Circulating long-chain acylcarnitine concentrations are not affected by exercise training in pregnant women with obesity

Abstract

The purpose of this study was to determine the effect of exercise during pregnancy in sedentary women with obesity on longitudinal changes in long-chain acylcarnitine (LC-AC) concentrations. We hypothesized that exercise training would significantly decrease circulating LC-ACs throughout gestation compared with a nonexercise control group. Pregnant women with obesity considered otherwise healthy [n = 80, means ± SD; body mass index (BMI): 36.9 ± 5.7 kg/m²] were randomized into an exercise (n = 40, aerobic/resistance 3 times/wk, ∼13th gestation week until birth) or a nonexercise control (n = 40) group. At gestation week 12.2 ± 0.5 and 36.0 ± 0.4, a submaximal exercise test was conducted, and indirect calorimetry was used to measure relative resting energy expenditure (REE), as well as respiratory exchange ratio (RER) at rest. Fasting blood samples were collected and analyzed for LC-AC concentrations. Fitness improved with prenatal exercise training; however, exercise training did not affect circulating LC-AC. When groups were collapsed, LC-ACs decreased during gestation (combined groups, P < 0.001), whereas REE (kcal/kg/day, P = 0.008) increased. However, average REE relative to fat-free mass (FFM) (kcal/kg FFM/day) and RER did not change. There was an inverse relationship between the change in RER and all LC-ACs (except C18:2) throughout gestation (C14: r = −0.26, P = 0.04; C16: r = −0.27, P = 0.03; C18:1: r = −0.28, P = 0.02). In summary, a moderate-intensity exercise intervention during pregnancy in women with obesity did not alter LC-ACs concentrations versus control, indicating that the balance between long-chain fatty acid availability and oxidation neither improved nor worsened with an exercise intervention.

NEW & NOTEWORTHY This research showed that a moderate-intensity prenatal exercise program, consisting of aerobic and resistance training, did not negatively impact normal alterations in substrate supply and demand for the mother and the offspring throughout gestation. Findings provide support for metabolic safety of exercise during pregnancy.

INTRODUCTION

In obesity, excess lipids in circulation can result in a mismatch between fuel availability (e.g., long-chain fatty acids, LCFA) and energy turnover (energy demand). LCFA availability that outpaces mitochondrial β-oxidation and/or alterations in the function of the machinery responsible for β-oxidation and oxidative phosphorylation (1, 2) can result in the accumulation of long-chain acylcarnitine moieties [LC-AC, e.g., myristoylcarnitine (C14), palmitoylcarnitine (C16), oleoylcarnitine (C18:1), linoleoylcarnitine (C18:2)], formed from the esterification of a LCFA to a molecule of carnitine, catalyzed by carnitine palmitoyltransferase 1 (3). LC-ACs that accumulate are eventually trafficked out of the mitochondria and ultimately into circulation (4, 5). Accordingly, LC-ACs can be used as markers of the coupling of fatty acid availability relative to the metabolic demand of energy in tissues such as skeletal muscle [“metabolic overload” (6)]. Maternal body mass index (BMI) is positively associated with circulating concentrations of LC-ACs (7), indicating that there may be an over-availability of fatty acids in an environment that does not match the demand. This high-supply environment during pregnancy could potentially have ramifications for the growing fetus, supported by the Developmental Origins of Health and Disease (DOHaD) theory, suggesting that the maternal environment directly influences the short- and long-term health of the offspring (8). Thus, it is critical to determine lifestyle practices that can be used to attenuate the metabolic burden of maternal obesity on the growing offspring.

Exercise training reduces circulating resting LC-AC concentrations in individuals with obesity (9, 10). However, to our knowledge, no reports have examined this relationship in pregnancy. Here, we investigated the impact of a moderate-intensity aerobic and resistance exercise intervention during pregnancy on circulating resting LC-AC levels in women with obesity. We hypothesized that exercise would reduce circulating resting LC-AC levels during pregnancy.

MATERIALS AND METHODS

Participants

This study is a subanalysis of 80 women from an ongoing randomized controlled trial (ClinicalTrials.gov Identifier: NCT02125149) at the Arkansas Children’s Nutrition Center (ACNC), and methods have been previously published (11). In the parent study, participants were recruited in the first 12 wk of gestation. After screening, participants were included if they satisfied the following criteria: prepregnancy obese body mass index (BMI) classification (≥30.0 kg/m²), singleton pregnancy, ≥18 yr of age, sedentary (did not engage in intentional physical activity, sedentary work activity level), and conception without assisted fertility treatments. Participants were excluded if they satisfied the following criteria: preexisting medical conditions (e.g., gestational diabetes mellitus, diabetes, hypertension), use of recreational drugs, tobacco or alcohol during pregnancy, and contraindications to exercise during pregnancy. After enrollment, participants were randomized to either an exercise or standard-of-care control group. An analysis concluded that 40 participants/group were needed to achieve sufficient power for the current substudy (see Statistical Analysis), the exercise group was chosen as the 40 women with the highest exercise compliance (>56%). The control group was matched to the exercise group based on BMI and race.

Before participation, participants were required to provide a signed release from their OB/GYN or primary care physician/nurse practitioner. All experimental procedures were conducted according to the Declaration of Helsinki and were approved by the Institutional Review Board at the University of Arkansas for Medical Sciences. Written informed consent was provided by each participant.

Fitness Testing

A submaximal graded walking treadmill test was conducted before the intervention (visit 1, 12.2 ± 0.5 wk gestation), and in the middle of the intervention after the second trimester (visit 2, 24.1 ± 0.6 wk gestation), as previously described (11). The percent grade [% grade at Borg Rating of Perceived Exertion (RPE)-15], oxygen uptake (V̇o₂) relative to body mass at RPE-15 (mL/kg/min), respiratory exchange ratio (RER at RPE-15), and work (watts) were recorded at both time points, and delta values were calculated (visit 2 – visit 1).

Exercise Training

The details of the exercise program used were described previously (11). In short, the moderate-intensity exercise protocol started at ∼13 wk of gestation and was periodized to increase the duration and intensity of exercise during the first 6 wk of training and thereafter to maintain the duration and intensity of exercise until birth, if possible. Participants in the exercise group completed ∼30–45 min of combined aerobic, resistance, and flexibility exercises thrice weekly. All exercise sessions were supervised by a physical trainer to coach the women one-on-one. Moderate intensity during aerobic exercise was ensured by tracking the participants’ Borg Rating of Perceived Exertion (RPE, kept between 12 and 14) (12), and heart rate using heart rate monitors (Polar Electro, Kempele, Finland), in addition to communicating with the participant regularly throughout the session using the talk test (able to hold a conversation). Compliance was defined as the percentage of exercise sessions completed relative to the goal of three exercise sessions per week from ∼13 wk gestation until the participant could no longer exercise, whether because of birth or physician recommendations. Participants in the exercise group were encouraged to gradually increase their average daily steps, tracked using a Garmin (Garmin Ltd., Olathe, KS) or Fitbit (Fitbit, San Francisco, CA) device.

Anthropometrics and Body Composition

Weight was measured to the nearest 0.1 kg using a tared standing digital scale (Perspective Enterprises, Portage, MI), and height was measured to the nearest 0.1 cm using a wall-mounted stadiometer (Tanita Corp., Tokyo, Japan). Gestational weight gain (GWG) and BMI were determined using standard equations. Body composition [fat-free mass (FFM)] was measured using air displacement plethysmography (ADP, BodPod, Concordia, CA).

Habitual Diet and Physical Activity Assessment

Habitual dietary intake was assessed before both visits using 3-day food records and the Nutrition Data System for Research (NDSR, Nutrition Coordinating Center, University of Minnesota, MN) software. Total energy intake (TEI, kcal/day), total macronutrient [carbohydrate (CHO), protein (PRO), and fat] intake relative to body mass (g/kg/day), and percent macronutrient intake (%) were assessed. An Actical accelerometer (Philips Respironics Co. Inc., Bend, OR) worn on the participant’s nondominant ankle was used to measure average daily activity counts (AC), daily steps, and time spent in sedentary, light, moderate, and strenuous activity before each visit.

Resting Energy Expenditure

Oxygen uptake relative to body mass (V̇o₂, mL/kg/min) and FFM (V̇o₂, mL/kg FFM/min), resting energy expenditure relative to body mass (REE, derived from V̇o₂, kcal/kg/day) and FFM (REE, derived from V̇o₂, kcal/kg FFM/day), and respiratory exchange ratio (RER) were measured by open-circuit indirect calorimetry (Moxus, AEI technologies, IL) with a ventilated hood. REE was expressed relative to body mass (kcal/kg/day) and to fat-free mass (FFM, kcal/kg FFM/day) because changes in whole body metabolism throughout pregnancy are largely driven by an increase in body mass (13).

Blood Collection and Analysis

Fasted blood samples from an antecubital vein following an overnight fast were obtained in the morning before the start of the intervention (visit 1, 12.2 ± 0.5 wk of gestation), and during the third trimester (visit 2, 36.0 ± 0.4 wk of gestation). Blood samples were collected in serum vacutainers (Becton, Dickinson & Company, Franklin Lakes, NJ) for the quantification of LC-AC (C14, C16, C18:1, and C18:2) concentrations.

LC-AC quantification.

Myristoyl-l-carnitine, palmitoyl-l-carnitine, oleoyl-l-carnitine, linoleoyl-l-carnitine, myristoyl-l-carnitine-d3, palmitoyl-l-carnitine-d3, and oleoyl-l-carnitine-d3 were purchased from Cayman Chemical (Ann Arbor, MI). All solvents used were of optima grade from Fisher Scientific (Pittsburg, PA). All stock standards (1 mg/mL) were prepared in methanol. Working calibration standards (0–1,000 ng/mL) were diluted in methanol:water (25:75) and spiked with deuterated compounds (100 ng/mL) as internal standards. Serum samples (100 µL) were spiked with internal standards (100 ng/mL, 100 µL). Methanol:water (50:50, 400 µL) was added and the samples vortexed for 20 s. Acetonitrile (800 µL) was then added and samples vortexed for another 20 s. Samples sat on ice for 20 min before centrifuging at 18,000 g for 10 min at 4°C. Supernatant was collected and dried down under a nitrogen stream. Extracts were reconstituted in 100 µL of methanol:water (25:75).

Chromatographic separation was performed on an UltiMate 3000 UHPLC system (Thermo Fisher Scientific, Waltham, MA) fitted with an Xselect CSH C18 column (100 × 2.1 mm, 2.5 µm; Waters, Milford, MA) kept at 30°C while samples in the autosampler were kept at 4°C. A flow rate of 200 µL/min and injection volume of 5 µL was used. Mobile phases consisted of 0.1% trifluoroacetic acid in water (A) and 0.1% trifluoroacetic acid in methanol (B) with a 15-min elution gradient as follows: hold 10% B for 1 min, ramp to 99% B over 4 min, hold 99% B for 2 min, return to 10% B in 1 min, hold 10% B for 7 min.

Identification was carried out on a SCIEX QTRAP 4000 (Framingham, MA) mass spectrometer with data acquisition and analysis performed using Analyst 1.7 software. Data were acquired by multiple reaction monitoring (MRM) in positive Turbo spray ionization mode. Nitrogen as curtain, CAD, GS1, and GS2 gas was set at 25, high, 45, and 45 units, respectively. Ion spray voltage and source temperature were at 5,000 V and 500°C.

Statistical Analysis

A sample size calculation was performed for a repeated-measures ANOVA using the GLMPOWER procedure in SAS version 9.4 (SAS Institute, Cary, NC). Based on a two-tailed α of 0.05 and a standard deviation of 0.006, the total n was determined to be 80 (i.e., 40 in each group). This yielded a power of 0.80 to detect a difference of 0.03 µmol/L in mean C16 and C18:1 LCAC concentration [based on a previous exercise study in women with obesity (9)].

Circulating LC-AC concentrations were analyzed with repeated-measures ANOVA, where time of gestation (12 and 36 wk) was repeated, and group (exercise or no exercise) was a factor in the model. All concentrations met the assumptions of normality and equal variances. SAS software, version 9.4 (SAS Institute, Cary, NC) was used to conduct analyses. Results were considered statistically significant if P < 0.05.

Analysis of dietary intake.

A linear model with time and group was used to determine if there were differences in dietary intake. Pearson’s correlations were used to determine if there were relationships between dietary intake and each LC-AC at 12 and 36 wk of gestation.

Analysis of circulating LC-AC concentrations and energy metabolism variables.

Differences in LC-ACs (C14, C16, C18:1, and C18:2) and metabolic variables (V̇o₂, REE, and RER) were analyzed using repeated-measures ANOVA. This model incorporated the main effect of group (exercise vs. control), the repeated effect of time (12 vs. 36 wk of gestation), and the time by group interaction. Early pregnancy BMI, early pregnancy V̇o₂ during submaximal exercise, the change in dietary fat intake relative to body mass throughout gestation, and gestational weight gain (GWG) were originally included in the model, but were not a significant source of variation, and were therefore removed. The outcome variables were examined for normality and homogeneity, and the homogeneity of the slopes with the covariates was tested. All assumptions of the ANOVA were met. A test of the main effect of group was also performed for δ LC-AC concentrations and energy metabolism variables. Pearson correlations were also used to assess the relationship between the change in metabolic variables (V̇o₂, REE, and RER) and the change in LC-ACs.

Analysis of performance variables.

A Student’s t test was used to compare the delta values of all performance variables between exercise and nonexercise groups.

Analysis after collapsing groups.

A one-factor analysis of covariance (factor: week) was run after combining groups.

RESULTS

There were no differences in baseline characteristics at 12 wk between groups (Table 1), as noted in our previous report from the same cohort (11). In the exercise group, the average compliance to the exercise program was 76%. As a result of the intervention, all exercise performance measures (%grade, V̇o₂, RER, and work at an RPE of 15) at 24 wk were improved compared with baseline (Table 2), and physical activity at 36 wk was higher in the exercise group compared with the control group, reported in our previous paper (11).

Table 1.

Baseline characteristics of participants within the exercise and control groups of sedentary pregnant women with obesity

Variable	Exercise	Control
n	40	40
Age, yr	29.2 ± 4.6	28.7 ± 4.4
Parity	2 ± 1	2 ± 1
Race
Caucasian	29 (72%)	29 (72%)
African American	9 (22%)	11 (28%)
More than one/Missing	2 (4%)	0 (0%)
Ethnicity (non-Hispanic)	37 (92%)	36 (90%)
BMI, 12 wk, kg/m2	36.2 ± 4.2	37.6 ± 6.8
BF%, 12 wk, %*	48.7 ± 5.2	48.5 ± 5.5
GWG, kg	10.0 ± 5.5	8.7 ± 5.1

Means ± SD; n = 80, except BF%. No statistical differences between exercise and control for any of the presented variables (P > 0.05).

^*n = 79 because of missing data. BF%, body fat percentage; BMI, body mass index; GWG, gestational weight gain.

Table 2.

Performance variables at a workload corresponding to the Borg scale rate of perceived exertion of 15 at 12 wk (preintervention), ∼24 wk (middle of the intervention) of gestation, and the delta values between the timepoints in sedentary women with obesity

	Control, n = 35			Exercise, n = 38
	12 Wk	24 Wk	Delta	12 Wk	24 Wk	Delta	P Value
% Grade at RPE-15	8 ± 3	7 ± 2	−1 ± 0.3	8 ± 2	10 ± 2	2 ± 0.4	<0.0001
V̇o2 at RPE-15, mL/kg/min	17.4 ± 3.0	16.3 ± 2.5	−1.1 ± 0.4	17.3 ± 3.1	18.3 ± 2.4	1.0 ± 0.4	0.0002
RER at RPE-15	1.1 ± 0.1	1.0 ± 0.1	−0.02 ± 0.02	1.0 ± 0.1	1.1 ± 0.1	0.04 ± 0.02	0.008
Work at RPE-15, W	100 ± 47	94 ± 35	−6 ± 5	109 ± 45	139 ± 9	30 ± 5	<0.0001

Values in the table are means ± SE. A Student’s t test was used to compare the delta values of all performance variables between exercise and nonexercise groups. P values denote delta values of control vs. exercise groups, % grade denote percent grade on treadmill. RER, respiratory exchange ratio (V̇co₂/V̇o₂); RPE, rate of perceived exertion; V̇o₂, oxygen uptake.

Results of Analysis of Dietary Intake

There were no significant interaction effects or main effects of time for any of the measured dietary variables, as reported previously (11). From Pearson’s correlation analysis comparing LC-ACs to dietary variables at 12 wk, there were positive relationships between C14 and TEI (r = 0.31, P = 0.01), absolute dietary fat intake (r = 0.32, P = 0.01), and absolute CHO intake (r = 0.28, P = 0.03), but there were no relationships with any other variable (absolute or relative to body mass). At 36 wk, no relationships between LC-ACs and dietary variables were observed.

Analysis of Circulating LC-AC Concentrations and Energy Metabolism Variables

The coefficient of variation for the measured variables was as follows: C14 = 2.6%, C16 = 1.9%, C18:1 = 2.4%, and C18:2 = 4.3%. The repeated-measures ANOVA revealed no significant group × time interactions for the LC-ACs or any other metabolic variable (Table 3). To look at the effect of time regardless of group, when groups were collapsed, all LC-ACs significantly decreased (all P < 0.001), and V̇o₂ relative to body mass (P = 0.010) and REE relative to body mass (P = 0.008) increased during pregnancy (Table 4). There was no effect of time on any other metabolic variable.

Table 3.

Metabolic measures in exercise and nonexercise control groups at baseline preintervention (12 wk) and 36 wk of gestation

	Control, n = 40			Exercise, n = 40
	12 Wk	36 Wk	P Value	12 Wk	36 Wk	P Value
C14, µmol/L	0.020 (0.001)	0.017 (0.001)	0.265	0.023 (0.001)	0.017 (0.001)	0.005
C16, µmol/L	0.113 (0.003)	0.091 (0.003)	<0.001	0.119 (0.003)	0.092 (0.003)	<0.001
C18:1, µmol/L	0.235 (0.009)	0.184 (0.009)	<0.001	0.250 (0.009)	0.187 (0.008)	<0.001
C18:2, µmol/L	0.252 (0.008)	0.203 (0.008)	<0.001	0.264 (0.008)	0.208 (0.008)	<0.001
V̇o2, mL/kg/min	2.85 (0.50)	3.03 (0.61)	0.086	2.08 (0.31)	3.02 (0.38)	0.004
V̇o2, mL/kg FFM/min	5.51 (0.82)	5.64 (1.08)	0.281	5.51 (0.58)	5.75 (1.18)	0.141
REE, kcal/kg/day	19.8 (3.15)	21.1 (4.11)	0.067	19.5 (2.13)	21.0 (2.63)	0.006
REE, kcal/kg FFM/day	38.3 (5.57)	39.4 (7.38)	0.238	38.4 (4.00)	39.9 (8.01)	0.153
RER	0.808 (0.010)	0.806 (0.010)	0.895	0.793 (0.010)	0.830 (0.009)	0.028

Values in the table are least squares means and (standard errors). Data was examined using a repeated-measures ANOVA. This model incorporated the main effect of group (exercise vs. control), the repeated effect of time (12 vs. 36 wk of gestation), and the time by group interaction. P values indicate within-group time effects. Bold values represent statistical significance. No delta values differed significantly between groups. Serum samples were used to quantify acylcarnitines (ACs). Indirect calorimetry was used to measure V̇o₂, REE, and RER. ANCOVA yielded no significant group × time interaction (NS). C14, myristoylcarnitine; C16, palmitoylcarnitine; C18:1, oleoylcarnitine; C18:2, linoleoylcarnitine; FFM: fat-free mass; REE, resting energy expenditure; RER, respiratory exchange ratio; V̇o₂, volume of oxygen consumption.

Table 4.

Circulating long-chain acylcarnitine concentrations and resting energy metabolism measures after combining exercise and control groups at 12 wk and 36 wk gestation in sedentary women with obesity

	12 Wk	36 Wk	P
C14, µmol/L	0.021 (0.001)	0.017 (0.001)	<0.001
C16, µmol/L	0.113 (0.002)	0.092 (0.002)	<0.001
C18:1, µmol/L	0.237 (0.006)	0.185 (0.006)	<0.001
C18:2, µmol/L	0.253 (0.005)	0.206 (0.005)	<0.001
V̇o2, mL/kg/min	2.82 (0.392)	3.02 (0.510)	0.010
V̇o2, mL/kg FFM/min	5.51 (0.711)	5.69 (1.13)	0.243
REE, kcal/kg/day	19.66 (0.37)	21.07 (0.38)	0.008
REE, kcal/kg FFM/day	38.36 (0.76)	39.69 (0.78)	0.221
RER	0.813 (0.006)	0.819 (0.006)	0.503

Values in the table are means and (standard deviations). A one-factor analysis of covariance (factor: week) was run after combining groups. Bold values represent statistical significance. Serum samples were used to quantify acylcarnitines (ACs). Indirect calorimetry was used to measure V̇o₂, REE, and RER. C14, myristoylcarnitine; C16, palmitoylcarnitine; C18:1, oleoylcarnitine; C18:2, linoleoylcarnitine; REE, resting energy expenditure; RER, respiratory exchange ratio; V̇o₂, volume of oxygen uptake.

Using Pearson correlations, there were no relationships between the change in V̇o₂ or REE and the change in LC-ACs. However, there was a significant negative relationship between RER and each of the LC-ACs (except C18:2), although the effect sizes were small (C14: r = −0.26, P = 0.04; C16: r = −0.27, P = 0.03; C18:1: r = −0.28, P = 0.02).

DISCUSSION

During pregnancy, there are two metabolic periods that facilitate the adaptation of the increasing energy demands of growing tissues (14, 15). During the anabolic first half of pregnancy, maternal energy storage is prioritized, and fat oxidation is reduced compared with prepregnancy (14, 15). However, during the second half of pregnancy, there is a metabolic shift characterized by increased lipolysis and fat oxidation (14, 15). Our data support this notion, whereby the reduction in LC-ACs with advancing gestation suggest a better matching between LCFA availability and oxidation of fatty acids within the mitochondrion (Table 3). These metabolic shifts during pregnancy ensure proper fuel partitioning to both the mother and the fetus. Of interest, although maternal obesity is associated with increased circulating LC-ACs (7), and exercise is known to reduce circulating LC-ACs in individuals with obesity (9, 10), to our knowledge, no studies to date have assessed the impact of prenatal exercise on LC-AC concentrations in women with obesity.

Thus, in the current study, we aimed to determine the effect of an exercise intervention during pregnancy in sedentary women with obesity on longitudinal changes in circulating LC-ACs. Exercise is known to improve the capacity of tissues such as skeletal muscle to oxidize LCFAs (9, 10). Thus, the premise of this study was that exercise during pregnancy would increase whole body lipid oxidation and concurrently reduce resting LC-AC levels. Zhang et al. demonstrated that there was a reduction in circulating concentrations of C16, C18:1, and C18:2 at rest, during acute exercise, and postexercise in women with obesity (preintervention BMI: 33.5 ± 0.6 kg/m²) that were sedentary and insulin-resistant, after a long-term (∼14 wk) aerobic training [weeks 1–4: 4 days/wk for 30 min at 60%–70% of maximal heart rate (HR); weeks 5–8: 4 days/wk for 40 min at 60%–70% of maximal HR; weeks 9 onward: 4 days/wk for 40 min at 75% of maximal HR) and energy restriction (500–600 kcal/day reduction) program (9). Thus, although the current study focused on the effects of chronic exercise on resting metabolism, future work should target metabolism during exercise because the intraexercise period could provide a window of optimization for significant clinical improvements in metabolic markers of health-compromised populations such as women with obesity and insulin resistance. Nonetheless, it seems as though baseline LC-ACs had limited bearing on the effect of the exercise intervention because baseline C18:1 and C18:2 concentrations were higher in the current study compared with baseline LC-ACs in the Zhang study (although concentrations of C16 were lower in the current study). However, weight loss may be a contributing factor to the magnitude of change in LC-ACs, because participants in the Zhang et al. study lost an average of ∼5.0 kg with the exercise/weight loss intervention, whereas the pregnant women in the exercise intervention in the current study gained ∼10 kg. In addition, the study by Zhang et al. used energy deficit in addition to an exercise intervention, which could have contributed to a reduction in LC-AC generation (e.g., energy deficit may reduce circulating levels of fatty acids). Energy restriction is not feasible in most pregnant populations; however, improving macronutrient intake (e.g., reducing fat intake, and increasing protein intake) may have potential for reducing resting LC-AC concentrations, although future work is required to support this theory. In addition, it may be that women who exercise during pregnancy have a lower energy balance because they are physically moving more. In fact, the women in the exercise group in the current study had higher average daily steps, moderate activity time, and vigorous activity time at 36 wk gestation compared with the control group. Overall, in the current study we found that prenatal exercise did not impact circulating LC-ACs, which confirms our previous report examining short-chain acylcarnitines, representative of the matching between branched-chain amino acid availability and utilization (11). Thus, it may be concluded that, regardless of exercise training, there is a tight regulation of these metabolites at rest. Resting acylcarnitine concentrations characteristically decrease as pregnancy progresses (16), likely the result of reduced maternal LCFA supply (from increased shunting to fetal tissues) and increased maternal/fetal tissue deposition during growth. The results of the current study show that this decrease in LC-FA concentrations was not further augmented by an exercise intervention during pregnancy, and furthermore, there was no accumulative effect of prenatal exercise on birth weight of the babies in this study demonstrated by no difference between groups in a post hoc analysis (Student’s t test, Control: 3.5 ± 0.3 kg; Exercise: 3.7 ± 0.2 kg; P = 0.141).

Of interest, the exercise protocol improved performance-related variables specific to aerobic performance but did not augment the reduction in LC-AC concentrations. After all, when performing moderate-intensity aerobic exercise, human skeletal muscle preferentially oxidizes fatty acids to a greater extent than carbohydrates (17, 18), and chronic moderate-intensity aerobic training increases resting fatty acid oxidation (19). Therefore, it could be surmised that if exercise training improves fatty acid oxidative efficiency both during exercise and at rest, and therefore more fatty acids are being more completely oxidized, it would subsequently reduce the production and trafficking of LC-ACs from the muscle into circulation. However, it should be noted that circulating concentrations of acylcarnitines may be derived from other sources than muscle tissue (6, 20). For instance, circulating concentrations of LC-AC are related to LC-AC production and concentrations in cardiac tissue (21) and peripheral blood mononuclear cells (22), indicating that other tissues, in addition to skeletal muscle, produce LC-ACs. Therefore, no conclusions can be drawn about where the circulating LC-ACs were derived from in the current study. Thus, future work should address this limitation using methodologies that allow for the determination of tissue-specific LC-AC production and trafficking.

Aside from the lack of intervention effect on concentrations of these circulating markers of fatty acid oxidation, there were no interaction effects of group and time on RER. However, exercise training may modulate substrate utilization during exercise and immediately postexercise more so than at rest. For instance, 12 mo of a jogging/walking program at 60% of heart rate reserve in normal weight sedentary individuals increased maximal fatty acid oxidation during exercise, and maximal fatty acid oxidation occurred at higher intensities after the exercise intervention compared with baseline in previously untrained men and women (23). However, there were no changes in resting substrate utilization with exercise training (23). Thus, exercise training improves metabolic flexibility during the exercise and the postexercise window, and the summation of this increased efficiency over a length of time may contribute to other health-related outcomes (e.g., reduction in disease risk) independent of changes in metabolism at rest. However, future work should assess further the impact of exercise training during pregnancy on both resting and exercise/postexercise energy expenditure and substrate use, especially because metabolic flexibility is impaired in pregnant women that are overweight or have obesity compared with pregnant women with normal weight (24). Furthermore, impaired metabolic flexibility in response to a meal is related to degree of inflammation and insulin resistance in pregnant women (24), whereas these relations have not been determined in response to exercise training in a population with obesity, such as the ones in the current study. Interestingly, in our previous report using the same group, we found that exercise training did not change the homeostatic model assessment for insulin resistance (HOMA2-IR, an indirect marker of insulin resistance using fasting glucose and insulin concentrations) (11). However, this group of women, although obese, were otherwise healthy and void of underlying metabolic impairments, such as gestational diabetes mellitus, which may explain the lack of effect on HOMA2-IR in this subpopulation of the ongoing RCT (or lack of power) (see materials and methods). Therefore, it may be that metabolic responsiveness to an exercise intervention during pregnancy is a function of the magnitude of the underlying metabolic disturbance, or other factors such as exercise intensity should be scrutinized. Also, the exercise intensity and duration are directly related to an individual’s physiological response to an exercise program (25), and this study may have been limited by the selection of optimal exercise dosage. However, maintenance of exercise intensity throughout pregnancy in previously sedentary women can be practically difficult. Because of the general lack of randomized controlled trials of exercise during pregnancy, it is difficult to quantify optimal exercise intensity to elicit beneficial metabolic changes, and more such trials are needed. Nevertheless, exercise intensity can be difficult to monitor during pregnancy compared with traditional ways of measuring intensity as a result of the physiological changes that occur during gestation. For instance, heart rate-based methods are commonly used to determine exercise intensity in nonpregnant individuals; however, blunted heart rate responses during exercise have been observed during pregnancy and therefore other means of measuring intensity (e.g., rate of perceived exertion) are recommended to be used (26).

In summary, chronic moderate-intensity exercise from ∼13 to 36 wk gestation in sedentary women with obesity did not influence circulating concentrations of LC-AC, suggesting that exercise does not modulate the coupling between LCFA availability and oxidation in tissue in the resting state. Thus, our current data suggest moderate-intensity aerobic and resistance training does not appreciably alter resting-state fuel partitioning during pregnancy.

GRANTS

This study was supported by Arkansas Children’s Research Institute and Arkansas Biosciences Institute Postgraduate Grant (to B. R. Allman and B. J. Spray), US Department of Agriculture—Agricultural Research Service Projects 6026-51000-010-05S and 6026-51000-012-06S (to B. R. Allman, R. S. Lan, A. Andres, and E. Børsheim), NIH/National Institute of Diabetes and Digestive and Kidney Diseases R01 DK107516 (to B. R. Allman A. Andres, and E. Børsheim), NIH/National Institute of General Medical Science COBRE 1P20GM109096-01A1 (to A. Andres and E. Børsheim), and NIH/National Center for Advancing Translational Sciences (NCATS) UL1TR003107 (to E. Børsheim).

DISCLAIMERS

The results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation, and the results of the present study do not constitute endorsement by American College of Sports Medicine (ACSM).

DISCLOSURES

B. R. Allman has a podcast about exercise and health-related outcomes (“BENT by Knowledge”) and is also the Senior Innovation Scientist for Breakout Lifestyle Fitness, Little Rock, a gym emphasizing resistance training and health-related outcomes. None of the other authors has any conflicts of interest, financial or otherwise, to disclose.

AUTHOR CONTRIBUTIONS

B.R.A., A.A., and E.B. conceived and designed research; R.S.L., A.A., and E.B. performed experiments; B.R.A., B.J.S., and R.S.L. analyzed data; B.R.A. interpreted results of experiments; B.R.A. and E.B. drafted manuscript; B.R.A., B.J.S., R.S.L., A.A., and E.B. edited and revised manuscript; B.R.A., B.J.S., R.S.L., A.A., and E.B. approved final version of manuscript.