- Email: [email protected]
The effects of alkylthioacetic acids (3-thia fatty acids) on fatty acid metabolism in isolated hepatocytes
Biochimica et Biophysica A cta, 1005 (1989) 296-302 Elsevier
296 BBALIP 53235
The effects of alkylthioacetic acids (3-thia fatty acids) on fatty acid metabolism in isolated hepatocytes Steinar Skrede, Michel N a r c e *, Steinar Bergseth and J o n Bremer Institute of Medical Biochemistry. University of 051o, 0$1o (Norway)
(Received 6 April 1989) (Revised manuscript received 3 July 1989)
Key words: Alkyithioaceticacid; Fatty acid metabolism; (Hepatocyte)
[amg-dudn alkylthioacetic acids (3-thia fatty acids) inhibit fatty acid synthesis from [l.t4Claeetate in isolated (Ah~.~utteojtes, while fatty acid oxidation is nearly unaffected or even stimulated. Desaturation of [l-UC]stearate nmme) is also unaffected. II-UCIDo&eeylthioacetie acid (a 3-thia fatty acid) is incorporated in triacylglycerol and in phoqt~lpids more efficiently than il-UClladmitate in isolated hepatocytes. The metabo!.~,, of [|-UCldode¢ y ~ acid to acid-soluble ImPacts (by ¢o-oxidatlen) is slow compared to the oxidation of II-UCJpaimitate. in hqmtocytet from ~ rats (rats fed tetradeeylthleaeetic acid for 4 days) the rate of il-UClpalmitate oxidation is laerem~ aml its rate of esterifleatien is deerem~ Stearate desaturation is also decreased. The rate of cyanide-insensitive pemxiszmal fatty acid ~oxidaflon is several-fold increased. The metabolic effects of long-chain 3-thia fatty acids are discussed and it is concluded that they behave essentially like nomud fatty acids except for their slow breakdown due to the sulfur atom in the 3 position, which blocks normal ,8-oxidation.
Intmtlnefam Alkylthioacetic acids (3-thia fatty acids) represent non.,8-oxidizable fatty acid analogues in which a S atom substitutes for the /~-methylene group in the chain. Their coenzyme A esters form complexes with acyi-CoA dehydrogenases and act as inhibitors of these enzymes [11. When long-chain 3-thia dicarboxylic acids and longchain 3-thia fatty acids are fed to rats they induce a
drop in plasma triacylglycerol and cholesterol, and induce increased peroxisomal #-oxidation and increased activity of ¢arnitine acyltransferases [2]; i.e., they have effects similar to those of clofibrate and other hypolipaemic drugs [3]. In the present paper we report some studies on the effects of alkylthioacefic acids on fatty acid metabolism in isolated hepatocytes to shed light on their metabolic and hypolipaemic effects in the whole animal.
*' Visiting scientist from: Laboratoire de Physiologie Animale et de la Nutrition, Facult,~des Sciences Mirande, Dijon, France. Cotrespoadeuce: J. Bremer, Institute of Medical Biochemistry, University of Oslo, P.O. Box 1112, Blindern, 0317 Oslo 3, Norway.
Methods Animals and hepatocytes Male Wistar rats of approx. 200 g were fed a standard pellet diet ad libitum until they were killed. Some animals were given tetradecylthioacetic acid (50 mg) in addition as a suspension in 25 carboxymethylcellulose (1 ml) by a gastric tube once a day for 4 days before their use in experiments in the morning the 5th day (adapted animals). Fasted-refed rats were fasted for 2 days and then fed white bread and 205 glucose in the drinking water for 2 days before they were used in experiments. Hepatocytes were prepared according to Seglen [4], and suspended in Krebs-Henseleit bicarbonate buffer containing 15 fatty acid free bovine serum albumin as previously described [5]. In experiments on palmitate oxidation the hepatocytes were preincubated with 1 mM (-)-carnitine for 20 min, thus avoiding insufficient intracellular levels of this compound [6]. In experiments on fatty acid synthesis and desaturation, no extra carnitine was added. Materials [1-~4C]Palmitate, [1-~4C]stearate, [1-]4C]acetate and iodo[1-14C]acetate were purchased from Amersham In-
0005-2760/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
297 ternational, U.K., and diluted to suitable specific activities with the corresponding acids. [1-14C]Palmitoyl-CoA was synthesized according to Kawagushi et al. [7]. Octyl-, dodecyl- and tetradecylthioacetic acid were synthesized from mercaptoaeetic acid and the corresponding alkylbromides as previously described [8]. [1-~4C]Dodecylthioacetic acid was synthesized from dodecanethiol and iodo[1-~4C]acetate. Dodecanethiol (25 #1) and KOH (25 mg) in methanol (1 ml) were added to the vial with iodo[1-~4C]acetate and left overnight at room temperature. The reaction mixture was acidified with HCI and the dodecylthioacetic acid was extracted with petroleum ether (3 × 1 ml). The petroleum ether was evaporated and the residue was put on a small silicic acid column (5 × 1 cm) in dry petroleum ether. The column was washed with a few ml of petroleum ether/diethyl ether (4: 1). The reaction product was then eluted with petroleum ether/diethyl ether/ acetic acid (4:1:0.2). Finally, a radioactive side-product was removed by HPLC on a Spherisorb ODS 5 #m column, 250 × 4.6 mm (Supelco, Bellafonte, U.S.A.) eluted with methanol/water/formic acid (85:15:0.5). Thin-layer chromatography on a silicic acid plate with petroleum ether/diethyl e t h e r / f o r m i c acid (70:30:4), and rechromatography in the above mentioned HPLC system both showed only one radioactive peak. Fatty acid free bovine albumin and collagenase type I were obtained from Sigma, St. Louis, U.S.A., and silica-gel aluminium thin-layer chromatography plates from Merck, Darmstadt, F.R.G.
Incubations The hepatocytes (20-25 mg protein/5 mi) were incubated at 37°C in Krebs-Henseleit bicarbonate buffer with 3.5% fatty acid free albumin and 0.5 mM [1-14C]palmitate or [1-14C]tetradecylthioacetic acid with or without the other acid unlabelled. The incubations were started by the addition of hepatocytes after gassing of the mixtures with 5% CO 2 in oxygen. Under these conditions 30-35% of the labelled palmitate had been taken up by the cells after 30 min, and 40-55% after 60 min. Assays Palmitate oxidation and esterification. At 30 and 60 min a sample of the incubation mixture (250 /~1) was added to 70 #1 2 M HCIO4, centrifuged to remove proteins and lipids, and assayed for acid soluble radioactivity. Under the incubation conditions used radioactive CO 2 amounts to only a few per cent as much as the acid soluble radioactivity. In the present study, the acid-soluble radioactivity therefore was taken as a sufficiently accurate measurement of oxidation. Other samples of the incubation mixture (1 ml) were cooled on ice and centrifuged, washed once with an
equal volume of incubation buffer to remove remaining free radioactive substrates (negligible amounts of esterifled acids had been secreted from the cells). The pellet was resuspended in 2 ml of water and the lipids were immediately extracted with 1 ml of n-butanol by vortexing and phase separation by centrifugation after addition of a little solid NaC1 to improve the phase separation. A sample of the butanol phase was used for total radioactivity determination. The sum of acid soluble radioactivity and radioactivity extracted from the cells with butanol amounts to total uptake by the cells. Another butanol sample was evaporated under a stream of air, the lipid residue was dissolved in chloroform/ methanol (3:1 v/v) and chromatographed on Silica gel aluminium thin layer chromatography plates. Two separate systems were run. In the first, using a mobile phase of hexane/diethyi ether/acetic acid (70:30:2), we separated phospholipids, diacyiglycerol, free fatty acids and triacylglycerol. Chromatography distribution is explained in Fig. 1. In the other we separated subclasses of phospholipids using a mobile phase of chloroform/methanol/acetic acid/water (65 : 2 5 : 4 : 4 ) as shown in Fig. 2. All samples were run in duplicate tracks. One track was exposed to iodine vapor, localizing lipids as brown spots; the other we cut into corresponding zones, immersing them in scintillation fluid for counting. Induction of peroxisomal fatty acid oxidation in adapted rats was confirmed by the method of Mannaerts et al. [9]. Desaturation and oxidation of stearate. Stearate desaturation was assayed by incubating 0.1 mM [1~4C]stearate with hepatocytes in the presence of 10 mM lactate and 5 mM glucose. The incubations were stopped by addition of 3.5 ml alcoholic KOH (12% w/v). After saponification at 75°C for 1 h and acidification with 1 ml of concentrated HCI, the fatty acids were isolated by suction of the reaction mixture through Sep-Pack C18 cartridges (Millipore Corporation, Milford, U.S.A.). The cartridge was washed with 15 ml methanol/water (20:80) to remove water-soluble products, followed by elution of fatty acids from the cartridge with 7 ml heptane. After evaporation of the heptane, the fatty acids were dissolved in acetonitrile and the conversion of stearate to oleate (A9 desaturation) was measured with high performance liquid chromatography on a Spherisorb ODS column (5/~m, 4.6 × 250 mrr,) c!uted at 1 ml/min with an acetonitrile/water gradiert (55-100% over 15 min) containing 0.5% acetic acid. Radioactivity was detected directly on line in a Raytest Ramona 5 LS radioactivity detector equipped with a 1 mi flow cell. Eluate and liquid scintillator were mixed 1:6. Fatty acid synthesis from acetate. Hepatocytes (1 mi) were incubated with 4 mM [1-14C]acetate (0.25 mCi/mmol) and 10 mM lactate. The incubations were
298
i e~
o
b
c
I
u
I
0
"0 0 L @
t
(Z
0
10
5
5
10 cm
¢m
i
._> U 0
.g
'10
|
10
0
g
0
cm
¢m
lo
Fi$, I. Incorporation of dodecylthioacetic acid and pa|mitic acid in triacylglyceroi in isolated hcpatocytes. Normal hepatocytes were incubated for 60 rain with [i-t4C]dodecylthioacetic acid (0.5 raM) or [1-Z4C]palmitic acid (0,5 raM) with or without the other acid (0.5 raM) unlabeled. The cell lipids were extracted and chromatolpraphed on thin4ayer sili¢ic acid plates with hexane/ethyl ether/acetic acid (70: 30: 2). The liquid front was at 17 cm, The heis,ht of the c-olunms 6ires the ~lative radioactivity of the spots, The lipids were separated into phospholipids ((a) see also Fig. 2), free fatty acids (b), diacylglycerol (c), and triacylglycerol (d), (A) [ |- t4 C]palmitic acid; (B) [ !- t4 CJpalmitic acid + unlabeiled dodecylthioacetic acid; (C) [l-t4C)dodecylthJoaceti¢ acid: (D) [!-t4C]dodecyithioacctic acid + unlabelled palmitic acid. (See Tables I and iil for total incorporation.)
B
,_> O
0
b
c
a
b
c
@
Cm
cm
Fig. 2. In¢~'l~ration of [1-Z4Cldodecyltldoacetic acid and [l°t4C]palmific acid in phospholipids. The pho.~p]~olipids at the origin of d~romato~rams shown m Fi& ] were eluted in chloroform/methanol (3: l) and z~hromato~raphed on silicic acid plates with chloroform/methanol/acetic acid/H 20 (65: 25: 4- 4). The iipids were separated into phosphafidylchollne (a), phosphatidylethanolamine (b) and unidentified radioactivity at the front (c). (A) []-z4C]Palmific add; B, [l- 14 C]dodccyltldoacetic acid.
299 stopped by addition of 2 ml alcoholic KOH (12% w/v). The mixtures were saponified, acidified and fractionated on Sep-Pack C!8 cartridges as in the desaturation assay. The heptane eluate was evaporated under a stream of air in counting vials and the radioactivity was determined after addition of scintillation fluid. Protein. Protein was determined according to Lowry et al. [10]. Results
Effects on fatty acid oxidation and esterification Table I shows that equimolar amounts of alkylthioacetic acids in the incubation medium have only moderate effects on the metabolism of palmitate in isolated normal hepatocytes. Since almost no free fatty acid was recovered from the cells, total uptake amounts to the sum of acid-soluble products plus esterified products. When these are added up in Table I, it is seen that total uptake of pahnitate was inhibited about 30% by dodecylthioacetic acid, le:,s by the ectyl and tetradecyl derivatives. The dodecyl and the tetradecyl derivatives had some inhibitory effects on the incorporation of palmitate into triaeylglycerol and particularly phospholipids. The effects on oxidation varied depending on the chain length of the fatty acid analogues. The octylthioacetic acid had no significant effect, dodecylthioacetic acid had a moderately inhibitory effect, while the tetradecyl derivative consistently had a small stimulatory effect on palmitate oxidation. (Significant only when calculated as per cent of control in each experiment.) This stimulatory effect of the tetradecylthioacetic acid on oxidation was much more pronounced with stearate as substrate in the presence of lactate (to increase stearate desaturation). Table II shows that the oxidation of 0.1 mM stearate was
nearly doubled in the presence oF 0.5 mM tetradecyltl'fioacetic acid. Tables I and II also show that palmitate and stearate oxidation was about doubled in cells isolated from adapted rats. The incorporation into diacylglycerol was also higher. The sum of reaction products in Table I shows that the total uptake increased about 20% in adapted cells. The incorporation into phospholipids, and particularly triacylglycerol, was decreased. Incorporation into phospholipids was further decreased by the addition of the fatty acid analogue to the incubation medium, and as in normal cells the oxidation of stearate was increased. Control experiments showed that the cyanide-insensitive oxidation of [1-14C]palmitoyl CoA was about 6-fold increased in liver homogenates from rats fed tetradecylthioaceti¢ acid, confirming that peroxisomal fl-oxidation had been induced [2].
Incorporation of alkylthioacetic acids in lipids In the thin-layer chromatograms for the assay of palrnitate incorporation in triacylglycerol we noticed that the radioactive profile of the triacylglycerol peak was markedly changed when dodecyl- and tetradecylthioacetic acid had been added to the incubations (Fig. 1). We have therefore synthesized (1-14C)-labelled dodecylthioacetic acid and have measured its incorporation into lipids. Fig. lc and d shows that radioactivity was recovered in the broad triacylglycerol spot in the chromatograms. Radioactive dodecylthioacetic acid was also recovered in the phospholipid peaks, mainly in phosphatidylethanolamine and phosphatidyicholine (Fig. 2B). Table III shows that dodecylthioacetic acid is incorporated in both triacylglycerol and phospholipids to even a greater extent than palmitate, whereas it is only
TABLE I Effects of alkylthioacetic acids (0. 5 raM) on oxidation and esterification of palmitate (0. 5 raM) in normal hepatocytes (normal) and hepatocytes from rats fed tetradecylthioacetic acid (adapted) The results are given as nmol palmitate oxidized or esterified in triacylglycerol or phospholipids per mg protein _+S.D. for the number of rats in parentheses. Acid-soluble products
Diacylglycerol
Triacylglycerol
normal
adapted
normal
adapted
normal
30 min None Octyl Dodecyi Tetradecyl
10.7_+2.9(7) 10A 5:1.7 (3) 7.7_+0.9 (7) 14.4_+1.8(4)
19.3__+5.7(8) 16.8_+5.9(3) 18.2+_5.8(4) 22.5+_6.8(3)
1.7_+0.7(4) 1.1+__0(2) 1.6__+0.6(3) 1.0+ 0.1 (2)
2.7_+0.9(6) 3.0_+0.6(3) 1.7-+0.5(4) 1.8_+0.3(3)
60 rain None Octyi Dodecyl Tetradecyl
17.0:t:4.2 (7) 17.7+4.3 (3) 13.7+2.0 (7) 23.1 _+3.8 (3)
30.2__+9.2(8) 27.1 +7.0 (3) 28.8-+7.7(4) 37.0+7.8 (3)
1.3_+0.1(4) 0.7_+0(2) 0.9_+0.3(4) 0.9+0.2 (2)
2.6_+1.2 (6) 1.9_+0.8(3) 1.2+0.6 (4) 1.4-+0.7(3)
Alkylthioacetic acid added
9.1_+2.7(4) 7.2_+1.6(2) 7.3_+ 1.7 (3) 7.1+3.3(2) 16.1 _+3.0 (4) 9.9_+0.1(2) 10.6-+2.9(3) 9.9-+ 3.5 (2)
Phospholipids adapted
normal
adapted
4.7_+2.3(6) 11.7+2.0(4) 10.7_+2.6(4) 5.0_+2.2(3) 8.7_+2.4(2) 10.4_+1.2(3) 5.0_+1.6 (4) 5.9 :l: 1.4 (3) 5.05:0.6 (4) 5.4_+1.6(3) 5.2_+1.5(2) 4.85:0.9(3) 7.9+4.7 (6) ,5.7_+4.7(3) 6.7-+2.5(4) 8.4_+4.9(3)
16.9 :t: 2.0 (4) 11.7+ 1.8 (2) 8.9-+2.8(3) 6.8_+ 1.8 (2)
17.2_+4.5(6) 15.2:1:2.7 (3) 6.5+ 1.5 (4) 5.4_+2.2(3)
300 TABLE I!
Effect of tetradecylthioacetic acid (0.3 raM) on oxidation and desaturation of stearic acid (0.1 raM) in isolated hepatocyte~ from normal rats and rats fed the tetradecylthioacetic acid for 4 days The standard incubation medium was fortified with 10 mM lactate and 10 mM glucose. The results are given as nmoi/lO mg protein 4- S.D. {[our experiments) Addition
Incubation
Oxidation
time
normal
None Tetradecyl thioacetic acid
60 rain
None Tetradecyl thioacetic acid
120 rain
~ ; m
A
Desaturation adapted
normal
adapted
11.5 4-2.0
18.5 4- 2.1
11.64- 2.5
10.25 "4-!.7
18.7 4- 0.9
22.4 4- 3.5
12.7 4-1.6
14.2 + !.6
17.9 4. 2.5
31,5 ± 2.9
21.4 4. 0.9
16.2 ± 2.0
28.1 4- 2.0
32.9 ~ 2.6
21.9 ± 2.0
6.25 4-1.35
slowly oxidized to acid-soluble products. Studies on the oxidation products (formed by ~0-oxidation) will be reported separately.
45
Effecton fattyacid synthesis
3O
[e <. c
InhtUit.or (raM)
00leote
I
0~.
0,4
InhibitOr (raM]
Fig, 3, Effects of tetradecylthioacetic acid and of oleic acid on the biosynthesis of fatty acids from [lJ4C]acetate (4 raM) in hepatocytes from normal and fasted-rcfed rats, The standard incubation medium was fortified with 10 mM lactate, The incubation time was 30 nun, The results are given as nmol acetate incorporated/rag protein :k S.D. of four experiments with parallels (eight values). (A) normal rats; (B) fasted-tared rats, o, oleic acid; e, tetradecylthioacetic acid,
Fig. 3 shows that tetradecylthioacetic acid has a strong inhibitory effect on the incorporation of acetate into fatty acids. With a concentration of 0.3 mM in the incubation medium the incorporation of acetate was nearly 90% inhibited, while the same concentration of oleate inhibited about 60~. In hepatocytes from fasted refed rats with a 4-5.fold higher incorporation of acetate in the fatty acids the inhibition was less pronounced, but again the effect of the alkylthioacetic acid was stronger than that of oleate. We also tested the long-term effect of tetradecylthioacetic acid on fatty acid synthesis in hepatocytes in tissue culture. The hepatocytes were kept in culture fo~ 48 h with 0.25 mM tetradecylthioacetic acid in the medium. Incorporation of acetate was then measured in the absence of the alkylthioacetic acid. The incorpora-
TABLE Ill Con~oa~son of ll J ~Clpalmitate fO.5 raM) and [l.l ~Cldodecylthioacetic acid f0.5 raM) metabolism in isolated normal hepatotytes Each labelled acid was incubated in the presence of the other acid unlabelled. The results are given as labelled acid (nmol/mg protein) recovered in acid-soluble products, in diacylglyceml + Iriacylglycerol, or in phospholipids, after 30 or 60 rain. Values are 1,lean + S.D. for the number of samples in parentheses. Substrata [IJ4ClPalmitate + dodecylthio. acetic acid [1J'ClDedecylthioacetic acid + palmitate
Acid-soluble products
Phospholipids
30 rain
60 rain
30 min
7.7 ::kO.9(7)
13.7+2,0 (7)
5.9+1.4(3)
8.9+2.8(3)
8.8+2.4(3)
11.6+3.1 (3)
2.0:k0.6 (3)
4.0+1.8(3)
12.0::I:2.9(2)
16.5 + 5.6 (3)
12.3+0.7(3)
16.5+3.9(3)
60 min
Tri- and diacylglycerol 30 min 60 rain
301 tion of acetate was 50% inhibited when compared with cells kept in culture in the absence of inhibitor. Apparently, besides the direct inhibition of fatty acid synthesis (Fig. 3), tetradecylthioacetic acid leo to a down-regulation of one or more enzymes in the biosynthesis of fatty acids. Effect on stearate desaturation (Z~9-desaturase) Table II shows that tetradecylthioacetic acid had no significant effect on the desaturation of stearate in isolated hepatocytes, even when the analogue is added in a concentration 5-times higher than that of stearate. However, in cells from rats fed 50 mg of tetradecylthioacetic acid for 4 days the desaturation was reduced about one-third. Again, addition of the analogue to the adapted cells had no effect. Since the alkylthioacetic acids are activated and incorporated in triacylglycerol and phospholipids, it is a possibility that they are also desaturated. So far we have not developed methods to show the presence of desaturated alkyltlfioacetic acids. D|scuss|on
It is striking that the long-chain 32-thia acids in several ways behave as long-chain fatty acids. When palmitate and dodeeylthioacetic acid are present together in the medium they are taken up by hepatocytes at similar rates (Table IIl). Their combined uptake rate is about 1.5-times that of palmitate (compare Table III and Table I). Since very little free fatty acid was found in the cells, it is evident that the uptake is limited by their rate of metabolism. When the concentration of palmitate is doubled from 0.5 to 1 mM in the medium, its uptake i~ nearly doubled [3]. Considering that the 3-thia acids are broken down only slowly, it is evident that the cells handle palmitate and dodecylthioacetic acid in a similar way except for their oxidation (Table III). It is striking that both are rapidly incorporated into lipids, and that this explains their rapid uptake. The incorporation of dodecylthioacetic acid even exceeded that of palmitate. Evidently, its coenzyme A ester accumulated in the cells to a greater extent than did that of palmitate, 'forcing' its incorporation into the lipids. In unpublished experiments we have found that the coenzyme A esters of the leng-chain 3-thia acids are good substrates for carnitine palmitoyltransferase. These CoA esters have been shown to form complexes with the acyI-CoA dehydrogenase [1]. It is surprising, therefore, that tetradecylthioaeetic acid (very similar to dodecylthioacetic acid) did not inhibit palmitate oxidation in hepatocytes. On the contrary, it stimulated the oxidation of palmitate and that of stearate even more. This cannot be explained by an increased uptake of fatty acid from the medium, since Table I shows that palmitate uptake was inhibited.
A possible explanation may be found in the effect of the accumulated 3-thiaacy~-CoA on fatty acid synthesis. Presumably, the CoA ester inhibks malonyl-CoA formation and, as the concentration of this intermediate drops, the inhibition of the outer carnitine acyltranso ferase I is relieved [11]. An increased activity of this enzyme may compensate for a direct inhibition of fatty acid oxidation. High concentrations of fat in the diet, especially fats which contain very-long-chain fatty acids, induce increased peroxisomal fatty aci~ oxidation [12,13]. In recent feeding experiments we have shown that bis(carboxymethylthio)decane, a thiadicarboxylic acid, and 3-thia fatty acids have the same effect [2]. This effect, similar to the effect of clofibrate and other hypolipaemic drugs, was reproduced in the present study, and the hepatocytes isolated from such adapted rats oxidized more and esterified less palmitate, like hepatocytes from rats fed high-fat diets or clofibrate [3,5]. The cells from adapted rats desaturated less stearate. Since no direct inlfibition of the z~9-desaturase was found, it is likely that the thia acids lead to a down-regulation of this enzyme. In this respect, tetradecylthioacetic acid apparently is like high-fat diets [14] and different from clofibrate and other peroxisome proliferators in its effect. They increase the activity of the A9-desaturase [15]. Altogether, our studies show that moderate amounts of the long-chain thia acids in the diet have similar and stronger effects on fatty acid metabolism than have fasting and high fat diets. We have also observed such a strong effect on fatty acid metabolism in hepatoma cells in tissue culture. The 3-thia fatty acids proved to be at least 20-fold more efficient in inducing peroxisomal fatty acid oxidation than were normal fatty acids [8]. Since the 3-tlfia fatty acids behave in several ways as normal fatty acids, it is evident that a sulfur atom substituting for the methylene group in the 3-position does not change the properties of the fatty acid drastically, except that it cannot be fl-oxidized. These fatty acid analogues therefore may be useful tools in studies of regulatory mechanisms in lipid metabolism. Acknowledgement This work has been supported by the Norwegian Research Council of Science and the Humanities and by The Norwegian Cancer Society. Re[~rences 1 Lau, S.-M., Brantley, R.K. and Thorpe, C. (1988) Biochemistry 27, 5089-5095. 2 Berge, R.K., Aarsland, A., Kryvi, H., Bremer, J. and Aarsaeter, N. (1989) Biochem. Pharmacoi., in press. 3 Christiansen, R.Z. (1978) Biochim. Biophys. Acta 530, 314-324. 4 Seglen, P.O. (1973) Exp. Ceil. Res. 82, 391-398.
302 5 lkrgseth, S., C. Christiansen, E.N. and Bremer, J. (1986) Lipids 21, 508-514. 6 Christiansen, R.Z. (1977) Biochim. Biophys. Acta 488, 249-262. 7 Kawagushi, A., Soskimura, T. and Okuda, S. (1980) J. Biochem. 89, 337-339. 8 Spydevold, O. and Bremer, J. (1989) Biochim. Biophys. Acta 1003, 72-79. 9 Mannaerts, G.P., Debeer, LJ., Thomas, J. and De Schepper, P.J. (1979) J. Biol. Chem. 254, 4585-4595.
10 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, RJ. (1951) J. Biol. Chem. 193, 265-275. 11 McGarry, J.D., Leatherman, G.F. and Foster, D.W. (1978). 12 Bremer, J. and Norum, K.R. (1982) .I. Lipid Res. 23, 243-256. 13 Chfistiansen, R.Z., Christiansen, E.N. and Bremer, J. (1979) Biochim. Biophys. Acta 573, 417-429. 14 Jeffcoat, P,. and James, A.T. (1977) Lipids 12, 469-474. 15 Kawashima, Y., Hanioka, N., Matsumura, M. and Kozuka, H. (1983) Biochim. Biophys. Acta 752, 259-264.
296 BBALIP 53235
The effects of alkylthioacetic acids (3-thia fatty acids) on fatty acid metabolism in isolated hepatocytes Steinar Skrede, Michel N a r c e *, Steinar Bergseth and J o n Bremer Institute of Medical Biochemistry. University of 051o, 0$1o (Norway)
(Received 6 April 1989) (Revised manuscript received 3 July 1989)
Key words: Alkyithioaceticacid; Fatty acid metabolism; (Hepatocyte)
[amg-dudn alkylthioacetic acids (3-thia fatty acids) inhibit fatty acid synthesis from [l.t4Claeetate in isolated (Ah~.~utteojtes, while fatty acid oxidation is nearly unaffected or even stimulated. Desaturation of [l-UC]stearate nmme) is also unaffected. II-UCIDo&eeylthioacetie acid (a 3-thia fatty acid) is incorporated in triacylglycerol and in phoqt~lpids more efficiently than il-UClladmitate in isolated hepatocytes. The metabo!.~,, of [|-UCldode¢ y ~ acid to acid-soluble ImPacts (by ¢o-oxidatlen) is slow compared to the oxidation of II-UCJpaimitate. in hqmtocytet from ~ rats (rats fed tetradeeylthleaeetic acid for 4 days) the rate of il-UClpalmitate oxidation is laerem~ aml its rate of esterifleatien is deerem~ Stearate desaturation is also decreased. The rate of cyanide-insensitive pemxiszmal fatty acid ~oxidaflon is several-fold increased. The metabolic effects of long-chain 3-thia fatty acids are discussed and it is concluded that they behave essentially like nomud fatty acids except for their slow breakdown due to the sulfur atom in the 3 position, which blocks normal ,8-oxidation.
Intmtlnefam Alkylthioacetic acids (3-thia fatty acids) represent non.,8-oxidizable fatty acid analogues in which a S atom substitutes for the /~-methylene group in the chain. Their coenzyme A esters form complexes with acyi-CoA dehydrogenases and act as inhibitors of these enzymes [11. When long-chain 3-thia dicarboxylic acids and longchain 3-thia fatty acids are fed to rats they induce a
drop in plasma triacylglycerol and cholesterol, and induce increased peroxisomal #-oxidation and increased activity of ¢arnitine acyltransferases [2]; i.e., they have effects similar to those of clofibrate and other hypolipaemic drugs [3]. In the present paper we report some studies on the effects of alkylthioacefic acids on fatty acid metabolism in isolated hepatocytes to shed light on their metabolic and hypolipaemic effects in the whole animal.
*' Visiting scientist from: Laboratoire de Physiologie Animale et de la Nutrition, Facult,~des Sciences Mirande, Dijon, France. Cotrespoadeuce: J. Bremer, Institute of Medical Biochemistry, University of Oslo, P.O. Box 1112, Blindern, 0317 Oslo 3, Norway.
Methods Animals and hepatocytes Male Wistar rats of approx. 200 g were fed a standard pellet diet ad libitum until they were killed. Some animals were given tetradecylthioacetic acid (50 mg) in addition as a suspension in 25 carboxymethylcellulose (1 ml) by a gastric tube once a day for 4 days before their use in experiments in the morning the 5th day (adapted animals). Fasted-refed rats were fasted for 2 days and then fed white bread and 205 glucose in the drinking water for 2 days before they were used in experiments. Hepatocytes were prepared according to Seglen [4], and suspended in Krebs-Henseleit bicarbonate buffer containing 15 fatty acid free bovine serum albumin as previously described [5]. In experiments on palmitate oxidation the hepatocytes were preincubated with 1 mM (-)-carnitine for 20 min, thus avoiding insufficient intracellular levels of this compound [6]. In experiments on fatty acid synthesis and desaturation, no extra carnitine was added. Materials [1-~4C]Palmitate, [1-~4C]stearate, [1-]4C]acetate and iodo[1-14C]acetate were purchased from Amersham In-
0005-2760/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
297 ternational, U.K., and diluted to suitable specific activities with the corresponding acids. [1-14C]Palmitoyl-CoA was synthesized according to Kawagushi et al. [7]. Octyl-, dodecyl- and tetradecylthioacetic acid were synthesized from mercaptoaeetic acid and the corresponding alkylbromides as previously described [8]. [1-~4C]Dodecylthioacetic acid was synthesized from dodecanethiol and iodo[1-~4C]acetate. Dodecanethiol (25 #1) and KOH (25 mg) in methanol (1 ml) were added to the vial with iodo[1-~4C]acetate and left overnight at room temperature. The reaction mixture was acidified with HCI and the dodecylthioacetic acid was extracted with petroleum ether (3 × 1 ml). The petroleum ether was evaporated and the residue was put on a small silicic acid column (5 × 1 cm) in dry petroleum ether. The column was washed with a few ml of petroleum ether/diethyl ether (4: 1). The reaction product was then eluted with petroleum ether/diethyl ether/ acetic acid (4:1:0.2). Finally, a radioactive side-product was removed by HPLC on a Spherisorb ODS 5 #m column, 250 × 4.6 mm (Supelco, Bellafonte, U.S.A.) eluted with methanol/water/formic acid (85:15:0.5). Thin-layer chromatography on a silicic acid plate with petroleum ether/diethyl e t h e r / f o r m i c acid (70:30:4), and rechromatography in the above mentioned HPLC system both showed only one radioactive peak. Fatty acid free bovine albumin and collagenase type I were obtained from Sigma, St. Louis, U.S.A., and silica-gel aluminium thin-layer chromatography plates from Merck, Darmstadt, F.R.G.
Incubations The hepatocytes (20-25 mg protein/5 mi) were incubated at 37°C in Krebs-Henseleit bicarbonate buffer with 3.5% fatty acid free albumin and 0.5 mM [1-14C]palmitate or [1-14C]tetradecylthioacetic acid with or without the other acid unlabelled. The incubations were started by the addition of hepatocytes after gassing of the mixtures with 5% CO 2 in oxygen. Under these conditions 30-35% of the labelled palmitate had been taken up by the cells after 30 min, and 40-55% after 60 min. Assays Palmitate oxidation and esterification. At 30 and 60 min a sample of the incubation mixture (250 /~1) was added to 70 #1 2 M HCIO4, centrifuged to remove proteins and lipids, and assayed for acid soluble radioactivity. Under the incubation conditions used radioactive CO 2 amounts to only a few per cent as much as the acid soluble radioactivity. In the present study, the acid-soluble radioactivity therefore was taken as a sufficiently accurate measurement of oxidation. Other samples of the incubation mixture (1 ml) were cooled on ice and centrifuged, washed once with an
equal volume of incubation buffer to remove remaining free radioactive substrates (negligible amounts of esterifled acids had been secreted from the cells). The pellet was resuspended in 2 ml of water and the lipids were immediately extracted with 1 ml of n-butanol by vortexing and phase separation by centrifugation after addition of a little solid NaC1 to improve the phase separation. A sample of the butanol phase was used for total radioactivity determination. The sum of acid soluble radioactivity and radioactivity extracted from the cells with butanol amounts to total uptake by the cells. Another butanol sample was evaporated under a stream of air, the lipid residue was dissolved in chloroform/ methanol (3:1 v/v) and chromatographed on Silica gel aluminium thin layer chromatography plates. Two separate systems were run. In the first, using a mobile phase of hexane/diethyi ether/acetic acid (70:30:2), we separated phospholipids, diacyiglycerol, free fatty acids and triacylglycerol. Chromatography distribution is explained in Fig. 1. In the other we separated subclasses of phospholipids using a mobile phase of chloroform/methanol/acetic acid/water (65 : 2 5 : 4 : 4 ) as shown in Fig. 2. All samples were run in duplicate tracks. One track was exposed to iodine vapor, localizing lipids as brown spots; the other we cut into corresponding zones, immersing them in scintillation fluid for counting. Induction of peroxisomal fatty acid oxidation in adapted rats was confirmed by the method of Mannaerts et al. [9]. Desaturation and oxidation of stearate. Stearate desaturation was assayed by incubating 0.1 mM [1~4C]stearate with hepatocytes in the presence of 10 mM lactate and 5 mM glucose. The incubations were stopped by addition of 3.5 ml alcoholic KOH (12% w/v). After saponification at 75°C for 1 h and acidification with 1 ml of concentrated HCI, the fatty acids were isolated by suction of the reaction mixture through Sep-Pack C18 cartridges (Millipore Corporation, Milford, U.S.A.). The cartridge was washed with 15 ml methanol/water (20:80) to remove water-soluble products, followed by elution of fatty acids from the cartridge with 7 ml heptane. After evaporation of the heptane, the fatty acids were dissolved in acetonitrile and the conversion of stearate to oleate (A9 desaturation) was measured with high performance liquid chromatography on a Spherisorb ODS column (5/~m, 4.6 × 250 mrr,) c!uted at 1 ml/min with an acetonitrile/water gradiert (55-100% over 15 min) containing 0.5% acetic acid. Radioactivity was detected directly on line in a Raytest Ramona 5 LS radioactivity detector equipped with a 1 mi flow cell. Eluate and liquid scintillator were mixed 1:6. Fatty acid synthesis from acetate. Hepatocytes (1 mi) were incubated with 4 mM [1-14C]acetate (0.25 mCi/mmol) and 10 mM lactate. The incubations were
298
i e~
o
b
c
I
u
I
0
"0 0 L @
t
(Z
0
10
5
5
10 cm
¢m
i
._> U 0
.g
'10
|
10
0
g
0
cm
¢m
lo
Fi$, I. Incorporation of dodecylthioacetic acid and pa|mitic acid in triacylglyceroi in isolated hcpatocytes. Normal hepatocytes were incubated for 60 rain with [i-t4C]dodecylthioacetic acid (0.5 raM) or [1-Z4C]palmitic acid (0,5 raM) with or without the other acid (0.5 raM) unlabeled. The cell lipids were extracted and chromatolpraphed on thin4ayer sili¢ic acid plates with hexane/ethyl ether/acetic acid (70: 30: 2). The liquid front was at 17 cm, The heis,ht of the c-olunms 6ires the ~lative radioactivity of the spots, The lipids were separated into phospholipids ((a) see also Fig. 2), free fatty acids (b), diacylglycerol (c), and triacylglycerol (d), (A) [ |- t4 C]palmitic acid; (B) [ !- t4 CJpalmitic acid + unlabeiled dodecylthioacetic acid; (C) [l-t4C)dodecylthJoaceti¢ acid: (D) [!-t4C]dodecyithioacctic acid + unlabelled palmitic acid. (See Tables I and iil for total incorporation.)
B
,_> O
0
b
c
a
b
c
@
Cm
cm
Fig. 2. In¢~'l~ration of [1-Z4Cldodecyltldoacetic acid and [l°t4C]palmific acid in phospholipids. The pho.~p]~olipids at the origin of d~romato~rams shown m Fi& ] were eluted in chloroform/methanol (3: l) and z~hromato~raphed on silicic acid plates with chloroform/methanol/acetic acid/H 20 (65: 25: 4- 4). The iipids were separated into phosphafidylchollne (a), phosphatidylethanolamine (b) and unidentified radioactivity at the front (c). (A) []-z4C]Palmific add; B, [l- 14 C]dodccyltldoacetic acid.
299 stopped by addition of 2 ml alcoholic KOH (12% w/v). The mixtures were saponified, acidified and fractionated on Sep-Pack C!8 cartridges as in the desaturation assay. The heptane eluate was evaporated under a stream of air in counting vials and the radioactivity was determined after addition of scintillation fluid. Protein. Protein was determined according to Lowry et al. [10]. Results
Effects on fatty acid oxidation and esterification Table I shows that equimolar amounts of alkylthioacetic acids in the incubation medium have only moderate effects on the metabolism of palmitate in isolated normal hepatocytes. Since almost no free fatty acid was recovered from the cells, total uptake amounts to the sum of acid-soluble products plus esterified products. When these are added up in Table I, it is seen that total uptake of pahnitate was inhibited about 30% by dodecylthioacetic acid, le:,s by the ectyl and tetradecyl derivatives. The dodecyl and the tetradecyl derivatives had some inhibitory effects on the incorporation of palmitate into triaeylglycerol and particularly phospholipids. The effects on oxidation varied depending on the chain length of the fatty acid analogues. The octylthioacetic acid had no significant effect, dodecylthioacetic acid had a moderately inhibitory effect, while the tetradecyl derivative consistently had a small stimulatory effect on palmitate oxidation. (Significant only when calculated as per cent of control in each experiment.) This stimulatory effect of the tetradecylthioacetic acid on oxidation was much more pronounced with stearate as substrate in the presence of lactate (to increase stearate desaturation). Table II shows that the oxidation of 0.1 mM stearate was
nearly doubled in the presence oF 0.5 mM tetradecyltl'fioacetic acid. Tables I and II also show that palmitate and stearate oxidation was about doubled in cells isolated from adapted rats. The incorporation into diacylglycerol was also higher. The sum of reaction products in Table I shows that the total uptake increased about 20% in adapted cells. The incorporation into phospholipids, and particularly triacylglycerol, was decreased. Incorporation into phospholipids was further decreased by the addition of the fatty acid analogue to the incubation medium, and as in normal cells the oxidation of stearate was increased. Control experiments showed that the cyanide-insensitive oxidation of [1-14C]palmitoyl CoA was about 6-fold increased in liver homogenates from rats fed tetradecylthioaceti¢ acid, confirming that peroxisomal fl-oxidation had been induced [2].
Incorporation of alkylthioacetic acids in lipids In the thin-layer chromatograms for the assay of palrnitate incorporation in triacylglycerol we noticed that the radioactive profile of the triacylglycerol peak was markedly changed when dodecyl- and tetradecylthioacetic acid had been added to the incubations (Fig. 1). We have therefore synthesized (1-14C)-labelled dodecylthioacetic acid and have measured its incorporation into lipids. Fig. lc and d shows that radioactivity was recovered in the broad triacylglycerol spot in the chromatograms. Radioactive dodecylthioacetic acid was also recovered in the phospholipid peaks, mainly in phosphatidylethanolamine and phosphatidyicholine (Fig. 2B). Table III shows that dodecylthioacetic acid is incorporated in both triacylglycerol and phospholipids to even a greater extent than palmitate, whereas it is only
TABLE I Effects of alkylthioacetic acids (0. 5 raM) on oxidation and esterification of palmitate (0. 5 raM) in normal hepatocytes (normal) and hepatocytes from rats fed tetradecylthioacetic acid (adapted) The results are given as nmol palmitate oxidized or esterified in triacylglycerol or phospholipids per mg protein _+S.D. for the number of rats in parentheses. Acid-soluble products
Diacylglycerol
Triacylglycerol
normal
adapted
normal
adapted
normal
30 min None Octyl Dodecyi Tetradecyl
10.7_+2.9(7) 10A 5:1.7 (3) 7.7_+0.9 (7) 14.4_+1.8(4)
19.3__+5.7(8) 16.8_+5.9(3) 18.2+_5.8(4) 22.5+_6.8(3)
1.7_+0.7(4) 1.1+__0(2) 1.6__+0.6(3) 1.0+ 0.1 (2)
2.7_+0.9(6) 3.0_+0.6(3) 1.7-+0.5(4) 1.8_+0.3(3)
60 rain None Octyi Dodecyl Tetradecyl
17.0:t:4.2 (7) 17.7+4.3 (3) 13.7+2.0 (7) 23.1 _+3.8 (3)
30.2__+9.2(8) 27.1 +7.0 (3) 28.8-+7.7(4) 37.0+7.8 (3)
1.3_+0.1(4) 0.7_+0(2) 0.9_+0.3(4) 0.9+0.2 (2)
2.6_+1.2 (6) 1.9_+0.8(3) 1.2+0.6 (4) 1.4-+0.7(3)
Alkylthioacetic acid added
9.1_+2.7(4) 7.2_+1.6(2) 7.3_+ 1.7 (3) 7.1+3.3(2) 16.1 _+3.0 (4) 9.9_+0.1(2) 10.6-+2.9(3) 9.9-+ 3.5 (2)
Phospholipids adapted
normal
adapted
4.7_+2.3(6) 11.7+2.0(4) 10.7_+2.6(4) 5.0_+2.2(3) 8.7_+2.4(2) 10.4_+1.2(3) 5.0_+1.6 (4) 5.9 :l: 1.4 (3) 5.05:0.6 (4) 5.4_+1.6(3) 5.2_+1.5(2) 4.85:0.9(3) 7.9+4.7 (6) ,5.7_+4.7(3) 6.7-+2.5(4) 8.4_+4.9(3)
16.9 :t: 2.0 (4) 11.7+ 1.8 (2) 8.9-+2.8(3) 6.8_+ 1.8 (2)
17.2_+4.5(6) 15.2:1:2.7 (3) 6.5+ 1.5 (4) 5.4_+2.2(3)
300 TABLE I!
Effect of tetradecylthioacetic acid (0.3 raM) on oxidation and desaturation of stearic acid (0.1 raM) in isolated hepatocyte~ from normal rats and rats fed the tetradecylthioacetic acid for 4 days The standard incubation medium was fortified with 10 mM lactate and 10 mM glucose. The results are given as nmoi/lO mg protein 4- S.D. {[our experiments) Addition
Incubation
Oxidation
time
normal
None Tetradecyl thioacetic acid
60 rain
None Tetradecyl thioacetic acid
120 rain
~ ; m
A
Desaturation adapted
normal
adapted
11.5 4-2.0
18.5 4- 2.1
11.64- 2.5
10.25 "4-!.7
18.7 4- 0.9
22.4 4- 3.5
12.7 4-1.6
14.2 + !.6
17.9 4. 2.5
31,5 ± 2.9
21.4 4. 0.9
16.2 ± 2.0
28.1 4- 2.0
32.9 ~ 2.6
21.9 ± 2.0
6.25 4-1.35
slowly oxidized to acid-soluble products. Studies on the oxidation products (formed by ~0-oxidation) will be reported separately.
45
Effecton fattyacid synthesis
3O
[e <. c
InhtUit.or (raM)
00leote
I
0~.
0,4
InhibitOr (raM]
Fig, 3, Effects of tetradecylthioacetic acid and of oleic acid on the biosynthesis of fatty acids from [lJ4C]acetate (4 raM) in hepatocytes from normal and fasted-rcfed rats, The standard incubation medium was fortified with 10 mM lactate, The incubation time was 30 nun, The results are given as nmol acetate incorporated/rag protein :k S.D. of four experiments with parallels (eight values). (A) normal rats; (B) fasted-tared rats, o, oleic acid; e, tetradecylthioacetic acid,
Fig. 3 shows that tetradecylthioacetic acid has a strong inhibitory effect on the incorporation of acetate into fatty acids. With a concentration of 0.3 mM in the incubation medium the incorporation of acetate was nearly 90% inhibited, while the same concentration of oleate inhibited about 60~. In hepatocytes from fasted refed rats with a 4-5.fold higher incorporation of acetate in the fatty acids the inhibition was less pronounced, but again the effect of the alkylthioacetic acid was stronger than that of oleate. We also tested the long-term effect of tetradecylthioacetic acid on fatty acid synthesis in hepatocytes in tissue culture. The hepatocytes were kept in culture fo~ 48 h with 0.25 mM tetradecylthioacetic acid in the medium. Incorporation of acetate was then measured in the absence of the alkylthioacetic acid. The incorpora-
TABLE Ill Con~oa~son of ll J ~Clpalmitate fO.5 raM) and [l.l ~Cldodecylthioacetic acid f0.5 raM) metabolism in isolated normal hepatotytes Each labelled acid was incubated in the presence of the other acid unlabelled. The results are given as labelled acid (nmol/mg protein) recovered in acid-soluble products, in diacylglyceml + Iriacylglycerol, or in phospholipids, after 30 or 60 rain. Values are 1,lean + S.D. for the number of samples in parentheses. Substrata [IJ4ClPalmitate + dodecylthio. acetic acid [1J'ClDedecylthioacetic acid + palmitate
Acid-soluble products
Phospholipids
30 rain
60 rain
30 min
7.7 ::kO.9(7)
13.7+2,0 (7)
5.9+1.4(3)
8.9+2.8(3)
8.8+2.4(3)
11.6+3.1 (3)
2.0:k0.6 (3)
4.0+1.8(3)
12.0::I:2.9(2)
16.5 + 5.6 (3)
12.3+0.7(3)
16.5+3.9(3)
60 min
Tri- and diacylglycerol 30 min 60 rain
301 tion of acetate was 50% inhibited when compared with cells kept in culture in the absence of inhibitor. Apparently, besides the direct inhibition of fatty acid synthesis (Fig. 3), tetradecylthioacetic acid leo to a down-regulation of one or more enzymes in the biosynthesis of fatty acids. Effect on stearate desaturation (Z~9-desaturase) Table II shows that tetradecylthioacetic acid had no significant effect on the desaturation of stearate in isolated hepatocytes, even when the analogue is added in a concentration 5-times higher than that of stearate. However, in cells from rats fed 50 mg of tetradecylthioacetic acid for 4 days the desaturation was reduced about one-third. Again, addition of the analogue to the adapted cells had no effect. Since the alkylthioacetic acids are activated and incorporated in triacylglycerol and phospholipids, it is a possibility that they are also desaturated. So far we have not developed methods to show the presence of desaturated alkyltlfioacetic acids. D|scuss|on
It is striking that the long-chain 32-thia acids in several ways behave as long-chain fatty acids. When palmitate and dodeeylthioacetic acid are present together in the medium they are taken up by hepatocytes at similar rates (Table IIl). Their combined uptake rate is about 1.5-times that of palmitate (compare Table III and Table I). Since very little free fatty acid was found in the cells, it is evident that the uptake is limited by their rate of metabolism. When the concentration of palmitate is doubled from 0.5 to 1 mM in the medium, its uptake i~ nearly doubled [3]. Considering that the 3-thia acids are broken down only slowly, it is evident that the cells handle palmitate and dodecylthioacetic acid in a similar way except for their oxidation (Table III). It is striking that both are rapidly incorporated into lipids, and that this explains their rapid uptake. The incorporation of dodecylthioacetic acid even exceeded that of palmitate. Evidently, its coenzyme A ester accumulated in the cells to a greater extent than did that of palmitate, 'forcing' its incorporation into the lipids. In unpublished experiments we have found that the coenzyme A esters of the leng-chain 3-thia acids are good substrates for carnitine palmitoyltransferase. These CoA esters have been shown to form complexes with the acyI-CoA dehydrogenase [1]. It is surprising, therefore, that tetradecylthioaeetic acid (very similar to dodecylthioacetic acid) did not inhibit palmitate oxidation in hepatocytes. On the contrary, it stimulated the oxidation of palmitate and that of stearate even more. This cannot be explained by an increased uptake of fatty acid from the medium, since Table I shows that palmitate uptake was inhibited.
A possible explanation may be found in the effect of the accumulated 3-thiaacy~-CoA on fatty acid synthesis. Presumably, the CoA ester inhibks malonyl-CoA formation and, as the concentration of this intermediate drops, the inhibition of the outer carnitine acyltranso ferase I is relieved [11]. An increased activity of this enzyme may compensate for a direct inhibition of fatty acid oxidation. High concentrations of fat in the diet, especially fats which contain very-long-chain fatty acids, induce increased peroxisomal fatty aci~ oxidation [12,13]. In recent feeding experiments we have shown that bis(carboxymethylthio)decane, a thiadicarboxylic acid, and 3-thia fatty acids have the same effect [2]. This effect, similar to the effect of clofibrate and other hypolipaemic drugs, was reproduced in the present study, and the hepatocytes isolated from such adapted rats oxidized more and esterified less palmitate, like hepatocytes from rats fed high-fat diets or clofibrate [3,5]. The cells from adapted rats desaturated less stearate. Since no direct inlfibition of the z~9-desaturase was found, it is likely that the thia acids lead to a down-regulation of this enzyme. In this respect, tetradecylthioacetic acid apparently is like high-fat diets [14] and different from clofibrate and other peroxisome proliferators in its effect. They increase the activity of the A9-desaturase [15]. Altogether, our studies show that moderate amounts of the long-chain thia acids in the diet have similar and stronger effects on fatty acid metabolism than have fasting and high fat diets. We have also observed such a strong effect on fatty acid metabolism in hepatoma cells in tissue culture. The 3-thia fatty acids proved to be at least 20-fold more efficient in inducing peroxisomal fatty acid oxidation than were normal fatty acids [8]. Since the 3-tlfia fatty acids behave in several ways as normal fatty acids, it is evident that a sulfur atom substituting for the methylene group in the 3-position does not change the properties of the fatty acid drastically, except that it cannot be fl-oxidized. These fatty acid analogues therefore may be useful tools in studies of regulatory mechanisms in lipid metabolism. Acknowledgement This work has been supported by the Norwegian Research Council of Science and the Humanities and by The Norwegian Cancer Society. Re[~rences 1 Lau, S.-M., Brantley, R.K. and Thorpe, C. (1988) Biochemistry 27, 5089-5095. 2 Berge, R.K., Aarsland, A., Kryvi, H., Bremer, J. and Aarsaeter, N. (1989) Biochem. Pharmacoi., in press. 3 Christiansen, R.Z. (1978) Biochim. Biophys. Acta 530, 314-324. 4 Seglen, P.O. (1973) Exp. Ceil. Res. 82, 391-398.
302 5 lkrgseth, S., C. Christiansen, E.N. and Bremer, J. (1986) Lipids 21, 508-514. 6 Christiansen, R.Z. (1977) Biochim. Biophys. Acta 488, 249-262. 7 Kawagushi, A., Soskimura, T. and Okuda, S. (1980) J. Biochem. 89, 337-339. 8 Spydevold, O. and Bremer, J. (1989) Biochim. Biophys. Acta 1003, 72-79. 9 Mannaerts, G.P., Debeer, LJ., Thomas, J. and De Schepper, P.J. (1979) J. Biol. Chem. 254, 4585-4595.
10 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, RJ. (1951) J. Biol. Chem. 193, 265-275. 11 McGarry, J.D., Leatherman, G.F. and Foster, D.W. (1978). 12 Bremer, J. and Norum, K.R. (1982) .I. Lipid Res. 23, 243-256. 13 Chfistiansen, R.Z., Christiansen, E.N. and Bremer, J. (1979) Biochim. Biophys. Acta 573, 417-429. 14 Jeffcoat, P,. and James, A.T. (1977) Lipids 12, 469-474. 15 Kawashima, Y., Hanioka, N., Matsumura, M. and Kozuka, H. (1983) Biochim. Biophys. Acta 752, 259-264.