The cold-induced lipokine 12,13-diHOME promotes fatty acid transport into brown adipose tissue

Abstract

Brown adipose tissue (BAT) and beige adipose tissue combust fuels for heat production in adult humans, and so constitute an appealing target for the treatment of metabolic disorders such as obesity, diabetes and hyperlipidemia. Cold exposure can enhance energy expenditure by activating BAT, and it has been shown to improve nutrient metabolism. These therapies, however, are time consuming and uncomfortable, demonstrating the need for pharmacological interventions. Recently, lipids have been identified that are released from tissues and act locally or systemically to promote insulin sensitivity and glucose tolerance; as a class, these lipids are referred to as 'lipokines'. Because BAT is a specialized metabolic tissue that takes up and burns lipids and is linked to systemic metabolic homeostasis, we hypothesized that there might be thermogenic lipokines that activate BAT in response to cold. Here we show that the lipid 12,13-dihydroxy-9Z-octadecenoic acid (12,13-diHOME) is a stimulator of BAT activity, and that its levels are negatively correlated with body-mass index and insulin sensitivity. Using a global lipidomic analysis, we found that 12,13-diHOME was increased in the circulation of humans and mice exposed to cold. Furthermore, we found that the enzymes that produce 12,13-diHOME were uniquely induced in BAT by cold stimulation. The injection of 12,13-diHOME acutely activated BAT fuel uptake and enhanced cold tolerance, which resulted in decreased levels of serum triglycerides. Mechanistically, 12,13-diHOME increased fatty acid (FA) uptake into brown adipocytes by promoting the translocation of the FA transporters FATP1 and CD36 to the cell membrane. These data suggest that 12,13-diHOME, or a functional analog, could be developed as a treatment for metabolic disorders.

Figure 1

Discovery of 12,13-diHOME, a cold-induced lipokine linked to BAT activation. (a) Volcano plot of 88 lipids, comparing the fold induction after cold challenge to the P value (paired Student’s t-test). The dashed line indicates a P value of 0.05. 12,13-diHOME is highlighted in red. n = 9 human subjects. (b) Individual plasma concentration of 12,13-diHOME before and after cold challenge. The P value for a paired Student’s t-test is shown. (c) Plasma 12,13-diHOME concentration plotted with BAT-standardized uptake value (SUV), as measured by a positron-emission tomography (PET) scan of radiolabeled fluorodeoxyglucose. r is the Spearman correlation coefficient between 12,13-diHOME and BAT activity. (d–i) Circulating 12, 13-diHOME concentration plotted against BMI (d), HOMA-IR (score) (e), circulating triglycerides (f), circulating ALAT (g), circulating HDL cholesterol (h) and LDL cholesterol (i). Throughout, r is the Spearman correlation coefficient between 12,13-diHOME and each parameter. Blue dots represent males; pink dots, females. n = 55 individuals (13M/42F). P value was calculated using algorithm AS 89.

Figure 2

The biosynthetic pathway of 12,13-diHOME is selectively increased in mouse BAT by cold exposure. (a) Serum 12,13-diHOME concentration in control mice, as compared to mice treated with norepinephrine (NE) for 30 min, and mice exposed to 4 °C–cold for 1 h. Data are means ± s.e.m.; n = 6 control mice, n = 6 treated with NE, n = 5 exposed to cold. *P < 0.05, **P < 0.005 by Student’s t-test. (b) Serum concentration of 12, 13-diHOME in male and female mice after a 7-d cold challenge and compared to mice housed at thermoneutrality. Data are means ± s.e.m.; n = 6 thermoneutral mice, n = 5 cold-exposed mice per group; *P < 0.05 by Student’s t-test. (c) Biosynthetic pathway of 12,13-diHOME production. (d) Ephx1 and Ephx2 mRNA expression measured by qPCR in BAT from control mice as opposed to that in mice exposed to 4 °C–cold for 1 h. Data are means ± s.e.m.; n = 6 control mice, n = 7 cold-exposed mice. *P < 0.05 by Student’s t-test. (e) Ephx1 gene expression, as measured by qPCR in BAT, sWAT and epididymal WAT (eWAT) from mice housed at thermoneutrality or 4 °C–cold for 7 d. Data are presented as normalized means ± s.e.m.; n = 4 per group; *P < 0.05 by Student’s t-test. (f) Ephx2 gene expression, as measured by qPCR in BAT, sWAT and eWAT of mice housed at either thermoneutrality or cold for 7 d. Data are presented as normalized means ± s.e.m.; n = 4 per group; *P < 0.05 by Student’s t-test. (g) 12,13-diHOME concentration in media from BAT and sWAT cultured ex vivo for 1 h, normalized to tissue weight. Data are means ± s.e.m.; n = 6 mice; *P < 0.05 by Student’s t-test. (h) 12,13-diHOME concentrations in BAT from wild-type and Myf5^CreBmpr1a^f/f mice housed at cold or thermoneutrality for 2 d or 11 d. Data are plotted as the normalized means ± s.e.m.; n = 5 WT thermoneutral males, 5 WT cold-exposed males, 4 Myf5^CreBmpr1a^f/f thermoneutral males, 5 Myf5^CreBmpr1a^f/f cold-exposed males, 3 WT thermoneutral females, 6 WT cold-exposed females, 4 Myf5^CreBmpr1a^f/f thermoneutral females, 6 Myf5^CreBmpr1a^f/f cold-exposed females; *P < 0.05 by Student’s t-test.

Figure 3

12,13-diHOME enhances cold tolerance and facilitates fatty acid uptake by BAT. (a) Body temperature in mice cold challenged at 4 °C for 90 min after pretreatment with 12,13-diHOME, 12,13-epOME or vehicle. Data are means ± s.e.m.; n = 5 mice per group; *P < 0.05 12,13-diHOME, as compared to vehicle, by analysis of variance (ANOVA) with post hoc Bonferroni test. (b,c) Total V(O₂) consumed and V(CO₂) produced (b) and average respiratory-exchange ratio (RER, c) measurements as measured by Comprehensive Lab Animal Monitoring System (CLAMS) for 1 h of cold (4 °C) in mice acutely treated with 12,13-diHOME or vehicle. Data are means ± s.e.m.; n = 6 mice per group; *P < 0.05 by Student’s t-test. (d) Serum triglycerides in mice fed a high-fat diet and treated with 12,13-diHOME or vehicle for 2 weeks. Data are means ± s.e.m.; n = 6 treated as compared to n = 5 controls; *P < 0.05 by Student’s t-test. (e) Radioactivity per 10 mg of liver, gastrocnemius muscle (Gastroc.), heart, interscapular BAT (iBAT), sWAT and epididymal white adipose tissue (eWAT) from mice treated with vehicle or 12,13-diHOME and then given an oral bolus of ³H-labeled triglyceride. Data are means ± s.e.m.; n = 6 per group; *P < 0.05 by Student’s t-test. (f) Radioactivity per 10 mg of tissues from mice treated with vehicle, norepinephrine (NE) or 12,13-diHOME and then given a bolus of ³H-labeled oleate. Data are means ± s.e.m.; n = 8 per group; *P < 0.05 12,13-diHOME as compared to vehicle by ANOVA with post hoc Bonferroni test. (g) Representative images of luciferase activity in Ucp1^Cre Rosa(stop)Luc mice injected intravenously with luciferin-conjugated fatty acid and 12,13-diHOME or vehicle. Data are representative images at 0 min, 10 min and 55 min. (h) Quantification of luciferase activity in (g). (i) Total luciferase counts from six individual experiments were averaged and plotted as the normalized means ± s.e.m.; n = 6 per group; *P < 0.05 12,13-diHOME as compared to vehicle treatment by Student’s t-test.

Figure 4

12,13-diHOME promotes fatty acid uptake in vitro by activating the translocation and oligomerization of FA transporters. (a) Fatty acid uptake in mature brown adipocytes constitutively expressing firefly luciferase that were pretreated with either 12,13-diHOME or vehicle, as measured by luciferase activity using 10 μM FFA-SS-Luc. Data are plotted as the normalized means ± s.e.m.; n = 6 technical-replicate wells per group; *P < 0.05 12,13-diHOME as compared to vehicle by ANOVA with post hoc Bonferroni test. (b) Radiolabeled ¹⁴C palmitic acid uptake in mature brown adipocytes pretreated for 15 min with either 12,13-diHOME or vehicle. The data were normalized by protein content. Data are presented as means ± s.e.m.; n = 10–11 technical-replicate wells per group; *P < 0.05 12,13-diHOME as compared to vehicle by Student’s t-test. (c) Radiolabeled ¹⁴C palmitic acid oxidation in mature brown adipocytes pretreated with either 12,13-diHOME or vehicle. The data were normalized by protein content. Data are presented as means ± s.e.m.; n = 10–11 technical-replicate wells per group. (d) Basal respiration of mature brown adipocytes treated with either 12,13-diHOME or vehicle. The data were normalized by protein content. Data are presented as means ± s.e.m.; n = 10–11 technical-replicate wells per group; *P < 0.05 12,13-diHOME as compared to vehicle by Student’s t-test. (e) Western blot analysis of membrane and cytosol fractions of differentiated brown adipocytes treated with 12,13-diHOME or vehicle. Data are presented as a representative image; n = 3 separate experiments. (f) The upper band corresponding to the oligomer form of FATP1 was measured by densitometry from immunoblots of three independent experiments. Data are presented as means ± s.e.m.; n = 3 separate experiments; *P < 0.05 12,13-diHOME as compared to vehicle by one-way ANOVA. (g) Densitometry of the low-glycosylation form of CD36 from immunoblots of three independent experiments. Data are presented as means ± s.e.m.; n = 3 separate experiments; *P < 0.05 12,13-diHOME as compared to vehicle by one-way ANOVA. (h) Proposed model of 12,13-diHOME biosynthesis and action in cold-activated BAT.