Mesenchymal stem cells prevent the progression of diabetic nephropathy by improving mitochondrial function in tubular epithelial cells

Abstract

The administration of mesenchymal stem cells (MSCs) was shown to attenuate overt as well as early diabetic nephropathy in rodents, but the underlying mechanism of this beneficial effect is largely unknown. Inflammation and mitochondrial dysfunction are major pathogenic factors in diabetic nephropathy. In this study, we found that the repeated administration of MSCs prevents albuminuria and injury to tubular epithelial cells (TECs), an important element in the progression of diabetic nephropathy, by improving mitochondrial function. The expression of M1 macrophage markers was significantly increased in diabetic kidneys compared with that in control kidneys. Interestingly, the expression of arginase-1 (Arg1), an important M2 macrophage marker, was reduced in diabetic kidneys and increased by MSC treatment. In cultured TECs, conditioned media from lipopolysaccharide-activated macrophages reduced peroxisomal proliferator-activated receptor gamma coactivator 1α (Pgc1a) expression and impaired mitochondrial function. The coculture of macrophages with MSCs increased and decreased the expression of Arg1 and M1 markers, respectively. Treatment with conditioned media from cocultured macrophages prevented activated macrophage-induced mitochondrial dysfunction in TECs. In the absence of MSC coculture, Arg1 overexpression in macrophages reversed Pgc1a suppression in TECs. These observations suggest that MSCs prevent the progression of diabetic nephropathy by reversing mitochondrial dysfunction in TECs via the induction of Arg1 in macrophages.

Fig. 1. MSC treatment prevented the progression of diabetic nephropathy.

Diabetic mice were treated with vehicle or MSCs 5, 9, and 13 weeks after STZ injection and sacrificed at 24 weeks. a Albuminuria, defined by the UACR, was markedly increased in diabetic mice 4 weeks after STZ injection. b Changes in plasma glucose, BUN, creatinine, and UACR levels in diabetic mice treated with or without MSCs. c Representative H&E, PAS, and MT staining of kidney sections. Scale bars, 50 µm. d–f MSC treatment prevented renal damage in diabetic mice. The mRNA levels of Desmin, a marker of podocyte injury d; α-SMA, and Fn1, markers of fibrosis e; and Kim-1 and Lcn2, markers of tubular epithelial injury f. Con, control non-diabetic CD1 mice; DM, diabetic CD1 mice; DM + MSCs, diabetic CD1 mice with MSC treatment. Data show the means ± SEMs of 6–9 mice. ^*P < 0.05, ^**P < 0.01, and ^***P < 0.001 versus Con; ^#P < 0.05, and ^##P < 0.01 versus DM

Fig. 2. MSC treatment decreased and increased the expression of M1 macrophages and Arg1, respectively, in diabetic kidneys.

a Real-time RT-PCR analysis of Ccl2, Vcam1, and Icam1 expression in the kidneys. b MSC treatment decreased the expression level of M1 macrophage markers. c MSC treatment increased Arg1 expression, whereas it did not alter Arg2 expression. Con, non-diabetic CD1 mice; DM, diabetic CD1 mice; DM + MSCs, diabetic CD1 mice with MSC treatment. Data show the means ± SEMs of six mice. ^*P < 0.05, ^**P < 0.01, and ^***P < 0.001 versus Con; ^#P < 0.05, ^##P < 0.01, and ^###P < 0.001 versus DM

Fig. 3. Coculture of RAW264.7 macrophages with MSCs suppressed LPS-induced M1 macrophage polarization and increased Arg1 expression.

MSCs were added to RAW264.7 cells at a ratio of 1:2 (MSCs: RAW264.7 ratio). After 24 h of incubation, LPSs were added to the cells at a final concentration of 10 ng/ml. cRAW, untreated RAW264.7 cells; aRAW, LPS-treated RAW264.7 cells; aRAW + MSCs, LPS-treated RAW264.7 cells cocultured with MSCs. Data are presented as the means ± SEMs (n = 4). ^***P < 0.001 versus control (cRAW); ^###P < 0.001 versus LPS-treated RAW264.7 cells (aRAW)

Fig. 4. Overexpression of Arg1 suppressed M1 polarization of macrophages.

RAW264.7 cells were transfected with control vector (Con vector) or arginase-1 vector (Arg1 vector) and treated with or without lipopolysaccharides (LPSs). a Nitrite concentration in the culture supernatant of RAW264.7 cells after treatment with LPSs for 24 h. b Expression of Arg1 and M1 markers determined using real-time PCR. Data represent means ± SEMs (n = 5). ^*P < 0.05 versus Con vector-treated RAW264.7 cells without LPS treatment; ^#P < 0.05 versus Con vector-treated RAW264.7 cells with LPS treatment

Fig. 5. MSCs reversed cytokine-mediated mitochondrial dysfunction in TECs.

Conditioned media from RAW264.7 cells (see Fig. 3.) was transferred to HK-2 cells for 24 h. a Effect of MSCs on the mRNA (left panel) and protein expression (right panel) of the mitochondrial biogenesis markers PGC-1α, mtTFA, and COX-IV in HK-2 cells. b Effect of MSCs on mitochondrial respiration. Real-time OCRs were measured using an XF24 extracellular flux analyzer. During the measurements, 1 μg/ml oligomycin (Oligo), 1 μm carbonyl cyanide p-(trifluoromethoxy)-phenyl-hydrazone (FCCP), and 1 μm rotenone (Rote) plus 2 μm antimycin A (AA) were sequentially added. c The area under the curve of the basal OCR (left panel) and the ECAR (right panel). All OCRs and ECARs were normalized based on the cell number. Data show the means ± SEMs (n = 5). ^*P < 0.05, ^**P < 0.01, and ^***P < 0.001 versus control (cRAW → HK-2); ^#P < 0.05 and ^##P < 0.01 versus LPS-treated cells (aRAW → HK-2). d MSC treatment increased the mitochondrial mass. Mitochondria stained with 100 nm MitoTracker Red are shown in red, and nuclei stained with DAPI are shown in blue. The treatment of HK-2 cells with conditioned media from aRAW cells significantly decreased mitochondrial mass. In contrast, the coculture of MSCs with conditioned media from aRAW cells increased mitochondrial mass. Scale bars, 10 µm (n = 4). e MSCs decreased mitochondrial ROS production in HK-2 cells, as measured using MitoSOX-based flow cytometry (n = 4). cRAW, untreated RAW264.7 cells; aRAW, LPS-treated RAW264.7 cells; aRAW + MSCs, aRAW cells cocultured with MSCs. aRAW → HK-2, HK-2 cells were treated with conditioned media from aRAW cells; aRAW + MSCs → HK-2, HK-2 cells treated with conditioned media from aRAW cells cocultured with MSCs

Fig. 6. The overexpression of Arg1 enhanced mitochondrial function in TECs.

a Real-time OCR rates were measured using an XF24 extracellular flux analyzer. b Basal OCRs and ECARs. c Relative mRNA levels of mitochondrial biogenesis markers. Data represent the means ± SEMs (n = 5). ^*P < 0.05 and ^**P < 0.01 versus Con vector-treated RAW264.7 cells without LPS treatment; ^#P < 0.05 and ^##P < 0.01 versus Con vector-treated RAW264.7 cells with LPS treatment

Fig. 7. MSC administration prevented mitochondrial changes in the TECs of diabetic kidneys.

a Representative TEM image of proximal TECs from non-diabetic mice. The normal shape of the mitochondria with a sustained normal pattern of cristae is shown. b Representative TEM image of TECs from diabetic mice. Mitochondria with abnormal shapes, such as small, rounded mitochondria and the loss of a normal cristae pattern, were observed. Some mitochondria exhibited an electron-dense structure owing to the aggregation of cristae membranes or focal enlargement of the intermembrane space (white arrows). c Representative TEM image of TECs from diabetic mice treated with MSCs. Scale bars, 500 nm. The lower panel is a TEM image at a higher magnification highlighting the region of interest boxed in the upper panel. d Quantification of abnormal mitochondria. Data show the means ± SEM, n = 20 images/mouse, n = 2–3 mice/group. ^**P < 0.01 versus control; ^#P < 0.05 versus diabetic mice

Fig. 8. Proposed mechanism by which MSCs prevent the progression of diabetic nephropathy.

a The M1 macrophage population is increased in diabetic kidneys, and pro-inflammatory cytokines suppress mitochondrial biogenesis to induce mitochondrial dysfunction in TECs. Increased ROS produced from dysfunctional mitochondria cause injury to TECs. In addition, diabetes may lead to the direct injury of TECs. TEC injury may in turn activate M1 macrophages to sustain the progression of diabetic nephropathy. b MSCs increase Arg1 expression in macrophages to suppress M1 macrophage polarization and prevent the deterioration of kidney function