Sage Journals: Discover world-class research

Abstract

Background:

Dietary and circulating branched-chain amino acids (BCAA) exhibited divergent associations with obesity, and this study evaluated the independent and joint associations of dietary intakes and circulating concentrations of BCAA with obesity.

Methods:

Energy-adjusted dietary and log-transformed circulating BCAAs from 540 Filipino women were categorized into quartiles and dichotomized by the median, low(<median) or high(≥median) to examine the joint association as follows: “low dietary and low circulating”BCAA, “low dietary and high circulating”BCAA, “high dietary and low circulating” BCAA and “high dietary and high circulating”BCAA. Multivariable-adjusted logistic regression was used to compute the odds ratio(OR) and 95% confidence intervals(CI) of obesity(BMI ≥25kg/m²) at a two-sided P < 0.05.

Results:

The odds of being obese by increasing quartile of dietary total BCAA were 1.00, 0.84 (0.49, 1.43), 0.68 (0.39, 1.19), and 0.56 (0.31, 1.02; P for trend = .05). However, the odds of being obese by increasing quartile of circulating total BCAA were 1.00, 2.62 (1.38, 5.00), 3.43 (1.83, 6.44), and 5.97 (3.19, 11.16; P for trend < .0001). The ORs (95% CI) for being obese by categories of the joint total dietary and circulating BCAA (using low dietary and low circulating BCAA as reference) were 2.06 (1.19, 3.56) for low dietary and high circulating, 0.54 (0.28, 1.02) for high dietary and low circulating and 1.84 (1.03, 3.28) for high dietary and high circulating (P for interaction = .61).

Conclusion:

Higher circulating BCAA profiles, independent of dietary BCAA intake levels, were associated with a higher prevalence of obesity, with no evidence of interaction in the divergent independent associations between dietary and circulating BCAAs and obesity.

Keywords

branched-chain amino acids body mass index waist circumference metabolomics diet

Introduction

Obesity is a prominent risk factor for multiple metabolic disorders and a significant public health problem¹ deserving exigent public health intervention, given the recently reported annual 0.48% increment in obesity-associated disability-life adjusted years between 2000 and 2019,² the predictable 4 trillion dollar upsurge in the economic implications of obesity management by 2035,³ the 39.8% predicted increment in obesity burden between 2020 and 2035,² and the potentially enervating impact on population health, life expectancy and human capital development worldwide.⁴

The influence of environmental exposures, including diet and lifestyle, and the changes in the metabolome in the pathophysiology of obesity manifestation have been well documented.^5
-9 Branched-chain amino acids (BCAA), including isoleucine, leucine and valine, are essential amino acids derived from protein sources, accounting for a large portion of the essential amino acids supply.¹⁰ They are critical for metabolism,^11,12 with a signature in the diet and metabolome that makes them potentially exhibit complicated associations in the pathophysiology of obesity manifestations. Several epidemiological reports have documented the independent associations of dietary^13
-15 and circulating^16
-18 BCAAs with obesity in diverse populations. Similarly, some studies have examined the associations of dietary BCAA intake and obesity with its associated complications,^19
-27 with most reports suggesting that higher dietary BCAA intakes were inversely linked with obesity, but this was not the case for obesity-associated complications. For example, a meta-analysis suggested that dietary BCAA intake exhibited an independent inverse relationship with obesity and a direct relationship with type 2 diabetes (T2D).¹⁴ Likewise, evidence suggesting elevated BCAA profiles as metabolic signatures in predicting the onset of obesity and its associated complications has been reported.^{16
-18,28
-41} Additionally, a meta-analysis of previously published studies revealed broad changes in amino acid signatures in the metabolome, including elevated BCAA levels, which correlated more closely with obesity events.¹⁷

BCAAs are principally acquired from diets,^10
-12 even though they can also be synthesized in minute proportions in the body,^11,42,43 and several studies that have explored the independent associations of dietary BCAA intake^13
-15 and circulating levels of BCAAs^16
-18 with obesity revealed discordant associations. For example, most studies have reported that dietary BCAA consumption exhibits an inverse association with obesity manifestation. However, serum BCAA has been reported to be positively associated with obesity. These differences in findings appear unclear, given that BCAA are from dietary consumption and absorbed into the vasculature and metabolized primarily in the liver and muscles. Additionally, most of these studies are limited, providing minimal information on joint associations between dietary and circulating BCAA levels within a single epidemiological dataset on the onset of metabolic disorders and chronic diseases. There is little information on the implications of the joint associations of dietary and circulating BCAAs in obesity manifestations. To date, a longitudinal study has reported that higher dietary and circulating BCAA levels are linked with incident T2D among women with a history of gestational diabetes in the United States.⁴⁴ There is limited information on how these associations affect obesity, given that obesity can lead to insulin resistance and, by extension, T2D.

It is to this effect that it was necessary to examine the joint relationships of dietary and circulating BCAAs within a single epidemiological population, because earlier epidemiological reports have only explored the associations either in diet or serum without taking into account the joint association. Additionally, the findings from this study will provide evidence-based information for relevant dietary guidelines and advisories on dietary BCAA consumption and its implications for health outcomes. Clarifying these associations is vital for understanding the significance of BCAA in the aetiology of obesity and its associated complications. Furthermore, the findings of this study would be pivotal in setting an agenda for dietary interventions to guide public health priorities for obesity prevention and management. It is also worthwhile to conduct such studies to extend the scope of understanding the diet-metabolome-disease relationship, thereby seeking answers for precision medicine approaches to obesity treatment. Similarly, such information would support efforts to promote appropriate guidance and advisories for lifestyle modification for the prevention of chronic diseases in diverse populations. Therefore, this study investigated the independent and joint relationships of dietary and circulating BCAA with obesity.

Materials and Methods

Study Population

The Filipino Women’s Diet and Health (FiLWHEL) is an ongoing population-based study that began in 2014 with the primary goal of evaluating the magnitude and lifestyle risk factors of non-communicable diseases among immigrant Filipino women (aged 19 years and above) in Korea. The design and methods of the FiLWHEL have been reported elsewhere.⁴⁵ Convenience sampling was adopted to recruit participants across several provinces in Korea. While the first wave of enrollment was between 2014 and 2016, the second wave began in 2019 (but was provisionally suspended in 2020 due to the COVID-19 pandemic) and continued in 2021. By May 2022, 700 Filipino women had been recruited at baseline through an interviewer-administered questionnaire, administered by trained staff eloquent in the Filipino language under the guidance of the principal investigators, to gather information on sociodemographics, lifestyle, medical history, acculturation, and quality of life, among other factors. Physical examinations were conducted by trained staff, who collected biospecimens, including blood samples and toenails, using a uniform standard protocol across all sites. All data collected was reviewed, clarified on-site, and double-checked for consistency before data entry. Overall, 540 of 700 (77.1%) participants were included in the final analysis of this study, after 160 participants were excluded for various reasons described in Figure S1. The Ethics review committees of the Sookmyung Women’s University (SMWU-1311-BR-012), Seoul National University (1904/002-011), and Hayang University (HYU-2020-03-008-6) approved the study, and participants provided informed consent.

Dietary Assessment of Branched-Chain Amino Acids

Dietary intake was evaluated using 1-day or 2-day 24-hour dietary recalls, where participants reported the quantity and portions of foods consumed the previous day using food models, including household measures, weight/volume, portion sizes, and standard units. Nutrient data was estimated from the foods and drinks consumed using the computer-aided analysis program (Can-Pro 4.0) for professionals by the Korean Society of Nutrition, Seoul, Korea,⁴⁶ food composition tables of the Food and Nutrition Research Institute of the Philippines (for Filipino ingredients),⁴⁷ Korean Rural Development Administration⁴⁸ and the United States Department of Agriculture database.⁴⁹ Dietary BCAA intakes, including isoleucine, leucine and valine, were estimated in grams/day, and total BCAA (tBCAA) was valued as the sum of all BCAA. Total and individual dietary BCAA intakes were adjusted for energy intake using the residual method⁵⁰ and classified into quartiles to ensure a practicable number of participants for statistical comparison.

Metabolite Profiling of Branched-Chain Amino Acids

Serum samples, derived from fasting blood samples (voluntarily provided by participants after an 8-hour overnight fast at recruitment) drawn through venipuncture by a trained phlebotomist, were processed and transported in cold packs for storage at −80°C for further analysis between 2014 and 2022. The serum samples were retrieved, sub-aliquoted, and dispatched to Nightingale Health Ltd, Helsinki, Finland, on dry ice for metabolite profiling using high-throughput proton (¹H) nuclear magnetic resonance (NMR) coupled with mass spectrometry. The ¹H NMR evaluated multiple low-molecular metabolites, including isoleucine, leucine, valine, and tBCAA, in millimoles per litre. Details of the blood collection and processing, as well as the principles, extraction, determination, quantification, and normalization of the ¹H NMR method, can be found elsewhere.^51
-55 Additionally, intra-assay reproducibility showed outstanding coefficients of variation, with values of 4.3% for isoleucine, 3.6% for leucine, and 3.5% for valine. Furthermore, circulating BCAA profiles were log-transformed and classified into quartiles to ensure a practicable number of participants for statistical comparison.

Joint Dietary and Circulating BCAA Exposure

EEnergy-adjusted dietary BCAA intakes and log-transformed circulating BCAA concentrations from the serum were dichotomized by the median value of each BCAA distribution as low (<median) or high (≥median), and participants were categorized as follows to clarify the joint association of both BCAA profiles; “low dietary and low circulating” BCAA (< median of both dietary and circulating BCAA), “low dietary and high circulating” BCAA (< median dietary BCAA and ≥ median circulating BCAA), “high dietary and low circulating” ( ≥ median dietary BCAA and < median circulating BCAA), and “high dietary and high circulating” BCAA (≥ median of both dietary and circulating BCAA). Participants with “low dietary and low circulating” BCAA were applied as a reference for statistical comparison with other categories to discern the significance of the joint exposure of both dietary and circulating BCAA in obesity manifestations.

Anthropometry and Obesity Ascertainment

According to standard protocol, the weight of participants was measured in kilograms using bioelectric impedance equipment (In Body 620, Biospace Company Limited, Seoul, Korea). Height and waist circumference (WC) were measured to the nearest 0.1 cm while standing without shoes using a stretch-resistant tape measure. WC was measured at the midpoint between the lowest border of the rib cage and the uppermost lateral border of the right iliac crest. Body mass index (BMI) was calculated as weight (in kilograms) divided by the square of height (in meters). General obesity was defined as a BMI ⩾ 25 kg/m², and abdominal obesity as WC ⩾ 80 cm.⁵⁶

Assessment of Other Covariates

Participants’ age and length of stay were reported in years. The highest education level was reported and categorized as “elementary and high school” and “college education and above.” Employment status was derived from self-reported current occupation and classified as “no” or “yes.” Smoking status was categorized as “yes,” where participants described smoking 100 cigarettes or more in a lifetime, else “no.” Alcohol use was defined as self-reported current use of any form of alcoholic drinks in the last 1 year, and physical activity was described as “vigorous” if participants reported spending at least an average of at least an hour daily of vigorous physical activity or else “no.”⁵⁷ The family history of type 2 diabetes was self-reported as ’no’ or “yes.” Additionally, type 2 diabetes was defined as one of the following conditions: fasting blood glucose ⩾ 126 mg/dl, glycated haemoglobin ⩾ 6.5%, a history of diagnosed type 2 diabetes, or current use of blood glucose-lowering medications.⁵⁸ Blood pressure was measured using a Sphygmomanometer (Mercury Sphygmomanometer and OMRON; HEM-7111, Omron Healthcare, Kyoto, Japan for participants recruited between 2014-2016 and 2019-2022, respectively) at intervals, and the average of the last 2 measurements was used to describe hypertension as any of the following conditions: systolic blood pressure ⩾ 140 mmHg, diastolic blood pressure ⩾ 90 mmHg, diagnosis by health personnel or the use of blood pressure lowering drugs.⁵⁹

Statistical Analysis

Characteristics of participants were presented by obesity status (BMI ⩾ 25kg/m² or WC ⩾ 80cm) and quartile of energy-adjusted dietary intakes and log-transformed circulating concentrations of tBCAA, using frequencies (percentages) for categorical data or mean±standard deviation (SD) or median (interquartile range – IQR) depending on the data distribution for continuous variables. Multivariable-adjusted logistic regression was applied to determine the odds ratio (OR) and 95% confidence interval (CI) of obesity by quartile of energy-adjusted dietary intakes and log-transformed circulating BCAA profiles independently. Covariates were included in the adjusted models based on empirical understanding in the literature. Changes in ORs were evaluated when determining which variables to include in the final model. The first model was adjusted for age (continuous, years) only. Model 1 was adjusted for years of stay in Korea (continuous, years), highest education completed (elementary and high school, college education and above), employment status (no, yes), ever smoked (no, yes), current alcohol use (no, yes), vigorous physical activity (no, yes), family history of T2D (no, yes), menopausal status (no, yes), hypertension (no, yes), and energy intakes (continuous, kcal/day) in addition to age (continuous, years). Model 2 was limited to the independent associations of log-transformed circulating BCAA, adjusted for energy-adjusted dietary BCAA intakes (continuous, grams/day), in addition to the covariates in Model 1. A test for trend was conducted by assigning the median value of each quartile as in a continuous model. Similarly, multivariable-adjusted OR and 95% CI for obesity by categories of joint association of energy-adjusted dietary and log-transformed circulating BCAA profiles; “low dietary and high circulating,” “high dietary and low circulating,” and “high dietary and high circulating” were estimated using participants with “low dietary and low circulating” BCAA as reference, adjusting for similar covariates in the independent associations. Furthermore, generalized linear regression was applied to estimate the least-squares (LS) means and 95% CI of BMI and WC profiles across the quartiles of independent and joint associations of energy-adjusted dietary and log-transformed BCAA profiles, adjusting for the same covariates in the logistic regression models. The likelihood ratio test was applied for the interaction test, comparing nested models with cross-product terms of energy-adjusted dietary and log-transformed circulating BCAA concentrations with the original model, excluding the term. Type I error for P-values in the regression model was assessed using Bonferroni’s correction, and all statistical analyses were conducted at a two-sided P < .05 using SAS 9.4 (SAS Institute Inc., Cary, NC, USA).

Results

Characteristics of Participants

Characteristics of participants by obesity status revealed that 170 (31.5%) had a BMI ⩾ 25 kg/m² and 233 (43.2%) had a waist circumference (WC)⩾ 80 cm (Table S1). The mean age and length of stay in Korea were generally higher among participants with obesity, with typically poor lifestyle characteristics. Participants’ characteristics by quartile of energy-adjusted tBCAA intake (Table S2) suggest that the prevalence of BMI ⩾ 25 kg/m² among those in the fourth and first quartile was 35 (25.9%) and 48 (35.6%), respectively, with a similar trend for WC ⩾ 80 cm; 58 (43.0%) and 61 (45.2%), respectively. The employment rate, level of vigorous physical activity, and hypertension rate were lower, but the prevalence of smoking, alcohol use, and family history of type 2 diabetes was higher among participants in the third or fourth quartile compared to those in the first or second quartile of energy-adjusted tBCAA intake. Overall, the median (IQR) energy-adjusted dietary tBCAA in grams/day was 3.30 (2.59, 3.75) – first quartile, 5.41 (4.83, 5.96) – second quartile, 8.01 (7.37, 8.78) – third quartile, and 12.09 (10.42, 14.41) – fourth quartile. However, the mean (SD) energy intakes in kcal/day across the quartiles of dietary tBCAA intakes were 2084.2 ± 658.4, 1862.4 ± 523.6, 1712.4 ± 508.0, and 1426.9 ± 480.9 for the first, second, third, and fourth quartiles, respectively. Analogous trends were observed across the quartiles of energy-adjusted isoleucine, leucine, and valine intakes.

In addition, the participant characteristics by quartile of log-transformed circulating tBCAA (Table S3) revealed the prevalence of BMI ⩾ 25 kg/m² among those in the first and fourth quartiles were 18 (13.3%) and 66 (48.9%), respectively, with a similar observation for WC ⩾ 80 cm; 33 (24.4%) and 88 (65.2%) respectively. Smoking rates, prevalence of type 2 diabetes and hypertension were higher among participants in the third or fourth quartile compared to the first quartile. The mean (SD) circulating tBCAA in mmol/L was 0.28 ± 0.02 – first quartile, 0.32 ± 0.01 – second quartile, 0.36 ± 0.01 – third quartile, and 0.43 ± 0.05 – fourth quartile. Energy intake was even across the quartiles of log-transformed circulating tBCAA. The trends were similar across the quartiles of log-transformed concentrations of circulating isoleucine, leucine, and valine.

The characteristics of participants, as determined by the joint association of energy-adjusted dietary intake and log-transformed circulating concentrations of tBCAA (Table S4), revealed that the mean age, length of stay in Korea, and blood pressure profiles were not significantly different. Participants with high circulating levels of tBCAA tended to have higher levels of metabolic markers, including BMI, WC, fasting glucose, hemoglobin, diabetes, and blood pressure, compared to those with low levels. The median (IQR) energy-adjusted tBCAA intakes in grams/day were 4.12 (3.34, 5.37), 4.44 (3.28, 5.48), 9.55 (7.63, 12.52), and 9.29 (8.39, 11.60) for “low dietary and low circulating” tBCAA, “low dietary and high circulating” tBCAA, “high dietary and low circulating” tBCAA, and “high dietary and high circulating” tBCAA, respectively. Also, the means ± SDs tBCAA concentration in mmol/L were 0.31 ± 0.03 – “low dietary and low circulating” tBCAA, 0.40 ± 0.04 – “low dietary and high circulating” tBCAA, 0.30 ± 0.03 – “high dietary and low circulating” tBCAA, and 0.40 ± 0.06 – “high dietary and high circulating” tBCAA.

Relationship of Energy-Adjusted Dietary BCAA Intake with Obesity

The OR (95% CI) of obesity by quartiles of energy-adjusted dietary BCAA intake (Table 1) in the age-adjusted regression model revealed that higher energy-adjusted dietary BCAA intake was inversely associated with the odds of general obesity. The multivariable-adjusted regression model revealed a similar trend; tBCAA [1.00, 0.84 (0.49, 1.43), 0.68 (0.39, 1.19), and 0.56 (0.31, 1.02; P for trend = .05)], isoleucine [1.00, 0.79 (0.46, 1.36), 0.73 (0.42, 1.27), and 0.56 (0.31, 1.01; P for trend = .06)], leucine [1.00, 0.84 (0.49, 1.43), 0.77 (0.44, 1.33), 0.54 (0.29, 0.98; P for trend = .04)], and valine [1.00, 0.71 (0.41, 1.21), 0.67 (0.39, 1.16), 0.55 (0.31, 0.99; P for trend = .06]. A similar but statistically insignificant trend was observed with abdominal obesity after adjusting for similar covariates.

Table 1.

Odds Ratios and 95% Confidence Intervals for Obesity^a by Quartile of Energy-Adjusted Dietary BCAA Intakes.^b

	General obesity (body mass index ⩾ 25 kg/m²)					Abdominal obesity (waist circumference ⩾ 80 cm)
BCAA profiles	Q1	Q2	Q3	Q4	P for trend^c	Q1	Q2	Q3	Q4	P for trend^c
Total BCAA
Cases/total	48/135	46/135	41/135	35/135		61/135	58/135	56/135	58/135
Dietary; median (IQR) g/d	3.30 (2.58, 3.75)	5.41 (4.83, 5.96)	8.01 (7.37, 8.78)	12.09 (10.42, 14.41)		3.30 (2.58, 3.75)	5.41 (4.83, 5.96)	8.01 (7.37, 8.78)	12.09 (10.42, 14.41)
Circulating; mean (SD) mmol/L	0.35 (0.06)	0.36 (0.06)	0.36 (0.07)	0.34 (0.06)		0.35 (0.06)	0.36 (0.06)	0.36 (0.07)	0.34 (0.06)
Age-adjusted	1.00	0.86 (0.52, 1.44)	0.73 (0.44, 1.23)	0.61 (0.36, 1.03)	.05	1.00	0.83 (0.51, 1.35)	0.79 (0.48, 1.29)	0.88 (0.54, 1.43)	.67
Model 1	1.00	0.84 (0.49, 1.43)	0.68 (0.39, 1.19)	0.56 (0.31, 1.02)	.05	1.00	0.78 (0.47, 1.30)	0.70 (0.41, 1.17)	0.72 (0.42, 1.26)	.28
Isoleucine
Cases/total	48/135	45/135	42/135	35/135		62/135	56/135	59/135	56/135
Dietary; median (IQR) g/d	0.83 (0.66, 0.93)	1.38 (1.22, 1.51)	2.03 (1.84, 2.24)	3.10 (2.63, 3.63)		0.83 (0.66, 0.93)	1.38 (1.22, 1.51)	2.03 (1.84, 2.24)	3.10 (2.63, 3.63)
Circulating; mean (SD) mmol/L	0.05 (0.01)	0.05 (0.01)	0.05 (0.01)	0.05 (0.01)		0.05 (0.01)	0.05 (0.01)	0.05 (0.01)	0.05 (0.01)
Age-adjusted	1.00	0.83 (0.50, 1.38)	0.75 (0.45, 1.26)	0.61 (0.36, 1.03)	.06	1.00	0.75 (0.46, 1.23)	0.82 (0.51, 1.35)	0.80 (0.49, 1.31)	.53
Model 1	1.00	0.79 (0.46, 1.36)	0.73 (0.42, 1.27)	0.56 (0.31, 1.01)	.06	1.00	0.70 (0.42, 1.17)	0.75 (0.44, 1.26)	0.66 (0.38, 1.13)	.20
Leucine
Cases/total	47/135	46/135	43/135	34/135		62/135	57/135	58/135	56/135
Dietary: median (IQR) g/d	1.49 (1.15, 1.68)	2.40 (2.16, 2.65)	3.57 (3.24, 3.88)	5.25 (4.46, 6.36)		1.49 (1.15, 1.68)	2.40 (2.16, 2.65)	3.57 (3.24, 3.88)	5.25 (4.46, 6.36)
Circulating; mean (SD) mmol/L	0.10 (0.02)	0.10 (0.02)	0.09 (0.02)	0.09 (0.02)		0.10 (0.02)	0.10 (0.02)	0.09 (0.02)	0.09 (0.02)
Age-adjusted	1.00	0.89 (0.54, 1.49)	0.80 (0.48, 1.33)	0.60 (0.35, 1.02)	.05	1.00	0.78 (0.48, 1.27)	0.79 (0.48, 1.29)	0.79 (0.48, 1.29)	.44
Model 1	1.00	0.84 (0.49, 1.43)	0.77 (0.44, 1.33)	0.54 (0.29, 0.98)	.04	1.00	0.69 (0.41, 1.16)	0.69 (0.41, 1.17)	0.63 (0.36, 1.09)	.15
Valine
Cases/total	49/135	43/135	42/135	36/135		65/135	54/135	55/135	59/135
Dietary; median (IQR) g/d	0.98 (0.79, 1.13)	1.64 (1.46, 1.78)	2.37 (2.18, 2.58)	3.63 (3.13, 4.24)		0.98 (0.79, 1.13)	1.64 (1.46, 1.78)	2.37 (2.18, 2.58)	3.63 (3.13, 4.24)
Circulating; mean (SD) mmol/L	0.21 (0.03)	0.21 (0.03)	0.21 (0.04)	0.21 (0.04)		0.21 (0.03)	0.21 (0.03)	0.21 (0.04)	0.21 (0.04)
Age-adjusted	1.00	0.74 (0.44, 1.24)	0.72 (0.43, 1.20)	0.61 (0.36, 1.02)	.08	1.00	0.63 (0.38, 1.03)	0.66 (0.40, 1.07)	0.79 (0.49, 1.29)	.56
Model 1	1.00	0.71 (0.41, 1.21)	0.67 (0.39, 1.16)	0.55 (0.31, 0.99)	.06	1.00	0.59 (0.35, 0.99)	0.57 (0.34, 0.96)	0.64 (0.37, 1.11)	.20

Model 1 was adjusted for age (continuous years), years of stay in Korea (continuous years), highest education completed (elementary and high school, college education and above), employment status (no, yes), ever smoked (no, yes), current alcohol use (no, yes), vigorous physical activity (no, yes), family history of diabetes (no, yes) menopause (no, yes) hypertension (no, yes) and energy intake (continuous, kcal/d).

General obesity was defined as body mass index ⩾ 25 kg/m², Abdominal obesity was defined as waist circumference ⩾ 80 cm.

Dietary amino acid intake was adjusted for energy intake using the residual method.

Bonferroni correction for P-value = .01

Relationship of Circulating BCAA with Obesity

The multivariable-adjusted OR (95% CI) of obesity by quartiles of log-transformed concentrations of circulating BCAA (Table 2) suggested higher BCAA concentrations were associated with higher odds of general obesity; tBCAA [1.00, 2.62 (1.38, 5.00), 3.43 (1.83, 6.44), 5.97 (3.19, 11.16; P for trend < .0001)], isoleucine [1.00, 6.10 (3.08, 12.08), 3.90 (1.96, 7.77), and 6.96 (3.54, 13.71; P for trend = < .0001)], leucine [1.00, 2.17 (1.16, 4.08), 3.02 (1.63, 5.59), and 4.70 (2.55, 8.65); P for trend < .0001)], and valine [1.00, 3.38 (1.73, 6.61), 3.56 (1.82, 6.93) and 7.59 (3.91, 14.70; P for trend < .0001)]. These associations remained after additional adjusting for energy-adjusted dietary BCAA intakes. Similarly, high circulating tBCAA concentrations were associated with higher odds of abdominal obesity: 1.00 – first quartile, 1.76 (1.01, 3.05) – second quartile, 2.88 (1.67, 4.94) – third quartile, and 5.90 (3.38, 10.29) – fourth quartile, P for trend < .0001. A similar trend was observed for circulating isoleucine, leucine and valine concentrations, even with additional adjustment for energy-adjusted dietary BCAA intake.

Table 2.

Odds Ratios and 95% Confidence Intervals for Obesity^a by Quartile of Log-transformed Circulating BCAA Profiles.^b

	General obesity (body mass index ⩾ 25 kg/m²)					Abdominal obesity (Waist circumference ⩾ 80 cm)
BCAA profiles	Q1	Q2	Q3	Q4	P for trend^c	Q1	Q2	Q3	Q4	P for trend^c
Total BCAA
Cases/total	18/135	38/135	48/135	66/135		33/135	47/135	65/135	88/135
Dietary; median (IQR) g/d	7.04 (4.20, 9.77)	6.68 (4.29, 9.66)	6.13 (4.01, 9.71)	6.47 (4.46, 9.07)		7.04 (4.20, 9.77)	6.68 (4.29, 9.66)	6.13 (4.01, 9.71)	6.47 (4.46, 9.07)
Circulating; mean (SD) mmol/L	0.28 (0.02)	0.33 (0.01)	0.36 (0.01)	0.43 (0.05)		0.28 (0.02)	0.33 (0.01)	0.36 (0.01)	0.43 (0.05)
Age-adjusted	1.00	2.65 (1.41, 4.70)	3.74 (2.02, 6.92)	6.42 (3.50, 11.79)	<.0001	1.00	1.73 (1.01, 2.96)	3.04 (1.79, 5.17)	6.16 (3.58, 10.59)	<.0001
Model 1	1.00	2.60 (1.36, 4.94)	3.43 (1.83, 6.44)	6.01 (3.22, 11.21)	<.0001	1.00	1.75 (1.01, 3.04)	2.88 (1.68, 4.95)	5.92 (3.40, 10.33)	<.0001
Model 2	1.00	2.62 (1.38, 5.00)	3.43 (1.83, 6.44)	5.97 (3.19, 11.16)	<.0001	1.00	1.76 (1.01, 3.05)	2.88 (1.67, 4.94)	5.90 (3.38, 10.29)	<.0001
Isoleucine
Cases/total	14/135	51/135	45/135	60/135		32/135	55/135	69/135	77/135
Dietary; median (IQR) g/d	1.80 (1.10, 2.55)	1.67 (0.96, 2.48)	1.55 (0.99, 2.53)	1.65 (1.15, 2.33)		1.80 (1.10, 2.55)	1.67 (0.96, 2.48)	1.55 (0.99, 2.53)	1.65 (1.15, 2.33)
Circulating; mean (SD) mmol/L	0.04 (0.00)	0.04 (0.00)	0.05 (0.00)	0.06 (0.01)		0.04 (0.00)	0.04 (0.00)	0.05 (0.00)	0.06 (0.01)
Age-adjusted	1.00	5.42 (2.80, 10.48)	4.42 (2.27, 8.59)	7.23 (3.75, 13.93)	<.0001	1.00	2.28 (1.33, 3.89)	3.52 (2.07, 6.00)	4.56 (2.67, 7.78)	<.0001
Model 1	1.00	6.18 (3.12, 12.24)	4.05 (2.04, 8.05)	7.10 (3.61, 13.97)	<.0001	1.00	2.40 (1.39, 4.14)	3.15 (1.83, 5.43)	4.40 (2.55, 7.61)	<.0001
Model 2	1.00	6.10 (3.08, 12.08)	3.90 (1.96, 7.77)	6.96 (3.54, 13.71)	<.0001	1.00	2.39 (1.38, 4.13)	3.12 (1.81, 5.39)	4.38 (2.53, 7.57)	<.0001
Leucine
Cases/total	20/135	38/135	49/135	63/135		31/135	56/135	66/135	80/135
Dietary; median (IQR) g/d	3.39 (2.15, 4.75)	2.90 (1.84, 4.12)	2.73 (1.71, 4.01)	2.81 (2.03, 4.07)		3.39 (2.15, 4.75)	2.90 (1.84, 4.12)	2.73 (1.71, 4.01)	2.81 (2.03, 4.07)
Circulating; mean (SD) mmol/L	0.07 (0.01)	0.09 (0.00)	0.10 (0.00)	0.12 (0.02)		0.07 (0.01)	0.09 (0.00)	0.10 (0.00)	0.12 (0.02)
Age-adjusted	1.00	2.37 (1.29, 4.36)	3.27 (1.80, 5.93)	5.10 (2.83, 9.20)	<.0001	1.00	2.57 (1.50, 4.40)	3.24 (1.90, 5.53)	5.08 (2.96, 8.72)	<.0001
Model 1	1.00	2.21 (1.18, 4.14)	3.13 (1.70, 5.78)	4.86 (2.65, 8.93)	<.0001	1.00	2.50 (1.44, 4.34)	3.23 (1.86, 5.59)	4.91 (2.82, 8.55)	<.0001
Model 2	1.00	2.17 (1.16, 4.08)	3.02 (1.63, 5.59)	4.70 (2.55, 8.65)	<.0001	1.00	2.50 (1.44, 4.34)	3.22 (1.85, 5.59)	4.89 (2.80, 8.55)	<.0001
Valine
Cases/total	15/135	43/135	43/135	69/135		28/135	57/135	60/135	88/135
Dietary; median (IQR) g/d	2.08 (1.24, 3.11)	1.90 (1.28, 2.77)	1.97 (1.33, 2.83)	1.94 (1.31, 2.70)		2.08 (1.24, 3.11)	1.90 (1.28, 2.77)	1.97 (1.33, 2.83)	1.94 (1.31, 2.70)
Circulating; mean (SD) mmol/L	0.17 (0.01)	0.20 (0.01)	0.22 (0.01)	0.26 (0.03)		0.17 (0.01)	0.20 (0.01)	0.22 (0.01)	0.26 (0.03)
Age-adjusted	1.00	3.75 (1.95, 7.19)	3.76 (1.96, 7.21)	8.36 (4.41, 15.85)	<.0001	1.00	2.82 (1.63, 4.88)	3.10 (1.79, 5.35)	7.30 (4.18, 12.74)	<.0001
Model 1	1.00	3.42 (1.76, 6.67)	3.48 (1.79, 6.77)	7.49 (3.88, 14.47)	<.0001	1.00	2.64 (1.51, 4.63)	2.91 (1.67, 5.09)	6.87 (3.86, 12.22)	<.0001
Model 2	1.00	3.38 (1.73, 6.61)	3.56 (1.82, 6.93)	7.59 (3.91, 14.70)	<.0001	1.00	2.63 (1.50, 4.61)	2.92 (1.67, 5.11)	6.87 (3.86, 12.23)	<.0001

Abbreviation: BCAA, branched-chain amino acids.

Model 2 was adjusted for energy-adjusted dietary BCAA intake (continuous, kcal) in addition to covariates in Model 1.

General Obesity was defined as body mass index ⩾ 25kg/m², Abdominal obesity was defined as waist circumference ⩾ 80 cm.

Circulating BCAA profiles were log-transformed to reduce skewness.

Bonferroni correction for P-value = .01.

Joint Relationship of Energy-Adjusted Dietary and Log-Transformed Circulating BCAA with Obesity

In the joint relationship of energy-adjusted dietary and log-transformed circulating BCAA profiles with obesity (Table 3), the multivariable-adjusted ORs (95% CIs) for general obesity by categories of energy-adjusted dietary and log-transformed circulating tBCAA profiles were 1.00 for “low dietary and low circulating” tBCAA, 2.06 (1.19, 3.56) for “low dietary and high circulating” tBCAA, 0.54 (0.28, 1.02) for “high dietary and low circulating” tBCAA, and 1.84 (1.03, 3.28) for “high dietary and high circulating” tBCAA, P for interaction = .61, after adjusting for potential confounding factors. Also, the multivariable-adjusted ORs (95% CIs) for abdominal obesity by categories of energy-adjusted dietary and log-transformed circulating tBCAA profiles were 1.00 for “low dietary and low circulating” tBCAA, 2.40 (1.42, 4.05) for “low dietary and high circulating” tBCAA, 0.69 (0.39, 1.22) for “high dietary and low circulating” tBCAA, and 2.61 (1.49, 4.57) for “high dietary and high circulating” tBCAA, P for interaction = .58. A similar trend was observed for the joint associations of dietary and circulating isoleucine, leucine and valine profiles with general and abdominal obesity.

Table 3.

Odds Ratios and 95% Confidence Intervals for Obesity^a by the Combined Association of Energy-Adjusted Dietary^b and Log-Transformed Circulating^c BCAA Profiles.

	General obesity (body mass index ⩾ 25 kg/m²)					Abdominal obesity (Waist circumference ⩾ 80 cm)
BCAA profiles	“Low dietary and low circulating”	“Low dietary and high circulating”	“High dietary and low circulating”	“High dietary and high circulating”	P for interaction^d	“Low dietary and low circulating” tBCAA	“Low dietary and high circulating” tBCAA	“High dietary and low circulating” tBCAA	“High dietary and high circulating” tBCAA	P for interaction^d
Total BCAA
Cases/total	33/127	61/143	23/143	53/127		41/127	78/143	39/143	75/127
Dietary; median (IQR) g/d	4.12 (3.34, 5.37)	4.44 (3.28, 5.48)	9.55 (7.63, 12.52)	9.29 (8.39, 11.60)		4.12 (3.34, 5.37)	4.44 (3.38, 5.48)	9.56 (7.63, 12.52)	9.29 (8.39, 11.60)
Circulating; mean (SD) mmol/L	0.31 (0.03)	0.40 (0.04)	0.30 (0.03)	0.40 (0.06)		0.31 (0.03)	0.40 (0.04)	0.30 (0.03)	0.40 (0.06)
Age-adjusted	1.00	2.14 (1.27, 3.60)	0.53 (0.29, 0.98)	2.03 (1.19, 3.48)	.61	1.00	2.57 (1.55, 4.28)	0.77 (0.45, 1.31)	3.09 (1.83, 5.23)	.58
Model 1	1.00	2.06 (1.19, 3.56)	0.54 (0.28, 1.02)	1.84 (1.03, 3.28)	.61	1.00	2.40 (1.42, 4.05)	0.69 (0.39, 1.22)	2.61 (1.49, 4.57)	.58
Isoleucine
Cases/total	34/129	59/141	31/141	46/129		43/129	75/141	44/141	71/129
Dietary; median (IQR) g/d	1.02 (0.84, 1.33)	1.09 (0.79, 1.41)	2.47 (2.00, 3.16)	2.39 (2.12, 2.91)		1.02 (0.84, 1.33)	1.09 (0.79, 1.41)	2.47 (2.00, 3.16)	2.39 (2.12, 2.91)
Circulating; mean (SD) mmol/L	0.04 (0.00)	0.06 (0.01)	0.04 (0.01)	0.06 (0.01)		0.04 (0.00)	0.06 (0.01)	0.04 (0.01)	0.06 (0.01)
Age-adjusted	1.00	2.07 (1.23, 3.48)	0.78 (0.45, 1.38)	1.54 (0.90, 2.64)	.81	1.00	2.39 (1.44, 3.96)	0.90 (0.54, 1.52)	2.50 (1.49, 4.18)	.20
Model 1	1.00	1.87 (1.08, 3.24)	0.79 (0.43, 1.44)	1.38 (0.77, 2.46)	.95	1.00	2.10 (1.25, 3.55)	0.80 (0.46, 1.39)	2.07 (1.20, 3.59)	.23
Leucine
Cases/total	34/124	59/146	24/146	53/124		45/124	74/146	42/146	72/124
Dietary; median (IQR) g/d	1.84 (1.55, 2.40)	1.96 (1.48, 2.41)	4.21 (3.54, 5.64)	4.09 (3.62, 5.03)		1.84 (1.55, 2.40)	1.96 (1.48, 2.41)	4.22 (3.54, 5.64)	4.09 (3.62, 5.03)
Circulating; mean (SD) mmol/L	0.08 (0.01)	0.11 (0.01)	0.08 (0.02)	0.11 (0.02)		0.08 (0.01)	0.11 (0.01)	0.08 (0.02)	0.11 (0.02)
Age-adjusted	1.00	1.77 (1.06, 2.99)	0.51 (0.28, 0.93)	1.88 (1.09, 3.22)	.71	1.00	1.79 (1.09, 2.94)	0.70 (0.41, 1.17)	2.32 (1.38, 3.90)	.30
Model 1	1.00	1.79 (1.04, 3.08)	0.53 (0.28, 0.99)	1.85 (1.03, 3.33)	.53	1.00	1.73 (1.04, 2.90)	0.63 (0.36, 1.10)	2.09 (1.19, 3.67)	.41
Valine
Cases/total	35/134	57/136	23/136	55/134		47/134	72/136	38/136	36/134
Dietary; median (IQR) g/d	1.25 (0.97, 1.57)	1.33 (0.99, 1.66)	2.88 (2.32, 3.65)	2.79 (2.48, 3.60)		1.25 (0.97, 1.57)	1.33 (0.99, 1.66)	2.88 (2.32, 3.65)	2.79 (2.48, 3.60)
Circulating; mean (SD) mmol/L	0.18 (0.02)	0.24 (0.03)	0.18 (0.02)	0.24 (0.03)		0.18 (0.02)	0.24 (0.03)	0.18 (0.02)	0.24 (0.03)
Age-adjusted	1.00	2.09 (1.24, 3.51)	0.58 (0.32, 1.05)	1.93 (1.14, 3.25)	.66	1.00	2.15 (1.31, 3.54)	0.72 (0.42, 1.21)	2.39 (1.45, 3.95)	.24
Model 1	1.00	2.10 (1.22, 3.61)	0.59 (0.31, 1.10)	1.75 (1.00, 3.05)	.66	1.00	2.15 (1.28, 3.59)	0.66 (0.38, 1.14)	2.02 (1.19, 3.43)	.27

Abbreviation: BCAA, branched-chain amino acids.

General Obesity was defined as body mass index ⩾ 25 kg/m², Abdominal obesity was defined as waist circumference ⩾ 80cm.

Dietary amino acid intake was adjusted for energy intake using the residual method.

Circulating BCAA profiles were log-transformed to reduce skewness.

Bonferroni correction for P-value = .01.

Furthermore, the LS means (95% CI) of BMI and WC did not vary by the distribution of energy-adjusted dietary BCAA intakes (Table S5). For example, the LS means of BMI (in kg/m²) by the quartile of energy-adjusted dietary tBCAA intakes were 24.36 (23.29, 25.48) – first quartile, 23.87 (22.84, 24.95) – second quartile, 23.78 (22.73, 24.87) – third quartile and 23.54 (22.53, 24.60) – fourth quartile, P for trend = .14. Similarly, the LS mean of WC (in cm) by the quartile of energy-adjusted dietary tBCAA intakes was 83.50 (80.98, 86.02) – first quartile, 82.32 (79.85, 84.80) – second quartile, 82.47 (79.94, 85.00) – third quartile and 81.63 (79.17, 84.10) – fourth quartile, P for trend = .17.

However, the LS means (95% CI) of BMI in kg/m² and WC in cm increased across the quartiles of log-transformed circulating BCAA concentrations (Table S6). The LS means (95% CI) of BMI (in kg/m²) by quartiles of log-transformed circulating BCAA concentrations were 21.81 (20.89, 22.87), 23.33 (22.36, 24.33), 23.89 (22.93, 24.90), and 25.47 (24.47, 26.52; P for trend < .0001) for the first, second, third and fourth quartile respectively, with additional adjustment for energy-adjusted for dietary BCAA intakes. Similarly, The LS means (95% CI) of WC (in cm) by quartiles of log-transformed circulating BCAA concentrations were 78.46 (76.02, 80.89), 80.95 (78.55, 83.35), 81.88 (79.54, 84.22), 86.18 (83.88, 88.48; P for trend < .0001) for the first, second, third, and fourth quartile respectively, with additional adjustment for energy-adjusted for dietary BCAA intakes. Analogous trends were observed for circulating concentrations of isoleucine, leucine and valine.

Additionally, there was no evidence of interaction between energy-adjusted dietary BCAA intake and log-transformed circulating BCAA concentrations in the association of BCAA with BMI (Table 4). The LS mean (95% CI) of BMI (in kg/m²) by the distribution of the joint relationship of energy-adjusted and log-transformed circulating profiles of tBCAA was 22.89 (21.91, 23.91) – “low dietary and low circulating” tBCAA, 24.68 (23.66, 25.74) – “low dietary and high circulating” tBCAA, 22.35 (21.39, 23.36) – “high dietary and low circulating” tBCAA, and 24.74 (23.73, 25.79) – “high dietary and high circulating” tBCAA, P for interactions = .70. Similarly, there was no evidence of interaction for WC.

Table 4.

Least-square Means and 95% Confidence Intervals of Body Mass Index and Waist Circumference Profiles by the Combined Association of Energy-Adjusted Dietary^a and Log-Transformed Circulating^b BCAA Profiles.

BCAA profiles	“Low dietary and low circulating” tBCAA	“Low dietary and high circulating” tBCAA	“High dietary and low circulating” tBCAA	“High dietary and high circulating” tBCAA	P for interaction
Body mass index^c (kg/m²)
Total BCAA
Age-adjusted	21.57 (20.98, 22.18)	23.50 (22.89, 24.11)	21.13 (20.58, 21.68)	23.69 (23.04, 24.35)	.73
Model 1	22.89 (21.91, 23.91)	24.68 (23.66, 25.74)	22.35 (21.39, 23.36)	24.74 (23.73, 25.79)	.70
Isoleucine
Age-adjusted	21.63 (21.03, 22.24)	23.37 (22.75, 24.00)	21.52 (20.95, 22.11)	23.28 (22.64, 23.94)	.47
Model 1	22.98 (21.97, 24.03)	24.46 (23.45, 25.51)	22.77 (21.75, 23.83)	24.37 (23.35, 25.44)	.55
Leucine
Age-adjusted	21.82 (21.21, 22.44)	23.25 (22.65, 23.86)	21.15 (20.60, 21.70)	23.71 (23.05, 24.38)	.47
Model 1	23.09 (22.10, 24.12)	24.51 (23.50, 25.56)	22.37 (21.41, 23.37)	24.90 (23.86, 25.98)	.56
Valine
Age-adjusted	21.73 (21.15, 22.33)	23.37 (22.75, 24.00)	21.10 (20.54, 21.68)	23.64 (23.01, 24.28)	.72
Model 1	23.01 (22.04, 24.01)	24.63 (23.60, 25.70)	22.31 (21.33, 23.33)	24.68 (23.68, 25.73)	.74
Waist circumference (cm)
Total BCAA
Age-adjusted	77.87 (76.32, 79.42)	81.63 (80.17, 83.09)	76.60 (75.14, 78.06)	82.55 (81.00, 84.10)	.97
Model 1	80.51 (78.03, 82.97)	83.77 (81.38, 86.16)	79.08 (76.60, 81.56)	84.37 (82.00, 86.73)	.91
Isoleucine
Age-adjusted	77.48 (75.92, 79.04)	81.83 (80.35, 83.32)	77.61 (76.12, 79.10)	81.58 (80.03, 83.14)	.68
Model 1	80.16 (77.66, 82.67)	83.83 (81.47, 86.19)	80.07 (77.52, 82.63)	83.52 (81.11, 85.94)	.73
Leucine
Age-adjusted	78.22 (76.63, 79.80)	81.22 (79.76, 82.68)	76.95 (75.49, 78.41)	82.33 (80.74, 83.92)	.77
Model 1	80.80 (78.32, 83.28)	83.59 (81.21, 85.97)	79.51 (77.03, 81.99)	84.45 (82.03, 86.87)	.84
Valine
Age-adjusted	78.47 (76.95, 79.98)	81.26 (79.76, 82.77)	76.54 (75.03, 78.05)	82.28 (80.76, 83.80)	.76
Model 1	81.04 (78.61, 83.46)	83.61 (81.19, 86.03)	78.95 (76.42, 81.48)	84.09 (81.72, 86.45)	.75

Abbreviation: BCAA, branched-chain amino acids

Dietary amino acid intake was adjusted for energy intake using the residual method.

Circulating BCAA profiles were log-transformed.

BMI was log-transformed to minimize skewness.

Discussion

This study evaluated the relationship of dietary and circulating BCAA with obesity among women and found that higher circulating BCAA levels were associated with a higher prevalence of both general and abdominal obesity, independent of higher dietary BCAA intake (which was associated with a lower prevalence of general obesity) in the same population, with no evidence of interaction between dietary and circulating BCAA profiles. Our study is among the latest epidemiological reports (if not the first, especially in an Asian population) to examine the joint relationship of BCAA in both diet and metabolome in the pathophysiology of obesity manifestation, combining evidence relating to environmental exposures from diet and data from the metabolome in discerning the significance of BCAA exposure in metabolic disorders.

The inverse association between dietary BCAA and obesity prevalence was in tandem with earlier reports,^13,60 but not without evidence on the impact of the dietary environment of the BCAA–obesity association.⁶¹ Studies in which BCAA sources were primarily plant-based reported inverse associations,^13,62 but others with mainly animal protein BCAA sources documented positive associations.^63,64 Similarly, the direct relationship between circulating BCAA and obesity prevalence was also in tandem with a previous report,¹⁷ highlighting the potential dysregulation of BCAA metabolism in a cascade of complex enzymatic processes and stimulation of the p70s6 K phosphorylation and insulin receptor substrate-1 in the mammalian target of rapamycin (mTOR) pathway.^34,65,66 Whereas BCAAs are essential amino acids primarily sourced from diets¹⁰ and metabolized in the liver, muscle, and brown fat tissues for homeostatic regulation,²⁹ the complex interplay between their dietary exposure and cellular metabolism in the pathophysiology of obesity manifestations remains to be clearly elucidated. Earlier efforts from different populations on this phenomenon^60,63 documented the independent association with obesity without evidence of joint associations. In this study, higher circulating BCAA was associated with obesity, independent of dietary BCAA consumption level, with no evidence of interaction. A similar finding emphasizing dysregulated BCAA metabolism on account of type 2 diabetes risk was observed among women with a history of type 2 diabetes in a longitudinal study,⁴⁴ and there are plausible explanations for these findings.

First, high circulating BCAA might have likely masked the protective dietary BCAA–obesity association, implying that dysregulated BCAA metabolism (more than any other factor(s), including dietary BCAA) accounts for the association. Even though increased protein intake has been observed to acutely oversaturate the BCAA degradation pathway postprandially, the potential implication of dietary BCAA intake on BCAA metabolism remains insignificant, particularly in light of less apparent changes on branched-chain alpha-keto acids⁶⁷ – a BCAA-related metabolite critical in the homeostatic regulation of the BCAA pathway, which may trigger metabolic disorders in the vasculature.⁶⁸ Whereas the diet-obesity relationship is more often protective, especially in the light of previous reports,¹⁴ primarily documenting the viability of improved cardio-metabolic health attributed to modest BCAA supplementation in physically healthy states, particularly among athletes.^69
-71 This observation is supported by an experimental study on whey protein regimen supplementation among women with obesity in a weight loss trial, which emphasizes the non-impairment of weight loss and insulin sensitivity, with a negligible difference in BCAA and its associated metabolites post-supplementation.⁷² However, these findings are yet to be replicated in extensive longitudinal studies and trials with large sample sizes.

Second, the metabolism of BCAA differs extensively across tissues in the whole body, and the modest impact of dietary BCAA has been extensively reported.^73
-76 For example, BCAA catabolism primarily occurs in the liver,⁷³ whereas BCAA oxidation takes place in the skeletal muscles.⁷⁴ Most whole-body BCAA disposal occurs in skeletal muscle.⁷⁵ However, discrepancies in the regulation of BCAA and associated metabolites (including post-transcriptional modification of enzymes and complex enzymatic processes interrelated to multiple metabolic pathways), especially in the adipose tissues, potentially promote impaired BCAA metabolism and the onset of obesity and its sequelae.^75,76 In line with this hypothesis, a recent hospital-based study found that circulating BCAA correlated more closely with visceral fat distribution among type 2 diabetes subjects in China.⁷⁷

Third, BCAAs are vital substrates for energy production and, at the same time, crucial regulators of multiple complex cascades of energy signaling via several BCAA-related metabolites in the adipocytes.⁷⁸ We hypothesize that the bidirectional inhibition of BCAA catabolism in adipocytes, or the excessive clearance via digestion into the vasculature and subsequent extension into the adipocytes, could impair their BCAA clearance through multiple complex pathways, including the mTOR pathway, among others, leading to obesity manifestation.³⁹ Taken together, perturbations primarily in the vasculature arising from a complex cascade of compromised metabolic health could trigger dysregulated BCAA metabolism and consequently aggravate the complex network of BCAA and related metabolites, leading to metabolic disorders, including obesity and its sequelae.

Limitations of this study include the inability to infer causal associations due to the cross-sectional design, the finding being limited to females only, the use of convenience sampling, and the potential residual confounding factors in the reported associations. Using BMI to classify obesity status is often accompanied by inherent limitations, including a lack of precision in discerning the overall fat distribution and sometimes the false classification of people with hidden obesity and people with high muscle mass. It was on this basis that this study conducted additional analyses using WC, a reliable proxy for discerning fat distribution, especially in the visceral fat deposit, to discriminate abdominal obesity. Dietary information was evaluated using 1-day or 2-day 24-hour dietary recalls in this study. A 2-day 24-hour dietary recall was planned, but some participants were only able to provide a 1-day 24-hour dietary recall. This study did not evaluate the association between plant- or animal-based sources of BCAA and obesity. We plan to establish a separate BCAA database for future research.

The study’s strong points include the determination of BCAA in diet and serum, adjustment for several covariates, and it is one of the earliest to highlight the implications of dietary exposure and BCAA metabolism in obesity manifestations within a single epidemiological cohort of apparently healthy women, especially those of Asian ancestry. Future longitudinal studies should consider the alterations in dietary BCAA exposure within the context of the dietary environment that could lead to changes in BCAA metabolism and its implications on obesity and its sequelae.

Conclusion

Higher circulating BCAA profiles were associated with obesity, independent of dietary BCAA intake levels, with no evidence of interaction between dietary and circulating BCAA in the associations. Longitudinal cohorts from diverse populations are necessary to verify these associations.

Supplemental Material

sj-docx-1-nmi-10.1177_11786388251395146 – Supplemental material for Circulating Branched-Chain Amino Acids are Associated with Higher Odds of Obesity: Findings from the FiLWHEL Study

Supplemental material, sj-docx-1-nmi-10.1177_11786388251395146 for Circulating Branched-Chain Amino Acids are Associated with Higher Odds of Obesity: Findings from the FiLWHEL Study by Akinkunmi Paul Okekunle, Heejin Lee, Sherlyn Mae P. Provido, Grace H. Chung, Sangmo Hong, Sung Hoon Yu, Chang Beom Lee and Jung Eun Lee in Nutrition and Metabolic Insights

Footnotes

Acknowledgements

The authors are grateful to all volunteers who participated in this study.

Author’s Note

Heejin Lee is now affiliated with Department of Food and Nutrition, College of Human Ecology, Seoul National University, Korea.

ORCID iDs

Akinkunmi Paul Okekunle

Jung Eun Lee

Ethical Considerations

The Ethics review committees of Sookmyung Women’s University (SMWU-1311-BR-012), Seoul National University (1904/002-011), and Hayang University (HYU-2020-03-008-6) approved the study.

Consent to Participate

All respondents provided written informed consent before participation. All procedures in this study were conducted in accordance with the ethical standards of the IRB of Sookmyung Women’s University, Seoul National University, Hayang University and the 1964 Helsinki Declaration, as amended, or with comparable ethical standards.

Author Contributions

APO conceptualized and designed the study; HJ,SMPP, SH, SHY, CBL, and JEL conducted the data acquisition and curation; APO conducted the analysis; SH, GHC, SHY, CBL, and JEL contributed to the interpretation; HJ contributed to the data analysis and interpretation; APO drafted the manuscript; HJ, CBL and JEL critically revised the manuscript for important intellectual content. All authors read and approved the final version to be published and agreed to be accountable for the work. The authors thank all volunteers and staff for participating in this study.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The Brain Pool Program supported this research work through the National Research Foundation of Korea, funded by the Ministry of Science and ICT (2020H1D3A1A04081265), Hanmi Pharmaceutical Co., Ltd. (No. 201300000001270), Chong Kun Dang Pharm., Seoul, Korea (No. 201600000000225), Handok Inc., Seoul, Korea. This research was supported by the Research Grants for Asian Studies funded by Seoul National University Asia Center (SNUAC) in 2021 (0448A-20210077). The funders had no role in the study design, data collection and analysis, manuscript preparation, or the decision to publish.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data for this study cannot be made publicly available because the FiLWHEL study is still ongoing. During the signing of consent, the participants were not informed that their information would be stored in a publicly accessible database. However, other researchers can collaborate with the study team by following the approval procedures posted on the study website (). Requests to access the data should be sent to the Data Access Committee at nutepid@gmail.com.

Supplemental Material

Supplemental material for this article is available online.

References

Afshin

Forouzanfar

Reitsma

, et al. Health effects of overweight and obesity in 195 countries over 25 years. N Engl J Med. 2017;377:13-27.

Chong

Jayabaskaran

Kong

, et al. Trends and predictions of malnutrition and obesity in 204 countries and territories: an analysis of the Global Burden of Disease Study 2019. med. 2023;57:101850.

Tim

Rachel

Jaynaide

, et al. World Obesity Atlas 2023. World Obesity Federation; 2023.

OECD. The heavy burden of obesity: The Economics of Prevention. OECD Health Policy Studies, OECD Publishing; 2019. https://doi.org/10.1787/67450d67-en.

Hopkins

Goncalves

Cantley

LC.

Obesity and cancer mechanisms: cancer metabolism. J Clin Oncol. 2016;34:4277-4283.

Magkos

Tetens

Bügel

, et al. The environmental foodprint of obesity. Obesity. 2020;28:73-79.

Flegal

Kruszon-Moran

Carroll

Fryar

Ogden

CL.

Trends in obesity among adults in the United States, 2005 to 2014. JAMA. 2016;315:2284-2291.

Peeters

Barendregt

Willekens

Mackenbach

Al Mamun

Bonneux

Obesity in adulthood and its consequences for life expectancy: a life-table analysis. Ann Intern Med. 2003;138:24-32.

Barness

Opitz

Gilbert-Barness

Obesity: genetic, molecular, and environmental aspects. Am J Med Genet A. 2007;143A:3016-3034.

10.

Harper

Miller

Block

KP.

Branched-chain amino acid metabolism. Annu Rev Nutr. 1984;4:409-454.

11.

Shimomura

Kitaura

Kadota

, et al. Novel physiological functions of branched-chain amino acids. J Nutr Sci Vitaminol. 2015;61 Suppl:S112-S114.

12.

Jung

Okekunle

Lee

Sung

Lim

YJ.

Role of branched-chain amino acid metabolism in tumor development and progression. J Cancer Prev. 2021;26:237-243.

13.

Okekunle

Feng

Sun

Higher intakes of energy-adjusted dietary amino acids are inversely associated with obesity risk. Amino Acids. 2019;51:373-382.

14.

Okekunle

Zhang

Wang

, et al. Dietary branched-chain amino acids intake exhibited a different relationship with type 2 diabetes and obesity risk: a meta-analysis. Acta Diabetol. 2019;56:187-195.

15.

Okekunle

Lee

Provido

SMP

, et al. Dietary branched-chain amino acids and odds of obesity among immigrant Filipino women: the Filipino women’s diet and health study (FiLWHEL). BMC Public Health. 2022;22:654.

16.

Wang-Sattler

Herder

, et al. Novel biomarkers for pre-diabetes identified by metabolomics. Mol Syst Biol. 2012;8:615.

17.

Okekunle

Liu

, et al. Abnormal circulating amino acid profiles in multiple metabolic disorders. Diabetes Res Clin Pract. 2017;132:45-58.

18.

Lee

Jang

, et al. Prediction of future risk of insulin resistance and metabolic syndrome based on Korean boy’s metabolite profiling. Obes Res Clin Pract. 2015;9:336-345.

19.

Okekunle

Lee

Provido

, et al. Dietary Intakes of Branched-Chain Amino Acids and Odds of Dyslipidaemia Among Filipino Women in Korea: The Filipino Women’s Diet and Health Study (FiLWHEL) . International Congress on Obesity and MEtabolic Syndrome. Korean Society for Study of Obesity; 2021.

20.

Wang

Shu

Han

Association of dietary intake of branched-chain amino acids with long-term risks of CVD, cancer and all-cause mortality. Public Health Nutr. 2021;25:1-11.

21.

Katagiri

Song

Zhang

, et al. Dietary intake of branched-chain amino acids and risk of colorectal cancer. Cancer Prev Res. 2020;13:65-72.

22.

Tobias

Chai

Tamimi

, et al. Dietary intake of branched chain amino acids and breast cancer risk in the NHS and NHS II prospective cohorts. JNCI Cancer Spectrum. 2021;5:3. doi:10.1093/jncics/pkab032

23.

Fukushima

Harada

Takeuchi

, et al. Association between dyslipidemia and plasma levels of branched-chain amino acids in the Japanese population without diabetes mellitus. J Clin Lipidol. 2019;13:932-939.e2.

24.

Yadao

MacKenzie

Bergdahl

Reducing branched-chain amino acid intake to reverse metabolic complications in obesity and type 2 diabetes. J Physiol. 2018;596:3455-3456.

25.

Mirmiran

Bahadoran

Esfandyari

Azizi

Dietary protein and amino acid profiles in relation to risk of dysglycemia: findings from a prospective population-based study. Nutrition. 2017;9:971.

26.

Mirmiran

Bahadoran

Ghasemi

Azizi

Contribution of dietary amino acids composition to incidence of cardiovascular outcomes: a prospective population-based study. Nutr Metab Cardiovasc Dis. 2017;27:633-641.

27.

Cocate

Natali

de Oliveira

Alfenas

Hermsdorff

HHM

. Consumption of branched-chain amino acids is inversely associated with central obesity and cardiometabolic features in a population of Brazilian middle-aged men: potential role of leucine intake. J Nutr Health Aging. 2015;19:771-777.

28.

Doestzada

Zhernakova

C. L. van den Munckhof

, et al. Systematic analysis of relationships between plasma branched-chain amino acid concentrations and cardiometabolic parameters: an association and Mendelian randomization study. BMC Med. 2022;20:485.

29.

Neinast

Jang

Hui

, et al. Quantitative analysis of the whole-body metabolic fate of branched-chain amino acids. Cell Metab. 2019;29:417-429.e4.

30.

She

Van Horn

Reid

Hutson

Cooney

Lynch

CJ.

Obesity-related elevations in plasma leucine are associated with alterations in enzymes involved in branched-chain amino acid metabolism. Am J Physiol Metab. 2007;293:E1552-E1563.

31.

Tobias

Mora

Verma

Billia

Buring

Lawler

PR.

Fasting status and metabolic health in relation to plasma branched chain amino acid concentrations in women. Metabolism. 2021;117:154391.

32.

Tobias

Lawler

Harada

, et al. Circulating branched-chain amino acids and incident cardiovascular disease in a prospective cohort of US Women. Circ Genom Precis Med. 2018;11:e002157.

33.

Ottosson

Ericson

Almgren

, et al. Postprandial levels of branch chained and aromatic amino acids associate with fasting glycaemia. J Amino Acids. 2016;2016:8576730.

34.

Newgard

CB.

Interplay between lipids and branched-chain amino acids in development of insulin resistance. Cell Metab. 2012;15:606-614.

35.

McCormack

Shaham

McCarthy

, et al. Circulating branched-chain amino acid concentrations are associated with obesity and future insulin resistance in children and adolescents. Pediatr Obes. 2013;8:52-61.

36.

Yamakado

Nagao

Imaizumi

, et al. Plasma free amino acid profiles predict four-year risk of developing diabetes, metabolic syndrome, dyslipidemia, and hypertension in Japanese population. Sci Rep. 2015;5:11918.

37.

Würtz

Mäkinen

Soininen

, et al. Metabolic signatures of insulin resistance in 7,098 young adults. Diabetes. 2012;61:1372-1380.

38.

Takashina

Tsujino

Watanabe

, et al. Associations among the plasma amino acid profile, obesity, and glucose metabolism in Japanese adults with normal glucose tolerance. Nutr Metab. 2016;13:5.

39.

Newgard

Bain

, et al. A branched-chain amino acid-related metabolic signature that differentiates obese and lean humans and contributes to insulin resistance. Cell Metab. 2009;9:311-326.

40.

Michaliszyn

Sjaarda

Mihalik

, et al. Metabolomic profiling of amino acids and β-cell function relative to insulin sensitivity in youth. J Clin Endocrinol Metab. 2012;97:E2119-E2124.

41.

Zeleznik

Balasubramanian

Ren

, et al. Branched-chain amino acids and risk of breast cancer. JNCI Cancer Spectrum. 2021;5:5. doi:10.1093/jncics/pkab059

42.

Holeček

Why are branched-chain amino acids increased in starvation and diabetes?

Nutr. 2020;12:3087.

43.

Mansfeld

Urban

Priebe

, et al. Branched-chain amino acid catabolism is a conserved regulator of physiological ageing. Nat Commun. 2015;6:10043.

44.

Tobias

Clish

Mora

, et al. Dietary intakes and circulating concentrations of branched-chain amino acids in relation to incident type 2 diabetes risk among high-risk women with a history of gestational diabetes mellitus. Clin Chem. 2018;64:1203-1210.

45.

Abris

Hong

Provido

Lee

CB.

Filipino women's diet and health study (FiLWHEL): design and methods. Nutr Res Pract. 2017;11:70-75.

46.

Korean Nutrition Society (KSN). Computer-Aided Nutrition Analysis Program (CAN-Pro 4.0). Korean Nutrition Society; 2011.

47.

Food and Nutrition Research Institute (FNRI). The Philippine Food Composition Tables. Food and Nutrition Research Institute (FNRI); 1997.

48.

Lim

Kim

Cho

Choi

Park

Kim

SN.

National Standard Food Composition Tables provide the infrastructure for food and nutrition research according to policy and Industry. Korean J Food Nutr. 2013;26:886-894.

49.

U.S. Department of Agriculture (USDA). FoodData Central. 2019. Accessed 09 February 2021. fdc.nal.usda.gov

50.

Willett

Howe

Kushi

LH.

Adjustment for total energy intake in epidemiologic studies. Am J Clin Nutr. 1997;65:1220S-1228S.

51.

Okekunle

Yie

G-E

Song

, et al. Association of lipid profile with obesity among breast cancer survivors: a cross-sectional study. Lipids Health Dis. 2022;21:66.

52.

Soininen

Kangas

Würtz

, et al. High-throughput serum NMR metabonomics for cost-effective holistic studies on systemic metabolism. Analyst. 2009;134:1781-1785.

53.

Inouye

Kettunen

Soininen

, et al. Metabonomic, transcriptomic, and genomic variation of a population cohort. Mol Syst Biol. 2010;6:441.

54.

Soininen

Kangas

Würtz

Suna

Ala-Korpela

Quantitative serum nuclear magnetic resonance metabolomics in cardiovascular epidemiology and genetics. Circ Cardiovasc Genet. 2015;8:192-206.

55.

Holmes

Millwood

Kartsonaki

, et al. Lipids, lipoproteins, and metabolites and risk of myocardial infarction and stroke. J Am Coll Cardiol. 2018;71:620-632.

56.

World Health Organization. The Asia-Pacific Perspective: Redefining Obesity and Its Treatment. Regional Office of the Western Pacific (WPRO). World Health Organization; 2000.

57.

Kim

Lee

Provido

SMP

, et al. Association between Diet Quality and prevalence of obesity, dyslipidemia, and insulin resistance among Filipino immigrant women in Korea: the Filipino women’s diet and Health Study. Front Public Health. 2021;9. doi:10.3389/fpubh.2021.647661

58.

American Diabetes Association Professional Practice Committee. 2. Classification and diagnosis of diabetes: standards of medical care in diabetes-2022. Diabetes Care. 2022;45:S17-S38.

59.

Chobanian

Bakris

Black

, et al. Seventh Report of the Joint National Committee on Prevention, detection, evaluation, and treatment of high blood pressure. Hypertension. 2003;42:1206-1252.

60.

Jennings

MacGregor

Pallister

Spector

Cassidy

Associations between branched chain amino acid intake and biomarkers of adiposity and cardiometabolic health independent of genetic factors: a twin study. Int J Cardiol. 2016;223:992-998.

61.

Okekunle

Duan

Feng

Sun

Dietary intakes of branched-chained amino acid and risk for type 2 diabetes in adults: the Harbin Cohort Study on Diet, Nutrition and chronic Non-Communicable Diseases Study. Can J Diabetes. 2018;42:484-492.e7.

62.

Okekunle

Lee

Provido

SMP

, et al. Dietary intakes of branched-chain amino acids and plasma lipid profiles among Filipino women in Korea: the Filipino women’s Diet and Health Study (FiLWHEL). Nutr J. 2023;22(1):34.

63.

Zhang

Rao

Zhang

, et al. Dietary and circulating branched chain amino acids are unfavorably associated with body fat measures among Chinese adults. Nutr Res. 2024;128:94-104.

64.

Zheng

, et al. Cumulative consumption of branched-chain amino acids and incidence of type 2 diabetes. Int J Epidemiol. 2016;45:1482-1492.

65.

Kwon

Cruz

Marshall

McDaniel

ML.

Metabolic regulation by leucine of translation initiation through the mTOR-signaling pathway by pancreatic beta-cells. Diabetes. 2001;50:353-360.

66.

Bloomgarden

Diabetes and branched-chain amino acids: what is the link?

J Diabetes. 2018;10:350-352.

67.

Newton-Tanzer

Can

Demmelmair

, et al. Apparent saturation of branched-chain amino acid catabolism after high dietary milk protein intake in healthy adults. J Clin Endocrinol Metab. 2024;110:e1793-e1801.

68.

Walejko

Christopher

Crown

, et al. Branched-chain α-ketoacids are preferentially reaminated and activate protein synthesis in the heart. Nat Commun. 2021;12:1680.

69.

Samuelsson

Moberg

Apró

Ekblom

Blomstrand

Intake of branched-chain or essential amino acids attenuates the elevation in muscle levels of PGC-1α4 mRNA caused by resistance exercise. Am J Physiol Endocrinol Metab. 2016;311:E246-E251.

70.

Murphy

Saddler

Devries

McGlory

Baker

Phillips

SM.

Leucine supplementation enhances integrative myofibrillar protein synthesis in free-living older men consuming lower- and higher-protein diets: a parallel-group crossover study. Am J Clin Nutr. 2016;104:1594-1606.

71.

Ispoglou

White

Preston

McElhone

McKenna

Hind

Double-blind, placebo-controlled pilot trial of L-leucine-enriched amino-acid mixtures on body composition and physical performance in men and women aged 65-75 years. Eur J Clin Nutr. 2016;70:182-188.

72.

Piccolo

Comerford

Karakas

Knotts

Fiehn

Adams

SH.

Whey protein supplementation does not alter plasma branched-chained amino acid profiles but results in unique metabolomics patterns in obese women enrolled in an 8-week weight loss trial. J Nutr. 2015;145:691-700.

73.

Paulusma

Lamers

Broer

van de Graaf

SFJ

. Amino acid metabolism, transport and signalling in the liver revisited. Biochem Pharmacol. 2022;201:115074.

74.

Brosnan

ME.

Branched-chain amino acids: enzyme and substrate regulation. J Nutr. 2006;136:207S-211S.

75.

Mann

Mora

Madu

Adegoke

OAJ

. Branched-chain amino acids: catabolism in skeletal muscle and implications for muscle and whole-body metabolism. Front Physiol. 2021;12:702826.

76.

Liu

W-J

Yang

Zhao

Liu

HX.

The role of branched-chain amino acids and branched-chain α-keto acid dehydrogenase kinase in metabolic disorders. Front Nutr. 2022;9:932670.

77.

Cai

Wang

, et al. BCAA was more closely associated with visceral fat area than subcutaneous fat area in patients of type 2 diabetes mellitus: a cross-sectional study. BMC Endocr Disord. 2024;24:236.

78.

Jiang

Guo

Xiao

Role of branched-chain amino acid catabolism in the regulation of adipocyte metabolism. Endocrinology. 2025;166. doi:10.1210/endocr/bqaf089

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB