Date of Award

Summer 2020

Document Type

Open Access Thesis


Exercise Science

First Advisor

Mark Sarzynski


Background: Elevated levels of circulating branched-chain amino acids (BCAA) and ketone bodies are recognized as biomarkers for cardiovascular disease (CVD) and other pathological conditions in type-2 diabetes mellitus (T2DM). Aerobic exercise interventions have been shown to decrease the levels of these markers, suggesting improved metabolic status and reduced risk of CVD. However, the efficacy of resistance training and concurrent programs in reducing BCAA and ketone body levels has not been well researched.

Methods: The current study was performed as a secondary analysis of the HART-D trial, a 9-month randomized, controlled exercise-training trial of 262 participants with T2DM. Participants were randomized to one of four groups: non-exercise control, aerobic training (AT), resistance training (RT), or a combined aerobic and resistance training (ATRT). The effects of the 9-month intervention on BCAAs (leucine, valine, and isoleucine) and ketone bodies (β-hydroxy-butyrate, BHB; acetoacetate, AcAc; and acetone) were quantified by nuclear magnetic resonance spectroscopy (NMR) at LabCorp (Morrisville, NC). Generalized linear models were used to examine effects of exercise training between groups with adjustments for age, sex, race, change in fat mass, glucose, and medication status and baseline trait value. Pearson correlation analysis was used to examine associations of the changes in BCAA and ketone levels with changes in concomitant cardiometabolic biomarkers.

Results:The ATRT group increased total BCAA and leucine levels compared to the AT group, and increased isoleucine compared to all other groups (all p<0.05). RT decreased BHB levels (p<0.05) compared to the AT group only. Across all exercise groups combined, changes in total ketone bodies (r=0.2), BHB (r=0.21), and Acetone (r=0.17) were weakly correlated with changes in HbA1c levels. Changes in total BCAAs (r=0.30) and valine (r=0.36) were moderately correlated with changes in fasting glucose levels, while isoleucine was weakly correlated with glucose (r=0.2) (all p<0.05).

Conclusions: Our results show that the ATRT group increased isoleucine levels compared to the control group in diabetics, the mechanism of which is unclear. Exercise induced changes in BCAA and ketone body levels are weakly to moderately related to some concomitant cardiometabolic biomarkers such as fasting glucose and HbA1c levels. Further research is needed to examine the association of exercise training on circulating BCAA and ketone body levels in diabetics.