Podocyte foot process retraction is a hallmark of proteinuric glomerulonephritis. Cytoskeletal rearrangement causes a redistribution of slit membrane proteins from the glomerular filtration barrier towards the cell body. However, the underlying signaling mechanisms are presently unknown. Recently, we have developed a new experimental model of immune-mediated podocyte injury in mice, the antipodocyte nephritis (APN). Podocytes were targeted with a polyclonal antipodocyte antibody causing massive proteinuria around day 10. Rho-kinases play a central role in the organization of the actin cytoskeleton of podocytes. We therefore investigated whether inhibition of Rho-kinases would prevent podocyte disruption. C57/BL6 mice received antipodocyte serum with or without daily treatment with the specific Rho-kinase inhibitor HA-1077 (5 mg/kg). Immunoblot analysis demonstrated activation of Rho-kinase in glomeruli of antipodocyte serum-treated mice, which was prevented by HA-1077. Increased Rho-kinase activity was localized to podocytes in APN mice by immunostainings against the phosphorylated forms of Rho-kinase substrates. Rho-kinase inhibition significantly reduced podocyte loss from the glomerular tuft. Periodic acid staining demonstrated less podocyte hypertrophy in Rho-kinase-inhibited APN mice, despite similar amounts of immune complex deposition. Electron microscopy revealed reduced foot process effacement compared with untreated APN mice. Internalization of the podocyte slit membrane proteins nephrin and synaptopodin was prevented by Rho-kinase inhibition. Functionally, Rho-kinase inhibition significantly reduced proteinuria without influencing blood pressure. In rats with passive Heymann nephritis and human kidney biopsies from patients with membranous nephropathy, Rho-kinase was activated in podocytes. Together, these data suggest that increased Rho-kinase activity in the podocyte may be a mechanism for in vivo podocyte foot process retraction.
- immune complex glomerulonephritis
- mouse model
- rho-kinase activity
- fasudil (HA-1077)
- podocyte injury
the glomerular filtration barrier consists of capillary endothelial cells, the glomerular basement membrane, and glomerular epithelial cells (podocytes). Podocytes provide the crucial barrier to urinary protein loss with hundreds of podocyte foot processes and their interposed slit diaphragm (43). Most proteinuric renal diseases are typically, although not exclusively, associated with podocyte damage. Disruption of podocyte morphology in form of foot process retraction (effacement) or slit diaphragm reorganization leads to massive loss of protein into the urine (proteinuria) and nephrotic syndrome. The signaling pathways that contribute to foot process effacement are not well understood. The cytoskeleton of podocytes consists of actin- and intermediate filaments and microtubules. Whereas the podocyte foot processes contain three major proteins of the contractile apparatus in muscle, i.e., actin, myosin, and the Z-line protein α-actinin-4, the podocyte cell body and major processes contain intermediate filaments and microtubules that are absent from foot processes (12). During podocyte cytoskeletal rearrangement, these proteins act together to retract foot processes, round up the podocyte cell body, and loosen the podocyte glomerular basement membrane connection.
The Rho-family small GTPases Rho (5), CDC42 (47), and Rac-1 (6) are effective mediators of the arborized podocyte phenotype (28). The activation of Cdc42 and Rac1 signaling in podocytes is thought to cause foot process effacement and proteinuria. Rho-kinases are one of the major molecular switches downstream of RhoA controlling several critical cellular functions, such as actin cytoskeleton organization, cell migration and adhesion, reactive oxygen species formation, and apoptosis (11, 37). Rho-kinases play a pivotal role in the development of experimental renal diseases such as ischemia and reperfusion injury (21), diabetic nephropathy (23), ureteral obstruction (30), and hypertensive kidney injury (19). Rho-kinases 1/β and 2/α (Rho-associated kinases/ROK/ROCK) are serine/threonine protein kinases with 64% amino acid sequence homology (90% in the kinase domain) and ubiquitous distribution (2). Rho-kinases control cytoskeletal rearrangement in leukocytes, endothelial and epithelial cells, and many others cell types (32, 40). Several specific Rho-kinase effectors such as ezrin/radixin/moesin (ERM), adducin, LIMK-1, or the regulatory subunit of the myosin-specific phosphatase have been identified (2). Upon activation, these effectors contribute to stress fiber formation and cell body contraction by various mechanisms (1, 14, 22). Many of the known effects of Rho-kinases on cytoskeletal rearrangement have been found by Rho-kinase inhibitor experiments. The Rho-kinase inhibitors HA-1077 (fasudil, 5-Isoquinolinesulfonyl-homopiperazine) or Y-27632 [(R)-(+)-trans-N-(4-pyridyl)-4-(1-amino-ethyl)-cyclohexanecarboxamide] inhibit Rho-kinase potently and specifically (31, 41, 45). Inhibition of the Rho-Rho-kinase pathway decreases myosin light chain (MLC) phosphorylation and actin stress fiber formation (31, 45) and thereby abolishes the ability for cytoskeletal rearrangement.
In podocytes, studies by Gao et al. (14) and Kobayashi et al. (22) demonstrated a significant role of the Rho/Rho-kinase signaling pathway in the reorganization of both microtubules and actin filaments during process formation in vitro. In their studies, Rho/Rho-kinase inhibition enhanced the formation of thick processes containing microtubular bundles and actin filament based thin projections, suggesting that downregulation of Rho-kinase activation allowed microtubule and actin assembly. Studies with Rho-kinase inhibitors in renal disease models demonstrated strong anti-inflammatory and antifibrotic effects (18, 33) and were shown to decrease proteinuria in puromycin- or daunorubicin-treated mice and rats, respectively (10, 46). These in vitro data suggested a positive effect of Rho-kinase inhibition in podocyte-dependent disease. However, a direct effect of Rho-kinase inhibition on proteinuria has not been shown in immune-mediated glomerulonephritis. Immune-complex- or primary antibody-mediated damage to podocytes has been implicated in several different human glomerular diseases that are associated with cytoskeletal rearrangement and foot process effacement, such as membranous glomerulonephritis (38), renal lupus erythematodes (48), and idiopathic nephrotic syndrome (29). We therefore hypothesized that increased activation of Rho-kinases leads to cytoskeletal rearrangement in podocytes in the course of antibody-mediated podocyte injury, culminating in foot process retraction, proteinuria, and podocyte loss into the urine and that this could be prevented by Rho-kinase inhibition. To test this hypothesis, we recently developed a mouse model of immune-mediated podocyte injury, in which a polyclonal antipodocyte antiserum was used to induce podocyte injury and proteinuria in mice (25, 26). Here, we show for the first time, that Rho-kinase inhibition prevented proteinuria and podocyte disruption in vivo in a model of immune complex-mediated glomerulonephritis.
Wild-type male C57BL/6N mice (Charles River, Wiga, Sulzfeld, Germany) and Sprague-Dawley rats were bred in house under flow sterile conditions. Animals had free access to water and standard animal chow. The animal experiments were performed according to National and Institutional Animal Care and Ethical Guidelines and were approved by the Veterinarian Agency of Hamburg and the Local Animal Care Committee (82/06 and 107/09).
Murine model of antipotocyte nephritis.
In four separate experiments with an n = 5–15 for each condition and time point, a total of 105 male C57BL/6N mice (25–30 g body wt, 12–14 wk old) were treated. Antipodocyte nephritis was induced by application of 300 μl sheep antipodocyte serum per intravenous injection into the tail vein as described in detail by Meyer et al. (26). Control mice were given 300 μl preimmune serum. Urine was evaluated for proteinuria on days 0, 2, 5, 8, and 14. Mice were weighed and killed at days 4, 9, and 14 by cervical neck dislocation. Following the collection of blood, kidneys were perfused through the aorta with PBS. Kidneys were cut into equal pieces for the isolation of mRNA and protein and for histology and electron microscopy.
Rat model of membranous nephropathy.
Passive Heymann nephritis (PHN) was induced in 200- to 250-g male Sprague Dawley rats by intravenous injection of 500 μl (t1) and 750 μl (t0) of a sheep anti-FX1A antiserum or control sheep preimmune serum. Anti-FX1A antiserum was induced by repeated immunizations of sheep with isolated brush borders of proximal tubuli from rat kidneys as described previously (40a). Kidneys were harvested at 18 days.
Rho-kinase was inhibited with the well-established Rho-kinase inhibitor HA-1077. Treatment was commenced 24 h before administration of antipodocyte serum. HA-1077 was administered at 500 μg in 250 μl H2O per mouse by gavage two times a day until the end of the experiment. Control animals received 250 μl H2O twice daily. In a different set of experiments, mice were continuously treated with HA-1077 (1 mg/24 h) by subcutaneous osmotic minipumps (2002, 12 μl/24 h; Alzet, Cupertino, CA) for 14 days. Gavage and HA-1077 were clinically well tolerated by the mice.
Isolation of glomeruli.
For the isolation of glomeruli from mice, mice were perfused with Dynabeads (Dynal, Oslo, Norway) as described previously (17, 44). For the isolation of glomeruli, rats were perfused with PBS. Glomeruli were isolated by conventional sieving. Glomeruli were homogenized in T-PER lysis buffer (Pierce, Bonn, Germany). After centrifugation (15,000 g for 15 min at 4°C), protein concentration was determined by BCA protein-assay (Bio-Rad). Equal amounts of protein were separated by SDS-PAGE.
Blood pressure measurement.
Systolic blood pressure was measured using tail-cuff impedance plethysmography (TSE Systems, Bad Homburg, Germany) in trained mice for 5–10 min without anesthesia. Blood pressure measurements were performed in antipodocyte nephritis (APN) mice and APN mice pretreated with HA-1077 (n = 5 for each condition).
Analysis of proteinuria.
Proteinuria was assessed by SDS PAGE of mouse urine collected over 6–8 h in 96-well plates. Five microliters of urine were loaded in 4× LDS sample buffer onto a 4–12% graded Bis-Tris NuPage Gel (Invitrogen, Karlsruhe, Germany) and run in 50 mM MOPS buffer for 1 h at 60 mA. Following electrophoretical separation, proteins were visualized by GelCode (Pierce, Bonn, Germany) according to the manufacturer's instructions and scanned.
Urine albumin content was quantified using a commercially available ELISA system (Bethyl) according to the manufacturer's instructions as described previously (25). Briefly, 96-well plates were coated 1:100 with goat anti-mouse albumin in binding buffer (0.05 M carbonate-bicarbonate pH 9.6) for 60 min at room temperature. After washes in 50 mM Tris, 0.14 M NaCl, 0.05% Tween-20 pH 8.0, the plates were blocked for 30 min at room temperature with 50 mM Tris, 0.14 M NaCl, and 1% BSA pH 8.0 and rewashed. Diluted urine was incubated for 60 min at room temperature. Following washes, the secondary antibody [1:20,000; horseradish peroxidase (HRP) goat anti-mouse albumin] was applied for 60 min at room temperature. After washes, enzyme substrate supplied by the kit was added and the color development was stopped after 20 min with 2 M sulphuric acid. Extinction was measured at 450 nm in an ELISA plate reader (BioTek, EL 808). The urinary albumin concentration was calculated according to the formula for absorption = (A − D)/1 + (x/C)B + D, where A and D are values from the standard curve. Regression values for the standard curve were calculated to assess the accuracy of the measured values. Standard curves with r values > 0.9950 were used. The urinary albumin values were standardized against urinary creatinine values of the same animals.
Immunoblots against Rho-kinase 1 and 2 (both Transduction Laboratories; 1:1000) and β-actin (Sigma, Steinheim, Germany; 1:3,000) as a loading control were performed in preimmune or antipodocyte serum-treated mice with or without Rho-kinase inhibition. Lysis of kidneys was performed in 100 μl/mg tissue in lysis buffer (T-PER; Pierce), followed by denaturation in LDS sample buffer (Invitrogen). Protein lysates were separated on a 4–12% Bis-Tris NuPage gel (Invitrogen) in 50 mM MOPS buffer for 1 h at 60 mA. Protein transfer to an activated PVDF membrane (Millipore, Eschborn, Germany) was performed in transfer buffer (25 mM Tris base, 0.192 M glycine, and 20% methanol in ddH2O) in a Novex Mini Cell (Invitrogen) for 1.5 h at 30 V and 4°C. The PVDF membranes were blocked in 5% NFM for 60 min before incubation with primary antibodies diluted in Superblock blocking reagent (Pierce). Primary antibody was removed by washing in TBS-Tween for 40 min and detected by incubation for 45 min with HRP-coupled secondary antibodies (Jackson Immunoresearch Laboratories, Hamburg, Germany) diluted 1:15.000 in 5% nonfat milk. After being washed in TBS-Tween for 40 min, protein was visualized with ECL SuperSignal (Pierce) according to the manufacturer's instructions on a Biomax Light Film (Kodak).
Human specimen collection.
The study was conducted according to the Declaration of Helsinki principles with approval from the Local Ethics Committee. Tissue samples were retrieved from the renal biopsy program of the Department of Pathology and Neuropathology, University of Hannover.
Whole kidneys were perfusioned and postfixed in 4% buffered formalin for 24 h and embedded in paraffin (Medim Histotechnologie, Buseck, Germany). For immunohistochemical stainings, 1-μm tissue sections were deparaffinized and rehydrated. Antigen retrieval was performed by incubation with proteinase type XXIV (5 mg/ml; Sigma) for 15 min at 37°C or by microwave antigen retrieval in citrate buffer, pH 6.1, for 25 min. After a 30 min block in 5% horse serum and 1:25 avidin D (avidin biotin blocking kit; Vector) and two short washes in PBS, the tissue was incubated with nephrin (guinea pig 1:50; Progen), synaptopodin (mouse mAB 1:50; Progen with M.O.M. kit; Vector), phospho-ERM (rabbit polyclonal; Cell Signaling; 1:400), phospho-MLC (rabbit polyclonal; Cell Signaling; 1:200), or affinity purified Cy2 mouse or sheep IgG (donkey; Jackson Immunoresearch Laboratories; 1:200) in 5% horse serum with 1:25 biotin for 2 h at room temperature or overnight at 4°C. Binding was visualized using biotinylated secondary antibodies (all Jackson Immunoresearch Laboratories) diluted 1:400 in 5% horse serum for 30 min. Stainings were evaluated under an Axioskop (Zeiss, Jena, Germany) and photographed with an Axiocam HRc (Zeiss) or by confocal microscopy with a LSM 510 beta microscope using the LSM software.
Electron microscopical analysis was performed on whole kidneys that were perfusion fixed in 4% buffered paraformaldehyde. Tissue was postfixed with 1% osmium in 0.1 M sodium-cacodylat buffer, stained with 1% uranylacetate, and embedded in epoxy-resin (Serva, Heidelberg, Germany). Ultrathin sections were cut (ultramicrotome; Reichert-Jung) and contrasted with uranyl acetate in methanol followed by lead citrate. Micrographs were generated with a transmission-electron microscope (JEM 1010; JEOL).
Measurement of glomerular area and podocyte number.
Kidney cortex was embedded in paraffin for light microscopy examination. One-micrometer-thick sections were cut on a rotation microtome and stained with periodic acid shift reagent (Sigma). Planimetric examinations of glomerular cross-sectional area were performed at magnification of ×200 by means of a Zeiss drawing tube in combination with a semiautomatic interactive image analysis system (Morphomat 30; Zeiss) as described previously (25). With the use of a serpentine movement from cortex to medulla and vice versa, the outlines of 50 consecutively encountered capillary tufts were traced manually and the mean glomerular random cross-sectional area (AG) was determined.
For the evaluation of podocyte number per mean glomerular cross-sectional areas, 3-μm-thick paraffin sections were stained for WT-1 (Santa Cruz; 1:800), a specific marker of podocyte nuclei. WT-1 positive nuclei were counted in 50 glomeruli systematically sampled at ×200 magnification to allow calculation of the podocyte nuclei number per mean glomerular cross-sectional area.
Total Rho-kinase activation assay.
For measurement of total Rho-kinase activity, kidney cortex or isolated glomeruli were lysed in T-Per (Pierce) according to the manufacturer's instructions as described before (27). For the activity assay, 500 ng myosin phosphatase target subunit 1 (MYPT1; 654–880; Upstate Biotechnology, Lake Placid, NY) and Tris/ATP (final concentration: 1 mM ATP, 50 mM Tris·HCl, 0.1 mM EGTA, 0.1 M DTT, and 10 mM MgCl2) were added to the lysates, and recombinant Rho-kinase 2 protein (MBL, Woburn, MA) was added for the positive control. The lysates were incubated for 30 min at 30°C. The reaction was stopped by addition of 4× LDS and by subsequent heating of the assays for 10 min at 72°C. Phosphorylation of MYPT1 by Rho-kinase was detected by running the assays on a SDS PAGE and by subsequent immunoblotting with an anti phospho-MYPT1 (Thr850; Upstate) 1:5,000 in 3% nonfat milk followed by the secondary antibody anti-rabbit HRP (1:15,000; Jackson Immunoresearch Laboratories) in 3% nonfat milk. As a loading control, internal MYPT1 was detected by immunoblotting with an anti-MYPT1 (BD Biosciences, Heidelberg, Germany) followed by the secondary antibody anti-mouse HRP (1:15,000; Jackson Immunoresearch Laboratories) in 5% nonfat milk (not shown).
Values are ± SE; n refers to the number of animals or individual measurements in separate samples. Differences between multiple experimental groups were compared by Kruskal-Wallis analysis with post hoc analysis, and single groups were compared by the Mann-Whitney U-test. Paired Student's t-test was used to compare mean values within one experimental series. A P value of 0.05 was accepted as statistically significant.
Rho-kinase is activated in podocytes in the course of antipodocyte nephritis in mice.
Five separate sets of experiments were performed with a total of 130 mice. Mice were pretreated for 24 h with the Rho-kinase-specific inhibitor HA-1077 (500 μg) per gavage or water before a single intravenous administration of 300 μl sheep antipodocyte serum (APN) or preimmune serum (control) as described before (Fig. 1A; see also Ref. 26). Along the time course, 500 μg HA-1077 were given by gavage twice per day to ensure continuous Rho-kinase inhibition. Control animals were gavaged with water. In a separate set of experiments, mice were continuously treated with HA-1077 (1 mg/24 h) by subcutaneous osmotic minipumps for 14 days, again treatment was commenced 24 h before administration of APN serum. Total Rho-kinase activity was measured in lysates of kidney cortex on day 14 with a Rho-kinase activation assay employing the Rho-kinase-specific effector target on MYPT1 (Fig. 1B). Cortical Rho-kinase activity was strongly upregulated in antipodocyte serum-treated mice compared with preimmune serum-treated mice and mice treated with preimmune serum and HA-1077. Increased Rho-kinase activity was attenuated to normal to low levels by HA-1077 treatment. Western blots against Rho-kinase 1 and 2 from the same cortical lysates showed stable Rho-kinase 1 levels, whereas Rho-kinase 2 increased in antipodocyte serum-treated mice compared with preimmune serum-treated mice. Treatment with HA-1077 lead to an increase in Rho-kinase 1 and Rho-kinase 2 levels in cortical lysates of preimmune and antipodocyte serum-treated mice. In a subsequent experiment, glomeruli were isolated with magnetic beads and glomerular lysates were examined by Rho-kinase activation assay (Fig. 1C). Glomerular Rho-kinase activity was significantly upregulated in APN mice compared with preimmune serum and preimmune serum + HA-1077-treated mice and increased Rho-kinase activity was abolished by HA-1077 in APN + HA-1077-treated mice.
Glomerular Rho-kinase activation was further localized by immunostaining against the Rho-kinase-specific phosphorylation sites [ezrin(Thr567)/radixin(Thr564)/moesin(Thr558)] on the cytoskeletal Rho-kinase substrate ERM and MLC on Ser19 on day 9 (Fig. 1D). Ezrin is expressed in podocyte foot processes where it complexes with podocalyxin, which mediates its link to the actin cytoskeleton (34). In preimmune serum-treated mice, glomerular ERM phosphorylation could be detected in a linear pattern surrounding glomerular capillaries. In antipodocyte serum-treated (APN) mice, ERM phosphorylation increased in podocytes (Fig. 1D, c). This increase of ERM phosphorylation was partly reversed in Rho-kinase-inhibited APN mice (Fig. 1D, d). Rho-kinase-specific phosphorylation of MLC at Ser19 is normally detected in the muscle wall of renal arteries and arterioles, but only little activity is detected in normal podocytes. However, MLC phosphorylation strongly increased in podocytes of APN mice (Fig. 1D, g), which was abolished by Rho-kinase inhibition (Fig. 1D, h) similar to ERM phosphorylation.
We analyzed whether Rho-kinase activation was a general pathological feature of proteinuric podocyte injury in membranous nephropathy. Glomerular Rho-kinase activity was measured in PHN, an established rat model of membranous nephropathy. Similar to antipodocyte nephritis, Rho-kinase activity was stronger in glomeruli of PHN rats (Fig. 2A). The measured increase of Rho-kinase activity localized to podocytes by immunohistochemistry against phospho-ERM and phospho-MLC (Fig. 2B). In addition, human biopsy samples of patients with IgG4-positive membranous nephropathy demonstrated a strong increase in phospho-ERM and phospho-MLC immunostaining in podocytes compared with normal control biopsies (Fig. 2C).
Taken together these data demonstrate an increase of glomerular Rho-kinase activity in podocytes during immune-complex nephritis that can be attenuated by Rho-kinase inhibition.
Proteinuria is decreased by Rho-kinase inhibition.
We next used Coomassie-stained urine gels to analyze whether Rho-kinase inhibition in antipodocyte serum-treated mice influenced proteinuria (Fig. 3A). Antipodocyte serum-treated mice developed massive proteinuria beginning around day 8 after disease induction whereas Rho-kinase-inhibited mice showed only small amounts of urinary protein. Preimmune serum or preimmune serum + HA-1077-treated mice showed no proteinuria (not shown). This finding was quantitated by urine albumin ELISA, showing similar results (Fig. 3B): Rho-kinase inhibition with HA-1077 in antipodocyte serum-treated mice significantly reduced the massive albumin loss into the urine of antipodocyte serum-treated mice on day 8 and 14; however, baseline values of preimmune serum-treated mice were not reached. Preimmune serum-treated mice with and without Rho-kinase inhibition did not develop proteinuria. In antipodocyte serum-treated mice, nephrotic syndrome lead to edema and ascites and an increase of body weight (Fig. 3C). HA-1077-treated antipodocyte serum-treated mice had significantly less body weight on day 14 compared with untreated APN mice. To account for possible systemic effects of HA-1077, the blood pressure from antipodocyte serum-treated mice with or without Rho-kinase inhibition was measured by tail-cuff impedance plethysmography (Fig. 3D). No significant changes were recorded between the two groups, indicating that the effect of Rho-kinase inhibition on antipodocyte serum-treated mice was unrelated to blood pressure.
Glomerular and podocyte hypertrophy and podocyte loss were decreased by Rho-kinase inhibition.
Podocytes are the primary target of the injected antipodocyte serum and were found to be severely injured in antipodocyte nephritis (26). Antipodocyte serum causes podocyte and mesangial swelling by deposition of subepithelial immune complexes (Fig. 4A, c; see also Ref. 26). Glomerular size increases and podocytes are lost from the glomerular tuft. We therefore analyzed the glomerular pathology in Rho-kinase-inhibited antipodocyte serum-treated mice. In periodic acid stainings, glomeruli and podocytes of Rho-kinase-inhibited mice were less swollen and less vacuolated in the course of disease (Fig. 4A, d). Tubuli exhibited less cytoplasmatic vacuoles suggestive of reduced protein loss into the tubular lumen. Glomerular size and podocyte number were determined in the four experimental groups (Fig. 4B). Glomeruli of antipodocyte serum-treated mice were significantly larger than of preimmune serum-treated mice. Rho-kinase inhibition decreased overall glomerular size of antipodocyte serum-treated mice significantly; however, glomeruli were still significantly larger than the glomeruli of preimmune serum- and HA-1077-treated mice. Podocytes were stained with WT-1 and counted per glomerular section in all four groups. Total podocyte number significantly decreased in antipodocyte serum-treated mice compared with preimmune serum-treated mice as shown previously (25, 26). Interestingly podocyte number was significantly less reduced in Rho-kinase-inhibited mice receiving antipodocyte serum, indicating that podocytes were protected from detachment by Rho-kinase inhibition. However, podocyte number was still significantly reduced in these mice compared with preimmune serum + HA-1077-treated mice.
The injected antipodocyte antibody was found in a linear staining pattern along the glomerular filtration barrier where it is thought to bind to podocyte antigen(s), thereby inducing mouse IgG and complement C3 deposition (25, 26). To test whether sheep and mouse IgG deposition differed in Rho-kinase-inhibited mice, both proteins were visualized by immunostaining (Fig. 4C, a–h). However, sheep- and mouse-IgG deposition was similar in glomeruli of antipodocyte serum injected mice 14 days after disease induction. A linear staining pattern was observed along the glomerular filtration barrier independent of Rho-kinase inhibition. Preimmune serum-injected mice demonstrated no sheep or mouse IgG deposition along the glomerular filtration barrier. Immunoblots quantitated mouse-IgG content in isolated glomeruli in all four groups (Fig. 4D). Whereas preimmune serum-injected mice demonstrated no glomerular mouse IgG, glomeruli of mice treated with antipodocyte serum with or without HA-1077 contained comparable amounts of mouse IgG.
Cytoskeletal rearrangement in podocytes was prevented by Rho-kinase inhibition.
Since podocyte loss was reduced and the activity of cytoskeletal regulators such as ERM and MLC was found to be decreased to normal levels by Rho-kinase inhibition in antipodocyte serum-treated mice, we analyzed the extent of cytoskeletal rearrangement by transmission electron microscopy and by immunofluorescence against slit membrane proteins (Fig. 5). Analysis of electron micrographs of antipodocyte serum-treated mice demonstrated severely swollen and vacuolated podocytes with 60–80% foot process effacement. In the subepithelial space, immune deposits were observed (Fig. 5A, f). Rho-kinase inhibition reduced foot process effacement to 20–30%, and podocytes were less hypertrophied and vacuolated (Fig. 5A, g and h). Interestingly, immune deposits were found in a comparable amount in the subepithelial space of antipodocyte serum-injected mice despite Rho-kinase inhibition.
Confocal analysis of the slit membrane protein nephrin and the foot process protein synaptopodin localized both proteins to podocytes in a crisp linear staining pattern independent of Rho-kinase inhibition in preimmune serum-treated mice (Fig. 5B). In antipodocyte serum-treated mice, nephrin and synaptopodin staining was severely disrupted. The linear staining pattern was replaced by a fragmented and granulated, cytoplasmic staining pattern indicative of partial internalization of both proteins from the slit membrane to the cytoplasm. In Rho-kinase-inhibited antipodocyte serum-treated mice, nephrin and synaptopodin staining was found in a normal linear staining pattern along the slit membrane and also in the cytoplasm of podocytes (Fig. 5B, d and h). Together, electron micrographs and immunostainings demonstrated that cytoskeletal rearrangement was partly but not completely prevented by Rho-kinase inhibition in antipodocyte serum-treated mice.
Podocyte foot process effacement is a hallmark of nephrotic kidney diseases such as focal segmental glomerulosclerosis or membranous glomerulonephritis and is thought to be caused by cytoskeletal rearrangements in injured podocytes (4). However, the signaling mechanisms that cause this severe alteration of morphology and function are only little understood and specific therapies preventing or treating podocyte foot process retraction are not established.
Evidence of Rho-kinase-mediated cytoskeletal arrangement has been found ubiquitously in cells that contain an actin/microtubule organized cytoskeleton (2). Highly arborized cell types such as brain neurons or glomerular podocytes depend on Rho-kinase activity for proper function and morphology (8, 42). However, further Rho-kinase activation under conditions of cellular stress causes cytoskeletal rearrangement, stress fiber formation and mostly loss of cellular integrity and function (16). Activation of Rho-kinases and cytoskeletal rearrangement (stress fiber formation) had been reported previously in cultured podocyte under sheer stress (13). Rho-kinase inhibition prevented these changes and was also shown to enhance process formation in culture (14).
In this work, we therefore tested the hypothesis that podocytes would activate Rho-kinases under stress in vivo, which would lead to foot process retraction and ultimately proteinuria. Further, Rho-kinase inhibition would prevent podocyte cytoskeletal rearrangement and proteinuria. We tested this hypothesis in a new mouse model of immune-mediated glomerulonephritis, in which podocytes are directly targeted by antipodocyte serum, which leads to heavy proteinuria (26). The benefit of this model is that podocytes are targeted by local inflammation from subepithelial immune-complex deposits and not by chemical assault by puromycin induced podocyte injury (46; Smeets, 2003 no. 4676). Podocyte injury in antipodocyte nephritis is characterized by podocyte hypertrophy and podocyte foot process effacement with internalization of slit membrane proteins leading to nephrotic syndrome in mice (25, 26). Inhibition of in vivo Rho-kinase activity with the reversible and specific inhibitor HA-1077 was used because this inhibitor is well established in rodent models (9). HA-1077 or another specific Rho-kinase inhibitor, Y-27632, has been successfully used in rats or mice to reduce renal fibrosis in a model of unilateral ureter obstruction (30) and reduce nephropathy and proteinuria in spontaneously hypertensive rats (19), angiotensin II-induced hypertension (39), diabetic nephropathy (15, 23), and ischemia reperfusion kidney injury (36). HA 1077 was highly effective in reducing podocyte loss, cytoskeletal rearrangement, and proteinuria in our model.
We used the phosphorylation of the Rho kinase effector proteins ERM and MLC as indicators of local Rho-kinase activation in podocytes in antipodocyte nephritis, rat Heymann nephritis, and human membranous nephropathy. Ezrin and MLC rearrange the cytoskeleton upon specific phosphorylation and are expressed in podocytes (34). Ezrin links the antiadhesin podocalyxin to the actin cytoskeleton through association with the cytoplasmic tail of podocalyxin and by this regulates podocyte morphology. Interestingly, once formed, ezrin/podocalyxin complexes were stable and insensitive to actin depolymerization or inactivation of Rho-kinases (34). Interestingly, several isoforms of myosin heavy chain are present in podocytes and rare mutations in myosin heavy chain 9 (MYH) had been associated with familial forms of focal glomerulonephritis with foot process effacement (3). In addition, several single nucleotide polymorphisms on MYH9 have been recently identified by two independent groups as risk loci for FSGS in African Americans (20, 24). MYH9-associated risk of kidney injury was later also established in Europe (35). The pathomechanism by which MYH9 disregulation injures podocytes has not been established yet. It is therefore intriguing to speculate on a Rho-kinase-associated activation of ERM proteins and myosin light and heavy chain proteins in podocytes that may introduce prolonged stress fiber formation, shape change, and loss of function under pathophysiological conditions such a membranous nephropathy.
Nephrin and synaptopodin redistribution from the foot process to the cytosol was prevented in Rho-kinase-inhibited antipodocyte nephritic mice. Nephrin is a slit membrane protein that has not been associated with Rho-kinase signaling. We therefore consider the redistribution of nephrin to be secondary to Rho-kinase-driven actin cytoskeletal reorganization. Synaptopodin, however, is a podocyte foot process protein that associates with actin and α-actinin-4 and was recently identified as a regulator of RhoA signaling in podocytes (5). Synaptopodin induces stress fibers by increasing RhoA levels by reducing ubiquitination and targeting of RhoA for proteasomal degradation (5). Presumably, Rho-kinase activity was increased through RhoA in these podocytes. Therefore, it is possible that increased Rho-kinase activity in APN mice was induced by synaptopodin, despite internalization. The prevention of synaptopodin internalization in Rho-kinase-inhibited injured podocytes may therefore reflect the inhibition of synaptopodin-triggered Rho-kinase activation and stress fiber formation. It will be interesting to identify the mechanism of synaptopodin activation in the future. Alternatively, Rho-kinase may have been activated by a non-RhoA-synaptopodin-dependent signaling pathway such as caspase-3 cleavage (7). However, such activation has not been described in the podocyte yet.
In summary, these data summarize the first in vivo demonstration of increased Rho-kinase activity in podocytes after immune-mediated glomerular injury. Podocyte disruption was attenuated by Rho-kinase inhibition, indicating that Rho-kinase activation is essential in podocyte effacement in immune-complex-mediated glomerular disease.
This study was supported by the Deutsche Forschungsgemeinschaft Grant Me 1760/4-1 and Me 1760/4-2.
No conflicts of interest, financial or otherwise, are declared by the author(s).
Author contributions: C.M.-S., S.D., F.T., and T.N.M. conception and design of research; C.M.-S., S.D., M.S., S.M., K.A., S.G., and J.U.B. performed experiments; C.M.-S., S.D., S.M., K.A., S.G., J.U.B., and T.N.M. analyzed data; C.M.-S., S.M., S.B., F.T., and T.N.M. interpreted results of experiments; C.M.-S. and T.N.M. prepared figures; C.M.-S. and T.N.M. drafted manuscript; C.M.-S. and T.N.M. edited and revised manuscript; C.M.-S., S.D., M.S., S.M., K.A., S.G., S.B., J.U.B., F.T., and T.N.M. approved final version of manuscript.
We thank Mariola Rezska for excellent help with histology preparations.
- Copyright © 2012 the American Physiological Society