SIRT3 Deficiency Sensitizes Angiotensin-II-Induced Renal Fibrosis

Background: Sirtuin 3 (SIRT3) has a crucial role in the cardiovascular diseases. Our previous study revealed that SIRT3 knockout (SIRT3KO) promoted cardiac pericyte–fibroblast transition. In this study, we investigated the involvement of pericyte and iron in angiotensin II (Ang-II)-mediated renal fibrosis in the SIRT3KO mice. Methods and Results: NG2-DsRed mice and NG2-DsRed-SIRT3 knockout (SIRT3KO) mice were infused with saline or Ang-II (1000 ng/kg/min) for 4 weeks. Renal fibrosis, iron content and reactive oxygen species (ROS) were measured. Masson’s trichrome staining showed that SIRT3KO enhanced Ang-II-induced renal fibrosis. Immunostaining showed that Ang-II treatment increased the number of NG2-DsRed+ cells in the kidney, and SIRT3KO further enhanced NG2-DsRed+ cells. Moreover, SIRT3KO promoted pericyte differentiation into fibroblasts as evidenced by co-staining NG2-DsRed/FSP-1. Furthermore, DsRed/FSP-1+ and DsRed/transforming growth factor-β1 (TGF-β1)+ fibroblasts were elevated by SIRT3KO after Ang-II infusion. Ang-II-induced collagen I and TGF-β1 expression was also enhanced in the SIRT3KO mice. SIRT3KO significantly exacerbated Ang-II-induced iron accumulation. This was accompanied by an increase in acetyl-p53, HO-1 and FPN expression. Further, SIRT3KO sensitized Ang-II-induced upregulation of p47phox and gp91phox together with increased ROS formation in the kidney. Conclusion: Our study suggests that SIRT3 deficiency sensitized Ang-II-induced renal fibrosis by the mechanisms involved in promoting differentiation of pericytes into fibroblasts, exacerbating iron overload and accelerating NADPH oxidase-derived ROS formation.


Introduction
Hypertension is one of the major causes of renal diseases [1][2][3]. The mechanisms of kidney disease induced by hypertension are diverse and have not been fully understood. Hypertension leads to renal fibrosis, which might be via transforming growth factor-β1 (TGF-β1) and fibroblast-specific protein 1 (FSP-1) associated pathways [4,5]. In our previous study, we demonstrated that the production of TGF-β1 by pericytes and the pericyte-fibroblast transition contributed to renal fibrosis in the endothelial-specific prolyl hydroxylase domain protein-2 (PHD2) knockout mice [6]. Pericytes are mesenchymal cells residing around vessels. They were identified as the progenitors of fibroblasts that contribute to the deposition of extracellular matrix [7]. Recent studies have proven that pericytes play an important role in the development of hypertensive nephropathy [8]. Iron overload, which causes accelerated oxidative stress [9], plays an important role in hypertension-induced renal injury [10]. Moreover, hypertension has been shown to cause endothelial dysfunction, which also contributes to renal injury [11].
Sirtuin 3 (SIRT3), a mitochondrial nicotinamide adenine dinucleotide (NAD)+-dependent deacetylase, plays a key role in hypertension-induced renal injury [12,13]. Recent studies have shown that SIRT3 activation could alleviate the development of mouse hypertensive renal fibrosis by the suppression of endothelial-to-mesenchymal transition [12] and that the activation of SIRT3 signaling could ameliorate mouse injury induced by hypertension [13]. In our previous study, we found that the ablation of SIRT3 caused mouse coronary microvascular dysfunction and impaired cardiac recovery post myocardial ischemia [14]. Our study also showed that SIRT3 knockout (SIRT3KO) mice increased reactive oxygen species (ROS) formation and reduced capillary density in the obesity heart [15]. Our study further demonstrated that specific knockout of SIRT3 in EC resulted in mouse diastolic dysfunction and significantly increased perivascular fibrosis in the mouse coronary artery [16]. Furthermore, our study suggested that angiogenic capacities and colony formation were significantly impaired in SIRT3KO-endothelial progenitor cells (EPCs) compared to wild-type EPCs and that the loss of SIRT3 further enhanced ROS formation and apoptosis in EPCs [17]. Using NG2 pericyte tracing reporter NG2-DsRed mouse crossed with SIRT3KO mice, we demonstrated that Ang-II induced cardiac fibrosis and cardiac remodeling partly by the mechanism involving SIRT3-mediated pericyte-myofibroblast/fibroblast transition [18].
However, whether pericytes are involved in the pathogenesis of hypertensive renal fibrosis in SIRT3-deficient mice has not been studied yet. In this study, using NG2-DsRed mice to trace pericytes, we aimed to investigate the mechanism of hypertension-induced renal fibrosis involving SIRT3-associated pericyte pathway.

Methods
All procedures conformed to the Institute for Laboratory Animal Research Guide for the Care and Use of Laboratory Animals. It was also approved by the Animal Care and Use Committee of University of Mississippi Medical Center (protocol identifier: 1280C). The investigation conformed to the National Institutes of Health (NIH, Bethesda, MD, USA) Guide for the Care and Use of Laboratory Animals (NIH Pub. No. 85-23, Revised 1996).

Histological and Immunofluorescence Analysis
The renal tissues (n = 3 mice/group) were fixed in neutral buffered 10% formalin solution (SF93-20; Fisher Scientific, Pittsburgh, PA, USA) and embedded in frozen optimal cutting temperature compound (4585; Fisher Health Care, Houston, TX, USA). Frozen sections and paraffin sections were prepared (8 µm in thickness). Masson's trichrome staining (paraffin sections) was performed to measure the degree of renal fibrosis (blue). ROS (frozen sections) was measured by dihydroethidium (DHE) staining. Iron content (frozen sections) was measured by Prussian blue staining (Abcam, Cambridge, MA, USA). The area percentage of fibrosis, iron content and ROS fluorescence intensity were quantified by measuring six random microscopic fields using image-analysis software (Image J, NIH, Bethesda, MD, USA).

Western Blot Analysis
Mouse renal cortex tissues (n = 3-4 mice/group) were collected and homogenized in lysis buffer. The homogenates were centrifuged at 16,000× g at 4 • C for 15 min. A BCA Protein Assay Kit (Pierce Co, Rockford, IL, USA) was used to analyze the protein concentrations. Equal amounts (20 µg) of the protein were separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride (PVDF) membrane. The membranes were blocked with 5% nonfat dry milk in Tris-buffered saline and incubated with the following primary antibodies . After washing, the membranes were incubated for 2 h with an anti-rabbit or anti-mouse secondary antibody coupled to horseradish peroxidase (1:5000; Santa Cruz, CA, USA). Densitometric analyses were carried out with image acquisition and analysis software (Bio-Rad).

Endothelial Cell Glycolysis Analysis
Endothelial cells (ECs) were isolated from thoracic and abdominal aorta of WT and SIRT3KO mice and maintained in the EGM-2 medium (Lonza, MD, USA), as previously described [14,16,20]. Endothelial glycolysis was determined using XFe24 extracellular flux analyzer from Seahorse Bioscience, following the manufacturer's instructions. ECs were seeded at 2.5 × 10 4 cells per well in a 24-well Seahorse XFe24 plate (Agilent, CA, USA) one day prior to the assay. The cells were then washed twice with unbuffered XF base medium (Agilent) and incubated at 37 • C without CO 2 for 1 h. Three initial baseline measurements were performed by using the XFe24 Seahorse analyzer followed by the injection of glucose (final concentration: 10 mM) and three additional measurements. Extracellular acidification rate (ECAR) was calculated in the Wave software (Agilent) and normalized to protein content in each well.

Statistical Analysis
All analyses were performed using Statistical Package for Social Sciences version 20.0 (SPSS, Inc., Chicago, IL, USA). Data are expressed as means ± S.E.M. The assumption of normality in comparison groups was determined by normality test. The significance of differences in the means of corresponding values among groups was determined by using the one-way ANOVA. The significance of differences between two values was determined using LSD test. All the statistical tests were two-sided, and p values < 0.05 were considered significant (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001).

Deficiency of SIRT3 Enhanced Ang-II-Induced Renal Fibrosis
Western blot analysis demonstrated that SIRT3 expression was absent in the kidneys of SIRT3KO mice. After infusion of Ang-II for 4 weeks, SIRT3 levels remained unchanged in the kidneys of wild-type (WT) mice compared to WT without Ang-II ( Figure 1). Next, we examined whether Ang-II infusion led to more severe renal fibrosis in the SIRT3KO mice. As shown in Figure 2, Masson's staining showed that knockout of SIRT3 significantly enhanced Ang-II-induced renal fibrosis (Figure 2A,B). Western blot analysis further showed that knockout of SIRT3 promoted Ang-II-induced increase in the expression of collagen I and TGF-β1 in the mouse kidneys ( Figure 2C,D).

Deficiency of SIRT3 Promoted Ang-II-Induced Recruitment and Differentiation of Pericytes
Accumulating evidence suggests an involvement of pericyte-myofibroblast transition in renal fibrosis. Using NG2-DsRed reporter mice, we investigated the potential contribution of NG2-DsRed pericytes in Ang-II-induced renal fibrosis. As shown in Figure 3, Ang-II infusion resulted in a significant increase in the number of NG2-DsRed+ cells (red) in the WT mouse kidneys. Interestingly, Ang-II-induced NG2-DsRed+ cells were further enhanced in the kidney of SIRT3KO mouse compared to WT mice with Ang-II infusion ( Figure 3A,B).

Deficiency of SIRT3 Exacerbated Ang-II-Induced Iron Overload
Our recent study demonstrated that knockout of SIRT3 promoted p53 acetylation, which has been shown to contribute to iron overload in mouse hearts (Su H et al. unpublished study). As shown in Figure 4, Prussian blue staining showed that Ang-II infusion increased the iron content (blue) and that was further increased in the SIRT3KO mice ( Figure 4A,B). Moreover, the levels of acetyl-p53 were significantly higher in the SIRT3KO mice infused with Ang-II compared to the WT mice infused with Ang-II ( Figure 4C). Western blot analysis showed that the iron-overload-associated proteins (HO-1 and ferroportin) were upregulated in the WT mice infused with Ang-II. The levels of HO-1 and ferroportin were further elevated in the SIRT3KO mice infused with Ang-II ( Figure 4D,E). We then explored the potential mechanism by which knockout of SIRT3 led to an increase in pericytes in mouse kidneys. Angiopoietins are the key regulators of pericyte recruitment during angiogenesis. Our Western blot analysis revealed that the expression of angiopoietin-1 and angiopoietin-2 was significantly upregulated in the mouse kidneys after Ang-II infusion and this upregulation was further elevated by knockout of SIRT3 ( Figure 3C,D). Immunostaining of TGF-β1 (green) revealed that Ang-II infusion resulted in a significant elevation of TGF-β1 production in the glomerulus, whereas knockout of SIRT3 significantly increased Ang-II-induced TGF-β1 production in the glomerulus. Co-staining of NG2/TGF-β1 and DsRed/TGF-β1 further showed that Ang-II infusion increased TGF-β1 production in pericytes (yellow) compared to saline infusion and knockout of SIRT3 further enhanced TGF-β1 production in pericytes ( Figure 3E-G). Similarly, immunostaining of FSP-1 (green) showed that mice infused with Ang-II had a significant increase of fibroblasts in the glomerulus, and knockout of SIRT3 enhanced Ang-II-induced accumulation of fibroblasts in the glomerulus. Co-staining of NG2/FSP-1 and DsRed/FSP-1 further showed that SIRT3KO promoted Ang-II-induced differentiation of pericytes into fibroblasts (yellow) in the glomerulus ( Figure 3H-J).

Deficiency of SIRT3 Exacerbated Ang-II-Induced Iron Overload
Our recent study demonstrated that knockout of SIRT3 promoted p53 acetylation, which has been shown to contribute to iron overload in mouse hearts (Su H et al. unpublished study). As shown in Figure 4, Prussian blue staining showed that Ang-II infusion increased the iron content (blue) and that was further increased in the SIRT3KO mice ( Figure 4A,B). Moreover, the levels of acetyl-p53 were significantly higher in the SIRT3KO mice infused with Ang-II compared to the WT mice infused with Ang-II ( Figure 4C). Western blot analysis showed that the iron-overload-associated proteins (HO-1 and ferroportin) were upregulated in the WT mice infused with Ang-II. The levels of HO-1 and ferroportin were further elevated in the SIRT3KO mice infused with Ang-II ( Figure 4D,E).

Deficiency of SIRT3 Accelerated Ang-II-Induced Endothelial Dysfunction
Our Western blot analysis also showed that deficiency of SIRT3 resulted in a significant suppression of the endothelial glucose transporter-1 (GLUT1) and glycolytic enzyme PFKFB3 when infused with Ang-II ( Figure 5A,B). This was accompanied by a significant reduction of glycolysis in ECs of SIRT3KO mice ( Figure 5C). Ang-II had little effect on the levels of GLUT1 and PFKFB3 in the WT mice ( Figure 5A,B). Interestingly, Ang-II infusion resulted in a significant increase in the expression of NADP oxidase subunits p47phox and gp91phox in the mouse kidneys. Ang-II further increased p47phox and gp91phox expression in SIRT3KO kidneys ( Figure 5D,E). Moreover, DHE staining showed that Ang-II infusion led to a significant increase in ROS formation in the WT mouse kidneys, while knockout of SIRT3 further exacerbated ROS formation ( Figure 5F). Cells 2020, 9, x FOR PEER REVIEW 10 of 14

Deficiency of SIRT3 Accelerated Ang-II-Induced Endothelial Dysfunction
Our Western blot analysis also showed that deficiency of SIRT3 resulted in a significant suppression of the endothelial glucose transporter-1 (GLUT1) and glycolytic enzyme PFKFB3 when infused with Ang-II ( Figure 5A,B). This was accompanied by a significant reduction of glycolysis in ECs of SIRT3KO mice ( Figure 5C). Ang-II had little effect on the levels of GLUT1 and PFKFB3 in the WT mice ( Figure 5A,B). Interestingly, Ang-II infusion resulted in a significant increase in the expression of NADP oxidase subunits p47phox and gp91phox in the mouse kidneys. Ang-II further increased p47phox and gp91phox expression in SIRT3KO kidneys ( Figure 5D,E). Moreover, DHE staining showed that Ang-II infusion led to a significant increase in ROS formation in the WT mouse kidneys, while knockout of SIRT3 further exacerbated ROS formation ( Figure 5F).

Discussion
In the present study, we revealed that deficiency of SIRT3 deteriorated Ang-II-induced renal fibrosis. SIRT3 knockout mice exhibited more severe renal fibrosis after Ang-II infusion. Using NG2-DsRed mice and NG2-DsRed-SIRT3 knockout (SIRT3KO) mice to trace pericytes, our data further showed that these abnormalities might result from the increased numbers of pericytes, the production of TGF-β1 in pericytes and the differentiation of pericytes into fibroblasts. Interestingly, Ang-II infusion caused an iron overload in the mouse kidneys, and knockout of SIRT3 further promoted iron overload, which might also contribute to the exacerbation of renal fibrosis. Additionally, knockout of SIRT3 caused an impairment of glycolytic gene expression and accelerated NADPH oxidase-derived ROS formation after Ang-II infusion.
Renal fibrosis is an important pathological manifestation of hypertensive nephropathy [21,22]. In the present study, knockout of SIRT3 in mice exhibited increased renal fibrosis together with an upregulation of collagen I expression. These results provide strong evidence of the involvement of SIRT3 in the development of renal fibrosis after Ang-II infusion. Pericytes are believed to function as mesenchymal cells, which reside around capillaries, precapillary arterioles, postcapillary venules and collecting venules [23,24]. Accumulating evidence suggests that the pericyte-myofibroblast transition plays a critical role in the development of tissue fibrosis and is a novel mechanism contributing to renal fibrosis in the CKD [18]. In our study, Ang-II infusion resulted in an increase in the numbers of NG2-DsRed+ pericyte. Moreover, Ang-II induced an increase in the numbers of NG2-DsRed+ pericyte in the SIRT3KO mouse kidneys. The immunostaining of NG2 and DsRed showed increased pericyte recruitment similar to that of the NG2 (pericyte) tracing reporter NG2-DsRed mice. Angiopoietins, which are secreted by pericytes, are the key regulators of recruitment of pericytes during angiogenesis [16]. In the present study, the expression of angiopoietins was increased in the WT mice infused with Ang-II, suggesting that Ang-II-induced pericyte recruitment may be mediated by upregulation of angiopoietins in the kidneys. As we expected, levels of angiopoietins were further increased in the SIRT3KO mice infused with Ang-II. These results indicated that deficiency of SIRT3 sensitized the Ang-II-induced increase of pericyte recruitments. TGF-β1 is another key mediator in the regulation of pericyte recruitment and renal fibrosis in chronic kidney diseases [25]. FSP-1, which

Discussion
In the present study, we revealed that deficiency of SIRT3 deteriorated Ang-II-induced renal fibrosis. SIRT3 knockout mice exhibited more severe renal fibrosis after Ang-II infusion. Using NG2-DsRed mice and NG2-DsRed-SIRT3 knockout (SIRT3KO) mice to trace pericytes, our data further showed that these abnormalities might result from the increased numbers of pericytes, the production of TGF-β1 in pericytes and the differentiation of pericytes into fibroblasts. Interestingly, Ang-II infusion caused an iron overload in the mouse kidneys, and knockout of SIRT3 further promoted iron overload, which might also contribute to the exacerbation of renal fibrosis. Additionally, knockout of SIRT3 caused an impairment of glycolytic gene expression and accelerated NADPH oxidase-derived ROS formation after Ang-II infusion.
Renal fibrosis is an important pathological manifestation of hypertensive nephropathy [21,22]. In the present study, knockout of SIRT3 in mice exhibited increased renal fibrosis together with an upregulation of collagen I expression. These results provide strong evidence of the involvement of SIRT3 in the development of renal fibrosis after Ang-II infusion. Pericytes are believed to function as mesenchymal cells, which reside around capillaries, precapillary arterioles, postcapillary venules and collecting venules [23,24]. Accumulating evidence suggests that the pericyte-myofibroblast transition plays a critical role in the development of tissue fibrosis and is a novel mechanism contributing to renal fibrosis in the CKD [18]. In our study, Ang-II infusion resulted in an increase in the numbers of NG2-DsRed+ pericyte. Moreover, Ang-II induced an increase in the numbers of NG2-DsRed+ pericyte in the SIRT3KO mouse kidneys. The immunostaining of NG2 and DsRed showed increased pericyte recruitment similar to that of the NG2 (pericyte) tracing reporter NG2-DsRed mice. Angiopoietins, which are secreted by pericytes, are the key regulators of recruitment of pericytes during angiogenesis [16]. In the present study, the expression of angiopoietins was increased in the WT mice infused with Ang-II, suggesting that Ang-II-induced pericyte recruitment may be mediated by upregulation of angiopoietins in the kidneys. As we expected, levels of angiopoietins were further increased in the SIRT3KO mice infused with Ang-II. These results indicated that deficiency of SIRT3 sensitized the Ang-II-induced increase of pericyte recruitments. TGF-β1 is another key mediator in the regulation of pericyte recruitment and renal fibrosis in chronic kidney diseases [25]. FSP-1, which synthesizes the extracellular matrix and collagen, is a specific marker for identification of fibroblasts. In the present study, the levels of TGF-β1 and FSP-1 were significantly increased in the glomerulus of WT mice after Ang-II infusion and were further enhanced in the SIRT3KO mice. Co-staining of NG2/TGF-β1 and DsRed/TGF-β1 indicated that SIRT3KO promoted Ang-II-induced TGF-β1 production in glomerular pericytes. By co-staining of NG2/FSP-1 and DsRed/FSP-1, our study further revealed Ang-II-induced pericyte differentiation into fibroblasts in the glomerulus and showed that SIRT3KO might result in more fibrosis in the kidneys via pericyte-fibroblast transition.
SIRT3 is a NAD+-dependent deacetylase in the mitochondria. In the mitochondria, heme is the main source of synthesized iron [26][27][28]. Excessive degradation of heme by HO-1 leads to iron overload, which causes oxidative stress. It has been reported that p53 acetylation has a critical role in iron overload [29]. SIRT3 deficiency has been shown to cause striking acetylation of p53 [30]. Ferroportin is an iron exporter and plays an essential role in the export of iron from cells to blood. Elevation of iron content results in an upregulation of ferroportin expression in the heart [31]. Prussian blue staining revealed that Ang-II infusion led to a significant increase in the iron content, which was further accentuated in the kidneys of SIRT3KO mice. This were accompanied by an increase in the expression of iron-overload-associated proteins, namely acetyl-p53, HO-1 and ferroportin. These results suggested that SIRT3KO might enhance renal fibrosis via promoting Ang-II-induced iron overload.
Endothelial dysfunction also plays a crucial role in hypertensive renal injury [11]. Recent studies revealed that SIRT3 activation could alleviate the development of mouse hypertensive renal injury by improving endothelial function [32]. In our previous studies, knockout of SIRT3 in endothelial cells exhibited a reduction in the expression of endothelial glycolytic enzyme PFKFB3 and glucose transporter GLUT1 [15,16]. Consistent with these findings, the levels of PFKFB3 and GLUT1 were reduced in the kidney of SIRT3KO mice infused with Ang-II. In contrast, the levels of NADPH oxidase (p47phox and gp91phox) and ROS formation were upregulated in SIRT3KO mice, suggesting that deficiency of SIRT3 reduced glucose metabolism and enhanced ROS formation which may result in endothelial dysfunction in the kidneys.
In summary, deficiency of SIRT3 sensitized Ang-II-induced renal fibrosis by enhancing Ang-II-induced pericyte recruitment and pericyte-fibroblast transition, by promoting Ang-II-induced iron overload and by accelerating NADPH oxidase-derived ROS formation ( Figure 6). synthesizes the extracellular matrix and collagen, is a specific marker for identification of fibroblasts.
In the present study, the levels of TGF-β1 and FSP-1 were significantly increased in the glomerulus of WT mice after Ang-II infusion and were further enhanced in the SIRT3KO mice. Co-staining of NG2/TGF-β1 and DsRed/TGF-β1 indicated that SIRT3KO promoted Ang-II-induced TGF-β1 production in glomerular pericytes. By co-staining of NG2/FSP-1 and DsRed/FSP-1, our study further revealed Ang-II-induced pericyte differentiation into fibroblasts in the glomerulus and showed that SIRT3KO might result in more fibrosis in the kidneys via pericyte-fibroblast transition. SIRT3 is a NAD+-dependent deacetylase in the mitochondria. In the mitochondria, heme is the main source of synthesized iron [26][27][28]. Excessive degradation of heme by HO-1 leads to iron overload, which causes oxidative stress. It has been reported that p53 acetylation has a critical role in iron overload [29]. SIRT3 deficiency has been shown to cause striking acetylation of p53 [30]. Ferroportin is an iron exporter and plays an essential role in the export of iron from cells to blood. Elevation of iron content results in an upregulation of ferroportin expression in the heart [31]. Prussian blue staining revealed that Ang-II infusion led to a significant increase in the iron content, which was further accentuated in the kidneys of SIRT3KO mice. This were accompanied by an increase in the expression of iron-overload-associated proteins, namely acetyl-p53, HO-1 and ferroportin. These results suggested that SIRT3KO might enhance renal fibrosis via promoting Ang-II-induced iron overload.
Endothelial dysfunction also plays a crucial role in hypertensive renal injury [11]. Recent studies revealed that SIRT3 activation could alleviate the development of mouse hypertensive renal injury by improving endothelial function [32]. In our previous studies, knockout of SIRT3 in endothelial cells exhibited a reduction in the expression of endothelial glycolytic enzyme PFKFB3 and glucose transporter GLUT1 [15,16]. Consistent with these findings, the levels of PFKFB3 and GLUT1 were reduced in the kidney of SIRT3KO mice infused with Ang-II. In contrast, the levels of NADPH oxidase (p47phox and gp91phox) and ROS formation were upregulated in SIRT3KO mice, suggesting that deficiency of SIRT3 reduced glucose metabolism and enhanced ROS formation which may result in endothelial dysfunction in the kidneys.
In summary, deficiency of SIRT3 sensitized Ang-II-induced renal fibrosis by enhancing Ang-IIinduced pericyte recruitment and pericyte-fibroblast transition, by promoting Ang-II-induced iron overload and by accelerating NADPH oxidase-derived ROS formation ( Figure 6).