International Journal of Obesity (2000) 24, 1158±1166 ß 2000 Macmillan Publishers Ltd All rights reserved 0307±0565/00 $15.00 www.nature.com/ijo

Endogenous fat oxidation during medium chain versus long chain triglyceride feeding in healthy women AA Papamandjaris1, MD White1, M Raeini-Sarjaz1 and PJH Jones1* 1

School of Dietetics and Human Nutrition, Faculty of Agricultural and Environmental Sciences, McGill University, Macdonald Campus, Ste-Anne-de-Bellevue, Quebec, Canada H9X 3V9

OBJECTIVE: To compare the effect of medium chain triglycerides (MCT) vs long chain triglycerides (LCT) feeding on exogenous and endogenous oxidation of long chain saturated fatty acids (LCSFA) in women. SUBJECTS: Twelve healthy female subjects (age 19 ± 26 y, body mass index (BMI) 17.5 ± 28.6 kg=m2) DESIGN AND MEASUREMENTS: In a randomized cross-over design, subjects were fed weight maintenance diets providing 15%, 45% and 40% of energy as protein, carbohydrate and fat, respectively, with 80% of this fat comprising either a combination of butter and coconut oil (MCT) or beef tallow (LCT). Following 6 days of feeding, subjects were given daily oral doses of 1-13C labelled-myristic, -palmitic and -stearic acids for 8 days. Expired 13CO2 was used as an index of LCSFA oxidation with CO2 production assessed by respiratory gas exchange. RESULTS: No difference in exogenous LCSFA oxidation was observed as a function of diet on day 7. On day 14, greater combined cumulative fractional LCSFA oxidation (16.9  2.5%=5.5 h vs 9.1  1.2%=5.5 h, P < 0.007), net LCSFA oxidation (2956  413 mg=5.5 h vs 1669  224 mg=5.5 h, P < 0.01), and percentage dietary LCSFA contribution to total fat oxidation (16.3  2.3%=5.5 h vs 9.5  1.5%=5.5 h; P < 0.01) were observed in women fed the MCT vs LCT diet. With the MCT diet, but not the LCT diet, combined cumulative fractional LCSFA oxidation (P < 0.03), net LCSFA oxidation (P < 0.03), and percentage dietary LCSFA contribution to total fat oxidation (P < 0.02) were increased at day 14 as compared to day 7. Day 14 results indicated increased endogenous LCSFA oxidation during MCT feeding. CONCLUSION: The capacity of MCT to increase endogenous oxidation of LCSFA suggests a role for MCT in body weight control over the long term. International Journal of Obesity (2000) 24, 1158±1166 Keywords: medium chain fatty acids; long chain saturated fatty acids; fat oxidation;

Introduction A growing body of literature suggests metabolic discrimination of fatty acids for oxidation vs storage based on chain length. Both human and animal data indicate that medium chain triglycerides (MCT), containing medium chain fatty acids (MCFA) composed of chains of 8 ± 12 carbon atoms, are preferentially oxidized as compared to long chain triglycerides (LCT) containing long-chain fatty acids (LCFA), composed of chains of 14 or more carbon atoms.1 ± 14 Observations at the whole body level have shown that animals overfed and refed MCT- vs LCT-containing diets had decreased body weight gain.1,3,4,7 Additionally, animal studies of substrate utilization have shown more extensive and rapid oxidation of MCT in comparison to LCT.6,8,13 In humans, greater propensity of MCT vs LCT for oxidation has also been

*Correspondence: PJH Jones, School of Dietetics and Human Nutrition, Faculty of Agricultural and Environmental Sciences, 21 111 Lakeshore Road, McGill University, Macdonald Campus, Ste-Anne-de-Bellevue, Quebec, Canada H9X 3V9. E-mail: [email protected] Received 1 September 1999; revised 23 February 2000; accepted 29 March 2000

13

C breath test; humans

observed.2,5,9 ± 12,14 Yet it still remains to be determined whether the presence of MCT in the diet can alter the oxidation of other meal fatty acids. In addition, the potential capacity of extended MCT feeding to affect endogenous oxidation, whether by changing the composition of the adipose tissue pool through altered endogenous availability, or through induction of enzymes and metabolic processes, has not been determined. The capability of MCT to increase endogenous fat oxidation could have implications for reduction of adipose tissue mass. Given the recent evidence indicating impaired oxidation of LCT yet normal oxidation of MCT in the obese,15 substitution of MCT for LCT may provide a means of reducing adipose tissue deposition and increasing adipose tissue mobilization. The goal of this study, therefore, was to examine effects of MCT- vs LCT-enriched diets on exogenous and endogenous oxidation of myristic (MA), palmitic (PA) and stearic (SA) acids over 8 days, following one week of prefeeding. The speci®c objective was to determine whether consumption of a MCT-enriched diet would affect oxidation of long-chain saturated fatty acids (LCSFA) contained in the diet in a manner different from an LCT-enriched diet. Myristic acid,

Fatty acid chain length and fat oxidation AA Papamandjaris et al

PA and SA were chosen as they represent common saturated fatty acids found in the North American diet and were the major LCSFA in the experimental diets.

Methods Subjects and study design

Twelve healthy, non-smoking female college students, aged 22.7 0.7 y with a height of 1.62  0.01 m and initial body weight of 56.2  1.5 kg, were recruited from Macdonald Campus, McGill University. Subjects were screened for normal blood lipid levels (total cholesterol < 5.3 mmol=l, triglycerides < 1.24 mmol= l), absence of chronic disease, exercise frequency and regular menstrual history. A randomized cross-over design was used involving two 2 week feeding periods separated by a 14 day washout. An equal number of subjects per treatment were tested on each diet cycle and subjects were blinded to diet sequence. All procedures were approved by the McGill University Ethics Committee and informed, written consent was obtained from all subjects prior to their entrance into the study. Experimental diets

Prepared solid food diets providing 15%, 45% and 40% of energy as protein, carbohydrate and fat, respectively, were fed over each 14 day period. Treatment fat, comprising of a combination of butter and coconut oil for the MCT treatment, or beef tallow for the LCT treatment, made up 80% of the dietary fat. Diets (Table 1) were provided as a 2 day rotating menu comprising three meals per day repeated every second day which were isoenergetic both in total energy and fat. The rotating menu was modi®ed to ensure that the same meals were fed on pre-test days, day 6 and 13, and on test days, day 7 and 14. Ingestion of foodstuffs other than those provided was prohibited, except for water, which was permitted ad libitum. All ingredients were weighed to the nearest 0.5 g. Energy intake was calculated to maintain energy balance using the Mif¯in equation16 and an activity factor,17 and diets were designed to meet the Recommended Nutrient Intakes for Canadians.18 Meals were prepared and consumed under supervision at the Mary Emily Clinical Nutrition Research Unit at McGill University.

1159

Protocol

On days 7 and 14 of each diet cycle, following a 12 h fast and overnight stay at the Unit, subjects were awakened at 7:00 am and baseline breath samples collected (Figure 1). Basal metabolic rate (BMR) was measured over 30 min using a Deltatrac Metabolic Monitor (Sensormedics, Anaheim, California). Subjects were then served a hot breakfast containing a treatment-speci®c 10 mg=kg mixture of 1-13C labelled MA, PA and SA (CDN Isotopes, Quebec, Canada) blended into the meal. Labelled LCSFA were given at levels proportional to the levels of tracees, MA, PA, and SA, contained in each diet (Table 2). Thus the fraction of label ingested appearing as 13CO2 represented oxidation of the majority of dietary LCSFA. Over 5.5 h following the breakfast meal, the volume of subjects' expired CO2 was obtained from respiratory gas exchange (RGE). Breath samples were also collected at hourly intervals during this time. Subjects were then fed lunch, delivered a post-lunch breath sample, and a ninth breath sample was collected predinner, at approximately 18:00 h. From days 8 ± 13, the same dose of isotope mixture was delivered with breakfast, and breath samples were collected prior to each meal. Throughout the entire study, subjects kept diaries recording dates of menstruation and type and duration of exercise. Whole body respiratory gas exchange measurements

On days 7 and 14, following single and repeated isotope doses, respectively, O2 consumption and CO2 production for each subject were measured using a Deltatrac Metabolic Monitor as described previously.19 Brie¯y, a transparent ventilated hood was placed over each subject's head, with a hose connecting the hood to the analyser system. After a minimum 30 min warm-up, reference gas standards were used to calibrate the monitor. Respiratory gas exchange measurements were collected for 0.5 h prior to breakfast following a 12 h fast, and for 5.5 h following termination of consumption of the scheduled breakfast.

Table 1 Sample menu used during the dietary interventiona Breakfast Vegetable omelette* Fruit and ®bre cereal 1% milk Orange juice Blueberry muf®n* Raisins a

Lunch Grilled ham and cheese* Diet soda Grapes Oatmeal cookie*

Dinner Spaghetti with meat sauce Parmesan cheese Garlic toast* Celery Fruit juice Date square*

Food items to which treatment fat was added are denoted by*.

Figure 1 Experimental protocol for each 14 day dietary treatment phase. International Journal of Obesity

Fatty acid chain length and fat oxidation AA Papamandjaris et al

1160

Table 2 Fatty acid pro®le and percentage of 1-13C myristic (C14 : 0), palmitic (C16 : 0), and stearic acids (C18 : 0) delivered in treatment dose on medium chain and long chain triglyceride dietary treatmentsa MCT

Fatty acid C8 : 0 C10 : 0 C12 : 0 C14 : 0 C16 : 0 C16 : 1 C18 : 0 C18 : 1 C18 : 1 C18 : 2 a

o7 o9 o7 o6

Percentage of total fatty acid spectrum 3.9 4 17.7 13.3 25.4 1.6 8.1 19.3 0.6 4.8

LCT

13 C labelled fatty acid (g=100 g of total C14 : 0, C16 : 0, C18 : 0)

Percentage of total fatty acid spectrum 0 0.4 1.4 4.4 27.3 2.9 16.9 35.9 0.8 8

28.4 54.2 17.3

13 C labelled fatty acid (g=100 g of total C14 : 0, C16 : 0, C18 : 0)

9 56.2 34.8

LCT ˆ long chain triglyceride; MCT ˆ medium chain triglyceride.

Collection, puri®cation and analysis of carbon dioxide in breath samples

Subjects collected breath samples by exhaling into sealed bags through a sealed 1.3 cm diameter tube equipped with a stopcock. At each collection time, subjects ®lled the bags twice, with only the second exhalation being used as sample. During RGE measurement periods, subjects delivered breath samples while under the hood. A syringe was used to extract 180 ml of air which was then bubbled through a glass condenser, 1.0 m in total length, containing 10 ml 1 M NaOH, to trap the CO2. The samples were then transferred to vials and stored at 720 C until analysis. To release the CO2, 2.5 ml of the NaOH ± CO2 solution at room temperature was injected into vacutainers containing 2.5 ml of o-phosphoric acid 85%. The vacutainers were shaken for 90 s to release the CO2. A SIRA isotope ratio mass spectrometer (SIRA 12, Isomass, Cheshire, UK) was used to determine the 13C enrichment of the samples in delta per mil (%). The mass spectrometer was corrected for 17O and calibrated daily using CO2 gas and standards of known isotopic composition. Enrichments were corrected for background 13C abundance by subtraction of baseline enrichment, the pre-isotope dose enrichment on either day 7 or day 14.

split ratio was 100 : 1. The total run time was 57 min. Individual fatty acids were identi®ed against authentic standards (Sigma Chemical Company, St Louis, MO) using retention times. Mean percentages of MA, PA and SA of total fatty acid spectrum for each dietary treatment were determined. The total combined percentage of MA, PA and SA was then calculated and each of these FA were expressed as a percentage of this total (Table 2). This proportion of these individual fatty acids in the diet was used to establish the mixture of 1-13C labelled MA, PA and SA delivered to the subjects in a 10 mg=kg body weight dose.

Data analysis

Fractional oxidation rates of breakfast meal mixture of 1-13C labelled LCSFA, based on the percentage of the total label ingested appearing in the breath, were determined using the following equation:14 Fractional LCSFA oxidation …%h† ˆ mmol excess 13 C=mmol CO2 mmol 13 C administered  …mmol CO2 excreted=h†  100  1:35 …1†

Fatty acid composition of test meals

Replicate portions of all test meals were homogenized using a commercial blender. Fatty acid content was determined using gas ± liquid chromatography after lipid extraction20 and boron tri¯uoride methylation.21 The gas chromatograph (Hewlett Packard 5890 Series II, California) was equipped with an autosampler and ¯ame ionization detector. Separation was achieved on an SP 2330 capillary column, 30 m  0.25 mm  0.2 mm. Samples were injected at 100 C. Oven temperature was then increased to 190 C at a rate of 3 C=min and was held at that level for 25 min. The International Journal of Obesity

where mmol excess 13C=mmol CO2 ˆ (13C enrichment above baseline)  R Pee Dee Belemnite Limestone (RPBD)  1073,12 RPBD ˆ 0.011273,22 and the factor of 1.35 was added to account for bicarbonate retention of 13C in body pools.23,24 Amount of CO2 excreted (mmol) was determined using RGE. The mmol 13C administered was determined as the total number of mmol delivered in the mixture of the 13C labelled MA, PA and SA. Cumulative fractional LCSFA oxidation was determined as the total fractional LCSFA oxidation over 5.5 h.

Fatty acid chain length and fat oxidation AA Papamandjaris et al

The total amount of combined MA, PA and SA in the breakfast meal was calculated for individual subjects based on the following formula:25 Breakfast meal fatty acid content …mg† ˆ Breakfast meal fat …mg†  concentration of fatty acid in the breakfast meal …percentage

…2†

fatty acids of the total fatty acid content† Net oxidation rates of combined dietary MA, PA, and SA were calculated as follows:25 Net LCSFA oxidation …mg=5:5 h† ˆ breakfast meal fatty acid content …mg†  cumulative fractional LCSFA oxidation

…3†

…%=5:5 h† Percentage contribution of combined dietary MA, PA and SA to total postprandial fat oxidation was calculated as:26 Percentage dietary LCSFA contribution to total fat oxidation …%=5:5 h† ˆ net LCSFA oxidation

1161

Results Study subjects

…mg=5:5 h†=total postprandial fat oxidation …mg†  100% …4† where total postprandial fat oxidation was determined using RGE and the equations of de Weir.27 Approach to calculations

For all dependent variables, results were calculated for day 7 using day 7 baseline values, and two sets of results were generated for day 14 using either day 7 or day 14 baseline values in the calculations. Day 7 results represent exogenous LCSFA oxidation, as subjects had only received one isotope dose and had no stored labelled LCSFA to release during endogenous oxidation. Enrichment values for day 7 were obtained by subtraction of the day 7 baseline values from subsequent breath enrichments. On day 14, one set of enrichment values was generated by subtraction of each subjects' initial baseline as determined on day 7. These values are taken to represent combined exogenous and endogenous LCSFA oxidation, as subjects had received repeated isotope doses and therefore could produce label through both exogenous and endogenous oxidation. The other set of day 14 enrichment values was generated by subtraction of the day 14 baseline value, determined prior to breakfast and delivery of isotope on day 14. These values are taken to represent exogenous LCSFA oxidation, as oxidation from endogenous sources has been accounted for through baseline correction. Statistical analysis

24), and diet by meal interaction was used to compare overall appearance of 13CO2 in the breath over 8 days. Fractional LCSFA oxidation was assessed using repeated measures ANOVA with factors of sequence (1 or 2), diet (MCT or LCT), hour (0.5 ± 5.5), and diet by hour interaction for each set of results, day 7, day 14 exogenous, and day 14 combined. For the dependent variables cumulative fractional LCSFA oxidation, net LCSFA oxidation, and percentage dietary LCSFA contribution to total fat oxidation, repeated measures ANOVA models with factors of sequence (1 or 2), diet (MCT or LCT), day (7 or 14), and diet-byday interaction were employed. This approach was applied twice, once using the day 14 exogenous enrichment data for the day 14 factor, and once using the day 14 combined enrichment data for the day 14 factor. Post-hoc comparisons were made using contrasts. Level of signi®cance was set at P < 0.05. All values are expressed as means  s.e.m.

A repeated measures ANOVA with factors of sequence (1 or 2), diet (MCT or LCT), meal (1 ±

During both dietary cycles, subjects were observed to have consumed all food provided. Two subjects' energy intakes were increased based on reported hunger over the ®rst 3 days of the ®rst dietary cycle. There were no signi®cant changes in body weight between start (56.2  1.5 kg) and end (56.7  1.4 kg) of the study, either between or within study groups. Subject diaries indicated regular menstruation, with mean cycle length of 28.5 0.6 days. No abnormalities in health or large ¯uctuations in activity patterns were reported. Meal compositional analysis

Mean fatty acid composition pro®les of all meals for each diet are shown in Table 2. The percentage of MA, PA and SA differed between diets. The relative proportions of MA, PA and SA calculated for the dosing mixture were 28.4 0.3%, 54.2  0.2% and 17.3 0.2% for the MCT diet and 9.0  0.2%, 56.2 0.3% and 34.8  0.2% for the LCT diet, respectively. Substrate utilization as measured using respiratory gas exchange

A full description of data obtained using RGE can be found elsewhere.19 Diet fat treatment did not have a signi®cant effect on total fat or carbohydrate oxidation on either day 7 or 14. In the postprandial period on day 7, repeated measures ANOVA showed an overall greater19 rate of fat oxidation on the MCT diet. Overall appearance of label in the breath

Comparison of pre-meal 13CO2 enrichment in breath above baseline over the 8 day isotope delivery period International Journal of Obesity

Fatty acid chain length and fat oxidation AA Papamandjaris et al

1162

showed a main effect (P < 0.001) of dietary treatment, with values of 13C in breath on the MCT diet higher than those on the LCT diet (Figure 2).

Fractional LCSFA oxidation

On day 7, the ®rst day of isotope delivery, there was a main effect (P < 0.02) of diet on fractional LCSFA oxidation, with greater oxidation on the MCT diet at 4.5 h (Figure 3A). Similarly, on day 14, a main effect (P < 0.001) of diet on combined fractional LCSFA oxidation was observed, with increased oxidation on the MCT diet at 1.5 and 3.5 h (Figure 3B). On day 14, a main effect (P < 0.006) of diet on exogenous fractional oxidation was also observed, with greater oxidation on the LCT diet at 2.5 and 5.5 h (Figure 3C). The interaction between diet and hour for day 14 exogenous fractional oxidation was not signi®cant (Figure 3C). Cumulative fractional LCSFA oxidation on day 7 on the MCT diet (10.7 2.4%=5.5 h) did not differ from oxidation on the LCT diet (6.1 1.8%=5.5 h, Figure 4A). Comparison of combined cumulative fractional oxidation between diets over 5.5 h on day 14 showed a main effect (P < 0.007) of diet with greater oxidation from the MCT diet (16.9  2.5%=5.5 h) than from the LCT diet (9.1 1.2%=5.5 h; Figure 4A). Within-diet comparison between day 7 and combined values for day 14 showed an increase (P < 0.03) in oxidation on the MCT diet by day 14 (Figure 4A). On day 14, exogenous cumulative fractional oxidation was not different between diet types (MCT, 5.3  2.8%=5.5 h; LCT, 10.6  3.6%=5.5 h); however, the interaction term of diet by day approached signi®cance (P ˆ 0.07) in the ANOVA model using day factors of 7 and 14 exogenous (Figure 4A).

Net long-chain saturated fatty acid oxidation

The net LCSFA oxidation from the breakfast meals is shown in Figure 4B. The results were similar to those found for both between and within diet comparisons for cumulative fractional LCSFA oxidation. Betweendiet comparison of net LCSFA oxidation for day 7 showed no signi®cant effect of diet (MCT, 1836  426 mg=5.5 h vs LCT, 1135  330 mg=5.5 h). On day 14, there was an effect (P < 0.01) of diet on combined net oxidation, with greater net oxidation on the MCT diet (2956 413 mg=5.5 h) than on the LCT diet (1669  224 mg=5.5 h). Exogenous net LCSFA oxidation on day 14 was not affected by diet type (MCT, 932  497 mg=5.5 h vs LCT, 1926 671 mg= 5.5 h); however, as with the cumulative fractional oxidation data, the interaction term of diet by day approached signi®cance (P ˆ 0.08) in the ANOVA model using day factors of 7 and 14 exogenous. Within the MCT diet, there was greater (P < 0.03) combined net LCSFA oxidation on day 14 as compared to day 7. Percentage dietary long-chain saturated fatty acid contribution to total fat oxidation

Results for percentage dietary LCSFA contribution to total fat oxidation (Figure 4C) re¯ected the pattern observed for between and within-diet comparisons for both cumulative fractional LCSFA oxidation and net LCSFA oxidation. The only signi®cant between-diet ®nding occurred with the combined results on day 14, with LCSFA from the MCT contributing to 16.3  2.3 %=5.5 h of total fat oxidation, which differed (P < 0.01) from the 9.5  1.5 %=5.5 h contributed by those fatty acids on the LCT diet. For within-diet comparisons, the day 7 contribution on the MCT diet (9.3  1.9 %=5.5 h) was smaller (P < 0.02) than the combined contribution on the MCT diet on day 14 (16.3  2.3 %=5.5 h). Again, no differences were observed in between-diet comparisons of day 7 and

Figure 2 Comparison of pre-meal % 13CO2 enrichment in breath over baseline for MCT vs LCT diets from day 7 to day 14 after ingestion of 13C labelled LCSFA at each breakfast meal in women. Values were determined by subtraction of initial baseline at day 7. Arrows indicate daily breakfast delivery of labelled LCSFA. Letters (A or B) indicate day of two day rotating menu cycle. Whole body oxidation of labelled LCSFA was greater (P < 0.001) during MCT vs LCT feeding. International Journal of Obesity

Fatty acid chain length and fat oxidation AA Papamandjaris et al

1163

Figure 3 (A) Fractional oxidation of labelled LCSFA on day 7 after ®rst oral administration of labelled LCSFA at breakfast (time 0); there was a main effect (P < 0.02) of diet on fractional oxidation with greater oxidation on the MCT diet at 4.5 h. (B) Fractional oxidation of labelled LCSFA on day 14, after daily oral administration of labelled LCSFA at breakfast (time 0) starting on day 7. Values represent combined oxidation as they were calculated using the day 7 baseline. There was a main effect (P < 0.001) of diet on combined fractional oxidation with greater oxidation on the MCT diet at 1.5 h and 3.5 h. (C) Fractional oxidation of labelled LCSFA on day 14 after repeated consumption of labelled LCSFA at breakfast (time 0) starting on day 7. Values represent exogenous oxidation as they were calculated using the day 14 pre-dose baseline. There was a main effect (P < 0.006) of diet on exogenous fractional oxidation with greater oxidation on the LCT diet at 2.5 and 5.5 h. Signi®cant difference between diets indicated by * (P < 0.05).

day 14 exogenous values, but the interaction term of diet by day approached signi®cance at P ˆ 0.10.

Discussion Repeated delivery of 1-13C labelled MA, PA and SA and assessment of breath 13CO2 enrichment over a period of 8 days permitted measurement of exogenous and endogenous oxidation of LCSFA. The novel ®nding presently is that the acyl structure of fatty acids within the diet other than those which were labelled, speci®cally MCT vs LCT, affected the capacity of the body to retain and subsequently release 13C, obtained from 1-13C labelled LCSFA, through oxidation. Therefore, rate of oxidation of fatty acids

Figure 4 Cumulative fractional oxidation over 5.5 h of labelled LCSFA (A), net LCSFA oxidation (B) and percent dietary LCSFA contribution to total fat oxidation (C) on both MCT and LCT treatments on day 7, day 14 combined and day 14 exogenous. Cumulative LCSFA fractional oxidation (P < 0.007), net LCSFA oxidation (P < 0.01), and percent dietary LCSFA contribution to total fat oxidation (P < 0.01) were different between dietary treatments for the combined day 14 values. Day 14 combined cumulative LCSFA fractional oxidation (P < 0.03), net LCSFA oxidation (P < 0.03), and percent dietary LCSFA contribution to total fat oxidation (P < 0.02) within the MCT dietary treatment were greater as compared to day 7. Different letters indicate signi®cant differences between diets (P < 0.05). Signi®cant within diet comparisons are indicated by * (P < 0.05).

depends not only on their structure, but also on the structure of other fatty acids present in the dietary fatty acid mix. The day 7 results indicate that total exogenous oxidation of LCSFA, following the ®rst exposure to isotope after 1 week of prefeeding, was not signi®cantly affected by dietary treatment. Other researchers11,12 who reported increased oxidation of 1-13C labelled fatty acids in subjects consuming MCT diets labelled the MCFA speci®cally and therefore were not assessing the impact of MCFA on oxidation of other fatty acids, as in the present study. In addition, in those studies,11,12 the increase in oxidation was observed following initial feeding of MCT, not following a 7 day prefeeding period, as was employed International Journal of Obesity

Fatty acid chain length and fat oxidation AA Papamandjaris et al

1164

presently. Therefore, it would appear that the presence of MCT in the present diet was not capable of signi®cantly altering the exogenous oxidation of dietary LCSFA following 7 days of diet administration. Lack of difference in exogenous oxidation between the two dietary treatments is supported by results determined using RGE measurements.19 On day 7, there was no effect of dietary treatment on the thermic effect of food or on total fat oxidation. Examination of exogenous oxidation following 2 weeks of feeding showed similar results. For day 14, as for day 7, exogenous cumulative fractional LCSFA oxidation, net LCSFA oxidation, and percentage dietary LCSFA contribution to total fat oxidation were not different between dietary treatments. These ®ndings are supported by the previously reported RGE results which showed no effect of MCT vs LCT consumption on the thermic effect of food or on total fat oxidation on day 14.19 Endogenous oxidation of LCSFA after 8 days of repeated isotope delivery was also assessed. The low percentage dietary LCSFA contribution to total fat oxidation and the low cumulative fractional LCSFA oxidation determined on day 7 indicated that most of the label delivered remained in the body, and was not oxidized postprandially. This is in accord with other researchers who observed that less than 1% of label was oxidized postprandially26 and that breath 13CO2 enrichment was low following delivery of label.14,28 Consequently, ingested LCSFA were deposited in the body and available for subsequent endogenous oxidation. As the same amount of labelled LCSFA had been consumed each day, it may be hypothesized that the exogenous component of 1-13C labelled LCSFA oxidation remained the same over that period. Thus, any increase in combined exogenous and endogenous oxidation of the 1-13C LCSFA seen by day 14, as compared to day 7, indicates contribution from endogenous sources. By day 14, following repeated isotope dosing, the combined amounts of cumulative fractional LCSFA oxidation, net LCSFA oxidation, and percentage dietary LCSFA contribution to total fat oxidation differed between the two diets, and were also signi®cantly greater than on day 7 within the MCT diet. These ®ndings support the hypothesis that the 1-13C labelled LCSFA delivered in the MCT diet on previous days had been deposited in metabolic compartments and were subsequently released during endogenous oxidation. Lack of a signi®cant difference between days 7 and 14 within the LCT diet indicates that endogenous oxidation of stored labelled LCSFA did not occur to the same degree as on the MCT diet. The presence of MCT in the diet thus had the capacity to in¯uence both the deposition and subsequent mobilization of LCSFA in a manner that the presence of LCT did not. The effect of dietary treatment on BMR as measured using RGE supports in part these results.19 On day 7, BMR was signi®cantly greater on the MCT diet as compared to the LCT diet. By day 14, despite a lack of statistical signi®cance, there was

International Journal of Obesity

a trend towards an elevated BMR on the MCT diet. An elevated BMR may indicate increased endogenous fat oxidation, as seen on day 14 using in the present experiment. The present protocol did not permit assessment of endogenous oxidation on day 7. In contrast, doubly labelled water (DLW) assessment of average daily total energy expenditure (TEE) did not show an effect of dietary treatment.29 However, the DLW method of measuring TEE may not allow enough accuracy and precision to detect the small differences that may occur during MCT vs LCT feeding. The possibility that 13C background enrichment ¯uctuation contributed signi®cantly to the difference seen in endogenous oxidation between diets on day 14 is unlikely. The prefeeding period and uniformity of the diet ensured that the baseline 13C breath measurement taken on day 7 was representative of the baseline during the entire dietary period. Any background shifts away from this measured baseline would have been systematic and constant for both diets as the carbohydrate content was identical.12 Additionally, 13 C breath enrichment due to oxidation of labelled fatty acids delivered at the present dose is measurably higher than that due to natural 13C abundance,30 making the impact of potential background shifts in 13C enrichment negligible. The retention factor for 13C within the bicarbonate pool used in the oxidation calculations may be important in the interpretation of the ®ndings. The retention factor used (1.35) was based on research in similar populations under similar conditions.23 ± 26 Nevertheless, results may have been affected by a shift in 13C retention in the bicarbonate pool between day 7 and day 14. However, even if 13C retention was altered, either similarly for both diets between day 7 and 14, or disparately between diets, such results would indicate the presence of a effect on the metabolic fate of 13 C. It is also possible that not all the 13C oxidized on day 14 came from stored FA; some may have come from the 13C labelled CO2 pool. In any such scenario, the data as presented suggest that MCT consumption alters the utilization of 1-13C labelled LCSFA, but that greater appearance of 13CO2 during the MCT phase does not speci®cally indicate greater mobilization of stored 13C fatty acids. Furthermore, differences in fatty acid composition across diets may exist as a contributing factor. Oxidation rates of other nonlabelled fatty acids could have decreased in response to the increased oxidation of MA, PA and SA, resulting in no overall change in total fat oxidation. Nonetheless, examination of the pattern of label appearance offers insight into what may be occurring during storage and oxidation of the labelled LCSFA. Except for the initial rise and subsequent overall higher level of appearance of label on the MCT diet, the pattern of label appearance for the two diets was similar. The peaks seen at each day's pre-lunch breath sampling were most likely the result of exogenous oxidation of labelled LCSFA delivered with the

Fatty acid chain length and fat oxidation AA Papamandjaris et al

breakfast meal. This rise was similar to that typically seen following ingestion of 13C labelled fatty acids.12,23 The immediate rise in enrichment in the breath on the MCT diet not present on the LCT diet may have indicated rapid maximal enrichment of a fatty acid storage pool following the initial dose with isotope. This pool may then have been maintained at maximal enrichment during subsequent isotope delivery with the stored labelled LCSFA constantly available as substrates for endogenous oxidation. Enrichment of this pool on the LCT diet may not have occurred to the same degree, resulting in less endogenous oxidation. On the LCT diet, the labelled LCSFA can be hypothesized to have been deposited in a second adipose pool which was much less metabolically active, and therefore did not contribute in the same way to the presence of label in the breath through endogenous oxidation. The existence of two storage pools for fat has been postulated previously. In 1960, Hirsch et al31 hypothesized the existence of two separate metabolic compartments, one a caloric depot, exchanging slowly with dietary fat, the other a small metabolically active compartment, which may be in close metabolic relation to dietary lipids. Such a two pool system has also been proposed by other researchers.32,33 Beynen et al32 theorized that the rapid turnover rate of plasma free fatty acids originating from adipose tissue was explained by the presence of a small, rapidly exchangeable pool. Pittet et al33 observed continued incorporation of label from 13-methyltetradecanoic acid (13-MTD), a structurally labelled fatty acid, into adipose tissue in humans following cessation of delivery of 13-MTD. The authors suggested that the continued appearance of label indicated the presence of a buffer pool for temporary fatty acid storage prior to de®nite incorporation into adipose tissue. These pools may exist in order to maintain a speci®c fatty acid pattern within adipose tissue.34 ± 36 Fatty acids not optimal with respect to adipose tissue composition and subsequent lipid synthesis may be shunted towards the rapidly exchanging pool, to be oxidized more readily rather than stored long term. If, as supported by the present results, the presence of MCT within the diet can direct more fatty acids towards temporary storage in the metabolically active pool, rather than towards more permanent storage in the inert pool, less fat could be stored during consumption of a diet containing MCT. Manipulation of dietary fatty acids to include more MCT could be important in the context of the development and control of obesity, as it may result in less adipose tissue deposition.

Conclusion These results demonstrate that MCT can measurably affect endogenous oxidation of LCSFA following 2 weeks of feeding compared to LCT as determined

using a repeated 13C LCSFA dosing paradigm. In addition, the results obtained support the existence of both a metabolically active pool and an inert pool of fatty acid storage. Acyl structure of dietary fatty acids, speci®cally chain length, may determine in which pool dietary fat is stored, and subsequently how rapidly fatty acids can undergo endogenous oxidation. Further research in this area could include examination of effects of diets with greater proportions of MCT and similar amounts of LCSFA between diets to control for possible differences in oxidation and bicarbonate retention of 13C stemming from disparate LCSFA pro®les. Such research, in combination with the present results, will help in the understanding of the capacity of MCT to increase endogenous fat oxidation, and therefore to potentially decrease adipose tissue pool size during extended feeding.

1165

Acknowledgements

The authors would like to thank Jayne Rop for her technical assistance. This work was supported by a grant from the Dairy Farmers of Canada. References

1 Crozier G, Bois-Joyeux B, Chanez M, Girard J, Peter J. Metabolic effects induced by long-term feeding of mediumchain triglycerides in the rat. Metabolism 1987; 36: 807 ± 814. 2 Demmelmair H, Sauerwald T, Koletzko B, Richter T. New insights into lipid and fatty acid metabolism via stable isotopes. Eur J Pediatr 1997; 156(Suppl1): S70 ± S74. 3 Dulloo AG, Mensi N, Seydoux J, Girardier L. Differential effects of high-fat diets varying in fatty acid composition on the ef®ciency of lean and fat tissue deposition during weight recovery after low food intake. Metabolism 1995; 44: 273 ± 279. 4 Geliebter A, Torbay N, Bracco EF, Hashim SA, Van Itallie TB. Overfeeding with medium-chain triglyceride diet results in diminished deposition of fat. Am J Clin Nutr 1983; 37: 1 ± 4. 5 Jakobs C, Kneer J, Mann D, Boulloche J, Brivet M, Poll-The BT, Saudubray JM. In vivo stable isotope studies in three patients affected with mitochondrial fatty acid disorders: Limited diagnostic use of 1-13C fatty acid breath test using bolus technique. Eur J Pediatr 1997; 156(Suppl1): S78 ± S82. 6 Johnson RC, Young SK, Cotter R, Lin L, Rowe WB. Mediumchain-triglyceride lipid emulsion: metabolism and tissue distribution. Am J Clin Nutr 1990; 52: 502 ± 508. 7 Lavau MM, Hashim SA. Effect of medium chain triglyceride on lipogenesis and body fat in the rat. J Nutr 1978; 108: 613 ± 620. 8 Lieber CS, Lefevre A, Spritz N, Feinman L, DeCarli LM. Difference in hepatic metabolism of long- and mediumchain fatty acids: the role of fatty acid chain length in the production of the alcoholic fatty liver. J Clin Invest 1967; 46: 1451 ± 1460. 9 Mascioli EA, Randall S, Porter KA, Kater G, Lopes S, Babayan VK, Blackburn GL, Bistrian BR. Thermogenesis from intravenous medium-chain triglycerides. J Parenter Enteral Nutr 1991; 15: 27 ± 31. 10 Metges C, Wolfram G. Medium-and long-chain triglycerides labelled with 13C: a comparison of oxidation after oral or parenteral administration in humans. J Nutr 1991; 121: 31 ± 36. 11 Paust H, Keles T, Park W, Knoblach G. Fatty acid metabolism in infants. In: Chapman et al (eds). Stable isotopes in paediatric nutritional and metabolic research. Intercept: UK, 1990, pp 1 ± 22. International Journal of Obesity

Fatty acid chain length and fat oxidation AA Papamandjaris et al

1166

12 Schoeller DA, Klein PD, Watkins JB, Heim T, Maclean Jr WC. 13C abundances of nutrients and the effect of variations in 13 C isotopic abundances of test meals formulated for 13CO2 breath tests. Am J Clin Nutr 1980; 33: 2375 ± 2385. 13 Souza PF, Williamson DH. Effects of feeding medium-chain triacylglycerols on maternal lipid metabolism and pup growth in lactating rats. Br J Nutr 1993; 69: 779 ± 787. 14 Watkins JB, Klein PD, Schoeller DA, Kirschner BS, Park R, Perman JA. Diagnosis and differentiation of fat malabsorption in children using 13C-labelled lipids: trioctanoin, triolein, and palmitic acid breath tests. Gastroenterology 1982; 82: 911 ± 917. 15 Binnert C, Pachiaudi C, Beylot M, Hans D, Vandermander J, Chantre P, Riou JP, Laville M. In¯uence of human obesity on the metabolic fate of dietary long-and medium-chain triacylglycerols. Am J Clin Nutr 1998; 67: 595 ± 601. 16 Mif¯in MD, St Jeur ST, Hill LA, Scott BJ, Daugherty SA, Koh YD. A new predictive equation for resting energy expenditure in healthy individuals. Am J Clin Nutr 1990; 51: 241 ± 247. 17 Bell L, Jones PJH, Telch J, Clandinin MT, Pencharz PB. Prediction of energy needs for clinical studies. Nutr Res 1985; 5: 123 ± 129. 18 Health and Welfare Canada. Nutrient recommendations: The report of the scienti®c review committee. Ministry of Supply and Services: Ottawa, Canada, 1990, pp 203 ± 204. 19 White MD, Papamandjaris AA, Jones PJH. Enhanced postprandial energy expenditure with medium chain fatty acid feeding is attenuated after 14 days in premenopausal women. Am J Clin Nutr 1999; 69: 883 ± 889. 20 Folch J, Lees M, Sloan S. A simple method for the isolation and puri®cation of the total lipids from animal tissues. J Biol Chem 1957; 226: 497 ± 509. 21 Al Makdessi S, Andrieu JL, Bacconin A, Fugier JC, Herilier H, Faucon G. Assay of lipids in dog myocardium using capillary gas chromatography and derivitization with boron tri¯uoride and methanol. J Chromatogr 1985; 339: 25 ± 34. 22 Mook WA, Grootes PM. The measuring procedure and corrections for high precision mass spectrometric analysis of isotopic abundance ratio, especially referring to carbon, oxygen, and nitrogen. Int J Mass Spectrom Ion Phys 1973; 12: 273 ± 279. 23 Irving C, Wong WW, Shulman RJ, O'Brian Smith E, Klein PD. [13C] bicarbonate kinetics in humans: intra- vs interindividual variations. Am J Physiol 1983; 245: R190 ± R202.

International Journal of Obesity

24 Jones AE. The gastrointestinal handling and metabolic disposal of dietary lipid. Doctoral thesis, University of Southampton, UK, 1996. 25 MacDougall DE, Jones PJ, Kitts DD, Phang PT. Effect of butter compared with tallow consumption on postprandial oxidation of myristic and palmitic acids. Am J Clin Nutr 1996; 63: 918 ± 924. 26 MacDougall DE, Jones PJ, Vogt J, Phang PT, Kitts DD. Utilization of myristic and palmitic acid in humans fed different dietary fats. Eur J Clin Nutr 1996; 26: 755 ± 762. 27 Weir JB. New methods for calculating metabolic rate with special emphasis to protein metabolism. J Physiol 1949; 226: 497 ± 509. 28 Jones PJH, Pencharz PB, Clandinin MT. Whole body oxidation of dietary fatty acids: Implications for energy utilization. Am J Clin Nutr 1985; 42: 769 ± 777. 29 Papamandjaris AA, White MD, Jones PJH. Components of total energy expenditure in healthy young women are not affected after 14 day feeding with medium versus long chain triglycerides. Obes Res 1999; 7: 273 ± 280. 30 Jones PJH, Pencharz PB, Bell L, Clandinin MT. Model for determination of 13C substrate oxidation rates in humans in the fed state. Am J Clin Nutr 1985; 41: 1277 ± 1282. 31 Hirsch J, Farquhar JW, Ahrens EH, Peterson ML, Stoffel W. Studies of adipose tissue in man. A microtechnique for sampling and analysis. Am J Clin Nutr 1960; 8: 499 ± 511. 32 Beynen AC, Hermus RJ, Hautvast JG. A mathematical relationship between the fatty acid composition of the diet and that of the adipose tissue in man. Am J Clin Nutr 1980; 33: 81 ± 85. 33 Pittet PG, Bessart T, Jequier E. Adipose tissue labelling in man, using a structurally-labelled fatty acid as tracer. Int J Vit Nutr Res 1983; 53: 115 ± 119. 34 Field CJ, Angel A, Clandinin MT. Relationship of diet to the fatty acid composition of human adipose tissue structural and stored lipids. Am J Clin Nutr 1985; 42: 1206 ± 1220. 35 Jones PJH, Toy BR, Cha MC. Differential fatty acid accretion in heart, liver and adipose tissues of rats fed beef tallow, ®sh oil, olive oil and saf¯ower oils at three levels of energy intake. J Nutr 1995; 125: 1175 ± 1182. 36 Ostwald R, Okey R, Shannon A, Tinoco J. Changes in tissue lipids in response to diet. I. Fatty acids of subcutaneous, mesenteric and interscapular fat. J Nutr 1962; 76: 341 ± 352.