UC-MSC administration ameliorates renal injury in DN rats
To explore the therapeutic effect of UC-MSCs, we established a STZ-induced DN model of rats. The experimental design was shown in Fig. 1a, and the difference between the two treatment groups was that UC-MSCs were transplanted at different stages of DN. We analyzed renal function and pathological changes among control rats, DN rats, and UC-MSC-treated DN rats. Compared with DN rats, UC-MSC-treated DN rats showed a significant decrease in serum creatinine (CRE), serum urea nitrogen (BUN) and 24h total urinary protein (Fig. 1b). Moreover, UC-MSC administration attenuated tubular dilatation, massive accumulation of inflammatory cells in the interstitial area, renal glomerular hypertrophy and sclerosis, and renal interstitial fibrosis in the DN kidney (Fig. 1c-1e). These results indicate that UC-MSCs ameliorate renal injury in DN rats, and transplantation of UC-MSCs at different stages of DN has a consistent therapeutic effect.
UC-MSC administration inhibits renal inflammation in DN rats
To investigate whether UC-MSCs modulate the inflammatory response in the DN rats, we determined the levels of pro- and anti-inflammatory cytokines and quantified inflammatory cell infiltration in the DN rats with or without UC-MSC treatment. Unlike in DN group which showed elevated mRNA levels of pro-inflammatory IL-1β, IL-6 and TNF-α, UC-MSC treatment in rats significantly reduced the expression of pro-inflammatory cytokines but augmented anti-inflammatory IL-10 expression (Fig. 2a). Consistent with this data, decreased IL-1β production and increased IL-10 production (Fig. 2b and 2c) were observed in kidneys and serum from UC-MSC-treated DN rats, as compared with untreated controls. Moreover, UC-MSC administration resulted in a significant decreased accumulation of Ly6G+ neutrophils (Fig. 2d) and F4/80+ macrophages (Fig. 2e), as determined by immunohistochemistry and immunofluorescence staining. These results indicate that UC-MSC administration reduces systemic and local renal inflammation in DN rats.
UC-MSCs shift macrophage polarization toward an M2 phenotype in vivo and in vitro
Since macrophage M1 activation is implicated in the pathogenesis of chronic inflammatory diseases including DN, we next determined whether UC-MSCs may regulate macrophage polarization in DN rats. The results showed that administration of UC-MSCs increased arginase1 (ARG1) expression, an M2 macrophage marker, but decreased inducible nitric oxide synthase (iNOS) expression, an M1 macrophage marker, at the mRNA and protein levels (Fig. 3a, 3b). Moreover, using double immunofluorescence staining, we found that UC-MSC administration enhanced M2 macrophage ARG1 expression (Fig. 3c).
Next, we used UC-MSC-conditioned medium (UC-MSC-CM) to treat the macrophage cell line RAW264.7 in response to stimulation with LPS (100 ng/ml) and IFN-γ (30 ng/ml). Clearly, UC-MSC-CM treatment resulted in significant decrease in the expressions of iNOS, IL-1β, TNF-α, and TRAF6, but increase in ARG1 and IL-10 expression in LPS/IFN-γ-stimulated RAW264.7, as compared with untreated controls (Fig. 3d, 3e). To further demonstrate the effect of UC-MSCs on macrophage polarization, we used human macrophage cell line THP1. Similarly, UC-MSC-CM treatment markedly reduced iNOS, IL-1β, TNF-α, TRAF6 and p-STAT1 expression, but increased ARG1 and IL-10 expression in LPS/IFN-γ-stimulated THP1 (Suppl Fig. 1). Taken together, these results suggest UC-MSCs regulate macrophage polarization and function during inflammatory response via a paracrine mechanism.
High-throughput sequencing analysis of miRNA profile in the DN kidney and UC-MSCs
miRNAs have been implicated as the key regulators of different molecules involved in the DN network . To explore the communication between miRNAs and macrophage polarization in DN, we examined the miRNAs in kidney tissues of normal and DN rats using high-throughput sequencing. Moreover, we measured the expression levels of various miRNAs between exosomes derived from HEK293T and UC-MSCs. Analysis of differentially expressed miRNAs revealed that miR-146a-5p was significantly down-regulated in the DN kidney and abundant in UC-MSC-derived exosomes (Fig. 4a, 4b). Further validation of miR-146a-5p expression in the kidney by qRT-PCR confirmed the expression of miR-146a-5p was down-regulated in the DN rats and restored in the UC-MSC-treated DN rats (Fig. 4c). We also found that miR-146a-5p expression was significantly increased in LPS/IFN-γ-stimulated UC-MSCs, as compared with unstimulated controls. (Fig. 4d). More importantly, the expression of miR-146a-5p in the kidney was negatively correlated with the concentration of Cre and BUN in serum (Fig. 4e, 4f). These results reveal the potential function and candidate biomarker attributes of miR-146a-5p in DN, and suggest that UC-MSC-derived miR-146a-5p may play an important role in protection against DN.
UC-MSCs-derived miR-146a-5p targets TRAF6 and facilitates M2 macrophage polarization
To identify the mechanism how miR-146a-5p regulates macrophage polarization during DN, we performed bioinformatics analyses according to TargetScan and microRNA.org. The web-based prediction software for targets of miRNAs, TRAF6 3’-UTR region contains the putative binding sequence of miR-146a-5p, could be a potential target gene of miR-146a-5p (Fig. 5a). Using the dual-luciferase assay, we observed that miR-146a-5p mimics significantly suppressed luciferase activity when wide type (WT) 3’UTR of TRAF6 was inserted downstream of the luciferase report plasmid, as compared with mutant 3’UTR (Fig. 5b).
Moreover, we transfected miR-146a-5p mimic, miR-146a-5p inhibitor, and their negative control (NC) into UC-MSCs, and collected UC-MSC-CM to treat LPS/IFN-γ-stimulated RAW264.7. As expected, transfection with miR-146a-5p mimic obviously increased miR-146a-5p expression in UC-MSCs, as compared with control UC-MSCs (Fig. 5c). We also observed that CM derived from miR-146a-5p mimic-transfected UC-MSCs markedly decreased the expressions of iNOS, IL-1β, TNF-α, TRAF6 and p-STAT1, but significantly increased ARG1 and IL-10 expressions compared to control UC-MSCs-CM. However, CM derived from miR-146a-5p inhibitor-transfected UC-MSCs exerted the opposite results (Fig. 5d, 5e). Similar results were observed in human macrophage cell line THP1. Transfection with miR-146a-5p mimic leaded to promote M2 macrophage polarization, and transfection with miR-146a-5p inhibitor resulted in facilitating M1 macrophage polarization in THP1 cells (Supplementary Fig. 1). Together, these results demonstrate that miR-146a-5p contributes to UC-MSCs-mediated M2 macrophage polarization.
TRAF6 is required for UC-MSC-miR-146a-5p-mediated M2 macrophage polarization.
To further elucidate whether miR-146a-5p regulated macrophage polarization by targeting TRAF6, we used TRAF6 siRNA to knockdown TRAF6 in macrophage cell line and cocultured with miR-146a-5p inhibitor transfected-UC-MSCs. The knockdown efficiency of TRAF6 in RAW264.7 and THP1 was detected by qRT-PCR (Suppl Fig. 2a, 2c) and western blot analyze (Suppl Fig. 2b, 2c). We found that TRAF6 downregulation resulted in decreased expressions of iNOS, IL-1β, TNF-α and p-STAT1, and elevated ARG1 expression in RAW264.7 cells treated with CM derived from miR-146a-5p inhibitor-transfected UC-MSCs (Fig. 6a, 6b). Similar results were observed in THP1, indicating the elevated anti-inflammatory factors and reduced pro-inflammatory factors in the presence of miR-146a-5p inhibitor when TRAF6 was silenced (Suppl Fig. 2e, 2f). These results demonstrate that TRAF6 plays essential role in miR-146a-5p-regulated macrophage polarization.
MiR-146a-5p-modified UC-MSCs enhance the efficacy of anti-inflammation and renal function improvement in DN rats.
To further confirm the regulatory role of miR-146a-5p in macrophage polarization in vivo, we transfected miR-146a-5p mimic or miR-146a-5p mimic NC into UC-MSCs, and transplanted the modified-UC-MSCs into DN rats via the tail vein. Subsequently, we determined the levels of inflammatory cytokines, quantified inflammatory cell infiltration and M1/M2 macrophage marker expression in the experimental rats. The results showed that administration of miR-146a-5p mimic-transfected UC-MSCs in DN rats obviously reduced Cre, BUN and 24h total urinary protein compared to miR-146a-5p mimic NC-transfected UC-MSCs (Fig. 7a). Moreover, miR-146a-5p mimic-transfected UC-MSC administration further attenuated kidney pathological damage (Suppl Fig. 3). As compared with miR-146a-5p mimic NC-transfected UC-MSC administration in DN rats, decreased IL-1βand increased IL-10 production (Fig. 7b and 7c) were observed in kidney tissues and serum from miR-146a-5p mimic-transfected UC-MSCs-treated DN rats. miR-146a-5p mimic-transfected UC-MSC administration in DN rats also resulted in a significant decreased accumulation of Ly6G+ neutrophils (Fig. 7d, 7e) and F4/80+ macrophages (Fig. 7f, 7g), as determined by immunohistochemistry and immunofluorescence staining. Moreover, using double immunofluorescence staining, we found that miR-146a-5p mimic-transfected UC-MSC administration enhanced M2 macrophage ARG1 expression in DN rats (Fig. 8a). Notably, western blot analysis indicated the expression of iNOS, TRAF6 and p-STAT1 were significantly decreased, but ARG1 expression was increased in miR-146a-5p mimic-transfected UC-MSC-treated DN rats (Fig. 8b). Collectively, these results reveal that miR-146a-5p-modified UC-MSCs enhance the efficacy of renal function improvement and regulation of macrophage polarization in DN.