A significant body of literature indicates that Sirt1 is both modified and inhibited by S-nitrosation (Table 3; Kornberg et al., 2010; Rodríguez-Ortigosa et al., 2014; Shinozaki et al., 2014; Kalous et al., 2016, 2020; Nakazawa et al., 2017; Hoth et al., 2018; Kim et al., 2018; Sen et al., 2018). Recombinantly purified Sirt1 shows direct transnitrosation in vitro following treatment with GSNO (Kornberg et al., 2010; Kalous et al., 2016, 2020; Goetz et al., 2020) or nitrosated GAPDH (Kornberg et al., 2010). Treatment with the transnitrosation donor S-nitroso-N-acetylpenicillamine (SNAP) (Shinozaki et al., 2014; Kalous et al., 2020) or NO-donating compounds (NONOates) (Kalous et al., 2016, 2020) also yields S-nitrosated Sirt1. A subsequent loss of Sirt1 deacetylase activity accompanies S-nitrosation (Kornberg et al., 2010; Shinozaki et al., 2014; Kalous et al., 2016, 2020). The S-nitrosation sites of Sirt1 were localized to Cys395 or Cys398 of the Zn²⁺-tetrathiolate via serine mutagenesis (Figures 1A,B; Shinozaki et al., 2014; Kalous et al., 2016). The structural integrity of the Zn²⁺-tetrathiolate is essential for Sirt1 deacetylase activity; treatment of recombinant Sirt1 with nitrosating agents or mutation of any of the four tetrathiolate cysteines to serine resulted in Zn²⁺ loss (Shinozaki et al., 2014; Kalous et al., 2016). A concomitant loss of secondary protein structure was seen via circular dichroism (Kalous et al., 2016). S-nitrosation of Sirt1 is reversible with thiol-based reductants (Shinozaki et al., 2014; Kalous et al., 2016); concurrent supplementation with free Zn²⁺ restores Sirt1 deacetylase activity in vitro (Kalous et al., 2016), implying restoration of the Zn²⁺-tetrathiolate motif secondary structure and that Sirt1 S-nitrosation may serve as a reversible signaling mechanism.

Sirt1 S-nitrosation and subsequent inhibition has been shown in multiple cultured cell lines and animal systems, and points to a close relationship with the activity of various nitric oxide synthase (NOS) isoforms. S-nitrosation of transfected Sirt1 was demonstrated in COS-7 cells in response to 300 μM SNAP or 600 μM GSNO treatment (Shinozaki et al., 2014). Disrupting the Zn²⁺-tetrathiolate by treating immunopurified Sirt1 from COS-7 cell lysates with the Zn²⁺ chelator TPEN resulted in inhibition of Sirt1 deacetylase activity (Shinozaki et al., 2014). S-nitrosation as an output of inflammation was further demonstrated in several cellular models, including C2C12 myotubes and Hepa1c1c7 hepatocytes treated with cytokines (Shinozaki et al., 2014). Moreover, S-nitrosation of Sirt1 was correlated with inhibition of Sirt1 deacetylase activity; increased Sirt1 S-nitrosation correlated with increased p53 and p65 acetylation, thereby activating apoptosis and promoting expression of proinflammatory genes (Shinozaki et al., 2014). Downstream effects of Sirt1 S-nitrosation were also investigated in cultured RAW 264.7 macrophages (Kim et al., 2018). Treatment of RAW 264.7 macrophages with 1 mM SNAP increased Sirt1 S-nitrosation, decreased Sirt1 deacetylase activity, and allowed HMGB1 release (Kim et al., 2018).

The primary source of NO in biological systems are the three NOS isoforms (Marletta, 1994). The two constitutively expressed NOS isoforms are the neuronal (nNOS or NOS1) and endothelial (eNOS or NOS3) isoforms (Knowles and Moncada, 1994). Both eNOS and nNOS activity respond to Ca²⁺ fluxes and are dependent on Ca²⁺/calmodulin-binding for activation (Stevens-Truss et al., 1997; Gribovskaja et al., 2005; Campbell et al., 2014). In HEK293 cells stably expressing nNOS, activation of nNOS with the Ca²⁺ ionophore A23187 induced Sirt1 S-nitrosation and increased PGC1α acetylation as a proxy for Sirt1 activity (Kornberg et al., 2010). The inducible NOS isoform (iNOS or NOS2) is induced in response to inflammatory inputs such as cytokines and lipopolysaccharide (LPS); iNOS activity increases in many inflammatory and age-related diseases (Anavi and Tirosh, 2019). In C2C12 myotubes and Hepa1c1c7 hepatic cells, pretreatment with cytokines induced iNOS expression, yielded S-nitrosated Sirt1, and increased p53 acetylation as a proxy for Sirt1 activity (Shinozaki et al., 2014).

Rodent models point to a whole organismal role of Sirt1 S-nitrosation. Primary hepatocytes from Wistar rats showed Sirt1 S-nitrosation when cultured in media supplemented with cholic acid to alter bile salt homeostasis (Rodríguez-Ortigosa et al., 2014). Furthermore, Sirt1 S-nitrosation was tied to iNOS and nNOS activity. In C57BL/6 mice injected with LPS, Sirt1 S-nitrosation was observed in the liver, along with increased acetylation and DNA binding of p53 and p65. These effects were not seen in iNOS knockout (iNOS^–/–) animals (Shinozaki et al., 2014). A subsequent study examined Sirt1 S-nitrosation in C57BL/6 mice but with burn injury as the original inflammatory stimulus (Nakazawa et al., 2017). Sirt1 S-nitrosation seen following burn injury correlated with iNOS expression and increased p53 and p65 acetylation, implying Sirt1 inhibition (Nakazawa et al., 2017). Sirt1 S-nitrosation and inhibition was observed and tied to iNOS activity in a “two-hit” model of lung injury, wherein a pulmonary injury was induced in mice by artificial contusion, followed by either infection with the Streptococcus pneumoniae TIGR4 strain or exposure to LPS (Hoth et al., 2018). Inhibiting iNOS with the small molecule iNOS inhibitor 1400W (Garvey et al., 1997) prior to lung injury prevented Sirt1 inhibition as measured by a commercial fluorescent assay; however, the impact on protein targets of Sirt1 deacetylase activity was not assessed (Hoth et al., 2018). Relevance of this signaling cascade was also investigated systemically in the lungs of BALB/c mice treated with LPS; both increased Sirt1 S-nitrosation and extracellular HMGB1 were observed but inhibited by pretreatment with the iNOS inhibitor 1400W (Table 4; Kim et al., 2018). Moreover, in an aged rat model, in vivo Sirt1 S-nitrosation increased with age in an iNOS-dependent manner when comparing skeletal muscle of 2- and 28-month-old F344 rats (Shinozaki et al., 2014). Intraperitoneal injection of aged rats with the iNOS inhibitor 1400W reduced the level of Sirt1 S-nitrosation (Shinozaki et al., 2014).

Sirt1 S-nitrosation was also demonstrated in neurodegenerative models. In Parkinson’s disease, excess NO may be constitutively present due to increased nNOS activity (Hantraye et al., 1996; Przedborski et al., 1996). In an induced Parkinson’s disease mouse model, increased Sirt1 S-nitrosation paralleled increased p53 and p65 acetylation, and Sirt1 S-nitrosation was reduced when nNOS was inhibited with 7-nitroindazole (Table 4; Shinozaki et al., 2014). In a mouse model of Alzheimer’s disease, nNOS was also tied to Sirt1 S-nitrosation. Isolated primary cortical neurons of C57BL/6 mice were exposed to aggregated amyloid-β 1–42 peptide (Aβ_1–42) (Sen et al., 2018). Aβ_1–42 is thought to be the pathogenic form that accumulates and induces toxicity within neurons in Alzheimer’s disease. Neuronal exposure to Aβ_1–42 induced S-nitrosation of GAPDH and Sirt1, decreased the Sirt1-tau interaction, and increased tau acetylation in an nNOS-dependent manner (Table 4; Sen et al., 2018).

Together, these animal models suggest Sirt1 S-nitrosation occurs in vivo under conditions of inflammation, aging, and neurodegeneration. Pharmacological prevention or reversal of Sirt1 S-nitrosation resulting in a net increase in Sirt1 deacetylase activity may be beneficial in these disease contexts.

Recombinantly purified Sirt2, like its nuclear/cytosolic counterpart Sirt1, can be inhibited by S-nitrosation in vitro (Kalous et al., 2020; Table 3). A concentration-dependent increase in Sirt2 S-nitrosation was observed by treatment with GSNO and NO (up to 100 μM), correlating with inhibition of Sirt2 demyristoylase activity. As demonstrated by the ratio of the maximum rate of inactivation to the apparent covalent inhibition constant (k_inact/K_I), Sirt2 was the most efficiently inactivated by GSNO compared to Sirt1 and Sirt6, suggesting Sirt2 is more likely to be inactivated by GSNO compared to Sirt1 or Sirt6 in a cellular context (Kalous et al., 2020). Mutagenesis of the four Zn²⁺-tetrathiolate cysteines to alanine abolished S-nitrosation, indicating the Zn²⁺-tetrathiolate is the likely site of S-nitrosation (Kalous et al., 2020). Future studies are needed to determine if Sirt2 is modified and inhibited by S-nitrosation in cellular or animal models.

In vitro exposure of Sirt3 to NO (100 μM released from a NONOate), but not 100 μM GSNO, resulted in cysteine S-nitrosation (Table 3; Kalous et al., 2020). However, Sirt3 deacetylase activity was not significantly inhibited by NO. Similarly, the in vitro sensitivity of Sirt5 to NO-derived oxidants demonstrated Sirt5 S-nitrosation, but Sirt5 desuccinylase activity was not inhibited (Kalous et al., 2020). As discussed in section “Sirt1 Can Be Modified and Inhibited by Glutathionylation,” Sirt1 S-nitrosation occurs selectively at the Zn²⁺-tetrathiolate leading to inhibition of Sirt1 deacetylase activity (Shinozaki et al., 2014; Kalous et al., 2016). Likewise, mutation of the Zn²⁺-tetrathiolate cysteine residues to alanine abolished S-nitrosation signal in Sirt3 and Sirt5 (Kalous et al., 2020), but failed to inhibit deacylase activity. This decoupling of Sirt3 and Sirt5 Zn²⁺-tetrathiolate S-nitrosation and inhibition suggests that instead of directly inhibiting deacetylase activity, S-nitrosation may serve as a degradation signal similar to effects of Sirt3 glutathionylation (discussed in section “Sirt3 Can Be Modified and Inhibited by Glutathionylation”). Alternatively, the Sirt3 or Sirt5 Zn²⁺-tetrathiolate may be modified, but the integrity of the Zn²⁺-tetrathiolate may not be critical for Sirt3 or Sirt5 deacylase activity or disruption of the Zn²⁺-tetrathiolate may only alter binding and deacetylation of specific protein substrates in vivo, an effect missed when using acetylated peptide substrates in vitro. Alternatively, Sirt3 and Sirt5 may have evolved insensitivity to oxidants to maintain the deacylase activity of these mitochondrially targeted isoforms, as the mitochondrial environment is a primary source of ROS and RNS.

In our in vitro NO-derived oxidant screen, Sirt6 displayed similar patterns of oxidative modification and inhibition to the other nuclear sirtuin isoforms Sirt1 and Sirt2. Sirt6 was S-nitrosated and demyristoylase activity inhibited in a concentration-dependent manner by NO and nitrosothiols such as GSNO (Table 3; Kalous et al., 2020). Mutagenesis of Zn²⁺-tetrathiolate cysteines to alanine indicated the Sirt6 Zn²⁺-tetrathiolate is the likely site of modification (Figures 1A,C; Kalous et al., 2020). Of the sirtuins examined in our study (Sirt1, Sirt2, Sirt3, Sirt5, and Sirt6), Sirt2 and Sirt6 were the only isoforms sensitive to inhibition by NO (released from a NONOate), where the second-order rate constant for Sirt6 inactivation measured by k_inact/K_I analysis was similar to that of Sirt2, suggesting that Sirt6 and Sirt2 are the most likely sirtuins to be inactivated by reaction with NO-derived oxidants compared to the other human sirtuins (Kalous et al., 2020). Future studies are needed to determine if Sirt6 is modified and inhibited by S-nitrosation in cellular or animal models.

In addition to low micromolar GSNO treatment (6–100 μM) yielding Sirt1 S-nitrosation and inhibition of deacetylase activity (Kalous et al., 2016, 2020), Sirt1 can also be glutathionylated upon high concentration (2 mM) GSNO exposure (Table 3; Zee et al., 2010). It is important to note that GSNO preparations can be contaminated with significant amounts of GSSG (Becker et al., 1995; Ji et al., 1999; Giustarini et al., 2005), such that glutathionylation may result from reaction with GSSG instead of GSNO especially when treating with high GSNO concentrations. Mass spectrometry of FLAG-purified Sirt1 treated with 2 mM GSNO from HEK293 or bovine aortic endothelial cell (BAEC) lysate identified Cys67, Cys268, Cys326, and Cys623 as potential glutathionylated residues (Zee et al., 2010). However, instead of using a nitrosothiol-selective reductant such as ascorbate to differentiate the relative contribution of GSNO to S-nitrosation versus glutathionylation, DTT was the only reductant used, which reduces both modifications. Indeed, although five sites of GSNO-dependent modification were initially detected, only one, Cys67 (Figures 1A,B), was definitively identified as glutathionylated upon treatment with 2 mM GSNO (Zee et al., 2010). Therefore, the other sites identified may be sites of S-nitrosation or other GSNO-dependent redox-dependent modification such as inter- or intra-molecular disulfide formation.

While both S-nitrosation and glutathionylation are possible outcomes of a reaction between GSNO and a protein, S-nitrosation is the more kinetically favorable reaction (Gallogly and Mieyal, 2007). To deconvolute the relative ability of GSNO to nitrosate versus glutathionylate Sirt1, recombinantly purified Sirt1 was incubated with increasing concentrations of GSNO and trends in S-nitrosation and glutathionylation were observed (Kalous et al., 2016). Sirt1 was both glutathionylated and S-nitrosated by GSNO; however, the concentration dependence was vastly different. While Sirt1 S-nitrosation occurred in a concentration-dependent manner that directly paralleled Sirt1 inhibition and began at ∼6 μM GSNO, Sirt1 glutathionylation was not observed until 100 μM GSNO, consistent with S-nitrosation being the more favorable reaction. Both S-nitrosation and glutathionylation may collectively contribute to Sirt1 inhibition at concentrations higher than 100 μM. However, as GSNO and other S-nitrosothiols likely to accumulate only to nanomolar to low micromolar concentrations in cells (Broniowska and Hogg, 2012; Seneviratne et al., 2013), S-nitrosation is likely the physiologically relevant Sirt1 modification.

The deacetylase activity of recombinant zebrafish Sirt1 treated with supraphysiological GSSG concentrations (5 mM) showed 17% the deacetylase activity of untreated Sirt1 (Bräutigam et al., 2013). Incubation of glutathionylated Sirt1 with a fivefold molar excess of the oxidoreductase enzyme glutaredoxin-2 (Grx2) removed nearly all glutathione moieties and restored enzymatic activity to 63% that of fully reduced Sirt1, suggesting glutathionylation is a reversible process. A primary site of Sirt1 glutathionylation was localized to Cys204 (corresponds to Cys268 in the human isoform; Figures 1A,B); the cysteine to serine mutant significantly reduced the glutathionylation signal and deacetylase activity but did not completely abolish either. Furthermore, incubation of glutathionylated Sirt1 with Grx2 did not fully deglutathionylate Sirt1 or fully restore deacetylase activity. Taken together, this suggests Sirt1 can be glutathionylated at more than one site.

Cell-based studies suggest glutathionylation can inhibit Sirt1 activity (Shao et al., 2014). In HepG2 cells under conditions mimicking metabolic stress (high palmitate and high glucose) or treatment with S-nitrosocysteine (CysNO), Sirt1 glutathionylation was demonstrated by immunoblot (Shao et al., 2014). Overexpression of the deglutathionylating enzyme glutaredoxin-1 (Grx1) maintained endogenous Sirt1 activity and prevented apoptotic signaling in metabolically stressed HepG2 cells. Mutating a subset of potential glutathionylation target residues initially identified by Zee et al. (2010) abolished glutathionylation in a Sirt1 triple Cys-to-Ser mutant (Cys67, Cys326, and Cys623) (Shao et al., 2014). The oxidation-resistant Sirt1 triple mutant was further used in a study examining a potential glutathionylation-dependent interaction between GAPDH and Sirt1 (Rizvi et al., 2021). Though a definitive site of modification on Sirt1 was not identified, the triple mutant reduced glutathionylation signal and interaction with glutathionylated GAPDH via immunoblot (Rizvi et al., 2021). It was surprising that an established transnitrosating agent such as CysNO increased glutathione adducts on Sirt1 (Shao et al., 2014). However, S-nitrosation may serve as a precursor to glutathionylation by GSH. Importantly, these glutathionylation assays were performed using DTT (Shao et al., 2014) or TCEP (Rizvi et al., 2021) as a reductant prior to biotin labeling of putative glutathionylated cysteines rather than employing a nitrosothiol-selective reductant such as ascorbate. This confounds specificity in the modification identified; DTT and TCEP reduce both nitrosothiols and disulfides (and the other cysteine modifications discussed herein), indicating that treatment with CysNO may in fact result in S-nitrosation.

In animal models, Sirt1 glutathionylation occurs during oxidative stress. Sirt1 glutathionylation was observed in livers of C57/B6J mice fed a high fat, high-sucrose diet, which promoted production of ROS (assessed via dichlorodihydrofluorescein fluorescence), inhibition of Sirt1 deacetylase activity, and apoptosis via increased p53 acetylation (Table 4; Shao et al., 2014). A similar phenomenon was demonstrated in zebrafish, where Sirt1 glutathionylation correlated with delayed and disordered blood vessel network formation (Table 4; Bräutigam et al., 2013). Glutathionylated Sirt1 was observed in lungs of C57BL/6J mice exposed to cigarette smoke (Caito et al., 2010). Inhaled cigarette smoke increased Sirt1 glutathionylation, which was further augmented in Grx1 knockout (Grx1^–/–) mice (Caito et al., 2010). Additionally, Grx1 deletion alone increased Sirt1 glutathionylation, even in the absence of cigarette smoke exposure (Caito et al., 2010).

Sirt3 glutathionylation has been observed in response to direct exposure of recombinantly purified Sirt3 to glutathionylating agents and in response to physiological inflammatory stimuli (Table 3; Dikalova et al., 2017; Kalous et al., 2020). In vitro exposure of recombinantly purified Sirt3 to 100 μM H₂O₂/GSH resulted in glutathionylation by anti-glutathione but no change in Sirt3 deacetylase activity (Kalous et al., 2020). Sirt3 is more sensitive to glutathionylation by 100 μM H₂O₂/GSH than by 100 μM GSSG, with significantly higher anti-glutathione signal observed in response to 100 μM H₂O₂/GSH (Dikalova et al., 2017; Kalous et al., 2020).

Sirt3 glutathionylation has been observed in animal models. Sirt3 is glutathionylated in vivo in the aorta and kidney of hypertensive mice, which correlated with increased acetylation of the Sirt3 substrate SOD2, suggesting Sirt3 deacetylase activity is inhibited in vivo under inflammatory conditions (Dikalova et al., 2017). In clinically hypertensive human subjects, although Sirt3 glutathionylation was not examined, SOD2 acetylation was significantly increased, suggesting that both Sirt3 deacetylase and SOD2 dismutase activity are inhibited in hypertension (Table 4; Dikalova et al., 2017). Although increased Sirt3 glutathionylation was observed cellularly, Sirt3 protein levels were also reduced by ∼50%, suggesting that, instead of directly inhibiting Sirt3 activity, glutathionylation may destabilize Sirt3 or serve as a degradation signal in vivo, resulting in loss of Sirt3 activity via depletion of Sirt3 protein levels (Dikalova et al., 2017). Taken together, Sirt3 glutathionylation may either promote degradation of Sirt3, alter the Sirt3 interactome, or as a direct inhibitory modification under inflammatory conditions.

Sirt5 has only been examined in vitro to assess glutathionylation status. Recombinantly purified Sirt5 was not glutathionylated by treatment with conditions (100 μM GSNO, GSSG, or H₂O₂/GSH) that yielded glutathionylated Sirt3 (Table 3; Kalous et al., 2020).

A single study investigated the glutathionylation state of recombinantly purified Sirt6. Anti-glutathione immunoblots showed Sirt6 glutathionylation when treated with either GSSG or H₂O₂/GSH (100 μM) (Table 3; Kalous et al., 2020). However, purified Sirt6 demyristoylase activity was not significantly inhibited by 100 μM GSSG treatment, suggesting glutathionylation is not an inhibitory modification.

In a recent study, RKO colon adenocarcinoma cells were treated with 500 μM H₂O₂, and A4331 epidermal carcinoma cells were stimulated with EGF to mimic inflammatory conditions. In both cases, robust Sirt6 sulfenylation was observed exclusively at a cysteine residue unique to Sirt6 (Cys18; Figures 1A,C) via DYn-2 probe followed by LC-MS/MS analysis (Table 3; Yang et al., 2014). Incubation of recombinantly purified Sirt6 and hypoxia-inducible factor 1-α (HIF1α) with 500 μM H₂O₂ demonstrated electromobility shifts in non-reducing gels for both Sirt6 and HIF1α, consistent with formation of a Sirt6-HIF1α dimer (Yang et al., 2014). Dimerization was mediated by a disulfide and dependent on cysteine sulfenylation, as the Sirt6-HIF1α dimer was not observed in the absence of H₂O₂, or in the presence of H₂O₂ under reducing conditions. Moreover, formation of a Sirt6-HIF1α dimer was prevented by cotreatment with 500 μM H₂O₂ and 10 mM dimedone, which selectively and covalently labels sites of cysteine sulfenylation. Mass spectrometry identified an intermolecular disulfide between Sirt6 Cys18 and HIF1α Cys800. An important caveat to these results is that this study employed a supraphysiological H₂O₂ concentration (500 μM) (Yang et al., 2014). Physiological H₂O₂ concentrations are in the low micromolar range at maximum (Sies and Jones, 2020). Indeed, a more conservative treatment of recombinantly purified Sirt6 with 100 μM H₂O₂ showed no sulfenylation (Kalous et al., 2020).

Sirt6 sulfenylation was also detected in THP-1 monocytes stimulated with LPS; however, this study localized the sulfenic acid to Cys144 (Figures 1A,C), one of the four Sirt6 Zn²⁺-tetrathiolate cysteines (Long et al., 2017). Sulfenylation correlated with reduced Sirt6 deacetylase activity and increased GLUT1 translocation to the cell membrane and increased glycolytic activity in THP1 monocytes (Long et al., 2017). Interestingly, HIF1α regulates GLUT1 transcription (Chen et al., 2001). These data therefore provide physiological evidence that Sirt6 plays a repressive role in modulating HIF1α signaling.

Sirt1 sulfhydration has been demonstrated in cellular systems. In a 1-methyl-4-phenylpyridinium (MPP+) induced neurotoxicity model, SH-SY5Y neuroblastoma cells treated with 500 μM of the H₂S donor NaHS prevented morphological and apoptotic effects of MPP+, and a modified biotin switch technique revealed increased sulfhydration of Sirt1 (Li et al., 2020). In the hepatic HepG2 cell line, increasing the output of H₂S-producing enzyme cystathionine gamma-lyase (CSE) or treatment with NaHS led to an increase in both Sirt1 expression levels and deacetylase activity (Du et al., 2019). A complementary study in the human kidney HK-2 cell line found treatment with the polysulfide donor Na₂S₄ reversed effects of high glucose treatment; glucose-induced reduction in Sirt1 protein level was reversed, along with reversal of glucose-induced increases in STAT3 and p65 acetylation (Sun et al., 2021). Sirt1 sulfhydration was detected by Western blot following modified biotin switch; mutagenesis identified the modification site as the Zn²⁺-tetrathiolate (Du et al., 2019; Sun et al., 2021).

Sirt3 sulfhydration has been reported in multiple systems. In a model of chemotherapy-induced toxicity, human kidney cells (HK-2) or AKI mice were treated with cisplatin; basal levels of Sirt3 sulfhydration decreased, but rebounded upon NaHS treatment (Yuan et al., 2019). Likewise, in a liver toxicity model, the rat liver cell line BRL-3 was treated with the pesticide paraquat, or paraquat in combination with NaHS; paraquat treatment reduced sulfhydrated Sirt3 whereas co-treatment with NaHS restored Sirt3 sulfhydration (Liu et al., 2020). Functional downstream effects of Sirt3 sulfhydration in HK-2 cells included activation of Sirt3-catalyzed deacetylation of OPA1, SOD2, and ATP synthase subunit β (Yuan et al., 2019). The Zn²⁺-tetrathiolate was identified as the modification site via mutagenesis (Yuan et al., 2019).

Although the precise site(s) of nitration were not identified, global Sirt1 tyrosine nitration was demonstrated in both the retinas of mice and rats with diabetic retinopathy via microscopy (Table 3; Duarte et al., 2016). Under high glucose conditions, Sirt1 deacetylase activity decreased, as measured via an in vitro fluorometric assay and an increase in p65 acetylation. Direct detection of tyrosine nitration of immunopurified Sirt1 was visualized by immunoblot. Additionally, silencing NADPH oxidase 4 (NOX4) under high glucose conditions prevented Sirt1 inhibition, suggesting NOX4 is the primary source of O₂•⁻ production responsible for ONOO^– formation (via radical recombination between O₂•⁻ and NO) and Sirt1 inhibition under these conditions (Table 4; Duarte et al., 2016). In mouse aorta treated with nicotine, a microscopic overlay of Sirt1 and anti-3-nitrotyrosine showed co-localization (Table 4; Ding et al., 2019). Sirt1 deacetylase activity was also decreased, and Zn²⁺ was released from recombinant Sirt1 upon 50 μM OONO^– treatment, indicating tyrosine nitration of Sirt1 may be an inhibitory modification (Ding et al., 2019).

Sirt2 tyrosine nitration was indirectly implicated in rat glia Schwannoma cells treated with high glucose. An increase of global nitration correlated with increased acetylated α-tubulin and decreased Sirt2 protein (Table 3; Gadau, 2013), consistent with Sirt2 inhibition. Acetylated α-tubulin also correlated with dysregulation of the microtubular network (Gadau, 2013), which represents a potential pathway through which the structural changes observed in diabetic neuropathy occur. Future studies are needed to determine if Sirt2 nitration is directly responsible for the observed increased α-tubulin acetylation in Schwannoma cells.

Recombinantly purified Sirt3 treated with 100–500 μM ONOO^– showed a concentration-dependent inhibition by an in vitro fluorometric assay (Table 3; Pérez et al., 2018). Increased Sirt3 nitration in mouse brain of aged (24-month-old) relative to young (3-month-old) animals has also been observed. The source of nitration was nNOS-dependent and activated by ROS-regulating adaptor protein p66^Shc, as whole-body deletion of p66^Shc resulted in significantly reduced mitochondrial protein nitration in aged p66^Shc knockout (p66^Shc–/–) mice. p66^Shc can activate NAPDH oxidase, thereby increasing the O₂•^– production (Pérez et al., 2018) necessary for ONOO^– formation. If sufficient ONOO^– is produced to inhibit Sirt3, increased acetylation and decreased activity of critical antioxidant and other mitochondrial metabolic enzymes may result, which requires further study.

Recombinantly purified Sirt6 showed tyrosine nitration and deacetylase inhibition when treated with the OONO^– donor SIN-1 at high concentrations (0.5 and 5 mM) (Table 3); the specific site of modification was identified as Tyr257 by mass spectrometry analysis (Figures 1A,C; Hu et al., 2015). Sirt6 tyrosine nitration and deacylase inhibition, as determined by a fluorometric assay, was likewise demonstrated in HEK293 cells and human retinal microvascular endothelial cells treated with 0.5 or 5 mM of SIN-1 (Hu et al., 2015). Nitration and inhibition of Sirt6 was also observed in the retina of mice where systemic inflammation was induced by injection of LPS, suggesting nitration may be a relevant inhibitory modification of Sirt6 in vivo (Hu et al., 2015). Oxidants such as NO and ONOO^– cause both single and double-stranded breaks in DNA (Burney et al., 1999), and Sirt6 facilitates DNA damage repair to increase the efficiency of strand break repair (Tian et al., 2019). However, if DNA damage is too extensive, protective mechanisms are shut down, and the cell is directed toward apoptosis (Roos and Kaina, 2013). In this case, Sirt6 inhibition by oxidants may serve as a mechanism to inhibit pro-survival responses once oxidant-dependent DNA damage becomes too great to repair.