The Expression of RANK-RANKL in Injuried Podocyte and Involved Mechanisms
|School||Southern Medical University,|
|Keywords||podocyte RANK RANKL puromycin aminonucleoside apoptosis|
BACKGROUNDPodocytes are terminally differentiated cells that line the outer aspect of the glomerular basement membrane (GBM). Podocyte dysfunction, injury, or loss is a common and determining factor in glomerular diseases. Extensive experimental and clinical literature points to the importance of podocyte injury in the development and progression of glomerular disease. Podocytes are believed to be the primary target of glomerular damage in so-called podocytopathies (minimal change disease (MCD), focal segmental glomerulosclerosis (FSGS), and membranous nephropathy (MN)), but they are also damaged in glomerular diseases of mesangial proliferation, including IgA nephropathy and lupus nephritis.Podocytes are injured by immune-and non-immune-mediated disease, resulting in damage to the glomerular filtration barrier. The fate of the podocyte then depends on several factors, such as reparative mechanisms. If these are present, and/or the initial injury is halted, there may be resolution. However, if injury persists, and/or there are inadequate repair mechanisms present, proteinuria persists, leading to reduced renal function. During the injury process, there is a critical period of coordinated gene expression that determines whether podocytes survive or are lost. In response to injury, podocytes secrete antioxidant enzymes, and abnormal proteins, such as desmin and glial cell line-derived neurotrophic factor (GDNF), which is a survival growth factor for injured podocytes. However, the pathogenesis of podocyte injury is not quite clear. We postulated that there are other survival factors that are expressed in response to podocyte injury, and act to support the recovery of injured podocytes.TNF and receptor superfamilies are important in the pathogenesis of podocyte injury and apoptosis. The ligand of receptor activator of NF-kappaB (RANKL) is a member of the TNF family that is produced by osteoblasts, myocardial  and stromal cells. RANKL is not only a transmembrane molecule but also secreted, and this particularly by activated T cells. Receptor activator of NF-kappaB (RANK) is a cognate receptor which is expressed by osteoclast-like cells (OCLS). We don’t know whether RANK-RANKL is a receptor-ligand complex for pathogenesis of podocyte injury and apoptosis. Several studies suggest that RANKL and RANK are involved in cell survival and apoptosis. RANK expression has been shown to suppress endothelial cell apoptosis through its activation by RANKL. In human myocardial cells, RANK-RANKL gene expression is upregulated by allergens and irritants. Little is known about RANK-RANKL expression in non-bone marrow-derived cells. RANKL is expressed in the renal glomeruli, convoluted tubules, and parenchyma of the developing fetal kidney, whereas RANKL is not detected in adult kidney. However, no studies have addressed the functional role of RANK-RANKL in normal renal physiology, or in glomerular disease. We report here for the first time that RANKL and RANK are induced significantly in animal models of podocyte injury. Moreover, RANKL, acting through RANK, is a potent survival factor for injured podocytes and promotes protection from injury.METHODS 1. Construction of podocyte injuried rat model:thirty male Sprague-Dawley rats were randomly divided into two groups:PAN model group (PAN) and normal control group (Con). PAN rat models were constructed by a single intravenous injection PAN of100mg.kg-1body weight and the Con group rats were injected equal volume physiological saline.2. Evaluation of24-hour urine protein and the renal tissue morphology:The rats were sacrificed at day3,7,14after PAN injection.24-hour urinary protein excretion was determined using utilizing traditional Coomassie Brilliant Blue. Serum creatinine (Cr) was determined using commercial kits (Beckman Coulter, USA) by an autoanalyzer. The renal tissue morphology was observed by light and electron microscope.3. Evaluation of the expression of Nephrin, RANK and RANKL:the expression of Nephrin, RANK and RANKL mRNA were evaluated by RT-PCR. The protein expression and location of Nephrin, RANK and RANKL were decided by confocal microscope.1.36Male Sprague-Dawley rats were grouped subtotal nephrectomy control group (n=12), and sham-operated group (n=12). After subtotal nephrectomy in treatment group, the SD rats were follow up.2. At4,8and12weeks after nephrectomy, the kidney was examined respectively by light microscopy. RANK and RANKL expression was examined by histochemisty, Western blot and RT-PCR in rat nephrectomy kidney. 1. We analyzed expression of RANK in normal kidney tissues (n=3), focal segmental glomerulosclerosis (FSGS, n=12), IgA nephropathy (n=14), and membranous nephropathy (MN, n=16), the latter three groups are with proteinuria.1. Cultivation and induction of conditionally immortalized podocyte: Conditionally immortalized mouse podocyte friendly present as a gift by Professor Danesh of Baylor Medical College. Podocytes were grown in RPMI1640mediumin in bottle coated with type Ⅰ collage that contained10%FBS. For passaging cells, podocytes were grown under "growth permissive" conditions, which involved growing cells at33℃in the presence of20u/ml IFN-γ. For podocytes to acquire a differentiated and quiescent phenotype, cells were grown in DMEM mediumin under "restrictive conditions" at37℃in95%air/5%CO2without IFN-γ for>12d. Cells expressed with nephrin and synaptopodin are a differentiated and quiescent phenotype podocytes.2. Expression of RANK in cultured differentiated podocytes. We used cultured differentiated podocytes, which express specific podocyte Synaptopodin proteins, to investigate whether RANK was involved in podocyte injury with PAN. Mouse podocytes were exposed to PAN (25ug/ml), which is known to induce podocyte injury in rodents.3. Silence of RANK gene:Differentiated podocytes inoculated coverslip which coated with type collagen and cultivated without serum for12h. And then podocytes transfected with RANK siRNA by Lipofectamine. Effect of RANK silence detected with Western blot and real-time-PCR after transfected24-48h. 4. To evaluate the expression of RANK+/+and RANK-/-cells without PAN and stimulated with PAN as well as changes of podocyte cytoskeleton protein F-actin by immunofluorescence.5. RANKL and RANK protects mouse podocytes from apoptosis. To better understand RANK function in podocytes, we used an in vitro RANK knockdown system to determine whether RANK was necessary for podocyte survival.6. To investigate whether the decrease in the [Ca2+]i transient induced by RANKL was through L-type calcium channels, we used the whole cell patch-clamp technique to test the effect of RANKL on the calcium current. To determine further the mechanisms whereby RANKL decrease [Ca2+]i in podocytes, we studied endoplasmic reticulum Ca2+-ATPase activity after PAN exposure in the presence or absence of RANKL. Ca2+-ATPase activity was measured by an optical assay in crude endoplasmic reticulum extracted from control and RANKL-treated podocytes.1) All data were expressed as mean±SEM. Continuous variables between groups at each time point were compared using Analysis of Variance. Significance was defined as p<0.05RESULTS1. Up-regulation of RANK expression by PAN introduction.1) PAN administration caused heavy proteinuria and hypoproteinemia and as well as loss of renal function--a classics nephritic syndrome symptoms. Inflammatory cell infiltration, a devil of a protein cast, renal interstitial edema, partly renal tubule atrophy and fibrosis and focal segmental gloumerular sclerosis was obvious in pathology.2) RANK expression was increased in PAN model rats. PAN induced by a single injection of PAN was characterized by an increase in urine protein levels at7 days of follow-up (135±29versus6.15±0.68mg/24h; p<0.01). We found that rats with PAN had markedly elevated levels of RANK protein in the kidney compared with controls, with persistently raised levels throughout the observation period. The levels of RANKL protein increased in PAN rat kidney as compared with control. RANK was localized in podocytes, as indicated by co-labeling with Synaptopodin, a marker of podocyte. Compared with controls, immunofluorescence staining for RANK was increased in PAN, and double immunofluorescence staining revealed that the increase in RANK protein expression was mainly attributed to its increase in podocytes. We also found low expression of RANK in proximal tubular cells that appeared unchanged under normal and disease conditions (data not shown). To define the subcellular localization of RANK within podocytes, we performed immunogold analysis of RANK in glomerular walls of normal and PAN rats. Under normal conditions, low RANK expression was observed in podocytes, but increased RANK labeling was located on the top of the foot process membrane and the cytoplasm of podocytes in PAN rats. Significantly elevated RANKL mRNA levels were observed in PAN rats after7days (2.6±0.4-fold higher than controls, n=6; p <0.01).1) At12weeks after nephrectomy, the24hours urine protein was19.0±1.5mg in nephrectomy group, and6.1±0.6mg in sham-operated group (p<0.01).2)Glomerular sclerosis degree in control group is gentler than that in nephrectomy group.3) RANK and RANKL were expressed in a few tubules, and were absent in glomeruli of sham-operated rat kidneys. However, protein and mRNA of RANK and RANKL were upregulated in glomeruli in nephrectomy group. 1) Confocal microscopy showed the co-localization of RANK and synaptopodin, indicating that podocytes contribute to glomerular RANK expression in podocyte injury diseases. We found low levels of RANK expression in individuals without glomerular disease. By contrast, individuals with biopsy-proven FSGS, IgA nephropathy and MN had a significant increase in RANK expression and synaptopodin.1) Podocytes take on the cobblestone appearance grown under "growth permissive" conditions. And stellate cells with processes emerges grown under "restrictive conditions" for14days. Podocytes at state of proliferation did not express Synaptopodin and Nephrin, but at state of differentiated express Synaptopodin and Nephrin.2) We used cultured differentiated podocytes, which express specific podocyte Synaptopodin proteins, to investigate whether RANK was involved in podocyte injury with PAN. Mouse podocytes were exposed to PAN (25ug/ml), which is known to induce podocyte injury in rodents. Confocal microscopy showed that the bulk of RANK localizes to the cell membrane and cytoplasm. RANK immunofluorescence staining showed membrane localization in SJRH30cells, which were RANK positive controls.3) For determination whether RANK levels increased after PAN (25ug/ml) exposure, RANK mRNA was measured by real-time RT-PCR at24h and48h after PAN exposure. A low abundance of RANK mRNA was detected in control cells that were not exposed to PAN. By contrast, RANK was upregulated upon podocyte injury with PAN, and the level of RANK peaked within24h. Densitometric analyses indicate a1.8-and1.5-fold induction of RANK at24and48h, respectively. 4) Whole-cell lysates were harvested24h or48h after PAN injury, and RANK protein expression was determined by using Western blot analysis with RANK antibody. Densitometric analyses revealed that RANK was upregulated by3.8-and2.4-fold at24h and48h after PAN injury, respectively. RANK immunoblotting revealed that RANK was upregulated in an autocrine manner upon PAN injury in podocytes, reaching a maximum after24h PAN treatment. RANK protein expression was confirmed by using Western blot analysis with RANK antibody in SJRH30cells.5) To better understand RANK function in podocytes, we used an in vitro RANK knockdown system to determine whether RANK was necessary for podocyte survival. We consistently achieved close to71.6%siRNA transfection efficiency in podocytes, as visualized by transfecting a fluorescently tagged Cy3-RANK siRNA. We did not observe any morphologic changes between cells with or without RANK siRNA knockdown during the4d after transfection.6) To test whether RANK was involved in the apoptosis of podocytes in vitro, we studied podocyte apoptosis before and after stable knockdown of RANK with siRNA. The knockdown of RANK alone did not induce podocyte apoptosis, but increased mildly the apoptosis of podocytes exposed to PAN. However, RANKL reduced apoptosis of podocytes transfected RANK siRNA exposed to PAN compared with control siRNA (RANK siRNA16.5±1.5%versus control siRNA24.0±1.8%, p<0.01).7) To investigate whether RANKL was associated with podocyte apoptosis, we pretreated differentiated mouse podocytes with exogenous RANKL (40ng/ml) before PAN (25μg/ml) treatment, and measured apoptosis. The percentage of apoptotic cells was measured by flow cytometry (FAC-Scan) in control podocytes (8.7±0.97%) and podocytes exposed to RANKL (40ng/ml) after48h (5.7±0.81%). PAN (25μg/ml) exposure increased apoptosis of podocytes after48h (26.3±3.6%). Exogenous RANKL protected podocytes from PAN-induced apoptosis (15.5±2.2%).8) To investigate whether the decrease in the [Ca2+]i transient induced by RANKL was through L-type calcium channels, we used the whole cell patch-clamp technique to test the effect of RANKL on the calcium current. In podocytes, incubation with40ng/ml RANKL for5min had no significant effect on the whole cell L-type calcium current. Podocytes were held at+40mV and voltage ramps from-100to100mV were applied for800ms every2s. Ca2+-dependent K+current voltage (I-V) relationships were determined from the voltage ramp commands in the presence of a free calcium concentration of1mM in the patch pipette. RANKL inhibited Ca2+-dependent K+currents from762pA to603pA at+100mV in podocytes. To characterize the mechanisms by which RANKL decreases [Ca2+]i in podocytes, Ca2+-dependent K+currents were performed in the presence of thapsigargin, an inhibitor of the endoplasmic reticulum Ca2+-ATPase. The inhibitory effects of RANKL were not observed in the presence of thapsigargin. RANKL did not affect the resting membrane potential of podocytes.9) To determine further the mechanisms whereby RANKL decrease [Ca2+]i in podocytes, we studied endoplasmic reticulum Ca+-ATPase activity after PAN exposure in the presence or absence of RANKL. Ca2+-ATPase activity was measured by an optical assay in crude endoplasmic reticulum extracted from control and RANKL-treated podocytes. PAN and the specific Ca2+-ATPase inhibitor thapsigargin depressed Ca2+-ATPase activity in a concentration-dependent manner in preparations from the control podocytes. Podocytes were incubated with RANKL (10,40,80or160ng/ml), the activity of the endoplasmic reticulum Ca2+-ATPase increased in a dose-dependent manner.CONCLUSIONS1) Compared with controls, RANK and RANKL were increased in human podocyte diseases and the rat PAN model, and RANK protein expression was mainly attributed to podocytes.2) RANK and RANKL may play a role in progressive renal injure following subtotal nephrectomy.3) RANK was upregulated in mouse podocytes in vitro after injury induced by puromycin aminonucleoside (PAN). Knockdown of RANK expression by small interference RNA (siRNA) exacerbated podocyte apoptosis induced by PAN.4) RANKL inhibited significantly the apoptosis of podocytes induced by PAN.5) These findings suggest the increase in RANK-RANKL expression is a response to podocyte injury, and RANK-RANKL may be a novel receptor-ligand complex for the survival response during podocyte injury.