Imbalances in RNA metabolic networks in vascular lesions associated with diabetes and sarcopenia

Abstract

Diabetes and sarcopenia have been shown to share common risk factors and are conditions that have been demonstrated to have a severely deleterious effect on the quality of life of older adults, and which have even been shown to increase the risk of death. Their bidirectional relationship is clear: Each condition raises the likelihood of developing the other. The interaction between these two conditions has been demonstrated to result in a significant increase in the risk and severity of vascular complications. Vascular lesions in the context of diabetes complicated by sarcopenia exhibit unique pathological characteristics, namely a higher risk of vascular lesions and a more severe course. In the comorbidity model, the resulting imbalance in the body’s RNA metabolic network exacerbates vascular lesions, leading to persistent vascular damage. This review summarizes how RNA metabolic imbalance in the context of comorbidity serves as a key molecular hub linking diabetic metabolic disorders, the progression of sarcopenia, and the onset of vascular lesions. Non-coding RNA regulatory networks represent a pivotal regulatory mechanism for comorbidity-related vascular lesions, playing an irreplaceable regulatory role in the formation and progression of the comorbidity vicious cycle. They provide novel molecular targets for precision-targeted therapy of diabetes, sarcopenia and associated vascular lesions.

Key Words: Diabetes; Sarcopenia; Vascular lesions; RNA metabolic network; Non-coding RNA; Mechanisms of comorbidity

Core Tip: This article highlights that imbalances in RNA metabolism represent a key molecular link between diabetes, sarcopenia and vascular pathology. It emphasizes that non-coding RNA networks act as indispensable regulatory factors in the vascular damage caused by comorbidity, thereby providing new targets for precision therapy in the treatment of vascular pathology associated with comorbidity in older adults.

INTRODUCTION

Diabetes and sarcopenia are both prevalent among the elderly, and there is a clear bidirectional relationship between the two. A review of the extant epidemiological data indicates that the prevalence of sarcopenia among elderly patients with type 2 diabetes mellitus (T2DM) is 13.8%[1], and a meta-analysis further confirms that the prevalence of sarcopenia in patients with T2DM is approximately 18%[2]. The prevalence of sarcopenia is considerably elevated in patients with diabetes, and the underlying mechanisms involve multiple factors, including insulin resistance, chronic hyperglycaemia, inflammatory responses, and mitochondrial dysfunction[3,4]. Hyperglycaemia, the key metabolic disorder in diabetes, has been shown to induce skeletal muscle atrophy via the WW domain-containing E3 ubiquitin protein ligase 1/Krüppel-like factor (KLF) 15 axis by upregulating the expression of the transcription factor KLF15[5]. Furthermore, enhanced alanine catabolism in the liver in T2DM exacerbates the vicious cycle of hyperglycemia and muscle atrophy by promoting skeletal muscle proteolysis[6]. Conversely, sarcopenia has also been demonstrated to exacerbate glycemic control in patients with diabetes; the loss of muscle mass can lead to increased insulin resistance, creating a “diabetes-sarcopenia” vicious cycle[7,8].

Research has demonstrated that elevated blood glucose levels can induce oxidative stress, inflammation, impaired protein metabolism, and mitochondrial dysfunction, which collectively underpin sarcopenia[9]. Consequently, diabetes and sarcopenia have been demonstrated to be causally linked, and their co-occurrence has been shown to significantly increase the complexity of clinical management.

In the context of comorbidity, the risk and severity of vascular complications increase significantly, exhibiting unique pathological characteristics. Diabetes itself is an independent risk factor for vascular complications, and when combined with sarcopenia, patients’ risk of microvascular and macrovascular complications increases further. Research has indicated that patients diagnosed with T2DM who also exhibit sarcopenia are observed to have considerably elevated rates of diabetic retinopathy, peripheral neuropathy, and vascular complications in comparison to those not manifesting sarcopenia[10]. From a pathophysiological perspective, the presence of hyperglycemia leads to a state of oxidative stress and the subsequent accumulation of advanced glycation end products. These phenomena have been demonstrated to exert a direct deleterious effect on the vascular endothelium, in addition to a concomitant exacerbation of muscle atrophy via the activation of protein degradation pathways[11]. Concurrently, the decline in muscle mass induced by sarcopenia weakens the muscle’s function as a “metabolic buffer” thereby exacerbating glucose and lipid metabolism disorders and accelerating the progression of vascular damage[12]. Vascular lesions in this comorbid state often manifest as a more extensive range of damage, earlier disease progression, and poorer prognosis. This suggests the presence of unique molecular mechanisms underlying these conditions.

In recent years, imbalances in the RNA metabolic network have been shown to be a key molecular mechanism linking comorbid metabolic disorders, muscle atrophy, and vascular damage. Non-coding RNAs (ncRNAs), such as microRNAs (miRNAs), long ncRNAs (lncRNAs), and circular RNAs (circRNAs), play an indispensable regulatory role in this process. In the pathophysiological process of diabetic sarcopenia, the expression profiles of various ncRNAs change significantly. These changes contribute to the onset and progression of muscle atrophy by regulating key processes, such as insulin signaling pathways, the balance between protein synthesis and degradation, mitochondrial function, and inflammatory responses[13]. For instance, under hyperglycemic conditions, certain miRNAs can promote muscle protein degradation by regulating forkhead box O (FoxO) transcription factors and the ubiquitin-proteasome system, such as atrogin-1 and muscle-specific RING finger protein 1[14]. Furthermore, lncRNAs and circRNAs can act as competitive endogenous RNAs (ceRNAs), indirectly regulating the expression of downstream target genes by “sponging up” miRNAs, thereby serving as “bridges” in the molecular network of diabetic sarcopenia. Abnormal expression of these ncRNAs affects metabolic homeostasis in skeletal muscle cells and may mediate intercellular communication via pathways such as exosomes. This transmits signals of metabolic dysregulation to vascular endothelial cells, contributing to vascular complications[15]. Thus, an imbalance in the RNA metabolic network is the molecular basis for vascular complications associated with diabetes and sarcopenia.

NcRNA regulatory networks play a central role in the formation and maintenance of the vicious cycle of comorbidities, providing new molecular targets for precision-targeted therapies for vascular pathologies. In diabetic sarcopenia, ncRNAs link upstream signals - such as hyperglycemia, insulin resistance, and inflammation - to downstream effects - such as muscle atrophy and vascular damage -through multilevel regulatory mechanisms. For example, the miR-29 family is upregulated in diabetic conditions and promotes muscle atrophy by inhibiting the insulin-like growth factor-1 signaling pathway[16]. At the same time, circRNAs such as circFoxO3 can regulate FoxO3 expression by interacting with miRNAs, thereby influencing autophagy and apoptosis pathways[17]. The abnormal expression of these ncRNAs not only exacerbates muscle protein degradation but also leads to muscle dysfunction by affecting mitochondrial biosynthesis and energy metabolism[18]. More importantly, certain ncRNAs (such as miR-126) have been shown to participate in the regulation of angiogenesis and endothelial function; their dysregulation may directly lead to the development of diabetic microvascular complications[15]. Therefore, targeting these key ncRNAs holds promise for simultaneously improving muscle atrophy and vascular damage, achieving a “two birds with one stone” therapeutic effect.

This article aims to systematically review the latest research on imbalances in RNA metabolic networks associated with vascular complications in sarcopenic diabetes, integrate molecular mechanisms and clinical findings, and establish a theoretical foundation for future therapeutic strategies. A particular focus is placed on the bidirectional crosstalk between skeletal muscle and vascular endothelium - a dimension that has received limited attention in previous reviews. While significant progress has been made in the study of diabetic sarcopenia in recent years, research on the RNA regulatory networks underlying its comorbidity with vascular complications is still in its early stages. Current evidence suggests that ncRNAs play a key role in linking metabolic disorders, muscle atrophy, and vascular damage. However, the specific regulatory networks and signaling pathways require further elucidation[19]. Furthermore, ncRNA-based therapeutic strategies, such as antisense oligonucleotides (ASOs), miRNA mimics, and inhibitors, have demonstrated the potential to improve muscle atrophy and vascular function in animal models[20,21]. However, translating these findings into clinical applications faces numerous challenges, including optimizing delivery systems, improving targeting specificity, and assessing long-term safety. Therefore, this article aims to provide new perspectives for a deeper understanding of the molecular mechanisms underlying vascular complications associated with sarcopenia in diabetes and lay a theoretical foundation for developing precision treatment strategies based on RNA metabolic networks by systematically reviewing the latest research advances in this field.

CHARACTERISTICS OF VASCULAR COMPLICATIONS AND CLINICAL CHALLENGES IN PATIENTS WITH COEXISTING DIABETES AND SARCOPENIA

High risk of vascular lesions and unique pathological manifestations in patients with comorbidities

Patients with comorbid diabetes and sarcopenia are at significantly higher risk for vascular complications, which exhibit unique pathological manifestations. Epidemiological data indicate that sarcopenia is more prevalent in patients with diabetes than in the general population, and poor glycemic control is positively and linearly correlated with the risk of developing sarcopenia[22]. There is now ample population-based evidence to show that, among individuals already living with T2DM, sarcopenia further significantly increases the risk of future cardiovascular events. A study employing a “rate advancement period” analysis found that, in patients with sarcopenia, the age at which cardiovascular disease occurs is approximately 14.5 years earlier than in those without sarcopenia[23]; furthermore, sarcopenic obesity may be a highly significant high-risk phenotype for the development of heart failure in patients with T2DM[24]. This comorbidity exacerbates the vicious cycle of vascular damage: On the one hand, the chronic hyperglycaemia and insulin resistance associated with diabetes are key drivers of endothelial dysfunction, basement membrane thickening and vascular sclerosis; on the other hand, the decline in muscle mass and function caused by sarcopenia further impairs glycaemic control and metabolic homeostasis[25]. Specifically, in patients with sarcopenia, a reduction in skeletal muscle - the primary organ responsible for glucose uptake - exacerbates insulin resistance, making blood glucose control more difficult. This creates a vicious cycle of “hyperglycaemia-sarcopenia-more severe hyperglycaemia”, accelerating vascular remodelling[12]. Furthermore, the chronic low-grade inflammation associated with sarcopenia combines with the inflammatory response to diabetic vascular complications, resulting in greater damage to vascular endothelial cells and promoting the progression of atherosclerosis[26]. Vascular complications in patients with diabetes and sarcopenia are not limited to large vessels; the incidence of microvascular complications, such as retinopathy, nephropathy and neuropathy, is also significantly increased. The vascular pathological changes associated with this comorbidity are more extensive and severe, manifesting as an increased inflammatory response, more pronounced vascular sclerosis and more widespread microvascular damage. This ultimately leads to accelerated vascular remodeling and a poorer clinical prognosis[27].

The impact of comorbidities on the prognosis of vascular disease and current treatment challenges

Sarcopenia complicating diabetes has an extremely adverse impact on the prognosis of vascular complications and poses a serious challenge to existing treatment strategies. Clinical studies have confirmed that patients with diabetes and sarcopenia have significantly higher rates of vascular recurrence, readmission, and all-cause mortality than patients with diabetes alone[27]. For instance, an analysis of the National Health and Nutrition Examination Survey database revealed that individuals with high body fat and low muscle mass had a mortality rate more than four times higher if they had diabetes, with heart disease being the leading cause of death[27]. This poor prognosis is partly due to changes in drug metabolism and tolerability caused by sarcopenia. Conventional treatments for hyperglycaemia, hypertension and hyperlipidaemia are often less effective in patients with comorbidities. Those with sarcopenia frequently suffer from malnutrition, weight loss and alterations in drug volume distribution, which may affect the efficacy and safety of these treatments[28]. Furthermore, blood glucose fluctuations are more prevalent among patients with comorbidities. Such fluctuations have been demonstrated to be an independent risk factor for sarcopenia, with significant associations observed with low muscle mass, low grip strength, and slow walking speed[29]. Currently, treatment options for sarcopenia complicating diabetes are limited. Existing treatment strategies tend to focus on a single condition, making it difficult to reverse both sarcopenia and vascular complications simultaneously. Although routine strict blood glucose control may, to some extent, delay the progression of vascular complications in diabetes mellitus alone[30], recent clinical studies have reached contradictory conclusions: Long-term excessive glycaemic control and restrictive diets may further deplete skeletal muscle reserves and accelerate muscle mass loss, thereby exacerbating the vicious cycle of metabolic dysfunction and vascular damage[31]. Similar controversies exist regarding exercise intervention strategies: Moderate resistance training can improve insulin sensitivity and muscle mass, and assist in restoring vascular endothelial function[32]; however, for patients with comorbidities who have already developed severe vascular sclerosis and luminal stenosis, high-intensity exercise is likely to induce vascular stress injury and exacerbate endothelial inflammation. This clinical paradox, where therapeutic benefits coexist with damage, demonstrates that neither the regulation of blood glucose alone nor the improvement of muscle phenotype can fundamentally halt the progression of comorbidities. It is therefore evident that the key to overcoming current therapeutic bottlenecks lies in moving away from the traditional approach of single-organ regulation and instead exploring the core molecular dysregulation mechanisms shared by muscle atrophy and vascular damage. As an imbalance in the RNA metabolic network serves as a central hub linking these two pathological processes, it holds promise as a precise target for achieving synergistic interventions that protect muscle and repair blood vessels. Consequently, there is an urgent clinical need to explore precision intervention strategies that target the core molecular mechanisms of both conditions simultaneously, in order to break this vicious cycle and improve patient outcomes[12].

IMBALANCES IN THE RNA METABOLIC NETWORK AS A MOLECULAR HUB FOR COMORBID VASCULAR LESIONS

The structure of the RNA metabolic network and global imbalances in comorbid conditions

The RNA metabolic network is a highly dynamic and finely tuned regulatory system encompassing the entire process of messenger RNA (mRNA) transcription, splicing, transport, translation and degradation, as well as the generation and functional regulation of ncRNAs, such as miRNAs, lncRNAs and circRNAs. In the comorbid state of diabetes and sarcopenia, a “multiple-hit” effect comprising hyperglycaemia, insulin resistance, chronic inflammation and oxidative stress acts in concert to cause transcriptome reprogramming, resulting in a global imbalance in RNA metabolism. This is primarily reflected in the abnormal expression of numerous mRNAs: For instance, reduced levels of mRNA for key metabolic regulatory genes directly impact cellular energy metabolism and function. Concurrently, there are also highly pronounced disruptions in RNA processing, including alterations in mRNA stability, abnormal splicing variant ratios and dramatic changes in the ncRNA expression profile. In particular, the impact of dysregulation of splicing factors on cardiac muscle and vascular tissue cannot be overlooked. For example, upon activation, the splicing factor splice factor 3b subunit 1 mediates the alternative splicing of ketohexokinase, resulting in the production of the high-affinity ketohexokinase-C subtype, which drives pathological proliferation of cardiomyocytes, ultimately leading to cardiac hypertrophy and heart failure[33]. High expression of serine/arginine splicing factor 1 can regulate the alternative splicing of activating transcription factor 3, thereby inhibiting the KLF2-sphingosine-1-phosphate receptor 1 pathway and enhancing endothelial angiogenesis[34]. The alternative splicing variant of the lncRNA CARMEN (CARMEN-201) binds to the transcriptional repressor RE1-silencing transcription factor, suppressing cardiac gene expression and inducing vascular smooth muscle cell differentiation; alternant splicing of Myocardin produces a myocardium-specific subtype that regulates smooth muscle cell gene expression and participates in vascular development; the loss of transcriptional function in the truncated splice variant of T-box transcription factor 5 induces congenital cardiovascular malformations accompanied by vascular developmental defects[35]. These dysregulations in splicing factors suggest that disruptions in RNA processing represent a critical aspect of RNA metabolic dysregulation that requires urgent attention. Research indicates that ncRNAs, particularly miRNAs, lncRNAs and circRNAs, play a key role in the onset and progression of diabetic vascular complications by inhibiting or promoting the expression of target genes and participating in the proliferation, differentiation and apoptosis of vascular endothelial cells[36]. For example, miRNA-126 is specifically expressed in endothelial cells, and its abnormal expression is closely associated with diabetic vascular remodelling[37]. Furthermore, the abnormal expression of lncRNA metastasis-associated lung adenocarcinoma transcript 1 (MALAT1) has been shown to impair vascular endothelial cell function via hyperglycaemic metabolic memory[38]. These imbalances in RNA metabolism provide a molecular basis for vascular pathology in comorbid conditions and offer a new perspective on its complex pathological mechanisms.

Disruption of molecular pathways mediated by RNA metabolic imbalance and vascular damage

Disruption of downstream signalling pathways and vascular damage: Disruptions in several key molecular pathways caused by a global imbalance in the RNA metabolic network directly lead to vascular damage. Firstly, in terms of metabolic regulation, a decrease in the mRNA levels of key genes such as glucose transporter type 4 and peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α) impairs glucose uptake and mitochondrial biosynthesis directly, leading to impaired intracellular energy metabolism and exacerbating the toxic damage caused by hyperglycaemia to endothelial cells further. Secondly, abnormal processing of mRNA for angiogenesis-related factors - such as changes in the proportion of vascular endothelial growth factor (VEGF) splicing variants - disrupts the balance of angiogenesis, leading to tissue ischemia. Specific ncRNAs, including miR-27, have been implicated in this process through their regulation of lipid metabolism, glucose metabolism and angiogenic pathways[39]. Furthermore, sustained amplification of the inflammatory response is a key mechanism of vascular damage. In the context of comorbidity, the stability of the mRNA of pro-inflammatory factors such as interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α) increases and their translation efficiency is enhanced. This leads to the persistence of the local inflammatory response within blood vessels and its continuous amplification. This inflammatory microenvironment damages endothelial cells directly and promotes the phenotypic transformation and calcification of vascular smooth muscle cells by activating various signalling pathways, such as nuclear factor κB (NF-κB). Studies have shown that macrovascular calcification in patients with type 2 diabetes is closely linked to oxidative stress, inflammatory responses, and the regulation of ncRNAs[40]. Consequently, imbalances in RNA metabolism collectively drive the progression of vascular complications associated with diabetes and sarcopenia by disrupting multiple key pathways, including metabolism, angiogenesis, and inflammation (Figure 1).

Figure 1 Imbalances in the RNA metabolic network as a molecular hub for comorbid vascular lesions. This figure is produced using Adobe Illustrator and Biorender. Hyperglycaemia, insulin resistance, chronic inflammation and oxidative stress collectively drive dysregulation of the RNA metabolic network, involving dysregulation of miRNA (miRNA-126), long ncRNAs (metastasis-associated lung adenocarcinoma transcript 1) and circRNA expression, as well as abnormalities in epigenetic modifications such as m6A. This pivotal disruption subsequently triggers three downstream effects: Impaired metabolic regulation (such as downregulation of glucose transporter type 4 and peroxisome proliferator-activated receptor gamma coactivator 1-alpha mRNA, compromising glucose uptake and mitochondrial biosynthesis), impaired angiogenesis (abnormal splicing of vascular endothelial growth factor mRNA) and exacerbated inflammation (upregulation of interleukin-6 and tumor necrosis factor-alpha mRNA). These three pathways ultimately act synergistically on the vascular wall, leading to endothelial damage, smooth muscle cell phenotypic transformation and calcification, thereby accelerating the progression of diabetic vascular complications. MALAT1: Metastasis-associated lung adenocarcinoma transcript 1; GLUT4: Glucose transporter type 4; PGC-1α: Peroxisome proliferator-activated receptor gamma coactivator 1-alpha; VEGF: Vascular endothelial growth factor; IL-6: Interleukin-6; TNF-α: Tumor necrosis factor-alpha; NF-kB: Nuclear factor kappa B.

Regulation of the m6A-modified transcriptome and vascular injury: Chemical modifications of RNA molecules are key epigenetic pathways that mediate imbalances in RNA metabolic networks. m6A is the most abundant internal modification of eukaryotic mRNA; it is co-regulated by the writing complex methyltransferase-like (METTL) 3/METTL14, the demethylation eraser proteins FTO/ALKBH5, and reading proteins such as YTHDF1/2/3 and YTHDC1, and is extensively involved in biological processes including RNA splicing, translation, intracellular localisation and transcript stability regulation. Existing research has confirmed that dysregulation of m6A modification permeates the entire pathological course of diabetes and is strongly associated with persistent hyperglycaemic stimulation, insulin resistance and impaired pancreatic β-cell function[41].

At the level of vascular endothelium, abnormal m6A modification is a key driver of diabetic vascular complications. A hyperglycaemic microenvironment upregulates FTO expression, leading to decreased overall cellular m6A methylation levels, which subsequently suppress endothelial cell migration and tubular formation and activate inflammatory responses. Conversely, FTO knockdown at the cellular level effectively alleviates endothelial damage induced by high-glucose exposure[42]. In models of microvascular complications such as diabetic nephropathy and diabetic retinopathy, tissue m6A modification levels are generally abnormally elevated[43,44]. This phenotypic change is significantly correlated with fluctuations in the expression of key modifying enzymes such as METTL3. Multiple studies have confirmed that m6A modification is involved in skeletal muscle atrophy. A global decrease in m6A levels in skeletal muscle is accompanied by muscle mass loss, and m6A can regulate muscle protein degradation through the ubiquitin-proteasome pathway[45]. METTL3, as a key methyltransferase, is an indispensable regulatory molecule for myogenic differentiation and skeletal muscle regeneration after injury[46]. Meanwhile, m6A can form a regulatory network with miRNAs[47], collectively governing skeletal muscle developmental homeostasis.

Based on the above hypotheses, in the context of diabetic sarcopenia comorbidity, a hyperglycaemic and chronically inflammatory microenvironment is thought to simultaneously disrupt the expression levels of m6A-modifying enzymes in skeletal muscle and the vascular endothelium. This causes abnormal degradation of mRNAs associated with muscle homeostasis, leading to skeletal muscle atrophy, and disrupts the normal transcriptional regulation of genes involved in vascular function, exacerbating microcirculatory dysfunction. Atrophied skeletal muscle can release m6A-modified ncRNA via extracellular vesicles; upon uptake by vascular endothelial cells, this reshapes the cellular ceRNA regulatory network. Meanwhile, inflammatory mediators continuously released by damaged blood vessels further exacerbate the skeletal muscle microenvironment, creating a vicious cycle of mutually reinforcing muscle–vascular damage. It must be emphasized that the complete cascade currently rests on fragmentary experimental evidence from individual steps, and direct validation of the entire regulatory axis is still lacking.

THE KEY ROLE OF NCRNAS REGULATORY NETWORKS IN COMORBIDITY-RELATED VASCULAR LESIONS

The regulatory role of miRNAs and their association with diabetes, sarcopenia and vascular disease

MiRNAs, as the primary negative regulators in the RNA metabolic network, bind to target mRNAs via base pairing to inhibit their translation or promote their degradation, thereby playing a key regulatory role in vascular complications associated with diabetes and sarcopenia. In models of co-morbidity, the expression of multiple miRNAs is significantly dysregulated; among these, miR-21, miR-126 and miR-133a are involved in the regulation of vascular endothelial function, muscle regeneration and inflammatory responses, respectively. Studies have shown that miR-21 is closely associated with cardiac development, angiogenesis and vascular lesions in cardiovascular disease; its upregulation promotes the proliferation and migration of vascular smooth muscle cells, thereby accelerating the progression of atherosclerosis[48]. At the same time, miR-126 is specifically expressed in endothelial cells and is extensively involved in processes such as cell differentiation and proliferation; it plays a significant role in diabetic vascular remodelling by regulating endothelial dysfunction[37]. In vascular endothelial cells, miRNA-126 is generally regarded as a miRNA that maintains vascular integrity; however, in animal models of diabetes, miR-126 levels first increase and then decrease in both the early and late stages of the disease. This pattern indicates that, in response to hyperglycaemic stimulation, the body initially employs compensatory mechanisms to protect the blood vessels. However, persistent hyperglycaemia subsequently causes endothelial cell damage, leading to failure of vascular repair. However, a frequently overlooked aspect in many studies is the variation in the expression of miR-126 in skeletal muscle. The expression of miR-126 derived from myocytes has been shown to promote repair and regeneration following muscle injury, thereby maintaining the function and integrity of the neuromuscular junction[49]. The findings outlined above demonstrate that RNA molecules exhibit cell-type specificity, spatiotemporal dynamics and disease-stage dependence. It is therefore proposed that research into RNA metabolism in diabetes complicated by sarcopenia should establish models incorporating temporal and spatial dimensions.

Furthermore, downregulation of miR-133a is closely associated with sarcopenic muscle atrophy, whilst abnormal miRNA expression in diabetes also involves the regulation of angiogenesis; for example, there is a decrease in the expression of miR-210, miR-93 and miR-126 - which promote angiogenesis - and an increase in the expression of miR-320 and miR-503 - which inhibit angiogenesis. These changes collectively contribute to the development of various vascular complications associated with diabetes[50]. Among the vascular complications of diabetes, miRNAs targeting KLF2, KLF4 and sirtuin (SIRT) 1 can cause endothelial dysfunction, whilst miR-223 can modulate inflammatory responses in vascular endothelial cells and reduce leukocyte adhesion[51]. Similarly, miRNAs can guide the phenotypic polarisation of macrophages; miR-155 and miR-146a, as two key molecules with relatively well-defined roles in macrophage polarisation, often exert opposing effects in a diabetic environment. In diabetes, an abnormal increase in miR-155 promotes the polarisation of macrophages towards the pro-inflammatory M1 phenotype, thereby exacerbating the inflammatory response[52]. In contrast, miR-146a typically exerts an anti-inflammatory effect in a diabetic environment and promotes the polarisation of macrophages towards the M2 phenotype[53]. It is worth noting that miRNAs also contribute to the onset and progression of diabetic cardiomyopathy by regulating pathophysiological processes such as cardiomyocyte apoptosis and cardiac fibrosis; among these, miR-29 plays a significant role in diabetic cardiovascular complications, and its expression patterns are closely associated with the diabetic state[54]. However, there remains significant controversy regarding the expression patterns of the miR-29 family within the pathophysiological context of diabetes complicated by sarcopenia: Animal studies have confirmed that the hyperglycaemic microenvironment in mouse models of diabetes upregulates the expression of the miR-29 family, and that this molecule induces skeletal muscle atrophy by specifically inhibiting the insulin-like growth factor-1 signalling pathway[16]. In contrast, clinical test results from Chinese elderly patients with sarcopenia and concomitant high-risk cardiovascular factors showed significantly reduced levels of miR-29b in peripheral plasma[55]. These differences in expression patterns may be influenced by organisation-specific factors, regulatory networks and feedback from downstream pathways; variations in the dominant regulatory factors lead to changes in the composition of cell populations, which ultimately affect the expression of miR-29. These findings suggest that miR-29 cannot be categorised simplistically as either an atrophy-promoting or a protective molecule; its expression levels and biological functions are regulated not only by factors such as species differences and comorbidities, but are also closely related to the different subtypes within the miR-29 family. Future research should therefore distinguish between and focus on the independent expression profiles and biological effects of each subtype. Furthermore, miR-27 plays a significant role in the progression of diabetic lower limb vascular disease by specifically regulating lipid metabolism, glucose metabolism and angiogenesis pathways[39]. The synergistic dysregulation of these miRNAs constitutes a complex regulatory network within the comorbidity of diabetes, sarcopenia and vascular disease, providing potential biomarkers and intervention targets for clinical diagnosis and treatment.

The molecular scaffolding role of lncRNAs in comorbidity signalling and their competitive interaction with endogenous RNAs

LncRNAs are defined as ncRNA molecules longer than 200 nucleotides. They perform a dual function as molecular scaffolds and ceRNAs in vascular complications associated with diabetes and sarcopenia. LncRNAs act as molecular scaffolds to recruit various protein complexes. They participate in chromatin remodelling, transcriptional activation or silencing. This influences the expression of genes associated with vascular protection or damage. Research has demonstrated that LncRNA MALAT1 fulfils a pathogenic role in type 2 diabetic microvascular complications. It has been shown to be involved in the regulation of gene expression through splicing and the control of epigenetic processes, and its abnormal expression is closely associated with the development of diabetic microvascular complications[56]. In the context of comorbidity, it has been observed that exosome release during skeletal muscle atrophy is enriched with specific lncRNAs. This has been shown to cause protective miRNAs, such as miR-126, in vascular endothelial cells to be “adsorbed like a sponge”. This process has been demonstrated to indirectly inhibit the VEGF signalling pathway and exacerbate microvascular damage. This ceRNA regulatory mechanism of the muscle-vascular axis is one of the key characteristics that distinguishes vascular lesions associated with comorbidity from those associated solely with diabetes. In atherosclerosis, lncRNAs have been demonstrated to play a pivotal role in the initiation and progression of the disease by modulating the “damage-response” mechanism. This mechanism encompasses processes such as lipid deposition, vascular endothelial inflammation, and the processes of cell proliferation and apoptosis[57]. Research has identified several lncRNAs associated with comorbidity-related vascular lesions, such as MALAT1, HOX transcript antisense RNA and nuclear enriched abundant transcript 1. These have been shown to contribute to vascular pathology by regulating endothelial-mesenchymal transition and inflammatory pathways. In the context of diabetic neurovascular complications, the role of lncRNAs in modulating target gene expression is of particular interest. These non-coding transcripts have been shown to regulate relevant cellular pathways, inflammatory genes and downstream cytokines through activation or inhibition, thereby contributing to the progression of the disease[58]. The majority of in vivo studies suggest that MALAT1 promotes angiogenesis; research has confirmed that the absence of MALAT1 reduces vascular growth[59], and therefore, in a vascular repair context, MALAT1 is considered a protective lncRNA. However, in cases of diabetic vascular complications, MALAT1 has been observed to transition from its role as an angiogenic promoter to that of an inflammatory promoter, thus driving vascular damage[58]. This is because differences in microenvironmental signals cause MALAT1 to exert its effects via different molecular mechanisms. Under normal physiological conditions or during tissue repair, MALAT1 downregulates pro-inflammatory factors such as TNF-α and IL-6, whilst upregulating pro-angiogenic factors, thereby protecting blood vessels; however, under the pathological stimulation of hyperglycaemia, MALAT1 instead activates inflammatory signalling pathways such as nuclear factor kappa B (NF-κB), and the resulting inflammatory response causes damage to blood vessels. This underscores the limitations of studies that directly categorise a specific RNA as either protective or pathogenic, as its function is contingent on the specific disease context. In skeletal muscle, the role of MALAT1 is even more complex, exhibiting contradictions that are dependent on the tissue. During the initial phase of differentiation, MALAT1 has been observed to impede myogenic differentiation while concurrently exhibiting a dual effect of promoting proliferation and inhibiting differentiation[60]. Conversely, in the later stages of differentiation, several studies have demonstrated a promotion of myogenic differentiation by MALAT1, a function that is entirely opposite to that observed in the early stages[61]. This paradoxical phenomenon arises because these molecules regulate different signalling pathways at different stages of differentiation. In the early stages of differentiation, MALAT1 acts as a molecular scaffold to recruit an epigenetic repressive complex to the myogenetic oligodeoxynucleotide-bound promoters of myogenic genes, thereby blocking the myogenic programme and maintaining cell proliferation; however, once differentiation is initiated, miR-181a enters the nucleus and degrades MALAT1, causing the complex to dissociate, allowing myogenetic oligodeoxynucleotide to regain its transcriptional activity, causing the cells to cease proliferation and initiate differentiation[60]. In the context of sarcopenia, depletion of MALAT1 is a key driver of atrophy and fibrosis in ageing skeletal muscle[62]. The function of MALAT1 in the context of diabetic sarcopenia remains unclear and warrants further investigation in future studies. Furthermore, the ncRNA growth arrest-specific transcript 5 plays a significant role in coronary heart disease-related pathologies, contributing to pathological processes such as atherosclerosis, myocardial infarction and ischaemia-reperfusion injury; it may serve as a potential biomarker of great significance for the diagnosis, treatment and prognostic assessment of coronary heart disease[63]. In diabetic cardiomyopathy, the aberrant expression of lncRNAs regulates pathological processes such as cardiomyocyte apoptosis, fibrosis and inflammatory responses at the molecular level. It has been hypothesised that certain of these non-coding ribonucleic acids may also serve as potential diagnostic markers for dilated cardiomyopathy[64]. These findings suggest that lncRNAs play a central regulatory role in the signalling pathways underlying the comorbidity of diabetes, sarcopenia and vascular disease through their molecular scaffold and ceRNA functions (Figure 2).

Figure 2 The key role of non-coding RNA regulatory networks in comorbidity-related vascular lesions. This figure is produced using Adobe Illustrator and Biorender. Dysregulated miRNAs, long non-coding RNAs and circular RNAs form an integrated competing endogenous RNA regulatory network in diabetes-sarcopenia-associated vasculopathy. Dysregulation of miRNAs (such as miRNA-21, miRNA-126, miRNA-133a, miRNA-29 and miRNA-27), combined with the molecular scaffold function of long non-coding RNAs, the miRNA sponge effect and the role of circular RNAs as stable competing endogenous RNA regulatory factors, collectively regulate target genes. These non-coding RNAs collectively regulate endothelial dysfunction, inflammation, angiogenesis, vascular remodelling, atherosclerosis and vascular calcification, ultimately promoting the development of vascular complications in the context of diabetes-sarcopenia comorbidity. lncRNAs: Long non-coding RNAs; circRNAs: Circular RNAs; ceRNA: Competing endogenous RNA; VSMC: Vascular smooth muscle cell; MALAT1: Metastasis-associated lung adenocarcinoma transcript 1; HOTAIR: HOX transcript antisense RNA; NEAT1: Nuclear enriched abundant transcript 1; GAS5: Growth arrest-specific 5; EndMT: Endothelial-to-mesenchymal transition.

The stability of circRNA and its regulatory networks in the comorbidity microenvironment

CircRNA is distinguished by its distinctive covalently closed ring structure, which endows it with elevated stability and resistance to degradation by nucleases. This renders it an optimal endogenous competitive RNA molecule, capable of long-term regulation of miRNA activity. It plays a significant role in regulating the microenvironment associated with vascular complications in diabetes complicated by sarcopenia. CircRNA possesses a number of biological characteristics, including spatiotemporal specificity, high evolutionary conservation, stability and ubiquity. Its biological functions include sponge-like adsorption, regulation of gene transcription, interaction with RNA-binding proteins (RBPs), and participation in protein translation, amongst others[65]. In the context of comorbidities, the expression profiles of specific circRNAs undergo changes; these circRNAs regulate the expression of target genes by binding to miRNAs associated with vascular function via a sponge-like mechanism. Research indicates that circRNAs have the capacity to modulate the function of pancreatic β-cells, thereby contributing to the onset and progression of diabetes. Furthermore, they have been associated with complications such as diabetic macrovascular disease, microvascular disease and peripheral neuropathy[65]. Moreover, circRNAs have been shown to interact with RBPs, influencing the phenotypic transformation of vascular smooth muscle cells and endothelial cell function, thereby exerting a bidirectional regulatory role in the progression of vascular lesions. In the context of diabetic vascular lesions, the role of circRNAs is to regulate multiple stages of diabetic microvascular and macrovascular lesions. These include pathological changes such as endothelial cell apoptosis and thickening of the basement membrane. This occurs through circRNAs competing with miRNAs for binding sites[66]. It is important to note that the stability of circRNAs enables them to exert a sustained regulatory role within the comorbid microenvironment; by forming complex ceRNA networks, they collaborate with lncRNAs and miRNAs to constitute a ncRNAs regulatory network, thereby amplifying or inhibiting signalling pathways. In the progression of diabetic neuropathy, the role of circRNAs, along with other ncRNA species such as lncRNAs and miRNAs, is increasingly being elucidated. These ncRNA species have been shown to play a pivotal regulatory role in the progression of neuropathy, thus highlighting their potential as novel clinical targets for the prevention and treatment of diabetic peripheral neuropathy. In addition, their function as biomarkers for predicting disease onset and prognosis is a promising area of research[58]. Furthermore, the role of circRNAs in the calcification of vascular smooth muscle cells is significant, with the involvement of these molecules in the vascular calcification process through mechanisms such as the regulation of osteogenesis-related proteins, calcification signalling pathways and changes in intracellular Ca²⁺ levels[67]. These findings reveal the key regulatory role of circRNAs in the comorbidity microenvironment of diabetes, sarcopenia and vascular disease, providing a crucial basis for the development of new diagnostic markers and therapeutic targets.

The paracrine regulation of vascular injury via extracellular vesicles-ncRNA signalling axis

Extracellular vesicles (EVs) are a class of vesicular particles with a double-layered membrane structure that are actively secreted by cells. They can stably carry a variety of bioactive substances and serve as key natural carriers for intercellular communication and the exchange of substances in the body. In the pathological microenvironment of diabetic sarcopenia, EVs can specifically load ncRNAs - such as miRNAs, lncRNAs, and circRNAs - secreted by source cells and deliver these molecules to distant tissues via the circulatory system. This intercellular transport pattern disrupts the isolated homeostasis of local tissues and establishes a paracrine communication bridge between skeletal muscle and vascular endothelium. The bidirectional pathological interactions between these two tissue types simultaneously drive skeletal muscle atrophy and vascular endothelial injury, ultimately promoting the initiation and progression of composite lesions in diabetic sarcopenia.

There is significant controversy in the current literature regarding the biological effects of skeletal muscle-derived extracellular vesicles in the pathological context of diabetic sarcopenia. In severe limb ischemia caused by diabetes, EVs released from skeletal muscle cells carry abnormally high levels of muscle-specific miR-99-5p. Upon uptake by endothelial cells, miR-499-5p significantly impairs angiogenic capacity by specifically inhibiting the expression of SRY-box transcription factor 6 and meis homeobox 1[68]. However, this skeletal muscle-to-endothelial cell injury signal is not static. It is modulated by changes in the endothelial microenvironment, particularly ischaemia and hypoxia. Exosomes secreted by skeletal muscle cells under hypoxic conditions exhibit a distinct functional phenotype compared with those under normoxic conditions: They significantly enhance endothelial cell proliferation and migration, reduce apoptosis, and induce angiogenesis[69]. Notably, the direction and effects of this skeletal muscle-endothelium EV communication can be modulated by physiological interventions. Skeletal muscle-derived exosomes undergo dynamic changes depending on the type of exercise (resistance training, endurance training, or high-intensity interval training) and, upon entering cardiovascular tissues, trigger a series of protective physiological effects, including enhancing angiogenesis, inhibiting pathological fibrosis, and improving myocardial repair[70]. Furthermore, as a novel biomarker, exosomal RNA holds great potential for combining with traditional metabolic indicators to construct more accurate diagnostic models. Analysis of receiver operating characteristic curves showed that the area under the curve for a combined model of traditional biomarkers, such as calf circumference, was 0.892[71]; whilst the area under the curve for the triglycerides-glucose-body mass index, atherogenic index of plasma and postprandial glucose excursion was 0.918. The combination of multiple indicators in these composite models can significantly improve diagnostic performance. Although there are currently no direct comparative data on exosome quantification, research in other fields suggests that exosomal RNA may provide earlier information and could have particular diagnostic value for diabetes complicated by sarcopenia.

The evidence above paints a dynamic picture of communication between skeletal muscle and endothelial cell-derived EVs. Within the pathological microenvironment of diabetic sarcopenia, skeletal muscle EVs act not only as “amplifiers” of injury signals, but also as “sensors” of microenvironmental stress and as “effectors” of exercise intervention. This bidirectional, malleable, and condition-dependent nature of EV regulation has led to conflicting conclusions across different studies conducted under varying pathological conditions and using different intervention modalities. Overall, this dynamic and highly variable myo-vascular EV communication network provides a comprehensive molecular theoretical framework for systematically elucidating the pathogenesis of diabetic sarcopenia complicated by endothelial dysfunction, resolving controversies in existing research, and developing integrated intervention strategies.

THE MOLECULAR MECHANISMS BY WHICH RNA METABOLIC IMBALANCES EXACERBATE VASCULAR DAMAGE

Oxidative stress and the activation of inflammatory signalling pathways

In the context of the comorbidity of diabetes and sarcopenia, an imbalance in RNA metabolism is a key factor in exacerbating vascular damage. A chronic hyperglycaemic environment has been demonstrated to significantly enhance oxidative stress through various mechanisms, including activation of the polyol pathway, non-enzymatic glycosylation of proteins, and activation of the protein kinase C pathway[72,73]. This state of oxidative stress directly leads to a decrease in the mRNA expression levels of antioxidant enzymes (such as superoxide dismutase and glutathione peroxidase), significantly impairing the ability of vascular endothelial cells to scavenge reactive oxygen species and thereby causing persistent damage to endothelial cells[74]. Concurrently, perturbations in RNA metabolism are also discernible at the level of regulation of inflammatory signalling pathways. Disruptions in the RNA-level regulation of key inflammatory pathways, such as NF-κB and Janus kinase - signal transducer and activator of transcription, have been demonstrated to promote the abnormal transcription of vascular endothelial cell adhesion molecules (such as vascular cell adhesion molecule-1 and intercellular cell adhesion molecule-1) and pro-inflammatory factors (such as TNF-α and IL-6). This, in turn, has been shown to exacerbate the inflammatory response and structural damage to the vascular wall[75]. It is noteworthy that this chronic inflammatory state is not a unidirectional process; it can, in turn, act upon RNA metabolism itself. Research has demonstrated that inflammatory signals have the capacity to influence the stability of specific mRNAs by modulating RNA methylation modifications (e.g., m6A modifications). For instance, this can be achieved by altering the half-life of mRNAs encoding antioxidant enzymes or anti-inflammatory factors, thereby establishing a positive-feedback vicious cycle that continuously amplifies oxidative stress and inflammatory damage[76]. This oxidative-inflammatory loop, driven by an imbalance in RNA metabolism, constitutes a key molecular basis for the progression of vascular complications in diabetes complicated by sarcopenia (Figure 3).

Figure 3 Schematic diagram of non-coding RNA regulatory networks contributing to vascular injury in diabetic sarcopenia. This figure is produced using Adobe Illustrator and Biorender. Under the combined stimulation of hyperglycaemia, oxidative stress and inflammatory cytokines, non-coding RNAs (miRNAs, long non-coding RNAs and circular RNAs) derived from exosomes of atrophic sarcopenic skeletal muscle form a dysregulated competitive endogenous RNA network. This disturbance triggers endothelial dysfunction and vascular smooth muscle dysfunction, ultimately driving vascular remodelling. MALAT1: Metastasis-associated lung adenocarcinoma transcript 1; GAS5: Growth arrest-specific 5; circRNA: Circular RNAs; HOTAIR: HOX transcript antisense RNA; NEAT1: Nuclear enriched abundant transcript 1; ROS: Reactive oxygen species.

Endothelial dysfunction and an imbalance in vascular regeneration

Imbalances in RNA metabolism that impair endothelial function represent a central pathological mechanism in vascular complications associated with diabetes and sarcopenia. Firstly, it is crucial to acknowledge that reduced mRNA transcription of endothelial nitric oxide synthase (eNOS) and decreased mRNA stability are pivotal factors that result in vasodilation dysfunction. The suppression of eNOS mRNA expression, coupled with the disruption of the regulatory elements in its 3’ untranslated region, results in a reduction in eNOS protein synthesis. This phenomenon occurs under the influence of oxidative stress and inflammatory factors. Consequently, this results in inadequate nitric oxide production and significantly impaired vasodilation[77]. Secondly, the dysregulation of the mRNA expression of angiogenic factors, and the disruption of their ncRNA regulatory networks, further disrupt the balance of vascular regeneration. Abnormalities in the mRNA expression of pro-angiogenic factors, such as VEGF and fibroblast growth factor, coupled with disruptions in the associated lncRNA and miRNA regulatory networks, result in angiogenesis that is either insufficient (leading to tissue ischaemia) or abnormal (resulting in structurally defective new blood vessels). This phenomenon is particularly characteristic of diabetic retinopathy[78]. Furthermore, the aberrant expression of mRNAs associated with endothelial-mesenchymal transition also plays a significant role. The microenvironment of hyperglycaemia and inflammation has been demonstrated to induce the upregulation of specific miRNA transcripts in endothelial cells, which have been shown to promote mesenchymal transformation. The most notable of these miRNAs are those which regulate the expression of transcription factors, such as Snail family transcriptional repressor 1 and Twist-related protein 1/2, which have been identified as key players in this process. It is this gradual loss of endothelial characteristics that ultimately results in the transformation of endothelial cells into mesenchymal cells. This process directly promotes fibrosis and stiffening of the vascular wall, accelerating pathological changes in both large and small blood vessels[79,80].

Metabolic abnormalities and mitochondrial dysfunction

The underlying causes of the exacerbation of vascular complications in diabetes complicated by sarcopenia are metabolic abnormalities and mitochondrial dysfunction resulting from an imbalance in RNA metabolism. At the level of the insulin signalling pathway, there is a significant reduction in the mRNA expression levels of key molecules such as insulin receptor substrate-1 and protein kinase B (AKT). This directly impairs the tissues’ ability to uptake and utilise glucose, exacerbating hyperglycaemia and thereby further intensifying hyperglycaemia-related vascular damage[81]. Concurrently, the expression of mitochondrial biosynthesis genes (e.g., PGC-1α, nuclear respiratory factor 1 and transcription factor A, mitochondrial) is downregulated, resulting in a decrease in mitochondrial number and functional impairment. As the cell’s “powerhouse”, impaired mitochondrial function triggers an energy metabolism crisis and releases large quantities of pro-apoptotic factors, ultimately leading to the apoptosis of vascular endothelial cells and smooth muscle cells, thereby accelerating the degenerative changes in the vascular wall[75]. Consequently, a growing body of research supports the view that mitochondrial quality control (including mitochondrial autophagy and mitochondrial dynamics) represents a highly promising therapeutic target for cardiovascular damage in diabetes. In the inflammatory environment associated with diabetes, the immune inflammatory mediator S100A9 has been shown to directly induce excessive mitochondrial fission and reduced autophagy in cardiomyocytes by activating the STAT3 signalling pathway. Research has demonstrated that the inhibition of S100A9, along with its associated downstream signalling pathways, has the potential to effectively restore mitochondrial quality control mechanisms and thereby enhance cardiac function[82]. SIRT3 is a key energy sensor and deacetylase within mitochondria; a hyperglycaemic environment significantly inhibits the expression and enzymatic activity of SIRT3 in cardiomyocytes and vascular endothelial cells[83]. The classical SIRT3-FoxO3 axis activates PTEN-induced putative kinase 1/parkin RBR E3 ubiquitin-protein ligase-mediated mitochondrial autophagy to clear damaged mitochondria, forming the core of cardiovascular protection[84]. The potential of these pathways to repair cardiovascular damage associated with diabetes makes them a viable therapeutic strategy for the condition. Furthermore, the occurrence of abnormal splicing of mRNAs associated with fatty acid oxidation is a factor that cannot be overlooked. In the context of diabetic conditions, the splicing patterns of mRNA precursors for key enzymes involved in fatty acid β-oxidation (e.g., carnitine palmitoyltransferase 1A and acyl-CoA dehydrogenase, very long chain) are subject to alteration, leading to the production of non-functional or abnormal protein isoforms. This results in a diminished capacity of vascular endothelial cells to metabolise fatty acids, leading to an aberrant accumulation of lipids within the vascular wall and consequently expediting the formation and progression of atherosclerotic plaques[76,79]. These multi-level metabolic disturbances, triggered by an imbalance in RNA metabolism, collectively constitute a vicious cycle - ranging from energy metabolism to structural damage - in vascular complications associated with diabetes and sarcopenia.

In summary, the majority of the molecular pathways summarised in this paper (such as the vascular damage caused by miRNA-126, the influence of miRNAs on macrophage polarisation, the role of MALAT1 in inflammation and muscle proliferation and differentiation, and the damage to the cardiovascular system caused by mitochondrial dysfunction) are derived primarily from two types of indirect evidence: (1) The following findings are derived from studies conducted on pure type 2 diabetes models or patients; and (2) The following findings are derived from models of skeletal muscle atrophy and sarcopenia. The effects of abnormal growth arrest-specific transcript 5 expression, circRNA-ceRNA networks, and epigenetic modifications on muscle satellite cells are derived from analyses of human transcriptomic data.

These results suggest the functional roles of molecular mechanisms in comorbid conditions and provide a reasonable theoretical framework for comorbidity research -namely, that RNA metabolic imbalances induced by diabetes can accelerate muscle loss - but this has not been conclusively verified at the experimental level. Consequently, this review will address the absence of comorbidity models as a deficiency in the prevailing field.

THE POTENTIAL OF RNA METABOLIC NETWORKS AS TARGETS FOR PRECISION THERAPY

Therapeutic strategies based on ncRNA

NcRNAs, including miRNAs, lncRNAs and circRNAs, have been shown to play a pivotal role in the pathophysiology of diabetes complicated by sarcopenia (Table 1). For example, in the context of diabetes, the expression of muscle-specific miRNAs such as miR-1-3p and miR-133a-3p is altered, and these are directly involved in the regulation of muscle atrophy[85]. Dysregulation of these ncRNAs constitutes a central component of the imbalance in the RNA metabolic network. Consequently, therapeutic strategies that target these ncRNAs show great potential. In the context of miRNAs, the development of miRNAs mimics or antagomirs constitutes the primary means of restoring dysfunctional levels of miRNAs to normal in this comorbid condition. Evidence shows that miR-126 mimics improve endothelial function and angiogenesis, and miR-133a inhibitors help delay sarcopenia development. For more complex classes of ncRNAs, the design strategy for ASOs should shift from single-target approaches to multi-tissue coordination. Furthermore, due to differences in tissue distribution between muscle and blood vessels, the development of differentiated delivery systems presents a unique challenge for the translation of ASO therapies for comorbidities. To illustrate this point, consider the case of a rat model of diabetic erectile dysfunction, in which several differentially expressed lncRNAs have been identified. These may be involved in disease pathogenesis by regulating processes such as redox enzyme activity and energy metabolism, thereby offering new avenues for targeted therapy[86]. Concurrently, circRNAs have been shown to function as “sponges” for miRNAs, thereby modulating downstream gene expression through the establishment of a circRNA-miRNA-mRNA ternary transcriptional network. This notion has been substantiated by investigations into the mechanisms underlying angiogenesis signalling pathways in murine models of atherosclerosis[87]. A number of preclinical studies have confirmed that modulating the levels of these ncRNAs has the capacity to influence multiple target genes in both muscle and vascular tissues, thereby establishing the basis for the development of multi-target therapeutic strategies for diabetes complicated by sarcopenia.

Table 1 The regulatory functions of non-coding RNA and epigenetic modifications.

NcRNA/modification	Expression pattern	Target genes	Downstream signaling pathways	Mainly affects cell/tissue	Experimental model
miR-126	Early-stage diabetes ↑ late-stage ↓	Genes associated with the VEGF pathway, SPRED1, PIK3R2	VEGF/PI3K-AKT/eNOS angiogenesis pathway	Endothelial cells, skeletal muscle cells	Animal models of vascular injury
miR-155	↑	SOCS1, SHIP1	NF-κB	Macrophages	In vitro and animal models of diabetes-related inflammation
miR-146a	↓ or lacks functionality	IRAK1, TRAF6	NF-κB negative feedback regulation	Macrophages	In vitro and animal models of diabetes-related inflammation
MALAT1	Diabetic vascular damage ↑	VEGF, inflammatory factors	ceRNA, epigenetic regulation	Endothelial cells, skeletal muscle	In vitro and animal models of diabetic microvascular disease
GAS5	Abnormal expression	Inflammation-related genes	ceRNA	Cardiovascular tissue	Human clinical samples
circRNA-ceRNA network	↑ or ↓	Multiple miRNAs	Genes involved in angiogenesis, inflammation and calcification	Endothelial cells, smooth muscle cells	Human clinical samples
m6A modification	Abnormally ↑/↓	MyoD mRNA	METTL3, FTO	Myosatellite cells	Human clinical samples
	↑	eNOS mRNA	YTHDF2	Endothelial cells	In vitro high-glucose endothelial cell model
m6A-RBP network	Abnormal expression	Inflammation, EMT-related mRNA	RBPs	Renal tubules, blood vessels	Human model of diabetic nephropathy

VEGF: Vascular endothelial growth factor; PI3K: Phosphoinositide 3-kinase; AKT: Protein kinase B; eNOS: Endothelial nitric oxide synthase; NF-κB: Nuclear factor kappa B; MALAT1: Metastasis-associated lung adenocarcinoma transcript 1; ceRNA: Competing endogenous RNA; GAS5: Growth arrest-specific 5; circRNA: Circular RNA; MyoD: Myoblast determination protein 1; RBP: RNA-binding protein; EMT: Epithelial-mesenchymal transition; RBPs: RNA-binding proteins; YTHDF2: YTH domain-containing family protein 2.

It is worth emphasising that ncRNA is not only a target for the exogenous interventions mentioned above, but also a key mediator through which existing antidiabetic drugs exert their endogenous regulatory effects. Research has demonstrated that metformin exerts its therapeutic effect in hyperglycaemic conditions by downregulating the expression of miRNAs, such as miR-378a-3p, thereby helping to regulate mitochondrial energy metabolism, the activity of key enzymes involved in gluconeogenesis, and mitochondrial quality control[88]. Metformin also markedly reduces the abnormally elevated expression of miR-133a in diabetic skeletal muscle[89]. It inhibits miR-21 in a concentration-dependent fashion to ameliorate insulin resistance and regulates glucose transport in skeletal muscle by suppressing lncRNA Dreh[90,91]. At the vascular level, metformin upregulates miR-1 in vascular smooth muscle cells through the AMP-activated protein kinase/transforming growth factor-beta signaling pathway, thereby regulating vascular smooth muscle cells phenotypic switching[92]. sodium-glucose linked transporter 2 inhibitors also exert cardiorenal protective effects through ncRNAs and their targets. Empagliflozin has been shown to upregulate the anti-fibrotic miR-29 family[93].

Collectively, these findings demonstrate that various hypoglycemic agents actively regulate RNA metabolic networks. Accordingly, the differential expression of ncRNAs observed in clinical samples from patients with this comorbidity cannot be simply attributed to pathological disease processes, drug intervention alone, or their combined interactive effects. Different hypoglycemic drugs may exert synergistic, additive, or antagonistic influences on RNA regulatory networks, and patients with diabetes complicated by sarcopenia frequently receive combined hypoglycemic regimens. Given such complicated confounding effects, medication history should be incorporated as a critical confounding variable into experimental design and statistical analyses when investigating therapeutic targets related to ncRNAs.

However, RNA therapeutics generally suffer from issues such as poor targeting, inadequate delivery efficiency, off-target toxicity, and long-term safety concerns[94]. The absence of standardised design guidelines for targeted delivery and a universal theoretical framework has further delayed the clinical implementation of nucleic acid therapeutics[95]. Furthermore, the physicochemical properties of carriers and the route of administration directly influence the tissue enrichment of drugs. Practical challenges that remain to be addressed for RNA therapies also include drug delivery barriers in tissues protected by specialized physiological barriers, as well as evaluations of pharmacological efficacy and in vivo tolerability[96,97].

One fundamental cause of the challenges mentioned above is the strong tissue specificity exhibited in the expression and function of ncRNAs. This specificity implies that an RNA target with therapeutic potential in one tissue may perform a completely different, or even opposite, physiological function in another tissue. At the level of circRNAs, a profiling study covering 20 clinically relevant human tissues (including vascular smooth muscle cells, human umbilical vein endothelial cells, arterial endothelial cells, etc.) also reveals highly tissue-specific circRNA expression across distinct tissues[98]. These observations imply that systemically administered RNA agents require tissue-specific delivery systems. Without such design, miRNAs that confer protective effects in skeletal muscle may unintentionally cause injury to vascular endothelium. It also complicates the design of treatment strategies for comorbidities, as RNA targets validated as effective in a single tissue cannot simply be extrapolated for systemic application.

In light of the aforementioned delivery challenges, EVs offer unique advantages as natural and endogenous carriers. Research on acute myocardial infarction has confirmed that miRNAs encapsulated within EVs are able to withstand degradation by endogenous nucleases in vivo and, relying on their natural intercellular transport capacity, deliver substances to sites of cardiovascular damage, thereby demonstrating the value of EVs as carriers in the field of nucleic acid delivery[99]. Compared with synthetic nanomaterials, EVs possess low immunogenicity, high biocompatibility, and a natural ability to cross biological barriers[100]. They keep encapsulated nucleic acids stable in circulating blood and shield these molecules from degradation by nucleases. Nevertheless, the composition of endogenous EVs derived from natural skeletal muscle cannot meet therapeutic requirements. Even native vesicles secreted by stem cells exhibit limited therapeutic efficacy unless artificially fortified with optimized ratios of pro-angiogenic and myogenic factors[101]. It is possible to stably express skeletal muscle-targeting peptide and arginine-glycine-aspartic acid (peptide targeting injured ischemic blood vessels) on the EV membrane surface by genetically engineering donor cells[102]. Furthermore, utilising the encapsulation properties of EVs, two functional nucleic acid molecules can be simultaneously loaded within the same class of engineered EV carriers: Anti-miR-29 is loaded to antagonise the abnormal overexpression of miR-29 and alleviate the process of myofibre atrophy[103]; co-delivery of miR-126a-3p significantly activates endothelial cell function and promotes vascular regeneration by enhancing eNOS-related signalling[104] (Figure 4). Existing in vivo animal studies have confirmed that this dual-load targeted engineered EV formulation addresses the current clinical limitations of small-molecule and single-nucleic acid drugs, which can only target a single pathogenic pathway and cannot simultaneously address two major pathological changes. It thus provides a reliable research direction and practical basis for an integrated targeted intervention strategy for diabetic sarcopenia[101].

Figure 4 Mechanistic scheme showing how RNA metabolic imbalance aggravates vascular injury in diabetic sarcopenia. This figure is produced using Adobe Illustrator and Biorender. Hyperglycaemia induces oxidative stress and activates nuclear factor kappa B/Janus kinase - signal transducer and activator of transcription inflammatory signalling. This creates a positive feedback loop to trigger aberrant m⁶A modification and dysregulate specific mRNA stability. Reduced endothelial nitric oxide synthase mRNA expression suppresses NO production; disordered vascular endothelial growth factor/fibroblast growth factor transcripts together with disturbed long non-coding RNAs-miRNA networks promote endothelial-mesenchymal transition and endothelial dysfunction. Downregulated insulin receptor substrate 1/protein kinase B mRNA impairs insulin signalling. Decreased expression of mitochondrial biogenesis genes leads to mitochondrial dysfunction and apoptosis of vascular endothelial cells and vascular smooth muscle cell. Aberrant splicing of carnitine palmitoyltransferase 1A/acyl-CoA dehydrogenase very long chain mRNA disrupts fatty acid oxidation and causes lipid accumulation. These sequential pathological changes drive vascular remodelling and accelerate atherosclerosis. SOD: Superoxide dismutase; GPx: Glutathione peroxidase; NF-κB: Nuclear factor kappa B; JAK-STAT: Janus kinase - signal transducer and activator of transcription; VEGF: Vascular endothelial growth factor; FGF: Fibroblast growth factor; eNOS: Endothelial nitric oxide synthase; lncRNA: Long non-coding RNA; IRS-1: Insulin receptor substrate 1; AKT: Protein kinase B; PGC-1α: Peroxisome proliferator-activated receptor gamma coactivator 1-alpha; NRF1: Nuclear respiratory factor 1; TFAM: Mitochondrial transcription factor A; CPT1A: Carnitine palmitoyltransferase 1A; ACADVL: Acyl-CoA dehydrogenase very long chain; EndMT: Endothelial-mesenchymal transition; VECs: Vascular endothelial cells; VSMCs: Vascular smooth muscle cell.

The potential therapeutic value of RNA modification and epigenetic regulation

At present, small-molecule drugs targeting m6A modifiers have demonstrated considerable preclinical therapeutic potential. The specific METTL3 inhibitor STM2457 effectively alleviates diabetic microvascular injury. In an streptozotocin (STZ)-induced diabetic nephropathy model, it mitigates podocyte damage by inhibiting the NLR family pyrin domain containing 3 inflammasome/pyroptosis pathway. In diabetic retinopathy, STM2457 reverses hyperglycaemia-induced retinal vascular endothelial dysfunction and exerts a sustained vasoprotective effect. The FTO inhibitor FB23-2 directly binds to FTO and selectively antagonizes its m6A demethylase activity, thereby inhibiting diabetes-induced abnormal endothelial angiogenesis. Through optimized nano-delivery systems, the targeted enrichment efficiency and in vivo therapeutic efficacy of FB23-2 have been validated in animal models.

Although m6A-targeted interventions have a sound theoretical and experimental foundation, their clinical translation still faces significant bottlenecks. At present, most m6A-targeted drugs remain in the preclinical research stage and have not yet entered clinical trials[105]. Although natural product screening and AI-assisted drug design provide emerging strategies for the development of m6A regulators[106,107], systematic clinical evidence verifying their efficacy and safety is still lacking. Meanwhile, novel gene-editing technologies represented by CRISPR-Cas13 enable precise targeting and regulation of the m6A pathway[108], while their clinical applicability and practical value remain to be further validated.

In a similar manner, in diabetic nephropathy, RBPs act as key regulators of post-transcriptional RNA networks; abnormalities in their expression also contribute to the onset and progression of the disease, through mechanisms involving inflammation, epithelial-mesenchymal transition and mitochondrial dysfunction, amongst others[109]. Moreover, it is important to note that epigenetic regulation is not limited to RNA modifications, but also encompasses histone modifications and DNA methylation. These modifications have the capacity to indirectly influence RNA homeostasis by modulating the transcription of genes associated with RNA metabolism. For instance, co-activator-associated arginine methyltransferase 1, functioning as a histone arginine methyltransferase, contributes to epigenetic regulation by modulating post-translational methylations and plays a significant role in diseases such as diabetes[110]. Research suggests that the active ingredients in traditional Chinese medicine may exert protective effects at multiple levels - including cell proliferation and oxidative stress - by regulating molecular functions such as histone deacetylation and the binding of DNA transcription factors[111]. The vasoprotective effects of epigenetic drugs, such as demethylases or histone deacetylase inhibitors, in models of diabetes with comorbidities are currently being investigated, providing a direction for the development of novel therapeutic strategies.

The future prospects of multi-targeted combination therapy and precision medicine

The clinical translation of precision medicine hinges on precise disease stratification, dynamic assessment, and individualised therapeutic follow-up; this process relies heavily on a sensitive and stable biomarker system. With their excellent structural stability in the circulatory system, EVs can effectively protect encapsulated nucleic acids from degradation by nucleases and accurately reflect the real-time pathological state of parental cells. They thus provide a highly advantageous non-invasive liquid biopsy carrier for elucidating the dynamic pathological characteristics of multi-organ interactive pathologies[112], such as diabetes complicated by sarcopenia. During the progression of hyperglycaemia-mediated muscular degeneration, the miRNA profile carried by EVs undergoes specific remodelling as the condition advances, playing a significant role in the development of sarcopenia. Circulating EV-encapsulated miR-193b is markedly upregulated in patients with type 2 diabetes, and its elevated abundance correlates positively with the degree of muscle mass loss[113]. This observation suggests that circulating EV-derived miR-193b may act as a molecular biomarker for grading the severity of diabetic muscular atrophy. In addition, EVs can serve as an early monitoring marker for the efficacy of targeted drug interventions. Compared with conventional imaging and histological examinations, the expression levels of nucleic acid molecules carried by EVs exhibit sensitive dynamic fluctuations at an earlier stage in response to therapeutic interventions. Following EV-targeted RNA intervention therapy, changes in the ncRNA cargo of peripheral circulating EVs can reflect trends in the improvement of the in vivo pathological microenvironment at an early stage, providing an immediate and precise molecular reference for clinicians to assess treatment response and dynamically optimise dosages and therapeutic regimens[114].

On this basis, it is evident that the development of diabetes complicated by sarcopenia is associated with imbalances in RNA metabolic networks at multiple levels and across multiple pathways. Consequently, single-target intervention strategies may yield limited efficacy, while the employment of multi-target combination therapies and the implementation of precision medicine hold greater promise for achieving a comprehensive therapeutic effect. It is evident that the simultaneous targeting of miRNAs and lncRNAs, or the joint regulation of RNA modification and protein translation, can more comprehensively restore the homeostasis of RNA metabolic networks, given the complexity of these comorbidities. For instance, resveratrol has been shown to influence a range of biological processes, including inflammation, oxidative stress, apoptosis, metabolism and cell proliferation. It does so by regulating the expression of miRNAs and lncRNAs in tissue cells under various disease states. This, in turn, affects signalling pathways such as phosphoinositide 3-kinase-AKT, NF-κB, Wnt/β-catenin and autophagy pathways. By modulating biological processes such as inflammation, oxidative stress, apoptosis, metabolism and cell proliferation, resveratrol has the potential to improve the pathological progression of insulin resistance, diabetic nephropathy, liver fibrosis, pulmonary hypertension, myocardial hypertrophy, osteoporosis and various cancers[115]. It has been demonstrated that the combination of aerobic exercise and vitamin D supplementation can influence the expression of certain miRNAs, including miR-33, miR-122, miR-29 and miR-9, within the liver tissue of rats diagnosed with type 2 diabetes. This regulatory effect on miRNAs can, in turn, impact a range of biological processes, such as lipid metabolism, insulin signalling, inflammatory responses, and the regulation of fibrosis. Furthermore, it has been demonstrated that these substances can enhance hepatic insulin resistance and lipid metabolism disorders by modulating downstream genes and signalling pathways, including AMP-activated protein kinase, PGC-1α, AKT, phosphoenolpyruvate carboxykinase and glucose-6-phosphatase[116]. The extant literature demonstrates the merits of multi-target, multi-pathway combined interventions. Conversely, precision medicine strategies advocate the utilisation of individual patients’ RNA expression profiles to select the optimal therapeutic targets, informed by the molecular subtypes of co-occurring vascular lesions. For instance, transcriptomic sequencing analysis has been shown to facilitate the identification of key regulatory genes shared by type 2 diabetes and coronary artery disease, including hypoxia-inducible factor 1 subunit alpha and mothers against decapentaplegic homolog 1, thus providing a foundation for personalised treatment regimens[117]. Moreover, advancements in nanocarrier and gene-editing technologies provide the technical foundation for the targeted delivery and highly efficient action of RNA-based therapies. For instance, the encapsulation of miRNA mimics or ASOs within nanoparticles facilitates their targeted accumulation in muscle or vascular tissues, thereby enhancing therapeutic efficacy and reducing adverse effects. Despite the considerable potential of RNA-targeted therapies in the context of diabetes and sarcopenia comorbidity, there are still several issues to be resolved in order to ensure the precise delivery of targeted therapeutic agents. These issues include the efficiency of RNA drug delivery, tissue distribution specificity and pharmacokinetic barriers.

Skeletal muscle and vascular tissues exhibit significant differences in terms of blood supply, cellular composition and barrier structures. In the context of comorbidity, local blood perfusion is affected by chronic inflammation and microcirculatory disorders, which complicates the uniform distribution of RNA drugs across atrophic and dysfunctional regions. It is evident that the low endocytic activity of skeletal muscle cells and the barrier function of vascular endothelial cells have the capacity to affect the uptake of RNA carriers, thereby limiting therapeutic efficacy. Consequently, the utilisation of specialised delivery systems, such as muscle-targeted lipid nanoparticles and vascular endothelial-selective delivery, is imperative[118].

It is evident that chronic inflammation and an abnormal metabolic environment can exert a significant influence on the pharmacokinetic behaviour of RNA therapeutics. It is evident that alterations in the composition of plasma proteins, the activation of the complement system, and modifications in the proportion of immune cells have the capacity to influence the circulation time of RNA nanocarriers[119]. For instance, during the process of circulation, lipid nanoparticles have been observed to adsorb apolipoproteins, thereby forming a protein crown that serves to modify the interaction between the particles and target cell receptors[120]. The clearance of exogenous nucleic acid carriers by macrophages in the liver and spleen also reduces the effective drug dose in target tissues, thereby diminishing therapeutic efficacy; furthermore, RNA itself can activate the innate immune system and trigger an inflammatory response. In addition, the immunogenicity of the delivery vehicle must be taken into account; for example, whilst PEGylated lipids can prolong circulation time, they may induce anti-PEG antibodies, leading to accelerated clearance from the bloodstream[121]. Current clinical data indicate that unmodified mRNA vaccines are ineffective; repeated injections of mRNA-lipid nanoparticle are affected by the immune response of B-lymphocyte subsets, resulting in reduced efficacy in mice; an mRNA therapy for ornithine transcarbamylase deficiency was also discontinued due to poor pharmacokinetics and safety concerns; and early ASO formulations exhibited hepatotoxicity, thereby limiting their clinical application. If these challenges can be addressed one by one, RNA delivery therapy could serve as a long-acting, convenient and targeted treatment.

Single-cell RNA sequencing technology facilitates high-resolution analysis of the heterogeneity of immune cells and molecular interaction networks within the transplant microenvironment, offering novel perspectives for the development of precision intervention strategies[122].

In the future, the integration of multi-omics data, dynamic monitoring technologies and advanced drug delivery systems to realise a bidirectional RNA delivery system that promotes both muscle regeneration and vascular repair is expected to drive the development of precision medicine strategies based on RNA metabolic networks. These strategies are predicted to provide more effective and safer treatment options for patients with diabetes complicated by sarcopenia (Figure 4).

TECHNICAL CHALLENGES AND FUTURE DIRECTIONS IN RESEARCH ON RNA METABOLIC NETWORKS

Applications of high-throughput and single-cell RNA analysis technologies

Whole-transcriptome sequencing (RNA-seq) provides a powerful tool for comprehensively analysing global changes in RNA metabolic networks associated with vascular lesions in diabetes complicated by sarcopenia. Using RNA-seq, the expression profiles of mRNA, lncRNA, miRNA and circRNA can be detected simultaneously, thereby identifying key regulatory nodes. For example, in a rat model of type 2 diabetes, qRT-PCR was used to analyse miRNA expression in the liver. It was found that aerobic exercise and high-dose vitamin D significantly downregulated miR-122 and miR-9, whilst upregulating miR-33 and miR-29, thereby improving insulin resistance and dyslipidaemia by regulating pathways associated with lipid metabolism, inflammatory responses and fibrosis[116]. In a similar vein, research into ischaemic cardiomyopathy has employed single-nucleus RNA sequencing in conjunction with bioinformatic analysis to map the transcriptomic profiles of nearly 100000 cell nuclei. This has resulted in the systematic revelation of differentially expressed gene profiles and altered pathways across various cell types (e.g., cardiomyocytes, fibroblasts, endothelial cells) within the intramyocardial compartment. Consequently, new insights have been provided into the common molecular mechanisms underlying end-stage heart failure and potential therapeutic interventions[123]. Single-cell RNA sequencing technology has the capacity to reveal specific alterations in RNA metabolism within different cell types, including endothelial cells, smooth muscle cells and myosatellite cells. This cellular-level resolution is crucial for understanding the role of cellular heterogeneity in vascular pathologies, as different cell types may exhibit markedly distinct RNA metabolic responses to pathological stimuli such as hyperglycaemia. Furthermore, spatial transcriptomics offers a novel perspective on the heterogeneous distribution of RNA metabolic imbalances within the vascular microenvironment. This technique combines gene expression data with anatomical location within the tissue, revealing the spatial expression patterns of RNA molecules across different layers of the vascular wall (such as the intima, media and adventitia), thereby enabling more precise localisation of key regions where RNA metabolic networks are disrupted. For instance, in atherosclerosis-associated endothelial cell damage, the lncRNA lung cancer associated transcript 1 functions as a ceRNA by binding to hsa-miR-6776-5p via a sponge-like mechanism, thus competitively counteracting the inhibitory effect of this miRNA on the target gene, leucine-rich repeat-containing protein 25 mRNA. Consequently, lung cancer associated transcript 1 exerts a protective role in oxidative stress-induced damage to human coronary artery endothelial cells[124]. The combined application of these high-throughput technologies offers unprecedented opportunities for the systematic analysis of imbalances in RNA metabolic networks in co-morbid vascular lesions, ranging from the global to the single-cell level and from the molecular to the spatial scale.

Challenges in in vivo modelling and clinical validation

At present, there is a lack of ideal animal models capable of simultaneously simulating the vascular complications associated with the co-morbidity of diabetes and sarcopenia, which severely limits the validity of verifying the mechanisms underlying RNA metabolic networks. Existing animal models of diabetes, such as the STZ-induced type 1 diabetes model or the high-fat diet combined with STZ-induced type 2 diabetes model, whilst capable of simulating hyperglycaemia and partial muscle atrophy phenotypes[125,126], often struggle to simultaneously reproduce the full characteristics of sarcopenia - such as reduced muscle strength and functional decline - as well as the complexity of vascular pathologies. Currently, most models of diabetes complicated by sarcopenia involve first establishing a diabetes model and then superimposing the sarcopenia phenotype. The following methods are commonly used in animal models to study this complication: (1) Use of db/db mice[127]; (2) db/db mice combined with activity restriction, such as sedentary mice[128]; (3) STZ induction alone; and (4) Induction via high-fat diet combined with low-dose STZ. Among these animal models, the use of db/db mice offers the advantages of being a well-established and convenient approach; they not only exhibit diabetic and muscular phenotypes but also display marked vascular dysfunction[129]. However, it should be noted that db/db mice do not equate to naturally occurring type 2 diabetes, and the causes of muscle loss must be clarified - for instance, distinguishing between diabetes-induced muscle loss and that resulting from obesity and reduced physical activity. Induction using STZ alone is suitable for T1DM, whilst the animal model established using HFD + STZ more closely approximates the metabolic environment of T2DM. However, in studies focusing on vascular damage, it should be noted that HFD itself is a factor contributing to vascular damage, which may easily be confused with diabetes-induced vascular damage.

The prevailing assumption in the field regarding comorbidity is that it represents the sum of the two conditions, with much of the existing evidence deriving from studies focusing on a single disease. It is not entirely accurate to equate muscle atrophy in diabetes with a sarcopenia model, or to attribute glucose metabolism abnormalities in sarcopenia to prediabetes. In the case of a single disease, all diagnostic criteria for that disease must be met simultaneously before the condition can be classified as a comorbid condition; for example, in the context of sarcopenia, the diagnosis of diabetes should be confirmed primarily on the basis of blood glucose or glycated haemoglobin levels. In the context of diabetes, monitoring muscle mass (using the skeletal muscle mass index and strength tests such as grip strength to determine the stage of sarcopenia) should be carried out, whilst taking care to rule out muscle weakness caused by other factors (such as spinal cord lesions or malnutrition). The lack of direct evidence means that the following key questions remain unsupported by robust evidence: Are the activation thresholds for signalling pathways lower in a state of comorbidity than in a single disease? Are there specific molecular switches in a state of comorbidity? Furthermore, these models differ from human disease in terms of disease progression, metabolic characteristics and vascular pathological changes, leading to challenges in translating the regulatory mechanisms of RNA metabolic networks identified in animal models to clinical practice.

The standardised collection and analysis of RNA expression profiles in clinical samples also present numerous difficulties. At the level of pathology and intervention, heterogeneity within patients’ own diseases also constitutes a significant source of interference - individual differences such as disease type, stage, duration, comorbidities, lifestyle habits, and blood glucose control can all significantly influence the results of RNA and m6A modification assays. Taking the pathological mechanisms of inflammatory diseases as an example, chronic inflammation can significantly alter the m6A modification profile - METTL14 and m6A levels are reduced in peripheral blood mononuclear cells from patients with rheumatoid arthritis, and m6A participates in the regulation of inflammation by modulating the mRNA stability of TNF-α induced protein 3[130]. In metabolic diseases, the lncRNA TMEM9B antisense RNA 1 is significantly downregulated in skeletal muscle from male patients with T2DM and individuals with sarcopenia[105]. It should be noted that metabolic drugs may also indirectly influence m6A modification by altering metabolite levels such as α-ketoglutarate; targeting malate dehydrogenase 2 induces the accumulation of α-ketoglutarate, thereby affecting the activity of the m6A demethylase ALKBH5[131]. Therefore, multiple confounding factors should be comprehensively taken into account during the standardized collection and analysis of RNA expression profiles from clinical samples. For example, in research on diabetic retinopathy, the construction of RNA methylation regulatory networks requires consideration of multiple variables, including patients’ blood glucose control levels and disease duration[44]. At the technical level, methods based on anti-m6A antibodies are susceptible to interference from factors such as nucleic acid secondary structure, RNA contamination, and bacterial contamination[132]. Furthermore, given the complex bidirectional regulatory network between m6A and ncRNA[133], a single detection index cannot fully characterize the actual landscape of post-transcriptional regulation. In addition, variations in the processing and storage conditions of in vitro samples constitute a significant source of bias, directly affecting the reliability and cross-comparability of sequencing data.

In the future, multicentre, large-scale cohort studies will need to be established to confirm the prognostic value of RNA metabolic biomarkers and the clinical feasibility of therapeutic targets. By adopting strictly standardised sample collection and processing protocols, combined with detailed clinical data recording, the impact of confounding factors can be minimised, thereby enhancing the reliability and reproducibility of research findings. For example, in studies of hormone-induced avascular necrosis of the femoral head, the construction of ceRNA networks has provided new strategies for early diagnosis and targeted therapy; however, their clinical validation still requires support from large-scale cohort studies[134]. Only through rigorous clinical validation can findings from basic research be truly translated into clinical applications.

The potential of systems biology and artificial intelligence in the study of RNA networks

Systems biology offers the dual benefits of both top-level strategic planning and subsequent simulation and modelling. By utilising network analysis and dynamical models to coordinate the overall research design, this approach enables planning for sequencing types, sample grouping, and data analysis methods, thereby charting a course of action for the entire omics study. Guided by this approach, single-cell RNA sequencing and RNA interactomics serve as molecular-level reconnaissance tools, capturing transcriptional information from skeletal muscle, vascular endothelium, and infiltrating immune cells, whilst simultaneously recording the actual interactions between RNA and its binding proteins. This yields extensive raw sequencing datasets documenting the interactive states of various molecules in a pathological environment. Building on the raw data obtained through this reconnaissance, the regulatory patterns of RBPs are analysed using systems-biology network analysis approaches. To illustrate this point, consider the realm of research focused on the systematic identification and analysis of RBPs. In this domain, the construction of RNA-protein interaction networks that span various species, cellular states and functional modules has brought to light the fact that RNA binding constitutes a system-level process that is widely embedded, dynamically regulated and functionally diverse within cellular molecular networks[135]. This network-based perspective facilitates the identification of key nodes and modules within the network, which may act as central drivers of disease onset and progression, as well as potential therapeutic targets.

The field of artificial intelligence (AI), and more specifically machine learning and deep learning, has exhibited considerable potential in the domains of RNA sequence feature extraction, target prediction and the analysis of drug-target interactions. It has been demonstrated that AI algorithms have the capacity to automatically learn complex patterns from vast amounts of RNA sequencing data. This ability enables the identification of disease-associated RNA features and the prediction of the potential functions of ncRNAs, including lncRNAs and circRNAs. For instance, in research on acute ischaemic stroke, bioinformatics methods combined with multi-database analysis were used to successfully construct a circRNA-mediated immune-related ceRNA network and identify circulating circRNAs (circPTP4A2 and circTLK2) with diagnostic potential[136]. The utilisation of AI has the capacity to expedite the identification and optimisation of RNA-targeted therapeutic interventions. This is achieved through the prediction of the binding affinity of small-molecule or nucleic acid drugs to RNA targets, thereby facilitating the identification of candidate molecules that possess therapeutic potential.

Subsequently, through integrated multi-omics analysis encompassing transcriptomics, proteomics, metabolomics and epigenomics, multidimensional molecular clues were pieced together to reconstruct how dysregulation of RNA expression progressively leads to protein abnormalities, metabolic imbalances and local immune dysregulation, ultimately revealing the complete pathological causal chain underlying the concurrent damage to muscle and blood vessels. For instance, in research on Alzheimer’s disease, multi-omics integration has revealed a close association between gut microbiota dysbiosis and abnormalities in glycerophospholipid metabolism. This has helped to systematically elucidate the complex mechanisms by which the gut microbiota influences Alzheimer’s disease pathology via the metabolic-immune axis[137]. The assembled global molecular regulatory network is fed back into the systems biology model to conduct simulation analyses, modelling the dynamic changes across the entire pathological network following intervention at core targets, and identifying the optimal targeted immunomodulatory strategies and dosing regimens. Based on the model's predictive results, we will conduct experimental studies, including the engineering of engineered EV vectors, cellular functional assays and in vivo pharmacodynamic evaluations in animals, to validate the theoretically predicted therapeutic effects using in vitro and in vivo data.

This will establish a complete closed-loop research process, ranging from the mining of basic RNA data, mechanism elucidation and target screening, to the design of targeted immune regulation and the validation of in vivo therapeutic translation. For example, in research on myocardial infarction, the key implementation points of this translational pathway were systematically summarised, confirming that the bioinformatics-based identification of immune targets, combined with carrier-mediated delivery systems for in vivo intervention, constitutes a reliable research and development approach for cardiovascular inflammatory diseases[138]. This comprehensive research paradigm effectively bridges the gap between basic mechanistic research and translational medicine, enhancing the forward-looking nature and clinical translational value of this study, and providing theoretical support for the development of precision treatments for vascular complications associated with diabetes and sarcopenia.

Based on the aforementioned multi-omics and systems biology research perspectives, this paper proposes a conceptual framework - the multi-organ extracellular vesicle interaction hub - to guide future research directions. This framework posits that EVs and the ncRNAs they carry constitute a systemic communication network, facilitating information exchange between skeletal muscle, vascular endothelium, adipose tissue, the liver, and the gut microbiota in diabetic sarcopenia. Within this framework, pathogenic RNA signals generated by a particular organ (such as muscle-derived miR-499-5p) can be transmitted via extracellular vesicles to distal tissues (such as the vascular endothelium); conversely, signals associated with metabolic dysregulation produced by other organs (such as fat-derived miR-27a and liver-derived miR-122) can in turn regulate the muscle–vascular axis. Traditional views have focused solely on ‘bidirectional interactions’ between muscle and blood vessels. In contrast, this multi-organ perspective further expands our current understanding: The vicious cycle formed by diabetic sarcopenia combined with vascular pathology is not confined to a closed loop between two organs, but rather constitutes an open multi-organ network; extracellular vesicle-mediated RNA interactions amplify systemic metabolic dysregulation. The realisation of this framework requires the integration of multi-omics technologies, lineage-tracing systems for extracellular vesicles, and animal models capable of tissue-specific extracellular vesicle clearance and reconstitution. This theoretical framework establishes a verifiable foundation for subsequent research, elucidating the systemic pathophysiology of diabetic sarcopenia and its vascular complications, and providing insights into the development of targeted therapies.

CONCLUSION

The comorbidity of diabetes and sarcopenia is not merely the superimposition of two conditions, but rather a vicious cycle of pathological interactions arising from a global imbalance in the body’s RNA metabolic network. From an expert perspective, this finding extends the understanding of the pathogenesis of vascular lesions beyond traditional metabolic disorders and haemodynamic abnormalities to a deeper level of molecular regulation. ncRNAs, including miRNAs, lncRNAs and circRNAs, play a central role as mediating molecules in this process; they act as signalling hubs, closely linking hyperglycaemia, insulin resistance and muscle atrophy with endothelial dysfunction, inflammatory storms and oxidative stress. Although different studies have focused on specific molecular pathways, there is a consensus that an imbalance in RNA metabolism is the common upstream mechanism driving persistent vascular damage and pathological remodelling. Consequently, simply controlling blood glucose levels or improving muscle mass is unlikely to break this vicious cycle mediated by the RNA network.

In this context, precise intervention strategies that target RNA metabolism have been shown to possess significant therapeutic potential. ncRNA therapeutics, RNA modification and regulation, and multi-target combination therapies have been identified as promising avenues for addressing the underlying causes of RNA metabolic network dysfunction, with the aim of restoring homeostasis. However, it is imperative to acknowledge the persistent discrepancy between fundamental research and its translation into clinical practice. The present body of research is predominantly supported by the utilisation of single cell lines or rudimentary animal models, which fall short in their capacity to emulate the intricacies of comorbidity states in a realistic manner. In future, therapeutic approaches targeting RNA metabolic networks should be rigorously validated in standardised, reproducible animal models of diabetes-sarcopenia comorbidity, and combined screening should be implemented in clinical cohort studies to address concerns regarding the specificity and safety of these targets arising from a lack of validation in clinical samples. Future research must strengthen systems biology analyses, integrate multi-omics data, construct RNA regulatory network maps under conditions of comorbidity, and develop novel delivery systems to enhance the targeting and stability of RNA therapeutics. It is imperative to translate targets related to the RNA metabolic network from the laboratory to the clinical setting, with the ultimate aim of achieving substantial improvements in vascular health and quality of life for patients with diabetes and sarcopenia.