Light-responsive adipose-hypothalamus axis controls metabolic regulation | Nature Communications

Nature Communications volume 15, Article number: 6768 (2024) Cite this article

3498 Accesses

8 Altmetric

Metrics details

Light is fundamental for biological life, with most mammals possessing light-sensing photoreceptors in various organs. Opsin3 is highly expressed in adipose tissue which has extensive communication with other organs, particularly with the brain through the sympathetic nervous system (SNS). Our study reveals a new light-triggered crosstalk between adipose tissue and the hypothalamus. Direct blue-light exposure to subcutaneous white fat improves high-fat diet-induced metabolic abnormalities in an Opsin3-dependent manner. Metabolomic analysis shows that blue light increases circulating levels of histidine, which activates histaminergic neurons in the hypothalamus and stimulates brown adipose tissue (BAT) via SNS. Blocking central actions of histidine and denervating peripheral BAT blunts the effects of blue light. Human white adipocytes respond to direct blue light stimulation in a cell-autonomous manner, highlighting the translational relevance of this pathway. Together, these data demonstrate a light-responsive metabolic circuit involving adipose-hypothalamus communication, offering a potential strategy to alleviate obesity-induced metabolic abnormalities.

In recent decades, the prevalence of obesity and its related metabolic cardiovascular disorders has reached epidemic proportions worldwide1,2. Obesity is characterized by excessive accumulation of adipose mass, including ectopic lipid accumulation, which can systemically influence metabolism and inflammation. There are different shades of adipose tissue that possess different functions. White adipose tissue (WAT) is the primary site of energy storage, and brown and beige adipose tissue (BAT) is a specialized thermogenic organ3,4,5. Adipose tissue also functions as an endocrine organ that regulates energy metabolism via the secretion of bioactive molecules6. The interplays between genetic and epigenetic predisposition and environmental factors, including diet, physical activity, temperature and light, can influence energy metabolism7. An imbalance of these interactions leads to obesity. Of these environmental factors, light is involved in many critical physiological or biochemical processes of humans and animals, such as visual sensing and production of vitamin D. Opsins are light-sensitive proteins that activate G protein-coupled receptor (GPCR) signaling when light hits the chromophore linked to the opsin protein8. Recent studies, including our own and Nayak et al. have independently discovered a non-visual role for Opsin3 (Opn3) in regulating adipocyte function9,10. Light activates Opn3-GPCR signaling in brown adipocytes, regulating fuel metabolism and mitochondrial respiration, and in white adipocytes, blue light induces lipolysis during cold exposure. Mice with adipocyte-specific deletion of Opn3 or raised in an environment without the blue light wavelengths fail to maintain normal body temperature in response to cold challenge. These data suggest a metabolic role of Opn3 in adipocytes. However, the underlying mechanisms driving the metabolic effects of adipose Opn3 in response to long-term light exposure, as well as its clinical relevance, even in cell-based evidence, have remained elusive.

Adipose tissues are highly innervated and engage in crosstalk with other organs. There are several means of inter-organ communication between adipose tissues and the brain. One is mediated through the sensory innervation of adipose tissues, where sensory nerves relay information from adipose tissue to the central nervous system (CNS)11. CNS conveys the response via sympathetic nerves innervating to metabolic organs, such as WAT, BAT, liver, and muscle, to regulate systemic metabolism12. Opsin5 (Opn5) in the hypothalamus has been linked to BAT thermogenesis via the sympathetic nervous system (SNS), indicating a hypothalamic–BAT axis13. Additionally, leptin is a representative bioactive molecule released from adipose tissues; it travels to the brain and acts as a humoral feedback system to regulate CNS functions and peripheral adipose tissue functions14.

Histidine is an essential amino acid that plays important roles in various physiological processes15. Its metabolic pathways include the formation of histamine and carnosine, and it can be obtained from both dietary sources and endogenous protein breakdown. Excess histidine can be catabolized through several pathways, such as the transamination pathway. Studies have shown that intraperitoneal (i.p.) administration of L-histidine can increase histidine decarboxylase (Hdc) activity and histamine release in the hypothalamus16, leading to increased histaminergic neuronal activity and sympathetic inputs to BAT and WAT17,18. Thus, histidine may play a role in regulating energy expenditure and glucose metabolism.

In this study, we discovered that blue light could reduce lipid accumulation in subcutaneous WAT (scWAT), leading to improved obesity-induced metabolic abnormalities. This phenotype is dependent on Opn3 in scWAT. Blue light also alters lipid metabolism and glycerol release in both human and murine white adipocytes in vitro. Upon light activation, scWAT releases histidine to activate the histaminergic neurons in the hypothalamus, which then signals to activate BAT via the SNS. These findings uncover a light-induced crosstalk between adipose tissue and the hypothalamus that could potentially counteract impaired metabolism in obesity.

While Opn3 belongs to the opsin photoreceptor superfamily, it is also expressed in non-visual cell types19. We reanalyzed a publicly available murine gene expression database across different tissues and organs20. Among the various metabolic organs, we found Opn3 exhibited high expression levels in both WAT and BAT, comparable to those in the lens of male C57BL/6 J mice (Supplementary Fig. 1a). Other Opsin genes did not exhibit significant expression in adipose tissues. Within the mouse WAT, Opn3 gene expression was predominantly localized in the adipocytes, although also present to a much lesser extent in other cell types, such as macrophages and adipose stem and progenitor cells (Supplementary Fig. 1b–c, data source from a publicly available single-nuclei RNA sequencing dataset21). This adipocyte-specific pattern distinguished Opn3 from other opsin gene families (Supplementary Fig. 1d). Furthermore, levels of Opn3 gene expression were significantly lower in the adipocyte fraction of mice fed with a high-fat (HF) diet compared to those in mice fed a standard chow diet (Supplementary Fig. 1e). These intriguing observations suggest a potential connection between Opn3 and the metabolic regulation associated with adipocytes in mice.

To directly illuminate mouse scWAT, we utilized a battery-free and implantable micro light emitting diode (μLED) device and a wireless closed-loop system, as previously described (Supplementary Fig. 2a)10,22,23. Based on the peak absorbance of vertebrate Opn3 at 470 nm24, we used blue μLED as the experimental wavelength and red μLED (630–650 nm) as the control wavelength (Supplementary Fig. 2b), because mice are naturally insensitive to red light due to the absence of long wavelength-sensitive Opn125,26. We chose the Illumination setting at 20% duty cycle (pulse frequency: 20 Hz, pulse duration: 10 msec, power: 10 W) because it allowed the maximal light transmission through the fat tissue without substantially affecting the device temperature and the fat tissue (Supplementary Fig. 2c–i). Under this setup, the light intensity emitted from the blue light device was measured at approximately 0.1 watts per square meter. Furthermore, using optical sensors to measure light intensity, we found lights with varying wavelengths could differentially penetrate a murine skin (Supplementary Fig. 2j–m) or a human finger (Supplementary Fig. 2n–p).

We first examined the impact of the duration of blue-light illumination (0, 4, or 8 days blue light) on metabolic phenotype in 12-15-week-old C57BL/6 J male mice implanted with the same blue μLED devices (Supplementary Fig. 3a). Wireless photonic devices were implanted on the left side of the scWAT of all mice. After the one-week recovery following implantation, light-treated groups received a discontinuous 24 hours of lighting pulse (20 Hz, 20%, 10 W). All mice were fed a HF diet (60% kilocalories from fat) for 15 days after surgery to induce metabolic abnormalities. Both 4 and 8 days of direct blue-light exposure reduced HF diet-induced body weight (BW) gain (Supplementary Fig. 3b, c). However, 8 days, but not 4 days, of blue light also reduced the levels of plasma leptin and insulin, which were known to correspond with adiposity (Supplementary Fig. 3d, e). Therefore, we employed 8 days of blue light treatment to scWAT in all further studies.

Next, we assessed the metabolic phenotype in C57BL/6 J male mice receiving 8 days of blue or red light or no light (BLUE light group, blue μLED implanted and light on; RED light group, red μLED implanted and light on; NO light group, blue μLED implanted and light off) (Fig. 1a). After 8 days of light treatment, the BLUE light group gained significantly less BW than the other two groups without differences in food intake (Fig. 1b, c, and Supplementary Fig. 4a). The reduction of BW was primarily due to a decrease in the relative tissue weight of perigonadal WAT (pgWAT) and scWAT, whereas the relative tissue weight of BAT, liver, and quadriceps muscle was not altered (Fig. 1d). Importantly, the relative tissue weight of the treated scWAT was significantly lower than that of the untreated site only in the BLUE light group (Fig. 1e). Compared to the RED light and NO light groups, the blue light-exposed mice showed significantly higher maximal thermogenic capacity as measured by heat production and VO2 consumption in response to norepinephrine (NE) stimulation (Fig. 1f, g and Supplementary Fig. 4b, c). Mice treated with 8 days of blue light exhibited better glucose tolerance (Fig. 1h, i) and lower fasting insulin level and Homeostatic Model Assessment for Insulin Resistance (HOMA-IR) scores (Fig. 1j, k). In addition, the BLUE light group had significantly lower circulating leptin and free fatty acid (FA) levels than the other groups (Fig. 1l and Supplementary Fig. 4d).

a–l Studies in 12-15-week-old C57BL/6 J male mice. a Schematic showing µLED device implantation in mouse scWAT (NO, RED, and BLUE light groups). b % BW change over 8 days (NO: n = 10, RED: n = 8, BLUE: n = 7). c Average food intake (NO: n = 7, RED: n = 4, BLUE: n = 4). d Tissue weight relative to BW (NO: n = 10, RED: n = 6, BLUE: n = 7). e Treated and untreated scWAT weight relative to BW (NO: n = 10, RED: n = 6, BLUE: n = 7). f Change in energy expenditure (ΔHeat) after norepinephrine (NE) injection (NO: n = 8, RED: n = 6, BLUE: n = 5). *P < 0.05, vs NO light group. g AUC quantification of ΔHeat (NO: n = 8, RED: n = 6, BLUE: n = 5). h Glucose levels during intraperitoneal glucose tolerance test (IPGTT) (NO, RED, BLUE: n = 6). *P < 0.05, vs NO light group, #P < 0.05, vs RED light group. (i) AUC of IPGTT (NO, RED, BLUE: n = 6). j–l Fasting plasma insulin (j) (NO: n = 8, RED: n = 6, BLUE: n = 5), HOMA-IR (k) (NO: n = 7, RED: n = 5, BLUE: n = 5), and plasma leptin levels (l) (NO: n = 9, RED: n = 7, BLUE: n = 6). m–u Studies in 20-25-week-old C57BL/6 J male DIO mice. (m) % BW change over 8 days (NO: n = 6, RED: n = 5, BLUE: n = 6). (n) Average food intake (NO, RED, BLUE: n = 5). o % changes in fat and lean mass over 8 days (DEXA scan) (NO: n = 6, RED: n = 5, BLUE: n = 6). p Treated and untreated scWAT weight relative to BW (NO: n = 6, RED: n = 5, BLUE: n = 5). (q) ΔHeat after NE injection (NO: n = 5, RED: n = 4, BLUE: n = 5). *P < 0.05 vs NO light group. r AUC quantification of ΔHeat (NO: n = 5, RED: n = 4, BLUE: n = 5). s Glucose levels during IPGTT (NO, RED, BLUE: n = 5). t AUC of IPGTT (NO, RED, BLUE: n = 5). u Fasting plasma insulin level (NO: n = 4, RED: n = 4, BLUE: n = 5). Statistics were performed by two-tailed paired Student’s t-tests (e and p), one-way ANOVA followed by Tukey’s post hoc test (b–d, g, i–o, r, t, and u), and two-way ANOVA followed by Tukey’s post hoc test (f, h, q, and s). n.s. indicates no significant difference. Data are represented as mean ± SEM. The diagram in a was created with BioRender.com, released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.

To investigate whether the light-induced metabolic alterations can counteract obesity-induced metabolic abnormalities, we fed mice with a HF diet for 16 weeks to introduce diet-induced obesity (DIO) and then implanted the μLED devices onto the left side of scWAT in 3 groups (BLUE, RED, and NO light group). Notably, 8 days of blue light treatment was sufficient to suppress BW gain (Fig. 1m and Supplementary Fig. 4f) and reduce total fat mass, including abdomen fat and visceral fat, as detected by dual-energy X-ray absorptiometry (DEXA) scan, without differences in food intake (Fig. 1n, o). The weight of the blue light-treated scWAT was significantly lower than the untreated site (Fig. 1p). No changes were observed with red light treatment. Moreover, DIO mice treated with 8 days of blue light displayed higher NE-induced VO2 consumption and heat production (Fig. 1q, r, and Supplementary Fig. 4g,h), better glucose tolerance (Fig. 1s, t), and lower plasma insulin levels (Fig. 1u). These results suggest that direct blue light to scWAT for 8 days can counteract HF diet-induced adiposity and metabolic abnormalities.

To examine whether the metabolic impacts of 8 days of blue light treatment were sex-specific, the same light device implantation and metabolic assessments were performed on female mice fed on a HF diet for 17 days. Eight-days of blue light treatment did not lead to significant changes in BW gain, tissue weight, fasting insulin, and leptin levels (Supplementary Fig. 4i–o). However, similar to the male cohorts, female mice treated with 8 days of blue light exhibited greater heat production in response to NE (Supplementary Fig. 4p, q) and significant improvement in glucose tolerance (Supplementary Fig. 4r, s) compared to control mice. These results suggest that the light-induced metabolic benefits may be more robust in male mice than in females, although the effects on NE-triggered maximal thermogenic capacity and glucose tolerance were equally efficacious in both sexes.

To understand how direct blue light caused the changes in scWAT mass and function, we examined tissue morphology and gene expression. The adipocytes in the treated scWAT of the BLUE light group displayed smaller diameter and size distribution than those in the treated scWAT of NO light or RED light groups (Fig. 2a–c). No sign indicative of cell death was observed, suggesting that the blue light exposure did not harm the tissue. Consistent with adipocyte morphology, levels of TG, which corresponds with lipid accumulation, were also significantly lower in the blue-light-treated scWAT compared with the scWAT from the red-light-treated or no-light-treated group (Fig. 2d). Strikingly, 8 days of blue light treatment suppressed the expression of several genes involved in lipid metabolism in scWAT. These included lipogenic genes, such as MLX interacting protein-like (Mlxipl, encoding ChREBP protein), sterol regulatory element binding transcription factor 1 (Srebf1), ATP-citrate lyase (Acly), Acetyl-CoA carboxylase alpha (Acaca), fatty acid synthase (Fasn), and stearoyl coenzyme A desaturase 1 (Scd1), as well as lipolytic genes, including patatin-like phospholipase domain containing 2 (Pnpla2, encoding ATGL protein) and lipase, hormone sensitive type (Lipe, encoding HSL protein). Furthermore, FA transport protein 1 (Fatp1), Cd36 as well as glucose transporter 1 (Glut 1) and Glut 4 genes were downregulated in treated scWAT by 8 days of blue light treatment, suggesting the decrease of intracellular uptake of FA and glucose (Fig. 2e).

a H&E staining of treated scWAT in 12-15-week-old C57BL/6 J male mice exposed to different light wavelengths (no light, red light, and blue light) for 8 days. Scale bars, 50 µm. b and c Adipocytes’ size distribution (b) and the average of adipocytes’ diameter (c) of treated scWAT after 8 days of light exposure (n = 4 per group). d Triglyceride levels in treated scWAT exposed to different light wavelengths. (NO light: n = 6, RED light: n = 5, and BLUE light: n = 5). e and f Relative mRNA expression of indicated lipid/glucose metabolism genes (e) and beiging genes (f) in treated scWAT in mice exposed to different light wavelengths. (NO light: n = 13, RED light: n = 8, and BLUE light: n = 10). g A schematic panel depicting the experimental design for mRNA (qPCR) and glycerol measurement conducted in in vitro differentiated murine white adipocytes (3T3-L1 cells) stimulated with/without blue light exposure. h Relative mRNA expression of lipid/glucose metabolism genes in murine white adipocytes exposed to blue light for a prolonged period (8 days) (h) or for a short-term (1 or 4 hours) (i) relative to dark condition (n = 4 technical replicates per group, three biologically independent replicates per experiment). j Glycerol levels in the culture medium of murine white adipocytes exposed to a short-term blue light exposure (1 or 4 hours) relative to dark condition (n = 4 technical replicates per group, three biologically independent replicates per experiment). Statistics were performed by two-tailed unpaired Student’s t-tests (h–j) and one-way ANOVA followed by Tukey’s post hoc test (c–f). n.s. indicates no significant difference. Data are represented as mean ± SEM. The diagram in g was created with BioRender.com, released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.

Previously, Nayak et al. described that ceiling blue light can induce the formation of beige adipocytes in WAT in neonatal mice9. However, this phenomenon was not observed in the adult mice exposed to direct blue light illumination of scWAT for 8 days. We thoroughly examined multiple sections of scWAT exposed to our blue light setting but did not observe any adipocytes with multi-locular lipid droplets, which is the typical morphology of beige adipocytes (Fig. 2a). Additionally, the mRNA levels of browning genes, such as uncoupling protein 1 (Ucp1), cell-death-inducing DNA fragmentation factor-α-like effector A (Cidea), cytochrome c oxidase subunit 7a1 (Cox7a1), Cox8b and peroxisome proliferator-activated receptor gamma coactivator 1-alpha (Ppargc1a) were not altered in the scWAT of mice treated with 8 days of blue light (Fig. 2f).

Moreover, to understand how direct blue light caused changes in scWAT mass between the treated and untreated side of the same mouse, we also compared tissue morphology and gene expression between blue-light treated and untreated scWAT. Adipocytes’ diameter and level of TG on the treated side were relatively smaller and lower than those on the untreated side, respectively (Supplementary Fig. 5a–c). Consistent with the morphological changes, the expression of several lipid metabolism genes tended to be downregulated by blue light (Supplementary Fig. 5d).

Adipose tissue consists of multiple cell types. To determine whether blue light directly alters gene expression in adipocytes, we exposed 3T3-L1 adipocytes to blue light in vitro using a black-box system previously established in our lab (Fig. 2g)10. Consistent with our in vivo findings, the expression of several lipogenic genes (Mlxipl, Acaca, Fasn, and Scd1), and lipolytic genes (Pnpla2 and Lipe) were downregulated by 8 days of blue light treatment, while browning genes were not changed (Fig. 2h and Supplementary Fig. 6a). Moreover, in order to unravel whether the transcriptional changes in response to blue light are dependent on the duration of light exposure, we conducted assessments of the effects of acute blue light exposures on 3T3-L1 adipocytes. Consistent with the previous report by Nayak et al.9 that short-term blue light treatment enhances lipolysis, we found that 1-hour blue light exposure led to an increase in the expression of lipolytic genes (Pnpla2, Lipe, and lipoprotein lipase (Lpl)) (Fig. 2i). However, when exposed to blue light for 4 hours, a clear reduction in the expression levels of lipolytic genes, along with several lipogenic genes (Srebf1, Acly, Acaca, Fasn, and Scd1) and mitochondrial genes (Cox7a1 and Cox8b), was evident (Fig. 2i and Supplementary Fig. 6b). This extended exposure also resulted in a significant decrease of glycerol levels in the culture medium of white adipocytes, a recognized marker of lipolytic activity (Fig. 2j).

For translational relevance, we employed the same methodology to expose human white adipocytes to blue light (Supplementary Fig. 6c). Intriguingly, a 2-hour exposure to blue light elicited an increase in the expression of PPARGC1A, CPT1B, ACLY, and LIPE gene and elevated glycerol levels (Supplementary Fig. 6d–f). Conversely, a 4-hour exposure to blue light led to a decrease in the expression of several lipogenic genes (ACLY, ACACA, FASN, and SCD), genes encoding glucose transporters (SLC2A1 and SLC2A4), as well as the lipolytic LIPE gene, resulting in a reduction in glycerol levels (Supplementary Fig. 6d–f). Furthermore, an extended 8-day exposure to blue light suppressed the expression of both lipolytic and lipogenic genes, such as LIPE and ACACA, and the glucose transporter gene SLC2A4, without affecting adipogenesis (Supplementary Fig. 6g–k). These findings demonstrate that both human and murine white adipocytes respond to direct blue light stimulation in a cell-autonomous manner, resulting in changes in cellular lipid and glucose metabolism.

Direct blue light exposure did not induce beiging in scWAT (Fig. 2a, f), yet the mice receiving 8 days of blue light illumination of scWAT displayed increased heat production and improved metabolism (Figs. 1f–l and 1q–u). To interrogate which organs contribute to the upsurge of heat production in the BLUE light group, we examined the morphology and the expression of thermogenic genes in BAT. Upon HF feeding, BAT from the RED light or NO light group showed large lipid droplets, while BAT from the BLUE light group remained a typical multi-locular phenotype (Fig. 3a) and contained lower TG levels (Fig. 3b) relative to the other groups. These data were supported by the increased expression of lipolytic genes (Pnpla2, Lipe, and Lpl) as well as FA transporter genes (Cd36 and Fatp1) and the decreased expression of lipogenic genes (Acly, Fasn, and Scd1) expression in BAT (Fig. 3c). In addition, mRNA levels of genes involved in thermogenic metabolism, such as Ucp1, PR domain containing 16 (Prdm16), and elongation of very long chain fatty acids protein 3 (Elovl3), were increased (Fig. 3d). Importantly, the NE levels in BAT and circulation were significantly higher in the BLUE light group compared to the control groups (Fig. 3e, f), but this was not observed in treated or untreated scWAT (Supplementary Fig. 4e), suggesting no SNS-driven NE increase in scWAT. Furthermore, there was no change in blood pressure or pulse rate in mice receiving 8 days of blue light illumination compared to the NO light group (Fig. 3g, h). These results suggest that blue light reduces adipocyte size in scWAT primarily through decreasing lipid synthesis rather than altering SNS activity, but it activates heat production and fuel utilization in BAT via increased sympathetic tone.

a H&E staining of BAT in 12-15-week-old C57BL/6 J male mice exposed to scWAT by different light wavelengths (no light, red light, and blue light) for 8 days. Scale bars, 50 µm. b–d Triglyceride levels (b) (NO light: n = 7, RED light: n = 8, and BLUE light: n = 4), and relative mRNA expression of indicated lipid/glucose metabolism genes (c) and thermogenesis genes (d) in BAT (NO light: n = 6, RED light: n = 6, and BLUE light: n = 6) in mice exposed to scWAT by different light wavelengths. (e and f) NE levels in BAT (NO light: n = 7, RED light: n = 7, and BLUE light: n = 6) (e), and circulation (NO light: n = 7, RED light: n = 7, and BLUE light: n = 7) (f). (g and h) Systolic and diastolic blood pressure (g) and pulse rate (h) before and after 8 days of light treatment (NO light: n = 6, and BLUE light: n = 5). Statistics were performed by two-tailed paired Student’s t-tests (g and h) and one-way ANOVA followed by Tukey’s post hoc test (b–f). n.s. indicates no significant difference. Data are represented as mean ± SEM.

Blue light can activate Opn324. Opn3-global knockout (Opn3-GKO) mice display normal physiology without obvious visual dysfunction10,27. We confirmed that scWAT dissected from adult Opn3-GKO mice lacked Opn3 protein expression (Supplementary Fig. 7a, b). To examine whether the metabolic effects of blue light depend on Opn3 in scWAT, 12-15-week-old Opn3-GKO and wild-type (WT) male mice were implanted with the blue μLED devices and split into two groups: the BLUE light group (blue μLED implanted and light on) and the NO light group (blue μLED implanted and light off) and received the same illuminating exposure as described above (Fig. 4a). In male Opn3-GKO mice, 8 days of blue light treatment did not alter the metabolic changes observed in the WT mice, including reduction of BW gain and fat mass as well as improved glucose tolerance (Fig. 4b–f, and Supplementary Fig. 7c). Circulating insulin and leptin levels were not altered by blue light exposure in Opn3-GKO mice (Supplementary Fig. 7d, e). Furthermore, the increases in NE-stimulated maximal thermogenic capacity (Fig. 1f, g) and NE levels in both plasma and BAT (Fig. 3e, f), as well as the increase in thermogenic and lipolytic genes transcription in BAT by blue light (Fig. 3c, d), were not observed in Opn3-GKO mice (Supplementary Fig. 7f–m). Additionally, the blue light-induced alterations in genes involved in lipid metabolism and reduced TG levels in the treated scWAT were not detected in Opn3-GKO mice (Fig. 4g and Supplementary Fig. 7n). The metabolic effects of blue light were also absent in female Opn3-GKO mice (Supplementary Fig. 8a–m). These results demonstrate that the metabolic effects of blue light are mainly through an Opn3-dependent mechanism in scWAT in both sexes.

Studies in 12-15-week-old Opn3-global knockout (GKO) and WT male mice. a Schematic showing blue µLED device implantation in mouse scWAT (NO light and BLUE light groups). b % BW change over 8 days (NO light in WT: n = 6, BLUE light in WT: n = 5, NO light in Opn3-GKO: n = 5, and BLUE light in Opn3-GKO: n = 5). c Average food intake (NO light in WT: n = 6, BLUE light in WT: n = 5, NO light in Opn3-GKO: n = 5, and BLUE light in Opn3-GKO: n = 5). d Treated and untreated scWAT weight relative to BW (NO light in WT: n = 6, BLUE light in WT: n = 5, NO light in Opn3-GKO: n = 5, and BLUE light in Opn3-GKO: n = 5). e Glucose levels during intraperitoneal glucose tolerance test (IPGTT) (NO light in WT: n = 3, BLUE light in WT: n = 3, NO light in Opn3-GKO: n = 5, and BLUE light in Opn3-GKO: n = 5). f AUC of IPGTT (NO light in WT: n = 3, BLUE light in WT: n = 3, NO light in Opn3-GKO: n = 5, and BLUE light in Opn3-GKO: n = 5). g Relative mRNA expression of lipid/glucose metabolism genes (NO light in WT: n = 3, BLUE light in WT: n = 3, NO light in Opn3-GKO: n = 5, and BLUE light in Opn3-GKO: n = 5) in treated scWAT of Opn3-GKO and WT mice. Statistics were performed by two-tailed paired Student’s t-tests (d), one-way ANOVA followed by Tukey’s post hoc test (b, c, f, and g), and two-way ANOVA followed by Tukey’s post hoc test (e). n.s. indicates no significant difference. Data are represented as mean ± SEM. The diagram in a was created with BioRender.com, released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.

Both BAT and WAT release bioactive molecules that can modulate the function of other organs5. We hypothesized that blue light triggered scWAT to release circulating factors that could convey metabolic benefits via the activation of BAT. To test this hypothesis, we performed metabolomics analysis to assess the changes of circulating metabolites in the BLUE light group compared to the RED and NO light groups. The Principal Components Analysis (PCA) plot showed distinct clusters for the BLUE, RED, and NO light groups (Fig. 5a). Metabolite set enrichment analysis revealed that many metabolic pathways were altered in the BLUE group compared to the NO light group or the RED light groups (Fig. 5b, c). Interestingly, eight out of nine overlapping pathways in both comparisons (BLUE vs. NO or BLUE vs. RED) with a p-value cutoff of less than 5% were associated with amino acid metabolism, including histidine metabolism. The Variable Importance in Projection (VIP) is a weighted sum of squares of the Partial Least-Squares loadings, taking into account the amount of explained Y-variation in each dimension. Histidine was a metabolite with the highest VIP score in circulation (F: 8.59, P-value: 0.0029, -log p10: 2.53, FDR: 0.18) by one-way analysis of variance (ANOVA) and post hoc analysis (Fig. 5d). Indeed, the relative value of circulating histidine in mice treated with blue light was significantly higher in the pathway involved in histidine metabolism (Fig. 5e). Due to a short half-life, plasma histamine was not detected. Carnosine, an endogenous dipeptide consisting of alanine and histidine, is the other metabolite derived from L-histidine via Carnosine synthase 1 (Carns1). However, no differences in circulating carnosine were found. Furthermore, through the use of enzyme-linked immunosorbent assay (ELISA), we validated the increase in circulating histidine levels triggered by blue light in both male and female mice (Fig. 5f and Supplementary Fig. 4t). Notably, the elevated levels of circulating histidine induced by blue light were diminished in Opn3-GKO for both sexes (Fig. 5g and Supplementary Fig. 8n), indicating the Opn3-dependency of this process.

a–d Principal components analysis (PCA) plot (a), metabolite set enrichment analysis (MSEA) of altered pathways in circulation in blue lighted groups compared with non-lightened group (b) or red lightened group (c), and variable importance in projection (VIP) (d) from metabolomics analysis of plasma from 12-15-week-old C57BL/6 J male mice exposed to scWAT by different light wavelengths for 8 days (NO light: n = 7, RED light: n = 6, and BLUE light: n = 6). The colored boxes on the right (d) indicate the relative concentrations of the corresponding metabolite in each group under study. e The relative abundance of circulating histidine and carnosine within the histidine metabolism pathway, employing metabolomics analysis (NO light: n = 7, RED light: n = 6, and BLUE light: n = 6). (f and g) Circulating histidine levels in C57BL/6 J male mice (NO light: n = 6, RED light: n = 6, and BLUE light: n = 6) (f) and Opn3-GKO male mice (NO light: n = 5, and BLUE light: n = 4) (g) with/without light treatment for 8 days, measured by ELISA. h and i Clustered heatmap from metabolomics analysis (h) and the relative abundance of histidine and carnosine (i) in treated scWAT with 8 days of blue light exposure relative to no light, employing metabolomics analysis (NO light: n = 6, and BLUE light: n = 5). (j) A positive correlation between the relative abundance of circulating histidine and the histidine in treated scWAT (n = 11). k–m Relative mRNA expression of histidine metabolism genes, Hdc and Carns1, in treated scWAT (n = 8 per group) (k), murine white adipocytes (n = 4 per condition, three biological replicates) (l), and human white adipocytes (n = 3 per condition, three biological replicates) (m) exposed to 8 days of blue light relative to dark condition. Statistics were performed by MSEA (b and c), VIP score (d) using MetaboAnalyst, two-tailed unpaired Student’s t-tests (g, i, and k–m), one-way ANOVA followed by Tukey’s post hoc test (e and f), and Spearman’s Rank correlation test (two-tailed) (j). n.s. indicates no significant difference. n.d. indicates no determined. Data are represented as mean ± SEM.

To investigate whether blue light directly alters histidine metabolism in scWAT, we analyzed metabolic profiles in the treated scWAT between the BLUE and NO light groups. Using hierarchical clustering analysis, we found that of the top 50 metabolites significantly altered by blue light exposure, 19 (38%) were involved in amino acid metabolism (Fig. 5h). Importantly, the relative levels of histidine in blue light-treated scWAT were significantly higher than those in no light-treated scWAT (Fig. 5i). Furthermore, a positive correlation was observed between circulating histidine levels and the relative abundance of histidine in treated scWAT (Fig. 5j). These findings imply that blue light may induce the release of histidine from scWAT. Histidine is converted into histamine by the enzyme HDC. It can also be converted to carnosine by the enzyme Carns1. Interestingly, 8 days of blue light exposure to scWAT suppressed the mRNA expression of Hdc and Carns1 in the treated scWAT (Fig. 5k). These results were further validated in vitro differentiated murine and human white adipocytes, where 8 days of blue light also suppressed Hdc gene expression (Fig. 5l, m), but did not alter Carns1 gene expression. Together, these data suggest that blue light directly triggers the biosynthesis and release of histidine from scWAT in an Opn3-dependent manner.

Previous studies have shown that i.p. administration of L-histidine leads to increased expression of HDC, as well as histamine release and activity in the hypothalamus, resulting in elevated histaminergic neuronal activity and sympathetic inputs to BAT and WAT16,17,18. Administration of alpha-Fluoromethylhistidine (FMH), an irreversible inhibitor of HDC28, can block the stimulatory effect of histidine on sympathetic activation of BAT (summarized in Fig. 6a)17. HDC-knockout (HDC-KO) mice do not show an increase in histamine action upon histidine injection29. Based on these findings, we hypothesize that the elevated circulating L-histidine activates BAT via the hypothalamic-SNS pathway, which contributes to the improved metabolic phenotypes observed in the blue-light exposed mice.

a Schematic of histidine metabolism pathway showing circulating histidine and FMH (HDC antagonist) mediates histaminergic neurons in the hypothalamus and regulates BAT thermogenesis via sympathetic nervous system (SNS). b Histidine decarboxylase (HDC) activity, normalized to total protein (right panel) or tissue weight (left panel), in the isolated brain tissues containing the hypothalamus in mice treated with/without 8 days of blue light exposure plus PBS or FMH injection. (NO light + PBS: n = 6, BLUE light + PBS: n = 3, and BLUE light + FMH: n = 4). c Immunostaining of histaminergic neurons in the hypothalamus in mice treated with/without 8 days of blue light exposure plus PBS or FMH injection. Sections at 1.7 mm and 2.7 mm posterior to the bregma were stained for mouse HDC (red). DAPI (4′,6-diamidino-2-phenylindole) was used to visualize nuclei (blue). Scale bar: 200 μm (two left columns), 20 μm (three right columns). d Relative tyrosine hydroxylase (Th) mRNA expression in BAT in mice treated with non- or 8 days of blue light exposure to scWAT plus PBS or FMH injection (NO light + PBS: n = 5, BLUE light + PBS: n = 6, and BLUE light + FMH: n = 6). e Th immunostaining (green) in BAT sections from mice treated with/without 8 days blue lighting plus PBS or FMH injection. DAPI (4′,6-diamidino-2-phenylindole) was used to visualize nuclei (blue). Scale bar: 50 μm. f Quantification of percentage Th density in BAT sections (NO light + PBS: n = 5, BLUE light + PBS: n = 5, and BLUE light + FMH: n = 3). Statistics were performed by one-way ANOVA followed by Tukey’s post hoc test. Data are represented as mean ± SEM.

To test this hypothesis, we first assessed changes in the expression of genes involved in thermogenesis and lipid metabolism in the BAT of C57BL/6 J mice receiving i.p. injection of various dosages of L-histidine (0.00175, 0.0175, or 0.035 μmol/g BW). Peripheral administration of L-histidine at 0.035 μmol/g BW significantly increased lipolytic (Pnpla2 and Lipe) and thermogenic (Ucp1 and Cidea) gene transcription in a dose-dependent manner at 1 hour (Supplementary Fig. 9a, b). These data were consistent with previous studies showing that BAT sympathetic nervous activity gradually increased and reached a peak around 1 hour after i.p. injection of L-histidine in vivo17. Additionally, we detected elevated levels of circulating histidine 1 hour after an i.p. injection of L-histidine by ELISA (Supplementary Fig. 9c). Notably, the circulating levels observed in response to L-histidine injection at the dosage of 0.035 μmol/g BW closely resembled the levels induced by 8 days of blue light treatment (Fig. 5f).

Next, we treated C57BL/6 J male mice with L-histidine (0.00175, 0.0175, or 0.035 μmol/g BW) i.p. daily for 8 days to assess the effects of L-histidine on metabolic phenotypes. Despite previous reports noting a significant drop in 24-hour food intake in rats given i.p. injections of L-histidine at 0.35 and 0.70 μmol/g BW16, our study using a lower dose found no notable changes in feeding behavior or food intake (Supplementary Fig. 9d). At 0.035 μmol/g BW, L-histidine effectively suppressed BW gain without affecting food intake (Supplementary Fig. 9e, f). There was a dose-dependent reduction of abdominal, visceral, and total fat detected by DEXA (Supplementary Fig. 9g). Importantly, simultaneous injection of the HDC inhibitor FMH completely blunted the L-histidine-induced reduction of BW and body fat (Supplementary Fig. 9h–k). In murine brown adipocytes differentiated in vitro, various dosages of L-histidine treatment (100 nM, 1 μM, or 10 μM) had no effect on the expression of genes involved in thermogenesis and FA β-oxidation (Supplementary Fig. 9l, m), suggesting that L-histidine does not directly activate brown adipocytes. Therefore, the effects of L-histidine on body weight and adiposity were likely via its conversion into neuronal histamine in the hypothalamus, which then induced sympathetic nervous innervation to activate BAT.

Since FMH can block central HDC action to inhibit the activity of histamine neurons, we hypothesized that blue light induced-metabolic phenotype is dependent on the histamine-responsive neuron in the hypothalamus (Fig. 6a). To test this hypothesis, we measured HDC protein levels in mice implanted with blue μLED devices with or without a light on, as an assessment of histamine-responsive neurons in the hypothalamus. Those mice were then treated with i.p. administration of PBS or HDC antagonist (FMH). There were 3 groups of animals: BLUE light + PBS group (blue μLED implanted and PBS injected), BLUE light + FMH group (blue μLED implanted and FMH injected), and NO light + PBS group (blue μLED implanted, light off and PBS injected). After normalizing to protein and tissue weight, we found that the levels of HDC in the brain, including the hypothalamus, were significantly increased by 8 days of blue light treatment, which were decreased by simultaneous administration of FMH (Fig. 6b). Considering the location and distribution of HDC-responsive neurons in the posterior hypothalamus, we proceeded with continuous sectioning of the hypothalamus from 2.80 mm to 0.80 mm posterior to the bregma. In sections from comparable locations (at 1.7 mm and 2.7 mm posterior to the bregma), the number of HDC-responsive neurons in the hypothalamus was elevated by 8 days of blue light treatment to scWAT and this increase was diminished by simultaneous administration of FMH (Fig. 6c). Tyrosine hydroxylase (Th) is a rate-limiting enzyme in catecholamine synthesis and a key component of both central and peripheral SNS. Th mRNA was highly increased in the BAT by 8 days of blue light exposure to scWAT, while daily injection of FMH abolished such induction (Fig. 6d). Immunofluorescent staining showed marked increases of Th-positive nerve fibers in the BAT of mice with 8 days of blue light treatment to scWAT, and those were almost completely gone by FMH treatment (Fig. 6e, f).

Taken together, these data suggest that increased circulatory histidine by blue light treatment to scWAT increased the number of histamine-synthesizing (responsive) neurons in the hypothalamus, which then increase both Th mRNA and protein corresponding to sympathetic tone in BAT. HDC antagonist blunted the blue light-induced histamine action and activation of BAT via SNS.

To address whether the central histidine-histaminergic neuronal pathway contributes to the changes in energy metabolism observed in blue-light-exposed animals, we assessed metabolic phenotypes in blue-light- or no-light-exposed mice receiving PBS or FMH administration. All the mice were implanted with the blue μLED device and the light exposure procedure was conducted as described above (Fig. 7a). Indeed, HDC antagonism with FMH blunted the reduction of fat accumulation and BW changes induced by blue light without altering food intake (Fig. 7b, c). The blue light-induced changes in BW were mainly caused by a reduction in tissue weight of pgWAT and scWAT (Fig. 7d). The relative tissue weight of the treated scWAT was significantly lower than that of the untreated site by blue light, but such differences were no longer observed when the mice received FMH treatment (Fig. 7e). The increases in NE-stimulated maximal thermogenic capacity (measured by heat production) by blue light were abolished by FMH injection (Fig. 7f). Similarly, the blue light-induced transcriptional changes in BAT, such as increased expression of thermogenic genes, Ucp1, Cidea, Cpt1b, and Prdm16 mRNA, as well as the expression of lipolytic genes, Lipe and Lpl mRNA, were also diminished by FMH treatment (Fig. 7g). In addition, 8 days of blue light treatment significantly increased the circulating NE levels, consistent with the previous cohort (Fig. 3f), and conversely, FMH injection decreased the light-induced NE levels (Fig. 7h). The remaining question was whether FMH altered circulating histidine levels. Thus, we performed metabolomics analysis in plasma samples from the blue light + PBS and the blue light + FMH group. We found no differences in circulating histidine levels between these two groups (Fig. 7i).

a Schematic model showing blue µLED implantation in mouse scWAT with/without light treatment plus PBS or FMH injection. b % BW change over 8-days (NO light + PBS: n = 6, NO light + FMH: n = 8, BLUE light + PBS: n = 5, and BLUE light + FMH: n = 6). c Average food intake (NO light + PBS: n = 4, and NO light + FMH: n = 7). d Tissue weight relative to BW (NO light + PBS: n = 6, NO light + FMH: n = 8, BLUE light + PBS: n = 4, and BLUE light + FMH: n = 6). e Treated and untreated scWAT weight relative to BW (NO light + PBS: n = 6, NO light + FMH: n = 8, BLUE light + PBS: n = 4, and BLUE light + FMH: n = 6). f Change of heat after norepinephrine (NE) injection (NO light + PBS: n = 7, NO light + FMH: n = 10, BLUE light + PBS: n = 7, and BLUE light + FMH: n = 9). *P < 0.05, vs NO light + PBS, #P < 0.05, vs NO light + FMH. (g) Relative mRNA expression of indicated thermogenic and lipolytic genes in BAT (NO light + PBS: n = 6, NO light + FMH: n = 7, BLUE light + PBS: n = 7, and BLUE light + FMH: n = 6). h Plasma NE levels (NO light + PBS: n = 6, NO light + FMH: n = 8, BLUE light + PBS: n = 7, and BLUE light + FMH: n = 6). i The relative abundance of circulating histidine and carnosine, employing metabolomics analysis. (BLUE light + PBS: n = 5, and BLUE light + FMH: n = 4). j Proposed model for the underlying mechanism of the light-responsive adipose-hypothalamus axis in metabolic regulation. Statistics were performed by two-tailed paired Student’s t-tests (e), two-tailed unpaired Student’s t-tests (c and i), one-way ANOVA followed by Tukey’s post hoc test (b, d, g and h), and two-way ANOVA followed by Tukey’s post hoc test (f). Data are represented as mean ± SEM. The diagrams in a and j were created with BioRender.com, released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.

Lastly, BAT denervation studies added further credence to these findings. As previously reported, surgical denervation of BAT impairs thermogenesis and reduces systemic energy expenditure, ultimately leading to an increase in fat mass30,31,32. After confirming the success of BAT denervation via Th immunostaining (Supplementary Fig. 10a–b), we assessed metabolic phenotype and thermogenic function in mice exposed to blue light or no light to scWAT, with denervation of BAT (Supplementary Fig. 10c). Surgical denervation of bilateral BAT lobes completely abolished the blue light-induced BAT activation, including enhancement in thermogenic capacity and elevated levels of thermogenic and lipolytic gene expression, resulting in no differences in changes of BW gain and glucose tolerance (Supplementary Fig. 10d–i). Of note, even with BAT denervation, blue light treatment directly reduces the treated scWAT mass compared to the intact contralateral scWAT (Supplementary Fig. 10j) and increases the circulating histidine levels (Supplementary Fig. 10k). These results demonstrate that the metabolic regulation in response to blue light is mainly mediated through direct SNS activation of BAT.

Collectively, these data demonstrate that administration with HDC antagonist did not alter circulating histidine, but decreased HDC activity in the hypothalamus, leading to reduced sympathetic tone and BAT activation, thereby diminishing the metabolic effects induced by blue light. Blocking the central actions of histidine and performing peripheral BAT denervation blunts the effects of blue light, revealing a light-responsive adipose-hypothalamus axis in metabolic regulation (Fig. 7j).

In mice, the Opn3 gene is expressed at notably higher levels in adipose tissues than in other metabolic organs, whereas other Opsin genes are barely detectable in adipose tissues. Within the adipose tissue, Opn3 is predominantly expressed in adipocytes. We and others have previously uncovered a non-visual role for light-responsive Opn3 in regulating mouse brown and white adipocytes9,10. In the present study, we provide evidence that extended blue light exposure to scWAT for 8 days through a wireless optogenetic device can alleviate high-fat diet-induced metabolic abnormalities in an Opn3-dependent manner. This is orchestrated via an adipose-hypothalamus axis involving histidine released from scWAT to activate BAT activity via the histaminergic neuronal pathway. Furthermore, our results demonstrate that prolonged blue-light exposure to scWAT in vivo and in vitro suppresses lipid synthesis and reduces TG levels, leading to smaller adipocyte distribution. Obesity is characterized by enlarged adipocytes with a reduced capacity for lipid storage, which leads to ectopic fat deposition and insulin resistance33,34. Thus, the blue light-induced decrease in adipocyte size could improve cellular function and metabolic health. Of note, blue light treatment increases circulating histidine levels and BAT activity via activating histaminergic neurons in the hypothalamus, which contributes to enhanced energy expenditure and suppression of HF diet-induced BW gain.

Recent studies have provided insights into the distinct effects of light with various wavelengths on metabolic regulation. These studies have identified two pathways involved in light sensing that connect the hypothalamus and BAT via the SNS13,35 The first pathway involves violet light activating Opn5, a deep brain photoreceptor expressed in glutamatergic warm-sensing preoptic area neurons, resulting in impaired thermogenic function in BAT13. The second pathway consists of a light-responsive neural axis that links photoreception in intrinsically photosensitive retinal ganglion cells in the retina to adaptive thermogenesis in BAT. This pathway involves a series of neural relays and has been shown to improve glucose tolerance in mice and humans35. Diverged from these findings, our study identifies a pathway where blue light stimulates Opn3 in scWAT, which releases histidine to activate histaminergic neurons in the hypothalamus, leading to BAT activation and improved systemic metabolism. These studies lay the groundwork for light-based therapies for obesity-associated metabolic disorders. Moreover, recent evidence has shown the safety and efficacy of light therapy in treating brain injuries in humans36, suggesting the potential for further translational research in this area.

Opsins are photoreceptive proteins attuned to specific wavelengths, with Opn3 and Opn4 uniquely responsive to blue light. Opn3 is the predominant opsin in mouse adipose tissue, whereas Opn4, known as melanopsin, is present in human scWAT37. Ondrusova et al. show that prolonged blue light exposure to in vitro differentiated white adipocytes led to reduced lipid accumulation and decreased lipid droplet size via an Opn4-dependent mechanism37. Opn4 acts as a Gq-protein-coupled receptor to stimulate phospholipase C, initiating the production of inositol triphosphate and diacylglycerol and engaging transient receptor potential canonical (TRPC) channels to enhance Ca2+ and Na+ influx. In contrast, Opn3 typically associates with Gi- or Go-protein, without an established link to TRPC channels. Using the Opn3 KO mice, we demonstrate that blue light-induced metabolic improvements in HF-fed mice are mediated through Opn3. However, the potential involvement of Opn4 in human scWAT when subjected to blue light suggests an area for future research.

In this study, using an illumination system for in vivo and in vitro experiments, we demonstrate a metabolic circuit involving light-mediated adipose-derived histidine that triggers central histaminergic actions and peripheral BAT activation through an adipose-hypothalamus axis. Opn3 binds to a chromophore molecule, 11-cis-retinal, which isomerizes to all-trans-retinal upon light absorption24,38. This isomerization induces a conformational change in the opsin protein, leading to the activation of intracellular signaling pathways that ultimately result in the organism’s response to light. Since the metabolic phenotypes associated with blue light exposure to scWAT were clearly diminished in the Opn3-GKO mice, it suggests that the alteration of the retinal alone may not be essential for metabolic regulation. These findings indicate that Opn3 is crucial for the downstream signaling and gene expression that lead to changes in cellular metabolism in response to blue light rather than simply mediating the effects of retinal isomerization.

The relationships between circulating histidine levels and metabolic health in humans have been closely studied. Obese and type 2 diabetes (T2D) patients have been found to have significantly lower plasma histidine levels than individuals with normal weight39. Conversely, higher circulating histidine levels have been associated with lower mortality risk in T2D patients40. Additionally, a study by Piro et al. reported that histidine levels in visceral fat were significantly lower in obese patients with metabolic syndrome compared to non-obese healthy individuals41. These studies suggest a negative association between obesity-associated pathologies and histidine levels in circulation and visceral fat.

Furthermore, Joanna Moro et al. conducted a systematic review to summarize the effects of daily supplemental histidine on physiological functions in humans and rodents42. The daily requirement for histidine in adult humans is 8-12 mg/kg of BW/day, and the average intake in typical adult diets worldwide is within the range of 30-35 mg/kg of BW/day43,44. Higher dietary histidine is inversely associated with energy intake, insulin resistance, inflammation, and oxidative stress in overweight and obese individuals, as well as a lower prevalence of overweight and obesity in northern Chinese adults45. However, histidine supplementation with a large dose exceeding 8 g/day has been found to induce several side effects, such as severe anorexia, weakness, drowsiness, nausea, and memory disorder42. In a 12-week double-blind randomized clinical trial involving non-diabetic participants who were overweight and obese, carnosine supplementation was found to lower fasting insulin and insulin resistance compared to placebo. Carnosine is an endogenous dipeptide consisting of alanine and histidine, and these findings suggest that it may be a potential strategy for the prevention of T2D46. In rodents, the maximal tolerable dose of histidine in the diet is considered to be 25 g/kg, as a higher dose of 50 g/kg of diet has been found to cause adverse effects, such as decreased body weight gain, food intake, and growth retardation. These findings suggest that histidine supplementation may hold health benefits, but careful consideration must be taken with dosing to avoid adverse events.

The mechanism by which circulating histidine regulates metabolism remains unexplained. Recent studies using cryo-electron microscopy have resolved the structure of heterodimeric transporter LAT1/4F2hc, also known as the L-type amino acid transporter (LAT1)47. LAT1, which is expressed at a relatively high level on both the luminal and abluminal sides of the brain capillary endothelial cells as well as on brain parenchymal cells at the blood-brain-barrier (BBB), plays a role in the transport of essential amino acids, such as histidine, across the cell membrane48. Thus, in our study, histidine released from scWAT in response to blue light, can enter the brain presumably through LAT1 at the BBB. In the brain, histidine is converted to histamine via the action of HDC, the key enzyme in histamine biosynthesis. HDC is expressed in histamine neurons in the hypothalamus49, particularly within the posterior hypothalamus, where five distinct cell types of histaminergic neurons are clustered and innervated throughout the brain50. The histaminergic neurons have been implicated in regulating sleep-wakefulness and arousal51, suppressing food intake via histamine H1 receptors in the hypothalamus52, and increasing energy expenditure by stimulating lipolysis in adipose tissue via SNS activation53.

Fulop et al. generated the HDC-KO mice, which are deficient in producing histamine, and confirmed the lack of Hdc mRNA, histamine immunoreactivity, and histaminergic innervation throughout the brain29. Aged (30-week-old) HDC-KO mice became obese, despite not being hyperphagic, and exhibited impaired glucose tolerance, hyperinsulinemia, and hyperleptinemia. They also displayed reduced Ucp1 mRNA expression in BAT, leading to impaired thermogenesis in response to cold. As an alternative way to impair HDC function, i.p. bolus FMH injection has been shown to block the histidine-induced increase of sympathetic activity in BAT of rats17. In agreement with this reference, we demonstrate that i.p. injection of FMH decreased the number of the histaminergic neurons and HDC activity in the brain, including the hypothalamus. Furthermore, in a low L-histidine diet-induced mouse model, insufficient intake of histidine reduces the brain histamine concentrations and triggers anxiety-like behaviors54. Taken together, the lack or reduction of HDC enzyme in the hypothalamus leads to attenuated BAT function, highlighting the role of histidine in the regulation of energy homeostasis.

The limitations of the current study include its human physiological relevance and direct translation to human application. Currently, there is scant evidence on whether exposure to natural sunlight can enhance circulating histidine levels through the light-opsin pathway in human scWAT. The light intensity emitted from the blue light device we utilized was substantially lower than estimated solar diffuse irradiances for blue light from the sunlight55,56. While blue light does not efficiently penetrate the skin, direct and prolonged exposure to sunlight may activate scWAT similar to our experimental setup, suggesting the physiological relevance of our findings within a photobiological context. Furthermore, the present study suggests potential applicability by demonstrating that artificial blue light can penetrate human skin, suggesting accessibility to Opn3 or Opn4 in human scWAT. Additionally, our in vitro experiments with differentiated human white adipocytes reveal a cell-autonomous response to direct blue light stimulation, emphasizing the translational relevance of this pathway. Interestingly, lean individuals tend to have higher circulating histidine levels than obese individuals39. The development of optogenetic device technologies, such as flexible multiregional optogenetic devices with micro-LEDs57 and wearable LED-based therapeutic devices58, highlights the feasibility of validating the efficacy and safety of light-based treatments for metabolic regulation in humans. The use of global Opn3-KO mice has limited us from distinguishing whether the observed effects are solely due to the absence of Opn3 in adipocytes. Thus, identifying the specific cell types within scWAT that facilitate histidine release when exposed to blue light necessitates additional research at the single-cell level. Further studies using high-resolution approaches, such as single-cell RNA-sequence or single-cell metabolomics, will shed light on the exact light-sensing cell type that produces histidine within the WAT.

In conclusion, these results uncover a light-responsive adipose-hypothalamus axis in an Opn3 and histidine-dependent manner and provide a molecular mechanism for developing light-based treatments for the growing obesity epidemic.

All animal experiments and care procedures were approved by the Institutional Animal Care and Use Committee at Joslin Diabetes Center. C57BL/6 J (Stock no. 000664) mice were purchased from the Jackson Laboratory. Opn3-GKO mice were generated in Dr. King-Wai Yau’s laboratory at Johns Hopkins University27 and characterized at the Joslin Diabetes Center. All mice were kept in a temperature- and humidity-controlled room (23 °C, 30%) on a 12 h light/dark cycle (lights on 06:30 am; off 06:30 pm) unless indicated with free access to food and water. C57BL/6 J and Opn3-GKO mice were fed chow diet (cat. no. 5020, LabDiet). For creating DIO mice, 6-week-old mice were fed a HF diet (60% kcal from fat, cat. no. D12492, Research Diets) for 14–19 weeks.

Before sacrifice, mice were fasted (the duration of fasting times: 4-6 h) by transferring mice to clean cages with no food or feces in hoppers or bottom of cages. Blood was obtained from each tail under anesthesia with inhalation of isoflurane (cat. no. NDC 66794-017-25, Piramal Critical Care) to determine fasting glucose concentrations and to calculate HOMA-IR (as described below) by using a blood glucose meter (Infinity, US Diagnostics). Blood was also collected by cardiac puncture, and subsequently, plasma was separated by centrifugation at 4 °C and stored at −80 °C until future analysis of insulin, leptin, TG, free FA, and norepinephrine, and metabolomics (see below). The pgWAT, scWAT, interscapular BAT, liver, and quadriceps muscle were dissected and weighed, then snap-frozen in liquid nitrogen and stored at −80 °C until further analysis.

For optogenetic stimulation to scWAT, prior to surgery mice were given a dose of analgesic (Banamine, 2.5 mg/kg BW, s.c. injection, cat. no. 07-859-1323, Patterson) to minimize and prevent postoperative pain and distress. Mice were fully anesthetized with inhalation of 2.5% isoflurane to incise the shaved skin right over the left scWAT around the lumber, and to place a single µLED device on the exposed left scWAT. The device (customized light device without indicator, NeuroLux, USA) was sutured to the muscle surrounding the left scWAT using 5-0 monofilament threads. After the device implantation, the open skin was sutured with 5-0 coated vicryl undyed braided threads. Following the surgery, mice were given access to a HF diet and water, ample bedding for nesting, and were monitored regularly for any signs of distress or illness. Mice were allowed 7 days of recovery before turning on the light via the wireless controller box (NeuroLux Optogenetics system, NeuroLux) (Supplementary Fig. 2a). The light was on for 24 h per day until the sacrifice. The light device is battery-free and was controlled and monitored by a wireless closed-loop system10. The information for this system, including the light devices and hardware, was obtained from NeuroLux (http://www.neurolux.org/). BW and food intake in each cage were monitored. Bedding in cages was changed every three days.

For measuring maximum thermogenic capacity, following the light treatment, mice were placed in the Comprehensive Lab Animal Monitoring System (CLAMS, Columbus Instruments) or the Sable Systems’ Promethion system cages to gather measurements of oxygen consumption rate (VO2), carbon dioxide production rate (VCO2), and energy expenditure (Heat) for 15 min until they are fully asleep by pentobarbital (65 mg/kg BW, i.p. injection, Virbac). Mice were subsequently injected with NE (1 mg/kg BW, s.c. injection, cat. no. A0937, Sigma-Aldrich) and monitored. Data for volume of VO2, VCO2, and Heat data were normalized to the basal level prior to NE injection (ΔVO2, ΔVCO2, and ΔHeat) and analyzed among groups.

For measurements of body composition, mice were anesthetized with 2.5% isoflurane and scanned by dual-energy X-ray absorptiometry (DEXA).

For i.p. glucose tolerance test (IPGTT), following the light treatment, mice were fasted for 6 h by transferring mice to clean cages without food. Mice had free access to drinking water. A baseline glucose level was determined by collecting blood from the tail of conscious mice before intraperitoneal glucose loading (2.0 g/kg BW except for 1.0 g/kg BW for DIO mice, i.p. injection). Subsequently, blood was collected from the tail 15, 30, 45, 60, and 120 min after injection. Glucose concentrations were determined using a blood glucose meter (US Diagnostics).

For L-Histidine injection in vivo studies, 12-15-week-old C57BL/6 J male mice were administered daily histidine (cat. no. 151688, Sigma-Aldrich), PBS or histidine plus FMH (cat. No. sc-220045, Santa Cruz Biotechnology) by i.p. injection. At 1 h or 8 days after injection, mice were sacrificed and tissues and plasma were collected.

For the BAT denervation study, mice were anesthetized with continuous inhalation of 2.5% isoflurane for induction and maintenance. When the bilateral BAT lobs were exposed, the intercostal nerve bundles innervating each BAT lobe were surgically cut59. Skin incisions were closed using 5-0 coated Vicryl undyed braided threads. Mice were monitored for any signs of distress or illness for a recovery period of 2 weeks.

Immortalized murine brown preadipocytes were cultured in growth media (DMEM high glucose media added of 10% FBS)60. The media was replaced every 2 days until the cells reached confluence. Brown adipocyte differentiation was induced by using induction media (DMEM supplemented with 10% FBS, 20 nM insulin, 1 nM triiodothyronine (T3), 0.125 mM indomethacin, 5 μM dexamethasone, and 500 μM isobutylmethylxanthine (IBMX) for 2 days. Induction media was replaced by differentiation medium (DMEM supplemented with 10% FBS, 20 nM insulin, and 1 nM T3) for 6 days, with medium being replaced every 2 days.

Mouse white preadipocytes (3T3-L1) were purchased from ATCC (cat #: CL173). To induce differentiation, white preadipocytes were allowed to reach confluency and treated with induction medium supplemented with 10% FBS, 20 nM insulin, 5 µM dexamethasone, and 0.5 mM IBMX for 2 days. Subsequently, cells were maintained in a differentiation medium supplemented with 10% FBS, and 20 nM insulin for an additional 6 days. During the differentiation process, the medium was changed every other day.

Human white preadipocytes were cultured in growth media comprising 10% FBS in DMEM/high glucose61,62. For adipocyte differentiation, cells were exposed to adipogenic induction media (DMEM-H with 10% FBS, 33 μM biotin, 0.5 μM human insulin, 17 μM pantothenate, 0.1 μM dexamethasone, 2 nM T3, 500 μM IBMX, and 30 μM indomethacin). Throughout the differentiation process, media were refreshed every other day.

Plasma insulin level was determined by Ultra-Sensitive Mouse Insulin ELISA kit (cat. no. 90080, Crystal Chem). Insulin resistance was assessed by calculating HOMA-IR ((fasting glucose × fasting insulin)/405). Plasma leptin, FA, and NE were determined by ELISA (Mouse/Rat Leptin Quantikine ELISA kit, cat. no. MOB00B, R&D systems; Free fatty acid assay kit, cat. no. ab65341, Abcam; and Norepinephrine ELISA kit, cat. no. MBS760375, MyBioSource), respectively.

Tissues (scWAT, pgWAT, and BAT) TG were extracted and quantified with a TG assay kit (cat. no. ab65336, Abcam) according to the manufacturer’s instructions and normalized to total protein concentration determined by BCA Protein Assay (cat. no. 23225; Thermo Fisher Scientific).

Isolated brain tissues containing the hypothalamus were minced and homogenized with PBS. Homogenates were centrifuged for 20 min at 17,000 x g at 4 °C. The supernatants were collected and quantified with a Mouse HDC ELISA kit (cat. no. MBS9330743, MyBioSource) according to the manufacturer’s instructions and normalized to total protein concentration.

Each cell culture dish was placed inside a black box that had blue LED lights (RTGS Products) inside of the box to illuminate the cells. The black box was kept in an incubator, and differentiated murine or human white adipocytes were cultured in the black box for 8 days. The culture medium was changed every other day. Steady-state photoluminescence spectra of LED were recorded on a JASCO FP-6500 (Jasco), and the spectra were corrected for detector nonlinearity as previously described10.

Immortalized murine brown preadipocytes were cultured in growth media (DMEM high glucose media added of 10% FBS)60. After differentiation, the cells were washed with PBS, and growth medium with or without L-Histidine (cat. no. 151688, Sigma-Aldrich) was added and incubated for 24 h at 37 °C in a humidified incubator with 5% CO2.

Prior to starting the experiment, a temperature data logger (SubCueTM)63 was programmed to automatically record the temperature at every 10 min. The information for this miniature temperature data logger and hardware was obtained from SubCue (http://www.subcue.com/). The temperature data logger was placed under the μLED device (Supplementary Fig. 2c). The temperature was recorded after non-lighting or lighting by either red or blue light for 12 h under 2 different illumination settings (pulse frequency: 20 Hz, power: 10 W, percent of illumination periods: 20% or 100%).

C57BL/6 J male mice were fully anesthetized by pentobarbital (50 mg/kg BW, i.p. injection). Prior to surgery, mice were given a dose of analgesic (Banamine, 2.5 mg/kg BW, s.c. injection) to minimize and prevent postoperative pain and distress. After a small incision on the shaved lumbar skin to expose the scWAT, the light device was placed on top of the scWAT (Supplementary Fig. 2h). The temperature of lighting fat tissue underlying the light device and non-lighting region was measured using a thermal probe connected to the thermocouple meter (TC-2000, Sable Systems International) during 1 hour of continuous blue and red lighting at 20% duty cycle (pulse frequency: 20 Hz, pulse duration: 10 msec, power: 10 W).

C57BL/6 J male mice were fully anesthetized by pentobarbital (50 mg/kg BW, i.p. injection). Prior to surgery, mice were given a dose of analgesic (Banamine, 2.5 mg/kg BW, s.c. injection) to minimize and prevent postoperative pain and distress. To insert an optical sensor (16 × 10 mm) under the lumber skin, a small incision and blunt dissection to scWAT were performed. An optical sensor connected to a power meter (LASER POWER METER LP1, SANWA electric instrument CO., LTD) was inserted under the skin. The light device was placed on the skin or shaved skin. The light intensity was measured by using a power meter (Supplementary Fig. 2j).

For measuring each photon number at the different wavelength, the AS7341 spectral sensor was connected to an Adafruit QT Py hosting an ATMEL SAMD21 Cortex M0 microprocessor programmed in CircuitPython to measure light in 5 wavelength channels (415, 445, 480, 590, and 680 nm) as described in Supplementary Fig. 2n. The information for the multi-channel spectrophotometer and the microprocessor unit were obtained from the following website: AS7341 eight-channel spectrophotometer, (https://www.adafruit.com/product/4698); Adafruit QT Py, (https://www.adafruit.com/product/4600); CircuitPython, (https://circuitpython.org/). The CircuitPython script can be downloaded from https://github.com/mdlynes/Minispectrophotometer. As described in Supplementary Fig. 2o, the photon number entered in each channel was recorded every 1 sec for 10 sec before and after blockade on the light sensor using the human finger-avoiding nails. Experiments were performed in ten replicates. Light transmission through a human finger was calculated as the following:

% of light transmission = P(a) / P(0) × 100

P(a): average of photon number for 10 sec under blockade on the light sensor.

P(0): average of photon number for 10 sec under no blockade on the light sensor.

Trizol reagent was used to extract total RNA from tissue and RNA was purified using a spin column kit (cat. no. R2052, Zymo Research). RNA (500 ng to 2 µg) was reverse-transcribed with a high-capacity complementary DNA reverse transcription kit (Applied Biosystems). Quantitative real-time PCR was performed using SYBR green PCR Master Mix (cat. no. A25778, Applied Biosystems) with 300 nM of each forward and reverse oligonucleotide primer in an ABI Prism 7900 sequence detection system (Applied Biosystems). Ribosomal protein lateral stalk subunit P0 (Rplp0) and 18 S rRNA expression were chosen as an internal standard in mouse and human samples, respectively. Real-time PCR primer sequences are listed in Supplementary Table 1.

ScWAT was lysed in T-PER Tissue Protein Extraction Reagent (cat. no. 78510, Thermo Fisher Scientific) supplemented with protease inhibitor cocktail (cat. no. P8340, Sigma-Aldrich). Homogenates were centrifuged for 20 min at 17,000 x g at 4 °C and the supernatants were collected. Protein concentration was determined using a Pierce BCA kit (cat. no. 23225, Life Technologies). For immunoblots, 10 µg proteins were loaded, and the following primary antibodies were used: anti-Opn3 (cat. no. ab75285, 1:1000) and anti-β-Tubulin (cat. no. 2146, 1:1000) were purchased from Abcam and Cell Signaling Technologies, respectively. Proteins were detected using SuperSignal™ West Femto Maximum Sensitivity Substrate (cat. no.34095, Thermo Fisher) and quantified using Image Lab Software (Bio-Rad). β-Tubulin was used as an endogenous control for normalization.

In vitro differentiated murine and human white adipocytes were serum-starved for 3 hours on the day of the experiment. All cells received fresh serum-free media before light stimulation. One set of cells was exposed to blue light, while another set was kept in darkness within the incubator. At the end of 1–4 hours of exposure to blue light or darkness, culture media were collected and immediately frozen on dry ice for storage at −80 °C until further use. Glycerol assays were conducted following the manufacturer’s instructions using the free glycerol assay kit (Abcam, ab65337).

Cells were washed twice with PBS and were subsequently fixed with 10% buffered formalin for 15 minutes at room temperature. Following fixation, cells were stained using a filtered Oil Red O working solution, composed of 3 parts of 0.5% Oil Red O in isopropanol and 2 parts of water, for a duration of 1 hour at room temperature. After staining, cells were subjected to multiple washes with distilled water and then visualized.

ScWAT, pgWAT, and BAT were freshly collected and immediately fixed overnight in 10% neutral buffered formalin. Tissue samples were dehydrated using gradient ethanol and embedded in paraffin. The sections were stained with H&E. Measurement of cell size was performed using Image J and Cell Profiler 3.0.

BAT was fixed in 10% formalin and paraffin-embedded. Multiple 5 μm sections were prepared and stained with Th antibody. Briefly, sections were deparaffinized and rehydrated, followed by an antigen retrieval step in a modified citrate buffer (Dako Target Retrieval Solution, pH 6.1, Agilent). Sections were then incubated in Sudan Black (0.3% in 70% ethanol) to reduce the autofluorescence signal. This was followed by blocking in Millipore blocking reagent (EMD Millipore) and then incubating with rabbit anti-Th antibody (cat. no. AB152, EMD Millipore, 1:50) overnight at 4 °C. The next day, slides were washed in PBS and were incubated with goat anti-rabbit immunoglobulin G (IgG) (H + L) secondary antibody conjugated with Alexa Fluor 488 (cat. no. A-11008, Invitrogen, 1:200). Slides were mounted in mounting medium with DAPI (4′,6-diamidino-2-phenylindole) (cat. no. D9564, Sigma-Aldrich). Images were collected on a Zeiss LSM 710 NLO confocal microscope and processed with Image J.

Mice were anesthetized with pentobarbital sodium (100 mg/kg BW, i.p. injection) and perfused with saline followed by 4% paraformaldehyde (cat. no. 15714, Electron microscopy Sciences). The brain was meticulously isolated and excised the region containing the hypothalamus in a sagittal manner. Subsequently, it was post-fixed in 4% paraformaldehyde (PFA) overnight at 4 °C and cryoprotected using a 30% sucrose solution. Using the cryostat (Microm HM550), a series of 40 continuous sections was then obtained every 50 µm, covering a width of 2 mm (from 2.80 mm to 0.80 mm posterior to the bregma) from the excised brain tissues. Free-floating sections (50 µm) for immunofluorescent staining were treated with an antigen retrieval step in a modified citrate buffer (Dako Target Retrieval Solution, pH 6.1, Agilent). Blocking was performed in Millipore blocking reagent (EMD Millipore), followed by incubating the section in rabbit anti-HDC antibody (cat. no. EUD2601, ORIGENE) in a dark humid chamber overnight at 4 °C. The next day, slices were washed with PBS and were incubated with goat anti-rabbit immunoglobulin G (IgG) (H + L) secondary antibody conjugated with Alexa Fluor 594 (cat. no. A-11037, Invitrogen, 1:200). Sections were mounted on the slides using the mounting medium with DAPI. Images were collected on a Zeiss LSM 710 NLO confocal microscope and processed with Image J and ZEN software.

Plasma and tissue samples for metabolomics analysis were performed as previously described64,65,66,67,68. Metabolite extraction was achieved using a mixture of isopropanol, acetonitrile, and water at a ratio of 3:3:2 v/v. Extracts were divided into three parts: 75 uL for gas chromatography combined with time-of-flight high-resolution mass spectrometry, 150 uL for reversed-phase liquid chromatography coupled with high-resolution mass spectrometry, and 150 uL for hydrophilic interaction chromatography with liquid chromatography and tandem mass spectrometry, and analyzed as previously described64,65,66,67,68. We used the NEXERA XR UPLC system (Shimadzu, Columbia, MD, USA), coupled with the Triple Quad 5500 System (AB Sciex, Framingham, MA, USA) to perform hydrophilic interaction liquid chromatography analysis, NEXERA XR UPLC system (Shimadzu, Columbia, MD, USA), coupled with the Triple TOF 6500 System (AB Sciex, Framingham, MA, USA) to perform reversed-phase liquid chromatography analysis, and Agilent 7890B gas chromatograph (Agilent, Palo Alto, CA, USA) interfaced to a Time-of-Flight Pegasus HT Mass Spectrometer (Leco, St. Joseph, MI, USA). The GC system was fitted with a Gerstel temperature-programmed injector, a cooled injection system (model CIS 4). An automated liner exchange (ALEX) (Gerstel, Muhlheim an der Ruhr, Germany) was used to eliminate cross-contamination from the sample matrix that was occurring between sample runs. Quality control was performed using a metabolite standards mixture and pooled samples applying the methodology previously described69,70,71,72. A quality control sample containing a standard mixture of amino and organic acids purchased from Sigma-Aldrich as certified reference material, was injected daily to perform an analytical system suitability test, and monitor recorded signals day to day reproducibility as it was described64,65,66,67,68. A pooled quality control sample was obtained by taking an aliquot of the same volume of all samples from the study and injecting daily with a batch of analyzed samples to determine the optimal dilution of the batch samples and validate metabolite identification and peak integration. Collected raw data were manually inspected, merged, imputed, and normalized by the sample median. Metabolite identification was performed using in house authentic standards analysis. Metabolite annotation was used utilizing recorded retention time and retention indexes, recorded MSn and HRAMSn data matching with METLIN, NIST MS, Wiley Registry of Mass Spectral Data, HMDB, MassBank of North America, MassBank Europe, Golm Metabolome Database, SCIEX Accurate Mass Metabolite Spectral Library, MzCloud, and IDEOM databases.

Metabolomic data were analyzed as previously described by Tolstikov et al.64. Identified metabolites were subjected to pathway analysis with MetaboAnalyst 5.0, using the Metabolite Set Enrichment Analysis (MSEA) module, which consists of an enrichment analysis relying on measured levels of metabolites and pathway topology and provides visualization of the identified metabolic pathways. Accession numbers of detected metabolites (HMDB, PubChem, and KEGG Identifiers) were generated, manually inspected, and utilized to map the canonical pathways. MSEA was used to interrogate functional relation, which describes the correlation between compound concentration profiles and clinical outcomes.

Data are shown as mean values ± standard error of the mean (SEM). All statistics were performed with Microsoft Excel and Prism 10 (GraphPad). Data were tested for a normal (Gaussian) distribution using Shapiro-Wilk normality test. Statistical outliers were determined by the Grubb’s test. Two group comparisons were analyzed by using the two-tailed paired or unpaired Student’s t-test. Multiple comparisons were performed by using One-way factorial ANOVA or Two-way repeated-measures ANOVA followed by Tukey’s post-hoc test. P values less than 0.05 were considered statistically significant.

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

The authors declare that the data supporting the findings of this study are presented within the paper and its Supplementary Information files. Additionally, the raw data underlying these findings are available as source data files. All accession codes used in this study are listed below: PXD054050, GSE10246, GSE1133, and SCP1376. Data analysis pipelines used in this study for processing of Single Cell Portal (study no. SCP1376) can be obtained from https://github.com/camara-h/Tsuji_2024. Source data are provided with this paper.

Collaborators, G. Co. D. Global, regional, and national age-sex specific mortality for 264 causes of death, 1980-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 390, 1151–1210 (2017).

Article Google Scholar

Collaborators, G. O. Health Effects of Overweight and Obesity in 195 Countries over 25 Years. N. Engl. J. Med 377, 13–27 (2017).

Article Google Scholar

Zwick, R. K., Guerrero-Juarez, C. F., Horsley, V. & Plikus, M. V. Anatomical, Physiological, and Functional Diversity of Adipose Tissue. Cell Metab. 27, 68–83 (2018).

Article CAS PubMed PubMed Central Google Scholar

Kahn, C. R., Wang, G. & Lee, K. Y. Altered adipose tissue and adipocyte function in the pathogenesis of metabolic syndrome. J. Clin. Invest 129, 3990–4000 (2019).

Article PubMed PubMed Central Google Scholar

Cypess, A. M. Reassessing Human Adipose Tissue. N. Engl. J. Med 386, 768–779 (2022).

Article CAS PubMed Google Scholar

Gavalda-Navarro, A., Villarroya, J., Cereijo, R., Giralt, M. & Villarroya, F. The endocrine role of brown adipose tissue: An update on actors and actions. Rev. Endocr. Metab. Disord. 23, 31–41 (2022).

Article CAS PubMed Google Scholar

Pillon, N. J., Loos, R. J. F., Marshall, S. M. & Zierath, J. R. Metabolic consequences of obesity and type 2 diabetes: Balancing genes and environment for personalized care. Cell 184, 1530–1544 (2021).

Article CAS PubMed PubMed Central Google Scholar

Koyanagi, M. & Terakita, A. Diversity of animal opsin-based pigments and their optogenetic potential. Biochim Biophys. Acta 1837, 710–716 (2014).

Article CAS PubMed Google Scholar

Nayak, G. et al. Adaptive Thermogenesis in Mice Is Enhanced by Opsin 3-Dependent Adipocyte Light Sensing. Cell Rep. 30, 672–686.e678 (2020).

Article CAS PubMed PubMed Central Google Scholar

Sato, M. et al. Cell-autonomous light sensitivity via Opsin3 regulates fuel utilization in brown adipocytes. PLoS Biol. 18, e3000630 (2020).

Article CAS PubMed PubMed Central Google Scholar

Mishra, G. & Townsend, K. L. The metabolic and functional roles of sensory nerves in adipose tissues. Nat. Metab. 5, 1461–1474 (2023).

Article PubMed Google Scholar

Guilherme, A., Henriques, F., Bedard, A. H. & Czech, M. P. Molecular pathways linking adipose innervation to insulin action in obesity and diabetes mellitus. Nat. Rev. Endocrinol. 15, 207–225 (2019).

Article CAS PubMed PubMed Central Google Scholar

Zhang, K. X. et al. Violet-light suppression of thermogenesis by opsin 5 hypothalamic neurons. Nature 585, 420–425 (2020).

Article ADS CAS PubMed PubMed Central Google Scholar

Pico, C., Palou, M., Pomar, C. A., Rodriguez, A. M. & Palou, A. Leptin as a key regulator of the adipose organ. Rev. Endocr. Metab. Disord. 23, 13–30 (2022).

Article CAS PubMed Google Scholar

Brosnan, M. E. & Brosnan, J. T. Histidine Metabolism and Function. J. Nutr. 150, 2570S–2575S (2020).

Article PubMed PubMed Central Google Scholar

Yoshimatsu, H., Chiba, S., Tajima, D., Akehi, Y. & Sakata, T. Histidine suppresses food intake through its conversion into neuronal histamine. Exp. Biol. Med (Maywood) 227, 63–68 (2002).

Article CAS PubMed Google Scholar

Yasuda, T. et al. L-histidine stimulates sympathetic nerve activity to brown adipose tissue in rats. Neurosci. Lett. 362, 71–74 (2004).

Article CAS PubMed Google Scholar

Yoshimatsu, H. et al. Histidine induces lipolysis through sympathetic nerve in white adipose tissue. Eur. J. Clin. Invest 32, 236–241 (2002).

Article CAS PubMed Google Scholar

Regard, J. B., Sato, I. T. & Coughlin, S. R. Anatomical profiling of G protein-coupled receptor expression. Cell 135, 561–571 (2008).

Article CAS PubMed PubMed Central Google Scholar

Lattin, J. E. et al. Expression analysis of G Protein-Coupled Receptors in mouse macrophages. Immunome Res 4, 5 (2008).

Article PubMed PubMed Central Google Scholar

Emont, M. P. et al. A single-cell atlas of human and mouse white adipose tissue. Nature 603, 926–933 (2022).

Article ADS CAS PubMed PubMed Central Google Scholar

Shin, G. et al. Flexible Near-Field Wireless Optoelectronics as Subdermal Implants for Broad Applications in Optogenetics. Neuron 93, 509–521.e503 (2017).

Article CAS PubMed PubMed Central Google Scholar

Ikeda, K., Tajima, K., Tanabe, Y., Poon, A. S. Y. & Kajimura, S. Activation of UCP1-Independent Ca(2+) Cycling Thermogenesis by Wireless Optogenetics. Methods Mol. Biol. 2448, 131–139 (2022).

Article CAS PubMed PubMed Central Google Scholar

Koyanagi, M., Takada, E., Nagata, T., Tsukamoto, H. & Terakita, A. Homologs of vertebrate Opn3 potentially serve as a light sensor in nonphotoreceptive tissue. Proc. Natl Acad. Sci. USA 110, 4998–5003 (2013).

Article ADS CAS PubMed PubMed Central Google Scholar

Jacobs, G. H., Neitz, J. & Deegan, J. F. 2nd. Retinal receptors in rodents maximally sensitive to ultraviolet light. Nature 353, 655–656 (1991).

Article ADS CAS PubMed Google Scholar

Jacobs, G. H., Williams, G. A. & Fenwick, J. A. Influence of cone pigment coexpression on spectral sensitivity and color vision in the mouse. Vis. Res 44, 1615–1622 (2004).

Article CAS PubMed Google Scholar

Buhr, E. D. et al. Neuropsin (OPN5)-mediated photoentrainment of local circadian oscillators in mammalian retina and cornea. Proc. Natl Acad. Sci. USA 112, 13093–13098 (2015).

Article ADS CAS PubMed PubMed Central Google Scholar

Rodriguez-Caso, C. et al. Local changes in the catalytic site of mammalian histidine decarboxylase can affect its global conformation and stability. Eur. J. Biochem 270, 4376–4387 (2003).

Article CAS PubMed Google Scholar

Fulop, A. K. et al. Hyperleptinemia, visceral adiposity, and decreased glucose tolerance in mice with a targeted disruption of the histidine decarboxylase gene. Endocrinology 144, 4306–4314 (2003).

Article PubMed Google Scholar

Liew, C. W. et al. Ablation of TRIP-Br2, a regulator of fat lipolysis, thermogenesis and oxidative metabolism, prevents diet-induced obesity and insulin resistance. Nat. Med 19, 217–226 (2013).

Article CAS PubMed PubMed Central Google Scholar

Townsend, K. L. et al. Reestablishment of Energy Balance in a Male Mouse Model With POMC Neuron Deletion of BMPR1A. Endocrinology 158, 4233–4245 (2017).

Article CAS PubMed PubMed Central Google Scholar

Blaszkiewicz, M., Willows, J. W., Johnson, C. P. & Townsend, K. L. The Importance of Peripheral Nerves in Adipose Tissue for the Regulation of Energy Balance. Biology (Basel) 8, https://doi.org/10.3390/biology8010010 (2019).

McLaughlin, T. et al. Inflammation in subcutaneous adipose tissue: relationship to adipose cell size. Diabetologia 53, 369–377 (2010).

Article CAS PubMed Google Scholar

Stenkula, K. G. & Erlanson-Albertsson, C. Adipose cell size: importance in health and disease. Am. J. Physiol. Regul. Integr. Comp. Physiol. 315, R284–R295 (2018).

Article CAS PubMed Google Scholar

Meng, J. J. et al. Light modulates glucose metabolism by a retina-hypothalamus-brown adipose tissue axis. Cell 186, 398–412.e317 (2023).

Article CAS PubMed Google Scholar

Figueiro Longo, M. G. et al. Effect of Transcranial Low-Level Light Therapy vs Sham Therapy Among Patients With Moderate Traumatic Brain Injury: A Randomized Clinical Trial. JAMA Netw. Open 3, e2017337 (2020).

Article PubMed PubMed Central Google Scholar

Ondrusova, K. et al. Subcutaneous white adipocytes express a light sensitive signaling pathway mediated via a melanopsin/TRPC channel axis. Sci. Rep. 7, 16332 (2017).

Article ADS PubMed PubMed Central Google Scholar

Sugihara, T., Nagata, T., Mason, B., Koyanagi, M. & Terakita, A. Absorption Characteristics of Vertebrate Non-Visual Opsin, Opn3. PLoS One 11, e0161215 (2016).

Article PubMed PubMed Central Google Scholar

Mihalik, S. J. et al. Metabolomic profiling of fatty acid and amino acid metabolism in youth with obesity and type 2 diabetes: evidence for enhanced mitochondrial oxidation. Diabetes Care 35, 605–611 (2012).

Article CAS PubMed PubMed Central Google Scholar

Welsh, P. et al. Circulating amino acids and the risk of macrovascular, microvascular and mortality outcomes in individuals with type 2 diabetes: results from the ADVANCE trial. Diabetologia 61, 1581–1591 (2018).

Article CAS PubMed PubMed Central Google Scholar

Piro, M. C. et al. Free-amino acid metabolic profiling of visceral adipose tissue from obese subjects. Amino Acids 52, 1125–1137 (2020).

Article CAS PubMed Google Scholar

Moro, J., Tome, D., Schmidely, P., Demersay, T. C. & Azzout-Marniche, D. Histidine: A Systematic Review on Metabolism and Physiological Effects in Human and Different Animal Species. Nutrients 12, https://doi.org/10.3390/nu12051414 (2020).

Iwasaki, M. et al. Validity of a Self-Administered Food-Frequency Questionnaire for Assessing Amino Acid Intake in Japan: Comparison With Intake From 4-Day Weighed Dietary Records and Plasma Levels. J. Epidemiol. 26, 36–44 (2016).

Article PubMed Google Scholar

Schmidt, J. A. et al. Plasma concentrations and intakes of amino acids in male meat-eaters, fish-eaters, vegetarians and vegans: a cross-sectional analysis in the EPIC-Oxford cohort. Eur. J. Clin. Nutr. 70, 306–312 (2016).

Article CAS PubMed Google Scholar

Li, Y. C. et al. Relationships of Dietary Histidine and Obesity in Northern Chinese Adults, an Internet-Based Cross-Sectional Study. Nutrients 8, https://doi.org/10.3390/nu8070420 (2016).

de Courten, B. et al. Effects of carnosine supplementation on glucose metabolism: Pilot clinical trial. Obes. (Silver Spring) 24, 1027–1034 (2016).

Article Google Scholar

Yan, R., Zhao, X., Lei, J. & Zhou, Q. Structure of the human LAT1-4F2hc heteromeric amino acid transporter complex. Nature 568, 127–130 (2019).

Article ADS CAS PubMed Google Scholar

Puris, E., Gynther, M., Auriola, S. & Huttunen, K. M. L-Type amino acid transporter 1 as a target for drug delivery. Pharm. Res 37, 88 (2020).

Article CAS PubMed PubMed Central Google Scholar

Haas, H. & Panula, P. The role of histamine and the tuberomamillary nucleus in the nervous system. Nat. Rev. Neurosci. 4, 121–130 (2003).

Article CAS PubMed Google Scholar

Panula, P., Yang, H. Y. & Costa, E. Histamine-containing neurons in the rat hypothalamus. Proc. Natl Acad. Sci. USA 81, 2572–2576 (1984).

Article ADS CAS PubMed PubMed Central Google Scholar

Brown, R. E., Stevens, D. R. & Haas, H. L. The physiology of brain histamine. Prog. Neurobiol. 63, 637–672 (2001).

Article CAS PubMed Google Scholar

Morimoto, T., Yamamoto, Y. & Yamatodani, A. Brain histamine and feeding behavior. Behav. Brain Res 124, 145–150 (2001).

Article CAS PubMed Google Scholar

Tsuda, K. et al. Hypothalamic histamine neurons activate lipolysis in rat adipose tissue. Exp. Biol. Med (Maywood) 227, 208–213 (2002).

Article CAS PubMed Google Scholar

Yoshikawa, T. et al. Insufficient intake of L-histidine reduces brain histamine and causes anxiety-like behaviors in male mice. J. Nutr. 144, 1637–1641 (2014).

Article CAS PubMed Google Scholar

O’Hagan, J. B., Khazova, M. & Price, L. L. Low-energy light bulbs, computers, tablets and the blue light hazard. Eye (Lond.) 30, 230–233 (2016).

Article PubMed Google Scholar

Moyano, D. B., Sola, Y. & González-Lezcano, R. A. Blue-light levels emitted from portable electronic devices compared to sunlight. Energies 13, 4276 (2020).

Article Google Scholar

Shang, X., Ling, W., Chen, Y., Li, C. & Huang, X. Construction of a Flexible Optogenetic Device for Multisite and Multiregional Optical Stimulation Through Flexible micro-LED Displays on the Cerebral Cortex. Small 19, e2302241 (2023).

Article PubMed Google Scholar

Lee, J. H. et al. Wearable Surface-Lighting Micro-Light-Emitting Diode Patch for Melanogenesis Inhibition. Adv. Health. Mater. 12, e2201796 (2023).

Article Google Scholar

Vaughan, C. H., Zarebidaki, E., Ehlen, J. C. & Bartness, T. J. Analysis and measurement of the sympathetic and sensory innervation of white and brown adipose tissue. Methods Enzymol. 537, 199–225 (2014).

Article CAS PubMed PubMed Central Google Scholar

Tseng, Y. H. et al. New role of bone morphogenetic protein 7 in brown adipogenesis and energy expenditure. Nature 454, 1000–1004 (2008).

Article ADS CAS PubMed PubMed Central Google Scholar

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).

Article CAS PubMed PubMed Central Google Scholar

Shamsi, F. & Tseng, Y. H. Protocols for Generation of Immortalized Human Brown and White Preadipocyte Cell Lines. Methods Mol. Biol. 1566, 77–85 (2017).

Article CAS PubMed PubMed Central Google Scholar

Johnston, N. A. et al. Strategies for refinement of abdominal device implantation in mice: strain, carboxymethylcellulose, thermal support, and atipamezole. J. Am. Assoc. Lab Anim. Sci. 46, 46–53 (2007).

CAS PubMed Google Scholar

Tolstikov, V., Nikolayev, A., Dong, S., Zhao, G. & Kuo, M. S. Metabolomics analysis of metabolic effects of nicotinamide phosphoribosyltransferase (NAMPT) inhibition on human cancer cells. PLoS One 9, e114019 (2014).

Article ADS PubMed PubMed Central Google Scholar

Drolet, J. et al. Integrated Metabolomics Assessment of Human Dried Blood Spots and Urine Strips. Metabolites 7 https://doi.org/10.3390/metabo7030035 (2017).

Baskin, A. S. et al. Regulation of Human Adipose Tissue Activation, Gallbladder Size, and Bile Acid Metabolism by a beta3-Adrenergic Receptor Agonist. Diabetes 67, 2113–2125 (2018).

Article CAS PubMed PubMed Central Google Scholar

Kiebish, M. A. et al. Multi-omic serum biomarkers for prognosis of disease progression in prostate cancer. J. Transl. Med 18, 10 (2020).

Article CAS PubMed PubMed Central Google Scholar

Ntranos, A. et al. Bacterial neurotoxic metabolites in multiple sclerosis cerebrospinal fluid and plasma. Brain 145, 569–583 (2022).

Article PubMed Google Scholar

Bajad, S. & Shulaev, V. LC-MS-based metabolomics. Methods Mol. Biol. 708, 213–228 (2011).

Article CAS PubMed Google Scholar

Dunn, W. B. et al. Procedures for large-scale metabolic profiling of serum and plasma using gas chromatography and liquid chromatography coupled to mass spectrometry. Nat. Protoc. 6, 1060–1083 (2011).

Article CAS PubMed Google Scholar

Yuan, M., Breitkopf, S. B., Yang, X. & Asara, J. M. A positive/negative ion-switching, targeted mass spectrometry-based metabolomics platform for bodily fluids, cells, and fresh and fixed tissue. Nat. Protoc. 7, 872–881 (2012).

Article CAS PubMed PubMed Central Google Scholar

Want, E. J. et al. Global metabolic profiling of animal and human tissues via UPLC-MS. Nat. Protoc. 8, 17–32 (2013).

Article CAS PubMed Google Scholar

Download references

This work was supported in part by U.S. National Institutes of Health (NIH) grants R01DK122808 and R01DK102898 (to Y.-H.T.), P30DK036836 and S10OD028568 (to Joslin Diabetes Center’s Diabetes Research Center), and by US Army Medical Research grant W81XWH-17-1-0428 (to Y.-H.T.). T. Tsuji was supported by the SUNSTAR Research fellowship (Hiroo Kaneda Scholarship, Sunstar Foundation, Japan) and the American Heart Association grant 903968. M.D.L was supported by NIH fellowship K01DK111714. We would like to thank A. Clermont and K. Park of the Joslin Diabetes Center Animal Physiology core for assisting with the CLAMS studies.

Section on Integrative Physiology and Metabolism, Research Division, Joslin Diabetes Center, Harvard Medical School, Boston, MA, USA

Tadataka Tsuji, Yang Zhang, Tian Lian Huang, Henrique Camara, Meghan Halpin, Matthew D. Lynes & Yu-Hua Tseng

BPGbio, Framingham, MA, USA

Vladimir Tolstikov, Niven R. Narain & Michael A. Kiebish

Solomon H. Snyder Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, MD, USA

King-Wai Yau

Center for Molecular Medicine, MaineHealth Institute for Research, Scarborough, ME, USA

Matthew D. Lynes

Harvard Stem Cell Institute, Harvard University, Cambridge, MA, USA

Yu-Hua Tseng

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

T.T. designed and directed research, performed experiments, analyzed data and wrote the paper. Y.Z. performed in vitro experiments. T.L.H. carried out the ELISA assay. H.C. analyzed single-nuclei RNA sequencing datasets. M.H. performed the physiological experiments by using the CLAMS, DEXA scan, and blood pressure measurement. V.T. and M.A.K. performed metabolomic analysis and experimentation. N.R.N. oversaw metabolomic experiments. M.D.L. created the instruments for light penetration. K.-W.Y. provided Opn3-GKO mice. Y.-H.T. designed and directed the research and co-wrote the paper. All authors have read and agreed to the published version of the manuscript.

Correspondence to Yu-Hua Tseng.

The authors declare the following competing interests: V.T., M.A.K., and N.R.N. are employees of BPGbio. The remaining authors declare no other competing interests.

Nature Communications thanks Jing Wang and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

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/.

Reprints and permissions

Tsuji, T., Tolstikov, V., Zhang, Y. et al. Light-responsive adipose-hypothalamus axis controls metabolic regulation. Nat Commun 15, 6768 (2024). https://doi.org/10.1038/s41467-024-50866-0

Download citation

Received: 16 October 2023

Accepted: 24 July 2024

Published: 08 August 2024

DOI: https://doi.org/10.1038/s41467-024-50866-0

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative