The inflammasome is a multiprotein complex responsible for the activation of caspases 1 or 5, which has been found to be switched on by some endogenous danger signals. The present study was designed to test whether activation of this inflammasome contributes to homocysteine (Hcys)‐induced inflammatory injury in podocytes. By RT‐PCR and immunocytochemistry, we detected the expression of 3 major inflammasome components in mouse podocytes including NALP3, apoptosis‐associated speck‐like protein (ASC), and caspase‐1. All these 3 molecules were abundant in cytoplasma, and ASC was located around nuclei. Under fluorescent confocal microscope, no colocalization of three proteins could be observed under control condition. However, incubation of podocytes with L‐Hcys (80 μM) for 12 hours significantly increased colocalization of NALP3 with ASC or caspase‐1, which was associated with ASC trafficking towards the cell periphery, indicating the formation of a NALP3 inflammasome. To determine the activity of this inflammasome, colorimetric measurement of caspase‐1 activity was conducted. It was found that L‐Hcys increased caspase‐1 activity by 6 folds. These results indicate that activation of inflammasomes to produce pro‐inflammatory factors may turn on the inflammatory response to Hcys in podocytes, producing podocytes dysfunction and glomerular injury (supported by NIH grant HL057244 and DK54927).

Inflammasomes have been reported to serve as intracellular machinery initiating and promoting the innate immune reaction and inflammatory response to various danger signals. The present study hypothesized that activation of inflammasomes is one of the important mechanisms mediating endothelial inflammatory response to adipokines in obesity. To test this hypothesis, we first characterized the expression of 3 major inflammasome components in mouse endothelial cells (ECs) including cryopyrin (also called NALP3), apoptosis‐associated speck‐like protein (ASC), and caspase‐1. By RT‐PCR and immunocytochemistry, we found that these 3 molecules were abundantly expressed in mouse ECs. Fluorescent confocal microscopy showed that these molecules could be detected in the cytoplasma of mouse ECs, but they were not colocalized under control condition. However, when these cells were incubated with adipokine, visfatin even for two hours, these molecules were aggregated to form a NALP3 inflammasome, indicating the assembling and activation of this inflammatory machinery. Colorimetric measurement of caspase‐1 activity demonstrated that visfatin significantly increased caspase‐1 activity by about 4 folds. These results suggest that activation of inflammasomes may be one of initiating mechanisms responsible for the action of adipokine‐visfatin to induce endothelial dysfunction and inflammatory injury (supported by NIH grant HL057244, HL091464 and HL075316)

Our recent studies have indicated that hyperhomocysteinemia (hHcys) may induce podocyte damage, resulting in glomerulosclerosis. However, the molecular mechanisms mediating hHcys-induced podocyte injury are still poorly understood. In the present study, we first demonstrated that an intact NADPH oxidase system is present in podocytes as shown by detection of its membrane subunit (gp91phox) and cytosolic subunit (p47phox). Then, confocal microscopy showed that gp91phox and p47phox could be aggregated in lipid raft (LR) clusters in podocytes treated with homocysteine (Hcys), which were illustrated by their colocalization with cholera toxin B, a common LR marker. Different mechanistic LR disruptors, either methyl-β-cyclodextrin (MCD) or filipin abolished such Hcys-induced formation of LR-gp91phox or LR-p47phox transmembrane signaling complexes. By flotation of detergent-resistant membrane fractions we found that gp91phox and p47phox were enriched in LR fractions upon Hcys stimulation, and such enrichment of NADPH oxidase subunits and increase in its enzyme activity were blocked by MCD or filipin. Functionally, disruption of LR clustering significantly attenuated Hcys-induced podocyte injury, as shown by their inhibitory effects on Hcys-decreased expression of slit diaphragm molecules such as nephrin and podocin. Similarly, Hcys-increased expression of desmin was also reduced by disruption of LR clustering. In addition, inhibition of such LR-associated redox signaling prevented cytoskeleton disarrangement and apoptosis induced by Hcys. It is concluded that NADPH oxidase subunits aggregation and consequent activation of this enzyme through LR clustering is an important molecular mechanism triggering oxidative injury of podocytes induced by Hcys.

We have reported that hyperhomocyteinemia (hHcys) serves as a pathogenic factor to induce glomerulosclerosis. However, it remains unknown whether homocysteine (Hcys) exerts its actions through specific receptors. The present study tested the hypothesis that NMDA receptor (NMDAR) activation is involved in the development of hHcys‐induced glomerulosclerosis. Uninephrectomized Sprague‐Dawley rats were fed a folate‐free diet for 8 weeks to produce hHcys and an antagonist for NMDAR, MK‐801, was administrated. Rats treated with a folate‐free diet exhibited significantly increased plasma Hcys level (from 5.041±0.478 to 16.187±3.172), upregulation of glomerular NMDAR‐1 and NMDAR‐2A, enhanced NADPH oxidase (Nox)‐dependent O2.‐ production (2.8 folds vs. control), and remarkable glomerulosclerosis. Administration of MK‐801 significantly reduced these renal functional and pathological changes induced by hHcys. Correspondingly, extracellular matrix (ECM) deposition in the glomeruli of hHcys rats was ameliorated by the treatment with MK‐801, which was accompanied by changes in the activity of ECM proteolytic enzymes such as tissue inhibitor of metalloproteinase‐1 and matrix metalloproteinase‐1 and −9, towards a normal level. These results suggest that NMDAR is essential in Hcys‐induced Nox activation and consequent development of glomerulosclerosis (supported by NIH grants DK54927 and HL075316).

Our previous studies have demonstrated that hyperhomocyteinemia (hHcys) induced glomerulosclerosis through the activation of NADPH oxidase (Nox). However, it remains unknown how homocysteine (Hcys) produces detrimental effects on podocytes. The present study tested the hypothesis that Hcys may induce epithelial‐to‐mesenchymal transition (EMT) in podocytes through the activation of Nox. Mouse podocytes were incubated with L‐Hcys, a pathogenic isoform of Hcys. It was found that L‐Hcys (80 μM) induced a dramatic increase in the expression of Nox subunits and subsequent superoxide (O2.‐) production (2.9 folds vs. control), which was accompanied by significantly decreased expression of slit diaphragm‐associated protein, P‐cadherin and zonula occludens‐1, a change consistent with loss of the epithelial feature. Meanwhile, L‐Hcys induced the expression of mesenchymal markers, fibroblast‐specific protein‐1 and α‐smooth muscle actin. Nox inhibitors significantly reversed these Hcys‐induced changes in podocytes. In functional studies, L‐Hcys was found to increase dextran permeability across podocyte monolayers by 1.9 folds, which was inhibited by Nox inhibitors. These results suggest that an EMT occurs in podocytes due to cellular oxidative stress during hHcys and that enhanced EMT may result in podocyte dysfunction leading to glomerulosclerosis (supported by NIH grants DK54927 and HL075316).

In this study, mice lacking the gp91(phox) gene were used to address the role of NADPH oxidase in hyperhomocysteinemia-induced podocyte injury. It was found that a folate-free diet increased plasma homocysteine levels, but failed to increase O(2)(-) production in the glomeruli from gp91(phox) gene knockout (gp91(-/-)) mice, compared with wild-type (gp91(+/+)) mice. Proteinuria and glomerular damage index (GDI) were significantly lower, whereas the glomerular filtration rate (GFR) was higher in gp91(-/-) than in gp91(+/+) mice when they were on the folate-free diet (urine albumin excretion, 21.23+/-1.88 vs 32.86+/-4.03 microg/24 h; GDI, 1.17+/-0.18 vs 2.59+/-0.49; and GFR, 53.01+/-4.69 vs 40.98+/-1.44 microl/min). Hyperhomocysteinemia-induced decrease in nephrin expression and increase in desmin expression in gp91(+/+) mice were not observed in gp91(-/-) mice. Morphologically, foot process effacement and podocyte loss due to hyperhomocysteinemia were significantly attenuated in gp91(-/-) mice. In in vitro studies of podocytes, homocysteine was found to increase gp91(phox) expression and O2(*)(-) generation, which was substantially inhibited by gp91(phox) siRNA. Functionally, homocysteine-induced decrease in vascular endothelial growth factor-A production was abolished by gp91(phox) siRNA or diphenyleneiodonium, a NADPH oxidase inhibitor. These results suggest that the functional integrity of NADPH oxidase is essential for hyperhomocysteinemia-induced podocyte injury and glomerulosclerosis.

Recent studies have reported that lipid raft (LR) redox signaling platforms associated with NADPH oxidase are involved in coronary endothelial dysfunction upon different injury factors. The present study was designed to test whether statins could interfere with the formation of LR redox signaling platforms to protect the coronary arterial endothelium from injury. By confocal microscopy, we first detected the formation of LRs clustering in human coronary arterial endothelial cells (HCAECs) exposed to FasL, a death receptor ligand. In these LR clusters, NADPH oxidase subunits, gp91phox and p47phox were aggregated with LR, which was almost completely blocked by statins (lovastatin, provastatin and simvastatin). To further explore the functional relevance of this action, we performed detergent resistant membrane floatation to isolate LRs. Electron spin resonance spectrometry showed that superoxide (O2.−) production was 5‐fold higher in the LR fractions from FasL‐treated HCAECs than that from untreated cells. This FasL‐induced enhancement of O2.− production in LR fractions was substantially blocked by pretreatment with statins. Our results indicate that blockade of LR redox signaling platform formation in endothelial cell membrane may be another important therapeutic mechanism of statins in preventing endothelial injury and atherosclerosis.(Supported by NIH grants HL057244, HL075316, and DK054927)

Lipid raft (LR) redox signaling platform associated with NADPH oxidase (NOX) was reported to mediate the actions of death receptor activation in different cells. It is interesting to know whether this LR redox signaling platform is also involved in podocytes injury during hyperhomocysteinemia. The present study first characterized the presence of the NOX membrane subunit‐gp91phox and it LR clusters in podocytes, and then tested whether this redox signaling platform contributes to homocysteine (Hcys)‐induced podocytes apoptosis. It was found that Hcys markedly increased the expression of gp91phox and stimulated NOX‐dependent superoxide (O2.−) production in a concentration‐dependent manner, as measured by electronic spin resonance (ESR). Using confocal microscopy, gp91phox was found to aggregate in LR clusters upon Hcys stimulation, which was inhibited by lipid‐raft disruptors, methyl‐β ‐cyclodextrin (MCD) and filipin. Functionally, increased O2.− production associated with these LR‐gp91phox platform was also blocked by MCD or filipin. Flow cytometry showed that Hcys induced podocytes apoptosis, which could be attenuated by gp91phox siRNA or LR disruptors. Our results indicate that the formation of gp91phox‐associated LR redox signaling platform importantly contributes to podocytes injury during exposure to high Hcys levels (supported by NIH grants DK54927, HL075316, and HL57244).

NADPH oxidase (NOX) has been reported to contribute to glomerulosclerosis induced by hyperhomocysteinemia (hHcys). In the present study, we used a mouse model lacking gp91phox (gp91−/−), a membrane subunit of NOX, to test whether the pathogenic role of NOX is associated with podocytes injury induced by hHcys. gp91−/− and wild‐type mice (gp91+/+) were fed on folate‐free (FF) or normal chow for 6 weeks. HPLC analysis showed that both strains of mice had higher plasma homocysteine levels on FF diet compared to the mice on normal chow (9.86 ± 1.06 vs. 4.98 ± 0.52 μM in gp91−/− mice and 10.53± 1.27 vs. 5.05 ± 0.68 µM in gp91+/+ mice). ESR analysis showed that hHcys failed to markedly increase NOX‐dependent superoxide production in kidneys from gp91−/− mice compared with gp91+/+ mice. Glomerular sclerosis index was markedly lower in gp91−/− than in gp91+/+ mice on FF diet. RT‐PCR showed that hHcys significantly decreased nephrin mRNA expression in gp91+/+ mice but not in gp91−/− mice. Correspondingly, desmin mRNA expression was less elevated in gp91−/− mice on FF diet. Immunohistochemistry demonstrated decreased nephrin and increased desmin staining in gp91+/+ mice on FF diet, rather than in gp91−/− mice. Our results suggest that knock‐out of gp91phox leads to a substantial reduction of local oxidative stress in glomeruli, which protects podocytes from hHcys‐induced injury (supported by NIH grants DK54927 and HL075316).