点击显示 收起
1 From the Research Department of Human Nutrition, the Centre for Advanced Food Studies, The Royal Veterinary and Agricultural University, Frederiksberg, Denmark; the Institutes of Biochemistry and Nutrition and of Biotechnology, the Technical University of Denmark, Lyngby, Denmark; and The Danish National Library of Science and Medicine, Copenhagen.
2 Supported by the Danish Research Councils (FELFO) and a joint grant from The Danish Ministry of Culture and The Danish Research Academy. 3 Address reprint requests to H Bendixen, Research Department of Human Nutrition, The Royal Veterinary and Agricultural University, Rolighedsvej 30-DK-1958, Frederiksberg C, Denmark. E-mail: hb{at}kvl.dk.
ABSTRACT
Background: Different dietary fats are metabolized differently in humans and may influence energy expenditure, substrate oxidation, appetite regulation, and body weight regulation.
Objective: We examined the short-term effects of 4 triacylglycerols (test fats) on subjective appetite, ad libitum energy intake, meal-induced thermogenesis, and postprandial substrate oxidation.
Design: Eleven healthy, normal-weight men (mean age: 25.1 ± 0.5 y) consumed 4 different test fats [conventional fat (rapeseed oil) and 3 modified fats (lipase-structured fat, chemically structured fat, and physically mixed fat)] in a randomized, double-blind, crossover design.
Results: No significant differences in appetite sensations or ad libitum energy intakes were observed between the 4 test fats. Overall, the 4 fats exerted different effects on energy expenditure (meal effect: P < 0.01) and substrate oxidation (interaction between meal and time: P < 0.05). In post hoc tests, the 3 modified fats resulted in significantly higher postprandial energy expenditure and fat oxidation than did the conventional fat (P < 0.008, Bonferroni adjusted); no significant differences were observed between the 3 modified fats.
Conclusions: Structured fats do not change short-term postprandial appetite sensations or ad libitum energy intakes but do result in higher postprandial energy expenditure and fat oxidation than do conventional fats and hence promote negative energy and fat balance. In humans, a physically mixed fat (trioctanoate + rapeseed oil) is metabolized as quickly as are structured fats. The position of medium-chain fatty acids on the glycerol backbone of triacylglycerols does not seem to affect energy expenditure or appetite.
Key Words: Obesity • dietary fat • medium-chain fatty acids • long-chain fatty acids • structured fats • modified fats • conventional fats • appetite • energy intake • energy expenditure • substrate oxidation • men
INTRODUCTION
The increasing prevalences of weight gain, obesity, and type 2 diabetes are causally related to physical inactivity and high-fat diets (1–3). A reduction in total dietary fat intake is recommended for preventing and treating these disorders. The theory that a reduction in dietary fat causes a modest, dose-dependent decrease in body weight is supported by both prospective observational studies (4) and meta-analyses of intervention studies (1, 2, 5). Although the proportion of dietary energy obtained from fat has decreased slightly in the general population in recent years, the preference for high-fat foods in large segments of the population makes it difficult to achieve a substantial reduction in fat intake. High-fat diets seem to play a unique role in promoting weight gain by inducing an unintentional passive overconsumption of energy (6–9). This effect of fat is partially linked to its high energy density, partly because of fat-specific effects on metabolism and appetite.
Observational studies and some metabolic studies suggest that differences in fat quality, such as the degree of saturation and fatty acid chain length, influence the ability of fat to be readily oxidized (10–18) and to induce satiety (19–21). These findings have created a strong interest in developing structured triacylglycerols, hereafter referred to as structured fats, that are more satiating and less likely to produce a positive energy balance in humans (22). Structured fats are synthesized triacylglycerols containing mixtures of fatty acids of varying chain length, typically medium-chain fatty acids (MCFAs) and long-chain fatty acids (LCFAs) attached to the same glycerol backbone. These fats may be synthesized by lipase-catalyzed interesterification, resulting in specific location of MCFAs in the sn-1,3-position or by chemical interesterification, resulting in random distribution of fatty acids on the same glycerol backbone (23).
No studies have compared the effects of different kinds of structured fats on human energy expenditure (EE), substrate oxidation, and appetite or of the effect on these 3 outcomes of fatty acid position on the glycerol backbone. Hence, the aim of the present study was to examine the short-term effects of different selected modified fats on appetite, ad libitum energy intake, meal-induced thermogenesis, and postprandial substrate oxidation in healthy subjects.
SUBJECTS AND METHODS
This study was carried out at the Research Department of Human Nutrition, The Royal Veterinary and Agricultural University, Frederiksberg, Denmark. The Municipal Ethical Commitee of Copenhagen and Frederiksberg approved the study, finding it to be in accordance with the Helsinki-II declaration. All subjects gave their written consent after the experimental procedure had been explained to them.
Subjects
Eleven healthy, nonsmoking men with a mean (±SEM) age of 25.1 ± 0.5 y (range: 22–28 y) were recruited by advertisement and enrolled. None of the subjects were elite athletes. The subjects' mean body weight was 77.0 ± 2.9 kg (range: 57.9–88.6 kg), height was 1.85 ± 0.03 m (range: 1.61–1.93 m), and body mass index (BMI; in kg/m2) was 22.5 ± 0.6 (range: 18.9–25.0). Body composition was estimated by using the bioimpedance method. Fat-free mass and percentage body fat were calculated by using Heitmann's equation (24) and were determined to be 62.9 ± 2.0 kg (range: 50.5–70.0 kg) and 18.0 ± 1.0% (range: 11.3–21.9%), respectively. The subjects' energy requirements were estimated by using World Health Organization tables according to age, weight, height, and sex (25); a multiplication factor of 1.78 was used to account for the medium physical activity of the subjects. Their mean energy requirements were calculated to be 13.7 ± 0.3 MJ/d (range: 12.0–15.0 MJ/d).
Test fats
Four test fats were used: 1 conventional fat (rapeseed oil) and 3 modified fats (lipase-structured fat, chemically structured fat, and physically mixed fat). The lipase-structured fat was made from rapeseed oil (Aarhus Oliefabrik A/S, Aarhus, Denmark) and octanoic acid (Sigma Chemical Co, St Louis) by interesterification in a batch reactor with Lipozyme IM (NovoNordisk A/S, Bagsværd, Denmark) as the active enzyme, as previously described (26). The chemically structured fat was made from rapeseed oil and octanoic acid by interesterification with sodium methoxide as the catalyst. The physically mixed fat was made by blending rapeseed oil and trioctanoate (Grünau GmbH, Illertissen, Germany). The conventional fat was rapeseed oil. The regiospecific locations of the fatty acids in the triacylglycerols were determined by Grignard degradation and gas-liquid chromatography of the sn-2-monoacylglycerols. The overall fatty acid profiles were determined by converting the fatty acids into methyl esters and using gas-liquid chromatography for the analyses (27).
The overall fatty acid profiles of the 3 modified fats were similar (Tables 1 and 2), but in the lipase-structured fat the octanoic acid was solely located in the sn-1,3-positions, whereas in the chemically structured fat the fatty acids were randomly distributed on the glycerol backbone. In the physically mixed fat, the octanoic acid was always in the form of trioctanoate. The conventional fat contained no octanoic acid.
View this table:
TABLE 1 . Mean contents of 18-carbon and 8-carbon fatty acids in the 4 different test fats
View this table:
TABLE 2 . Fatty acid profiles: total fatty acid composition and the sn-2 position in the 4 different test fats
Diets
The computer database of foods from the National Food Agency of Denmark (DANKOST 2.0; Danish Food Tables 1989; Levnedsmiddeltabeller) was used to calculate the energy and nutrient composition of the diets.
Standard diet
On the day before the test day, each subject was given a standardized, carbohydrate-rich, weight-maintenance diet. The composition of the standard diet was 50% of energy from carbohydrate (ratio of sugars to starch: 0.08), 37% of energy from fat, and 13% of energy from protein. The subjects were instructed to adhere strictly to the diet, which consisted of ordinary food items. The standard diet was prepared at the research department according to each subject's individual energy requirements, adjusted to the nearest 0.5 MJ.
Test meal
The test meal was given as a breakfast. The percentages of total energy from the macronutrients were 34% carbohydrate (ratio of sugars to starch: 0.6), 60% fat (91% of the fat was the test fat), and 6% protein. The meal consisted of bread, jam, 2 slices of orange, and a chocolate drink, which contained the test fat. The drink was served in a closed container with a straw and was shaken thoroughly before serving. The amount of test fat given corresponded to 0.9 g/kg body weight, measured on each test day. The total energy content of the breakfast was 4698 ± 174 kJ (range: 3558–5618 kJ), which corresponded to 35.0 ± 0.7% (range: 29.7–37.5%) of the subject's daily energy needs.
Ad libitum lunch
The ad libitum lunch, consumed on the test day, was a homogeneous, mixed, hot meal that did not contain any test fat. The percentages of total energy from the macronutrients were 50% carbohydrate (ratio of sugars to starch: 0.08), 37% fat, and 13% protein. The meal included pasta, tomato purée, minced meat, green pepper, carrots, squash, onions, corn, and cream. The subjects could eat these foods ad libitum.
Experimental protocol
Each subject participated in 4 different tests separated by 2–4 wk in a randomized, double-blind, 4-way, crossover design. The sequence of the test fats corresponded to an orthogonal Latin-squared design.
The variation between measurements was minimized by having the subjects follow exactly the same procedures on each test day. The subjects were instructed to abstain from strenuous physical activity and alcohol consumption during the 2 d preceding each test day to ensure that glycogen stores and macronutrient balances would be equivalent on all test days (28).
On the test day, the subjects arrived at the Research Department of Human Nutrition in the morning. They had used the least strenuous method of transportation possible and had fasted for 12 h. Subjects voided and were weighed to the nearest 100 g. Their body composition was measured with bioimpedance by using an Animeter (HIS-Engineering Inc, Odense, Denmark). A tensiometer (Polar Accurex Plus TM; Polar Electro, Oulu, Finland) was attached to each subject to measure the pulse continuously during the test day.
The subjects then rested on a bed in the supine position with their heads slightly elevated; the bed was covered with an antidecubitus mattress. After a 30-min rest period, the subjects' fasting EE was measured for 45 min by using a ventilated-hood system. The test meal was served and eaten within 15 min. Each subject spent exactly the same amount of time consuming each of the 4 meals (13.1 ± 0.5 min on average).
After consumption of the test meal, EE was measured. During the postprandial measurements, the subjects were allowed to watch movies for light entertainment or listen to the radio. Each hour, the subjects had a 10-min break from wearing the hood, during which time they rested on the bed. The subjects were allowed to drink water (maximum of 200 mL total) and to visit the toilet during the breaks. The exact hour, type of activity, and amount of water consumed on the first of the test days were recorded and were then repeated on the next 3 test days. Urine was collected during the day (morning urine was subtracted from the total).
Postprandial EE was measured for 5 h, after which the ad libitum lunch was served and was eaten within 15 min. The subjects were instructed to eat until they reached a point of comfortable satisfaction, and the amount of food consumed was recorded. The subjects could have more servings on request. Each subject spent the same amount of time eating each of the 4 lunches (11.5 ± 0.8 min on average).
Questionnaires
Each subject completed questionnaires to assess subjective hunger, satiety, fullness, prospective food consumption, thirst, comfort, and desire to eat something fatty or something savory. Subjects completed these questionnaires immediately before the test meal (breakfast) and 15, 30, 45, 60, 90, 120, 150, 180, 210, 240, 270, 300, and 330 min after the test meal. Ratings were made on 100-mm visual analogue scales with words at each end that expressed the 2 most extreme ratings, as described by Flint et al (29). Immediately after consumption of the test meal and the ad libitum lunch, subjects rated the palatability of the meals (appearance, smell, taste, aftertaste, and overall palatability) by using visual analogue scales. The questionnaires were booklets that showed one question at a time.
Energy expenditure and substrate oxidation
EE was measured by indirect calorimetry with an open-air-circuit, computerized, ventilated-hood system (30). Carbon dioxide was measured with a Uras 10P analyzer (Hartmann and Braun, Frankfurt, Germany) and oxygen was measured with a Servomex 1100A paramagnetic analyzer (Servomex Ltd, Crowborough, United Kingdom). Ventilation through the system was measured with a mass flow meter (type HFM 201-100; Teledyne Hastings-Raydist, Hampton, VA).
The gas analyzers were calibrated before every run by using atmospheric gas and 2 span scaling gases. During the measurements, the analyzers were corrected for drift every hour by using the atmosphere as a reference.
EE and oxidation of carbohydrate, fat, and protein were calculated from the gas exchange and urinary nitrogen measurements by using the constants of Elia and Livesey (31). Protein oxidation was assumed to be constant during the test day.
RESULTS
Appetite, thirst, comfort, and desire for specific types of food
The response curves for the 4 appetite sensations (satiety, hunger, fullness, and prospective food consumption) and for thirst, comfort, and desire to eat something fatty or something savory are shown in Figures 1 and 2. The profiles were similar after the 4 test meals, and no significant differences between test fats were found. The postprandial means for subjective ratings of the 4 appetite sensations and of thirst, comfort, and desire to eat something fatty or savory also did not differ significantly between the test fats (Table 3). Fasting values for all variables in Table 3 did not differ significantly between test fats. There were no significant differences in peak or nadir values or slopes of the regression lines of the appetite variables.
FIGURE 1. . Mean subjective appetite scores (satiety, hunger, fullness, and prospective food consumption) in 11 healthy, normal-weight men before (0 min) and after consumption of 4 different test fats: , conventional fat (rapeseed oil); , lipase-structured fat (lipase interesterified rapeseed oil + octanoic acid); , chemically structured fat (chemically interesterified rapeseed oil + octanoic acid); and x, physically mixed fat (physical mixture of rapeseed oil + trioctanoate). There were no significant differences between the test fats by ANOVA. VAS, visual analogue scale.
FIGURE 2. . Mean subjective scores for the desire to eat something fatty or something savory and for thirst and comfort in 11 healthy, normal-weight men before (0 min) and after consumption of 4 different test fats: , conventional fat (rapeseed oil); , lipase-structured fat (lipase interesterified rapeseed oil + octanoic acid); , chemically structured fat (chemically interesterified rapeseed oil + octanoic acid); x, physically mixed fat (physical mixture of rapeseed oil + trioctanoate). There were no significant differences between the test fats by ANOVA. VAS, visual analogue scale.
View this table:
TABLE 3 . Subjective ratings (scores) on a visual analogue scale (VAS) of 4 appetite sensations and of thirst, comfort, and desire to eat something fatty or savory 5 h after consumption of the test meals containing 4 different test fats 1
Ad libitum energy intake
No significant differences in ad libitum energy intakes between the test fats were seen. These intakes averaged 3.65 ± 0.38, 3.41 ± 0.30, 3.42 ± 0.32, and 3.34 ± 0.40 MJ after the conventional, chemically structured, lipase-structured, and physically mixed fats, respectively.
Palatability
Ratings of the appearance, smell, taste, aftertaste, and overall palatability of the 4 test meals and the corresponding ad libitum lunches are shown in Table 4. There were no significant differences in these palatability ratings between test fats, although a significant interaction between treatment and period was found for taste of both the test meal and the ad libitum lunch.
View this table:
TABLE 4 . Palatability ratings (scores) on a visual analogue scale (VAS) of the test meals containing the 4 different test fats and of the corresponding ad libitum lunches1
Energy expenditure
Fasting EE was not significantly different before the 4 different test meals were consumed, averaging 4.84 ± 0.15, 4.83 ± 0.12, 4.87 ± 0.14, and 4.83 ± 0.18 kJ/min before the conventional, chemically structured, lipase-structured, and physically mixed fats, respectively.
EE increased after all 4 test fats, but the increase was significantly smaller after the conventional fat than after the 3 modified fats (P < 0.01) (Figure 3). DIT averaged 309 ± 24 kJ/5 h (6.5 ± 0.5%), 364 ± 30 kJ/5 h (7.6 ± 0.5%), 346 ± 18 kJ/5 h (7.2 ± 0.3%), and 325 ± 22 kJ/5 h (6.8 ± 0.4%) after the conventional, chemically structured, lipase-structured, and physically mixed fats, respectively (treatment effect: P < 0.05). This corresponded to a 5–17% greater DIT after the 3 test meals containing modified fats than after the meal containing the conventional fat. The difference between the conventional fat and the chemically structured fat was significant (P = 0.005), whereas the other pairwise comparisons were not significant.
FIGURE 3. . Left panel: mean energy expenditure in 11 healthy, normal-weight men before (0 min) and after consumption of 4 different test fats: , conventional fat (rapeseed oil); , lipase-structured fat (lipase interesterified rapeseed oil + octanoic acid); , chemically structured fat (chemically interesterified rapeseed oil + octanoic acid); and x, physically mixed fat (physical mixture of rapeseed oil + trioctanoate). There was a significant treatment effect (P = 0.0022) and time effect (P < 0.0001) by ANOVA. The treatment x time interaction was not significant. Right panel: Diet-induced thermogenesis (DIT). , conventional fat (rapeseed oil); , chemically structured fat (chemically interesterified rapeseed oil + octanoic acid); , lipase-structured fat (lipase interesterified rapeseed oil + octanoic acid); , physically mixed fat (physical mixture of rapeseed oil + trioctanoate). There was a significant treatment effect (P < 0.05) by ANOVA. In the post hoc tests, there was a significant difference between the conventional fat and the chemically structured fat (P = 0.005), but the other pairwise comparisons were not significant.
An overall significant difference in the time when peak values of EE occurred was observed between test fats by ANOVA: conventional fat, 180 min; chemically structured fat, 60 min; lipase-structured and physically mixed fats, 120 min. Post hoc tests comparing the test fats showed a significant difference between the conventional fat and the chemically structured fat (P = 0.005), whereas the other pairwise comparisons were not significant.
Substrate oxidation
Fat oxidation and carbohydrate oxidation did not differ significantly before consumption of the 4 test meals. Significant interactions between treatment and time were observed for postprandial fat and carbohydrate oxidation (Figure 4). At the first postprandial time point (60 min), a decrease in fat oxidation was observed after the conventional fat meal, whereas increases in fat oxidation were observed after the other 3 test fats. All 4 test fats resulted in increases in postprandial carbohydrate oxidation, with the highest increase occurring after the conventional fat meal. The overall differences in substrate oxidation were the result of significant differences between the conventional fat and lipase-structured, chemically structured, and physically mixed fats at 60 min and to a lesser extent at 120 min, the second postprandial time point.
FIGURE 4. . Left panels: mean fat and carbohydrate oxidation in 11 healthy, normal-weight men before (0 min) and after consumption of 4 different test fats: , conventional fat (rapeseed oil); , lipase-structured fat (lipase interesterified rapeseed oil + octanoic acid); , chemically structured fat (chemically interesterified rapeseed oil + octanoic acid); and x, physically mixed fat (physical mixture of rapeseed oil + trioctanoate). For fat oxidation, there was a significant treatment x time interaction (P = 0.0025) by ANOVA. In the post hoc tests, the comparisons between the conventional fat and the lipase-structured, chemically structured, and physically mixed fats were as follows: 0 min, P < 0.001, P < 0.0001, and P < 0.0001, respectively; 60 min, P = 0.10, P < 0.001, and P < 0.008, respectively (significance was defined as P < 0.008). For carbohydrate oxidation, there was a significant treatment x time interaction (P = 0.0287) by ANOVA. In the post hoc tests, the comparisons between the conventional fat and the lipase-structured, chemically structured, and physically mixed fats were as follows: 0 min, P < 0.008, P = 0.02, and P < 0.0001, respectively; 60 min, P = 0.89, P = 0.26, and P = 0.04, respectively (significance was defined as P < 0.008). Right panels: net fat and carbohydrate oxidation in 11 healthy, normal-weight men before (0 min) and after consumption of 4 different test fats: , conventional fat (rapeseed oil); , chemically structured fat (chemically interesterified rapeseed oil + octanoic acid); , lipase-structured fat (lipase interesterified rapeseed oil + octanoic acid); and , physically mixed fat (physical mixture of rapeseed oil + trioctanoate). The treatment effect for both net fat and carbohydrate oxidation was not significant (ANOVA).
Net postprandial fat oxidation averaged 1.65 ± 1.4, 3.5 ± 1.3, 3.1 ± 1.2, and 3.3 ± 1.6 g/5 h after the conventional, chemically structured, lipase-structured, and physically mixed fat meals, respectively (NS; Figure 4). Net postprandial carbohydrate oxidation averaged 13.9 ± 3.5, 13.1 ± 2.7, 12.9 ± 2.9, and 11.1 ± 2.9 g/5 h after the conventional, chemically structured, lipase-structured, and physically mixed fat meals, respectively (NS; Figure 4). Total protein oxidation averaged 6.0 ± 0.4, 5.8 ± 0.4, 5.7 ± 0.4, and 5.7 ± 0.3 g/5 h after the conventional, chemically structured, lipase-structured, and physically mixed fat test meals, respectively (NS).
DISCUSSION
In recent years, research has increasingly indicated that different types of fat are metabolized differently by the human body. Thus, different dietary fat compositions exert different effects on energy expenditure, substrate oxidation, and appetite, which in turn influence body weight regulation. Fatty acid composition varies in terms of chain length and degree of saturation Furthermore, the positions of individual fatty acids on the glycerol backbone may vary.
In the present study, consumption of meals containing modified fats with MCFAs instead of conventional fats resulted in higher postprandial EE and fat oxidation and lower carbohydrate oxidation in healthy humans; these are beneficial effects with regard to body weight regulation. These differences were mostly a result of differences between the conventional fat and the chemically structured fat. However, the differences between these 2 fats were of limited magnitude; therefore, we do not consider the chemically structured fat to be metabolized faster or more easily than the other 2 modified fats.
Our results for EE and substrate oxidation are generally consistent with previous findings that consumption of a meal containing medium-chain triacylglycerols (MCTs; 6–13 carbons) results in higher postprandial EE than does consumption of an isoenergetic meal containing long-chain triacylglycerols (LCTs; 14–24 carbons) (10, 12, 15, 16, 18). White et al (35) also found higher postprandial EE after 7 d of an MCT-supplemented diet than after 7 d of an LCT-supplemented diet, although this effect was attenuated after 14 d. The observed differences between MCTs and LCTs have been explained in terms of the different metabolic properties of these triacylglycerols (36–38).
Thus, MCTs are absorbed more quickly and efficiently than are LCTs, mainly because of the hydrophilic properties of the MCFAs. MCFAs are absorbed into the intestinal cell and are further directed to the portal blood because they do not form CoA esters and are therefore not used for triacylglycerol formation. From the portal blood, the MCFAs are transported directly to the liver. Because long-chain-fatty-acid–CoA ligase (EC no. 6.2.1.3) has low activity toward MCFAs, they do not form CoA esters and thus cannot be incorporated into triacylglycerols. Instead, they are directed toward mitochondrial oxidation because they enter the mitochondria independent of carnitine. In contrast, the LCFAs are used during absorption for triacylglycerol synthesis and are subsequently incorporated into chylomicrons and transported with the lymph.
Normally, only small amounts of MCFAs occur in tissues because of the limited formation of triacylglycerols and the preferred mitochondrial oxidation. Because MCTs are more readily absorbed and oxidized, MCTs may suppress appetite to a greater extent than do LCTs. In the present study, the oxidation of the 3 modified fats was faster than the oxidation of the conventional fat. It was therefore not surprising that energy intakes at the subsequent ad libitum meal tended to be lower after the test meals containing the 3 modified fats than after the meal containing the conventional fat, although these differences were not significant. This finding contrasts with results from previous research in which an MCT-rich diet resulted in significantly lower ad libitum energy intakes over a 2-wk period than did an isoenergetic LCT-rich diet (20). This was also found in a short-term study in which an MCT-rich breakfast was associated with lower ad libitum energy intakes at a subsequent meal than after an LCT-rich breakfast (19) or after liquid test meals containing MCTs or LCTs (21). The discrepancy between these findings and our results could be attributed to differences in study designs and in the composition and properties of the fats used.
In the present study, appetite sensations after consumption of the 4 test fats did not differ significantly; this was consistent with the ad libitum energy intakes. However, when investigating subjective appetite sensations and energy intake, it is important to keep in mind that it can be difficult to detect differences between treatments because of type II error or a low study power resulting from a relatively low number of subjects (29), as was the case in the present study. On the other hand, the response curves for the 4 test fats were very similar for each appetite sensation. It is therefore doubtful that inclusion of more subjects in this study would have led to significant differences in appetite between the test fats. However, a greater number of subjects might have resulted in significant differences in ad libitum energy intakes.
No studies have compared the effects of different kinds of structured triacylglycerols on human EE, substrate oxidation, and appetite. Furthermore, no experiments have investigated the effect on these 3 variables of the positions of fatty acids on the glycerol backbone. As described by Bell et al (22), the fatty acid that is in the sn-2 position on the glycerol molecule is absorbed from the intestine as the 2-monoglyceride, whereas fatty acids in the sn-1 and sn-3 positions are absorbed as free fatty acids. The resynthesis of LCTs proceeds via the monoglyceride pathway by using the sn-2 monoglyceride as a template, forming a new population of triacylglycerol molecules. When structured lipids are absorbed, MCFAs are directed toward the liver through the portal vein. Therefore, it is necessary to use endogenous fatty acids to acylate the absorbed monoglycerides and form new triacylglycerols. The fatty acids located in the sn-2 position of the dietary fat are retained in chylomicron lipid particles. These lipid particles are transported by the lymph to the systemic circulation, to the liver, and eventually to the peripheral tissues. The chylomicrons are degraded by lipoprotein lipase, releasing the fatty acids in the sn-1,3 positions for extrahepatic deposition or oxidation. Hence, the fatty acid in the sn-2 position is least likely to be lost by extrahepatic oxidation. Conceivably, a difference in the nutrient effect of the individual triacylglycerol could be derived from the deviating metabolic pathways followed by MCFAs during absorption and the fatty acids located in the sn-2 position compared with the fatty acids in the sn-1,3 positions.
In the present study, no differences in EE, substrate oxidation, ad libitum energy intakes, or appetite were found between the 3 modified fats. These results indicate that the position of MCFAs on the glycerol backbone of triacylglycerols does not have an effect on these variables. Furthermore, our results showed that in humans, the modified fat consisting of a physical mixture of trioctanoate and a conventional fat was metabolized as quickly as were the structured fats. This might be important in the future use of modified fats because the synthesis of modified fats as physical mixtures may have considerable economic and practical advantages. In conclusion, structured fats do not change short-term postprandial appetite sensations or ad libitum energy intakes but do result in higher postprandial EE and fat oxidation than do conventional fats and hence promote negative energy and fat balance.
ACKNOWLEDGMENTS
We thank John Lind, Inge Timmermann, Charlotte Kostecki, Karina G Rossen, Berit Hoeilt, Yvonne Rasmussen, and Grete Peitersen for their excellent technical assistance.
REFERENCES