Role of human liver lipogenesis and reesterification in triglycerides secretion and in FFA reesterification

Transcription

1 Role of human liver lipogenesis and reesterification in triglycerides secretion and in FFA reesterification FREDERIQUE DIRAISON 1 AND MICHEL BEYLOT 1,2 1 Laboratoire de Physiopathologie Métabolique et Rénale and the 2 Centre de Recherche en Nutrition Humaine de Lyon, Lyon, France Diraison, Frederique, and Michel Beylot. Role of human liver lipogenesis and reesterification in triglycerides secretion and in FFA reesterification. Am. J. Physiol. 274 (Endocrinol. Metab. 37): E321 E327, To measure 1) the contribution of hepatic de novo lipogenesis (DNL) and plasma free fatty acid (FFA) reesterification to plasma triglyceride (TG) secretion, and 2) the role of oxidation and hepatic and extrahepatic reesterification in FFA utilization, five normal subjects drank deuterated water and were infused (postabsorptive state) with [1-13 C]palmitate and [1,2,3-2 H 5 ]glycerol. Total lipid oxidation (L ox ) was measured by indirect calorimetry. FFA oxidation ( µmol kg 1 min 1 ) accounted for 45% of FFA turnover rate (R t ) (1.04 µmol kg 1 min 1 ) and 91% of L ox ; FFA reesterification was µmol kg 1 min 1. Fractional and absolute TG R t were h 1 and µmol kg 1 min 1. DNL accounted for % of TG secretion, and hepatic FFA reesterification accounted for %; this last process represented a utilization of FFA of µmol kg 1 min 1. We conclude that, in the postabsorptive state, 1) DNL and FFA reesterification account for only 50 55% of TG secretion, the remaining presumably being provided by stored lipids or lipoproteins taken up by liver, 2) most reesterification occurs in extrahepatic tissues, and 3) oxidation and reesterification each contribute about one-half to FFA utilization; FFA oxidation accounts for almost all L ox. mass spectrometry; indirect calorimetry; oxidation; stable isotopes; lipids THE IN VIVO METABOLISM of triglycerides (TG) has received renewed interest in the past few years. This was motivated by the demonstration that high TG levels contribute to the development of insulin resistance (24, 30) as well as atherosclerosis (12). Therefore, TG metabolism as free fatty acid (FFA) metabolism has implications for hyperlipidemias, obesity, and diabetes. In the postabsorptive state TG are secreted by the liver. These TG are synthesized by use of fatty acids either produced by hepatic lipogenesis [de novo lipogenesis (DNL)] or provided by the uptake of plasma FFA (reesterification) or the degradation of lipoproteins taken up by the liver. The fractional contribution of DNL to TG secretion can be estimated by use of [ 13 C]acetate infusion (18) or deuterated water ingestion (10, 25, 26). Methods for measuring absolute rates of reesterification and secretion have been proposed (19, 22). These methods combine infusion of labeled palmitate to measure FFA turnover rate (R t ) with indirect calorimetry to measure total net lipid oxidation (L ox ). The difference between FFA R t and L ox is assumed to represent reesterification (19, 22). The addition of labeled acetate infusion proposed by Hellerstein et al. (19) allows the simultaneous measurement of DNL and therefore the correction of the net lipid oxidation rate measured by indirect calorimetry for this lipogenesis. However, these approaches rely on several assumptions that have to be tested. First, they assume that liver is the only tissue responsible for the reesterification of plasma FFA, which is questionable because splanchnic FFA uptake was measured at 25 30% of FFA appearance rate (36), far less than current estimates of plasma FFA reesterification. Moreover, there is evidence that other tissues, such as muscles, can also reesterify FFA (7). Second, the contribution of the breakdown of lipoproteins taken up by liver is considered negligible. This is questionable, despite the evidence that plasma FFA are an important source of liver TG synthesis (2, 27), and this potential contribution of lipoprotein breakdown by liver has never been estimated to our knowledge. Third, the calculation of FFA reesterification (difference between FFA R t and L ox ) assumes that plasma FFA are the only source of fatty acids oxidized. This last assumption received recent experimental support. Sidossis et al. (33) found that, when labeled carbon exchanges in the citric acid cycle are taken into account, the classic discrepancy (14, 24) between plasma FFA oxidation (measured by the excretion rate of labeled CO 2 during infusion of a labeled fatty acid) and total lipid oxidation is no longer observed in postabsorptive subjects, i.e., plasma FFA and total lipid oxidation values are nearly equal. However, there is also evidence that fatty acids from both tissues (8) and plasma TG (38) are also oxidized. Aarsland et al. (1) recently described another method based on the kinetics of the appearance of labeled fatty acids in plasma TG during [1,2-13 C 2 ]acetate infusion. However, the only directly measured value is DNL; calculation of total TG secretion rate requires extrapolation for plateau value, and the contribution of the reesterification of plasma FFA was not measured. In the present study, the simultaneous use of [ 13 C]palmitate and deuterated glycerol infusions, deuterated water ingestion, and indirect calorimetry allowed us to measure more directly most parameters of plasma FFA and TG metabolism in postabsorptive subjects. Lipogenesis was calculated from deuterium incorporation in the palmitate of TG (10). The kinetics of the appearance of [ 13 C]palmitate in TG during the infusion of the tracer and of its disappearance after interruption of the infusion allowed the calculation of, respectively, plasma FFA reesterification by liver and the R t of plasma TG. The comparison of the R t of glycerol and FFA gave the intracellular reesterification rate (4, 22) (i.e., the reesterification of fatty acids released by breakdown of tissue TG directly within tissue, without any appear /98 $5.00 Copyright 1998 the American Physiological Society E321

2 E322 HUMAN PLASMA FFA AND TRIGLYCERIDES METABOLISM ance in the circulating plasma pool), and the comparison of FFA R t and either FFA or total L ox gave upper and lower limits for extracellular reesterification rate (i.e., the reesterification of fatty acids released in plasma by one tissue and taken up by another). Comparison of this last value with liver reesterification gave the contribution of hepatic and extrahepatic tissues to total extracellular reesterification. The results confirm that lipogenesis is a minor contributor to TG secretion in postabsorptive subjects; however, reesterification of plasma FFA contributes only about one-half to TG secretion. Last, most extracellular reesterification takes place in extrahepatic tissues. METHODS Subjects. Informed written consent was obtained from five normal volunteers (three men and two women, aged yr). All had a normal body weight (body mass index: ) and normal plasma glucose, triglyceride, and cholesterol values. None was taking any medication. They consumed a weight-maintaining diet with 200 g carbohydrate and abstained from heavy physical activity or alcohol consumption for the week before the studies. The last meal was consumed between 7:00 and 8:00 PM on the day before the tests. Protocols. The protocol of the study was approved by the Ethical Committee of Lyon and by the Institut National de la Santé et de la Recherche Médicale, and the study was conducted according to the Hurriet law. All tests were performed in the Centre de Recherche en Nutrition Humaine de Lyon. On the day before the tests, the subjects drank a loading dose of deuterated water (3 g/kg body water, one-half after the evening meal and one-half at 10:00 PM; Eurisotop, Saint Aubin, France). Then until the end of the study, they drank only water enriched with 2 H 2 O (4.5 g 2 H 2 O/l of drinking water). All tests were initiated in the postabsorptive state. At 7:30 AM an indwelling catheter was inserted in a forearm vein for the infusion of tracers. Another catheter was placed in a dorsal vein of the opposite hand, which was kept at 55 C, to obtain arterialized blood. After an initial blood sampling and collection of expired gas in the initial state, a bolus of NaH 13 CO 3 (1.0 µmol/kg) was injected, and primed-continuous infusions of [1-13 C]palmitate (Mass Trace Woburn, MA; 0.03 µmol kg 1 min 1 after a bolus of 0.3 µmol/kg) and [1,2,3-2 H 5 ]glycerol (Mass Trace; 0.05 µmol kg 1 min 1 after a bolus of 0.5 µmol/kg) were initiated and continued for 4 h. Blood samples were collected at 60, 120, 180, 200, 220, and 240 min and during the 6 h after the interruption of tracer infusion. Expired gas samples were collected every hour, except for the period of min, when samples were collected every 15 min. Respiratory gas exchanges were measured from 120 to 240 min (Deltatrac metabolic monitor, Datex, Heksinki, Finland). Urine samples were collected from 0 to 240 min for determination of nitrogen excretion. Analytic procedures. Metabolites were assayed with enzymatic methods (4, 5) on neutralized perchloric extracts of plasma (glucose, glycerol) or on plasma (FFA, TG). Plasma insulin and glucagon levels were determined by radioimmunoassay (4). Deuterium enrichment in plasma water was measured by isotope ratio mass spectrometry (optima 18, Fisons Instruments, Middlewich, UK) (40) and expressed as mole percent excess. Glycerol was purified by sequential ionexchange chromatography, and its enrichment was determined by gas chromatography-mass spectrometry by use of the triacetate derivative (3); samples were injected into a gas chromatograph (HP5890, Hewlett-Packard, Palo Alto, CA) equipped with a 25-m fused silica capillary column (OV1701, Chrompack, Bridgewater, NJ) interfaced with a mass spectrometer (HP5871A, Hewlett-Packard) working in the electron impact mode. Ions of mass-to-charge ratio (m/z) were selectively monitored; standard curves containing known amounts of natural and labeled glycerol were run with the biological samples. After addition of heptadecanoic acid as internal standard for the determination of plasma palmitate concentration, plasma lipids were extracted by the method of Folch et al. (11). TG and FFA were separated by thin-layer chromatography, scraped off the silica plates, and eluted from silica with ether (10). The methylated (FFA) and transmethylated (TG) derivatives of palmitate were prepared according to Morrison and Smith (28). Total enrichment (i.e., 13 C and deuterium) of the palmitate of FFA and of TG was measured by gas chromatography-mass spectrometry, monitoring ions of m/z (10). 13 C enrichment was selectively determined by gas chromatography isotope ratio mass spectrometry (6). Deuterium enrichment was calculated by subtracting 13 C enrichment from the total enrichment. Palmitate concentrations were calculated from the area ratio of the peaks corresponding to palmitate and to heptadecanoate; standard curves containing known concentration ratios of these two fatty acids were run with the samples. 13 C enrichment in the CO 2 of expired gas was measured by gas chromatography isotope ratio mass spectrometry (17). Calculations. Glycerol and palmitate turnover rates were calculated from their respective enrichments in deuterium and 13 C measured during the 180- to 240-min period with steady-state equations. Plasma FFA R t was calculated from palmitate R t by use of the ratio of palmitate to FFA concentration. The fractional contribution of lipogenesis to TGpalmitate was calculated from the deuterium enrichments in plasma water and in the palmitate of plasma TG, as previously described (10). In short, the deuterium enrichment that would be obtained if lipogenesis were the only source of TG-palmitate was calculated from plasma water enrichment; the comparison of the actual enrichment observed with that theoretical value gives the contribution of lipogenesis to the circulating pool of TG-palmitate. TG kinetics were calculated from the decay of 13 C enrichment in TG-palmitate after the end of the infusion of [1-13 C]palmitate; this approach is comparable to the one used with radioactively labeled tracers (35). A single exponential was fitted to the data: IE IE 0 e kt, where IE is the isotopic enrichment and k is the fractional turnover rate (FR t ). The half-life time of the plasma TG pool is then calculated as t 1/ /FR t. The absolute TG R t is calculated as TG R t FR t M, where M is the plasma pool of TG obtained by multiplying the TG concentration by the plasma volume estimated to 37 ml/kg (9). Absolute de novo lipogenesis is obtained as DNL DNL% TG R t 3, where the factor 3 acknowledges the fact that there are three fatty acids per molecule of TG. The fractional contribution of the hepatic reesterification (FRE t ) of FFA to the plasma TG pool is calculated from the 13 C enrichment of plasma palmitate and the increase of 13 C enrichment in TG-palmitate during the infusion of [1-13 C]palmitate by use of the formula FRE t [(IE t2 IE t1 )/(t 2 -t 1 )]/IE Pal, where t 1 and t 2 are the times when samples are taken, IE is the corresponding IE of TGpalmitate, and IE Pal is the enrichment at plateau of plasma palmitate. The comparison of FRE t with FR t gives the percent contribution of reesterification to TG turnover rate: RE% FRE t /FR t. The absolute contribution of reesterification, RE, to the TG secretion is then RE RE% TGR t. Three times RE gives the rate of reesterification of plasma FFA by the liver (RE hep ).

3 HUMAN PLASMA FFA AND TRIGLYCERIDES METABOLISM E323 Table 1. Hormone and metabolite concentrations Plasma palmitate oxidation was calculated as Pal ox F 13 CO 2 /(IE Pal fc), where F 13 CO 2 is the excretion rate of labeled CO 2 produced by palmitate oxidation and fc is the factor correcting for incomplete label recovery. Pal ox was calculated using either the bicarbonate recovery factor (0.80) or the acetate recovery factor (0.56) measured by Sidossis et al. (33). Total FFA ox was calculated from Pal ox and the ratio of palmitate over FFA concentration. Oxidation rate for total lipids was calculated from the indirect calorimetry data by use of stoichiometric equations (13). O 2 consumption and CO 2 production were the average values of three 20-min periods of steady-state measurements during the 3rd and 4th h of the protocol. Total lipid oxidation was converted into fatty acid oxidation by assuming an average TG molecular weight of 860 (13) and multiplying by three, because there are three molecules of fatty acid for each molecule of TG. Intracellular reesterification rate (RE inc ) was calculated by subtracting plasma FFA R t from three times glycerol R t (4, 22). Extracellular reesterification rate (RE exc ) was calculated by subtracting from plasma FFA R t their oxidation rate calculated from either respiratory gas exchange (total lipid oxidation) or labeled CO 2 excretion rate (plasma FFA oxidation). Subtracting RE hep from RE exc gives the contribution of extrahepatic tissue to plasma FFA reesterification (non-re hep ). All results are shown as means SE. RESULTS Glucose, mmol/l FFA, µmol/l Glycerol, µmol/l TG, mmol/l Insulin, mu/l Glucagon, ng/l Values are means SE. FFA, free fatty acids; TG, triglycerides. The concentrations of plasma insulin and glucagon and of metabolites during the 180- to 240-min period are shown in Table 1. There was during the remaining 6 h of the study a progressive rise of FFA, which reached µmol/l at 600 min without significant variations of the other parameters. The evolution of plasma palmitate and glycerol enrichments is shown in Fig. 1. Table 2. Plasma FFA and glycerol turnover rates Glycerol R t, µmol kg 1 min FFA R t, µmol kg 1 min FFA R t /glycerol R t Values are means SE. R t, turnover rate Plateau levels of enrichment and concentration were obtained during the 120- to 240-min period, and kinetic parameters were calculated with equations for steadystate conditions. In addition, when measuring glycerol enrichment, we found no evidence for a partial loss of one or two deuterium atoms from the labeled glycerol infused, contrary to what was observed by Previs et al. (29) in the perfused rat liver system (i.e., in the plasma glycerol samples there were increases only in the 148/145 ion ratios and no increases in the 146 or 147/145 ratios). Therefore, there is no evidence that glycerol R t was artifactually overestimated. The FFA over glycerol R t ratio is shown in Table 2. This ratio was near to three, and therefore there was little RE inc ( µmol kg 1 min 1 ). The evolution of CO 2 enrichments is shown in Fig. 2. FFA oxidation rate, calculated from labeled CO 2 excretion rate by use of either the classical bicarbonate correction factor or the acetate correction factor of Sidossis et al. (33), is compared in Table 3 with L ox. This L ox, calculated from the respiratory gaseous exchange data, was corrected for the measured lipogenesis to convert the net L ox to the total L ox rate. This correction was almost negligible given the low lipogenic rate of normal postabsorptive subjects (see next paragraph). Use of the acetate correction factor raised the contribution of FFA oxidation to the FFA disappearance rate from 32 to 45% and to total L ox from 62 to 91% (Table 3). Therefore the non-ffa oxidation rate (considered to be the oxidation of fatty acids from plasma or tissue TG) was very low ( µmol kg 1 min 1 ) with the use of the acetate correction factor. Table 4 shows the kinetics of plasma TG and the contribution of lipogenesis and plasma FFA reesterification to TG secretion. Reesterification was calculated from the increase of 13 C enrichment in TG-palmitate Fig. 1. Evolution of 13 C enrichment (left vertical axis) of the palmitate of plasma free fatty acids (FFA, s) and plasma triglycerides (r) and of the deuterium enrichment of glycerol (l, right vertical axis) during and after a 4-h (0 240 min) infusion of [1-13 C]palmitate. MPE, mole percent excess. Fig. 2. Evolution of 13 C enrichment of CO 2 in expired gas during and after a 4-h (0 240 min) infusion of [1-13 C]palmitate. APE, atoms percent enrichment.

4 E324 HUMAN PLASMA FFA AND TRIGLYCERIDES METABOLISM Table 3. Plasma FFA and total lipid oxidation rates during the infusion of labeled palmitate. This increase was linear with time (Fig. 1), and the intercept of the regression line with the time axis shows a delay of 30 min for the appearance of 13 C in TG-palmitate. There was a similar delay between the end of labeled palmitate infusion and the beginning of the decline of enrichment in TG-palmitate. The FR t of plasma TG was /h, and their turnover rate was µmol kg 1 min 1. The fractional contribution of reesterification (FRE) was /h. Thus RE contributed only one-half to TG secretion. Deuterium enrichment in plasma water was stable throughout the study at % (Fig. 3). The mean deuterium enrichment of palmitate in plasma TG during the 0- to 240-min period of the study was %. The calculated contribution of lipogenesis was low ( %, Table 4), representing an absolute synthetic rate of fatty acids of µmol kg 1 min 1. Therefore, nearly 50% of the TG secretion rate was not accounted for by lipogenesis and RE. The contribution of hepatic reesterification to plasma FFA R t was calculated to represent µmol kg 1 min 1 (Table 5). This hepatic reesterification was compared with the total RE exc calculated by subtracting from FFA R t the L ox rate obtained either from the excretion rate of 13 CO 2 or gaseous exchanges; whatever the calculation, liver was a minor contributor to RE exc. DISCUSSION Bicarbonate correction Acetate correction FFA oxidation µmol kg 1 min %offfar t non-ffa oxidation µmol kg 1 min L ox, µmol kg 1 min FFA oxidation was calculated from labeled CO 2 excretion rate by use of either bicarbonate or acetate correction factor. Total lipid oxidation (L ox ) was calculated from gaseous exchanges and corrected for lipogenesis. Using a combination of stable isotope-labeled tracer and indirect calorimetry, we measured in normal subjects most aspects of plasma FFA (turnover rate, oxidation, and reesterification) and TG (turnover rate, contribution of reesterification and lipogenesis to secretion) kinetics. In agreement with previous reports (1, 3), we found that the intracellular reesterification of fatty acids was almost negligible in normal postabsorptive subjects. This result is based on the comparison of plasma glycerol and fatty acid turnover rates. We used Table 4. Kinetic parameters of plasma TG FR t,h 1 t,h R t, Reesterification, µmol kg 1 min 1 %oftgr t Lipogenesis, %oftgr t Values are means SE. FR t, fractional turnover rate. t, half-life time. Fig. 3. Deuterium enrichment (MPE) in plasma water during 10-h protocol. deuterated glycerol to measure glycerol kinetics. Previs et al. (29) showed recently in perfused rat livers the presence of a substrate cycling between the infused glycerol and liver triose phosphates. When [1,2,3,- 2 H 5 ] glycerol is infused, this cycling results in the release of glycerol molecules having lost one to four excess deuterium atoms. If this loss of deuterium occurs to a significant extent in vivo, it would induce an artifactual decrease of the enrichment of glycerol measured during infusion of deuterated tracer and an overestimation of glycerol turnover rate. This possibility seemed unlikely, because we previously found comparable glycerol turnover rates in rats infused with [2-13 C]glycerol or [1,2,3,- 2 H 5 ] glycerol (4). Our present finding that there was no detectable appearance of glycerol molecules having lost one or more deuterium in human subjects infused with deuterated glycerol confirms the validity of this tracer for determination of glycerol kinetics in humans. Extracellular reesterification of FFA is calculated as the difference between FFA disappearance rate and oxidation (1, 19). It relies thus on the way this oxidation is estimated. One can use either lipid oxidation measured by indirect calorimetry or the oxidation rate calculated from the excretion in expired gas of labeled CO 2 during the infusion of a 13 C- or 14 C-labeled fatty acid. Several authors repeatedly found a large discrep- Table 5. Total extracellular reesterification rate and contribution of liver and extrahepatic tissues to reesterification Reesterification Extracellular Hepatic Extrahepatic FFA R t FFA ox Bicarbonate Acetate FFA R t L ox Values are means SE. Oxidation of FFA (FFA ox ) was calculated from labeled CO 2 excretion rate by use of bicarbonate or acetate correction factor. L ox, total lipid oxidation rate measured by indirect calorimetry and corrected for lipogenesis.

5 HUMAN PLASMA FFA AND TRIGLYCERIDES METABOLISM E325 ancy between these two measurements (14, 24, 33). This was generally interpreted as showing that FFA are only a part of the lipid oxidized and that fatty acids from plasma or tissue TG are also oxidized. If this interpretation is correct, to calculate the extracellular reesterification one should use the oxidation rate obtained from the excretion of labeled CO 2. However, Sidossis et al. (32, 33) provided evidence that plasma FFA oxidation measured with tracer was underestimated because of the loss of labeled carbon through exchanges in the citric acid cycle during the oxidation process before the release of the labeled carbon as CO 2. This was based on the comparison of the labeled CO 2 kinetics in expired gas after infusion of labeled acetate and palmitate was stopped and on the demonstration of the appearance, with both tracers, of labeled carbons in plasma compounds such as lactate or glutamate (reflecting exchanges at the oxaloacetate and -keto-glutarate levels in the citric acid cycle) (32). They proposed to use, instead of the classic bicarbonate recovery factor, an acetate recovery factor that would also take into account these isotopic exchanges (33). With this factor they reported that FFA oxidation accounted in postabsorptive subjects for all (33) or 80% (34) of lipid oxidation measured by indirect calorimetry, instead of 50 60%. Our present finding that FFA oxidation represented 90% of total lipid oxidation when the acetate factor was used (instead of 62% with the bicarbonate factor) would agree with their results. Therefore, if Sidossis et al. are correct, one could use for calculating extracellular fatty acid recycling either tracer-determined plasma FFA oxidation or lipid oxidation measured by indirect calorimetry [although the latter has to be corrected for lipogenesis when significant lipogenesis occurs (19)]. On the other hand, the results from Sidossis et al. are difficult to reconcile with the previous demonstration that, at least in animals, fatty acids from very low density lipoprotein (VLDL)-TG contribute to lipid oxidation (8, 38, 39): when fatty acid-labeled VLDL-TG were infused, labeled CO 2 was excreted without any significant appearance of labeled fatty acid in the plasma FFA pool. This contribution of VLDL-TG was found to be as important as that of FFA in rats (38). In the present experiments, labeled fatty acids were incorporated in plasma TG with an enrichment reaching at the end of the 4-h infusion of [1-13 C]palmitate one-third to one-half of that in plasma FFA. If these fatty acids from plasma TG were oxidized, their contribution to lipid oxidation could not be distinguished from that of FFA (unless dual tracer experiments or short-term infusion of labeled fatty acid were performed). Therefore, although one could argue that, if plasma TG significantly contributes to lipid oxidation, the decay of CO 2 enrichment in expired gas after labeled palmitate and acetate infusions are stopped should be different, contrary to what was observed (33), additional studies are needed to clarify these points. Whatever the correct value to be used for its calculation, extracellular reesterification is an important process, because its upper and lower estimates were 70 and 55% of FFA turnover rate (Table 3). Plasma TG kinetics were calculated from the decay of label in palmitate of plasma TG after the [1-13 C]palmitate infusion was stopped. The values obtained agree with those reported previously for VLDL-TG using radioactive tracers or other approaches with stable isotope-labeled tracers (1, 15, 18, 21, 31). DNL, expressed as a relative or an absolute value, was a minor contributor to TG secretion, in agreement also with previous reports (10, 18, 26). Plasma FFA reesterification is considered to be the main or nearly exclusive source (2, 27) of liver TG synthesis and secretion. Actually we found that hepatic reesterification of plasma FFA was a large contributor but accounted for only one-half of the turnover rate of plasma TG. This value is lower than the one (80 100%) reported by Barter and Nestel (2) in normal subjects when these investigators compared the plateau values of the enrichment of palmitate in plasma FFA and TG during infusion of tritiated palmitate. However, they measured enrichments in plasma drawn from a superficial vein draining mainly subcutaneous adipose tissue. FFA enrichment is already decreased in mixed venous blood compared with arterial samples (20); this dilution is probably greater in a superficial vein. Therefore, Barter and Nestel may have overestimated the contribution of plasma FFA reesterification. If lipogenesis and plasma FFA reesterification do not account for all liver TG secretion, the remaining fatty acids could be provided by the breakdown of plasma lipoprotein taken up by liver (presumably VLDL remnants) or by the mobilization of fatty acids of previously stored TG and phospholipids. Our data do not allow us to determine which of these two potential sources is the more important, but there is in vitro evidence in support of the latter (37). Last, we compared the total extracellular and the hepatic reesterification rates. Whatever the calculation chosen for determining the total extracellular reesterification, the contribution of liver appeared minor, showing that most of this reesterification takes place in extrahepatic tissues (presumably muscles and adipose tissue). It should be stressed that, had liver reesterification accounted for the totality of plasma TG secretion, this contribution to extracellular reesterification would anyway have remained minor. This conclusion agrees with the recent estimate from FFA splanchnic balance and glycerol kinetics data that splanchnic FFA uptake represents only 25% of the amount of plasma FFA to be reesterified (23). Therefore, the assumption (19) that all extracellular reesterification occurs in liver is not verified. In conclusion, using simple and noninvasive methods we were able to measure most aspects of plasma FFA and TG kinetics, especially with regard to liver lipid metabolism. We show that liver reesterification is only a minor part of the metabolic fate of plasma FFA and, although a major contributor to it, does not account for the totality of TG secretion. This approach will be

6 E326 HUMAN PLASMA FFA AND TRIGLYCERIDES METABOLISM useful for investigating abnormalities of lipid metabolism in situations such as obesity, diabetes, and stress. We thank C. Pachiaudi for help in the measurement of enrichments with isotope ratio mass spectrometry. We thank also the nurses of the Centre de Recherche en Nutrition Humaine for their help in performing the tests, and all the subjects who volunteered for this study. This work was supported by grants from the Societé Francophone de Nutrition Entérale et Parentérale, the Association Francaise des Diabetiques, and the Fondation de France. Address for reprint requests: M. Beylot, Laboratoire de Physiopathologie Métabolique et Rénale, Faculté R. Laennec, Rue G. Paradin, Lyon Cedex 08, France. Received 2 June 1997; accepted in final form 28 October REFERENCES 1. Aarsland, A., D. Chinkes, and R. R. Wolfe. Contributions of de novo synthesis of fatty acids to total VLDL-triglyceride secretion during prolonged hyperglycemia/hyperinsulinemia in normal man. J. Clin. Invest. 98: , Barter, P. J., and P. J. Nestel. Precursors of plasma triglyceride fatty acids in obesity. Metabolism 22: , Beylot, M., C. Martin, B. Beaufrere, J. P. Riou, and R. Mornex. Determination of steady-state and nonsteady-state glycerol kinetics in humans using deuterium-labeled tracer. J. Lipid Res. 28: , Beylot, M., C. Martin, M. Laville, J. P. Riou, R. Cohen, and R. Mornex. Lipolytic and ketogenic flux in hyperthyroidism. J. Clin. Endocrinol. Metab. 73: 42 49, Beylot, M., J. P. Riou, F. Bienvenu, and R. Mornex. Increased ketonemia in hyperthyroidism; evidence for a -adernergic mechanism. Diabetologia 19: , Binnert, C., M. Laville, C. Pachiaudi, V. Rigalleau, and M. Beylot. Use of gas chromatography isotope ratio mass spectrometry to study triglycerides metabolism in humans. Lipids 30: , Budohoski, L., J. Gorski, K. Nazar, H. Kaciuba-Uscilko, and R. L. Terjung. Triacylglycerol synthesis in the different skeletal muscle fiber sections of the rat. Am. J. Physiol. 271 (Endocrinol. Metab. 34): E574 E581, Dagenais, G. R., R. G. Tancredi, and K. L. Zierler. Free fatty acid oxidation by forearm muscle at rest and evidence for an intramuscular lipid pool in the human forearm. J. Clin. Invest. 58: , Dagher, J. F., J. H. Lyons, D. C. Finlayson, J. Shamsai, and F. D. Moore. Blood volume measurement: a critical study. Adv. Surg. 1: , Diraison, F., C. Pachiaudi, and M. Beylot. Measuring lipogenesis and cholesterol synthesis in humans with deuterated water: use of simple gas chromatography-mass spectrometry techniques. J. Mass Spectrom. 32: 81 86, Folch, J., M. Lees, and G. H. Sloane Stanley. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226: , Fontbonne, A., E. Eschwege, and F. Cambien. Hypertriglyceridemia as a risk factor of coronary heart disease in subjects with impaired glucose tolerance or diabetes. The Paris prospective study. Diabetologia 32: , Frayn, K. N. Calculation of substrate oxidation rates in vivo from gaseous exchange. J. Appl. Physiol. 55: , Groop, L. C., R. C. Bonadonna, M. Shank, A. S. Petrides, and R. A. DeFronzo. Role of free fatty acids and insulin in determining free fatty acid and lipid oxidation in man. J. Clin. Invest. 87: 83 89, Grundy, S. M., H. Y. I. Mok, L. Zech, D. Steinberg, and M. Berman. Transport of very low density lipoprotein triglycerides in varying degrees of obesity and hypertriglyceridemia. J. Clin. Invest. 63: , Guilly, R., E. Billion-Rey, C. Pachiaudi, S. Normand, J. P. Riou, E. S. Jumeau, and J. L. Brazier. On-line purification and carbon 13 isotopic analysis of carbon dioxide in breath: evaluation of on-line gas chromatography isotope ratio mass spectrometry. Anal. Chim. Acta 259: , Havel, R. J., J. P. Kane, E. O. Balasse, N. Segel, and L. V. Basso. Splanchnic metabolism of free fatty acids and production of triglycerides of very low density lipoproteins in normotriglyceridemic and hypertriglyceridemic humans. J. Clin. Invest. 49: , Hellerstein, M. K., M. Christiansen, S. Kaempfer, C Kletke, K. Wu, J. S. Reid, K. Mulligan, N. S. Hellerstein, and C. H. L. Shackleton. Measurement of de novo lipogenesis in humans using stable isotopes. J. Clin. Invest. 87: , Hellerstein, M. K., R. A. Neese, and J. M. Schwarz. Model for measuring absolute rates of hepatic de novo lipogenesis and reesterification of free fatty acids. Am. J. Physiol. 265 (Endocrinol. Metab. 28): E814 E820, Jensen, M. D., P. J. Rogers, M. G. Ellman, and J. M. Miles. Choice of infusion-sampling mode for tracer studies of free fatty acid metabolism. Am. J. Physiol. 254 (Endocrinol. Metab. 17): E562 E565, Kissebah, A. H., S. Alfarsi, and P. W. Adams. Integrated regulation of very low density lipoprotein triglyceride and apoprotein-b kinetics in man: normolipemic subjects, familial hypertriglyceridemia and familial combined hyperlipidemia. Metabolism 30: , Klein, S., V. R. Young, L. A. Blackburn, B. R. Bistrian, and R. R. Wolfe. Palmitate and glycerol kinetics during brief starvation in normal weight young adult and elderly subjects. J. Clin. Invest. 78: , Landau, B. R., J. Wahren, S. F. Previs, K. Ekberg, V. Chandramouli, and H. Brunengraber. Glycerol production and utilization in humans: sites and quantification. Am. J. Physiol. 271 (Endocrinol. Metab. 34): E1110 E1117, Laville, M., V. Rigalleau, J. P. Riou, and M. Beylot. Respective role of plasma nonesterified fatty acid oxidation and total lipid oxidation in lipid induced insulin-resistance. Metabolism 44: , Lee, W. N. P., S. Bassilian, Z. Guo, D. Schoeller, J. Edmond, E. A. Bergner, and L. O. Byerley. Measurement of fractional lipid synthesis using deuterated water and mass isotopomer analysis. Am. J. Physiol. 266 (Endocrinol. Metab. 29): E372 E383, Leitch, C. A., and P. H. J. Jones. Measurement of triglycerides synthesis in humans using deuterium oxide and isotope ratio mass spectrometry. Biol. Mass Spectrom. 20: , Lewis, G. F., K. D. Uffelman, L. W. Szeto, B. Weller, and G. Steiner. Interaction between free fatty acids and insulin in the acute control of very low density lipoproteins in humans. J. Clin. Invest. 95: , Morrison, W. R., and L. M. Smith. Preparation of fatty acid methyl esters and dimethylacetals from lipids with boron fluoridemethanol. J. Lipid Res. 5: , Previs, S. F., C. A. Fernandez, D. Yang, M. V. Soloviev, F. David, and H. Brunengraber. Limitations of mass isotopomer distribution analysis of glucose to study gluconeogenesis: substrate cycling between glycerol and triose-phosphates in liver. J. Biol. Chem. 270: , Reaven, G. R. The fourth musketeer: from Alexandre Dumas to Claude Bernard. Diabetologia 38: 3 13, Reaven, G., D. B. Gill, R. C. Gross, and J. W. Farquhar. Kinetics of triglyceride turnover of very low density lipoproteins of human plasma. J. Clin. Invest. 44: , Sidossis, L. S., A. R. Coggan, A. Gastaldelli, and R. R. Wolfe. Pathway of free fatty acid oxidation in human subjects. Implication for tracer studies. J. Clin. Invest. 95: , Sidossis, L. S., A. R. Coggan, A. Gastaldelli, and R. R. Wolfe. A new correction factor for use in tracer estimations of plasma fatty acid oxidation. Am. J. Physiol. 269 (Endocrinol. Metab. 32): E649 E656, Sidossis, L. S., C. A. Stuart, G. I. Shulman, G. D. Lopaschuk, and R. R. Wolfe. Glucose plus insulin regulate fat oxidation by controlling the rate of fatty acid entry into the mitochondria. J. Clin. Invest. 98: , 1996.

7 HUMAN PLASMA FFA AND TRIGLYCERIDES METABOLISM E Taskinen, M. R., W. F. Beltz, and I. Harper. The effect of non-insulin-dependent diabetes on VLDL triglycerides and VLDL apob metabolism. Studies before and after sulfonylurea therapy. Diabetes 35: , Wahren, J., S. Efendic, R. Luft, L. Hagenfelt, O. Bjorkman, and P. Felig. Influence of somatostatin on splanchnic glucose metabolism in postabsorptive and 60-h fasted humans. J. Clin. Invest. 59: , Wiggins, D., and G. F. Gibbons. Origin of hepatic very low density lipoprotein triacylglycerol: the contribution of cellular phospholipid. Biochem. J. 320: , Wolfe, R. R., and M. J. Durkot. Role of very low density lipoproteins in the energy metabolism of the rat. J. Lipid Res. 26: , Wolfe, R. R., J. H. F. Shaw, and M. J. Durkot. Effect of sepsis on VLDL kinetics: responses in basal state and during glucose infusion. Am. J. Physiol. 248 (Endocrinol. Metab. 11): E732 E740, Wong, W. W., L. S. Lee, and P. D. Klein. Deuterium and oxygen-18 measurements on microliter samples of urine, plasma, saliva and human milk. Am. J. Clin. Nutr. 45: , 1987.