Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Feb 1.
Published in final edited form as: Am J Biol Anthropol. 2023 Sep 23;183(2):e24848. doi: 10.1002/ajpa.24848

When the Cold Gets Under Your Skin: Evidence for Brown Adipose Tissue Activity in Samoan Adults

Alexandra Niclou 1,2,*, Lupesina Vesi 3, Maria Arorae 3, New Caledonia Naseri 3, Kima Faasalele Savusa 3, Take Naseri 4, James P DeLany 5, Stephen T McGarvey 6, Anna C Rivara 7, Cara Ocobock 2,8
PMCID: PMC10843446  NIHMSID: NIHMS1937634  PMID: 37740598

Abstract

OBJECTIVES

Brown adipose tissue (BAT) is a heat-producing organ aiding non-shivering thermogenesis (NST) during cold stress. Due to its potential cold-adaptive role BAT has been predominantly studied in cold and temperate climate populations, but not among warm-climate adults. This work explores if BAT activity can be inferred in Samoans.

MATERIALS AND METHODS

We inferred BAT activity by comparing metabolic rate and surface heat dissipation using indirect calorimetry and thermal imaging between room temperature and cold exposure among Samoans (N=61, females: n=38) from 'Upolu Island, Samoa. BAT activity was inferred using ANOVA linear regression models with the variables measured at cold exposure as outcomes. T-tests were used to compare changes in surface temperature between room temperature and cold exposure.

RESULTS

Metabolic rate significantly increased after cooling. In both the supraclavicular area, a known BAT location, and the sternum, a non-BAT location, temperatures decreased significantly upon cold exposure. Differences in supraclavicular temperatures between room temperature and cold were significantly smaller than differences in sternum temperatures between exposures. These results suggest that BAT thermogenesis occurred in known BAT-locations and thus contributed to NST during cooling.

CONCLUSIONS

This study adds to our understanding of BAT activity across different populations and climates. Further study may illuminate whether the cold-adaptive properties of BAT may have played a role in the successful expansion of populations across the globe, including warm-climate groups.

Keywords: Brown Adipose Tissue, Metabolic Rate, Thermoregulation, Polynesia

1. Introduction

Over the last decade, research on brown adipose tissue (BAT) in humans has gained considerable traction due to the tissue’s thermogenic and metabolic properties (Lee et al., 2010; Ouellet et al., 2012; Tam et al., 2012; Trayhurn 2017). While shivering can be difficult to detect at low levels, BAT activity is believed to be the main source of non-shivering thermogenesis (NST) in mammals, generating heat under mild cold conditions without the onset of intense shivering (Foster and Frydman, 1978; van der Lans et al., 2013; Devlin, 2021). In humans, NST allows for comfortable living in moderately cold conditions and represents one of the principal physiological adaptations to cold (Cannon and Nedergaard, 2004). Due to its heat producing capacity and role in mammalian hibernation, most of our current understanding of BAT activity stems from studies on rodents, human infants, and cold-adapted or cold-acclimated human populations (Himms-Hagen, 1985; Cannon and Nedergaard, 2004; Yoneshiro et al., 2016; Levy et al., 2018; Niclou and Ocobock, 2021; Ocobock et al., 2022). While prior studies have shown BAT activation in temperate climate populations, only recently have groups examined BAT activity in a population rarely exposed to prolonged or seasonal cold (Cypess et al., 2009; van Marken Lichtenbelt et al., 2009; Monfort-Pires et al., 2022; Niclou and Ocobock, 2022; Yoneshiro et al., 2016;). The present study is the first to explore the effects of BAT on NST by inferring BAT activity through simultaneous measures of metabolic rate and heat dissipation before and during sustained mild cold exposure among Polynesian adults from Samoa. Evidence of BAT activity in a tropical population could provide evidence of its role as a ubiquitous trait contributing to thermogenesis across all populations.

Human BAT was first recorded in infants, who despite lacking enough muscle mass to produce substantial heat from shivering, maintain core temperature through NST (Cannon and Nedergaard, 2004). While shivering is achieved by contracting small muscles, minimally increasing body temperature while significantly increasing energy expenditure, BAT does not rely on muscle contraction (Parsons, 2007). Instead, BAT releases heat by interrupting the electron transport chain using an uncoupling protein. This uncoupling within brown adipocytes accelerates the oxidation process, increasing proton leakage from mitochondria resulting in the dissipation of heat rather than mechanical energy (Sell, Deshaies & Richard, 2004). BAT activity thus generates heat without muscle activation or by acting synergistically with muscles resulting in a metabolically more efficient mechanism of heat production (Tam et al., 2012). While insignificant at thermoneutrality, BAT activation results in significant changes in metabolic rate (MR) and total energy expenditure at mild cold exposure (15-19°C) (Ouellet et al., 2012; Ocobock et al., 2022).

The thermogenic capacity of BAT suggests that it may play a role in cold adaptation and may contribute to differences in metabolic responses to cold exposure (Cypess et al., 2009; Lee et al., 2010; Levy et al., 2018; Niclou and Ocobock, 2022). In circumpolar environments, for example, BAT may have contributed to the survival and thriving of humans by providing an additional source of heat production, reducing the energetic costs of shivering and vasoconstriction (Andersen et al., 1960; Hart et al., 1962). In the cold-adapted Yakut of Siberia, temperatures measured in the supraclavicular area, the main and the most superficial, as in closest to the skin, BAT location in human adults, did not change significantly after being exposed to cold, while simultaneously, measurements of temperature changes at the sternum, in which little to no BAT is detected in adults, dropped significantly after cold exposure (Levy et al., 2018). The decrease in heat dissipation noted in the sternum compared to the relative constant temperature measured at the supraclavicular region, despite a cooling environment, suggests the activation of NST in prominent BAT locations (Levy et al., 2018; Niclou and Ocobock, 2022; Ocobock et al., 2022).

This approach to inferring BAT activity by comparing the changes in skin temperature between the supraclavicular and the sternum area is supported by evidence from PET/CT studies. When comparing individuals with no detectable BAT with individuals with detectable BAT using a thermal imaging camera similar to the one used in this study, Jang and colleagues noted a significant decrease in temperature in the chest compared to the supraclavicular area in those without BAT while no differences were detected between the same two locations in those with BAT (Jang et al. 2014). The noted difference in temperature between the supraclavicular and the sternum in this sample was associated with an 85% predictive value for the presence of BAT in the supraclavicular region as observed using PET/CT scans after injection of a glucose analogue, the gold standard for BAT detection (Jang et al., 2014).

Evidence of NST in individuals exposed to periodic long-term cold environments, such as winter in the Northern Hemisphere, hints at the possible role BAT plays in cold acclimatization. A study comparing BAT activity at mild cold exposure between Yakut adults and a group of temperate climate residents from Illinois, demonstrated that heat dissipation in the supraclavicular area was significantly greater in the Yakut compared to the temperate climate population (Levy et al., 2022). This is consistent with findings from seasonal BAT activation data gathered among a group of North Americans from Albany, New York. Comparing BAT activity between seasons in a New York state sample. Niclou and Ocobock (2022) observed that NST through BAT activity during mild cold exposure was increased in the winter months compared to summer. Similarly, a study in Japan demonstrated that individuals with quantifiable BAT activity, measured using PET/CT scanning, at room temperature had greater glucose uptake and fat oxidation in the winter compared to summer, increasing NST during the colder season (Yoneshiro et al., 2016). This documented acclimatization to changing seasonal temperatures further highlights the important role BAT plays in the human evolutionary response to cold.

There is, however, little currently available data on BAT activity in warm climate populations (Monfort-Pires et al., 2022). Here we aim to address this gap by investigating the effects of an experimental mild cold exposure on NST and metabolic rate in a Polynesian group from Samoa where mean annual temperatures ranges from 26 - 31°C (Samoan Meteorology Division). With little seasonal fluctuation in temperatures and conditions generally described as tropical, Samoa is a well-suited location for the examination of BAT activity in a population with an ostensibly reduced need for NST. This study is the first to examine the possible presence of BAT activity in adult Samoans. Given that BAT activity affects metabolism in cold-adapted and temperate climate populations, we predict that BAT activation will also contribute to changes in metabolic rate and heat dissipation in Samoans. There is precedence for this as BAT activity was inferred in Samoan infants, providing a rationale for studying metabolic and thermal capacities of BAT in Samoan adults (Oyama et al., 2021). However, BAT activity significantly decreases throughout childhood among cold and temperate climate populations raising the question whether BAT activity can be identified in warm-climate adults with little day-to-day exposure to cold (Cannon and Nedergaard, 2004).

We hypothesized that metabolic rate would increase in Samoan adults upon cold exposure, reflecting observations made in Samoan infants and non-Polynesian adult samples from cold and temperate-climate environments (Levy et al., 2018; Niclou and Ocobock, 2022; Oyama et al., 2021; Ocobock et al., 2022). During cold exposure we also expected to see a greater difference in sternum temperatures between ambient room temperature and the experimental cold exposure compared to the difference in supraclavicular temperatures between the two exposures. This study will also investigate the associations between body composition such as fat free mass (FFM) and body fat percentage (BF%) on BAT activity; previous work demonstrated that BAT is inversely correlated with fat mass (Saito et al., 2009), while others provide evidence of similar BAT activity between individuals with and without obesity (Hanssen et al., 2016).

By measuring changes in metabolic rate and surface temperatures after mild cold exposure and assessing the association between BAT activity and body composition, this work investigates the physiological response of potential BAT activity in an adult warm-climate group. The outcomes of this work will provide insight into the variation in metabolic effects of BAT between populations and climates, potentially exposing BAT activity as a shared trait across all populations rather than a trait found only among cold-adapted or -acclimated groups.

2. Material and Methods

2.1. Setting

Samoa is an independent nation in the Polynesian region of the South Pacific. In 2020, the official census population was 202,506 with approximately 80% of individuals living in rural areas and 20% in the urban center (Samoa Bureau of Statistics, 2021). Evidence from Lapita pottery and other archaeological findings estimate that Samoa was first peopled between 2,750 and 2,880 years ago (Petchey, 2001). Genetic studies in modern-day Samoans indicate a strong Austronesian lineage as well as a possible partial population replacement by other Oceanic populations around 900 to 1,050 years ago, reflecting genetic admixture (Harris et al., 2020). The rapid economic and social transitions tied to Westernization, especially in the last 100 years, have had adverse effects on health in the region. Shifts from more traditional lifestyles based in maritime culture and local food consumption to increased sedentary behaviors and consumption of high-calorie and high-processed imported foods are associated with increased risks of obesity, diabetes, hypertension, cholesterol, and psychosocial stress in the Samoan archipelago through urbanization, dietary changes, lowered physical activity, food dependence and aid. (Baker, Hanna, & Baker, 1986; DiBello et al., 2009a; DiBello et al., 2009b; McGarvey, 2001; McGarvey et al., 2005).

2.2. Study Sample

Participants were selected from individuals who had participated in previous genetic studies of obesity and related cardiometabolic traits conducted by several of our study authors (Hawley et al. 2014; Hawley et al., 2020; LaMonica et al., 2021; Minster et al. 2016). Potential participants (n=150) for this study were selected from lists of prior research participants who had consented to contact for future research studies and who had previously measured HbA1C < 6.5%. Of these, 50 screened individuals did not participate in the study after moving out of the greater Apia region (n=22), being unable for family or work-related reasons including declining to participate with no specific explanations (n=24), or not fitting the study inclusion criteria (n=4). Pregnant women and individuals with prior diabetes diagnoses were excluded. All participants (n=100) were over 18 years old and healthy, defined as not suffering from any acute illness at the time of data collection. No restrictions or limitations were put on BMI, age, or sample sex ratio. Subjects were recruited from an existing participant pool from previous studies. By design of these earlier studies, all participants were < 55 years old. 49 participants enrolled in the same study underwent a slightly different protocol and were omitted from the analysis and results for this paper. Data was analyzed and interpreted for a sample of n=61.

Data collection was conducted from January to July 2021 at the Obesity, Lifestyle and Genetic Adaptations (OLaGA) research center within the Samoan Ministry of Health in Apia (the capital city). Participants were residents of 'Upolu Island. Participants were asked to arrive having fasted overnight and not have smoked for a minimum of 12 hours. All participants were of Polynesian ancestry (a requirement of the genetic studies into which they were originally recruited). Assessments were conducted in Samoan or English based on participant preference. The study was reviewed and approved by the Samoan Ministry of Health and by the Yale University and University of Notre Dame Boards of Ethics (IRB # 19-07-5453).

2.3. Anthropometry

Using standard protocols, anthropometric measurements were completed on all participants (Lohman, Roche, & Martorell, 1988). A stadiometer (Seca Corporation, Hanover, MD) was used to measure height to the nearest millimeter and body mass index (BMI) was determined by calculating the ratio of individual body weight over height squared. Percent body fat (BF%), skeletal muscle mass (SMM) and fat free mass (FFM) were measured with a Quantum V Segmental bioelectrical impedance analyzer (RJL Systems, Clinton, MI) using preset standard NHANES-III equations. No menstrual data was collected for this sample despite symptoms of perimenopause and menopause potentially affecting thermogenesis (Rossmanith and Ruebberdt, 2009)

2.4. BAT Activity

BAT activity was inferred by comparing metabolic rate and heat dissipation in the supraclavicular and sternum areas at thermoneutrality and at mild cold exposure following the protocol by Levy (2019). Thermoneutral measurements were taken at room temperature (25-28°C). Participants were then exposed to mild cold exposure (15-19°C) by wearing cooling suits (Allen-Vanguard, Ottawa, Ontario) covering the entire body surface area from ankles to wrists and neck, lined with sewn-in tubing through which temperature-controlled water was pumped. Mean water temperature was approximately 9°C as measured using a digital water thermometer (Amazon Aquaneat, Seattle, WA) in the water bottle feeding the pump resulting in participants experiencing mild cold exposure (15-19°C) as the flow of the water through the tubes of the cooling suit offers a protective layer between water and skin (Levy, 2019). The part of the suit covering the torso and arms was fitted tightly on participants to minimize wrinkling, which could potentially lead to changes in the water temperature. To ensure that the water temperature did not change in the suit while water is flowing away from its cold source, an unworn cooling suit was flooded, and temperature of the water was compared using thermal imaging. This unpublished data provided the necessary reassurance that changes in water temperature within the suit were negligible.

Participants wore an empty (inactivated) cooling suit and laid on a cot with their torso at an incline of 35°, following best practices to measure metabolic rate in heavier participants as described by Miles-Chan et al. (2014). Participants rested for 30 minutes before metabolic rate (MR, kcal/day) measurements were taken, allowing for them to be completely rested before initial measurements were taken. Following the rest period, MR was measured for 30 minutes under thermoneutral condition (MRRT) using a K5 portable calorimetry unit (Cosmed, Chicago, IL), calibrated before use on each participant. The K5 unit measures O2 consumption and CO2 production by breath-by-breath analysis, using a mask with bi-lateral unidirectional valves sealed around the participant’s mouth and nose. A mask was used instead of a potentially more accurate hood for MR measurements as the larger hood would have interfered with thermal images taken of the supraclavicular and sternal areas. MRRT was calculated and recorded by the COSMED Omnia software. The last 10 minutes of MRRT data was averaged and used for analysis. All measurements were repeated at mild cold exposure (MRC) for 30 minutes by activating the cooling suit. Data used for analysis was averaged from the last 10 minutes of MRC. During the cold exposure, none of the participants showed signs of shivering for the duration of MRC.

Heat dissipation, inferring BAT activity, was measured in tandem with MRRT and MRC measurements. Supraclavicular (at room temperature: SupraTempRT; at mild cold exposure: SupraTempC) and sternum (at room temperature: SternTempRT; at mild cold exposure: SternTempC) temperatures were measured using an E76 thermal imaging camera (FLIR, Wilsonville, OR). The camera was placed approximately 30cm from the area being captured, which in the case of the supraclavicular area consisted of the area delineated inferiorly by the clavicle, medially by the sternocleidomastoid, and laterally by the margin defining the participant’s body from their surroundings. The sternum, used as a control for tissue heat dissipation given its lack in superficial BAT that could be detected via thermal imaging was demarked as the area inferior to the chin, superior to the diaphragm, and medially between the pectoral muscles. All pictures were taken over the cooling suit, which was controlled for by capturing images over the same area of the suit under both conditions. Figure 1 shows the same section of the supraclavicular area captured in visible light and infrared by the thermal imaging camera. The juxtaposition of visible and infrared light images as well as position of the tubing in relationship to the shoulder and neck for the supraclavicular region, and torso for the sternum allowed for the appropriate determination of the anatomical landmarks used to capture thermal images in the regions of interest without removing the suit. Differences in heat dissipation between left and right supraclavicular areas have been noted before, which may be due to the placement of the major arteries or differences in musculature between dominant arms. For the purposes of this paper and following methodologies used in similar studies, left and right supraclavicular temperatures were averaged at room temperature and cold exposure and will be referred to as supraclavicular temperature in the text. Thermal images of the left and right supraclavicular area and of the sternum were repeatedly taken at 5-minute intervals during the 30 minutes of metabolic rate measurements at thermoneutrality and again at cold exposure. Images taken within the last 10 minutes of either exposure were used for analysis. Comparing metabolic rate and heat emission in the supraclavicular and sternal areas at room temperature and at mild cold exposure records differences in heat generation and calorie expenditure associated with BAT activity between the two exposures. Comparing heat dissipation in the sternal, a superficial BAT-free location, and supraclavicular, the main BAT location in adults, allows for comparison of BAT-specific heat production. Significant increases in heat dissipation in the supraclavicular area suggest heat production from BAT activation, pointing to increased NST capacities after cooling.

Figure 1. Supraclavicular view.

Figure 1

Open in a new tab

View of participant’s right supraclavicular area from which thermal images were taken (left). The middle and right images are thermal images of the target area of the supraclavicular area (red circle) highlighted on left image. Thermal images of right supraclavicular area at room temperature (25-28°C) (middle) and at mild cold exposure (15-18°C) (right). Top left corner shows highest recorded temperature for heat dissipation in area.

2.5. Statistical Analysis

Statistical analyses were performed in RStudio (Version 3.6.2). Results are presented as means ± standard deviation. Variables were tested for normal distribution by plotting them into histograms. MR, heat dissipation in BAT (SupraTemp), and heat dissipation in non-BAT (SternTemp) locations were measured for 61 participants (38 females, 23 males). Paired sample t-tests were used to determine differences in outcome variables for males and females. Anthropometric measurements and dependent variables (MRR, SupraTempR, SternTempR, MRC, SupraTempC, SternTempC, Δsupraclavicular, and Δsternum) were compared between males and females using one-way analysis of variance (ANOVA). To infer BAT activity, linear regression models were run using ANOVA with cold exposure residual score variables (MRC, SupraTempC, and SternTempC) as each outcome. Ranges in environmental temperatures measured during the room temperature and the cold exposure phases of the protocol were not correlated with inferred BAT activity measures nor based in clinically validated measurements resulting in their limited use for statistical analysis and have thus been omitted in the statistical models. The differences between measurements at room temperature and mild cold exposure were recorded for the supraclavicular area (Δsupraclavicular), the sternum (Δsternum), and MR (ΔMR). Δsupraclavicular and Δsternum were compared using a paired t-test. Linear regression models were run using MRRT, and MRC, with the following independent variables: age, BF%, FFM, Δsupraclavicular, Δsternum. Finally, linear regression models were run using Δsupraclavicular, Δsternum with age, BF%, FFM, and Δsternum and Δsupraclavicular, respectively as independent variables. Results were considered statistically significant at P-value ≤ 0.05. Results in tables for descriptive statistics are reported as means +/− standard deviation.

3. Results

3.1. Descriptive Statistics

Table 1 displays descriptive statistics for 61 participants. Males were between the ages of 34-51 years and ages ranged between 31-54 years for females. While males were significantly taller and had significantly greater FFM and SMM (FFM: P <0.001; SMM: P =0.004), females had significantly greater BMI and percent body fat (BMI: P =0.004; BF% P <0.001).

Table 1.

Descriptive statistics for age and anthropometric measurements in males and females. * One-way ANOVA test of differences by sex are significant at P < 0.025. ** One-way ANOVA test of differences by sex are significant at P < 0.001. BMI: body mass index, BF%: body fat percentage, FFM: fat free mass, SMM: skeletal muscle mass

Females
(n = 38)		 Males
(n = 23)		 P-value
Age (years) 42.6 ± 5.7 43.2 ± 5.7 0.679
Height (cm) 163.3 ± 5.6 173.3 ± 6.0 <0.001**
Body Weight (kg) 104.1 ± 19.9 99.9 ± 22.6 0.454
BMI 39.2 ± 7.7 33.3 ± 7.3 0.004*
BF% 44.5 ± 9.8 32.5 ± 8.4 <0.001**
FFM (kg) 56.1 ± 4.2 66.1 ± 10.7 <0.001**
SMM (kg) 22.7 ± 3.6 27.9 ± 9.7 0.004*
Open in a new tab

3.2. Metabolic Rate and Surface Temperatures

Tables 2 shows mean metabolic rate, supraclavicular temperature, and sternum temperature by exposure for the overall sample and by sex. Figure 2 depicts the differences between the variation in MRRT and MRC. Average MRRT was 1785 ± 510 Kcal/day and MRC was 1847 ± 543 Kcal/day. MRC was significantly elevated compared to MRRT (P < 0.001). Neither average MRRT (P = 0.082) nor average MRC (P = 0.225) were significantly different between sexes. MR increased among n=37 participants (average increase: 239 ± 216 Kcal/day) and MR decreased among n=24 (average decrease: 212 ± 260 Kcal/day) following mild cold exposure. 47.8% of males and 34.2% of females saw a decrease in their metabolic rate after cold exposure, while 52.2% of males and 65.8% of females saw an increase in their metabolic rate after being cold exposed. Δsupraclavicular and Δsternum, neither separately (for Δsupraclavicular: P = 0.877, for Δsternum: P = 0.945) nor combined (P = 0.997), were predictors for ΔMR.

Table 2.

Mean metabolic rate, supraclavicular, and sternum temperature for overall sample and by sex at room temperature and mild cold exposure. * Paired sample t-tests indicating significant changes between exposures at P < 0.05.

Room
Temperature
(25-28°C)		 Mild Cold
Exposure
(15-19°C)		 P-value
Overall Sample (N = 61)
Metabolic Rate (Kcal/day) 1785 ± 510 1847 ± 543 < 0.001*
Supraclavicular Temperature (°C) 33.26 ± 0.82 32.33 ± 0.90 < 0.001*
Sternum Temperature (°C) 32.00 ± 0.98 29.74 ± 1.05 < 0.001*
Females (n = 38)
Metabolic Rate (Kcal/day) 1688 ± 416 1776 ± 483.19 0.056
Supraclavicular Temperature (°C) 33.00 ± 0.78 32.02 ± 0.82 < 0.001*
Sternum Temperature (°C) 31.68 ± 0.94 29.46 ± 1.05 < 0.001*
Males (n = 21)
Metabolic Rate (Kcal/day) 1946 ± 611 1963 ± 623 0.835
Supraclavicular Temperature (°C) 33.68 ± 0.72 32.81 ± 0.74 < 0.001*
Sternum Temperature (°C) 32.56 ± 0.78 30.22 ± 0.90 < 0.001*
Open in a new tab

Figure 2. Metabolic rate variation and mean at room temperature and cold exposure.

Figure 2

Open in a new tab

Significant increase in MR at cold exposure compared to room temperature (P < 0.001).

Both supraclavicular and sternum temperatures decreased significantly after cold exposure, however, changes in temperatures were significantly smaller for the supraclavicular area compared to the sternum (P < 0.001). Figure 3 shows the variation and mean for SupraTempRT, SternTempRT, SupraTempC, and SternTempC. Average SupraTempRT was 33.26 ± 0.82°C, while average SupraTempC was 32.33 ± 0.90 °C. The decrease in heat dissipation from the supraclavicular area between exposures was significant (P < 0.001). Decreases in supraclavicular temperatures after cooling were significant in sex-stratified analyses as well (females: P < 0.001, males: P < 0.001). Males had significantly greater SupraTempRT (P = 0.001) and SupraTempC (P = 0.002) compared to females. SternTempRT averaged 32.00 ± 0.98°C and SternTempC averaged 29.74 ± 1.05°C for the overall sample; a significant decrease between exposures (P < 0.001). This significant decrease in sternum temperature was reflected in sex-stratified analyses as well (females: P < 0.001, males: P < 0.001).

Figure 3. Supraclavicular and sternum surface temperature variation and mean at room temperature and cold exposure.

Figure 3

Open in a new tab

Significant decreases in sternum temperature at cold exposure compared to room temperature.

Average SternTempRT and SternTempC were significantly colder than average SupraTempRT and SupraTempC for the overall sample (all P < 0.001). Stratified by sex, both male and female SternTempRT and SternTempC were also significantly colder compared to SupraTempRT and SupraTempC (all P < 0.001). Furthermore, Figure 4 shows how Δsternum was significantly greater than Δsupraclavicular (P < 0.001) for the overall sample. Similarly, Δsternum was significantly greater than Δsupraclavicular both in females and males (females: P < 0.001, males: P < 0.001), suggesting a greater drop in heat dissipation in the non-superficial BAT location compared to BAT-locations in both men and women. Neither Δsupraclavicular (P = 0.209) nor Δsternum (P = 0.405), however, were significantly different by sex.

Figure 4. Differences in supraclavicular and sternum surface temperature between room temperature and cold exposure.

Figure 4

Open in a new tab

Δsupraclavicular was significantly smaller than Δsternum, suggesting greater heat dissipation in the supraclavicular area during cold exposure.

3.3. Predictors of BAT Activity

Multiple linear regression analyses, which incorporated age, sex, and body composition characteristics were performed for MRRT, MRC, Δsupraclavicular, and Δsternum. Table 3 displays the results of the multiple regression analyses for Δsupraclavicular, and Δsternum. FFM and BF% were predictors for MRRT and MRC. Adding Δsupraclavicular and Δsternum as independent variables to the multiple linear regression model with MRC as a dependent variable (Table 3), weakened the models emphasizing that Δsupraclavicular and Δsternum were not significant predictors of, nor associated with MR at cold exposure.

Table 3.

Multiple linear regression analysis testing for predictors of MRRT and MRC. ** indicates significant predictors at P < 0.001. BF%: body fat percentage, FFM: fat free mass.

MRRT
Adj. R2 = 0.416
p < 0.001**		 MRC
Adj. R2 = 0.371
p < 0.001**		 MRC
Adj. R2 = 0.316
p < 0.001**
Beta Coef. P – values Beta Coef. P – values Beta Coef. P – values
Age (years) −6.644 0.476 −19.444 0.624 −18.007 0.101
BF% 16.230 < 0.001** 18.786 < 0.001** 15.728 < 0.001**
FFM (kg) 35.225 < 0.001** 30.010 < 0.001** 29.110 < 0.001**
Δsupraclavicular - - - - 150.493 0.285
Δsternum - - - - −30.478 0.666
Open in a new tab

Table 4 shows predictors of change in supraclavicular and sternum surface temperature between room temperature and cold exposure. Δsternum and Δsupraclavicular were added to the models with Δsupraclavicular and Δsternum as the dependent variable, respectively. No significant predictors of change in supraclavicular or sternum surface temperature were found except for Δsternum significantly predicting Δsupraclavicular and Δsupraclavicular significantly affecting Δsternum.

Table 4.

Multiple linear regression analysis testing for predictors of SupraTempC and SternTempC. * indicates significant predictor of dependent variable at P < 0.05. BF%: body fat percentage, FFM: fat free mass.

Δsupraclavicular
Adj. R2 = 0.0002
p = 0.3978		 Δsupraclavicular
Adj. R2 = 0.0794
p = 0.0836		 Δsternum
Adj. R2 = −0.0331
p = 0.7526		 Δsternum
Adj. R2 = 0.0247
p = 0.265
Beta
Coef. 		P -
values 		Beta
Coef. 		P -
values 		Beta
Coef. 		P -
values 		Beta
Coef. 		P -
values
Age (years) −0.001 0.927 −0.001 0.903 0.002 0.984 0.006 0.789
BF% 0.007 0.206 0.010 0.058 0.005 0.919 −0.015 0.179
FFM (kg) −0.006 0.348 −0.005 0.448 −0.068 0.314 −0.003 0.848
Δsupraclavicular - - - - - - 0.556 0.042*
Δsternum - - 0.141 0.042* - - - -
Open in a new tab

4. Discussion

This study provides evidence for BAT activity in an adult Polynesian population in Samoa using a field-based BAT-inferring method. Surface temperature in the supraclavicular area decreased significantly less than the temperature measured on the sternum during mild cold exposure, suggesting BAT activation at NST. Heat dissipation from BAT cannot be captured through thermal imaging at the sternum given BAT’s deeper location in the body in the chest cavity. MR significantly increased after cold exposure as a potential response to BAT thermogenesis. Additionally, more than 50% of the sample saw an increase in MR after cold exposure, further highlighting that MR may increase upon BAT activation. The BAT-inferring methodology used for this study does not allow for individual assessment of BAT thermogenesis. However, this field-based approach provides initial evidence of BAT activity in an adult Polynesian population, which could lead to individualized BAT measuring protocols, such as PET/CT scanning technology, among participants from warm-climate environments in the future.

4.1. Body Composition and Inferred BAT Activity in Samoans

The study sample was generally categorized as having obesity following Samoan-specific BMI equations (BMI > 32kg/m2). Females were heavier and had significantly greater BMIs and BF% compared to males, as found in prior work among contemporary Samoans (Hawley et al., 2014; Heinsberg et al., 2023). FFM and BF% were positively associated with MRRT and MRC but not with changes to measured surface temperatures. Both FFM and BF% were used in multivariate models testing for predictors of BAT activity as FFM is generally found to be a strong predictor of metabolic rate, an association replicated in the present study, while BF% is often found to be inversely correlated with BAT activity – albeit in leaner samples (BMI < 25kg/m2) compared to Samoan participants in this study (Cypess et al., 2009; Saito et al., 2009; van Marken Lichtenbelt et al., 2009; Nielsen et al., 2000; Ravussin et al., 1986; Yoneshiro et al., 2011b). In similar work among the cold-adapted Yakut of Siberia (males: mean BF% = 21.4, mean FFM = 55.8 kg; females: mean BF% = 33.2, mean FFM = 41.0 kg), a positive association was found between BF% and changes in supraclavicular temperatures after mild cold exposure, suggesting that greater fat reserves represent greater availability of fuel for BAT thermogenesis in the Yakut (Levy et al. 2018). This association, directly linking BAT thermogenesis and BF% was not reproduced in this Samoan sample, suggesting different potential fuel sources for BAT activation in Samoans. The expected association between BAT thermogenesis and BF% may become more apparent in a larger sample size.

Neither Δsupraclavicular nor Δsternum were associated with MRC in regression models, and their inclusion weakened the model, albeit only slightly. It is important to note these effects of Δsupraclavicular and Δsternum on the model as they potentially suggest that changes in temperature indicative of BAT activity may not be contributing to increased NST in this sample. Instead, instances of deep shivering which may not be detected through visual examination or by the participant may have contributed to increases in MR during cold exposure. However, given the relatively small sample size, the addition of extra covariates to the model reduces the power of the statistical analysis, possibly contributing to a reduced weakness of the model. Repeating the regression analyses in a larger sample in the future should clarify the importance of the role of Δsupraclavicular nor Δsternum as predictors of MRC.

None of the independent variables of interest (age, BF%, FFM) were correlated with ΔMR, thus no statistical model was created. In existing literature, a reduction in MR with age is well documented and believed to be in part the result of a decline in FFM with age, albeit none of these studies looked at within participant change in MR (Fukagawa et al., 2010; Lazzer et al., 2010; Piers et al., 1998). The results of this study do not support evidence that changes in MR after cold exposure may be affected by age as observed in previous work (Cypess et al., 2009; Yoneshiro et al., 2011a). The present study sample was relatively young and homogenous in age, which may not provide enough variation to demonstrate the full effects of age on changes in MR upon possible BAT activation. Δsupraclavicular and Δsternum were also not significant predictors of change in MR, which suggests that BAT activity may not be associated with changes in metabolic rate during cold exposure Given the relatively small sample size and the associated limited power for analysis of multiple covariates, results were not included in the table.

Neither Δsupraclavicular nor Δsternum were significantly different by sex, suggesting that males did not differ from females in the change in surface temperature experienced after cold exposure. While inconsistent with previous findings suggesting that women have higher rates of BAT activity, especially in middle age and older, compared to men, this association may be due to differences in body composition observed in this sample (Bredella, 2017; van Marken Lichtenbelt et al., 2009; Pfannenberg et al., 2010). Female participants were significantly heavier and had higher BF% which, given prior research results, may have contributed to inhibited heat dissipation from BAT (Saito et al., 2009; Vijgen et al., 2011; Matsushita et al., 2014).

4.2. Increased MR as a Result of BAT Thermogenesis

MR significantly increased between exposures for the sample. While not all individuals experienced increases in MRRT and MRC, 61% had higher MR after cooling. Among those who experienced increases in MR between exposures, average measurements were similar to those observed in female Finnish reindeer herders and overweight adults from North America (Niclou and Ocobock, 2022; Ocobock et al., 2022).

Previous investigations into changes in MR after cooling have provided differing results, which may be reflective of differences in inter-population variation in MR and BAT thermogenesis. In a temperate New York population for which measurements were taken both in the winter and in the summer, MR significantly increased after 30 minutes of mild cold exposure irrespective of season (Niclou and Ocobock, 2022). In Finnish reindeer herders, MR increased significantly after cooling. Interestingly, women displayed significantly greater room temperature MR measurements compared to men. In turn, 15% of the male sample saw declines in MR after cold exposure (Ocobock et al., 2022). In the similarly cold-adapted Yakut of Siberia, the sample as a whole saw no significant changes in MR after mild cold exposure. By sex, however, MR in men decreased significantly, while no significant changes were noted in women (Levy et al. 2018). In a study measuring energy expenditure and resting MR in a small, all-male sample exposed to chronic cold, energy expenditure did not change while MR decreased slightly after cold acclimation (Hanssen et al., 2016). Given that our participant group lives in a fairly consistent warm climate, the greater change in MR after cold exposure is unsurprising as previous work showed an increase in MR in non-acclimatized individuals during the summer but no changes in MR in the same sample after several months of acclimatization to cold in the winter (Niclou and Ocobock, 2022).

For the 40% of the study sample who experienced a decrease in MR after cooling, this may be due to vasoconstriction responses such as habituation or the Q10 effect. The Q10 effect is the ratio between changes in temperature and their direct association with changes in metabolic rate. Mild cold exposure may lead to increased vasoconstriction of vessels near the surface of the body resulting in reduced MR (i.e., Q10 effect). Simultaneously, MR at the core of the body may increase as a response to cold exposure. If the peripheral Q10 effects outweigh the changes in core MR, the overall measured MR may decrease after cooling (Barany, 1967; Bennet, 1985; Wakabayashi et al., 2017). Habituation may also result in blunted vasoconstriction and metabolic rate response (Young, 1996; Mäkinen et al., 2004).

A habituation or acclimatization response to cold exposure resulting in a blunted MR response despite increases in BAT activity has been previously observed when comparing BAT activity between seasons within a population (Niclou and Ocobock, 2022). While lifestyle questions such as inquiring about day-to-day cold exposure were not recorded for this specific sample, habituation is unlikely to be the cause for a lack of change in MR, given the lack of large seasonal variation, independent of behavioral patterns such as fishing or exposure to air conditioning, in temperatures in Samoa. The 30 minutes exposure to cold during the cooling protocol may have contributed to the Q10 effect, especially since only the last 10 minutes of the data collected during the cooling phase were used for data analysis. Future work should look at the variation in MR during the cooling period to determine if the Q10 effect may have contributed to a portion of the sample experiencing no changes in MR measurements at the end of cold exposure.

4.3. Heat Dissipation and NST

Average changes in supraclavicular temperature and the significant decreases in supraclavicular temperature between exposures were similar to those observed in studies using PET/CT-scanning (van der Lans et al., 2015). Supraclavicular measurements were also similar to those recorded in a New York sample taken across seasons and using the same thermal imaging and cooling suit methodology as the present study (Niclou and Ocobock, 2022). Average Δsupraclavicular in a Siberian population was significantly smaller, however, and the decrease in supraclavicular temperature between room temperature and mild cold exposure was not significant in the Yakut (Levy et al., 2018). Among Finnish reindeer herders, another cold-climate group, decreases in supraclavicular temperature were significant after cooling, aligning with the present Samoan sample (Ocobock et al., 2022). While smaller Δsupraclavicular may suggest greater thermogenesis in BAT-locations, Δsternum was significantly greater than Δsupraclavicular in the present Samoan sample, as well as in the cold-adapted Siberian and Finnish samples (Levy et al., 2018; Ocobock et al., 2022).

Larger temperature drops in non-BAT locations, such as the sternum, compared to the supraclavicular area were also observed in PET/CT-scanning studies in individuals with greater BAT activity (van der Lans et al., 2015; Chondronikola et al., 2016). These significant differences in heat dissipation between known BAT deposit areas and those without detectable BAT in addition to increases in MR at mild cold exposure suggest BAT activation promotes NST in this Polynesian sample.

4.3. BAT Activity Among Warm-climate Populations

Given its thermogenic capacities, BAT has been proposed to play an important role in human cold adaptation as observed in circumpolar populations, while simultaneously potentially benefitting their metabolic health (c.f. Ocobock and Niclou, 2022). The presence of BAT thermogenesis in Samoans and other warm-climate and temperate-climate populations point towards the presence of BAT across all populations, irrespective of their current climate (Monfort-Pires et al., 2021; Niclou and Ocobock, 2022; Saito et al., 2009; Vijgen et al., 2011). Along with cultural adaptations, BAT may have aided in the migration of populations across the globe from circumpolar regions to the tropics by providing thermogenic protection during bouts of cold exposure. Sub-Artic populations experience month-long cold periods, while desert populations often endure cold nighttime temperatures and tropical populations may experience cold bouts when wet during the rainy season or simply when crossing bodies of water for food or travel. Varying levels of cold stress impact most populations, supporting the need and retention of BAT activated NST. The existing body of work ranging from assessing BAT activity among temperate climate groups (van der Lans et al., 2015 Saito et al., 2009; Vijgen et al., 2011), to comparing BAT activity between temperate and cold-climate populations (Levy et al., 2022), and providing evidence for acclimatization through seasonal changes in BAT activated NST (Niclou and Ocobock, 2022; Yoneshiro et al., 2016) highlights the breadth of roles BAT plays across different populations and climates. BAT thermogenesis may potentially be a fairly old trait, retained among humans as a measure of protection against hypothermia during cold stress. The uncoupling protein associated with BAT thermogenesis is found across different mammal species, even if not highly conserved in many of them (Gaudry and Campbell, 2017; Speakman, 2019). In humans, it may have been retained given its repeated need by humans as they migrate through different climates and encounter different degrees of cold stress. These variations in environmental exposures result in warm-climate populations having some kind of BAT presence despite the current reduced need of a cold adaptive trait in the population.

Like our findings, a study investigating the effects of dietary olive oil intake on BAT activity among a sample from Brazil, provides further support for BAT thermogenesis in a warm-climate sample (Monfort-Pires et al., 2021). While evidence to date demonstrates that BAT can be absent in individuals, prior study outcomes support the presence and thermogenic activation of BAT across populations exposed to varying instances and degrees of cold stress. Future work should aim to further investigate BAT activity among different populations with varying environmental exposures and population histories. This could help provide further evidence of BAT activity as a shared trait and help identify the variation in its metabolic and thermogenic effects between and within populations. Furthermore, non-thermogenic properties of BAT activity, such as its endocrine effects also need to be further studied among warm-climate populations as BAT’s role in cold adaptation may not be the sole reason it has been retained across different populations (Villaroya et al., 2019).

4.4. Limitations

BAT activity and NST were inferred by combining changes in MR and heat dissipation measurements in known BAT and non- BAT areas of the body. This method does not provide direct evidence of BAT activity such as observed in PET/CT-scanning studies but instead follows the standard for non-invasive BAT activity measurements set by Levy (2019). Despite not measuring direct BAT activity, measurements recorded for this study were comparable to those obtained through PET/CT-scanning analysis (Salem et al. 2015; van der Lans et al., 2015; Chondronikola et al., 2016; Yoneshiro et al., 2016).

Water temperature pumped through the cooling suit was measured before and after the 30-minute mild cold exposure and was not monitored throughout the cooling period. We noted a large range in cold exposure between participants. While it would have been useful to detect possible continuous effects of these different cold exposure temperatures on inferred BAT activity, water temperature pumped through the suit was only measured at the beginning and end of the 30-minute cold exposure period and cold exposure temperature was determined using estimates from Levy (2019), limiting the evidence for variation in temperature across the measurement period. Similarly, room temperature was collected at the beginning of the day, independent of time of data collection, using a wall thermometer. This low accuracy method limited the inclusion of room temperature as a potential variable affecting changes in inferred BAT activity between participants. Furthermore, the water flowing through the tubes of the suit may have created some interference, artificially decreasing the temperature measured due to the water being cold, with the thermal images taken of the supraclavicular and sternum areas over the suit as the skin was not exposed. These interferences may have been mitigated as the rate of water flow was constant throughout the suit and measurements were only collected once the entire suit was flooded with water.

No menstrual cycle information was collected, restricting our understanding of the possible role of menopause or perimenopause associated changes in body temperature on heat dissipation recorded in some females (Rossmanith and Ruebberdt, 2009). Future work should include menopausal and menstrual cycle data to best determine the possible changes in heat dissipation associated with these factors when inferring BAT activity from surface heat dissipation. Heart rates were not monitored during room temperature or mild cold exposure, preventing measures of anxiety that could have contributed to changes in MR. Participants were repeatedly asked about their comfort level during the protocol, ensuring minimal anxiety and shivering. Furthermore, participants were given a detailed description of the protocol and were encouraged to ask any questions or voice any concerns during the recruitment visits, possibly easing some of the worries associated with an unknown protocol. Finally, all participants had previously participated in an energetic study employing similar methods of MR measurements and were thus familiar with part of the protocol.

5. Conclusion

The present study provides evidence for possible BAT activity in a Samoan cohort living in a tropical climate. This work suggests an increase in NST as a response to cold in this sample by demonstrating increases in MR and significantly greater surface temperatures in the supraclavicular area compared to temperatures recorded at the sternum upon mild cold exposure. BAT is thus seemingly activated after prolonged mild cold exposure in this non-cold acclimated sample. The evidence of BAT activity in Samoans adds to recent findings of BAT activity in warm-climate populations, and adds important insight into the similar functioning of BAT across populations living in differing environments. Future research on BAT will help identify whether BAT NST activity is a shared trait found across all populations independent of climate and shed light on the variation in its thermogenic and metabolic effects between populations.

6. Acknowledgments

The authors would like to acknowledge the research team and the participants who made this wok possible and extend our gratitude to the Samoan Ministry of Health. The authors would also like to thank Drs. Lee Gettler, Stephanie Levy, Mark Golitko, and Nicola Hawley for their helpful feedback on earlier drafts of this paper. We are very grateful to Jessica Young and Ouyang Zhu for their help with statistical analyses. This research was made possible by funds provided by the National Science Foundation (grant #1945331), the Wenner-Gren Foundation (grant #10016), the Notre Dame Institute for Scholarship in the Liberal Arts, and the National Institutes of Health (grant #R01HL140570). Niclou is supported by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health under Award Number T32DK064584. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Footnotes

7.

Conflict of Interest

The authors declare no conflicts of interest.

8. Data Availability Statement

The data supporting the findings of this study are available from the corresponding author upon reasonable request.

9. References

  1. Andersen KL, Løyning Y, Nelms JD, Wilson O, Fox RH, & Bolstad A (1960). Metabolic and thermal response to a moderate cold exposure in nomadic Lapps. Journal of applied physiology, 15(4), 649–653. [DOI] [PubMed] [Google Scholar]
  2. Baker PT, Hanna JM, & Baker TS (1986). The Changing Samoans. Behavior and Health in Transition. [Google Scholar]
  3. Barany M. (1967). ATPase activity of myosin correlated with speed of muscle shortening. The Journal of General Physiology, 50(6), Suppl), 197–218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bennett AF (1985). Temperature and muscle. The Journal of Experimental Biology, 115(1), 333–344. [DOI] [PubMed] [Google Scholar]
  5. Bredella MA (2017). Sex differences in body composition. Sex and gender factors affecting metabolic homeostasis, diabetes and obesity, 9–27. [Google Scholar]
  6. Cannon B, & Nedergaard JAN (2004). Brown adipose tissue: function and physiological significance. Physiological reviews. [DOI] [PubMed] [Google Scholar]
  7. Chondronikola M, Volpi E, Børsheim E, Porter C, Annamalai P, Enerbäck S, ... & Sidossis LS (2014). Brown adipose tissue improves whole-body glucose homeostasis and insulin sensitivity in humans. Diabetes, 63(12), 4089–4099. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cypess AM, Lehman S, Williams G, Tal I, Rodman D, Goldfine AB, ... & Kahn CR (2009). Identification and importance of brown adipose tissue in adult humans. New England journal of medicine, 360(15), 1509–1517. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Devlin MJ (2021). Brown Adipose Tissue, Nonshivering Thermogenesis, and Energy Availability. In Evolutionary Cell Processes in Primates (pp. 131–160). CRC Press. [Google Scholar]
  10. DiBello JR, Baylin A, Viali S, Tuitele J, Bausserman L, & McGarvey ST (2009a). Adiponectin and type 2 diabetes in Samoan adults. American Journal of Human Biology: The Official Journal of the Human Biology Association, 21(3), 389–391. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. DiBello JR, McGarvey ST, Kraft P, Goldberg R, Campos H, Quested C, ... & Baylin A (2009b). Dietary patterns are associated with metabolic syndrome in adult Samoans. The Journal of nutrition, 139(10), 1933–1943. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Foster DO, & Frydman ML (1978). Brown adipose tissue: the dominant site of nonshivering thermogenesis in the rat. In Effectors of thermogenesis (pp. 147–151). Birkhäuser, Basel. [DOI] [PubMed] [Google Scholar]
  13. Fukagawa NK, Bandini LG, & Young JB (1990). Effect of age on body composition and resting metabolic rate. American Journal of Physiology-Endocrinology and Metabolism, 259(2), E233–E238. [DOI] [PubMed] [Google Scholar]
  14. Gaudry MJ, & Campbell KL (2017). Evolution of UCP1 Transcriptional Regulatory Elements Across the Mammalian Phylogeny. Frontiers in Physiology, 8, 670. 10.3389/fphys.2017.00670 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hanssen MJ, van der Lans AA, Brans B, Hoeks J, Jardon KM, Schaart G, ... & van Marken Lichtenbelt WD (2016). Short-term cold acclimation recruits brown adipose tissue in obese humans. Diabetes, 65(5), 1179–1189. [DOI] [PubMed] [Google Scholar]
  16. Harris DN, Kessler MD, Shetty AC, Weeks DE, Minster RL, Browning S, ... & TOPMed Population Genetics Working Group. (2020). Evolutionary history of modern Samoans. Proceedings of the National Academy of Sciences, 117(17), 9458–9465. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hart JS, Sabean HB, Hildes JA, Depocas F, Hammel HT, Andersen KL, ... & Foy G (1962). Thermal and metabolic responses of coastal Eskimos during a cold night. Journal of Applied Physiology, 17(6), 953–960. [DOI] [PubMed] [Google Scholar]
  18. Hawley NL, & McGarvey ST (2015). Obesity and diabetes in Pacific Islanders: the current burden and the need for urgent action. Current diabetes reports, 15(5), 1–10. [DOI] [PubMed] [Google Scholar]
  19. Hawley NL, Minster RL, Weeks DE, Viali S, Reupena MS, Sun G, ... & McGarvey ST (2014). Prevalence of adiposity and associated cardiometabolic risk factors in the samoan genome-wide association study. American Journal of Human Biology, 26(4), 491–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hawley NL, Pomer A, Rivara AC, Rosenthal SL, Duckham RL, Carlson JC, ... & McGarvey ST (2020). Exploring the paradoxical relationship of a Creb 3 regulatory factor missense variant with body mass index and diabetes among Samoans: Protocol for the Soifua Manuia (good health) observational cohort study. JMIR Research Protocols, 9(7), e17329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Heinsberg LW, Hawley NL, Duckham RL,, Pomer A, Rivara AC, Naseri T, Reupena MS, Weeks DE, McGarvey ST, Minster RL . (2023) Validity of anthropometric equation-based estimators of fat mass in Samoan adults. American Journal of Human Biology, 35: e23838. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Himms-Hagen J. (1985). Brown adipose tissue metabolism and thermogenesis. Annual review of nutrition, 5(1), 69–94. [DOI] [PubMed] [Google Scholar]
  23. Jang C, Jalapu S, Thuzar M, Law PW, Jeavons S, Barclay JL, & Ho KKY (2014). Infrared thermography in the detection of brown adipose tissue in humans. Physiological Reports, 2(11), e12167. 10.14814/phy2.12167 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. van der Lans AA, Vosselman MJ, Hanssen MJ, Brans B, & van Marken Lichtenbelt WD (2016). Supraclavicular skin temperature and BAT activity in lean healthy adults. The Journal of Physiological Sciences, 66(1), 77–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. LaMonica LC, McGarvey ST, Rivara AC, Sweetman CA, Naseri T, Reupena MS, ... & Hawley NL (2022). Cascades of diabetes and hypertension care in Samoa: Identifying gaps in the diagnosis, treatment, and control continuum–a cross-sectional study. The Lancet Regional Health-Western Pacific, 18, 100313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lazzer S, Bedogni G, Lafortuna CL, Marazzi N, Busti C, Galli R, ... & Sartorio A (2010). Relationship between basal metabolic rate, gender, age, and body composition in 8,780 white obese subjects. Obesity, 18(1), 71–78. [DOI] [PubMed] [Google Scholar]
  27. Lee P, Greenfield JR, Ho KK, & Fulham MJ (2010). A critical appraisal of the prevalence and metabolic significance of brown adipose tissue in adult humans. American Journal of Physiology-Endocrinology and Metabolism, 299(4), E601–E606. [DOI] [PubMed] [Google Scholar]
  28. Levy SB (2019). Field and laboratory methods for quantifying brown adipose tissue thermogenesis. American Journal of Human Biology, 31(4), e23261. [DOI] [PubMed] [Google Scholar]
  29. Levy SB, Klimova TM, Zakharova RN, Federov AI, Fedorova VI, Baltakhinova ME, & Leonard WR (2018). Brown adipose tissue, energy expenditure, and biomarkers of cardio-metabolic health among the Yakut (Sakha) of northeastern Siberia. American Journal of Human Biology, 30(6), e23175. [DOI] [PubMed] [Google Scholar]
  30. Levy SB, Klimova TM, Zakharova RN, Fedorov AI, Fedorova VI, Baltakhinova ME, ... & Leonard WR (2022). Brown adipose tissue thermogenesis among young adults in northeastern Siberia and Midwest United States and its relationship with other biological adaptations to cold climates. American Journal of Human Biology, e23723. [DOI] [PubMed] [Google Scholar]
  31. Lohman TG, Roche AF, & Martorell R (1988). Anthropometric standardization reference manual (Vol. 177, pp. 3–8). Champaign: Human kinetics books. [Google Scholar]
  32. Maäkinen TM, Pääkönen T, Palinkas LA, Rintamäki H, Leppäluoto J, & Hassi J (2004). Seasonal changes in thermal responses of urban residents to cold exposure. Comparative Biochemistry and Physiology - Part A, 139, 229–238. [DOI] [PubMed] [Google Scholar]
  33. van Marken Lichtenbelt WD, Vanhommerig JW, Smulders NM, Drossaerts JM, Kemerink GJ, Bouvy ND, ... & Teule GJ (2009). Cold-activated brown adipose tissue in healthy men. New England Journal of Medicine, 360(15), 1500–1508. [DOI] [PubMed] [Google Scholar]
  34. Matsushita M, Yoneshiro T, Aita S, Kameya T, Sugie H, & Saito M (2014). Impact of brown adipose tissue on body fatness and glucose metabolism in healthy humans. International Journal of Obesity, 38(6), 812–817. [DOI] [PubMed] [Google Scholar]
  35. McGarvey ST (2001). Cardiovascular disease (CVD) risk factors in Samoa and American Samoa, 1990-95. Pacific health dialog, 8(1), 157–162. [PubMed] [Google Scholar]
  36. McGarvey ST, Bausserman L, Viali S, Tufa J. (2005). Prevalence of the metabolic syndrome in Samoans. American Journal of Physical Anthropology, 40(Supplement), 14–15. [Google Scholar]
  37. Miles-Chan JL, Sarafian D, Montani JP, Schutz Y, & Dulloo AG (2014). Sitting comfortably versus lying down: Is there really a difference in energy expenditure?. Clinical nutrition, 33(1), 175–178. [DOI] [PubMed] [Google Scholar]
  38. Minster RL, Hawley NL, Su CT, Sun G, Kershaw EE, Cheng H, ... & McGarvey ST (2016). A thrifty variant in CREBRF strongly influences body mass index in Samoans. Nature genetics, 48(9), 1049–1054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Monfort-Pires M, Regeni-Silva G, Dadson P, Nogueira GA, Mueez UD, Ferreira SR, ... & Velloso LA (2022). Brown fat triglyceride content is associated with cardiovascular risk markers in adults from a tropical region. Frontiers in endocrinology, 13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Mongillo JF, & Zierdt-Warshaw L (2000). Encyclopedia of environmental science. University Rochester Press. [Google Scholar]
  41. Niclou A & Ocobock C (2022), Weather Permitting: Increased Seasonal Efficiency of Non-shivering Thermogenesis through BAT Activation in the Winter. American Journal of Human Biology, e23716. [DOI] [PubMed] [Google Scholar]
  42. Nielsen S, Hensrud DD, Romanski S, Levine JA, Burguera B, & Jensen MD (2000). Body composition and resting energy expenditure in humans: role of fat, fat-free mass and extracellular fluid. International journal of obesity, 24(9), 1153–1157. [DOI] [PubMed] [Google Scholar]
  43. Ocobock C, Soppela P, Turunen M, Stenback V, & Herzig KH (2020). Evidence for brown adipose tissue activation among male and female reindeer herders from sub-arctic Finland. American Journal of Physical Anthropology, 171, 205. [Google Scholar]
  44. Ocobock C, Soppela P, Turunen M, Stenbäck V, & Herzig KH (2022). Brown adipose tissue thermogenesis among a small sample of reindeer herders from sub-Arctic Finland. Journal of Physiological Anthropology, 41(1), 1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Ouellet V, Labbé SM, Blondin DP, Phoenix S, Guérin B, Haman F, ... & Carpentier AC (2012). Brown adipose tissue oxidative metabolism contributes to energy expenditure during acute cold exposure in humans. The Journal of clinical investigation, 122(2), 545–552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Oyama S, Arslanian KJ, Levy SB, Ocobock CJ, Fidow UT, Naseri T, & Hawley NL (2021). Feasibility of using infrared thermal imaging to examine brown adipose tissue in infants aged 18 to 25 months. Annals of Human Biology, 48(5), 374–381. [DOI] [PubMed] [Google Scholar]
  47. Parsons K. (2007). Human thermal environments: the effects of hot, moderate, and cold environments on human health, comfort and performance. CRC press. [Google Scholar]
  48. Petchey FJ (2001). Radiocarbon determinations from the Mulifanua Lapita site, Upolu, western Samoa. Radiocarbon, 43(1), 63–68. [Google Scholar]
  49. Pfannenberg C, Werner MK, Ripkens S, Stef I, Deckert A, Schmadl M, ... & Stefan N (2010). Impact of age on the relationships of brown adipose tissue with sex and adiposity in humans. Diabetes, 59(7), 1789–1793. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Piers LS, Soares MJ, McCormack LM, & O’Dea K (1998). Is there evidence for an age-related reduction in metabolic rate?. Journal of Applied Physiology, 85(6), 2196–2204. [DOI] [PubMed] [Google Scholar]
  51. Population & Demography Indicator Summary. Samoa Bureau of Statistics - Population & Demography. (n.d.). Retrieved November 22, 2021, from https://web.archive.org/web/20190403053500/http:/www.sbs.gov.ws/index.php/population-demography-and-vital-statistics. [Google Scholar]
  52. Ravussin E, Lillioja S, Anderson TE, Christin L, & Bogardus C (1986). Determinants of 24-hour energy expenditure in man. Methods and results using a respiratory chamber. The Journal of clinical investigation, 78(6), 1568–1578. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Rossmanith WG, & Ruebberdt W (2009). What causes hot flushes? The neuroendocrine origin of vasomotor symptoms in the menopause. Gynecological Endocrinology, 25(5), 303–314. [DOI] [PubMed] [Google Scholar]
  54. Saito M, Okamatsu-Ogura Y, Matsushita M, Watanabe K, Yoneshiro T, Nio-Kobayashi J, ... & Tsujisaki M (2009). High incidence of metabolically active brown adipose tissue in healthy adult humans: effects of cold exposure and adiposity. Diabetes, 58(7), 1526–1531. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Salem V, Engbeaya CI, Jayasinghe SM, Thomas D, Coello C, Comninos A, ... & Dhillo W (2015). Thermal imaging as a novel non-invasive method to measure human brown adipose tissue activity in humans. In Society for Endocrinology BES 2015 (Vol. 38). BioScientifica. [Google Scholar]
  56. Samoa Meteorology Division (2018) Climate of Samoa. Retrieved from http://www.samet.gov.ws/index.php/climate-of-samoa.
  57. Sell H, Deshaies Y, & Richard D (2004). The brown adipocyte: update on its metabolic role. The international journal of biochemistry & cell biology, 36(11), 2098–2104. [DOI] [PubMed] [Google Scholar]
  58. Speakman JR (2019). Fifty shades of brown: The functions, diverse regulation and evolution of brown adipose tissue. Molecular aspects of medicine, 68, 1–5. [DOI] [PubMed] [Google Scholar]
  59. Tam CS, Lecoultre V, & Ravussin E (2012). Brown adipose tissue: mechanisms and potential therapeutic targets. Circulation, 125(22), 2782–2791. [DOI] [PubMed] [Google Scholar]
  60. Trayhurn P. (2017). Origins and early development of the concept that brown adipose tissue thermogenesis is linked to energy balance and obesity. Biochimie, 134, 62–70. [DOI] [PubMed] [Google Scholar]
  61. Vijgen GH, Bouvy ND, Teule GJ, Brans B, Schrauwen P, & van Marken Lichtenbelt WD (2011). Brown adipose tissue in morbidly obese subjects. PloS one, 6(2), e17247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Villarroya J, Cereijo R, Gavaldà-Navarro A, Peyrou M, Giralt M, & Villarroya F (2019). New insights into the secretory functions of brown adipose tissue. Journal of Endocrinology, 243(2), R19–R27. [DOI] [PubMed] [Google Scholar]
  63. Wakabayashi H, Nishimura T, Wijayanto T, Watanuki S, & Tochihara Y (2017). Effect of repeated forearm muscle cooling on the adaptation of skeletal muscle metabolism in humans. International Journal of Biometeorology, 61(7), 1261–1267. [DOI] [PubMed] [Google Scholar]
  64. Yoneshiro T, Aita S, Matsushita M, Kameya T, Nakada K, Kawai Y, & Saito M (2011b). Brown adipose tissue, whole-body energy expenditure, and thermogenesis in healthy adult men. Obesity, 19(1), 13–16. [DOI] [PubMed] [Google Scholar]
  65. Yoneshiro T, Aita S, Matsushita M, Okamatsu-Ogura Y, Kameya T, Kawai Y, ... Saito M (2011a). Age-related decrease in cold-activated brown adipose tissue and accumulation of body fat in healthy humans. Obesity, 19, 1755–1760. [DOI] [PubMed] [Google Scholar]
  66. Yoneshiro T, Matsushita M, Nakae S, Kameya T, Sugie H, Tanaka S, & Saito M (2016). Brown adipose tissue is involved in the seasonal variation of cold-induced thermogenesis in humans. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, 310(10), R999–R1009. [DOI] [PubMed] [Google Scholar]
  67. Young AJ (1996). Homeostatic responses to prolonged cold exposure: Human cold acclimatization. In Fregly MJ & Blatteis CM (Eds.), Handbook of Physiology: Environmental Physiology (pp. 419–438). Bethesda, MD: American Physiological Society. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data supporting the findings of this study are available from the corresponding author upon reasonable request.

RESOURCES