Triptolide protects podocytes from puromycin aminonucleoside induced injury in vivo and in vitro

Extracts of Tripterygium wilfordii Hook F have been used to treat glomerulonephritis for more than 30 years in China with dramatic antiproteinuric effects. Triptolide, a diterpene triepoxide, is one of the major active components of these extracts. To clarify its antiproteinuric effects we induced podocyte injury by puromycin aminonucleoside. Triptolide effectively reduced the proteinuria induced by puromycin in nephrotic rats without reducing the glomerular filtration rate. The antiproteinuric effect was associated with improvement in the foot process effacement, a decrease in the podocyte injury marker desmin as well as the restoration of nephrin and podocin expression and distribution. In cultured mouse podocytes triptolide pretreatment prevented the puromycin-induced disruption of the actin cytoskeleton and microfilament-associated synaptopodin while protecting nephrin and podocin expression. Triptolide suppressed reactive oxygen species generation and p38 mitogen-activated protein kinase activation while restoring RhoA signaling activity. These results show that triptolide ameliorates puromycin aminonucleoside-mediated podocyte injury in vivo and in vitro. Extracts of Tripterygium wilfordii Hook F have been used to treat glomerulonephritis for more than 30 years in China with dramatic antiproteinuric effects. Triptolide, a diterpene triepoxide, is one of the major active components of these extracts. To clarify its antiproteinuric effects we induced podocyte injury by puromycin aminonucleoside. Triptolide effectively reduced the proteinuria induced by puromycin in nephrotic rats without reducing the glomerular filtration rate. The antiproteinuric effect was associated with improvement in the foot process effacement, a decrease in the podocyte injury marker desmin as well as the restoration of nephrin and podocin expression and distribution. In cultured mouse podocytes triptolide pretreatment prevented the puromycin-induced disruption of the actin cytoskeleton and microfilament-associated synaptopodin while protecting nephrin and podocin expression. Triptolide suppressed reactive oxygen species generation and p38 mitogen-activated protein kinase activation while restoring RhoA signaling activity. These results show that triptolide ameliorates puromycin aminonucleoside-mediated podocyte injury in vivo and in vitro. Proteinuria is the main clinical manifestation of podocyte diseases including minimal change disease, focal segmental glomerular sclerosis, and membranous nephropathy.1.Cattran D.C. Idiopathic membranous glomerulonephritis.Kidney Int. 2001; 9: 1983-1994Abstract Full Text Full Text PDF Scopus (124) Google Scholar, 2.Ly J. Alexander M. Quaggin S.E. A podocentric view of nephrology.Curr Opin Nephrol Hypertens. 2004; 13: 299-305Crossref PubMed Scopus (65) Google Scholar, 3.de Zoysa J.R. Topham P.S. Podocyte biology in human disease.Nephrology. 2005; 10: 362-367Crossref PubMed Scopus (18) Google Scholar It is believed that podocyte injury is a major contributor to severe proteinuria. There are expanding literatures elucidating the molecular events of podocyte injury, although, the treatment for podocyte diseases are far from satisfactory. Therefore, more effective drugs are desirable to improve the treatment for patients with podocyte diseases. Extracts of Tripterygium wilfordii Hook F (TWHF) have been used in the treatment of glomerulonephritis for more than 30 years in China. Tablets made from the extracts of TWHF showed very dramatic effects on decreasing proteinuria in patients with minimal change disease, focal segmental glomerular sclerosis and membranous nephropathy.4.Li L.S. Zhang X. Chen G.Y. Clinical study of Tripterygium wilfordii Hook in the treatment of nephritis.Chin J Intern Med. 1981; 20: 216-220Crossref PubMed Scopus (10) Google Scholar Additionally, extracts of TWHF could alleviate glomerular albumin permeability induced by protamine, tumor necrosis factor (TNF-α) and the serum from patients with focal segmental glomerular sclerosis in vitro.5.Sharma M. Li J.Z. Sharma R. et al.Inhibitory effect of Tripterygium wilfordii multiglycoside on increased glomerular albumin permeability in vitro.Nephrol Dial Transplant. 1997; 12: 2064-2068Crossref PubMed Scopus (18) Google Scholar It is an intriguing possibility that further characterization of the effective component of TWHF on podocyte lesion will provide a new mechanism based medicine for the treatment of podocyte diseases. Triptolide, a diterpene triepoxide, was identified as one of the major active components of TWHF. Recent reports showed that triptolide has strong immunosuppressive and anti-inflammatory activities.6.Yang Y. Liu Z. Tolosa E. et al.Triptolid induces apoptotic death of T lymphocyte.Immunopharmacology. 1998; 40: 139-149Crossref PubMed Scopus (140) Google Scholar, 7.Liu Q. Chen T. Chen G. et al.Immunosuppressant triptolide inhibits dendritic cell-mediated chemoattraction of neutrophils and T cells through inhibiting Stat3 phosphorylation and NF-κB activation.Biochem Biophys Res Commun. 2006; 345: 1122-1130Crossref PubMed Scopus (65) Google Scholar, 8.Sylvester J. Liacini A. Li W.Q. et al.Tripterygium wilfordii Hook F extract suppresses proinflammatory cytokine induced expression of matrix metalloproteinase genes in articular chondrocytes by inhibiting activating protein-1 and nuclear factor kappaB activities.Mol Pharmacol. 2001; 59: 1196-1205PubMed Google Scholar, 9.Kim Y.H. Lee S.H. Lee J.Y. et al.Triptolide inhibits murine inducible nitric oxide synthase expression by down regulating lipopolysaccharide-induced activity of nuclear factor-kappa B and c-Jun NH(2)-terminal kinase.Eur J Pharmacol. 2004; 494: 1-9Crossref PubMed Scopus (97) Google Scholar, 10.Dai Y.Q. Jin D.Z. Zhu X.Z. et al.Triptolide inhibits COX-2 expression via NF-kappa B pathway in astrocytes.Neurosci Res. 2006; 55: 154-160Crossref PubMed Scopus (49) Google Scholar In previous work, we found that triptolide could effectively reduce proteinuria, alleviate glomerular immune injuries, and remarkably improve podocyte lesion in rat model with passive Heymann nephritis.11.Qin W. Liu Z. Zeng C. et al.Therapeutic effect of triptolide on podocyte injury in passive Heymann nephritis.Chin J Nephrol Dial Transplant. 2007; 16: 101-109Google Scholar All these findings implicated that beneficial therapeutic effects of triptolide on proteinuria might be mediated, at least in part, by a protective effect on podocytes. Injection of puromycin aminonucleoside (PAN) to rats produces severe proteinuria and mimics the lesions of minimal change or focal segmental glomerular sclerosis.12.Hagiwara M. Yamagata K. Capaldi R.A. et al.Mitochondrial dysfunction in focal segmental glomerulosclerosis of puromycin aminonucleoside nephrosis.Kidney Int. 2006; 69: 1146-1152Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar It was found that PAN specifically injured podocytes, leading to foot process effacement, actin cytoskeleton disorganization, decreased expression and abnormal distribution of slit diaphragm proteins, including nephrin and podocin,13.Guan N. Ding J. Deng J. et al.Key molecular events in puromycin aminonucleoside nephrosis rats.Pathol Int. 2004; 54: 703-711Crossref PubMed Scopus (42) Google Scholar, 14.Mundel P. Reiser J. Zúñiga Mejía Borja A. et al.Rearrangements of the cytoskeleton and cell contacts induce process formation during differentiation of conditionally immortalized mouse podocyte cell lines.Exp Cell Res. 1997; 236: 248-258Crossref PubMed Scopus (723) Google Scholar, 15.Oh J. Reiser J. Mundel P. Dynamic (re)organization of the podocyte actin cytoskeleton in the nephrotic syndrome.Pediatr Nephrol. 2004; 19: 130-137Crossref PubMed Scopus (67) Google Scholar resulting in a well-described model of podocyte injury. To address the question of whether triptolide has a direct action on podocytes, an animal model of PAN was used to evaluate the effect of triptolide in vivo. In addition, an in vitro study was performed using immortalized podocytes to confirm the direct protective effects of triptolide on podocyte injuries and to explore the underlying mechanism of triptolide action. Proteinuria emerged at 5 days after single PAN injection, reached the peak at 10 days, was persistent at 14 days, and decreased at 21 days. The antiproteinuria effect of triptolide was evaluated by its preventing and treatment effect, respectively. It was found that proteinuria was significantly reduced at 5 days in the triptolide prevention group compared with PAN rats (P<0.01). The reducing effect of triptolide on proteinuria was persistent obviously at 10 days (P<0.01), 14 days (P<0.01), and maintained at 21 days (P<0.01). At 21 days, proteinuria was restored to normal (P>0.05, prevention group versus normal control). The similar effect on proteinuria was observed in the triptolide treatment group. There was no significant difference between the triptolide prevention group and treatment group (Table 1).Table 1The effect of triptolide on proteinuria in PAN nephrosis rats (mg/24 h)5 days10 days14 days21 daysNormal control3.59±1.904.06±1.173.99±1.463.64±2.02PAN model29.22±4.46**P<0.01 versus normal control88.74±14.16**P<0.01 versus normal control40.19±8.18**P<0.01 versus normal control12.71±4.77**P<0.01 versus normal controlPrevention12.03±3.72**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).20.24±4.06**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).12.32±2.73**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).5.35±1.70P<0.01 versus the PAN model (n=5 animals per group).Treatment14.97±4.54**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).22.91±5.90**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).14.38±3.81**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).4.13±2.43P<0.01 versus the PAN model (n=5 animals per group).PAN, puromycin aminonucleoside.P<0.05 versus normal control** P<0.01 versus normal control▴▴ P<0.01 versus the PAN model (n=5 animals per group). Open table in a new tab PAN, puromycin aminonucleoside. P<0.05 versus normal control The restoration of serum albumin level after treatment paralleled to that of decrease of proteinuria. The serum albumin level was markedly elevated at 5 days in triptolide prevention group compared with PAN rats (P<0.05). The significant effect was observed at 10 days (P<0.01). In the triptolide prevention group, serum albumin level was restored to normal at 21 days. Meanwhile, restoration of serum albumin in the triptolide treatment group was similar to that in the prevention group, and there was no significant difference between the two groups (Table 2).Table 2The effect of triptolide on serum albumin level in PAN nephrosis rats (g/l)5 days10 days14 days21 daysNormal control34.32±1.2634.16±1.6233.73±1.9435.46±0.76PAN model24.14±3.93**P<0.01 versus normal control17.3±3.23**P<0.01 versus normal control23.55±4.34**P<0.01 versus normal control30.96±2.17**P<0.01 versus normal controlPrevention28.36±4.18**P<0.01 versus normal control▴P<0.05 versus the PAN model25.23±3.07**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).28.35±1.99**P<0.01 versus normal control▴P<0.05 versus the PAN model35.33±0.88▴P<0.05 versus the PAN modelTreatment27.32±4.30**P<0.01 versus normal control▴P<0.05 versus the PAN model23.06±3.79**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).27.74±3.28**P<0.01 versus normal control▴P<0.05 versus the PAN model34.52±2.10▴P<0.05 versus the PAN modelPAN, puromycin aminonucleoside.P<0.05 versus normal control** P<0.01 versus normal control▴ P<0.05 versus the PAN model▴▴ P<0.01 versus the PAN model (n=5 animals per group). Open table in a new tab PAN, puromycin aminonucleoside. P<0.05 versus normal control In addition, the level of triglyceride in both the triptolide prevention and treatment groups was significantly decreased compared with those in PAN rats at 10 and 14 days (P<0.01), and returned to normal range at 21 days (Table 3). The levels of cholesterol decreased at 14 days and was back to normal at 21 days both in the triptolide prevention and treatment groups (Table 4).Table 3The effect of triptolide on triglyceride level in PAN nephrosis rats (mmol/l)5 days10 days14 days21 daysNormal control1.66±0.221.55±0.151.69±0.191.62±0.23PAN model3.21±1.11**P<0.01 versus normal control7.31±2.96**P<0.01 versus normal control4.81±1.05**P<0.01 versus normal control2.00±0.78Prevention2.14±0.44*P<0.05 versus normal control4.61±1.54**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).1.87±0.36P<0.01 versus the PAN model (n=5 animals per group).1.65±0.14Treatment2.11±0.46*P<0.05 versus normal control4.91±1.63**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).2.01±0.74P<0.01 versus the PAN model (n=5 animals per group).1.64±0.40PAN, puromycin aminonucleoside.* P<0.05 versus normal control** P<0.01 versus normal control▴▴ P<0.01 versus the PAN model (n=5 animals per group). Open table in a new tab Table 4The effect of triptolide on cholesterol level in PAN nephrosis rats (mmol/l)5 days10 days14 days21 daysNormal control2.37±0.342.23±0.432.28±0.292.30±0.29PAN model3.54±1.03*P<0.05 versus normal control4.83±1.33**P<0.01 versus normal control (n=5 animals per group).3.79±1.28*P<0.05 versus normal control2.44±0.44Prevention2.74±0.923.77±1.16*P<0.05 versus normal control3.09±0.992.34±0.35Treatment2.77±0.913.89±1.27*P<0.05 versus normal control3.02±1.092.30±0.42PAN, puromycin aminonucleoside.* P<0.05 versus normal control** P<0.01 versus normal control (n=5 animals per group). Open table in a new tab PAN, puromycin aminonucleoside. PAN, puromycin aminonucleoside. No statistically significant difference in serum creatinine levels was found among the three groups at the time when rats were killed (data not shown). The levels of aspartate aminotransferase and alanine aminotransferase remained normal in triptolide-treated rats (data not shown). To exclude the possibility that triptolide reduced proteinuria by decreasing glomerular filtration rate (GFR), we measured GFR in normal control, PAN model, triptolide prevention, and treatment groups (five rats in each group) at 5, 10, and 14 days, respectively. We found no significant changes of GFR in the PAN model, triptolide prevention, and treatment groups at 5 and 14 days of PAN injection, compared with normal control. And at 10 days, GFR was significantly lower in the PAN model than in normal control (P<0.05); however, no significant differences were found between PAN model and the triptolide prevention and treatment groups (P>0.05). GFR of both triptolide prevention and treatment groups did not differ from that of normal level (P>0.05; Figure 1). Therefore, the results confirmed that triptolide reduced proteinuria by the protective and reversing effect on podocyte injuries in PAN nephrosis rats, rather than by decrease of GFR. Light-microscopy examination showed no histological changes of global or focal segmental glomerulosclerosis, interstitial fibrosis, or tubular atrophy in all the three groups (Figure 2). Foot process effacement could be seen at 5 days after PAN injection in PAN rats (P<0.01 versus normal control). The most extensive foot process effacement that developed at 10 days paralleled with massive proteinuria. The effaced foot processes were just like sheet covering the glomerular basement membrane, and the slit diaphragm gap disappeared. The change persisted at 14 days and got recovery at 21 days (Figure 3a–m). As showed in Table 5, at each time point (5, 10, 14, and 21 days), foot process widths were significantly decreased in the triptolide prevention group compared with that in PAN model rats (P<0.01), but were still wider than normal controls (P<0.01). The foot process size was restored to normal at 21 days in triptolide prevention group. The similar improvement of foot process effacement was observed in triptolide treatment group (Table 5; Figure 3a–m).Table 5The effect of triptolide on foot process width in PAN nephrosis rats (nm)5 days10 days14 days21 daysNormal control271±64268±60267±57277±66PAN model1808±316**P<0.01 versus normal control4929±1002**P<0.01 versus normal control2231±664**P<0.01 versus normal control876±264**P<0.01 versus normal controlPrevention863±209**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).1307±279**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).881±211**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).270±54P<0.01 versus the PAN model (n=5 animals per group).Treatment880±218**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).1444±291**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).906±245**P<0.01 versus normal controlP<0.01 versus the PAN model (n=5 animals per group).274±61P<0.01 versus the PAN model (n=5 animals per group).PAN, puromycin aminonucleoside.** P<0.01 versus normal control▴▴ P<0.01 versus the PAN model (n=5 animals per group). Open table in a new tab PAN, puromycin aminonucleoside. By fluorescence microscopy, the expression intensity and distribution pattern of nephrin and podocin in glomeruli were observed. The staining of nephrin and podocin was revealed as a linear pattern along glomerular capillary wall in the normal control rats. At 5 days, the expression of nephrin and podocin was significantly reduced and the distribution became dot-like. At 10 days, the peak point of proteinuria, the changes of expression intensity and distribution both of nephrin and podocin were distinctly obvious. The linear staining pattern disappeared with remarkably decreased expression intensity of nephrin and podocin. At 14 days, the expression intensity was increased, but distribution still remained dot-like both in nephrin and podocin. At 21 days, the distribution of nephrin and podocin got better. Triptolide could significantly improve the expression of nephrin and podocin, and reverse the redistribution of nephrin and podocin. Compared with PAN model rats, the expressions of nephrin and podocin were significantly increased in both of the triptolide prevention and treatment groups at 10 days, and distribution appeared linear at 14 days. The expression and distribution of nephrin and podocin were significantly recovered at 14 days and completely restored to normal at 21 days (Figure 4a–m, n–z). In normal control rats, trace amount of desmin was found in glomeruli. In PAN model rats, the expression of desmin increased significantly at 5 days, and reached its peak at 10 days. At 14 days, desmin expression began to decrease, and at 21days further decreased but remained still higher than that in normal controls. Compared with PAN rats, expression of desmin was significantly decreased in both the triptolide prevention and treatment groups at 5, 10, 14, and 21 days. At 21 days, expression of desmin returned to normal levels (Figure 5a–m). F-actin filaments in cultured podocytes were distributed as stress fiber-like bundles along the axis or into the process of cells. PAN caused podocytes cytoskeleton reorganization in a dose-dependent manner. Treatment of podocytes with 25 and 50 μg/ml PAN for 24 h only caused F-actin disorderliness and rarefaction of filaments. Treatment with 100 μg/ml PAN resulted in cell retraction and dramatic loss of actin stress-fiber organization (Figure 6a–d), and hence, cytoskeletal changes were accompanied by loss of synaptopodin staining (Figure 6q–s). However, when cells were preincubated with triptolide for 30 min before exposure to PAN, podocytes avoided the above changes. The protective effect of triptolide on cytoskeleton was also dose dependent (Figure 6i–l). At the dose of 3 ng/ml, triptolide almost completely restored normal cytoskeleton in podocytes without affecting cell survival (Figure 6m–p). To ensure that cytoskeleton disruption effect of PAN were not due to the apoptosis of podocytes, we performed Hoechst staining. Our results showed no significant difference in apoptosis between PAN-treated (<0.2% of total cells), triptolide pretreatment (<0.3% of total cells), and untreated (0%) podocytes (n>300 cells for each treatment) (Figure 6e–h and m–p). For subsequent experiments, triptolide was used at concentration of 3 ng/ml and PAN 100 μg/ml. To assess whether triptolide could recover podocytes from prior injury, podocytes were pretreated with 100 μg/ml PAN for 24 h. Then cells were transferred to triptolide medium (3 ng/ml) for an additional 24 or 72 h. Podocytes incubated with triptolide for only 24 h following PAN treatment had sparse stress fibers, and staining of F-actin in them was stronger than in those without triptolide incubation. After 72 h of incubation with triptolide medium, podocytes reformed robust network of actin stress fibers (Figure 7). Synchronously, disrupted distribution of synaptopodin by PAN was restored after treatment with triptolide (Figure 8).Figure 8Triptolide reversed podocyte synaptopodin distribution in PAN-induced injury. Untreated podocytes (a). Podocytes were treated with PAN (100 μg/ml, 24 h) followed by culture medium for further 72 h (b). Podocytes were treated with PAN (100 μg/ml, 24 h) followed by triptolide medium for further 72 h (c). Green fluorochrome corresponds to synaptopodin and red to the nucleus. Original magnification × 400.View Large Image Figure ViewerDownload (PPT) The effects of triptolide on the expression and distribution of nephrin and podocin were further confirmed in cultured podocytes. PAN treatment (100 μg/ml, 24 h) decreased the expression of nephrin and podocin in cultured podocytes. However, triptolide pretreatment could protect podocytes against PAN-induced injuries (Figure 9a–c, d–f). These observations were consistent with the results of nephrin and podocin expression analyzed by flow cytometry (Figure 10a and b).Figure 10Effects of triptolide pretreatment on pococytes nephrin and podocin expression analyzed by flow cytometry. MFI of nephrin (a). MFI of podocin (b). Panels on the left represent untreated podocytes. Middle panels represent PAN (100 μg/ml) treated podocytes. Panels on the right represent podocytes preincubated for 30 min with triptolide before PAN exposure.View Large Image Figure ViewerDownload (PPT) Additionally, triptolide could recover PAN-induced injuries on nephrin and podocin expression in podocytes (Figure 11a–f). Podocytes were treated with PAN (100 μg/ml) for 24 h followed by triptolide medium (3 ng/ml) for further 72 h; the expressions of nephrin and podocin were significantly improved after triptolide treatment. This result was also confirmed by flow cytometry (Figure 12a and b).Figure 12Triptolide restored podocytes nephrin and podocin expression analyzed by flow cytometry. MFI of nephrin (a). MFI of podocin (b). Panels on the left represent untreated podocytes. Middle panels represent podocytes were treated with PAN (100 μg/ml, 24 h) followed by culture medium without triptolide for further 72 h. Panels on the right represent podocytes were treated with PAN (100 μg/ml, 24 h) followed by culture medium with triptolide for further 72 h.View Large Image Figure ViewerDownload (PPT) To unravel the mechanisms underlying the protective effect of triptolide on podocytes, PAN-induced intracellular production of reactive oxygen species (ROS) was examined. PAN (100 μg/ml) significantly increased ROS generation in podocytes. This effect was observed at 30 min and maintained for 6 h (Figure 13a). ROS level at 30 min was twofold higher than that of basal level after PAN treatment. Pretreatment of podocytes with triptolide (3 ng/ml) or antioxidant N-acetylcysteine (NAC, 10 mmol/l) before PAN exposure led to a significant reduction in the cellular ROS level (Figure 13b). To characterize the intracellular signaling pathway associated with the protective effects of triptolide in podocytes, phosphorylation of p38 mitogen-activated protein kinase (MAPK) was analyzed. Treatment of podocytes with PAN induced a strong increase in phosphorylation of p38 MAPK. It increased at 3 h and was maintained for 12 h after PAN treatment (Figure 13c). Pretreatment of triptolide (3 ng/ml) effectively suppressed PAN-induced phosphorylation of p38 MAPK (Figure 13d). Triptolide did not affect basal phosphorylation of p38 MAPK. We next examined the role of ROS and p38 MAPK phosphorylation in modulating PAN-induced actin reorganization described in Figure 6. Western blots showed that NAC (10 mmol/l) significantly suppressed PAN-induced phosphorylation of p38 MAPK. Both NAC (10 mmol/l) and p38 inhibitor SB-203580 (25 μmol/l) effectively inhibited PAN-induced cytoskeleton disarrangement (Figure 13e–h). RhoA has been suggested to play an important role in cytoskeleton reorganization. To investigate whether RhoA-signaling pathways are involved in the effect of triptolide on podocytes, we performed RhoA activation assay. As shown in Figure 14a, PAN treatment induced strong decline in RhoA activity at 30 min and the low level of activity was maintained till 12 h. When cells were pretreated with triptolide prior to PAN exposure, RhoA activity was not decreased and normal level was maintained. Total protein level of RhoA was not affected by PAN and triptolide treatment during the time of the test. However, the increase of RhoA activation was markedly inhibited by the specific RhoA inhibitor, C3 exoenzyme (1 μg/ml). Immunofluorescence staining was consistent with the result from western blotting, showing that inhibition of RhoA activity by C3 exoenzyme abolished the protective effect of triptolide on PAN-induced F-actin dissociation (Figure 14b–c). These results strongly suggested that restoration of RhoA activity mediated the protective effect of triptolide. We further examined whether p38 MAPK pathway cross-talked with RhoA-signaling pathways in regulating PAN-induced cytoskeleton disruption. Western blot results showed that C3 exoenzyme (1 μg/ml) did not alter p38 MAPK phosphorylation in response to PAN, and neither NAC (10 mmol/l) nor SB-203580(25 μmol/l) altered RhoA activity (Figure 15). Our data suggested that p38 MAPK and RhoA are two independent signaling pathways involved in regulation of PAN-induced cytoskeleton disorganization. The concept that podocyte has the major role in the development of proteinuria and progression of glomerulosclerosis, leads us to find new approaches to target podocyte lesions. These efforts have generated numerous findings showing that certain reagents, such as retinoids, fluvastatin and darbepoetin, decreased proteinuria by ameliorating podocyte injury.16.Moreno-Manzano V. Mampaso F. Sepúlveda-Muñoz J.C. et al.Retinoids as a potential treatment for experimental puromycin-induced nephrosis.Br J Pharmacol. 2003; 139: 823-831Crossref PubMed Scopus (55) Google Scholar, 17.Shigeru S. Miki N. Toshiro F. Fluvastatin ameliorates podocyte injury in proteinuric rats via modulation of excessive Rho signaling.J Am Soc Nephrol. 2006; 17: 754-764Crossref PubMed Scopus (103) Google Scholar, 18.Eto N. Wada T. Inagi R. et al.Podocyte protection by darbepoetin: preservation of the cytoskeleton and nephrin expression.Kidney Int. 2007; 72: 455-463Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar Here, we reported another reagent, triptolide, showing potent efficacy in reducing proteinuria and restoring podocyte injury in PAN-induced nephrosis. Results of this study demonstrated that triptolide effectively reduced proteinuria and ameliorated podocyte foot process effacement in PAN nephrosis rats. Recovery from podocyte injury was further proved by reduction of desmin expression in PAN rats after treatment with triptolide. In addition, the alterations of slit diaphragm proteins, nephrin and podocin, induced by PAN were remarkably restored by triptolide. In vitro studies further confirmed that triptolide protected and reversed PAN-induced cytoskeleton disruption, as well as distribution of nephrin and podocin expression, in podocytes. It was found that the above action of triptolide may be mediated through ROS–p38 MAPK and RhoA pathway. PAN-induced nephrosis is characterized by heavy proteinuria and podocyte foot process effacement. Studies have correlated podocyte slit diaphragm proteins and cytoskeletal abnormalities with the onset of proteinuria in PAN-induced nephrosis.13.Guan N. Ding J. Deng J. et al.Key molecular events in puromycin aminonucleoside nephrosis rats.Pathol Int. 2004; 54: 703-711Crossref PubMed Scopus (42) Google Scholar,19.Saleem M.A. Ni L. Witherden I. et al.Colocalization of nephrin, podocin, and the actin cytoskeleton: evidence for a role in podocyte foot process formation.Am J Pathol. 2002; 161: 1459-1466Abstract Full Text Full Text PDF PubMed Scopus (140) Google Scholar Recently, it has been revealed that the slit diaphragm proteins complex (nephrin–podocin–CD2AP) is an important component in maintaining glomerular filtration barrier. In PAN-induced nephrosis models, the expression and distribution of nephrin and podocin were found abnormal before the onset of proteinuria.20.Nakatsue T. Koike H. Han G.D. et al.Nephrin and podocin dissociate at the ons