Diverse Consequences of Hexosamine Biosynthesis in Cardiovascular Disease

In the stressed heart, metabolic remodeling precedes most, if not all, other pathophysiological changes [1]. When the heart is not stressed, it relies on fat for energy provision, with glucose being an additional energy source, while lactate, ketone bodies, and amino acids contribute as well, but do so only to a small extent. As a metabolic omnivore, the heart switches its nutrient preference towards more glucose under stress. After uptake, glucose is converted to pyruvate in the glycolytic pathway (Fig. 1). Pyruvate either enters the mitochondria for oxidation, or it is reduced to lactate. In addition to glycolysis, the carbons of glucose may be shunted into ancillary metabolic pathways, including hexosamine biosynthetic pathway (HBP), pentose phosphate pathway, glycogen synthetic pathway, and serine biosynthetic pathway.

Since its first discovery in 1984, HBP has been shown to play critical roles in various diseases, including cardiovascular disease, diabetes, and cancer. HBP flux is governed by its rate-limiting enzyme GFAT (Glutamine:fructose-6-phosphate aminotransferase) (Fig. 1). As the end product of HBP, UDP-GlcNAc serves as substrate for both O-GlcNAcylation and N-glycosylation, two common posttranslational modifications. O-GlcNAcylation is regulated by two enzymes, O-GlcNAc transferase (OGT) and O-GlcNAcase (OGA), which conjugate the GlcNAc moiety to and removes it from serine and threonine residues of target proteins, respectively. In addition to targets like NFAT, CAMKII, HDAC4, and FoxO1, the above three key O-GlcNAcylation enzymes themselves can be modified by O-GlcNAcylation, yielding a tight and precise control of O-GlcNAcylation.

Previous studies show that HBP and O-GlcNAcylation play a crucial role in pathological cardiac hypertrophy and heart failure [2]. In cultured cardiomyocytes, O-GlcNAcylation is upregulated by multiple pro-hypertrophic stimuli, including phenylephrine, and angiotensin II, while inhibition of this modification represses hypertrophic growth. Potential mechanisms include NFAT O-GlcNAcylation to promote its nuclear translocation and O-GlcNAcylation upregulation to activate the pro-hypertrophic ERK signaling pathway. Consistently, in vivo data from failing human hearts and hearts of animal models with heart failure confirm that O-GlcNAcylation is increased in heart failure, whereas reducing O-GlcNAcylation exerts a cardioprotective role. Recently, Gelinas et al. found AMPK activation protects mouse heart from pressure overload-induced cardiac hypertrophy by promoting the inhibitory phosphorylation of GFAT at serine 243, downregulating OGT expression, and inhibiting O-GlcNAcylation [3]. Using cardiac specific GFAT1 overexpression and knockout mouse models, we demonstrated that prolonged elevation of HBP and O-GlcNAcylation exacerbates cardiac dysfunction by sustained activation of mTOR pathway under chronic hemodynamic stress [4]. Taken together, these findings suggest that long-term induction of HBP by hemodynamic stress is detrimental to the heart.

In contrast, myocardial infarction and ischemia/reperfusion (I/R) inflict an acute injury to the heart. Previous studies show that elevation of O-GlcNAcylation protects cardiomyocytes from acute ischemic injury [5]. In an in vitro model of simulated I/R using cardiomyocytes, glucosamine treatment augments O-GlcNAcylation and decreases cell death by promoting mitochondrial Bcl-2 translocation and restoring mitochondrial membrane potential. In addition, pharmacological upregulation of O-GlcNAcylation attenuates both hydrogen peroxide triggered cardiomyocyte death in vitro and I/R injury in vivo, which can be explained, at least partially, by increased O-GlcNAc-modified voltage-dependent anion channel, preservation of membrane potential as well as reduced calcium-induced mPTP activation [6]. Along these lines, Watson et al. reported that inhibition of O-GlcNAcylation by cardiac specific OGT deletion aggravates cardiac dysfunction after myocardial infarction [7]. Taken together, these findings suggest that acute activation of HBP is cardioprotective in ischemic heart disease.

O-GlcNAcylation also participates in the pathogenesis of diabetes related cardiomyopathy. Recently, Lu et al. reported that high glucose activates O-GlcNAcylation of CaMKIIδ, leading to acute production of reactive oxygen species in cardiomyocytes [8]. This is consistent with a finding in type I diabetes showing overexpression of OGA improves calcium handling and heart function [9]. On the other hand, Backs and colleagues found that O-GlcNAcylation of HDAC4 protects the diabetic heart by counteracting pathological CaMKII signaling [10]. Collectively, the role of O-GlcNAcylation in diabetes related cardiomyopathy still remains incompletely understood.

In conclusion, O-GlcNAcylation, as a nutrient and stress sensor, has diverse consequences in cardiovascular disease (Fig. 1). It is generally agreed that prolonged elevation of HBP and O-GlcNAcylation in chronic heart disease provokes cardiac dysfunction, whereas acute activation of HBP is adaptive. The difference may be explained by intensity, duration, and targets of O-GlcNAcylation under different disease conditions. While HBP represents a promising target, some critical issues remain. For example, it is challenging to determine global O-GlcNAcylation profile due to its labile nature. In addition, the effect of O-GlcNAcylation may lie on multiple O-GlcNAcylated proteins. Moreover, O-GlcNAcylation of different targets may exert opposite effects on the same signaling pathway. Therefore, caution should be exercised when manipulating HBP for therapeutic gain against cardiovascular disease of different etiologies.