Abstract
Thermogenic adipose tissue, consisting of brown and beige fat, regulates nutrient utilization and energy metabolism. Human brown fat is relatively scarce and decreases with obesity and aging. Hence, inducing thermogenic differentiation of white fat offers an attractive way to enhance whole-body metabolic capacity. Here, we show the role of endothelin 3 (EDN3) and endothelin receptor type B (EDNRB) in promoting the browning of white adipose tissue (WAT). EDNRB overexpression stimulates thermogenic differentiation of human white preadipocytes through cAMP-EPAC1-ERK activation. In mice, cold induces the expression of EDN3 and EDNRB in WAT. Deletion of EDNRB in adipose progenitor cells impairs cold-induced beige adipocyte formation in WAT, leading to excessive weight gain, glucose intolerance, and insulin resistance upon high-fat feeding. Injection of EDN3 into WAT promotes browning and improved whole-body glucose metabolism. The findings shed light on the mechanism of WAT browning and offer potential therapeutics for obesity and metabolic disorders.
Similar content being viewed by others
Introduction
Obesity and related complications such as diabetes, cardiovascular diseases, and nonalcohol fatty liver diseases are rapidly increasing worldwide and causing healthcare burdens. Developing preventive and therapeutic strategies for obesity and its complications is of great importance to communities1,2. Mammals have two main types of adipose tissue, brown adipose tissue (BAT) and white adipose tissue (WAT). WAT contains unilocular lipid droplets and is responsible for storing excess energy, whereas BAT possesses multilocular lipid droplets and is specialized for consuming energy3. Increased amount or activity of BAT promotes energy expenditure and inversely correlates with body mass index, the circulating levels of glucose and triglycerides, as well as the prevalence of diabetes and cardiovascular dysfunctions in humans, pointing to the therapeutic potential of BAT in combating metabolic diseases4,5,6,7,8,9.
BAT expresses uncoupling protein 1 (UCP1), which dissipates energy by uncoupling the proton motive force from ATP production to generate heat in response to cold exposure. Although UCP1 expression is restricted to BAT under basal conditions, prolonged cold exposure or β3-adrenergic stimulation can increase UCP1-mediated thermogenic capacity in BAT, and activate the recruitment or differentiation of brown-like beige/brite, adipocytes in WAT that express UCP1 or other futile thermogenic machinaries10,11,12 to produce heat, in a process called browning or thermogenic differentiation. In adult humans, WAT is distributed throughout the body, while BAT is present in small quantities and restricted regions, such as cervical, supraclavicular, and paravertebral regions13. Considering abundance and location, WAT is more easily reachable and manipulatable. Thus, thermogenic differentiation of WAT may provide an opportunity for preventing or treating obesity and obesity-related complications. However, the drivers and detailed mechanisms underlying the thermogenic differentiation of WAT call for further exploration.
G-protein coupled receptors (GPCRs) regulates a broad spectrum of biological processes, and serve as the most druggable targets in biology14. GPCRs have been strongly linked to thermogenic adipocyte differentiation and function15. These include the GPCRs involved in canonical thermogenic activation, such as the β-adrenergic receptors, and several recently identified GPCRs with known or unknown ligands, including GPR316, GPR12017, opsin318 and others. Using clonal human brown and white preadipocytes, we previously discover that the expression levels of G protein-coupled endothelin receptor type B (EDNRB) in human adipose precursor cells are positively correlated to thermogenic potential in the differentiated adipocytes19, indicating the potential role of this receptor and its ligand in promoting thermogenic adipocyte differentiation.
Initially identified as regulators of vasoconstriction/vasodilation, endothelins are now known to modulate diverse physiological functions20,21. In mammals, there are two types of endothelin receptors, type A (EDNRA) and type B (EDNRB), which are targeted by three endothelins, known as endothelin 1-3 (EDN1-3)22,23. EDN1 and EDN2 activate both EDNRA and EDNRB with equal affinity, whereas EDN3 has a lower affinity for EDNRA, and mainly activates EDNRB22,23. Although both EDNRA and EDNRB are linked with Gq signaling, which stimulates intracellular Ca2+ mobilization, they exert opposite biological actions. The EDNRA signaling regulates vasoconstriction, cellular proliferation, growth, and inflammation, resulting in adverse vascular outcomes. In contrast, EDNRB mediates vasodilation and inhibition of growth and inflammation24,25. The opposing roles of EDNRA and EDNRB and their differential affinities to different endothelin ligands highlight the complexity of the EDN-EDNRs system and reiterate their importance in normal and pathophysiological conditions.
Regarding downstream signaling pathways, while both EDNRA and EDNRB are traditionally known to activate Gq signaling, EDNRB has been linked to other Gs or Gi proteins26,27 and other signaling systems, such as the mitogen-activated protein kinases/extracellular signal-regulated kinase (MAPK/ERK) signaling pathway28,29. Klepac et al. showed inhibition of autocrine EDN1-EDNRA-Gq signaling in murine or human adipose precursor cells enhances thermogenic adipocyte differentiation30. Ablation of EDNRB in human brown preadipocytes causes marked defects in the expression of thermogenic genes, such as UCP1 and DIO2, without alteration in lipid accumulation19. Given the opposing effects of EDNRA and EDNRB, it is conceivable that activation of two receptors triggers distinct signaling pathways to regulate thermogenic adipocyte differentiation. However, it remains unclear whether EDNRB activation and its ligand stimulation are sufficient to determine thermogenic differentiation in human white preadipocytes.
In the present study, we demonstrate that EDNRB activation promotes browning of human white preadipocytes via cAMP-EPAC1-ERK activation. In mice, loss of EDNRB in preadipocytes impaired cold-induced beige adipocyte formation, leading to excessive weight gain, glucose intolerance, and insulin resistance upon high-fat feeding. Conversely, EDN3 administration increased thermogenic adipocyte differentiation and improved glucose tolerance in obesity. These findings highlight that the EDN3-EDNRB signaling plays a critical role in thermogenic adipocyte differentiation and metabolic regulation.
Results
Activation of EDNRB promotes thermogenic differentiation of white preadipocytes in vitro
Previously, our laboratory demonstrated that a set of gene signatures positively correlates with thermogenic differentiation potential in clonal human preadipocytes19. Among this, the levels of EDNRB and EDN3, the principal ligand of EDNRB, were positively correlated with UCP1 expression after adipogenic induction, while EDN1 and EDN2 were not significantly correlated (Supplementary Fig. 1a). The fact that the ligands and receptors were expressed in human preadipocytes suggest that these EDNs may regulate preadipocyte function in an autocrine/paracrine manner. To interrogate the role of EDNs and their receptors in human adipose tissue at single-cell resolution31, we analyzed two publicly available single-cell RNA sequencing (scRNA-seq) datasets from human subcutaneous WAT (scWAT)32,33, and found that individuals with obesity (BMI ≥ 30) expressed significantly lower levels of EDNRB in their adipose progenitors compared to lean and overweight individuals whose BMI is less than 30. In contrast, the expression pattern of EDNRA in the same cohort was completely opposite, with higher levels of expression in scWAT adipose progenitors in individuals with obesity (Supplementary Fig. 1b, c).
To further elaborate the role of EDNRB in thermogenic adipocyte differentiation, we investigated whether activation of EDNRB in human white preadipocytes could induce a browning process after differentiation. To this end, we adapted the CRISPR activator (CRISPR-SAM) system34 to increase endogenous EDNRB expression in human white preadipocytes. We designed 3 different gRNAs (#1-3) targeting around 50–150 bp upstream of the human EDNRB gene and delivered them into human white preadipocytes carrying the CRISPR-SAM machinery35 (dCas9-VP64 and MS2-p65-HSF1) (Fig. 1a). Delivering CRISPR-SAM system combined with gRNA #3 into human white preadipocytes resulted in a significant 2-fold increase in mRNA and protein levels of EDNRB compared with delivering empty vector of gRNA or other sequences of gRNAs (Supplementary Fig. 2a, b), but it did not alter the expression of EDNRA, another isoform of endothelin receptor (Supplementary Fig. 2c).
a Schematic of EDNRB gene activation using CRISPR-SAM system. b–g Human white preadipocytes were transfected with the CRISPR-SAM system and either empty vector (EV) or sgRNA#3 targeting EDNRB (EDNRB OE) and were differentiated to mature adipocytes. b Relative EDNRB and EDNRA mRNA levels in human white adipocytes. n = 3 biological replicates/group. c Relative EDNRB protein levels in human white adipocytes. n = 4 biological replicates/group. d Relative mRNA expressions in adipogenic and thermogenic genes in human white adipocytes. n = 3 biological replicates/group. e Fatty acid-dependent oxygen consumption rate (OCR) in human white adipocytes. Quantification of maximal OCR in the right panel. EV: n = 3, EDNRB OE: n = 4, EV-ETO: n = 4, EDNRB OE-ETO: n = 3 biological replicates. Pal, palmitate-BSA. OA, oligomycin A. AA, antimycin A. ETO, etomoxir. f Forskolin (FSK)-stimulated OCR in human white adipocytes. n = 5 biological replicates/group. g Thermogenesis capacity monitored by ERthermAC dye in human white adipocytes. Area above the curve (AAC) of thermogenesis in the right panel. EV: n = 21 cells, EDNRB OE: n = 22 cells. Data are presented as mean ± SEM. P values were determined using unpaired two-tailed t tests: (b–g). Source data are provided as a Source Data file. Panel (a) was created with BioRender.com released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.
Next, we tested whether EDNRB overexpression can promote thermogenic differentiation potential in white preadipocytes (Supplementary Fig. 2d). Human white preadipocytes containing CRISPR-SAM and gRNA #3 (EDNRB OE) maintained high levels of EDNRB mRNA and protein expression after being induced to adipogenic differentiation compared to the preadipocytes delivered with an empty vector of gRNA (EV) (Fig. 1b,c). EDNRB overexpression did not change lipid droplet formation after or during adipogenic differentiation (Supplementary Fig. 2e, f). In addition, the expression of general adipogenic differentiation markers, such as PPARγ and AP2, was not altered (Fig. 1d). Intriguingly, overexpression of EDNRB in human white preadipocytes significantly increased the expression of UCP1 and other thermogenic genes, such as iodothyronine deiodinase 2 (DIO2), cell death-inducing DNA fragmentation factor-like effector A (CIDEA), and carnitine palmitoyltransferase 1B (CPT1B), after differentiation (Fig. 1d). These results mirrored our previous findings showing knocking out EDNRB in human brown preadipocytes impaired expression of thermogenic genes without altering adipogenic differentiation19.
In addition to the increased UCP1 levels, white adipocytes differentiated from EDNRB OE preadipocytes acquired functional phenotypes of beige adipocytes. The adipocytes displayed elevated levels of carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP)-induced maximal oxygen consumption rate (OCR) and glucose-induced extracellular acidification rate (ECAR), which indicated high mitochondrial activity and glucose utilization (Supplementary Fig. 2g, h). To examine fatty acid utilization in adipocytes, we provided palmitate-BSA as the fuel and monitored OCR with and without the addition of etomoxir (ETO), which is an inhibitor of fatty acid oxidation. The differences in OCR between the adipocytes with and without ETO represent fatty acid-dependent OCR, which was significantly elevated in EDNRB OE adipocytes (Fig. 1e). Moreover, upon forskolin (FSK) treatment, EDNRB OE human adipocytes showed a higher OCR (Fig. 1f) and increased heat production as measured by the ERthermAC dye, in which reduced fluorescence indicated increased cellular temperature36 (Fig. 1g and Supplementary Fig. 2i). These results demonstrated that EDNRB overexpression in human white preadipocytes is sufficient to increase mitochondrial activity and promote thermogenic function in differentiated adipocytes.
EDN3-EDNRB signaling promotes thermogenic differentiation of white preadipocytes via cAMP-EPAC1-ERK activation
To investigate whether EDN3-EDNRB signaling enhances thermogenic function in the differentiation of human white preadipocytes, we pretreated EDNRB OE and control EV preadipocytes with either vehicle or EDN3 for 3 days, followed by a 12-day adipogenic differentiation process (Fig. 2a). Pre-treatment of 100 nM EDN3 in EV white preadipocytes was able to slightly induce UCP1 induction after differentiation. Notably, EDN3 pretreatment further activated UCP1 expression in EDNRB OE white preadipocytes after differentiation (Fig. 2b, c). Levels of adipogenic transcription factor PPARγ and fatty acid binding protein AP2 were not altered in EV and EDNRB OE preadipocytes upon EDN3 pretreatment (Fig. 2c and Supplementary Fig. 3a, b).
a–d Human white preadipocytes were pretreated with vehicle (Veh) or 100 nM EDN3 and with or without inhibitors for 3 days, followed by a 12-day adipogenesis. a Schematic of pretreatment experiments. UCP1 mRNA (b) and protein (c) levels in mature adipocytes (EV and EV and EDNRB OE). d UCP1 mRNA levels in mature adipocytes (EV and EDNRB KO). e Cytosolic calcium levels in human white preadipocytes after 100 nM EDN3 treatment. f, g Intracellular cAMP levels (f) and oxygen consumption rate (OCR; g) in human white preadipocytes after 6 hours of Veh or EDN3 treatment. Maximal OCR was quantified in the right panel. h–j EV and EDNRB OE preadipocytes were pretreated with Veh or EDN3 combined with or without melittin (ME) or YM254890 (YM) for 3 days, followed by a 12-day adipogenesis. UCP1 mRNA (h) and protein (i) levels in mature adipocytes pretreated with ME. j UCP1 mRNA levels in mature adipocytes pretreated with YM. k The protein levels in preadipocytes after 6 hr of Veh or EDN3 treatment. The quantification of protein bands was in the right panels. l The EPAC1 protein in EDNRB OE preadipocytes after 48 hr of EPAC1 siRNA treatment. m EDNRB OE preadipocytes with or without EPAC1 knockdown were pretreated with Veh or EDN3 for 3 days, followed by a 12-day adipogenesis. Protein levels in mature adipocytes. n EV or EDNRB OE preadipocytes were pretreated with Veh, 1 μM or 10 μM of PD98059 (PD-1 or PD-10) for 3 days and were differentiated to mature adipocytes. UCP1 and DIO2 mRNA levels in human white adipocytes. n = 3 biological replicates/group in (b–e), (i), (k–n); n = 4 biological replicates/group in (f–h), (j). Data are presented as mean ± SEM. P values were determined using unpaired two-tailed t tests: (l); one-way ANOVA with Tukey’s multiple-comparison test: (b–d), (f–k), (m), (n); two-way ANOVA with Bonferroni’s multiple-comparison test: (e). Source data are provided as a Source Data file. Panel (a) was created with BioRender.com released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.
To ascertain whether the induction of UCP1 by EDN3 is dependent on EDNRB, we generated EDNRB knockout (EDNRB KO) in human white preadipocytes and observed that EDN3 pretreatment failed to elicit UCP1 expression following adipogenic differentiation in these cells (Fig. 2d). We then extended EDN3 treatment across the entire adipogenic differentiation span of 15 days. This extended treatment led to a modest increase in UCP1 protein levels, yet the effect was less pronounced compared to what was achieved with the 3-day pretreatment regimen (Supplementary Fig. 3c,d). These findings suggest that the EDN3/EDNRB axis may predominantly function to prime adipose precursors for thermogenic differentiation.
EDNRB is a G protein-coupled receptor (GPCR) that typically signals through Gs or Gq subunits37,38, initiating intracellular calcium and cAMP signaling pathways. These pathways regulate various cellular processes such as cell proliferation, differentiation, and motility in neural progenitor cells28 and melanoma cells29, often via calcium-mediated MAPK/ERK pathway activation. Upon treating both EDNRB OE and control human white preadipocytes with EDN3, we observed elevated levels of cytosolic calcium and cAMP (Fig. 2e, f). Notably, the EDNRB OE preadipocytes exhibited more pronounced responses to EDN3 compared to control cells. This acute treatment with EDN3 resulted in the upregulation of mitochondrial genes, including PGC1β, TFAM, and COX7a1, consequently enhancing mitochondrial DNA (mtDNA) content and mitochondrial respiration (Supplementary Fig. 3e, f and Fig. 2g).
To elucidate the signaling mechanisms underlying EDN3/EDNRB-mediated thermogenic differentiation, cells treated with EDN3 were simultaneously co-treated with G-protein inhibitors, Melittin (ME) or YM254890 (YM), to block the Gs or Gq pathways, respectively. Interestingly, inhibition of the Gs pathway with ME effectively blocked EDN3-mediated UCP1 induction without altering adipocyte differentiation (Fig. 2h, i and Supplementary Fig. 3g), while inhibition of the Gq pathway by YM did not yield a significant impact (Fig. 2j). The cAMP-PKA pathway is a canonical signaling pathway to trigger UCP1 expression during thermogenic differentiation39,40. However, EDN3/EDNRB signaling did not activate PKA (Supplementary Fig. 3h). In addition to PKA, cAMP can act through an alternative pathway involving exchange protein activated by cAMP (EPAC) protein, leading to ERK activation41, which has recently been shown to play a role in the proliferation and differentiation of adipose precursors for both brown and beige adipocytes42. Notably, EDNRB OE white preadipocytes exhibited elevated levels of ERK and CREB phosphorylation compared to empty vector (EV) cells even without the addition of exogenous ligands; and the addition of EDN3 further augmented this activation (Fig. 2k). As a result of ERK activation, EDNRB OE white preadipocytes demonstrated increased cell proliferation compared to EV cells, with further enhancement observed upon EDN3 treatment (Supplementary Fig. 3i).
To explore the involvement of EPAC in EDN3/EDNRB-mediated thermogenic differentiation, we knocked down EPAC1 in EDNRB OE cells using siRNA (Fig. 2l), and observed that EDN3 failed to induce ERK and CREB phosphorylation as well as UCP1 expression in those cells (Fig. 2m), revealing the essential role of EPAC in this process. Furthermore, to validate the contribution of the ERK signaling in EDN3/EDNRB-induced thermogenic differentiation, we pre-treated EDNRB OE preadipocytes with a MEK/ERK inhibitor, PD98059 (PD), followed by adipogenic differentiation. While blocking ERK activation with 1 or 10 μM PD treatment did not affect adipocyte differentiation (Supplementary Fig. 3j), it did impede the induction of UCP1 and DIO2 expression in EDNRB OE white preadipocytes post-differentiation (Fig. 2n). Collectively, these findings highlight that EDN3/EDNRB primes human white preadipocytes through the activation of the cAMP-EPAC1-ERK pathway, thereby conferring upon them the potential for thermogenic differentiation.
Treating the EDNRB OE and control human white preadipocytes with EDN1 led to increased cytosolic calcium levels as similar to EDN3 (Supplementary Fig. 4a). Notably, the EDNRB OE preadipocytes displayed greater responses to EDN3 than those to EDN1, reinforcing the notion that EDN3 is the specific ligand for EDNRB. In contrast, 100 nM EDN1 pretreatment impaired thermogenic adipose differentiation in both EV and EDNRB OE human white preadipocytes as shown by reduced UCP1 and PPARγ expression (Supplementary Fig. 4b). In addition, acute EDN1 treatment reduced the level of ERK1/2 phosphorylation, contrasting with the effects observed with EDN3 treatment (Supplementary Fig. 4c). These results are consistent with a previous study by Klepac et al. showing the EDN1-EDNRA signaling inhibited differentiation of mouse brown adipocytes30.
To examine the role of EDN3-EDNRB signaling in human brown adipocyte differentiation, we pretreated EV and EDNRB knockout (KO) human brown preadipocytes with EDN1 or EDN3, followed by adipogenic differentiation. EDN3-triggered calcium flux was almost completely blunted in EDNRB KO brown preadipocytes; however, EDN1-mediated cytosolic calcium mobilization was slightly increased in human brown preadipocytes with EDNRB KO (Supplementary Fig. 4d). Ablation of EDNRB in human brown preadipocytes also abolished EDN3-triggered thermogenic activity in differentiated brown adipocytes as UCP1 expression was markedly reduced, without altering adipocyte differentiation, as indicated by unchanged PPARγ expression (Supplementary Fig. 4e).
Loss of EDNRB in preadipocytes impairs glucose and energy metabolism in mice
PDGFRα and PDGFRβ serve as markers for adipose progenitor cells for beige differentiation43,44,45. Analysis of a publicly available scRNA-seq dataset revealed that in mouse scWAT, a higher percentage of adipose stem and progenitor cells (ASPC) expressed Pdgfrα compared to those expressing Pdgfrβ (40% vs. 10%, Supplementary Fig. 5a). Notably, Pdgfrα expression was specifically confined to ASPC, while Pdgfrβ was predominantly expressed in pericytes (Supplementary Fig. 5a). Additionally, there was a strong positive correlation between PDGFRα expression and EDNRB expression in human preadipocytes, whereas PDGFRβ did not show a similar correlation (Supplementary Fig. 5b, c). Thus, to interrogate whether EDNRB is required for thermogenic adipocyte differentiation in mice, we inducibly knocked out EDNRB in PDGFRα+ preadipocytes by crossing Ednrb floxed mice with Pdgfrα-CreERT2 mice (iPreAd-EDNRB KO; iKO). Ednrb floxed mice without Pdgfrα-CreERT2 serve as the control mice (CTL). After tamoxifen (TAM) injection 5 times within 7 days to induce expression of Cre recombinase, the Ednrb gene would be deleted in PDGFRα+ preadipocytes (Supplementary Fig. 5d). We found the expressions of Ednrb were notably decreased in the stromal vascular fraction (SVF) of BAT, subcutaneous WAT (scWAT), and perigonadal WAT (pgWAT) in iKO mice compared with control mice after 1 week of TAM induction (Supplementary Fig. 5e). However, no differences in glucose tolerance, energy metabolism, heat production and body temperature were observed between control and iKO mice (Supplementary Fig. 5f–m).
Cold stimulation is a potent way to induce thermogenic adipocyte differentiation and function in BAT and scWAT and leads to improved metabolism46,47. Intriguingly, we found both mRNA and protein levels of Ednrb were significantly elevated in scWAT, but not in BAT, of mice housed in a cold (5 °C) environment for 7 days compared with mice kept at room temperature (RT, 22 °C) or thermoneutrality (TN, 30 °C) (Fig. 3a,b and Supplementary Fig. 6a). The level of Ednra expression was higher in scWAT compared to BAT at room temperature and thermoneutrality, but it was dramatically down-regulated upon 7-day cold stimulation (Fig. 3c). Similar findings were also found a published RNA-sequencing dataset48 (see re-analyzed data in Supplementary Fig. 6b,c). These data highlight the distinct expression patterns of Ednrb vs. Ednra in response to temperature changes in different fat depots, and further support a potential role of EDNRB in thermogenic regulation of scWAT in vivo.
a–c C57BL6/J male mice were housed at thermoneutral temperature (30 °C, TN) or room temperature (22 °C, RT) or cold (5 °C) for 7 days. The Ednrb mRNA (a) EDNRB protein (b) and Ednra mRNA (c) in BAT and scWAT. TN: n = 4; RT: n = 5; Cold: n = 7 in (a) and (c). TN: n = 3; RT: n = 3; Cold: n = 3 in (b). d Schematic of the experimental design. Male mice were intraperitoneally injected by tamoxifen 5 times within 7 days to induce EDNRB KO in PDGFRα+ preadipocytes (iKO) and were then housed at 5 °C for 7 days to trigger thermogenic differentiation in scWAT. After induction of scWAT browning, GTT, cold test and CLAMS were measured. e Glucose levels during glucose tolerance test (GTT) after 7 days of cold exposure. Area under the curve (AUC) of GTT in the right panel. CTL: n = 5, iKO: n = 6. f, g Heat production in mice after 7 days of cold exposure. Quantification results in the right panels (f). ANCOVA analysis of heat production per body weight of mice (g). n = 8 mice/group. h, i VO2 in mice after 7 days of cold exposure. Quantification results in the right panels (h). ANCOVA analysis of VO2 per body weight of mice (i). n = 8 mice/group. j Acute cold test in mice after 7 days of cold exposure. Body temperature measured by a rectal probe in mice during acute cold test. n = 6 mice/group. Data are presented as mean ± SEM. P values were determined using unpaired two-tailed t tests: AUC of GTT in (e); one-way ANOVA with Tukey’s multiple-comparison test: (a–c), (f), (h); two-way ANOVA with Bonferroni’s multiple-comparison test: GTT curve in (e), (j); two-tailed ANCOVA analysis: (g), (i). Source data are provided as a Source Data file. Panel (d) was created with BioRender.com released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.
Next, to examine the metabolic impact of Ednrb deletion in preadipocytes, we placed the iPreAd-EDNRB KO and control mice in a cold environment for 7 days (Fig. 3d). Body weight of iKO and control mice displayed no difference (Supplementary Fig. 6d). Intriguingly, after 7-day cold exposure, the male mice with EDNRB iKO in adipose progenitor cells became glucose intolerance compared to control mice in 4 independent cohorts (Fig. 3e and Supplementary Fig. 6e-g). Female iKO mice also displayed a trend toward glucose intolerant but did not reach statistical significance (Supplementary Fig. 6h). In addition, the iPreAd-EDNRB KO mice showed a significant reduction in heat production and oxygen consumption (Fig. 3f–i and Supplementary Fig. 6i, j) and were not able to maintain body temperature under acute cold stimulation (Fig. 3j). Importantly, after 24 hrs of cold exposure, while the control animals were able to active full thermogenic program to return to their normal body temperature, the iKO mice still displayed a reduced body temperature, further revealing their defects in adaptive thermogenesis (Fig. 3j). These data suggest that loss of EDNRB in preadipocytes leads to defects in 7-day cold-mediated improvement of glucose and energy metabolism.
Loss of EDNRB in preadipocytes causes defects in cold-induced thermogenic differentiation in scWAT
To investigate whether the dysregulations in glucose and energy metabolism resulted from the impairment of cold-induced thermogenic adipocyte differentiation in iPreAd-EDNRB KO mice, we performed gene expression and histological analysis in scWAT, pgWAT, and BAT of mice after 7-day cold exposure. Compared to the control mice, thermogenic genes, such as Prdm16, Ucp1 and Cidea, were downregulated in scWAT of iPreAd-EDNRB KO mice (Fig. 4a). UCP1 protein levels were also markedly decreased in scWAT of iKO mice compared to those in the control mice (Fig. 4b). Histological analyzes showed the formation of UCP1-positive and multilocular beige adipocytes in scWAT of control animals after 7-day cold exposure; however, this was largely diminished in the scWAT of the iPreAd-EDNRB KO mice (Fig. 4c, d). These phenotypes were found in both male and female mice (Supplementary Fig. 7a–c). In concordance with the in vitro findings, the knockout of EDNRB in preadipocytes resulted in a decrease of ERK1/2 phosphorylation in scWAT after 7 days of cold exposure (Fig. 4e) with no obvious changes in the protein levels of PPARγ and AP2 (Supplementary Fig. 7d–f). Notably, the morphology, mRNA and protein levels of UCP1 in BAT and pgWAT were not altered (Supplementary Fig. 7g-k).
a, b Male mice were intraperitoneally injected by tamoxifen 5 times within 7 days to induce EDNRB KO in PDGFRα+ preadipocytes (iKO) and were then housed at 5 °C for 7 days. a The mRNA levels of thermogenic genes in scWAT. CTL: n = 5, iKO: n = 6. b The protein levels of UCP1 in scWAT. The quantification of protein bands in right panel. n = 4 mice/group. c, d Control and iKO mice were housed at thermoneutral temperature (30 °C, TN) or cold (5 °C) for 7 days. Images of immunohistochemical stain for UCP1 proteins (UCP1 IHC; c) and HE stain (d) in scWAT. High magnification images were in the right panels. Scale bar=100 μm. Two independent experiments were repeated with similar results, as shown in Supplementary Fig. 7a and 7b. e The phosphorylated ERK and total ERK protein in scWAT of mice after 7 days of cold exposure. The quantification of protein bands in right panel. n = 3 mice/group. f, g PDGFRα+ cell lineage-tracing control and EDNRB iKO mice were housed at thermoneutral temperature (30 °C, TN) or cold (5°C) for 7 days. f The immunofluorescent images from the section of scWAT. Scale bar=100 μm. g Quantification of GFP signal in each section of scWAT. CTL at TN: n = 6, iKO at TN: n = 6, CTL at Cold: n = 6, iKO at Cold: n = 7, 4 slides/mouse. h The mRNA levels of Ednrb, Pdgfrα and Ebf2 in SVF of scWAT from mice housed at TN or cold for 7 days. CTL at TN: n = 3, iKO at TN: n = 3, CTL at Cold: n = 3, iKO at Cold: n = 4. Data are presented as mean ± SEM. P values were determined using unpaired two-tailed t tests: (a), (b), (e); one-way ANOVA with Tukey’s multiple-comparison test: (g), (h). Source data are provided as a Source Data file.
To verify whether the decrease of beige adipocyte formation in the iPreAd-EDNRB KO mice resulted from the deletion of Ednrb in PDGFRα+ preadipocytes, we established a lineage-tracing system in our mouse model. The mTmG mice have a transgene expressing membrane-bound tdtomato (mT) which is driven by Rosa26 promoter. After recombination by Cre recombinase, the transgene expresses membrane-bound GFP (mG) in cells. By crossing mTmG mice with Pdgfrα-CreERT2 mice or Pdgfrα-CreERT2/Ednrbf/f mice, we performed a pulse-chase analysis to label the PDGFRα+ preadipocytes with membrane GFP by treating the animals with TAM for 1 week and followed by quantifying GFP+ adipocytes after 7-day cold exposure using fluorescent microscopy and flow cytometry (Supplementary Fig. 8a). Cold significantly increased the frequency of GFP+ adipocytes that were derived from Pdgfrα-expressing adipose progenitors labeled during the 1-week pulse period in scWAT of control mice, but this effect was reduced by half in the scWAT of the iPreAd-EDNRB KO mice (Fig. 4f, g and Supplementary Fig. 8b,c). Cold-stimulated expression of Ednrb, Pdgfrα and Ebf2 was significantly impaired in the SVF of scWAT from the iKO mice (Fig. 4h). Thus, EDNRB appears to play an essential role in cold-induced thermogenic adipocyte differentiation from the PDGFRα+ preadipocytes in scWAT.
Mice with loss of EDNRB in adipose progenitor cells develop metabolic abnormalities upon high-fat diet feeding
In humans, EDNRB levels were significantly lower in scWAT progenitor cells of individuals with obesity (Supplementary Fig. 1b,c). Consistent with these findings, we found the expression Ednrb and Edn3 was downregulated in scWAT of mice fed with a high-fat diet (HFD, 60% kilocalories from fat) for 14 weeks (Fig. 5a), but their expression was not altered in BAT of the same mice (Supplementary Fig. 9a).
a Mice were fed with chow diet (CD) and high-fat diet (HFD) at room temperature for 14 weeks. Ednrb and Edn3 mRNA in scWAT of mice. CD: n = 4, HFD: n = 5. b Schematic of experimental design in DIO mice. Mice were intraperitoneally injected by tamoxifen 5 times within 7 days to induce EDNRB KO in PDGFRα+ preadipocytes (iKO). Mice were fed with HFD at room temperature for 12 weeks. Additional tamoxifen injection was done after 6 weeks of HFD. After HFD feeding, GTT, ITT and CLAMS were measured. c Body weight (BW) in mice during HFD treatment. CTL: n = 5, iKO: n = 6. d Heat production in mice after 12 weeks of HFD treatment. Quantification results in the right panel. n = 6 mice/group. e ANCOVA analysis of heat production per body weight of mice. f, g GTT (f) and insulin tolerance test (ITT; g) in control (CTL) and iKO mice after 12 weeks of CD or HFD treatment. AUC of GTT and AAC of ITT in the right panels. CTL-CD: n = 5, iKO-CD: n = 6, CTL-HFD: n = 5, iKO-HFD: n = 5 in GTT. CTL-CD: n = 5, iKO-CD: n = 6, CTL-HFD: n = 4, iKO-HFD: n = 5 in ITT. Data are presented as mean ± SEM. P values were determined using unpaired two-tailed t tests: (a); one-way ANOVA with Tukey’s multiple-comparison test: (d), AUC of GTT in (f), AAC of ITT in (g); two-way ANOVA with Bonferroni’s multiple-comparison test: (c); two-tailed ANCOVA analysis: (e). Source data are provided as a Source Data file. Panel (b) was created with BioRender.com released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.
To determine whether mice carrying EDNRB deficiency in the adipose progenitors would alter their metabolic responses to excess calories, we first induced Ednrb deletion in PDGFRα+ preadipocytes by TAM injection and fed them with an HFD for 12 weeks. To ensure the KO efficiency could be sustained for over 12 weeks, we gave the mice a second set of TAM injection at 6 weeks of HFD (Fig. 5b). The mice with EDNRB deficiency in preadipocytes gained significantly more weight during HFD treatment (Fig. 5c). To account for the difference in body weight between the two mouse groups, we employed the analysis of covariance (ANCOVA) regression method49,50,51, considering body mass when analyzing metabolic parameters such as energy expenditure. The pronounced obesity phenotype observed in iPreAd-EDNRB KO mice on HFD may stem from reduced energy expenditure/heat production (Fig. 5d, e) and a decreased preference for fat consumption, as reflected by the increased RER (Supplementary Fig. 9b, c). Interestingly, no disparities were observed in oxygen consumption (VO2), carbon dioxide production (VCO2), activity levels, or food intake between control and iPreAd-EDNRB KO mice (Supplementary Fig. 9d–g). Moreover, following 12 weeks of HFD treatment, control mice developed glucose intolerance and insulin resistance compared to the chow-fed control mice, whereas iPreAd-EDNRB KO mice fed with HFD diet displayed exacerbated symptoms across different independent cohorts (Fig. 5f, g and Supplementary Fig. 9h-j). In summary, these results suggest that the loss of EDNRB-mediated thermogenic differentiation in scWAT exacerbates obesity phenotypes and metabolic disruptions.
EDN3-EDNRB signaling promotes thermogenic differentiation of progenitor cells in scWAT and benefits metabolism
The fact that both EDN3 and EDNRB mRNAs were detected in scWAT of mice (Figs. 3a and 5a) suggests that this ligand-receptor interaction may involve cell-cell communication within the WAT microenvironment. To test this hypothesis, we measured the levels of EDN3 protein by ELISA in mouse serum and various tissues collected from mice housed in TN, RT or cold for 7 days. Intriguingly, cold exposure specifically increased the levels of EDN3 in scWAT and pgWAT, but not in serum or other tissues such as heart and BAT (Fig. 6a and Supplementary Fig. 10a), suggesting that EDN3 might function as a local cold-induced adipose niche factor within WAT. To further investigate the source of EDN3, we isolated SVF and adipocytes from scWAT of mice housed at TN or cold conditions and identified that the SVF, but not mature adipocytes, expressed Edn3 mRNA and its levels were increased by three folds upon cold stimulation (Fig. 6b). Isolated SVF cells from scWAT of mice exposed 7-day cold expressed and secreted more EDN3 protein (Fig. 6c and Supplementary Fig. 10b). Expression of EDNRB was also upregulated in SVF cells of scWAT upon cold stimulation (Supplementary Fig. 10c).
a The protein levels of EDN3 in scWAT and BAT of mice housed at thermoneutral temperature (30 °C, TN) or room temperature (22 °C, RT) or cold (5 °C) for 7 days. TN: n = 4; RT: n = 5; Cold: n = 7. b, c Mice were housed at thermoneutral temperature (30 °C, TN) or cold (5 °C) for 7 days. b The SVF cells and adipocytes of scWAT were isolated and Edn3 mRNA was measured. n = 3 mice/group. c The isolated of SVF cells were incubated with DMEM medium for 6 h. The EDN3 protein was measured in the secreted medium. n = 5 mice/group. d Schematic of daily injection of either PBS or EDN3 (10 µg) into scWAT of mice for 7 days. e–i PBS and 10 µg EDN3 were daily injected into scWAT of chow diet-fed mice (CD) for 7 days. e Images of H&E stain in scWAT of mice. The quantification of adipocyte size in the right panel. n = 8 mice/group. 4 slides/mouse. Scale bar= 100 μm. f The mRNA levels of thermogenic genes and adipogenic genes in scWAT of mice. The protein levels of UCP1, PPARγ (g) phosphorylated ERK1/2 and total ERK1/2 (h) in scWAT of mice. The quantification of protein bands in the right panel. n = 3 mice/group. i GTT in mice. AUC of GTT in the right panel. n = 4 mice/group. j, k PBS and 10 µg EDN3 were daily injected into scWAT of DIO mice for 7 days. j The protein levels of UCP1, PPARγ in scWAT of DIO mice. The quantification of protein bands in right panel. n = 3 mice/group. k GTT in mice. AUC of GTT in the right panel. n = 4 mice/group. Data are presented as mean ± SEM. P values were determined using unpaired two-tailed t tests: (b), (c), (e–h), (j), AUC of GTT in (i), (k); one-way ANOVA with Tukey’s multiple-comparison test: (a); two-way ANOVA with Bonferroni’s multiple-comparison test: GTT curve in (i,) (k). Source data are provided as a Source Data file. Panel (d) was created with BioRender.com released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.
Upon re-analysis of a publicly available scRNA-seq dataset32, we found that Edn3 expression was specifically confined to the ASPC population within mouse scWAT (Supplementary Fig. 10d). Thus, we hypothesize the presence of an autocrine effect of EDN3/EDNRB signaling in adipose progenitor cells during cold exposure within scWAT. To investigate this hypothesis, we treated human white preadipocytes with norepinephrine (NE) or forskolin (FSK) for 1 day to mimic cold-mediated signaling. Interestingly, both stimulations resulted in an upregulation of EDN3 and EDNRB expression in adipose progenitor cells (Supplementary Fig. 10e, f). Collectively, these findings suggest that EDN3 may function as an autocrine factor, activating EDNRB signaling in adipose progenitors, thereby contributing to cold-induced browning and remodeling of scWAT.
To examine whether EDN3 administration is sufficient to promote thermogenic differentiation in scWAT, we daily injected recombinant EDN3 (10 µg/mouse) or PBS into both sides of scWAT in mice fed with a regular chow diet (CD) (Fig. 6d). Following 1 week of EDN3 injection, there was a significant elevation in the EDN3 protein within scWAT, with no corresponding increase detected in circulation (Supplementary Fig. 11a). Notably, numerous adipocytes in scWAT exhibited the acquisition of smaller and multi-locular lipid droplets (Fig. 6e and Supplementary Fig. 11b). Furthermore, these adipocytes displayed increased levels of thermogenic marker genes, including Ucp1, Dio2, and Cidea (Fig. 6f) in contrast to scWAT injected with PBS. Additionally, local EDN3 administration led to increased protein levels of UCP1 and phosphorylated ERK1/2, while leaving PPARγ mRNA and protein unchanged in scWAT (Fig. 6g, h and Supplementary Fig. 11c). Notably, there were no alterations in tissue weights or gene expression in other adipose tissues such as BAT and pgWAT (Supplementary Fig. 11d–f).
Moreover, the localized injection of EDN3 into scWAT resulted in improved glucose tolerance (Fig. 6i). This improvement was accompanied by increased glucose-stimulated insulin secretion and a reduction in the RER (Supplementary Fig. 11g, h), with no discernible changes in energy expenditure and oxygen consumption (Supplementary Fig. 11i–k). Consequently, the local administration of EDN3 to scWAT induced favorable effects in mice, manifesting as enhanced glucose homeostasis and fat utilization.
To examine whether EDN3 administration could counteract obesity, we performed daily EDN3 injections into scWAT in mice with diet-induced obesity (DIO) to increase EDN3 protein locally, not in circulation (Supplementary Fig. 12a). scWAT exhibited a shift towards thermogenic differentiation in its gene expression profile, particularly evidenced by an increase in UCP1 levels (Fig. 6j and Supplementary Fig. 12b, c). Moreover, we observed reduced tissue weight and cell size in scWAT, but no adipocytes with multilocular lipid droplets were found (Supplementary Fig. 12d, e). Notably, no significant alterations were noted in the gene expression profiles of BAT and pgWAT (Supplementary Fig. 12f, g).
Consistent with observations in mice fed a chow diet, DIO mice receiving EDN3 injection for 1 week exhibited better glucose tolerance (Fig. 6k), augmented glucose-stimulated insulin secretion (Supplementary Fig. 12h) and a reduction in the RER (Supplementary Fig. 12i). Notably, there were no significant changes in heat production and oxygen consumption in the DIO mice treated with EDN3 (Supplementary Fig. 12j–l). Taken together, these findings suggest that administration of EDN3 could potentially induce the browning of scWAT and promote glucose-stimulated insulin secretion and a shift in fuel utilization towards lipids. These effects collectively contribute to improvements in whole-body glucose metabolism.
To further clarify the role of the EDN3-EDNRB axis in enhancing glucose tolerance and promoting browning of scWAT, we administered PBS or EDN3 injections to both control mice and iPreAd-EDNRB KO mice (Supplementary Fig. 13a). In control mice, EDN3 administration improved glucose tolerance, an effect not observed in the iPreAd-EDNRB KO mice (Supplementary Fig. 13b–d). Additionally, EDN3-injected scWAT showed increased mRNA levels of thermogenic genes such as Ucp1 and Dio2 (Supplementary Fig. 13e), as well as elevated UCP1 protein levels (Supplementary Fig. 13f) in control mice. Conversely, the browning effect of EDN3 was inhibited in the iPreAd-EDNRB KO mice (Supplementary Fig. 13e,f). These findings underscore the crucial role of EDNRB in precursor cells for mediating the browning effects of EDN3 on scWAT.
Discussion
Although the endothelin system was originally identified for vaso-regulatory function, now they are known to regulate multiple cellular processes in different organs in health and diseases. In this study, we unravel the role of EDN3/EDNRB signaling in priming the thermogenic differentiation of both human and murine white adipose precursor cells via activating intracellular cAMP and EPAC1-ERK-mediated pathways. We discover that cold exposure specifically induces EDN3 and EDNRB expression in scWAT, and EDN3/EDNRB is crucial for the cold-induced beige adipocyte formation from the PDGFRα+ adipose precursor cells. This highlights the cell type-specific effects of the endothelin system and provides cellular mechanistic insights into its diverse function.
In addition to adipose progenitor cells, a wide range of mesenchymal cells in various tissues express PDGFRα. It needs to consider the potential contribution from non-adipose tissues in our iPreAd EDNRB KO mouse model. Importantly, our model is inducible, and there were no significant metabolic changes observed under basal conditions. Moreover, metabolic impairments became evident under cold exposure and high-fat diet feeding, which are closely associated with the regulation of EDN3/EDNRB levels, specifically in preadipocytes of scWAT. Most importantly, the EDNRB level is highly positively correlated with Pdgfra expression in preadipocytes. Apart from adipose tissue, fibroblasts of the heart have the second-highest number of cells expressing Pdgfra, as indicated by scRNA-seq data from the Tabula Muris52. Nonetheless, despite this observation, we did not detect any changes in EDN3 secretion within the heart under cold exposure, nor did we observe any discernible functional declines in heart function. Hence, while we cannot completely dismiss the influence of deleting EDNRB in other tissues expressing PDGFRα, our data strongly suggest the observed metabolic effects in our iPreAd EDNRB KO largely originate from adipose tissues.
Indeed, EDN3 and EDNRB are found present in various tissues to regulate myriad biological functions such as cell proliferation/migration53, cell adhesion54, and differentiation55. Moreover, the levels of EDN3 and EDNRB are dramatically downregulated in pancreas56,57 and WAT57 of patients with type 2 diabetes, respectively. An EDN3-like peptide can regulate glucose homeostasis in DIO mice via stimulating GLP-1 secretion in enteric cells and inhibiting gluconeogenesis in hepatocytes58. Insulin induces endothelial EDNRB expression that leads to attenuating diabetes-derived atherosclerosis25. In the present study, we uncover the role of EDN3-EDNRB in white adipose tissue to acquire the thermogenic function. Existing findings strongly suggest that the formation and activation of beige adipocytes within WAT play crucial roles in influencing systemic glucose homeostasis and insulin sensitivity59,60,61. This phenomenon occurs through the secretion of various factors that exert regulatory effects on other tissues throughout the body. In our model, the thermogenic differentiation of scWAT induced by EDN3 links to the increased insulin secretion from beta cells in response to glucose challenge, potentially influencing systemic glucose homeostasis. Further investigation is needed to elucidate the potential secreted factors that mediate communication between scWAT and beta cells.
Local EDN3 administration on scWAT enhanced the browning and glucose tolerance in both lean and DIO mice. Since EDNRB is expressed in mature adipocytes and other cell types, such as endothelial cells, it is important to assess potential side effects from actions in these other cell types. Our findings showed that the effect of EDN3 was prevented in iPreAd-ENDRB knockout mice, highlighting the role of the EDN3-EDNRB axis in regulating the browning of PDGFRα+ adipose progenitor cells and systemic glucose homeostasis. Nevertheless, this experiment was conducted using mouse groups of mixed genders, further investigation is necessary.
We find that EDN3/EDNRB signaling activates the cAMP-EPAC1-ERK pathway, leading to the proliferation of preadipocytes and upregulation of UCP1, which subsequently differentiate into thermogenically active beige adipocytes within the WAT. Moreover, EPAC1 has recently been recognized as a crucial molecule, operating independently of the PKA pathway, in the proliferation and differentiation of thermogenic preadipocytes under β3-adrenergic signaling or cold stimulation42. Here, we present an alternative physiological cue for activating EPAC1 via EDN3/EDNRB signaling during cold-induced beige adipogenesis. In addition to the cAMP-EPAC1-ERK pathway, previous findings by Park et al. suggest that EDNRB activates eNOS/NO signals, promoting the differentiation of perivascular progenitor cells into brown adipocytes62.
While previous research has shown that blocking the ERK pathway enhances WAT browning in animals with obesity, it’s crucial to consider that these findings were derived from the adiponectin-Cre mouse model and primarily examined the impact of ERK in mature adipocytes, along with its upstream factor, Cdk563. In contrast, our study reveals that ERK activation induced by EDN3/EDNRB in the PDGFRα+ preadipocyte promotes thermogenic differentiation, particularly under EPAC1 signaling. These results further support the idea that EDN3-EDNRB can activate different signaling pathwaysα to regulate various biological effects, depending on the specific cellular context.
Within the WAT niche, EDN3 is produced by SVFs of scWAT. Although the basal level of EDN3 is low in WAT, it can be further induced upon cold exposure. According to scRNA-seq data, EDN3 expression may be predominantly confined to adipose progenitor cells, consistent with our previous findings in human clonal preadipocytes where both EDN3 and its receptor EDNRB were expressed and positively correlated with thermogenic differentiation potential19. In this study, we present evidence that EDN3 expression was detected in human white preadipocytes and was induced by norepinephrine and cAMP stimulation. However, the specific cellular origins of EDN3 within adipose tissue and the precise mechanisms governing cold-induced upregulation of EDN3 remain to be determined and merit further investigation.
In the current investigation, we observed that EDNRB is essential for thermogenic differentiation of human brown preadipocytes in vitro. However, its impact on mouse BAT in vivo seems minimal. The discrepancy may stem from the unique microenvironment within mouse BAT, which is not replicated in an in vitro setting. Considering species differences, our in vitro experiments on human cells may explain variations from mouse BAT. Given human brown adipocytes share similarities with mouse beige adipocytes, our findings not only highlight the significance of the EDN3/EDNRB pathway in human thermogenic function but also reveal its potential involvement in beiging of WAT in mice.
In humans, mutations in EDNRB and EDN3 genes interfere with the differentiation and development of neural crest cells into two distinct cell types, namely melanocytes and enteric neural cells. These mutations are the causative of Waardenburg-Shah syndrome and Hirschsprung disease23,64. These diseases are typically regarded as focal ailments primarily affecting the distal colon and rectum. However, defects in motility of the upper gastrointestinal tract and nervous system suggest that the condition may extend beyond the distal gastrointestinal tract. A study revealed that adult patients with Hirschsprung disease may exhibit prediabetic characteristics, hinting at a potential involvement of EDNRB in glucose metabolism65. Given these findings, further investigation into the thermogenic function and browning of scWAT in these patients is warranted.
In summary, using the human preadipocytes and cell-specific transgenic mouse model, we elucidate the role of EDN3/EDNRB signaling in the priming of white adipose progenitor cells towards acquiring thermogenic differentiation potential. Our results indicate that the EDN3/EDNRB signaling system plays a pivotal role in this process. Notably, adipose progenitors in scWAT of individuals with obesity exhibit lower levels of EDNRB expression32,33, and our findings present a plausible explanation for the reduced thermogenic response in WAT of individuals with obesity. These results offer promising prospects for developing effective therapeutic interventions against obesity and its associated metabolic complications.
Methods
Immortalization of human white and brown preadipocytes
All cellular experiments comply with the relevant ethical regulations of Joslin Diabetes Center and China Medical University. Primary SVF cells from human white and brown fat were infected with a retrovirus containing the pBABE-hTERT-Hygro plasmid (Addgene no. 1773, Cambridge, MA) in the presence of 4 μg/ml Polybrene until reaching 90% confluence. Subsequently, cells were subjected to hygromycin selection (400 μg/ml) in DMEM/H medium supplemented with 10% FBS and antibiotics. Upon completion of drug selection, cells were maintained in culture medium containing 50 μg/ml hygromycin for 2 weeks.
Generation of human white preadipocytes with EDNRB overexpression
Endogenous EDNRB gene expression was activated by the CRISPR synergistic activation mediator (CRISPR-SAM) system established by Konermann et al.34. Different single guide RNAs (sgRNAs) #1-3 targeting the human EDNRB promoter were designed using the online tool: http://sam.genome-engineering.org/database/ and cloned into the lentiviral backbone vector. The sequences of different sgRNAs are listed in Supplementary Table 1.
Lentiviruses expressing dCas9-VP64, MS2-P65-HSF1, and sgRNAs A-D were generated in 293 T cells by co-transfection of the packaging plasmids psPAX.2 and psMD2.G using PolyJet DNA in vitro transfection reagent (SignaGen Laboratories, Rockville, MD). Virus-containing supernatant was harvested 48 h post transfection and filtered through a 0.45 μm filter (Fisher Scientific, Pittsburgh, PA) followed by concentration using Lenti-X concentrator (Clontech).
Immortalized human white preadipocytes were infected with dCas9-VP64 and MS2-P65-HSF1 lentiviruses in the presence of polybrene (8 µg/ml). Two days later, the cells were selected with blasticidin (5 µg/ml) and hygromycin (250 µg/ml) for 6 days to generate CRISPR-SAM cells. Then CRISPR-SAM cells were further infected with lentiviruses containing empty vector (EV) or sgRNAs #1-3 targeting to human EDNRB promoter, followed by Zeocin (100 µg/ml) selection to generate EV or EDNRB OE #1-3 (sgRNA #1-3) cells.
Generation of EDNRB KO in human white and brown preadipocytes
An sgRNA (5’-CTGCGGCCTGTCGCGGATCT-3’) targeting the human EDNRB gene was cloned into the pSpCas9(BB)-2A-Puro (PX459) V2.0 plasmid. Immortalized human white and brown preadipocytes were transiently transfected with empty plasmid or plasmid containing gRNA targeting EDNRB. Two days later, the cells were selected with puromycin (1 µg/ml) for 4 days to generate EDNRB knockout cells.
Culture and differentiation of human white and brown preadipocytes
Human preadipocytes were plated and grown in high glucose-DMEM (DMEM/H) medium supplemented with 10% FBS. For adipocyte differentiation, cells were grown for 3 days until reaching confluence (EDN3 pretreatment was administered in this period) and then treated with the adipogenic induction medium containing 10% FBS, 0.5 mM isobutylmethylxanthine, 0.1 µM dexamethasone, 0.5 µM human insulin, 2 nM T3, 30 µM indomethacin, 17 µM pantothenate, 33 µM biotin (Sigma-Aldrich, Dallas, TX) for another 12 days. Cells were routinely checked for mycoplasma contamination and all the cells used in this study were free of mycoplasma.
Gene expression
Total RNA was extracted from cells or tissues with Trizol and purified using a spin column kit (Zymo Research). RNA (500 ng) was reverse transcribed with a high-capacity complementary DNA (cDNA) reverse transcription kit (Applied Biosystems). Real-time PCR was performed in a solution containing cDNA (6 ng), forward and reverse oligonucleotide primers (300 nM each) and SYBR green PCR Master Mix (Roche). Fluorescence was determined and analyzed in an ABI 7900 sequence detection system (Applied Biosystems). 18 S rRNA and acidic ribosomal phosphoprotein P0 (ARBP) expression were used to normalize gene expression in human and mouse cells, respectively. Real-time PCR primer sequences are listed in Supplementary Table 2.
Protein expression
Cells were scraped from tissue-culture plates with RIPA buffer. After centrifugation (12,000 g, 15 min), protein lysates were harvested and stored at -80°C until further use. Protein concentrations were determined by using the Pierce BCA kit (Life Technologies), according to the manufacturer’s instructions. For immunoblots, lysates were diluted into Laemmli buffer, boiled, and loaded onto 4-15% Tris gels for SDS–PAGE. After complete separation of the proteins, they were transferred onto a PVDF membrane (Amersham Biosciences) and were blocked in western blocking buffer (Roche). Primary antibodies for EDNRB (NBP130599, Novus Biologicals), UCP1 (ab23841, abcam), phospho-ERK1/2 (Thr202/Tyr204) (#4370, Cell Signaling), total ERK1/2 (#4695, Cell Signaling), phospho-CREB (Ser133) (#9198, Cell Signaling), total CREB (#9197, Cell Signaling), Phospho-PKA Substrate (#9624, Cell Signaling), PPARγ (GTX32803, GeneTex), FABP4/AP2 (GTX636995, GeneTex), EPAC1 (#4155, Cell Signaling), TUBULIN (#2146, Cell Signaling), VINCULIN (sc-25336, Santa Cruz) or ACTIN (MAB1501, Millipore) were applied in blocking buffer over night at 4°C. After washing 3 times for 10 min with TBS-T, secondary antibodies were applied for 1 h in blocking buffer. Membranes were washed again 3 times for 10 min in TBS-T and developed using chemiluminescence (ThermoFisher). For quantification of scanned immunoblots, integrated pixel density of interested bands was measured using ImageJ software. Total actin was used as an endogenous control for normalization.
Oil Red O staining
The cells were washed twice with PBS and subsequently fixed with 10% buffered formalin for 15 min at room temperature. After fixation, the cells were stained with a working solution of filtered Oil Red O. This working solution was prepared by combining 3 parts of 0.5% Oil Red O in isopropanol with 2 parts of water. The staining process was conducted for 1 hour at room temperature. Following the staining, the cells were washed several times with distilled water and then subjected to visualization.
Seahorse bioanalyzer
The oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) were measured using the Seahorse XF24 Flux analyzer (Agilent Technologies) following the manufacturer’s protocol. Cells were cultivated and differentiated in XF24 cell culture microplates. To assess the glucose-dependent OCR, cells were incubated with serum-free DMEM containing 25 mM glucose and 1 mM pyruvate. Different drugs were administered to measure various parameters: 2 µM oligomycin (ATP production respiration), 10 µM forskolin (FSK-dependent respiration), 2 µM FCCP (maximal respiration), and 2 µM antimycin A (non-mitochondrial respiration). To determine the fatty acid-dependent OCR, cells were incubated in running media (2.5 mM glucose, 0.5 mM carnitine, 5 mM HEPES, and 200 µM palmitate-BSA in KRB buffer) with or without 60 µM etomoxir (ETO), an inhibitor of carnitine palmitoyltransferase-1 (CPT1), followed by sequential addition of oligomycin, FCCP, and antimycin A. In order to measure the ECAR, cells were incubated with serum-free DMEM. Different drugs were applied to evaluate different parameters: 10 mM glucose (glucose-dependent ECAR), 2 µM oligomycin (maximal glycolysis), and 50 mM 2-deoxy-D-glucose (inhibition of glycolysis). Both OCR and ECAR values were normalized using protein content as a reference.
Thermogenic capacity
Differentiated human adipocytes were cultured in DMEM/H medium and treated with 250 nM of ERthermAC (SCT057, Sigma-Aldrich) at 37 °C for 30 minutes. Before imaging, the medium was replaced with fresh 90 µl of phenol red-free DMEM/H. Fluorescence in the stained cells was observed using a Zeiss LSM710NLO confocal microscope, operating in point scan mode with a 40x objective lens. The microscope setup included a stage incubator and CO2 supply. Images of ERthermAC were captured using a 520 nm laser, and the emission band was set between 580–640 nm. The temperature inside the microscope was maintained at a constant 25 °C. After measuring a point of basal fluorescence, 10 µM forskolin (F6886, Sigma-Aldrich) was added to initiate thermogenesis. Images were recorded every 5 minutes for a total duration of 100 minutes. The results were interpreted based on relative intensity, where the fluorescence intensity was normalized to the basal measurements.
Cytosolic calcium
Cells were seeded into each well of a 96-well plate. After growth medium was removed, the cells were incubated with 100 μl of FLUOFORTE® reagent (ENZ-51016, ENZO) at 37 °C, 5% CO2 incubator for 1 hour. After washed with Hank’s Buffer with 20 mM HEPES (HHBS), the cells were incubated with 100 μl HHBS, and the fluorescence was then immediately detected with a GloMax Discover Multimode Detection System (Promega) using 490 nm excitation and emission at 525 nm. Basal fluorescence readings were obtained from three separate points. Following this, either 100 nM EDN1 or EDN3 was added to induce calcium mobilization. Fluorescence was recorded at 20-second intervals over a period of 200 s. The results were analyzed as relative intensity, with the fluorescence intensity normalized to the basal measurements.
Cell proliferation
A total of 10,000 cells were seeded into each well of a 96-well plate. Following incubation with growth medium for 3 days, the cells were trypsinized, and the cell number was determined using an Automated Cell Counter (Countess 3, ThermoFisher Scientific).
Animals
All animal experiments adhered to the relevant ethical regulations governing the use of small rodents and received approval from the Institutional Animal Care and Use Committees at Joslin Diabetes Center and China Medical University. C57BL6J mice (Stock no. 000664), Ednrb flox mice (Stock no. 009063), Pdgfrα-CreERT2 mice (Stock no. 032770), and homozygous Rosa26-mTmG mice (Stock no. 007676) were obtained from the Jackson Laboratories.
To generate inducible deletion of EDNRB in PDGFRα+ preadipocytes (iPreAd-EDNRB KO), Ednrb flox mice were crossed with Pdgfrα-CreERT2 mice; the Ednrd flox mice without Pdgfrα-CreERT2 serve as control mice. To create Rosa26-mTmG reporter mice in PDGFRα+ preadipocytes, either with or without EDNRB KO, homozygous Rosa26-mTmG mice were initially crossed with Pdgfrα-CreERT2 mice. Subsequently, the mice with the Rosa26-mTmG reporter in PDGFRα+ preadipocytes were crossed with either wild-type or Ednrb flox mice. For the induction of Cre expression, mice received intraperitoneal injections of tamoxifen (Sigma-Aldrich) at a dosage of 100 mg/kg, dissolved in a solution consisting of 10% ethanol and 90% peanut oil, for a total of five injections over a period of seven days.
During cold exposure or thermoneutral (TN) housing experiments, mice were housed in a controlled environmental diurnal chamber (Caron Products & Services) set to either 5 °C (cold) or 30 °C (TN).
Mice had ad libitum access to food and water and were maintained on a normal chow diet (Mouse Diet 9 F 5020, PharmaServ) containing 22% of calories from fat, 23% from protein, and 55% from carbohydrates. The mice were subjected to a 12-h light-dark cycle with 30% humidity. Euthanasia was performed using isoflurane (catalog no. NDC 66794-017-25, Piramal Critical Care) by inhalation.
Isolation of the stromovascular (SVF) and adipocyte fraction from adipose tissues
Adipose tissues were dissected, minced, and digested using a cocktail consisting of type 1 collagenase (1.5 mg/ml; Worthington Biochemical), dispase II (2.5 U/ml; Stemcell Technologies), and fatty acid-free BSA (2%; Gemini Bio-Products) in Hanks’ balanced salt solution (Corning Hanks’ Balanced Salt Solution with calcium and magnesium). The digestion process was carried out for 45 min at 37 °C with gentle shaking. The digested tissue was then centrifuged at 500 g and 4 °C for 10 min. The top layer containing the adipocyte fraction was carefully collected. After removing the remaining supernatant, the pellet was collected as the SVF fraction. The adipocyte fraction was passed through a 200 μm cell strainer, washed with 3% BSA in PBS, and centrifuged at 30 g for 5 minutes to remove debris and impurities. The SVF fraction was incubated with ammonium-chloride-potassium lysis buffer (ACK Lysing Buffer, Lonza) to lyse red blood cells. The resulting cell suspension was then filtered through a 40 μm cell strainer, washed with 1.5% BSA in PBS, and centrifuged at 500 g for 7 min to obtain a purified SVF fraction.
Indirect calorimetry
Mice were individually housed in metabolic cages of a Comprehensive Lab Animal Monitoring System (CLAMS) or a Sable Systems’ Promethion system at room temperature. After a 12 h acclimation period, animals were monitored for 24 h in order to obtain measurements for the volume of oxygen consumption (VO2), the volume of carbon dioxide production (VCO2) and respiratory exchange ratio (RER), which was calculated as the ratio of total VCO2 produced to total VO2 consumed, heat production, activity and accumulated food intake. To account for differences of body weight among mice in metabolic parameters, the analysis of covariance (ANCOVA) was conducted using the Energy Expenditure Analysis Webpage provided by NIDDK Mouse Metabolic Phenotyping Centers (MMPC) (http://www.mmpc.org/shared/regression.aspx).
Cold tolerance test
For cold tolerance tests, mice were singly housed and placed at an ambient temperature of 5 °C. Rectal temperature measurements were done using an aRET-3 rectal probe (Physitemp).
Glucose and insulin tolerance tests
For the glucose tolerance test (GTT), animals were fasted for 6 hours (7AM to 1PM) with free access to drinking water. A baseline blood sample was collected from the tail of fully conscious mice, followed by an i.p. injection of glucose (2.0 g/kg body weight), and blood was taken from the tail at 15, 30, 60, and 120 minutes after injection. For the insulin tolerance test (ITT), animals were fasted for 6 h (7AM–1PM) with free access to drinking water. Baseline blood samples were collected from the tail of fully conscious mice. Insulin (1 U/kg body weight) (HumulinO; Eli Lilly) was administered by i.p. injection, and blood samples were taken from the tail at 15, 30, 60, and 90 minutes after injection. Glucose concentrations were determined from blood using an InfinityO Blood Glucose Meter (US Diagnostics).
Histological analysis
Tissues were fixed for 24 hours in 10% formalin and embedded in paraffin. Paraffin blocks were cut into 5 µm sections and stained with hematoxylin and eosin (H&E). For UCP1, PPARγ, AP2 IHC staining, the sections were deparaffinized and rehydrated. Antigen retrieval was performed using a modified citrate buffer (Dako Target Retrieval Solution, Agilent). The sections were then processed according to the instructions provided in the Vectastain ABC kit (PK-6105, Vector Lab). Blocking was carried out using blocking serum, followed by incubation with the primary Ucp1 antibody overnight at 4 °C. On the following day, the slides were washed with PBS and incubated with secondary antibodies at a 1:200 dilution for one hour. Subsequently, the slides were incubated with the vectastain ABC reagent for 30–45 minutes, followed by the addition of DAB substrate (SK-4100, Vector Lab) for 2–10 minutes. After a hematoxylin counterstain (S-3309, Agilent), the slides were mounted and imaged.
Flow cytometry
The mice were subjected to either thermoneutral (TN) housing or cold exposure for a duration of 7 days. Adipocytes were isolated from subcutaneous white adipose tissue (scWAT). The adipocytes were collected from the top layer, passed through a 200 μm cell strainer, washed with 3% BSA in PBS, and centrifuged at 30 g for 5 minutes. This washing step was repeated three times. Subsequently, the GFP and tdTomato signals within the adipocytes were promptly analyzed using a BD FACSAria flow cytometer (Becton Dickinson).
Measurement of EDN3 using ELISA
100 mg of tissue was washed with PBS and homogenized on ice using a glass homogenizer. To disrupt the cell membranes, the homogenate underwent three freeze-thaw cycles. Afterward, the homogenate was centrifuged at 5000 g, 2–8 °C for 5 minutes, and the resulting supernatant was collected for analysis. Serum samples were obtained by centrifuging the blood at 1000 g for 15 minutes. EDN3 levels in BAT, scWAT, pgWAT, heart and serum were measured using a sandwich enzyme immunoassay (OKEH01505, Edn3 ELISA Kit Mouse, Aviva Systems Biology).
Statistics
All statistics were calculated using Microsoft Excel and GraphPad Prism. Unpaired Student’s t tests were performed to compare only 2 groups. One-way and two-way ANOVA followed by Tukey’s post-hoc tests were performed for multiple comparisons. Correlations were established based on Spearman’s correlation tests, and the Spearman’s correlation coefficient was provided. Significance of ANCOVA analysis was calculated directly by the Energy Expenditure Analysis Webpage of MMPC. All experiments involving mice were conducted with comparisons and statistical analysis within littermates. P values less than 0.05 were considered statistically significant.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
The authors declare that the data supporting the findings of this study are presented within the paper and its Supplementary Information files. Source data are provided with this paper.
References
Kopelman, P. G. Obesity as a medical problem. Nature 404, 635–643 (2000).
Cornier, M. A. et al. The metabolic syndrome. Endocr. Rev. 29, 777–822 (2008).
Ouellet, V. et al. Brown adipose tissue oxidative metabolism contributes to energy expenditure during acute cold exposure in humans. J. Clin. Investig. 122, 545–552 (2012).
Blondin, D. P. et al. Contributions of white and brown adipose tissues and skeletal muscles to acute cold-induced metabolic responses in healthy men. J. Physiol. 593, 701–714 (2015).
Cypess, A. M. et al. Cold but not sympathomimetics activates human brown adipose tissue in vivo. Proc. Natl Acad. Sci. USA 109, 10001–10005 (2012).
Cypess, A. M. et al. Identification and importance of brown adipose tissue in adult humans. N. Engl. J. Med. 360, 1509–1517 (2009).
van Marken Lichtenbelt, W. D. et al. Cold-activated brown adipose tissue in healthy men. N. Engl. J. Med. 360, 1500–1508 (2009).
Becher, T. et al. Brown adipose tissue is associated with cardiometabolic health. Nat. Med. 27, 58–65 (2021).
Soberg, S. et al. Altered brown fat thermoregulation and enhanced cold-induced thermogenesis in young, healthy, winter-swimming men. Cell Rep. Med. 2, 100408 (2021).
Ikeda, K. et al. UCP1-independent signaling involving SERCA2b-mediated calcium cycling regulates beige fat thermogenesis and systemic glucose homeostasis. Nat. Med. 23, 1454–1465 (2017).
Kazak, L. et al. A creatine-driven substrate cycle enhances energy expenditure and thermogenesis in beige fat. Cell 163, 643–655 (2015).
Guan, H. P. et al. A futile metabolic cycle activated in adipocytes by antidiabetic agents. Nat. Med. 8, 1122–1128 (2002).
Leitner, B. P. et al. Mapping of human brown adipose tissue in lean and obese young men. Proc. Natl Acad. Sci. USA 114, 8649–8654 (2017).
Hauser, A. S., Attwood, M. M., Rask-Andersen, M., Schioth, H. B. & Gloriam, D. E. Trends in GPCR drug discovery: new agents, targets and indications. Nat. Rev. Drug Discov. 16, 829–842 (2017).
Sveidahl Johansen, O., Ma, T. & Gerhart-Hines, Z. Leveraging GPCR signaling in thermogenic fat to counteract metabolic diseases. Mol. Metab. 60, 101474 (2022).
Sveidahl Johansen, O. et al. Lipolysis drives expression of the constitutively active receptor GPR3 to induce adipose thermogenesis. Cell 184, 3502–3518 (2021).
Quesada-Lopez, T. et al. The lipid sensor GPR120 promotes brown fat activation and FGF21 release from adipocytes. Nat. Commun. 7, 13479 (2016).
Sato, M. et al. Cell-autonomous light sensitivity via Opsin3 regulates fuel utilization in brown adipocytes. PLoS Biol. 18, e3000630 (2020).
Xue, R. et al. Clonal analyses and gene profiling identify genetic biomarkers of the thermogenic potential of human brown and white preadipocytes. Nat. Med. 21, 760–768 (2015).
Luscher, T. F. & Barton, M. Endothelins and endothelin receptor antagonists: therapeutic considerations for a novel class of cardiovascular drugs. Circulation 102, 2434–2440 (2000).
Dhaun, N. & Webb, D. J. Endothelins in cardiovascular biology and therapeutics. Nat. Rev. Cardiol. 16, 491–502 (2019).
Davenport, A. P. et al. Endothelin. Pharm. Rev. 68, 357–418 (2016).
Baynash, A. G. et al. Interaction of endothelin-3 with endothelin-B receptor is essential for development of epidermal melanocytes and enteric neurons. Cell 79, 1277–1285 (1994).
Schneider, M. P., Boesen, E. I. & Pollock, D. M. Contrasting actions of endothelin ET(A) and ET(B) receptors in cardiovascular disease. Annu Rev. Pharm. Toxicol. 47, 731–759 (2007).
Park K. et al. Insulin decreases atherosclerosis by inducing endothelin receptor B expression. JCI Insight 1, e86574 (2016).
Eguchi, S., Hirata, Y., Imai, T. & Marumo, F. Endothelin receptor subtypes are coupled to adenylate cyclase via different guanyl nucleotide-binding proteins in vasculature. Endocrinology 132, 524–529 (1993).
Schinelli, S. et al. Stimulation of endothelin B receptors in astrocytes induces cAMP response element-binding protein phosphorylation and c-fos expression via multiple mitogen-activated protein kinase signaling pathways. J. Neurosci. 21, 8842–8853 (2001).
Shinohara, H. et al. Gi2 signaling enhances proliferation of neural progenitor cells in the developing brain. J. Biol. Chem. 279, 41141–41148 (2004).
Bagnato, A. et al. Endothelin B receptor blockade inhibits dynamics of cell interactions and communications in melanoma cell progression. Cancer Res 64, 1436–1443 (2004).
Klepac, K. et al. The Gq signalling pathway inhibits brown and beige adipose tissue. Nat. Commun. 7, 10895 (2016).
Maniyadath, B., Zhang, Q., Gupta, R. K. & Mandrup, S. Adipose tissue at single-cell resolution. Cell Metab. 35, 386–413 (2023).
Emont, M. P. et al. A single-cell atlas of human and mouse white adipose tissue. Nature 603, 926–933 (2022).
Hildreth, A. D. et al. Single-cell sequencing of human white adipose tissue identifies new cell states in health and obesity. Nat. Immunol. 22, 639–653 (2021).
Konermann, S. et al. Genome-scale transcriptional activation by an engineered CRISPR-Cas9 complex. Nature 517, 583–588 (2015).
Wang C. H. et al. CRISPR-engineered human brown-like adipocytes prevent diet-induced obesity and ameliorate metabolic syndrome in mice. Sci Transl Med. 12, eaaz8664 (2020).
Kriszt, R. et al. Optical visualisation of thermogenesis in stimulated single-cell brown adipocytes. Sci. Rep. 7, 1383 (2017).
Deliu, E. et al. Intracellular endothelin type B receptor-driven Ca2+ signal elicits nitric oxide production in endothelial cells. J. Biol. Chem. 287, 41023–41031 (2012).
Shihoya, W. et al. Activation mechanism of endothelin ETB receptor by endothelin-1. Nature 537, 363–368 (2016).
Kajimura, S., Spiegelman, B. M. & Seale, P. Brown and beige fat: physiological roles beyond heat generation. Cell Metab. 22, 546–559 (2015).
Reverte-Salisa, L., Sanyal, A. & Pfeifer, A. Role of cAMP and cGMP signaling in brown fat. Handb. Exp. Pharm. 251, 161–182 (2019).
de Rooij, J. et al. Epac is a Rap1 guanine-nucleotide-exchange factor directly activated by cyclic AMP. Nature 396, 474–477 (1998).
Reverte-Salisa, L. et al. EPAC1 enhances brown fat growth and beige adipogenesis. Nat. Cell Biol. 26, 113–123 (2024).
Lee, Y. H., Petkova, A. P., Mottillo, E. P. & Granneman, J. G. In vivo identification of bipotential adipocyte progenitors recruited by beta3-adrenoceptor activation and high-fat feeding. Cell Metab. 15, 480–491 (2012).
Vishvanath, L. et al. Pdgfrbeta+ mural preadipocytes contribute to adipocyte hyperplasia induced by high-fat-diet feeding and prolonged cold exposure in adult mice. Cell Metab. 23, 350–359 (2016).
Benvie, A. M. et al. Age-dependent Pdgfrbeta signaling drives adipocyte progenitor dysfunction to alter the beige adipogenic niche in male mice. Nat. Commun. 14, 1806 (2023).
Shamsi, F., Wang, C. H. & Tseng, Y. H. The evolving view of thermogenic adipocytes - ontogeny, niche and function. Nat Rev Endocrinol, 12, 726–744 (2021).
Cohen, P. & Kajimura, S. The cellular and functional complexity of thermogenic fat. Nat. Rev. Mol. Cell Biol. 22, 393–409 (2021).
Roh, H. C. et al. Warming induces significant reprogramming of beige, but not brown, adipocyte cellular identity. Cell Metab. 27, 1121–1137.e1125 (2018).
Tschop, M. H. et al. A guide to analysis of mouse energy metabolism. Nat. Methods 9, 57–63 (2011).
Muller, T. D., Klingenspor, M. & Tschop, M. H. Revisiting energy expenditure: how to correct mouse metabolic rate for body mass. Nat. Metab. 3, 1134–1136 (2021).
Mina, A. I. et al. CalR: a web-based analysis tool for indirect calorimetry experiments. Cell Metab. 28, 656–666 e651 (2018).
Tabula Muris, C. et al. Single-cell transcriptomics of 20 mouse organs creates a Tabula Muris. Nature 562, 367–372 (2018).
Goto, A. et al. GDNF and endothelin 3 regulate migration of enteric neural crest-derived cells via protein kinase A and Rac1. J. Neurosci. 33, 4901–4912 (2013).
Gazquez, E. et al. Endothelin-3 stimulates cell adhesion and cooperates with beta1-integrins during enteric nervous system ontogenesis. Sci. Rep. 6, 37877 (2016).
Watanabe, Y. et al. Differentiation of mouse enteric nervous system progenitor cells is controlled by endothelin 3 and requires regulation of ednrb by SOX10 and ZEB2. Gastroenterology 152, 1139–1150.e1134 (2017).
Marselli, L. et al. Gene expression profiles of beta-cell enriched tissue obtained by laser capture microdissection from subjects with type 2 diabetes. PLoS One 5, e11499 (2010).
Rahimi, M. et al. Age-related obesity and type 2 diabetes dysregulate neuronal associated genes and proteins in humans. Oncotarget 6, 29818–29832 (2015).
Bartfai T. et al. EDN3-like peptides and uses thereof. US Patent US20130345392 (2013).
Cohen, P. et al. Ablation of PRDM16 and beige adipose causes metabolic dysfunction and a subcutaneous to visceral fat switch. Cell 156, 304–316 (2014).
Shao, M. et al. Cellular origins of beige fat cells revisited. Diabetes 68, 1874–1885 (2019).
Czech, M. P. Mechanisms of insulin resistance related to white, beige, and brown adipocytes. Mol. Metab. 34, 27–42 (2020).
Park, K. et al. Endothelial cells induced progenitors into brown fat to reduce atherosclerosis. Circ. Res. 131, 168–183 (2022).
Banks, A. S. et al. An ERK/Cdk5 axis controls the diabetogenic actions of PPARgamma. Nature 517, 391–395 (2015).
Bondurand, N., Dufour, S. & Pingault, V. News from the endothelin-3/EDNRB signaling pathway: role during enteric nervous system development and involvement in neural crest-associated disorders. Dev. Biol. 444, S156–S169 (2018).
Medhus, A. W., Bjornland, K., Emblem, R., Haug, E. & Husebye, E. Are patients with Hirschsprung’s disease prediabetic? Metabolism 50, 1389–1390 (2001).
Acknowledgements
This work was supported in part by US National Institutes of Health (NIH) grants R01DK132469, R01DK133528, R01DK102898 (to Y.-H.T.), K01DK125608 (to F.S.), and P30DK036836 and S10OD028568 (to Joslin Diabetes Center’s Diabetes Research Center, DRC). C.-H.W. was supported by grants from National Science and Technology Council, Taiwan (MOST110-2320-B-039-063-MY3, NSTC112-2320-B-039-023-MY3 and NSTC113-2314-B-039-068) and intramural fundings (CMU109-YTY-01, CMU112-IP-01, and 10951M1) from China Medical University, Taiwan. This work was also financially supported by the “Drug Development Center, China Medical University” from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan. Experiments and data analysis were performed in part through the use of the Medical Research Core Facilities, Office of Research & Development at China Medical University, Taiwan. T.T. was supported by the SUNSTAR Research Fellowship (Hiroo Kaneda Scholarship, Japan), the American Heart Association Grant 903968, and the Eleanor and Miles Shore Program Award from Harvard Medical School.
Author information
Authors and Affiliations
Contributions
C.-H.W. designed and directed the research, conducted most of the experiments, analyzed the data, and wrote the paper. T.T. conducted EDN3 injection experiments in mice, and L.-H.W. carried out EDN3 acute treatment experiments in cells. C.-Y.Y. and T.L.H. performed immunohistochemistry staining. M.S. maintained Ednrb floxed mice. F.S. generated the PDGFRα+ cell lineage-tracing mouse model. Y.-H.T. co-directed the research and co-wrote the paper. All authors read and approved the final version of the manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks the anonymous, reviewers for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Source data
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Wang, CH., Tsuji, T., Wu, LH. et al. Endothelin 3/EDNRB signaling induces thermogenic differentiation of white adipose tissue. Nat Commun 15, 7215 (2024). https://doi.org/10.1038/s41467-024-51579-0
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-024-51579-0
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.