Histidine triad nucleotide-binding 2 (HINT2)is a member of the histidine triad superfamily of nucleotide hydrolase and transferase enzymes. This family of proteins catalyzes reactions on nucleotide-containing substrates in pathways important for cell growth, apoptosis, and DNA, RNA, and carbohydrate metabolism.1 HINT2 has high-sequence homology (61% identical) to HINT1, a purine nucleotide-binding protein that hydrolyzes adenosine monophosphate–NH2 and lysyl-adenylate.2 HINT1 is ubiquitously expressed and has tumor suppressor activity in liver.3 In contrast to HINT1, HINT2 has adenosine phosphoramidase activity and shows expression restricted to the liver, pancreas, and adrenal gland. HINT2 is localized exclusively in the mitochondria4 and has unknown biological function.5 In this issue of Hepatology, Martin et al.6 report a study in which they ablated the Hint2 gene in mice to uncover a physiological role for hepatic HINT2. The resultant Hint2 knockout (KO) mice display a complex metabolic phenotype consistent with marked mitochondrial dysfunction in the liver.
By 20 weeks of age, Hint2 KO mice develop fatty liver, decreased glucose tolerance, delayed recovery from insulin-induced hypoglycemia, and increased levels of plasma insulin, leptin, adiponectin, and cholesterol. In the mitochondria, Hint2 KO mice have decreased respiration and display morphologically deformed, fused mitochondria. Additionally, Hint2 KO mice have increased HIF-2a and reactive oxygen species levels. Martin et al. concluded that HINT2 plays a major role in the regulation of mitochondrial lipid metabolism, respiration, and glucose homeostasis. One particularly striking feature of the Hint2 KO mouse is a global increase in mitochondrial protein acetylation.
Several studies have demonstrated a role for mitochondrial protein acetylation in the regulation of many metabolic pathways (for a recent review, see Anderson and Hirschey7). Ablation of the mitochondrial NAD+-dependent deacetylase sirtuin-3 (Sirt3) in mice results in mitochondrial dysfunction. Sirt3 KO mice have hyperacetylated mitochondrial proteins and, similar to Hint2 KO mice, develop fatty liver, glucose intolerance, decreased respiration, and decreased resistance to oxidative stress.8-12 Using this mouse model, the role of hyperacetylation in regulating several metabolic pathways has been mechanistically defined. For example, hyperacetylation of the fatty acid oxidation enzyme, long-chain acyl coenzyme A (CoA) dehydrogenase, reduces its activity and leads to reduced lipid oxidation and the development of fatty liver.8 Additionally, reduced mitochondrial respiration and adenosine triphosphate (ATP) production occurs due to hyperacetylation of multiple components of complexes I and II, which inhibits their function.13,14 Sirt3 KO mice are less resistant to oxidative stress due to hyperacetylation and inhibition of manganese superoxide dismutase, which is a critical component of the antioxidant system.9,12 The similarity between the metabolic phenotypes in the Hint2 and Sirt3 KO mice is compelling and suggests that hyperacetylation may contribute, at least in part, to the hepatic metabolic phenotype in the Hint2 KO animals. Sirt3 KO mice accumulate hyperacetylated mitochondrial proteins because the enzyme responsible for reversing this posttranslational modification is absent. However, Martin et al. show that Hint2 KO mice express normal levels of enzymatically active SIRT3, suggesting an alternative mechanism leading to mitochondrial protein hyperacetylation.
A second well-established way to induce protein hyperacetylation is by altering nutrient status in mice. High-fat diet-feeding is one way to induce hyperacetylation and can be explained by a corresponding decrease in SIRT3 expression.15,16 Less SIRT3 expression and less deacetylase activity results in mitochondrial protein hyperacetylation. Additionally, mitochondrial protein hyperacetylation is induced by calorie restriction. Because calorie restriction increases SIRT3 expression, the mechanism for increased acetylation in this condition remains unknown.9,17 An additional way to induce hyperacetylation is through chronic ethanol feeding paradigms. In ethanol-fed mice, SIRT3 protein expression remains normal, but deacetylase activity is dampened by protein carbonylation of SIRT3.18,19 Protein carbonylation, thought to occur as a consequence of ethanol metabolism, allosterically inhibits SIRT3, and this inhibition has been hypothesized to explain the increase in acetylation. However, none of these dietary manipulations can explain the elevation in mitochondrial protein acetylation in Hint2 KO mice under normal standard-diet feeding.
Elevated mitochondrial protein acetylation has been observed in one additional mouse model, the Friedreich's ataxia mouse, in which the gene encoding frataxin has been ablated.20 Frataxin is a mitochondrial, iron-binding protein required for the assembly of iron-sulfur cluster proteins, including complexes I, II, and III of the respiratory chain, and aconitase of the trichloroacetic acid cycle (for a review, see Martelli et al.21). Frataxin deficiency severely compromises respiration and mitochondrial function leading to energetic stress and ATP deficiency. Wagner et al.20 showed that hyperacetylation of mitochondrial proteins in this model is caused by inhibition of SIRT3 activity due to an 85-fold decrease in NAD+/NADH. Furthermore, they showed that SIRT3 was carbonylated, which was previously shown to reduce its activity.18 Martin et al. suggest that inhibition of SIRT3, and the resulting protein hyperacetylation, is a feedback mechanism to limit oxidative metabolism when mitochondrial respiration is compromised.
The overlapping metabolic phenotypes of the Hint2 KO mice and the Sirt3 KO mice suggest that mitochondrial protein hyperacetylation in the Hint2 KO mice could explain some of the metabolic abnormalities. However, what remains unknown is the mechanism of acetylation in Hint2 KO mice. Although Martin et al. tested for possible reductions in SIRT3 and found no difference in deacetylase activity from mitochondrial extracts of wild-type and Hint2 KO mice using a kit-based assay, a more rigorous assessment may be prudent in order to conclude that SIRT3 activity is unaffected by deletion of Hint2. The substrate used in their assay was not specific for SIRT3, and an acetylated fluorescent peptide could be deacetylated equally well by contaminating nuclear or cytoplasmic deacetylases.
Additionally, SIRT3 activity could be dampened by reductions in local NAD+ concentration in Hint2 KO mice, as suggested by Martin et al. While the NAD+ or NADH levels were not measured, Martin et al. showed the activities of two other NAD+-dependent enzymes were reduced: glutamate dehydrogenase (GDH) and short-chain 3-hydroxyacyl-CoA dehydrogenase. These data suggest that reduced NAD+, leading to reduced SIRT3 activity, could explain the hyperacetylation observed in Hint2 KO mice. However, Martin et al. show that exogenous SIRT3 deacetylates GDH, as described,22 and stimulates more GDH activity in Hint2 KO mice compared with wild-type mice. These findings show that addition of SIRT3, in the absence of exogenous NAD+, can rescue GDH activity. Therefore, the mitochondrial protein acetylation in Hint2 KO mice cannot be fully explained by reduced NAD+ levels.
Thus, more work is required to identify the mechanism of mitochondrial protein acetylation in Hint2 KO mice. One exciting possibility is that deletion of Hint2 stimulates acetylation of mitochondrial proteins by a novel direct, or indirect, mechanism (Fig. 1A). A second possibility is that Hint2 deletion reduces the NAD+ pool, thereby reducing SIRT3 activity (Fig. 1B). The histidine triad superfamily has nucleotide-binding properties, which suggests that the NAD+ pool could be modulated by HINT2. Third, HINT2 could regulate SIRT3 activity directly, and its absence leads to reduced mitochondrial protein deacetylation (Fig. 1C). For example, HINT1 has lysine deadenylase activity, and if true for HINT2, then SIRT3 could be a direct substrate for HINT2 to modulate its activity.
The study by Martin et al. demonstrates an important role for Hint2 in hepatic function. Without Hint2, mitochondrial function is disrupted. In this setting, mitochondrial protein hyperacetylation appears to be a major functional consequence of Hint2 ablation. Further studies on HINT2 to determine its substrates and enzymatic activity, as well as its influence on the NAD+ pool, will uncover how Hint2 ablation induces mitochondrial protein acetylation.