Transient Receptor Potential Vanilloid 4 Channel Deficiency Aggravates Tubular Damage after Acute Renal Ischaemia Reperfusion

Transient receptor potential vanilloid 4 (TRPV4) cation channels are functional in all renal vascular segments and mediate endothelium-dependent vasorelaxation. Moreover, they are expressed in distinct parts of the tubular system and activated by cell swelling. Ischaemia/reperfusion injury (IRI) is characterized by tubular injury and endothelial dysfunction. Therefore, we hypothesised a putative organ protective role of TRPV4 in acute renal IRI. IRI was induced in TRPV4 deficient (Trpv4 KO) and wild–type (WT) control mice by clipping the left renal pedicle after right–sided nephrectomy. Serum creatinine level was higher in Trpv4 KO mice 6 and 24 hours after ischaemia compared to WT mice. Detailed histological analysis revealed that IRI caused aggravated renal tubular damage in Trpv4 KO mice, especially in the renal cortex. Immunohistological and functional assessment confirmed TRPV4 expression in proximal tubular cells. Furthermore, the tubular damage could be attributed to enhanced necrosis rather than apoptosis. Surprisingly, the percentage of infiltrating granulocytes and macrophages were comparable in IRI–damaged kidneys of Trpv4 KO and WT mice. The present results suggest a renoprotective role of TRPV4 during acute renal IRI. Further studies using cell–specific TRPV4 deficient mice are needed to clarify cellular mechanisms of TRPV4 in IRI.

Transient receptor potential (TRP) channels were initially identified in a mutant strain of Drosophila melanogaster as a novel component required in phototransduction 1 . Subsequently, they were identified in vertebrates and found to be ubiquitously expressed in many cells and tissues. Currently the mammalian TRP channel family comprises 28 members which share some structural similarity with each other 2 . Mutated TRP genes are causal for human hereditary diseases, the so-called 'TRP channelopathies' 3 , for some TRP genes.

Results
The Impact of TRPV4 Deficiency on Renal Damage after IRI. To gain insights into the (patho)physiological role of TRPV4 in renal IRI, we performed comparative in vivo studies using TRPV4 deficient (Trpv4 KO) mice and wild-type (WT) control mice. Although Trpv4 KO mice exhibited slightly higher serum concentrations of sodium and glucose, and blood haemoglobin and haematocrit levels at baseline, the kidney function markers creatinine and urea nitrogen were initially indistinguishable between Trpv4 KO and WT mice (Table 1). Already six hours after renal IRI, however, Trpv4 KO mice showed higher serum creatinine and urea nitrogen levels compared to WT mice; this difference in creatinine levels persisted over a period of twenty-four hours after ischaemia (Fig. 1A). Other lab parameters summarised in Supplementary Table 1 showed no significant differences at these time points after renal ischaemia. Sham-operated Trpv4 KO mice had similar creatinine levels compared to sham-operated WT mice (Supplemental Fig. 1). Kidney sections were haematoxylin and eosin (HE) stained to assess the degree of tubular damage after ischaemic AKI. Histological analyses and semi-quantitative scoring revealed aggravated cortical tubular damage in IRI-damaged kidneys of Trpv4 KO mice compared to IRI-damaged kidneys of WT mice (Fig. 1B,C). Of the five examined parameters tubular necrosis, tubular degeneration and brush border loss showed the most prominent differences between IRI-damaged kidneys of WT and Trpv4 KO mice, whereas sham-operated mice did not differ in any of the examined parameters (Fig. 1D). Extensive necrosis of tubular cells with brush border was verified by proximal tubular cell specific staining for aquaporin-1 (Supplemental Fig. 2). Additionally, we verified the existence of functional TRPV4 channels in the proximal tubular system. Proximal tubular cells (PTCs) were isolated from WT and Trpv4 KO mice and were stimulated in vitro with the TRPV4 agonist GSK1016790A at both 10 and 100 nM concentrations after loading with the calcium indicator fluo-4 AM. Calcium influx was detected in PTCs of WT mice after application of GSK1016790A but not in PTCs of Trpv4 KO mice (Supplemental Fig. 3). Histological analysis of the outer stripe of the outer medulla, a region characterized by a relative paucity of intertubular capillaries and therefore particularly prone to ischemic injury, showed less distinctive, but similar differences (Supplemental Fig. 3A-C). Surprisingly, renal mRNA levels of the AKI marker lipocalin 2 (Lcn2; also known as neutrophil gelatinaseassociated lipocalin [Ngal]) and kidney injury molecule-1 (Kim1, also named Tim1 and Havcr1), as well as renal mRNA levels of the pro-inflammatory cytokines interleukin (Il)-1 beta, Il-6 and tumor necrosis factor (Tnf) did not differ between sham or IR-injured kidneys of Trpv4 KO and WT mice ( Fig. 2A-E).
Expression of TRPV4 in the Renal Tubular System. We were somewhat surprised of the massive tubular necrosis and brush border loss seen in the renal cortex of Trpv4 KO mice, as it is believed that TRPV4 is not expressed in proximal tubular cells. Therefore, we used the advantage of our KO mice and tested several commercially available antibodies against TRPV4, however, only one of the tested commercially available antibodies worked in our hands for immunofluorescence detection of TRPV4 (Supplemental Fig. 4). A detailed analysis of renal tubular TRPV4 expression in WT kidneys revealed that TRPV4 is co-expressed in aquaporin-1 (proximal tubule), aquaporin-2 (collecting duct) and partly in sodium-chloride symporter (distal convoluted tubule) positive tubules but not in sodium-potassium-chloride co-transporter 2 (thick ascending limb) positive tubules ( Fig. 3A-D).
The Impact of TRPV4 Deficiency on Renal Apoptosis after IRI. In renal ischaemia, there is mismatch of local tissue oxygen supply and demand and cells accumulate metabolic waste products. As a result of this imbalance, tubular epithelial cells undergo death by apoptosis or necrosis depending on injury severity. Moreover, Figure 1. Renal function and damage after renal ischaemia/reperfusion injury (IRI). (A) Serum creatinine levels of Trpv4 KO (▪) and WT (○) mice at baseline, 6 and 24 hours after renal IRI. Please note that at baseline all serum creatinine levels were below the measurement limit (18 µmol/L). n = 6 for both WT and Trpv4 KO mice, repeated measurement two-way ANOVA (P Time < 0.0001, P Genotype = 0.0007, P Interaction < 0.0001), Sidak's multiple comparisons test, *P < 0.05, # P < 0.001. (B) Representative cortical images of haematoxylin and eosin stained sections of sham and I/R-injured kidneys of WT and Trpv4 KO mice (×200). * on the histological images represent necrosis and # represent tubular cast formation. (C) Semi-quantification of cortical tubular injury in sham and I/R-injured kidneys of WT and Trpv4 KO mice, two-way ANOVA (P Treatment < 0.0001, P Genotype = <0.0001, P Interaction < 0.0001), Sidak's multiple comparisons test, # P < 0.001, n = 6 for both WT and Trpv4 KO mice. (D) Detailed analysis of cortical tubular damage, one-way ANOVA, Tukey post-hoc test, # P < 0.001, n = 6 for both WT and Trpv4 KO mice. renal tubular epithelial apoptosis can initiate reperfusion-induced inflammation and subsequent tissue injury 35 . To assess tubular epithelial apoptosis, TUNEL labeling was performed on paraffin sections of IRI-kidneys of WT and Trpv4 KO mice. Interestingly, injured kidneys of Trpv4 KO mice showed similar rates of apoptosis as injured kidneys of WT mice (Fig. 4A). Quantification by direct counting of TUNEL-positive tubular cells confirmed this observation (Fig. 4B). Nevertheless, the concentration of the tissue damage marker lactate dehydrogenase (LDH) was significantly higher in IRI-injured Trpv4 KO mice compared to WT mice, supporting the histological finding of more extensive renal necrosis in Trpv4 KO mice (Fig. 4C). Contralateral kidneys of Trpv4 KO and WT mice showed virtually no TUNEL-positive tubuli (Supplemental Fig. 5).
The Impact of TRPV4 Deficiency on Renal Cellular Infiltration after IRI. Renal IRI is associated with infiltration of granulocytes and monocytes/macrophages. These cells contribute to inflammation and subsequent repair of the injured kidney. Therefore, we characterized renal granulocyte and macrophage infiltration after IRI by flow cytometry in WT and Trpv4 KO. Whole kidney cell suspensions were immunolabeled for Ly6G and F4/80 as markers for granulocytes and macrophages, respectively. Among pre-gated singlet live cells (Fig. 5A), Ly6Gpositive & F4/80-negative granulocytes as well as Ly6G-negative & F4/80-positive macrophages were detected (Fig. 5B). No significant differences in granulocyte or macrophage populations were detected between IRI-damaged kidneys of WT and Trpv4 KO mice at 24 hours after IRI (Fig. 5C,D). Renal cell populations were also similar in kidneys of Trpv4 KO and WT mice which underwent sham surgery (Fig. 5C,D). We also performed immunohistochemistry to detect IRI-induced granulocyte infiltration, and counted the average number of Ly6B-positive cells in the outer medulla after IRI. Similarly to our flow cytometry results, no differences in granulocyte counts were detected between WT and Trpv4 KO mice (Supplemental Fig. 6A,B).
The Impact of TRPV4 Deficiency on Primary PTCs Treated with Cobalt Chloride (CoCl 2 ). The stabilization of hypoxia-inducible factor (HIF)1α protein is a hallmark of hypoxia. To mimic hypoxia in vitro, WT and Trpv4 KO PTCs were exposed to the HIF1α stabilizing agent CoCl 2 in a low glucose medium for 24 hours. mRNA expression of the HIF1α target gene vascular endothelial growth factor (Vegfa) was upregulated by CoCl 2 treatment in both WT and Trpv4 KO PTCs, indicating successful HIF1α-stabilization. We next measured mRNA levels of the stress cytokine interleukin 6 (Il6), of nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha (IkBa), a bona fide NF-κB target gene, and of the endoplasmic reticulum stress marker CCAAT/enhancer-binding protein homologous protein (Chop). In both WT and Trpv4 KO PTCs CoCl 2 treatment caused a similar upregulation of these stress genes ( Fig. 6A-D). We subsequently counted the average number of cells per view-field of untreated and CoCl 2 -treated WT and Trpv4 KO PTCs by microscopic analysis. Percentage of remaining PTCs were calculated and percentage of apoptotic PTCs were determined using Annexin V staining (Fig. 6E-G and Supplemental Fig. 8). To corroborate these data caspase 3/7 activity was determined In all cases P Treatment < 0.001, and P Genotype and P Interaction are not significant.  using the Caspase-Glo 3/7 Assay (Fig. 6H) and number of viable cells were determined by the CellTiter-Glo Assay (Fig. 6I). In agreement with our in vivo data, we did not find any differences in the rate of apoptosis between WT and Trpv4 PTCs, whereas fewer Trpv4 PTCs could be detected after CoCl 2 treatment when compared to WT PTCs suggesting an increased rate of necrotic cell death in Trpv4 KO PTCs. Low-glucose environment by itselfwithout CoCl 2 -did not lead to increased cell death in Trpv4 PTCs compare to WT PTCs (Supplemental Fig. 9).

Discussion
Acute kidney injury is a global public health problem in developed and developing countries. Recent data show that AKI is responsible for more than 1.5 million deaths yearly worldwide 36 . Thus, there is an urgent need for new therapeutic approaches. Recent pre-clinical data indicate that TRPV1 agonists can improve the outcome of ischaemic AKI 37,38 . Besides TRPV1, TRPV4 has been recently identified to serve as a major calcium influx pathway in endothelial cells to control vascular function 26,39 . Very recently, we were able to show that TRPV4 channels can promote endothelium-dependent relaxation in different vessels in the kidney, including vasa recta 25 . Others have demonstrated that TRPV4 plays an important role in the preservation of vasodilatation under hypoxic conditions 27 . Moreover, TRPV4 channels are highly expressed in renal tubular cells as shown in this study and by others 40 . We therefore hypothesized that loss-of-function of the osmotically-sensitive ion channel TRPV4, would impact the ability of tubular cells to recover from IRI.
In this study, we explored the pathophysiological role of TRPV4 in renal IRI. We uncovered massive necrosis of PTCs after renal IRI and provide evidence -contrasting to earlier reports -that PTCs express functional TRPV4. As early as 6 hours after renal IRI Trpv4 KO mice showed higher serum creatinine levels compared to WT mice, which was maintained over 24 hours after renal ischaemia. Surprisingly, renal expression of the AKI markers Ngal and Kim1, which are expressed by distal and proximal tubular cells, respectively, did not differ between WT and Trpv4 KO mice. In the kidney, NGAL and KIM1 have been reported to be expressed specifically in injured nephrons 19,41 . On the other hand, the expression profile of NGAL and KIM1 in the injured kidney has been described to be similar to TRPV4 expression in the intact kidney [18][19][20]41 . Therefore, one possible explanation for the lack of an increased Ngal and Kim1 expression Trpv4 KO mice after renal IRI could thus be the higher necrosis rate of TRPV4 deficient tubular cells, which will in parallel deplete the number of NGAL and KIM1 expressing cells making the evaluation of renal Ngal and Kim1 expression ambiguous in this study. Noteworthy, histological analysis of the IRI-damaged kidneys of Trpv4 KO mice revealed massively increased cortical tubular damage, in particular tubular necrosis and loss of aquaporin-1 positive proximal tubular cells, compared to the kidneys of WT mice which underwent the same surgical procedure. This finding surprised us for two reasons; first, proximal tubular cells were thus far unknown to express TRPV4 channels, and second, the renal KO mice, two-way ANOVA, Sidak's multiple comparisons test. In both cases P Treatment < 0.0001, and P Genotype and P Interaction are not significant.
cortex is known to be less prone to ischemic injury as compared to the outer stripe of the outer medulla 18,20,41 . Therefore, we performed diligent co-immunostainings which revealed that proximal tubular cells indeed highly express TRPV4. While our study was underway Janas and co-workers reported on the expression of TRPV4 in the mouse kidney by using qPCR on isolated nephron segments, immunohistochemistry, and a Trpv4 reporter mouse model 40 . Similarly to our findings they found TRPV4 to be expressed in the proximal tubule, in the distal convoluted tubule, and in the collecting duct, but not in the glomerulus and the thick ascending limb. These Tissue injury resulting from hypoxia can result in both necrosis and apoptosis. Apoptosis occurs in normal and disease states, whereas necrosis is induced only when cells or tissues are exposed to severe and acute injury 42 . Lack of TRPV4 has recently been linked to increased apoptosis and inhibited autophagy in a rat liver stellate cell line 43 . Therefore, we addressed the question whether a higher rate of apoptosis might be responsible for the extensive renal tubular damage in Trpv4 KO mice. Interestingly, injured kidneys of Trpv4 KO mice showed similar numbers of TUNEL-positive tubular cells as injured kidneys of WT mice. Moreover, in vitro treatment of WT and Trpv4 KO PTCs with the HIF1α-stabilizing agent CoCl 2 caused similar levels of apoptosis as measured by either the caspase 3/7 assay or by counting the number of Annexin V positive PTCs. In addition, mRNA levels of stress genes like Il6, IkBa and Chop were similarly upregulated in CoCl 2 -treated WT and Trpv4 KO PTCs. However, CoCl 2 treatment induced more cell death in Trpv4 KO PTCs compared to WT PTCs as assessed by direct counting and by luminescent cell viability assay. These data suggest that TRPV4 deficiency induces necrosis but does not interfere with cell apoptosis in a hypoxia model of HIF1α stabilization. To the best of our knowledge no HIF binding sites on the Trpv4 gene has been established. Nevertheless, it has recently been shown that HIF1α transcriptionally regulates Trpa1 expression presumably via binding to an unusual hypoxia-response element-like motif within the Trpa1 gene, which subsequently leads to suppressed secretion of interleukin-6 and interleukin-8 44 . One might speculate that a similar mechanism could regulate Trpv4 expression in response to hypoxia. This in turn might be an indispensable mechanism for successful adaptation of PTCs to hypoxia.
Renal tubular epithelial apoptosis can initiate reperfusion-induced inflammation which is a hallmark of renal IRI 35,45 . Similar to the results for TUNEL-positive renal tubular cells, we found comparable degrees of infiltrating neutrophil granulocytes and F4/80-positive macrophages as well as mRNA expression levels of pro-inflammatory cytokines in IRI-kidneys of WT and Trpv4 KO mice. These finding in combination with our histological and immunofluorescence results strongly suggests that the lack of functional TRPV4 does not increase inflammation and apoptosis, but directly compromises the ability of the renal epithelium to recover from ischaemic injury leading to prompt necrosis. Higher serum LDH levels in Trpv4 KO mice compared to WT mice after renal IRI support this concept.
Calcium entry is a hallmark of the cellular reaction to ischemia in numerous cell types, including tubular epithelial cells. TRPV4 channels have been shown to be involved in ischemia-induced calcium entry in a number of cells, including reactive astrocytes and cardiac myocytes 46,47 . One may therefore expect that lack of TRPV4 channels should be rather beneficial. On the other hand, pharmacological inhibitors of TRPV4 channels are able to reduce cytokine production and preserve endothelial cell function, including Ca 2+ signaling, and endothelium-dependent vasodilation, resulting in increased survival e.g. in septic mice 48 . Additionally, Trpv4 KO mice exhibit higher blood sodium concentrations and present impaired osmotic sensitivity upon challenge 17 . Sublethal tubular damage results in impaired proximal sodium reabsorption which is expected to activate the tubular glomerular feedback mechanism leading to increased afferent vascular resistance and a concomitant decrease in glomerular filtration rate 49 . The impaired tubular transport of sodium in Trpv4 KO mice might also contribute to the progression of IRI in a tubular glomerular feedback-dependent manner. In a recent study, the exogenous TPRV4 activator apigenin reduced hypertension-induced renal fibrosis through the AMP-activated protein kinase/sirtuin 1 pathway in a deoxycorticosterone acetate (DOCA)-salt rat model. Knockout of TRPV4 in mice abolished the beneficial effects of apigenin that were observed in the DOCA-salt hypertensive rats, suggesting a protective and anti-fibrotic role of TRPV4 in chronic renal disease 50 . Here, our data suggest that lack of TRPV4 impairs recovery after acute renal IRI and induces massive-scale necrosis, especially in the renal cortex.
We demonstrated earlier that activation of TRPV4 promotes endothelial relaxation in renal resistance arteries, renal conduit arteries and in the medullary vasa recta 25 . During renal IRI the most commonly injured epithelial cells are the proximal tubular cells. These cells are particularly susceptible to ischemic insult partly due to their high metabolic rate required for mediating ion transport and a limited capacity to undergo anaerobic glycolysis 51 . As Trpv4 KO mice exhibited significantly more necrosis of proximal tubular cells after renal IRI compared to WT mice, it is tempting to speculate that vascular mechanisms may also be responsible for the observed effect.
A limitation of our study is that we used Trpv4 pan-null mice. Global lack of TRPV4 can lead to a number of pathologies e.g. impaired osmotic and mechanosensing 17 which become more prominent in stress situations. Although we carefully designed our study to circumvent possible confounding effects caused by the baseline phenotype of the Trpv4 KO mice, we cannot exclude that the above-mentioned limitation of global TRPV4 deficiency could have influenced the outcome of our study. This might apply particularly for haemodynamic changes after renal ischemia, since TRPV4 has an established role in the maintenance of vascular tone 39 . In addition, the increased sensitivity to ischaemia may not only be due directly to TRPV4 loss-of-function, but due to developmental-compensatory and other secondary mechanisms caused by genetically-engineered global absence on Trpv4. Indeed, TRPV4 have been recently described as a sensor of cell stress in the diseased fatty liver. These data lend solid support to the emerging concept of TRPV4 protecting a key epithelial cell -hepatocytes -from oxidative injury 52 .
In summary, genetically-encoded Trpv4 knockout (pan-null) aggravates acute renal tubular injury which is associated with enhanced renal epithelial necrosis. Although the results of our study suggest that this necrosis may be a direct effect of TRPV4 deficiency, additional studies are needed to clarify the exact molecular mechanism. Further studies using cell-specific TRPV4 mouse models are needed to clarify the precise pathophysiological sequence of events, and to clarify how the IRI-induced, TRPV4-dependent tubular damage influences long term renal recovery and survival.

Methods
Mice. Trpv4 KO mice have been generated and characterized previously. 17 Wild-type (WT) control mice as littermates on the same genetic background were used. Mice were allowed free access to standard chow and water and were kept in a 12:12-h light-dark cycle. Animal studies were approved by the Berlin Animal Review Board in 2012 (No. G0082/12) and conducted in accordance with the standards of the American Physiological Society.

Renal IRI Model.
Renal IRI was induced as described earlier 53 . Briefly, male mice (aged 14-17 weeks) were anaesthetized by isoflurane (2.3%) in air (350 mL/min). Mice were operated individually to ensure similar exposure to isoflurane (33.8 ± 3.0 min, mean ± SD) 54 . Body temperature was maintained at 37 °C and monitored during surgery using a temperature controller with a heating pad (TCAT-2, Physitemp Instruments). Ischaemia was induced after right-sided uni-nephrectomy by clipping the pedicle of the left kidney for 20 minutes with non-traumatic aneurysm clips (FE690K, Aesculap). Reperfusion was confirmed visually and the abdomen and the skin were sutured separately with a 5/0 braided-silk suture. After surgery mice had free access to water and chow. Body-warm sterile physiological saline solution (1 mL) and pre-emptive analgesia with tramadol (1 mg/ kg) were applied subcutaneously to every mouse. Sham operation was performed in a similar manner, except for clamping the renal pedicle. Mice with bleeding during surgery, with incomplete renal reperfusion, with excessive exposure to isoflurane for any reason, with significant temperature fluctuation during surgery, or with signs of infection 24 h after IRI, were immediately euthanized and not used for further analysis. Twenty-four hours after reperfusion, mice were sacrificed, and kidney and blood samples were collected for further analysis. The kidneys were divided into two portions. Half of the kidney was immersed in 4% phosphate-buffered saline (PBS)-buffered formalin for histology, and the other half was snap-frozen in liquid nitrogen for RNA preparation.

Primary Proximal Tubular Cell Isolation and Calcium concentration (Ca 2+ ) measurement. Cells
were obtained and cultured as described previously 53 . Briefly, WT and Trpv4 KO mice aged 10-12 weeks were euthanized by an overdose of isoflurane and the kidneys were flushed with 10 ml ice-cold PBS (Sigma). Renal cortices were dissected visually and minced into small pieces which were transferred through two layers of stainless steel sieves (pore size 125 μm and 106 μm; Linker). Tubular fragments caught by the 106 μm sieve were flushed in the reverse direction with PBS and centrifuged for 5 minutes at 170 × g, washed, and then resuspended into the appropriate amount of culture medium: 1:1 DMEM/F12 (D 8437; Sigma) supplemented with fetal calf serum (FCS) 10%, penicillin 100 IU/ml and streptomycin 100 µg/ml buffered to pH 7.4. Plates were incubated in a standard humidified incubator at 5% CO 2 . Medium was changed two days later and maintained every other day until the monolayer of cells reached 90% confluence; at this time, over 90% of the isolated cells Quantitative Real-Time (qRT)-PCR. qRT-PCR was performed as described earlier 53 . Briefly, total RNA was isolated from snap-frozen kidneys after homogenization with a Precellys 24 homogenizer (Peqlab) using RNeasy RNA isolation kit (Qiagen). RNA quality and concentration were determined by a NanoDrop-1000 spectrophotometer (Thermo Fisher Scientific). Two micrograms of total renal RNA were transcribed to cDNA (Applied Biosystems). Quantitative analysis of target mRNA expression was calculated using the relative standard curve method. TaqMan and SYBR green analysis was conducted using an Applied Biosystems 7500 Sequence Detector (Applied Biosystems). The expression levels were normalized to the purinergic housekeeping gene hypoxanthine-guanine phosphoribosyltransferase. Primer sequences are provided in Supplementary Table 2.

Serum Measurements.
To allow for repeated measurement of lab parameters within a short time interval in the same mouse, 90 μL blood was taken from the facial vein and parameters were measured using an i-STAT system with Chem8+ cartridges (Abbott). Serum creatinine levels of sham-operated mice were measured by Labor 28 GmbH (Berlin, Germany). Lactate dehydrogenase was measured at the Pathophysiological Platform of the Max-Delbrück Center for Molecular Medicine (Berlin, Germany) using a Fuji DRI-CHEM automated clinical chemistry analyser.

Microscopy.
Histology. Formalin-fixed, paraffin-embedded, 5 µm thick kidney sections were stained with haematoxylin (Sigma) and 1% eosin (Sigma). Semi-quantitative scoring of tubular damage was performed in a blinded manner in that 10 to 18 images at 200× magnification per sample were scored on a 1 to 10 scale according to percentage (where 0 = 0% and 10 = 100%) of tubular dilation, brush border loss, tubular necrosis, tubule degeneration and tubular cast formation. The total score was calculated as sum of all five morphological parameters.
CoCl 2 and untreated primary proximal tubular cells isolated from WT and Trpv4 KO mice (see below) were stained with Annexin V (51-65875×; BD Pharmingen) and DAPI (final concentration 1 µg/mL). Annexin V positive cells and negative cells were counted in 7-10 randomly selected high-power fields (400×) using of a Zeiss Axio Imager.M2 fluorescence upright microscope. TUNEL Assay. TMR In Situ Cell Death Detection Kit (Roche) was used according to the manufacturer's instructions on formalin-fixed, paraffin-embedded 2 μm thick sections. In order to visualize cell nuclei slides were counterstained with the nuclear dye 4′,6-diamidino-2-phenylindole (DAPI, Sigma). The number of TUNEL-positive tubular cells was determined as mean of at least 5 non-overlapping fields in the cortex by use of a Zeiss Axio Imager.M2 fluorescence upright microscope using a 20× Plan-Apochromat objective. HIF1α Stabilization in Primary PTC. Primary proximal tubular cells isolated from WT and Trpv4 KO mice were treated with 300 μM CoCl 2 to stabilize HIF1α or left untreated in a low-glucose DMEM medium (D6046, Sigma) for 24 hours. Low glucose medium was used because it is known that cell death in proximal tubular cell cultures is HIF1α-dependent only when glucose availability is limited 55 . After 24 hours medium was removed and cells were harvested for further analysis. To count Annexin V and DAPI-positive WT and Trpv4 KO PTC, cells were seeded at a density of 10 4 cells/well in a chamber slide with a removable 8 well silicone chamber (Ibidi). Untreated PTC and PTC treated with 300 μM CoCl 2 for 24 hours were stained with Annexin V (100 µL binding buffer +5 µL Annexin V/well) after washing with PBS. After 15 minutes incubation in the dark at room temperature the staining solution was removed and PTC were fixed with pre-warmed (37 °C) 4% formaldehyde (200 µL/well) for 30 minutes in the dark at room temperature. Thereafter the PTC were washed, the chamber was removed and the slide was subjected to microscopic analysis. Caspase-Glo 3/7 and CellTiter-Glo Luminescent Cell Viability Assays. 96-well clear plates were seeded with 1 × 10 4 WT or Trpv4 KO PTC/well. Cells were allowed to attach overnight. On average, cells were at 50% confluency before addition of 300 μM CoCl 2 . Twenty-four hours after CoCl 2 treatment the cells were assayed by the Caspase-Glo 3/7 or by the CellTiter-Glo Luminescent Cell Viability Kit according to manufacturer's instructions (Promega). After 10 min on a plate shaker at room temperature, 90% of the lysate volume was transferred to a 96-well solid-white plate (Nunc). Cell lysates were analysed in a SpectraFluor Plus microplate reader (Tecan), and blank values were subtracted from data points. Relative fluorescence intensity of treated cells was normalized by their respective untreated control cells to normalize for possible seeding density differences between WT and Trpv4 KO PTC. In case of the CellTiter-Glo Luminescent Cell Viability Assay an ATP standard curve was measured in every experiment.
Statistics. Statistical analysis was performed using GraphPad Prism 5.04 (GraphPad Software). Normality of the data was evaluated by Kolmogorov-Smirnov test. Grubbs' test was used to detect outliers. Data were analysed by repeated measurement or regular two-way ANOVA with Sidak's multiple comparisons test, as appropriate. Data with two groups were tested by two-sided unpaired t-test (data with normal distribution). Data are presented as boxplots, showing median, 25 and 75 percentiles and minimum and maximum. Detailed tubular injury score data are presented as mean ± SD. P values < 0.05 were considered as statistically significant.
Data availability. All data generated or analysed during this study are included in this published article (and its Supplementary Information files). Raw data is available to interested parties upon request.