Energy expenditure and substrates oxidative patterns, after glucose, fat or mixed load in normal weight subjects

Transcription

1 European Journal of Clinical Nutrition (1997) 51, 370±374 ß 1997 Stockton Press. All rights reserved 0954±3007/97 $12.00 and substrates oxidative patterns, after glucose, fat or mixed load in normal weight subjects E Bobbioni-Harsch 1, F Habicht 1, T Lehmann 1, RW James 2, F Rohner-Jeanrenaud 3 and A Golay 1 1 Teaching Diabetic Division, Geneva University Hospital, 1211 Geneva 14, Switzerland; 2 Clinical Diabetes Unit, Geneva University Hospital, Geneva, Switzerland; and Laboratoires de Recherches MeÂtaboliques, Geneva Medical School, Geneva, Switzerland Objectives: To evaluate energy balance after three isocaloric oral loads of different composition and to establish possible relationships between the substrates oxidative patterns and the modi cations of insulin and free fatty acids (FFA) plasma pro les. Design: Each subject received, in a randomized order, three oral loads of kj ( Kcal) either as glucose, lipids (cream) or a mixture (glucose cream). Setting: The experiments were performed at the University Hospital of Geneva. Subjects: Ten normal body-weight volunteers. Methods: (EE) and substrates oxidation were measured by indirect calorimetry during 8 h following each load. Plasma glucose, insulin and FFA were also measured. Results: EE was , and KJ over 8 h after glucose, mixed and lipids load, respectively. Glucose oxidation was the highest after oral glucose as compared to mixed and lipids load, while the highest value of lipids oxidation was measured after fat load. A signi cant relationship linked fat oxidation to plasma FFA (r ˆ 0.54, P < 0.002) as well as to insulin (r ˆ 70.40, P < 0.002). Conclusions: (a) The energetic cost of glucose and fat intake is the same; (2) after each load, the main source of energy corresponds to the substrate administered; (3) both plasma insulin and FFA in uence the substrate oxidative patterns observed after each load; (4) alimentary fat may contribute to fat oxidation by maintaining elevated plasma FFA levels. Descriptors: energy expenditure; glucose and lipids oxidation; insulin; FFA Introduction Knowledge of the mechanisms that regulate the homeostasis of body energy and body composition is essential when trying to elucidate the pathogenesis of obesity. The maintenance of a constant body weight depends directly upon the ability of the organism to adapt its substrate oxidation to both quality and quantity of substrate intake. It has been demonstrated (Moeri et al, 1988) that carbohydrate ingestion stimulates both glucose oxidation and storage and, by this mechanism, carbohydrate balance is achieved. Dietary fat seems to be handled differently by the organism. In fact, alimentary fat does not stimulate fat oxidation (Schutz et al, 1989) and, as shown by Flatt et al (1985), alimentary lipids, when added to a standard mixed meal, are almost completely stored. Furthermore, the energetic cost of a high fat meal is lower than that of a high carbohydrate meal (Schwartz et al, 1985; Lean & James, 1988). Finally, Tremblay et al (1989) have demonstrated that normal weight subjects develop hyperphagia when fed a high fat diet and this effect was in part attributed to a slow adaptation of lipid metabolism to a high fat diet. The results of these studies suggested the hypothesis that a diet rich in fat could contribute to the development of obesity. It should be noted, however, that in these studies Correspondence: Dr E Bobbioni-Harsch Received 30 September 1996; revised 3 February 1997; accepted 14 February 1997 fat consumption was associated either experimentally (Flatt et al, 1985) or spontaneously (Tremblay et al, 1989), with increased caloric intake. It is therefore dif cult to evaluate the relative impact of the quality and quantity of a diet on both energy balance and substrate oxidative patterns. In fact, when caloric intake is kept constant, as reported by Hill et al (1991), normal weight subject were able to modify their substrate oxidation in function of diet composition, without alterations of energy balance. These results suggest that excess caloric intake, rather than diet composition, is responsible for fat deposition. The aims of the present study were to evaluate the energetic cost of three isocaloric oral loads composed of glucose, fat or glucose and fat and to establish possible correlations between the substrate oxidative patterns and the modi cation of circulating insulin and FFA. Materials and methods Ten healthy, non-obese volunteers (5 males, 5 females: age y; body weight Kg; BMI kg/m 2 ) participated in this study. Before the tests, the subjects were informed about the purpose of the study and its experimental protocol (approved by the ethical committee of the Geneva University Hospital) and gave their written consent. The subjects were instructed by a dietician to consume, for the three days before the tests, a diet composed of 50% carbohydrates, 30% lipids and 20% proteins which represented a daily intake of 117 kj (28 Kcal)=kg body weight. Each volunteer was subjected

2 to the three oral loads in a randomized order at intervals of at least one week. The experiments begun in the morning after a 10 h fast. After voiding, the subjects were placed in a recumbent position, an in-dwelling catheter was placed into an antecubital vein and kept open by a slow saline infusion. Blood samples were drawn under basal conditions and every hour during 8 h after the ingestion of each load. Gas exchanges were measured as described by JeÂquier (1981) (for a review, see Ferranini, 1988), by placing the head of the subjects in a ventilated hood (Deltatrac, Datex Corp, Helsinky, Finland). VO2 and VCO2 were measured under basal conditions for 30 min before each test load. After the ingestion of the load, VO2 and VCO2 were measured for 8 h with two intervals of an hour between the third and the fourth hour and between the sixth and the seventh hour. During these intervals, subjects were allowed to seat but not to run. Urine was collected throughout the test for nitrogen determination. Protein oxidation, which was assumed to remain constant throughout each experiment, was calculated as N, where N is nitrogen excretion (g/min) in urine. The non-protein respiratory quotient (RQ) was determined from VO2 and VCO2 measurements after substraction of protein oxidation; glucose and lipid oxidation rates were obtained from the tables of Lusk (1924). Energy expenditure (EE) was calculated from the rates of glucose, lipid and protein oxidation. The loads supplied 1547 kj (370 Kcal)/m 2 of body surface either under the form of glucose (glucose load), cream (lipid load: composition in g ˆ 94% lipids, 3% carbohydrates, 3% proteins) or glucose and cream (mixed load: composition in g ˆ 47% lipids, 51.5% glucose and 1.5% proteins). Plasma glucose was determined by the glucose oxidase method (Caraway, 1987), insulin was determined by radio immunoassay according to Herbert et al (1965) and FFA were measured enzymatically (Dole & Meinertz, 1960)., glucose and lipid oxidation as well as the modi cation of insulin secretion and FFA plasma concentrations over 8 h period were calculated as the surface area under the curve (s). Dynamic modi cations of the different parameters measured in this study were analyzed by 2 way ANOVA for repeated measures and by one way ANOVA followed by Scheffe F-test when the comparison was made between the three different experimental conditions at each timepoint or when comparing the surface areas. Signi cant relationship between different parameters measured during the tests was evaluated by single regression analysis. Results Table 1 summarizes the non protein energy balance in the three experimental conditions., which was the same in basal conditions through the three experiments, was increased to the same extent after the ingestion of each load. When calculated as percent of energy intake, the increment in energy expenditure averaged 8±9% either after glucose, mixed and fat load. As shown in Figure 1a, the Respiratory Quotient (RQ) reached its highest values after glucose load, while it remained almost unchanged after fat load (2 way ANOVA: load 6 time P ˆ ). The calculation of 371 Table 1 Non-protein energy balance in the three experimental conditions Glucose load Mixed load Lipids load EI a (KJ) (638 20) (627 18) (613 15) Basal EE b (KJ over (8 h) (366 26) (378 20) (375 28) Post-load EE (KJ over 8 h) (425 26) (435 30) (427 28) D EE (KJ over 8 h) (59 8) (58 6) (52 8) D EE (% of EI) a EI ˆ Energy intake EE ˆ. Values into parentheses indicate the equivalent in Kcal. Figure 1 Non protein Respiratory Quotient (a), glucose (b) and lipid (c) oxidation in basal conditions and after glucose, mixed and lipid loads $ P < 0.05; $ P < 0.01; $ P < vs lipids np < 0.05; nnp < 0.001; nnnp < vs glucose * P < 0.05; ** P < 0.001; *** P < vs mixed.

3 372 Figure 2 Percentage of non-protein energy expenditure linked to glucose (u) and to lipid (j) oxidation in basal conditions and after glucose, mixed and lipid loads. gluocse (Figure 1b) and lipids (Figure 1c) oxidation derived from RQ indicated that carbohydrates utilization was stimulated by the ingestion of both glucose and mixed load (2 way ANOVA: load 6 time P ˆ ) concomitantly to a decrease in fat utilization. After lipids load, fat oxidation was similar to the one measured in basal, fasting conditions while it decreased after glucose and mixed load (2 way ANOVA: load 6 time P ˆ ). No signi cant differences in protein oxidation were measured through the three experiments (namely , and g over 8 h respectively after glucose, mixed and lipid load). When evaluating the substrate contribution to the non-protein energy expenditure (Figure 2), it appeared that glucose and lipid oxidation supplied respectively and % of the energy after glucose load, and after mixed load, and % after lipid load. Plasma glucose and insulin (Figures 3a and b) rose after both glucose and mixed load, while they were almost unchanged after fat ingestion (2 way ANOVA for glucose: load 6 time P ˆ 0.001; for insulin: load 6 time P ˆ ). Plasma FFA concentrations, which were elevated in fasting conditions, declined promptly after glucose but less markedly after a mixed load. In both cases FFA levels rose again from the fth hour to reach values similar or even higher than basal levels by the end of the studies. After lipids load, FFA remained unchanged during 1, then they progressively increased between the second and the fth hour and remained elevated until the end of the study (2 way ANOVA for FFA: load 6 time P ˆ ). Plasma TG (Figure 3d) tended to decrease after oral glucose while they clearly increased after both mixed and lipids load. In this last condition the peak value was reached 5 h after the ingestion of the load; plasma TG concentrations then decreased to return at the basal levels at the end of the experimentation (2 way ANOVA for TG: load 6 time P ˆ ). Simple regression analysis showed a signi - Figure 3 Plasma glucose (a), insulin (b), FFA (c) and TG (d) in basal conditions and after glucose (s), mixed (u) and lipid (n) load $ P < 0.05; $ P < 0.01; $ P < vs lipids n P < 0.05; nn P < 0.01; nn P < vs glucose * P < 0.05; ** P < 0.01; *** P < vs mixed.

4 Figure 4 (a) Relationship between plasma FFA concentrations and glucose oxidation: r ˆ 70.74, P < ; (b) Relationship between plasma FFA concentrations and lipid oxidation: r ˆ 054, P < cant relationship between the plasma FFA concentrations, calculated as FFA surface area, and the amounts of both glucose (r ˆ 70.74, P < ; Figure 4a) and fat (r ˆ 0.54 P < 0.002; Figure 4b) oxidized. The total amount of insulin secreted (insulin surface area) was signi cantly linked to glucose (r ˆ 0.70, P < ) and to fat (r ˆ 0.54; P < ) oxidation. Finally, insulin and FFA surface areas were signi cantly related (r ˆ 0.40, P < 0.002). Discussion The results of our study indicate that the energetic cost of glucose and lipid utilization and/or storage is equivalent, namely between 8±9% of the calories ingested. A lower themogenicity of alimentary fat has been reported by some authors (Schwartz et al, 1985; Lean & James, 1988) but not by all (Abbott et al, 1990). In normal weight male subjects, Schwartz et al (1985) measured energy expenditure during 6 h following the ingestion of either a high fat or high carbohydrate meal. These authors found a lower effect of fat, as compared to carbohydrates. A possible explanation of the different results reported in Schwartz's and our study could be due to the presence of both sexes in our study group, while Schwartz et al investigated only male subjects. Another possible explanation could be linked to the different duration of the measurements (namely 6 h and 8 h respectively in Schwartz's and our study). In fact, these authors reported that in none of their experimental groups EE had returned to the basal values at the end of the measurements (i.e. after 6 h) and, therefore, diet induced thermogenesis was to some degree underestimated. As recently demonstrated by Reed et al (1966), the thermic effect of food is in uenced by several factors such as the duration of the measurement and meal size and composition. Furthermore, it is known that intestinal absorption of fat is slower than that of carbohydrates and thus it is likely that, in the study by Schwartz et al, the underestimation of food-induced thermogenesis mainly concerned the high fat meal. This hypothesis is supported by two observations: rstly in our study the energetic cost of fat would be lower than glucose (6.5% vs 8.6% respectively) if DIT was evaluated over a period of 6 h. The more prolonged, 8 h measurements performed in our study can explain the discrepancies between the two studies. Secondly, when looking at the plasma glucose and TG pro les, respectively after glucose and lipids load, it can be seen that glycaemia returned to the basal values 4 h after the load, while TG remained elevated until the seventh hour of the test. This further suggests that fat absorption took longer while than glucose to be completed. In our study, the calculation of substrate oxidation following each load showed that, in normal weight subjects, the main source of energy corresponded to the substrate ingested. In fact, after the glucose load, glucose oxidation represented % of the non proteinderived EE; when calories were administered as glucose and lipids, the oxidation of these substrates supplied respectively and % of the energy consumed after the load. Finally, fat oxidation yielded, after oral lipids, % of the calories expended. Since no lipid turnover study was performed in our experiments, the question of whether, after mixed or lipid load, endogenous or alimentary fat were utilized remains open. Flatt et al (1985) suggested that dietary fat is primarily stored rather than oxidized, based on the observation that the addition of 50 g of margarine to a high carbohydrate meal did not enhance fat oxidation. As just mentioned, in these experiments fat was added to a 2006 kj (480 Kcal) containing meal: it therefore represented a supplementary source of energy. This could explain why alimentary fat was preferentially stored. Bennett et al (1992) offered the same conclusion, since they did not observe any increase in fat oxidation after adding 50 g fat to a standard breakfast. On the other hand, it has been shown (Hill et al, 1991) that, when caloric intake is kept constant, it is possible to observe a rapid adaptation of substrate oxidative patterns to the diet composition, without any alteration of energy balance. This implies that at least normal weight subjects are able to ef ciently utilize alimentary fat, as long as the caloric intake does not exceed the energetic needs of the subject. In a recent study, Grif ths et al (1994) observed that, after a mixed meal, plasma FFA remained relatively elevated, despite the simultaneous presence of a high plasma concentration of insulin. They concluded that circulating FFA were probably of dietary origin and derived from the hydrolysis of chylomicron-triacylglycerol by endothelial lipoprotein lipase. In our study after mixed load, plasma FFA remained signi cantly higher than after glucose alone (Figure 3c), despite the rise of circulating insulin to concentrations that have been demonstrated to inhibit lipolysis (Chen et al, 373

5 ; Bonadonna et al, 1990). This suggests that, in this experimental condition, plasma FFA were, at least in part, or alimentary origin. After the lipid load, plasma FFA progressively increased from the second hour, and remained elevated until the end of the test. The elevation of plasma FFA probably resulted from the contribution of both endogenous lipolysis and chylomicron-triacylglycerol hydrolysis as reported by Binnert et al (1996), who conclusively demonstrated the entry of exogenous lipids into the NEFA pool. On the basis of these observations, we propose that, in our study, alimentary fat contributed to the smaller decrease (mixed load) or to the increase (lipid load) of circulating FFA. In turn, plasma FFA modi cations (as evaluated by surface area) were closely (P < 0.002) linked to the modi cations of fat oxidation. We therefore suggest that dietary fat in uences fat oxidation by modifying the concentration of plasma FFA. The role of circulating FFA in the regulation of fat oxidation has been demonstrated by Groop et al (1991) in a study performed in steady state (clamp) conditions. From our results it would appear that plasma FFA levels are also involved in the regulation of lipid oxidation in dynamic conditions which are closer to the physiological situation, for example the post-prandial period. In our study, total lipid oxidation was only slightly (10%) increased after fat ingestion. One could, therefore, speculate that, compared to glucose, alimentary fat does not stimulate fat oxidation. However, Binnert et al (1996) have elegantly demonstrated that fat ingestion is followed by exogenous fat oxidation (which contributes up to 70% of the total lipid oxidation) and a reduction of endogenous lipid oxidation. This could explain why, after a lipid load, total fat utilization is only slightly modi ed. We can therefore conclude that the ingestion of lipids, as well as glucose, induces a reduction of endogenous lipid oxidation. In other terms, the absorption of exogenous energy, whatever the origin (carbohydrates or lipids), permits the organism to spare its own energy reserves (lipids stored in adipose tissue). Finally, it should be noted that in fasting basal conditions fat oxidation is already very elevated and supplies about 75% of the non-protein derived energy: one could speculate that this represents a near-maximal stimulation of lipid oxidation. Similarly, the highest level of glucose oxidation, measured after oral glucose, does not exceed 75% of the total energy expenditure. In our study, plasma insulin was signi cantly correlated both to glucose and lipid oxidation and to circulating FFA. This suggests that the control exercised by insulin on substrate oxidation is in part due to the regulation of FFA plasma levels. In fact, insulin, by inhibiting lipolysis, reduces the levels of circulating FFA and therefore lipid oxidation; this, in turn, further facilitates glucose disposal by muscle. Conclusions The results of the present study do not support the hypothesis that alimentary fat, in the absence of fat over-consumption could contribute to the development of obesity via its low thermogenicity. The modi cations of plasma insulin and FFA levels, induced by a meal, largely in uence the subsequent patterns of substrates oxidation in normal weight subjects. We cannot exclude, however, that obese individuals (Swanimathan et al, 1985; Thomas et al, 1992) or those with a genetic predisposition to obesity (Astrup, 1993; Astrup, 1994) fail to ef ciently utilize alimentary fat and, as a consequence, accumulate lipid in adipose tissue. References Abbott WGH, Howard BV, Ruotolo G & Ravussin E (1990): Energy expenditure in humans: Effect of dietary fat and carbohydrate. Am. J. Physiol. 258, E347±E35. Astrup A (1993): Dietary composition, substrate balances and body fat in subjects with a predisposition to obesity. Int. J. Obes. 17 (Suppl 3), 532±536. Astrup A, Buemann B, Western P, Toubro S, Raben A & Christiansen N (1994): Obesity as an adaptation to high-fat diet: evidence from a crosssectional study. Am. J. Clin. Nutr. 59, 350±355. Bennett C, Reed GW, Peters JC, Abumrad NN, Sun M & Hill JO (1992): Short-term effects of dietary fat ingestion on energy expenditure and nutrient balance. Am. J. Physiol. 270, E445±E450. Bonadonna RC, Groop LC, Zych K, Shank M & DeFronzo R (1990): Dose dependent effect of insulin on plasma free fatty acid turnover and oxidation in humans. Am. J. Physiol. 259, E736±E750. Caraway WT (1987): Carbohydrates. In Tietz NW (ed). Fundamentals of Clinical Chemistry 2nd edn. Philadelphia: WB Saunders Co, p 427. Chen YDI, Golay A, Swislocki ALM & Reaven GM (1987): Resistance to insulin suppression of plasma free fatty acid concentrations and insulin stimulation of glucose uptake in non-insulin dependent diabetes mellitus. J. Clin. Endocr. Metab. 64, 17±21. Dole VP & Meinertz H (1960): Microdetermination of long chain fatty acids in plasma and tissues. J. Biol. Chem. 235, 2595±2599. Ferranini E (1988): The theoretical bases of indirect calorimetry: a review. Metabolism 287±301. Flatt JP, Ravussin E, Acheson KJ & JeÂquier E (1985): Effects of dietary fat on postprandial substrate oxidation and on carbohydrate and fat balances. J. Clin. Invest. 76, 1019±1024. Grif ths J, Humpfreys S, Clark M, Fielding B & Frayn K (1994): Immediate metabolic availability of dietary fat in combination with carbohydrate. Am. J. Clin. Nutr. 59, 53±59. Groop L, Bonadonna R, Shank M, Petrides A & DeFronzo R (1991): Role of free fatty acids and insulin in determining free fatty acid and lipids oxidation in man. J. Clin. Invest. 87, 83±89. Herbert V, Lau KS, Gottlieb CW & Bleicher SY (1965): Coated charcoal immunoassay of insulin. J. Clin. Endocrinol. Metab. 25, 1175±1184. Hill J, Peters J, Reed G, Schlundt D, Sharp T & Greene H (1991): Nutrient balance in humans: effects of diet composition. Am. J. Clin. Nutr. 54, 10±17. JeÂquier E (1981): Long term measurement of energy expenditure. Direct or indirect calorimetry? In: BjoÈntorp P, Cairella M, Howard AN (eds). Advances in Obesity III. London: J Libbey, pp 130±135. Lean MEJ & James WPI (1988): Metabolic effect of isoenergetic nutrient exchange over 24 hours in relation to obesity in women. Int. J. Obes. 12, 15±27. Lusk G (1924): Animal calorimetry: analysis of the oxidation of mixtures of carbohydrate and fat. J. Biol. Chem. 59, 41±42. Moeri R, Golay A, Schultz Y, Telmer E, JeÂquier E & Felber JP (1988): Oxidative and non oxidative glucose metabolism following graded doses of oral glucose in man. DiabeÁte & MeÂtabolisme 14, 1±7. Reed GW & Hill JO (1996): Measuring the thermic effect of food. Am. J. Clin. Nutr. 63, 164±169. Schutz Y, Flatt JP & JeÂquier E (1989): Failure of dietary fat intake to promote fat oxidation: a factor favoring the developing of obesity. Am. J. Clin. Nutr. 50, 307±314. Schwartz R, Ravussin E, Massari M, O'Connel M & Robbins D (1985): The thermic effect of carbohydrate versus fat feeding in man. Metabolism 34, 285±293. Swanimathan R, King RFGJ, Holm Field J, Siwek RA, Baker M & Wales JK (1985): Thermic effect of feeding carbohydrate, fat, protein and mixed meal in lean and obese subjects. Am. J. Clin. Nutr. 42, 177±181. Thomas C, Peters JC, Reed GW, Abumrad NN, Sun M & Hill JO (1992): Nutrient balance and energy expenditure during ad libitum feeding of high-fat and high-carbohydrate diets in humans. Am. J. Clin. Nutr. 55, 934±942. Tremblay A, Plourde G, Despres JP & Bouchard C (1989): Impact of dietary fat content and fat oxidation on energy intake in humans. Am. J. Nutr. 49, 799±805.