DETERMINING EXOGENOUS GLUCOSE OXIDATION DURING MODERATE EXERCISE by ANNA ELENA BOZAC B.Sc, University of Ottawa, 1981 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF PHYSICAL EDUCATION in THE FACULTY OF GRADUATE STUDIES Department of Physical Education We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 1990 © Anna Elena Bozac , 1990 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Physical Education The University of British Columbia Vancouver, Canada Date March 29, 1990. DE-6 (2/88) ii ABSTRACT The purpose of this study was to determine the quantity of a glucose drink oxidized during cycle ergometer exercise at 60% V02max for 75 minutes. A second purpose was to determine if the glucose drink improved sprint time to exhaustion at 90% V02max after 75 minutes of exercise. Six trained male cyclists (V02imax > 60 ml-kg"1-min"1) exercised on three occasions during which they ingested either water ad lib (W), 13C-cornsyrup (100 g, 2.02 M) + water ad lib (CS), or NaH 1 2 C03/NaH 1 3 C03 mixture (5 mg-kg"1, 1% 13c-enriched) + water ad lib (B). Treatments B and CS were ingested after 5 minutes of cycling at 60% V02max. During exercise, there was no difference between treatments in plasma lactate response, changes in plasma volume, sprint time to exhaustion, or in respiratory exchange ratio (RER), VO2, or VCO2. RER showed a significant decline (p< .01) from 5 minutes (1.00±0.05, X+SD) to 75 minutes (0.96±0.05), and VO2 showed a significant positive shift (p< .01) from 3.15(±0.29) to 3.52(±0.45) l-min"1. A transient rise in plasma glucose was observed with CS. Changes from rest in 1 3 C / 1 2 C ratio (9 1 3C) showed a significant increase (p< .01) following CS. Peak glucose oxidation rate was 7.26 g-15 min"1 which occurred after 75 minutes. Total dose of exogenous 1 3C-glucose recovered as 13C02 (above baseline) was 22%. These observations suggest that (1) during moderate exercise of 75 minutes duration, oxidation of exogenous glucose occurs within 15 minutes but contributes marginally to total carbohydrate utilization as RER continued to fall with or without CS, and (2) sprint time to exhaustion after 75 minutes of cycling is not improved with glucose ingestion. TABLE OF CONTENTS iii Abstract ii Table of Contents iii List of Tables iv List of Figures v Acknowledgement vi CHAPTER 1. STATEMENT OF THE PROBLEM 1 CHAPTER 2. REVIEW OF LITERATURE 10 CHAPTER 3. METHODS AND PROCEDURES 45 CHAPTER 4. RESULTS 62 CHAPTER 5. DISCUSSION 75 CHAPTER 6. SUMMARY AND CONCLUSIONS 84 REFERENCES 87 APPENDICES Appendix A. Instructions to Subjects 95 Appendix B. Laboratory Procedures 97 Appendix C. Recovery of Breath 1 3C02 Using NaH 1 3 C03 100 Appendix D. Raw Data 102 iv LIST OF TABLES TABLE I. Physical characteristics, maximal O2 uptake and workload of the subjects 46 TABLE II. Order of conditions for each subject 56 TABLE III. Treatment composition, osmolality and amount of carbohydrate delivered 58 TABLE IV. Sprint time responses to experimental conditions 73 TABLE V. Exogenous glucose oxidation during exercise: results from published human studies 76 TABLE VI. Fractional recovery of 1 3C or 1 4 C in breath from ingested bicarbonate: results from published human studies 81 V LIST OF FIGURES Figure 1. C3 and C4 photosynthetic pathways 29 Figure 2. Schematic diagram of an isotope ratio mass spectrometer 32 Figure 3. Diagrammatic presentation of the experimental protocol 51 Figure 4. The statistical design of the experimental protocol 59 Figure 5A. Changes in plasma glucose concentration 63 Figure 5B. Changes in plasma lactate concentration 63 i Figure 6. Changes in plasma volume 64 Figure 7A. Changes in RER 66 Figure 7B. Changes in O2 consumption 66 Figure 7C. Changes in CO2 production 66 Figure 8. Changes in 3 1 3 C 68 Figure 9A. Changes in 1 3 c / l 2 C from rest 69 Figure 9B. Changes in 1 3 C / 1 2 C from rest w/o JG 69 Figure 10. Best-fitted line for the two conditions 71 Figure 11 A. Exogenous glucose oxidation 72 Figure 11B. Total exogenous glucose recovered 72 Acknowledgement I would like to express my thanks and appreciation to the subjects for their participation and cooperation in this study. I am indebted to my thesis committee - Drs. Angelo Belcastro, Peter Jones, and especially Don McKenzie, for their guidance and patience. A special "thank you" to my husband for his continual support and understanding throughout the study, to Maria Koskolou for her technical assistance, and to Han Joo Eom for his statistical advice. 1 CHAPTER 1 STATEMENT OF THE PROBLEM Introduction The common practice of ingesting carbohydrate (CHO) solutions during prolonged exercise is based on its supposed benefits of preventing dehydration, reduction of blood glucose levels, and/or depletion of muscle glycogen stores, all of which can lead to the development of fatigue (Alborg et al.,1976; Bjorkman et al.,1984; Costill et al.,1971; Coyle et al.,1983). Avoidance of hypoglycemia has been proposed as another important reason for ingesting carbohydrates during endurance athletic events. However, there appears to be a tight control of blood glucose during exercise (Felig and Wahren, 1975); whether hypoglycemia actually occurs and is indeed a cause of fatigue remains controversial (Felig et al., 1982). The overall importance of exercise as a stimulus to CHO utilization is shown by the observation that total body turnover of glucose is increased 2, 3 and 5 times, respectively, during mild, moderate and severe leg exercise (Wahren et al.,1971). In exercise of about 75 minutes duration, both muscle glycogen and blood-borne glucose predominate as energy-yielding substrates (Felig and Wahren, 1975). Consequently, many investigators have attempted to either increase and/or "spare" existing muscle glycogen stores in an effort to postpone fatigue. This forms the basis of two commonly used dietary practices: 1) CHO loading, and 2) CHO supplementation before and/or during exercise. Varying the composistion of a diet affects the relative availability of lipids and carbohydrates during subsequent exercise (Costill, 1988). This makes CHO loading for endurance events an effective means of increasing muscle glycogen stores and benefits performance. 2 In contrast, the administration of CHO supplements during exercise have not consistently shown a sparing of muscle glycogen stores (Coyle et al., 1986; Hargreaves and Briggs, 1988; Flynn et al., 1987). The benefit of this practice may be its ability to supplement the energy supply of blood glucose. Blood glucose can also contribute to the metabolic substrate pool during muscular activity. By 40 minutes of exercise, the fraction of energy derived from blood glucose increases and may account for 75-90% of the estimated CHO oxidation (Wahren et al., 1971). Blood glucose homeostasis during exercise is accomplished primarily by increased hepatic glucose production. However as exercise extends beyond about 40 minutes, splanchnic glucose output appears to plateau despite increases in glucose utilization (Alborg et al., 1974). Beyond 90 minutes of exercise, glucose production fails to keep pace with glucose utilization and blood glucose concentration slightly declines (Alborg et al., 1974). The availability of exogenous carbohydrates at this stage are believed to maintain blood glucose levels and thus supply glucose requirements. To determine the oxidation rate of a CHO (or glucose) load given orally during or before exercise, previous techniques involved either invasive procedures such as catherization (Alborg and Felig, 1977, 1976), or the administration of radioactive isotopes (Van Handel et al.,1980). Both approaches have limitations for technical and ethical reasons. Although oxidative activity is often implicated from increases in the respiratory exchange ratio (RER), an increase in RER may be due to factors other than increased CHO oxidation. Hyperventilation, decreasing pH and an increasing blood lactic acid concentration release body CO2 which affects RER values. Similarly, blood glucose concentration is frequently used to indicate current glucose status particularly following glucose ingestion. Several studies suggest 3 blood glucose response is not correlated with the glucose load administered (Mosora et al., 1981; Pirnay et al., 1982; Massicotte et al., 1986). Over the past 15 years, the use of stable isotope tracers in metabolic studies has increased. This technique allows for a quantitative measure of substrate oxidation using a noninvasive procedure. Statement of the Problem The purpose of this study is to determine the oxidation rate of an exogenous 13C-labelled glucose beverage administered to trained cyclists during a 75 minute cycle test at 60% V02max. Of particular interest is 1) how much of the exogenous glucose is utilized? 2) what is the time course of exogenous glucose metabolism?, and 3) whether the changes over time in blood glucose concentration and respiratory exchange ratio (RER) parallel those of 1 3 C abundance in expired air? A second purpose is to determine whether sprint time to exhaustion at 90% V02max following the cycle test is improved with glucose ingestion. Subproblems The subproblems are: 1) To determine the effects of moderate exercise on baseline 1 3 C abundance of expired air. 4 2) To establish values for the recovery of labelled 13(X)2 from a NaH 13C03 bolus administered to trained cyclists during a 75 minutes cycle test at 60% V02max. Hypotheses The hypotheses are: 1) 13c-glucose given orally during a 75 minute cycle test at 60% V02max increases the rate of carbohydrate utilization. This is based on a review of the literature where some researchers report that carbohydrate ingested during exercise is rapidly oxidized and provides a blood-borne glucose fuel source thereby preventing depletion of glycogen stores. Others report no effect of glucose ingestion on the rate of carbohydrate utilization. The main procedural difference between studies is exercise duration and intensity. Theory suggests an increased rate of carbohydrate oxidation based on the significant substrate shifts observed during prolonged aerobic exercise. 2) After ingestion of a glucose load during moderate exercise, blood glucose concentration is not a valid indicator of glucose oxidation. This is hypothesized from data which suggest that while a glucose load provides significantly greater amounts of glucose to the blood than does water, the percent contribution to oxidative metabolism is not proportional to measured values of blood glucose concentration (Costill et al., 1980). 3) The time course for the changes in the enrichment of exhaled CO2 with 1 3 C 0 2 is delayed relative to that of RER during moderate exercise. 5 Whereas shifts in VO2 and VCO2 adjust within 4-6 min after the onset of exercise, additional time is required for redistribution of 13C02 throughout the body bicarbonate pools which delay its appearance in expired CO2, (Barstow et al., 1989). 4) Exercise at 60% V02imax will increase baseline 1 3 C enrichment of expired CO2. This is hypothesized since during exercise carbohydrate is oxidized at a faster rate, relative to fat, than at rest, and the naturally occuring enrichment of carbon in carbohydrate is higher than in fat. 5) The amount of exogenous 1 3C-glucose oxidized is underestimated due to a low turnover rate of the bicarbonate pools. This is based on the delay observed in the appearance of 1 3 C abundance in expired air following administration of a l 3 C label (Barstow et al., 1989). Definition of Terms The following definitions reflect the common usage of the key words used in the literature on this topic. 1 3C-glucose - A non-radioactive or stable carbon isotope frequently used as a "label" in breath tests, and monitored by ^ C / ^ C mass ratio measurements. Available artificially from commercial preparations, or naturally from plants that utilize the C4 photosynthetic pathway (Smith and Epstein, 1971). 6 3 1 3 C - Refers to the 1 3 C abundance of an isolated CO2 sample calculated from mass spectrometer measurements comparing the 1 3 C / 1 2 C ratios in a sample to that in a standard under identical measurement conditions. Results are expressed in relative delta (9) per mil (%>) according to the formula: a13C(%„) = ( 1 3 C / 1 2 C ) sample -1 ( 1 3 C / 1 2 C ) standard 1,000 (Craig, 1957) Atom percent excess (APE) - An alternate unit used to express the enrichment of 1 3 C in a CO2 sample. To measure small differences in the isotopic composition of gases, such as CO2 , APE is less convenient than 3 since it is a much larger unit. Relative to the PDB standard used for carbon analysis, 1.123 x 10- 3 APE = 1 3. PDB standard - An obselete belemnite limestone standard found in the PeeDee range of North Carolina. Most natural levels contain less 1 3 C than the standard PDB, so values are usually negative with respect to this standard. Baseline - Refers to the abundance of an isotope in any situation where that isotope is not being infused or ingested concurrently. It provides "background level(s)" (Wolfe, 1984, p. 59). Isotope Enrichment - A measurement generated following the administration of an isotope-labelled substrate. Isotope Abundance - Refers to the quantity of isotope present in a sample. 7 Isotope dilution - The basic principle of tracer methodology where the rate of appearance (Ra) of unlabelled molecules is determined by the dilution of the infused labelled molecules (Wolfe, 1984, p. 113). Hypoglycemia - A condition characterized by blood glucose levels less than 50 mg-dl"1 (2.8 mmol-l*1) for whole blood analysis (Lagua et al.,1974). Osmolality - Refers to the concentration of a solution, ie. the number of particles in solution. Osmolality (mOsm-kg H2O"1) and osmolarity (mOsm-l"1) are often used interchangeably; however, the osmolality of a solution is usually higher than is the osmolarity (Selkurt, 1975, pp. 9-14). Respiratory Exchange Ratio (RER^ - The ratio of the volume of carbon dioxide expired per minute (VCO2) to the volume of oxygen consumed during the same time interval (VO2) At the cellular level, this ratio is termed the respiratory quotient and it reflects the proportion of carbohydrate to lipid oxidation (McArdle, Katch and Katch, 1986). VO?max - Maximal oxygen consumption defined as the point at which oxygen consumption plateaus and shows no further increase (or increases only slightly) with an additional workload (McArdle, Katch and Katch, 1986). Delimitations 1) The study is delimited to highly trained endurance cyclists (V02max >60 ml-kg- 1-min~1) all of whom are registered racers with the Bicycling Association of British Columbia (B.A.B.C.). 8 2) The study is delimited to a 2.02 molar concentration of glucose beverage delivered 5 minutes after the onset of exercise at 60% V02max. 3) The study is delimited to a 75 minute period of exercise intensity at a workload corresponding to each subject's 60% VC^max, followed by a sprint to exhaustion at a workload corresponding to each subject's 90% V02imax. 4) The study is delimited to a five week data collection period for each subject. All data will be collected in a laboratory. Assumptions The following assumptions are made: 1) Subjects are selected from a normally distributed population of competitive cyclists. 2) Each subject's response is independent of the responses of all other subjects.and of previous trials. Due to the high caliber of cyclist employed, a training effect is not expected. 3) The variances of the responses by the selected population under all treatments are equal. Since highly trained competitive cyclists are used, their optimal cycling style would be well established and would show minimal intraindividual variability, ie. they are able to cycle in a similar fashion using the same cadence and pace, and under similar conditions of diet and activity for 72 hours prior to each trial. 9 4) Cycling for 75 minutes at 60% V02max describes a steady state condition. This assumption is monitored by measurement of VO2 throughout the exercise bout. 5) Protein contribution is assumed to be less than 5% of total energy as has been previously reported (Butterfield and Calloway, 1984; Millward et al., 1982). Significance At present, there seems to be some controversy regarding the utilization of exogenous glucose when ingested during a moderate exercise bout for 75 minutes. Whereas some investigators state that glucose ingested orally during exercise is of little importance for muscle metabolism and remains in an unoxidized glucose pool, other investigators report almost 100% utilization. A second debatable issue is whether glucose ingestion provides any advantage over water in a performance sprint ride. Due to improved analytical techniques, it is possible to quantitate glucose oxidation using glucose naturally labelled with the stable isotope 1 3 C . This investigation was thus designed to determine the oxidation rate of an exogenous glucose drink ingested by trained cyclists at the onset of a 75 minute moderate exercise bout. It was also designed to determine if glucose ingestion improves sprint time to exhaustion following the 75 minutes of exercise. The significance of this study is to evaluate whether the practice of ingesting a glucose drink during this exercise protocol is beneficial. 10 CHAPTER 2 REVIEW OF LITERATURE Introduction In view of the importance of carbohydrate (CHO) metabolism during exercise, considerable interest has been focused on the potential ergogenic benefits of exogenous CHO ingestion and subsequent metabolic effects. It is known that the level of body stores as well as the amount of CHO ingested during the preceeding days modify the energy sources during an exercise bout (Bergstrom et al., 1967). The ingestion of CHO during exercise has been shown to illicit significant changes in hormonal and metabolic response relative to those observed with exercise alone. Of particular interest is whether they can better maintain and/or improve performance. In earlier studies, respiratory gas exchange of oxygen (O2) and carbon dioxide (CO2) was used to demonstrate reliance on CHO as an important muscle fuel. The mechanisms underlying the shift towards CHO oxidation were later examined using more invasive measurements (Alborg and Felig, 1977; Hargreaves and Briggs, 1988) and with isotopes (Van Handel et al., 1980; Pirnay et al., 1977a, 1981). The following review will discuss the role of both endogenous and exogenous CHO during muscular activity. In addition, specific attention will directed towards stable isotope 1 3 c methodology used to determine glucose oxidation. 11 Carbohydrate Metabolism During Exercise Carbohydrate Stores of the Body During exercise, normal man expends 10 to 20 times more energy that at rest. Consequently, the blood glucose pool, which amounts to about 20 g, corresponding to 80 kcal (330 kJ), is not large enough to support prolonged work. Additional CHO is stored in the form of glycogen in the liver (80-90 g) and in skeletal muscle (300-400 g). Thus the total tissue stores of CHO represent less than 2050 kcal (9000 kJ), a small energy reserve compared with the potential stored as triglyceride in adipose tissue (approximately 15 kg or 135,000 kcal; 560,000 kJ). Although liver and muscle glycogen reserves are small, they perform important functions to maintain fuel homeostasis. Liver glycogen is the primary source of blood glucose that is rapidly mobilized in response to exercise. Muscle on the other hand cannot directly contribute to blood glucose since it lacks the enzyme glucose-6-phosphatase required for glucose entry into metabolic pathways. Its contribution to glucose homeostasis is limited to the Cori cycle. Muscle glycogen is of use in meeting the energy needs of the muscle fibers in which it is contained. Resting Muscle Free fatty acids (FFA) are the predominant fuel in the resting postabsorptive state as indicated by a respiratory quotient close to 0.7 (Andres et al., 1956). In the resting state, uptake of glucose accounts for less than 10% of total O2 consumption by muscle. The main site of glucose consumption in this state is the brain. 12 Exercising Muscle In addition to FFA, muscle glycogen and blood-borne glucose largely contribute to the energy demands of exercise. The relative contribution of each depends somewhat on exercise duration and intensity. Several good reviews describe the relationship between these fuel sources during exercise (Felig and Wahren, 1975; Wahren, 1979; Costill, 1988). Basically, during the early stages of exercise, muscle glycogen is the major fuel consumed (Wahren et al., 1971; Costill, 1988). As exercise continues, muscle blood flow increases thereby allowing blood-borne substrates to contribute as energy sources. Wahren et al. (1971) showed that during upright exercise for 40 minutes on a bicycle ergometer, glucose uptake by leg muscles increased 7- to 20-fold above basal levels, depending on exercise intensity. After 40 minutes of exercise, this accounted for 28-37% of the total oxidative metabolism determined from the arterial-femoral venous (A-FV) oxygen difference, and as much as 75-90% of CHO metabolism estimated from ventilatory respiratory quotient. The same study reported an increase in splanchnic glucose output, above resting state, of 2-, 3- and 5-fold during 40 minutes of exercise at 65, 130 and 195 Watts, respectively. Even during more prolonged exercise of 3 to 4 hours, the rate of glucose utilization accounts for 35-40% of total fuel utilization (Alborg et al., 1974). Thus blood glucose is a quantitatively important fuel for contracting muscle during short-term as well as prolonged exercise. Glucose Production and Blood Glucose Regulation The stimulation of glucose utilization that occurs with exercise depletes the blood glucose pool. To maintain homeostasis, an increase in 13 glucose production occurs via hepatic glycogenolysis and gluconeogenesis. During short term exercise of 40 minutes, increased glucose production results almost entirely from hepatic glycogenolysis as splanchnic uptake of the gluconeogenic precursors alanine, lactate, pyruvate and glycerol remains unchanged from resting state (Wahren et al., 1971). Increasing exercise intensity from 65 to 195 Watts, increases the demand on hepatic glycogenolysis by about 10%. The amount of hepatic glucose mobilized represents 20-25% of the total hepatic glycogen store. Beyond 40 minutes of exercise, an increased reliance on hepatic gluconeogenesis is observed. During prolonged mild exercise at 30% V02max for 4 hours, the relative contribution from gluconeogenesis to overall hepatic glucose output, estimated from splanchnic substrate balances, increased from 25% in the resting state to 45% (Alborg et al., 1974). Coinciding with this is a further mobilization of hepatic glycogen resulting in a 75% depletion. Despite the dramatic drop in hepatic glycogen observed during exercise, blood glucose levels do not fall until hepatic glycogen stores are very low. This tight control of blood glucose appears secondary to an increased tissue sensitivity to circulating insulin (King et al., 1988). A similar response is apparent following a glucose load where a smaller rise in blood glucose is observed under exercising versus resting conditions (Costill, 1988). Regulation of Glycogen Mobilization and Glucose Uptake by Skeletal Muscle The rate of glycogenolysis in skeletal muscle is a function of the relative intensity of the exercise (Saltin and Karlsson, 1971). The rate of CHO utilization and degradation of muscle glycogen at the same absolute or relative oxygen uptake is influenced by the intensity of exercise, exercise duration, the 14 concentration of glycogen in the muscle, and the state of physical training (Gollnick, 1986). One possible mechanism for the activation of glycogenolysis during exercise involves the elevation of intracellular concentrations of cyclic AMP in response to the release of epinephrine. Epinephrine can initiate a cascade of reactions that activate the enzyme phosphorylase (phos) b kinase. This enzyme in turn converts phos b, the enzyme form that is most abundant in quiescent muscle and is normally inactive in the absence of AMP, to phos a , a form of the enzyme that is active in the absence of AMP. A flux generator for muscle glycogen breakdown is at the level of glycogen phosphorylase. Another mechanism that may be involved in producing the rapid response in glycogenolysis at the onset of exercise is an increase in the free calcium concentration within the cytosol. The rate of muscle glycogenolysis is somewhat dependent on pre-exercise glycogen concentration. At identical tension development, glycogen-supercompensated muscles break down more glycogen than control muscles (Richter and Galbo, 1986). Glucose uptake into muscle tissue can be induced by muscle contraction perse (Richter, 1987). This is at least in part due to a contraction-induced increase in muscle membrane permeability (Ploug et al., 1984). The mechanism behind the contraction-induced increase in membrane permeability to glucose is still unclear. The release of calcium during muscle contraction may initiate this event. It has been suggested that blood glucose concentration stimulates glucose uptake into muscle at a threshold level =8-10 mmol-M via a seperate glucose carrier. Others have suggested that the energy state of muscle, as reflected by ATP/ADP ratios and phosphocreatine levels, regulates muscle glucose 15 uptake. According to this theory glucose uptake is negatively correlated with the energy state of muscle. Insulin is not necessary for muscle contractions to increase muscle glucose uptake. Findings to support this include the translocation of glucose transporters from an intracellular storage site to plasma membrane in the kpresence of insulin and an insulin-induced decrease in the Km for glucose transport (Richter, 1987). Although not necessary for glucose uptake and transport into muscle, it has been shown in red muscle that insulin and muscle contractions have an additive effect on glucose transport. Hormonal Effects on Substrate Utilization During Exercise A primary consideration during the stress of exercise is the maintenance of nearly "normal resting" levels of blood and cellular metabolites. In particular, the maintenance of blood glucose levels at about 4 mmol-M is critical. The coordinated physiological response to maintain blood glucose homeostasis during exercise is governed by the autonomic nervous system (ANS) and the endocrine (hormonal) system. At the onset of exercise, impulses from motor centers in the brain as well as from working muscles, elicit a workload-dependent increase in sympatho-adrenal activity and in release of some pituitary hormones. These changes control the changes in secretion of endocrine cells. The state of the organism prior to exercise is important for the magnitude of the hormonal response (Galbo and Kjaer, 1987). Exercise capacity influences the response, which depends on the relative rather than absolute work intensity. During prolonged exercise, the hormonal changes may be gradually intensified due to feedback from metabolic error signals, among which is a decrease in glucose 16 availability, as well as from non-metabolic error signals sensed by pressure, volume and temperature receptors. At rest, circulating levels of glucose in blood and of insulin are adequate to inhibit hepatic glycogenolysis; such that normally, liver glycogen is not utilized as a fuel source. Similarly, low levels of circulating epinephrine inhibit muscle glycogenolysis. At rest, muscle is primarily dependent of fatty acid oxidation. As exercise begins, there is an increase in AMP, IMP and Pi secondary to ATP utilization. These modulators stimulate muscle phos a and b to increase muscle glycogen breakdown. A subsequent increase in glucose-6-phosphate serves as another stimulus of muscle phos a . After 40 minutes of endurance-type exercise, there is little change in the hormonal mileu from rest and both muscle glycogen and fatty acids are the major fuel sources. After 180 minutes of exercise, increased circulating levels of glucagon and epinephrine stimulate phos b in muscle and liver as well as phos a in liver. Also, low circulating levels of insulin stimulate phos b in liver. These changes increase hepatic production of glucose by glycogenolysis and result in an increase in blood glucose release from the liver. At this stage, blood-borne glucose and fatty acids predominate as substrates for oxidation. Glucose Ingestion in Connection With Exercise Some form of nutrition, usually a CHO source, is frequently consumed during prolonged and/or moderate to strenuous exercise bouts. It is of interest to examine whether extra glucose will modify muscle substrate availability and/or utilization. 17 Over the past two decades, researchers have attempted to address this issue; however, the results remain controversal. Differences in exercise protocol, training status of subjects, metabolic status of subjects, intensity of exercise and measurement of glucose oxidation are but a few of the factors that vary among studies. In those studies involving trained athletes (V02max 56-70 ml-kg-1 •min"1) exercising at intensities 60-70% V02max, of foremost interest is whether performance is improved with CHO ingestion, and whether any improvements observed are due to glycogen "sparing" and/or the prevention of hypoglycemia. In an early study by Ivy et al. (1979), a significant increase in work output over the final 30-40 minutes of a 2-hour cycling bout at 70% V02max was observed when the subjects were given CHO feedings at 15 minute intervals. A total glucose concentration of approximately 90 g was consumed over a 90 minute period During a control trial, blood glucose fell steadily throughout exercise reaching a mean value of 3.9 mmol-l"1 (50 mgdl"1) after 2 hours. CHO feedings on the other hand, increased blood glucose to 4.4 - 6.0 mmol-l"1 (80-110 img-dl"1), which remained elevated throughout the activity. The effects of the CHO feedings on general fatigue were inconsistent among the subjects. Coyle et al. (1983) showed that during cycle exercise at 75% V02max, the ingestion of a glucose polymer given as 1 g k g - 1 after 20 minutes, followed by 0.25 g-kg -1 after 60, 90 and 120 minutes increased time to fatigue over a control group. Fatigue was not found to be associated with hypoglycemia in either the treatment or control groups, although Coyle suggested that it may be a cause for fatigue in certain susceptible individuals. Muscle glycogen utilization was not examined. Instead, they suggested that CHO administration during exercise maintains blood glucose and insulin levels higher which may result in 18 increased utilization of blood glucose with a proportionate slowing of muscle glycogen depletion. In a subsequent study by Coyle et al. (1986), administration of 2 gkg~1 CHO after the first 20 minutes, followed by 0.5 g-kg"1 every 20 minutes thereafter did not lower glycogen utilization in the vastus lateralis muscle after 105 minutes of exercise at 70% V02max. They reached similar conclusions using another group of subjects who exercised until fatigued at the same work intensity. Once again, they attributed the increased time to fatigue with CHO feeding to a high rate of blood glucose oxidation which was calculated from the respiratory excahnge ratio (RER). Hargreaves and Briggs (1988) supported Coyle's conclusions in their study. They measured muscle glycogen concentration in the vastus lateralis muscle during a 2 hour cycle exercise with and without a glucose polymer given at a dose of 60 g-h~1. They observed no effect of CHO ingestion on the rate of muscle glycogen utilization. In contrast, Bjorkman et al. (1984) reported a significantly lower rate of glycogen breakdown (p< .05) in 7 out of 8 subjects during prolonged exercise when glucose was ingested intermittently at a rate of 17.5 g every 20 minutes. They exercised 8 subjects until exhaustion at 68±24% V02imax (X±SD). Mean duration of exercise was 137 minutes following glucose ingestion which was significantly longer (p< .01) compared to 116 minutes with water. An important observation was the varied response to glycogen depletion between subjects. When all 8 subjects were included, the rate of glycogen breakdown between treatments ceased to be significant. Such individual variations on glycogen utilization, when measured from muscle samples, has been previously reported (Costill et al., 1977). 19 Flynn's approach to performance appraisal involved measurement of work output during a two hour self-paced cycle (Flynn et al., 1987). In contrast to the previous studies, the subjects had CHO-loaded. Varying compositions of glucose polymer feedings providing 22.5 - 45 g CHO-hr 1 failed to increase total work output, or work performed in the last 30 minutes or final 5 minutes of exercise. Muscle glycogen utilization was similar in the water and in all CHO trials, and at no time during any of the trials did blood glucose approach hypoglycemic levels. It is possible that the demand on exogenous sources of CHO was adequately met by the available muscle and liver glycogen reserves. In contrast Erickson et al. (1987) observed glycogen sparing following glucose ingestion. They used 5 competitive cyclists (V02max = 65.2±2.7 ml-kg-1 •min-1) to determine the effects of 1 g-kg"1 glucose during exercise on blood glucose and FFA concentration, and on muscle glycogen utilization and RER. Exercise consisted of 90 min of cycling at 65-70% V02imax. When compared to a control trial using artificially sweetened and flavored water, exercised cyclists demonstrated a reduction in muscle glycogen utilization, an increase in blood glucose concentration, and a decrease in FFA concentration following the exercise period. None of the parameters were measured during exercise. No significant differences between the two trials was observed for RER. The authors contribute decreased muscle glycogen use to the glucose treatment. In conclusion, these studies have suggested that ingestion of CHO during moderate exercise at or above 60% V02max, may delay time to fatigue in non-CHO-loaded athletes. This effect appears mediated by maintaining blood glucose levels and utilization, rather than by slowing muscle glycogen use. Averaged hourly rates ranged from 42-125 g CHO-hr 1 . In CHO-loaded athletes, these effects remain questionable. 20 Determining Exogenous Glucose Oxidation The previous section reviewed studies in which an exogenous CHO source was administered during moderate to intense exercise. In neither of these studies was the rate of oxidation of the exogenous glucose assessed. Although frequently implicated from measurements of respiratory exchange ratio (RER), the reliability of this measurement is limited. The fuel oxidized during basal steady-state conditions can be determined by the "respiratory quotient" defined as the volume of expired CO2 divided by the volume of O2 consumed. During the nonsteady-state condition of exercise, this ratio is termed the "respiratory exchange ratio" (RER) and does not indicate the fuels being oxidized. During exercise, RER indicates immediate alterations of respiratory gas exchange. Just prior to the onset of exercise some people hyperventilate, anticipating the respiratory demands to follow. This results in a pumping out of CO2 stores and may temporarily increase RER. Further increases in RER may be seen during exercise as lactic acid begins to build up in active cells and diffuse into the extracellular spaces (body fluids). Lactic acid acting as a fixed acid forms carbonic acid, which in turn breaks down into water and carbon dioxide: Muscle Activity ' > Lactic acid + N a C 0 3 ^ Na lactate + H2CO3 production H2CO3^= i H20 + C02 expired The CO2 is then expired as compensation for metabolic acidosis thereby raising RER. A final point is that RER does not distinguish between the respective proportions of endogenous or exogenous carbohydrates involved. More accurate determinations require the use of radioisotopes or more invasive procedures using 21 multiple arterial catherizations. For both technical and ethical reasons, these techniques have limitations in human subjects. Two studies using 1 4C-glucose (Costill et al., 1973; Van Handel et al., 1980) did not observe a significant contribution of exogenous glucose to oxidative metabolism during either mild or moderate exercise. In one study (Costill et al., 1973), two subjects cycled at 60% V02max and five subjects ran at 65-72% V02max for 30 minutes prior to ingesting a 31.8 g glucose load, after which they exercised for an additional 60 minutes. At the end of exercise, 8.4% of the total 1 4 C dose had been recovered in expired air. In a later study (Van Handel et al., 1980), six competitive cyclists exercised on an ergometer for 3 hours at 50% V02max. After 2 hours they ingested a 10 or 40 g glucose load and continued to cycle for another 60 minutes. At the end of exercise, 8.9% and 6.4%, respectively, of the total dose had been recovered in expired air. In both studies, some 1 4 C 0 2 was present in expired air 5 minutes after feeding, though it was a negligible amount of the total dose. It has been convincingly argued (Pirnay et al., 1982; Krzentowski et al., 1984a) that at the time of glucose ingestion after 60 or 120 minutes of exercise, the rate of lipid oxidation is high and is coincident with high levels of circulating free fatty acids which have been found to decrease glucose uptake and oxidation by muscle (Costill et al., 1977; Ravussin et al., 1979). Also, the amounts of glucose administered were quite small. The ingestion of more concentrated glucose solutions at the beginning of exercise have been found to decrease FFA concentration and increase exogenous glucose oxidation by the working muscles (Alborg et al., 1976; Ravussin et al., 1979; Wahren, 1979). A number of studies over the past decade have measured exogenous CHO oxidation using "naturally labelled" 1 3C-glucose. An early report by Lacroix et al. (1973) introduced this novel approach. In six healthy male volunteers, the administration of glucose resulted in a marked, reproducible rise in the isotopic 22 ratio of 13C/12C in expired CO2 . Next, they quantitated the amount of glucose oxidized under normal resting conditions (Mosora et al., 1976). Following ingestion of 100 g maize glucose, 28.64±11.52 g (X±SD) were completely oxidized into CO2 in 7 hours. To determine if the amount oxidized was correlated with the increase in blood glucose concentration also observed and/or the magnitude of the insulin response, they manipulated the loading dose to 33, 66 or 100 g (Mosora et al., 1981). Generally, the changes in blood glucose response were not significantly different between the three doses. In contrast, the plasma insulin response and the amount of exogenous glucose converted to expired CO2 were proportional to the loading dose and significantly correlated (r=0.72, p< .001). Since total glucose oxidation, derived from indirect calorimetry, was similar under each experimental condition, they concluded a lesser contribution of endogenous glucose when the amount of exogenous glucose given is increased. They attributed this to a greater enrichment of the systemic glucose pool with high loads of exogenous 1 3 C -glucose which results from increased retention of splanchnic glucose second to increasing insulin responses (Felig and Wahren, 1971). Although they observed utilization of less than 1/3 of the 100 g glucose after 7 hours for subjects at rest, the results were very different for the same subjects during prolonged exercise at 50% V02imax (Pirnay et al., 1977b). In this instance 94.8±4.2 g (X±SD) of glucose was oxidized after 4 hours. The utilization of exogenous glucose was observed 15 minutes after ingestion and reached a maximal rate of 40 g-h"1 between the first and second hour, whereas in resting conditions (Mosora et al., 1976) the maximal rate was 6 g-h"1 and occurred between the third and fourth hour. In a series of systematic studies performed by the same group, several variants involving exogenous glucose ingestion and mild, prolonged exercise were investigated. These include the influence of repeated administration 23 of smaller quantities of oral glucose (Pirnay et al., 1981), time of glucose ingestion (Krzentowski et al., 1984a), physical training (Krzentowski et al., 1984b), osmolality of the glucose solution (Jandrain et al., 1989) and use of sucrose (Gerard et al., 1986) on exogenous CHO utilization during exercise. The results have suggested that ingested CHO quickly participates in cellular oxidation regardless of the experimental protocol, and that after 4 hours about 90% of the administered dose is oxidized. Similar conclusions were reported by Ravussin et al. (1979) who utilized 100 g of naturally labelled 1 3C-glucose in glycogen depleted and control subjects. Bicycle ergometer was performed for 2 hours at 40% V02max. Depleted subjects used 1 3C-glucose almost to the same extent as controls, 38±10 and 41 ±5 g (X±SD), respectively. The relative contribution of 13C-glucose to the total energy expenditure estimated from RER was 19% for glycogen depleted subjects and 25% for controls (p< .02). They attributed this to the inhibitory role of high plasma FFA levels, which were 2- to 3-times higher in the glycogen depleted group, on muscle glucose uptake and oxidation. More recently, Massicotte et al. (1986) quantified the amount of ingested 1 3C-labelled glucose and fructose oxidized during 3 hours of exercise on a cycle ergometer at 50+5% V02max. Tests were conducted 2 hours after a light breakfast of 400 kcal containing 50 g CHO. Seven percent glucose or fructose solutions were administered every 20 minutes at a rate of 46.5 g CHO-h" 1. Over the 3 hour period, 75% of the ingested glucose was oxidized compared with 56% for fructose (p< .05). As computed from RER and VO2, both solutions contributed to decrease endogenous CHO utilization. The higher fat utilization observed with fructose ingestion was entirely compensated by a lower oxidation of fructose; thus, endogenous CHO utilization was similar with both sugars. When a glucose polymer solution was ingested (Massicotte et al., 1989), no additional advantage 24 over a glucose solution was observed. Exogenous substrate utilization from 0-120 minutes for a 7% solution containing 100 g of glucose or glucose polymer, was 70±15 and 64±18 g (X±SD), respectively. The studies by Massicotte et al. differ from those perviously discussed in that subjects were not postabsorptive upon commencement of exercise. A one-day fast in humans (Dohm et al., 1986) and in rats (Dohm et al, 1983) has shown increased fat mobilization and utilization. During treadmill exercise at 70% V02max, fasting was associated with higher plasma concentrations of FFA and glycerol and lower RER values than was observed when subjects exercised in a fed state (Dohm et al., 1986). Costill et al. (1977) have demonstrated that increased availability of plasma FFA slows glycogen depletion in the gastrocnemius muscle during 30 minutes of treadmill exercise at 68% V02max. An overnight or postabsorptive fast does not cause this marked increase in FFA utilization during mild exercise (Knapik et al., 1988). Cycle ergometer exercise at 45% V02max for about 2 hours in a postabsorptive state showed a decline in blood glucose concentration to below resting values and an increase in FFA levels of around 0.5 mM. No significant changes in blood lactate, gycerol and alanine suggested that gluconeogenesis played a minor role in maintaining blood glucose concentration. RER dropped slightly from 0.88 at the onset of exercise to 0.86 after 2 hours. It is likely that during exercise of mild intensity, a postabsorptive or fed state minimally affects substrate use since lipid is the preferential fuel and is not supply-limited. Thus, results from the studies of Pirnay and Mosora et al. (1977-1989) and those of Massicotte et al. (1986, 1989) are comparable. At higher work intensities, metabolic status is of greater importance. Loy et al. (1986) looked at the effects of a 24 hour fast on cycling endurance time during intense exercise at 79-86% V02max. On a separate occasion, subjects 25 performed the same exercise 3 hours after a liquid meal of 355 kcal containing 50g CHO. Fasting had no effect on heart rate, RER or level of glycogen depletion. However, time to exhaustion was significantly decreased by fasting. Blood glucose levels dropped and blood lactate levels rose independent of metabolic state. Exercise caused an increase in FFA values with fasting values being significantly higher. The reduced time to fatigue after fasting was explained secondary to reduced liver glycogen content and subsequent inability to maintain blood glucose. That fasting did not spare muscle glycogen content as previously reported (Costill et al., 1977) is likely due to exercise duration which was 5-7 times longer. It appears that during more intense exercise, greater reliance on CHO for energy necessitates a fed state to increase its availability. When exercise intensity is increased, the contribution of exogenous glucose is altered. Using four volunteers, Pirnay et al. (1982) investigated the influence of four different work intensities (22, 39, 51 and 64% V02max) on the utilization of ingested glucose during a 105-minute treadmill exercise. Between 22 and 51% V02max, exogenous glucose oxidation was linearly correlated (r=0.81, p< .001) with workload. If the more intensive workload was included, exogenous glucose oxidation was poorly correlated with workload (r=0.51, p>.05). The cumulative value of exogenous glucose oxidized at 64% V02max was 44±12.4 g, (X+SD) which was not significantly different from 41.2±25.6 g measured at 51% V02max. The authors speculated a lesser contribution of exogenous glucose during intense exercise due to impaired gastric emptying and/or altered glucose uptake by working muscles. The former conclusion warrants further discussion. During intermittent exercise of 12-20 minutes at an intensity of less than 70% V02max, gastric emptying, based on the volume of fluid remaining in the stomach, was not significantly different between water and CHO-containing beverages (Costill and Saltin, 1974; Mitchell et al., 1988). Carbohydrates ranged 26 in amounts from 25-75 g, and in osmolalities from 278-1390 mOsm-kg H2O"1. In contrast, during more prolonged exercise at 70% Wmax or 250±34 Watts (X±SD), drink composition was a strong inhibitor of gastric emptying (Rehrer et al., 1989). These investigators tested 8 trained cyclists and 8 untrained individuals at two different work intensities of 50% Wmax and 70% Wmax. Differences in training or experience in drinking during exercise had no effect on gastric emptying or secretion. Increased exercise intensity resulted in increased gastric secretion (p<.05) but did not affect gastric emptying. Drink composition was observed to be a strong effector of gastric emptying. The slowest drink emptied was a hypertonic glucose drink, 1060 mOsm-kg H2O"1; however, a calorically richer maltodextrin-fructose mixture of 444 mOsm-kg H2O"1 also emptied significantly slower than either an isotonic maltodextrin-sucrose mixture of 296 mOsm-kg H2O"1 or an artificially sweetened placebo of 40 mOsm-kg H2O"1. Wheeler and Banwell (1986) compared intestinal water and electrolyte flux during the perfusion of two CHO-electolyte solutions to that of water in human subjects at rest. The intestinal water absorption rate for the plain water drink was not significantly different to a CHO-electrolyte drink of 240 mOsm-kg H2O"1 containing 50 g-M of a glucose polymer and 20 g-H of fructose. However, replacement of a portion of the glucose polymer with 16 g-l"1 of sucrose changed the osmolality to 260 mOsm-kg H2O"1 and slowed net water absorption when compared with plain water. There were no significant differences between groups in absorption rates of sodium, potassium or chloride. Results of this investigation may not be applicable under exercising conditions. The above results suggest that gastric emptying during continuous exercise at moderate intensities can be influenced by the CHO composition of the drink and/or its osmolality. It should be emphasized that mode of exercise , ie. running vs cycling, may play a major role in the volume of drink emptied (Neufer et al., 1986). It appears that the osmolality of a glucose solution ranging from 439-1204 mOsm-kg H2O"1 does not affect its metabolic availability during 4 hours of mild treadmill exercise (Jandrain et al., 1989). However, this has not be ascertained using a more hyperosmolar solution or at moderate exercise intensities. 28 13C-Tracer Methodology Sources of 1 3 C There are many types of labelled 13C-compounds prepared commercially by companies in both North America and Europe. However, artificial enrichment is not always necessary. Knowing that the 1 3 C / 1 2 C ratio in exhaled CO2 reflects the natural abundance of these two isotopes in the diet, and that plants following the C4 photosynthetic pathway are slightly richer in 1 3 C than the more commonly ingested C3 plants, naturally labelled substrates derived from C4 plants can be used as 13C-tracers in metabolic studies. The two photosynthetic pathways and food chains from which all human nutrients are derived are the Calvin-Benson route (Calvin, 1962) and the Hatch-Slack route (Hatch and Slack, 1970), commonly referred to as C3 and C4, respectively. In the C3 route, CO2 penetrates through the stomates, dissolves in cytosol and is used to carboxylate one molecule of ribulose-1,5 diphosphate (RuDP) to produce two molecules that have three carbon atoms, hence the name C3. This newly formed substance is 3-phosphoglycerate. Thereafter, photosynthesis proceeds according to the Calvin cycle (see Fig. 1A). Isotopic fractionation occurs at the initial carboxylation step in which 12C02 is preferentially incorporated rather than 13CC>2. In about 10% of plants, the initial fixing of atmospheric CO2 is accomplished in quite a different way. CO2 enters the leaves through the stomates and then penetrates an outer layer of cells, called the mesophyll cells, where it dissolves and equilibrates with bicarbonate. Subsequent reaction with phosphoenolpyruvate (PEP), catalyzed by the enzyme PEP carboxylase, yields Fig. 1. C3 AND C4 PHOTOSYNTHETIC PATHWAYS A. CC-2 1.100% 1 3 C ATMOSPHERIC AIR STOMATE (OPEN) EPIDERMIS *** RuDP ^ - R u D P CARBOXYLASE - CARBOHYDRATE 1.080% 1 3 C PGA INTERCELLULAR AIR-SPACE CELL WALL PHOTOSYNTHETIC CELL B. ATMOSPHERIC AIR C02 1.100% 1 3 C —t EPIDERMIS INTRACELLULAR AIR-SPACE HCO3" PEP. OAA RuDP CARBOXYLASE * PEP CARBOXYLASE • MESOPHYLL CELL WALL BUNDLE SHEATH CELLS CARBOHYDRATE 1.090% 1 3 C A. 3-carbon Calvin-Benson cycle with significant isotopic effect (***). B. 4-carbon Hatch-Slack pathway with minor isotopic effect (*). PYR=pyruvate; PEP=phosphoenolpyruvate; OAA=oxalo-acetate; RuDP=ribulose diphosphate; PGA=phosphoglycerate. (Modified from Bjorkman and Berry, 1973). 30 oxaloacetate (OAA) from which the C4 compounds malic acid and aspartic acid are formed (see Fig. 1B). For that reason, these plants are called C4 plants. These four-carbon acids are transported into bundle sheath cells located further inside the leaves. Malic and aspartic acid are subsequently broken down enzymatically regenerating pyruvate and CO2. Pyruvate returns to the mesophyll cells and CO2 is incorporated as in the C3 plants to form 3-phosphoglycerate. As far as isotopic effects are concerned, a small effect occurs at the PEP carboxylase step but is much smaller than that in C3 plants. Also, since the bundle sheath cells act a "closed" compartment, exchange or loss of CO2 does not occur. Isotopic discrimination (or fractionation) is thus minimal. As a result of these two pathways of photosynthesis, carbon in C4 plants is distinctly more abundant in 1 3 C than carbon in C3 plants. Although these differences are impressive when expressed in d 1 3 C , in absolute terms they are a few hundredths of a percent. In addition to fractionation during CO2 fixation, metabolic fractionation of carbon occurs within the plant (Park and Epstein, 1961). 1 3 C / 1 2 C ratio analysis of chemical fractions from several plant phyla show in all cases the lipid fraction is depleted in 1 3 C compared to the whole plant, and that the degree of l 3 C depletion is inversely related to the amount of lipid in the plant (Park and Epstein, 1961). Schoeller et al. (1980) have analyzed a series of foods and published their 1 3 C isotopic abundances. Basically, all cereals, fruits and vegetables are of C3 origin except for corn-derived products and sugar cane which are of C4 origin. Animal products, ie. meat, milk and cheese, will vary in 1 3 C natural abundance depending on the animal's food source (grasses vs maize). 1 3 C / 1 2 C ratio analysis is conventionally performed with a dual-inlet isotope ratio mass spectrometer. 31 Isotope Ratio Mass Spectrometer (IRMS) Instrumentation (Wolfe, 1984; Hachey et al., 1987) IRMS is used to measure isotopic abundances of elements in gaseous form. A schematic diagram is illustrated in Fig. 2. The inlet system contains holding cells for both a standard gas and a sample gas to be measured. Capillary "leak" lines admit the sample and the standard alternately to the ion source under high vacuum. This dual-inlet system allows all mass spectrometric conditions to be maintained at constant settings during the analysis of both the sample and the standard. Once in the source, the gas molecules are bombarded with electrons such that some lose an outer electron and become positively charged. Repelling electrodes force these ions into the magnetic flight tube where they are radially deflected onto collector plates according to their masses. In contemporary IRMS instrumentation, after introduction of the standard and sample gases, the entire process of valve sequencing, pressure matching, ion source tuning and focusing, vacuum monitoring and data collection is controlled by computer. The radial path (r) of each molecular ion varies with the square root of its mass (m) according to the formula: r 2 = m2V where V= the accelerating voltage, eB 2 e= the ion charge, and B= the magnetic field strength. Each mass is different enough to allow simultaneous collection of both the heavier and lighter isotopes on seperate collector plates. For example, 1 2 C 0 2 at mass 44 and 1 3 C 0 2 at mass 45 each create a different ion current. The ratio of these currents gives the isotope ratio which is then compared to that of the standard. Instrument internal precision is generally less than 0.05%© for 1 3 C . Fig. 2. SCHEMATIC DIAGRAM OF AN ISOTOPE RATIO MASS SPECTROMETER INLET ION SYSTEM SOURCE INLET SYSTEM CONTROL] ION SOURCE CONTROL MAGNET FLIGHT TUBE PUMPING SYSTEM ANALYZER CONTROL • > COMPUTER MAGNETIC CURRENT CONTROL COLLECTOR COLLECTOR SIGNAL PROCESSOR AMPLIFIERS t-1 AMPLIFIERS < (Modified from Hachey et al., 1987) 33 13c02 Determination Typically, 1 3 C is present in all forms of carbon to around 11,110 ppm (1.1%), and the typical IRMS can determine its absolute concentration to within 1-2 ppm. Over short ranges of comparison in the region of natural levels of 1 3 C , it is customary to compare the 1 3 C / 1 2 C ratio in a sample to that in a given standard. For convenience, this difference is multiplied by 1000% and referred to as part per thousand (del per mil) or simply "delta". It is determined according to the Craig formula (1957): d13C(%„) = ft13C/12C) sample -1 1000 _( 1 3 C/ 1 2 C) standard where sample and standard are of mass 45 and 44, respectively. The geochemical origins of these measurements are reflected in the use of a now obselete belemnite limestone standard found in the PeeDee range of North Carolina, referred to as PDB. Most natural levels contain less l 3 C than the PDB standard, so values are usually negative with respect to this standard. A 3 value of -10 means that the 1 3 C / 1 2 C ratio of the sample is less than that of the standard by 10 per mil or 1%. Analysis of Respiratory CO2 Collection of respiratory CO2 for 1 3 C isotopic abundance can be accomplished by a variety of techniques. 1) Direct trapping can be achieved by having subjects exhale directly into a spiral trap immersed in liquid nitrogen (Lacroix et al., 1973). This requires that liquid nitrogen be available at the collection site as well as the vacuum apparatus used during breath collection. This approach is poorly suited for most experimental protocols. 34 2) Human breath can also be collected in evacuated serological tubes (Schoeller and Klein, 1978). Subjects exhale into a collection bag from which a 20-50 ml sample is injected into an evacuated tube for storage. Although extremely convenient, this technique contributes to isotopic variability of breath CO2 due to the presence of a contaminant originating in the rubber septa in different batches of Vacutainer® tubes (Milne and McGaw, 1987). 3) Absorption of breath in a NaOH solution can trap CO2 as CO32" (Schoeller et al., 1977). Static absorption involves exhalation through tubing into an empty collection flask to which NaOH is added and the flask is then stoppered and shaken for 30 minutes. Dynamic trapping is quicker and more convenient since the subject exhales through tubing directly into the NaOH solution for about 5 minutes. In both instances, the CO2 (as CO32") can be transferred to a screw-top vial for transportation and storage. When ready for analysis, the CO2 is released from the basic solution by acidification and trapped in a liquid nitrogen bath at under vacuum. It is then purified cryogenically as is also necessary for any of the above techniques. In this distillation process at less than 5 mm Hg, water vapor is removed with a liquid nitrogen-methanol slush (-78°C), the CO2 is frozen in liquid nitrogen (-196°C) and all other gases are pumped away. The purified CO2 is then sealed in an evacuated collection tube to await mass spectrometric analysis. Dynamic collection of CO2 as CO32" in 2 N NaOH was shown to result in isotopic fractionation, ie. the lighter 1 2 C isotope is trapped in preference to the heavier 1 3 C isotope (Schoeller et al., 1977). When compared to samples collected directly in a liquid-nitrogen trap, a nonfractionating method, the magnitude of fractionation was established to be -13±0.8%»(X±2SEM). In earlier work, Yemm and Bidwell (1969) observed a 2% discrimination against 14C02 in the absorption of CO2 by a 0.1 N NaOH solution. They attributed this effect to the 35 relative rates of diffusion of l 2 c 0 2 and ^ ^C02 across the boundary layer of the absorbing solution. It is of interest to note that values for a "discrimination factor" (DF) calculated for each of 5 seperate experiments were extremely consistent (SD=0.003). Although a 2% discrimination against 1 4 C 0 2 would introduce an error of 2% in 1 4 C measurements, a similar discrimination against l 3 C 0 2 would result in a 2 0 % change in the 1 3 C 0 2 abundance ratio. This is usually much greater than the peak appearance of labelled 13C-substrate. Fortunately, since we are dealing with ratios of 1 3 C / 1 2 C , fractionation per se does not cause difficulties provided the variations in the degree of fractionation from sample to sample are small. Schoeller et al. (1977) observed the variation as insignificant (SD=0.17%«) when dynamic trapping occurred in a 250 ml round-bottom flask. Alternately, the static collection technique eliminated fractionation; however, the 30 minute shaking period significantly added to the time required to collect each breath sample. Determination of 1 3 C in Organic Samples by Combustion The preferred technique is called "sealed tube combustion" in which a small sample (2-5 mg) is sealed in a quartz tube containing copper oxide and silver wire. The tube is then combusted at 650°C for 2 hours and allowed to cool to room temperature in the closed furnace. The copper oxide is a source of oxygen for the formation of CO2 and H2O. In addition, above 500°C the elemental copper promotes the conversion of nitrogen oxides to N2 and sulfer oxides to CUSO4. The slow cool-down period eliminates these contaminates and below 500°C the copper eliminates excess O2. The silver wire removes halogens and converts sulfer oxides to Ag2S04. Once cooled, the sealed tubes are cracked under vacuum and the CO2 purified cryogenically as previously described. 36 Important Considerations in 1 3C-Tracer Studies in Humans A. Variations in 1 3 C / 1 2 C Ratios It has been reported that isotopic differences exist between similar foodstuffs from different geographical areas. Protein in meat samples from the United States were calculated to be enriched in 1 3 C relative to Japanese and German samples (Nakamura et al., 1982). This to a large degree reflects the use of corn-derived animal feeds in the United States. In addition, systematic isotopic differences were also observed. For example, German grains and produce were sytematically 2%o lighter than the American and Japanese products. The incorporation of isotopically "light" CO2 from fossil fuels used in nearby industrial areas may have occurred. A great deal of variation in 9 1 3 C can also occur within a plant species (Smith and Epstein, 1971). Growth under different environmental conditions may necessitate an adaptation for more efficient photosynthetic carbon fixation which could be reflected in high 1 3 C / 1 2 C ratios. Under conditions of low CO2 concentration, the Hatch-Slack route provides a very efficient means of fixing atmospheric CO2 (Bjorkman and Berry, 1973). Also, the stage of plant development affects its relative 1 3 C / 1 2 C ratio. During plant growth, respired CO2 is enriched in 1 3 C whereas during plant decomposition, the respired CO2 is enriched in 1 2 C (Park and Epstein, 1961). In most metabolic studies using naturally enriched 13C-substrates, the above variations can be minimized by administering to all subjects a common source. However, the effect of substrate intake and the physiological state of each subject can induce significant natural variation in the 1 3 C / 1 2 C ratio. Physiological changes alter the carbon isotope ratio in response to changes in endogenous carbohydrates. During fasting, increases in lipid oxidation 37 cause the isotope ratio of breath CO2 in rats and humans (Schoeller et al., 1984) to decrease because lipids are isotopically lighter than carbohydrates. Conversely, during exercise, increased CHO utilization is reflected by an increase in the 1 3 C / 1 2 C ratio of breath CO2 (Schoeller et al., 1984; Wolfe et al., 1984b; Barstow et al., 1989). Administration of a substrate can further alter the carbon isotope ratio due to oxidation of the naturally labelled substrate and/or shifts in the source of substrate for energy metabolism. Unlike 1 4 C tests in which the isotope background is low, 1 3 C tests are performed against a large background or natural abundance of approximately 1.1% 1 3 C . Because of natural background, the labelled CO2 that is measured, is actually the amount of 13C02 in excess of the 13c02 abundance before the labelled substrate is administered. Also, the background value can change during the period of the test affecting the detection of excess 13C02 generated from the labelled substrate. This requires administration of large enough quantities of labelled substrate to generate l3C02 levels much larger than those generated from changes in the endogenous 13c/12c ratio. In the resting state, variations in 13c/12c ratio have been noted between individuals within a single population as well as within a single individual (Schoeller et al., 1980; Barstow et al., 1989; Jones et al., 1985). Measurement error accounts for a relatively small amount of this variation (Schoeller er al., 1977), whereas the biological variation is quite large. When diet is controlled, differences in body composition, resting energy expenditure and rates of food absorption and digestion can contribute to interindividual variability (Jones et al., 1985). Intraindividual variability may arise from subtle differences in energy expenditure and/or intake, as well as from the variability in the 13c/12C ratio of the metabolic fuel. 38 B. Baseline Measurements To correct for the natural abundance of 1 3 C , background levels are needed. Previously this was accomplished by obtaining a resting background sample before isotope infusion or ingestion, and subtracting it from all samples collected after isotope infusion or ingestion. This approach is satisfactory for experiments in which no aspect of the metabolic state changes from the time at which background and enriched samples are collected. Maintaining 1 3 C abundance at this steady state level can be achieved by fasting and/or using a test protocol that does not produce a change in endogenous 13C02 abundance of expired air other than that resulting from label oxidation. Schoeller et al. (1980) reported that the combination of resting and fasting minimizes the variation in 1 3 C / 1 2 C ratios in respiratory CO2 owing to a relatively constant rate of lipid oxidation and therefore producing CO2 of nearly constant isotopic abundance. Unfortunately, withholding food is not always possible or desirable. Isotopically neutral test meals containing foods which are low in 1 3 C have also been used to minimize breath 1 3 C levels (Schoeller et al., 1980). Alternately, a normal diet can be used if a background profile for each subject on that diet is obtained (Jones et al., 1985). Subsequent subtraction from corresponding values of a test substrate for each subject will provide a valid estimate of 1 3C-label response. Of utmost importance is the maintenance of the initial testing conditions during subsequent tests. A similar approach can be utilized under exercising conditions where changes in background enrichment can occur without substrate intake owing to a change in the mix of endogenous substrates being oxidized (Wolfe et al., 1984b; Barstow et al., 1989; Schoeller et al., 1984). The need to correct this arises if the 39 enrichment in 13C02 from test substrate is in the same range as the change in background enrichment resulting from the experimental perturbation. This can be determined by looking at the changes in background enrichment that occur with exercise alone, ie. without isotope administration. C. Determining Substrate Oxidation Rate Based on 13C02 Output A principal goal of many tracer kinetic experiments is to determine the oxidation rate of the traced substances by the appearance in breath of labelled C originating from the tracer. Two major concerns are 1) determining the fractional recovery of labelled 1 3 C in breath CO2, and 2) determining VCO2. Fractional recovery is usually assumed to be about 80%, based on a number of studies that have measured the recovery of label following administration of a bicarbonate tracer. These studies have assumed that CO2 released from bicarbonate, a reaction catalyzed by carbonic anhydrase in blood and cytosol, and CO2 released from oxidation of fuels, a predominantly mitochondrial process, are handled similarly by the body. Bicarbonate pool sizes and kinetic parameters are needed to accurately determine nutrient oxidation. They are required to account for the effect of transit through the bicarbonate pools on the rate of appearance of labelled carbon in breath, and to quantitate the total recovery of labelled respiratory CO2. When labelled carbon is given as 1 3 C - or 14C-bicarbonate, it has been well documented that a fraction of the labelled carbon liberated from the tracer is not recovered in breath as labelled CO2, at least during the short period of the tracer experiment. The fraction retained is believed to represent CO2 fixation in metabolic pathways or loss to slowly exchanging bicarbonate pools. Several models have been proposed to define whole body bicarbonate metabolism (Winchell et al., 1970; Steele, 1955). They are 40 constructed from estimates of arterial and venous CO2 levels, organ blood flow rates and tissue levels, and production rates of bicarbonate and CO2. In general, most models consist of at least a rapidly turning over pool, in equilibrium with blood and well-perfused tissues, and more slowly turning over pools that are comprised of less well-perfused tissues. The more rapidly turning over pools are essentially in equilibrium with the body water pool, whereas the slowest turning over pool is presumed to be a "sink" that is not in ready equilibrium with blood bicarbonate. Bone has been proposed as the major sink for label, based on its large carbonate and bicarbonate content. To date, two experimental approaches characterized bicarbonate kinetics. The first uses plateau enrichment of breath 1 3 C 0 2 during continuous infusion of labelled bicarbonate to obtain an estimate of overall flux of bicarbonate (Robert et al., 1987; Wolfe et al., 1984a). The second uses multicompartmental analysis of the decay of 1 3 C 0 2 in breath after a bolus dose of NaH 1 3 C03 (Irving et al., 1983). The latter approach allows for stipulation of nutrient pool sizes and transfer rates using a detailed mathematical description of bicarbonate kinetics. Single-bolus studies have revealed that the pattern of decay of CO2 enrichment in breath is very complex and best described by multiexponential equations (Issekutz et al., 1968; Irving et al., 1983). In the study by Irving et al. (1983), five subjects received a bolus injection of 10 mmol-kg-1 of NaH 1 3CC*3 afterwhich isotopic abundance measurements were determined for 6 hours. Decay of breath 1 3 C 0 2 was shown to follow a 3-exponential decay function with a negative baseline drift. When compared to the composite models described above, their multicompartmental model included a larger amount of freely exchangeable bicarbonate but smaller fractional rate constants for exchange of bicarbonate between fast and slow peripheral pools. The smaller fractional rate constants suggest that the assumption of complete HC03"-C02 exchange between 41 the vascular and tissue compartments is incorrect. Fractional recovery was around 50% which was much less than the most widely used fractional recovery term of 81%. Large inter- and intraindividual variations occurred in bicarbonate losses. In a subsequent study (Irving et al., 1984) these were minimized by carrying out a determination of bicarbonate kinetics simultaneously with oxidation measurements. The results obtained for recovery of labelled bicarbonate using the single-bolus approach have a wide range between 50-90%. This likely reflects the complexity of the system and the mathematics. With the constant infusion technique, the recovery is generally around 80%. Exactly where the isotope is lost, is not well documented. Some may be diverted to bone and/or biosynthetic pathways due to isotopic exchange in the TCA cycle. The rate constants of these metabolic pathways may change under different conditions, therefore it is reasonable to expect that fractional recovery would also change. When compared with intravenous routes of administration, fractional recoveries were not significantly different using an intragastric route (Hoerr et al., 1989). Their study also investigated whether a postabsorptive or continuously fed state alters l 3 c retention. Mean fractional recovery for the postabsorptive period was significantly less (p< .0001) than that in the fed state, 0.72 and 0.81, respectively. An interesting finding in this study was that label recovery from bicarbonate tracer was linearly related to VO2 (r=0.85, p< .0001) and VCO2 (r=0.82, p< .0001). When these parameters increased during feeding, label recovery also increased. One would expect a similar increase in label recovery during exercise since both VO2 and VCO2 increase. However, during exercise, the rate constants and pools sizes may be substantially altered (Slanger et al., 1970; Wolfe et al., 1984a). Wolfe et al. (1984a) observed an increase in recovery from 78% at rest to 42 106% after 20 minutes of mild exercise at 30% V02max. This was attributed to an increase in total VC02- Since the rate of 13C02 recovery is equal to VCO2 times enrichment, to maintain equilibrium a decrease in CO2 enrichment with 1 3 C must accompany the increased VCO2. In exercise, compensatory hyperventilation probably accounted for some of the extra CO2 excretion, and since some was labelled from prior isotope infusion, extra 13C02 was lost. Also, the change in the background enrichment of expired CO2 from exercise alone contributed to the total enrichment measured. Variability in CO2 measurements pose a serious threat in determining substrate oxidation rates. The basic principle in determining VCO2 is quite simple: the difference between the concentration of CO2 in inspired and expired air is multiplied by the volume of air expired corrected to STPD. In practice, determination of VCO2 is the largest source of error in determining total 13C02 excretion. In addition to error in the measurement of CO2 concentration and expired volume, alterations in breathing patterns can cause transient changes. Also, changes in VCO2 are observed after feeding second to stimulation of the metabolic rate. During a 6 hour glucose infusion, the percent of VCO2 from glucose did not change overtime but a progressive increase in the absolute rate of glucose oxidation was observed along with an increase in the percent of glucose uptake (Shaw and Wolfe, 1983). Both were due to the increase in VCO2. When lipid was added to the glucose infusion, no change in VCO2 resulted. These observations would lead to the conclusion that the increased FFA levels inhibited the ability of tissues to oxidize glucose. If, however, the uncontrolled variable, VCO2, is factored out, lipid infusion had no effect on glucose oxidation. In other words, by expressing the data as a function of VCO2, the fact that VCO2 was 43 different in the two situations is accounted for. This is accomplished by dividing the oxidation data by VCO2. The transition from rest to exercise represents a dynamic state associated with increases in CHO oxidation, CO2 production and O2 consumption. Under these circumstances, measurements of labelled 1 3 C in CO2 may not represent utilization of the labelled substrate. In fact, an initial decrease in isotope ratios is observed early in exercise (Schoeller et al., 1984; Barstow et al., 1989) despite evidence that CHO is being preferentially used (Wahren et al., 1971). This decrease likely reflects the time required for distribution of labelled l 3 C as 13C02 throughout the body bicarbonate pools. Because the preferred substrate for resting skeletal muscle is lipid, the bicarbonate in this pool is isotopically "light". Even if the fractional recovery of the labelled substrate is high, until a steady state of 13C02 exchange is achieved, breath collections will be isotopically "light". In the resting state, this can take as long as 60-90 minutes (Irving et al., 1983). On the other hand, VCO2 and thus total CO2 in these pools approaches a steady state within 4-6 minutes after the onset of exercise (Barstow et al., 1989), thereby adjusting much more rapidly than l3C02 in the transition from rest to exercise. Since the enrichment of exhaled CO2 with l3C02 is the ratio of 13C02 recovered to CO2 production, an initial decrease occurs. This implies that early in exercise, or in exercise of short duration, changes in the enrichment of exhaled CO2 with 13C02 may not reflect the underlying oxidative fuel mix since 13C02 equilibrates much more slowly than VCO2. In conclusion, 13C-tracer methods are useful for estimating in vivo rates of substrate oxidation in human subjects. However, excess CO2 produced by oxidation must be measured in the presence of a natural background of 1 3 C . This 44 requires knowledge of changes in background that can occur over the duration of the study, and the administration of a large enough dose of labelled substrate. The fractional recovery of labelled 1 3 C in breath CO2 should be calculated since it can change under various physiological circumstances. The sensitivity of both fractional recovery and total 13C02 excretion, to VCO2, warrants accurate measurement of the latter parameter. Also, caution should be used in selecting the best way of expressing oxidation data. Inter- and intraindividual variability should be determined and possible sources explained. These include variation in the method of respiratory CO2 collection and analysis, and, if a naturally labelled 1 3 C substrate is used, geographic and/or species-specific variation. 45 CHAPTER 3 METHODS AND PROCEDURES Subjects Six male subjects, age 23-31 years, volunteered to participate in the study, after having all the procedures explained to them and a written consent. All are competitive cyclists from the Lower Mainland. They were recruited by a written advertisement circulated at a local bike race, or through the Sports Medicine Clinic at the University of British Columbia. Three of the subjects were university students and three were employed full-time. All six were highly trained (V02max > 60 ml •kg"1-min"1). Physical characteristics are described in Table I. Following a V02max test, each subject was asked to complete three 75 minute cycle bouts, one week apart, at 60% V02max under three experimental conditions. All tests were performed on the same weekday for each subject, and within one hour of the same time. Two subjects completed the experimental trials in four weeks: one subject incurred an injury during a race after Trial 1 which required a week of convalescence; another subject had a scheduled university exam coincident with Trial 3. Selection of highly trained cyclists is based on experience that competitive athletes are generally willing and able to withstand considerable discomfort and to exercise until the development of physiological signs of exhaustion (Coyle et al., 1983). This requires a very motivated group which in itself may constitute a bias selection. Apparatus All tests were performed on a Mijnhardt cycle ergometer, Type KEM 3, equipped with toe clips and straps (or the subject's own pedals), and a racing TABLE I. Physical characteristics, maximal 02 uptake and workload of the subjects (X±SD) Age, years Height, cm Weight, kg V02max, l-min"1 V02max, ml-kg'^-mirr1 Workload(60% V02max),Watts Workload(90% V02max),Watts 26.8 ±3.06 186.5 ±4.89 80.8 ±8.02 4.96 ±0.45 61.2 ±1.81 243.0 ±22.7 396.0 ±35.1 47 saddle. This ergometer has an electronically-controlled braking system and hyperbolic operation which allows for workloads to be adjusted independent of pedalling speeds. A Medical Graphics Corporation (MGC) Exercise System 2001 was used for measurement of ventilatory parameters. In this system, expired gas flow is measured with a pneumotachometer and a pressure volume transducer. Volume is obtained from integration of the flow signal by a waveform analyzer and converted to BTPS. The expired O2 concentration is measured by a zirconia fuel-cell sensor. It is linear from 0-100% O2 with an accuracy of ±0.05% O2. Expired CO2 concentration is measured using an infrared absorption analyzer which is linear from 0-10% CO2 with an accuracy of ±0.02% CO2. The expired gas was sampled continuously from the mouthpiece and computed values for each breath were extapolated to a 15-second or 60-second average. Calibration of the MGC Exercise System 2001 was done daily and/or prior to each trial. The calibration procedure involves: 1) Determination of a volume conversion constant. This consists of adjustment for zero flow and volume calibration using a 3-liter calibration syringe. Criteria for an accurate zero flow value is a number less than ±9 ml-sec 1. Criteria for an accurate conversion factor is a number less than 2%. Volume calibration did not differ by more than 100 ml-sec^-volr1 between calibrations. 2) Calibration of system analyzers. This consists of adjustment for an O2 cell zero and calibration against high quality, analyzed and guaranteed gases supplied by Medigas Pacific, Ltd. (Vancouver, B.C., Canada). Criteria for an accurate O2 cell zero is a reading of 0.00±0.03%. Criteria for accurate O2 and CO2 concentrations were values that did not differ by more than 0.15% from the percentage of O2 and CO2 in the calibration gas mixture. 48 3) O2 and CO2 phase calibration. This determines the phase delay or time difference between measurement of volume by the pneumotach, which is instantaneous, and measurement of gas concentration by the O2 and CO2 analyzers. Once determined, the phase delay is used to match the time course of each volume sampled to the time course of its corresponding gas concentration to arrive at a correct cross-integral. Criteria for an accurate O2 and CO2 phase calibration are repeat values within ±0.01 seconds (10 msec) of their respective previous values. Preliminary Assessment Using an incremental cycling protocol (30 watt increase per minute), a V02max test was performed on the cycle ergometer interfaced with a Medical Graphics Corporation (MGC) Exercise System 2001 for measurement of ventilatory parameters. Heart rate was recorded at each incremental stage using a 3-lead LIFEPACK6 (Physio-Control™) monitor. The criteria defined for attainment of V02max (Fox and Mathews, 1981), of which at least three were obtained, is: 1) a levelling or decrease in VO2 with increasing work, 2) a plateau in heart rate, 3) RQ> 1.15, 4) volitional fatigue. An oral screening test indicated no family history of diabetes or other sugar intolerance, and none were on any prescribed medications. One subject took a multiple vitamin mineral supplement every second day for the past six months. 49 Dietary Assessment Dietary consistency throughout the experimental protocol was important for two major reasons: 1) to control for pre-trial glycogen stores, 2) to control for pre-trial 3 1 3 C values. Each subject was asked to follow a similar diet for 3 days prior to each trial. A 3-day food record (see APPENDIX A) was completed and brought to the first trial. A photocopy was provided, and thereafter each subject was asked to follow the same diet for the 3 days prior to all subequent trials. To reinforce its importance, a 3-day food record was also requested at all subsequent trials. All subjects consumed a small meal containing 75 g of CHO two hours prior to all trials. Meals were individualized to accomodate food preferences. Carbohydrate content was determined from published food values (Pennington, 1987). The same meal was ingested prior to each trial. To minimize the intake of 13C-containing foods, subjects were provided with a list of "Foods To Avoid" for the duration of the study (see APPENDIX A). All sugar in B.C. is refined from sugar cane which is a C4 plant with a high natural abundance of 1 3 C . To minimize excessive 1 3 C intake from this source, beet sugar, a C3 plant with a much lower 1 3 C abundance, was generously donated by the Alberta Sugar Company for the subjects' consumption. Experimental Protocol Unless otherwise stated, the following protocol applied to all experimental conditions. Subjects reported to the laboratory 2 hours after ingesting a 75 g CHO meak. In two of the trials, an indwelling catheter was inserted into the cephalic vein and kept patent with an infusion of heparin-treated saline. Resting blood 50 samples were collected for determination of glucose and lactate concentration, as well as a "pre-test" blood sample for Hb and Hct. Once positioned on the cycle ergometer, resting values for V02.VC02, and RER were obtained using the MGC Exercise System 2001 metabolic cart. A sample of expired air was collected for isotopic determination o f 1 3 C . After a 5 minute warm-up at 50% V02max, the subjects cycled for 75 minutes at a workload corresponding to 60% V02max, followed by a ride to exhaustion at 90% V02imax. Room temperature was maintained within ±2°C by the use of a fan. Two of the experimental conditions required the ingestion of a treatment beverage at time=5 which was after 5 minutes of cycling at 60% V02max. The NaH 1 3C03/NaH 1 2 C03 beverage mixture was consumed within a 30 second period, and the 1 3C-glucose drink within a 2 minute period. Water was allowed ad libitum for all three experimental conditions. Blood samples1, air samples, and respiratory measurements were collected at: -10 min - RESTING sample 5 min - after 5 minutes of cycling at 60% V02max 20 min - after 20 minutes of cycling at 60% V02max 35 min - after 35 minutes of cycling at 60% V02max 50 min - after 50 minutes of cycling at 60% V02max 60 min - after 60 minutes of cycling at 60% V02max 70 min - after 70 minutes of cycling at 60% V02imax A diagrammatic presentation of the experimental protcol is illustrated in Figure 3. 1For hemoglobin and hematocrit determination, only "pre-test" and "post-test" samples were obtained. Fig. 3. DIAGRAMMATIC PRESENTATION OF THE EXPERIMENTAL PROTOCOL REST » < EXERCISE 50%-* < -60% V02max — • 90% INGEST TREATMENT ^10 I 5 6 5 # 20 35 50 65 75 TIME* Vertical arrows indicate collection times for blood samples, air samples and respiratory measurements. 52 Sample Preparation 1) Preparation of blood glucose and lactate samples: 5 ml blood samples were drawn from the cephalic vein into "grey top" 7 ml draw Vacutainer® tubes and kept under ice for the duration of each trial. Each tube contained 14 mg potassium oxide and 17.5 mg sodium floride for glycolytic inhibition. The tubes were then spun in an IEC Model CLINI-COOL (Damon/International Equipment Company) refigerated centrifuge at 700 rpm for 10 minutes or until plasma was clear. The plasma was seperated from the RBC, placed in a 4 ml glass tube and frozen until all trials were completed. 2) Preparation of hemoglobin and hematocrit samples: 5 ml blood samples were drawn from the cephalic vein into "lavender top" 7 ml draw Vacutainer® tubes which contain 10.5 mg EDTA(K3) as anti-coagulant. Samples were kept at room temperature for the duration of the trial, after which they were transported to the Department of Laboratory Medicine at the University Hospital, U.B.C. Site. 3) Dynamic trapping of CO2 from expired air samples: From an outport of the metabolic cart, expired air samples were collected in previously evacuated Douglas bags for 1 minute (2 minutes for resting samples) and stoppered. Using a Beckman™ pump equipped with an E/C Meter (Brooks), 2 liters of expired air were bubbled through 4 ml of a 1.0 N NaOH solution at a rate of 400 ccmin"1 to trap the CO2 as CO32" in NaOH 1 . The solution was transferred to a 5 ml plastic tube, capped and frozen until all trials were completed. 3) Purification of CO2 from expired air samples: Thawed samples of CO2 trapped in NaOH were liberated in an evacuated Rittenberg tube by acidification with 1 Dynamic collection of CO2 as CO32- in NaOH was shown to result in fractionation of the carbon isotope, ie. the lighter isotope, (12C02) is trapped in preference to the heavier one,(13C02). The magnitude of this fractionation was established by Schoeller et al.(1977) to be -13±0.8%«and is highly reproducible. Since all calculations of 1 3 C abundance deal with 1 3 C to 1 2 C ratios, knowledge of the absolute abundance of 1 3 C is not required. 53 orthophosphoric acid (H2PO4), and then purified cyrogenically. This process involves removal of water vapor by a methanol-liquid N2 slush (-78°C), freezing in liquid N2 (-196°C) and subsequent removal of other gases with a 50 mTorr vacuum (see APPENDIX B for a detailed outline of this procedure). Sample Analysis 1) Blood glucose and lactate concentration: Thawed plasma samples were recentrifuged for 10 minutes (2200 rpm, 4°C) in a HERMLE Z 360K centifuge. An enzymatic, calorimetric assay was used to determine concentration (Sigma Diagnostics). For plasma glucose concentration, the standard deviation (SD) and coefficient of variation (V) were ±5.062 mg-dl"1 and 5.8%, respectively. For plasma lactate concentration, SD and V were ±0.070 mmol-l"1 and 2.62%, respectively. A SHIMADZU UV-160 narrow-bandwidth spectrophotometer was used for absorbance readings (see APPENDIX B for a detailed outline of both procedures). 2) Hemoglobin and hematocrit determination: "pre-test" and "post-test" blood samples were analyzed using by the Department of Laboratory Medicine, University Hospital, U.B.C. Site using a Coulter Counter SI 1986 (Coulter Electronics of Canada Limited). For hemoglobin concentration, SD and V were ±0.17 g-dl"1 and 1.3%, respectively. For hematocrit values, SD and V were ±0.025% and 5.2%, respectively. 3) 1 3 C abundance in expired CO2: The 1 3 C to 1 2 C ratio in the purified CO2 sample was analyzed by isotope-ratio mass spectrometry. A dual inlet, triple collector PRISM #5 mass spectrometer (VG ISOGAS) was used for analysis of air samples for four subjects. Analysis was done through Department of Oceanography, U.B.C. For analysis of air samples for two subjects, a dual inlet, double collector SIRA #10 mass spectrometer (VG ISOGAS) was used. Analysis was done through the Department of Physics, U. Calgary. For both mass 54 spectrometers, instrument internal precision was less than ±0.05% . Reliability between the two instruments was high (r=0.95, p<.01). Due to cost restraints, half of the collected samples were analyzed in duplicate, the other half by a single analysis. 4) Respiratory gases: A mouthpiece with a breath-by-breath sample line allowed for direct measurement of VO2, VCO2 and RER by open-circuit indirect calorimetry using the MGC Exercise System 2001 metabolic cart Computations 1) Plasma Glucose (mg-dM) = Atest" Ablank x 100 A std - Ablank where A= absorbance at 450 nm (Sigma Diagnostics, 1987). 2) Plasma Lactate (mmol/L) = (Atest - Ablank) x 7.23 where A= absorbance at 340 nm (Sigma Diagnostics, 1988). 3) Changes in Plasma Volume (APV): Estimated from hemoglobin and hematocrit values taken Pre- and Post-test according to the formula (Costill et al.,1974): %APV = [HbA/HbB(1-HctA) - (1-HctB)] x 100 (1-HctB) * where B and A are Pre-test and Post-test values, respectively 4) Isotopic Composition of Expired CO2: Expressed as the delta per 1,000 difference between the 1 3 c to 1 2 C ratio in the sample and in a known standard using the Craig formula (Craig, 1957): d13c (%) = (13c/12Qsample (13c/12C)standard x 1,000 The 3 1 3 C was then expressed by reference to the International Standard PDB. 5) Glucose Oxidation (g-min-1) =|3 1 3C e - d ^ C p ] x VCO2 X 1.35 |_a13Cg-a13CoJ 0.666 8l3ce= the 8 1 3 C value of the CO2 sampled in each collection period following glucose administration, 55 313Co= the 3 1 3C value of the CO2 sampled in each collection period during the "no isotope" (BASELINE) trial, 313Cg= the d 1 3 C value of the glucose given, 0.666= the fractional recovery of 1 3 C or 1 4 C in expired CO2 from published human studies using a bolus administration of bicarbonate (Hoerr et al., 1989), and 1.35= the amount of glucose converted to CO2 (1 liter of CO2 corresponds to 1.35 g of glucose oxidized). Dependent Variables The dependent variables served one of 2 functions: 1) Measures of homogeneity and/or consistency between trials (Group A) 2) Measures of treatment outcome (Group B) Trie Group A dependent variables are: VO2, blood lactate and hemoglobin concentration, hematocrit The Group B dependent variables are: 3 l 3C, blood glucose concentration, RER, VCO2, glucose oxidation rate, sprint time to exhaustion at 90% V02imax Experimental Conditions Three experimental conditions were tested which gave rise to 21 measures of each dependent variable for each subject. The three conditions were assigned to three columns (trials) of a Latin Square repeated twice. The specific ordering of conditions are outlined in Table II. Subjects were assigned randomly to one of three rows of the Latin Square, with the constraint that only two people could be assigned to the same condition per trial. 56 TABLE II. Order of conditions for each subject Subject Trial: 1 2 3 1 a b c 2 c a b 3 b c a 4 b a c 5 a c b 6 c b a 57 The three conditions were: A) "No isotope" which provided BASELINE measurements, B) NaHl3c03/NaH1 2C03 mixture, and C) 1 3C-glucose drink. The composition of the two treatment beverages (B,C) are described in Table III. Experimental Design Analysis of 913C obtained from Treatment B indicated that insufficient levels of 13C enrichment were given in the NaH13C03/NaH12C03 mixture. Any isotope effects were masked by the large volumes of C02 produced during the experimental protocol. Since the amount of bicarbonate given was too small to cause any physiological effects, this treatment served as a second "No isotope" experiment (Treatment A). Thus values obtained for the dependent variables on Treatment B were averaged with those from Treatment A. This adjustment gave rise to 7 measures for each variable in each experimental conditions: No isotope (Treatment A/B) and 13C-glucose (Treatment C). In the analysis of the REST data, the experimental design can be treated as one way repeated measures. For the EXERCISE component, the design is a 2 x 6 factorial with repeated measures on both factors. The factors are then Treatments with two levels, and Time with 7 levels. The statistical design is diagrammed in Figure 4. To test for a treatment effect, a 2-way analysis of variance with repeated measures on both factors was performed. Post hoc comparisons utilizing orthogonal polynomials followed a significant F. In addition, the following comparisons were made to test the specific hypothesis: 58 TABLE III. Treatment compostiton, osmolality and amount of carbohydrate delivered Treatment NaHl3c03/NaHl2c03 mixture 1 3C-glucose Composition 5mg/kg of 1% NaH"l3c03 dissolved in 75ml H2O 100g of naturally 1 3 C enriched cornsyrup mixed in 275ml H2O Osmolality g CHO (mOsm/kg H2O) delivered 1.57/kg 0 2020 100 Fig. 4. THE STATISTICAL DESIGN OF THE EXPERIMENTAL PROTOCOL EXPERIMENTAL CONDITION: EXERCISE DATA: B T2 T3 T4 T5 T6 T7 T2 T3 T4 T5 T6 T7 2x6 Repeated Measures ANOVA T2 T3 T4 T5 T6 T7 REST DATA: T1 T1 T1 one-way Repeated Measures ANOVA 60 Test of Hypothesis 1 Hypothesis one states that the rate of carbohydrate oxidation increases when an oral glucose load is administered. This would be indicated by a significant linear or quadratic Treatment by Time interaction for both RER and ai3c. Test of Hypothesis 2 Hypothesis 2 states that after 1 3C-glucose ingestion, changes in plasma glucose concentration over time differ from 9 1 3 C . This would be tested with orthogonal polynomials following a significant Time main effect for both variables. Test of Hypothesis 3 Hypothesis 3 states that following 1 3 C ingestion, RER adjusts more rapidly than 3 1 3 C . This would be tested using orthogonal polynomials following a significant Time main effect for both variables. Test of Hypothesis 4 Hypothesis 4 states during exercise baseline 1 3 C enrichment of expired CO2 increases. This would be indicated by a significant linear or quadratic trend for Time main effect on Treatment A/B. Data Analysis For the RESTING component, a one way analysis of variance for repeated measures was performed plasma glucose concentration1, 9 1 3 C , VCO2, 1 n=5 due to loss of one sample. 61 and RER. A 0.05 level of significance was used. Data were analyzed on the U.B.C. MTS mainframe computer system using the UCLA Biomedical Programme BMD P2V (BMDP Statistical Software). For the EXERCISE component, a 2-way analysis of variance with repeated measures on both factors and trend analysis was performed on those dependent variables mentioned above in addition to plasma lactate concentration, VO2 and VCO2. A 0.01 level of significance was used. The computer programme BMD P2V was used to give a repeated measures analysis of variance with an orthogonal breakdown of each source of variation to test for trend. All preplanned comparisons were tested at a 0.01 level of significance. Sprint time and changes in plasma volume were examined using a one way repeated measures analysis of variance. Both were tested at a 0.05 level of significance.using STATVIEW 512+™, Version 1.0, 1986 (Brainpower, Inc.). Tables and Graphs were constructed using Cricket Graph, Version 1.10, 1986 (Cricket Software, Inc.), and/or Statview 512+™, Version 1.0, 1986 (Brainpower, Inc.). 62 CHAPTER 4 RESULTS Blood Parameters Rest values for plasma glucose concentration did not differ significantly (Fi ,5 = 2.15, p>.05) between the two treatment conditions1, X = 134 mg-dl"1 (Fig. 5A). Exercise, before glucose ingestion, did not alter plasma glucose response. Ingestion of the glucose load (Treatment C) resulted in higher peak values; however, by 60 minutes of exercise, plasma glucose levels were at, or near, pre-exercise values (Fig. 5A). The analysis of variance of this exercise component shows only the main effect of Time as significant ^5,25 = 4.08, p= /.0076). Most of the variance in the Time main effect when averaged over the two conditions was accounted for by the cubic trend of plasma glucose concentration over Time (Fi.5 = 24.45, p= .0043). Changes in plasma glucose concentration between the two conditions over the collection times were not significant (F5.25 = 2.66, p= .0680). Plasma lactate concentration showed an initial increase from rest values with exercise which remained unaffected with glucose ingestion (Fig. 5B). Without glucose ingestion, mean exercise lactate values increased from 2.24 to 3.34 mmol-l"1 over 70 minutes, but this did not reach significance at 0.01 level of significance. The large variability in lactate response (range = .0835 - 1.769) is the likely cause. Reductions in plasma volume were similar under both experimental conditions (Fi,4 = 0.16, p>0.05) as illustrated in Fig. 6. 1 The two conditions are Treatment A (water ad lib) and Treatment C (100 g cornsyrup + water ad lib). Fig. 5A. CHANGES IN PLASMA GLUCOSE CONCENTRATION (X±SD) TIME (min) 65 Calorimetric Results For both experimental conditions1 , RER decreased over 75 minutes of exercise suggesting that the contribution of carbohydrates to the total energy demand declined independent of glucose supplementation (Fig. 7A). An analysis of variance of the exercise component shows a significant Time main effect (F5(25 = 7.88, p= .0091) when averaged over both conditions. Eighty-three percent of the variability is accounted for by a significant linear trend (F-^ 5 = 48.18, p= .001), as evidenced by the steady linear decrease in mean RER from 1.00 at 5 minutes to 0.96 after 75 minutes. There was no difference in RER response between the two conditions. There was a significant difference between the two conditions in resting RER values (Fi,5 = 13.08, p= .0153). This is due to the variability in the resting data which was 5-10 times greater than in the exercise data. This variability is secondary to fluctuations in tidal volume. These fluctuations are proportionately larger in subjects who are at rest. Variability in O2 consumption was less than 0.5 l-min-1 at any collection time under any experimental condition. When averaged over the two conditions, there was a significant Time main effect (F5;25 = 12.07, p< .0001) which can best be interpreted using trend analysis. A significant linear trend (F-^5 = 25.50, p= .0039) accounted for 98% of the variability. Fig. 7B shows an increase in mean VO2 during exercise from 3.14 to 3.51 l-min - 1; however, the response between the two conditions did not differ. Whereas O2 consumption showed a gradual upward shift with exercise, CO2 production remained fairly stable at 3.2 - 3.3 l-min-1 throughout exercise and under all experimental conditions (Fig. 7C). 1. The two conditions are Treatment A/B and Treatment C (refer to Chapter 3, Experimental Design, p. 57). Fig. 7A. CHANGES IN RER (X+SD) 1.6 r WATER CORNSYRP 7B. CHANGES IN 02 CONSUMPTION (X±SD) 4 r [ h _ _ _ ! -4 •+* 3|-2 h — WATER — CORNSYRP -20 0 20 40 60 80 TIME TIME 67 Isotopic Results Dynamic collection of CO2 as CO32" in NaOH was shown to result in fractionation of the carbon isotope (Fig. 8). However, Schoeller et al. (1977) have demonstrated that in NaOH sampling techniques, the isotopic fractionation is highly reproducible and does not affect calculations of excess 1 3 C 0 2 excretion. Resting values for 1 3 C / 1 2 C in expired CO2 were not significantly different (F1 ,5 = 0.01, p= .9395) between the two experimental conditions1, X = -36.695%o(Fig. 8). The treatment effects on 3 1 3 C , plotted as the change in 3 1 3 C from resting values, is shown in Figure 9A. 8 1 3 C fell significantly 5 minutes into exercise (p< .05) for both experimental conditions. Following glucose ingestion (Treatment C), the 1 3 C / 1 2 C in expired CO2 increased above resting values and "exercise baseline" values (Treatment A/B) beginning at 35 minutes. On the contrary, 1 3 C / 1 2 C was not significantly modified without glucose ingestion (Treatment A/B). The analysis of variance of the exercise component showed: i) a significant Time main effect (F5t25 = 39.37, p< .0001), and ii) a significant Treatment x Time interaction (F5(25 = 7.42, p= .0002). Ninety percent of the variance in the Time main effect was accounted for by the linear trend of 9 1 3 C over Time. When averaged over Treatments, there is a significant increase in 3 1 3 C (F-^5 = 68.36, p= .0004) of 1.98%0 between 5 and 75 minutes of exercise (Fig. 9A). Although a plateau effect appears to occur under both conditions, at a 0.01 level of significance, a Time quadratic effect was not significant (F^s = 15.81, p= .0106) when averaged over both treatments. The significant Treatment x Time interaction suggests the nature of change over the 6 collection times differed between the two conditions. This effect 1. The two conditions are Treatment A/B and Treatment C (refer to Chapter 3, Experimental Design, p. 57). Fig. 8. CHANGES IN 913C (X±SD) -32 r WATER CORNSYRUP -20 0 20 40 60 80 TIME (min) Fig. 9A. CHANGES IN 13C/12C FROM REST (X±SD) TIME (min) 70 is especially pronounced after 35 minutes where a change in slope on Treatment C is evident (Fig. 9A). When one subject's data is eliminated, the interaction is more clearly visualized (Fig. 9B) This subject showed higher 3 1 3 C values on both water and bicarbonate treatments than on the cornsyrup treatment. Using trend analysis (n=6), we observe a significant linear interaction (F-^ 5 = 18.80, p= .0075) which accounts for 92% of the variability in the interaction effect. When the best-fitted straight line is drawn for the two conditions, it appears that Treatment A/B shows a gradual linear improvement in 3 1 3 C over 75 minutes, whereas with Treatment C this incline is more sharply positive (Fig. 10). The quadratic interaction effect was not significant (F-\ts = 0.05, p>.05). Fig 11A shows the exogenous glucose oxidation rate in g-15 min"1. There is a sharp rise in the oxidation rate between 30-45 minutes following glucose ingestion, which continues to increase for the remaining exercise period. At the end of exercise, 22.15 g (22%) of the total glucose dose administered had been recovered as 13C02-(Fig. 11B). Sprint Time Mean sprint time increased over the three trials from 184.5 seconds in the first trial to 270.8 seconds in the last trial independent of treatment (Table IV). When corrected for this sequence effect, there was no significant difference (F2J0 = 1.189, p>.05) between the experimental conditions in sprint time. Test of Hypothesis 1 A nonsignificant Treatment x Time interaction for RER (Fig. 7A) does not support Hypothesis 1 which predicted an interacting effect for both RER and 3 1 3c. Fig. 10. BEST-FITTED LINE FOR THE TWO CONDITIONS B WATER • CORNSYRUP I i I i l i 1 1 1 0 20 40 60 80 TIME (min) Fig. 11 A. EXOGENOUS GLUCOSE OXIDATION CORNSYRUP 0 20 40 60 80 TIME (min) TABLE IV. Sprint time responses to experimental conditions (seconds) Subject Trial: 1 2 3 1 265 285 336 2 324 364 608 3 92 125 128 4 246 316 254 5 113 107 160 6 67 116 139 Mean 184.5 218.8 270.8 ±SD 43.7 47.2 74.9 74 Test of Hypothesis 2 The test applied to this hypothesis was a comparison of orthogonal polynomials, following a significant Time main effect, between plasma glucose concentration and 9 1 3 C while on Treatment C. For plasma glucose concentration, the cubic component was significant (Fi ,5 = 43.01, p= .0012), whereas for d 1 3 C the linear component was significant (F-1,5 = 79.49, p= .0003) suggesting the peak effect of the glucose drink differed between the two variables. This supports Hypothesis 2. Test of Hypothesis 3 The test applied to hypothesis 3 was a comparison between RER and 3 1 3 C of the orthogonal polynomials following a significant Time main effect on Treatment C. For RER, none of the polynomials were significant at a 0.01 level of significance; for 3 1 3 C the first polynomial, linear trend, was significant (F1,5 = 79.49, p= .0003). This would suggest that RER adjusted to the glucose load within 15 minutes of exercise while 3 1 3 C showed a more gradual change which continued beyond 75 minutes. This gives support for Hypothesis 3. Test of Hypothesis 4 A significant linear trend (F-^5 = 49.76, p= .0009) for the Time main effect gave support for Hypothesis 4 which predicted an increase in baseline 9 1 3 C during exercise (Fig. 8). 75 CHAPTER 5 DISCUSSION Seventy-five minutes of cycling at a workload corresponding to 60% V02max may be regarded as a moderate submaximal exercise bout for trained subjects. Under these conditions, both muscle glycogen and blood-borne glucose are in a state of flux; that is, blood-borne glucose utilization increases while muscle glycogen utilization decreases from 0 - 75 minutes. Although the contribution of free fatty acids is also important, its proportion of total oxygen uptake by muscle remains relatively constant for this exercise duration (Alborg et al., 1974). The results of this study indicate that ingestion of 275 ml of a 2.02 molar (M) glucose solution during 75 minutes of exercise provides a peak glucose oxidation rate of 4.83 g.15 min"1. It is unclear whether a plateau or decrease in oxidation rate occurs beyond 75 minutes. It is possible that impaired gastric emptying may have lessened the contribution of the exogenous glucose administered. Although it appears that the osmolality of a glucose solution does not affect its metabolic availability during prolonged mild exercise (Jaindrain et al., 1989), with increased exercise intensity drink composition is a strong effector of gastric emptying (Rehrer et al., 1989). Glucose ingestion did not improve sprint time to exhaustion in a performance ride at 90% V02max following the 75 minute cycle test. Sprint time improved for most subjects from the first to last trial independent of the treatment administered. It is possible that knowledge of their previous sprint time was adequate incentive to improve sprint time at subsequent trials. It also suggests that the final sprint ride involved a strong psychological component. Of the total glucose dose administered, 22% was recovered as 1 3 C 0 2 . This is in agreement with the amounts recovered after 75 minutes in a series of studies by Pirnay et al. and Krzentowski et al. (see Table V) following the TABLE V. Exogenous glucose oxidation during exercise : results from published human studies Investigator Protocol %V02max g exogenous glucose oxidized in 75 minutes Gerard (1986) treadmill 50 45 (sucrose) Pirnay(1981) treadmill 45 24.5 Pirnay (1977a) treadmill 50 27.2 Pirnay (1977b) treadmill 50 30.5 Krzentowski (1984a) treadmill 45 34 Krzentowski (1984b) bicycle 30-50 26 Pirnay (1982) treadmill 64 35.6 Present study bicycle 60 22 77 administration of 100 g of naturally labelled 13c-glucose. The work load in these studies corresponded to 45-50% V02max for a treadmill protocol and 40% V02max for a bicycle protocol. When compared to a study by Pirnay et al. (1982), with an exercise intensity adjusted to induce a VO2 close to 64% of the subjects' V02max and a glucose load of 100 g, 13% more of the total glucose dose was recovered, ie.34.5%. Two important differences between their study and the present one were subject selection and mode of exercise. It has been shown that both these factors can influence the measurement of V02max (Bouchard et al., 1979). This would also influence the workloads selected in many of these studies since they are extrapolated from V02max. In a group of untrained men, Bouchard et al. (1979) showed that V02max varied according to the type of maximal exercise performed, where walking uphill on a treadmill to exhaustion elicited the highest V02max. It was concluded that for non-athletes V02max was higher when maximal exercise involved large muscle groups. In trained athletes, V02max is higher when performing the specific sport activity (Stromme et al., 1977). In the studies outlined on Table V, the subjects were described as "healthy male subjects used to practice various sports but without scheduled training". It is quite possible that increased muscle VO2, associated with treadmill exercise in these untrained subjects, was accompanied by a concomitant increase in fuel demand. This demand would be further augmented if muscle glycogen stores were low. In contrast, the present study used athletes specifically adapted to bicycle training. At submaximal workloads, their total muscle VO2 may be lower for two reasons: 1) there is less muscle mass involvement in cycling vs treadmill exercise, and 2) there is increased muscle group specificity with trained subjects. It is also believed that their muscle glycogen stores were not depleted, although this 78 was not determined. One might speculate that with a lower muscle VO2, in conjunction with sufficient muscle glycogen stores, the uptake of glucose by muscle would be less than with a higher muscle VO2 requirement and/or suboptimal muscle glycogen stores. This is demonstrated in the study by Krzentowski et al. (1984b) where only 26% of exogenous glucose was recovered after 75 minutes of cycle exercise. This is less than the mean of 33% reported in studies that used treadmill exercise (see Table V). Alborg and Felig (1976) have shown that ingesting concentrated glucose solutions of 1.66 M during cycling significantly increased glucose uptake and oxidation by the legs accounting for 60% of leg O2 consumption. In the present study ingestion of the glucose solution had no effect on the rate of total carbohydrate utilization as RER values fell throughout the exercise protocol whether or not glucose was consumed (Fig. 7A). Although we did not measure changes in leg VO2, it is quite possible that most of the exogenous glucose oxidized was delivered to the leg muscles. Muscle glycogen content was not measured; consequently, it cannot be determined whether ingestion of the glucose beverage resulted in liver and/or muscle glycogen sparing. Analysis of plasma glucose concentration showed an increased effect following glucose ingestion. However, even 30 minutes after ingestion of the glucose solution, when the contribution of the drink to plasma glucose was highest, the accumulated amount of 1 3C02 recovered represented less than 3% of the total dose administered (see Fig.11B). Clearly the oxidation of blood-borne glucose by muscle is not solely predicated on blood concentration. Dynamic trapping of expired air in NaOH resulted in fractionation of the carbon isotopes. This could have been avoided by having the subject exhale 79 directly into a spiral trap immersed in liquid nitrogen; however, this approach is not feasible for most experimental protocols. When comparing isotope ratios of samples collected alternately in NaOH and a liquid nitrogen-cooled trap, Schoeller et al. (1977) established the magnitude of fractionation to be-13 ±0.8%o. They also compared the amount of CO2 collected per unit time by the two methods and found the trapping efficiency of the NaOH to be between 10-20%. Fortunately, the presence of fractionation does not in itself effect calculations of the amount of excess 13C02 excreted provided variations in the degree of isotope fractionation from one sample to another are small. Schoeller et al. (1977) determined the variability was insignificant (SD=0.17%o) even though the absolute depletion of 13C02 was -13%»in all samples. The cause of fractionation presumably results from the relative isotopic rates of diffusion across the boundary layer of the absorbing NaOH solution (Yemm and Bidwell, 1969). Yemm and Bidwell also reported a consistent preference for 12C02 which caused no serious systematic errors in the isotope measurements. With the 1 3 C technique employed, total body 13C02 production was recovered. It might be argued that recycling of 1 3 C in the form of lipid via lipogenesis or in the form of glucose via the Cori cycle might influence the interpretation of the results. Alborg and Felig (1976) and Wahren (1979) have shown that ingestion of glucose in a large quantity prior to or during exercise results in an increase in insulin level which has an inhibitory effect on lipolysis. This would result in a decrease in plasma FFA concentration with a concomitant increase in exogenous glucose oxidation by the working muscles. It has been demonstrated that there is no significant increase in the splanchnic lactate uptake during exercise preceeded by the ingestion of a glucose load (Alborg and Felig, 1976, 1977). Also, at submaximal workloads, lactic acid would be expected to 80 show little if any change in highly trained subjects. Thus, during exercise, lipogenesis and the recycling of l 3 C via the Cori cycle can be excluded and the 13c enrichment in expired CO2 can be considered a reflection of exogenous glucose oxidation. After oxidation of 1 ^C-glucose, the 1 3C02 produced mixes in the total pool of CO2 contained in blood and bicarbonate. It has been documented that a fraction to the labelled carbon is not recovered due to these slowly exchanging pools (Irving et al., 1983; Winchell et al., 1970). Most models that have been proposed consist of at least one rapidly turning over pool in equilibrium with blood and well-perfused tissues, and more slowly turning over pools that are comprised of less well-perfused tissues such as bone. Based on a number of studies that have measured the recovery of label in expired breath following administration of a bolus bicarbonate tracer, 66.6% was recovered (Table VI). An assumption made is that CO2 released from bicarbonate and CO2 released from oxidation of fuels are handled similarly by the body. The former is a reaction catalyzed by carbonic anhydrase, whereas the latter is predominantly a mitochondrial process. In a recent study by Hoerr er al. (1989), an important finding was that label recovery from bicarbonate tracer is a function of VO2 and VCO2, and as these parameters increase, label recovery also increases. Wolfe et al. (1984a) have shown that during mild exercise at 30% V02max, recovery from labelled bicarbonate infusion increased from 78% at rest to 106% after 20 minutes of exercise. In the same study, simply the change in endogenous substrate metabolism associated with exercise was sufficient to induce a significant increase in the enrichment of expired CO2. They reasoned that during exercise CHO is oxidized at a faster rate relative to fat, and that the naturally occurring enrichment of carbon in CHO is higher than in fat (Jacobson et al., 1970). 81 TABLE VI. Fractional recovery of 13C or 14C in breath from Ingested bicarbonate: results from published human studies Investigator Protocol Metabolic % of dose status recovered Irving (1983) 1 3 C bolus postabsorptive 52 Irving (1984) 1 3 C bolus postabsorptive 74 Issekutz (1968) 1 4 C bolus postabsorptive 79 Yang (1983) 1 3 C bolus postabsorptive 72 1 3 C bolus postabsorptive 56 Mean ±SD 66.6 11.9 82 In the present study, the bicarbonate tracer trial was unsuccessful due to incorrect calculation of the minimal detectable tracer dose (see APPENDIX C). Since VO2 and VCO2 were significantly increased from rest values, and the recovery of bicarbonate tracer is a function of VO2 and VCO2, it is conceivable that 100% recovery of bicarbonate label occurred. In that situation, only 13.14 g (13%) of the administered glucose dose would have been oxidized. More research dealing with bicarbonate recovery during moderate to high intensity exercise is necessary to clarify discrepancies in this area. Unlike the results of Wolfe et al. (1984a), this study did not demonstrate a significant increase in breath 13C02 above rest values with exercise alone. The explanation for this is two-fold. First of all, the subjects reported to the lab two hours following a high CHO snack. Their resting RER when interpreted as RQ is reflective of carbohydrate oxidation. Secondly, with the exercise intensity selected, and using highly trained athletes, lipid may have been the fuel preferentially oxidized and thus we would not see the large increases in breath l3C02 with exercise. Indeed, after the initial increase in 1 3 C enrichment from 5-35 minutes a plateau effect occurs which would coincide with the onset of FFA utilization (Felig and Wahren, 1975). Both RER abd 3 1 3 C showed a decrease early in exercise, followed by either a slow rise in 9 l 3 C or a further gradual decrease in RER. The time course for changes in RER however were much faster than for 3 1 3 C (Fig. 7A and 8). A recent study by Barstow et al. (1989) investigated the time course of changes in RER and d 1 3 C . They suggested that the initial drop in both parameters observed early in exercise is not a reflection of increased lipid oxidation but instead signifies an adjustment from rest to exercise in VO2, VCO2, and 13C02. They summarized that VO2 adjusts slightly more rapidly than VCO2 in the transition from rest to exercise due to a greater storage capacity for CO2 relative to O2, whereas 13C02 83 adjusts the slowest due to redistribution of 1 3 C 0 2 throughout the body C O 2 pools. The practical significance of this lies in the application of both RER and 3 1 3 C over time. During exercise less than 40 minutes duration, changes in 1 3 C enrichment of breath C O 2 will be less likely to reflect the underlying oxidative fuel than will be estimations made from RER. 84 CHAPTER 6 SUMMARY AND CONCLUSIONS Summary The main purpose of this study was to determine the oxidation rate of an exogenous glucose beverage administered during a 75 minute cycle test at 60% V02max. An isotopic tracer technique was employed using naturally labelled 1 3C-glucose. A second purpose was to determine whether the glucose drink improved sprint time to exhaustion at 90% V02max following the 75 minute cycle test. Other problems investigated in this study were: (1) the effects of moderate exercise on baseline 1 3 C abundance in expired air, (2) the changes in blood glucose concentration and respiratory exchange ratio (RER) for the same test, and (3) the recovery of a Na H 1 3 C 0 3 bolus during moderate exercise. Six well-trained male cyclists volunteered as subjects. All subjects were tested under three experimental conditions. The experimental task involved three 75 minute cycle bouts, one week apart, at 60% V02max. All tests were performed on the same apparatus, and at the same time on the same weekday for each subject. After a warm-up, the subjects cycled for 75 minutes at a workload corresponding to 60% V02max, followed by a sprint to exhaustion at 90% V02max. Treatments were administered after 5 minutes of cycling at 60% V02max. The three treatment conditions were (1) water ad lib, (2) NaH 1 2 C03/NaH 1 3 C03 mixture + water ad lib, and (3) 1 3 C -cornsyrup drink + water ad lib. Under all three conditions, respiratory assessment was analyzed calorimetrically, and expired air samples were collected for 1 3 C 0 2 determination. Blood samples were collected under two of the conditions. 85 The results indicated that there was no difference between trials in any of the blood or calorimetric parameters analyzed. Expired 1 3 C 0 2 increased with exercise alone, and was significantly higher than rest values following 1 3 C -glucose ingestion. Peak glucose oxidation was 7.26 g.15 min - 1 after 75 minutes. A plateau or drop-off effect was not observed. Twenty-two percent of the glucose dose administered was recovered after 75 minutes. Sprint times to exhaustion were not affected by the treatments administered. The percent recovery of N a H 1 3 C 0 3 could not be determined due to error in calculation of the minimal detectable dose. Conclusions The investigation suggests the following conclusions: 1. During moderate exercise, exogenous glucose is oxidized within 15-30 minutes of ingestion; however peak oxidation cannot be ascertained within 75 minutes. 2. Changes in plasma glucose concentration do not follow the same time course of exogenous glucose oxidation making it a poor indicator of exogenous glucose utilization. 3. RER adjusts more rapidly than does 3 1 3 C from rest to exercise making it a valuable indicator of substrate oxidation in the early stages of exercise. 4. Baseline breath 1 3 C 0 2 changes with exercise and should be adjusted accordingly in calculations of substrate oxidation. 5. Exogenous glucose did not improve sprint time to exhaustion in this experimental protocol. Suggestions for Further Research Without correcting for the recovery of label, the amount of exogenous glucose recovered after 75 minutes of moderate exercise is =13%. Therefore, the determination of l 3C-label recovery during moderate exercise using NaH 1 3 C03 should be assessed to correctly quantitate glucose oxidation. It is also suggested that this experiment be repeated with a larger sample size before rejecting this practice. 87 REFERENCES Andres, R., G. Cader and K.L. Zieler. The quantitatively minor role of carbohydrate in oxidative metabolism by skeletal muscle in intact man in the basal state: measurement of oxygen and glucose uptake and carbon dioxide and lactate production in the forearm. J. Clin. Invest. 35:671-682, 1956. Alborg, G. and P. Felig. Substrate utilization during prolonged exercise preceded by ingestion of glucose. Am. J. Physiol. 233:E188-E194, 1977. Alborg, G., and P. Felig. Influence of glucose ingestion on fuel-hormone response during prolonged exercise. J. Appl. PhysiolAl: 683-688, 1976. Alborg, G., P. Felig, L. Hagenfeldt, R. Hendler and J. Wahren. Substrate turnover during prolonged exercise in man; splanchnic and leg metabolism of glucose, free fatty acids, and amino acids. J. Clin. Invest. 53:1080-1090, 1974. Barstow, T.J., D.M. Cooper, S. Epstein and K. Wasserman. Changes in breath 13c02 / 1 2 C 0 2 consequent to exercise and hypoxia. J. Appl. Physiol. 66:936-942, 1989. Bergstrom, J., L. Hermansen, E. Hultman and B. Saltin. Diet, muscle glycogen and physical performance. Acta. Physiol. Scand. 71:140-150, 1967. Bjorkman, O., K. Sahlin, L. Hagenfeldt, and J. Wahren. Influence of glucose and fructose ingestion on the capacity for long term exercise in well-trained man. Clin. Physiol. Oxf. 4:483-494, 1984. Bjorkman, O., and J. Berry. High-efficiency photosynthesis. Sci. Amer. 229:80-93, 1973. Bouchard, C , P. Godbout, J.-C. Mondor and C. Leblanc. Specificity of maximal aerobic power. Eur. J. Appl. Physiol. 40:85-93, 1979. Butterfield, G.E. and D.H. Calloway. Physical activity improves protein utilization in young men. Br. J. Nutr. 51:171-184, 1984. Calvin, M. The path of carbon in photosynthesis. Science 135:879-889, 1962. Costill, D.L. Carbohydrates for exercise: dietary demands for optimal performance. Int. J. Sports Med. 9:1-18, 1988. Costill, D.L, and J.M. Miller. Nutrition for endurance sport: carbohydrate and fluid balance. Int. J. Sports Med. 1:2-14, 1980. Costill, D.L., E. Coyle, G. Dalsky, W. Evans, W. Fink and D. Hoopes. Effects of elevated plasma FFA and insulin on muscle glycogen usage during exercise. J. Appl. Physiol. 43:695-699, 1977. 88 Costill, D.L and W.J. Fink. Plasma volume changes following exercise and thermal dehydration. J. Appl. Physiol. 37:912-916, 1974. Costill, D.L. and B. Saltin. Factors limiting gastric emptying during rest and exercise. J. Appl. Physiol. 37:679-683, 1974. Costill, D.L, A. Bennett, G. Branam and D. Eddy. Glucose ingestion at rest and during prolonged exercise. J. Appl. Physiol. 34:764-769, 1973. Costill, D.L., K. Sparks, R. Gregor, and C. Turner. Muscle glycogen utilization during exhaustive running. J. Appl. Physiol. 31:353-356, 1971. Coyle, E.F., A.R. Coggan, M.K. Hemmert and J.L. Ivy. Muscle glycogen utilization during prolonged strenuous exercise when fed carbohydrate. J. Appl. Physiol. 61:165-172, 1986. Coyle, E.F., J.M. Hagberg, B.F. Hurley, W.H. Martin, A. Ehsani, and J.O. Holloszy. Carbohydrate feeding during prolonged strenuous exercise can delay fatigue. J. Appl. Physiol. 55:230-235, 1983. Craig, H. Isotopic standards for carbon and oxygen and correction factors for mass-spectrometric analysis of carbon dioxide. Geochim. Cosmochm. Acta. 12:133-149, 1957. Dohm, G.L, R.T. Beeker, R.G. Israel and E.B. Tapscott. Metabolic responses to exercise after fasting. J. Appl. Physiol. 61:1363-1368, 1986. Dohm, G.L., E.B. Tapscott, H.A. Barakat and G.J. Kasperek. Influence of fasting in rats on glycogen depletion during exercise. J. Appl. Physiol. 55:830-833, 1983. Erickson, M.A., R.J. Schwarzkopf and R.D. McKenzie. Effects of caffeine, fructose, and glucose ingestion on muscle glycogen utilization during exercise. Med. Sci. Sports Exerc. 19:579-583, 1987. Felig, P., H. Cheriff, A. Minagawa, and J. Wahren. Hypoglycemia during prolonged exercise in normal men. A7. Engl. J. Med. 306:895-900, 1982. Felig, P., and J. Wahren. Fuel homostasis in exercise. N. Engl. J. Med. 293:1078-1084, 1975. Felig, P. and J. Wahren. Influence of endogenous insulin secretion on splanchnic glucose and amino acid metabolism in man. J. Clin. Invest. 50:1702-1711, 1971. Flynn, M.G., D.L. Costill, J.A. Hawley, W.J. Fink, P.D. Neufer, R.A. Fielding and M.D. Sleeper. Influence of selected carbohydrate drinks on cycling performance and glycogen use. Med. Sci. Sports Exerc. 19:37-40, 1987. Fox, E.L. and D.K. Mathews. The Physiological Basis of Physical Education and Athletics, Third Edition. Saunders College Publishing. New York, NY, 1981. 89 Galbo, H. and M.Kjaer. Endocrinology and metabolism in exercise: future research directions. Can. J. Spt. Sci. 12(Suppl 1):102S-107S, 1987. Gerard, J., B. Jandrain, F. Pirnay, N. Pallikarakis, G. Krzentowski, M. Lacroix, F. Mosora, A.S. Luyckx and P.J. Lefebvre. Utilization of oral sucrose load during exercise in humans. Diabetes 35:1294-1301, 1986. Gollnick, P.D. Metabolic regulation in skeletal muscle: influence of endurance training as exerted by mitochondrial protein concentration. Acta. Physiol. Scand. 128(Suppl 556):53-66, 1986. Hachey, D.L, W.W. Wong, T.W. Boutton and P.D. Klein. Isotope ratio measurements in nutrition and biomedical research. Mass Spectrom. Rev. 6:289-328, 1987. Hargreaves, M. and C.A. Briggs. Effect of carbohydrate ingestion on exercise metabolism. J. Appl. Physiol. 65:1553-1555, 1988. Hatch, M.D.and C.R. Slack. Photosynthetic CC-2-fixation pathways. Ann. Rev. Plant Physiol. 21:141-162, 1970. Hoerr, R.A., Y.M. Yu, D.A. Wagner, J.F. Burke and V.R. Young. Recovery of 1 3 C in breath from N a H l 3 C 0 3 infused by gut and vein: effect of feeding. Am. J. Physiol. 257:E426-E438, 1989. Irving, C.S., W.W. Wong, R.M. Wong, T.W. Boutton, R.J. Shulman, C.L Lifschitz, E.W.Malphus, H. helge and P.D. Klein. Rapid determination of whole-body bicarbonate kinetics by use of a digital infusion. Am. J. Physiol. 247:R709-R716, 1984. Irving, C.S., W.W. Wong, R.J. Shulman, E. O'Brian Smith and P.D. Klein. [13C]bicarbonate kinetics in humans; intra- vs interindividual variations. Am. J. Physiol. 245:R190-R202, 1983. Issekutz, Jr., B., P. Paul, H.I. Miller and W.M. Bortz. Oxidation of plasma FFA in lean and obese humans. Metabolism 17:62-73, 1968. Ivy, J .L, W. Miller, V. Dover, L G . Goodyear, W.M. Sherman, S. Farrel, and H. Williams. Endurance improved by ingestion of a glucose polymer supplement. Med. Sci. Sports Exercise 15:466-471, 1983. Ivy, J .L, D.L. Costill, W. Fink and R.W. Lower. Influence of caffeine and carbohydrate feedings on endurance performance. Med. Sci. Sports Exerc. 11:6-11, 1979. Jacobson, B.S., B.N. Smith, S. Epstein and G.C. Laties. The prevalence of carbon" 1 3 in respiratory carbon dioxide as an indicator of the type of endogenous substrate. J. Gen Physiol. 55:1-17, 1970. 90 Jandrain, B.J., F. Pirnay, M. Lacroix, F. Mosora, A.J. Scheen and P.J. Lefebvre. Effect of osmolality on availability of glucose ingested during prolonged exercise in humans. J. Appl. Physiol. 67:76-82, 1989. Jones, P.J.H., P.B. Pencharz, L. Bell and M.T. Clandinin. Model for determination of 1 3 C substrate oxidation frates in humans in the fed state. Am. J. Clin. Nutr. 41:1277-1282, 1985. King, D.S., G.P. Dalsky, W.E. Clutter, D.A. Young, M.A. Staten, P.E. Cryer and J.O. Holloszy. Effects of exercise and lack of exercise on insulin sensitivity and responsiveness. J. Appl. Physiol. 64:1942-1946, 1988. Klein, P.D. and E.R.Klein. Applications of stable isotopes to pediatric nutrition and gastroenterology: measurement of nutrient absorption and digestion using 1 3c. J. Pediatr. Gastroenteral. Nutr. 4:9-19, 1985. Klein, S. and R.R. Wolfe. The use of isotopic tracers in studying lipid metabolism in human subjects. Baill. Clin. Endocrinol. Met. 1:797-816, 1987. Knapik, J.J., C.N. Meredith, B.H. Jones, L. Suek, V.R. Young and W.J. Evans. Influence of fasting on carbohydrate and fat metabolism during rest and exercise in men. J. Appl. Physiol. 64:1923-1929, 1988. Krzentowski, G., B. Jandrain, F. Pirnay, F. Mosora, M. Lacroix, A.S. Luyckx and P.J. Lefebvre. Availability of glucose given orally during exercise. J. Appl. Physiol. 56:315-320, 1984a. Krzentowski, G., F. Pirnay, A.S. Luyckx, M. Lacroix, F. Mosora and P.J. Lefebvre. Effect of physical training on utilization of a glucose load given orally during exercise. Am. J. Physiol. 246:E412-E417, 1984b. Lacroix, M., F. Mosora, M. Pontus, P.L efebvre, A. Luyckx and G. Lopez-Habib. Glucose naturally labeled with carbon-13; use for metabolic studies in man. Science 181:445-446, 1973. Lagua, R.T., V.S. Claudio and V.F. Thiele. Nutrition and Diet Therapy: Reference Dictionary, Second Edition. The C.V Mosby Co., St. Louis, MO, 1974. Loy, S.F., R.K. Conlee, W.W. Winder, A.G. Nelson, D.A. Arnall and G. Fisher. Effects of 24-hour fast on cycling endurance time at two different intensities. J. Appl. Physiol. 61:654-659, 1986. McArdle, W.D., F.I. Katch and V.L. Katch. Exercise Physiology, Second Edition. Lea & Febiger, Philadelphia, PA, 1986. Massicotte, D., F. Peronnet, G. Brisson, K. Bakkouch and C. Hillaire-Marcel. Oxidation of a glucose polymer during exercise: comparison with glucose and fructose. J. Appl. Physiol. 66:179-183, 1989. 91 Massicotte, D., F. Perronnet, C. Allah, C. Hillaire-Marcel, M. Ledoux and G. Brisson. Metabolic response to [13C]glucose and [13C]fructose ingestion during exercise. J. Appl. Physiol. 61:1180-1184, 1986. Medical Graphics Corporation. User Manual for Medical Graphics Corporation Exercise System 2001. Revision A. St. Paul, MN, 1985. Millward, D.J., C.T.M. Davies, D. Halliday, S.L Wolman, D. Matthews and M. Rennie. Effect of exercise on protein metabolism in humans as explored with stable isotopes. Federation Proc. 41:2686-2691, 1982. Milne, E. and B.A. McGaw. The applicability of evacuated serological tubes for the collection of breath for isotopic analysis of CO2 by isotope ratio mass spectrometry. Biomed. Environ. Mass Spectrom. 15:467-472, 1987. Mitchell, J.B., D.L. Costill, J.A. Houmard, M.G. Flynn, W.J. Fink and J.D Beltz. Effects of carbohydrate ingestion on gastric emptying and exercise performance. Med. Sci. Sports Exerc. 20:110-115, 1988. Mosora, F.,M. Lacroix, A. Luyckx, N. Pallikarakis, F. Pirnay, G. Krzentowske and P. Lefebvre. Glucose oxidation in relation to the size of the oral glucose loading dose. Metabolism 30:1143-1149, 1981. Mosora, F., P. Lefebvre, F. Pirnay, M. Lacroix, A. Luyckx and J. Duchesne. Quantitative evaluation of the oxidation of an exogenous glucose load using naturally labeled 1 3C-glucose. Metabolism 25:1575-1582, 1976. Nakamura, K., D.A. Schoeller, F.J. Winkler and H.-L. Schmidt. Geographical variations in the carbon isotope composition of the diet and hair in contemporary man. Biomed. Mass Spectrom. 9:390-394, 1982. Neufer, P.D., D.L. Costill, W.J. Fink, J.P. Kirwan, R.A. Fielding and M.G. Flynn. Effects of exercise and carbohydrate composition on gastric emptying. Med. Sci. Sports Exerc. 18:658-662, 1986. Park, R., S. Epstein. Metabolic fractionation of C 1 3 & C 1 2 in plants. Plant Physiol. 36:133-138, 1961. Pennington, J.A. Bowes & Church's Food Values of Portions Commonly Used, 15th Edition. J. B. Lippincott Co., Philadelphia, PA, 1987. Pirnay, F., J.M. Crielaard, N. Pallikarakis, M. Lacroix, F. Mosora, G. Krzentowski, A. Luyckx and P. Lefebvre. Fate of exogenous glucose during exercise of different intensities in humans. J. Appl. Physiol. 53:1620-1624, 1982. Pirnay, F., G. Krzentowski, J.M. Crielaard, N. Pallikarakis, M. Lacroix, F. Mosora, A. Luyckx and P. Lefebvre. Oxidation of orally administered naturally labeled [13C]glucose during prolonged muscular exercise. In Biochemistry of Exercise, Volume 4, Part A. Poortsmans, J. and G. Niset (eds), 1981. 92 Pirnay, F., M. Lacroix, F. Mosora, A. Luyckx and P. Lefebvre. Glucose oxidation during prolonged exercise evaluated with naturally labelled [13C]glucose. J. Appl. Physiol. 43:258-261, 1977a. Pirnay, F., M. Lacroix, F. Mosora, A. Luyckx and P. Lefebvre. Effect of glucose ingestion on energy substrate utilization during prolonged muscular exercise. Europ. J. Appl. Physiol. 36:247-254, 1977b. Ploug, T., H. Galbo and E.A. Richter. Increased muscle glucose uptake during contractions: no need for insulin. Am. J. Physiol. 247:E726-E731, 1984. Ravussin, E., P. Pahud, A. Dorner, M.J. Arnaud and E. Jequier. Substrate utilization during prolonged exercise preceded by ingestion of 1 3C-glucose in glycogen depleted and control subjects. Pfugers Arch. 382:197-202, 1979. Rehrer, N.J., E. Beckers, F. Brouns, F.Ten Hoor and W.H.M. Saris. Exercise and training effects on gastric emptying of carbohydrate beverages. Med. Sci. Sports Exerc. 21:540-549, 1989. Richter, E.A. Influence of hormones and exercise on metabolism in isolated muscle. Can. J. Spt Sci. 12(Suppl 1):108S-112S. 1987. Richter, E.A. and H. Galbo. High glycogen levels enhance glycogen breakdown in contracting, isolated perfused muscle. J. Appl. Physiol. 61:827-831, 1986. Robert, J.J., J. Koziet, D. Chauvet, D. Darmaun, J.F. Desjeux and V.R. Young. Use of 1 3C-labeled glucose for estimating glucose oxidation: some design considerations. J. Appl. Physiol. 63:1725-1732, 1987. Saltin, B. and J. Karlsson. Muscle glycogen utilization during work of different intensities. In Pernow, B. and B. Saltin (eds.): Muscle Glycogen During Exercise, pp. 289-299. Plenum, New York, NY, 1971. Schoeller, D.A., C. Brown, N. Nakamura, A. Nakagawa, R.S. Mazzeo, G.A. Brooks and T.F. Budinger. Influence of metabolic fuel on the 1 3 C / 1 2 C ratio of breath C02- Biomed. Mass Spectrom. 11:557-561, 1984. Schoeller, D.A., P.D. Klein, J.B. Watkins, T. Heim and W.C. MacLean, Jr. 1 3 C abundances of nutrients and the effect of variations in 1 3 C isotope abundances of test meals formulated for 1 3 C 0 2 breath tests. Am. J. Clin. Nutr. 33:2375-2385, 1980. Schoeller, D.A., and P.D. Klein. A simplified technique for collecting breath CO2 for isotope ratio mass spectrometry. Biomed. Mass Spectrom. 5:29-31, 1978. Schoeller, D.A., J.F. Schneider, N.W. Solomons, J.B. Watkins and P.D. Klein. Clinical diagnosis with the stable isotope 1 3 C in CO2 breath tests: methodology and fundamental considerations. J. Lab. Clin. Med. 90:412-421, 1977. 93 Selkurt, E.E.(Ed). Basic Physiology for the Health Sciences. Little, Brown and Co., Boston, MA, 1975. Shaw, J.H.F. and R.R. Wolfe. Determination of glucose turnover and oxidation in normal volunteers and septic patients using stable and radio-isotopes: the response to glucose infusion and total parenteral feeding. Austr. N.Z. J. Surg. 56:785-791, 1986. Shaw, J.H.F. and R.R. Wolfe. Glucose-FFA interactions during glucose and/or lipid infusions (Abstract). Fed. Proc. 42:673, 1983. Sigma Diagnostics. Lactate. St. Louis MO, 1988. Sigma Diagnostics. Glucose. St. Louis, MO, 1987. Slanger, B.H., N Kusubov and H.S. Winchell. Effect of exercise on human CO2-HCO3- kinetics. J. Nucl. Med. 11:716-718, 1970. Smith, B.N. and S. Epstein. Two categories of 1 3 C / 1 2 C ratios for higher plants. Plant Physiol. 47:380-384, 1971. Steele, R. The retention of metabolic radioactive carbonate. Biochemistry 60:447'-453, 1955. Stromme, S.B., F. Ingjer and H.D. Meen. Assessment of maximal aerobic power in specifically trained athletes. J. Appl. Physiol. 42:833-837, 1977. Van Handel, P.J., W.J. Fink, G. Branam and D.L. Costill. Fate of 1 4 C glucose ingested during prolonged exercise. Int. J. Sports Med. 1:127-131, 1980. Wahren, J. Glucose turnover during exercise in healthy man and in patients with diabetes mellitus. Diabetes 28(Suppl 1):82-88, 1979. Wahren, J., P. Felig, G. Ahlborg, and L. Jorfeldt. Glucose metabolism during leg exercise in man. J. Clin. Invest. 50:2715-2725, 1971. Wheeler, K.B. and J.G. Banwell. Intestinal water and electrolyte flux of glucose-polymer electrolyte solutions. Med. Sci. Sports Exerc. 18:436-439, 1986. Winchell, H.S., H. Stahelin, N. Kusubov, B. Slanger, M. Fish, M. Pollycove and J.H. Lawrence. Kinetics of CO2-HCO3" in normal adult males. J. Nucl. Med. 11:711-715, 1970. Wolfe, R.R. Tracers in Metabolic Research: Radioisotope and Stable Isotope/Mass Spectrometry Methods. Alan R. Liss, Inc., New York, NY, 1984. Wolfe, R.R., M.H. Wolfe, E.R. Nadel and J.H.F. Shaw. Isotopic determination of amino acid-urea interactions in exercise in humans. J. Appl. Physiol. 56:221-229, 1984a. 94 Wolfe, R.R., J.H.F. Shaw, E.R. Nadel and M.H. Wolfe. Effect of substrate intake and physiological state on background 13CC>2 enrichment. J. Appl. Physiol. 56:230-234, 1984b. Yang, R.D., C.S. Irving, W.W. Wong, J.H. Hoffer, V.R. Young and P.D. Klein. The effect of diet and meal ingestion on whole body 1 3 C bicarbonate kinetics in young men (Abstract). Federation Proc. 42:825A, 1983. Yemm, E.W. and R.G.S. Bidwell. Carbon dioxide exchanges in leaves. I. Discrimination between 1 4 C 0 2 and 1 2 C 0 2 in photosynthesis. Plant Physiol. 44:1328-1334, 1969. 95 APPENDIX A. Instructions to Subjects 1 3C-glucose Study This Study involves a stable isotope, carbon-13 ( 1 3 C). It is important to avoid food sources that are rich in carbon-13 for at least 3 days prior to each trial. The two more common sources of carbon-13 found in B.C. are cane sugar and maize (ie. corn). FOODS TO AVOID which contain these sources include: - Sugar (all kinds) and foods high in sugar such as chocolate, candies, condensed milk, jams, jellies, molasses, syrups, regular soft drinks*, doughnuts, pastries, iced cakes, Jello® NB. You will be given a source of beet sugar which you can consume ad lib. - Corn, including creamed corn, corn-based soups, sweet corn, baby corn, and corn products such as corn syrup, cornmeal, corn chips, tacos and taco chips, maize, Corn Flakes®, Corn Bran® * Diet soft drinks are allowed To ensure that dietary consistency exists between each trial, a 3-day food record is requested prior to Trial 1. A copy of this food record will be returned, and you will be asked to follow the same type of diet for 3 days prior to the subsequent 2 trials. Use the provided worksheets to list EVERYTHING YOU ATE OR DRANK for EACH DAY to be recorded. Be sure to include: 1) ALL FOOD AND DRINKS, for example: snacks; beet sugar (or honey) and cream (or milk) in coffee; sandwich fillings; types of different vegetables; Diet soft drinks; alcohol; etc. 2) THE AMOUNT OF FOOD that you ate (ounces, slices, cups, teaspoons, etc. in whole numbers or fractions). 3) Please record the NUMBER OF HOURS SPENT TRAINING for EACH DAY recorded. NOTE: it is important that you consume a 75 g CHO meal 2 hours PRIOR TO EACH TRIAL. I will work on a suitable meal choice with you. An example of such a meal is: Sandwich (with 2 slices of bread) 8 oz. fruit juice 1 apple (or 1 muffin) Please feel free to call me if you have any questions or concerns (731-5402). 3-DAY FOOD AND ACTIVITY RECORD DATE: ACTIVITY: LENGTH OF TIME: FOOD/BEVERAGE INTAKE: TIME OF DAY AMOUNT FOOD OR BEVERAGE APPENDIX B. Laboratory Procedures DETERMINATION OF PLASMA LACTATE CONCENTRATION A. Deproteination of Sample 1. Thaw frozen plasma. 2. Recentrifuge plasma for 10-12 minutes at 2900 rpm, 4°C. 3. Meanwhile, label the required TEST cuvet(s). 4. Pump some trichloroacetic acid (TCA) into a spare test tube. Keep cold. 5. Into each labelled cuvet, pipet 0.5 ml of TCA. 6. QUICKLY pipet 0.25 ml (250/J) of spinned plasma in the TCA-containing cuvet(s). Vortex for 30 seconds. 7. Refrigerate for 5 minutes, then centrifuge for 12 minutes at 2900 rpm, 4°C. B. Lactate Procedure 1. Label the required TEST and BLANK cuvets. 2. Reconstitute the appropriate number of NAD vials needed1 by pipeting into each: 2.0 ml Glycine Buffer 4.0 ml Water 0.1 ml (100/J) Lactate dehydrogenase Cap and invert vials several times to dissolve the NAD. Combine contents of vials if more than one are being used. 3. Pipet 1.4 ml of this solution into BLANK and TEST cuvets. 4. To BLANK cuvet add: 0.1 ml (100/*\) of TCA. Mix by gentle inversion. 5. To TEST cuvets add: 0.1 ml (100 /A) of supernatant fluid (from Procedure A) Mix by gentle inversion. 6. Allow BLANK and TEST cuvets to stand at room temp for more than 30 minutes. 7. Read absorbance of BLANK and TEST cuvets at 340 nm. 8. Calculate directly from absorbance readings as: Blood Lactate (mmol/L) = AA340 x 7.23, or, Blood Lactate (mg/dl) = AA340 x 65.1 where AA340= (TEST absorbance - BLANK absorbance). NOTE: Multiply result by two since only half of the required NAD-containing solution was used, ie. 1.4 ml instead of 2.8 ml. 1To calculate the amount needed, determine the total number of samples (N) to be analyzed. Multiply by 1.5, then divide by 6. eg. N=15 (15 x 1.5) + 6 = 3.75 Therefore, 4 vials are needed. 98 DETERMINATION OF PLASMA GLUCOSE CONCENTRATION A. REAGENT PREPARATION 1. ENZYME SOLUTION: Add contents of 1 PGO enzyme capsule to 100 ml of distilled water. Stir gently to dissolve. Store in an amber bottle. Keep refrigerated at 2-5°C. Stable for up to 1 month unless turbidity develops. 2. COLOR REAGENT SOLUTION: Reconstitute o-dianisidine diHCI vial with 20 ml distilled water. Stable for 3 months refrigerated at 2-6°C. WARNING: Harmful if swallowed, inhaled or absorbed through skin. AVOID ALL CONTACT. CARCINOGEN. Causes irritation. 3. COMBINED ENZYME-COLOR REAGENT SOLUTION: Combine 100 ml of ENZYME SOLUTION and 1.6 ml of COLOR REAGENT SOLUTION. Mix by inverting several times. Solution is stable for up to 1 month refrigerated at 2-6°C unless turbidity or color forms. B. GLUCOSE PROCEDURE 1. Thaw frozen plasma. 2. Recentrifuge plasma for 10-12 minutes at 2900 rpm, 4°C. 3. Meanwhile, label 3 or more tubes (16 per trial) as follows: BLANK, STANDARD, TEST 1 A, TEST 1B, TEST 2A, TEST 2B 4. To BLANK add: 0.5 ml water To STANDARD add: 25/A (0.025 ml) of STANDARD GLUCOSE SOLUTION + 0.5 ml water. (20-fold dilution) To each TEST add: 25 >l (0.025 ml) of plasma + 0.5 ml water. (20-fold dilution) 5. To each tube add 2.5 ml of COMBINED ENZYME-COLOR REAGENT SOLUTION and vortex each sample for 15 sec. 6. Incubate all tubes at room temperature (18-26°C) for 45 minutes. NOTE: Avoid exposure to direct sunlight or bright daylight. (Store in cupboard). 7. Read absorbance (A) of all tubes at 450 nm. NOTE: Readings should be completed within 30 minutes. 8. Calculate TEST values as follows: Serum glucose (mg/dl) = ATEST - ABLANK X 100 ASTD - ABLANK NOTE: Multiply result by two since only half of the required COMBINED ENZYME-COLOR REAGENT was used, ie. 2.5 ml instead of 5.0 ml. 99 C02 EXTRACTION collector tube sample. • IOO'C 1. Open all valves to evacuate the entire system. Pressure should be below 50 mTorr. 2. Remove stop-cock from the Rittenberg tube. 3. Add 7 drops of orthophosphoric acid to one side of the tube. 4. Add 5 drops of sample trapped in 1.0 N NaOH to the other side of the tube. 5. Replace stop-cock and close the SAMPLE VALVE. Heat the stop-cock to help create a good seal. Remove any moisture by wiping inside the top section of tube. 6. With VALVES 1 & 2 closed, add the sample tube to the apparatus. 7. Open VALVES 1 & 2 and wait until pressure drops below 80 mTorr. 8. Meanwhile, freeze sample with liquid N2 for approximately 1 minute. 9. Keeping the sample tube frozen in liquid N2, evacuate the sample tube by opening the SAMPLE VALVE. Wait until pressure drops below 80 mTorr. 10. Close theSAMPLE VALVE, then close VALVES 1 & 2. Remove sample and replace with a blank collector tube. Reopen VALVES 1 & 2 to evacuate the sample line. 11. Heat sample tube with hands, then react the phosphoric acid with the sample KEEP SAMPLE VALVE CLOSED. 12. Close VALVES 1 & 2, remove blank collector tube and add the sample tube on the vacuum apparatus. Open all valves except for the SAMPLE VALVE to evacuate the entire system. Wait until the pressure drops below 80 mTorr. 13. Close VALVES 1 & 5 and set up the water and CO2 traps. NB. The water trap is a methanol/liquid N2 syrup, and the CO2 trap is liquid N2. 14. Open the SAMPLE VALVE and let the CO2 transfer run for 2 minutes. NB. This step freezes the CO2 in the liquid N2 trap. Open VALVE 5 for 1 minute to eliminate other gases. 15. Close VALVES 3 & 5 to localize the trapped CO2. 16. Remove both traps. Close VALVE 2 and the SAMPLE VALVE to prevent moisture in the water trap or sample tube from entering the vacuum apparatus. 17. Freeze the collector tube in liquid N2 for 2 minutes. Heat the CO2 trap to ensure all CO2 in the sample is transferred. 18. Open VALVE 5 to evacuate any air and quickly pinch off collector tube using a blow torch. NB. Keep collector tube frozen in liquid N2 throughout this step. 19. Label collector tube appropriately. 100 APPENDIX C. Recovery of Breath 1 3 C 0 2 Using N a H 1 3 C 0 3 Any isotopic effects from the NaH 1 3C03/NaH 1 2C03 mixture were masked by the effects of exercise alone (75 minutes, 60%VO2max) on 1 3 C02 production. To illustrate this, subject CF will serve as example: A. Calculation of the moles of 1 3 C02 produced with exercise alone from 5 to 50 minutes. It was postulated that the dose of NaH 1 3C03 administered would be fully recovered within 45 minutes of ingestion, ie. after 50 minutes of exercise. Using the water trial (EXPERIMENTAL CONDITION A), V C 0 2 at Time= 20, 35 and 50 minutes is 2,998, 2,916 and 3.101 l-min-1, respectively. Converting to molesmin -1 and multiplying by 15 minutes (the number of minutes in each collection period) gives the moles of C O 2 produced in each collection period: Time VC02(l/min) VC02(mole/min) 5-20 2.998 0.1338 20-35 2.916 0.1302 35-50 3.101 0.1384 moles C02/15min 2.0076 1.9527 2.0766 The 3 1 3 C value for each collection period is converted to Ru (the ratio of 1 3 C : 1 2 C in each breath sample). Ru is then multiplied by moles C02/15min for each collection period to obtain moles 13C02/15min. Summing the later values gives the total moles 1 3 C02 produced with exercise alone in the time interval 5 to 50 minutes: Time d 1 3 C 5-20 -36.707 20-35 -36.594 35-50 -36.250 Ru .0108245 .0108258 .0108297 moles C02/15min 2.0076 1.9527 2.0766 moles 1 3C0 2/15min .0217313 .0211395 •0224889 .0653597 B. Calculation of moles 1 3 C given in NaH 1 3C03/NaH 1 2C03 mixture. 5 mg/kg of the mixture was administered. For CF, this was equivalent to 0.3725g. Converting this value to moles and multiplying by Ru of the mixture gives the number of moles of 1 3 C administered. .3725q bicarbonate x 0.12100175 = 0.000536 moles 1 3 C 83.99587 g/mole C. Amount of 1 3 C administered relative to amount of 1 3 C02 produced: .0005366 = .00821 -0.8% .0653597 101 D. Calculation of moles 1 3 C02 produced with exercise alone from 35 to 75 minutes. From Fig. 9A the oxidation rate of the exogenous glucose beverage was significantly increased after 35 minutes. Infact, most was oxidized from 35 to 75 minutes of exercise. As in (A), the water trial allows calculation of the moles of 1 3CC*2 produced with exercise alone in this time interval. V C O 2 is used to calculate the moles of C O 2 produced: Time VC02(l/min) VC02(mole/min) 35-50 3.101 0.1384 50-65 2.904 0.1296 65-75 3.262 0.1456 *NB. The last time interval is moles CCVIOmin. moles C02/15min" 2.0766 1.9446 1.4560 As in (A), convert 3 1 3 C to Ru, multiply by moles CC>2/15min and sum to obtain total moles 1 3 C02 produced from 35 - 75 minutes: Time 3 1 3 C Ru moles C02/15min* moles 1 3C0 2/15min* 35-50 -36.250 .0108297 2.06766 .0224889 50-65 -35.728 .0108355 1.9446 .0210708 65-75 -35.461 .0108385 1.4560 .0157809 S .0593406 kNB. The last time interval is moles C02(or 13CO2)/10min. E. Calculation of moles 1 3 C given in 100g cornsyrup beverage. Convert 100g to moles and multiply by Ru to obtain moles 1 3 C given: 100g cornsvrup x .011113266 = .006174 moles 1 3 C 180g/mole F. Amount of 13C-cornsyrup administered relative to amount of 1 3 C02 produced: .0061740 = .10404 =10.4% .0593406 From the above calculations, the amount of 13C-bicarbonate administered relative to the amount of 1 3CC>2 produced from exercise alone was =1/10 less than the amount of 13C-cornsyrup administered relative to the amount of 1 3 C02 produced from exercise. Giving 10 times more NaH 1 3C03 would have alleviated this problem. APPENDIX D. Raw Data TREATMENT A (water ad lib) Subj Time 9 1 3 C vco 2 V O 2 RER [giu] [lact] [Hb] Hct (min) (%4 (l/min) (l/rrin) (mg/dl) (mmol/L) (mg/dl) -10 -35.150 0.407 0.419 0.971 141.8 1.75 150 .430 5 -37.209 2.923 2.776 1.05 153.6 2.70 20 -36.707 2.998 2.739 1.09 159.9 2.71 35 -36.594 2.916 2.811 1.04 164.7 3.03 50 -36.250 3.101 3.095 1.00 142.0 2.44 65 -35.728 2.904 2.918 0.995 126.1 2.65 75 -35.461 3.262 3.049 1.07 116.9 2.44 164 .483 -10 -35.554 0.499 0.437 1.14 198.3 2.62 131 .388 5 -37.000 3.074 3.249 0.946 125.7 2.22 20 -36.403 3.096 3.232 0.958 143.8 2.36 35 -35.800 2.939 3.062 0.960 145.8 2.02 50 -35.770 2.690 2.920 0.921 178.8 2.06 65 -35.520 2.979 3.229 0.922 126.3 1.90 75 -35.325 2.829 3.065 0.923 185.9 2.52 139 .408 -10 -36.256 0.693 0.506 1.37 109.4 0.78 133 .392 5 -37.362 3.900 3.590 1.09 173.5 2.34 20 -37.226 3.910 3.730 1.05 178.2 2.70 35 -36.496 3.735 3.595 1.04 200.9 4.40 50 -36.744 3.930 3.760 1.04 199.5 4.87 65 -36.331 4.040 3.920 1.03 200.0 4.89 75 -36.458 4.100 4.120 0.995 207.5 4.99 144 .410 -10 -35.700 0.589 0.478 1.23 144.6 2.41 133 .394 5 -37.940 2.796 3.030 0.923 136.3 2.27 20 -37.490 2.884 2.963 0.973 153.4 1.34 35 -37.070 2.730 3.050 0.895 135.5 1.52 50 -37.010 2.639 3.039 0.868 191.2 2.23 65 -37.100 2.725 3.007 0.906 187.9 1.81 75 -36.990 2.858 3.079 0.928 191.7 2.62 139 .412 -10 -34.660 0.780 0.478 1.63 124.6 1.46 145 .417 5 -36.034 3.530 3.310 1.07 117.0 1.84 20 -36.196 3.040 2.920 1.04 100.6 1.30 35 -35.289 3.580 3.580 1.00 131.4 1.97 50 -34.503 3.610 3.690 0.978 103.0 2.43 65 -34.738 3.750 3.790 0.989 96.2 1.65 75 -34.630 4.010 3.920 1.02 161.4 4.30 162 .471 -10 -35.433 0.459 0.492 0.932 132.2 1.50 151 .445 5 -37.107 3.070 3.062 1.00 123.1 2.05 20 -36.584 3.054 3.269 0.934 109.7 1.70 35 -35.777 3.245 3.514 0.923 130.6 1.97 50 -35.978 3.404 3.616 0.941 120.7 2.43 65 -36.098 3.145 3.521 0.893 120.0 1.65 75 -36.242 3.466 3.740 0.927 130.0 4.30 155 ,450 Sprint (sec) JD JG JL JM TC 265 324 92 246 113 67 103 TREATMENT B (NaH 1 3 C03/NaH 1 2 C03 mixture + water ad lib) CF JD JG JL JM TC Time d13C V C O 2 V O 2 RER (min) (l/min) (l/min) -10 -36.659 0.290 0.272 1.07 5 -38.363 2.830 2.820 1.00 20 -36.410 2.610 2.737 0.954 35 -35.132 2.890 3.017 0.958 50 -37.192 2.690 2.898 0.928 65 -36.716 2.705 2.962 0.913 75 -36.264 2.820 3.130 0.901 -10 -35.788 0.508 0.502 1.01 5 -37.737 3.130 3.146 0.995 20 -36.492 3.106 3.220 0.965 35 -35.663 3.040 3.182 0.955 50 -36.202 3.135 3.320 0.944 65 -36.430 3.492 3.641 0.959 75 -36.524 3.198 3.470 0.922 -10 -36.519 0.591 0.438 1.35 5 -37.730 3.630 3.550 1.02 20 -35.993 3.800 3.760 1.01 35 -36.242 3.640 3.750 0.971 50 -35.054 3.600 3.790 0.950 65 -36.203 4.090 4.120 0.993 75 -36.092 4.230 4.280 0.988 -10 -35.600 0.435 0.427 1.02 5 -33.960 2.772 2.926 0.947 20 -35.300 2.914 2.960 0.985 35 -36.000 2.874 3.165 0.908 50 -36.210 2.768 3.046 0.909 65 -36.750 2.627 2.986 0.880 75 -36.610 3.100 3.074 1.01 -10 -36.167 0.859 0.620 1.39 5 -36.177 3.500 3.280 1.07 20 -35.240 3.560 3.340 1.07 35 -35.790 3.560 3.450 1.03 50 -36.448 3.745 3.720 1.01 65 -35.212 3.920 3.700 1.06 75 -35.765 3.650 3.710 0.983 -10 -35.120 0.137 0.144 0.950 5 -36.722 3.301 3.380 0.977 20 -34.843 3.520 3.660 0.962 35 -34.937 3.350 3.722 0.900 50 -34.203 3.540 3.850 0.919 65 -33.535 3.289 3.686 0.892 75 -34.699 3.228 3.727 0.866 Sprint (sec) 285 364 125 316 107 116 TREATMENT C (13C-glucose drink + water ad lib) Subj Time 3 1 3 C VCC-2 V O 2 RER [giu] [lact] [Hb] Hct Sprint (min) (%* (l/min) (l/min) (mg/dl) (mmol/L) (mg/dl) (sec) -10 -35.973 0.217 0.268 0.810 103.8 0.92 154 .442 5 -37.879 2.813 2.910 0.967 108.7 1.34 20 -37.583 2.940 2.990 0.983 114.9 1.78 35 -36.093 2.880 2.957 0.974 141.5 1.34 50 -35.227 3.100 3.207 0.967 133.9 1.51 65 -34.318 3.079 3.189 0.966 131.3 1.54 75 -34.544 3.065 3.150 0.973 150.8 1.40 163 .477 336 -10 -35.534 0.416 0.421 0.988 153.4 1.70 135 .394 5 -36.472 2.801 2.903 0.965 122.5 2.16 20 -36.352 3.156 3.211 0.983 187.7 1.58 35 -36.034 2.954 2.991 0.988 243.2 1.48 50 -34.144 3.190 3.258 0.979 226.8 1.28 65 -33.825 2.910 2.941 0.989 177.7 1.17 75 -32.995 3.210 3.357 0.956 181.0 1.29 138 .403 608 -10 -36.660 0.672 0.580 1.16 124.0 0.91 133 .394 5 -38.999 3.626 3.483 1.02 158.5 1.92 20 -38.391 3.563 3.567 0.999 215.7 2.52 35 -38.114 3.730 3.635 1.03 232.7 1.98 50 -37.572 3.758 3.732 1.01 214.7 2.55 65 -36.692 4.082 3.985 1.02 188.9 2.36 75 -36.745 3.870 3.950 0.980 192.9 2.77 144 .411 128 -10 -35.410 0.466 0.406 1.15 118.2 1.73 138 .400 5 -37.830 2.596 2.665 0.974 130.0 2.54 20 -35.840 2.790 2.839 0.983 161.0 2.00 35 -36.180 2.603 2.730 0.954 146.4 1.66 50 -34.720 2.496 2.809 0.889 134.1 1.31 65 -34.200 2.646 2.923 0.905 146.2 1.42 75 -34.260 2.689 2.908 0.925 165.2 1.96 147 .426 254 -10 -37.718 0.843 0.616 1.37 109.7 0.77 144 .413 5 -37.157 3.208 3.309 0.970 125.8 2.02 20 -36.102 3.498 3.510 0.997 160.5 1.97 35 -35.122 3.720 3.804 0.979 164.3 2.62 50 -34.146 3.595 3.717 0.967 145.4 1.91 65 -35.368 3.543 3.723 0.952 151.5 2.43 75 -35.338 3.502 3.803 0.921 167.4 1.94 163 .472 160 -10 -35.091 0.271 0.354 0.766 148.2 1.48 160 .458 5 -35.681 3.400 3.400 1.00 139.0 2.41 20 -34.225 3.358 3.479 0.965 216.4 3.12 35 -33.211 3.480 3.679 0.946 214.9 2.29 50 -32.839 3.609 3.899 0.926 166.4 2.58 65 -32.217 3.571 3.750 0.953 148.2 2.49 75 -32.498 3.557 3.763 0.945 109.5 2.26 175 .503 139 CF JD JG JL JM TC