Variations of lipid and fatty acid contents during the reproductive cycle of the anthozoan Renilla koellikeri.

Lipid and fatty acid contents were monitored in the sea pansy Renilla koellikeri during its reproductive cycle. The fluctuations of lipids and fatty acids were closely related to sexual development and gonads represented the principal site of lipid storage in the colony. Between April and May, fatty acids and lipids greatly accumulated just prior to spawning. The subsequent decrease of lipid and fatty acid contents in June probably resulted from loss due to spawning. Fatty acid composition was assessed for the whole animal and was compared between male and female gonads, and between gonads and somatic tissues. Arachidonic acid (AA) was the principal fatty acid in the whole colony followed by palmitic acid and eicosapentaenoic acid (EPA). The examination of fatty acid variations in gonads showed that EPA and AA were highly incorporated in eggs during the month preceding spawning, whereas in the same period, EPA was the only fatty acid to accumulate in spermatophores. The differences of fatty acid levels observed in gonads reflected sex differences observed in colonies. The seasonal variations of the omega6/omega3 ratio of the colonies were largely influenced by EPA and AA, and appeared as a reliable indicator of the sexual maturity of the sea pansy.

An isomeric mixture of conjugated linoleic acids but not pure cis-9, trans-11-octadecadienoic acid affects body weight gain and plasma lipids in hamsters.

We report the effect of an atherogenic diet supplemented with cis-9, trans-11-octadecadienoic acid (c9t11), linoleic acid (LA) or an isomeric mixture of conjugated linoleic acids (CLA) on plasma lipids, weight gain and food intake of male Golden Syrian hamsters. Animals were assigned to three diet groups (n = 10), and fed nonpurified diet, supplemented with 10% hydrogenated coconut oil and 0.05% cholesterol for 6 wk. The first diet group was further supplemented with 1% CLA (CLA group), the second diet group with 0.2% c9t11 (c9t11 group) and the third group with 0.2% LA (LA group). The diets were designed to have equivalent levels of c9t11 in the CLA and c9t11 groups. At 2 and 6 wk of feeding, the CLA group had significantly lower plasma triglyceride and total cholesterol concentrations than either the c9t11 or the LA groups. HDL-cholesterol did not differ among diet groups. The CLA group had significantly lower weight gain but greater food intake than either the c9t11 or the LA groups. There were no significant differences between the c9t11 and the LA groups in any of the variables measured. We conclude that under our experimental conditions of short-term feeding, c9t11, thought to be the active compound in CLA, does not produce the same effect as the isomer mixture.

Uptake and metabolism of structured triglyceride by Caco-2 cells: reversal of essential fatty acid deficiency.

Structured lipids have been proposed as efficient vehicles for the supplementation of essential fatty acids (EFA) to patients with malabsorption. We investigated how a novel structured triglyceride (STG), containing purely octanoic acid in the sn-1/sn-3 and [14C]linoleic acid in the sn-2 positions, was incorporated into different lipid classes in Caco-2 cells. We also evaluated the contribution of gastric lipase in the uptake and metabolism of [14C]linoleic acid from the STG. We furthermore determined the potential of the STG to correct EFA deficiency induced in Caco-2 cells. The absorption of STG by Caco-2 cells was significantly greater compared with that of triolein. The addition of human gastric lipase significantly enhanced cellular uptake of the labeled substrate, reflecting the stereoselectivity of gastric lipase to hydrolyze medium chain FA. Analysis of the intracellular lipids synthesized revealed a predominance of phospholipids-monoglycerides. Most of the radioactivity in the lipoproteins isolated from Caco-2 cells was recovered in TG-rich lipoproteins (45%) and to a lesser extent in the high-density lipoprotein (36%) and low-density lipoprotein (17%) fractions. The administration of STG to Caco-2 cells rendered EFA deficient produced a marked increase of the cellular level of linoleic and arachidonic acids. This resulted in a lower ratio of 20:3(n-9) to 20:4(n-6), reflecting the correction of EFA deficiency in Caco-2 cells. Our data demonstrate that STG, in the presence of gastric lipase, have beneficial effects on lipid incorporation, lipoprotein production, and EFA status, utilizing Caco-2 cells as a model of EFA deficiency.

The role of dietary choline in the beneficial effects of lecithin on the secretion of biliary lipids in rats.

Earlier studies showed that dietary soybean lecithin increases biliary lipid secretion, which mainly comes from the contribution of high density lipoprotein (HDL) and hepatic microsomal pools of phosphatidylcholine and cholesterol. In addition, a lecithin diet enhances bile secretion and prevents bile acid-induced cholestasis. This study evaluated the contribution of choline, a component of lecithin, to the observed effect of lecithin on biliary secretory function. Rats were fed either a control diet (CD), a choline diet (ChD) or a lecithin-enriched diet (LD) for 2 weeks. Results showed that like LD, ChD induced an increase in bile flow and bile acid secretion rate when compared with the control diet. However, unlike LD, ChD did not significantly increase biliary phospholipids and cholesterol output. An increase of hydrophilic bile acids (i.e. ursodeoxycholic and muricholic acids) in bile of rats fed choline could explain why the biliary phospholipid and cholesterol secretion was not increased. During taurocholic acid infusion, both experimental diets increased bile flow and the bile acid secretion rate maximum (BASRm). The cholestasis usually observed after the BASRm is reached was inhibited by ChD and LD. Both diets induced a decrease in plasma cholesterol (total and HDL), however, only LD induced statistically significant changes. Analysis of total cholesterol and phospholipid content of microsomes and canalicular membranes indicated no statistically significant difference between control and experimental groups either under basal conditions or after bile acid infusion. Similarly, the phospholipid classes and fatty acid composition of biliary phosphatidylcholine were not altered by feeding ChD and LD. We conclude that choline contributes to the beneficial effect of a lecithin diet on bile secretion. It is postulated that this effect may be attributed to modulation of HDL and an enhancement of the cholesterol and phospholipid pools destined for biliary secretion.

The effects of cholesterol uptake from high-density lipoprotein subfractions on biliary sterol secretion in rats with essential fatty-acid deficiency.

High-density lipoprotein (HDL) participates in the transfer of cholesterol to the liver, in which it is subsequently excreted into bile as bile acid and cholesterol. In this study, the effect of essential fatty-acid (EFA) deficiency on cholesterol contribution from HDL subfractions to bile was investigated. Rats that were rendered EFA-deficient over 4 weeks displayed changes in their plasma HDL subfractions and liver tissue fatty acids. Plasma linoleic (18:2n6), linolenic (18:3n3,) and arachidonic (20:4n6) acids decreased, whereas palmitoleic (16:1n7) and eicosatrienoic (20:3n9) acids increased. EFA deficiency was confirmed by an elevation of the 20:3(n-9)/20:4(n-6) index. To examine the hepatic handling of lipoprotein-derived cholesterol, HDL2 and HDL3 from donor rats were isolated, labeled with [14C]-cholesterol, and injected iv into EFA-deficient and normal rats with a bile fistula. In HDL subfractions from control rats, no significant variations were noted in the specific activity of cholesterol output in both groups of EFA recipient rats; however, the output of biliary bile acids was significantly decreased in EFA-deficient rats following the administration of labeled HDL3. In HDL2 and HDL3 originating from EFA-deficient rats, a decrease in the specific activity of both biliary cholesterol and bile acid output was recorded in EFA-deficient rats. Concomitant with the defective HDL2- and HDL3-[14C] cholesterol translocation into bile of EFA-deficient rats, increased hepatic very-low-density lipoprotein (VLDL)-[14C] cholesterol secretion was observed in vivo. HDL2 and HDL3 particles, derived from EFA-deficient rats, had an altered composition including a depletion in apo A-I and an enrichment in apo E isoforms, which are the the two major HDL apolipoproteins involved in the delivery of cholesterol to the liver. Taken together, these results show that normal EFA status is necessary for efficient HDL-cholesterol processing by the liver.

Iron-salicylate complex induces peroxidation, alters hepatic lipid profile and affects plasma lipoprotein composition.

Iron overload, with its associated toxic effects, has serious health consequences and results in damage to the liver, heart and other organs. Salicylate may be used as the lipophilic carrier, transporting more iron into hepatocytes. In this study, we examined the effect of the combined administration of these compounds on plasma lipid profile and lipoprotein composition, as well as on hepatic lipid concentration. Male Spraque-Dawley rats were injected i.p. with Fe (15 mg/kg weight). This injection was repeated 24 h later with a gavage of sodium salicylate (700 mg/kg). Control rats received 0.9% NaCl only. The peroxidation indices TBARS (P < 0.001) and conjugated dienes (P < 0.05) significantly increased in the blood (50 and 122%, respectively) and liver (333 and 101%, respectively) of Fe salicylate-treated rats. Concomitantly, blood and liver arachidonic acid content was diminished by iron treatment. In parallel, the plasma lipid profile was markedly affected in Fe-salicylate treated-rats. Lower plasma concentrations of total cholesterol (25%, P < 0.0001) cholesteryl ester, (34%, P < 0.001) and high-density lipoprotein-cholesterol (50%, P < 0.001) were observed. Lipoprotein composition analysis revealed enrichment of free cholesterol and depletion of cholesterol ester in very low-density, intermediate-density, low-density and high-density (HDL2, HDL3) lipoproteins. Furthermore, SDS-polyacrylamide gel electrophoresis revealed several alterations in the apolipoprotein distribution of these lipoproteins. The activity of lecithin:cholesterol acyltransferase was unchanged and could not account for the reduction of cholesterol esterification. As for the plasma, the liver exhibited a significant (P < 0.001) decrease in total cholesterol (2.42 +/- 0.07 versus 1.89 +/- 0.06 mg/g liver), essentially due to a reduction in cholesteryl ester (0.93 +/- 0.07 versus 0.51 +/- 0.03 mg/g, P < 0.001). Again, the activity of ACAT (dpm/mg microsomal protein) was not lower (12,700 +/- 1250) than that of controls (9650 +/- 1080). Thus, the iron-salicylate was able to induce peroxidation and to profoundly affect the intravascular and intrahepatic lipid, and plasma lipoprotein metabolism. Additional work is needed to elucidate the mechanisms involved in the underlying lipid and lipoprotein abnormalities.

Metabolism of 4,7,10,13,16,19-docosahexaenoic acid in isolated perfused adult and newborn pig eyes.

We performed open-circuit perfusions of newborn and adult pig eyes to study the age-dependent metabolism of 4,7,10,13,16,19-docosahexaenoic acid (DHA) in this organ. DHA taken up by the perfused eyes was partitioned into glycerolipids, beta-oxidation, and the intracellular nonesterified fatty acid pool. In newborn eyes, DHA was incorporated into structural lipids to a greater extent than in adult eyes. Competition experiments suggest that the adult eye is more selective for DHA than the newborn eye. Finally, pulse-chase data indicate that DHA transport from the circulation across the retinal pigment epithelium and into the retina is more rapid in adult than in newborn eyes. The results are discussed with respect to the rapid accumulation of retinal DHA in early life and the avid retention of this fatty acid by the adult retina.

Impact of essential fatty acid deficiency on hepatic sterol metabolism in rats.

The major aim of the current investigation was to define whether essential fatty acid (EFA) deficiency modifies the intrahepatic metabolism and biliary output of sterols in rats. EFA-deficient diet caused an impoverishment in linoleic, arachidonic, and docosahexaenoic acids, and a marked enrichment in the eicosatrienoic acid of the plasma, liver, and hepatic microsomes. During a short term of biliary drainage, a significant decline of the pool size of biliary sterols was noted in EFA-deficient rats compared with control rats. To assess the biosynthesis of biliary components, the common bile duct was cannulated and the pool size depleted (24 hours). Subsequently, a 6-hour bile collection disclosed a significant decrease (nmoles/min/g liver) in bile acids (4.8 +/- 0.3 vs. 8.4 +/- 0.7, P < .005), cholesterol (0.26 +/- 0.01 vs. 0.34 +/- 0.02, P < .05), and phospholipids (1.49 +/- 0.11 vs. 2.82 +/- 0.32, P < .005) in EFA-deficient rats compared with controls (n = 6/group). When cholesterogenesis was measured by the incorporation of [14C]acetate and 3H20 into cholesterol, using liver slices, it was also found to be significantly (P < .001) reduced in EFA-deficient rats. The activity of hepatic 3-hydroxy-3-methylglutaryl coenzyme A reductase, the rate-limiting enzyme in cholesterol biosynthesis, was consistently lower (80 percent, P < .001) in EFA-deficient rats. In parallel experiments, the direct measurement of microsomal acyl-CoA:cholesterol acyl-transferase (ACAT) showed a decrease averaging 52 percent (P < .001). This is in striking contrast to the elevated activity (157 percent, P < .005) of cholesterol 7 alpha-hydroxylase, the rate-limiting enzyme in bile acid biosynthesis. Current experiments also suggest that the enzyme regulation involving phosphorylation and dephosphorylation is modified by EFA deficiency. Among the structural alterations observed in the morphology of hepatocytes in EFA-deficient rats, the lumen of bile canaliculi was reduced in size. These results underline the importance of EFA in the major mechanisms involved in the maintenance of hepatocyte sterol balance.

Malondialdehyde-modified HDL leads to accumulation of cholesterol in rat liver endothelial cells.

In order to study in vivo the effect of modified high density lipoprotein (HDL) on the movement of free cholesterol to liver cells and bile, we injected i.v. into rats, native or malondialdehyde modified HDL labelled with [14C]cholesterol. Bile analysis indicated that the contribution of labelled cholesterol to bile acid secretion was diminished in the group receiving MDA-modified HDL when compared to control group. On the other hand, the liver analysis revealed higher radioactivity in the treated group. A separation of liver cells into parenchymal, endothelial, and kupffer at 90 min after the injection of MDA-modified HDL or native HDL indicated that the endothelial cell uptake of labelled free cholesterol from MDA-modified HDL was 2.6-fold higher than for native HDL. It is suggested that liver endothelial cells may be involved in the protection against atherogenoic oxidized lipoprotein. However, with regard to our finding, the uptake of cholesterol from modified HDL was detrimental to bile acid secretion.

Malondialdehyde-modified high-density lipoprotein cholesterol: plasma removal, tissue distribution and biliary sterol secretion in rats.

Evidence has been accumulating for the putative role of chemically or oxidatively altered lipoproteins in accelerating events in the atherogenic process. In this study, the movement of free cholesterol from native high density lipoprotein (HDL) and malondialdehyde (MDA)-modified HDL to the liver for biliary cholesterol secretion and bile acid transformation was examined in vivo. To this end, HDL from normal donor rats was isolated, conjugated with MDA, labelled with [14C]cholesterol and injected i.v. into rats with bile diversion. While the 6 h collection revealed no substantial differences in bile flow, less 14C excretion was recovered in the fresh bile of animals receiving MDA-modified HDL. Bile analysis indicated that a significant decline in labelled bile acid secretion characterized these rats. Compared with controls, MDA-modified HDL also caused an enhanced accumulation of [14C]cholesterol in the liver and the kidneys, with reduced delivery to the sites of steroidogenesis, i.e., the adrenals and testes. No plasma removal differences were noted in the HDL of both groups of rats. Thus, modification of HDL by MDA seems to impair the tissue distribution of its cholesterol moiety, particularly in the liver, where it accumulates at the expense of bile acid transformation.

Modulation of GTP-dependent fusion by linoleic and arachidonic acid in derivatives of rough endoplasmic reticulum from rat liver.

The effect of modulation of the content of unsaturated free fatty acids on GTP-dependent fusion of stripped rough microsomes from rat liver was determined. Cytidine monophosphate, CDP and CTP were all observed to be able to stimulate free fatty acid accumulation and coincident membrane fusion. GTP was required for membrane fusion in the presence of cytidine nucleotide but was not required for free fatty acid accumulation. In the presence of GTP and cytidine nucleotide, the addition of ATP and CoA led to the synthesis of triacyglycerol and marked inhibition of both free fatty acid accumulation and membrane fusion. Delipidated bovine serum albumin also inhibited both free fatty acid accumulation and membrane fusion. Analysis by gas chromatography indicated that linoleic acid and arachidonic acid were the most actively fluctuating of the accumulated free fatty acids. Comparison by quantitation indicated a high correlation between GTP-dependent membrane fusion and changes in amount of unesterified linoleic acid and arachidonic acid. The results suggest that polyunsaturated free fatty acids may be required for GTP-dependent membrane fusion.

Lipoprotein abnormalities associated with cholesteryl ester transfer activity in cystic fibrosis patients: the role of essential fatty acid deficiency.

The purpose of this study was to elucidate the roles of lecithin:cholesterol acyltransferase (LCAT) and cholesteryl ester transfer protein (CETP) in the lipoprotein derangement of cystic fibrosis (CF) patients with respect to their essential fatty acid (EFA) status. Triglyceride enrichment and cholesteryl ester (CE) depletion were observed in the lipoproteins of 22 CF patients. The abnormal chemical composition was more severe in 12 EFA-deficient (EFAD) than in 10 EFA-sufficient (EFAS) patients. Expressed in nmol.L-1.h-1, LCAT activity was higher (P < 0.05) in both EFAS (mean +/- SE, 92.7 +/- 1.9) and EFAD (108.8 +/- 3.0) patients than in control subjects (65.2 +/- 0.9). An equal CE transfer was recorded in the lipoprotein-deficient serum, as a source of CETP activity, in all groups studied by using normal exogenous low-density lipoprotein (LDL) and high-density lipoprotein (HDL). However, in contrast to the maximal amount of CE transferred from endogenous HDL to endogenous apolipoprotein B (apo B) in control subjects, a reduction in CETP activity was seen in CF patients and more pronounced in the EFAD group. These findings indicate that impaired lipoprotein composition may have marked effects on the transfer of CE between HDL and apo B in EFAD CF patients.

Adipose hormone-sensitive lipase preferentially releases polyunsaturated fatty acids from triglycerides.

Rat adipose hormone-sensitive lipase-mediated release of fatty acids from triglycerides was studied in three model systems: i) cultured preadipocytes containing polyunsaturated fatty acid-enriched triglyceride; ii) perfused epididymal fat pads; and iii) in vitro incubations of crude preparations of hormone-sensitive lipase with synthetic triglyceride-analogues as substrates. We found that cultured preadipocytes challenged with 10 microM norepinephrine tended to release more omega 6 and omega 3 polyunsaturated fatty acids than saturated fatty acids. Fat pads perfused with 10 microM norepinephrine preferentially released arachidonate and alpha-linolenate but tended to retain oleate and linoleate. Finally, crude preparations of hormone-sensitive lipase released from the triglyceride-analogue substrates alpha-linolenate twice as fast as oleate. We conclude that rat adipose hormone-sensitive lipase preferentially releases polyunsaturated fatty acids from triglycerides. We suggest that this may be a mechanism by which these fatty acids are kept from being trapped in fat depots and maintained in the circulation.

Essential fatty acid deficiency lowers the activity of the acetylated low density lipoprotein receptor of rat peritoneal macrophages.

We compared phospholipid fatty acid composition, cholesterol ester accumulation, and receptor-mediated binding, internalization, and degradation of acetylated low-density lipoprotein (acetyl-LDL) in cultured peritoneal macrophages from rats fed an essential fatty acid deficient or control diet. The deficient diet increased the 5,8,11-eicosatrienoic acid and decreased the omega 6 fatty acid content of macrophage phospholipid relative to control. The deficient diet did not affect macrophage uptake of [1-14C]oleate; however, it lowered the accumulation of intracellular labelled cholesteroyl oleate to 66% of the control. This effect was attributed to a diminution of the specific binding of acetyl-LDL, and not to acetyl-LDL internalization nor to degradation. The results demonstrate the sensitivity of the acetyl-LDL receptor to changes in its membrane environment, brought about through dietary means.

Modulation of polyunsaturated fatty acid content of triglycerides in rat pre-adipocytes in culture.

Rat peri-renal and epididymal pre-adipocytes in culture undergoing triglyceride (TG) accumulation were incubated with oleic (18:1), linoleic (18:2), alpha-linolenic (18:3 omega 3), arachidonic (20:4) and 4,7,10,13,16,19-docosahexaenoic (22:6 omega 3) acids in the presence of 0.8 microM insulin. The fatty acids were incorporated in cellular TG with relative enrichments over control from 1.4-fold for 18:1 to greater than 40-fold for 18:3 omega 3. Greater than 80% of fatty acids taken up were incorporated into cellular TG. The balance was distributed, in decreasing amounts, into phospholipids, unidentified intracellular constituents, and ketone bodies. The P/S ratio of cellular TG was at least an order of magnitude lower than that of the external milieu for both cell types and for all treatment groups, including controls. Doubling the concentration of treatment fatty acid increased its incorporation into cellular TG. However, it did not affect the accumulation of the other fatty acids in TG. Epididymal cells consistently acquire a higher proportion of treatment fatty acids in cell TG than peri-renal cells. Pre-adipocytes with polyunsaturated fatty acids (PUFA)-enriched TG is a potential model for the study of PUFA metabolism in these types of cells.

Effect of 4,7,10,13,16,19-docosahexaenoic acid on triglyceride accumulation and secretion in rat hepatocytes in culture.

We have investigated the effect of oleic (18:1) and 4,7,10,13,16,19-docosahexaenoic (22:6 omega 3) acids on triglyceride (TG) accumulation, secretion and reuptake in rat hepatocytes in culture. We also calculated the percentage of total TG, TG-esterified 18:1 and TG-esterified 22:6 omega 3 that were secreted relative to the total accumulation (intra + extracellular TG). Both fatty acids were incorporated mainly in the intracellular TG fraction. Treatment with 18:1 but not with 22:6 omega 3 increased the quantity of TG secreted into the culture medium relative to controls. Treatment with 22:6 omega 3 caused a greater accumulation of intracellular TG than 18:1. This arises in part from the preferential retention of 22:6 omega 3-enriched TG by the hepatocytes. At 24 hr, there was no longer any difference in the net secretion of TG by 18:1 and 22:6 omega 3-treated cells, which may be a consequence of the reuptake of TG-esterified 18:1. There was no reuptake of TG-esterified 22:6 omega 3. We conclude that inhibition of hepatocyte TG synthesis is not obligatory for 22:6 omega 3-induced diminution of TG secretion.

Rat liver outer mitochondrial carnitine palmitoyltransferase activity towards long-chain polyunsaturated fatty acids and their CoA esters.

The activity of the overt form of rat liver mitochondrial carnitine palmitoyltransferase or CPT0 (EC 2.3.1.21) towards different fatty acid substrates was studied. The following non-esterified fatty acids (NEFA) and their CoA esters in the presence of 1% bovine serum albumin (BSA) were tested: 16:0, 18:0, 18:1, 18:2, 18:3 omega 3, 20:4, 20:5 omega 3 and 22:6 omega 3. The data fit a square hyperbolic model for enzyme catalysis (p less than 0.001, non-linear regression). Asymptotic Vmax and K0.5, substrate concentration at one-half Vmax, were calculated using total concentrations of acyl-CoA, or unbound concentrations of NEFA. BSA was found to act as a true substrate reservoir for NEFA in that the dissociation of the NEFA-BSA complex was 10-330 times faster than the CPT0 reaction. Regardless of form (NEFA or CoA ester), 18:3 omega 3 gave the highest, while 22:6 omega 3 and 18:0 gave the lowest rates of acylcarnitine synthesis. Except for 18:3 omega 3 and 18:2, Vmax for NEFA was generally lower than for acyl-CoA, with the greatest differences observed for 20:4, 20:5 omega 3 and 22:6 omega 3, suggesting that acyl-CoA synthesis may also be important in the control of the entry of these fatty acids into the mitochondria. The data provide an enzymatic rationale for the relatively low content of 18:3 omega 3 in esterified lipid.

Lipogenesis from ketone bodies in perfused livers from streptozocin-induced diabetic rats.

Production of ketone bodies and their contribution to lipogenesis were measured in isolated livers from normal and streptozocin-induced diabetic (STZ-D) rats perfused with tracer amounts of 3H2O and (R)-3-hydroxy[3-14C]butyrate. Diabetes decreased by 80-95% the total rates of fatty acid and 3-beta-hydroxysterol synthesis in perfused livers and livers of live rats. The activity of cytosolic acetoacetyl-CoA synthetase was slightly (17%) decreased in livers from STZ-D rats. The incorporation of ketone bodies into fatty acids and sterols was markedly inhibited in perfused livers from STZ-D rats despite the stimulation of ketogenesis by diabetes and the presence of oleate. Treatment of the rats with insulin before liver perfusion led to a normalization of the rates of ketogenesis and fatty acid synthesis. The rates of sterol synthesis were only partially normalized by insulin treatment. We conclude that in STZ-D, ketosis does not stimulate hepatic lipogenesis via cytosolic activation of acetoacetate.

Production of acetone and conversion of acetone to acetate in the perfused rat liver.

The utilization of millimolar concentrations of [2-14C]acetone and the production of acetone from acetoacetate were studied in perfused livers from 48-h starved rats. We devised a procedure for determining, in a perfused liver system, the first-order rate constant for the decarboxylation of acetoacetate (0.29 +/- 0.09 h-1, S.E., n = 8). After perfusion of livers with [2-14C]acetone, labeled acetate was isolated from the perfusion medium and characterized as [1-14C]acetate. No radioactivity was found in lactate or 3-hydroxybutyrate. After 90 min of perfusion with [2-14C]acetone, the specific activity of acetate was 30 +/- 4% (n = 13) of the initial specific activity of acetone. We conclude that, in perfused livers from 2-day starved rats, acetone metabolism occurs for the most part via free acetate.