretreated with G-CSF/AMD3100 than in those pretreated with G-CSF alone. TUNEL assays of DNA fragmentation further con rmed the positive effects of G-CSF/AMD3100 on BMSC mobilization. Taken together, our ndings are consistent with those of previous studies demonstrating that mesenchymal stem cells treatment could alleviate the cisplatin-induced renal injury and I/R-induced renal failure, at least in part, by attenuating apoptosis and tubular injury and by promoting tubular regeneration 14,[39][40][41] .
Numerous studies have demonstrated that BMSCs exert potent immunomodulatory effects in vitro and in vivo [42][43][44] . In the current study, we observed the levels of IL-10 mRNA expression were high in cisplatintreated animals pretreated with G-CSF/AMD3100 compared with animals pretreated with G-CSF alone.
Milwidet al. 43 reported that BMSC-secreted IL-10 contributed to the attenuation of severe cisplatininduced AKI. Moreover, the levels of the pro-in ammatory cytokines IL-1ß, IL-6 and TNF-α are increased in renal tubular cells during cisplatin-induced AKI 45,46 . Our results for IL-6 and TNF-α mRNA expression are consistent with these ndings. We observed signi cantly lower levels of IL-6 and TNF-α following G-CSF/AMD3100 or G-CSF pretreatment, consistent with the results of Overath et al. 47 , who observed that adipose-derived MSCs pretreated by exposure to hypoxia signi cantly decreased pro-in ammatory cytokine levels and signi cantly attenuated cisplatin-induced renal injury in mice.

Conclusions
Some reagents can prevent kidney damage caused by toxic molecules. Our ndings provide some evidences for the potential of combination of G-CSF/AMD3100 as a therapy protocol for cisplatininduced AKI. This combination maybe enhance the migration and homing of BMSCs to sites of renal tissue damage to produce bene cial effects. All experiments were performed using C57BL/6J mice and were conducted according to the Huazhong University of Science and Technology Guide for the Care and Use of Laboratory Animals. All experimental animal procedures were approved by the Institutional Animal Care and Use Committee of Huazhong University of Science and Technology, Wuhan, China.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analyzed during this study are included in this article.

Competing interests
The authors declare no con icts of interest. Funding