Exploring the key role of DNA methylation as an epigenetic modulator in oxidative stress related islet cell injury in patients with type 2 diabetes mellitus: a review

Abstract

Type 2 diabetes mellitus (T2DM) is a multifactorial metabolic disorder characterised by impaired insulin secretion and action, often exacerbated by oxidative stress. Recent research has highlighted the intricate involvement of epigenetic mechanisms, particularly DNA methylation, in the pathogenesis of T2DM. This review aims to elucidate the role of DNA methylation as an epigenetic modifier in oxidative stress-mediated beta cell dysfunction, a key component of T2DM pathophysiology. Oxidative stress, arising from an imbalance between reactive oxygen species (ROS) production and antioxidant defence mechanisms, is a hallmark feature of T2DM. Beta cells, responsible for insulin secretion, are particularly vulnerable to oxidative damage due to their low antioxidant capacity. Emerging evidence suggests that oxidative stress can induce aberrant DNA methylation patterns in beta cells, leading to altered gene expression profiles associated with insulin secretion and cell survival. Furthermore, studies have identified specific genes involved in beta cell function and survival that undergo DNA methylation changes in response to oxidative stress in T2DM. These epigenetic modifications can perpetuate beta cell dysfunction by dysregulating key pathways essential for insulin secretion, such as the insulin signalling cascade and mitochondrial function. Understanding the interplay between DNA methylation, oxidative stress, and beta cell dysfunction holds promise for developing novel therapeutic strategies for T2DM. Targeting aberrant DNA methylation patterns may offer new avenues for restoring beta cell function and improving glycemic control in patients with T2DM. However, further research is needed to elucidate the complex mechanisms underlying epigenetic regulation in T2DM and to translate these findings into clinical interventions.

Introduction

The prevalence of type 2 diabetes mellitus (T2DM) is rising globally [1–3]. Currently, no pharmaceutical treatment exists to permanently cure, prevent, or halt the progression of diabetes [4]. The fundamental problem with existing diabetes medications is that they focus on treating the symptoms rather than the underlying cause of diabetes. Maintaining and restoring surviving pancreatic beta cells is the biggest challenge in treating diabetes, irrespective of the type of diabetes. The growing epidemic of type 2 diabetes cannot be controlled by hypoglycemia medication alone [5]. Therefore, a greater comprehension of the underlying molecular mechanisms of diabetes is crucial in order to create improved therapeutics, especially given the disease’s fast rising prevalence.

Genetic testing has not been found to be a reliable means of predicting a patient’s clinical risk or the pathological implications that may result from their diabetes, even though genome-wide association studies (GWAS) have identified multiple genes connected to T2DM susceptibility [6, 7]. The etiology and progression of type 2 diabetes, however, may be significantly influenced by epigenetic processes, such as DNA methylation, post-translational histone modification, and non-coding RNAs, which may act as a crucial mediator between genetic predisposition and environmental factors [8–10]. These factors alter the degree of gene expression, either alone or in combination [11]. The main environmental factors that contribute to the development of T2DM are low birth weight, obesity, oxidative stress, advanced age, and an unconducive in utero environment [12].

Oxidative stress has a major role in the regulation of epigenetic regulatory systems and is a pathophysiological factor for diabetes [13]. Aberrant DNA methylation patterns can result from alterations in DNA protein and protein-protein interactions brought on by oxidative stress-induced DNA strand breakage and damage [14]. Guanine oxidation within CpG islands results in 8-hydroxy-2’-deoxyguanosine (8-OHdG), which decreases the affinity of methyl-domain binding proteins (MBPs) and increases the methylation of neighbouring cytosine in DNA. In pancreas, islet cell formation, viability, and function are all dependent on this epigenetic control. On the other hand, oxidative stress disrupts such epigenetic markings in a diabetic milieu, leading to islet cell death and malfunction. Therefore, this review sheds light on the role of DNA methylation as an epigenetic modifier in oxidative stress mediated islet cell dysfunction in patients of T2DM.

Epigenetics

A heritable characteristic resulting from changes in gene expression that do not change the underlying DNA sequence is known as an epigenetic trait [15]. It has been suggested that there are three types of signals that induce an epigenetic state: epigenators, epigenetic initiators, and epigenetic maintainers (Fig. 1). An intracellular pathway is activated by the external stimulus that produces the epigenator signal. The activation of the latent epigenetic initiator might be released by a modification-based event or a protein-protein interaction. Temperature is one instance of an epigenator signal in plants. The intracellular epigenetic initiator signal receives and translates the extracellular epigenator signal by starting the necessary processes to establish the signaled chromatin state at the chromatin spot. Non-coding RNAs, DNA-binding proteins, and/or any other entity that can recognize DNA sequences and specify the desired chromatin state can act as the initiator signal. The chromatin state is preserved throughout successive generations by an epigenetic maintainer signal, however not to the extent necessary to initiate it. Throughout the cell cycle and in terminally differentiated cells, this signal maintainer maintains the epigenetic profile.These traits are reversible modifications passed down vertically through both mitosis and meiosis. However, they can also occur sporadically as a result of numerous environmental variables or gene related changes, such as those affecting the enzymes that catalyse these epigenetic modifications [16, 17]. Examples include methylation of DNA, alterations to the posttranslational histone tails such as acetylation, phosphorylation, and methylation, and regulatory RNAs such as the non-coding RNAs shown in Fig. 2(C). [18].

Fig. 1.

The three signal categories and the epigenetic process. The start of the epigenetic process is caused by the extracellular epigenator signal. After receiving the signal from the epigenator, the epigenetic initiator establishes the chromatin state to start the epigenetic pathway, which is maintained by the epigenetic maintainer over the course of multiple generations

Fig. 2.

A concise synopsis of the major epigenetic alterations. (A) Nucleosomes are formed when chromosomal DNA bundles up with histone proteins. Acetylation and methylation are examples of post translational changes that control nucleosome opening and accessibility to nuclear factors. Histone acetyl transferase (HAT) masks the positive charges of amino acid side chains during the acetylation step, reducing DNA packing and enabling transcription. Conversely, transcription is inhibited by deacetylation processes caused by histone deacetylase (HDAC), which enhances chromatin packing. (B) DNA methylation results in the insertion of a methyl group by DNA methyl transferase (DNMT) enzymes at the fifth carbon position of DNA cytosine, causing gene silence. (C) Non-coding RNAs, such as siRNA, miRNA, LncRNA, CircRNA, and others, are expressed as housekeeping molecules and function as the main post-transcriptional regulators of gene expression

Usually, chromatin structure is altered by the interaction and modulation of individual epigenetic changes, resulting in an epigenetic profile that alters gene expression and the functioning of the genome. The particular epigenetic profile can either activate or mute genes by packing the chromatin structure tightly and blocking access to the DNA, or by loosening the structure and letting the transcription machinery to access the DNA [17]. Consequently, it has been demonstrated that the epigenetic profile modifies gene expression in a variety of cell types, developmental phases, in both healthy and diseased states [19].

The key epigenetic alterations that are critical to the expression of genes are histone modifications such as acetylation, deacetylation, DNA methylation, and non-coding RNAs like miRNA, circRNA, lncRNA, and siRNA. These modifications are depicted in Fig. 2 below.

DNA methylation

DNA methylation, which comprises covalent attachment of a methyl group (-CH3) to the 5’ position of the cytosine residue in the dinucleotide 5’-cytosine-phosphodiester bond guanine-3’ (5’-CpG-3’) resulting in 5-methylcytosine (5mC), occurs naturally during DNA replication. This CpG DNA methylation is found to be reversible [17, 20]. Although they can also be found in enhancer and intragenic regions, CpG islands are often found in the promoter region of genes [16, 17]. Their CpG density is high (> 50%). While the majority of CpG dinucleotides in humans are methylated, the promoters of somatic cells and germ-line cells have comparatively unmethylated CpG dinucleotides [20].

Typically, transcriptional activation is correlated with promoter CpG island hypomethylation, and transcriptional repression is correlated with hypermethylation. Yet, gene expression control by DNA methylation is even more intricate than simple promoter methylation since methylation of CpG sites in regulatory regions other than the gene promoter also controls gene expression in a tissue-specific way. DNA methylation-mediated gene silencing is essential for several biological processes, including cellular differentiation, genomic imprinting, X chromosome inactivation, and retrotransposon silencing.

There are currently two schools of thoughts explaining how DNA methylation prevents the production of certain genes. According to the first mechanism, transcription factors are prevented from binding to cytosine in the primary DNA groove by methylation. According to the second mechanism, certain histone modifiers, including histone deacetylases, are attracted to the 5mC by methylation, and this modification of the chromatin state leads to the formation of a compact chromatin structure.

The enzyme family of DNA methyltransferases (DNMTs) facilitates the methylation of CpG. Mammals DNMTs generate the 5mC of the CpG by using S-adenosyl methionine (SAM) as the methyl group donor. Four types of DNMTs are found in humans: DNMT1, DNMT3A, DNMT3B, and DNMT3L. DNMT1 is responsible for preserving DNA methylation patterns during the mitotic process.

By replicating the parental DNA strands’ pre-existing DNA methylation profile onto the daughter DNA strands, DNMT1 methylates hemi-methylated CpGs. They are categorized as de novo DNMTs, DNMT3A and DNMT3B. They recognize CpGs that are not methylated and respond to various stimuli, such oxidative stress, by being methylated, creating unique DNA methylation patterns. Therefore, during the early stages of embryonic development, DNMT3A and DNMT3B are in charge of establishing new DNA methylation profiles. Germ cell-resident DNMT3L enhances methyl donor binding and supports de novo methylation [17, 20], and [21].

DNA demethylation (Fig. 3) occurs actively through the use of ten-eleven translocation (TET) enzymes to oxidize the methylated base or by activation-induced deaminase (AID) to deaminate the methylated or adjacent base. Both of these methods are independent of DNA replication. Next, the mutant nucleotide and the surrounding nucleotides are repaired via base excision repair (BER). By weakening DNA methylation signals during replication, DNA demethylation occurs passively. Nevertheless, there is no direct conversion of 5 mC into cytosine [22].

Fig. 3.

The methylation and demethylation mechanism of cytosine. The activity of DNA methyltransferases (DNMTs) produces 5methyl cytosine (5mC)). Under certain conditions, maybe prompted by DNA-binding transcription factors specific to a given tissue, TET enzymes oxidize specific 5 mC residues to make 5hmC (5-hydroxymethyl cytosine), which can then be further oxidized to form 5 fC (5-formyl cytosine) and 5caC (5-carboxy cytosine). Alternatively, 5fC and 5caC can be converted to cytosine via the base excision repair mechanism (BER) and removed by thymidine DNA glycosylase (TDG). [23]

DNA methylation in T2DM

It is believed that only around 10% of the estimated heritability of T2DM is explained by any of the genetic risk factors for the disease that GWASs have found thus far. In an effort to explain part, if not all, of the missing heritability, researchers have therefore been looking into the role of epigenetics in T2DM [17]. According to recent studies, the expression of T2DM genetic risk factors is influenced by T2DM environmental risk factors, and this in turn affects the particular intracellular signaling pathways that are implicated in the onset and progression of T2DM. The subsequent section focusses on the current understanding of the role of DNA methylation in the development and pathophysiology of type 2 diabetes, although the precise mechanisms by which DNA methylation contributes to T2DM pathogenesis yet lack clarity.

DNA methylation and islet cell dysfunction

Insulin

Pancreatic β cells release insulin, a peptide hormone, in response to the absorption of nutrients. By promoting glucose absorption and metabolism in skeletal and adipose tissues, it controls blood glucose levels [24]. Insulin promotes the translocation of the intracellular glucose transporter GLUT4 to the cell membrane during this process. By increasing the rate of glycogen synthesis in the liver, skeletal muscle, and adipose tissue, insulin also controls blood glucose levels by reducing the rate of glycogen breakdown in these tissues and blocking hepatic gluconeogenesis and glycogenolysis [25].

The insulin promoter has several long-range cis-acting regulatory elements. Numerous widely distributed and tissue-specific transcription factors have been demonstrated to bind to these sites. A variety of transcription factors can be bound by the cyclic adenosine monophosphate (cAMP) responsive element (CRE) which are one of the regulatory elements [26]. In the insulin promoter, rodents have one CRE site and humans have four, with CRE2 being the only CRE that is shared by both species.

Two transcription factors that are related with CRE are activating transcription factor-2 (ATF-2) and cAMP responsive element-binding protein-1 (CREB-1). On the one hand, ATF-2 stimulates transcription, while CREB-1 inhibits it. Remarkably, the ATF-2 action was suppressed by CRE2 mutations rather than CREB-1 mutations [26]. Additionally, the CRE site is essential for the regulation of the insulin gene [16]. The areas upstream of the transcription start site (TSS) of the mouse Ins2 and human insulin (INS) promoters have been found to include three and nine CpG sites, respectively. Subsequent examination has demonstrated that the CpG sites found in the promoters of the mouse Ins2 and human INS were distinctly unmethylated in β cells when contrasted with other tissues. It has been demonstrated that the methylation of these CpG sites suppresses the production of the insulin-promoter-driven reporter gene by over 90% using the NIT-1 mouse insulinoma cell line [27]. Individual CpG sites were methylated in the mouse Ins2 promoter and investigated in order to evaluate the impact of individual methylation events on insulin gene expression.

It has been demonstrated that the methylation of the CpG site inside CRE independently suppresses the activity of the insulin promoter by around 50%. Furthermore, methylation of one of the other two CpG sites did not affect the production of the insulin gene, while methylation of the other CpG site nearly doubled the expression of the insulin gene. These results reveal that other mechanisms likely work in harmony with DNA methylation to reduce the expression of the insulin gene, as methylation-dependent inhibition of the insulin promoter is not merely additive. Additional investigation revealed that methylation of CRE CpG prevents CREB from binding. According to these results, methyl CpG binding protein 2 (MeCP2) binds to the methylation promoter to prevent insulin transcriptional activators from binding, which lowers the production of the insulin gene [27]. The insulin gene is totally methylated, according to a recent in vitro analysis of Ins2 in mouse embryonic stem cell cultures; however, as the cells transform into β cells that produce insulin, the gene demethylates. All of these results point to the importance of demethylation of the insulin promoter CpG sites in β cell development and tissue-specific insulin gene expression [27].

Using human islets from donors with and without diabetes, the function of DNA methylation was investigated in further detail in a different study. According to the results, the pancreatic islets of T2DM donors had lower levels of insulin, insulin mRNA, and glucose-stimulated insulin secretion (GSIS) than those of non-T2DM donors. The T2DM pancreatic islets also displayed elevated DNA methylation at four CpG loci (− 234, − 180, −102, and + 63). Moreover, three of these CpG sites (− 234, − 180, and + 63) showed a negative correlation between the expression of insulin mRNA and the level of methylation. Next, using clonal rat insulinoma-derived INS 832/13 β cells, researchers examined the impact of hyperglycemia on the DNA methylation of the insulin promoter.

β cells showed increased DNA methylation at two CpG sites in the insulin gene promoter after being cultivated under high glucose conditions for 3 days [28]. Using Zucker diabetic fatty rats and rat pancreatic insulin-producing β cell line (INS-1 cells) cultured under normal or increased glucose concentrations for 14 days, the effects of excess nutrition on the status of DNA methylation of the rat insulin-1 (Ins1) gene promoter were assessed. The researchers discovered that DNA methyl transferase 1 mRNA expression levels and activity were elevated by the high glucose concentration. As a result, Ins1 mRNA expression was suppressed and the five CpG sites within the Ins1 promoter, including the CRE, had enhanced DNA methylation [28].

Also, pancreatic islets extracted from Zucker diabetic fatty rats showed enhanced Ins1 promoter methylation. Moreover, the artificial methylation of the Ins1 promoter dramatically reduced the activity of the reporter gene (luciferase) regulated by the insulin promoter. A DNA methylation inhibitor markedly enhanced Ins1 mRNA suppression by high glucose concentration. Furthermore, it was discovered that the high glucose experimental circumstances dramatically increased DNA methyltransferase activity and decreased TET activity. It has been demonstrated that the first line of treatment for type 2 diabetes, metformin, dramatically suppresses insulin promoter DNA methylation and upregulates the production of Ins1 mRNA [29].

Aristaless-related homeobox

The aristaless-related homeobox (ARX) gene is an X linked transcription factor that controls the development of the pancreas and other tissues in both humans and animals. It has been discovered that Arx is expressed in the mouse pancreas at every stage of embryogenesis [30]. Therefore, Arx is first shown in the pancreatic anlage, then the pancreatic endocrine cells’ developing progenitors, and lastly the resulting α cells [23]. In animals with Arx deficiency, mature endocrine α cells are lost, whereas the number of δ and β cells increases concurrently [19] to preserve the mass of the pancreatic endocrine part overall [31]. The absence of α cells in humans with ARX-null mutations highlights the critical function of ARX in the specification and differentiation of α cells in mammals.

To assess the methylation status of the CpG sites in α and β cells, respectively, a study was conducted utilizing a mouse pancreatic α cell line (α-TC1) and a mouse pancreatic insulinoma β cell line (Min6). Two primary groupings of CpG-rich sites were identified in the regulatory region of the Arx gene’s lineage de termination: UR1 (CpG-rich sites near the TSS and within the proximal promoter) and UR2 (CpG-rich sites 2 Kb upstream of the TSS). The bulk of the CpG sites in the UR2 region were discovered by the authors to be methylation in the β cells but unmethylated in the α cells [32].

Additionally, using mice and the Cre/loxP system, it has been shown that β cells lacking Dnmt1 can gradually develop into α cells. This alteration apparently links with the DNA methyl transferase1 small interfering RNA (siRNA) on Min6 cells and the methylation status of the Arx UR2 region. This region is typically methylated in adult β cells, which results in silent Arx. But in mature α and mature β cells that do not have Dmnt1, the area gets hypomethylated, leading to the expression of Arx. The methylation Arx UR2 region has been found to bind to MeCP2, which draws other proteins, notably the histone H3R2 methylase Prmt6, to further restrict Arx synthesis. The etiology of T2DM and the methylation status of Arx have not been proven to be directly linked, although this study suggests otherwise [32].

Pancreatic and duodenal homeobox-1

The differentiation of all pancreatic cell lineages requires a transcription factor known as pancreatic and duodenal homeobox-1 (PDX-1). PDX-1 expression is maintained at low levels in exocrine cells, however its precise function is yet unclear. It is abundantly expressed in β cells, where it increases the transcription of the insulin gene, thereby facilitating the synthesis of insulin [33]. DNA methylation and mRNA expression levels of the PDX-1 gene plays a substantial role in the differentiation of β cells. This procedure was examined using clonal rat insulinoma-derived INS 832/13 β cell line and human pancreatic islets from T2DM and non-T2DM donors. It was shown that, in comparison to non-T2DM, pancreatic islets derived from T2DM showed reduced levels of PDX-1 mRNA expression. This decrease was assumed to be caused by the hypermethylation of 10 CpG sites in the PDX-1 distal promoter and enhancer regions.

As a result, a positive association was discovered between the expression of GSIS and insulin mRNA and the decline in PDX-1 mRNA. Furthermore, a negative association was seen between the expression of PDX-1 and the hypermethylation of CpG sites, which was supported by clonal β cells producing less reporter gene. Hyperglycemia was only associated with a large increase in DNMT1 expression in the cultivated β cell line, despite the fact that the data revealed that it was the origin of the hypermethylation in β cells from both persons and the cell line [34].

Glucagon-like peptide-1 receptor

The glucagon-like peptide-1 receptor (GLP-1R) is widely distributed in pancreatic islets [35]. GLP 1R activation plays a role in the temporary increase of GSIS in β cells. Although it cannot be directly verified in people, it is believed that continuous GLP-1R activation promotes β cell neogenesis and proliferation as well as total insulin production [36]. Reduced GLP-1R expression in pancreatic islets has been observed in studies on T2DM in humans and rats; however, at first, it was not apparent if this reduction was brought on by other mechanisms or by a change in the DNA methylation state of the GLP-1R gene.

Using pancreatic islets or beta cells from T2DM and non-T2DM human donors, the mRNA expression levels of MECP2, DNMT1, DNMT3A, and DNMT3B, as well as the DNA methylation status of 12 CpG sites (five upstream and seven downstream) related to the GLP-1R TSS, were further examined. Two of these CpG sites, at + 199 and + 205 bp from the TSS, were subsequently investigated as a single CpG unit according to the characteristics of the GLP-1R sequence [35]. This study found that the pancreatic beta cells from T2DM donors expressed less GSIS and GLP-1R than those from non-T2DM donors.

Additionally, the CpG unit indicated a little increase in the total amount of DNA methylation between the T2DM and non-T2DM beta cells. This increase was judged to be too slight to persist after multiple testing correction, and it was unable to demonstrate a statistically significant correlation with GLP-1R mRNA expression. Furthermore, DNMT3A expression in the T2DM islets was shown to be comparatively lower than in the non-T2DM beta cells. However, there was no difference in the expression levels of MECP2, DNMT1, and DNMT3B between the T2DM and non-T2DM beta cells [35].

Moreover, the GLP 1R promoter’s DNA methylation state was examined using isolated α and β cells from the pancreatic islets. The DNA methylation level of the GLP-1R CpG site at position − 376 was shown to be inversely correlated with GLP-1R expression and was significantly higher in α cells compared to β cells [35].

Peroxisome proliferator-activated receptor γ coactivator-1 α

The peroxisome proliferator-activated receptor γ coactivator-1 α (PPARGC1A) is a transcriptional coactivator of many transcription factors and nuclear receptors, such as the peroxisome proliferator activated receptor γ (PPARG or PPARγ) [37, 38]. This protein regulates a large number of transcription factors implicated in various physiological and metabolic processes, including as glucose transport, fatty acid oxidation, PPARGC1A-regulated transcription factors, glycogenolysis, and gluconeogenesis [39].

Among the tissues with high energy requirements where PPARGC1A is mostly expressed are the pancreas, liver, and skeletal muscle [37]. A study was conducted to determine whether the expression of the PPARGC1A gene is altered in T2DM pancreatic islets, and if so, whether DNA methylation (among other variables) is responsible for the alteration. Also investigated was the effect of experimentally downregulating PPARGC1A on insulin secretion in human islets using islets from organ donors with and without type 2 diabetes. The results indicate a significant decrease in PPARGC1A mRNA expression in T2DM islets compared to non-T2DM islets. A two-fold increase in DNA methylation and a tendency toward an inverse connection between PPARGC1A mRNA levels and PPARGC1A promoter methylation expression were also found out in this investigation. Additionally, the PPARGC1A promoter (four CpG sites were examined) showed that DNA methylation in the T2DM islets had increased by a factor of two.

Moreover, a positive correlation was seen between reduced GSIS and decreased expression of PPARGC1A mRNA in the T2DM islets. These findings were corroborated by the experimental downregulation of PPAR GC1A expression in human islets by siRNA, which decreased insulin mRNA expression and secretion and therefore established a link between PPARGC1A expression level and GSIS [40] (See Table 1).

Table 1.

List of Beta cell-associated genes and their status of methylation in pancreatic beta cells of human subjects suffering from T2DM:

Genes targeted	Sample size	Outcome	Reference
SREBF1	Healthy controls = 34, T2DM = 15	Increased methylation	[41]
EXOC3L2	Healthy controls = 34, T2DM = 15	Increased methylation	[42]
CDKN1A	Healthy controls = 34, T2DM = 15	Decreased methylation	[42]
PDE7B	Healthy controls = 34, T2DM = 15	Decreased methylation	[42]
SEPT9	Healthy controls = 34, T2DM = 15	Decreased methylation	[42]
SLC2A2	Healthy controls = 8, T2DM = 6	Increased methylation	[43]
MSI2	Healthy controls = 16, T2DM = 2	Decreased methylation	[44]

Oxidative stress in islet cell dysfunction

The ultimate stage of T2DM is pancreatic beta cell failure, which is brought on by the activation of multiple pathophysiological pathways including inflammation, autoimmune response, insulin resistance, hyperglycemia/glucotoxicity, lipotoxicity, and de-/trans-differentiation [45]. Many clinical and experimental investigations have demonstrated that the underlying mechanism by which these mechanisms impair the structure and function of pancreatic beta cells and result in the development of diabetes complications is non-physiological oxidative stress [46, 47]. Figure 4 summarizes the function of oxidative stress in causing different forms of death of beta cell.

Fig. 4.

Pathophysiological mechanisms initiated by oxidative stress generate various forms of islet cell death in diabetic conditions. Islet cell death brought on by oxidative stress, which comprises ferroptosis, autophagy, necroptosis, and apoptosis, was the consequence of pathophysiological processes that included glucolipotoxicity, insulin resistance, inflammation, autoimmune reaction, and hyperglycemia

Oxidative stress in beta cells contributes to the downregulation of important insulin expression regulators as well as other genes specific to beta cells [48, 49]. The low amount of antioxidant capability of beta cells makes the negative effects of oxidative stress on them more severe than on other cell types.

When antioxidative enzyme gene expression in various cell types of islets prepared from human pancreatic sections was compared, it was found that beta cells expressed less of the (Hydrogen peroxide-H₂O₂), H₂O₂-eliminating glutathione peroxidase (GPx1), catalase (CAT) (15-fold), and O-2-eliminating superoxide dismutases CuZn/Mn SOD (1.4-fold) than non-beta cells [50].

According to recent research, beta cells may be less vulnerable to oxidative damage than previously believed because they can withstand micromolar concentrations of H₂O₂ via a thioredoxin/reductase-dependent mechanism [51, 52]. This assumption is based on the human Endo C bH1 beta cell line’s documented capacity to use the peroxiredoxin/thioredoxin antioxidant system to detoxify physiologically generated levels of H₂O₂ continuously for four hours at a dose of 50 µM. The line is identical to primary human beta cells. Nevertheless, in the same investigation, EndoC-bH1 beta cells were given a bolus of 100 µM H₂O₂, which decreased cell viability, caused DNA damage, and depleted cellular energy [51]. In light of these findings, questions about the effectiveness of the peroxiredoxin/thioredoxin system in protecting beta cells during extended exposure to physiologically relevant H₂O₂ concentrations still persists [53].

However, because of their low antioxidant capacity, beta cells are more vulnerable to signals from reactive oxygen species (ROS) that controls glucose-stimulated insulin secretion (GSIS) under physiological settings [54]. Under conditions of high glycolytic and tricarboxylic acid (TCA) flux, reactive species are mostly formed in the mitochondrial respiratory chain, which supplies adenosine triphosphate (ATP) for GSIS [49]. In rat islet beta cells, it has been demonstrated that ROS induce the activation of ryanodine receptors (RyR), which is necessary for the rise in intracellular free Ca2 + during GSIS [55].

Nonetheless, a persistent rise in glycolytic flux brought on by the pathophysiology of diabetes causes an excess of ROS and RNS to be produced in beta cells, which may have harmful effects [49]. Increased ROS levels cause beta cell damage and decrease insulin expression and secretion, indicating the importance of maintaining oxidative stress tolerance and maintaining redox equilibrium for healthy beta cell function [56].

Insulin resistance and glucotoxicity related Beta cell death

In T2DM, compensatory hyperinsulinemia and beta cell enlargement are brought on by high glucose conditions, insulin resistance, and increasing insulin requirements. Ultimately, it results in loss of beta cell mass [57]. While glucose metabolites are mostly affected by oxidative phosphorylation in normoglycemic situations, hyperglycemia stimulates other biosynthetic pathways, such as glucose autooxidation, which is the primary pathway through which reactive oxygen species (ROS) are produced in diabetic environment [58].

Highly reactive hydroxyl ions (HO-) are produced when glucose oxidation reacts with transition metals to produce O − 2 and hydrogen peroxide (H₂O₂). Nuclear factor kappa B protein (NF-kB) is activated by hyperglycemia, which also enhances the expression of NAD(P)H oxidase (NOX) and inducible nitric oxide synthase (iNOS), producing ONO − 2 and NO• leading to the production of highly noxious ·O − 2 [59].

When single strand breaks are formed by peroxynitrite, poly (ADP-ribose) polymerase 1 (PARP-1) is activated, leading to damage to DNA [60]. The protein PARP-1 attaches itself to damaged DNA and catalyzes the breakdown of NAD + into nicotinamide and ADP-ribose. These two compounds are subsequently used to form poly (ADP-ribose) polymers and form covalent bonds with proteins [61]. Glyceraldehyde-3 phosphate dehydrogenase (GAPDH) has a decrease in activity when poly ADP ribosylation occurs [62]. Glucose, glucose-6-phosphate, fructose-6-phosphate, fructose-1,6-diphosphate, glyceraldehyde-3-phosphate, and dihydroxyacetone phosphate are examples of GAPDH upstream glycolytic intermediates which increase the activity of sorbitol and hexosamine metabolism, protein kinase C activation, enediol formation, polyol pathway, and glycation, and contribute to the increased production of reactive oxygen species (ROS) [63].

Furthermore, nicotinamide adenine dinucleotide phosphate (NADPH) concentration is lowered by activating the polyol pathway and aldose reductase enzyme activity, which weakens beta cell antioxidant mechanisms [64]. Hyperglycemia-associated ROS overproduction causes beta cells to release more pro-inflammatory markers including interleukin 1beta (IL-1b) by activating caspase 1 and producing an inflammasome complex. This leads to pyroptosis, a type of programmed cell death [65, 66]. Proapoptotic and inflammatory activities are also triggered by prolonged hyperglycemia. These actions are mediated by nitric oxide (NO) and species associated with the generation of NO. Beta cell failure is encouraged by NF-kB activation and IL-1b-stimulated iNOS production [67].

In cell models for the disease, such as adipocytes treated with TNF and dexamethasone, ROS overproduction has been linked to insulin resistance [68]. The oxidation rate of the redox-sensitive dye dichloro-fluorescein (DCF) was 50% greater in insulin-resistant cells treated with TNF-a than in cells treated with dexamethasone (65% higher). Protein carbonylation was shown to be elevated in adipocytes treated with dexamethasone (by 110%), and in those treated with TNF (by 50%). This finding indicated that insulin-resistant cells have accumulated oxidative stress, and that the rise in ROS levels both precedes and initiates the emergence of discernible insulin resistance. It is important to note that insulin resistance depletes beta cells by increasing their need for insulin and causing compensatory hyperinsulinemia, which ultimately compromises their ability to survive [63].

In T2DM, elevated plasma fatty acid levels can exacerbate insulin resistance in addition to other negative outcomes. Since diabetes-related lipotoxicity is preceded by a chronically raised glucose concentration, the term “glucolipotoxicity” is used to characterize the detrimental action of lipids on beta cells [63]. When MIN6 cells were treated with oleate, their H₂O₂ levels increased. However, when INS-1E cells were treated with palmitate, their mitochondrial · O − 2 production increased [68, 69]. Reduced insulin expression in isolated islets and beta cells exposed to high fatty acid concentrations is associated with oxidative stress, GSIS suppression, and an acceleration of apoptotic cell death [70].

Both experimental and clinical studies have shown positive association between oxidative indicators and reduced GSIS in islets in diabetic patients, as well as higher levels of oxidative stress measures in circulation [71]. Higher amounts of oxidative stress markers, such as DNA base oxidation, 8 hydroxy-guanine, hydroperoxides, and 8-epi-PGF2a, have been found in the blood of diabetes patients [72, 73]. Human islets treated to high glucose concentrations showed excess production of reactive oxygen species (ROS) and decreased insulin content and GSIS, whereas rat islets exposed to high concentrations of the glucose metabolite D-glyceraldehyde showed the opposite effects [74].

The oxidative stress indicators 8-hydroxy deoxyguanosine (8-OHdG) and nitro-tyrosine were found in much higher amounts in the pancreatic islets that were collected from T2D cadaveric organ donors. Additionally, GSIS and reduced insulin expression were associated with these markers [75]. Through ferroptosis, an iron-related programmed cell death mechanism, iron excess linked to elevated glucose levels damages beta cells and results in insulin resistance [76]. Lipid peroxidation and mitochondrial shrinkage are the main characteristics of ferroptosis, which is carried out by the glutathione (GSH)/glutathione peroxidase 4 (Gpx4) pathway [77]. Consequently, a deficiency in GSH or inhibition of GSH synthesis may cause ferroptosis; however, this type of cell death can be prevented by antioxidants and lipophilic/iron-chelating substances [78].

Stancic and colleagues [79] have investigated the capacity of diabetogenic drugs, such as high glucose (HG), proinflammatory cytokines, H₂O₂, and streptozotocin (STZ), to induce ferroptosis in the Rin-5 F beta cell line in vitro. They also investigated the role of ferroptosis in beta cell injury in vivo using STZ-diabetic mice. All assessed in vitro treatments decreased the viability of Rin-5 F cells and were associated with elevated iron, ROS, lipid peroxides, and inactivation of NF-E2-related factor 2 (Nrf-2). Furthermore, the expression of Gpx4 and MMP decreased. In beta cell treatments with HG, H₂O₂, and STZ, the ferroptosis inhibitor ferrostatin 1 (Fer-1) decreased those effects; however, in Rin-5 F cells treated with cytokines, it had no effect on mitochondrial membrane potential (MMP) or cell survival. Additionally, Fer-1 decreased the amount of lipid peroxide accumulation and macrophage infiltration, increasing the number of insulin-positive cells in the islets of STZ diabetic mice. These findings strongly suggest that, rather than proinflammatory cytokines, high glucose levels and excessive reactive oxygen species (ROS) production cause beta cell death by ferroptosis in diabetic conditions [79].

A growing body of evidence indicates that aberrant autophagy may be connected to insulin resistance, obesity, and beta cell dysfunction in diabetes [80, 81]. The macro-, crino-, and micro-autophagy components of the lysosomal degradation machinery preserve cellular homeostasis by recycling damaged macromolecules and organelles into new proteins and energy [81]. A accumulation of extra proteins and their detrimental effects can result from defective autophagy in beta cells exposed to oxidative stress [75]. Therefore, the increase in autophagy in the islets of diabetic db/db and C57BL/6mice fed a high-fat diet (HFD) suggests that autophagy plays a role in the beta cells’ adaptive response to combat insulin shortage and resistance [82].

Autophagy may also protect cells from the negative effects of increased PARP-1 activity and single- or double-strand DNA breaks caused by H₂O₂ [83]. Inhibiting proinsulin synthesis and increasing AIF translocation from mitochondria to the nucleus, independent of the caspase cascade, activation of PARP-1 and the consequent intracellular depletion of NAD + and ATP [61, 84] results in chromatin condensation, DNA fragmentation, and programmed necrotic cell death. Huang et al.‘s research, however, showed that PARP-1 might trigger autophagy via stimulating the AMPK-AMP-activated protein kinase (AMPK) mammalian target of rapamycin (mTOR) pathway, which is a serine/threonine protein kinase LKB1. This implies that the balance between autophagy and necrosis, which is regulated by two different PARP-1-pathways, is necessary for beta cell survival.

Oxidative stress and DNA damage are two examples of the several cellular stressors that cause beta cells to go through autophagy [85]. Through a number of methods, ROS causes autophagy by activating mitogen-activated protein kinases (MAPKs) such as c-Jun NH2-terminal kinase 1 (JNK1), which suppresses insulin receptor substrate 1 (IRS1) [86]. Hydrogen peroxide may activate the enzyme PKR-like eIF2a kinase (PERK). Long-term autophagy requires the synthesis of LC3, which is increased when PERK phosphorylates the general autophagy regulator initiation factor 2a (eIF2a) [87]. By blocking IkBa translation, PERK additionally elevates NF-kB, a factor that contributes to autophagy [88]. Hydrogen peroxide instantly oxidizes Atg4 proteases, enhancing their activity and facilitating faster synthesis of proteolytically mature LC3 [89].

Beta cell identity and de-/trans-differentiation

The conventional view that the primary cause of the beta cell loss linked to diabetes is beta cell death has recently been called into question. De- and trans-differentiation processes lead to the loss of beta cell identity, which is increasingly associated with beta cell dysfunction [57]. Research on rodents in vitro and in vivo indicates that several transcription factors crucial for the development, maturation, regeneration, and appropriate functioning of beta cells regulate the preservation of beta cell identity.

Dedifferentiation (Fig. 5) is defined by the upregulation of genes inaccessible to beta cells, such as progenitor Neurogenin3, SRY-box transcription factor 9 (SOX9), or monocarboxylate transporter MCT1, and lactate dehydrogenase genes, and the downregulation of important markers of beta cells, such as Pdx-1, musculoaponeurotic fibrosarcoma oncogene homolog A (MafA), homeobox protein (Nkx6.1), neurogenic differentiation factor 1 (NeuroD1), forkhead box protein O1 (Foxo1), and Pdx-1 [90].

Fig. 5.

Oxidative stress impairs beta cell identity by altering the expression of genes linked to beta cell maturity and failure. De- and trans-differentiation processes lead to a loss of beta cell identity, which contributes to beta cell failure in a diabetic milieu, reduced expression of oxidative stress-related beta cell-fate markers, including duodenal homeobox factor-1 (Pdx-1Progenitor markers such as SRY-box transcription factor 9 (SOX9) and Hairy enhancer of split-1 (HES1) exhibit enhanced expression, whereas the muscle-aponeurotic fibrosarcoma oncogene homolog A (MafA), paired box 4 (Pax4), and homeobox protein Nkx6.1 suppress the expression of insulin (Ins) and the glucose transporter GLUT2, suggesting that the process of dedifferentiation is taking place. Additionally, oxidative stress can raise the aristaless-related homeobox gene factor (Arx), which is unique to alpha cells. This can lead to a decrease in Pdx-1 expression, which in turn causes beta cells to transdifferentiate into alpha cells

Contrarily, transdifferentiation (Fig. 5) postulates that beta cells can transform into distinct islet cell types that make hormones. Mice lacking the Foxo1 transcription factor in beta cells showed reductions in beta cell mass because beta cells dedifferentiated into progenitor-like cells, alpha-like cells, Neurogenin3-, Nanog-, octamer-binding transcription factor 4 (Oct4-), and cell cycle regulator L-Myc-expressing cells [91]. Hyperglycemia and hyperglucagonemia were the outcomes of this.

likewise, independent studies have shown that loss of transcription factors critical to maintaining the identity of mature beta cells can also cause dedifferentiation in human beta cells. Pdx1, MafA, and Nkx6.1 markers were specifically identified in the pancreatic islets of T2D cadaveric organ donors, and their selective loss was associated with a significant drop in insulin [92, 93]. Beta cells most likely employ dedifferentiation as a defensive tactic to prevent an autoimmune reaction [94]. Evidence showing that the autoimmune response and the onset of diabetes are inhibited in NOD mice when genes linked to beta cell maturity and acceptance of the stemness-like properties of beta cells are deleted [95].

When the beta cell-specific transcription factor aristaless-related homeobox gene (Arx) is overexpressed or beta cell-specific Pdx-1 is deleted in rats, the amount of α cells rises and the amount of β cells falls (98). Diabetic db/db mice had similar alterations in beta cell-fate and were also demonstrated to have reduced levels of Pdx-1, MafA, and insulin [90]. Pdx-1 is important for pancreatic growth and differentiation as well as maintaining optimal beta cell function, whereas MafA transcription factor primarily regulates insulin gene expression and also contributes to beta cell proliferation and development [48, 49].

When Pdx1 was specifically removed from beta cells, the majority of the deleted beta cells underwent reprogramming to become alpha-like cells, which included the de-repression of the MafB transcription factor that is exclusive to alpha cells. This resulted in severe hyperglycemia in PKO animals [96]. Based on these findings, it seems likely that Pdx1 controls beta cell identity by simultaneously activating genes related to beta cell fate and suppressing genes related to alpha cell fate. Beta cells that lack NKX6.1 are more likely to acquire the delta-cell phenotype [97], whereas those that lack PAX6 or NKX2.2 develop polyhormonal or epsilon-like cells, respectively [98, 99].

Various findings suggest that oxidative stress and redox-related mechanisms trigger de- or trans-differentiation processes in beta cells by obstructing the expression of transcription factors associated with beta cell maturity [100], as depicted in Fig. 2. Early research revealed that loss of Pdx-1 and MafA binding to the insulin promoter during extended culture of insulin-secreting HIT-T15 cells at high glucose concentrations (11.1 mM) hampered insulin gene expression and GSIS [101, 102].

These glucotoxicity effects were linked to posttranscriptional loss of Pdx-1 mRNA and posttranslational loss of MafA protein caused by oxidative stress. These findings were further corroborated by research demonstrating that antioxidants like N-acetylcysteine (NAC) or aminoguanidine improved insulin gene expression in high-glucose cultured HIT-T1 cells by maintaining Pdx1 and MafA binding to the insulin promoter [103]. Subsequent research has demonstrated that H₂O₂ inhibits Pdx-1’s binding to the insulin promoter, hence reducing the production of the insulin gene in rat islets [104]. It is likely that the JNK pathway and the Foxo1 transcription factor mediate the detrimental effects of ROS on Pdx-1 activity [44]. Specifically, Foxo1 undergoes nuclear translocation with JNK activation, while beta cells are shielded from ROS-linked reduced expression of insulin when JNK inhibits it.

Accordingly, Foxo1 expression in H₂O₂-treated HIT-T15 beta cells is siRNA-inhibited while maintaining Foxo1’s cytoplasmic localization and Pdx-1’s nuclear localization in addition to insulin expression [105]. Similarly, H₂O₂ causes MafA to localize in the cytoplasm and inhibits its activating influence on insulin gene expression [106]. Consequently, it is thought that the cytoplasmic translocation of MafA caused by hyperglycemia serves as a precursor to beta cell malfunction. The data presented clearly imply that the subcellular location of important insulin transcription factors and their regulators is influenced by redox state [48].

Leenders and colleagues (2010) examined the impact of H₂O₂-mediated oxidative stress on the identity and function of human beta cells in diabetes circumstances using human islets and the human EndoC-bH1cell line. In human EndoC-bH1 cells, they discovered that treatment with H₂O₂ increased the stress response of the cells and decreased the expression of insulin and glucose transporter GLUT1. These effects were linked to a decrease in the expression of genes related to beta cell fate, such as Pax6, Nkx2.2, NeuroD1, Kir6.2, MafB, and Foxa2, as well as beta cell fate markers MafA, Pdx-1, paired box4 (Pax4), and Nkx6.1. In H₂O₂-treated EndoC-bH1 cells, simultaneous upregulation of the progenitor cell-specific SOX9, hairy, and enhancer of split-1 (HES1) genes points to the possibility of dedifferentiation processes [94].

Accordingly, using primary human islets and human islets transplanted into mice fed a high-fat diet showed that tacrolimus, a commonly used immunosuppressant in organ transplantation therapies, causes beta cell maturity features to be lost more quickly by activating Foxo1 in response to metabolic stress and reducing MafA expression [107]. Results demonstrating that Gpx1 overexpression or treatment with Gpx mimetic drugs could restore expression of Pdx-1 and/or MafA in islets of diabetic mice provide support for the idea that oxidative stress is linked to the loss of beta cell identity [108, 109]. However, those results also suggest that focusing on antioxidant pathways may be important for maintaining beta cell identity and function in diabetic patients.

Oxidative stress induced epigenetic modulations in T2DM

Diabetes mellitus raises blood glucose levels, which impact metabolic and biochemical pathways and are mirrored in epigenetic changes [110]. In diabetics, decreased levels of S-adenosyl methionine (SAM), a methyl group donor, may prevent methyltransferase activity and methylation of DNA and histone proteins [111, 112]. The assessment of the histone modification role in the diabetes cohort showed a correlation between the monocytes’ levels of glycohemoglobin and H3K9 acetylation, along with changes in the DNA methylation of the 3′ UTR of the thioredoxin-interacting protein, indicating the possibility of using this biomarker to detect oxidative stress [113, 114].

A lower amount of DNA methyltransferases (DNMTs) in leucocytes was found in an investigation of older T2DM patients [115]. The development of type 2 diabetes is thought to be linked to changes in DNA methylation brought on by an increase in short chain fatty acid levels. These changes cause pro-inflammatory cytokine genes to become dysregulated and hypomethylated 116]. Likewise, reduced levels of trimethylated lysine 9 residue of histone 3 (H3K9me3) at the interleukin-6 promoter and Suv39H1 methyltransferase were found in the vascular smooth muscle cells of T2DM mice [117]. Changes in miRNA levels have been linked to the pathophysiology of diabetes. For instance, a decrease in miR-126 has been linked to the development of T2DM [118, 119]. This suggests that regulation of epigenetic regulatory systems is greatly influenced by oxidative stress [120].

Atypical DNA methylation patterns can result from alterations in DNA protein and protein-protein interactions brought on by oxidative stress-induced DNA strand breakage and damage [14]. 8-OHdG, which is generated when guanine within CpG islands oxidizes, decreases the affinity of methyl-domain binding proteins and increases the methylation of neighbouring cytosine in DNA.

Reactive molecules can also promote DNA methylation by accelerating up the interaction between DNA and SAM donor and cytosine deprotonation (127). Oxidative stress reduces the activity of HDACs, including Silent Information Regulator 1 (Sirt1), and enhances the interaction between DNMT1 and histone deacetylase (HDAC) 1 [121]. Furthermore, oxidative stress can have detrimental effects on various lysine residues on histone domains [122]. Undoubtedly, these alterations in epigenetic modifications cause disruptions in the expression and activity regulation of antioxidant enzymes and their regulators, which in turn favours events of oxidative stress. Increased H4K20 methylation, H3K9 acetylation, and DNA methylation all decrease the expression of the SOD gene in diabetes conditions [123, 124]. Histone modifications and miRNAs epigenetically regulate Nrf2, a key regulator of genes associated with antioxidant defence which is regulated by Keap1 [110].

Oxidative stress and epigenetic targets aimed by diabetes therapy

The correlation between diabetes pathophysiology and epigenetic mechanisms has incentivized the creation of novel treatment approaches capable of undoing epigenetic alterations, including epigenome editing and epidrugs [125, 126]. As a result, artificial substances have been thoroughly investigated as possible epigenetic regulators, mostly in cancer but also in other clinical conditions like diabetes. In addition to being currently approved medications for the treatment of hematological malignancies in elderly patients, DNMTs inhibitors such as 5-azacytidine (Aza, Vidaza) and its derivative 5-aza-2′ deoxycytidine (DAC, Decitabine) have also been demonstrated to be effective in the treatment of inflammatory diseases [127, 128], and [129]. Application of DAC to HFD fed ob/ob mice enhanced insulin sensitivity, while treatment of non-obese diabetic (NOD) mice with DAC delayed diabetes development by cyclophosphamide [130, 131].

HDAC inhibitors trichostatin and valproic acid In diabetic rats, there was a decrease in apoptosis and an increase in beta cell proliferation and function [132, 133]. Medication like metformin and fenofibrate, which function as agonists of the HDAC enzyme Sirt1, increased insulin production and enhanced glucose metabolism in diabetic mice [134, 135].

Antioxidants become a potent alternative strategy in epigenetic therapy when one considers how oxidative stress influences the development of diabetes through altering epigenetic processes. Apart from their well-established anti-inflammatory and antioxidant properties, polyphenolic substances are also becoming more widely acknowledged for their impact on epigenetic regulation. Research has demonstrated that polyphenolic substances has the capacity to regulate many epigenetic mechanisms, such as histone modification, DNA methylation, and miRNA levels [136].

While most studies focused on the benefits of natural polyphenols in cancer treatment, it is clear that epigenetic regulatory systems have a role in a number of disorders. These findings are undoubtedly useful guidelines for future study in other areas. Thus far, a restricted quantity of research has documented a causal association between dietary polyphenolic chemicals and epigenetic alterations in the pathophysiology of diabetes. Nevertheless, it is indisputable that worldwide alterations in DNA and/or histone methylation patterns may be impacted by dietary deficiencies or excesses in methyl donors [120].

For the regeneration of primary methyl donor SAM from S-adenosyl homocysteine (SAH) [111], [139SAH buildup in the cell may prevent the methylation of DNA or histone proteins because it has a negative regulatory influence on methyltransferase activity. These nutrients include zinc, folate, choline, betaine, or vitamins B2, B6, and B12. Polyphenols have the ability to inhibit DNMTs either directly (by reducing DNMT expression and activity) or indirectly (by lowering SAM levels). While curcumin and epigallocatechin-3-gallate (EGCG) directly inhibit DNMT1, catechol-containing polyphenols such as quercetin, epicatechin, rutin, luteolin, and caffeic acid act as non-competitive DNMTs inhibitors by inducing SAH overproduction through the use of SAM as a methyl donor for their own methylation [136–138].

Oxidative stress and DNA methylation in T2DM

Ortiz et al. [139] found that high FKBP5 intron 2 methylation at a specific CpG-dinucleotide region was positively associated with elevated HbA1c levels (8.3%) in patients with Type 2 Diabetes Mellitus (T2DM) (r = 0.535, p = 0.003). Roshanzamir and Hassan-Zadeh [140] further investigated the effect of glycemia on the DNA methylation of pro-inflammatory genes (IL-1β and IL1R1), discovering that hyperglycemia led to hypomethylation of IL-1β and hypermethylation of IL1R1. Sanger sequencing revealed that hyperglycemia affects two specific CpG sites in IL-1β and one of three CpG sites in IL1R1. These findings suggest that DNA methylation at metabolic pathway-associated CpG sites may serve as an epigenetic marker of chronic inflammation and T2DM. However, assuming that changes in DNA methylation at single or few CpG sites are functionally significant is precarious, especially if the absolute changes are minor. Thus, these results should be interpreted with caution until corroborated by mechanistic data.

Pinzon-Cortes et al. [141] examined global DNA methylation and hydroxymethylation in peripheral blood mononuclear cells (PBMCs) from well- and poorly-controlled T2DM patients, finding an increase in 5hmC in the latter group. The correlation between 5mC and 5hmC with glycated hemoglobin indicates a direct link between hyperglycemia and epigenomic alterations. Their work suggests that TET isoenzymes are activated by increased glucose levels, a hypothesis confirmed by Dhliwayo et al. [142] who demonstrated that hyperglycemia activates TET enzymes and demethylates genomic cytosine. They also found that inhibiting poly(ADP-ribose) polymerase (Parp) prevents demethylation via the TET-dependent oxidation pathway (5mC-5caC), implying that hyperglycemia activates Parp, which in turn increases TET activity and DNA methylation. Similarly, Yuan et al. [143] utilized HPLC-MS/MS and qPCR to measure global 5mC and 5hmC, as well as histone deacetylase SIRT6 and TET activity, in fasting blood samples from a T2DM population. They found that genomic 5mC decreased and 5hmC increased with overexpression of TET2, TET3, and SIRT6. Correlation analysis indicated a positive relationship between SIRT6 and TET2 activity (r = 0.277, p < 0.001), suggesting that hyperglycemia activates NAD + and α-KG, leading to the oxidation of 5mC to 5hmC by SIRT6 and TET2. Pharmacological inhibition of SIRT6 improved glycemic control in T2DM, supporting its role in metabolic regulation [144].

Contrary to these findings, Wu et al. [145] present data that challenge the activation of TET enzymes by hyperglycemia, as suggested by Dhliwayo et al. [142] and Yuan et al. [143]. Further research is necessary to elucidate the relationship between metabolism and DNA methylation in β-cells via TET enzymes. Since no TET enzyme has been identified in terminally differentiated β-cells, it is possible that changes in 5mC in stem cells, which differentiate into β-cells in the pancreas or blood, could lead to epigenetic modifications in the T2DM model. This hypothesis aligns with disease development models where DNA methylation changes originate in stem cells before appearing in differentiated cells. Rigorous experimentation is needed to validate this concept.

The debate continues regarding whether poor glycemic control is a cause or consequence of epigenetic regulation. Kim [146] emphasizes that DNA methylation may act both as a cause and consequence of T2DM. Numerous studies suggest that DNA methylation drives T2DM development, partly due to accumulated errors leading to altered transcriptional responses [146]. While aberrantly methylated genes in T2DM are linked to β-cell dysfunction and disrupted insulin and glucagon secretion [21], other studies indicate differential methylation patterns in tissues such as adipose tissue, skeletal muscle, and liver when compared to non-T2DM samples [147–149].

T2DM is characterized by oxidative stress, which modifies DNA, RNA, lipids, and proteins, producing reactive oxygen species (ROS) [150, 151]. Chronic hyperglycemia over activates metabolic pathways for protein kinase C, polyol, hexosamine, and glucose autoxidation, generating ROS [152]. ROS production from electron leakage at complexes I-III in the electron transport chain, p66Shc, and monoamine oxidase is implicated in T2DM pathogenesis [153]. Mitochondria play a significant role in increased ROS production, which is linked to diabetes onset, progression, and complications such as retinopathy, neuropathy, nephropathy, and cardiovascular disease [154]. T2DM and prolonged hyperglycemia deplete systemic and intracellular ascorbic acid, exacerbating oxidative stress and impairing glucose control [155]. Recent research indicates that ascorbic acid supplementation improves glycemic control in T2DM [156]. Hyperglycemia and ROS activate the H3K4 methyltransferase SET7, reducing the expression and activity of the Nrf2 antioxidant enzyme [154].

Two important aspects emerge from this: (1) Ascorbic acid regulates TET isoenzymes, and (2) ROS-induced DNA damage may disrupt epigenetics. Ascorbic acid enhances TET-mediated 5mC oxidation in vivo [157, 158]. Despite glucose availability affecting α-KG production, reduced ascorbic acid levels in T2DM may partially inactivate TET, leading to hypermethylation and potentially creating an oncogenic environment. Additionally, ROS directly modify DNA bases, with hydroxyl free radicals (OH) generated by Fenton reactions involving the reduction of H₂O₂ with ferrous or copper ions [159]. OH-mediated DNA alterations involve hydrogen abstraction or interaction with DNA bases, such as the oxidation of 5mC to 5hmC [154]. OH radicals can also directly impact DNA methylation by oxidizing guanosine to 8-oxo-2′-deoxyguanosine (8-oxodG). In the short-patch base excision repair (BER) pathway, 8-oxoguanine DNA glycosylase (OGG1) removes 8-oxodG residues [154, 160]. The accumulation of 8-oxodG reduces neighboring cytosine methylation, leading to hypomethylation and transcriptional activation [161]. Zhou et al. [162] propose that OGG1 binds to TET1 at oxidized lesions to demethylate DNA, suggesting that oxidative stress recruits the OGG1/TET1 complex to 8-oxodG, converting 5mC to 5caC near ROS-induced damage.

Conclusion and future scope

One of the key events in the etiology of type 2 diabetes is the loss of bulk and function in the pancreatic beta cells. This review examined many aspects of oxidative stress’s role in the structural and functional disruption of pancreatic beta cells to highlight the need of focusing on redox-related pathways in diabetes therapy. Unmistakably, oxidative stress has a negative impact on beta cell identity and survival through its involvement in a number of pathogenic pathways that lead to beta cell dysfunction, de-/trans differentiation, and death, according to both experimental and clinical research. Oxidative stress can cause harm to beta cells directly by the oxidative destruction of macromolecules or indirectly through disruptions of several regulatory networks, one of which includes epigenetic mechanisms. The inhibition of Arx by DNA methylation is one of the regulatory processes that determines the identity of beta cells. On the other hand, there is a negative correlation between insulin production and the degree of DNA methylation of the insulin promoter. Therefore, clinical studies have demonstrated that altered methylation of the insulin gene promoter is associated with type 2 diabetes (T2DM), indicating that DNA methylation is important for both insulin secretion and beta cell dysfunction. Diabetes-related oxidative stress-induced damage to DNA and proteins can result in aberrant DNA methylation patterns and histone changes. This provides a path for the creation of a novel therapeutic strategy centered on repairing epigenetic abnormalities by targeting oxidative stress. The experimental data demonstrates that pro-survival pathways in beta cells under diabetes conditions can be stimulated by antioxidants originating from plants, such as polyphenolic chemicals, which also modify redox-related signaling and alter ROS production. Furthermore, antioxidants are used in epigenetic therapy because polyphenolic substances can modify mechanisms associated to epigenetics, including DNA methylation, histone changes, and miRNA level. In the very least, this strategy might help preserve beta cells in the early stages of diabetes or stimulate the growth of new beta cell mass in the later stages of the condition. In a nutshell epigenetics may have multiple applications in ameliorating the conditions associated with diabetes. The differentially methylated insulin gene has been identified as a biomarker of progressive beta cell loss, so the identification of novel blood-based epigenetic biomarkers may be utilized to forecast the risk of developing diabetes and its complications.

Moreover, a promising direction for diabetes treatment in the future may be the creation of novel epidrugs that operate as activators or repressors of different epigenetic enzymes important for DNA or histone modifications and control of miRNA production. Despite these advancements in the field of diabetes, very few studies of cross sectional design could not effectively establish causality or satisfy hill’s criteria in establishing true cause and effect relationship between oxidative stress and epigenetic modifiers in the pathogenesis of T2DM. Therefore, further longitudinal or cohort studies and larger epidemiological studies are required to unveil the precise role of DNA methylation in oxidative stress mediated islet or beta cell dysfunction in the pathogenesis of T2DM.

Also, future research should examine if epigenetic pathways may be targeted for the treatment of T2DM and how important epigenetic biomarkers are for predicting the disease. Future research should take into account the methylation of repetitive areas and non-CpG methylation, as these are suggested to be important factors in the pathophysiology of T2DM.

Funding

This work is not funded by any organisation.

Data availability

Data will be made available on request.

Declarations

Ethical approval

This study doesn’t involve human participants, their data or biological material.

Conflict of interest

The author declares that there is no conflict of interest with regard to the publishing of this work.