Post‐translational modifications in diabetic cardiomyopathy

Abstract

The increasing attention towards diabetic cardiomyopathy as a distinctive complication of diabetes mellitus has highlighted the need for standardized diagnostic criteria and targeted treatment approaches in clinical practice. Ongoing research is gradually unravelling the pathogenesis of diabetic cardiomyopathy, with a particular emphasis on investigating various post‐translational modifications. These modifications dynamically regulate protein function in response to changes in the internal and external environment, and their disturbance of homeostasis holds significant relevance for the development of chronic ailments. This review provides a comprehensive overview of the common post‐translational modifications involved in the initiation and progression of diabetic cardiomyopathy, including O‐GlcNAcylation, phosphorylation, methylation, acetylation and ubiquitination. Additionally, the review discusses drug development strategies for targeting key post‐translational modification targets, such as agonists, inhibitors and PROTAC (proteolysis targeting chimaera) technology that targets E3 ubiquitin ligases.

1. INTRODUCTION

Post‐translational modifications (PTMs) enhance the functional diversity of the proteome by introducing covalent additions of functional groups or proteins, proteolytic cleavage of regulatory subunits or protein degradation. ¹ The utilization of mass spectrometry‐based proteomics, particularly electrospray ionization mass spectrometry, has progressively unveiled the magnitude and prevalence of PTMs, highlighting its complexity. ² To date, over 200 PTMs have been identified. These chemical modifications, such as phosphorylation, glycosylation, ubiquitination, nitrosylation, methylation, acetylation, lipidation and protein hydrolysis, play a crucial role in cellular and subcellular compartments. These modifications can be reversible, depending on their nature. ³ They dynamically modulate homeostasis in response to environmental changes and regulate multiple cellular signals. Consequently, studying these PTMs is of great significance, particularly in the context of chronic diseases such as cardiovascular disease, cancer, neurodegenerative diseases and diabetes.

Diabetic cardiomyopathy (DCM) is a pathophysiological condition resulting from diabetes mellitus (DM) that ultimately culminates in heart failure (HF). DCM can manifest in both type 1 and type 2 diabetes, with type 2 accounting for the majority of cases (90%–95%). ⁴ Among patients with DM, cardiovascular disease, especially coronary artery disease and ischemic cardiomyopathy, is the primary cause of mortality. ⁵ However, DCM is defined as myocardial dysfunction in the absence of coronary artery disease, hypertension or valvular heart disease. ⁶ Patients diagnosed with DCM initially present with myocardial diastolic dysfunction, which is clinically evident as HF with preserved ejection fraction. ⁷ , ⁸ Irrespective of the preservation or reduction in cardiac function, the most prominent features of DCM are increased extracellular matrix production and left ventricular (LV) hypertrophy. ⁹ , ¹⁰ , ¹¹ The pathogenesis of DCM is intricate, encompassing systemic insulin resistance and metabolic disorders, impaired Ca²⁺ handling, reduced nitric oxide bioavailability, mitochondrial dysfunction, elevated oxidative stress, myocardial fibrosis, cardiac autonomic neuropathy and microvascular dysfunction ⁷ (Figure 1).

FIGURE 1.

Pathogenesis associated with diabetic cardiomyopathy. An intricate interplay among systemic insulin resistance and metabolic disorders, impaired Ca²⁺ handling, mitochondrial dysfunction, oxidative stress, myocardial hypertrophy and myocardial fibrosis.

An increasing number of evidence shows a strong correlation between DCM and both genetic and environmental factors. Moreover, it has been suggested that the impaired contraction of the myocardium may be attributed, at least in part, to PTMs of specific proteins. ⁴ Furthermore, the therapeutic options available for diastolic dysfunction are limited, thus emphasizing the urgency of comprehending the molecular mechanisms underlying diabetic cardiac dysfunction. This review profoundly summarizes and discusses the common PTMs implicated in the pathogenesis of DCM, including O‐GlcNAcylation, phosphorylation, methylation, acetylation and ubiquitination. The aim is to establish a novel theoretical framework and therapeutic target for the management of DCM.

2. PROTEIN O‐GlcNAcylation IN DIABETIC CARDIOMYOPATHY

O‐GlcNAcylation is a form of protein modification where O‐linked β‐N‐acetylglucosamine (O‐GlcNAc) is attached to serine and threonine residues. ¹² This process involves two enzymes: O‐GlcNAc transferase (OGT) and O‐GlcNAcase (OGA, also referred to as NCOAT or MGEA5). ¹³ , ¹⁴ OGT covalently attaches O‐GlcNAc to the protein from the glycosylated donor uridine diphosphate N‐acetylglucosamine (UDP‐GlcNAc, the end product of metabolic hexosamine biosynthetic pathway), which means that catalyses the transfer of O‐GlcNAc from UDP‐GlcNAc to serine and threonine, while OGA hydrolyzes O‐GlcNAc from the protein. ¹³ This process is intricately linked to glucose metabolism and arises from an overload of the hexosamine biosynthesis pathway. ¹⁵ As a result, a correlation exists between elevated O‐GlcNAcylation levels and diabetes, ultimately contributing to the pathogenesis of various cardiovascular diseases. ¹⁶

In DCM, the primary pathophysiological manifestations are cardiac insulin resistance and subsequent metabolic disturbances. Unlike the normal adult heart, which primarily relies on fatty acid oxidation for energy production, the diabetic heart exhibits the ability to redirect acetyl coenzyme A (CoA) towards ketone body synthesis, resembling the metabolic pattern typically observed in the foetal heart. ¹⁷ , ¹⁸ This metabolic alteration is evidenced by the significantly elevated levels of cardiac acetoacetate and β‐hydroxybutyrate in diabetic mice ¹⁹ (Figure 2). Acetoacetate and β‐hydroxybutyrate are the two principal ketone bodies that can be produced through the catabolism of acetyl CoA by the enzymes SCOT (succinyl‐CoA: 3‐oxoacid CoA transferase) and BDH1 (β‐hydroxybutyrate dehydrogenase). ²⁰ It was observed that BDH1, the enzyme responsible for β‐hydroxybutyrate dehydrogenase activity, undergoes direct modification by O‐GlcNAc, leading to suppressed expression. This finding suggests that increased O‐GlcNAcylation may hinder the utilization of ketone bodies ¹⁹ (Figure 2). Consequently, O‐GlcNAcylation could serve as a potential mechanistic link connecting glucose and ketone body metabolism. The current consensus in academia is that acute increases in O‐GlcNAcylation have a protective effect against cardiac injury. Conversely, chronic increases in O‐GlcNAcylation have been linked to impaired calcium handling and contractile properties, mitochondrial dysfunction, myocardial hypertrophy and myocardial fibrosis, thereby exacerbating the development of DCM. The role of Ca²⁺ release and uptake by the sarcoendoplasmic reticulum in myocardial contraction and relaxation is widely recognized in academic literature. It has been established that the rate of calcium sequestration into the sarcoendoplasmic reticulum is controlled by the sarcoendoplasmic reticulum ATPase 2a (SERCA2a) gene. ²¹ Notably, heightened O‐GlcNAcylation has been found to reduce the expression of the SERCA2a gene, partially through its impact on nuclear transcription factors. This process ultimately led to compromised diastolic Ca²⁺ sequestering ²² (Figure 2). Interestingly, the presence of O‐GlcNAcylated myofilaments had a significant impact on myocardial contractility, potentially due to a decrease in cardiac myofilament Ca²⁺ sensitivity resulting from increased O‐GlcNAcylation levels. ²³ Calcium/calmodulin (Ca²⁺/CaM) dependent protein kinase II (CaMKII), a multifunctional serine/threonine kinase, is known to respond to acute β‐adrenergic activation and contribute to cardiac remodelling under pathological stress. ²⁴ , ²⁵ Numerous studies have demonstrated the involvement of CaMKII in various heart diseases, including myocardial hypertrophy, ²⁵ HF, ²⁵ , ²⁶ ischemia/reperfusion injury, ²⁷ and myocardial infarction. ²⁸ In recent years, research has shown that under conditions of diabetic hyperglycaemia, CaMKII can be modified by O‐GlcNAc, resulting in its activation and subsequent augmentation of sarcoplasmic reticulum Ca²⁺ release ²⁹ (Figure 2). Specifically, the increased CaMKII activity was consistent with O‐GlcNAcylated CaMKII S279. ²⁹ Meanwhile, the increased O‐GlcNAcylation, coupled with CaMKII activity, prompted the generation of reactive oxygen species (ROS) in mouse cardiomyocytes, further aggravating DCM ³⁰ (Figure 2). While the inhibition of CaMKII activity might restore the systolic and diastolic function of the myocardium in individuals with type 2 diabetes. ³¹ In addition, O‐GlcNAcylation induced by DM may lead to mitochondrial dysfunction associated with DCM, and over 88 mitochondrial proteins can be O‐GlcNAcylated. ³² , ³³ The main manifestation was not only a more than twofold increase in mitochondrial protein O‐GlcNAcylation in DCM with an increase in OGT and a decrease in OGA, but also mislocalized OGT in diabetic mitochondria with impaired activity of complex IV. ³² Molecular insights implicated that elevated O‐GlcNAcylation of 8‐oxoguanine DNA glycosylase (Ogg1, the main DNA glycosylase responsible for repair of the ROS‐induced mutagenic DNA lesion) increased mitochondrial DNA damage ³⁴ (Figure 2). Regarding myocardial hypertrophy, previous studies have demonstrated that O‐GlcNAcylation can induce the activation of transcription factors, including NFAT (nuclear factor of activated T‐cells), GATA4 or MEF2C (myocyte enhancer factor 2C). ³⁵ , ³⁶ A recent study proved that elevated O‐GlcNAc modification is associated with cardiac fibrosis and upregulated expression of pro‐hypertrophic genes such as Myh‐7 and Nppa ³⁷ (Figure 2). In humans, there is a correlation between elevated O‐GlcNAcylation and LV dysfunction. Additionally, the density of LV O‐GlcNAc is positively correlated with blood glucose levels and inversely correlated with LV ejection fraction. ³⁷ The impaired cardiac phosphatidylinositol 3‐kinase (PI3K) (p110α)/AKT signalling is suggested as a potential mechanism behind this phenomenon. O‐GlcNAcylation often contributes to cardiac damage, but it has also been demonstrated that certain molecules modified by O‐GlcNAcylation can provide protection against HF in DCM. For instance, it was shown that the re‐expression of the N‐terminal proteolytic fragment of Histone Deacetylase 4 (HDAC4‐NT) under high O‐GlcNAc conditions can serve as a preventive measure against HDAC4‐dependent DCM, whereas mice lacking HDAC4 (HDAC4‐KO) develop HF. ³⁸ The effects of O‐GlcNAcylation on the heart appear to be inconsistent and depend on specific changes at certain sites.

FIGURE 2.

Pathogenesis associated with O‐GlcNAcylation and diabetic cardiomyopathy. The level of cardiac acetoacetate and β‐hydroxybutyrate was striking increased; BDH1 was directly modified by O‐GlcNAc and then the expression was suppressed, inhibiting ketone body utilization. In addition, SERCA2a and CaMKII can be modified by O‐GlcNAc, leading to impaired diastolic Ca²⁺ sequestering and increased sarcoplasmic reticulum Ca²⁺ release. Meanwhile, enhanced CaMKII activity by O‐GlcNAcylation induced ROS production. Elevated Ogg1 expression increased mitochondrial DNA damage. Furthermore, Myh‐7 and Nppa pro‐hypertrophic gene expression increased under O‐GlcNAcylation, presenting cardiac fibrosis.

The intricate nature of protein O‐GlcNAcylation in cardiovascular disease gives rise to highly nuanced effects that are contingent upon the specific target proteins involved. As mentioned above, modifications of different genes result in different outcomes in Ca²⁺ handling, leading to impaired myocardial contraction or relaxation function. Promising strategies for restoring physiological O‐GlcNAc homeostasis and improving cardiac function include targeted interventions at specific loci and regulatory modifications, such as adeno‐associated viral vector gene therapy. ³⁷

3. PROTEIN PHOSPHORYLATION IN DIABETIC CARDIOMYOPATHY

Phosphorylation, a well‐researched PTMs, exerts regulatory control over cell growth, differentiation, apoptosis and signal transduction. Unlike O‐GlcNAcylation, phosphorylation is governed by a multitude of specific kinases and phosphatases, responsible for the addition and removal of phosphate groups, respectively. ³⁹ The serine/threonine kinases comprise the largest subset of these enzymes. In contrast to O‐GlcNAcylation, phosphorylation indirectly impairs cardiac function by affecting various signalling pathways. Notably, the phosphorylation of PI3K/AKT assumes a crucial role in insulin resistance and the pathogenesis of DCM. ⁴⁰

The classical signalling cascade of the insulin receptor (IR) commences with its binding to insulin, followed by autophosphorylation of IR and subsequent activation of various kinases, including the PI3K‐AKT signalling pathway. ⁴¹ Then, the downstream signals regulate cellular physiological processes, such as glucose transport, glycogen synthesis and protein translation by activating numerous downstream target proteins. ⁴¹ Research demonstrated that the association between β‐adrenergic signalling and insulin resistance exhibited a biphasic effect. ⁴² Short‐term stimulation increased insulin‐stimulated glucose uptake through PKA/Ca²⁺‐dependent and PI3K‐independent pathways mediated AKT phosphorylation. In contrast, long‐term stimulation resulted in the inhibition of insulin‐stimulated glucose uptake and insulin‐induced autophosphorylation of IRs in cardiac myocytes, while concurrently promoting threonine phosphorylation of IRs. Insulin resistance is additionally linked to the activation of the renin‐angiotensin system. ⁴³ Studies have shown that the administration of the renin‐angiotensin system inhibitor captopril could potentially rectify insulin signalling and regulate substrate utilization within the myocardium, thereby enhancing cardiac function. ⁴⁴ The underlying mechanisms involved may entail heightened phosphorylation of AKT and reduced activation of adenosine mono phosphate‐activated protein kinase (AMPK), ultimately leading to the restoration of insulin sensitivity and the improvement of myocardial energetics ⁴⁴ (Figure 3). The transcription factor forkhead box‐containing protein O (FoxO) is widely recognized as a crucial component in regulating insulin signalling, acting downstream of insulin. ⁴⁵ In the setting of systemic insulin resistance, the targeted removal of FoxO1 and FoxO3 specifically in cardiomyocytes was associated with the preservation of cardiac function. ⁴⁶ Molecular insights implicated that the sustained activation of FoxO1 contributed to the elevation of pIRS1 Ser levels, ultimately resulting in the development of insulin resistance (Figure 3). Meanwhile, FoxO1‐dependent downregulation of IRS1 led to reduced AKT signalling and insulin resistance. ⁴⁶ The insulin‐stimulated PI3K/AKT signalling pathway regulates the glucose transporter type 4 (GLUT4), which is essential for glucose uptake and metabolism. ⁴⁷ As said before, the binding of insulin to the sarcolemmal IR triggers the activation of the PI3K/AKT signalling pathways. This activation facilitated the uptake of glucose by myocardial tissue through GLUT4 located at the plasma membrane, thereby playing a critical role in sustaining cardiac energy supply ⁴⁸ , ⁴⁹ (Figure 3). In contrast, decreased GLUT4 protein expression and translocation to the cell surface in cardiomyocytes was considered to be among the pathogenic mechanisms of DCM. ⁴⁹ These observations collectively suggest the involvement of AKT in the regulation of insulin sensitivity. Carvacrol, a monoterpenic phenol isolated from various mints, exhibited notable effects on the phosphorylation of PI3K, PDK1 and AKT, while concurrently reducing PTEN phosphorylation, thereby restoring the PI3K/AKT signalling pathway ⁵⁰ (Figure 3). Of particular significance, carvacrol effectively enhanced GLUT4 translocation to the cell membrane of cardiomyocytes, exerting an anti‐DCM effect ⁵⁰ (Figure 3). Additionally, Chinese herbal Rhynchophylline (Rhy) has the anti‐DCM effect as well. Molecular insights implicated that Rhy can inhibit the overactivation of the sarcoplasmic reticulum Ryanodine receptor 2 (RyR2), which was known to cause increased Ca²⁺ leakage. This inhibition was primarily achieved by antagonizing RyR2 phosphorylation thereby regulating calcium homeostasis ⁵¹ (Figure 3). As mentioned previously, CaMKII Ser280 can be activated by O‐GlcNAcylation to release Ca²⁺ and, in addition, CaMKII can be activated by phosphorylation ⁵² (Figure 3). There is also cross‐talking between phosphorylation and ubiquitination. For instance, Y‐box binding protein‐1 (YB‐1), a member of the highly conserved cold shock domain protein family, exacerbated DCM when its total protein levels were significantly reduced, accompanied by elevated protein phosphorylation levels. ⁵³ The stability of YB‐1 was regulated by phosphorylation, which facilitated its degradation via otubain‐1 (OTUB1)‐dependent ubiquitination. This mechanism was mediated by the upstream ERK/RSK signalling pathway, and inhibition of the ERK pathway can ameliorate DCM. ⁵³ In relation to myocardial fibrosis, it was hypothesized that Sal B, the active components of Salvia miltiorrhiza, could enhance cardiac function in mice with DCM by attenuating hyperglycemia‐induced myocardial remodelling and myocardial fibrosis. ⁵⁴ , ⁵⁵ Mechanistically, Sal B was found to augment the phosphorylation of AKT and ERK by inhibiting the expression of insulin‐like growth factor‐binding protein 3 (IGFBP3), activating these signalling pathway and thus promoting angiogenesis. ⁵⁵

FIGURE 3.

Pathogenesis associated with phosphorylation and diabetic cardiomyopathy. Under physiological conditions, upon insulin binding to the sarcolemmal insulin receptor, activation of PI3K/AKT signalling occurs upstream. This activation leads to glucose uptake by myocardial tissue through GLUT4 at the plasma membrane. Sustained activation of FoxO1 increased pIRS1 Ser levels, ultimately leading to insulin resistance. While captopril can restore insulin signalling and increase AKT phosphorylation, then normalizing substrate utilization. Carvacrol can promote the phosphorylation of PI3K and AKT, and enhance GLUT4 translocation to the cell membrane of cardiomyocytes as well. In addition, rhynchophylline can antagonize RyR2 phosphorylation to inhibit Ca²⁺ leakage.

In brief, the regulation of glucose metabolism is significantly influenced by the involvement of PI3K/AKT. The diminished phosphorylation and activation of PI3K and AKT hinder the insulin‐mediated glucose uptake, exerting a substantial effect on insulin actions within the cardiac system. Site‐specific targeted activation of AKT phosphorylation may play a role in alleviating insulin resistance and potentially reversing DCM. Given that phosphorylation is regulated by the competing activities of protein kinases and phosphatases, ⁵⁶ a considerable array of pharmaceutical agents drugs targeting protein kinases/phosphatases has been devised. The ROCK kinases (Rho‐associated coiled‐coil containing kinases), which belong to the serine/threonine protein kinase family, exert a crucial influence on regulating the actin cytoskeleton to influence cell motility and regulate vascular tone. ⁵⁷ Fasudil, a ROCK inhibitor, has demonstrated efficacy in ameliorating symptoms associated with cardiovascular conditions, including hypertension, angina pectoris and ischemic stroke, among others. ⁵⁸ In addition, CP‐91149, an inhibitor of glycogen phosphorylase, has been studied in type 2 diabetes to increase glucose availability and meet high energy demands. ⁵⁹

4. PROTEIN METHYLATION IN DIABETIC CARDIOMYOPATHY

As an important epigenetic regulation, methylation refers to the transfer of active methyl groups into target chemicals, catalysed by methyltransferases, to form methylation products. The process is reversible and involves two enzymes, methyltransferase and demethylase. Methylation includes DNA methylation, RNA methylation and protein methylation (subdivided into histone and non‐histone methylation). ⁶⁰ , ⁶¹ , ⁶² , ⁶³ In cardiovascular disease, the role of methylation has been recognized and relevant preclinical studies and drugs targeting DNA methyltransferases and histone methyltransferases have emerged. ⁶⁴ In this review, we mainly discuss histone methylation, which usually occurs at the arginine and lysine residues of histone 3 (H3) and histone 4 (H4) and affects the transcriptional activity of related genes.

Insulin resistance and metabolic disorders are fundamental parts in the development of DCM. Decreased EHMT2 (euchromatic histone lysine methyltransferase 2) in the liver of leptin receptor gene‐deficient mice can downregulate HMGA1 gene expression, impairing IR transcription. ⁶⁵ It has been reported to cause insulin resistance in the liver, which in turn led to systemic insulin resistance (Figure 4A). Classical nuclear factor kappa‐B (NF‐κB) plays a crucial role in the immune response, and activation of NF‐κB‐p65 is also a feature of DCM. Under high glucose induction, the recruitment of the methyltransferase Set7 led to an increase in H3K4 methylation, which resulted in a sustained increase in NF‐κB‐p65 gene expression (Figure 4A). And this process may be caused by the release of ROS from exposure to high glucose. ⁶⁶ , ⁶⁷ Moreover, in a diabetic mouse model, renal failure increased dimethylation of cardiac histone H3K4, which promoted cardiomyocyte hypertrophy and exacerbated DCM. ⁶⁸ In DCM, the mechanisms underlying histone methylation lack specificity, and targeted intervention studies are currently lacking. However, several studies have demonstrated that modulating histone methylation modifications in adipocytes can effectively inhibit lipid storage capacity. ⁶⁹ , ⁷⁰ , ⁷¹ Lipid deposition can damage myocardial calmodulin, which in turn affects the diastolic and systolic functions of cardiomyocytes. ⁷² Exploring the potential impact of such interventions in cardiomyocytes may offer a therapeutic approach to ameliorate insulin resistance and metabolic disorders. Intervening in the response to histone methylation‐induced inflammation is also one of the strategies. ⁷³ Galangin can ameliorate DCM by reducing oxidative stress and inflammation. ⁷⁴ , ⁷⁵ Hao et al. proposed that DNA methylation and histone modifications may play an important role in microvascular complications of DCM and thereby represent a potential therapeutic target. ⁷⁶

FIGURE 4.

Pathogenesis associated with methylation and acetylation and diabetic cardiomyopathy. (A) EHMT2 mediated histone methylation can cause insulin resistance. The recruitment of the methyltransferase Set7 led to an increase in H3K4 methylation, which resulted in a sustained increase in NF‐κB‐p65 gene expression. (B) Resveratrol could activate SIRT1 and alleviate endoplasmic reticulum stress‐mediated apoptosis; resveratrol could ameliorate oxidative stress through deacetylation of NF‐κB‐p65 and histone 3; resveratrol could upregulate SERCA2a expression.

5. PROTEIN ACETYLATION IN DIABETIC CARDIOMYOPATHY

Acetylation is the process of transferring an acetyl group to an amino acid side‐chain group, with histone acetylation being the most prevalent form. ⁷⁷ Histone acetylation is mediated by coactivator complexes containing histone acetyltransferases (HATs), while deacetylation is mediated by co‐repressor complexes containing HDACs. ⁷⁷ In addition to this, acetylation can also occur at the non‐histone or N‐terminal regions of proteins. Histone acetyltransferases can be classified into four major groups based on their structure and properties, namely CBP/p300, GCN5, MYST and SRC/p160, while the NAD⁺‐dependent sirtuin (SIRT) family represents the most prevalent deacetylase. ⁷⁸ Extensive research has demonstrated the crucial role of acetylation and deacetylation of functional proteins in cardiomyocyte differentiation, cardiac remodelling and various cardiovascular disorders, such as DCM.

The reduced NAD+/NADH ratio and consequent inactivation of sirtuins have been identified as contributing factor to the decreased cardiac metabolic rate observed in individuals with diabetes. ⁷⁹ Numerous studies have been conducted on HDACs, revealing the regulatory roles of SIRT1 and SIRT3 in DCM through their influence on cardiomyocyte metabolism. Notably, SIRT3 overexpression attenuated hyperglycemia‐induced glycolysis and oxidative stress damage by inhibiting p53 acetylation and TP53‐induced expression of glycolytic and apoptotic regulators. ⁸⁰ Conversely, the impact of SIRT1 on cardiac function appeared to be contingent upon the dosage administered. Specifically, high doses of SIRT1 induced cardiomyopathy, whereas mild to moderate levels of SIRT1 expression did not exhibit this adverse effect. ⁸¹ Furthermore, even though high expression of SIRT1 increased oxidative stress, moderate SIRT1 induced resistance to oxidative stress and apoptosis. Resveratrol could activate SIRT1 and alleviate endoplasmic reticulum stress‐mediated apoptosis in DCM (Figure 4B), and PERK/eIF2α, ATF6/CHOP and IRE1α/JNK signalling pathways may be involved in this process. ⁸² , ⁸³ According to Bagul et al., in the presence of resveratrol, SIRT1 ameliorated oxidative stress in DCM through deacetylation of NF‐κB‐p65 and histone 3 ⁸⁴ (Figure 4B). Sulaiman et al. suggested that resveratrol adjusted calcium homeostasis to improve cardiac function by activating SIRT1 and upregulating SERCA2a expression ⁸⁵ (Figure 4B). Therefore, SIRT1 is considered as a potential target for the treatment of cardiovascular diseases, especially DCM. ⁸⁶ Resveratrol can activate SIRT2, SIRT3 and SIRT5 as well, which played an indispensable role in diabetic cardiomyopathy. ⁸⁷ Transforming growth factor beta (TGF‐β) is a pro‐sclerotic cytokine that is associated with cardiac fibrosis and hypertrophy. ⁸⁸ Under high glucose conditions, p300 activity was increased, while p300 enhanced TGF‐β activity via Smad2 acetylation, which was involved in fibrotic interstitium. ⁸⁹ Hyperglycemia‐induced advanced glycation end products (AGEs) impaired Na⁺‐K⁺‐ATPase activity by decreasing AMPK and SIRT1, leading to contractile dysfunction of the heart. ⁹⁰ , ⁹¹ Meanwhile, further SIRT1‐downregulation enhanced endothelial cell death, one of the main factors of the increased endothelial permeability in DM. ⁹² It was mentioned in the section on phosphorylation modifications that reduced glucose uptake in the hearts of mice with DCM was associated with decreased expression of GLUT1 and GLUT4, as well as a decrease in insulin‐stimulated GLUT4 translocation to the sarcolemma. ⁴⁹ HDAC inhibition can upregulate the GLUT1 and GLUT4 expression, accompanied by increased GLUT1 acetylation and p38 phosphorylation. ⁹³ It can not only reduce cardiac hypertrophy but also increase angiogenesis, ultimately improving cardiac function. HDAC knockout diabetic mice developed HF. However, the re‐expression of the HDAC4 N‐terminal fragment prevented HDAC4‐dependent DCM. The production of the HDAC4 N‐terminal fragment relied primarily on dependent on PTMs of HDAC4 by O‐GlcNAcylization at serine (Ser)‐642 to counteract pathological CaMKII signalling. ³⁸

Currently, there are many studies on acetylation affecting the mitochondrial NAD⁺/NADH redox state. Berthiaume et al. proposed a method based on alternate mitochondrial electron transport that normalizes the mitochondrial redox state and improves DCM. ⁹⁴ Dietary NAD+ supplementation or pharmacological activation of sirtuins may restore the NAD+/NADH ratio, improving mitochondrial function and cardiac performance. ⁹⁵ , ⁹⁶ In addition, exercise has been shown to have pleiotropic benefits on cardiometabolic homeostasis as a strategy for non‐pharmacological treatment. Treadmill exercise significantly attenuated diabetes‐induced cardiac insufficiency, accompanied by reduced mitochondrial damage and increased cardiac mitochondrial enzyme activity. Molecular insights implicated that the exercise response factor, fibroblast growth factor 21 (FGF21), maintained normal mitochondrial function by inducing the AMPK/FoxO3/SIRT3 signalling axis, thereby reversing diabetes‐induced hyperacetylation and dysfunction of the mitochondrial enzyme cluster. ⁹⁷

6. PROTEIN UBIQUITINATION IN DIABETIC CARDIOMYOPATHY

Ubiquitin is a low molecular weight protein with 76 amino acid residues that can be tagged on the surface of a specific protein substrate by the action of enzymes and then recognized and degraded by organelles or multi‐enzyme complexes. ⁹⁸ The ubiquitin proteasome pathway refers to the degradation of ubiquitinated target proteins by the 26S proteasome in an ATP‐dependent manner. ⁹⁹ The process is specifically modified by a cascade reaction of three enzymes, including E1 ubiquitin activating enzyme, E2 ubiquitin binding enzyme and E3 ubiquitin ligase. ¹⁰⁰ There are more than 600 types of E3 ubiquitin ligases that have been identified. The classical E3 ubiquitin ligases are divided into three main types, RING (Really Interesting New Gene) type, HECT (Homologous to E6AP C‐terminus) type and RBR (RING‐between‐RING) type. ¹⁰¹ The specific recognition of substrates by E3 ubiquitin ligases makes their role in the ubiquitin pathway particularly important and much more studied.

Several E3 ligases are involved in the development of insulin resistance through different pathways. One mechanism is the direct degradation of IRs (especially hyperglycemia‐induced) and related substrates via the ubiquitin proteasome pathway. ¹⁰² For example, IR substrate (IRS) can exert biological effects by tyrosine phosphorylation in response to insulin and insulin‐like growth factor (Figure 5). Mice lacking IRS1 or IRS2 exhibited peripheral insulin resistance. ¹⁰³ Inhibition of E3 ubiquitin ligase MG53 (also known as tripartite motif‐containing 72, TRIM72) mediated ubiquitination degradation of IRS‐1 contributed to amelioration of insulin resistance ¹⁰⁴ (Figure 5). E3 ubiquitin ligase, Cullin7, regulated insulin signalling through the mTOR‐S6K‐IRS1 signalling axis, as evidenced by enhanced insulin sensitivity in Cullin7^+/− or Fbxw8^+/− mice. ¹⁰⁵ Fbxw8 is an essential component of Cullin7 and is responsible for substrate specific recognition. ¹⁰⁶ Insulin resistance is always presented with lipid overload, causing mitochondrial dysfunction and myocardial toxicity. ¹⁰⁷ Lipid overload increased AKAP121 (A kinase anchoring protein 121, a mitochondrial outer membrane scaffold protein) ubiquitination, modulated DRP1 (dynamin related protein 1, mitochondrial dynamic regulation protein) phosphorylation and altered OPA1 (optic atrophy 1, mitochondrial dynamic regulation protein) processing, eventually resulting in impaired mitochondrial energetics. ¹⁰⁷ , ¹⁰⁸ , ¹⁰⁹ , ¹¹⁰ This process may be mediated by ROS production. In addition to regulating metabolism through target protein degradation via the ubiquitin proteasome pathway, E3 ubiquitin ligases are involved in the regulation of gene transcription. ¹¹¹ These transcription factors include PPARα (peroxisome proliferators‐activated receptor α), FOX‐O, Nrf2 (nuclear factor derived‐2) and NF‐κB. ¹¹² , ¹¹³ PPARα was the first transcription factor identified to be associated with DCM. ¹¹⁴ In mice with cardiac specific PPARα overexpression, expression of target genes involved in cardiac fatty acid uptake and oxidation pathways was increased, while expression of genes involved in glucose transport and utilization was decreased. ¹¹⁴ Consistently, these mice had increased rates of myocardial fatty acid oxidation and decreased glucose uptake and oxidation, similar to the metabolic phenotype of the DCM. ¹¹⁴ MG53 can positively upregulate the expression level of PPARα, which triggers a cascade of events leading to DCM (Figure 5). E3 ubiquitin ligases are also involved in the pathogenesis of cardiac hypertrophy and myocardial fibrosis. Atrogin‐1 (also named Muscle‐atrophy F‐box), an F‐box protein with skeletal and cardiac muscle specificity, is a component of the SCF family of E3 ubiquitin ligases. ¹¹⁵ Atrogin‐1 promoted ubiquitination of calcineurin, thereby inhibiting cardiomyocyte hypertrophy ¹¹⁶ (Figure 5). According to Ye et al., the deubiquitinating enzyme ubiquitin‐specific peptidase 25 (USP25) can combined with SERCA2a and prevent proteasomal pathway from degrading it. ¹¹⁷ This process not only maintained calcium homeostasis in cardiomyocytes but also contributed to the inhibition of cardiac hypertrophy. Moreover, increased E3 ubiquitin ligase WWP2 expression was related with an elevated pro‐fibrotic gene expression program in diseased heart. WWP2, particularly the WWP2‐N isoform containing the N‐terminal C2 and WW1 domains of WWP2, regulated cardiac fibrosis by modulating SMAD2 signalling ¹¹⁸ (Figure 5).

FIGURE 5.

Pathogenesis associated with ubiquitination and diabetic cardiomyopathy. MG53 can upregulate PPARα expression, decreasing genes involved in glucose transport and utilization. While inhibition of MG53 mediated ubiquitination degradation of IRS‐1 contributed to amelioration of insulin resistance. Atrogin‐1 promoted ubiquitination of calcineurin, thereby inhibiting cardiomyocyte hypertrophy. WWP2 can regulate cardiac fibrosis by modulating SMAD2 signalling.

Currently, therapeutic approaches targeting the ubiquitin proteasome or E3 ubiquitin ligase are advancing, and targeting E3 ubiquitin ligase is considered to offer higher specificity and less toxicity compared to E1 and E2. As a proteasome inhibitor, MG132 can reverse pathogenic cardiac manifestations, including myocardial hypertrophy, fibrosis and reduced LV ejection fraction. ¹¹³ Molecular insights suggest that MG132 increased expression of Nurf2 and downstream antioxidant gene, while decreased nuclear accumulation of cardiac NF‐κB and its DNA binding activity. ¹¹² The small molecule MyoMed‐205 can target MuRF1 and improve the diastolic dysfunction of the myocardium. ¹¹⁹ Alongside agonists and inhibitors, PROTAC (proteolysis targeting chimaera) technology has garnered considerable attention and hold promise for the development of more specific therapeutic regimens.

7. DISCUSSION

In summary, the widespread and dynamic nature of PTMs implicates their involvement in various aspects of the pathogenesis of DCM, with different types of PTMs interacting to coordinate the intricate pathophysiological functions. PTMs at different sites of the same protein may have different effects on the disease. Meanwhile, because of the cross‐talk between post‐translational modifications, it is difficult to target a single specific modification to achieve a better therapeutic effect. This highlights the importance of clear mechanisms. We found that in DCM, insulin resistance and Ca²⁺ handling disorders seem to be more involved or more studied (Table 1). For one thing, insulin resistance and metabolic disorders play a central role in DCM development. Insulin resistance in the heart disrupts insulin signalling‐mediated substrate utilization, increasing the likelihood of cardiac insufficiency. ¹²⁰ Diabetic patients exhibit increased cardiac fatty acid utilization, changed glucose uptake and oxidation and increased fructose content, which may lead to mitochondrial structural remodelling and accelerate O‐GlcNAylation and AGEs formation. ¹⁰⁷ , ¹²¹ , ¹²² , ¹²³ Besides, a number of PTMs can alter the function of the mitochondrial permeability transition pore (mPTP) opening, affect energy metabolism and indirectly participate in the pathophysiology process of DCM. ¹²⁴ Among the various PTMs, phosphorylation and acetylation are overwhelmingly dominant. ¹²⁴ Dichloroacetate has shown promise in restoring myocardial substrate selection balance by enhancing pyruvate dehydrogenase flux, thereby reversing diastolic dysfunction in DCM. ¹²⁵ HDAC inhibitors, such as chidamide, have received a lot of attention in the treatment of haematological diseases, especially peripheral T‐cell lymphoma. ¹²⁶ HDAC inhibitors are also currently thought to relieve systemic insulin resistance and glucose handling, ⁹³ , ¹²⁷ while further studies are still needed regarding the specific regulation of HDAC on the heart. Therefore, normalizing the metabolic profile in DCM by modulating PTMs may be an important target to improve diabetic cardiac function.

TABLE 1.

Involvement of post‐translational modifications in insulin resistance and impaired Ca²⁺ handling impacts diabetic cardiomyopathy.

PTMs	Related mechanism
	Insulin resistance & metabolic disorders	Impaired Ca2+ handling
O‐GlcNAcylation	BDH1 can be modified by O‐GlcNAc and then the expression was suppressed, then inhibiting ketone body utilization 19	O‐GlcNAcylation decreased SERCA2a gene expression, leading to impaired diastolic Ca2+ sequestering 22 CaMKII can be modified by O‐GlcNAc, leading to CaMKII activation and increased sarcoplasmic reticulum Ca2+ release 29
Phosphorylation	Long‐term β‐adrenergic signalling resulted in the inhibition of insulin‐stimulated glucose uptake and insulin‐induced autophosphorylation of insulin receptors 43 Captopril can increase AKT phosphorylation and decrease AMPK activation, restoring insulin sensitivity and normalizing substrate utilization 44 Sustained activation of FoxO1 increased pIRS1 Ser levels, leading to insulin resistance 46	Rhynchophylline can antagonize RyR2 phosphorylation to regulate Ca2+ homeostasis 51 CaMKII can be activated by phosphorylation, leading to increased sarcoplasmic reticulum Ca2+ release 29
Methylation	EHMT2 can cause insulin resistance in the liver, then leads to systemic insulin resistance 65	–
Acetylation	SIRT3 overexpression attenuated hyperglycemia‐induced glycolysis by inhibiting p53 acetylation 81	Re‐expression of the HDAC4 N‐terminal fragment counteract pathological CaMKII signaling 38 Resveratrol adjusted Ca2+ homeostasis by activating SIRT1 and upregulating SERCA2a expression 86 β‐lap exerts cardio‐protective effects at least partially through SIRT1‐mediated SERCA2a deacetylation 125
Ubiquitination	Inhibition of MG53 mediated ubiquitination degradation of IRS‐1 contributed to amelioration of insulin resistance 102 Cullin7 regulated insulin signal through the mTOR‐S6K‐IRS1 signalling axis, as evidenced by enhanced insulin sensitivity in Cullin7+/− or Fbxw8+/− mice 103 MG53 can upregulate the PPARα expression, with increased rates of myocardial fatty acid oxidation and decreased glucose uptake and oxidation 108	USP25 can prevent SERCA2a ubiquitination degradation, maintaining Ca2+ homeostasis 111

For another, impaired calcium handling emerges as a critical feature of DCM. In normal conditions, 80%–90% of intracellular calcium is released from the sarcoplasmic reticulum, while 10%–20% enters through L‐shaped calcium channels. ¹²⁸ The dynamic balance of sarcoplasmic reticulum calcium involves release mediated by RyR2 receptors and return mediated by SERCA ATP consumption, respectively. ¹²⁹ , ¹³⁰ Various PTMs related to calcium handling disturbance, especially O‐GlcNAcylation, contributing to prolonged diastolic SERCA2‐mediated calcium removal from the cytoplasm, increased RyR2‐mediated calcium leakage from the sarcoplasmic reticulum and alterations in myofilament calcium sensitivity. In DCM, the SERCA2a expression and activity are reduced, whereas the acetylation level is significantly increased, which is also a feature of heart failure. This alteration may be associated with increased levels of O‐GlcNAcylation and decreased levels of SIRT1. With the exception of the SIRT1 activator resveratrol, studies have shown that β‐lap (a quinone‐containing natural compound) exerts cardio‐protective effects at least partially through SIRT1‐mediated SERCA2a deacetylation. ¹³¹ , ¹³² Regarding RyR2, phosphorylation of it may contribute to DCM while causing inflammatory responses. ¹³³ Tian et al. proposed that interactions among RyR1, RyR2, Ca²⁺ signalling and immune‐related molecules may be relevant to DCM. ¹³³ Although RyR1 is expressed at relatively low levels in the heart, it appears to be a starting point in this interaction. Recent clinical studies have reported that cardiac resynchronization therapy responders exhibit a significantly reduced RyR1 glycation in peripheral lymphocytes. ¹³⁴ Thus, modulating PTMs to restore calcium homeostasis holds significant therapeutic potential for DCM.

AUTHOR CONTRIBUTIONS

Zhi Li: Writing – original draft (lead). Jie Chen: Writing – original draft (lead). Hailong Huang: Writing – original draft (supporting). Qianru Zhan: Writing – original draft (supporting). Fengzhi Wang: Writing – original draft (supporting). Zihan Chen: Writing – review and editing (equal). Xinwei Lu: Writing – review and editing (equal). Guozhe Sun: Project administration (lead).

FUNDING INFORMATION

The study is supported by the Natural Science Foundation of Liaoning Province of China (2022‐MS‐213).

CONFLICT OF INTEREST STATEMENT

The authors declare no competing interests.

Li Z, Chen J, Huang H, et al. Post‐translational modifications in diabetic cardiomyopathy. J Cell Mol Med. 2024;28:e18158. doi: 10.1111/jcmm.18158