SIRT1 and Kidney Function

In 1986, the silent information regulator 2 (Sir2) gene was isolated from budding yeast [1]. Thirteen years later, the Sir2 gene was identified as being associated with the life span of yeast. Sinclair and Guarente [2] found that the accumulation of extrachromosomal ribosomal DNA (rDNA) circles is a major cause of ageing in yeast, which is generated by homologous recombination between rDNA repeats. Sir2 overexpression silences transcription at silent mating loci, telomeres and the rDNA and suppresses extrachromosomal rDNA circle formation to extend the life span of yeast, while deletion of Sir2 shortens the life span. This discovery was validated by a recent quantitative genetic analysis of whole genomes of different strains of yeast, which showed that, as the first quantitative trait locus, the Sir2 gene is responsible for the majority of the effects on yeast's replicative life span [3]. During this period, the antiageing effect of Sir2 and its homologs was also observed in Drosophila, Caenorhabditis elegans and mice when their expression levels are increased [4,5,6,7].

In mammals, the sirtuin family is the homolog of the Sir2 gene in yeast, consisting of at least 7 isoforms, i.e. SIRT1-SIRT7. As a commonality of the sirtuin family, the 7 isoforms have the same 275-amino-acid-sized catalytic core region and a diverse subcellular localization. SIRT1, SIRT6 and SIRT7 are mainly found in the nucleus, and SIRT2 is in the cytoplasm, while SIRT3, SIRT4 and SIRT5 are localized in the mitochondria [8]. SIRT1 is the most extensively studied family member.

The dietary intervention of caloric restriction (CR) was first experimented with in rats in 1934 [9], and it showed a positive impact by increasing their life span without malnutrition. Afterwards, the effect of delaying ageing and extending the life span were demonstrated in diverse species including primates [10,11]. In addition to this, CR benefits patients with a wide range of age-related diseases such as diabetes, cancer, cardiovascular disease, neurodegenerative diseases and age-associated renal injury [10,12,13,14]. Interestingly, SIRT1 overexpression and CR share phenotypes of antiageing and antiageing-related diseases [15,16]. The study of yeast helps to link the two together. When the glucose concentration is reduced from 2 to 0.5%, the average life span of yeast (the number of generations) increases from 21.1 to 26 [17]. A similar life extension phenomenon cannot be repeated in the Sir2 mutant strain, suggesting that CR-induced longevity is dependent on activation of the Sir2 protein. As for mammals, in several studies [18,19] SIRT1 was proven to be the key regulator of physiological adaption when subjected to different diets including fasting, CR and a high-fat diet. CR, the mechanism of which remains complicated and not yet fully understood, is proposed to benefit rodents at least in part by activating SIRT1; in vivo effects of SIRT1-activating compounds further support this theory. Resveratrol (trans-3,5,4′-trihydroxystilbene) and its analogs are SIRT1 activators and show life-extending, anti-inflammatory, antioxidant, antidiabetic and cancer-preventive effects [20].

The first clue about how SIRT1 performs its function also comes from a study on yeast. Overexpression of yeast Sir2 proteins, together with the presence of nicotinamide adenine dinucleotide (NAD+), promotes global deacetylation of histones, especially lysines 9 and 14 of histone H3 and lysine 16 of histone H4, which are acetylated in active chromatin and hypoacetylated in silenced chromatin [21]. This enzyme activity of histone deacetylases categorizes sirtuins into the class III family of histone deacetylase enzymes, for it requires NAD+ as a cofactor instead of zinc in class I and II histone deacetylases. Therefore, sirtuins are able to monitor metabolic homeostasis via variation of the NAD+ level, which is sensitive to the intracellular state of energy and redox change. In order to catalyze the conversion of an acetylated substrate to a deacetylated substrate, NAD+ is firstly cleaved into nicotinamide and O-acetyl-ADP-ribose, the latter of which will covalently attach to the acetyl moiety of the substrate. Then, hydrolysis of the acetyl-lysine bond liberates O-acetyl-ADP-ribose and makes itself biologically active [22]. The deacetylation reaction yields nicotinamide, which in turn serves as an inhibitor to the reaction itself and becomes a salvage pathway in which NAD+ is resynthesized from nicotinamide [8].

In the sirtuin family, SIRT1, SIRT2 and SIRT3 are NAD+-dependent deacetylases. In addition to its deacetylase function, SIRT1 also has mono-ADP-ribosyltransferase activity (transfer of ADP-ribose to other proteins), which is of minor importance for its biologic function, while SIRT4 primarily functions as a mono-ADP-ribosyltransferase, SIRT5 possesses desuccinylase and demalonylase activities [23], and SIRT6 can remove long-chain fatty acids from selected substrate proteins [24].

Another important characteristic of sirtuins is that they target a variety of transcriptional factors for deacetylation besides histone. Lysine acetylation activity is a dynamic, reversible and evolutionarily conserved protein posttranslational modification. A recent proteomic study [25] highlights the significant role of SIRT1 in regulating diverse substrates and cellular pathways. A total number of 4,623 lysine acetylation sites in 1,800 proteins were identified and quantified via HPLC-MS/MS analysis using mouse embryonic fibroblasts isolated from wild-type and SIRT1 knockout mice, of which 485 lysine acetylation sites are enhanced by more than 100% after SIRT1 knockout. Interestingly, not all of these sites are necessarily direct SIRT1 targets, because a number of acetyltransferases and major acetyltransferase complexes such as Kat5 (Tip60), Kat8 (Myst1) and p300 are targeted and regulated by SIRT1-mediated deacetylation. With regard to the direct substrates deacetylated by SIRT1, acetylated modification of these transcription factors and transcriptional coregulatory proteins by SIRT1 may have an influence on them in different ways, including altering their stability, activity, subcellular localization, DNA-binding ability and protein-protein interactions [26,27]. Notably, the results of this modification by SIRT1 may vary, depending on the kind of stimuli, the stimulated protein and its acetylation site. Take p53 as an example. In response to DNA damage, SIRT1 binds and deacetylates the p53 protein with specificity for its C-terminal Lys382 residue [28]. Deacetylated p53 results in nuclear translocation and reduced DNA-binding ability, which lead to either growth arrest or apoptosis [29].

Since SIRT1 is such an essential metabolic sensor, its activity is regulated dynamically in order to allow for adaptions to any alteration to the cellular metabolic state. Therefore, nutritional, hormonal and environmental signals, as well as the NAD+ level and the SIRT1-interactacting proteins responding to those signals, compose the regulation network of SIRT1.

With a high-fat and high-glucose diet, SIRT1 expression is decreased, while during starvation and nutrient deprivation, SIRT1 expression is increased [30]. In experiments of acute nutrient withdrawal, the forkhead transcription factor Foxo3a is activated and is in physical interaction with p53 [31]. Being mediated through two p53 binding sites present in the SIRT1 promoter, Foxo3a stimulates SIRT1 transcription. SIRT1 expression cannot be induced in either Foxo3a knockdown cells or starved p53-deficient mice. Thus, SIRT1 is regulated in a nutrient-sensing pathway involving the reciprocal action of relevant proteins.

During stress response, SIRT1 links chromatin dynamics/gene expression to environmental stimuli. SIRT1 induced by acute stress ensures genome integrity [32]. SIRT1 mediates heterochromatin formation by deacetylating histone polypeptides with a preference for histone H4 lysine 16 (H4-K16Ac) and H3 lysine 9 (H3-K9Ac), followed by recruitment and deacetylation of histone H1 at lysine 26. This facultative heterochromatin results in a mark restricting silenced chromatin with hypomethylation of H3-K79 [33]. Also, SIRT1 enhances methyltransferase activity. Following DNA damage, SIRT1 may generate aberrant methylation of the CpG islands in promoters, promoting heritable gene silencing [34]. However, the endogenous mechanisms that regulate the level and activity of SIRT1 are not clear. There are studies suggesting both posttranscriptional and posttranslational modifications of SIRT1. The RNA-binding protein HuR, regulating the stability of many target mRNAs, associated with the 3′ untranslated region of the mRNA encoding SIRT1, stabilizes the SIRT1 mRNA and increases SIRT1 expression levels, thus functioning in a posttranscriptional way [35]. Sumoylation of SIRT1 at Lys 734 increased its deacetylase activity. Conversely, mutation of SIRT1 at Lys 734 or desumoylation by SENP1, a nuclear desumoylase, reduces its deacetylase activity. Stress-inducing agents promote the association of SIRT1 with SENP1, and cells depleted of SENP1 (but not of SENP1 and SIRT1) are more resistant to stress-induced apoptosis than control cells [36]. Another posttranslational modification is phosphorylation. Mass spectrometry shows that the cell cycle-dependent kinase cyclin B/Cdk1 forms a complex with and phosphorylates SIRT1 at threonine 530 and serine 540 [37].

Given the fact that NAD+ serves as the core of the SIRT1 deacetylase enzymatic reaction, NAD biosynthesis critically regulates its activity [38]. NAD+ is generated from tryptophan and vitamin B₃, which is also known as nicotinic acid or nicotinamide, from dietary intake de novo [39]. Interestingly, nicotinamide, a by-product of SIRT1 and another enzyme-mediated reaction which NAD+ participates in as a cofactor, serves as an endogenous inhibitor of SIRT1 and is also involved in the resynthesis of NAD+ in the salvage pathway. This resynthesis is of great significance, because it ensures a SIRT1-mediated reaction via lowering the concentration of nicotinamide and raising NAD+ levels. First, nicotinamide is converted into nicotinamide mononucleotide (NMN) by nicotinamide phosphoribosyltransferase (Nampt). Subsequently, NMN adenylyltransferase (Nmnat) regenerates NAD+ from NMN. Nampt is the rate-limiting enzyme of the NAD+ salvage re-cycle and a link to the circadian clock cycle. CLOCK:BMAL1 (circadian locomotor output cycles kaput-brain and muscle aryl hydrocarbon receptor nuclear translocator-like 1) is physically and specifically associated with the E-boxes on the Nampt promoter, and active Nampt induces SIRT1 by regenerating NAD+ [38]. Moreover, CLOCK has an intrinsic acetyltransferase activity, which enables circadian chromatin remodeling by acetylating histones and nonhistone proteins, and SIRT1 acts as the histone deacetylase that counterbalances the histone acetyltransferase function of CLOCK. Thus, a loop of the cellular metabolism and the circadian clock emerges, and any attacks on the loop may have an impact on SIRT1. As a result, lack of NAD+ decreases SIRT1 activity, and possible reasons for NAD+ shortage include serious nutrition insufficiency, excessive consumption by activated poly-ADP-ribose polymerases in stress response and disturbed recycling from nicotinamide in the salvage pathway.

SIRT1 has been found to be an essential factor in the regulation of systemic metabolic homeostasis [40] as well as a metabolic sensor in the modulation of gene expression in response to changes in cellular energy states in multiple organs including the kidney [41]. Increased SIRT1 activity by either pharmacologic or genetic means can prolong the life of obese animals and significantly improves their metabolic parameters such as glucose tolerance, fasting blood glucose levels and insulin resistance [42,43]. Since the kidney is very likely to be attacked in metabolic disorders such as diabetes mellitus, and diabetes is associated with reduced SIRT1 expression in the kidney, it is likely that the kidney may benefit from SIRT1 activation. This is supported by recent studies. Hasegawa et al. [44] used mouse models of diabetes mellitus to show that decreased SIRT1 in proximal tubules represents the onset of diabetic nephropathy. In their experiment, albuminuria could be prevented by overexpressing SIRT1 in the proximal tubules of a transgenic mouse model, while in the proximal tubule-specific SIRT1 conditional knockout mouse model, streptozotocin-induced albuminuria showed no signs of improvement. This is due to the cross talk between SIRT1 and claudin-1. In the diabetic milieu, reduced expression of SIRT1 causes downregulation of intracellular-type Nampt, presumably by direct transcriptional mechanisms, which leads to a reduction of the levels of NMN secreted from tubular cells. As a result, the basal level of NMN around podocytes will no longer be maintained, so that SIRT1 expression is decreased and fails to modulate claudin-1 epigenetically. Activated claudin-1 in podocytes impairs the glomerular barrier function through downregulating synaptopodin or podocin expression and induces albuminuria [44]. Podocyte SIRT1 is also involved in the pathophysiology of diabetic nephropathy. Podocyte-specific SIRT1 knockout mice show increased urinary albumin excretion in diabetes [Guan and Hao, unpubl. data]. It is suggested that in the diabetic kidney, SIRT1 in the podocytes functions as a deacetylase and inactivates the p65 subunit of NF-κB and STAT3 in order to protect the podocytes from injury [45]. Also, our laboratory reported that resveratrol, a SIRT1 activator, attenuated diabetic kidney injury via modulating abnormal angiogenesis in a type 1 diabetic rat model [46].

Hypertension is frequently clustered with obesity, hyperlipidemia and hyperinsulinemia, collectively termed the ‘metabolic syndrome', because the regulation of systemic blood pressure is strongly associated with energy metabolism. It has been documented that reduced caloric dietary intake decreases arterial blood pressure in mildly hypertensive patients or reduces the number of antihypertensive drugs required to control hypertension. Recent studies have suggested that SIRT1 is involved in blood pressure through the regulation of renal sodium reabsorption in the collecting duct and the regulation of vascular tone.

Aldosterone increases renal tubular Na⁺ absorption by enhancing the activity of the epithelial Na⁺ channel α-subunit (α-ENaC) expressed in the apical membrane of principal cells of the collecting duct. Zhang et al. [47] explored that a complex containing the histone H3K79 methyltransferase disruptor of telomeric silencing-1 (Dot1) is associated with and represses the α-ENaC promoter in murine inner medullary collecting duct cells, and SIRT1 overexpression can inhibit basal α-ENaC mRNA expression and α-ENaC promoter activity by functionally and physically interacting with Dot1 to enhance the distributive activity of Dot1 on H3K79 methylation, surprisingly in a deacetylase-independent manner, and independent of mineralocorticoid receptor signaling as well. Since the effects of SIRT1 on α-ENaC transcription are independent of its enzymatic activity, changes in nicotinamide levels or the use of SIRT1 activators are unlikely to modulate sodium reabsorption in the collecting duct [47].

Several studies have demonstrated that SIRT1 promotes endothelium-dependent vasodilation by targeting endothelial nitric oxide synthase (eNOS) for deacetylation [48,49]. SIRT1 and eNOS colocalize and coprecipitate in endothelial cells, and SIRT1 deacetylates eNOS, stimulating eNOS activity and increasing endothelial nitric oxide (NO). The SIRT1-induced increase in endothelial NO is mediated through lysines 496 and 506 in the calmodulin-binding domain of eNOS. Inhibition of SIRT1 in the endothelium of arteries inhibits endothelium-dependent vasodilation and decreases bioavailable NO. Moreover, vascular smooth muscle cell-specific SIRT1 transgenic mice prevented the increase in systolic blood pressure caused by angiotensin II infusion without vascular remodeling in murine thoracic and renal aortas [50]. SIRT1 overexpression significantly inhibited reactive oxygen species (ROS) generation, vascular inflammation and collagen synthesis in arterial walls and decreases the angiotensin II-increased binding of NF-κB on its specific binding sites on the TGF-β₁ promoter. Ichiki et al. [51] and Miyazaki et al. [52] likewise showed that administration of resveratrol, a SIRT1 activator, not only suppresses AT1R expression in the aorta but also significantly blunts angiotensin II-induced hypertension in mice.

Renal tubules are easy to be affected by acute hemodynamic changes or toxins, which may lead to acute kidney injury (AKI). In a cisplatin-induced AKI model, the effect of SIRT1 was investigated by using transgenic mice overexpressing SIRT1 specifically in proximal tubules. SIRT1-overexpressing mice can preserve both the number of peroxisomes and their catalase activity, which is severely decreased by cisplatin; thus, SIRT1 exerts protective action in tubular cell damage and kidney injury [53]. Elevated ROS levels within mitochondria are deleterious to both the mitochondria and mitochondrion-rich tubular cells, and damaged mitochondria in turn produce more ROS, which leads to cell apoptosis and necrosis in AKI. This excessive oxidative stress and mitochondrial dysfunction can be altered by SIRT1 activation. SRT1720, a specific SIRT1 activator, can deacetylate and activate peroxisome proliferator-activated receptor-γ coactivator 1α in order to expedite the recovery of mitochondrial protein expression and function, as well as normal mitochondrial biogenesis [54,55]. Also, high SIRT1 expression contributes to cell regeneration following ischemia/reperfusion-induced AKI. SIRT1 heterozygotes are more susceptible to ischemia/reperfusion injury and hold a significantly higher level of the cyclin-dependent kinase (CDK) inhibitor p21. p21, which is downstream of p53, binds to and inhibits the activity of cyclin-CDK2 or cyclin-CDK4 complexes and thus functions as a regulator of cell cycle progression at G₁. Either SIRT1 or SIRT1-induced lower p53 levels might suppress p21 and result in increased cell proliferation [56].

Renal fibrogenesis is the common outcome of various diseases damaging the kidney, and it leads to end-stage renal failure. The antifibrotic function of SIRT1 has been demonstrated using several kinds of renal fibrosis animal model. In the unilateral ureteral obstruction model, SIRT1, abundantly expressed in murine renal medullary interstitial cells, can attenuate oxidative stress-induced COX2 expression to rescue the interstitium from inflammation and fibrogenesis [57].

TGF-β can induce extracellular matrix production and promote fibrogenesis, which signals through the heteromeric complex of TGF-β type I receptor with TGF-β type II receptor to activate and assemble Smad2, Smad3 and Smad4 into heteromeric complexes that translocate into the nucleus [58,59]. Thus, these complexes in the nucleus can interact with various transcription factors and regulate the expression of TGF-β target genes [60]. In the obstructed kidney, Smad3 acetylation, as well as its phosphorylation, is an early response in the induction of interstitial fibrosis. Resveratrol can prevent Smad3 acetylation, but not phosphorylation, resulting in less interstitial fibrosis [61]. The underlying mechanism was elucidated by Huang et al. [62], using a 5/6 nephrectomy model: deacetylation of Smad3 directly induced by SIRT1 activation can attenuate renal damage by inhibiting TGF-β-Smad signaling. SIRT1 heterozygotes suffer from worse kidney function and end up with more serious kidney fibrosis. Simic et al. [63] examined the progression from AKI to chronic kidney disease and reached a similar conclusion by performing the ischemia/reperfusion procedure on kidney-specific knockout mice and kidney-specific transgenic mice. The kidney-specific knockout mice subsequently showed worse kidney function and more kidney fibrosis, whereas the kidney-specific transgenic mice showed improvement. Furthermore, endothelial cell-specific SIRT1 knockout mice have impaired angiogenesis, reduced matrilytic activity and retention of the profibrotic cleavage substrate tissue transglutaminase as well as endoglin-accompanied MMP-14 suppression, and are susceptible to folic acid-induced kidney injury. Restoration of MMP-14 expression in SIRT1-depeleted mice improved the angiogenic and matrilytic functions of the endothelium, prevented renal dysfunction and attenuated nephrosclerosis [64]. So far, SIRT1 seems to be a promising target of intervention for delaying the development of nephrosclerosis.

As mentioned before, SIRT1 induces kidney recovery from external stress such as ischemia/reperfusion injury by inducing renal epithelial cell proliferation and reducing its apoptosis. Recently, Li and colleagues [65] suggested that, compared to mice with single conditional knockout of Pkd1, a classic autosomal dominant polycystic kidney disease animal model, with double conditional knockout of Pkd1 and SIRT1, showed delayed renal cyst formation in postnatal murine kidneys. Further, inhibiting SIRT1 with nicotinamide or the SIRT1-specific inhibitor EX-527 delayed cyst formation in Pkd1-null murine embryonic kidneys and in Pkd1-conditional knockout postnatal kidneys. This finding is in accordance with previous studies demonstrating that SIRT1 can promote the recovery of the kidney from injury, because autosomal dominant polycystic kidney disease is a disease with continuously abnormal cyst formation; in either damaged tubules or cysts, mechanistically, SIRT1 promotes growth of the epithelium by deacetylation and inactivation of Rb and p53.

CR consists of established dietary interventions that have been shown to increase both the median and the maximum life spans of a variety of species, including yeast, fish and mammals [17,66]. SIRT1 activity is suggested to be increased by CR, and SIRT1 is one of the molecules that mediate the beneficial effects of CR. SIRT1-overexpressing transgenic mice show a phenotype that resembles mice under CR [15], and SIRT1-deficient mice failed to show life span prolongation under CR [67]. Kume et al. [68] demonstrated that the kidney also benefits from CR. Ageing leads to significantly decreased Sirt1 expression and enhanced PI3K, which suppress both autophagy and cell cycle arrest, resulting in the accumulation of oxidative stress and subsequent apoptosis through acetylated Foxo3-mediated Bim expression, whereas a 12-month CR diet recues SIRT1 downregulation in the aged kidney. Preserved SIRT1 function promotes cell adaptation to hypoxia through deacetylated Foxo3-enhanced Bnip3-mediated autophagy and p27Kip1 expression. Our research team is also interested in renal ageing. We find that younger mice are more resistant to both ischemia/reperfusion [56] and cisplatin-induced kidney injury [Guan et al., unpubl. data]. Both SIRT1 expression and the NAD+ cycle decline as age advances. Losing one allele of SIRT1 makes SIRT1 heterozygotes similarly vulnerable to injury as aged mice. Specific agonists of SIRT1 as well as pharmaceutically refilled NAD+ can activate SIRT1 to lower the risk of kidney injury following acute stress. The responsible mechanism involves the effect of SIRT1 on cell survival and mitochondrial function.

Few studies show SIRT1 to be involved in human kidney disease despite the fact that SIRT1 plays an important role in the pathogenesis of kidney disease in mammals. This might be due to difficulty acquiring human kidney tissue specimens. Hasegawa et al. [44] showed by immunohistochemistry of human kidney biopsy sections that lack of glomerular SIRT1 epigenetically increased the expression of the tight junction protein claudin-1, resulting in serious proteinuria.

A Japanese case-control study on type 2 diabetes [69] identified 4 single nucleotide polymorphisms (SNPs) within the SIRT1 gene which have been nominally associated with susceptibility to diabetic nephropathy; a haplotype consisting of the 11 SNPs in the SIRT1 locus had a stronger association. Another Japanese study [70] enrolled 219 hemodialysis patients and 803 control subjects and revealed that the SIRT1 polymorphisms rs7069102 and rs2273773 are associated with abnormal cholesterol metabolism and coronary artery calcification, respectively, in Japanese HD patients; whether they have an impact on survival is not yet clear.

Since SIRT1 is an antiageing molecule and has a protective effect on multiple organs such as the heart, neurons and kidneys, the scientific community keeps seeking its activator. SRT1720, a synthetic SIRT1-activating compound, has been proven to attenuate damage induced by ischemia/reperfusion and obstructive kidney disease [56,57]. Resveratrol, a natural antioxidant present in red wine, is a more widely recognized candidate for activating SIRT1. Resveratrol extends the life span of yeast, worms and flies [71], protects rodent kidneys from fibrogenesis and diabetic nephropathy [46,62] and improves metabolic parameters in aged mice and obese humans [42,72]. A molecule as small as 228 Da in size, resveratrol is likely to regulate the activity of other cellular proteins in addition to SIRT1. For example, resveratrol binds to a hydrophobic pocket of F1-ATPase and inhibits this key enzyme in mitochondrial respiration [73], which may indirectly activate AMP-activated protein kinase. Nevertheless, a recent experiment led by Sinclair and colleagues [74] ends the debate on whether resveratrol directly activates SIRT1 and supports the model that some of the biological effects of resveratrol and SIRT1-activating compounds are strictly dependent on SIRT1. Many clinical trials of resveratrol have been carried out or are going on at the moment. Due to the limitations of resveratrol regarding its bioavailability, pharmacokinetics and target specificity, the results are expected to be diverse. So far, at least three human studies have shown protective effects of resveratrol by systolic blood pressure reduction and improved insulin sensitivity [72,75,76].

Once again, the SIRT1 activator is back on stage, and further, exquisitely designed studies are required to elucidate its role in a pharmacological approach to renal protection and healthy ageing.

CMH received research funds from Natural Science Foundation of China (81130075, 31471101), the National Basic Research Program of China (2011CB94001).

The authors declare no conflicts of interest.