Regulation of NLRP3 inflammasome activation in glomeruli by ASM during obesity
To test whether ASM in podocytes plays an important role in obesity-induced NLRP3 inflammasome activation in glomeruli of mice, we fed WT/WT, Smpd1−/−, and Smpd1trg/Podocre mice with normal diet (ND) or high fat diet (HFD) for 12 weeks. In Smpd1−/− mice, Smpd1 gene which encodes ASM was knocked out in all cells. In Smpd1trg/Podocre mice, Smpd1 gene was specifically deleted in podocytes [28]. By confocal microscopy, we observed remarkable elevation of NLRP3-ASC colocalization in glomeruli of WT/WT mice on HFD compared with ND-fed WT/WT mice, indicating the abundant formation of NLRP3 inflammasome induced by obesity. Such enhanced NLRP3 inflammasome formation in glomeruli was abolished by Smpd1 gene knockout in Smpd1−/− mice, but significantly amplified by podocyte-specific Smpd1 gene overexpression in Smpd1trg/Podocre mice compared to WT/WT mice (Fig. 1A). As the products of NLRP3 inflammasome, IL-1β and IL-18 in glomeruli were measured as well. HFD treatment obviously increased IL-1β production in glomeruli of WT/WT mice compared to ND-fed WT/WT mice (Fig. 1B). Such elevation of glomerular IL-1β production was prevented by Smpd1 gene deletion in Smpd1−/− mice. On the contrary, podocyte-specific Smpd1 gene overexpression remarkably enhanced glomerular IL-1β production in Smpd1trg/Podocre mice compared with WT/WT mice on both diets. Similar tendencies were demonstrated in glomerular IL-18 production in these mice (Fig. 1C).
Control of obesity-induced inflammatory exosome release from podocytes by ASM
To explore the mechanism by which NLRP3 inflammasome products are secreted out of podocytes to trigger glomerular inflammation during obesity, we tested whether exosomes mediate the release of NLRP3 inflammasome products from podocytes. By nanoparticle tracking analysis (NTA), we found that HFD remarkably increased urinary exosomes of WT/WT mice. Such effect of HFD on urinary exosome excretion was totally blocked by Smpd1 gene deletion in Smpd1−/− mice. In contrast, podocyte-specific Smpd1 gene overexpression significantly elevated urinary exosome excretion in Smpd1trg/Podocre mice on both diets (Fig. 2A and 2B). To confirm the origin of urinary exosomes detected by NTA, we measured CD63 (exosome marker) and podocin (podocyte marker) in urine samples. Mouse CD63 ELISA kit (Assay Genie, Dublin, Ireland) and mouse podocin ELISA kit (Biomatik, Cambridge, Canada) were used to detect the levels of CD63 and podocin in purified urinary exosomes. It was found that HFD-fed WT/WT mice had evidently higher levels of CD63 and podocin in their urine compared to control mice. In Smpd1-/- mice, however, such elevations of CD63 and podocin were significantly attenuated by Smpd1 gene knockout. On the contrary, Smpd1 gene overexpression enhanced obesity-induced elevations of CD63 and podocin in urine of Smpd1trg/Podocre mice compared with WT/WT mice (Fig. 2C and 2D). Furthermore, we confirmed whether NLRP3 inflammasome products were cargos of urinary exosomes of these mice. As shown in Fig. 2E and 2F, HFD markedly elevated the amount of NLRP3 inflammasome products, IL-1β and IL-18, in the urinary exosomes of WT/WT mice. Such obesity-induced elevations of IL-1β and IL-18 in urinary exosomes were blocked by Smpd1 gene deletion in Smpd1−/− mice but exaggerated by podocyte-specific Smpd1 gene overexpression in Smpd1trg/Podocre mice.
Inhibition of immune cell infiltration in glomeruli by Smpd1 gene deletion during obesity
We then examined whether HFD-induced immune cell infiltration in glomeruli was determined by ASM activity in podocytes. Immunofluorescent staining of CD8, a T cell marker, revealed greater infiltration of T cells in the glomeruli of WT/WT mice on HFD compared with control mice. Such enhancement of T cell infiltration in glomeruli was prevented by Smpd1 gene deletion in Smpd1−/− mice. Nevertheless, podocyte-specific Smpd1 gene overexpression significantly amplified T cell infiltration in glomeruli in Smpd1trg/Podocre mice on both diets (Fig. 3A). Immunohistochemical staining of F4/80, a macrophage marker, was performed to detect macrophage infiltration in glomeruli of these mice. As shown in Fig. 3B, greater macrophage infiltration in glomeruli was detected in glomeruli of WT/WT mice on HFD compared to ND-fed WT/WT mice. Podocyte-specific Smpd1 gene overexpression significantly enhanced macrophage infiltration in glomeruli of Smpd1trg/Podocre mice on both diets. On the contrary, Smpd1 gene deletion blocked obesity-induced macrophage infiltration in glomeruli of Smpd1−/− mice.
Contribution of ASM activity to podocyte injury in ORG
Next, we tested whether obesity-induced podocyte injury is affected by ASM. By TEM, we observed the ultrastructure of podocytes in different groups of mice. It was found that HFD treatment induced foot process effacement in podocytes of WT/WT mice compared to control mice. Such pathological change was not observed in Smpd1−/− mice on HFD. Podocyte-specific Smpd1 gene overexpression, however, worsened obesity-induced foot process effacement in podocytes of Smpd1trg/Podocre mice (Fig. 4A). Also, we found that HFD-fed WT/WT mice had much less expression of podocin (a protein component of the filtration slits of podocytes) in glomeruli compared with ND-fed WT/WT mice. On the contrary, glomerular expression of desmin, a marker of podocyte injury, was obviously increased by HFD treatment in WT/WT mice. Such pathological changes in podocin and desmin were totally blocked by Smpd1 gene knockout, but significantly amplified by podocyte-specific Smpd1 gene overexpression (Fig. 4B and 4C).
Glomerular damage and proteinuria prevented by Smpd1 gene deletion
Moreover, morphological and functional changes of glomeruli were assessed in different groups of mice. By PAS staining, morphological examinations showed sclerotic changes in glomeruli of WT/WT mice on HFD. HFD treatment significantly increased the glomerular damage index in WT/WT mice. Smpd1 gene knockout blocked the glomerular damage induced by obesity. Podocyte-specific Smpd1 gene overexpression, however, enhanced glomerular damage in Smpd1trg/Podocre mice on both diets (Fig. 5A). Meanwhile, HFD-fed WT/WT mice on HFD exhibited proteinuria and albuminuria compared to ND-fed WT/WT mice. Proteinuria and albuminuria due to obesity were prevented by Smpd1 gene deletion in Smpd1−/− mice. On the contrary, podocyte-specific Smpd1 gene overexpression significantly aggravated obesity-induced proteinuria and albuminuria in Smpd1trg/Podocre mice.
Visfatin-induced NLRP3 inflammasome activation in podocytes determined by ASM
As a pro-inflammatory adipokine, visfatin has been reported to play an important role in obesity-induced chronic inflammation [51]. Therefore, we tested whether visfatin is involved in NLRP3 inflammasome activation and associated inflammatory exosome release in podocytes. By confocal microscopy, we demonstrated that visfatin induced formation of NLRP3 inflammasomes as indicated by increased colocalization of NLRP3 (green fluorescence) and ASC (red fluorescence) in WT/WT podocytes compared to control cells. In podocytes isolated from Smpd1−/− mice, visfatin-induced NLRP3 inflammasome activation was blocked by Smpd1 gene deletion. In contrast, Smpd1 gene overexpression enhanced visfatin-induced NLRP3 inflammasome activation in podocytes of Smpd1trg/Podocre mice (Fig. 6).
Elevation of inflammatory exosome release from podocytes by visfatin
To test whether visfatin can induce inflammatory exosome release from podocytes, we observed Rab7a, an MVB marker, and IL-1β, a NLRP3 inflammasome product, in podocytes by confocal microscopy. As shown in Fig. 7A and 7B, there was much more colocalization of Rab7a (green fluorescence) and IL-1β (red fluorescence) in WT/WT podocytes treated with visfatin compared to control cells, indicating formation of MVBs containing IL-1β. Such increase in colocalization of Rab7a and IL-1β was prevented by Smpd1 gene knockout in podocytes of Smpd1−/− mice. However, Smpd1 gene overexpression amplified such change in podocytes of Smpd1trg/Podocre mice. By NTA, we also measured exosome release from different podocytes. It was found that visfatin evidently increased exosome secretion from WT/WT podocytes. Such elevation of exosome release was blocked by Smpd1 gene deletion but aggravated by Smpd1 gene overexpression (Fig. 7C and 7D). Moreover, mouse CD63 ELISA kit (Assay Genie, Dublin, Ireland), mouse IL-1β ELISA kit (Assay Genie, Dublin, Ireland), and mouse IL-18 ELISA kit (Assay Genie, Dublin, Ireland) were used to detect the levels of CD63, IL-1β, and IL-18 in purified exosomes. As shown in Fig. 7E-G, CD63, IL-1β, and IL-18 were elevated by visfatin in purified exosomes. Such changes were significantly attenuated by Smpd1 gene deletion in podocytes of Smpd1−/− mice. On the contrary, Smpd1 gene overexpression in podocytes enhanced visfatin-induced elevations of CD63, IL-1β, and IL-18 in purified exosomes.
Inhibition of lysosome-MVB interaction in podocytes by visfatin
To explore the molecular mechanism by which visfatin affects exosome release from podocytes, we observed Rab7a, an MVB marker, and Lamp-1, a lysosome marker, in podocytes by super-resolution microscopy. As shown in Fig. 8, there was considerable amount of colocalization of Rab7a (green fluorescence) and Lamp-1 (red fluorescence) in WT/WT podocytes under control condition, indicating normal lysosome-MVB interaction. Visfatin obviously decreased the colocalization of Rab7a and Lamp-1 in WT/WT podocytes, suggesting reduction of lysosome-MVB interaction. Such change of lysosome-MVB interaction was prevented by Smpd1 gene deletion in podocytes of Smpd1−/− mice. However, Smpd1 gene overexpression amplified visfatin-induced reduction of lysosome-MVB interaction in podocytes of Smpd1trg/Podocre mice.