DNA-dependent Protein Kinase Catalytic Subunit (DNA-PKcs) Drives Chronic Kidney Disease Progression in Male Mice Ⅱ

DNA-PKcs-mediated phosphorylation of TAF7 aggravates renal fibrosis

To further dissect the possible molecular mechanisms by which DNAPKcs regulate the development of CKD, phosphoproteomics was performed to reveal possible substrates of DNA-PKcs. As shown in Fig. 5a, most of the differentially phosphorylated proteins were nuclear proteins. The phosphorylation levels of only 6 proteins of these proteins were significantly decreased in kidney tissues of DNA-PKcs−/− mice compared to WT mice, as shown by the volcano plots (Fig. 5a). Heatmap analysis of differentially phosphorylated proteins between DNA-PKcs−/− mice and WT controls showed that phosphorylation levels of the indicated sites of Slc4a1, Ampd2, Hbb-b1, Gas2, Ank1 and TAF7 were significantly decreased after DNA-PK knockout (Fig. 5a). Through subcellular localization analysis and knockdown of these proteins, we found that DNA-PKcs mediates renal fibrosis, perhaps through phosphorylation of TAF7, which is a subunit of TFIID. Coimmunoprecipitation studies revealed a possible direct interaction between DNA-PKcs and TAF7 (Fig. 5b). Protein-protein docking between DNA-PKcs and TAF7 was analyzed through an online database ClusPro39,40. The docking results showed a possible direct interaction between DNA-PKcs and TAF7, and the 213-site serine of TAF7 is one of the nearest sites perhaps binding to DNA-PKcs (Supplementary Fig. 6a). An in vitro kinase assay was performed to analyze whether TAF7 was phosphorylated by DNA-PKcs directly. The products of the kinase assay were analyzed by mass spectrometry. The results showed the S213, T137, Y24, and S171 sites of human TAF7 were phosphorylated by DNA-PKcs (Supplementary Fig. 6b). Furthermore, our results revealed a positive association between the protein levels of p-DNA-PKcs and TAF7 in human CKD kidney tissues via immunofluorescence co-staining of pDNA-PKcs with TAF7 (Fig. 5c).

CLICK HERE TO LEARN KNOW HERBAL FORMULATION CISTANCHE FOR KIDNEY

A previous study found the calculated molecular weight of TAF7 is about 40 kDa, but phosphorylated TAF7 is approximately 55 kDa41. Western blot and immunohistochemical staining results showed that levels of TAF7 were greatly induced by UUO in the kidney tissues of WT controls (Fig. 5d and Supplementary Fig. 6d). DNA-PKcs knockout inhibited the phosphorylation of TAF7 in both the baseline and UUO models compared to WT controls (Fig. 5d). The phosphorylation of TAF7 in UUO mice was also confirmed by Phos-tag SDS-PAGE42 (Supplementary Fig. 6c). Similar results were also observed in UIR model (Fig. 5e). Treatment with the DNA-PKcs inhibitor NU7441 also inhibited the upregulated phosphorylation of TAF7 induced by UIR in mice (Fig. 5f). Additionally, protein levels of TAF7 analyzed by Western blot showed that both DNA-PKcs knockout in HK-2 cells (Fig. 5g) and treatment with NU7441 in HK2 (Fig. 5h) or NRF-49F cells (Fig. 5i) inhibited the upregulated phosphorylation of TAF7 induced by TGFβ1 in vitro.

Since DNA-PKcs is involved in the phosphorylation of TAF7 in renal fibrosis, we examined whether TAF7 deficiency may affect the dedifferentiation of renal epithelial cells or the activation and proliferation of interstitial fibroblasts induced by TGFβ1. TAF7 knockout mPTC cells and NRK-49F cells were generated using CRISPR/Cas9, and successful construction of TAF7 knockout cells was confirmed by Western blot (Supplementary Fig. 6e). Our results showed that TAF7 knockout ameliorated the dedifferentiation of renal epithelial cells, as shown by low levels of FN and collagen I production after exposure to TGFβ1 (Fig. 5j, l). Similarly, TAF7 knockout blunted the profibrotic phenotype of NRF-49F cells induced by TGFβ1, as shown by low levels of FN production (Fig. 5k), collagen I (Fig. 5m), α-SMA (Fig. 5n) as well as low levels of EdU staining (Fig. 5o) in TAF7−/− cells. To determine whether overexpression of TAF7 in renal epithelial cells is sufficient to drive a profibrotic phenotype, HK-2 cells or NRK49F cells were transfected with a human TAF7 or its mutant’s overexpression plasmids. Western blot results showed that the profibrotic phenotype of HK-2 or NRF49F cells was indeed enhanced by overexpression of TAF7 or its mutants (Supplementary Fig. 6h,j). However, the profibrotic effects of TAF7 were almost entirely blocked by DNA-PKcs knockout in HK-2 cells (Supplementary Fig. 6i). Furthermore, we examined whether TAF7 deficiency may affect the profibrotic effects of DNA-PKcs. Interestingly, TAF7 deficiency inhibited the profibrotic effects of DNA-PKcs overexpression (Supplementary Fig. 6f) and blocked the anti-profibrotic effects of NU7441 in mPTC cells (Supplementary Fig. 6g). Recently, Wang S, et al. found cytoplasmic DNA-PKcs were increased and interacted with Fis1 and phosphorylated it at Thr34 in its TQ motif, which increased the affinity of Fis1 for Drp1 and induced mitochondrial fragmentation in AKI43. In our study, nuclear extraction protein analysis results showed DNA-PKcs was mainly localized and increased in the nucleus in kidneys of UUO mice which was different from AKI model (Supplementary Fig. 6k). And knocking down Fis1 promoted the profibrotic response in renal tubular epithelial cells challenged with TGFβ1, suggesting that Fis1 did not mediate the profibrotic role of DNA-PKcs in CKD (Supplementary Fig. 6l). Taken together, these data indicated that DNA-PK mediated phosphorylation of TAF7 aggravates renal fibrosis.

TAF7 promotes mTORC1 activation by upregulating the expression of RAPTOR 

To investigate the mechanisms of the profibrotic effects of TAF7, RNA-Seq was performed to analyze transcriptomic changes in NC and TAF−/− mPTC cells treated with or without TGFβ1. RNA-Seq results showed that TAF7 knockout ameliorated TGFβ1-induced fibrosis-associated gene expression in mPTCs (Supplementary Fig. 7a, b). KEGG pathway enrichment analysis showed that the mTOR signaling pathway was the top pathway between TAF−/− and NC mPTC cells without TGFβ1 treatment (Fig. 6a). Gene set enrichment analysis (GSEA) revealed a significant upregulation of the mTOR signaling pathway after treatment with TGFβ1, which was significantly decreased after TAF7 knockout in mPTC cells compared to NC control cells (Fig. 6a). Heatmap analysis of mTOR signaling demonstrated that RPTOR (RAPTOR), a positive regulator of mTORC144,45, was increased by treatment with TGFβ1, which was inhibited by knockout of TAF7 in mPTC cells (Fig. 6b).

Protein levels of RPTOR and phosphorylation of mTOR were both significantly increased after treatment with TGFβ1 in mPTC cells as analyzed by Western blot. However, TAF7 deficiency decreased the protein levels of RPTOR and the phosphorylation of mTOR induced by TGFβ1 in mPTC cells compared to the NC controls (Fig. 6c). MRNA levels of mTOR and its protein partners, including Raptor, Rictor, and Mlst8, were analyzed by qRT-PCR, and the results showed that TGFβ1 treatment upregulated the mRNA levels of Rptor, Rictor, and Mlst8 but not mTOR. TAF7 deficiency significantly decreased the mRNA levels of Raptor both with and without TGFβ1 treatment in mPTC cells compared to NC control cells but not Rictor, Mlst8, or mTOR (Supplementary Fig. 7c). Similarly, DNA-PKcs deficiency or inhibition significantly decreased mRNA levels of Raptorin kidney tissues of UUO mice and UIR mice (Supplementary Fig. 7d, e). Additionally, we constructed two different residues 213 serine site mutants of TAF7, TAF7- S213D, and TAF7-S213A to analyze the phosphorylation of serine 213 of TAF7 on transcriptional activation of human Raptor induced by TGFβ1 in HK-2 cells. The luciferase assay results showed that overexpression of all these TAF7 mutants increased transcriptional activation of human Raptor, although the relative luciferase activity of the TAF7- S213A group was lower than that of the TAF7-WT and TAF7-S213D groups (Fig. 6d). Moreover, ChIP assay results showed that TAF7 could bind to Rptor promoter directly (Fig. 6e). These results indicated that TAF7 mediates transcriptional activation of RPTOR induced by TGFβ1, perhaps not only through phosphorylation of residue 213. Furthermore, our results showed that protein levels of RPTOR and phosphorylation of mTOR were both increased in kidney tissues of UUO and UIR mice and were significantly decreased after DNA-PKcs knockout in both baseline and model mice (Fig. 6f–i). Similar results were also observed in response to DNA-PKcs knockout HK-2 cells and NU7441 treatment in NRF-49F cells (Fig. 6j, k). Additionally, Western blot results showed that protein levels of RPTOR and phosphorylation of mTOR were increased after overexpression of either DNA-PKcs or TAF7 (Supplementary Fig. 7f,g). Moreover, the knockdown of RPTOR markedly inhibited the phosphorylation of mTOR induced by TGFβ1 (Supplementary Fig. 7h). Taken together, these results indicate that DNA-PKcs-mediated phosphorylation of TAF7 promotes mTORC1 activation by upregulating the expression of RPTOR.

Fig. 6 | DNA-PKcs-mediated phosphorylation of TAF7 promotes mTORC1 activation by upregulating RPTOR expression. an RNA-seq showing that mTOR signaling (indicated by red font) was the top differentially expressed signaling factor between the TAF7−/− and NC groups. GSEA showed that mTOR signaling was upregulated in the NC + TGFβ1 group and suppressed in the TAF7−/− + TGFβ1 group (n = 2), TGFβ1 (5 ng/ml) treatment for 24 h. b Heatmap image showing that Rptor (indicated by red font) was decreased in the TAF7−/− groups compared to the NC groups with or without TGFβ1 (5 ng/ml) treatment for 24 h (n = 2 of each group). c Protein levels of RPTOR, mTOR, and phosphorylated mTOR analyzed by Western blot in TAF7−/− and NC mPTCs treated with TGFβ1 (5 ng/ml) for 24 h. Bars represent quantification results (mean ± SD, n = 3 independent experiments). d Luciferase reporter assay in HK-2 cells with TGFβ1 (5 ng/ml) for 24 h. Bars represent the results of 3 independent experiments (mean ± SD). e ChIP assay was performed to analyze TAF7 binds to the promoter of Raptor. Bars represent quantification results (mean ± SD, n = 3 independent experiments). f Protein levels of RPTOR, mTOR and phosphorylated-mTOR were analyzed by Western blot in kidney tissues of DNAPKcs−/− and WT control mice subjected to UUO (day 7) or UIR (day 21) (g), NU7441 (40 mg/kg) and vehicle treatment mice subjected to UUO (h) or UIR (i), in DNAPKcs−/− or NC HK-2 cells (j), NU7441 (0.1 μM) or vehicle treatment NRK-49F cells (k) treated with TGFβ1 (5 ng/ml) for 24 h. Bars represent quantification results (mean ± SD, n = 6 mice of each group and n = 3 independent experiments for cell models). One-way ANOVA followed by Tukey’s multiple comparisons test was used to determine the p-values for d, e. Two-way ANOVAs followed by Šídák’s multiple comparisons test were used to determine the p-values for c, f–k. Source data are provided as a Source Data file.

Inhibition of DNA-PK corrects metabolic reprogramming in vivo and in vitro 

According to several previous studies9,46, aberrant activation of mTORC1 promotes the progression of renal fibrosis by mediating metabolic reprogramming in fibrotic kidneys. Thus, we examined whether inhibition of DNA-PK or TAF7 corrects metabolic reprogramming in fibrotic kidneys. Based on RNA-Seq analysis of transcriptomic changes in TGFβ1-treated NC and TAF−/− mPTC cells, KEGG pathway enrichment analysis showed that the metabolic signaling pathway was one of the most differentially expressed pathways between TGFβ1-treated TAF−/− and NC mPTC cells (Fig. 7a). Furthermore, GSEA revealed significant upregulation of oxidative phosphorylation and the fatty acid oxidation (FAO) pathway after treatment with TGFβ1 in TAF7 knockout mPTC cells compared to NC control cells (Fig. 7a). The glycolysis pathway was significantly decreased when TAF7 was knocked out in mPTCs compared to the NC control (Fig. 7a). These results indicated that TAF7 deficiency corrects TGFβ1-induced metabolic reprogramming in PTC cells.

To understand the role of DNA-PKcs in regulating metabolic reprogramming in fibrotic kidneys. First, the protein and mRNA levels of key enzymes involved in fatty acid oxidation (CPT1, CPT2, ACOX1, and ACOX2) were analyzed by Western blot and qRT-PCR. The results showed that expression of these enzymes was significantly reduced in the kidneys of WT mice under UUO or UIR conditions, but their expression was significantly restored in DNA-PKcs knockout mice (Fig. 7b, c and Supplementary Fig. 8). Second, mitochondrial morphology and function are necessary for oxidative phosphorylation, and electron microscopy revealed obvious fragmentation of the mitochondria in kidney tubule cells of WT mice induced by UUO, which was significantly improved in kidney tubule cells of DNA-PKcs−/− mice (Fig. 7f). These results revealed that the abnormal oxidative phosphorylation and fatty acid metabolism induced by UUO in WT mice is improved by DNA-PKcs knockout.

Glycolysis is a feature of renal fibrosis and promotes the development of renal fibrosis. Protein levels of hexokinase 2 (HK2) and lactate dehydrogenase A (LDHA), which are rate-limiting enzymes for glycolysis, were analyzed by Western blot. The results showed that levels of HK2 and LDHA were significantly upregulated in kidney tissues of UUO mice and UIR mice, indicating increased glycolytic activity in fibrotic kidneys. In contrast, glycolysis was significantly inhibited in kidney tissues of DNA-PKcs-deficient mice compared to WT mice in both the UUO and UIR models, as indicated by lower levels of HK2 and LDHA (Fig. 7d, e). Extracellular acidification rate (ECAR) assays were performed to analyze glycolytic activity in TGFβ1-treated DNA-PKcs−/− HK-2 cells and NU7441-treated NRK-49F cells in vitro. Both knockout of DNA-PKcs and treatment with NU7441 significantly decreased the ECAR, which was upregulated in TGFβ1-treated cells compared to control cells (Fig. 7g–h). Knockout of TAF7−/− also significantly decreased the upregulation of ECAR in mPTCs induced by TGFβ1 (Fig. 7i). Additionally, extracellular lactate levels in the culture medium of DNA-PKcs−/− and NU7441-treated cells were also decreased (Fig. 7j, k).

To understand how metabolic reprogramming was improved in DNA-PKcs−/− mice, we used metabolomics profiling to analyze metabolite changes following UUO surgery (Fig. 8a). UUO induced impaired glucose metabolism in the kidneys of WT mice, as evidenced by the downregulation of glucose metabolites, such as pyruvate, in WT UUO kidneys, which was restored by knockout of DNA-PKcs (Fig. 8a). Additionally, increased accumulation of lactate in the kidneys of WT mice subjected to UUO was also observed, indicating a metabolic shift from oxidative phosphorylation to enhanced glycolysis in kidney cells. However, in the injured kidneys of DNA-PKcs−/− mice, glycolysis was significantly inhibited (Fig. 8a). Moreover, accumulation of carnitine-conjugated long-chain fatty acids, including lauroyl carnitine (Fig. 8a) and myristoyl carnitine (Fig. 8a), was observed in the kidneys of WT UUO mice, suggesting defects in fatty acid metabolism. In contrast, DNA-PKcs knockout largely corrected all these abnormalities induced by UUO. Taken together, these results indicate that inhibition of DNAPK or TAF7 corrects metabolic reprogramming induced by injury or TGFβ1 in vivo and in vitro.

Discussion 

Epithelial dedifferentiation and myofibroblast activation and proliferation, all initiated by cellular injury, promote the progression of CKD2. Although the DNA damage response (DDR) often correlates with renal epithelial injury47, little is known about its potential role in epithelial dedifferentiation and myofibroblast activation in progressive CKD. In this study, our results revealed that the expression of DNAPKcs, a DNA-dependent protein kinase catalytic subunit, increased in fibrotic kidneys and promoted the development of CKD. Although DNA-PKcs is activated by DNA double-stranded breaks (DSBs)24,25 or ROS26, our results showed DNA-PKcs expression was also activated by TGFβ1-SMAD signaling, while DNA-PKcs knockout inhibited SMAD2/ SMAD3. These results suggest a new possible pathway mediating TGFβ1-SMAD signaling activation in fibrosis.

Next, we generated global prkdc gene-knockout mice. Our results revealed that the deletion of DNA-PKcs attenuated renal tubular injury and the progression of renal interstitial fibrosis in both UUO and UIR mouse models. We cannot conclude that DNA-PK promotes renal fibrosis independent of lymphocytes, as the most prominent phenotype of DNA-PKcs−/− mice is lymphocyte deficiency27,48. To exclude the potential effect of lymphocyte deficiency, we constructed proximal renal tubular epithelial cell with specific DNA-PKcs knockout in vivo by using CRISPR/cas9 knock-in mice. Our results revealed that renal tubular-specific deletion of DNA-PKcs also hampered the progression of renal interstitial fibrosis in UUO. Additionally, our results also showed that DNA-PKcs deficiency preserved the tubular epithelial cell phenotype and regulated interstitial fibroblast activation in vitro. These results indicate that DNA-PKcs mediates epithelial dedifferentiation and myofibroblast activation, probably without any direct relationship with lymphocyte deficiency. Additionally, we investigated the antifibrotic effects of a highly specific DNA-PKcs inhibitor, NU7441, and the results showed that NU7441 treatment markedly attenuated the progression of renal interstitial fibrosis in both UUO and UIR mouse models. NU7441 is a partial inhibitor of DNA-PK at physiological doses. A previous study indicated that NU7441 treatment did not significantly affect B cell function in adult mice33. Moreover, we confirmed the specific action of NU7441 on DNA-PKcs by treating DNA-PKcs knockout mice with NU7441 in UUO mouse model. These results suggest a potential transformation of NU7441 for the clinical treatment of CKD, although further research is needed.

Our results showed that DNA-PKcs deficiency did not aggravate DSBs in vivo and in vitro, suggesting that DNA-PKcs mediates renal injury and renal interstitial fibrosis, probably not through its role in DDR. Phosphoproteomics analysis showed that phosphorylation of TAF7 was decreased in the kidney tissues of DNA-PKcs−/− mice. TAF7 deficiency inhibited the profibrotic phenotype of renal epithelial cells and fibroblasts induced by TGFβ1. The profibrotic effects of TAF7 were almost entirely blocked by DNA-PKcs deficiency in renal epithelial cells. Our results indicate that TAF7 is a substrate for DNA-PKcs kinase activity and that DNA-PKcs-mediated phosphorylation of TAF7 aggravates renal fibrosis.

RNA-Seq results showed that TAF7 promotes mTORC1 activation by upregulating the expression of Raptor, which is a positive regulator of mTORC149,50. Phosphoproteome results showed that phosphorylation of TAF7 on serine-213 was significantly decreased in the kidneys of DNA-PKcs−/− mice, compared to that in WT controls. Mass spectrometry analysis indicated the S213, T137, Y24, and S171 sites of human TAF7 were phosphorylated by DNA-PKcs in vitro. Moreover, transfection of different TAF7 mutants, including TAF7-WT, TAF7-S213D, and TAF7-S213A, increased transcriptional activation of human Rptor in HK-2 cells. These results indicate that TAF7 mediates transcriptional activation of Rptor induced by TGFβ1, probably not only through phosphorylation of its 213-site. TAF7 is not a core transcription factor but interacts with other TAFs, such as TAF141,51, and regulates the enzymatic activities of transcription factors, which could explain why overexpression of TAF7 alone did not activate the transcription of human RPTOR in HK-2 cells without TGFβ1 treatment. Several serine residues of TAF7 can be phosphorylated to activate or inhibit transcription41. For instance, a previous study found that phosphorylation of TAF7 on serine 264 disrupted TAF7 binding to TAF1, resulting in transcriptional upregulation of cyclin D141. In addition to TAF1-dependent transcription, TAF7 also regulates TAF1-independent transcription51. In this study, ChIP assay results demonstrated that TAF7 could bind to Rptor promoter directly, suggesting TAF7 perhaps collaborates with other transcription factors to initiate Rptor expression. Although the mechanisms of TAF7-mediated transcriptional upregulation of Rptor need further research, our results indicate that DNA-PKcs mediates the phosphorylation of TAF7 to promote mTORC1 activation by upregulating the expression of Rptor.

Several previous studies have reported that mTORC1 signaling is activated in dedifferentiated epithelial cells and myofibroblasts from fibrotic kidneys9,52. Although DNA-PK has also been reported to phosphorylate and activate AKT/mTOR signaling in tumor cells in response to DNA breaks53, the precise mechanism is unknown. Our results revealed that DNA-PKcs-mediated phosphorylation of TAF7 promotes mTORC1 activation by upregulating the expression of Raptor, providing a possible mechanism of mTORC1 activation in fibrotic kidneys. Chronic activation of mTORC1 promotes metabolic reprogramming in many diseases, including CKD9,54. Recent studies have highlighted that both renal tubular cells and fibroblasts alter metabolic phenotypes in response to CKD9,10,13. The shift from fatty acid oxidative pathways and oxidative phosphorylation to glycolysis is the primary characteristic of metabolic reprogramming9,10. As mTOR plays a central role in maintaining metabolic homeostasis, inhibition of mTOR displays side effects. Moreover, long-term inhibition of TGF-β signaling is also associated with unacceptable adverse effects57. In this study, we found that the activity of DNA-PKcs was almost undetectable in normal kidneys, in contrast to its robust enhancement under CKD. Thus, inhibition of DNA-PKcs corrects metabolic reprogramming in fibrotic kidneys.

Fig. 8 | Deletion of DNA-PK corrects metabolic reprogramming. a Metabolomics analysis of kidney tissues from each group as indicated. Heatmap image showing relative levels of metabolites in the glycolysis pathway, fatty acid metabolism, and Krebs cycle in kidneys from each group (n = 4). Bars represent statistical analysis of representative metabolites in the kidneys of each group (mean ± SD, n = 4 mice of each group). One-way ANOVA followed by Tukey’s multiple comparisons test was used to determine the p-values. b Working model (the template was created with BioRender.com) illustrating in which DNA-PKcs mediates activation of Raptor/mTORC1 signaling through phosphorylation of TAF7 and promotes metabolic reprogramming in injured epithelial cells and myofibroblasts.

References 

1. Romagnani, P. et al. Chronic kidney disease. Nat. Rev. Dis. Primers 3, 17088 (2017). 

2. Humphreys, B. D. Mechanisms of renal fibrosis. Annu. Rev. Physiol. 80, 309–326 (2018). 

3. Nath, K. A. The tubulointerstitium in progressive renal disease. Kidney Int. 54, 992–994 (1998). 

4. Zhou, D. & Liu, Y. Renal fibrosis in 2015: Understanding the mechanisms of kidney fibrosis. Nat. Rev. Nephrol. 12, 68–70 (2016). 

5. Wynn, T. A. & Ramalingam, T. R. Mechanisms of fibrosis: therapeutic translation for fibrotic disease. Nat. Med. 18, 1028–1040 (2012). 

6. Meng, X. M., Nikolic-Paterson, D. J. & Lan, H. Y. TGF-beta: the master regulator of fibrosis. Nat. Rev. Nephrol. 12, 325–338 (2016).

Supportive Service Of Wecistanche-The largest cistanche exporter in the China:

Email:wallence.suen@wecistanche.com 

Whatsapp/Tel:+86 15292862950

Shop For More Specifications Details:

https://www.xjcistanche.com/cistanche-shop

CLICK HERE TO GET NATURAL ORGANIC CISTANCHE EXTRACT WITH 25% ECHINACOSIDE AND 9% ACTEOSIDE FOR KIDNEY