Targeting adipose remodeling: Synergistic mechanisms of drugs and adipose-derived stem cells in obese type 2 diabetes mellitus

Abstract

Type 2 diabetes mellitus, particularly when accompanied by obesity, has become a major global public health burden. Visceral adipose tissue accumulation contributes to insulin resistance, lipotoxicity, and chronic inflammation, thereby accelerating metabolic deterioration. Although pharmacological agents such as pioglitazone and metformin are effective in modulating fat distribution and improving metabolic parameters, their roles in adipose tissue remodeling remain insufficiently elucidated. Recent advances in regenerative medicine have highlighted the therapeutic potential of adipose-derived stem cells, owing to their differentiation capacity, anti-inflammatory secretory profile, and involvement in metabolic homeostasis. This review summarized current pharmacological and stem cell-based strategies targeting adipose tissue dysfunction in patients with obesity and type 2 diabetes mellitus with a particular focus on the mechanistic roles of adipokines, mitochondrial dysfunction, and extracellular matrix remodeling in visceral adipose tissue. It further discussed the potential synergistic benefits of adipose-derived stem cell-based combination interventions. Finally, the review envisioned future directions for integrating molecularly targeted drugs with cell therapies in the personalized management of metabolic disorders.

Key Words: Adipose-derived stem cells; Fat redistribution; Insulin resistance; Metabolic inflammation; Pioglitazone; Type 2 diabetes mellitus

Core Tip: This review explored how pharmacological agents (pioglitazone, metformin, glucagon-like peptide-1 receptor agonist, sodium-glucose cotransporter-2 inhibitors) and adipose-derived stem cells synergistically remodel dysfunctional adipose tissue in obesity-associated type 2 diabetes. It highlighted the roles of adipose browning, mitochondrial restoration, anti-inflammatory paracrine effects, and fat redistribution. By integrating metabolic drug actions with regenerative cell therapies, the review proposed a precision intervention strategy targeting insulin resistance and metabolic inflammation at the tissue level, offering translational insights for individualized treatment of diabetic obesity.

INTRODUCTION

Diabetes has become one of the most severe metabolic disorders worldwide with the incidence of type 2 diabetes mellitus (T2DM), especially when accompanied by obesity, and it is continuing to rise. According to the latest Diabetes Atlas published by the International Diabetes Federation in 2025, approximately 589 million adults aged 20-79 years were living with diabetes globally in 2024, a number projected to increase to 853 million by 2050. The associated healthcare burden is also intensifying with annual diabetes-related expenditures exceeding one trillion United States dollars[1]. A concurrent Lancet systematic analysis based on Global Burden of Disease 2021 data indicated that T2DM accounts for approximately 96% of all diabetes cases and has shown a steady increase from 1990 to 2021. If current trends persist, it is estimated that by 2050, one in ten people worldwide will have diabetes[2].

Compared to overall obesity, visceral adipose tissue (VAT) expansion plays a critical role in promoting insulin resistance, lipotoxicity, and chronic low-grade inflammation. Recent studies have shown that VAT has a high fatty acid mobilization rate, releasing large quantities of free fatty acids (FFAs) that reach the liver directly via the portal vein, thereby triggering hepatic glucose production, ectopic fat deposition, and systemic inflammatory amplification[3,4]. Moreover, VAT remodeling is characterized by adipocyte hypertrophy, hypoxia, and stromal fibrosis, all of which further impair insulin signaling pathways, ultimately accelerating T2DM progression and increasing the risk of cardiovascular complications.

The insulin sensitizer pioglitazone [a peroxisome proliferator-activated receptor gamma (PPARγ) agonist] has been shown to reduce epicardial and pericardial fat area by approximately 9% over 24 weeks in patients with T2DM along with simultaneous improvements in left ventricular diastolic function and systemic insulin sensitivity[5]. Metformin promotes thermogenesis in brown adipose tissue via the adenosine monophosphate-activated protein kinase (AMPK)/fibroblast growth factor 21 pathway, contributing to moderate reductions in body weight and VAT[6]. However, both drugs primarily regulate adipocyte metabolism rather than directly reversing adipose tissue fibrosis and immune imbalance, thus offering limited efficacy in long-term VAT remodeling. Additionally, high doses or prolonged use may be accompanied by adverse effects such as fluid retention and gastrointestinal discomfort.

In recent years adipose-derived stem cells (ADSCs) have demonstrated distinct advantages in enhancing tissue insulin sensitivity and remodeling the adipose microenvironment due to their robust differentiation potential and immunomodulatory properties. Evidence suggests that the paracrine effects of ADSCs when combined with insulin-sensitizing agents such as pioglitazone and metformin may synergistically reduce VAT volume and inflammatory activity, enhance thermogenic and mitochondrial function in brown/beige adipose tissue, and effectively delay T2DM progression through combined cellular and pharmacological actions[7,8].

Based on this rationale, the present review focused on the vertical biological axis of adipose tissue remodeling-insulin resistance-stem cell regulation. Within this context we systematically examined the complementary mechanisms and translational potential of pharmacological and ADSC-based therapies in T2DM and obesity, aiming to establish a solid theoretical foundation for the development of future combined and precision-targeted drug + cell intervention strategies.

ADIPOSE TISSUE DYSFUNCTION

Metabolic differences between visceral and subcutaneous fat and their significance

In the context of obesity, VAT and subcutaneous adipose tissue (SAT) exhibit markedly different metabolic profiles. Studies have shown that VAT secretes significantly higher levels of proinflammatory factors compared with SAT. As reported by Kahn et al[9], conditioned media derived from VAT of individuals with obesity is enriched with proinflammatory cytokines such as interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α) as well as matrix remodeling proteins. These secretions can impair insulin sensitivity in hepatocytes and skeletal muscle cells and activate inflammatory signaling pathways.

In contrast, adipocytes in SAT are smaller, more insulin-sensitive, and secrete greater amounts of protective adipokines, such as adiponectin. According to a review by Dhokte and Czaja[3], SAT plays a buffering role by absorbing postprandial circulating FFAs, whereas VAT adipocytes are hypertrophic, prone to insulin resistance, exhibit elevated basal lipolysis, and have a reduced capacity to take up FFAs. Additionally, VAT is more susceptible to hypoxia, mitochondrial dysfunction, and fibrosis, leading to excessive production of reactive oxygen species (ROS) and macrophage recruitment, which further amplifies local inflammation[10]. Taken together, the differences between VAT and SAT in adipokine secretion, insulin sensitivity, and inflammatory status make VAT a greater contributor to systemic insulin resistance and the progression of T2DM.

Under normal physiological conditions SAT is the primary source of adiponectin, a key adipokine that enhances insulin sensitivity. In contrast, VAT secretes higher levels of proinflammatory adipokines such as resistin and visfatin, and in the obese state VAT becomes a major source of TNF-α and IL-6, exacerbating systemic inflammation and insulin resistance[10,11]. Regarding insulin sensitivity, SAT adipocytes are smaller and highly responsive to insulin, efficiently absorbing and storing FFAs. VAT adipocytes, however, are larger and less insulin-sensitive, resulting in greater FFA flux into the portal circulation and contributing to insulin resistance in the liver and skeletal muscle[3].

Furthermore, VAT is more prone than SAT to chronic low-grade inflammation, characterized by abundant macrophage infiltration and the formation of crown-like structures and accompanied by sustained secretion of inflammatory mediators such as TNF-α and IL-6[9,10]. In contrast, SAT contains fewer inflammatory cells and exhibits a relatively mild inflammatory response. At the metabolic level expansion of VAT adipocytes is often associated with tissue hypoxia and mitochondrial dysfunction, leading to increased ROS production and activation of cellular stress responses. In contrast, mitochondrial function in SAT remains relatively intact, supporting superior metabolic regulation[3,10]. These structural and functional disparities underscore the pivotal role of VAT in obesity-associated T2DM pathogenesis, identifying it as a hidden culprit driving insulin resistance and metabolic dysregulation (Table 1).

Table 1 Metabolic and inflammatory characteristics of visceral adipose tissue vs subcutaneous adipose tissue.

Dimension	VAT	SAT
Adipocyte size/insulin sensitivity	Large adipocytes, insulin-resistant	Small adipocytes, insulin-sensitive
Proinflammatory/anti-inflammatory profile	IL-6 (↑), TNF-α (↑), resistin (↑)	Adiponectin (↑)
Lipolysis/FFA output	High basal lipolysis, FFA enters portal circulation first	Lower lipolysis, buffers postprandial FFA
Hypoxia/ROS and mitochondrial function	Marked hypoxia, ROS (↑), mitochondrial dysfunction	Relatively preserved mitochondrial function
ECM remodeling and fibrosis	ECM deposition (↑), prone to fibrosis	Mild
Macrophage infiltration	Increased M1 macrophages, crown-like structures prominent	Lower inflammatory infiltration

VAT: Visceral adipose tissue; SAT: Subcutaneous adipose tissue; IL-6: Interleukin 6; TNF-α: Tumor necrosis factor alpha; FFA: Free fatty acid; ROS: Reactive oxygen species; ECM: Extracellular matrix; M1: Classically activated macrophage (proinflammatory phenotype).

Roles of adipokines and inflammatory signaling pathways

Obesity leads to the excessive secretion of proinflammatory adipokines from adipose tissue, triggering chronic low-grade systemic inflammation that interferes with insulin signaling pathways. In the obese state elevated levels of leptin can stimulate monocytes and macrophages to secrete proinflammatory cytokines such as TNF-α and IL-6 while also enhancing its own proinflammatory signaling[12]. This forms a positive feedback loop in which inflammatory mediators like TNF-α and IL-1β further stimulate adipocytes to produce more leptin, perpetuating the activation of inflammatory signaling pathways and sustaining chronic inflammation[12]. In contrast, adiponectin is a prototypical anti-inflammatory and insulin-sensitizing adipokine that under healthy conditions enhances insulin sensitivity in peripheral tissues by activating pathways such as AMPK and suppressing nuclear factor kappa B (NF-κB) signaling. However, in obesity adiponectin levels decline, diminishing its protective effects and thereby exacerbating defects in insulin signaling[10].

Moreover, obesity is often accompanied by elevated plasma levels of FFAs, which can activate toll-like receptor 4 (TLR4)/NF-κB and c-Jun N-terminal kinase (JNK) pathways in adipose tissue, liver, and skeletal muscle cells. This activation leads to serine phosphorylation of insulin receptor substrate 1 [IRS-1(Ser-P)], weakening downstream phosphoinositide 3-kinase (PI3K)/protein kinase B (Akt) signaling and ultimately resulting in insulin resistance[3,13]. In summary leptin, adiponectin, FFAs, and proinflammatory cytokines such as TNF-α and IL-6 interact through multiple intersecting inflammatory pathways, collectively disrupting downstream insulin receptor signaling and contributing to systemic metabolic dysregulation[12,13].

Further studies have shown that not only do the levels of adipokines and inflammatory mediators change significantly in obesity, but their coordinated signaling interactions also play crucial roles in obesity-related insulin resistance. Elevated leptin levels can continuously stimulate macrophages to express activation markers such as cluster of differentiation (CD)25 and CD69 and to secrete TNF-α and IL-6 while these inflammatory cytokines in turn upregulate leptin expression, reinforcing a stable positive feedback loop within the leptin-inflammation axis and maintaining chronic inflammation in adipose tissue[12]. Meanwhile, the significant reduction of adiponectin in obesity weakens its original functions of regulating glucose and lipid metabolism via AMPK and PPAR-α activation and of suppressing NF-κB and JNK activity, thereby rendering insulin signaling more vulnerable to inflammatory inhibition[10].

Additionally, high FFA levels, a hallmark of obesity, can activate TLR4-NF-κB and JNK pathways in the liver and skeletal muscle, leading to increased activity of IκB kinase (IKK) and JNK. This results in IRS-1(Ser-P), thereby impairing PI3K/Akt-mediated insulin signaling[3,13]. Proinflammatory cytokines TNF-α and IL-6 are not only produced by hypertrophic adipocytes but also by infiltrating M1-polarized macrophages. TNF-α directly activates IKKβ/NF-κB and JNK signaling, promoting IRS-1 phosphorylation and weakening insulin receptor signaling capacity. IL-6 on the other hand indirectly suppresses insulin signaling by inducing suppressor of cytokine signaling 3 (SOCS3) expression[14,15]. These signaling pathways intersect and amplify one another in a cascade-like fashion, ultimately resulting in systemic impairment of insulin signaling, a molecular basis for obesity-associated insulin resistance.

Lipotoxicity and ectopic fat deposition

Lipotoxicity refers to cellular dysfunction caused by the abnormal accumulation of fatty acids in non-adipose tissues[16]. In obesity impaired adipose tissue function leads to excessive release of FFAs, which undergo excessive β-oxidation in mitochondria, generating large amounts of ROS and metabolic intermediates that trigger endoplasmic reticulum (ER) stress, oxidative stress, and inflammatory responses[17,18]. For instance, the saturated fatty acid palmitic acid can induce ER stress and elevate ROS levels in pancreatic β cells, impairing mitochondrial function and ultimately triggering β cell apoptosis and disrupting insulin secretion[19].

Prolonged elevation of FFAs also leads to the accumulation of lipid metabolites such as long-chain acyl-CoA, diacylglycerol, and ceramide in hepatic and muscular tissues. These lipids activate protein kinase C and protein phosphatase 2A, thereby inhibiting the Akt/IRS signaling pathway and directly impairing insulin signaling[3,13]. In addition, excess lipids are ectopically deposited in the liver, skeletal muscle, and pancreatic β cells, causing organ damage: Lipid overload in the pancreas leads to β cell apoptosis and secretory dysfunction; intramuscular lipid accumulation reduces glucose uptake capacity; and hepatic steatosis induces cellular stress and disrupts glucose metabolism[12]. Together, these mechanisms contribute to a deleterious lipotoxic effect central to the development of obesity-associated T2DM[12].

Under high-fat conditions mitochondrial β-oxidation becomes overloaded, resulting in excessive ROS production that triggers oxidative stress, disrupting cellular structural integrity and functional homeostasis. ROS not only directly damage organelle membranes but also activate classical inflammatory pathways such as NF-κB and JNK, thereby enhancing the secretion of proinflammatory factors and promoting apoptosis[20]. Concurrently, excessive intake of saturated fatty acids induces ER stress by increasing the protein-folding burden, activating the unfolded protein response, and ultimately triggering programmed cell death, a phenomenon particularly pronounced in pancreatic β cells, leading to marked suppression of insulin synthesis and secretion[21,22].

As adipose tissue gradually reaches its storage capacity, FFAs are transported and ectopically deposited in non-adipose tissues such as the liver, skeletal muscle, and pancreas, further increasing the metabolic burden. Lipid accumulation in the pancreas impairs β cell function and survival. Lipid buildup in skeletal muscle inhibits glucose uptake and oxidation. Hepatic steatosis enhances gluconeogenesis and may lead to hepatocellular necrosis[19]. These pathological processes collectively drive systemic insulin resistance and glucose metabolic disturbances, forming a key mechanistic foundation for obesity-related type 2 diabetes.

PHARMACOLOGICAL STRATEGIES TARGETING FAT DISTRIBUTION

Pioglitazone

Pioglitazone is a thiazolidinedione (TZD) class insulin sensitizer, and its primary mechanism of action is the activation of PPARγ[23]. PPARγ is a master regulator of adipocyte differentiation. Upon activation by pioglitazone PPARγ promotes the differentiation of preadipocytes and increases the formation of new small adipocytes (adipocyte hyperplasia), particularly within SAT[24]. Although pioglitazone treatment may increase overall fat mass, this fat is preferentially redistributed into the subcutaneous fat depot while visceral and ectopic fat deposition is relatively reduced[25,26]. Some studies have shown that pioglitazone shifts fat from ectopic/visceral sites such as the liver and muscle to subcutaneous storage, thereby reversing lipotoxic metabolic damage[27]. In other words, pioglitazone converts an unhealthy apple-shaped fat distribution into a metabolically safer pear-shaped pattern (considered a key mechanism by which it improves insulin resistance)[25].

Through PPARγ-mediated adipose tissue remodeling, pioglitazone exerts multiple effects to enhance insulin sensitivity[26,28]. On one hand, the expansion of the subcutaneous fat depot increases the body’s capacity for safe lipid storage, lowering plasma FFAs levels and reducing ectopic fat accumulation, which alleviates insulin resistance in muscle and liver[25]. On the other hand, pioglitazone upregulates beneficial adipokines such as adiponectin while downregulating adverse ones like leptin[29]. Increased adiponectin enhances hepatic and muscular insulin responsiveness and exerts anti-inflammatory effects. In fact, TZD treatment has been shown to improve the antilipolytic effect of insulin on adipose tissue, significantly reducing excessive lipolysis and FFA release[29]. Furthermore, PPARγ activation modulates immune cell phenotypes; pioglitazone has been shown to activate anti-inflammatory pathways and suppress proinflammatory signals such as NF-κB, thereby attenuating obesity-related chronic adipose tissue inflammation[30].

In summary, pioglitazone reprograms adipose tissue by promoting adipocyte differentiation and redistributing fat, effectively transforming it into a metabolic buffer for insulin action, thereby improving the overall metabolic status of patients with T2DM[27]. Figure 1 illustrates the mechanism by which pioglitazone activates PPARγ, promotes preadipocyte differentiation, increases subcutaneous fat accumulation, and shifts fat away from visceral depots, ultimately enhancing insulin sensitivity and reducing inflammation.

Figure 1 Mechanistic illustration of pioglitazone-induced adipose tissue remodeling via peroxisome proliferator-activated receptor gamma activation. This schematic illustrates the core mechanisms by which pioglitazone exerts its insulin-sensitizing effects. Peroxisome proliferator-activated receptor gamma, a key transcription factor that promotes the differentiation of preadipocytes into mature adipocytes, particularly in subcutaneous fat depots. This expansion of subcutaneous fat enhances lipid storage capacity, facilitating the redistribution of lipids away from visceral and ectopic sites (e.g., liver, muscle). The resulting reduction in lipotoxicity and inflammation improves insulin sensitivity and metabolic homeostasis. Additionally, peroxisome proliferator-activated receptor gamma activation upregulates adiponectin and suppresses proinflammatory pathways, contributing to the anti-inflammatory effects of pioglitazone. PPARγ: Peroxisome proliferator-activated receptor gamma.

Metformin

Metformin is a first-line treatment for T2DM known for its favorable safety profile and significant glucose-lowering effect. It is frequently associated with modest weight loss or at minimum weight neutrality[31]. The primary mechanism of metformin centers on the regulation of cellular energy metabolism, primarily through the activation of AMPK[32]. Activated AMPK functions as a cellular energy sensor, promoting catabolic pathways while inhibiting anabolic ones. Specifically, AMPK activation suppresses the mechanistic target of rapamycin complex 1 signaling pathway[33]. Mechanistic target of rapamycin is a critical regulator of cell growth and lipogenesis; inhibition by metformin reduces hepatic lipid synthesis and export while also suppressing gluconeogenesis, thereby improving hepatic glucose homeostasis and glycemic control[29].

Within adipose tissue AMPK activation inhibits lipogenic enzymes (such as fatty acid synthase) and promotes fatty acid oxidation. Several studies have shown that metformin can promote the browning of white adipose tissue (WAT) and enhance thermogenesis by upregulating the expression of uncoupling protein 1 (UCP1) and UCP3 (among other mechanisms), thereby increasing fatty acid combustion for heat production[25,34]. This effect is particularly significant in visceral fat and is believed to underlie the preferential reduction of visceral adiposity by metformin[25]. Notably, some reports indicate that metformin can substantially reduce hepatic fat content without significantly lowering overall body weight. This suggests that metformin can remodel the metabolism of visceral and ectopic fat, reducing its accumulation and thereby improving metabolic health even in the absence of major weight loss[35,36].

Beyond its impact on energy metabolism pathways, metformin also directly improves the inflammatory and endocrine function of adipose tissue. Chronic inflammation is a key contributor to insulin resistance in VAT. Metformin has been shown to suppress activation of the NOD-like receptor protein 3 (NLRP3) inflammasome in VAT, leading to reduced expression of downstream proinflammatory cytokines such as TNF-α, IL-6, IL-1β, and monocyte chemoattractant protein 1 (MCP-1)[37]. This underscores the significant anti-inflammatory effects of metformin and its capacity to improve the adipose microenvironment in obesity and diabetes. Additionally, metformin modulates lipid mobilization within adipose tissue. It partially inhibits catecholamine-induced lipolysis by reducing cyclic adenosine monophosphate (cAMP) production and protein kinase A (PKA) activity, which decreases hormone-sensitive lipase phosphorylation on lipid droplets, thereby suppressing excessive lipolysis[29]. This results in lower circulating FFA levels and helps mitigate FFA-induced disruption of insulin signaling (lipotoxicity). Equally important, metformin influences adipokine secretion profiles. For instance, it has been shown to reduce plasma levels of plasminogen activator inhibitor-1 and leptin. Although findings regarding its effect on adiponectin levels vary across studies, animal experiments have demonstrated that metformin upregulates adiponectin receptors AdipoR1 and AdipoR2[29,38].

Overall, metformin reprograms the metabolic functions of adipocytes and hepatocytes through AMPK activation, reducing lipid synthesis and accumulation, enhancing fatty acid oxidation, and alleviating inflammation in adipose tissue. These multitarget effects not only improve glycemic control but also enhance adipose tissue function and distribution in patients with obesity and T2DM, offering long-term benefits for insulin sensitivity and cardiometabolic risk reduction. Figure 2 illustrates the schematic model of the action of metformin on the AMPK/mechanistic target of rapamycin signaling pathway.

Figure 2 Metformin-mediated metabolic reprogramming via adenosine monophosphate-activated protein kinase/mechanistic target of rapamycin complex 1 pathway. This diagram illustrates the core mechanisms by which metformin exerts its metabolic benefits in obesity and type 2 diabetes. Metformin activates adenosine monophosphate-activated protein kinase, a key energy sensor that downregulates anabolic processes and promotes catabolism. Activated adenosine monophosphate-activated protein kinase inhibits the mechanistic target of rapamycin complex 1, thereby suppressing hepatic gluconeogenesis and lipogenesis while enhancing fatty acid oxidation. These actions collectively contribute to reduced hepatic glucose output, decreased lipid accumulation, and diminished visceral fat mass. Moreover, through its impact on adipocyte metabolism and inflammation, metformin supports insulin sensitivity and improves adipose tissue function. AMPK: Adenosine monophosphate-activated protein kinase; mTORC1: Mechanistic target of rapamycin complex 1.

Combination therapy

Since individual drugs target different aspects of metabolic dysfunction, combining agents with distinct mechanisms of action can yield synergistic effects by addressing adipose tissue pathology from multiple angles. This multipronged approach may more effectively improve metabolic outcomes in patients with obesity and T2DM[39]. Recent studies and clinical practice support the use of combination pharmacotherapy to promote fat redistribution and enhance insulin sensitivity[40,41]. Below we focused on several representative combination regimens and their mechanisms and effects on adipose tissue remodeling.

Pioglitazone combined with metformin: Combining a PPARγ agonist (pioglitazone) with an AMPK activator (metformin) aims to simultaneously address two major pathological processes: Insulin resistance and excessive hepatic glucose production. Clinical studies have shown that the combination of pioglitazone and metformin achieves better glycemic control and lipid metabolism improvements compared with monotherapy. In a randomized controlled trial, patients with T2DM who received 30 mg of pioglitazone combined with metformin (and sulfonylureas) for 24 weeks showed a significant increase in subcutaneous fat mass without a corresponding increase in intra-abdominal fat[26]. This suggests that metformin may offset to some extent the risk of central fat accumulation associated with pioglitazone, leading to overall weight gain being confined mainly to the metabolically safer subcutaneous fat depot[42,43].

This combination improves insulin sensitivity through complementary mechanisms: Pioglitazone expands lipid storage capacity and reduces ectopic fat burden; and metformin decreases hepatic glucose output and promotes visceral fat oxidation. Together they contribute to improved glycemic and lipid profiles[27]. In clinical applications the weight-lowering tendency of metformin may partially counteract the weight gain associated with pioglitazone, enhancing treatment tolerability and yielding a more favorable metabolic benefit profile[44].

Glucagon-like peptide-1 receptor agonists combined with metformin: Glucagon-like peptide-1 (GLP-1) receptor agonists (e.g., liraglutide, semaglutide) improve glycemic control by stimulating insulin secretion, delaying gastric emptying, and suppressing appetite while also producing significant weight loss. Notably, GLP-1 receptor agonists (GLP-1RAs) have a preferential effect on reducing visceral fat. In patients with obesity and T2DM, GLP-1RAs reduce both subcutaneous and visceral fat, but the reduction in visceral fat is more pronounced[29].

Combining GLP-1RAs with metformin is a widely adopted strategy in managing patients with obesity and T2DM, offering synergistic benefits for glycemic control, weight loss, and adipose tissue optimization. For example, in patients on background metformin therapy, the addition of liraglutide at 1.8 mg/day significantly reduced visceral fat volume[45]. A placebo-controlled trial in South Asian patients with T2DM found that liraglutide treatment led to a marked reduction in intra-abdominal fat, and the extent of visceral fat loss positively correlated with glycated hemoglobin (HbA1c) reduction, suggesting a direct link between fat loss and metabolic improvement[45].

Other studies have reported that adding semaglutide (a GLP-1RA) to oral antidiabetic therapy for 52 weeks led to significant reductions in both visceral and hepatic fat content. Similarly, oral semaglutide also demonstrated significant visceral fat reduction after 26 weeks of treatment[46]. Thus, GLP-1RA-based combination therapy not only enhances glycemic control but also contributes to cardiometabolic health by promoting weight loss and reducing visceral fat accumulation. Importantly, GLP-1RAs may also improve adipose tissue endocrine function. For instance, liraglutide added to metformin therapy has been shown to elevate levels of the anti-inflammatory adipokine omentin-1, which may help modulate the inflammatory microenvironment in adipose tissue. Given their superior glycemic and weight-lowering effects, clinical guidelines recommend GLP-1RAs in combination with metformin and/or other agents for patients with T2DM requiring weight management to achieve maximum therapeutic efficacy[47,48].

Combination of sodium-glucose co-transporter 2 inhibitors with metformin: Sodium-glucose co-transporter 2 (SGLT-2) inhibitors (e.g., dapagliflozin, empagliflozin) promote urinary glucose excretion by inhibiting renal glucose reabsorption, resulting in an average daily caloric loss of approximately 300 kcal and thereby contributing to sustained weight reduction[49]. SGLT-2 inhibitors alone can reduce body fat with studies showing that most of the weight loss is due to fat mass reduction, particularly targeting metabolically harmful VAT. For example, compared with metformin 100 mg canagliflozin daily for 12 weeks significantly reduced VAT volume in patients with T2DM[50].

Combining SGLT-2 inhibitors with other antidiabetic agents often enhances the beneficial effects on fat distribution. In an open-label controlled trial comparing the addition of dapagliflozin (10 mg/day) to metformin vs metformin monotherapy, the combination group experienced greater weight loss (8.6% vs 5.7%) and a more significant reduction in waist circumference after 6 months[41]. Notably, both groups showed an approximately 21% reduction in omental fat thickness (an indicator of intra-abdominal fat), but only the combination group exhibited a further significant reduction in perirenal fat thickness, suggesting that SGLT-2 inhibitors may have a more pronounced effect on deep visceral fat depots, such as retroperitoneal fat[41].

In addition, dapagliflozin combined with other oral agents has demonstrated advantages in fat redistribution. For instance, dapagliflozin plus saxagliptin combined with metformin significantly reduced hepatic steatosis and total body fat in patients with nonalcoholic fatty liver disease. In patients with T2DM with suboptimal glycemic control on metformin, adding dapagliflozin for 12 weeks not only reduced total, visceral, and subcutaneous fat volumes but also increased plasma adiponectin levels[39,51].

SGLT-2 inhibitors also improve adipose tissue inflammation. In T2DM rat models empagliflozin significantly downregulated proinflammatory cytokines such as MCP-1, TNF-α, IL-1β, and IL-6 in VAT and reduced macrophage infiltration into adipose tissue. This anti-inflammatory remodeling further enhanced insulin sensitivity in adipose tissue. Thus, combining SGLT-2 inhibitors with other drugs (especially metformin or GLP-1RAs) can act via a dual mechanism (increasing energy expenditure and improving the inflammatory microenvironment) to more effectively reduce harmful fat accumulation and improve metabolism.

Multiagent combinations and translational prospects: With growing insight into the multifactorial pathogenesis of T2DM, early implementation of combination therapies targeting multiple pathways is increasingly advocated to correct metabolic derangements comprehensively[52]. The combination of GLP-1RAs and SGLT-2 inhibitors has shown additive and synergistic effects on lowering glucose and body weight and improving fat distribution, including greater reductions in visceral and hepatic fat as well as cardiovascular and renal benefits[53]. Based on these findings, some experts have proposed adding a PPARγ agonist (e.g., pioglitazone) to the GLP-1RA + SGLT-2 inhibitor regimen to create a triple therapy aimed at further enhancing insulin sensitivity and β cell function, reversing residual lipotoxicity, and inhibiting inflammation, ultimately reducing cardiovascular risk to the greatest extent possible[54,55].

A recent systematic review and meta-analysis (involving 19 studies and over 25000 patients in real-world settings) provided strong empirical support: Compared with using newer antidiabetic drugs alone, adding pioglitazone to either GLP-1RA or SGLT-2 inhibitor significantly further reduced HbA1c (by approximately 1.0% and 0.5%, respectively) and body weight without increasing hypoglycemia risk[56]. Notably, the pioglitazone + SGLT-2 inhibitor combination resulted in an additional mean weight loss of 2.3 kg compared with SGLT-2 inhibitor monotherapy[56]. More importantly, combination therapy has been associated with a reduced risk of adverse cardiovascular outcomes during follow-up. For example, real-world cohort data showed significantly reduced risks of cardiovascular events (adjusted hazard ratio = 0.76, 95% confidence interval: 0.66-0.88) and heart failure (HF) (adjusted hazard ratio = 0.67, 95% confidence interval: 0.55-0.82) with pioglitazone + SGLT-2 inhibitor therapy[55].

These findings highlight the significant clinical potential of combination therapies targeting fat distribution and metabolic pathways. In summary, the rational combination of agents acting on different targets, PPARγ, AMPK, GLP-1 receptors, and renal glucose reabsorption, can comprehensively remodel adipose tissue in patients with obesity-associated T2DM. This includes reducing harmful visceral and ectopic fat accumulation, expanding metabolically safe subcutaneous fat depots, and improving adipokine profiles and inflammation, thereby significantly enhancing insulin sensitivity and metabolic health[29]. This multitarget strategy offers promising translational potential in the treatment of obesity-associated T2DM. Table 2 summarizes the main targets, signaling pathways, adipose regions affected, weight change trends, and metabolic benefits of currently used monotherapies and combination regimens in clinical practice, showcasing their mechanistic differences, shared benefits, and synergistic potential in adipose remodeling and insulin sensitivity enhancement.

Table 2 Summary of major drugs targeting fat distribution and their combined therapeutic strategies and mechanisms of action.

Drug/regimen	Primary target	Key signaling pathway	Main affected fat depot	Body weight impact	Key metabolic benefits
Pioglitazone	PPARγ	PPARγ: Promotes adipocyte differentiation, increases adiponectin, anti-inflammatory	Subcutaneous fat (↑), visceral fat (↓)	Increased (mainly subcutaneous)	Improves insulin sensitivity, redistributes fat, alleviates lipotoxicity and inflammation
Metformin	AMPK	AMPK: Inhibits mTORC1, promotes FA oxidation, anti-inflammatory	Visceral and hepatic fat (↓)	Decreased or weight-neutral	Lowers glucose, anti-lipogenesis, promotes lipid oxidation, reduces inflammation, and improves adipokine profile
Pioglitazone + metformin	PPARγ + AMPK	Fat redistribution (PPARγ) + improved glucose metabolism (AMPK)	Subcutaneous fat (↑), visceral fat (↓)	Balanced	Complementary effects improve fat deposition and insulin resistance, broad metabolic improvement
GLP-1RA + metformin	GLP-1R + AMPK	Appetite suppression, insulin promotion (GLP-1R) + lipid oxidation (AMPK)	Visceral fat (↓) > subcutaneous fat (↓)	Decreased significantly	Antihyperglycemic + weight loss + visceral fat optimization, improved adipokines, anti-inflammatory
SGLT-2i + metformin	SGLT-2 + AMPK	Calorie loss via glycosuria (SGLT-2) + anti-lipogenesis, lipid oxidation (AMPK)	Abdominal, hepatic, and perirenal fat (↓)	Decreased (mainly fat reduction)	Glycemic control + fat loss + anti-inflammatory, reduces lipotoxicity, enhances insulin sensitivity
GLP-1RA + SGLT-2i	GLP-1R + SGLT-2	Dual mechanism: Appetite suppression + urinary glucose excretion	Visceral and hepatic fat (marked) (↓)	Decreased (more pronounced)	Multitarget fat reduction, weight loss, metabolic enhancement, cardiovascular and renal benefits
Pioglitazone + GLP-1RA or SGLT-2i	PPARγ + GLP-1R/SGLT-2	Fat redistribution reprogramming (PPARγ) + fat reduction/glycemic control (GLP-1RA/SGLT-2i)	Fat in multiple depots (↓), subcutaneous storage (↑)	Decreased (via GLP-1RA or SGLT-2i)	Multi-action: Glucose lowering, anti-inflammatory, improved adipokines/Lipotoxicity, better cardiovascular outcomes

PPARγ: Peroxisome proliferator-activated receptor gamma; AMPK: AMP-activated protein kinase; mTORC1: Mammalian target of rapamycin complex 1; FA: Fatty acid; GLP-1RA: Glucagon-like peptide-1 receptor agonist; GLP-1R: Glucagon-like peptide-1 receptor; SGLT-2: Sodium-glucose cotransporter 2; SGLT-2i: Sodium-glucose cotransporter 2 inhibitor.

ADSCS AND THEIR METABOLIC APPLICATIONS

Biological characteristics of ADSCs

ADSCs are a subtype of mesenchymal stem cells (MSCs) that characteristically express high levels of MSC surface markers such as CD73, CD90, and CD105 while lacking or exhibiting low expression of hematopoietic markers such as CD34 and CD45[57]. Originating from the mesoderm, ADSCs possess multipotent differentiation potential, enabling them to differentiate into various cell lineages, including adipocytes, osteoblasts, chondrocytes, cardiomyocytes, myocytes, and even neurons and hepatocytes[58,59]. ADSCs derived from different adipose depots exhibit variations in phenotype and differentiation potential. For example, ADSCs derived from brown adipose tissue display a stronger proliferative capacity and higher differentiation potential into osteogenic, adipogenic, and myogenic lineages compared with those derived from WAT[60]. Additionally, ADSCs can give rise to white, beige, or brown adipocyte lineages, providing new therapeutic avenues for obesity-related disorders. The multipotent differentiation capacity and expression of MSC markers make ADSCs an ideal cell source for regenerative medicine[57,58]. Table 3 summarizes the phenotypic differences and major differentiation potentials of ADSCs from different origins.

Table 3 Phenotypic differences and differentiation potential of adipose-derived stem cells from various sources.

Source	Typical surface markers	Differentiation potential/characteristics
Subcutaneous fat	CD73+, CD90+, CD105+; CD34-/Low, CD45-	Strong adipogenic differentiation; moderate osteogenic/chondrogenic potential; suitable for tissue engineering
Visceral (intra-abdominal) fat	Same as subcutaneous; some studies report SC-ASCs (high CD10) vs VS-ASCs (high CD200)	Slight variation in adipogenic potential; possibly different impacts on metabolic regulation
Brown adipose tissue	Same as MSC markers; commonly express Sca-1, CD29, CD49 in experimental settings	Stronger differentiation capacity; high proliferation; excellent adipogenic, osteogenic, and myogenic potential
Others (e.g., retroperitoneal, periovarian fat)	Similar MSC markers; may include more vascular-related markers (e.g., CD146)	Broad differentiation spectrum; capable of differentiating into mitochondria-rich cell lineages; less studied

CD73: Cluster of differentiation 73; CD90: Cluster of differentiation 90; CD105: Cluster of differentiation 105; CD34: Cluster of differentiation 34; CD45: Cluster of differentiation 45; SC-ASCs: Subcutaneous adipose-derived stem cells; CD10: Cluster of differentiation 10; VS-ASCs: Visceral adipose-derived stem cells; MSC: Mesenchymal stem cell; Sca-1: Stem cell antigen-1; CD29: Cluster of differentiation 29; CD49: Cluster of differentiation 49; CD146: Cluster of differentiation 146.

Anti-inflammatory and paracrine effects in insulin-sensitive tissues

ADSCs exert paracrine regulatory functions by secreting a variety of factors and extracellular vesicles (EVs), thereby improving inflammation and insulin signaling in adipose tissue, liver, muscle, and other insulin-sensitive tissues. It has been reported that ADSCs can secrete anti-inflammatory cytokines such as IL-10 as well as proangiogenic and tissue repair factors including transforming growth factor-β (TGF-β), vascular endothelial growth factor (VEGF), placental growth factor, and angiopoietin-1[58]. IL-10 suppresses inflammation via the Janus kinase 1/tyrosine kinase 2-signal transducer and activator of transcription 3 (STAT3) pathway[61,62]. TGF-β binds its receptor to activate mothers against decapentaplegic homolog (SMAD) signaling and regulate gene expression[63]. VEGF acts on endothelial cells to promote their proliferation and survival[64].

Moreover, ADSC-derived EVs are rich in microRNAs (miRNAs) and growth factors (such as VEGFA, fibroblast growth factor, insulin-like growth factor, and platelet-derived growth factor), which can be taken up by target cells (e.g., adipocytes, hepatocytes, and myocytes) to modulate gene expression and improve the metabolic microenvironment[58,65]. For instance, ADSC-derived EVs have been shown to suppress the expression of proinflammatory cytokines such as IL-6, TNF-α, and MCP-1 in target tissues, thereby alleviating chronic inflammation[65]. Collectively, the anti-inflammatory cytokines (e.g., IL-10), tissue-repair factors [e.g., VEGF, hepatocyte growth factor (HGF)], and miRNA-containing EVs secreted by ADSCs act synergistically to inhibit macrophage-mediated inflammation, activate insulin signaling, and enhance insulin sensitivity[58,66].

Through the secretion of various paracrine factors and miRNA-enriched EVs, ADSCs exert broad metabolic regulatory effects across multiple tissues. Their anti-inflammatory cytokines (e.g., IL-10) inhibit inflammatory responses via STAT3 activation while growth and immunomodulatory factors such as TGF-β, VEGF, HGF, and MCP-1 participate in tissue repair and immune regulation[67-69]. Among the bioactive miRNAs in ADSC-derived EVs, miR-223-3p has been shown to downregulate the profibrotic gene E2F transcription factor 1 in hepatocytes, thereby attenuating hepatic lipid accumulation and fibrosis. Other miRNAs promote adipose tissue browning and adiponectin secretion by activating the AMPK/PPARγ coactivator 1-alpha (PGC-1α) pathway, thereby enhancing systemic metabolic function[66,70].

At the target tissue level, ADSC-derived immunomodulators can suppress macrophage infiltration in adipose tissue and enhance insulin signaling[71,72]. In the liver VEGF and angiopoietin-1 jointly promote angiogenesis and tissue regeneration while EV-derived anti-inflammatory components reduce inflammatory and oxidative stress[73]. In skeletal muscle IL-10 and related factors alleviate inflammation and improve glucose uptake[74]. Overall, ADSCs remodel the microenvironment of insulin-sensitive tissues through the secretion of a spectrum of bioactive molecules, thereby cooperatively enhancing metabolic function at multiple levels (Table 4).

Table 4 Secretory factors of adipose-derived stem cells and their metabolic and immunomodulatory effects in target tissues.

Category of secretion	Representative molecules/miRNAs	Main target tissues or cells	Typical receptors/signaling pathways	Expected metabolic and inflammatory effects
Cytokines	IL-10, TGF-β, VEGF, HGF	VAT, SAT, liver, skeletal muscle, macrophages	JAK1/Tyk2-STAT3; SMAD2/3; VEGFR2-PI3K/Akt	Inhibit NF-κB and JNK-mediated inflammation; promote angiogenesis and tissue repair; enhance insulin signaling throughput
Exosomal miRNAs	miR-223-3p, miR-29a, miR-155, miR-210	Hepatocytes, adipocytes, macrophages	Target E2F1, SOCS1/STAT1, SHIP1, IRAK1/NF-κB	Alleviate hepatic steatosis and fibrosis; modulate macrophage polarization; promote adipose browning and adiponectin secretion
Growth/chemotactic factors	Ang-1, FGF, IGF-1, MCP-1	Vascular endothelium, fibroblasts, immune cells	Tie2-Akt; FGFR-MAPK; CCR2	Promote vascular remodeling; chemotactic regulation of immune cells; improve tissue microcirculation
Antioxidant regulatory axis	PGC-1α-induced upregulation of SOD2, UCP1	White/brown adipocytes	PGC-1α-TFAM; UCP1-uncoupling	Enhance mitochondrial biogenesis and oxidative phosphorylation; reduce ROS load; increase energy expenditure

miRNA: MicroRNA; IL-10: Interleukin 10; TGF-β: Transforming growth factor beta; VEGF: Vascular endothelial growth factor; HGF: Hepatocyte growth factor; VAT: Visceral adipose tissue; SAT: Subcutaneous adipose tissue; JAK1: Janus kinase 1; Tyk2: Tyrosine kinase 2; STAT3: Signal transducer and activator of transcription 3; SMAD2/3: Mothers against decapentaplegic homolog 2/3; VEGFR2: Vascular endothelial growth factor receptor 2; PI3K: Phosphoinositide 3-kinase; Akt: Protein kinase B; NF-κB: Nuclear factor kappa B; JNK: C-Jun N-terminal kinase; E2F1: E2F transcription factor 1; SOCS1: Suppressor of cytokine signaling 1; SHIP1: SH2-containing inositol phosphatase 1; IRAK1: Interleukin-1 receptor-associated kinase 1; Ang-1: Angiopoietin 1; FGF: Fibroblast growth factor; IGF-1: Insulin-like growth factor 1; MCP-1: Monocyte chemoattractant protein 1; Tie2: Tyrosine kinase with immunoglobulin-like and endothelial growth factor-like domains 2; FGFR: Fibroblast growth factor receptor; MAPK: Mitogen-activated protein kinase; CCR2: C-C chemokine receptor type 2; PGC-1α: Peroxisome proliferator-activated receptor gamma coactivator 1-alpha; SOD2: Superoxide dismutase 2; UCP1: Uncoupling protein 1; TFAM: Transcription factor A, mitochondrial; ROS: Reactive oxygen species; miR: MicroRNA.

Evidence from animal and clinical studies in metabolic syndrome

Numerous animal studies have demonstrated that ADSC therapy can significantly improve obesity and metabolic disorder parameters. In high-fat diet-fed mice ADSC infusion led to a marked reduction in body fat percentage (from approximately 46.7% to 32.3%)[75] alongside improvements in blood glucose, glucose tolerance, and HbA1c levels, suggesting a decrease in Homeostasis Model Assessment of Insulin Resistance and enhanced insulin sensitivity[75]. In another study intravenous administration of ADSCs into type 2 diabetic rats significantly reduced fasting glucose and insulin levels, improved insulin resistance index, decreased liver enzymes (alanine aminotransferase, aspartate aminotransferase), and attenuated hepatic fibrosis[76]. In a nonalcoholic steatohepatitis model, ADSC-derived exosomes carrying miR-223-3p inhibited hepatic lipid accumulation and the expression of fibrotic markers, significantly lowering hepatic triglyceride/cholesterol content and alanine aminotransferase/aspartate aminotransferase activities[70]. In a polycystic ovary syndrome rat model, intraperitoneal injection of ADSC-conditioned medium led to reductions in body weight and blood glucose, decreased ovarian volume, and partial restoration of estrogen levels and normal estrous cycles, indicating improvements in both metabolic and reproductive functions[77].

Although limited in number, some clinical studies have also shown the potential benefits of ADSC therapy. Animal models and early-phase clinical trials suggest that ADSC transplantation may promote endogenous β cell regeneration, protect residual β cells, modulate immune responses, and enhance insulin sensitivity in peripheral tissues[70]. While phase II human studies remain scarce, reports have indicated that ADSC therapy can improve insulin resistance and glycemic control in patients with type 2 diabetes (e.g., reductions in Homeostasis Model Assessment of Insulin Resistance and increases in adiponectin levels). Overall, the current literature provides high-quality evidence supporting the therapeutic role of ADSCs in obesity, T2DM, nonalcoholic fatty liver disease, and polycystic ovary syndrome (diseases related to metabolic syndrome). The underlying mechanisms involve immunomodulation, inflammation suppression, and restoration of endocrine functions[70,75,76]. Future efforts should focus on large-scale clinical trials to validate the safety and long-term efficacy of ADSC therapy while exploring optimized delivery routes and preconditioning strategies to enhance its translational potential.

THE ROLE OF ADSCS IN ADIPOSE TISSUE REMODELING IN OBESITY AND DIABETES

Beiging and thermogenic regulation of adipose tissue

Adipose tissue beiging refers to the process whereby white adipocytes acquire characteristics of brown adipocytes, including the expression of mitochondrial UCPs such as UCP1, thereby enhancing energy expenditure. This process plays a critical role in combating obesity and improving glucose metabolism. ADSCs have a dual role in this transformation process[78]. On one hand, ADSCs inherently possess the potential to differentiate into beige or brown adipocytes. For instance, Silva et al[79] found that ADSCs isolated from mediastinal fat expressed brown adipocyte markers such as PR domain zinc finger protein 16 (PRDM16) and UCP1 during differentiation, ultimately becoming thermogenically active brown-like adipocytes. On the other hand, exogenous stimulation of ADSCs can induce their conversion into a brown-like phenotype[80]. For example, treatment of human ADSCs with a combination of metformin and vitamin D significantly upregulated the expression of the thermogenic protein UCP1 at both the mRNA and protein levels while concurrently inhibiting adipogenesis, resulting in a brown-like phenotype[81].

These findings suggest that pharmacological or signaling pathway modulation of ADSCs can activate thermogenic transcriptional networks, including PGC-1α, PRDM16, and UCP1, thereby promoting differentiation or transdifferentiation of ADSCs into brown/beige adipocytes. Additionally, ADSCs exert regulatory effects on surrounding cells via the secretion of various cytokines or exosomes. Recent studies have shown that exosomes derived from healthy mouse ADSCs when injected into obese mice markedly promote the polarization of adipose tissue macrophages toward the anti-inflammatory M2 phenotype [characterized by high IL-10 and arginase-1 (ARG1) expression] and induce beiging of WAT, thereby contributing to metabolic improvement[82]. Mechanistically, ADSC-derived exosomes facilitate M2 macrophage polarization (IL-10/TNF-stimulated gene-6 ↑) and in vivo white-fat beiging, leading to reduced adipose inflammation and improved systemic insulin sensitivity. These effects have been demonstrated to attenuate obesity and drive UCP1+ remodeling in WAT[68,83].

Pharmacologic priming of human ADSCs with metformin plus vitamin D further enhances PGC-1α/PRDM16/UCP1 signaling programs and induces a brown-like phenotype[81]. Notably, in T2DM the beiging capacity of ADSC-derived beige adipocytes is impaired with reduced UCP1 expression and increased mitochondrial ROS, reinforcing the causal link between thermogenic failure and insulin resistance[84]. Collectively, these findings connect ADSC-mediated M2 polarization and beige induction with UCP1-mitochondrial activation, culminating in the alleviation of insulin resistance at both the tissue and whole-body levels. In summary, ADSCs and their secreted products play a key role in the browning of WAT by upregulating thermogenic regulators including PRDM16, PGC-1α, and UCP1 and activating mitochondrial biogenesis and thermogenic programs[81,85].

Interactions between ADSCs, adipokines, and the immune microenvironment

Obesity and diabetes are commonly accompanied by chronic inflammation of adipose tissue, which is closely associated with dysregulated adipokine secretion. In the obese state anti-inflammatory factors such as adiponectin are downregulated while proinflammatory cytokines including leptin, IL-6, and TNF-α are markedly elevated, resulting in a persistent low-grade inflammatory milieu within adipose tissue[86]. This proinflammatory environment alters ADSC function. ADSCs derived from individuals with obesity are often reprogrammed toward a proinflammatory phenotype characterized by a reduced capacity to suppress T cell proliferation and an increase in the secretion of inflammatory cytokines such as IL-6 and TNF-α[87]. These alterations not only impair the regenerative potential of ADSCs but also exacerbate local inflammation.

In contrast, ADSCs derived from healthy donors exhibit pronounced anti-inflammatory properties. Studies have shown that ADSCs from normal mice can modulate the immune microenvironment of adipose tissue by secreting anti-inflammatory factors such as IL-10 and releasing exosomes that promote macrophage polarization toward the M2 phenotype, marked by high expression of IL-10 and ARG1[88,89]. For example, injecting exosomes from healthy donor-derived ADSCs into obese mouse models significantly increased the proportion of M2 macrophages in adipose tissue and markedly improved insulin sensitivity. M2 macrophages possess anti-inflammatory functions, secreting cytokines such as IL-10 to suppress the expression of proinflammatory signals like IL-6 and TNF-α. This leads to elevated adiponectin levels, attenuation of chronic inflammation in adipose tissue, and normalization of lipid metabolism[68,71].

It is important to note that a bidirectional interaction between ADSCs and macrophages can trigger and sustain an inflammatory positive feedback loop during obesity progression. Specifically, hypertrophic adipocytes release exosomes enriched with proinflammatory mediators (e.g., MIF, RBP4) and several miRNAs (e.g., miR-29a, miR-155, miR-34a), which promote macrophage polarization toward the classically activated M1 phenotype. These M1 macrophages are characterized by elevated expression of inducible nitric oxide synthase and TNF-α, thereby intensifying tissue inflammation[90,91]. M1-derived cytokines such as IL-6 and TNF-α along with M1-secreted exosomes containing miR-155, miR-29a, and miR-210, inhibit the differentiation and regenerative capacity of ADSCs and impair their immunomodulatory function. This reciprocal amplification ultimately establishes a persistent inflammatory loop, aggravating the immunometabolic dysfunction of adipose tissue[82,92]. At the molecular level proinflammatory cues (e.g., TLR4 → IKKβ/NF-κB, JNK) drive IRS-1(Ser-P), weakening PI3K-Akt-glucose transporter type 4 (GLUT4) signaling and propagating adipose and systemic insulin receptor. NLRP3 inflammasome activation further amplifies IL-1β/IL-18 signaling in adipose immunometabolism[93-95]. Obesity-derived exosomal miR-29a/miR-155 from adipose macrophages directly induces the insulin receptor across adipocytes, myocytes, and hepatocytes, establishing an EV-mediated endocrine axis[83,96]. In contrast, ADSC-EVs dampen TLR-NF-κB/NLRP3 activity and restore adipokine balance (adiponectin ↑, IL-6/TNF-α ↓), thereby reducing IRS-1(Ser-P) and normalizing insulin signaling[83,97]. These mechanistic links connect ADSC immunomodulation with quantifiable insulin-signaling rescue.

In summary, ADSCs, adipokines, and immune cells form a regulatory network. Healthy ADSCs secrete anti-inflammatory cytokines and ADSC-derived EVs that promote M2 macrophage polarization and suppress IL-6/TNF-α signaling while ADSCs in an obese environment tend to adopt a proinflammatory phenotype, favoring M1 polarization and enhanced cytokine production[82,87]. This regulatory mechanism highlights the central role of ADSCs in obesity-associated adipose tissue inflammation and provides insights into their therapeutic potential in metabolic disease management (Tables 5 and 6).

Table 5 Key adipokines and inflammatory mediators associated with obesity and diabetes.

Molecule	Category	Major source	Action on insulin signaling	Regulation trend (in obesity)	Regulation evidence by drugs/ADSCs
Leptin	Proinflammatory adipokine	VAT > SAT	Promotes IRS-1 Ser P, inhibits PI3K/Akt	↑	Downregulated by GLP-1RA; indirectly inhibited by ADSC EVs
Adiponectin	Anti-inflammatory/sensitizing	SAT	Activates AMPK/PPARα	↓	Upregulated by pioglitazone, SGLT-2 inhibitors
TNF-α	Classical proinflammatory cytokine	M1 macrophages, hypertrophic adipocytes	IKKβ/NF-κB AND JNK → promotes IRS-1 Ser P	↑	Downregulated by pioglitazone, metformin, ADSC EVs
IL-6	Proinflammatory cytokine	Same as above	SOCS3 inhibits IRS-1	↑	Suppressed via ADSC-derived IL-10 axis
IL-10	Anti-inflammatory cytokine	M2 macrophages, ADSCs	Inhibits NF-κB/JNK, enhances insulin signaling	↓	Healthy ADSC EVs → promote M2 macrophage polarization↑

ADSCs: Adipose-derived stem cells; VAT: Visceral adipose tissue; SAT: Subcutaneous adipose tissue; IRS-1: Insulin receptor substrate 1; Ser P: Serine phosphorylation; PI3K/Akt: Phosphoinositide 3-kinase/protein kinase B; GLP-1RA: Glucagon-like peptide-1 receptor agonist; EVs: Extracellular vesicles; AMPK: Adenosine monophosphate-activated protein kinase; PPARα: Peroxisome proliferator-activated receptor alpha; SGLT-2: Sodium-glucose cotransporter 2; TNF-α: Tumor necrosis factor alpha; M1: Classically activated macrophage (proinflammatory phenotype); IKKβ: IκB kinase beta; JNK: C-Jun N-terminal kinase; IL-6: Interleukin 6; IL-10: Interleukin 10; SOCS3: Suppressor of cytokine signaling 3; M2: Alternatively activated macrophage (anti-inflammatory phenotype); NF-κB: Nuclear factor kappa B.

Table 6 Targeted pathways and metabolic effects of adipose-derived stem cell-derived exosomal microRNAs.

miRNA	Major direct target genes/pathways	Target tissues/cells	Metabolic or inflammatory effects	Evidence type1
miR-223-3p	E2F1 → Inhibits adipogenesis/fibrosis	Hepatocytes	Reduces hepatic lipid accumulation and fibrosis	HFD mice (in vivo)/HepG2 (in vitro)
miR-29a	SOCS1/STAT1	Macrophages	Inhibits M1 polarization, decreases TNF-α/IL-6	LPS-stimulated RAW264.7/mice
miR-155	SHIP1/SOCS1	Macrophages	Promotes M1 polarization, amplifies inflammation	Functional antagonism in vitro
miR-34a	SIRT1/FGF21	Liver and adipose tissue	Inhibits browning, promotes adipogenesis	DIO mice (in vivo)
miR-210	IRAK1/NF-κB	Liver, skeletal muscle	Reduces inflammation, improves insulin sensitivity	db/db mice (in vivo)
miR-126	IRS-1/PI3K-Akt	Endothelial cells	Promotes angiogenesis, improves insulin signaling	STZ rats (in vivo)/HUVEC

¹Evidence types include in vivo animal models and in vitro cell line experiments. miRNA: MicroRNA; E2F1: E2F transcription factor 1; SOCS1: Suppressor of cytokine signaling 1; STAT1: Signal transducer and activator of transcription 1; SHIP1: SH2-containing inositol phosphatase 1; SIRT1: Sirtuin 1; FGF21: Fibroblast growth factor 21; IRAK1: Interleukin-1 receptor-associated kinase 1; NF-κB: Nuclear factor kappa B; IRS-1: Insulin receptor substrate 1; PI3K: Phosphoinositide 3-kinase; Akt: Protein kinase B; TNF-α: Tumor necrosis factor-alpha; IL-6: Interleukin 6; HFD: High-fat diet; HepG2: Human hepatocellular carcinoma cell line; LPS: Lipopolysaccharide; RAW264.7: Murine macrophage cell line; DIO: Diet-induced obesity; db/db: Diabetic obese (leptin receptor-deficient) mouse model; STZ: Streptozotocin; HUVEC: Human umbilical vein endothelial cells; M1: Classically activated macrophage (proinflammatory phenotype); miR: MicroRNA.

Effects of ADSCs on mitochondrial function in adipose tissue

Mitochondrial dysfunction in adipocytes is characterized by reduced mitochondrial content, impaired oxidative phosphorylation, and excessive ROS accumulation. It is a key pathological basis of obesity and insulin resistance. ADSCs can improve mitochondrial biogenesis and function in adipocytes through multiple mechanisms[98,99].

On one hand, ADSCs secrete various factors, including growth factors and miRNAs carried by microvesicles or exosomes that activate the mitochondrial biogenesis regulator PGC-1α pathway in adipocytes. PGC-1α promotes mitochondrial biogenesis by activating nuclear respiratory factors [nuclear factor erythroid 2-related factor 1 (Nrf1) and Nrf2] and estrogen-related receptor alpha, which in turn induce transcription factor A, mitochondrial (TFAM), and DNA polymerase γ, facilitating mitochondrial DNA replication[100]. Activation of this pathway increases mitochondrial number, enhances expression of respiratory chain complexes, and boosts oxidative phosphorylation capacity. During MSC differentiation into adipocytes, upregulation of PGC-1α is accompanied by increased mitochondrial content and improved oxidative capacity, indicating that PGC-1α overexpression significantly enhances mitochondrial oxidative metabolism[101]. By secreting factors that upregulate PGC-1α, ADSCs can simulate this process, thereby increasing PGC-1α levels in target adipocytes and initiating TFAM-mediated mitochondrial DNA replication.

On the other hand, PGC-1α also induces the expression of mitochondrial antioxidant enzymes, thereby reducing ROS-induced mitochondrial damage[102,103]. For example, activation of PGC-1α upregulates superoxide dismutase 2 (SOD2) in mitochondria, scavenging free radicals and maintaining mitochondrial homeostasis. During MSC differentiation it has been confirmed that PGC-1α regulation of SOD2 expression helps suppress ROS accumulation, suggesting that ADSCs may similarly protect adipocyte mitochondria from oxidative stress through activation of this pathway[101].

Furthermore, during ADSC-induced beige/brown adipocyte differentiation, the upregulation of UCPs such as UCP1 can lower mitochondrial membrane potential, reduce ROS production, and increase energy expenditure. Experiments have shown that treating human ADSCs with a combination of metformin and vitamin D significantly upregulates UCP1 expression in differentiated adipocytes, indicating that ADSC-mediated mitochondrial improvements are associated with enhanced UCP1 expression[81]. Beyond biogenesis ADSC-EVs activate PGC-1α → Nrf1/TFAM and mitochondrial-fusion programs (OPA1/MFN1/MFN2), increasing mitochondrial DNA replication and respiratory capacity in metabolic injury models[100].

By enhancing fatty-acid oxidation and UCP1-linked uncoupling, ADSC interventions lower lipotoxic intermediates (e.g., ceramides), a causal brake on Akt signaling. Conversely, pharmacologic ceramide-synthesis inhibition prevents the lipid-induced insulin receptor in vivo, underscoring this mediator pathway[104,105]. Finally, UCP1 status and redox control are intertwined: Defective UCP1 programs associate with ROS accumulation in ADSC-derived beige adipocytes from T2DM; and thermogenic activation can rebalance mitochondrial ROS and insulin signaling[84,106]. These mechanisms provide testable mediators (PGC-1α/TFAM, UCP1, mt-fusion markers, ceramide panels) linking ADSC therapy to insulin-signaling rescue.

In summary, ADSCs enhance mitochondrial biogenesis and oxidative phosphorylation activity in adipocytes by upregulating the PGC-1α/TFAM pathway and reducing ROS-induced mitochondrial damage by inducing antioxidant enzymes such as SOD2. These effects contribute to improved mitochondrial function in adipocytes affected by obesity, thereby increasing energy metabolism and insulin sensitivity. This provides a solid theoretical basis for the therapeutic application of ADSCs in metabolic diseases.

Integrated secretome model: Linking metabolism-inflammation-regeneration

To clarify the cooperative roles of cytokines, growth factors, and exosomal miRNAs derived from ADSCs, we proposed an integrated secretome model in which metabolism, inflammation, and tissue regeneration are jointly regulated through a limited number of key crosstalk nodes. Anti-inflammatory cytokines such as IL-10 and TGF-β activate STAT3 and SMAD2/3 signaling, which suppresses the TLR-IKKβ/JNK-NF-κB pathway. This inhibition reduces IRS-1(Ser-P) and restores PI3K-Akt-GLUT4-mediated glucose uptake in adipocytes[107]. Simultaneously, proregenerative factors including insulin-like growth factor-1, HGF, and VEGF promote endothelial repair and increase capillary density, thereby reducing hypoxia-induced inflammation and fibrosis[108].

Exosomal miRNAs provide an additional, rapid regulatory layer. For example, miR-223-3p and miR-146a suppress NLRP3/TGF-β-activated kinase 1 (TAK1)-TNF receptor-associated factor 6 (TRAF6) activation, whereas obesogenic miRNAs such as miR-155 and miR-29a enhance M1 macrophage polarization and insulin resistance. ADSC-derived EVs shift this balance toward an M2 phenotype characterized by high ARG1 and IL-10 expression along with increased adiponectin, which collectively favor healthy hyperplastic expansion over maladaptive hypertrophy.

On the metabolic axis ADSCs activate the AMPK/SIRT1-PGC-1α-PRDM16-UCP1 pathway, increasing mitochondrial biogenesis and fatty acid oxidation while reducing ceramide and diacylglycerol accumulation[109]. This alleviates lipotoxic inhibition of Akt signaling. Together these actions of the ADSC secretome coordinate immune homeostasis, extracellular matrix softening and angiogenesis, and thermogenic remodeling. This integrated mechanism provides a coherent explanation for how ADSCs improve insulin sensitivity across adipose tissue, liver, and muscle (Figure 3).

Figure 3 Stem cell secretome. Integrated adipose-derived stem cell secretome suppresses inflammatory signaling and restores insulin action in adipose tissue with interleukin-10, transforming growth factor-β activate signal transducer and activator of transcription 3, and mothers against decapentaplegic homolog 2/3, which repress the toll-like receptor-IκB kinase β/c-Jun N-terminal kinase-nuclear factor kappa B axis, lowering serine-phosphorylated insulin receptor substrate-1 and re-establishing phosphoinositide 3-kinase-protein kinase B-glucose transporter 4-dependent glucose uptake in adipocytes. Adipose-derived stem cell-derived extracellular vesicles deliver miR-223-3p and miR-146a to inhibit NOD-like receptor family pyrin domain containing 3 and transforming growth factor-β-activated kinase 1/tumor necrosis factor receptor-associated factor 6 signaling, further dampening innate immune activation. By contrast, obesogenic miR-155 and miR-29a (dashed) promote M1 polarization and exacerbate insulin resistance. The net effect is a shift of the adipose immune milieu toward resolution with improved insulin signaling and glucose uptake. Solid arrows indicate activation; dashed lines denote context-dependent or deleterious routes. IL-10: Interleukin-10; TGF-β: Transforming growth factor-β; STAT3: Signal transducer and activator of transcription 3; SMAD2/3: Mothers against decapentaplegic homolog 2/3; TLR: Toll-like receptor; IKKβ: IκB kinase β; JNK: C-Jun N-terminal kinase; NF-κB: Nuclear factor kappa B; IRS-1: Insulin receptor substrate-1; PI3K: Phosphoinositide 3-kinase; Akt: Protein kinase B; GLUT4: Glucose transporter 4; ADSCs: Adipose-derived stem cells; EVs: Extracellular vesicles; NLRP3: NOD-like receptor family pyrin domain containing 3; TAK1: Transforming growth factor-β-activated kinase 1; TRAF6: Tumor necrosis factor receptor-associated factor 6; miR: MicroRNA.

FUTURE DIRECTIONS: INTEGRATED APPLICATION OF PHARMACOLOGICAL AND CELL THERAPIES

Synergistic potential of ADSCs and insulin sensitizers

The combined application of ADSCs with commonly used insulin-sensitizing agents, such as metformin, TZDs, GLP-1RAs, and SGLT-2 inhibitors may offer complementary benefits across multiple pathways. For example, an animal study demonstrated that simultaneous administration of ADSCs and the GLP-1 analog exenatide in diabetic rats significantly improved renal function and tissue structure with inflammation, fibrosis, and apoptosis markers further reduced compared to either monotherapy group[110]. Similarly, in a diabetic nephropathy model, repeated infusion of ADSCs combined with the SGLT-2 inhibitor empagliflozin resulted in more pronounced protection of renal function and ultrastructural integrity, whereas monotherapy showed inferior effects[111].

Additionally, coadministration of bone marrow MSCs with pioglitazone or GLP-1RA has been shown to significantly reduce insulin resistance and inflammation through various molecular mechanisms[7]. These findings suggest that insulin-sensitizing drugs may enhance ADSC viability, paracrine secretion, and regenerative capacity by improving the tissue metabolic microenvironment by reducing insulin resistance and modulating adipokine profiles, thereby achieving synergistic therapeutic effects. It is important to note that such synergism may vary depending on intervention timing and disease model. For example, in a murine model of radiation-induced skin fibrosis, preventive combination therapy did not yield additional benefit, suggesting that drug-ADSC interactions must be contextually assessed under different pathological conditions[112].

In summary, current experimental data support the promising potential of combining ADSCs with agents like GLP-1RA and SGLT-2 inhibitors[110,111] while the synergistic effects of metformin and other drugs remain less clear. This highlights the need for future studies to further elucidate the mechanisms and optimal timing of pharmacological-stem cell combination strategies.

Personalized therapy and omics-guided intervention strategies

Within the framework of precision medicine, transcriptomic, proteomic, and single-cell omics analyses of ADSC therapy can reveal individualized response markers and subtype differences[113,114]. For example, RNA sequencing comparing ADSCs from patients with diabetic nephropathy with those from healthy controls revealed significant differences in the expression of angiogenesis-related genes and miRNAs, suggesting that a hyperglycemic and uremic environment may alter ADSC function. However, functional assays indicated that their proangiogenic capacity remained intact, supporting the feasibility of autologous ADSC therapy with individualized transcriptomic-based adjustments[115].

Single-cell RNA sequencing provides even more refined insights into cellular lineage composition. A recent study performing single-cell RNA sequencing on clinical-grade ADRCs (adipose-derived stromal vascular fractions) found high consistency in cell composition across different donors with minimal influence of sex or body mass index on the proportion of MSCs[113]. This low heterogeneity facilitates the reproducibility of batch preparation and simplifies recipient selection.

Genomic studies have also highlighted the role of a patient’s genetic background in shaping treatment response. For instance, a specific single nucleotide polymorphism identified in adipocyte-like cells derived from fat MSCs was found to affect the PPARγ binding site and the induction of ATP binding cassette transporter A1 expression by TZDs. This single nucleotide polymorphism also predicted an increased risk of cholesterol elevation following TZD therapy, implying that genotyping or transcriptomic analysis could be used to anticipate individual responses to drug or cell-based therapies[116].

In summary, integrating multiomics data, including single-cell transcriptomics, genomics, and proteomics, enables the identification of molecular markers and response subtypes associated with ADSC efficacy. These insights can guide optimal therapeutic timing and regimen selection. Although this research area is still in its early stages, existing evidence suggests that omics-based stratification based on metabolic status, genetic background, or ADSC subpopulation characteristics holds promise for the precision optimization of ADSC therapies (e.g., determining ideal administration windows or identifying suitable patient cohorts)[113,115].

Safety, scalability, and ethical considerations in clinical translation

Safety is a core consideration in the clinical application of ADSCs.

Immunogenicity: ADSCs are characterized by low immunogenicity, typically lacking expression of major histocompatibility complex class II costimulatory molecules. As such they elicit minimal immune rejection when used in allogeneic infusions[117]. Recent clinical trials (e.g., in patients with coronavirus disease 2019) have shown that repeated intravenous administration of allogeneic ADSCs did not result in treatment-related serious adverse events, indicating that short-term multiple dosing is safe and well-tolerated[117]. Nevertheless, rare infusion-related reactions or coagulation abnormalities should still be monitored.

Tumorigenic potential: ADSCs expanded in vitro are generally non-tumorigenic in vivo. Previous animal studies involving high-dose ADSC administration (with follow-up periods of up to 6 months) have not observed tumor formation[118]. Similarly, human clinical trials with follow-ups of up to 3 months have not reported tumorigenesis related to ADSC transplantation[118]. Due to their proliferative capacity and potential to promote tumor growth in tumor-prone microenvironments, rigorous oversight remains necessary.

Batch variability and donor heterogeneity: ADSCs derived from different donors or anatomical sites may exhibit interbatch variability. To reduce such inconsistencies researchers recommend the implementation of strict good manufacturing practice (GMP)-grade production and quality control systems. Standardized protocols including uniform liposuction procedures, cell sorting, and expansion conditions can significantly improve the reproducibility of ADSC preparation[119]. Single-cell omics data suggest that under the same manufacturing protocol, the major cell subpopulations of ADSCs remain relatively stable across different patients[113]. In addition, proper donor selection (e.g., infection-free, appropriate age) and repeated safety screenings (e.g., karyotype stability, phenotypic profiling) are essential quality control measures.

GMP-compliant manufacturing and oversight: All therapeutic ADSCs must be produced under GMP guidelines to ensure sterility, consistency in cell quantity and viability, and full traceability throughout the manufacturing process[119]. Research protocols must be reviewed and approved by independent ethics committees and regulatory agencies to ensure compliance with ethical and safety standards.

Long-term follow-up: Although short-term safety appears favorable, patients receiving ADSC therapy require long-term monitoring to detect delayed adverse events (e.g., tumor development, immune reactions, or other complications) and to assess the durability of therapeutic effects. In conclusion, clinical translation of ADSCs must balance innovation with safety. While advancing efficacy validation, equal emphasis must be placed on rigorous quality control, ethical review, and long-term surveillance to ensure both patient safety and treatment reliability[118,119].

POTENTIAL OF COMBINED THERAPY TO MITIGATE MAJOR T2DM COMPLICATIONS

The burden of T2DM is driven primarily by macrovascular and microvascular complications, which frequently outweigh glycemic indices in determining morbidity and mortality[120]. Our unified framework of adipose tissue remodeling-insulin resistance-stem cell regulation complications can be viewed as downstream manifestations of chronic adipose inflammation, endothelial dysfunction, and fibrotic remodeling across organs. Here, we synthesized how combined pharmacologic-ADSC strategies may modify shared pathobiology and improve outcomes in cardiovascular disease, diabetic kidney disease (DKD), and neuropathy (Table 7).

Table 7 Mapping of combined therapy components to targeted complication pathways and endpoints.

Combination	Dominant mechanism (within unified axis)	Target complication	Key endpoints	Notes
GLP-1RA + ADSC/EV	Reducing adipose inflammation → improved endothelial function; proangiogenic repair	ASCVD	MACE, FMD, hs-CRP	Weight loss enables ADSC efficacy
SGLT-2i + ADSC/EV	Hemodynamic unloading; anti-fibrosis + tubular regeneration	HF/CKD	HF hospitalization, eGFR slope, UACR	Monitor volume status
Metformin + ADSC/EV	AMPK activation; mitochondrial support + neurotrophic cues	Neuropathy	NCS, QST, CCM metrics	Consider B12 monitoring
Low-dose PPARγ + ADSC/EV	Adipose remodeling; anti-inflammation + anti-fibrosis	Multiorgan	Composite renal/cardiac endpoints	Consider edema risk

GLP-1RA: Glucagon-like peptide-1 receptor agonist; ADSC: Adipose-derived stem cell; EV: Extracellular vesicle; SGLT-2i: Sodium-glucose cotransporter-2 inhibitor; PPARγ: Peroxisome proliferator-activated receptor gamma; AMPK: Adenosine monophosphate-activated protein kinase; ASCVD: Atherosclerotic cardiovascular disease; HF: Heart failure; CKD: Chronic kidney disease; MACE: Major adverse cardiovascular events; FMD: Flow-mediated dilation; hs-CRP: High-sensitivity C-reactive protein; eGFR: Estimated glomerular filtration rate; UACR: Urinary albumin-to-creatinine ratio; NCS: Nerve conduction study; QST: Quantitative sensory testing; CCM: Corneal confocal microscopy; B12: Vitamin B12.

Cardiovascular complications of T2DM (atherosclerotic cardiovascular disease and HF)

Adipose inflammation and maladaptive adipokine signaling in T2DM drive atherogenesis and myocardial remodeling by impairing endothelial nitric oxide, amplifying oxidative stress, and promoting fibrosis, fitting our unified axis of adipose tissue remodeling-insulin resistance-stem cell regulation. Visceral/perivascular adipose dysfunction links directly to endothelial injury and plaque progression, underscoring therapy beyond glycemic targets[121,122]. GLP-1RAs reduce major adverse cardiovascular events independent of glucose lowering and are recommended by current American Diabetes Association Standards for patients with atherosclerotic cardiovascular disease risk[123-125].

SGLT-2 inhibitors consistently lower HF hospitalization and, in some trials cardiovascular death with benefits observed across the HF spectrum and supported by natriuresis improved energetics and anti-inflammatory/anti-fibrotic actions[126,127]. Complementarily, ADSCs and their EVs attenuate vascular inflammation, promote endothelial repair and angiogenesis, and limit adverse cardiac remodeling via paracrine cargo (including sphingosine-1-phosphate/sphingosine kinase-1/sphingosine-1-phosphate receptor-1 signaling and proreparative macrophage polarization), providing a biologic partner to GLP-1RAs/SGLT-2 inhibitors to stabilize plaque/microvascular dysfunction and restrain NLRP3/TGF-β-driven fibrosis[128-130].

For complication-focused studies endpoints should pair event reduction with remodeling mediation, including major adverse cardiovascular event and HF hospitalization alongside endothelial function measured by flow-mediated dilation, structural/matrix indices, and inflammatory biomarkers, anchored to American Diabetes Association guidance and cardiovascular outcome trial conventions[123].

Beyond simple additive effects ADSCs and conventional pharmacological agents appear to converge on shared molecular hubs, creating the potential for mechanistic synergy. GLP-1RAs via cAMP/PKA-cAMP response element-binding protein-PGC-1α signaling) and metformin (via AMPK activation) stimulate the PGC-1α-PRDM16-UCP1 thermogenic program. When combined with ADSC-mediated or EV-mediated induction of adipocyte beiging and mitochondrial repair, these effects may intensify thermogenic remodeling and attenuate the accumulation of lipotoxic intermediates. SGLT-2 inhibitors improve hemodynamic status and dampen systemic inflammatory tone. In parallel, ADSC-EV-derived miRNAs (e.g., miR-223-3p, miR-146a) inhibit NLRP3/TAK1-TRAF6 signaling, potentially enhancing anti-inflammatory reprogramming and limiting fibrosis through modulation of the TGF-β/SMAD pathway. At the level of insulin signaling, suppression of SOCS3 expression and IRS-1(Ser-P) by the ADSC secretome could augment GLP-1RA-induced improvements in Akt activation and GLUT4 translocation, an interaction amenable to experimental verification through paired adipose tissue biopsies and EV cargo profiling.

Renal complications of T2DM: DKD

In T2DM ectopic lipid deposition and adipose-derived inflammation drive glomerular hyperfiltration, tubulointerstitial injury, and fibrosis, which are mechanisms aligned with our unified axis and with guideline emphasis on complication-focused care beyond glycemia[131]. Robust renal outcome trials show that SGLT-2 inhibitors slow estimated glomerular filtration rate decline and reduce kidney failure and cardiorenal events across broad chronic kidney disease (CKD) populations. The American Diabetes Association and Kidney Disease: Improving Global Outcomes recommend SGLT-2 inhibitors on top of renin-angiotensin-aldosterone system blockade with GLP-1RAs added for residual albuminuria and cardiometabolic risk, and the FLOW trial demonstrated renoprotection with semaglutide in T2DM with CKD[131-134].

Preclinical and early clinical data indicate ADSCs and their EVs mitigate renal inflammation/fibrosis, promote tubular repair and microvascular integrity (e.g., via VEGF/HGF, TGF-β modulation, Nrf2/Kelch-like ECH-associated protein 1), and are feasible in progressive DKD. Combining SGLT-2 inhibitors (hemodynamic/anti-fibrotic) with ADSC/EV therapy (pro-regenerative/anti-inflammatory) targets complementary nodes of remodeling while underscoring GMP standardization and potency assays for EV products[135-137]. Complication-oriented trials should pair event reduction with mechanistic mediation: EGFR slope; ≥ 40%-50% decline, kidney failure, or cardiorenal composites; urinary albumin-to-creatinine ratio; and imaging/biomarkers of perfusion/fibrosis (magnetic resonance imaging indices, neutrophil gelatinase-associated lipocalin, kidney injury molecule-1), consistent with the American Diabetes Association/Kidney Disease: Improving Global Outcomes conventions[131-133].

Neuropathic complications of T2DM: Peripheral and autonomic neuropathy

In T2DM lipotoxicity, mitochondrial dysfunction, and microvascular ischemia converge to cause axonal loss and demyelination. Current guidance emphasizes early, complication-focused screening and management of peripheral and autonomic neuropathies[138]. Emerging clinical evidence suggests GLP-1RA can improve structural and electrophysiological markers of diabetic peripheral neuropathy (e.g., reversal of nerve morphological abnormalities with semaglutide/dulaglutide in a prospective cohort) while broader reviews outline anti-inflammatory and mitochondrial mechanisms[139,140]. Metformin may confer ancillary neuroprotective effects but warrants vitamin B12 monitoring during long-term use per updated recommendations[141,142]. For SGLT-2 inhibitors small studies and reviews indicate potential benefit on autonomic indices (e.g., heart rate variability, skin sympathetic nerve activity), aligning with their favorable cardiorenal physiology although confirmatory trials are still needed[143].

Regenerative approaches add complementary biology. ADSCs/MSC-derived EVs dampen neuroinflammation, deliver neurotrophic cargo (e.g., nerve growth factor/brain-derived neurotrophic factor), and enhance endoneurial perfusion in preclinical models with early human evidence supporting feasibility and a shift toward standardized, cell-free EV products[144,145]. Complication-oriented studies should pair symptoms/events with mechanistic mediation using nerve conduction studies, quantitative sensory testing, corneal confocal microscopy for small-fiber integrity, and autonomic testing (e.g., sudomotor measures, heart rate variability/Ewing battery), aligning endpoints with guideline conventions[138,146].

Patient selection, safety, and implementation

A clinically pragmatic strategy would be to initiate treatment with pharmacological agents, such as GLP-1RAs, SGLT-2 inhibitors, and where appropriate metformin for an initial period of approximately 8-12 weeks[147,148]. This phase aims to reduce edema and systemic inflammation while improving endothelial function[149]. Once tissue hypoxia and fibrosis have sufficiently resolved, administration of ADSCs or their EVs can be introduced to promote tissue repair. Long-term maintenance would then be achieved through continued use of evidence-based, guideline-recommended therapies[65,150,151]. This sequential approach is designed to optimize both cell engraftment and paracrine activity while aligning with our proposed triad framework in which immune quiescence enables metabolic remodeling that supports sustained tissue regeneration.

Risk-stratified deployment should start from the complication profile. People with established or high-risk atherosclerotic cardiovascular disease favor GLP-1RAs with proven cardiovascular benefit. Those with HF or CKD should prioritize SGLT-2 inhibitors (on top of renin-angiotensin-aldosterone system blockade) while individuals with obesity-predominant insulin resistance benefit from weight-centric incretin therapy ideally integrated with ADSC/EV-based regeneration within our unified axis. These priorities align with the American Diabetes Association Standards of Care 2025 for cardiovascular risk and pharmacotherapy[123,152] and with the Kidney Disease: Improving Global Outcomes 2024 recommendations for CKD[131]. For HF the American Heart Association/American College of Cardiology/Heart Failure Society of America 2022 guidelines embed SGLT-2 inhibitors in guideline-directed medical therapy[153].

Safety monitoring should include volume status and periprocedural holding of SGLT-2 inhibitors (generally 3 days or 4 days for ertugliflozin) to mitigate euglycemic diabetic ketoacidosis risk[154]. For GLP-1RAs patients should be counseled on gastrointestinal intolerance and rare pancreatitis signals while watching for gallbladder disease[155,156]. If pioglitazone is used as an adipose-remodeling adjunct, weight edema/HF risk and fracture propensity should be considered[157]. For implementation of ADSC/EV therapeutics, clinicians should adhere to the International Society for Stem Cell Research clinical translation standards (GMP-grade inputs, traceability, potency assays) and rigorous donor-screening/quality systems[158]. For EV products clinicians should follow MISEV2023 for identity, purity, and potency characterization to reduce batch variability and improve safety/read-across into trials[159,160]. Finally, practical selection should respect estimated glomerular filtration rate thresholds for SGLT-2 inhibitor initiation (generally ≥ 20 mL/minute/1.73 m² per the American Diabetes Association) and integrate weight-management guidance to match phenotype and treatment goals[47,152].

Synergistic pathways linking ADSCs with standard therapies

Beyond merely additive effects the ADSC secretome and its EVs intersect with the pharmacological mechanisms of conventional agents at several key nodes. Metformin through activation of AMPK acts synergistically with ADSC-EV-delivered miR-126, which promotes VEGF/PI3K-Akt/endothelial nitric oxide synthase signaling, thereby improving vascular endothelial function[161-164]. GLP-1RAs activate the cAMP/PKA-cAMP response element-binding protein pathway, aligning with ADSC-mediated AMPK/sirtuin 1-PGC-1α-PRDM16/UCP1 signaling to enhance thermogenesis and repair mitochondrial damage[165,166]. SGLT-2 inhibitors combine the reduction of circulating volume and cardiac preload/afterload with the anti-inflammatory effects of ADSC-EV-derived miR-146a and miR-223-3p, which inhibit IL-1 receptor-associated kinase 1/TAK1-TRAF6-NF-κB/NLRP3 signaling, thereby further reinforcing anti-fibrotic responses[167-169].

Pioglitazone via PPARγ activation increases ADSC survival, migration, and VEGF secretion, creating favorable microenvironmental conditions prior to cell or EV administration. At the insulin-signaling level ADSC-mediated downregulation of SOCS3 and reduced IRS-1(Ser-P) act synergistically to restore PI3K-Akt-GLUT4 translocation[170,171]. These mechanistic interactions can be experimentally validated through paired adipose tissue biopsies and analysis of EV molecular cargo (Figure 4).

Figure 4 Adipose-derived stem cell-drug synergy map. Adipose-derived stem cell secretome/extracellular vesicles target drug-action pathways to improve vascular, metabolic, and anti-inflammatory outcomes. Mechanistic intersections whereby adipose-derived stem cell (ADSC) secretome/extracellular vesicles (EVs) augment and complement conventional agents. Metformin (adenosine monophosphate-activated protein kinase) cooperates with ADSC-EV-delivered miR-126 to enhance vascular endothelial growth factor → phosphoinositide 3-kinase/protein kinase B → endothelial nitric oxide synthase signaling and improve endothelial function. Glucagon-like peptide-1 receptor agonist (cyclic adenosine monophosphate/protein kinase A → cyclic adenosine monophosphate response element-binding protein) aligns with ADSC-driven adenosine monophosphate-activated protein kinase/sirtuin-1 → peroxisome proliferator-activated receptor γ coactivator-1α → PR domain zinc finger protein 16/uncoupling protein-1 programs to promote beiging, mitochondrial repair, and fatty-acid oxidation, reducing ceramide/diacylglycerol lipotoxicity. The hemodynamic/anti-inflammatory effects of sodium-glucose cotransporter-2 inhibitors combine with EV miR-146a and miR-223-3p to suppress interleukin-1 receptor-associated kinase-1/transforming growth factor-β-activated kinase-1-tumor necrosis factor receptor-associated factor-6-nuclear factor kappa B → NOD-like receptor protein-3 signaling and reinforce anti-fibrotic responses (mothers against decapentaplegic homolog 2/3 modulation). Peroxisome proliferator-activated receptor γ activation by pioglitazone increases ADSC survival, migration, and vascular endothelial growth factor secretion, providing microenvironmental optimization for subsequent cell/EV therapy. At the insulin-signaling node ADSC effects (suppressor of cytokine signaling-3 ↓ and reduced serine-phosphorylated insulin receptor substrate-1) synergize with drug actions to restore phosphoinositide 3-kinase/protein kinase B/glucose transporter-4 translocation and adipose glucose uptake. These interactions are testable by paired adipose biopsies and serial EV-cargo profiling (e.g., miR-126/146a/223), together with functional readouts (flow-mediated dilation, estimated glomerular filtration rate slope, adipose peroxisome proliferator-activated receptor γ coactivator-1α/uncoupling protein-1, phosphorylated insulin receptor substrate-1). Solid arrows indicate activation/flow; T-bars or dashed lines indicate inhibition or context-dependent routes. IRAK1: Interleukin-1 receptor-associated kinase-1; TAK1: Transforming growth factor-β-activated kinase-1; TRAF6: Tumor necrosis factor receptor-associated factor-6; NF-κB: Nuclear factor kappa B; NLRP3: NOD-like receptor protein-3; PPARγ: Peroxisome proliferator-activated receptor γ; ADSC: Adipose-derived stem cell; EVs: Extracellular vesicles; AMPK: Adenosine monophosphate-activated protein kinase; SGLT-2i: Sodium-glucose cotransporter-2 inhibitor; VEGF: Vascular endothelial growth factor; PI3K: Phosphoinositide 3-kinase; Akt: Protein kinase B; miR: MicroRNA.

CONCLUSION

The co-occurrence of obesity and T2DM is driven by underlying mechanisms including VAT accumulation, insulin resistance, lipotoxicity, chronic low-grade inflammation, and mitochondrial dysfunction. Existing pharmacological agents such as pioglitazone and metformin have demonstrated significant efficacy in improving fat distribution and metabolic profiles, particularly through mechanisms involving fat redistribution, adiponectin secretion, AMPK activation, and anti-inflammatory modulation. The emergence of novel therapies such as GLP-1RA and SGLT-2 inhibitors has further expanded therapeutic targets, offering promising avenues for multifaceted adipose tissue remodeling and systemic metabolic improvement. Simultaneously, ADSCs by virtue of their differentiation capacity, paracrine activity, and immunomodulatory properties are becoming powerful tools in regenerative medicine to combat obesity-related metabolic disorders. ADSCs can not only induce adipose beiging and enhance mitochondrial function but also reshape the adipokine profile and inflammatory microenvironment through exosome-mediated intercellular communication. Importantly, ADSCs exhibit substantial synergistic potential when combined with various insulin-sensitizing agents, enabling coordinated regulation of insulin sensitivity, inflammation, and lipid metabolism across multiple tissues. Taken together, the integration of pharmacologic and stem cell-based interventions provide a robust mechanistic rationale and demonstrates encouraging preclinical efficacy.

Looking ahead, successful clinical translation will rely on individualized strategies informed by multiomics profiling, standardized ADSC manufacturing and quality-control protocols, and comprehensive long-term safety evaluation. To ensure conceptual coherence, we adopted a single precision-based, combined framework centered on adipose tissue remodeling-insulin resistance-stem cell regulation. Within this schema inflammation resolution is treated as a facet of adipose tissue remodeling, and tissue regeneration is encompassed by stem cell regulation. Grounded in this unified axis, targeted combinations hold the potential to deliver transformative prevention and management of obesity-associated T2DM.