Comparison of indirect calorimetry and a new breath 13C/12C ratio method during strenuous exercise J. A. ROMIJN,

E. F. COYLE,

J. HIBBERT,

AND R. R. WOLFE

Metabolism Unit, Shriners Burns Institute, and Departments of Anesthesiology and Surgery, University of Texas Medical Branch, Galveston 77550; and Human Performance Laboratory, Department of Kinesiology and Health, The University of Texas at Austin, Austin, Texas 78712 Romijn,

J. A., E. F. Coyle,

J. Hibbert,

and R. R. Wolfe.

Comparisonof indirect calorimetry and a new breath l:C/‘*C ratio method during strenuous exercise. Am. J. Physiol. 263 (Endocrinol. Metab. 26): E64-E71, 1992.-A new stable isotope method for the determination of substrate oxidation rates in vivo is describedand comparedwith indirect calorimetry at rest and during high-intensity exercise (30 min at 80-85% maximal 0, uptake capacity) in six well-trained cyclists. This method usesthe absolute ratios of 13C/12Cin expired air, endogenous glucose,fat, and protein in addition to 0, consumption and is independent of CO, production (VCO,). Carbohydrate and fat oxidation rates at rest, calculated by both methods, were not significantly different. During exercisethe breath 13C/12Cratio increasedand reacheda steady state after 15-20 min. Carbohydrate oxidation rates during exercisewere 39.4 & 5.2 and 41.7 k 5.7 mg kg-l min-l [not significant (NS)], and fat oxidation rates were 7.3 & 1.3 and 6.9 & 1.2 mgekg-l=min-l (NS), using indirect calorimetry and the breath ratio method, respectively. We concludethat the breath 13C/12Cratio method can be used to calculatesubstrateoxidation under different conditions, such asthe basalstate and exercise.In addition, the resultsobtained by this new method support the validity of the underlying assumption that indirect calorimetry regards%o., asa reflection of tissue CO, production, during exercise in trained subjects, even up to 80-85% maximal 0, uptake. stable isotopes;massspectrometry; substrate oxidation l

l

underestimation of substrate oxidation (19). In the application of these tracer methods, account must also be taken of changes in bicarbonate recovery between rest and exercise (2, 17, 21). To circumvent the use of VCO, and the problem of isotopic exchange in the TCA cycle, we have developed an alternative approach to calculate substrate oxidation rates at rest and during exercise. It involves the measurement of the absolute 13C/12C ratios in expired breath and in endogenous glucose, fat, and protein. With the measurement of these absolute 13C/12C ratios and of Vo2, the absolute rates of substrate oxidation can be calculated, independently of VCO~. This method relies on only a small difference in the absolute 13C/12C ratios of endogenous substrates. To amplify the naturally occurring difference between endogenous carbohydrates, protein, and fat, we used an exhaustive exercise protocol on the day before the actual measurements to deplete endogenous glycogen stores, which was followed by glycogen repletion with 13C-enriched cornstarch. This new breath ratio method was used to assess the validity of indirect calorimetry for estimation of carbohydrate and fat oxidation rates during high-intensity exercise. METHODS

TWO METHODS are available to quantify whole body substrate oxidation rates at rest and during exercise: indirect calorimetry and tracer techniques. Indirect calorimetry relies on the assumption that 0, consumption (Vo2) and CO2 production (VCO,), as measured in expired air, reflect gas exchange at the tissue level (7). This is probably true under both conditions for 02, of which there are no large stores in the body. In contrast, VCO~ is only a reliable estimate of tissue CO2 production in the presence of a stable bicarbonate pool. However, during exercise, even of moderate intensity, bicarbonate kinetics are markedly altered (2). During exercise above the lactate threshold, a depletion of the bicarbonate pool may result in an overestimation of tissue CO2 production and consequently of carbohydrate oxidation (with a concomitant underestimation of fat oxidation). A second approach to quantify substrate oxidation uses administration of 13C- or 14C-enriched substrates, such as glucose and amino acids (17). The oxidation rate is calculated from the rate of excretion of labeled CO2 and the enrichment of the substrate in the plasma. This tracer method, however, is hampered by the same potential limitations with respect to bicarbonate metabolism as indirect calorimetry. In addition, there may be exchange of labeled CO2 within the tricarboxylic acid (TCA) cycle, which will result in decreased recovery of CO2 in the breath and thus in an

PRESENTLY,

E64

0193~1849/92

$2.00

Copyright

Subjects

Six highly trained endurancecyclists [age 24 t 1 yr, weight 76.2 . k 3.2 kg, height 1.79 t 3 m, maximal 0, uptake capacity wo 2 max)62 & 3 ml kg-l. min-‘1 volunteered for this study. All subjectswere healthy, asindicated by medicalhistory and physical examination. They were consuminga weight-maintaining diet containing at least 250 g carbohydrates daily. To2 maxhad been determined several weeks before the present protocol, while the subjects cycled on a stationary ergometer (Monark 819). \io, max was determined during an incremental cycling protocol lasting 7-10 min. The study was approved by the institutional review boards of the University of Texas at Galveston and Austin. l

Experimental

Protocol

The experiment was performed on two consecutive days in four subjects and on three consecutive days in two subjects (Fig. 1). The 1st day served to increasethe 13Cenrichment of endogenouscarbohydrates by a combined glycogen depletionrepletion protocol, using IV-enriched cornstarch. To deplete glycogenin the cycling musculatureand in the liver as much as possible,the subjects exercised until fatigued on the 1st day after fasting for the previous 12 h. They first cycled for 30 min at 80-85% VO, max and then for 2-3 h at 65-70% VO, m8Xuntil blood glucoseconcentration declined below 3 mM and they began to experience leg fatigue. At this point they performed 8-12 bouts of exerciseat the highest intensity they could maintain for 2 min (70-90% \I~o, ,,,) followed by 2 min of recovery

0 1992 the American

Physiological

Society

Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on December 24, 2018.

SUBSTRATE DAY

OXIDATION

1

Exercise

E65

IN EXERCISE Intervals

(O?‘Oof V02 max)

k

85% -------------------I

70% -I

I Start 13CCarbohydrate Repletion

Time (Minutes)

N=6

Intervals

DAY 2

I-

Exercise [6,6 -d2Glucose]

+

N=6

I

4 I

70%

85% ----------m-v------I

I

I

Fig. 1. Schematic representation of study design. VO, max, maximal 0, uptake. n, no. of subjects.

Start ’ 3 C-Carbohydrate Repletion I

1 N=4

Breath Measurements Time (Minutes)

-, o 17/

I

i,

N=6

;o

1

I

’

I

N=2

I

II

II

*

DAY 3 Exercise Breath Measurements Time (Minutes) b

exerciseat 50% V02 mBx.This protocol resultsin near-complete depletion of glycogen in the vastus lateralis muscle(5). Immediately after exercise,cornstarch (Polycose; RossLaboratories, Columbus,OH) wasingestedin equalamounts20,18,16,14,12, 10, and 8 h before exercise. The total amount consumedwas 8.6 g/kg. Cornstarch hasa high natural abundanceof 13C,dueto isotopic selectionwithin corn (13). To further increasethe 13C enrichment of cornstarch, 1 g of [UJ3C6]glucose(99%enriched; Isotech, Miamisburg, OH) was added to each 700 g of glucose polymers. In addition to enriched cornstarch, the subjectsconsumed a carbohydrate-free diet containing at least 800 kcal fat and 500 kcal proteins in four equal aliquots, which were ingested20, 17, 14, and 11 h before exercise. On the day after glycogendepletion-repletion, substrate oxidation was calculated by both indirect calorimetry and by the breath ratio method in the postabsorptive state at rest and during exercise (30 min cycling at 80-85% vo2 max;Fig. 1). To validate the assumptionthat the enrichment of the muscle glycogen, which was oxidized during exercise, was derived from the cornstarch, two subjectsperformed the glycogen depletion-repletion protocol on two consecutive days, and the breath ratio method was compared with indirect calorimetry on the second and third consecutive days during 30 min of cycling exercise at 80-85% VO, max(Fig. 1). In the four other subjects,the rate of glucosetissue uptake was quantified during exerciseby infusion of [6,6-2H,lglucose to assessthe relative contribution of plasma glucose to total carbohydrate oxidation (Fig. 1). Indirect

Calorimetry

Indirect calorimetry was performed at rest during at least 15 min and during the 30 min of exercise. The resting values were obtained after the subjectshad been laying on a bed for at least 1 h. The subjectsbreathed through a Daniel’s valve while inspired volume of air was measuredusing a dry-gas meter (Parkinson-Cowan CD-4). Expired gaseswere sampledfrom a mixing chamberand analyzed for 0, (Applied Electrochemistry S3A) and CO2 (Ametek CD-3A). Analog outputs from the instruments were directed to a laboratory computer for calculation of VO, and VCO,. The gas analyzers were calibrated with

gasesof known concentration. The dry-gas meter wascalibrated against a Tissot spirometer. Collection

of Expired

Air

Expired air was collected in 3-liter anesthesiabagsat rest (3 in total, with an interval of 5 min) and at l- 10,12,14,16,18, 20, 23, 26, and 30 min after the start of exercise. From these bags, samplesof expired air were collected in lo-ml evacuated tubes for determination of the absoluteratio of 13C/12C. Isotope Infusion

In the four subjects, in whom glucoseturnover was determined, Teflon catheters were placed percutaneously in an antecubital vein, and a sampling catheter was inserted in the dorsal hand vein of the contralateral side. The heated hand technique was used to obtain arterialized blood samples(11). The subjectswere restedfor 1 h after catheter placement.Then, after a blood samplewas drawn to determine background enrichment, a primed constant infusion of [6,6-2H2]glucose(99% enriched; Merck, Rahway, NJ) was started at the rate of 0.22 pmol kg-l min-l (prime 17.6 pmol/kg) using a calibrated syringe pump (Harvard Apparatus, South Natick, MA). The exact infusion rate in each experiment was determinedby measuring the glucoseconcentration in the infusates. The subjectsrested for the first 2 h of the [6,6-2H2]glucoseinfusion. When exercise started, the rate of isotope administration wasdoubledto minimize changesin isotope enrichment, resulting from the stimulation of glucoseproduction (20). l

l

Blood Sampling

Blood samplesfor the analysis of 13Cenrichment of plasma glucose,protein, and free fatty acids were obtained before the glycogen depletion-repletion protocol and before and after 30 min of exerciseon the day of the actual measurements.It was assumedthat these enrichmentsrepresentedthe corresponding enrichmentsof glucose,protein, and fat in the remainderof the

Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on December 24, 2018.

E66

SUBSTRATE

OXIDATION

body. After 30 min of exercise, a blood sample was obtained to assess the absolute 13C/12C ratio of plasma glucose. When [6,6-2H2]glucose was infused, blood was taken before starting the isotope infusion and 105, 110, 115, and 120 min after the start of the [6,6-2H2]glucose infusion, to measure resting kinetics, and at 5, 10, 15,20, 25, 30 min of exercise. All samples were collected in lo-ml vacutainers containing lithium heparin and were placed on ice. Plasma was separated by centrifugation at 4°C and frozen until further processing. Analysis Plasma glucose concentration was measured on a glucose analyzer (Beckman Instruments) by use of the glucose oxidase method. Plasma lactate concentration was measured enzymatically on a lactate analyzer (Yellow Springs Instruments, Yellow Springs, OH). The enrichment of [6,6-2H2]glucose in plasma was determined as previously described (17). Briefly, plasma was deproteinated with barium hydroxide and zinc sulfate solutions. Glucose was extracted by mixed-bed anion-cation exchange chromatography (AG-l-X8 and AG 5OW-X8; Bio-Rad Laboratories, Richmond, CA) and reacted with acetic anhydride and pyridine to form the penta-acetate derivative. Gas chromatography-mass spectrometry (GC-MS; model 5985B; Hewlett-Packard, Fullerton, CA) used electronic impact ionization, selectively monitoring ions at mass-to-charge ratio (m/e) 202, 201, and 200. Correction was made for the contribution of singly labeled molecules (m/e 201) to the apparent enrichment at m/e 202. For the determination of the 13C/12C ratio, glucose was extracted from plasma by barium hydroxide and zinc sulfate and by anion-cation exchange chromatography (see above). After drying the sample in a Speed Vat concentrator (Speed Vat, Farmingdale, NY), the sample was resuspended in 0.02 N H,SO, and was applied to a high-performance liquid chromatography system with two Bio-Rad Aminex HPX-87P columns (Bio-Rad, Richmond, CA) in series. The mobile phase was 0.02 N H,SO, with a flow rate of 0.6 ml/min (3). For extraction of proteins, the barium hydroxide-zinc sulfate precipitate of the plasma was washed three times with 5 ml deiodinized water to extract all water-soluble substrates. Subsequently, protein was extracted with 5 ml potassium hydroxide (0.05 N), of which 200 ~1 were used for combustion. Plasma free fatty acids and triglycerides were extracted from plasma and isolated by thin-layer chromatography (17). The isolated samples were placed in quartz tubes, containing cupric oxide, which were then sealed under high vacuum and placed in a combustion furnace at 590°C. The resulting CO2 gas was purified on a vacuum prep line and then analyzed by a 6-in. 60 triple-collector isotope ratio mass spectrometer (IRMS) (Nuclide, State College, PA) to determine the absolute 13C/12C ratio. The absolute 13C/12C ratio in breath was determined on an IRMS (SIRA II; VG Isotech, Cheshire, UK). Both IRMS instruments were calibrated using a standard with a known enrichment vs. Pee Dee Belamite-limestone obtained from the International Atomic Energy Commission, Vienna, Austria. The coefficient of variation was