Genetic evidence for involvement of β2-adrenergic receptor in brown adipose tissue thermogenesis in humans

Abstract

Background: Sympathetic activation of brown adipose tissue (BAT) thermogenesis can ameliorate obesity and related metabolic abnormalities. However, crucial subtypes of the β-adrenergic receptor (AR), as well as effects of its genetic variants on functions of BAT, remains unclear in humans. We conducted association analyses of genes encoding β-ARs and BAT activity in human adults.

Methods: Single nucleotide polymorphisms (SNPs) in β1-, β2-, and β3-AR genes (ADRB1, ADRB2, and ADRB3) were tested for the association with BAT activity under mild cold exposure (19 °C, 2 h) in 399 healthy Japanese adults. BAT activity was measured using fluorodeoxyglucose-positron emission tomography and computed tomography (FDG-PET/CT). To validate the results, we assessed the effects of SNPs in the two independent populations comprising 277 healthy East Asian adults using near-infrared time-resolved spectroscopy (NIR_TRS) or infrared thermography (IRT). Effects of SNPs on physiological responses to intensive cold exposure were tested in 42 healthy Japanese adult males using an artificial climate chamber.

Results: We found a significant association between a functional SNP (rs1042718) in ADRB2 and BAT activity assessed with FDG-PET/CT (p < 0.001). This SNP also showed an association with cold-induced thermogenesis in the population subset. Furthermore, the association was replicated in the two other independent populations; BAT activity was evaluated by NIR_TRS or IRT (p < 0.05). This SNP did not show associations with oxygen consumption and cold-induced thermogenesis under intensive cold exposure, suggesting the irrelevance of shivering thermogenesis. The SNPs of ADRB1 and ADRB3 were not associated with these BAT-related traits.

Conclusions: The present study supports the importance of β2-AR in the sympathetic regulation of BAT thermogenesis in humans. The present collection of DNA samples is the largest to which information on the donor's BAT activity has been assigned and can serve as a reference for further in-depth understanding of human BAT function.

Fig. 1. Association analyses in the FDG-PET/CT population.

^aIdentifiers in the Single Nucleotide Polymorphism database at the National Center for Biotechnology Information are shown. ^bAncestral allele/derived allele of each SNP is indicated. ^cThe tested genetic models are as follows: A additive model, D dominant model, R recessive model, DAF derived allele frequency, P-HW p values for Hardy–Weinberg test, OR odds ratio, CI confidence interval; A binary variable representing the presence (1) or absence (0) of active BAT was used as the dependent variable. As independent variables other than genotypes, sex (0 = female, 1 = male), age, the interaction term of age and sex (female: 0*age, male: 1*age), and a dummy variable representing the test month (0 = December and March, 1 = January and February) were included. For each SNP, three genetic models (additive, dominant, and recessive) of the derived allele were tested. The black dots and horizontal bars represent the ORs of each model and their 95% CIs, respectively.

Fig. 2. rs1042718 genotypes and cold-induced thermogenesis (CIT).

CIT values adjusted for fat-free mass in (a) 50 male participants of the FDG-PET/CT population and (b) 42 male participants of the artificial climate chamber experiment are shown. Differences between genotype groups were tested using multiple linear regression models. *p < 0.05; N.S. not significant. The box plot shows median values (central line), mean values (cross mark), and 75th and 25th percentiles (upper and lower boundaries). The largest and smallest values are represented as whiskers drawn from the ends of the boxes to the values. Outliers are indicated as dots.

Fig. 3. Effects of ADRB2 and ADRB3 SNP genotypes on BAT activity in the group of NIR_TRS or IRT.

a Representative thermal images for each genotype (CC vs. A allele carrier) in the IRT experiment. Images at 27 °C baseline (left) and after 90 min of cold exposure (right) are shown. Consent to publish thermal images was obtained from the participants; b, d The box plot shows the distribution of [total-Hb]_sup in genotype groups of the NIR_TRS population; c, e The box plots show the distribution of ΔTemp in the genotype groups of the IRT population after 90 min of the cold exposure. ΔTemp refers to supraclavicular temperature minus chest temperature. b, c Associations of ADRB2 rs1042718 are indicated; d, e Associations of ADRB3 rs4994 genotype (G allele carrier vs. AA) are indicated. Genotypes, age, sex, test season (IRT population experimented in winter and summer; winter = 1, summer = 0), and ethnicity (only the IRT population comprising Japanese and Chinese participants) were included as independent variables. *p < 0.05; N.S. not significant. The box plot shows median values (central line), mean values (cross mark), and 75th and 25th percentiles (upper and lower boundaries). The largest and smallest values are represented as whiskers drawn from the ends of the boxes to the values. Outliers are indicated as dots.