Single and Transient Ca2+ Peaks in Podocytes do not induce Changes in Glomerular Filtration and Perfusion

Chronic alterations in calcium (Ca2+) signalling in podocytes have been shown to cause proteinuria and progressive glomerular diseases. However, it is unclear whether short Ca2+ peaks influence glomerular biology and cause podocyte injury. Here we generated a DREADD (Designer Receptor Exclusively Activated by a Designer Drug) knock-in mouse line to manipulate intracellular Ca2+ levels. By mating to a podocyte-specific Cre driver we are able to investigate the impact of Ca2+ peaks on podocyte biology in living animals. Activation of the engineered G-protein coupled receptor with the synthetic compound clozapine-N-oxide (CNO) evoked a short and transient Ca2+ peak in podocytes immediately after CNO administration in vivo. Interestingly, this Ca2+ peak did neither affect glomerular perfusion nor filtration in the animals. Moreover, no obvious alterations in the glomerular morphology could be observed. Taken together, these in vivo findings suggest that chronic alterations and calcium overload rather than an induction of transient Ca2+ peaks contribute to podocyte disease.

Diseases of the glomerular filter are a leading cause of end stage renal failure. Podocyte injury is the common final pathway of any glomerular disease. Podocytes are highly specialized epithelial cells and part of the three-layered kidney filtration barrier as they enwrap the glomerular capillaries with their primary and secondary foot processes completely 1 . This barrier consists of the glomerular fenestrated endothelium, the glomerular basement membrane and the podocytes 1 . The latter form an interdigitating pattern of foot processes of adjacent cells. The only cell-cell contact of adjacent podocytes is the slit diaphragm, which is located in between the foot processes 1 . The slit diaphragm does not only serve as a filter to prevent loss of macromolecules from blood into primary urine, but also acts as a signalling platform [2][3][4] . In the last two decades several genes have been identified to be crucial to mediate signal transduction in podocytes [5][6][7][8][9] . In particular, the regulation of the actin-based cytoskeleton via several actin-binding proteins like alpha-actinin-4 and CD2AP 7,10,11 appears of specific importance as cytoskeletal changes precede the development of foot process effacement and subsequently the onset of proteinuria as observed in patients with podocyte diseases [12][13][14] . Calcium was identified as an important second messenger in states of podocyte health and disease 15 . As early as 1978 Kerjaschki et al. reported that foot process effacement might be Ca 2+ -dependent 16 . The impact of Ca 2+ for podocyte biology was further validated in the following years. Overexpression of the G-protein-coupled angiotensin II type 1 receptor, which results in an increased Ca 2+ flux, causes glomerulosclerosis in rats 17 . Moreover, gain-of-function mutations in the non-selective cation channel TRPC6 were identified to cause familial FSGS 8,18 . In addition, TRPC6 was found to be upregulated in different glomerular diseases, e.g. minimal change disease or membranous glomerulonephritis 19 . TRPC channels themselves are tightly regulated 20 and activated by various stimuli (such as e.g. angiotensin II AT1 receptor). Some reports see a role for TRPC channels to regulate the contractile and motile phenotype of podocytes, as TRPC5 activates Rac1 and TRPC6 RhoA through Ca 2+ signalling 20 . If this balance is disturbed, podocytes develop either an increased motility through overactivation of Rac1 or a contractile phenotype through overactivation of RhoA. In vivo, expression of either constitutively active Rac1 or RhoA both caused albuminuria and foot process effacement 21,22 . Additional findings on the Ca 2+ dependent degradation of the actin-binding protein synaptopodin via the calcineurin/cathepsin-L pathway further emphasise the importance of Ca 2+ in podocytes 23,24 .
Taken together, numerous reports highlight the role of Ca 2+ dependent signalling in podocytes. However, most studies only provide indirect evidence that indeed Ca 2+ signalling is required to regulate podocyte function in vivo as they focus on e.g. receptor function and do not study Ca 2+ levels in vivo. Due to technical limitations most experiments performed to investigate the role of Ca 2+ have been performed on cultured podocytes ex vivo. To further study the role of Ca 2+ dependent signalling events in podocytes in vivo and its implication on podocyte function we applied the DREADD (Designer Receptor Exclusively Activated by a Designer Drug) concept (t-test: *p < 0.05). (C,D) Representative images of Ca 2+ imaging with Fluo-8 (scale bar = 20 μ m). FLAG hM 3 D expressing podocytes were used for Ca 2+ imaging with Fluo-8. Administration of 2 μ M CNO leads to an immediate increase of intracellular Ca 2+ levels. The time point of CNO administration is indicated with a red tick in panel one. Ionomycin was used as positive control (indicated with a green tick in panel one). 12 cells are depicted from which some reached saturation of the fluorescence signal. Fluorescence is depicted as F/F 1-5 , where the mean value of the first five measurements (F 1-5 ) prior to CNO stimulation was used for normalisation. Experiments were performed in three biological replicates. (E) Statistical analysis revealed a significant increase of intracellular Ca 2+ after CNO stimulation. 45 cells from three biological replicates were included in the statistics. (t-test: p < 0.0001). (F) FLAG.hM 3 D expressing podocytes were used for repetitive Ca 2+ imaging with Fluo-8. Podocytes were treated with CNO at 0 and 24 hrs. At both time points an immediate increase of intracellular Ca 2+ levels could be observed. Fluorescence is depicted as F/F 1-5 , where the mean value of the first five measurements prior to CNO stimulation was used for normalisation. and generated a novel podocyte-specific transgenic mouse model in which we were able to induce increased intracellular Ca 2+ levels by administration of a specific chemical compound. To this end, we use a previously well characterized mutant human muscarinic type 3 receptor (hM 3 D), which leads to an exclusive activation by binding of the inert compound Clozapine-N-Oxide (CNO) 25,26 . With this in vivo approach we demonstrate that single transient Ca 2+ peaks do not affect glomerular function.

Results
Conditional expression of a functional hM 3 D in murine immortalized podocytes. We first validated the hM 3 D construct in murine immortalized mouse podocytes and generated cell lines either expressing FLAG-tagged hM 3 D or GFP as control. Immunofluorescence stainings revealed a clear membrane localization of the receptor (Fig. 1A). To verify functionality of the receptor we administered CNO and studied phosphorylation of MAP kinases as a readout for intracellular Ca 2+ signalling. CNO led to a significantly increased phosphorylation of the MAP-Kinases MEK1/2 (Fig. 1B). In addition, we performed Ca 2+ -imaging studies with the Ca 2+ indicator dye Fluo-8. Treatment with 2 μ M CNO increased intracellular calcium levels in the hM 3 D-expressing cells, but not in control cells (Fig. 1C,D). Depicted are the Ca 2+ signals of 12 different hM 3 D-expressing cells (Fig. 1D). Of note, some cells showed a single peak upon stimulation with CNO ( Fig. 1D upper panel) while others revealed an oscillating response only ( Fig. 1D middle panel) or did not react at all (Fig. 1D lower panel). This is most likely a technical artefact due to the lentiviral system used to express the hM 3 D transgene in the podocyte cell line. While the use of this retroviral gene transfer system allows expression of transgenes even in non-proliferating podocytes, which are hardly to transfect, it does not allow to control for expression levels as integration is random and can occur only once or several times. Experiments were performed in three biological replicates and a total of 45 cells were included in the statistical analysis shown in Fig. 1E. We analysed the fluorescence signal after CNO stimulation in comparison to the basal level prior to CNO stimulation. We observed a significant increase in intracellular Ca 2+ upon CNO stimulation in hM 3 D expressing cells (p < 0.0001). We also excluded receptor desensitization or internalization by repetitive Ca 2+ imaging experiments at 0 hrs and 24 hrs after the initial stimulation (Fig. 1F). This data clearly demonstrates the functionality of the hM 3 D-receptor and the ability to induce an increase of intracellular calcium levels in vitro.
Increased intracellular Ca 2+ levels after CNO stimulation cause Rac1 activation. Small GTPases are known to be crucial for podocyte function and maintenance 21,22 and are under tight control of intracellular Ca 2+ levels 27 . Moreover, GTPases play a crucial role for the organisation of the actin cytoskeleton and cell migration 28 . To investigate the effects of increased intracellular Ca 2+ levels after CNO stimulation on small GTPases in podocytes, we performed GTPase pull-down assays for active Rac1 and RhoA. Podocytes were treated with 1 μ M CNO for 15 min and the GTP-bound active Rac1 and RhoA were pulled down ( Fig. 2A). Comparing the GFP-expressing control cells with the FLAG.hM 3 D-expressing podocytes revealed no statistically significant differences of active Rac1 or RhoA levels in the receptor expressing cells in comparison to control cells, although a trend towards increased active Rac1 and decreased RhoA levels could be observed ( Fig. 2A).
Generation of a transgenic podocyte-specific hM 3 D mouse model. Next, we generated a knock-in mouse model to manipulate Ca 2+ levels in podocytes specifically in vivo. To this end, a ROSA26 targeting vector was generated in which expression of the C-terminal FLAG-tagged hM 3 D construct is driven by a CAGS promotor as published previously 29 . To ascertain cell type specific expression of the receptor transgene a loxP-flanked Neo.Stop cassette was inserted upstream of the receptor sequence. In addition, we included an IRES.GFP to visualize transgene expression (Fig. 3A). C57/BL6 ES cells of the Bruce 4 lineage were transfected with the targeting vector and positive clones were identified via selection and southern blot analysis (Fig. 3B). Positive clones were used to generate chimeras that were intercrossed with C57/BL6 mice to establish the ROSA26 LoxP-STOP-LoxP hM 3 D mouse strain. Mice carrying the hM 3 D construct in the ROSA26 locus were mated with Nphs2. Cre mice to achieve a podocyte specific receptor expression after Cre mediated recombination and deletion of the STOP codon ( Fig. 3A) 30 . We further validated expression of the receptor by isolating and culturing primary glomerular cells in vitro. Podocytes from hM 3 D-transgenic mice also expressed GFP and could easily been identified (Fig. 3C). To validate the functionality of the hM 3 D receptor in primary podocytes we labelled the cells with the Ca 2+ indicator dye Fluo-8 and stimulated the cells with 100 μ M CNO. hM 3 D/GFP-positive cells showed a strong increase in the intracellular Ca 2+ -levels upon CNO treatment, while hM 3 D-negative cells did not respond to CNO treatment (Fig. 3D). Experiments were performed in three biological replicates. Analyses of the signal intensities revealed an immediate increase of intracellular Ca 2+ with a constant decline of the signal (Fig. 3E). Of note, the strong oscillatory phenotype observed in some virus transduced cultured podocytes was not seen in primary cells. Other glomerular cells, like endothelial or mesangial cells, which were co-cultured in this mixed primary cell culture system, served as internal controls and did not respond to CNO nor did the outgrowing glomerular cells from hM 3 D negative control animals. Thus, our data clearly shows the podocyte-specific expression of the hM 3 D receptor and its functionality in isolated primary podocytes.

Administration of CNO leads to increased Ca 2+ -levels in podocytes in vivo.
To assess the functionality of the receptor in vivo we mated our ROSA hM3D/wt ; Nphs2.Cre tg/wt mice with the Ca 2+ -reporter strain GCaMP3 tg/wt 31 . Using Multi-Photon-microscopy we were able to study the effects of CNO administration on podocytes in vivo. After i.a. injection of 5 mg/kg bodyweight CNO an immediate elevation of the intracellular Ca 2+ levels was observed in podocytes ( Fig. 4A,B). All podocytes responded simultaneously and showed a significant single transient Ca 2+ peak (Fig. 4B,C).
Having shown the feasibility of our transgenic mouse model to transiently increase Ca 2+ levels in podocytes we set out to study the effects of CNO and a subsequent increase of intracellular Ca 2+ -levels in podocytes on glomerular filtration and glomerular perfusion. First, we performed transcutaneous glomerular filtration rate measurements to study the glomerular filtration rate (GFR) as published by Schreiber et al. 32 . Using this method, we were able to record the glomerular filtration rate after CNO administration in freely moving mice over 60 minutes. We did not observe any significant differences in GFR between hM 3 D-expressing animals in comparison to their littermate controls (Fig. 5A).
Second, we investigated the impact of increased intracellular Ca 2+ -levels on glomerular perfusion. As podocytes harbour a contractile actin-myosin-based machinery 33 we hypothesized, that podocytes adapt their contractile behaviour due to the changes in Ca 2+ concentration, which in turn might alter vascular resistance. To this end, we performed ex vivo perfusion experiments as published previously 34 . Kidneys of hM 3 D-expressing mice and their littermate controls were subjected with increasing CNO concentrations, starting at 0.01 μ M up to 100 μ M. However, we could not observe a change in the vascular resistance after CNO administration (Fig. 5B). Administration of angiotensin II, which leads to constriction of the vas afferens, resulted in an immediate and strong decrease of the perfusion rate and served as positive control. Taken together, these results show that administration of CNO and a subsequent increase in intracellular Ca 2+ does not change the glomerular filtration rate nor renal vascular resistance.
Increased intracellular Ca 2+ -levels do not cause glomerular disease. After investigating short-term effects of increased intracellular Ca 2+ levels on podocyte function, we next investigated effects after repetitive administration of CNO. Animals expressing the hM 3 D construct and their littermate controls received CNO for 5 consecutive days (5 mg/kg bodyweight i.p.). However, we could not observe any signs of glomerular disease based on histological analyses as well as urinary albumin-to-creatinine ratios (Fig. 6B-D). Immunofluorescence stainings for the actin-binding protein synaptopodin as well as the slit diaphragm protein podocin did also not reveal differences between the hM 3 D-expressing and control mice (Fig. 6C). Next, to investigate if a prolonged application of CNO causes glomerular disease, we administered CNO (5 mg/kg bodyweight i.p.) to mice expressing the hM 3 D construct and their control hM 3 D-negative littermates for 18 consecutive days. Again, urinary and  histological analyses did not reveal any signs of glomerular disease (Fig. 7B-D). In conclusion, even a repetitive administration of CNO for 18 days failed to induce glomerular disease.

Discussion
In this study, we present a novel transgenic mouse model to study Ca 2+ signalling in podocytes in vivo. We used a modified DREADD, a Gq-coupled human muscarinic type 3 receptor (hM 3 D). Due to two point mutations (Y149C/A239G) this receptor can solely be activated by the chemical compound CNO but no longer by its native ligand acetylcholine 25,26 . Moreover, great advantages of the usage of CNO are its bioavailability for rodents 35 , the natural affinity of its parent compound clozapine to M 3 receptors, which makes CNO a potent agonist of the mutated M 3 receptor 26 and its pharmacological inert behaviour 25,36 . The highly specific binding of CNO to the receptor causes activation of the Phospholipase C (PLC) pathway in the targeted cell type. This activation leads to binding of IP 3 to the IP 3 -receptor at the ER resulting in a strong Ca 2+ release from the ER into the cytoplasm. This allows studying effects of increasing intracellular Ca 2+ levels in vivo in a novel, highly specific manner. As we inserted a LoxP-STOP-LoxP cassette in the ROSA26 locus upstream of the hM 3 D sequence, this mouse model may be used to study effects of Gq-coupled receptor activation in a variety of tissues and cell types. Here, we mated the ROSA hM3D/wt mice to a podocyte-specific Cre recombinase driver line (NPHS2.Cre tg/wt ) 30 to achieve exclusive expression of the receptor transgene in podocytes. Doing so, the increase of intracellular Ca 2+ after CNO administration solely occurred in podocytes.
Surprisingly, despite the proven functionality of the hM 3 D receptor in podocytes as demonstrated in vitro, ex vivo and in vivo, activation of the PLC pathway did not affect glomerular filtration or perfusion nor caused glomerular disease even after a prolonged stimulation for 18 consecutive days. Apparently, short and transient Ca 2+ peaks as visualized by calcium imaging in vivo in this study are not sufficient to cause overt morphological alterations or functional changes at the renal filtration barrier. We did not observe any effect on glomerular perfusion, filtration or the onset of albuminuria. Our findings are in line with previously reported effects of G-protein coupled receptors in regulating glomerular biology. Expression of a constitutively active Gq-coupled receptor mutant in podocytes did also not result in any glomerular dysfunction under physiological conditions 37 . Only when a second hit was present, e.g. the PAN nephrosis model or diabetic conditions, glomerular malfunction was aggravated to a greater extent as compared to wildtype controls 37 .
These findings do not allow the assumption that Ca 2+ signalling events are dispensable for podocyte function. As we activated the PLC pathway, increased Ca 2+ levels primarily derive from the ER. This might be of particular importance in podocytes with their delicate cytoskeletal architecture. Increased Ca 2+ levels close to the ER, which is primarily found in perinuclear regions in the podocyte cell body, might not resemble the intracellular Ca 2+ concentration as observed in patients with TRPC6 gain-of-function mutations 8 . Moreover, podocytes rely on a stringent control of their intracellular Ca 2+ levels as they harbor several homeostasis mechanisms including the Na + -Ca 2+ -exchanger, the ATP-dependent plasma membrane Ca 2+ pump (PMCA) as well as Ca 2+ -buffers like calbindin and parvalbumin 38 . It is well conceivable that increasing intracellular Ca 2+ levels due to the CNO administration might be compensated by these homeostasis mechanisms. In addition, the slit diaphragm might serve as a Ca 2+ microdomain orchestrating Ca 2+ -dependent proteins like the TRPC channels, GTPases and synaptopodin. Furthermore, administration of CNO did only result in a transient Ca 2+ peak in podocytes.
One might also speculate, whether the hM 3 D-transgene gets rapidly internalized or undergoes desensitization upon stimulation. However, studies performed in hippocampal neurons expressing a similar hM 3 D-transgene showed repetitive activation of the receptor 24 hrs after the initial CNO treatment 25 , suggesting that receptor internalisation and desensitization does not occur. Similarly, we could observe a repetitive activation of the receptor and subsequent increase of intracellular Ca 2+ levels by stimulating podocytes with CNO at 0 and 24 hrs. It is therefore easily conceivable, that a sustained Ca 2+ influx rather than a single Ca 2+ peak is required to induce cytoskeletal changes to cause foot process effacement in podocytes.
In conclusion, we present a novel DREADD mouse model to study Ca 2+ signalling events in vivo. As the DREADD sequence is integrated into the ROSA26 locus it can be easily applied for all murine tissues by mating the loxP-STOP-loxP hM 3 D mice with tissue specific Cre recombinase mouse lines. In podocytes, a transient Ca 2+ peak alone is not sufficient to impair podocyte function challenging the current view of transient Ca 2+ peak as regulators of glomerular biology.

Material and Methods
Cell culture. Immortalized mouse podocytes were cultured with RPMI-1640 medium (Sigma-Aldrich, St. Louis, USA) with 10% fetal bovine serum, 5% sodium pyruvate solution 100 mM (Sigma-Aldrich, St. Louis, USA) and 5% HEPES buffer solution 1 M (Life Technologies, Carlsbad, USA) as previously reported 39 . The undifferentiated and proliferating cells were cultured at 33 °C in the presence of 2.5 μ l murine IFN-gamma (Provitro,   Berlin, Germany). To induce podocyte differentiation the cells were shifted to 37 °C for 14 days in the absence of IFN-gamma. To generate podocyte cell lines with stable expression of FLAG-tagged hM 3 D or GFP we used the commercially available pLenti6/V5-dest vector (Thermo Fisher, Waltham, USA) as previously published 40 . To this end, podocytes were grown at 33 °C and transduced with a Lentivirus containing the hM 3 D or GFP construct respectively.
Immunofluorescence on cells. Immortalized mouse podocytes were grown at 37 °C for 14 days. The cells were fixed with 4% paraformaldehyde, further blocking was performed with 5% normal donkey serum for 30 min. The cells were washed with 1 x PBS (phosphate buffered saline: 137 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4 and 2 mM KH 2 PO 4 ) containing 1 mM CaCl 2 and 0.5 mM MgCl 2 for three times. The cells were stained with an anti-FLAG antibody over night at 4 °C. The primary antibody was diluted in 1 x PBS with 1 mM CaCl 2 , 0.5 mM MgCl 2 and 0.1% Triton-X and with 5% NDS. Prior to incubation with the secondary antibody, cells were washed again three times with 1 x PBS with 1 mM CaCl 2 and 0.5 mM MgCl 2 . The respective secondary antibody was diluted in 1 x PBS with 1 mM CaCl 2 , 0.5 mM MgCl 2 and 0.1% Triton-X. Mounting was performed with Prolong Gold with DAPI (Thermo Fisher, Waltham, USA). Antibody dilution and commercial source are listed in Table 1. All images were taken with an Axiovert 200 M microscope/C-Apochromat 63x/1.20 W water immersion objective (all from Carl Zeiss MicroImaging GmbH). Images were further processed with ImageJ/Fiji Software and Photoshop CS4 version 11.0. Western Blot. Immortalized mouse podocytes were stimulated with 1 μ M CNO (Enzo Life Sciences, Farmingdale, USA) diluted in medium for 10 min. Afterwards the cells were harvested using a cell scraper and lysed in 100 μ l 1% Trition-X 100 buffer (20 mM Tris-HCl pH 7.5, 50 mM NaCl, 50 mM NaF, 15 mM Na 4 P 2 O 7 ) and 1 mM of the serine protease inhibitor PMSF Phenylmethanesulfonyl fluoride (Sigma-Aldrich, St. Louis, USA) and 2 mM of the phosphatase inhibitor Na 3 VO 4 (Sigma-Aldrich, St. Louis, USA). Following incubation on ice for 15 min and subsequent centrifugation (14.000 × g, 4 °C, 15 min) the supernatant was used for further analyses. The diluted lysates (with 2x SDS-PAGE buffer) were separated by SDS-PAGE. For gel electrophoresis the XCell SureLockTM Mini-Cell System was used. Electrophoresis was performed with 70 V for 30 min, followed by 25 mA/gel for 1:45 h. Blotting was performed on PVDF membranes (Carl Roth, Karlsruhe, Germany) for 1 h at 12 V. After the transfer, membranes were blocked with 5% bovine serum albumin (BSA) (Sigma-Aldrich, St. Louis, USA) for 30 min, followed by three washing steps. The primary antibody was diluted in washing buffer (30 mM Tris, 300 mM NaCl and 0.3% (v/v) Tween20, pH 7.5) and incubation was performed over night at 4 °C. The respective secondary antibody was diluted in washing buffer and incubated 45 min at room temperature, followed by visualization with chemiluminescence with a Fusion Solo S (Vilber Lourmat, Eberhardzell, Germany). Antibody dilution and commercial source are listed in Table 1. Transgenic mouse models. Generation of the ROSA26 hM3D/wt mice is described in the results section in detail.
Briefly, to generate mice expressing the hM 3 D transgene in podocytes, we mated ROSA26 hM3D/wt mice to NPHS2. Cre tg/wt mice 30 . Multi-Photon-imaging was performed using the Ca 2+ reporter mouse GCaMP3 tg/wt 31 . ROSA hM3D/wt ; NPHS2.Cre tg/wt mice were mated with GCaMP3 tg/wt to achieve a simultaneous hM 3 D and Ca 2+ reporter expression in podocytes exclusively. All animals were on a pure C57/Bl6 background. The mouse holding was done in the University of Cologne animal facility according to standardized specific pathogen-free conditions. All animal For immunofluorescence analysis frozen kidney sections (5 μ m) were used. After fixation with paraformaldehyde sections were blocked with 5% normal donkey serum in phosphate buffered saline with 0, 1% Triton-X (PBS-T) and subsequently incubated with primary antibody diluted in PBS-T with 5% NDS overnight at 4 °C. The respective secondary antibody was diluted in PBS-T and incubation was performed for 45 min at room temperature. Mounting was done with Prolong Gold antifade with DAPI (Thermo Fisher, Waltham, USA). All images were taken with an Axiovert 200 M microscope/C-Apochromat 63x/1.20 W (all from Carl Zeiss MicroImaging GmbH, Jena, Germany). Images were further processed with ImageJ/Fiji Software and Photoshop CS4 version 11.0. Antibody dilution and commercial source are listed in Table 1.
Multi-Photon -Calcium imaging. 4 weeks old ROSA hM3D/wt ; NPHS2.Cre tg/wt ; GCaMP3 tg/wt mice were anaesthetized using isoflurane and buprenorphine (0.1 mg/kg). A tube was placed into the trachea to facilitate breathing, and the right carotid artery was cannulated for dye infusion and injection of CNO. 70 kDa Texas red dextrane was injected i.a. to label the vasculature. The left kidney was exteriorized via a small incision at the left flank of the animal. The mouse was placed onto a heated mouse holder and the kidney was stabilized and covered with a coverslip for imaging. Body temperature was maintained during the imaging time. Calcium imaging was performed with a TSC SP8 upright Multi-Photon microscope (Leica) using an IR Apo L25x/0.95 W objective and a Coherent Chameleon Vision II laser at a wavelength of 940 nm. For acquisition two external hybrid detectors were used. The time courses of calcium signals were acquired at 1 frame per second over 90 seconds. After 10 seconds of baseline recording, 5 mg/kg bodyweight CNO was injected. The images were further processed with Leica Application Suite X, Version 1.1.0.

Transcutaneous glomerular filtration rate measurement.
To study the glomerular filtration rate we made use of a transcutaneous measurement approach 32 . Therefore ROSA hM3D/wt ; NPHS2.Cre tg/wt mice were anaesthetized with isofluran and a sensor was taped directly on their skin. After injecting 5 mg/kg CNO i.p., we injected 15 mg/100 g FITC-Sinistrin i.v. and measured the GFR for 60 min in the freely moving mice. Calculation of the GFR was performed according to Schreiber et al. 32 .
Isolated perfused mouse kidney. Kidneys from ROSA hM3D/wt ; NPHS2.Cre tg/wt mice and their littermate controls were perfused ex-situ at a constant perfusion pressure (100 mmHg) as described in detail previously 34 . Perfusion medium consisted of a modified Krebs-Henseleit buffer supplemented with bovine serum albumin (6 g/100 ml) and human erythrocytes (10% hematocrit). The renal vein was cannulated and samples of the venous perfusate were collected and weighed every 2 minutes for the determination of renal blood flow. Three samples were taken during each experimental period and the last two values were averaged for statistical analysis.
Statistical analysis. All results are expressed as means ± SEM, except for the GFR measurement, which is depicted as mean. Statistical significance was evaluated using GraphPad Prism version 6 for Windows (GraphPad Software, San Diego, CA). For two groups t-test was applied and a P-value < 0.05 was considered significant. For two groups with two independent variables 2 way ANOVA combined with Bonferroni's multiple comparison test was applied and a P-value < 0.05 was considered significant.