Role of pyruvate kinase M2-mediated metabolic reprogramming during podocyte differentiation

Abstract

Podocytes, a type of highly specialized epithelial cells, require substantial levels of energy to maintain glomerular integrity and function, but little is known on the regulation of podocytes' energetics. Lack of metabolic analysis during podocyte development led us to explore the distribution of mitochondrial oxidative phosphorylation and glycolysis, the two major pathways of cell metabolism, in cultured podocytes during in vitro differentiation. Unexpectedly, we observed a stronger glycolytic profile, accompanied by an increased mitochondrial complexity in differentiated podocytes, indicating that mature podocytes boost both glycolysis and mitochondrial metabolism to meet their augmented energy demands. In addition, we found a shift of predominant energy source from anaerobic glycolysis in immature podocyte to oxidative phosphorylation during the differentiation process. Furthermore, we identified a crucial metabolic regulator for podocyte development, pyruvate kinase M2. Pkm2-knockdown podocytes showed dramatic reduction of energy metabolism, resulting in defects of cell differentiation. Meanwhile, podocyte-specific Pkm2-knockout (KO) mice developed worse albuminuria and podocyte injury after adriamycin treatment. We identified mammalian target of rapamycin (mTOR) as a critical regulator of PKM2 during podocyte development. Pharmacological inhibition of mTOR potently abrogated PKM2 expression and disrupted cell differentiation, indicating the existence of metabolic checkpoint that need to be satisfied in order to allow podocyte differentiation.

Fig. 1. Metabolomics analysis revealed higher lactate production in differentiated podocytes.

a Immunofluorescence staining for nephrin (red), phalloidine for F-actin (green) and DAPI for nuclear (blue) in undifferentiated podocytes (UDPs) or differentiated podocytes (DPs) as indicated (n = 5). Scale bar=5 μm. b Representative western blotting results of nephrin, podocin and synaptopodin confirm the differentiation of podocytes (n = 3). c Representative map of ¹H-NMR spectra in the extracellular medium incubated with the presence of different podocytes (n = 6). UDM: undifferentiated podocyte medium; DM: differentiated podocyte medium. The map shows the significance of metabolites variations between these two classes. Peaks in the positive direction indicate metabolites that are more abundant in the UDM groups. Consequently, metabolites that are more abundant in DM are presented as peaks in the negative direction.

Fig. 2. Glycolysis-related genes and proteins were upregulated in differentiated podocytes.

a Glycolytic pathway with the assayed glycolytic genes in dark blue. b Real-time PCR analysis of glycolysis-related mRNAs performed in UDPs and DPs. mRNAs were normalized to actin and compared to UDPs (n = 3). c Representative blot images of glycolysis-related proteins (n = 3). d All proteins were normalized to tubulin and compared to UDPs. *P < 0.05, **P < 0.01, determined by t test. Data are shown as the means ± SD.

Fig. 3. Differentiation of podocytes stimulated mitochondrial function.

a Representative confocal and electron microscopy (EM) images showing alterations in mitochondrial morphologies between podocytes as indicated. In the confocal images, cells are labeled with MitoTracker Red (red) for mitochondria and DAPI (blue) for nuclear. Left scale bar=2 µm. Right scale bar=500 nm. Pictures show representative fields of over 10 cells photographed. Statistical analyses showing the average size of mitochondria (b) and the proportion of total mitochondrial in podocytes (c), and data were measured by ImageJ. d Mitochondrial mass stained by MitoTracker Red and measured by Flow Cytometer (n = 3). e Mitochondrial membrane potential labeled with the fluorescent dye JC-1 and measured by Flow Cytometer (n = 3). Real-time PCR analysis of mRNAs involved in mitochondrial dynamics (Opa-1 and Drp-1, F) and mitochondrial biogenesis (Pgc-1α and Tfam, G) in cultured podocytes (n = 3). h, i Representative western blotting results of relevant protein levels (n = 3). *P < 0.05, **P < 0.01, determined by t test. Data are shown as the means ± SD.

Fig. 4. Differentiated podocytes preferentially relied on OXPHOS for their energy demands.

a Oxygen consumption rate (OCR) of podocytes, followed by sequential treatments with oligomycin, FCCP and rotenone/antimycin A. b–e Statistical analyses of baseline respiratory capacity, ATP-coupled respiratory capacity, maximum respiratory capacity and reserve respiratory capacity in OCR. f Extracellular acidification rate (ECAR) in cultured podocytes, followed by sequential treatments with glucose, oligomycin A, and 2-deoxyglucose (2-DG). g–i Statistical analyses of glycolysis, glycolytic capacity and glycolytic reserve in ECAR. j Intracellular ATP level was normalized by protein content, and undifferentiated podocyte ATP was used as control. k–l ATP level in cultured podocyte was measured in response to glycolysis inhibitor oxamate or complex I inhibitor rotenone treatment for 45 min, individually. ATP level was normalized by protein content and expressed as % of control, which was defined as the baseline value in cells exposed only to vehicle (n = 3). m Quantification of lactate and pyruvate ratio in different podocytes (n = 3). n–o ATP levels in primary podocyte was normalized by protein content and expressed as % of control (n = 3). *P < 0.05, **P < 0.01, ^#P < 0.05, ^##P < 0.01, determined by t test. Data are shown as the means ± SD.

Fig. 5. Regulation of PKM2 during in vitro podocyte differentiation.

a Representative blot image of PKM1, PKM2, PKLR proteins (n = 3) in podocytes. b Bar charts show means of optical density (O.D.), and normalized to undifferentiated podocytes. c Real-time PCR analysis of Pkm1, Pkm2 and Pklr mRNAs performed in UDPs and DPs. Relative expression level was compared to Pkm1 in UDPs (n = 3). d PK activity in podocytes as indicated (n = 3). e Representative blot image of cross-linked podocytes (n = 3). f Tetramer, dimer and monomer PKM2 were compared to tubulin and normalized with undifferentiated podocytes (n = 3). g Immunofluorescence staining for PKM2 (green), mitochondria (red) and DAPI (blue) in podocytes as indicated. Mitochondria and nuclear from mature podocytes showed more staining of PKM2 (n = 10). Scale bar=2 μm. h Representative blot image of phosphorylated PKM2 on Tyrosine 105 (n = 3). i–j Mitochondria and nuclear fractions were prepared from undifferentiated or differentiated podocyte. Western blot analysis of cytosolic, mitochondrial and nucleonic fractions was performed to evaluate translocation of PKM2 from the cytosolic compartment to mitochondria and nuclear. *P < 0.05, **P < 0.01, determined by t test. Data are shown as the means ± SD.

Fig. 6. The effect of Pkm2 depletion on differentiation, mitochondria OXPHOS and glycolysis.

a Representative western blotting results of PKM2 and PKM1 in control (shCtrl) and Pkm2-knockdown (shPkm2) podocytes (n = 3). b Bar charts show means of optical density (O.D.), and normalized to cells transfected with shCtr-RNAi lentivirus. c Real-time PCR analysis of Pkm2 and Pkm1 mRNA levels in the absence or presence of Pkm2-RNAi lentivirus (n = 3–6). d PK activity in different podocytes as indicated (n = 3). e ATP level was normalized by protein content, and used shCtrl podocyte as control. f Representative western blotting results of nephrin and synaptopodin expression level in cultured podocytes as indicates (n = 3). g Representative western blotting results of PGC-1α and TFAM expression level (n = 3). h Representative blot images of OPA-1 and DRP-1 (n = 3). i Representative blot images of F6PK and LDHA (n = 3). j Immunofluorescence staining for nephrin (red), phalloidine for F-actin (green) and DAPI for nuclear (blue). Scale bar=5 μm. k Representative immunofluorescence and electron microscopy (EM) images showing alterations in mitochondrial morphologies between different podocytes as indicated. In the immunofluorescence images, cells are labeled with MitoTracker Red (red) for mitochondria and DAPI (blue) for nuclear. Left scale bar = 2 µm. Right scale bar = 500 nm. Pictures show representative fields of over 10 cells photographed. l–n Average mitochondrial area, mitochondrial mass and MMP. o, p Relative lactate production and glucose consumption in culture medium derived from podocytes, and shCtrl podocyte was used as control (n = 3). q Effects of PKM2 on OCR in podocytes (n = 4). OCR traces were obtained using a Seahorse XF96 Analyzer. r Statistical analyses of baseline respiratory capacity, ATP-coupled respiratory capacity, maximum respiratory capacity and reserve respiratory capacity in OCR. s Effects of PKM2 on ECAR in podocytes (n = 4). t Statistical analyses of glycolysis, glycolytic capacity and glycolytic reserve in ECAR.

Fig. 7. Pkm2 deletion in podocytes aggravated adriamycin-induced podocyte injury.

a Representative western blotting results of PKM2 and PKM1 in primary podocytes isolated from Pkm2^fl/fl and Pkm2^−/− mice (n = 3). b Bar charts show means of optical density (O.D.), and normalized to Pkm2^fl/fl podocytes. c Real-time PCR analysis of Pkm2 and Pkm1 mRNA levels (n = 5–6). d PK activity in different podocytes as indicated (n = 5). e ATP level was normalized by protein content, and used Pkm2^fl/fl podocyte as control (n = 5). f Real-time PCR analysis of glycolysis-related mRNAs. mRNAs were normalized to actin, and Pkm2^fl/fl podocyte was used as control (n = 6). g Real-time PCR analysis of mRNAs involved in mitochondrial biogenesis (Pgc-1α and Tfam) and mitochondrial dynamics (Opa-1 and Drp-1) in primary podocytes (n = 6). h Representative immunofluorescence and electron microscopy (EM) images showing alterations in mitochondrial morphologies between different podocytes as indicated. In the immunofluorescence images, cells are labeled with MitoTracker Red (red) for mitochondria and DAPI (blue) for nuclear. Pictures show representative fields of over 10 cells photographed. Left scale bar=2 µm. Right scale bar=500 nm. i Immunofluorescence staining for nephrin (red), phalloidine for F-actin (green) and DAPI for nuclear (blue) in primary podocytes as indicated. Scale bar=5 μm. j Effects of PKM2 on ECAR in primary podocytes. k Statistical analyses of glycolysis, glycolytic capacity and glycolytic reserve in ECAR. l Effects of PKM2 on OCR in primary podocytes. m Statistical analyses of baseline respiratory capacity, ATP-coupled respiratory capacity, maximum respiratory capacity and reserve respiratory capacity in OCR. n Urinary albumin levels in podo-Pkm2^fl/fl and podo-Pkm2^−/− mice after ADR injection (25 mg/kg body weight) at day 5 (n = 4–6). o Quantitative determination of WT1-positive cells in the glomeruli in different groups as indicated. p Immunofluorescence staining for Wilm’s tumor 1 (WT-1) and nephrin in different groups as indicated. Scale bar=20 μm. q Representative electron microscopy (EM) shows podocyte foot process effacement after ADR injection. Top scale bar=1 µm. Bottom scale bars =500 nm.

Fig. 8. mTOR signaling pathway regulated PKM2 expression and podocyte differentiation.

a IHC detection of p-S6 in renal cortex during postnatal development. Scale bar=20 μm. b Representative western immunoblot analysis of p-S6 in renal cortex. c Representative western immunoblot analysis of p-S6 in cultured podocyte during differentiation. d Western immunoblot analysis of p-S6 and PKM2 expression in Tsc1^fl/fl and Tsc1^−/−-podocytes. e Rapamycin caused a significant reduction in the protein levels of p-S6 and PKM2 in cultured podocyte. f Western immunoblot analysis of nephrin and synaptopodin with 10 nM rapamycin treatment for 14 days.