WAT is primarily composed of white adipocytes along with other cell types including stem cells, preadipocytes, and immune cells. Its vascular and neural innervation density is only 1/5 to 1/6 of that in BAT (26, 27). WAT is distributed in subcutaneous regions (abdomen, thighs, buttocks) and visceral depots (pericardium, gonads, mesentery, ligamentum teres hepatis, and retroperitoneum) (12). The spherical morphology of white adipocytes is characterized by a single, large lipid droplet that occupies approximately 90% of the cellular volume. Their primary physiological function is to store excess energy in the form of triglycerides to meet the body’s metabolic demands (26, 28, 29). Additionally, WAT serves an endocrine function through the secretion of adipokines that regulate various physiological processes (27, 28). Among these, adiponectin and leptin are particularly noteworthy. Adiponectin enhances insulin sensitivity while suppressing cell death and inflammation (30), whereas leptin reduces appetite and counteracts obesity (31).

BAT is composed of uncoupling protein 1 (UCP1) expressing brown adipocytes, abundant capillaries, and adrenergic nerve fibers (26, 28, 29). UCP1 is a transmembrane protein exclusively expressed in the inner mitochondrial membrane of brown adipocytes and beige adipocytes (32). BAT is more abundant in newborns and relatively scarce in adults, primarily distributed in specific anatomical regions such as the paraclavicular, paravertebral, and periadrenal areas (12, 26, 28). Multilocular lipid droplets and numerous large mitochondria packed with dense cristae are characteristic features of brown adipocytes (26, 28). The primary function of BAT is to generate heat through UCP1 mediated proton leak (33, 34). Beyond UCP1 dependent adaptive thermogenesis, brown adipose tissue utilizes additional thermogenic pathways. For example, calcium cycling facilitates thermogenesis via uncoupling of the sarco/endoplasmic reticulum Ca²⁺ ATPase (SERCA) calcium pump and its regulatory protein sarcolipin, while creatine enhances mitochondrial respiration by disrupting the adenosine triphosphate (ATP)/adenosine diphosphate (ADP) stoichiometric balance, significantly amplifying heat production under ADP limited conditions (33). Similar to WAT, BAT also secretes adipokines, referred to as “batokines” (35). Notably, neuregulin 4 (Nrg4), a secretory factor enriched in brown adipocytes, is significantly upregulated during their differentiation and has been shown to inhibit hepatic fatty acid synthesis (36).

In addition to WAT and BAT, WAT contains a distinct cell type termed “beige” or “brite” adipocytes. These adipocyte precursors typically exhibit characteristics similar to white adipocytes under basal conditions but acquire features resembling classical brown adipocytes upon specific stimulation (18, 37, 38). Emerging evidence also suggests that beige/brite adipocytes may directly transdifferentiate from mature white adipocytes (39). The distinction between beige and brite adipocytes lies in their lipid droplet morphology: beige cells are multilocular, whereas brite cells are paucilocular (40). Classical brown adipocytes and stimulus induced UCP1 expressing beige adipocytes originate from divergent lineages—the former deriving from myogenic factor 5 (Myf-5) positive myogenic precursors, and the latter arising from non Myf-5 lineages. Despite their developmental differences, both cell types co-express PR/SET domain 16 (PRDM16) and UCP1, functionally permitting the classification of beige adipocytes as “brown like” cells within white adipose depots (37, 38, 41, 42).

As mentioned earlier, WAT browning refers to the process in which brown like adipocytes appear within WAT (13). Specifically, when white adipocytes or beige adipocyte precursors are stimulated by certain conditions such as cold exposure, temperature receptors transmit signals to the hypothalamus, activating the sympathetic nervous system centrally and releasing norepinephrine to bind β3-adrenergic receptors on adipocyte membranes. This subsequently activates the adenylate cyclase-protein kinase A (AC-PKA) signaling pathway, leading to the activation of PPARγ coactivator-1α (PGC-1α). PGC-1α promotes UCP1 expression while free fatty acids (FFAs) released from triglycerides undergo aerobic oxidation in the respiratory chain, releasing H +. UCP1 acts as an H+ transporter, allowing H+ to flow along its concentration gradient into the mitochondrial matrix, uncoupling substrate oxidation from ADP phosphorylation and converting electrochemical potential energy into heat. Notably, beige adipocytes exhibit UCP1 expression levels comparable to classical brown adipocytes, thereby acquiring thermogenic capacity (33, 37, 38, 43). The induction of browning is influenced by multiple stimuli, which can be categorized into: environmental conditions (e.g., cold, physical activity); synthetic compounds (e.g., PPAR agonists (14, 18, 38, 44), β3-adrenergic receptor agonists (18, 37, 39, 43), irisin (18, 37); and nutrients (e.g., carotenoids, capsaicin, arginine) (18). Browning occurs more frequently in subcutaneous adipose tissue (18).

Insulin resistance leading to hepatic FFA deposition constitutes a core pathogenic mechanism in MASLD (9, 45). Substantial evidence demonstrates that WAT browning significantly enhances energy expenditure and improves systemic metabolism, manifesting as reduced body weight, improved insulin sensitivity, and attenuated hepatic steatosis and inflammation, particularly under high fat diet conditions (18, 46–51). The mechanistic basis involves browning induced generation of beige/brite adipocytes in WAT, which elevates thermogenesis through upregulated UCP1 expression and enhanced mitochondrial oxygen consumption, thereby promoting FFA catabolism and reducing hepatic lipid accumulation (18, 34, 39, 44, 49, 50, 52, 53). This is particularly relevant given that excessive intrahepatic triglyceride deposition represents a fundamental pathological feature of MASLD (54). Experimental studies show that n-3 polyunsaturated fatty acids (PUFAs) may induce adipocyte browning via PPARγ activation while increasing adipose Nrg4 production, collectively preventing hepatic steatosis. Similarly, PPARα stimulates hepatic fibroblast growth factor 21 (FGF21) production to promote WAT browning, increase energy expenditure, and alleviate hepatic steatosis (47). Beyond improving hepatic steatosis, browning inducing interventions in high fat diet fed mice also reduce hepatic inflammation, as evidenced by decreased proinflammatory cytokines and chemokines, elevated antioxidant gene expression, and increased populations of anti-inflammatory M2 macrophages (55–59). Concurrently, these treatments ameliorate liver fibrosis by suppressing profibrotic genes and facilitating the phenotypic transition of M1 Kupffer cells toward M2 subtypes (57–59). Although most studies attribute these anti-inflammatory effects to secondary metabolic improvements from browning (e.g., reduced steatosis and insulin resistance), emerging evidence directly implicates UCP1+ adipocytes in mitigating hepatic inflammation through reducing extracellular succinate levels. This metabolite normally activates succinate receptor 1 (SUCNR1) a G protein coupled receptor highly expressed on dendritic cells and macrophages to potentiate proinflammatory responses (32).

Numerous studies have investigated the browning of white adipose tissue in rodent models and isolated human cells. However, clinical trials focusing on white adipose tissue browning remain limited. These studies—utilizing morphological and immunohistochemical analyses, among other methods—have demonstrated that various activating factors can induce the browning phenomenon in human subcutaneous white adipose tissue. Nevertheless, they have not thoroughly explored the systemic metabolic implications of this browning process (60–62). One study showed that treatment with the β3-adrenergic receptor agonist mirabegron improved insulin resistance in subjects, increased the expression of beige adipocyte specific genes in subcutaneous WAT, and revealed a correlation between UCP1 protein levels and changes in insulin sensitivity (63). Another study found that sitagliptin enhanced [¹⁸F] FDG uptake in subcutaneous WAT of overweight prediabetic patients while improving glucose tolerance and lipid metabolism, suggesting that these metabolic benefits might be linked to adipose tissue browning (64). However, neither of these studies performed biopsies to directly confirm the presence of browning.

In summary, WAT browning can convert excess fatty acids into heat energy, thereby improving metabolic function. While numerous preclinical studies have demonstrated this effect, clinical research remains limited and insufficiently comprehensive. Further investigation is needed to determine the feasibility of this approach in humans. Nevertheless, WAT browning holds significant potential as a therapeutic strategy for ameliorating MASLD.

With the widespread application of PPAR agonists in the treatment of MASLD, comprehensive consideration of their safety profiles and the optimization of efficacy evaluation strategies have become particularly important. Previous studies have reported, especially for PPARγ agonists such as thiazolidinediones (TZDs), risks of congestive heart failure, edema, weight gain, and fractures (154–156). Animal studies have shown that upregulation of hepatic PPARγ may promote hepatic steatosis (138). Compared with placebo, elafibranor was more likely to cause abdominal pain, diarrhea, nausea, and vomiting in patients with primary biliary cholangitis (116). Aleglitazar demonstrated a higher incidence of safety issues, including heart failure, gastrointestinal bleeding, and renal impairment, which led to the early termination of the trial (148).

Regarding the efficacy evaluation of PPAR agonists for MASLD treatment, both histological examination (such as liver biopsy) and non-invasive tests (such as magnetic resonance elastography (MRE)) have their own advantages and disadvantages. Liver biopsy can directly observe liver pathology and is the most accurate diagnostic method, but it is an invasive procedure with associated risks and is not convenient for repeated testing. Noninvasive tests like MRE are simple to perform and can be repeated, making them suitable for long term monitoring, but they can only indirectly assess the condition and their accuracy may be affected by various factors. For example, in clinical trials of pemafibrate, reliance solely on MRE data may have compromised the reliability of the results (99). Therefore, future studies should strive to utilize both methods simultaneously to improve the accuracy of evaluation.

In summary, all three PPAR isoforms ameliorate MASLD through mechanisms including the reduction of hepatic lipid deposition, improvement of inflammation, and attenuation of fibrosis ( Table 1 ). However, the clinical efficacy of PPARα agonists remains controversial (102–105); clinical studies on PPARβ/δ agonists are still limited, and the safety profile of PPARγ agonists requires careful consideration. While PPARs represent potential therapeutic targets for MASLD, their specific clinical benefits warrant further investigation.

PPARα facilitates WAT browning. PPARα controls PRDM16 transcription and induces PGC-1α gene expression. PRDM16 cooperates with PGC-1α to regulate the browning process, providing essential conditions for brite adipocyte formation (18). PRDM16, a zinc finger protein, activates PGC-1α and PGC-1β through direct physical binding when expressed in white preadipocytes, broadly activating the brown adipocyte differentiation program. Adipose tissue specific overexpression of PRDM16 in mice promotes WAT browning (157). In human white adipocytes, PPARα overexpression or treatment with PPARα agonists increases the expression of brown adipocyte specific genes, including PRDM16, PGC-1α, and UCP1, demonstrating PPARα’s ability to promote white adipocyte browning (44, 158). PPARα mediated WAT browning is also associated with irisin (18, 159–161). Irisin induces PPARα to promote white adipocyte browning. Treatment of mouse primary white adipocytes with the PPARα antagonist GW6471 reduces UCP1, PGC-1α, and Cidea levels and attenuates irisin’s effects (159). Cidea is another BAT specific gene (162, 163). Fenofibrate treatment promotes WAT browning in mice on both standard and high fat diets, increasing brown adipocyte specific gene expression and irisin levels (160). The PPARα agonist Wy14643 improves insulin resistance in high fat diet fed mice, induces the appearance of beige adipocyte clusters in WAT, and elevates plasma irisin levels (161). However, some studies indicate that PPARα does not affect cold induced browning in mice but promotes β3-adrenergic receptor stimulation induced adipose tissue browning. This may relate to different stimulation mechanisms or compensatory effects of PPARγ during pharmacological activation (164). Dual PPARα/γ agonists more effectively induce WAT browning in obese mice. PPARα increases plasma FGF21 levels, which crosses the blood brain barrier to enhance β-adrenergic signaling. This process interacts with PPARγ activation to synergistically promote WAT browning (25). The mechanisms of PPARγ mediated WAT browning will be discussed later.

In BAT, PPARβ/δ activation induces the expression of genes associated with fatty acid oxidation and thermogenesis to exert its thermogenic effects (20, 165). However, research on whether PPARβ/δ can promote WAT browning remains limited. Some evidence suggests PPARβ/δ may facilitate WAT browning. In the WAT of obese mice, PPARβ/δ induces UCP1 to promote thermogenesis, which may be related to its interaction with PGC-1α. WAT specific PPARβ/δ overexpression mice exhibited significant histological changes in WAT, yet PPARβ/δ agonists failed to produce similar outcomes, potentially due to insufficient treatment duration (110). Leptin promotes browning of epididymal WAT in rats, a process involving PPARβ/δ. Treatment with a PPARβ/δ antagonist attenuates this browning effect, reducing expression of PPARγ and PRDM16 as well as UCP1 protein levels. This regulation may be mediated through FGF21 (166), which has been shown to directly modulate white adipocyte browning (25). However, this study lacked histological examination of rat adipose tissue. However, this study lacked histological examination of rat adipose tissue. Contradictorily, other research demonstrates that the PPARβ/δ agonist GW0742 does not promote WAT browning in mice fed either standard or high fat diets (161). In conclusion, whether PPARβ/δ promotes WAT browning requires further investigation.

As early as 1998, studies demonstrated that PPARγ agonists could increase UCP1 mRNA expression in human preadipocytes, confirming the presence of brown adipocytes within WAT isolated from perirenal fat depots (167). PPARγ promotes WAT browning through several mechanisms. PPARγ binds to the PGC-1α promoter to induce expression of brown adipose specific genes (168, 169). The PPARγ agonist rosiglitazone facilitates the conversion of white preadipocytes into brite adipocytes, accompanied by elevated levels of PGC-1α and UCP1 (38). Rosiglitazone also extends PRDM16 half-life through the ubiquitin proteasome pathway, thereby promoting WAT browning in mice (170). Additional studies suggest that PPARγ activation promotes white adipocyte browning by suppressing “visceral white” genes such as resistin and angiotensinogen. This effect is mediated through PPARγ’s recruitment of carboxy terminal binding proteins 1 (CtBP1) and CtBP2 into complexes containing C/EBPα at relevant promoters (171). Post translational modifications of PPARγ also significantly influence its browning inducing capacity. SIRT1 induces white adipocyte browning both in vivo and in vitro by deacetylating PPARγ at Lys293 and Lys268, thereby promoting PRDM16 recruitment. This process appears to involve sympathetic innervation, as both SIRT1 overexpressing mice and those lacking endogenous SIRT1 inhibitors exhibit enhanced cold induced white adipose browning (172). β3-adrenergic receptors have been shown to mediate this process in mouse white adipocytes (39). Furthermore, PRMT4 methylates PPARγ at Arg240, facilitating PRDM16 binding and initiating WAT browning and thermogenesis in mice (173). Inhibition of cyclin dependent kinase 5 (CDK5) mediated phosphorylation at PPARγ Ser273 by roscovitine promotes brite adipocyte formation in WAT (40).

In summary, activation of either PPARα or PPARγ promotes the emergence of beige/brite adipocyte clusters in WAT through mechanisms including induction of PRDM16 and PGC-1α expression, thereby exerting thermogenic and systemic metabolic regulatory effects. Additionally, PPARα mediated WAT browning is associated with irisin, while the post translational modification status of PPARγ determines its browning inducing capacity. Whether PPARβ/δ can promote WAT browning requires more direct experimental evidence. Although both PPARα and PPARγ can induce browning in human white adipocytes in vitro (44, 158, 167), whether they can elicit WAT browning in vivo requires further clinical investigation. Importantly, such studies would need to include histological examination of WAT in human subjects to confirm the occurrence of browning.