Biochimica et Biophysiea Acta. 1085(19()') 381-384
381
© 1991 ElsevierScience PublishersBA All rights reserved 0005-2760/91/$03.50 ADONIS 0005276091002f)2L BBALIP 50328
Rapid Report
Oxidation of very-long-chain fatty acids in rat brain: cerotic acid is/3-oxidized exclusively in rat brain peroxisomes W e s s e l L a g e w e g ~, J a n e E . C . S y k e s -~ M a t t h i j s L o p e s - C a r d o z o and Ronald J.A. Wanders i,
2
I Department of Pediatrics. Unit'ersity of Amsterdam, Academic Medical Centre. Amsterdam (The Netherlands ) and " Laboratory of Veterinary Bioehemia'try. State Unirersity of Utrecht. Utrecht (The Netherlands)
(Received tO Jul~ 1991)
Key words: Beta-oxidation:Glial cell: Pcroxisome:(Rat brain) We studied the effect of sodium 2-[5-(4-chlorophenyl)pentylloxirane-2-carboxylate (POCA), a potent inhibitor of mitochondrial carnitine palmitoyltransferuse !, on fatty acid oxidation by rat brain cells. In cultured glial cells as well as in dissociated brain cells from adult rats palmitic acid (16:0) oxidation was inhibited by about 85% of control values when 2 5 / z M POCA was added to the medium, whereas no inhibition of cerotic acid (26: 0) oxidation was observed. Furthermore, omission of carnitine from the culture medium resulted in a 57.7% decrease in palmitic acid oxidation ~n cultured glial cells, whereas cerotic acid oxidation was not influenced. These results indicate that rat brain pero:~.!somes contribute only little (about 15%) to palmitic acid oxidation and provide conclusive evidence that cerotic acid is oxidized exclusively in rat brain peroxisomes.
Long-chain fatty acids (LCFAs) can be fl-oxidized in mitochondria as well as in peroxisomes [1,2]. fl-Oxidation of saturated very-long-chain fatty acids (VLCFAs), however, occurs in peroxisomes rather than in mitochondria of rat i!ver as first reported by Kawamura et ai. [3]. This conclusion was based upon the finding that the fl-oxidation of lignocerie acid (24: 0) was found to occur in a fraction lighter than mitochondria and to be insensitive to inhibition by cyanide, which depressed palmitie acid (16:0) fl-oxidation drastically. Experiments by Singh et al. [4], who used highly purified sltbcellular fractions of rat liver, have shown that VLCFA oxidation occurs virtually exclusively in peroxisomes. Surprisingly, mitochondria from rat liver [4] and human skin fibroblasts [5] were found to oxidize the CoA-esters of lignoceric acid at an appreciable rate. Recently, a number of reports have appeared which suggest that oxidation of VLCFAs is mltoehondrial
Abbreviations: POCA, 2-[5-(4-chlomphenyl)pentyl]oxirane-2-carboxylate; CPT-I, carnitine palmitoyltransferaseI; VLCFA(s),very-longchain fatty acid(s): LCFA(s). long-chainfatty acid(s). Correspondence:R.J.A.Wanders, Departmentof Pediatrics(F0-224), Universityof Amsterdam. Academic Medical Centre, Meibergdreef 9, 1105AZ Amsterdam,
rather than peroxisomal [6-8]. This was concluded front experiments with homogenates of liver, heart and muscle from the rat [6,8] and with human skin fibroblasts [7] in which antimycin and rotenone were used to inhibit mitochondrial/]-oxidation. There is equal ambiguity with regard to the subcellular site of VLCFA t-oxidation in the brain [9-11]. Indeed, Singh and Singh [9] reported that lignoceric acid oxidation in rat brain homogenates was unaffected by cyanide and carnitine, which led these authors to suggest that peroxisomes are the primary site of lignoceric acid /3-oxidation, a conclusion strengthened by recent work from the same group [10]. Poulos and co-workers [11], however, suggested that the oxidation of lignoceric acid is primarily mitochondrial in rat brain. Furthermore, these authors reported that rat brain mitochondria display significant VLCFA activating activity. Resolution of this question is of great importance since it has been suggested that the primary defect in X-linked adrenoleukodystrophy, a genetic disease in man, is due to an impaired peroxisomal capacity to activate VLCFAs leading to their deficient oxidation. Most of the studies described above were done with crude homogenates and crude subcellular fractions. In various studies inhibitors of the respiratory chain were used to discriminate between mitoehondrial and perox-
382 isomal fl-oxidation. In an attempt to gain conclusive evidence for the contribution of the peroxisomes and mitochondria to the degradation of LCFAs and VLCFAs by fl-oxidation in rat brain, we decided to use a different approach based upon the use of intact cells rather than homogenates, in which the intracellular organization is disrupted. Furthermore, we used 2-[5(4-chlorophenyl)pentyl]oxirane-2-carboxylate (POCA), a powerful inhibitor of carnitine palmitoyltransferase I (CPT-I) [12], rather than antimycin a n d / o r rotenone to inhibit mitochondrial ,t3-oxidation. Glial ceils were isolated from the cerebra of 7-dayold rat pups and cultured in a chemically defined medium as described earlier [13]. After 5 days the cultures were transferred to chemically defined m e d i u m from which bovine serum albumin had been omitted. After an overnight incubation, the m e d i u m was removed and the cells were w a s h e d with physiological salt solution (0.9% NaCI; 2 × 2 ml). Then 2.5 ml of the same serum albumin-free m e d i u m with or without P O C A (Byk G u l d e n Pharmazeutika, Konstanz, F.R.G.) was added and the cells were incubated for 2 h at 37°C to enable binding of P O C A to CPT-I. Fatty acid oxidation reactions were started by adding [1-v~C]palmitic acid or [1-~4C]cerotic acid (both 4 p,M) to the medium. Fatty acids were dissolved in a solution containing 0.1 M Tris-HCl (pH 8.5), and 10 m g / m l a-cyclodextrin (Sigma, St. Louis). Incubations lasted 2 h at 37°C. The reactions were stopped by adding 0.45 ml of 10% bovine serum albumin and 0.6 ml 3 M trichloroacetic acid. C O , was trapped overnight in vials containing 0.3 ml 2 M NaOH. The m e d i u m was centrifuged and unrcacted fatty acid in the s u p e r n a t a n t was removed by extraction of 1.5 ml supernatant with 3 × 5 ml of nheptane. Radioactive d e g r a d a t i o n products in the acid-soluble layer as well as radioactive C O 2 were counted. Dissociated brain cells were p r e p a r e d by the m e t h o d of Wiesmann et al. [14] and Siegrist et al. [15], as modified by R o e d e r et al. [16]. More than 95% ef the cells were intact based on the exclusion of Trypan blue dye. ,B-Oxidation of radi,_,active substrates by dissociated brain cells was m e a s u r e d essentially as described above, except that the reaction m e d i u m was phos-
TABLE
-
-
.
.
.
0
1
2
3
4
TIME (houra)
Fig. I. Time-course of [l-14C]palmitic acid and [I-;4C]cerotie acid fl-oxidation activities in rat brain glial cell cultures, fl-Oxidation activities were measured as described in the text. 14C-labelled water-soluble degradation products and CO 2 increased linearly up to 4 h. Palmitic acid oxidation (e) and eerotie acid oxidation ( • ) .
phate-buffered saline (PBS) containing 10 m M of glucose and the reaction volume was reduced to 0.3 ml. In Fig. 1 we show the time-course of [1-t4C]palmitic acid and [l-14C]cerotic acid oxidation by glial cells in culture. 14C-labelled acid-soluble d e g r a d a t i o n products in the m e d i u m as well as CO2 increased linearly with time up to 4 h for both [1-t4C]palmitic acid and [l~4C]eerotic acid oxidation. For both substrates more label a p p e a r e d as acid-soluble metabolites t h a n as C O 2. After 4 h of incubation, C O 2 accounted for 22% of the total counts in the case of palmitie acid oxidation and 7 % in the case of cerotic acid oxidation (data not shown). E d m o n d et al. [17] showed that carnitine stimulated [1-t4C]palmitie acid oxidation approximately 3-fold in primary cultures of astroeytes. W e therefore analysed the effect of omitting carnitine from the growth m e d i u m on palmit:e acid and cerotie acid oxidation. Cells were transferred to carnitine-free m e d i u m 16 h prior to the oxidation m e a s u r e m e n t s . Omission of eatnitine resulted in a 2-fold decrease in palmitic acid/3-oxidation
I
Effi'et of carnitbte on the [3-oxidanon of [ I- t4 Clpahnitic acid and [ I- t4 C]cerotic acid in rat brain glial cell cultures
Palmitic acid [3-oxidation(nmol/h per mg)
Cerotic acid/J-uxidation (nmol/h per mg)
+ carnitine 3.26+ 1.20(n = 7)
+ carnitine 1.69_+0.41 (n = 7)
- carniline 1.38_+031 OI = 3) (42.3'~)
- carnitine 1.62_+0.08 (n = 3) (95.9%)
[3-Oxidation activities were measured as described in the text. Cells were put on carnitine-free medium 16 h prior to the oxidation measurements. Data are presented as mean_+S.D. Percentage resiL,al activity without earnitine is shown in parentheses.
383 whereas only a small effect on cerotic acid B-oxidation could be observed (Table 1). W e next examined the effect of different concentrations of P O C A on [IJ'~C]fatty acid oxidation by cultured glial ceils. At a concentration of 1 /.tM of POCA, palmitie acid oxidation was already inhibited by about 80% compared to control experiments without P O C A (Fig. 2). Maximal inhibition was found at PC'CA concentrations of 15 p.M. The specific activities for palmitic acid ,0-oxidation were 3.26 + 1.20 n m o l / h per mg protein (mean + S.D. n = 7) without P O C A and 0.53 + 9.10 n m o l / h per mg (mean +_ S.D. n = 7) with 15/.tM POCA. Cerotic acid oxidation was not inhibited by P O C A in concentrations up to 25 /xM (Fig. 2). The specific activities for cerotic acid oxidation were 1.69 + 0.41 n m o l / h per mg (mean _+ S.D. n = 7) without P O C A and 1.39 + 0.29 n m o l / h per mg (mean + S.D. n = 7) with 1 5 / z M POCA. The data p r e s e n t e d in this p a p e r are in accordance with earlier findings with h u m a n fibroblasts [18-20] and clearly indicate that in intact viable rat brain glial cells the B-oxidation of V L C F A s occurs in peroxisomes. Cerotic acid /3-oxidation was not only unaffected by carnitine, but also proved to be insensitive to P O C A , while palmitie acid oxidation was stimulated by carnitine and inhibited by POCA. The conclusion that the oxidation of V L C F A s is peroxisomal rather than mitochondrial is in accordance with recent results from Singh and co-workers [9,10]. Poulos and coworkers [1 I], however, reported that the bulk of lignoceric acid oxidation m e a s u r e d in vitro in rat brain h o m o g e n a t e s is
120
--~
1001
SO
40
,o
k_
POCA
(IJM)
Fig. 2. Effect of POCA on [I-'4C]palmitic acid and [l-14C'lcerotic acid/~-oxidation in rat brain glial cell cultures. POCA w~,sadded to the medium at final concentrations of 1.5, 15, 25 and 5(1 ,~M. Each point represents the average of duplicate incubations from one representative experiment. /~-Oxidation activities in the absence of POCA are expressed as 1110'7~..Palmitic acid oxidation (el and cerotic acid oxidation ( • ).
u.t
lOOl
__L
--
~1
eo
6o
4O
0 0
, 10
, 2o
, 3o
, 4o
, 5o
P O C A (pM)
Fig. 3. Effect of POCA on [IJ~Clpalmitic acid and [IJa(']cerotic acid ~-oxidation in dis.sociated brain cells from adult rats. P()CA was added to the medium at final concentrations of I. 5. 15.25 and 51) #M. Each point represents the average of duplicate incub;ttions from two independent experiments. Palmilic acid oxidation tel and cerotic acid oxidation ( • ).
mitochondrial. Assuming that the in vitro observations reflect the in vivo situation, these authors suggested that V L C F A oxidation in the brain occurs in mitochondria [11]. As pointed out by Lazo et al. [10], the mitocbondrial preparation of Poulos and coworkers might have been contaminated with acyI-CoA synthetase activities from other cellular m e m b r a n e s a n d / o r myelin. Poulos and co-workers used brain homogenates from adult rats [11]. W e (this paper) and others [9,10], however, have used material from developing rat brains. To rule out the possibility that the differences observed were caused by an age-related effect we repeated the P O C A inhibition studies with intact brain cells isolated from adult rats. In these cells, as in cultured glial cells. the /3-oxidation of V L C F A s proved to be peroxisomal (Fig. 3). To our knowledge this is the first report concerning V L C F A /3-oxidation in cultured glial cells. There are only a few reports about the oxidation of palmitic acid by rat brain glial cells (astrocytes) [17,21]. According to E d m o n d et al. [17] astrocytes are the only cell population in the brain that can oxidize fatty acids. In fact, these cells exhibit a preference for medium- and longchain fatty acids rather than ketone bodies or glucose as their primary metabolic fuel. Neurones and oligodendrocytcs rely primarily on ketone bodies for oxidative metabolism [17]. Since our glial cell cultures contain astrocytcs as well as oligodendrocytes the contribution of both cell types to the/3-oxidation of LC- and V L C F A s remains unclear and is currently being invcstigatcd.
384 References 1 Lazarow, P. and De Duve. C. 11976) Proc. Natl. Acad. Sci. USA 73, 2043-2046. 2 Mannaerts, G.P. and Debeer, LJ. 11982) Ann. N.Y. Acad. Sci. 386, 30-39. 3 Kawamura, N., Moser, l-l.W, and Kishimoto, Y. (1981) Bioeh~m. Biophys. Rer. Commun. 99. 1216-1225. 4 Singh, H., Derwas, N. and Poulos, A. (1987) Arch. Biochem. Biophys. 259, 382-389. 5 Singh, H., Derwas, N. and Poulos, A. (1987) Arch. Biochem. Biophys. 254, 526-533. 6 Reubsaet, F.A.G., Veerkamp, J.H., Trijbels, J.M.F. and Monnens, LA.H. (1989) Lipids 24, 945-950. 7 Reubsaet, F.A.G., Veerkamp, J.H., Brilckwilder, M.LP., Trijbels. J.M.F., Hashimoto, T. and Monnens, LA.H. (1991) Biochim. Biophys. Acta 1083, 305-309. 8 Veerkamp, J.H. and Van Moerkerk, H.T.B. 11986) Biochim. Biophys. Acta 875, 301-310. 9 Singh, R.P. and Singh, I. (1986) Neurochem. Res. I I, 281-289. l0 Lazo, O., Singh, A.K. and Singh, L (1991) J. Neurochem. 56, 1343-1353. II Singh, H., Usher, S. and Poulos, A. (1989) J. Neurochem. 53, 1711-1718.
!2 TurnbulL D.M., Bart!e!t. K.. Younan. S I.M. and Sherrat!, H.S.A (1984) Biochem. Pharmacol. 33, 475-481. 13 Koper, J.W., Lopes-Cardozo, M., Romijn, H.J. and Van Golde, L.M.G. 11984)J. Neurosci. Meth. 10, 157-169. 14 Wiesn,ann, U.N., Hofmann, K., Burkhart, T. and Herschkowitz. N. 11975) Neurobiology 5, 305-315. 15 Siegrist. I-I.P.. Bologa-Sandru, L., Burkhart, T.. Wiesmann, U., Hofmann. K. and Herschkowitz, N. 11981) J. Neurosci. Res. 6, 293-301. 16 Roeder. L.M., Tildon, J.T. and Holman. D.C. (1984) Biochem. J. 219. 131-135. 17 Edmond, .I., Robbins, R.A., Bergstrom, J.D., Cole, R.A. and Vellis, J. ('987) J. Neurosci. Res. 18, 551-561. 18 Wanders, R.J.A, van Roermund. C.W.T., Van Wijland, M.J.A., Schutgens. R.B.H.. Van den Bosch, H. and Tager, J.M. (1989) in Molecular basis of membrane-associated diseases (Azzi, A., Drahota, Z. and Papa, S., eds.), pp 407-419, Springer-Verlag, Berlin. 19 Suzuki, Y., Shimozawa, N., Yajima, S., Yamaguchi, S., Orii, T. and Hashimoto, T. 11991) Biochem. Pharmacol. 41,453-456. 20 Jakobs, B.S. and Wanders, RJ.A. 11991) Bioehem. Biophys. Res. Commun., in press. 21 Aaestad, N., Korsak, R.A., Morrow, J.W. and Edmond, J. 11991) J. Neurochem. 56, 1376-1386.
381
© 1991 ElsevierScience PublishersBA All rights reserved 0005-2760/91/$03.50 ADONIS 0005276091002f)2L BBALIP 50328
Rapid Report
Oxidation of very-long-chain fatty acids in rat brain: cerotic acid is/3-oxidized exclusively in rat brain peroxisomes W e s s e l L a g e w e g ~, J a n e E . C . S y k e s -~ M a t t h i j s L o p e s - C a r d o z o and Ronald J.A. Wanders i,
2
I Department of Pediatrics. Unit'ersity of Amsterdam, Academic Medical Centre. Amsterdam (The Netherlands ) and " Laboratory of Veterinary Bioehemia'try. State Unirersity of Utrecht. Utrecht (The Netherlands)
(Received tO Jul~ 1991)
Key words: Beta-oxidation:Glial cell: Pcroxisome:(Rat brain) We studied the effect of sodium 2-[5-(4-chlorophenyl)pentylloxirane-2-carboxylate (POCA), a potent inhibitor of mitochondrial carnitine palmitoyltransferuse !, on fatty acid oxidation by rat brain cells. In cultured glial cells as well as in dissociated brain cells from adult rats palmitic acid (16:0) oxidation was inhibited by about 85% of control values when 2 5 / z M POCA was added to the medium, whereas no inhibition of cerotic acid (26: 0) oxidation was observed. Furthermore, omission of carnitine from the culture medium resulted in a 57.7% decrease in palmitic acid oxidation ~n cultured glial cells, whereas cerotic acid oxidation was not influenced. These results indicate that rat brain pero:~.!somes contribute only little (about 15%) to palmitic acid oxidation and provide conclusive evidence that cerotic acid is oxidized exclusively in rat brain peroxisomes.
Long-chain fatty acids (LCFAs) can be fl-oxidized in mitochondria as well as in peroxisomes [1,2]. fl-Oxidation of saturated very-long-chain fatty acids (VLCFAs), however, occurs in peroxisomes rather than in mitochondria of rat i!ver as first reported by Kawamura et ai. [3]. This conclusion was based upon the finding that the fl-oxidation of lignocerie acid (24: 0) was found to occur in a fraction lighter than mitochondria and to be insensitive to inhibition by cyanide, which depressed palmitie acid (16:0) fl-oxidation drastically. Experiments by Singh et al. [4], who used highly purified sltbcellular fractions of rat liver, have shown that VLCFA oxidation occurs virtually exclusively in peroxisomes. Surprisingly, mitochondria from rat liver [4] and human skin fibroblasts [5] were found to oxidize the CoA-esters of lignoceric acid at an appreciable rate. Recently, a number of reports have appeared which suggest that oxidation of VLCFAs is mltoehondrial
Abbreviations: POCA, 2-[5-(4-chlomphenyl)pentyl]oxirane-2-carboxylate; CPT-I, carnitine palmitoyltransferaseI; VLCFA(s),very-longchain fatty acid(s): LCFA(s). long-chainfatty acid(s). Correspondence:R.J.A.Wanders, Departmentof Pediatrics(F0-224), Universityof Amsterdam. Academic Medical Centre, Meibergdreef 9, 1105AZ Amsterdam,
rather than peroxisomal [6-8]. This was concluded front experiments with homogenates of liver, heart and muscle from the rat [6,8] and with human skin fibroblasts [7] in which antimycin and rotenone were used to inhibit mitochondrial/]-oxidation. There is equal ambiguity with regard to the subcellular site of VLCFA t-oxidation in the brain [9-11]. Indeed, Singh and Singh [9] reported that lignoceric acid oxidation in rat brain homogenates was unaffected by cyanide and carnitine, which led these authors to suggest that peroxisomes are the primary site of lignoceric acid /3-oxidation, a conclusion strengthened by recent work from the same group [10]. Poulos and co-workers [11], however, suggested that the oxidation of lignoceric acid is primarily mitochondrial in rat brain. Furthermore, these authors reported that rat brain mitochondria display significant VLCFA activating activity. Resolution of this question is of great importance since it has been suggested that the primary defect in X-linked adrenoleukodystrophy, a genetic disease in man, is due to an impaired peroxisomal capacity to activate VLCFAs leading to their deficient oxidation. Most of the studies described above were done with crude homogenates and crude subcellular fractions. In various studies inhibitors of the respiratory chain were used to discriminate between mitoehondrial and perox-
382 isomal fl-oxidation. In an attempt to gain conclusive evidence for the contribution of the peroxisomes and mitochondria to the degradation of LCFAs and VLCFAs by fl-oxidation in rat brain, we decided to use a different approach based upon the use of intact cells rather than homogenates, in which the intracellular organization is disrupted. Furthermore, we used 2-[5(4-chlorophenyl)pentyl]oxirane-2-carboxylate (POCA), a powerful inhibitor of carnitine palmitoyltransferase I (CPT-I) [12], rather than antimycin a n d / o r rotenone to inhibit mitochondrial ,t3-oxidation. Glial ceils were isolated from the cerebra of 7-dayold rat pups and cultured in a chemically defined medium as described earlier [13]. After 5 days the cultures were transferred to chemically defined m e d i u m from which bovine serum albumin had been omitted. After an overnight incubation, the m e d i u m was removed and the cells were w a s h e d with physiological salt solution (0.9% NaCI; 2 × 2 ml). Then 2.5 ml of the same serum albumin-free m e d i u m with or without P O C A (Byk G u l d e n Pharmazeutika, Konstanz, F.R.G.) was added and the cells were incubated for 2 h at 37°C to enable binding of P O C A to CPT-I. Fatty acid oxidation reactions were started by adding [1-v~C]palmitic acid or [1-~4C]cerotic acid (both 4 p,M) to the medium. Fatty acids were dissolved in a solution containing 0.1 M Tris-HCl (pH 8.5), and 10 m g / m l a-cyclodextrin (Sigma, St. Louis). Incubations lasted 2 h at 37°C. The reactions were stopped by adding 0.45 ml of 10% bovine serum albumin and 0.6 ml 3 M trichloroacetic acid. C O , was trapped overnight in vials containing 0.3 ml 2 M NaOH. The m e d i u m was centrifuged and unrcacted fatty acid in the s u p e r n a t a n t was removed by extraction of 1.5 ml supernatant with 3 × 5 ml of nheptane. Radioactive d e g r a d a t i o n products in the acid-soluble layer as well as radioactive C O 2 were counted. Dissociated brain cells were p r e p a r e d by the m e t h o d of Wiesmann et al. [14] and Siegrist et al. [15], as modified by R o e d e r et al. [16]. More than 95% ef the cells were intact based on the exclusion of Trypan blue dye. ,B-Oxidation of radi,_,active substrates by dissociated brain cells was m e a s u r e d essentially as described above, except that the reaction m e d i u m was phos-
TABLE
-
-
.
.
.
0
1
2
3
4
TIME (houra)
Fig. I. Time-course of [l-14C]palmitic acid and [I-;4C]cerotie acid fl-oxidation activities in rat brain glial cell cultures, fl-Oxidation activities were measured as described in the text. 14C-labelled water-soluble degradation products and CO 2 increased linearly up to 4 h. Palmitic acid oxidation (e) and eerotie acid oxidation ( • ) .
phate-buffered saline (PBS) containing 10 m M of glucose and the reaction volume was reduced to 0.3 ml. In Fig. 1 we show the time-course of [1-t4C]palmitic acid and [l-14C]cerotic acid oxidation by glial cells in culture. 14C-labelled acid-soluble d e g r a d a t i o n products in the m e d i u m as well as CO2 increased linearly with time up to 4 h for both [1-t4C]palmitic acid and [l~4C]eerotic acid oxidation. For both substrates more label a p p e a r e d as acid-soluble metabolites t h a n as C O 2. After 4 h of incubation, C O 2 accounted for 22% of the total counts in the case of palmitie acid oxidation and 7 % in the case of cerotic acid oxidation (data not shown). E d m o n d et al. [17] showed that carnitine stimulated [1-t4C]palmitie acid oxidation approximately 3-fold in primary cultures of astroeytes. W e therefore analysed the effect of omitting carnitine from the growth m e d i u m on palmit:e acid and cerotie acid oxidation. Cells were transferred to carnitine-free m e d i u m 16 h prior to the oxidation m e a s u r e m e n t s . Omission of eatnitine resulted in a 2-fold decrease in palmitic acid/3-oxidation
I
Effi'et of carnitbte on the [3-oxidanon of [ I- t4 Clpahnitic acid and [ I- t4 C]cerotic acid in rat brain glial cell cultures
Palmitic acid [3-oxidation(nmol/h per mg)
Cerotic acid/J-uxidation (nmol/h per mg)
+ carnitine 3.26+ 1.20(n = 7)
+ carnitine 1.69_+0.41 (n = 7)
- carniline 1.38_+031 OI = 3) (42.3'~)
- carnitine 1.62_+0.08 (n = 3) (95.9%)
[3-Oxidation activities were measured as described in the text. Cells were put on carnitine-free medium 16 h prior to the oxidation measurements. Data are presented as mean_+S.D. Percentage resiL,al activity without earnitine is shown in parentheses.
383 whereas only a small effect on cerotic acid B-oxidation could be observed (Table 1). W e next examined the effect of different concentrations of P O C A on [IJ'~C]fatty acid oxidation by cultured glial ceils. At a concentration of 1 /.tM of POCA, palmitie acid oxidation was already inhibited by about 80% compared to control experiments without P O C A (Fig. 2). Maximal inhibition was found at PC'CA concentrations of 15 p.M. The specific activities for palmitic acid ,0-oxidation were 3.26 + 1.20 n m o l / h per mg protein (mean + S.D. n = 7) without P O C A and 0.53 + 9.10 n m o l / h per mg (mean +_ S.D. n = 7) with 15/.tM POCA. Cerotic acid oxidation was not inhibited by P O C A in concentrations up to 25 /xM (Fig. 2). The specific activities for cerotic acid oxidation were 1.69 + 0.41 n m o l / h per mg (mean _+ S.D. n = 7) without P O C A and 1.39 + 0.29 n m o l / h per mg (mean + S.D. n = 7) with 1 5 / z M POCA. The data p r e s e n t e d in this p a p e r are in accordance with earlier findings with h u m a n fibroblasts [18-20] and clearly indicate that in intact viable rat brain glial cells the B-oxidation of V L C F A s occurs in peroxisomes. Cerotic acid /3-oxidation was not only unaffected by carnitine, but also proved to be insensitive to P O C A , while palmitie acid oxidation was stimulated by carnitine and inhibited by POCA. The conclusion that the oxidation of V L C F A s is peroxisomal rather than mitochondrial is in accordance with recent results from Singh and co-workers [9,10]. Poulos and coworkers [1 I], however, reported that the bulk of lignoceric acid oxidation m e a s u r e d in vitro in rat brain h o m o g e n a t e s is
120
--~
1001
SO
40
,o
k_
POCA
(IJM)
Fig. 2. Effect of POCA on [I-'4C]palmitic acid and [l-14C'lcerotic acid/~-oxidation in rat brain glial cell cultures. POCA w~,sadded to the medium at final concentrations of 1.5, 15, 25 and 5(1 ,~M. Each point represents the average of duplicate incubations from one representative experiment. /~-Oxidation activities in the absence of POCA are expressed as 1110'7~..Palmitic acid oxidation (el and cerotic acid oxidation ( • ).
u.t
lOOl
__L
--
~1
eo
6o
4O
0 0
, 10
, 2o
, 3o
, 4o
, 5o
P O C A (pM)
Fig. 3. Effect of POCA on [IJ~Clpalmitic acid and [IJa(']cerotic acid ~-oxidation in dis.sociated brain cells from adult rats. P()CA was added to the medium at final concentrations of I. 5. 15.25 and 51) #M. Each point represents the average of duplicate incub;ttions from two independent experiments. Palmilic acid oxidation tel and cerotic acid oxidation ( • ).
mitochondrial. Assuming that the in vitro observations reflect the in vivo situation, these authors suggested that V L C F A oxidation in the brain occurs in mitochondria [11]. As pointed out by Lazo et al. [10], the mitocbondrial preparation of Poulos and coworkers might have been contaminated with acyI-CoA synthetase activities from other cellular m e m b r a n e s a n d / o r myelin. Poulos and co-workers used brain homogenates from adult rats [11]. W e (this paper) and others [9,10], however, have used material from developing rat brains. To rule out the possibility that the differences observed were caused by an age-related effect we repeated the P O C A inhibition studies with intact brain cells isolated from adult rats. In these cells, as in cultured glial cells. the /3-oxidation of V L C F A s proved to be peroxisomal (Fig. 3). To our knowledge this is the first report concerning V L C F A /3-oxidation in cultured glial cells. There are only a few reports about the oxidation of palmitic acid by rat brain glial cells (astrocytes) [17,21]. According to E d m o n d et al. [17] astrocytes are the only cell population in the brain that can oxidize fatty acids. In fact, these cells exhibit a preference for medium- and longchain fatty acids rather than ketone bodies or glucose as their primary metabolic fuel. Neurones and oligodendrocytcs rely primarily on ketone bodies for oxidative metabolism [17]. Since our glial cell cultures contain astrocytcs as well as oligodendrocytes the contribution of both cell types to the/3-oxidation of LC- and V L C F A s remains unclear and is currently being invcstigatcd.
384 References 1 Lazarow, P. and De Duve. C. 11976) Proc. Natl. Acad. Sci. USA 73, 2043-2046. 2 Mannaerts, G.P. and Debeer, LJ. 11982) Ann. N.Y. Acad. Sci. 386, 30-39. 3 Kawamura, N., Moser, l-l.W, and Kishimoto, Y. (1981) Bioeh~m. Biophys. Rer. Commun. 99. 1216-1225. 4 Singh, H., Derwas, N. and Poulos, A. (1987) Arch. Biochem. Biophys. 259, 382-389. 5 Singh, H., Derwas, N. and Poulos, A. (1987) Arch. Biochem. Biophys. 254, 526-533. 6 Reubsaet, F.A.G., Veerkamp, J.H., Trijbels, J.M.F. and Monnens, LA.H. (1989) Lipids 24, 945-950. 7 Reubsaet, F.A.G., Veerkamp, J.H., Brilckwilder, M.LP., Trijbels. J.M.F., Hashimoto, T. and Monnens, LA.H. (1991) Biochim. Biophys. Acta 1083, 305-309. 8 Veerkamp, J.H. and Van Moerkerk, H.T.B. 11986) Biochim. Biophys. Acta 875, 301-310. 9 Singh, R.P. and Singh, I. (1986) Neurochem. Res. I I, 281-289. l0 Lazo, O., Singh, A.K. and Singh, L (1991) J. Neurochem. 56, 1343-1353. II Singh, H., Usher, S. and Poulos, A. (1989) J. Neurochem. 53, 1711-1718.
!2 TurnbulL D.M., Bart!e!t. K.. Younan. S I.M. and Sherrat!, H.S.A (1984) Biochem. Pharmacol. 33, 475-481. 13 Koper, J.W., Lopes-Cardozo, M., Romijn, H.J. and Van Golde, L.M.G. 11984)J. Neurosci. Meth. 10, 157-169. 14 Wiesn,ann, U.N., Hofmann, K., Burkhart, T. and Herschkowitz. N. 11975) Neurobiology 5, 305-315. 15 Siegrist. I-I.P.. Bologa-Sandru, L., Burkhart, T.. Wiesmann, U., Hofmann. K. and Herschkowitz, N. 11981) J. Neurosci. Res. 6, 293-301. 16 Roeder. L.M., Tildon, J.T. and Holman. D.C. (1984) Biochem. J. 219. 131-135. 17 Edmond, .I., Robbins, R.A., Bergstrom, J.D., Cole, R.A. and Vellis, J. ('987) J. Neurosci. Res. 18, 551-561. 18 Wanders, R.J.A, van Roermund. C.W.T., Van Wijland, M.J.A., Schutgens. R.B.H.. Van den Bosch, H. and Tager, J.M. (1989) in Molecular basis of membrane-associated diseases (Azzi, A., Drahota, Z. and Papa, S., eds.), pp 407-419, Springer-Verlag, Berlin. 19 Suzuki, Y., Shimozawa, N., Yajima, S., Yamaguchi, S., Orii, T. and Hashimoto, T. 11991) Biochem. Pharmacol. 41,453-456. 20 Jakobs, B.S. and Wanders, RJ.A. 11991) Bioehem. Biophys. Res. Commun., in press. 21 Aaestad, N., Korsak, R.A., Morrow, J.W. and Edmond, J. 11991) J. Neurochem. 56, 1376-1386.
