| Abstract |
Renal epithelial cells (podocytes) and their slit diaphragms (SD) that cover the slit pores among adjacent foot processes (FP) of the podocytes are crucial constituent of the primary barrier for ultrafiltration of plasma in renal glomeruli. They ensure the integrity of the glomerular basement membrane (GBM) and prevent the urinary loss of proteins. A putative distraction of the SDs, of its main components, nephrin and podocin is involved in the immune mediated glomerulonephritis.
We have previously reported that decreases of the podocyte SD proteins nephrin and podocin represent early events in the podocytopathy of lupus nephritis (LN). The latter is a prototype of autoimmune disease with a various histological and clinical picture with proteinuria been a major clinical manifestation. We asked whether immunosuppressive agents such as glucocorticoids (GC) and cyclophosphamide (CY), may have direct effects on SD and subsequently on podocytes.
We used NZB/W F1 female mice in our experiments. NZB/W are thought to be a representative model of spontaneous development of LN. Animals were divided in groups of 6 and 9 months old, as well as in groups of: a) untreated (control) mice (n = 24), b) GC–treated mice, which received dexamethasone p.o. for 3 (n = 4) or 6 (n = 5) consecutive months, and c) CY–treated mice, which received CY i.p. for 3 (n = 4) or 6 (n = 4) consecutive months. C57Bl/6 mice (n = 20) were used as normal controls. Comparisons were made with the parental strains of NZB (n = 5) and NZW (n = 4) as well.
Assessment of 24h proteinuria was made by Bradford assay. After mice were euthanized tissue from the left kidney was used for histologic studies: light microscopy, IgG immunofluorescence and electron microscopy. Right kidney was used for biochemical studies: glomerular expression of nephrin and podocin in the protein level was studied with western blot. Nephrin expression in kidney tissue was also examined with immunofuorescence. To determine whether changes in nephrin and podocin expression in NZB/W treated mice correlate with altered transcription in corresponding genes, we performed quantitative real-time PCR. Finally, anti-dsDNA antibodies in mice serum was examined with ELISA.
Data are expressed as the mean ± standard error of the mean (SEM) values. Comparisons were performed with the Mann-Whitney U test for numerical data. The Spearman’s rho (ρ) test was used for correlation analysis.
Treatment of NZB/W mice with GC or CY prevented the development of proteinuria, whereas untreated littermates developed overt proteinuria by the age of 6 months.
Three-month-old NZB/W mice demonstrated normal histology or findings of mild mesangial LN (MMLN) in light microscopy. In contrast, all untreated 6- and 9-month-old mice developed focal or diffuse proliferative glomerulonephritis (FPLN and DPLN, respectively) associated with numerous IgG deposits. In GC– and CY–treated mice, glomerulonephritis was milder with markedly reduced interstitial inflammation and mesangial hyperplasia; seven out of eight
7
(7/8) NZB/W mice treated for 6 months had normal LM histology. Accordingly, IgG kidney deposition was significantly reduced in treated mice being undetectable in mice treated with CY for 6 months.
Electron microscopy examination of kidney sections from untreated 6- and 9-month-old mice with FPLN or DPLN revealed extensive mesangial, subendothelial, and subepithelial EDD fusion of podocytes and effacement of FPs, and destruction of slit diaphragms. Conversely, in mice treated with GC or CY, there were only few sub-endothelial EDD. Both SD and podocytes FPs and SD were well preserved.
Next the glomerular expression of the main SD proteins was examined with western blot. Nephrin was increased in treated mice compared to untreated littermates both in GC and CY groups (p<0,05). Glomerular nephrin expression showed negative association with the histological class of nephritis (ρ=-0.69, p<0.001), and positive association with podocin protein levels (ρ=0.53, p=0.007). Treatment of NZB/W mice for 3–6 months resulted also in higher glomerular podocin protein levels compared to untreated aged-matched mice with FPLN or DPLN. Similar to nephrin, podocin levels correlated inversely with the histological class of nephritis (ρ=-0.62, p=0.001). Nephrin expression and localization was further assayed by IF in kidney sections from NZB/W mice. In accordance with the WB results, nephrin IF staining was stronger in treated mice which was reduced in older untreated mice with FPLN or DPLN, and became diminished in 9-month-old diseased mice. Nephrin IF score correlated with the nephritis histological class (ρ=0.55, p=0.008) and nephrin expression assayed by WB (ρ=0.71, p=0.004).
To determine whether changes in nephrin and podocin expression in NZB/W treated mice correlate with altered transcription in corresponding genes, we performed quantitative real-time PCR in total RNA. Nephrin mRNA was significantly increased after 3–6 of GC and CY therapy compared to age-matched untreated littermates (p<0,05 in the GC group and p<0,01 in the CY group). Podocin mRNA levels was increased in treated mice but differences were statistically significant only after 6 months of treatment. We observed a significant correlation between nephrin and podocin mRNA both in untreated (ρ=0.67, p=0.005, n = 16) and in GC / CY-treated (ρ=0.72, p=0.002, n = 16) mice.
Next, we addressed whether electron microscopy findings in kidney biopsies correlated with differential expression of SD proteins. Indeed, electron dense deposits showed significant inverse correlation with glomerular nephrin (ρ correlation coefficient ranging from -0.75 to -0.85) and podocin (ρ ranging from -0.88 to -0.93) protein expression. Similarly, podocyte FP effacement showed negative association with both nephrin (ρ = -0.77, p = 0.009) and podocin (ρ= -0.93, p < 0.001). With respect to mRNA levels, deposits of any type and podocyte FP effacement showed also inverse association with nephrin – but not podocin – expression.
We also examined the SD protein expression and their mRNA levels in treated mice compared to normal controls and their parental strains.
Our results present the effect of early treatment of LN on podocytes SD proteins. Treatment with GC or CY prevented the development of proteinuria, and halted the histologic alterations
8
in light and electron microscopy, where deposits of any type were diminished and podocytes structure was preserved. Glomerular nephrin and podocin expression significantly increased in both treated groups up to normal levels. Nephrin immunofluorescence confirmed these results. The findings above and the increased mRNA levels may imply that the SD alterations at the protein level precede the alterations in the mRNA level.
The previous results are confirmed by the correlation of SD proteins with the histologic findings.
Podocytopathy may be due to immunological mechanisms that involve elements of both the innate and adaptive immunity (CD8+ T cells, anti-dsDNA antibodies, co-stimulatory molecules such as B7-1 and Toll-like receptors).
Immunosuppressive agents, in addition to the effect on the immune system, may directly influence the unique structure and function of podocytes. Our findings further emphasize the role of podocytes and the slit diaphragm in the pathogenesis of immune-mediated glomerulonephritis, and point toward the potential benefits of development of selective, antiproteinuric, and podocyte-protective drugs.
|
Treatment effect on the expression of podocyte protein in systemic lupus erythematosus nephritis
