Emerging Insights into Brown Adipose Tissue Crosstalk With Pancreatic β-Cells in Metabolic Regulation

Abstract

Brown adipose tissue (BAT), traditionally recognized for its role in thermogenesis, has emerged as an active endocrine organ that coordinates systemic energy expenditure with glucose homeostasis. This review explores the emerging concept of bidirectional crosstalk between BAT and pancreatic β-cells, focusing on potential mechanisms through which BAT may regulate insulin secretion and β-cell survival. In addition to its thermogenic function, BAT serves as a metabolic sink and secretes various hormones (batokines), metabolites, and exosomes that can influence β-cell function directly or indirectly. Key batokines such as fibroblast growth factor 21, IL-6, ependymin-related protein 1, neuregulin 4, and phospholipid-transfer protein have shown potential in the preservation of β-cell health, although their clinical relevance requires further investigation. Emerging evidence also points to BAT-derived exosomes and microRNAs, including miR-26a, as novel regulators of insulin secretion. Neural mechanisms may contribute to this interorgan communication via sympathetic and sensory innervation, and BAT-derived neurotrophic factors may modulate autonomic inputs to peripheral tissues, including the pancreas. Conversely, β-cells influence BAT activation via hormonal (eg, insulin, glucagon), exosomal, and central pathways, forming a proposed BAT-brain-islet axis. This bidirectional communication appears disrupted in obesity and diabetes, where BAT dysfunction and β-cell stress exacerbate metabolic decline. Despite growing interest, mechanistic insights into BAT-islet crosstalk remain incomplete. Future research using omics technologies, co-culture systems, and in vivo manipulation models will be critical to identify novel mediators and clarify their roles in metabolic regulation. Understanding this interorgan communication may offer new therapeutic avenues for obesity and diabetes.

Coordinated Metabolism and the Need for Interorgan Communication

Maintaining energy and glucose homeostasis requires coordinated communication among organs that regulate energy intake, expenditure, and storage. Pancreatic β-cells, the primary source of insulin, play a critical role by adjusting insulin secretion in response to dynamic metabolic demands. They integrate signals from diverse tissues, including white adipose tissue (WAT), skeletal muscle, and liver to maintain glycemic control (1). Recently, brown adipose tissue (BAT), traditionally recognized as a thermogenic organ, has emerged as a potential regulator of this interorgan network (2). Beyond its role in heat production, BAT has now gained recognition as an endocrine and metabolic regulator that may communicate with β-cells to align systemic energy expenditure with glucose availability. Alterations in BAT function, mediated through sympathetic innervation, hormonal signals, and metabolic pathways, could potentially affect pancreatic β-cell function (2). Understanding this BAT-β-cell crosstalk could uncover novel mechanisms of metabolic regulation.

BAT as a Thermogenic Organ

BAT, characterized by abundant mitochondria, mediates nonshivering thermogenesis primarily via uncoupling protein 1 (UCP-1) (3, 4). In rodents, BAT is predominantly located in the interscapular region, whereas in adult humans, it is found in the supraclavicular area as well as deeper depots near the kidneys, spinal cord, and cervical area (5-8). BAT thermogenesis is tightly regulated by sympathetic innervation (Fig. 1): cold exposure triggers noradrenaline (NE) release, activating β-adrenergic receptors (β-ARs), which in turn stimulate lipolysis and release free fatty acids (FFAs), particularly long-chain fatty acids (9, 10). These long-chain fatty acids activate UCP-1, thus diverting stored energy from ATP synthesis to heat generation (11-13). Beige adipose tissue, induced within WAT under similar stimuli, shares thermogenic properties and represents an additional target for antiobesity therapies (14, 15). Recent discoveries of UCP-1-independent mechanisms, including creatine- and calcium-dependent pathways (14, 16-18), highlight the metabolic versatility of thermogenic fat.

Figure 1.

Dual thermogenic and endocrine functions of BAT. The upper panel illustrates the thermogenic function of BAT, which is activated by sympathetic nerves releasing NE that binds to β3-adrenergic receptors. This stimulates lipolysis and increases FFAs, which fuel UCP-1-mediated thermogenesis in mitochondria, leading to increased heat production, energy expenditure, insulin sensitivity, and glucose uptake. The lower panel depicts the endocrine function of BAT, which secretes a variety of factors, including cytokines (eg, IL-6), batokines (eg, IGF-1, FGF21, Nrg4, EPDR1, BMPs, GDF15, Adissp, and BATSP1), as well as metabolites and BAT-derived exosomes. These circulating factors influence interorgan communication and systemic metabolic homeostasis. The illustration was created with BioRender.com.

Abbreviations: Adissp, adipose-secreted signaling protein; β3-AR, β3-adrenergic receptor; BAT, brown adipose tissue; BATSP1, brown adipose tissue secreted peptide 1; BMP, bone morphogenetic protein; EPDR1, ependymin-related protein 1; FFA, free fatty acid; FGF21, fibroblast growth factor 21; GDF15, growth differentiation factor 15; NE, noradrenaline; Nrg4, neuregulin 4; UCP-1, uncoupling protein 1.

In addition to generating heat, BAT functions as a metabolic sink for circulating glucose and lipids (19), particularly under conditions of increased energy demand such as cold exposure (20, 21). Due to its inherent insulin sensitivity, BAT effectively lowers circulating glucose levels and takes up triglycerides, thereby reducing overall insulin demand and β-cell stress (21). Indeed, individuals with active BAT have lower fasting glucose levels and improved insulin sensitivity compared to those without detectable BAT (22).

BAT as an Endocrine Organ and a Systemic Metabolic Regulator

BAT is increasingly recognized as a secretory organ, releasing adipokines (termed batokines), cytokines, metabolites, and exosomes, collectively referred to as the “BAT secretome” (2, 23). These factors influence local and systemic energy metabolism, glucose homeostasis, and insulin sensitivity (24). Notably, mouse studies have shown that BAT transplantation significantly improves glucose tolerance and insulin sensitivity, effects partly mediated by BAT-derived IL-6 (25, 26). Other batokines, such as fibroblast growth factor 21 (FGF21) (27), neuregulin 4 (Nrg4) (28), IGF-1 (29), growth differentiation factor 15 (GDF15) (30), bone morphogenetic proteins (31), BAT-secreted peptide 1 (32), adipose-secreted signaling protein (33), and ependymin-related protein 1 (EPDR1) (34), have been shown to influence distant tissues including the liver, skeletal muscle, bone, heart, and pancreas (2, 24, 34). These endocrine activities highlight BAT's broader physiological role in maintaining metabolic health and suggest potential mechanisms by which BAT might influence pancreatic β-cell function.

Pancreatic β-cells and Physiological Basis of BAT-β-cell Crosstalk

Pancreatic β-cells are finely tuned to respond to dynamic nutrient and hormonal cues, adjusting insulin secretion in response to glucose, FFAs, gut hormones, and neuronal inputs. This regulatory flexibility ensures glucose homeostasis under varying metabolic states. However, insulin secretion must also be matched to changes in systemic energy expenditure, such as during BAT activation. Cold exposure or β-adrenergic stimulation increases BAT glucose and lipid uptake, rapidly lowering circulating glucose and insulin levels (35). To maintain homeostasis, β-cells likely receive and respond to BAT-derived signals indicating thermogenic demands. This implies the existence of regulatory circuits, likely hormonal, metabolic, and neuronal, that synchronize energy utilization with insulin dynamics.

Growing evidence suggests a bidirectional BAT-β-cell communication axis (Fig. 2). BAT-derived hormones [eg, FGF21, Nrg4, EPDR1, phospholipid-transfer protein (PLTP)], metabolites [eg, retinoic acid, 12,13-dihydroxy-9Z-octadecenoic acid, 5-oxoproline (5OP)], exosomes, and neuronal inputs may directly or indirectly influence β-cell survival and function (2). In parallel, pancreatic hormones, including insulin, glucagon, and amylin, as well as islet-derived exosomes may act on BAT to regulate thermogenic activity, substrate utilization, and insulin sensitivity (36, 37). Together, these reciprocal signals likely form a BAT-brain-islet axis that coordinates systemic energy balance and glucose metabolism under physiological conditions.

Figure 2.

Bidirectional interactions between BAT and pancreatic β-cells via secretory and neural pathways. BAT may influence pancreatic β-cell function through multiple mechanisms. BAT secretes endocrine factors including FGF21, EPDR1, GDF15, PLTP, and Nrg4, which travel systemically to potentially modulate β-cell functions. Nrg4 also promotes WAT beigeing. BAT-secreted metabolites (eg, MOVA and atRA), along with exosomes and lncRNAs, further contribute to the regulation of β-cell activity. Afferent sensory nerves arising from BAT relay environmental signals (eg, via Opsin 3) to the brain, thereby potentially affecting efferent autonomic nerves innervating the islets, including the SNS and PSNS, to regulate β-cell function. BAT-derived factors such as BMP8b, S100b, and Nrg4 promote local SNS neurite growth. Conversely, pancreatic islet hormones—insulin and amylin from β-cells and glucagon from α-cells—help regulate BAT glucose uptake and thermogenesis. In addition, exosomes released by β-cells may also modulate BAT activity. The illustration was created with BioRender.com.

Abbreviations: atRA, all trans retinoic acid; BAT, brown adipose tissue; BMP8b, bone morphogenetic protein 8b; Clstn3β, calsyntenin 3β; EPDR1, ependymin-related protein 1; FGF21, fibroblast growth factor 21; GDF15, growth differentiation factor 15; lncRNAs, long noncoding RNAs; MOVA, 3-methyl-2-oxovaleric acid; NPM3, nucleophosmin 3; Nrg4, neuregulin 4; PLTP phospholipid-transfer protein; PSNS, parasympathetic nervous system; S100b, S100 calcium-binding protein B; SNS, sympathetic nervous system; WAT, white adipose tissue.

Disruption of BAT-β-cell Communication in Obesity

In obesity, this finely tuned crosstalk becomes dysregulated. Initially, β-cells adapt to nutrient overload by enhancing insulin secretion, but chronic exposure to elevated glucose and fatty acids induces glucolipotoxicity, leading to oxidative stress, inflammation, endoplasmic reticulum stress, mitochondrial dysfunction, and ultimately β-cell apoptosis (38-44). Concurrently, obesity compromises BAT function through “whitening,” which is characterized by increased lipogenesis (45) and inflammation, which diminish BAT's insulin sensitivity and thermogenic capacity (46). These impairments worsen systemic insulin resistance, placing further burden on β-cells. Disrupted BAT-β-cell communication might thus accelerate β-cell failure and the progression of type 2 diabetes (T2D). Understanding how this interorgan crosstalk is altered in obesity and T2D, and whether it can be restored, represents a key direction for future research.

BAT Crosstalk With β-cells via Hormonal and Metabolite Signals

Emerging evidence indicates that BAT can influence pancreatic β-cell function through a range of hormonal and metabolic signals (2, 23). Although direct clinical evidence remains limited, preclinical findings from both in vivo and in vitro studies have shed light on these potential interorgan communication pathways. Where possible, we distinguish human-relevant observations from experimental findings in animals or cellular models, while also noting mechanistically plausible but as-yet unconfirmed hypotheses.

FGF21

FGF21, a well-characterized batokine, has been shown to preserve β-cell viability under glucolipotoxic stress. In diabetic db/db mice where β-cells are largely exhausted, FGF21 enhances insulin secretion (47), reduces β-cell apoptosis, and improves glycemic control (48-50). Mechanistically, its protective effects likely involve activation of ERK1/2 and Akt signaling pathways and modulation of the GH axis through upregulation of PPARγ and cytokine-inducible SH-2 containing (CIS) protein (49, 50). Although FGF21 is found to inhibit β-cell proliferation and expansion in healthy mice (49), it maintains β-cell viability in diabetic models (49), suggesting a context-dependent role.

IL-6

IL-6 is a pleiotropic cytokine produced by several tissues, including skeletal muscle and adipose tissues (51). Clinically, muscle-derived IL-6 released during exercise improves glucose homeostasis by promoting peripheral glucose disposal (52). IL-6 is also classified as a batokine, with expression confirmed in rodent BAT (25) and human differentiated beige adipocytes (53). Transplantation of BAT from IL-6-knockout mice into wild-type recipients failed to confer the same metabolic benefits, such as improved glucose tolerance, observed with wild-type BAT grafts, indicating that BAT-derived IL-6 is crucial for maintaining glucose balance (25). However, its direct effects on β-cell function and insulin secretion are yet to be explored.

EPDR1

EPDR1, a batokine recently identified in human brown adipocytes (34), has been shown to modulate β-cell function (54). Treatment of human pancreatic islets and the β-cell line EndoC-βH1 with recombinant human EPDR1 enhanced glucose-stimulated insulin secretion (GSIS), the gold-standard measure of β-cell function. Conversely, silencing of EPDR1 expression in human islets and INS1 (832/13) cells reduced GSIS and pyruvate-stimulated insulin secretion but not potassium-stimulated secretion (54), indicating nutrient-specific regulation. Mechanistically, EPDR1 is reported to activate the cAMP-PKA signaling cascade, increase ATP/ADP ratio, and promote pyruvate into the TCA cycle (54), a central mitochondrial pathway that generates ATP and intermediates required for amplifying insulin secretion. Of note, EPDR1 is also found in β-cells, whose expression was elevated in individuals with T2D and obesity (54), although the relative contributions of BAT-derived vs islet-derived EPDR1 remain to be clarified.

Nrg4

Nrg4 is another BAT-enriched factor whose expression increases with cold exposure (28). Nrg4-knockout mice developed hyperinsulinemia, impaired glucose tolerance, and reduced insulin sensitivity during high-fat feeding (28). By contrast, adipose-specific Nrg4 overexpression mice showed lower fasting glucose and insulin concentrations (28), highlighting the importance of adipose-derived Nrg4 in lowering circulating insulin demand and preserving β-cell function. Mechanistic studies in MIN6 cells, a mouse β-cell line, show that Nrg4 prevented palmitate-induced apoptosis by activating mTOR-dependent autophagy and sustaining mitochondrial pyruvate flux, thereby conferring a β-cell-sparing effect in vitro (55). Although elevated Nrg4 levels are shown to be beneficial in mouse models, clinical data remain inconsistent. Some studies have reported higher circulating Nrg4 levels in T2D (56-58), while others show reduced circulating (59) and adipose Nrg4 (28), particularly in individuals with gestational diabetes mellitus (60, 61). These discrepancies likely reflect disease stage and interindividual variability. Thus, longitudinal or case-control studies are needed to define Nrg4′s clinical utility as a biomarker and its regulatory role in β-cell physiology, while mechanistic insights will benefit from diabetic db/db and streptozotocin (STZ)-induced mice.

PLTP

PLTP, enriched in BAT and detectable in its secretome (62), has been implicated in systemic glucose homeostasis. In a mouse model, AAV-PLTP was injected directly into BAT, and plasma PLTP activity rose sharply (62). Intravenous AAV-PLTP delivery improved both glucose and insulin tolerance and lowered fasting insulin levels in high-fat diet-fed mice, suggesting a protective effect on systemic glucose homeostasis (62). Supporting its BAT origin, mice lacking BAT showed >50% reduction in plasma PLTP activity (62), further implicating BAT as a major source of circulating PLTP. Proteomic profiling of the conditioned medium from clonally derived human thermogenic adipocytes identified PLTP as 1 of the most abundant secreted proteins, confirming its classification as a BAT-derived factor in humans (62). Clinically, circulating PLTP activity was shown to be positively associated with metabolic parameters such as waist circumference (63), fasting glucose, fasting insulin (64), insulin resistance, and hemoglobin A1c (65). Importantly, PLTP levels were disproportionately elevated in individuals with both diabetes and central obesity (63). A 10-year prospective cohort study further found that higher baseline PLTP levels predicted later development of T2D (64). These findings suggest PLTP may be upregulated in a context-dependent or compensatory manner. However, whether PLTP directly affects β-cells remains untested, requiring islet-specific gain- and loss-of-function studies.

Metabolite Mediators

BAT releases a variety of metabolites during thermogenesis, especially in the cold. Many of these molecules act in autocrine, paracrine, and endocrine fashions to influence systemic metabolism (2, 66). The lipokine 12,13-dihydroxy-9Z-octadecenoic acid exemplifies this role by boosting fatty acid uptake into BAT and lowering circulating triglycerides (67). BAT also secretes retinoic acid. One of its isomers, all-trans-retinoic acid (atRA), is a vitamin A derivative that functions as a hormone-like regulator at nanomolar concentrations (68). Subcutaneous administration of atRA in diabetic rats has been reported to restore islet morphology and increase both islet number and area (69). While BAT-derived atRA likely contributes to these effects, direct evidence for BAT-secreted atRA acting on pancreatic β-cells is still lacking. 4 BAT-enriched metabolites: 3-methyl-2-oxovaleric acid (MOVA), 5OP, β-hydroxyisobutyric acid, and β-hydroxyisovaleric acid have been found to be inversely correlated with body mass index and positively correlated with UCP-1 mRNA in humans (70). Experimentally, while each induced browning of human adipocytes in vitro, 5OP and β-hydroxyisobutyric acid improved both glucose tolerance and insulin sensitivity in mice, but MOVA selectively enhanced insulin sensitivity (70). However, insulin levels were not measured during glucose tolerance tests, leaving the effects on β-cells unclear.

Other Adipokines

BAT also secretes nonexclusive adipokines such as adiponectin and GDF15 (23, 30). Adiponectin has been found to increase insulin secretion in mouse islets, possibly via the adiponectin–HNF4α–PPARα pathway (71). In addition, GDF15, primarily secreted by liver and kidney (72), has been shown to enhance GSIS in human islets by activating the canonical insulin release pathway, highlighting its potential role in β-cell regulation and metabolic homeostasis (73). Although not BAT-specific, GDF15 expression in mouse BAT increased significantly (>10-fold) after acute cold exposure (30), raising the possibility that BAT-derived GDF15 contributes to islet regulation during thermogenic activation. Whether cold-induced GDF15 exerts functional effects on β-cells remains to be investigated.

Despite increasing evidence, definitive proof of BAT-β-cell communication remains incomplete. The functional relevance can be established using islet-targeted gain- and loss-of-function models. Multiomics integration of proteomics, metabolomics, and transcriptomics in BAT-β-cells co-culture systems will help identify novel mediators. Transcriptomic analysis of β-cells exposed to BAT-conditioned medium could reveal functional pathways linking BAT signals to β-cell regulation. These efforts will be critical for establishing the physiological and pathophysiological significance of BAT-β-cell crosstalk.

Potential BAT Crosstalk With β-cells via Exosomes

Exosomes are nanoscopic extracellular vesicles that facilitate intracellular communication by transporting bioactive molecules, including miRNAs, proteins, and DNA. Multiple organs, including skeletal muscle, liver, WAT, and BAT, secrete exosomes that mediate autocrine, paracrine, and endocrine signaling, contributing to systemic metabolic regulation. Clinically, it is established that circulating exosomes released during exercise facilitate muscle-liver crosstalk, promoting glucose metabolism and systemic metabolic adaptations in humans (74).

Established Clinical and Experimental Evidence of Adipose Tissue Exosomes

Adipose tissue-secreted exosomes or adipocyte-derived extracellular vesicles (ADEVs), isolated from lean human adipose tissue, have been shown to protect human islets and improve GSIS in both human and mouse β-cells through mechanisms requiring their uptake and internalization by β-cells (75). In contrast, ADEVs from individuals with obesity carried proinflammatory signals that impaired β-cell function, contributing to glucolipotoxicity and increased β-cell apoptosis (75, 76). Similarly, another mouse study found that treatment with ADEVs isolated from diet-induced obesity (DIO) mice significantly increased GSIS (77). This effect is likely due to the uptake of ADEVs by β-cells, which increases phosphorylation and the abundance of insulinotropic proteins (77), implicating ADEVs as potential mediators linking adipose tissue dysfunction to β-cell failure in T2D. Experimental mouse models support this, as IV injection of ADEVs derived from adipose tissue macrophages of obese mice into normal C57BL/6 mice leads to elevated insulin levels, insulin resistance, and impaired glucose tolerance (78). Mechanistically, ADEVs enriched with miR-155, secreted by adipose tissue macrophages in obese mice, decreased GLUT4 expression and impaired insulin signaling in adipocytes (78). Despite these findings, direct clinical evidence validating such detrimental effects remains limited and warrants further investigation.

Emerging Experimental Observations of BAT-derived Exosomes

Emerging studies have suggested BAT exosomes (BAT-exos) as novel regulators of metabolic health, insulin sensitivity, and thermogenesis (79). Typically isolated from conditioned media of minced BAT, BAT-exos administered intravenously have been shown to lower basal blood glucose levels and improve glucose tolerance and insulin sensitivity in various mouse models of metabolic dysfunction, including aging, polycystic ovary syndrome, and DIO (80-82). Fluorescent tracing using PKH26-labeled BAT-exos has revealed their systemic distribution, with preferential uptake in the liver, ovaries, uterus, lungs, and spleen (80). These findings suggest that BAT-exos may exert metabolic effects through actions on peripheral tissues, though their impact on the pancreas remains unclear, as it was not the focus of these studies. Functionally, BAT-exos have demonstrated therapeutic potential comparable to BAT transplantation. For example, in the polycystic ovary syndrome mouse model, BAT-exos improved metabolic and reproductive parameters via the STAT3-GPX4 signaling pathway (81). Proteomic profiling of BAT-exos has further revealed an enrichment of mitochondrial proteins and metabolic-pathway components compared with serum-derived exosomes, correlating with enhanced cellular oxygen consumption, increased energy expenditure, and lowered baseline glucose levels in obese mice (82). Despite these systemic metabolic benefits of BAT-exos, their direct impact on pancreatic islets or insulin secretion remains experimentally unexplored.

Mechanistic Insights and Candidate Effectors in BAT-exos

Experimental studies further dissecting BAT-exos contents have identified bioactive components that may mediate their beneficial effects. Long non-coding RNAs (lncRNAs), such as AK029592, humanlincRNA1030, and ENSMUST00000152284, are enriched in BAT-exos compared to WAT-exos (83). Correlation analysis between these lncRNAs and mRNAs demonstrated that all 3 lncRNAs were associated with commonly enriched pathways, including the insulin signaling, insulin resistance, and T2D pathways (83). Loss-of-function studies revealed that knockdown of AK029592 in BAT impaired BAT thermogenic function and glucose tolerance in mice, suggesting a functional role for BAT-exosomal lncRNAs in metabolic regulation (84). In addition, proteins like nucleophosmin 3, identified as highly enriched in BAT-exos, have been shown to regulate glucose homeostasis indirectly; siRNA-mediated depletion of nucleophosmin 3 in BAT-exos leads to higher fasting glucose levels and impaired glucose tolerance and insulin sensitivity in DIO mice receiving these exosomes intravenously (85, 86). However, direct measures of insulin secretion or β-cell health were not conducted, highlighting a key knowledge gap concerning the direct impact on β-cell function.

Potential Roles of BAT-exosomal miRNAs

BAT-exosomal miRNAs also appear to modulate systemic metabolism. Recent sequencing studies in mice identified miRNAs specifically enriched in BAT-exos, many of which overlap with exercise-induced plasma exosomal miRNAs (87). In adipose tissue-specific dicer knockout mice (ADicerKO), which lack miRNA processing, circulating exosomal miRNAs were reduced by 88% (88), indicating that adipose miRNAs are major sources of serum exosomal miRNAs. Moreover, BAT transplantation in ADicerKO mice restored over 50% of circulating exosomal miRNAs, reduced circulating insulin levels by approximately 20%, and markedly improved glucose tolerance (88), suggesting that BAT-exos miRNAs may influence β-cell function and systemic glucose homeostasis. One example is miR-26a (89), which is highly expressed in BAT and correlates negatively with body mass index, Homeostatic Model Assessment for Insulin Resistance, fasting glucose, and insulin levels in humans (89). In rodents, miR-26a was expressed 2- to 4-fold higher in BAT than in inguinal WAT (90). Its expression was reduced in both high-fat diet-fed mice and genetically obese ob/ob mice. Transgenic overexpression of miR-26a in pancreatic islets (3-fold increase) lowered fasting glucose levels, improved glucose tolerance, and reduced insulin secretion initially at 30 minutes during a glucose tolerance test. However, after prolonged 12-week high-fat feeding, these mice eventually showed diminished GSIS and reduced β-cell mass (89), indicating a direct and time-dependent role for miR-26a in β-cell regulation. Importantly, BAT transplantation in ADicerKO mice elevated circulating miR-26a by ∼318-fold (88), supporting BAT as a key source of this regulatory miRNA.

While current evidence strongly supports a systemic metabolic role for BAT-exos, their direct role in regulating β-cell function remains hypothetical. Future studies should investigate insulin secretion directly in isolated islets or β-cells following BAT-exos exposure and identify key bioactive exosomal components. Co-culture experiments, targeted exosome transfer, and multiomics profiling will be critical to uncovering specific mechanisms of BAT-β-cell crosstalk and evaluating translational potential.

BAT Crosstalk With β-cells via the Nervous System

Neural Communication Between BAT and the Central Nervous System

BAT is extensively innervated by the sympathetic nervous system, though parasympathetic innervation is limited to 2 minor BAT depots—pericardial and mediastinal—but absent in major depots like interscapular BAT (91, 92). Sympathetic activation releases NE, which binds to β3-AR on brown adipocytes to trigger UCP-1-mediated thermogenesis (93). Multiple brain nuclei, including the preoptic area (94), dorsomedial hypothalamus (95), ventromedial hypothalamus (96), and arcuate nucleus (ARC) (97), regulate BAT activity and systemic energy balance. ARC neuropeptide Y neurons, which detect peripheral metabolic signals through the semipermeable blood-brain barrier (98), inhibit BAT thermogenesis by suppressing tyrosine hydroxylase (TH)-expressing neurons in the paraventricular nucleus (97).

Experimental Evidence of Sensory Innervation in BAT

Unlike the well-studied sympathetic innervation, sensory innervation of adipose tissue, particularly BAT, remains less explored. Recent experimental studies have revealed robust sensory innervation from dorsal root ganglia neurons in adipose tissue, influencing local thermogenic and lipogenic responses (99, 100). Inguinal WAT-specific ablation of sensory innervation selectively increased beige adipocyte differentiation and thermogenic gene expression without affecting systemic sympathetic tone or temperature perception (99), indicating a localized sensory regulatory mechanism. Importantly, this thermogenic effect was abolished by concurrent sympathetic denervation, suggesting that sensory neurons act upstream to modulate local sympathetic function (99). These foundational studies provide a proof-of-principle for investigating the potential role of dorsal root ganglia sensory neurons in BAT regulation.

Although somatosensory innervation of BAT is relatively less extensive compared to WAT (99), it is nonetheless present. Sensory neuropeptides, such as substance P and calcitonin gene-related peptide, have been detected in BAT of rats (101, 102). Anterograde transneuronal viral tracing studies in rodents have identified sensory projections from BAT to brain regions such as the brainstem, forebrain, and hypothalamus (103, 104), areas critical for sympathetic regulation. Functionally, capsaicin-mediated sensory nerve ablation in BAT reduced calcitonin gene-related peptide by 55% to 72% and decreased BAT and core body temperature during cold exposure, demonstrating a direct role for sensory nerves in thermoregulation (103). In addition, BAT-expressed transient receptor potential vanilloid 1 channels, which integrated thermosensory and nociceptive signals, suppressed thermogenesis by dampening sympathetic nerve activity (105, 106). Other unique BAT-mediated pathways, such as Opsin 3-mediated histidine release, may activate hypothalamic histaminergic neurons, further linking BAT to central sympathetic regulation (107). Together, these observations support the hypothesis that BAT can sense environmental cues, such as temperature, and relay sensory feedback to the central nervous system (CNS), thereby influencing lipid storage or mobilization and systemic energy homeostasis. This suggests the existence of bidirectional BAT-CNS communication (103). However, direct evidence of BAT sensory communication affecting pancreatic islets remains limited, necessitating further investigation. Given the absence of classical parasympathetic innervation in adipose tissue, sensory innervation may provide a local modulatory mechanism similar to the parasympathetic counterbalance observed in other visceral organs.

Batokine-mediated Modulation of Neural Circuits

BAT-derived endocrine signals (batokines) may modulate neural circuits controlling metabolic homeostasis. Bone morphogenetic protein 8b (BMP8b) and Nrg4 have been shown to stimulate sympathetic neurite growth, promoting sympathetic innervation (31, 108). Specifically, adipose-specific BMP8b-transgenic mice display increased TH-positive fibers, whereas BMP8b-knockout mice exhibit reduced innervation and neurovascular remodelling in adipose tissues (31). These findings suggest that BMP8b released from BAT may also enhance sympathetic innervation of pancreatic β-cells, a hypothesis that could be tested by evaluating TH staining in the pancreas of BAT-specific BMP8b-overexpressing mice. Another BAT-enriched factor, calsyntenin 3β, is highly expressed in BAT and beige adipocytes and facilitates the secretion of S100b, a neurotrophic factor capable of enhancing sympathetic neurite growth and activity (109). While speculative, these findings suggest that batokines may modulate systemic metabolism by shaping neural inputs to tissues such as the pancreas.

Influence of BAT on Islets via Neural Pathways

Pancreatic islets are densely innervated by autonomic nerves, with parasympathetic stimulation enhancing insulin secretion and sympathetic input typically suppressing it (110). Recent human imaging studies, employing combined fluorodeoxyglucose-positron emission tomography and blood oxygenation level dependent functional magnetic resonance imaging techniques, provided clinical evidence linking BAT activation to CNS modulation, particularly in higher-order brain regions associated with food intake control (111). Specifically, infusion of the gut hormone secretin reduced glucose uptake in reward-related regions like the caudate nucleus while increasing glucose uptake in BAT, thereby enhancing satiety-related circuits. These findings suggest that secretin-induced BAT activation sends feedback signals to the brain, influencing feeding behavior through CNS pathways. This study provides compelling evidence that BAT can modulate more than thermogenic and sympathetic outputs.

Supporting these clinical observations, rodent studies demonstrated that the secretin receptor (SCTR) is highly enriched in BAT, where secretin binding activated the SCTR-PKA-ATGL/HSL pathway, leading to UCP-1-mediated thermogenesis. Crucially, UCP-1-knockout mice, which lack functional BAT, changed their feeding behavior, confirming that BAT thermogenesis is an essential relay for the secretin-driven BAT-to-brain satiety signal (112). It is plausible that BAT-generated heat may serve as a new and alternative afferent signal to the brain that modulates feeding and possibly insulin secretion. Although intriguing, direct connections between BAT-induced CNS activity and pancreatic β-cell function remain speculative, necessitating further studies explicitly testing insulin secretion responses.

Altered Islet Innervation in Diabetes

Findings on islet innervation in diabetes are inconsistent. Some studies using neuronal marker neurofilament 200 reported elevated nerve density in human T2D pancreas, nonobese diabetic mice and STZ-induced diabetic mice (113), including increased sympathetic TH staining in β-cells of diabetic db/db mice (114). Others reported reduced central islet innervation concurrent with progressive insulitis in nonobese diabetic mice using AchE and PGP9.5 staining (115). These discrepancies likely result from marker-specific limitations and methodological differences in these studies, highlighting the need for standardized methodologies in characterizing pancreatic islet innervation.

Recent data reveal increased sympathetic input to pancreatic islets from both diabetic mouse models and human patients (113), which may reflect a compensatory neural attempt to maintain tighter glucose regulation. BAT-derived neurotrophic factors, such as S100b, could contribute to this remodeling. BAT-derived sensory signals may also reach central brain areas that could alter autonomic outputs to pancreatic islets. BAT-derived heat may act as an afferent signal that the brain integrates and, in turn, uses to modulate insulin secretion. Thus, a proposed BAT-brain-islet neural axis could serve as a comprehensive integrative pathway coordinating metabolic balance, promoting BAT thermogenesis and insulin sensitivity to minimize insulin demand while concurrently fine-tuning insulin secretion from pancreatic β-cells. This concept, while compelling, remains hypothetical and requires rigorous testing.

Bidirectional Interaction Between Pancreatic Islets and BAT

While the primary focus of this review is on BAT's influence on pancreatic β-cells, growing evidence supports a reciprocal interaction in which pancreatic islets can also impact BAT function via hormonal, metabolite, exosomal, and neuronal pathways.

Established Roles of Pancreatic Hormones on BAT Function

Insulin

Under cold exposure or β3-AR stimulation, circulating insulin levels increase significantly, largely due to enhanced lipolysis in WAT. This insulin surge plays a crucial role in supporting BAT activity. Clinical fluorodeoxyglucose-positron emission tomography/computed tomography studies show that insulin significantly increases glucose uptake in human BAT, attributed to high GLUT4 expression levels relative to WAT (36). In animal models, insulin promotes glucose uptake, stimulates lipoprotein lipase activity for fatty acid uptake, and supports anabolic processes required for sustained thermogenesis. These mechanisms ensure sufficient substrate availability for thermogenesis. However, insulin does not directly drive thermogenic gene expression (eg, UCP-1) (116), which is primarily regulated by β3-adrenergic signaling. In support with this, another study using fat-specific insulin receptor and IGF-1 receptor knockout mice has found that while thermogenic genes (eg, UCP-1, PGC1α, PRDM16) in BAT are significantly downregulated under basal conditions, their expression can still be robustly induced by cold exposure (117), suggesting that insulin's role in BAT is to supply substrates, glucose, and FFAs, rather than directly initiate thermogenic programs.

Cold-induced BAT thermogenesis also requires integration with lipolysis in WAT for FFAs, establishing a WAT-β-cell-BAT axis. A central mediator of the crosstalk between lipolysis and insulin is fatty acid-binding protein 4 (FABP4), predominantly secreted from adipocytes during sympathetic stimulation or fasting (118). Adipocyte-released FABP4 enters the circulation, where it forms a complex with extracellular enzymes (eg, adenosine kinase and nucleoside diphosphate kinase) to regulate local ATP/ADP ratios and potently stimulates β-cell insulin secretion (119). Therefore, FABP4 serves as an endocrine lipid signal conveying the state of ongoing lipolysis to pancreatic β-cells, augmenting insulin release as part of a feedback loop (119). Interestingly, a recent study found that β-adrenergic stimulation alone in vivo can provoke robust FABP4 release from adipocytes even in ATGL-deficient adipose tissue lacking active lipolysis (120), suggesting a proactive metabolic coordination whereby insulin secretion is anticipatorily enhanced before substantial lipolysis occurs. This feed-forward aspect of FABP4 release further tightens the coordination between neural activation of fat and hormonal facilitation of nutrient flux.

Glucagon

Glucagon receptor activation in BAT of mice promotes lipolysis and thermogenesis via UCP-1-independent but FGF21-dependent mechanisms (37). However, BAT-specific glucagon signaling appears dispensable for overall regulation of energy expenditure or glucose regulation (37), indicating that its role in BAT might be context-dependent or compensatory rather than essential.

Emerging Evidence for β-Cell-Derived Exosomal Regulation of BAT

Exosomes secreted from pancreatic β-cells may represent a novel pathway for interorgan communication. In diabetic mouse models, β-cell-derived exosomes (β-cell-exos) isolated from pancreatic β-cells (MIN6) preserved islet architecture, enhanced islet survival, and improved glucose tolerance (121). Under conditions of elevated FFAs, β-cell-exos enriched in bioactive molecules such as miR-29 have been shown to exacerbate hepatic insulin resistance and glucose intolerance (122), suggesting their potential to influence distant tissues and systemic metabolism beyond the pancreas. Although their direct effects on BAT have not yet been demonstrated, it is plausible that these β-cell-exos, particularly those carrying miR-29, may similarly impair BAT insulin sensitivity and thermogenic function, contributing to broader metabolic dysfunction. This possibility warrants further investigation.

Pancreatic Hormone Regulation of BAT via the CNS

Neuronal pathways provide an established route through which pancreatic hormones modulate BAT activity. In STZ-induced diabetic rats, central activation of BAT thermogenesis is impaired and cannot be restored by peripheral insulin administration, suggesting central insulin resistance as a key factor underlying cold intolerance in diabetes (123). Central administration of insulin produces variable effects on BAT thermogenesis, either stimulatory or inhibitory, depending on the brain region targeted, insulin dose, and glucose status (124). Insulin signaling within hypothalamic regions, particularly in the ARC, suppresses neuropeptide Y neurons, thus relieving their inhibition of sympathetic neurons that activate BAT. Similarly, centrally administered glucagon enhances sympathetic outflow and stimulates BAT thermogenic activity (125). In addition, amylin, co-secreted with insulin, acts centrally via receptor activity-modifying protein 1 to further modulate neuronal control of BAT thermogenesis and energy expenditure (126). These findings highlight the complexity of pancreatic hormone signaling within the CNS and its downstream effects on BAT function.

Summary and Future Directions

Emerging evidence supports the existence of a complex communication network between BAT and pancreatic β-cells, involving hormonal, metabolic, neuronal, and exosomal pathways. This interorgan crosstalk may coordinate insulin secretion with energy expenditure under physiological conditions. BAT-derived factors, including batokines, metabolites, and exosomal bioactive molecules, potentially influence β-cell survival and insulin secretion. Conversely, pancreatic β-cells may also reciprocally modulate BAT function through endocrine signals, exosomes, and central neural circuits. Together, these observations suggest the presence of a bidirectional BAT-brain-islet axis involved in maintaining systemic energy and glucose balance.

Despite growing interest, a mechanistic understanding of this crosstalk remains limited. Much of the current evidence is indirect or correlative, with relatively few studies directly investigating the causal and bidirectional nature of this communication. Key knowledge gaps include the molecular contents and mechanisms by which BAT-derived exosomes affect β-cell regulation and the specific neural circuits mediating BAT-brain-islet interactions. These limitations highlight the need for refined experimental approaches. Future research should integrate physiologically relevant paradigms, such as cold exposure, fasting, and dietary stress, with advanced experimental models, including co-culture systems, tissue-specific knockouts, and neuronal circuit-mapping techniques (eg, opto- or chemogenetics). Multiomics platforms will be important in identifying key molecular pathways underpinning BAT-β-cell communication. Clinical investigations assessing the translational relevance of BAT-derived molecules, particularly in patients with obesity and diabetes, will be equally critical. Elucidating the mechanisms of BAT-β-cell crosstalk may open new avenues for preserving β-cell health and developing more precise treatments for obesity and T2D.

Funding

This work was supported by the National Health and Medical Research Council, Australia (no. 1144286, no. 1162276) to Y.C.S. and a UNSW University International Postgraduate Award scholarship to C.X.Y.

Disclosures

The authors have nothing to disclose.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.