TRPV4-mediated Ca2+ inux is essential to glomerular endothelial inammation in sepsis associated acute kidney injury

Sepsis-associated acute kidney injury (S-AKI) is a frequent complication of critical patients and is associated with high morbidity and mortality. The glomerular endothelial cell injury is the main characteristics during S-AKI. Ca 2+ inux is a key step in the establishment of endothelial injury. Transient receptor vanilloid subtype 4 (TRPV4) ion channels are permeable to Ca 2+ and are widely expressed in endothelial cells. However, the role of TRPV4 on glomerular endothelial inammation in S-AKI has remained elusive. overload and NF-κB/IRF-3 activation. These discoveries may provide novel pharmacological strategies for the treatment of glomerular endothelial dysfunction and kidney injury during endotoxemia, sepsis, and other inammatory diseases.


Introduction
Sepsis is a life threatening clinical syndrome characterized by organ dysfunction caused by maladaptive host response to infection. Acute kidney injury (AKI) is a frequent complication of sepsis, and ranks as the tenth cause of mortality in the US, especially in patients admitted to the intensive care unit (ICU) (12).
Moreover, 35-65% of ICU patients suffered from AKI (24,28). The impairment of the glomerular ltration barrier (GFB), which is composed of glomerular endothelium, the glomerular basement membrane (GBM) and the podocyte layer is the most important clinical feature of AKI (29). The glomerular endothelium is a semipermeable membrane formed by glomerular endothelial cells (GECs), which are a cell type with round shape and fenestrations(26).Glomerular endothelial cells are susceptible to be damaged by circulating pathogens, endotoxins, proin ammatory cytokines, and reactive oxygen. Previous studies have shown that glomerular endothelial in ammation occurs in the early stage of sepsis (37). Thus, the damage of GECs seems to play a pivotal role in S-AKI.
TRPV4, a Ca 2+ permeable cation channel, can be activated by diverse stimuli including moderate heat, cell swelling and endogenous and exogenous ligands (18,38,39). Our previous study demonstrated that TRPV4-mediated Ca 2+ rise induces the increase of nitric oxide(NO) production in the mesenteric artery endothelial cells, resulting in a restoration of ow-induced dilation in mesenteric arteries of aged rats (10). In airway epithelial cells, TRPV4-de cient mice display exacerbated ventilatory changes and recruitment of polymorphonuclear leukocytes into the airways after LPS challenge (1). TRPV4 can also regulate vascular permeability in acute lung injury (5). Excessive activation of TRPV4 causes endothelial detachment from the basement membrane, which leads to disruption of the pulmonary endothelial barrier, pulmonary edema, and alveolar ooding (3). Furthermore, pharmacological inhibition of TRPV4 channels reduce cytokine production, restore endothelial function and increase survival in septic mice (9).
Despite the association of TRPV4 function with key characteristics of sepsis, including systemic in ammatory responses, endothelial dysfunction and survival rate, the molecular mechanism of TRPV4 channels involving in the development of S-AKI is unknown. Here we investigated the contribution of TRPV4 channels to the development of S-AKI, and determined whether pharmacological blockade or knock down of TRPV4 exerts a protective effect. In this study, we provide the rst evidence that blocking TRPV4 diminishes LPS-induced cytosolic Ca 2+ -elevations, which are essential for glomerular endothelial in ammation and barrier function. Furthermore, inhibition of TRPV4 restrains NF-κB and IRF-3 nuclear translocation. Clamping intracellular Ca 2+ mimics the LPS response seen in the absence of TRPV4. In vivo, pharmacological blockade or knock down of TRPV4 reduced the in ammatory response of glomerular endothelial cells, increased survival rate and improved renal function in cecal-ligation-andpuncture (CLP)-induced sepsis without altering renal cortical blood perfusion and NO production. Since TRP channels are highly attractive drug targets (25), these discoveries may provide novel pharmacological strategies for the treatment of glomerular endothelial in ammation and kidney injury during endotoxemia, sepsis, and other in ammatory diseases.

Material And Methods
Animals 200× to evaluate tissue damage. The grader of the histology slides was examined blindly by a pathologist. The percentage of histological changes in the cortex and medulla was scored using a semiquantitative scale designed to evaluate the degree of the necrosis in tubular and glomerular areas, tubular vacuolization and cast formation on a ve point scale based on injury area of involvement as previous study (7). The scale is as follows: 0≤10%; 1=10-25%; 2=25-50%; 3=50-75%; and 4=75-100%.

Immunohistochemistry
The kidney tissues were xed with 4% formalin, embedded in para n and cut into slices. The sections were incubated with the primary antibodies, ICAM1 (1:1000, Abcam, USA), (E-selectin, 1:100; Beyotime, Haimen, China). Every mouse kidney sample was stained in one slide, and ten viewing elds of the cortex in each slide at 200x magni cation were randomly captured for next calculation. The positive cell count of one viewing eld in glomerulus was calculated by Image J.

Immuno uorescence staining
Mice were anesthetized with iso urane and perfused through the left ventricle with normal saline followed by 4% paraformaldehyde. After perfusion, kidney tissues were dissected and cut into sections at 10 µm with Leica freezing microtome. Slices were washed with 0.01 M of PBS and blocked in 2% BSA for 1 hours at room temperature. After blocking, slices were incubated with primary antibody: anti-CD68(1:100, Abcam, USA ), anti-NF-κB(1:100, Cell Signaling, USA), anti-IRF3(1:100, Beyotime, China) followed by the secondary antibody (1:200, Beyotime,China). The nuclei were stained with DAPI. Finally, the uorescence was detected using light microscopy and uorescence microscopy. MRGECs were seeded onto circular coverslips and xed with 4% formaldehyde for 10 min, followed by permeabilization with 0.1% Triton X-100 in PBS. The cells were blocked with 2% BSA at room temperature for 60 min and then incubated with primary anti-VE-cadherin(1:100, R&D systems Inc.,MN, USA), anti-NF-κB(1:100, Cell Signaling, USA), anti-IRF3(1:100, Beyotime, China) at 4°C overnight. Following incubation, samples were washed with PBS and incubated with secondary antibody. The coverslips were mounted using 90% glycerol in PBS, and the uorescence was detected with light microscopy and uorescence microscopy.

NO detection
The levels cellular NO were measured by a commercial NO assay kit (Beyotime, Haimen, China), according to the manufacturer's instructions. Brie y, the cells were lysed and centrifuged. RLU/OD was detected on 540nm by spectrum detection. The NO contents were calculated by constrast with the standard curve.

ROS detection
The levels of kidney and cellular ROS production were evaluated using standardized methods by ROS assay kit, according to the manufacturer's instructions(Beyotime, Haimen, China). Production of ROS was evaluated by changes in the uorescence intensity resulted from oxidation of the intracellular uoroprobe 5-(and -6)-chloromethyl-2',7'-dichlorodihydro uorescein diacetate (CM-H2DCF-DA, Molecular Probes, Eugene, OR, USA). ROS contents were observed or detected by confocal or uorescence microscopy.
The measurement of blood urea and creatine.
Mice blood were collected before kidney harvest and kept at room temperature for 2 hours. Serum was isolated by 3500g for 10 minutes centrifugation. The serum was detected by ELISA blood urean and creatine assay kit(Westang Biotech, Shanghai,China) according to the manufacturer's instruction.

Doppler hemodynamics
The PeriCam PSI Zoom System was used to collect the blood ow perfusion data. Kidney blood ow was detected by PeriCam PSI Zoom 24 hours after CLP. The mice were anesthetized by iso urane. Then kidneys were exposed and kidney blood ow was detected by PeriCam PSI Zoom.

Ca 2+ image
Cytoplasmic calcium concentration was monitored using Zeiss 710 confocal confocal laser system. siRNA transfection siRNA and negative control were synthesized by GenePharma (Shanghai, China),siRNA sequences were as bellows: 5'-AAGACUUGUUCACGAAGAAAU-3'; 5'-UUCUUCGUGAACAAGUCUUUG-3'. siRNA transfection in vitro was performed by lipo8000(Beyotime, Shanghai, China), according to the manufactural instructions. The siRNA transfection in vivo was performed as follows: The TRPV4 siRNA or scrambled siRNA (5 mg/kg) was diluted in 50 uL of EntransterTM-in vivo RNA transfection reagent (Engreen Biosystem Co., Ltd., Beijing, China) and 10% glucose mixture according to the manufacturer's instructions. A volume of 50 uL siRNA or siTRPV4 mix was continuously subjected to mice by intravenous injection 3 days before CLP according to the reagent's instructions

Data analysis
All data are represented as mean ± SD. Statistical analysis between two groups was performed using student t test, and one-way ANOVA was used to compare the signi cance among three or more groups.
Bonferroni method was used to evaluate the signi cance conservatively. The calculations and data processing were performed using Sigmaplot 14.0.
Firstly, we measured whether LPS-induced Ca 2+ entry was dependent on TRPV4. The MRGEC was stimulated by LPS after pretreated with HC or transfected with TRPV4siRNA. As expected, MRGEC pretreated with both HC and TRPV4siRNA displayed a signi cant reduction in LPS-induced cytosolic [Ca 2+ ] entry, relative to control group( Fig. 1a-f). As TLR4 recognizes LPS and drives the in ammatory response, which contributes to sepsis. Next, we reasoned that if other TLR ligand (TLR 1/2 ligand) induced Ca 2+ entry was dependent on TRPV4. Consistent with these observation, a robust rise in cytosolic [Ca 2+ ] was observed after stimulation with CU-T12-9 (TLR 1/2 ligand) (10µM), but we failed to see the cytosolic [Ca 2+ ] reduction after treated by HC (Fig. 1g, h). Overall, these results clearly demonstrate that TRPV4 controls the Ca 2+ -entry triggered by LPS. Considering that a Ca 2+ entry mechanism involving store-operated Ca 2+ entry (SOCE), and heteromeric TRPV4-C1 channels participate in mediating SOCE (22), we wondered that if blocking TRPV4 affects SOCE in MRGEC. In the next step, thapsigargin(TG) was used to delete Ca 2+ store in endoplasmic reticulum, we found that inhibition of TRPV4 did not affect SOCE occurred (Fig. 1i, j).These results demonstrated that TRPV4 regulates the Ca 2+ -entry triggered by LPS, but not the Ca 2+ -entry triggered by CU-T12-9 or SOCE.

LPS-induced phosphorylation and translocation of NF-κB and IRF-3 is regulated by TRPV4.
LPS-initiated signaling triggers multiple post-translational modi cations in the p65 subunit of NF-κB (NF-κB p65) (36). The phosphorylation of NF-κB p65 is associated with nuclear translocation(42) and transcriptional transactivation(40), respectively. Likewise, phosphorylation at IRF3 is crucial for the induction of Type I IFNs (19). Next, we examined if LPS-induced phosphorylation of either NF-κB p65 or IRF3 was impaired after inhibition or knockdown of TRPV4. LPS stimulation (1µg/ml) induced the phosphorylation and translocation of NF-κB and IRF-3, and this effect was signi cantly decreased compared to HC treatment or TRPV4 transfection(( Fig. 2. g, h, j, k, l, n). As a proximal readout, degradation of IκBα may indicate the degree of signal transduction upstream of NF-κB p65. We also measured the degradation of IκBα, we did not observe signi cant change in IκBα protein levels after HC treatment or TRPV4siRNA transfection, suggesting that TRPV4 regulates LPS signaling downstream of IκBα( Fig. 2. g, i, k, m). Thus, in the absence of TRPV4, LPS-induced phosphorylation of the key transcription factors, NF-κB and IRF-3 are signi cantly compromised.
TRPV4 inhibition reduced glomerular endothelial in ammation and enhanced endothelial cell barrier function.
To determine if inhibition of TRPV4 attenuated LPS-induced endothelial in ammation, MRGEC was pretreated with HC before stimulated by LPS. Our data showed that LPS caused signi cant increase of endothelial adhesion molecules including VCAM1, ICAM1 and E-selectin, which was inhibited by HC treatment (Fig. 3a, b). Next, phalloidin and VE-cadherin were stained to assess cytoskeletal remodeling and adherens junction integrity in endothelial cells, respectively. These results showed that LPS induced massive formation of actin stress bers and intercellular gaps due to cell contractions in MRGEC. These changes were largely reduced by HC treatment (Fig. 3c). During sepsis, the production of ROS injures the endothelium and damages extracellular structures such as cell membranes and glycocalyx directly. In addition, it also can impair the endothelium-dependent vasoreactivity (4).The DCFH-DA detection showed that LPS increases ROS production strongly, which was suppressed by HC treatment (Fig. 3d).It is general accept that the NO system is a major contributor to endothelial injury, thus we detected the expression of eNOS in the next step. It can be seen clearly that LPS signi cant decreased phosphorylation of eNOS, while HC treatment did not increase the level of p-eNOS as well was the production of NO (Fig. 3eg),suggesting that TRPV4 inhibition reduced glomerular endothelial in ammation and enhanced cell barrier function without altering the NO production.
Clamping intracellular Ca 2+ mimics the defects in LPS response seen in the absence of TRPV4.
It is reported that stimulation of the T cell receptor (TCR), but not tumor necrosis factor receptor (TNFR), induces a Ca 2+ -dependent phosphorylation of NF-κB p65 (20). It is, however, unknown if LPS-induced phosphorylation of NF-κB p65 requires intracellular Ca 2+ or [Ca 2+ ] i elevations in MRGEC. We depleted and clamped [Ca 2+ ] i by loading MRGEC with the high a nity membrane permeable Ca 2+ -chelator BAPTA-AM and then stimulated them with LPS. Preventing [Ca 2+ ] i elevations decreased the phosphorylation and translocation of NF-κB p65 induced by LPS stimulation (Fig. 4a-c). We then measured whether LPSinduced increase of endothelial adhesion molecules was diminished by clamping intracellular Ca 2+ . We found that loading MRGEC with BAPTA-AM caused decrease of VCAM1, ICAM1 and E-selectin (Fig. 4d,e).
As we veri ed before that Ca 2+ -entry triggered by CU-T12-9 was not dependent on TRPV4, we want to investigate whether TRPV4 participates in CU-T12-9-induced endothelial in ammation. Consistent with previous results, our data showed that CU-T12-9 inhibition of TRPV4 has no signi cant impact on CU-T12-9-induced endothelial in ammation (Fig. 4.f-h).
Inhibition of TRPV4 reduced glomerular endothelial in ammation and retained translocation of NF-κB and IRF-3 in CLP-induced sepsis.
To further investigate whether TRPV4 contribute to glomerular endothelial in ammation in vivo, HC (11,33), the highly selective inhibitors which have been shown to be well tolerated in vivo with minimal side effects was used (35). We employed the widely used polymicrobial sepsis model, CLP. As showed in Fig. 5a-d, glomerulus of mice pretreated with HC showed lower expression of VCAM1 and E-selectin than CLP-operated mice. Consistent with this, mice with HC treatment exhibited signi cantly reduced levels of CD68 positive macrophages in ltration into the endothelium (Fig. 5e). As we have demonstrated that TRPV4 regulated LPS-induced translocation of NF-κB and IRF-3 in MGREC, in the next we detected the distribution of NF-κB and IRF-3 in glomerulus of mice. Our data showed that mice after CLP operated exhibited increased translocation of NF-κB and IRF-3, which was repressed by HC treatment (Fig. 5f,g). These results suggested that of TRPV4 mitigated glomerular endothelial in ammation possibly through inhibition of NF-κB and IRF-3 and their signal transduction downsteam.
Inhibition or knockdown of TRPV4 ameliorates CLP-induced kidney injury and renal function.
To evaluate the role of TRPV4 in sepsis, HE staining was performed to observe the extent of kidney injury. Our results revealed that 24 hours after CLP caused greater in ammatory cell in ltration in the glomeruli and tubulointerstitium, tubular epithelial cells swell and vacuolation, cast formation in the tubules, and glomerulus atrophy in the kidneys compared with those in sham-operated WT mice (Fig. 6a). However, mice administrated with HC or exhibited much weaker renal damage and much lower Jablonski Grade scores than those of CLP-operated mice (Fig. 6b). As renal hypoperfusion is one of the main characteristics of S-AKI, we wondered that whether TRPV4 inhibition improves CLP-induced renal hypoperfusion. Color doppler ow imaging was used to detect local renal blood ow, our result showed that 24 hours after mice received CLP operation exhibited signi cant reduction of renal perfusion (Fig.   Supplemental). TRPV4 inhibition did not improve renal hypoperfusion (Fig. 6d, e), which was contrary to our previous study (10). Mice subjected to CLP died within 3 days, with a survival rate of 43% after 24 hours. In contrast, the 24-hour survival rate of mice that received HC, 1 hour prior to the CLP surgery was 85% (Fig. 6c). To further verify the protective effect of TRPV4 inhibition in sepsis, we applied siRNA interference to downregulate TRPV4 in vivo. Mice were received the NCsiRNA or TRPV4 siRNA 3 days prior to the CLP surgery.24 hours after subjected to CLP, HE staining revealed that knockdown of TRPV4 ameliorates renal damage and exhibited much lower Jablonski Grade scores than those of CLP-operated mice pretreated with NCsiRNA (Fig. 6f, g). The survival rate of NCsiRNA treated mice was 37.5% and that of 75% of TRPV4siRNA transfection (Fig. 6h). However, local renal blood ow was not affected after knockdown of TRPV4 as showed in Fig. 6.k,l. Furthermore, we measured the serum levels of BUN and creatinine to functionally identify the renoprotective effect of TRPV4 knockdown in mice with CLP. Mice with CLP-operated after NCsiRNA-pretreated showed signi cant increased levels of BUN (Fig. 6i) and creatinine (Fig. 6j), while TRPV4 knockdown strongly alleviated these effects.

Discussion
In the present study, we have uncovered a novel TRPV4-mediated Ca 2+ -signaling pathway that is essential for LPS-induced glomerular endothelial in ammation and S-AKI and is, therefore, critical to innate immunity. TRPV4 mediates cytosolic Ca 2+ elevations in response to LPS, and this Ca 2+ in ux is necessary for glomerular endothelial cells activation. An important reason for this phenomenon is the Ca 2+ -in ux also regulates the activation and nuclear translocation of NF-κB and IRF-3. Thus, TRPV4 controls the transcriptional programs mediated by these key transcription factors during CLP-induced sepsis associated AKI.
The role of TRPV4 in ECs under endotoxin stimulation has been demonstrated previously (23). The mechanism involving in glomerular endothelial function and in ammation induced by sepsis has remained elusive. Firstly, we investigated whether TRPV4 suppression signi cantly attenuated the increase in [Ca 2+ ] i induced by the LPS, as expected, treatment with HC or transfected with TRPV4siRNA prevented LPS-induced cytosolic [Ca 2+ ] i elevations (Fig. 1a-f). To examine whether TRPV4 was required for signaling through other TLRs, we stimulated MGREC with CU-T12-9, which is TLR1/2 agonist and measured whether CU-T12-9 induced Ca 2+ entry was dependent on TRPV4. We found that MGREC stimulated with CU-T12-9 displayed a signi cant increase of cytosolic [Ca 2+ ] entry, relative to control group. However, CU-T12-9-induced Ca 2+ entry occurred independently of TRPV4 (Fig. 1g, h). It is documented that heteromeric TRPV4-TRPC1 channels contribute to SOCE in vascular endothelial cells (22).TRPV4 associated with Ca 2+ -activated potassium channels contribute to modulating SOCE with physiological relevance (27). Furthermore, endothelial cells, which fail to trigger SOCE limits LPS-induced vascular in ammation (14). Based on these, we wondered whether inhibition of TRPV4 affect the response of SOCE in MRGEC. Our results showed that blocking TRPV4 did not affect SOCE (Fig. 1i-j), indicating that TRPV4 mainly participates in TLR4 signaling, mediates intracellular Ca 2+ overload without altering SOCE.
Remarkably, a thorough study demonstrated that inhibition of the NF-κB in choroid plexus epithelial cells, involved in the production and regulation of several in ammatory cytokines, inhibited TRPV4-mediated activity, suggesting a link between TRPV4 and cytokine production (31). However, another study reported that TRPV4 is dispensable for LPS-induced nuclear translocation of NF-κB in in urothelial cells. (2).
According to these previous study, in our next study, we found that TRPV4 participates in LPS-induced phosphorylation and tranlocation of NF-κB and IRF3in MRGEC. Inhibition or knockdown of TRPV4 repressed the activation of NF-κb and IRF3( Fig. 2e-g,h,j,k,l,n). Interestingly, IκBα degradation appears to be unaffected after inhibition or knockdown of TRPV4 in MRGEC (Fig. 2g,i,k,m). This observation suggests that TRPV4-dependent Ca 2+ -signaling can greatly modify NF-κB signaling from different aspects. The local Ca 2+ -in ux promotes the TLR4 signaling which is critical to downstream signaling, including IRF-3 activation.
In the skin, TRPV4 contributes to the development and maturation of cell-cell junctions in epithelia of the skin (32). It directly interacts with β-catenin and E-cadherin, thereby results in strengthened cell-cell adhesion structures between keratinocytes and the basal membrane (21,32), thus forming a physical barrier against external environmental insults. Of note, TRPV4 is prominently expressed in endothelial cells, where it is involved in the hyper-in ammatory response and mortality associated with sepsis(9). In the next study, we wondered whether TRPV4 play a critical role in LPS-induced cell-cell junction disruption and production of endothelial adhesion molecules. Our results demonstrated that LPS stimulation caused an increase in VCAM1, ICAM1 and E-selectin, which were reduced by HC treatment (Fig. 3a, b).This fall in line with previous study that higher plasma levels of E-selectin and/or ICAM-1 correlate with damage of organs, severity of sepsis, and mortality(16, 30).And a decrease in ICAM-1 and E-selectin expression correlates with decreased neutrophil in ltration in lung, liver and kidney and a decrease in tissue damage (41). Inhibition of TRPV4 also enhanced endothelial cell barrier function as observed in immuno uorescence (Fig. 3c), indicating that TRPV4 is excessive activated during sepsis, thus promotes intercellular permeability. What we should consider is, a recent and comprehensive study demonstrated that TRPC6 de ciency prevented lung endothelial hyperpermeability, and reduced mortality up to 70% after LPS challenge (34). While pointing to a different TRP channel, these observations may be complementary with our results. Accordingly, inhibition of TRPV4 reduced the LPS-induced ROS production (Fig. 3d).These observation provided a strong envidence that TRPV4 participates in mediating kidney injury during an in ammatory systemic insult. It is reported that TRPV4-mediated rise in Ca 2+ induces the activation of nitric oxide synthases (eNOS and iNOS) expressed in epithelial cells, increasing NO production in the luminal compartment, resulting in a bactericidal effect(6). On the other hand, NO would decrease leukocyte adhesion to endothelial cells through down-regulation of cell adhesion molecules (8, 17). Thus we detected whether inhibition of TRPV4 affects the expression of eNOS and the production of NO. As can be seen from Fig. 3e-f, HC treatment did not increase the level of p-eNOS as well was the production of NO, suggesting that TRPV4 inhibition alleviated glomerular endothelial injury without improving the NO decrease.
Acute kidney injury is a frequent and serious complication during sepsis. To better understand the nature of LPS-induced [Ca 2+ ] i mobilization, MRGEC were loaded with BAPTA-AM and then stimulated with LPS, BAPTA-AM-loaded cells showed a drastic reduction of VCAM1, ICAM1 and E-selectin as well as the p-NF-κB and p-IRF-3 (Fig. 4a,c-e).The nuclear translocation of NF-κB was also prevented after diminishing cytosolic [Ca 2+ ] i elevations (Fig. 4b), con rming that Ca 2+ -signaling is essential for LPS-induced response.
To further investigate whether TRPV4 is critical to glomerular endothelial in ammation in vivo, we used a speci c TRPV4 antagonist, HC, and found that TRPV4 inhibition reduced CLP-induced increase of VCAM1, E-selectin in glomerulis of mice( Fig. 5a-d). Accordingly, mice with HC treatment exhibited signi cant reduced levels of CD68 positive macrophages in ltration into the endothelium in glomeruli (Fig. 5e). Furthermore, the translocation of NF-κB and IRF-3 glomeruli was also depressed by HC treatment (Fig. 5f,g).
As these observation, we wondered whether inhibition of TRPV4 ameliorate AKI in sepsis, HC was used to block TRPV4 and we found that TRPV4 inhibition ameliorates CLP-induced kidney injury (Fig. 6a,b).We also detected the renal blood ow after inhibiting TRPV4, surprisingly, our result showed that TRPV4 inhibition did not improve renal hypoperfusion (Fig. 1d, e),indicating that the protective effect of TRPV4 inhibition on S-AKI was not dependent on renal hypoperfusion. Besides, the 24-hour survival rate of mice that received HC was improved (Fig. 6c). Next, we applied speci c siRNA targeting TRPV4 to downregulate TRPV4 expression and investigated whether TRPV4 is involved in CLP-induced AKI. Our results showed that mice administrated with TRPV4siRNA decreased tubulointerstitial damage, renal dysfunction, serum level of BUN and CRE (Fig. 6f,g,i,j). The survival rate of mice was also increased, which is congruent with the effect of HC (Fig. 6h). However, the local renal blood ow after CLP was still unaffected by TRPV4siRNA administration (Fig. 6k, l). In our view, there may be a variety of mechanisms involving in regulating local renal blood ow, alterations of renal blood ow after sepsis are due to insu cient hemodynamic optimization or speci c microvascular injuries induced by sepsis (endothelial dysfunction, increased leukocyte adhesion, rheological abnormalities, glycocalyx alterations and functional shunting) (15). Consistent with our study, previous study reported that TLR4 inhibitor is able to reverse a manifest impairment in renal function caused by sepsis, but it does not involve in improving macro-circulation or micro-circulation, enhancing renal oxygen delivery, or attenuating tubular necrosis (13). Thus, our in vivo experimental results revealed that inhibition or knockdown of TRPV4 exerted a protective effect of CLPinduced AKI without improving local renal blood ow. However, the deep molecular mechanism of TRPV4 involving in S-AKI injury needs to be de ned.

Conclusion
Overall, in this study, we provide the rst evidence that inhibition of TRPV4 ameliorates glomerular endothelial in ammation, kidney dysfunction, and increased mortality via mediating Ca 2+ overload and NF-κB/IRF-3 activation under sepsis situation. Therefore, from a clinical perspective, it should be carefully considered the application of TRPV4 antagonists as a treatment option in the future.

Declarations
Availability of data and materials All data generated or analyzed in this study are included in this published article and its supplementary information les.
Ethics approval and consent to participate Animals were handled in accordance with procedures approved by Medical Ethics Committee of Shanghai Jiaotong University, and were performed in strict accordance with approved guidelines.

Consent for publication
Not applicable.
Phalloidin and VE-cadherin staining for assessing cytoskeletal remodeling and adherens junction integrity in MRGEC 24 hours after LPS and HC treatment. magni cation, ×200. Arrows in the Phalloidin and VE-cadherin staining represent gaps between the cells. d. The DCFH-DA staining showed the production of ROS. magni cation, ×50. e, f. The protein level of p-eNOS and eNOS was determined by western blot. *P < 0.05, **P < 0.01, n = 6. g. The NO production in supernatant was detected by NO assay kit. *P < 0.05, n = 6. LPS signaling and NF-κB translocation are abrogated by clamping intracellular Ca2+. a, b. The protein level of p-NF-κB p65, NF-κB p65 was determined. Cells were treated as indicated. **P < 0.01, n = 6. c.
Immuno uorescence staining of NF-κB was performed after treated with BPATA-AM. d, e The protein level of VCAM1, ICAM1 and E-selectin was determined by western blot. Cells were loaded with BPATA-AM before stimulated by LPS (1μg/mL). **P < 0.01, n = 6. f-h. The protein level of ICAM1 and E-selectin was determined by western blot. Cells were treated with HC before stimulated by CU-T12-9(10μM). *P < 0.05, n = 6.  TRPV4 inhibition or knockdown of TRPV4 ameliorates CLP-induced kidney injury and renal vascular in ammation. HC (10mg/kg) was intraperitoneally injected 1 hour before CLP operation. a. Representative images of HE. Images were captured at magni cation, ×200, scale bar=100 μm; b. Group data summarizes the results for HE staining scores. Histopathological grading of tissue injury was assessed using the 0-to 4-point scoring system. **P < 0.01, ***P < 0.001, n = 6. c. The survival rate of each group was recorded at 6, 12, 18 and 24 hours after CLP. d, e. 24 hours after CLP, renal blood ow was evaluated via laser doppler imaging and quanti ed as the average perfusion for each animal. *P < 0.05, n = 6. f. The NCsiRNA and TRPV4siRNA (5 mg/kg) were transfected in vivo via tail vein injection 3 days before CLP. Representative images of HE. Images were captured at magni cation, ×200, scale bar=100 μm; g. Group data summarizes the results for HE staining scores. Histopathological grading of tissue injury was assessed using the 0-to 4-point scoring system. *P < 0.05, ***P < 0.001, n = 6. h. The survival rate of each group at 6, 12, 18 and 24 hours after CLP was recorded. i, j. Serum creatinine and urea nitrogen levels in mice of each group were measured 24 h after CLP. *P <0.05, n=6. k, l. renal blood ow was evaluated via laser doppler imaging and quanti ed as the average perfusion for each animal.