Metformin and Adipose Tissue: A Multifaceted Regulator in Metabolism, Inflammation, and Regeneration

Abstract

Metformin, a first-line treatment for type 2 diabetes mellitus (T2DM), has garnered attention due to its pleiotropic effects beyond glycemic control. In addition to improving insulin sensitivity and inhibiting hepatic gluconeogenesis, metformin modulates inflammation, oxidative stress, and cellular metabolism, particularly within adipose tissue. Adipose tissue is a crucial endocrine organ that plays a central role in metabolic homeostasis and is integral to obesity and T2DM pathogenesis. This review discusses the actions of metformin on white and brown adipose tissues, extracellular matrix remodeling, mature adipocytes, and adipose-derived stem cells. Through both AMP-activated protein kinase-dependent and independent pathways, metformin influences adipogenesis, lipolysis, adipokine secretion, fibrosis, and cellular senescence. We also focus on how the drug concentration influences its biological effects, emphasizing the necessity of physiologically relevant experimental models. These insights deepen our understanding of metformin’s therapeutic potential in metabolic and age-related disorders.

GRAPHICAL ABSTRACT

INTRODUCTION

Metformin has been widely utilized in clinical practice for decades as a first-line therapy for type 2 diabetes mellitus (T2DM). It effectively controls blood glucose levels by increasing insulin sensitivity, inhibiting hepatic glucose production, and promoting glucose uptake in peripheral tissues. Recent research indicates that metformin may possess additional therapeutic properties beyond traditional antidiabetic effects, such as anti-cancer [1], anti-inflammatory, and antioxidant properties [2], and these possibilities have attracted extensive scientific interest.

Adipose tissue, historically viewed merely as a passive energy storage organ, is now recognized as an active endocrine organ critical for maintaining systemic metabolic homeostasis [3]. Metformin exerts multifaceted effects on adipose tissue by modulating adipocyte differentiation, lipid metabolism, and adipokine secretion; this provides a mechanistic link between adipose tissue remodeling and the broad metabolic benefits of metformin.

Although metformin has been extensively studied, a comprehensive review encompassing its developmental history, clinical applications, molecular mechanisms, newly discovered therapeutic functions, and specific mechanisms in adipose tissue remains lacking. Thus, this review aims to provide a systematic overview of metformin, particularly emphasizing its recently discovered therapeutic potential and its effects on adipose tissue. It seeks to establish mechanistic connections between adipocyte pathophysiology and systemic metabolic outcomes, thereby informing translational research strategies for therapeutic optimization.

OVERVIEW OF METFORMIN

Brief developmental history

Metformin, a classic hypoglycemic drug used for nearly a century, traces its origins back to botanical therapies in medieval Europe [4]. Although diabetes pathophysiology was poorly understood historically, plants rich in guanidine compounds, such as Galega officinalis (goat’s rue), were utilized as herbal remedies for polyuria symptoms [5]. Early 20th-century scientific advances revealed guanidine derivatives as active hypoglycemic constituents in G. officinalis [6]. Subsequently, various guanidine derivatives—including monoguanidine, diguanidine, and phenformin—were artificially synthesized or isolated [7]. Initially, metformin and related guanidine compounds were withdrawn from clinical use due to concerns about biological toxicity and heightened risks of lactic acidosis. Additionally, the introduction of insulin as a direct hypoglycemic treatment relegated metformin to obscurity for half a century [8,9]. Scientific and clinical interest in metformin gradually re-expanded due to its safe hypoglycemic properties and potential applications in other diseases like influenza and malaria [8,9]. Through sustained efforts from visionary clinicians such as Jean Sterne, clinical trials conclusively addressed previous safety concerns regarding metformin [10-12]. On December 29, 1994, the U.S. Food and Drug Administration officially approved metformin for sale in the United States [7]. Today, metformin is globally acknowledged as a first-line diabetes treatment, used by nearly 70% of diabetes patients worldwide, either alone or in combination with other antidiabetic agents [13]. With a large patient population and increasing clinical studies, a broader spectrum of metformin’s effects has gradually emerged.

Clinical use

As a first-line treatment for T2DM, metformin has unequivocally demonstrated efficacy in lowering blood glucose. It can be administered as monotherapy or combined with other hypoglycemic drugs. When used alone, metformin exhibits superior glucose control compared to other oral hypoglycemic agents [14]. Monotherapy can lower glycosylated hemoglobin A1C (HbA1c) levels by approximately 1% to 1.5% (14 mmol/mol) [15,16]. Furthermore, metformin more efficaciously controls blood glucose and body weight than dipeptidyl peptidase 4 inhibitors [17]. Recent clinical studies suggest that combining metformin with glycine and liraglutide can improve HbA1c control [18]. Metformin is particularly effective in preventing diabetes progression among prediabetic populations, notably patients with histories of gestational diabetes, severe obesity, or elevated fasting blood glucose [19,20]. Despite the emergence of newer antidiabetic agents such as glucagon-like peptide-1 (GLP-1) receptor agonists and sodium-glucose cotransporter-2 inhibitors, metformin remains the recommended first-line therapy due to its affordability, consistent efficacy, and favorable safety profile. Additionally, metformin has shown potential therapeutic benefits for chronic kidney disease, cardiovascular disease prognosis, polycystic ovary syndrome (PCOS), cognitive decline and Alzheimer’s disease, and even osteoarthritis [19,21-24].

Compared to other glucose-lowering medications, metformin poses a lower risk of hypoglycemia and lactic acidosis. Gastrointestinal intolerance is its primary side effect, and gradual dose titration effectively mitigates this issue [25]. The minimum recommended dose is 500 mg daily, with an optimal effective dose of 2,000 mg daily. For adults, the maximum dose of regular tablets is 2,550 mg daily, while the recommended maximum dose for extended-release formulations, which better reduce gastrointestinal adverse effects, is 2,000 mg daily [26]. Recent clinical guidelines from the American Diabetes Association have revised precautions regarding metformin use, affirming that it is safe to use in patients with an estimated glomerular filtration rate ≥30 mL/min/1.73 m² [25]. Notably, metformin may contribute to vitamin B12 deficiency; hence, periodic vitamin B12 screening is advisable [27].

Pharmacokinetics of metformin

The physicochemical properties of metformin (1,1-dimethylbiguanide hydrochloride) originate from its molecular structure, which exists predominantly as an organic cation under physiological pH. Initially, metformin absorption primarily takes place in the jejunum, where it enters intestinal epithelial cells through the luminal side via the plasma membrane monoamine transporter (PMAT, encoded by the solute carrier family 29 member 4 [SLC29A4] gene) [28]. Recent studies have shown that this transporter protein is closely associated with gastrointestinal adverse reactions [29]. Subsequently, metformin reaches the systemic circulation by crossing the basolateral membrane of intestinal epithelial cells through organic cation transporter 1 (OCT1, encoded by the SLC22A1 gene) [28]. However, the types of membrane protein transporters involved vary among different organs and cell types. For instance, OCT1 is expressed not only in intestinal epithelial cells but also on hepatocyte cell membranes, alongside OCT3 (encoded by the SLC22A3 gene) [30,31]. Moreover, OCT1 is expressed on adipocytes, and its expression is notably elevated in obese individuals [32]. OCT2 (encoded by the SLC22A2 gene) is predominantly present in renal epithelial cells, sharing a similar cation recognition mechanism with OCT1 [33]. Additionally, human multidrug and toxin extrusion proteins 1 and 2 (MATE1 and MATE2, encoded by the SLC47A1 and SLC47A2 genes, respectively) have been identified on the apical membrane of proximal renal tubule cells, playing a critical role in metformin excretion [34-36]. Notably, OCTs and MATEs have also been identified in endometrial epithelial cells, implicating them in PCOS treatment [37]. At the intracellular organelle level, metformin regulates lipid metabolism through mitochondrial oxidized nicotinamide adenine dinucleotide (NAD⁺) transporters, particularly SLC25A47 [38,39]. Therefore, the distribution and function of metformin transport proteins expressed in the plasma and organelle membranes across various tissues significantly influence metformin’s therapeutic efficacy.

Debates regarding experimental concentrations of metformin

Researchers have drawn attention to the concentrations of metformin utilized in experimental research as a concern [40]. Numerous previous in vitro studies employed relatively high metformin concentrations, often inconsistent with the actual plasma concentrations observed in vivo [41-43]. This discrepancy has raised questions about the relevance of proposed mechanisms. For example, metformin only inhibits mitochondrial respiratory chain complex I at supratherapeutic concentrations [44]. Clinical evidence supports this concern, as the recommended metformin dose is approximately 2 g/day. Given its short half-life and limited oral bioavailability (around 50%), plasma concentrations rarely surpass 40 μM [41,45]. Animal studies indicate that therapeutic metformin concentrations range from approximately 40 to 70 μM in the portal vein and from 10 to 40 μM in the systemic circulation [42]. Researchers are increasingly using lower, physiologically relevant metformin concentrations to better understand its mechanisms within realistic in vivo contexts [46,47]. However, replicating prolonged therapeutic blood concentrations remains challenging in tissue and cell culture systems. Therefore, although comparable transient concentrations may be attainable, further research is needed to elucidate the mechanisms underpinning metformin’s sustained effects. Tissue distribution studies employing isotopic labeling, such as with ¹¹C-labeled metformin, have shown predominant accumulation in the kidneys and liver, with hepatic uptake unaffected by pathological changes seen in non-alcoholic fatty liver disease [48]. Metformin also accumulates in the salivary glands, skeletal muscle, and intestinal tissues. These insights significantly enhance our understanding of the tissue-specific effects and potential therapeutic targets of metformin [49,50].

Molecular mechanisms

Recent research into the subcellular mechanisms of metformin primarily focuses on mitochondria and lysosomes. Although the exact concentration at which metformin exerts its effects remains debated, the classical view suggests that it inhibits respiratory chain complex I on the mitochondrial membrane [44]. Complex I catalyzes the dehydrogenation of reduced nicotinamide adenine dinucleotide (NADH) to NAD⁺, subsequently facilitating the generation of adenosine triphosphate (ATP) through the Krebs cycle. Inhibition of complex I reduces the intracellular ratio of ATP to adenosine monophosphate (AMP) [51]. Changes in the ratios of ATP, adenosine diphosphate (ADP), and AMP affect the phosphorylation of threonine 172 on the α subunit of AMP-activated protein kinase (AMPK) and thus affect AMPK activity [52]. In hepatic cells, activation of AMPK partly involves the protein-threonine kinase liver kinase B1 (LKB1), a critical component of metformin’s molecular mechanism [53]. This pathway significantly contributes to conditions such as metastatic prostate cancer, cardiovascular disease, and non-small cell lung cancer, a fact that provides insights into the anti-cancer effects of metformin [54-56]. In 2017, another mechanism was identified, where the glycolytic intermediate fructose-1,6-bisphosphate (FBP) regulates AMPK activity through aldolase enzymes, independently of ATP, ADP, and AMP concentrations [57]. Furthermore, it was discovered that aldometanib, an aldolase inhibitor, activates AMPK through the lysosomal pathway, and notably prolongs the healthy lifespan of nematodes and mice [58]. AMP accumulation can also inhibit the conversion of ATP to cyclic AMP (cAMP), subsequently decreasing protein kinase A (PKA) activity [59]. The role of this pathway in metformin-induced suppression of triple-negative breast cancer stem cells has also been validated [60]. An analysis of metformin’s mechanisms using phosphoprotein enrichment and quantitative proteomics indicated that 44% of proteins function via an LKB1-dependent mechanism [61]. These findings suggest that alternative pathways, such as aldolases and cAMP-related signaling, remain promising areas for further investigation. Additionally, a regulatory effect on the cellular redox state via non-competitive inhibition of mitochondrial glycerophosphate dehydrogenase, leading to reduced blood glucose levels, has been confirmed [62]. Importantly, this gluconeogenesis-inhibiting mechanism occurs at low metformin concentrations [63].

Traditionally, lysosomes have been recognized primarily as cellular degradation centers. However, deeper insights now acknowledge their involvement in regulating cellular metabolism and signal transduction [64]. A pivotal study in 2016 demonstrated that metformin promotes the translocation of scaffold protein AXIN, recruiting LKB1 to the lysosomal surface by directly interacting with vacuolar-type H(+)-ATPase (V-ATPase) regulators [65]. This interaction subsequently activates AMPK while concurrently inhibiting mechanistic target of rapamycin complex 1 (mTORC1) [65]. v-ATPase is a eukaryotic protein complex responsible for proton transport from the cytoplasm to lysosomes against concentration gradients [66]. Subsequent research further confirmed that at low concentrations, metformin inhibits v-ATPase activity by facilitating the binding of presenilin enhancer 2 (PEN2) with ATPase H⁺ transporting accessory protein 1 (ATP6AP1), an auxiliary v-ATPase factor, thereby activating AMPK independently of AMP or ADP accumulation [47].

Research on the subcellular mechanisms of metformin has predominantly emphasized mitochondria and lysosomes, and these pathways frequently converge on the key signaling node, AMPK. Indeed, AMPK plays a central role in mediating metformin’s effects, regulating cellular glucose metabolism, lipid metabolism, amino acid metabolism, autophagy, and mitochondrial function [67]. A promising focus in ongoing research may be the mechanisms by which metformin acts upon lysosomes at physiologically relevant concentrations. Further exploration of these pathways may uncover additional therapeutic potential of metformin in treating conditions such as diabetes, cancer, and aging.

METABOLIC FUNCTION OF ADIPOSE TISSUE

The primary function of adipose tissue is lipid storage and metabolism. Fatty acids stored within adipose tissue predominantly originate from triacylglycerols (TAGs), either transported by plasma lipoproteins or synthesized de novo within adipocytes. Lipoprotein lipase (LPL), secreted by adipocytes or muscle cells, binds to glycosylphosphatidylinositol-anchored high-density lipoprotein-binding protein 1 (GPIHBP1) located on the endothelial surface of capillaries, facilitating the hydrolysis of circulating lipoproteins. LPL enzymatically cleaves triglycerides within chylomicrons and very-low-density lipoproteins (VLDLs), releasing free fatty acids that are subsequently absorbed by adipocytes through CD36-positive fatty acid translocase on the plasma membrane [68]. Glucose uptake into adipocytes primarily occurs via insulin-responsive glucose transporter 4 (GLUT4) localized on the cell membrane. Under low plasma insulin conditions, GLUT4 remains within intracellular vesicles. Upon insulin stimulation, GLUT4 relocates to the plasma membrane, where it facilitates glucose entry into cells [69].

Within adipocytes, glucose serves as a primary energy source and supplies 3-phosphoglycerol for TAG synthesis through glycolysis and the tricarboxylic acid cycle, while also providing acetyl coenzyme A for de novo lipogenesis [70]. The conversion of 3-phosphoglycerol and fatty acids—from glycolysis, endogenous adipocyte stores, or exogenous sources—is catalyzed by enzymes such as glycerol-3-phosphate acyltransferase (GPAT) and 1-acylglycerol-3-phosphate O-acyltransferase (AGPAT), forming diacylglycerol (DAG). DAG is subsequently esterified by diacylglycerol acyltransferase (DGAT) into TAG. Under conditions of elevated energy demand, such as fasting, exercise, or stress, hormones including glucagon, catecholamines, natriuretic peptides, thyroid hormones, and tumor necrosis factor-α (TNF-α) stimulate lipolysis by increasing intracellular cAMP and cyclic guanosine monophosphate (GMP). This mobilization of stored fat begins with adipose triacylglyceride lipase (ATGL, encoded by patatin like domain 2, triacylglycerol lipase [PNPLA2]), which hydrolyzes TAG into DAG and a free fatty acid. Hormone-sensitive lipase (HSL, encoded by lipase E, hormone sensitive type [LIPE]) further hydrolyzes DAG to monoacylglycerol (MAG) and another fatty acid [71]. Lastly, monoacylglycerol lipase (MAGL) hydrolyzes MAG into glycerol and an additional fatty acid [71].

EFFECTS OF METFORMIN ON ADIPOSE TISSUE

In recent decades, adipose tissue has increasingly been recognized as playing important endocrine roles, in addition to its function in energy storage and thermal insulation. Advances in basic science and analytical techniques have shown that adipose tissue comprises diverse cell types, including immune cells, stromal stem cells, and others [72]. Since the discovery of leptin in the 1990s, our understanding of adipose tissue has expanded beyond its role in energy storage and expenditure to include its endocrine functions.

The impact of metformin on various categories of adipose tissue

From functional and morphological standpoints, adipose tissue is classified into white adipose tissue (WAT) and brown adipose tissue (BAT), each primarily composed of its respective adipocyte subtype.The distribution and characteristics of different types of adipose tissue and the effects of metformin on them are illustrated in Fig. 1.

Fig. 1.

The distribution and characteristics of different types of adipose tissue and the effects of metformin on them. ↑ represents promotion, ↓ represents inhibition. White adipose tissue is widely distributed subcutaneously throughout the body, especially in the abdomen, buttocks, thighs, and upper arms (subcutaneous fat), surrounding organs such as the liver and intestines (visceral fat), and in eye sockets, breasts, etc. (other special areas). Brown adipose tissue is mainly distributed in the supraclavicular area, deep neck, perivascular area (such as around the aorta and renal artery), paraspinal area, and perirenal area. UCP1, uncoupling protein 1.

Brown adipose tissue

BAT is typically located in highly vascularized areas in both humans and mice [73]. Neonates exhibit BAT analogous to that found in mice, notably in the scapular region [74], although this BAT diminishes with age. Positron emission tomography imaging has revealed residual BAT in adults, particularly around the neck, clavicle, and vertebral regions [75]. BAT, which is predominantly composed of brown adipocytes, shares immunological markers similar to skeletal muscle and uniquely expresses myogenic factor 5 (Myf5) and paired box 7 (Pax7), which are absent in white adipocytes. This indicates that brown adipocytes derive from a distinct lineage [76]. The transcriptional regulator PR domain-containing 16 (PRDM16) activates peroxisome proliferator-activated receptor gamma (PPAR-γ) expression, facilitating differentiation into brown adipocytes and skeletal muscle cells, but not white adipocytes [77]. PRDM16 is pivotal in thermogenesis and broader metabolic regulation [78]. Morphologically, brown adipocytes feature multiple small lipid droplets dispersed throughout their cytoplasm and abundant mitochondria, essential for their thermogenic function. Uncoupling protein 1 (UCP1), a key thermogenesis mediator, disrupts oxidative phosphorylation upon activation, inhibiting ATP production and dissipating energy as heat via non-shivering thermogenesis [77].

Given BAT’s unique energy-expenditure role, metformin may exert therapeutic benefits by upregulating its thermogenic function. Unlike energy-storing adipose tissue, BAT predominantly functions in energy consumption and heat production, and metformin promotes BAT proliferation, differentiation, and metabolic capacity [79]. Metformin enhances BAT activity by upregulating BAT-specific genes such as UCP1 via the nuclear factor erythroid 2-related factor 2 (Nrf2)/heme oxygenase-1 (HO-1) signaling pathway [80]. Additionally, metformin administration in obese mouse models increases BAT-specific markers, including UCP1, ELOVL fatty acid elongase 3 (Elovl3), and Cidea [81]. Its thermogenic capacity can also be maintained through activation of the deacetylase sirtuin 1 (SIRT1), reducing brown adipocyte senescence. Studies in progeroid mice have demonstrated that metformin-induced SIRT1 activation significantly reduces mitochondrial oxidative stress in brown adipocytes, alleviates pericellular inflammation, and delays senescence-related functional decline [82]. Furthermore, metformin increases BAT activity and the clearance of VLDL absorbed by BAT, intensifying excess energy consumption [83]. Metformin also attenuates the inflammatory response in BAT induced by hypoxia-inducible factor-α (HIF-α), restoring insulin sensitivity in brown adipocytes [84]. Long-term metformin use reduces inflammation in BAT, increases systemic energy expenditure, and promotes BAT activation. Investigation into metformin’s positive impact on BAT differentiation and metabolism highlights its therapeutic potential, making BAT a promising intervention target [85].

However, some studies suggest that metformin may not significantly increase BAT activity; notably, these studies were short-term (less than 2 weeks) in mouse models [85]. However, it is worth noting that, except for the study on VLDL uptake conducted in animal models, the aforementioned findings were largely derived from experimental models using metformin at millimolar concentrations. It remains unclear whether these supraphysiological levels can be replicated in vivo or may lead to unforeseen side effects. Moreover, given the limited distribution of BAT in the human body, whether its functional changes are sufficient to induce significant metabolic improvements remains a matter of debate.

White adipose tissue

WAT is the predominant adipose depot in the human body, widely distributed across visceral and subcutaneous regions, along with distinct localized fat reservoirs. It primarily consists of white adipocytes, characterized by a single large lipid droplet and relatively fewer mitochondria [3]. The principal functional role of white adipocytes is lipid storage within intracellular droplets during periods of energy surplus and mobilization of these energy reserves via metabolic cascades in response to physiological demands [86]. This pivotal function contributes significantly to overall energy homeostasis. Additionally, WAT provides essential thermal insulation, aiding in maintaining stable body temperature.

After 14 weeks of metformin treatment in mice, fibroblast growth factor 21 expression in WAT significantly increased, suppressing white adipocyte differentiation [81]. Concurrently, metformin reduced interleukin-17 (IL-17) levels, thereby alleviating the inflammatory state in mice [81]. Together, these effects contributed to enhanced glucose metabolism and reduced metabolic disturbances. Metformin also lowers the levels of various pro-inflammatory cytokines through decreased macrophage infiltration and inhibition of macrophage polarization, thus improving insulin sensitivity in adipose tissue [87]. Metabolomic and transcriptomic analyses indicate that short-term metformin administration downregulates adipose tissue genes involved in fatty acid synthesis, altering circulating fatty acid flux. Additionally, metformin suppresses genes related to AMPK inhibition, enhancing adipose insulin sensitivity [88]. Beyond metabolic effects, clinical research has demonstrated that metformin-induced weight loss, improved insulin sensitivity, and reduced pro-inflammatory cytokines in patients with metabolic syndrome correlate with increased adiponectin and decreased leptin and resistin secretion [89]. Notably, metformin’s effects on increasing adiponectin and reducing leptin secretion are especially pronounced in obese individuals with a body mass index exceeding 30 kg/m² [90]. These adipokines play crucial roles in appetite regulation, insulin sensitivity, and systemic inflammation. Thus, metformin’s direct impact on adipose tissue modulates systemic metabolic homeostasis by regulating adipokine secretion.

Metformin exerts multifaceted actions on WAT, encompassing anti-inflammatory effects, metabolic improvements, and promoting adipose browning. These benefits primarily result from AMPK pathway activation, modulation of macrophage polarization, and regulation of adipokine balance. Together, these mechanisms underpin metformin’s anti-obesity and antidiabetic properties mediated through adipose tissue [91]. However, the precise molecular targets responsible for these effects remain unclear. Moreover, numerous studies have reported that metformin inhibits adipocyte differentiation and downregulates fatty acid synthesis—findings that seemingly contradict the assumption that glucose-lowering effects involve increased energy substrate uptake in adipose tissue [81,88]. The mechanisms underlying these paradoxical observations require further investigation.

Anatomical distribution of adipose tissue

In addition to morphological distinctions, adipose tissue’s anatomical distribution has significant clinical implications. Adipose tissue is typically categorized into visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT). VAT accounts for approximately 10% to 20% of total body fat mass, primarily surrounding the mesentery and digestive organs. Conversely, SAT constitutes over 80% of body fat, predominantly located in the gluteal, femoral, and abdominal regions. Adult men generally possess higher VAT proportions than women. Clinically, VAT has a stronger association with hepatic steatosis, insulin resistance, and cardiovascular diseases than SAT, likely due to its ability to deliver substantial lipid amounts directly to the liver through the portal vein, causing metabolic disturbances in hepatocytes [71,92]. Owing to limited stem cell recruitment and reduced adipogenic capacity in VAT, adipocytes in this depot tend to hypertrophy. In contrast, SAT adipocytes tend toward proliferation due to an abundance of mesenchymal progenitor cells and elevated adipogenic activity [3,93]. In contrast, dyslipidemia has been associated with VAT, whereas blood glucose and insulin abnormalities have been linked to SAT [94].

Drug therapies typically target SAT, aligning well with metformin’s therapeutic profile [46]. In clinical studies, metformin-induced weight loss correlated significantly with SAT reduction, whereas its effects on visceral or hepatic fat were comparatively modest [95]. Metformin may exert beneficial effects by improving energy metabolism and insulin sensitivity in SAT through mechanisms such as reducing adipocyte hypertrophy and alleviating inflammatory responses [89,96]. Within VAT, metformin increases metabolic function primarily by reducing pro-inflammatory cytokines (e.g., TNF-α and IL-6) and oxidative stress, consequently improving systemic inflammation and insulin resistance [97].

Metformin exhibits a “dual effect” on SAT and VAT. In SAT, metformin predominantly reduces tissue volume and enhances metabolic function, whereas in VAT, its actions mainly involve modulation of inflammation and fibrosis rather than pronounced volume reduction. These divergent effects may stem from inherent biological differences between these adipose depots and their distinct roles in systemic metabolism.

REPERCUSSIONS OF METFORMIN ON VARIOUS CELL TYPES WITHIN ADIPOSE TISSUE

Adipose tissue is a multifunctional organ predominantly composed of adipocytes, but it also includes diverse cell types, such as adipocyte precursor cells, adipose-derived stem cells (ASCs), and immune cells. These components constitute the stromal vascular fraction. Disruptions or alterations in these cellular constituents or in the extracellular matrix (ECM) can impair adipose tissue functionality and contribute to disease progression. Pathological conditions modulate the adaptive and maladaptive responses of adipose tissue, thus influencing systemic metabolic homeostasis. Effects of metformin on them are illustrated in Fig. 2.

Fig. 2.

Effects of metformin on various cell types in adipose tissue, including extracellular matrix (ECM), mature adipocytes, and adipose-derived stem cells (ASCs). ↑ represents promotion, ↓ represents inhibition. MMP, matrix metalloproteinase; MSC, mesenchymal stem cell; GLUT4, glucose transporter 4; ROS, reactive oxygen species.

Extracellular matrix

The ECM, which is primarily composed of collagen, fibrin, and proteoglycans, constitutes an essential microenvironment for various cell types [98]. Current evidence highlights a strong association between collagen VI expression, adipocyte expansion, and systemic metabolic regulation. Under hypoxic conditions, increased expression of collagen I and III occurs, largely due to HIF-1α activation [99]. It is reasonable to postulate that the ECM likely influences adipose tissue metabolism through pathological hypoxia and inflammatory responses, driving upregulation of various ECM components. This hinders orderly adipocyte expansion and disrupts adipocyte metabolic function. Local tissue dysfunction—characterized by adipocyte death, impaired angiogenesis, and chronic inflammation—directly diminishes insulin sensitivity and metabolic performance of adipocytes, thereby contributing to systemic metabolic dysregulation [100]. Matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases exert a pivotal role in stimulating ECM synthesis [100].

Metformin has demonstrated the capability to inhibit transforming growth factor-beta1 (TGF-β1)-induced fibrosis via AMPK activation, thus mitigating adipocyte senescence and insulin resistance arising from ECM abnormalities [101]. MMPs are also therapeutic targets of metformin, with significantly reduced expression levels of major fibrogenic MMPs following treatment [102]. Regarding fibrosis origins, metformin limits mesenchymal stem cell (MSC) differentiation into myofibroblasts through AMPK activation and inhibition of the phosphoinositide 3-kinase (PI3K) pathway, restoring redox homeostasis [102]. Similar effects were observed in mouse SAT [103]. Therapeutically, inhibition of the downstream PI3K/Akt signaling pathway reduces adipogenesis and improves ECM microenvironments, potentially ameliorating T2DM [104]. Whereas autophagy exerts beneficial cellular functions, deficiencies in autophagy genes alter ECM remodeling induced by obesity [105]. High metformin concentrations can reduce type I collagen deposition and inflammatory cytokine expression in adipose tissue, promoting improved wound healing after burn injuries [106]. However, whether the high concentrations (100 mM) used in in vitro experiments correlate to comparable effects in vivo remains an open question that warrants further investigation. In addition to adipose tissue, it is noteworthy that the regulatory effect of metformin on ECM has been harnessed in targeted nanomedicine to enhance the efficacy of targeted therapy, thereby substantiating its regulatory impact on ECM generation [107].

Metformin regulates multiple cellular targets and signaling pathways within the ECM, exerting effects both indirectly through modulation of cell differentiation and directly by structurally improving the ECM microenvironment. These mechanisms contribute to restoring ECM metabolic homeostasis, positioning metformin as a promising therapeutic agent for obesity- and diabetes-associated adipose fibrosis.

Mature adipocytes

Mature adipocytes constitute the predominant component of adipose tissue and are essential for executing its overall functions. The primary roles of adipocytes include metabolic regulation and distinct secretory activities.

Experimental evidence dating back to 1993 showed that metformin effectively attenuates chronic insulin exposure-induced reductions in GLUT4 expression on adipocyte membranes, indirectly supporting its potential in countering insulin-induced adipocyte senescence [108]. Similar results have been confirmed in human adipocytes [109]. The underlying mechanisms involve metformin’s modulation of insulin signaling pathways, AMPK activation, and enhanced GLUT4 translocation to the plasma membrane [110]. Metformin significantly reduces lipid accumulation in adipocytes by inhibiting insulin-induced lipogenesis and suppressing lipid droplet fusion [111]. It also attenuates adrenaline-or cAMP-stimulated lipolysis, reducing glycerol and free fatty acid release, thereby limiting adipocyte energy mobilization [112]. Even when hyperuricemia elevates free fatty acid levels and induces insulin resistance, metformin can mitigate these effects in white adipocytes [113]. Meanwhile, hyperinsulinemia and elevated blood glucose in T2DM contribute to adipocyte senescence. Classical senescence markers and additional indicators confirm the detrimental impact of senescence on adipocyte metabolic function and insulin sensitivity [114]. Under hyperinsulinemic conditions, metformin alleviates insulin-induced cellular senescence and reduces secretion of senescence-associated secretory phenotype factors and inflammatory cytokines [46]. These effects improve adipocyte function and may delay T2DM progression. Although direct evidence remains limited, metformin may indirectly stimulate differentiation of white adipocytes into beige adipocytes via AMPK pathway activation, enhancing energy expenditure [115]. Regarding exosome secretion, metformin attenuates exosome release from obese mouse white adipocytes through the AMPK pathway, mitigating fatty liver resulting from obesity-induced AMPK inhibition [116]. The molecular mechanisms underlying metformin’s direct regulation of adipocyte-derived exosomes remain unclear. It has yet to be determined whether metformin influences exosome biogenesis via classical pathways such as AMPK/mTOR or modifies exosomal cargo through epigenetic alterations [117]. Current research predominantly relies on cellular experiments or murine models, with scarce clinical data from human populations [118,119]. Moreover, heterogeneity among exosomes derived from different adipose depots has not been adequately explored [120]. The absence of standardized protocols for exosome isolation and characterization likely contributes to inconsistent reproducibility among studies.

By modulating metabolic activities, inflammatory responses, and differentiation processes in mature adipocytes, metformin exerts integrated effects such as reduced lipid storage, improved glucose metabolism, suppression of excessive lipolysis, and regulation of adipokine and exosome secretion. These actions underpin its therapeutic efficacy in obesity and diabetes. Additionally, metformin may influence systemic insulin sensitivity and cognitive function by regulating adipocyte-mediated metabolic pathways.

Adipose-derived stem cells

ASCs are a type of MSC with multipotent differentiation capabilities, providing significant potential in regenerative medicine, immunomodulation, and disease treatment. Owing to their accessibility, pluripotency, and robust paracrine activity, ASCs have become central tools in regenerative medicine, playing key roles in tissue repair, immune regulation, metabolic modulation, and potentially antifibrotic activities [121].

Metformin significantly inhibits the adipogenic differentiation of ASCs, reducing lipid droplet fusion and growth. This effect has been demonstrated in ASCs derived from both epididymal and inguinal fat depots. The underlying mechanism may involve the downregulation of adipogenesis-related proteins such as Cidec, Perilipin1, and Rab8a [122]. These observations align with earlier findings indicating that metformin suppresses differentiation and lipogenesis in 3T3–L1 cells [123]. Metformin administration significantly influences ASC morphology, proliferation rate, differentiation potential, and senescence, generally enhancing their proliferation and function. A murine study reported that prolonged metformin administration (8 weeks, 2.8 mg/day) increased ASC viability and proliferation while reducing their population doubling time [124]. These results align with findings from horse-derived ASCs, suggesting that metformin’s positive impact on viability, DNA synthesis, and metabolic efficiency correlates with increased mitochondrial membrane potential, reduced apoptosis, and elevated WNT3A/β-catenin expression [125]. Another study similarly demonstrated that metformin (5 mM) reduced the expression of lipogenic genes and lipid droplet accumulation, concurrently activating AMPK and thereby inhibiting human preadipocyte differentiation [32]. However, one specific study produced contrasting outcomes depending on metformin concentration: lower concentrations (1.25 to 2.5 mM) promoted adipogenesis potentially via an AMPK-independent pathway, whereas higher concentrations (5 to 10 mM) strongly inhibited adipogenesis through AMPK activation in 3T3–L1 cells [126]. Notably, metformin effectively prevents oxidative stress-induced senescence and dysfunction during ASC proliferation, thus restoring adipogenesis capacity [127]. In vitro studies have indicated that treatment with metformin significantly reduces oxidative stress in adipose tissue-derived precursor cells, especially those showing passage-related senescence from elderly donors, enhancing their adipogenic capacity and insulin sensitivity. In contrast, metformin exerts relatively modest effects in precursor cells from younger individuals. These anti-aging effects may partly involve AMPK phosphorylation and activation [127]. Furthermore, metformin increases the synthesis of membrane-derived vesicles, which contain numerous regenerative growth factors. Metformin-treated ASCs exhibit extensive networks of filopodia and lamellipodia [124]. These structures play a pivotal role in promoting both cell migration and extension, which are valuable characteristics in cells intended for regenerative medicine applications.

Metformin exerts multi-target regulatory effects on the metabolism, differentiation, immunomodulation, and regenerative capacity of ASCs, demonstrating therapeutic potential in obesity, diabetes complications, tissue engineering, and anti-aging therapies. Nevertheless, the precise mechanisms underlying these effects—particularly at physiologically relevant concentrations and the involvement of AMPK-dependent or independent pathways—remain unclear. Additionally, stem cell-based therapies have emerged as a promising area, yet detailed mechanistic studies and clinical validation in humans are necessary. Given metformin’s concentration-dependent influence on adipocyte differentiation and its differential anti-aging effects depending on cellular senescence status, carefully designed multicenter human studies with adequate sample sizes and ethical oversight are essential. Studies involving obese participants or individuals with diabetes must monitor both metabolic changes in transplanted ASCs and systemic metabolic improvements post-transplantation [128]. Long-term surveillance should include assessing abnormal tissue proliferation or tumor formation at transplantation sites, alongside evaluations of telomerase activity and senescence-associated genes such as p16 and p21 [129].

CONCLUSIONS

Metformin exerts profound and multidimensional effects on adipose tissue by modulating cellular metabolism, differentiation, inflammation, and regenerative processes. These actions occur through direct effects on mature adipocytes, ECM remodeling, and regulatory impacts on ASCs. Metformin improves insulin sensitivity, reduces excessive lipogenesis and lipolysis, alleviates inflammation, and delays cellular senescence primarily via AMPK activation. Specifically, metformin enhances metabolic function and limits hypertrophy in WAT, whereas it promotes thermogenesis and energy expenditure in BAT. The capacity of metformin to remodel adipose tissue at both cellular and molecular levels underlies its broad therapeutic potential in obesity, T2DM, and tissue engineering.

Despite promising findings, several challenges and questions persist. First, discrepancies between in vitro and in vivo studies— particularly regarding metformin concentrations—limit translational relevance. Many observed cellular effects occur at supraphysiological concentrations unattainable under standard clinical dosing, raising concerns about physiological applicability. Future studies should prioritize pharmacologically relevant concentrations and utilize advanced models such as organoids, three-dimensional cultures, or in vivo systems that better replicate systemic environments. Second, although AMPK is well-recognized as a critical target, increasing evidence indicates involvement of AMPK-independent mechanisms (e.g., lysosomal and redox regulation). Clarifying these additional pathways and their tissue-specific roles will enhance understanding of metformin’s mechanisms of action. Third, individual variability—including age, obesity status, genetic background, and adipose depot origin—may significantly influence metformin’s efficacy and mechanism of action. Implementing precision medicine approaches integrating genomics and metabolomics will help tailor metformin therapy across diverse populations. Finally, given the emerging significance of ASCs in regenerative medicine, understanding how metformin regulates their fate and function might pave the way for novel applications in tissue repair and anti-aging interventions. Nonetheless, long-term safety, dosage optimization, and potential off-target effects of metformin in stem-cell-based therapies require thorough evaluation.

In recent years, several novel antidiabetic agents, such as GLP-1 receptor agonists, have demonstrated both glucose-lowering and weight-reducing effects. Concurrently, GLP-1 receptor agonists have been observed to reduce pro-inflammatory cytokines (e.g., IL-1β and IL-6), thereby attenuating chronic inflammation. While metformin activates AMPK to enhance mitochondrial function, GLP-1 receptor agonists might further facilitate cellular repair via the PI3K/Akt signaling pathway. Combination therapies involving metformin and GLP-1 receptor agonists in diabetes models show superior metabolic improvements compared to monotherapy. These agents potentially exert complementary effects via distinct yet intersecting pathways like AMPK and PI3K/Akt, warranting further investigation into their molecular interactions.

In summary, although metformin remains foundational in managing metabolic diseases, expanding understanding of its role in adipose tissue biology and regenerative medicine opens new therapeutic avenues. Future research should aim to refine mechanistic insights, optimize dosing strategies, and explore novel therapeutic indications to fully harness the therapeutic potential of this century-old drug.