Influence
of Dietary
Fat Polyunsaturated Utilization
to Saturated in Obesity
Ratio on Energy
Substrate
Peter J.H. Jones, Julie E. Ridgen, P. Terry Phang, and C. Laird Birmingham The effect of the polyunsaturated to saturated (P:S) ratio of dietary fat on preprandial and postprandial macronutrient oxidation was studied in normal-weight and obese individuals. Total thermogenic response and fat and carbohydrate oxidation rates were determined by duplicate respiratory gas exchange measurements after test breakfasts, in seven normal and eight overweight subjects who consumed self-selected diets containing fat of high or low P:S ratio. Dietary intake records and erythrocyte linoleic to oleic (L:O) acid ratio changes were used as indicators of dietary compliance. No diet- or weight-related differences were observed in resting fat or carbohydrate oxidation rates, or in protein-free basal energy expenditure. Obese subjects consuming low P:S ratio diets exhibited reduced (P < .05) contribution of fat oxidation to the thermogenic response, compared with lean individuals consuming high or low P:S ratio diets. However, total calories associated with the thermogenic response, and total fat and carbohydrate oxidation after the test breakfasts, did not differ significantly across groups. These findings suggest that, in obesity, whole-body postprandial disposal of dietary fat is influenced by the long-chain fatty acid composition. Copyright 0 1992 by W.B. Saunders Company
D
IETARY TRIGLYCERIDES largely contain longchain fatty acids which vary in carbon chain length, degree of unsaturation, and isomeric configuration of double bonds. The mixture of fatty acids consumed thus depends on the nature of the food selected. Growing evidence suggests that the blend of ingested long-chain fatty acids influences whole-body macronutrient oxidation.‘-” Oxidation studies using labeled fats in animals1-3 and humans4W6indicate that dietary polyunsaturated fatty acids are oxidized as fuel sources more rapidly than are saturated long-chain fatty acids. This concept of differential partitioning of long-chain fatty acids for oxidation versus storage is supported by energy balance studies in animals7-9 and respiratory gas exchange data in humans.” These findings suggest structure-dependent discrimination of fatty acids for energy production versus retention within storage pools. As positive caloric balance over extended periods is the principal underlying cause of obesity, qualitative fat consumption may exist as a factor in regulation of caloric balance and body weight. Presently, considerable debate exists as to whether the thermogenic response to food is altered in obesity.“-” Both diminished1’.14 and unchanged”“’ thermogenesis postprandially have been reported; however, much variation exists in the composition of test meals and the relative caloric load given in these studies. Thus, some of the inconsistency in thermic response may be due to meal composition. Whether the long-chain fatty acid composition of consumed fat
From the Division of Human Num’tion, School of Family and Num’tional Sciences and the Departments of Surgery and Internal Medicine, Faculq of Medicine, University of Britkh Columbia, Vancouver, BC, Canada. Supported by grants from the British Columbia Medical Services Foundation and Bristol-Meyers, Canada. Personal funding for P. TP. wasprovidedby the Bn’tkh Columbia Health Research Foundation. Address reprint requests to Peter J.H. Jones, PhD, Division of Human Nutrition, 2205 E Mall, University of British Columbia, Vancouver, British Columbia V6T 1 W5, Canada. Copyright 0 1992 by W.B. Saunders Company 0026-0495192/4104-0009$03.00/O
396
influences total energy expenditure or substrate utilization postprandially in obesity has not been fully examined. The present study was undertaken to investigate the effect of the polyunsaturated to saturated (P:S) fatty acid ratio of dietary fat on macronutrient oxidation in normalweight and obese individuals. The first objective was to examine whether macronutrient oxidation is altered by differences in the P:S ratio of consumed dietary fat in lean and obese individuals. The second objective was to examine whether any such alterations in macronutrient oxidation may be associated with changes in total caloric postprandial thermogenesis. Fat and carbohydrate oxidation rates and thermogenic responses were determined for 270 minutes after test breakfasts, in normal and overweight subjects consuming diets differing in the P:S ratio of fat. SUBJECTS AND METHODS Subjects Seven lean and eight obese subjects, aged 19 to 35 years, participated (Table 1). Criteria for acceptability included limitation to moderate physical activity and absence of reported chronic disease, including diabetes and the use of medications. Body fat content of subjects was determined from skinfold measurements at bicep, tricep, subscapular, and suprailiac sites, using Harpenden calipers.” Obese men and women had greater than 28% and 30%
body fat, respectively. Non-obese subjects’ body weights were within normal ranges for height and frame size, according to Metropolitan Tables, 1983. The study protocol was approved by the Clinical Experimentation Ethical Review Committee of the University of British Columbia. Diets and Protocol
Subjects were tested using a randomized cross-over design. Each subject underwent two experimental dietary periods of 14 days, separated by 7 days of habitual food intake. During dietary periods, the free-living subjects self-selected foods, with supervision, containing either a high or low P:S fat ratio. Target dietary intakes were 4.5%, 40%, and 15% of calories as fat, carbohydrate, and protein, respectively. High and low P:S fat ratio targets were 2.00 and 0.3, respectively. Alcohol and caffeine consumption were prohibited during dietary periods. Food items were chosen by subjects using detailed written and verbal instructions provided to them by a consultant dietitian before and during the study. Specific informaMetabolism,
Vol41, No 4 (April), 1992: pp 396-401
FAT P:S RATIO AND ENERGY METABOLISM
IN OBESITY
Table 1. Subject Description, Repotted Caloric Intakes, and P:S Fatty Acid Ratios of Dietary Fat, and Erythrocyte Fatty Acid L:O Acid Ratio Changes of Lean and Obese Subjects Consuming High and Low P:S Fat Ratio Diets Subject Characteristics
Lean
23.4 ir 1.5
Age W
30.8 IT 1.9
Sex 4
M b)
3
4 4
F (n) Height (cm)
171.2 + 2.2
172.6 f 4.4
Weight (kg)
66.6 it: 2.7
97.9 +- 8.0
Body fat (%)
18.7 2 2.4
36.1 2 1.3
Body weight change (kg) High P:S
0.35 r 0.09
1.2 f 0.59
Low P:S
0.25 + 0.12
0.48 * 0.22
Reported caloric intakes (kcal/d) High P:S
2,541 f 352
2,689 + 302
Low P:S
2,491 r 294
2,572 f 476
Reported P:S ratio’ of fat consumed High P:S
1.13 + 0.16
0.91 + 0.12
Low P:S
0.26 + 0.05
0.28 f 0.08
a transparent ventilated hood. The specific monitor used was initially validated with a lung model, as previously reported.*’ This lung model determined oxygen and carbon dioxide inaccuracies to be 1.9% and 1.5%, respectively?’ Prior to each energy expenditure measurement period, the unit was calibrated after warm-up, using appropriate reference gas standards. Computed values for oxygen and carbon dioxide exchange were corrected at standard temperature, pressure, and humidity. For each subject, respiratory gas exchange measurements before and after breakfast meals on days 11 and 14 were averaged for each of the two dietary periods. Carbohydrate and fat oxidation rates were determined each minute, using the Weir formulas, with subtraction of gas exchange associated with protein oxidation.** A constant protein oxidation rate of 0.7 g protein kg fat-free mass-’ d-’ was assumed, and nonprotein energy expenditure and macronutrient utilization rates (per minute) were calculated, as described previously.” The theoretical value was used for protein oxidation, instead of actual measurement of nitrogen excretion, since even significant inaccuracies in assumed nitrogen excretion propagate into only small errors on overall calculated substrate oxidation rates. Energy expenditure and macronutrient oxidation data were expressed as 30-minute averages. Analytical
Erythrocyte L:O ratio difference
between low P:S and
high P:S ratio diets
0.268 f 0.082
0.177 + 0.044
NOTE: Values are means + SEM. *Values represent the averages of food records of two periods of 3 days for each subject on each diet.
tion was provided regarding types of foods and oils appropriate for each dietary period. Regular contact was maintained with each subject over each period, to maximize adherence to target intakes. Except during days of testing, subjects did their normal work and recreational activities. Food intakes of subjects were monitored during initial and final 3-day segments of each 14-day dietary period, using food intake records. Nutrient intakes were calculated using a computerized nutrient composition program (Canadian Nutrient File, 1986). The degree of adherence to each dietary fat treatment was also assessed by measurement of differences in erythrocyte cell membrane linoleic to oleic (L:O) acid ratios during diet periods, from blood sampled on day 14 of each period.” To determine basal energy expenditure (BEE) and the thermic effect of food (TEF) of a standardized breakfast meal, respiratory gas exchange measurements were performed on days 11 and 14 during each of the two dietary periods. Subjects, having fasted during the previous 12 hours, reported to the Metabolic Facility at 7~00AM. Continuous respiratory gas exchange measurements were performed for 30 minutes before, and for 270 minutes following, the commencement of consumption of the breakfast test meal. The breakfast meal was composed of solid foods and designed to meet the target macronutrient and P:S fat ratio for each respective dietary period. The meal contained one third of each subject’s calculated daily energy requirement, as determined from their height and estimated ideal body weight, using the Harris-Benedict formula adjusted for activity with a factor of 1.7.” No additional foods were consumed during the testing period. Subjects remained supine, reading or lying quietly in a temperature- and noisecontrolled environment during the measurement period. Respiratory Gas Exchange Methodology Respiratory gas exchange measurements were performed with a Deltatrac Metabolic Monitor (Sensormedics, Anaheim, CA), using
Fatty acid compositional analysis of erythrocytes was determined by gas liquid chromatography (Hewlett Packard model 5750, Palo Alto, CA) of fatty acid methyl esters, after lipid extraction and transesterification.” Nitrogen carrier gas flowed (10 cc per minute) through a 0.53 mm x 30 m capillary column packed with Carbowax-10 (Supelco, Bellefonte, PA). The column was operated isothermally at 17o”C, with injector and flame ionization detector temperatures set at 220°C and XX, respectively. Chromatography peaks were identified by comparison of their retention data with those of authentic fatty acid methyl esters (Supelco, Bellefonte. PA). Statistical Analysis Substrate oxidation data were compared using a two-factor, two-way ANOVA, followed by Tukey’s tests, where appropriate. Comparisons were performed using a statistical software package (Systat, Evanston, IL). Results are expressed as means -+ SEM. RESULTS
Lean and obese groups were sex-matched, although obese subjects were slightly older (Table 1). Body weights and body fat content, as determined by skinfold measurements, were greater for obese subjects. Mean group body weight changes over each 1Cday study period were slightly positive in direction. There were no diet or weight effects on reported caloric intakes. Test breakfast meal sizes for lean (943 2 66 kcal) and obese (935 -C 66 kcal) subjects were similar in each diet. Reported dietary P:S fat ratios, averaged from two 3-day food records, indicated substantial shifts in qualitative fat intake for both lean and obese groups. Calculated P:S fat ratios were lower in obese subjects, compared with normal subjects consuming high P:S fat ratio diets. Mean erythrocyte fatty acid LO acid ratio changes were supportive of differences observed in reported food intake. All lean subjects showed L:O acid ratios appropriate’ for the shift in dietary fat, whereas three of the eight obese subjects showed slight changes in the direction opposite to what would be predicted.
JONES ET AL
For pairs of measures on the same diet and subject, the mean coefficient of variation (CV) for the resting nonprotein respiratory quotient (RQ) and total fat and carbohydrate oxidation were 3.9%, 4.3%, and 4.6%, respectively. For resting energy expenditure and calories associated with TEF, mean variations were 4.7% and 5.8%, respectively. No significant effects of body weight or dietary fat were observed on fasting nonprotein RQ values (Fig 1). Fat and carbohydrate oxidation rates before, and for 270 minutes after, the breakfast meal for lean and obese groups consuming high and low P:S fat ratio diets are shown in Figs 2 and 3, respectively. There were no significant effects of diet or subject group on basal fat or carbohydrate oxidation rates. Subjects on each diet treatment responded to the breakfast meal in a similar manner, although obese subjects tended to oxidize progressively less fat, and more carbohydrate, with increasing time after the meal. The nadir and peak for fat and carbohydrate oxidation, respectively, occurred between 60 and 180 minutes postprandially. Comparison of the changes in fat oxidation rates (g . min-‘) between prebreakfast and 270-minute timepoints showed no significant difference between high (-0.007 r 0.005 v -0.01 t 0.006 g. min-‘) or low (-0.003 ? 0.009 v -0.018 f 0.008 g . min-‘) P:S fat ratio diets for lean and obese subjects, respectively. However, cumulative fat oxidation relative to the basal level was influenced by body weight (P < .05) on the low P:S fat ratio diet (Fig 2). Obese subjects consuming a low P:S fat ratio diet exhibited significantly (P < .Ol) lower cumulative fat oxidation relative to the basal level over 270 minutes (-3.96 c 1.54 g .270 min-‘), compared with lean subjects consuming high (0.06 2 1.134 g. 270 min-‘) or low (1.15 + 1.64 g. 270 min-‘) P:S fat ratio diets. This effect was due to a combination of a slightly elevated resting, and lower subsequent postprandial, fat oxidation rate in the low P:S fat ratio obese group, compared with the normal weight group. For carbohydrate oxidation, no significant differences were observed in cumulative oxidation over basal levels between groups or diets (Fig 3). Regression analysis comparing percent body fat content in the obese subjects with the contributions of fat and carbohydrate oxidation to TEF were .36 (NS) and .45 (NS), respectively.
I High P:S
LOW P:S
LEAN
High P:S
LOW P:S
OBESE
Fig 1. Respiratory quotients before consumption of breakfast test meals in lean and obese subjects consuming high and low P:S fat ratio diets. Values are means of day 11 and day 14 measurements.
2 a IL
0.09
OBESE
BT
-30
0
30
80
90
120
150
180
210
240
]
270
MINUTES +
HIGH P:S
+
LOW P:S
b
HIGH P:S
LOW P:S
LEAN
HIGH P:8
LOW P:S
OBESE
Fig 2. Fat oxidation rates in lean (A) and obese (B) subjects during BEE (-30 minutes) and at 30-minute intervals after the breakfast test, while consuming high and low P:S fat ratio diets. Time = 0 was the point at which the subject commenced consumption of the breakfast meal. (C) Also shown is cumulative fat oxidation over basal for 270 minutes following the meal for each case. “Significantly different from obese low P:S (P c .05). “Significantly different from obese low P:S (P < .Ol). ‘Significantly different from obese low P:S (P < .05). Values are means * SEM. 0, High P:S; -•-, low P:S.
Total fat and carbohydrate oxidation data for the 270minute test period are presented in Table 2. No differences were observed in fat or carbohydrate oxidation, expressed as net amount oxidized or as percent of the meal content. Nonprotein calculated energy expenditure data, before and following the breakfast meal, for lean and obese subject groups consuming high and low P:S fat ratio diets are shown in Fig 4. In all cases, energy expenditure remained quite constant from 90 to 270 minutes, after an initial postprandial increase. Caloric expenditure associated with BEE was not statistically different across diet or subject groups. Figure 5 shows the TEF expressed as percent of test breakfast meal energy and as total uncorrected energy over the 270-minute measurement period in each group. No significant differences were observed in corrected or net rates of energy production. However, a trend (P < .lO) toward reduced net TEF caloric expenditure was suggested in obese subjects on the low P:S fat ratio diet. For high and low P:S fat ratio diets, respectively, net TEF calories oxidized were 51.8 + 8.3 and 53.8 + 6.9 kcal for lean
399
FAT P:S RATIO AND ENERGY METABOLISM IN OBESITY
Table 2. Fat and Carbohydrate
Oxidation Expressed as Total and
Percent of Meal Fat and Carbohydrate
in Lean and Obese Subjects
After Breakfast Meals Consuming High and Low P:S Fat Ratio Diets Ll?.Tl
Obese
(n = 7)
In = 8)
High P:S
16.1 1: 2.8
20.0 2 2.0
Low P:S
17.7 + 3.0
19.2 2 3.5
Total fat oxidation (g/270 min)
Percent of meal fat
0.051
:
:
:
:
:
:
:
:
:
:
:
-30
0
30
60
90
120
150
180
210
240
270
I
MINUTES +
HIQH P:S +
LOW P:S
High P:S
33.6 + 6.2
43.2 + 4.4
Low P:S
36.7 -t 5.3
42.0 + 5.8
Total carbohydrate oxidation (g/270 min) High P:S
49.2 ? 6.2
43.5 + 4.0
Low P:S
45.7 k 5.2
46.0 + 5.5
High P:S
52.5 t 6.1
48.6 -c 5.7
Low P:S
49.9 2 6.1
51.5 2 7.2
Percent of meal carbohydrate
NOTE. Values are means + SEM.
LEAN
OBESE
Fig 3. Carbohydrate oxidation rates in lean (A) and obese (B) subjects during BEE (-30 minutes) and at 30-minute intervals after the breakfast test while consuming high and low P:S fat ratio diets. (C) Also shown is cumulative carbohydrate oxidation over basal for 270 minutes following the meal for each case. Values are means f SEM. 0, High P:S; -_*-, low P:S.
subjects, and 57.8 + 8.8 and 36.9 ? 5.4 kcal for obese subjects.
sion in fat oxidation, resulting in a trend toward reduced dietary thermogenesis. However, this trend was statistically insignificant, due to considerable variation in TEF response within groups. The negative values observed for fat oxidation above basal levels in obese subjects reflect a postprandial shift in fat metabolism to favor storage and lipogenesis, compared with the positive values in lean individuals that indicate less chaneling of dietary substrates into stored fat. Katzeff and Danforth” also report negative 4-hour postprandial fat oxidation rates relative to basal levels, in lean and obese subjects on weight maintenance diets. Our present findings support their data, which show a significant reduction in lipid oxidation relative to basal levels in obese subjects compared with lean subjects. We presently extend this observation, suggesting that such a difference occurs with consumption of saturated, but not unsaturated, fat.
DISCUSSION
Animal’” and humat? studies using tracer labels suggest that long-chain polyunsaturated fatty acids are preferentially utilized for oxidation, compared with saturated fatty acids. Additional support for the concept of structurerelated discrimination for energy expenditure of dietary long-chain fatty acids is found in energy balance studies.7-’ To assess whether dietary fat composition plays a role in the pathogenesis of obesity, components of TEF after fixed meals varying only in fatty acid content were compared in lean and obese subjects. Our findings indicate that obese individuals consuming a meal high in saturated fats oxidize less fat relative to basal levels during the immediate postprandial period, in comparison with normal weight individuals. This effect appears to be due to a combination of opposing shifts in both basal and postprandial fat oxidation. The observed difference was not evident with consumption of a meal high in polyunsaturated fat. In the overweight group consuming saturated fat, carbohydrate oxidation did not fully compensate for the relative depres-
I!
:
:
:
-30
0
30
:
:
:
:
:
;
SO
SO
120
150
180
210
:
240
;
270
J
MINUTES +
HIGH P:S
+
LOW P:S
Fig 4. Protein free energy expenditure of lean (A) and obese (B) subjects during BEE (-30 minutes) and at 30.minute intervals after the breakfast test meal while consuming high and low P:S fat ratio diets. 0, High P:S; -_*-, low P:S.
JONES ET AL
“IS”
P:S
LOW I?*
LEAN
Me”
P:S
LOW P:8
OBESE
Fig 5. Cumulative caloric expenditure above basal over the 270minute test period for individuals of each group on each diet, expressed as percent of meal energy oxidized and as net energy oxidized. Values are means f SEM.
Although postprandial shifts in fat oxidation were observed between obese and normal weight subjects, no significant diet effect on the resting RQ or thermogenic response was seen in our normal weight individuals, as was reported in a similar previous study.” In the present study, it would be anticipated that the added precision of two measurement periods, in comparison to the earlier experiments restricted to a single measure, should have enabled reproduction of the former study’s RQ and energy utilization differences. However, our subjects self-selected their own diets and arranged their own transport to the testing facility on the morning of each trial. Thus, perhaps either the more rigorous dietary intervention approach or the overnight stay before each test day used in the earlier study” was responsible for a metabolic response of adequate proportion to exceed analytical and random biological error. Presently, the within-subject variability in RQ on each diet was high (CV, 3.9%), approximately half of which was attributable to biological variation. The respiratory gas exchange analyzer used presently had overall measurement errors for oxygen and carbon dioxide of less than 2%.‘l Thus, the inconsistency of our results in normal-weight individuals with those of the similar study performed previously”’ likely relate to factors other than the gas analyzer errors. The reasons for the observed changes in fat partitioning observed in obese, but not in normal-weight, individuals, as shown previously,” remain to be determined. Obesity is associated with sympathetic nervous system activity alterations in animals,24sZ and with lower circulating catecholamine levels in overweight humans.” Perhaps dietary long-chain fatty acids evoke a structure-dependent response of sympathetic activity or catecholamine release. Alternatively, intrinsic genetic differences between individuals may result in an altered metabolic response toZ dietary saturated fat in those predisposed to obesity. The trend in basal fat oxidation rates in the obese to be higher with
saturated versus unsaturated fat feeding may arise from one of these effects. A similar elevation in resting fat oxidation was previously observed in a study performed in normal subjects consuming diets high in saturated fat.“’ The controversial notion that impaired postprandial thermogenesis contributes to obesity is not supported by the results of this study, although there was a trend toward reduced TEF caloric expenditure in obese subjects on the low P:S fat ratio diet (Fig 5). Several reports demonstrate a significantly diminished rate of increase in energy expenditure following a meal in overweight subjects”~L4; however, others show no difference.“.” The response to dietary fat of fat oxidation above basal levels, observed presently suggests that diet composition may be a factor responsible for the disparate findings of these studies. Although not significantly different, trends suggest that obesity may be associated with impaired postprandial thermogenesis when diets high in saturated fatty acids are consumed, yet may manifest a response more similar to that of lean subjects in diets with polyunsaturated fats. Also, it cannot be ruled out that the differences in the postprandial measurement interval, and the use of liquid test diets, may influence results of these conflicting studies. In comparing metabolic responses of individuals differing in body composition, the size of the dietary challenge to which TEF is determined has been shown to be critical.” The magnitude of the metabolic response to a meal clearly depends on its caloric content.27.28 In the present study, energy content of the meal was based on ideal body weight, reflected in the mean meal caloric content differing by less than 1% between groups. However, meal size varied within each group; therefore, data comparisons were made between groups on an absolute and relative load basis. Segal et al” have demonstrated that whether meals are fed as a constant caloric load or in a manner dependent on body weight, thermogenic response differences between lean and obese subjects are consistent. It was therefore considered appropriate that, to minimize errors in estimation of exact body composition, subjects be fed test meals containing calories proportioned relative to ideal body weight. Subjects were putatively consuming diets similar in composition to those of meals given during test periods. As a means of verification, 3-day food records were maintained, which, for both lean and obese individuals, indicated substantial shifts in the nature of the fat consumed. As an independent index of compliance, erythrocyte L:O acid ratio responses to diet were examined. L:O acid ratio changes have been suggested as a reliable indicator of recent dietary fat intake patterns.” With the exception of three obese subjects, the L:O acid ratio of all individuals changed in the direction consistent with reported food record intakes. The interpretation, in the cases of absence of anticipated change, was not clear. At first glance, lack of response likely signifies lack of compliance. However, in the obese, erythrocyte fatty acid profiles may require a longer interval than that presently provided to reflect dietary fat patterns. Although significant changes in erythrocyte fatty acids have been reported over 8 days,29 a period of greater
FAT P:S RATIO AND ENERGY METABOLISM
401
IN OBESITY
than 3 weeks is required in normal-weight individuals for erythrocyte fatty acids to fully equilibrate to dietary fat intake.” More data are required to delineate the utility of erythrocyte fatty acid profiles in obese subjects. Presently, since food intake records indicated the appropriate P:S fat ratio changes between periods, the three obese nonresponders were included in the obese group. Statistical comparison without this subgroup produced identical significant differences in fat oxidation over the basal level. In summary, although our data do not show defective thermogenesis in obesity, a trend toward a blunted TEF in
obese subjects given the low P:S fat ratio breakfast was observed. The reduced contribution of fat oxidation to TEF observed in obese subjects consuming the low P:S fat ratio diets suggests that, although the total thermogenic response remains unaffected, overweight individuals partition less dietary saturated fat for oxidation postprandially, compared with individuals of normal body weight. ACKNOWLEDGMENT
The technical assistance of Laurel Aberhardt, Alexander Benson, Terri Broughton, and May Lee is greatly appreciated.
REFERENCES
1. Cenedella RJ, Allen A: Differences between the metabolism of linoleic and palmitic acid: Utilization for cholesterol synthesis and oxidation to respiratory CO,. Lipids 4:155-158,1969 2. Toorop AI, Romsos DR, Leveille GA: The metabolic fate of dietary 1-[‘4C]linoleate and 1-[14C] palmitate in meal fed rats. Proc Sot Exp Biol Med X0:312-316,1979 3. Leyton J, Drury PJ, Crawford MA: Differential oxidation of saturated and unsaturated fatty acids in vivo in the rat. Br J Nutr 57:383-393,1987 4. Forsgren L: Expiratory pattern of 14C0, in man after feeding “‘C-1abelled fatty acids. Arch Kemi 30~355-360,1968 5. Watkins JB, Klein PD, Schoeller DA, et al: Diagnosis and differentiation of fat malabsorption in children using 13Clabeled lipids; trioctanoin, triolein and palmitic acid breath tests. Gastroenterology 82911-917, 1982 6. Jones PJH, Pencharz PB, Clandinin MT: Whole body utilization of dietary fatty acids: Implications for energy utilization. Am J Clin Nutr 42:169-777,1985 7. Shimomura Y, Tamura T, Suzuki M: Less body fat accumulation in rats fed a safflower oil diet than rats fed a beef tallow diet. J Nutr 120:1291-1296,199O 8. Mercer SW, Trayhurn P: Effect of high fat diets on energy balance and thermogenesis in brown adipose tissue of lean and genetically obese ob/ob mice. J Nutr 117:2147-2153,1987 9. Jones PJH: Effect of fatty acid composition of dietary fat on energy balance and expenditure in hamsters. Can J Physiol Pharmacol67:994-998,1989 10. Jones PJH, Schoeller DA: Polyunsaturated:saturated ratio of dietary fat influences energy substrate utilization in the human. Metabolism 37:145-151,198s 11. Katzeff HL, Danforth E: Decreased thennic effect of a mixed meal during overnutrition in human obesity. Am J Clin Nutr 50:915-921,1989 12. Segal KR, Edano A, Bland0 L, et al: Comparison of thermic effects of constant and relative caloric loads in lean and obese men. Am J Clin Nutr 51:14-21,199O 13. Shetty PS, Jung RT, James WPT, et al: Postprandial thermogenesis in obesity. Clin Sci 60:519-525,198l 14. Kaplan ML, Leveille GA: Caloric response in obese and nonobese women. Am J Clin Nutr 29:1108-1113,1976 15. Welle SL, Campbell RG: Normal thermogenic effect of food in obese women. Am J Clin Nutr 37:87-92,1983
16. Schutz Y, Bessard T, Jequier E: Exercise and postprandial thermogenesis in obese women before and after weight loss. Am J Clin Nutr 45:1424-1432,1987 17. D’Alessio DA, Kavle EC, Mozzoli MA, et al: Thermic effect of food in lean and obese men. J Clin Invest 81:1781-1789,1988 18. Durnin JVGA, Womersley J: Body fat assessed from total body density and its estimation from skinfold thickness: Measurements on 481 men and women aged from 16 to 72 years. Br J Nutr 32:77-97,1974 19. Glatz JFC, Soffers AEMF, Katan MB: Fatty acid composition of serum cholesterol esters and erythrocyte membranes as indicators of linoleic acid intake in man. Am J Clin Nutr 49:269216,1989 20. Bell L, Jones PJH, Telch J, et al: Prediction of energy needs for clinical studies. Nutr Res 5:123-129,1985 21. Phang PT, Rich T, Ronco J: A validation and comparison study of two metabolic monitors. J Parent Enteral Nutr 14:259-261, 1989 22. Weir WE: New methods for calculating metabolic rate with special emphasis to protein metabolism. J Physiol 109:1-9,1949 23. Bannon CD, Craske JD, Hai NT, et al: Analysis of fatty acid methyl esters with high accuracy and reliability. II. Methylation of fats and oils with boron trifluoride-methanol. J Chromotgr 247:6369,1982 24. Sakaguchi T, Arase K, Bray GA: Sympathetic activity and food intake of rats with vetromedial hypothalamic lesions. Int J Obes 12:43-49,1988 25. Peterson HR, Rothschild M, Weinberg CR, et al: Body fat and the activity of the autonomic nervous system. N Engl J Med 318:1077-1083,1988 26. Bouchard C, Tremblay A: Genetic effects in human energy expenditure components. Int J Obes 14:45-54,199O 27. Hill JO, Heymsfield SB, McManus C, et al: Meal size and thermic response to food in male subjects as a function of maximal aerobic capacity. Metabolism 33:743-759,1984 28. Belko AZ, Barbieri TF, Wong EC: Effect of energy intake and protein intake and exercise intensity on the thermic effect of food. Am J Clin Nutr 43:863-869,1986 29. Farquhar JW, Amens EH: Effects of dietary fats on human erythrocyte fatty acid patterns. J Clin Invest 42675-685, 1963
of Dietary
Fat Polyunsaturated Utilization
to Saturated in Obesity
Ratio on Energy
Substrate
Peter J.H. Jones, Julie E. Ridgen, P. Terry Phang, and C. Laird Birmingham The effect of the polyunsaturated to saturated (P:S) ratio of dietary fat on preprandial and postprandial macronutrient oxidation was studied in normal-weight and obese individuals. Total thermogenic response and fat and carbohydrate oxidation rates were determined by duplicate respiratory gas exchange measurements after test breakfasts, in seven normal and eight overweight subjects who consumed self-selected diets containing fat of high or low P:S ratio. Dietary intake records and erythrocyte linoleic to oleic (L:O) acid ratio changes were used as indicators of dietary compliance. No diet- or weight-related differences were observed in resting fat or carbohydrate oxidation rates, or in protein-free basal energy expenditure. Obese subjects consuming low P:S ratio diets exhibited reduced (P < .05) contribution of fat oxidation to the thermogenic response, compared with lean individuals consuming high or low P:S ratio diets. However, total calories associated with the thermogenic response, and total fat and carbohydrate oxidation after the test breakfasts, did not differ significantly across groups. These findings suggest that, in obesity, whole-body postprandial disposal of dietary fat is influenced by the long-chain fatty acid composition. Copyright 0 1992 by W.B. Saunders Company
D
IETARY TRIGLYCERIDES largely contain longchain fatty acids which vary in carbon chain length, degree of unsaturation, and isomeric configuration of double bonds. The mixture of fatty acids consumed thus depends on the nature of the food selected. Growing evidence suggests that the blend of ingested long-chain fatty acids influences whole-body macronutrient oxidation.‘-” Oxidation studies using labeled fats in animals1-3 and humans4W6indicate that dietary polyunsaturated fatty acids are oxidized as fuel sources more rapidly than are saturated long-chain fatty acids. This concept of differential partitioning of long-chain fatty acids for oxidation versus storage is supported by energy balance studies in animals7-9 and respiratory gas exchange data in humans.” These findings suggest structure-dependent discrimination of fatty acids for energy production versus retention within storage pools. As positive caloric balance over extended periods is the principal underlying cause of obesity, qualitative fat consumption may exist as a factor in regulation of caloric balance and body weight. Presently, considerable debate exists as to whether the thermogenic response to food is altered in obesity.“-” Both diminished1’.14 and unchanged”“’ thermogenesis postprandially have been reported; however, much variation exists in the composition of test meals and the relative caloric load given in these studies. Thus, some of the inconsistency in thermic response may be due to meal composition. Whether the long-chain fatty acid composition of consumed fat
From the Division of Human Num’tion, School of Family and Num’tional Sciences and the Departments of Surgery and Internal Medicine, Faculq of Medicine, University of Britkh Columbia, Vancouver, BC, Canada. Supported by grants from the British Columbia Medical Services Foundation and Bristol-Meyers, Canada. Personal funding for P. TP. wasprovidedby the Bn’tkh Columbia Health Research Foundation. Address reprint requests to Peter J.H. Jones, PhD, Division of Human Nutrition, 2205 E Mall, University of British Columbia, Vancouver, British Columbia V6T 1 W5, Canada. Copyright 0 1992 by W.B. Saunders Company 0026-0495192/4104-0009$03.00/O
396
influences total energy expenditure or substrate utilization postprandially in obesity has not been fully examined. The present study was undertaken to investigate the effect of the polyunsaturated to saturated (P:S) fatty acid ratio of dietary fat on macronutrient oxidation in normalweight and obese individuals. The first objective was to examine whether macronutrient oxidation is altered by differences in the P:S ratio of consumed dietary fat in lean and obese individuals. The second objective was to examine whether any such alterations in macronutrient oxidation may be associated with changes in total caloric postprandial thermogenesis. Fat and carbohydrate oxidation rates and thermogenic responses were determined for 270 minutes after test breakfasts, in normal and overweight subjects consuming diets differing in the P:S ratio of fat. SUBJECTS AND METHODS Subjects Seven lean and eight obese subjects, aged 19 to 35 years, participated (Table 1). Criteria for acceptability included limitation to moderate physical activity and absence of reported chronic disease, including diabetes and the use of medications. Body fat content of subjects was determined from skinfold measurements at bicep, tricep, subscapular, and suprailiac sites, using Harpenden calipers.” Obese men and women had greater than 28% and 30%
body fat, respectively. Non-obese subjects’ body weights were within normal ranges for height and frame size, according to Metropolitan Tables, 1983. The study protocol was approved by the Clinical Experimentation Ethical Review Committee of the University of British Columbia. Diets and Protocol
Subjects were tested using a randomized cross-over design. Each subject underwent two experimental dietary periods of 14 days, separated by 7 days of habitual food intake. During dietary periods, the free-living subjects self-selected foods, with supervision, containing either a high or low P:S fat ratio. Target dietary intakes were 4.5%, 40%, and 15% of calories as fat, carbohydrate, and protein, respectively. High and low P:S fat ratio targets were 2.00 and 0.3, respectively. Alcohol and caffeine consumption were prohibited during dietary periods. Food items were chosen by subjects using detailed written and verbal instructions provided to them by a consultant dietitian before and during the study. Specific informaMetabolism,
Vol41, No 4 (April), 1992: pp 396-401
FAT P:S RATIO AND ENERGY METABOLISM
IN OBESITY
Table 1. Subject Description, Repotted Caloric Intakes, and P:S Fatty Acid Ratios of Dietary Fat, and Erythrocyte Fatty Acid L:O Acid Ratio Changes of Lean and Obese Subjects Consuming High and Low P:S Fat Ratio Diets Subject Characteristics
Lean
23.4 ir 1.5
Age W
30.8 IT 1.9
Sex 4
M b)
3
4 4
F (n) Height (cm)
171.2 + 2.2
172.6 f 4.4
Weight (kg)
66.6 it: 2.7
97.9 +- 8.0
Body fat (%)
18.7 2 2.4
36.1 2 1.3
Body weight change (kg) High P:S
0.35 r 0.09
1.2 f 0.59
Low P:S
0.25 + 0.12
0.48 * 0.22
Reported caloric intakes (kcal/d) High P:S
2,541 f 352
2,689 + 302
Low P:S
2,491 r 294
2,572 f 476
Reported P:S ratio’ of fat consumed High P:S
1.13 + 0.16
0.91 + 0.12
Low P:S
0.26 + 0.05
0.28 f 0.08
a transparent ventilated hood. The specific monitor used was initially validated with a lung model, as previously reported.*’ This lung model determined oxygen and carbon dioxide inaccuracies to be 1.9% and 1.5%, respectively?’ Prior to each energy expenditure measurement period, the unit was calibrated after warm-up, using appropriate reference gas standards. Computed values for oxygen and carbon dioxide exchange were corrected at standard temperature, pressure, and humidity. For each subject, respiratory gas exchange measurements before and after breakfast meals on days 11 and 14 were averaged for each of the two dietary periods. Carbohydrate and fat oxidation rates were determined each minute, using the Weir formulas, with subtraction of gas exchange associated with protein oxidation.** A constant protein oxidation rate of 0.7 g protein kg fat-free mass-’ d-’ was assumed, and nonprotein energy expenditure and macronutrient utilization rates (per minute) were calculated, as described previously.” The theoretical value was used for protein oxidation, instead of actual measurement of nitrogen excretion, since even significant inaccuracies in assumed nitrogen excretion propagate into only small errors on overall calculated substrate oxidation rates. Energy expenditure and macronutrient oxidation data were expressed as 30-minute averages. Analytical
Erythrocyte L:O ratio difference
between low P:S and
high P:S ratio diets
0.268 f 0.082
0.177 + 0.044
NOTE: Values are means + SEM. *Values represent the averages of food records of two periods of 3 days for each subject on each diet.
tion was provided regarding types of foods and oils appropriate for each dietary period. Regular contact was maintained with each subject over each period, to maximize adherence to target intakes. Except during days of testing, subjects did their normal work and recreational activities. Food intakes of subjects were monitored during initial and final 3-day segments of each 14-day dietary period, using food intake records. Nutrient intakes were calculated using a computerized nutrient composition program (Canadian Nutrient File, 1986). The degree of adherence to each dietary fat treatment was also assessed by measurement of differences in erythrocyte cell membrane linoleic to oleic (L:O) acid ratios during diet periods, from blood sampled on day 14 of each period.” To determine basal energy expenditure (BEE) and the thermic effect of food (TEF) of a standardized breakfast meal, respiratory gas exchange measurements were performed on days 11 and 14 during each of the two dietary periods. Subjects, having fasted during the previous 12 hours, reported to the Metabolic Facility at 7~00AM. Continuous respiratory gas exchange measurements were performed for 30 minutes before, and for 270 minutes following, the commencement of consumption of the breakfast test meal. The breakfast meal was composed of solid foods and designed to meet the target macronutrient and P:S fat ratio for each respective dietary period. The meal contained one third of each subject’s calculated daily energy requirement, as determined from their height and estimated ideal body weight, using the Harris-Benedict formula adjusted for activity with a factor of 1.7.” No additional foods were consumed during the testing period. Subjects remained supine, reading or lying quietly in a temperature- and noisecontrolled environment during the measurement period. Respiratory Gas Exchange Methodology Respiratory gas exchange measurements were performed with a Deltatrac Metabolic Monitor (Sensormedics, Anaheim, CA), using
Fatty acid compositional analysis of erythrocytes was determined by gas liquid chromatography (Hewlett Packard model 5750, Palo Alto, CA) of fatty acid methyl esters, after lipid extraction and transesterification.” Nitrogen carrier gas flowed (10 cc per minute) through a 0.53 mm x 30 m capillary column packed with Carbowax-10 (Supelco, Bellefonte, PA). The column was operated isothermally at 17o”C, with injector and flame ionization detector temperatures set at 220°C and XX, respectively. Chromatography peaks were identified by comparison of their retention data with those of authentic fatty acid methyl esters (Supelco, Bellefonte. PA). Statistical Analysis Substrate oxidation data were compared using a two-factor, two-way ANOVA, followed by Tukey’s tests, where appropriate. Comparisons were performed using a statistical software package (Systat, Evanston, IL). Results are expressed as means -+ SEM. RESULTS
Lean and obese groups were sex-matched, although obese subjects were slightly older (Table 1). Body weights and body fat content, as determined by skinfold measurements, were greater for obese subjects. Mean group body weight changes over each 1Cday study period were slightly positive in direction. There were no diet or weight effects on reported caloric intakes. Test breakfast meal sizes for lean (943 2 66 kcal) and obese (935 -C 66 kcal) subjects were similar in each diet. Reported dietary P:S fat ratios, averaged from two 3-day food records, indicated substantial shifts in qualitative fat intake for both lean and obese groups. Calculated P:S fat ratios were lower in obese subjects, compared with normal subjects consuming high P:S fat ratio diets. Mean erythrocyte fatty acid LO acid ratio changes were supportive of differences observed in reported food intake. All lean subjects showed L:O acid ratios appropriate’ for the shift in dietary fat, whereas three of the eight obese subjects showed slight changes in the direction opposite to what would be predicted.
JONES ET AL
For pairs of measures on the same diet and subject, the mean coefficient of variation (CV) for the resting nonprotein respiratory quotient (RQ) and total fat and carbohydrate oxidation were 3.9%, 4.3%, and 4.6%, respectively. For resting energy expenditure and calories associated with TEF, mean variations were 4.7% and 5.8%, respectively. No significant effects of body weight or dietary fat were observed on fasting nonprotein RQ values (Fig 1). Fat and carbohydrate oxidation rates before, and for 270 minutes after, the breakfast meal for lean and obese groups consuming high and low P:S fat ratio diets are shown in Figs 2 and 3, respectively. There were no significant effects of diet or subject group on basal fat or carbohydrate oxidation rates. Subjects on each diet treatment responded to the breakfast meal in a similar manner, although obese subjects tended to oxidize progressively less fat, and more carbohydrate, with increasing time after the meal. The nadir and peak for fat and carbohydrate oxidation, respectively, occurred between 60 and 180 minutes postprandially. Comparison of the changes in fat oxidation rates (g . min-‘) between prebreakfast and 270-minute timepoints showed no significant difference between high (-0.007 r 0.005 v -0.01 t 0.006 g. min-‘) or low (-0.003 ? 0.009 v -0.018 f 0.008 g . min-‘) P:S fat ratio diets for lean and obese subjects, respectively. However, cumulative fat oxidation relative to the basal level was influenced by body weight (P < .05) on the low P:S fat ratio diet (Fig 2). Obese subjects consuming a low P:S fat ratio diet exhibited significantly (P < .Ol) lower cumulative fat oxidation relative to the basal level over 270 minutes (-3.96 c 1.54 g .270 min-‘), compared with lean subjects consuming high (0.06 2 1.134 g. 270 min-‘) or low (1.15 + 1.64 g. 270 min-‘) P:S fat ratio diets. This effect was due to a combination of a slightly elevated resting, and lower subsequent postprandial, fat oxidation rate in the low P:S fat ratio obese group, compared with the normal weight group. For carbohydrate oxidation, no significant differences were observed in cumulative oxidation over basal levels between groups or diets (Fig 3). Regression analysis comparing percent body fat content in the obese subjects with the contributions of fat and carbohydrate oxidation to TEF were .36 (NS) and .45 (NS), respectively.
I High P:S
LOW P:S
LEAN
High P:S
LOW P:S
OBESE
Fig 1. Respiratory quotients before consumption of breakfast test meals in lean and obese subjects consuming high and low P:S fat ratio diets. Values are means of day 11 and day 14 measurements.
2 a IL
0.09
OBESE
BT
-30
0
30
80
90
120
150
180
210
240
]
270
MINUTES +
HIGH P:S
+
LOW P:S
b
HIGH P:S
LOW P:S
LEAN
HIGH P:8
LOW P:S
OBESE
Fig 2. Fat oxidation rates in lean (A) and obese (B) subjects during BEE (-30 minutes) and at 30-minute intervals after the breakfast test, while consuming high and low P:S fat ratio diets. Time = 0 was the point at which the subject commenced consumption of the breakfast meal. (C) Also shown is cumulative fat oxidation over basal for 270 minutes following the meal for each case. “Significantly different from obese low P:S (P c .05). “Significantly different from obese low P:S (P < .Ol). ‘Significantly different from obese low P:S (P < .05). Values are means * SEM. 0, High P:S; -•-, low P:S.
Total fat and carbohydrate oxidation data for the 270minute test period are presented in Table 2. No differences were observed in fat or carbohydrate oxidation, expressed as net amount oxidized or as percent of the meal content. Nonprotein calculated energy expenditure data, before and following the breakfast meal, for lean and obese subject groups consuming high and low P:S fat ratio diets are shown in Fig 4. In all cases, energy expenditure remained quite constant from 90 to 270 minutes, after an initial postprandial increase. Caloric expenditure associated with BEE was not statistically different across diet or subject groups. Figure 5 shows the TEF expressed as percent of test breakfast meal energy and as total uncorrected energy over the 270-minute measurement period in each group. No significant differences were observed in corrected or net rates of energy production. However, a trend (P < .lO) toward reduced net TEF caloric expenditure was suggested in obese subjects on the low P:S fat ratio diet. For high and low P:S fat ratio diets, respectively, net TEF calories oxidized were 51.8 + 8.3 and 53.8 + 6.9 kcal for lean
399
FAT P:S RATIO AND ENERGY METABOLISM IN OBESITY
Table 2. Fat and Carbohydrate
Oxidation Expressed as Total and
Percent of Meal Fat and Carbohydrate
in Lean and Obese Subjects
After Breakfast Meals Consuming High and Low P:S Fat Ratio Diets Ll?.Tl
Obese
(n = 7)
In = 8)
High P:S
16.1 1: 2.8
20.0 2 2.0
Low P:S
17.7 + 3.0
19.2 2 3.5
Total fat oxidation (g/270 min)
Percent of meal fat
0.051
:
:
:
:
:
:
:
:
:
:
:
-30
0
30
60
90
120
150
180
210
240
270
I
MINUTES +
HIQH P:S +
LOW P:S
High P:S
33.6 + 6.2
43.2 + 4.4
Low P:S
36.7 -t 5.3
42.0 + 5.8
Total carbohydrate oxidation (g/270 min) High P:S
49.2 ? 6.2
43.5 + 4.0
Low P:S
45.7 k 5.2
46.0 + 5.5
High P:S
52.5 t 6.1
48.6 -c 5.7
Low P:S
49.9 2 6.1
51.5 2 7.2
Percent of meal carbohydrate
NOTE. Values are means + SEM.
LEAN
OBESE
Fig 3. Carbohydrate oxidation rates in lean (A) and obese (B) subjects during BEE (-30 minutes) and at 30-minute intervals after the breakfast test while consuming high and low P:S fat ratio diets. (C) Also shown is cumulative carbohydrate oxidation over basal for 270 minutes following the meal for each case. Values are means f SEM. 0, High P:S; -_*-, low P:S.
subjects, and 57.8 + 8.8 and 36.9 ? 5.4 kcal for obese subjects.
sion in fat oxidation, resulting in a trend toward reduced dietary thermogenesis. However, this trend was statistically insignificant, due to considerable variation in TEF response within groups. The negative values observed for fat oxidation above basal levels in obese subjects reflect a postprandial shift in fat metabolism to favor storage and lipogenesis, compared with the positive values in lean individuals that indicate less chaneling of dietary substrates into stored fat. Katzeff and Danforth” also report negative 4-hour postprandial fat oxidation rates relative to basal levels, in lean and obese subjects on weight maintenance diets. Our present findings support their data, which show a significant reduction in lipid oxidation relative to basal levels in obese subjects compared with lean subjects. We presently extend this observation, suggesting that such a difference occurs with consumption of saturated, but not unsaturated, fat.
DISCUSSION
Animal’” and humat? studies using tracer labels suggest that long-chain polyunsaturated fatty acids are preferentially utilized for oxidation, compared with saturated fatty acids. Additional support for the concept of structurerelated discrimination for energy expenditure of dietary long-chain fatty acids is found in energy balance studies.7-’ To assess whether dietary fat composition plays a role in the pathogenesis of obesity, components of TEF after fixed meals varying only in fatty acid content were compared in lean and obese subjects. Our findings indicate that obese individuals consuming a meal high in saturated fats oxidize less fat relative to basal levels during the immediate postprandial period, in comparison with normal weight individuals. This effect appears to be due to a combination of opposing shifts in both basal and postprandial fat oxidation. The observed difference was not evident with consumption of a meal high in polyunsaturated fat. In the overweight group consuming saturated fat, carbohydrate oxidation did not fully compensate for the relative depres-
I!
:
:
:
-30
0
30
:
:
:
:
:
;
SO
SO
120
150
180
210
:
240
;
270
J
MINUTES +
HIGH P:S
+
LOW P:S
Fig 4. Protein free energy expenditure of lean (A) and obese (B) subjects during BEE (-30 minutes) and at 30.minute intervals after the breakfast test meal while consuming high and low P:S fat ratio diets. 0, High P:S; -_*-, low P:S.
JONES ET AL
“IS”
P:S
LOW I?*
LEAN
Me”
P:S
LOW P:8
OBESE
Fig 5. Cumulative caloric expenditure above basal over the 270minute test period for individuals of each group on each diet, expressed as percent of meal energy oxidized and as net energy oxidized. Values are means f SEM.
Although postprandial shifts in fat oxidation were observed between obese and normal weight subjects, no significant diet effect on the resting RQ or thermogenic response was seen in our normal weight individuals, as was reported in a similar previous study.” In the present study, it would be anticipated that the added precision of two measurement periods, in comparison to the earlier experiments restricted to a single measure, should have enabled reproduction of the former study’s RQ and energy utilization differences. However, our subjects self-selected their own diets and arranged their own transport to the testing facility on the morning of each trial. Thus, perhaps either the more rigorous dietary intervention approach or the overnight stay before each test day used in the earlier study” was responsible for a metabolic response of adequate proportion to exceed analytical and random biological error. Presently, the within-subject variability in RQ on each diet was high (CV, 3.9%), approximately half of which was attributable to biological variation. The respiratory gas exchange analyzer used presently had overall measurement errors for oxygen and carbon dioxide of less than 2%.‘l Thus, the inconsistency of our results in normal-weight individuals with those of the similar study performed previously”’ likely relate to factors other than the gas analyzer errors. The reasons for the observed changes in fat partitioning observed in obese, but not in normal-weight, individuals, as shown previously,” remain to be determined. Obesity is associated with sympathetic nervous system activity alterations in animals,24sZ and with lower circulating catecholamine levels in overweight humans.” Perhaps dietary long-chain fatty acids evoke a structure-dependent response of sympathetic activity or catecholamine release. Alternatively, intrinsic genetic differences between individuals may result in an altered metabolic response toZ dietary saturated fat in those predisposed to obesity. The trend in basal fat oxidation rates in the obese to be higher with
saturated versus unsaturated fat feeding may arise from one of these effects. A similar elevation in resting fat oxidation was previously observed in a study performed in normal subjects consuming diets high in saturated fat.“’ The controversial notion that impaired postprandial thermogenesis contributes to obesity is not supported by the results of this study, although there was a trend toward reduced TEF caloric expenditure in obese subjects on the low P:S fat ratio diet (Fig 5). Several reports demonstrate a significantly diminished rate of increase in energy expenditure following a meal in overweight subjects”~L4; however, others show no difference.“.” The response to dietary fat of fat oxidation above basal levels, observed presently suggests that diet composition may be a factor responsible for the disparate findings of these studies. Although not significantly different, trends suggest that obesity may be associated with impaired postprandial thermogenesis when diets high in saturated fatty acids are consumed, yet may manifest a response more similar to that of lean subjects in diets with polyunsaturated fats. Also, it cannot be ruled out that the differences in the postprandial measurement interval, and the use of liquid test diets, may influence results of these conflicting studies. In comparing metabolic responses of individuals differing in body composition, the size of the dietary challenge to which TEF is determined has been shown to be critical.” The magnitude of the metabolic response to a meal clearly depends on its caloric content.27.28 In the present study, energy content of the meal was based on ideal body weight, reflected in the mean meal caloric content differing by less than 1% between groups. However, meal size varied within each group; therefore, data comparisons were made between groups on an absolute and relative load basis. Segal et al” have demonstrated that whether meals are fed as a constant caloric load or in a manner dependent on body weight, thermogenic response differences between lean and obese subjects are consistent. It was therefore considered appropriate that, to minimize errors in estimation of exact body composition, subjects be fed test meals containing calories proportioned relative to ideal body weight. Subjects were putatively consuming diets similar in composition to those of meals given during test periods. As a means of verification, 3-day food records were maintained, which, for both lean and obese individuals, indicated substantial shifts in the nature of the fat consumed. As an independent index of compliance, erythrocyte L:O acid ratio responses to diet were examined. L:O acid ratio changes have been suggested as a reliable indicator of recent dietary fat intake patterns.” With the exception of three obese subjects, the L:O acid ratio of all individuals changed in the direction consistent with reported food record intakes. The interpretation, in the cases of absence of anticipated change, was not clear. At first glance, lack of response likely signifies lack of compliance. However, in the obese, erythrocyte fatty acid profiles may require a longer interval than that presently provided to reflect dietary fat patterns. Although significant changes in erythrocyte fatty acids have been reported over 8 days,29 a period of greater
FAT P:S RATIO AND ENERGY METABOLISM
401
IN OBESITY
than 3 weeks is required in normal-weight individuals for erythrocyte fatty acids to fully equilibrate to dietary fat intake.” More data are required to delineate the utility of erythrocyte fatty acid profiles in obese subjects. Presently, since food intake records indicated the appropriate P:S fat ratio changes between periods, the three obese nonresponders were included in the obese group. Statistical comparison without this subgroup produced identical significant differences in fat oxidation over the basal level. In summary, although our data do not show defective thermogenesis in obesity, a trend toward a blunted TEF in
obese subjects given the low P:S fat ratio breakfast was observed. The reduced contribution of fat oxidation to TEF observed in obese subjects consuming the low P:S fat ratio diets suggests that, although the total thermogenic response remains unaffected, overweight individuals partition less dietary saturated fat for oxidation postprandially, compared with individuals of normal body weight. ACKNOWLEDGMENT
The technical assistance of Laurel Aberhardt, Alexander Benson, Terri Broughton, and May Lee is greatly appreciated.
REFERENCES
1. Cenedella RJ, Allen A: Differences between the metabolism of linoleic and palmitic acid: Utilization for cholesterol synthesis and oxidation to respiratory CO,. Lipids 4:155-158,1969 2. Toorop AI, Romsos DR, Leveille GA: The metabolic fate of dietary 1-[‘4C]linoleate and 1-[14C] palmitate in meal fed rats. Proc Sot Exp Biol Med X0:312-316,1979 3. Leyton J, Drury PJ, Crawford MA: Differential oxidation of saturated and unsaturated fatty acids in vivo in the rat. Br J Nutr 57:383-393,1987 4. Forsgren L: Expiratory pattern of 14C0, in man after feeding “‘C-1abelled fatty acids. Arch Kemi 30~355-360,1968 5. Watkins JB, Klein PD, Schoeller DA, et al: Diagnosis and differentiation of fat malabsorption in children using 13Clabeled lipids; trioctanoin, triolein and palmitic acid breath tests. Gastroenterology 82911-917, 1982 6. Jones PJH, Pencharz PB, Clandinin MT: Whole body utilization of dietary fatty acids: Implications for energy utilization. Am J Clin Nutr 42:169-777,1985 7. Shimomura Y, Tamura T, Suzuki M: Less body fat accumulation in rats fed a safflower oil diet than rats fed a beef tallow diet. J Nutr 120:1291-1296,199O 8. Mercer SW, Trayhurn P: Effect of high fat diets on energy balance and thermogenesis in brown adipose tissue of lean and genetically obese ob/ob mice. J Nutr 117:2147-2153,1987 9. Jones PJH: Effect of fatty acid composition of dietary fat on energy balance and expenditure in hamsters. Can J Physiol Pharmacol67:994-998,1989 10. Jones PJH, Schoeller DA: Polyunsaturated:saturated ratio of dietary fat influences energy substrate utilization in the human. Metabolism 37:145-151,198s 11. Katzeff HL, Danforth E: Decreased thennic effect of a mixed meal during overnutrition in human obesity. Am J Clin Nutr 50:915-921,1989 12. Segal KR, Edano A, Bland0 L, et al: Comparison of thermic effects of constant and relative caloric loads in lean and obese men. Am J Clin Nutr 51:14-21,199O 13. Shetty PS, Jung RT, James WPT, et al: Postprandial thermogenesis in obesity. Clin Sci 60:519-525,198l 14. Kaplan ML, Leveille GA: Caloric response in obese and nonobese women. Am J Clin Nutr 29:1108-1113,1976 15. Welle SL, Campbell RG: Normal thermogenic effect of food in obese women. Am J Clin Nutr 37:87-92,1983
16. Schutz Y, Bessard T, Jequier E: Exercise and postprandial thermogenesis in obese women before and after weight loss. Am J Clin Nutr 45:1424-1432,1987 17. D’Alessio DA, Kavle EC, Mozzoli MA, et al: Thermic effect of food in lean and obese men. J Clin Invest 81:1781-1789,1988 18. Durnin JVGA, Womersley J: Body fat assessed from total body density and its estimation from skinfold thickness: Measurements on 481 men and women aged from 16 to 72 years. Br J Nutr 32:77-97,1974 19. Glatz JFC, Soffers AEMF, Katan MB: Fatty acid composition of serum cholesterol esters and erythrocyte membranes as indicators of linoleic acid intake in man. Am J Clin Nutr 49:269216,1989 20. Bell L, Jones PJH, Telch J, et al: Prediction of energy needs for clinical studies. Nutr Res 5:123-129,1985 21. Phang PT, Rich T, Ronco J: A validation and comparison study of two metabolic monitors. J Parent Enteral Nutr 14:259-261, 1989 22. Weir WE: New methods for calculating metabolic rate with special emphasis to protein metabolism. J Physiol 109:1-9,1949 23. Bannon CD, Craske JD, Hai NT, et al: Analysis of fatty acid methyl esters with high accuracy and reliability. II. Methylation of fats and oils with boron trifluoride-methanol. J Chromotgr 247:6369,1982 24. Sakaguchi T, Arase K, Bray GA: Sympathetic activity and food intake of rats with vetromedial hypothalamic lesions. Int J Obes 12:43-49,1988 25. Peterson HR, Rothschild M, Weinberg CR, et al: Body fat and the activity of the autonomic nervous system. N Engl J Med 318:1077-1083,1988 26. Bouchard C, Tremblay A: Genetic effects in human energy expenditure components. Int J Obes 14:45-54,199O 27. Hill JO, Heymsfield SB, McManus C, et al: Meal size and thermic response to food in male subjects as a function of maximal aerobic capacity. Metabolism 33:743-759,1984 28. Belko AZ, Barbieri TF, Wong EC: Effect of energy intake and protein intake and exercise intensity on the thermic effect of food. Am J Clin Nutr 43:863-869,1986 29. Farquhar JW, Amens EH: Effects of dietary fats on human erythrocyte fatty acid patterns. J Clin Invest 42675-685, 1963
