YAP1 preserves tubular mitochondrial quality control to mitigate diabetic kidney disease

Abstract

Renal tubule cells act as a primary site of injury in diabetic kidney disease (DKD), with dysfunctional mitochondrial quality control (MQC) closely associated with progressive kidney dysfunction in this context. Our investigation delves into the observed inactivation of yes-associated protein 1 (YAP1) and consequential dysregulation of MQC within renal tubule cells among DKD subjects through bioinformatic analysis of transcriptomics data from the Gene Expression Omnibus (GEO) dataset. Receiver operating characteristic curve analysis unequivocally underscores the robust diagnostic accuracy of YAP1 and MQC-related genes for DKD. Furthermore, we observed YAP1 inactivation, accompanied by perturbed MQC, within cultured tubule cells exposed to high glucose (HG) and palmitic acid (PA). This pattern was also evident in the tubulointerstitial compartment of kidney sections from biopsy-approved DKD patients. Additionally, renal tubule cell-specific Yap1 deletion exacerbated kidney injury in diabetic mice. Mechanistically, Yap1 deletion disrupted MQC, leading to mitochondrial aberrations in mitobiogenesis and mitophagy within tubule cells, ultimately culminating in histologic tubular injury. Notably, Yap1 deletion-induced renal tubule injury promoted the secretion of C-X-C motif chemokine ligand 1 (CXCL1), potentially augmenting M1 macrophage infiltration within the renal microenvironment. These multifaceted events were significantly ameliorated by administrating the YAP1 activator XMU-MP-1 in DKD mice. Consistently, bioinformatic analysis of transcriptomics data from the GEO dataset revealed a noteworthy upregulation of tubule cells-derived chemokine CXCL1 associated with macrophage infiltration among DKD patients. Crucially, overexpression of YAP1 via adenovirus transfection sustained mitochondrial membrane potential, mtDNA copy number, oxygen consumption rate, and activity of mitochondrial respiratory chain complex, but attenuated mitochondrial ROS production, thereby maintaining MQC and subsequently suppressing CXCL1 generation within cultured tubule cells exposed to HG and PA. Collectively, our study establishes a pivotal role of tubule YAP1 inactivation-mediated MQC dysfunction in driving DKD progression, at least in part, facilitated by promoting M1 macrophage polarization through a paracrine-dependent mechanism.

Keywords: Chemokine; Diabetic kidney disease; Hippo signaling pathway; Macrophage; Mitochondria; Yes-associated protein 1.

Graphical abstract

Fig. 1

Yes-associated protein 1 (YAP1) and mitochondrial quality control (MQC)-related mediators were disrupted in renal tubule cells from diabetic kidney. (A, B) Expression of YAP1 and MQC-related genes in tubulointerstitial compartment of subjects with diabetic kidney disease (DKD) in GSE104954. (C) The diagnostic performance of YAP1 and each individual MQC-related genes for DKD. (D) The diagnostic performance of a comprehensive predictive model consisted of YAP1 and MQC-related genes for DKD. (E, F) Representative Western blots and quantification of p-YAP1, YAP1, MQC-related mediators, and interleukin 6 (IL6) in HK-2 cells exposed to palmitic acid and high glucose (PA + HG). (G) Transmission electron microscopy images of mitochondria within HK-2 cell with or without PA + HG treatment. Bar = 200 nm. (H) Transcripts of YAP1, MQC-related mediators, and proinflammatory genes were measured by quantitative real-time polymerase chain reaction. (I, J) Representative plots and quantification of flow cytometry analysis for JC-1 staining in HK-2 cells with or without PA + HG treatment. The JC-1 red/green fluorescence ratio was calculated to represent mitochondrial membrane potential. (K, L) Representative plots and quantification of flow cytometry analysis for MitoSox Red staining in HK-2 cells with or without PA + HG treatment. The mean fluorescence intensity of MitoSox Red was calculated to represent mitochondrial reactive oxygen species level. Data are expressed as the mean ± SEM. All experiments were repeated at least three times. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001. PGC1ɑ, peroxisome proliferator-activated receptor γ coactivator α; TFAM, mitochondrial transcription factor A; TOMM20, translocase of the outer mitochondrial membrane complex subunit 20; PINK1, tensin homolog induced kinase 1; LC3B, microtubule-associated protein 1 light chain 3B; p62, ubiquitin-binding protein sequestosome 1; IL8, interleukin 8; p-YAP1, phosphorylated YAP1.

Fig. 2

Hippo signaling pathway was activated in tubule cells under the context of diabetic kidney disease (DKD). (A) Representative images of the immunofluorescence staining for yes-associated protein 1 (YAP1) in cultured HK-2 cells with or without palmitic acid and high glucose (PA + HG) treatment. Bar = 50 μm. (B, C) Representative Western blots and quantification of YAP1 in cytoplasmic and nuclear fraction of HK-2 cells exposed to PA + HG. (D) Representative images of immunohistochemistry staining for YAP1 and p-YAP1 in kidney sections from control donors, early-stage DKD patients, and late-stage DKD patients. Bar = 50 μm. (E, F) Quantitative assessment for percentage of YAP1 and p-YAP1 expression (n = 8 for each group). (G) Transcripts of genes involved in Hippo signaling pathway were measured by quantitative real-time polymerase chain reaction in cultured HK-2 cells with or without PA + HG treatment. (H, I) Representative Western blots and quantification of key molecules involved in Hippo signaling pathway in HK-2 cells exposed to PA + HG. (J) Schematic diagram showing the activation of Hippo signaling pathway and the ultimately degraded YAP1 in tubule cells under the context of DKD. Data are expressed as the mean ± SEM. All experiments were repeated at least three times. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001. MST1, serine/threonine kinase 4; MST2, serine/threonine kinase 3; p-MST1/2, phosphorylated MST1/2; MOB1, MOB kinase activator 1; p-MOB1, phosphorylated MOB1; LATS1, large tumor suppressor kinase 1; LATS2, large tumor suppressor kinase 2; p-LATS1/2, phosphorylated LATS1/2; TAZ, Tafazzin; TEAD, TEA domain transcription factor; DAPI, 4′,6-diamidino-2-phenylindole; p-YAP1, phosphorylated YAP1.

Fig. 3

Renal tubule cell-specific yes-associated protein 1 (Yap1) deletion aggregated kidney injury via disrupting mitochondrial quality control (MQC) in diabetic kidney disease (DKD) mice. (A) The schematic diagram of in vivo experiment. (B–I) Body weight, kidney weight/body weight, blood glucose, serum triglyceride, cholesterol, low-density lipoprotein cholesterol (LDL-c), and creatinine and urinary albumin-to-creatinine ratio (UACR) in the indicated mouse groups. (J) Representative images of histologic images of Periodic acid-Schiff (PAS) staining, Masson's trichrome (MASSON) staining, as well as immunohistochemistry staining for kidney injury molecule-1 (KIM1) and ɑ-smooth muscle actin (ɑ-SMA) in kidney sections. Black arrows indicated expanded mesangial expansion and tubular atrophy, dilation, or brush border loss. Bar = 50 μm. (K-M) Quantitative assessment for tubular injury score, percentage of collagen staining and percentage of KIM1 expression. (N, O) Representative Western blots and quantification of p-YAP1, YAP1, MQC-related mediators, and interleukin 6 (IL6) in renal cortex from indicated mouse groups. (P) Representative transmission electron microscopy images of mitochondria with abnormal morphology in tubule cells from indicated mouse groups. Damaged mitochondria were observed with matrix swelling and collapsed cristae. An early phase of mitophagy was seen as the formation of double membrane, wrapped around a mitochondrion. The red arrowheads indicate autophagic vacuoles containing damaged mitochondria, or autophagosomes surrounded by a double-membrane with undigested damaged mitochondria inside. Bar = 1 μm. Mt, mitochondrion; Av, autophagic vacuole; N, nucleus. Data are expressed as the mean ± SEM. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001 (n = 6 for each group). ND, normal diet; HFD, high-fat diet; STZ, streptozotocin; TKO, tubule cell-specific knockout; Veh, vehicle; PGC1ɑ, peroxisome proliferator-activated receptor γ coactivator α; TFAM, mitochondrial transcription factor A; TOMM20, translocase of the outer mitochondrial membrane complex subunit 20; PINK1, tensin homolog induced kinase 1; p62, ubiquitin-binding protein sequestosome 1; LC3B, microtubule-associated protein 1 light chain 3B; p-YAP1, phosphorylated YAP1.

Fig. 4

Renal tubule cells-specific yes-associated protein 1 (Yap1) deletion boosted CXCL1 secretion and affected macrophage polarization in diabetic kidney. (A) Overlapped differentially expressed genes (DEGs) recognized among GSE104954, GSE175759 and GSE30529. (B) Gene ontology enrichment analyses of overlapped DEGs. (C) Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of overlapped DEGs. (D) Renal expression of chemokines C-X-C motif chemokine ligand 1 (CXCL1), C–C motif chemokine ligand (CCL19) and CXCL6 in 3 selected GSE datasets (GSE30529, GSE104954, and GSE175759). (E) Transcripts of CXCL1, CCL19 and CXCL6 were measured by quantitative real-time polymerase chain reaction in cultured HK-2 cells with or without palmitic acid and high glucose treatment (PA + HG). (F) Representative images of the immunofluorescence staining for CXCL1 in cultured HK-2 cells with or without PA + HG treatment. Bar = 20 μm. (G) Representative images of the immunofluorescence staining for CXCL1 in kidney sections from indicated mouse groups. Bar = 15 μm. (H) Representative Western blots and quantification of CXCL1 in renal cortex from indicated mouse groups. (I) Representative images of the immunofluorescence staining for CD68 and CD206 in kidney sections from indicated mouse groups. Bar = 15 μm. Data are expressed as the mean ± SEM. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001 (n = 6 for each group). ND, normal diet; HFD, high-fat diet; STZ, streptozotocin; TKO, tubule cell-specific knockout; Veh, vehicle; DAPI, 4′,6-diamidino-2-phenylindole.

Fig. 5

C-X-C motif chemokine ligand 1 (CXCL1) expression and macrophage infiltration were enhanced in human diabetic kidney. (A) The identified cell clusters by t-distributed Stochastic Neighbor Embedding (t-SNE) analysis in single nucleus RNA-sequencing (snRNA-seq) dataset merged by GSE131882, GSE151302 and GSE195460. PT, proximal convoluted tubule; PTVCAM1, vascular cell adhesion molecule 1 (VCAM1)-positive proximal convoluted tubule; PEC, parietal epithelial cells; ATL, ascending thin limb; TAL, thick ascending limb; DCT, distal convoluted tubule; PC, principal cells; ICA, type A intercalated cells; ICB, type B intercalated cells; PODO, podocyte; ENDO, endothelial cells; MES, mesangial cells; FIB, fibroblasts; LEUK, leukocytes. (B-D) Bubble charts show the expression of chemokine genes (CXCL1, CCL19, and CXCL6) in tubule cells from snRNA-seq dataset merged by GSE131882, GSE151302 and GSE195460. (E) The distribution of immune cells in kidney from snRNA-seq dataset merged by GSE131882, GSE151302 and GSE195460. Red dashed circle shows macrophage. (F) The distribution of macrophage, T cell, B cell and plasma cell in kidney from snRNA-seq dataset merged by GSE131882, GSE151302 and GSE195460. Red dashed circle shows macrophage. (G) The expression patterns of macrophage, T cell, B cell and plasma cell. UMAP, uniform manifold approximation and projection; CXCL1, C-X-C motif chemokine ligand 1; CCL19, C–C motif chemokine ligand; CXCL6, C-X-C motif chemokine ligand 6.

Fig. 6

Yes-associated protein 1 (Yap1) activation counteracted diabetes-induced kidney injury via maintaining mitochondrial quality control (MQC) homeostasis in diabetic kidney. (A) The schematic diagram of in vivo experiment. (B–I) Body weight, kidney weight/body weight, blood glucose, serum triglyceride, cholesterol, low-density lipoprotein cholesterol (LDL-c), and creatinine and urinary albumin-to-creatinine ratio (UACR) in the indicated mouse groups. (J) Representative images of histologic images of Periodic acid-Schiff (PAS) staining, Masson's trichrome (MASSON) staining, as well as immunohistochemistry staining for kidney injury molecule-1 (KIM1) in kidney sections. Black arrows indicated expanded mesangial expansion and tubular atrophy, dilation, or brush border loss. Bar = 50 μm. (K-M) Quantitative assessment for tubular injury score, percentage of collagen staining and percentage of KIM1 expression. (N, O) Representative Western blots and quantification of p-YAP1, YAP1, MQC-related mediators, and interleukin 6 (IL6) in renal cortex from indicated mouse groups. (P) Representative images of the immunofluorescence staining for C-X-C motif chemokine ligand 1 (CXCL1) in kidney sections from indicated mouse groups. Bar = 15 μm. (Q-R) Representative Western blots and quantification of CXCL1 in renal cortex from indicated mouse groups. (S) Representative transmission electron microscopy images of mitochondria with abnormal morphology in tubule cells from indicated mouse groups. Damaged mitochondria were observed with matrix swelling and collapsed cristae. An early phase of mitophagy was seen as the formation of double membrane, wrapped around a mitochondrion. The red arrowheads indicate autophagic vacuoles containing damaged mitochondria, or autophagosomes surrounded by a double-membrane with undigested damaged mitochondria inside. Bar = 1 μm. Mt, mitochondrion; Av, autophagic vacuole; N, nucleus. Data are expressed as the mean ± SEM. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001 (n = 6 for each group). HFD, high-fat diet; STZ, streptozotocin; DKD, diabetic kidney disease; Veh, vehicle; PGC1ɑ, peroxisome proliferator-activated receptor γ coactivator α; TFAM, mitochondrial transcription factor A; TOMM20, translocase of the outer mitochondrial membrane complex subunit 20; PINK1, tensin homolog induced kinase 1; LC3B, microtubule-associated protein 1 light chain 3B; p62, ubiquitin-binding protein sequestosome 1; DAPI, 4′,6-diamidino-2-phenylindole; p-YAP1, phosphorylated YAP1.

Fig. 7

Yes-associated protein 1 (YAP1) maintained mitochondrial quality control (MQC) homeostasis in cultured HK-2 cells. (A, B) Representative Western blots and quantification of p-YAP1, YAP1, MQC-related mediators, and interleukin 6 (IL6) in adenovirus (Ad)-infected HK-2 cells with or without palmitic acid and high glucose (PA + HG) treatment. (C, D) Representative plots and quantification of flow cytometry analysis for JC-1 staining in Ad-infected HK-2 cells with or without PA + HG treatment. The JC-1 red/green fluorescence ratio was calculated to represent mitochondrial membrane potential. (E, F) Representative plots and quantification of flow cytometry analysis for MitoSox Red staining in Ad-infected HK-2 cells with or without PA + HG treatment. The mean fluorescence intensity of MitoSox Red was calculated to represent mitochondrial reactive oxygen species level. (G) The relative mitochondrial DNA (mtDNA) copy numbers in Ad-infected HK-2 cells with or without PA + HG treatment were measured by qPCR. (H) Representative confocal images of mt-Keima in Ad-infected HK-2 cells with or without PA + HG treatment. Confocal microscopy was analyzed to detect the mt-Keima located in mitochondria (mitochondria at neutral pH, green fluorescence) and the mt-Keima delivered to lysosomes (mitochondria at acidic pH, red fluorescence). The zoom images were magnified from boxed areas in overlay images. Bar = 40 μm. (I) Quantification of mitophagy index by mt-Keima imaging. Mitophagy index was determined by analyzing the ratio of red/green fluorescence. (J) Real-time measurements of the oxygen consumption rate (OCR) in Ad-infected HK-2 cells with or without PA + HG treatment via Seahorse XF96. After basal OCR was obtained, oligomycin (1.5 μM) was added to obtain ATP-linked OCR. Then, the uncoupler FCCP (1 μM) was added to obtain maximal OCR. Finally, none-mitochondrial OCR was obtained after adding Antimycin A + rotenone (0.5 μM each) to inhibit the electron transport chain. Mitochondrial spare respiratory capacity was calculated by subtracting basal respiration from maximal respiratory capacity. Each point in the lines represents the average measurements of six different wells. (K) OCR results from the Seahorse analysis of the mitochondrial respiration parameters: basal respiration, spare respiratory capacity, maximal respiration, and ATP production. (L) Mitochondrial respiratory chain complex enzyme activities in Ad-infected HK-2 cells with or without PA + HG treatment. (M) ELISA of CXCL1 level for Ad-infected HK-2 cells with or without PA + HG treatment. (N) ELISA of CXCL1 level for Ad-infected HK-2 cells with or without mitophagy inhibitor Mdivi-1 treatment. Data are expressed as the mean ± SEM. All experiments were repeated at least three times. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001. NC, negative control; PGC1ɑ, peroxisome proliferator-activated receptor γ coactivator α; TFAM, mitochondrial transcription factor A; TOMM20, translocase of the outer mitochondrial membrane complex subunit 20; PINK1, tensin homolog induced kinase 1; p62, ubiquitin-binding protein sequestosome 1; IL6, interleukin 6; CXCL1, C-X-C motif chemokine ligand 1; p-YAP1, phosphorylated YAP1; FCCP, carbonyl cyanide-p-trifluoromethoxyphenylhydrazone; ELISA, enzyme-linked immunosorbent assay.

Fig. 8

Schematic diagram showing the yes-associated protein 1 (YAP1) as a critical molecular pattern for diabetic kidney disease (DKD). The initiation of DKD within tubule cells triggers the activation of the Hippo signaling pathway, leading to the degradation of the downstream core mediator YAP1. This cascade results in the disruption of YAP1-mediated mitochondrial quality control, characterized by impaired mitobiogenesis and mitophagy. Consequently, there is an elevation in the production of mitochondrial reactive oxygen species (mtROS), a decline in mitochondrial membrane potential (MMP, ΔΨm), mtDNA copy numbers, oxygen consumption rate, and the enzymatic activity of mitochondrial respiratory chain complexes. These molecular alterations stimulate the synthesis of inflammatory cytokines (such as IL6) and chemokines (CXCL1, CXCL6, and CCL19) by tubule cells, instigating the recruitment and activation of macrophages. This intricate process significantly contributes to the pathogenesis of DKD. MST1, serine/threonine kinase 4; MST2, serine/threonine kinase 3; MOB1, MOB kinase activator 1; LATS1, large tumor suppressor kinase 1; LATS2, and large tumor suppressor kinase 2; TAZ, Tafazzin; TEAD, TEA domain transcription factor; PGC1ɑ, peroxisome proliferator-activated receptor γ coactivator α; TFAM, mitochondrial transcription factor A; PINK1, tensin homolog induced kinase 1; LC3B-Ⅱ, microtubule-associated protein 1 light chain 3B-Ⅱ; p62, ubiquitin-binding protein sequestosome 1; IL6, interleukin 6; CXCL1, C-X-C motif chemokine ligand 1; CCL19, C–C motif chemokine ligand; CXCL6, C-X-C motif chemokine ligand 6.