THE JOURNAL OF CHEMISTRY Vol. 267, No 28, ,5, PP .20065 ... · THE JOURNAL OF BIOLOGICAL CHEMISTRY Q 1992 by The American Society for Biochemistry and Molecular Biology, Inc. Vol.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Q 1992 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 267, No . 28, Issue of October ,5 , PP .20065-20074,1992 Printed in U. S. A .

Substrate Specificities of Rat Liver Peroxisomal Acyl-CoA Oxidases: Palmitoyl-CoA Oxidase (Inducible Acyl-CoA Oxidase), Pristanoyl-CoA Oxidase (Non-inducible Acyl-CoA Oxidase), and Trihydroxycoprostanoyl-CoA Oxidase*

(Received for publication, April 6, 1992)

Paul P. Van VeldhovenS, Geertrui VanhoveQ, Stanny AssselberghsT, Hendrik J. EyssenT, and Guy P. Mannaerts From the Afdeling Farmacologie, Katholieke Uniuersiteit Leuuen, Campus Gasthuisberg, B-3000 Leuuen and the 7Rega Instituut, Minderbroedersstraat, 8-3000 Leuuen, Belgium

Rat liver peroxisomes contain three acyl-CoA oxidases: palmitoyl-CoA oxidase, pristanoyl-CoA oxidase, and trihydroxycoprostanoyl-CoA oxidase. The three oxidases were separated by anion-exchange chromatography of a partially purified oxidase preparation, and the column eluate was analyzed for oxidase activity with different acyl-CoAs.

Short chain mono (hexanoyl-) and dicarboxylyl (glutaryl-)-CoAs and prostaglandin E2-CoA were oxidized exclusively by palmitoyl-CoA oxidase. Long chain mono (palmitoyl-) and dicarboxylyl (hexadecanedioyl-) -CoAs were oxidized by palmitoyl-CoA oxidase and pristanoyl-CoA oxidase, the former enzyme catalyzing -70% of the total eluate activity. The very long chain lignoceroyl-CoA was also oxidized by palmitoyl-CoA oxidase and pristanoyl-CoA oxidase, the latter enzyme catalyzing -65% of the total eluate activity. Long chain 2-methyl branched acyl-CoAs (2-methylpalmitoyl-CoA and pristanoyl-CoA) were oxidized for -90% by pristanoyl-CoA oxidase, the remaining activity being catalyzed by trihydroxycoprostanoyl-CoA oxidase. The short chain 2-methylhexanoyl-CoA was oxidized by trihydroxycoprostanoyl-CoA oxidase and pristanoyl- CoA oxidase (-60 and 40%, respectively, of the total eluate activity). Trihydroxycoprostanoyl-CoA was oxidized exclusively by trihydroxycoprostanoyl-CoA oxidase. No oxidase activity was found with isovaleryl- CoA and isobutyryl-CoA.

Substrate dependences of palmitoyl-CoA oxidase and pristanoyl-CoA oxidase were very similar when assayed with the same (common) substrate. Since the two oxidases were purified to a similar extent and with a similar yield, the contribution of each enzyme to substrate oxidation in the column eluate probably re- flects its contribution in the intact liver.

The first enzyme of peroxisomal @-oxidation is an acyl-CoA

*This work was supported by grants from the Geconcerteerde Onderzoeksacties van de Vlaamse Gemeenschap, from the Belgian Fonds voor Geneeskundig Wetenschappelijk Onderzoek, and from the Onderzoeksfonds van de Katholieke Universiteit Leuven. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ To whom correspondence should be addressed: Dept. of Phar- macology, Campus Gasthuisberg (0 & N), B-3000 Leuven, Belgium.

§ Aspirant of the Belgian Nationaal Fonds voor Wetenschappelijk Onderzoek.

oxidase (1). In the rat, extrahepatic peroxisomes contain two and liver peroxisomes three acyl-CoA oxidases (2, 3). A first enzyme, palmitoyl-CoA oxidase, oxidizes the CoA esters of straight chain fatty acids. The enzyme is induced in liver and to a lesser extent in some extrahepatic tissues (kidney, intes- tinal mucosa, heart) by treatment of the animals with peroxisome proliferators (2, 3). Palmitoyl-CoA oxidase has been purified by Osumi et al. (4) and by Inestrosa et al. ( 5 ) . Its native molecular mass is 150 kDa, and it consists of subunits of 72, 52, and 21 kDa. The latter two subunits are formed in vivo by posttranslational proteolytic cleavage of the 72-kDa subunit (6,7). The amino acid sequence of the 72-kDa subunit has been reported by Miyazawa et al. (8).

A second enzyme, pristanoyl-CoA oxidase, oxidizes the CoA esters of 2-methyl branched fatty acids such as the synthetic 2-methylpalmitic acid and the naturally occurring pristanic acid, but it also shows activity toward the CoA esters of straight long chain fatty acids (3). The enzyme is not induced by treatment of the animals with peroxisome proliferators (3, 9). Pristanoyl-CoA oxidase has been purified in our laboratory. Its native molecular mass is 420 kDa, and it consists of identical subunits of 70 kDa (3).

Liver peroxisomes contain a third acyl-CoA oxidase, trihydroxycoprostanoyl-CoA oxidase, which oxidizes the CoA esters of the bile acid intermediates di- and trihydroxycoprostanic acids (2). The enzyme, which is not induced by peroxisome proliferators, has been partially purified in our laboratory. Its native molecular mass is 139 kDa, and it consists of identical subunits of 69 kDa (2).

Palmitoyl-CoA oxidase, pristanoyl-CoA oxidase, and trihydroxycoprostanoyl-CoA oxidase are also present in the human, but the enzymes have been less well characterized than the rat enzymes (10, ll).’

Isolated rat liver peroxisomes are capable of &oxidizing a wide variety of CoA esters: those of medium, long, and very long chain fatty acids, 2-methyl branched fatty acids, medium and long chain dicarboxylic fatty acids, and the carboxyl side chains of the bile acid intermediates di- and trihydroxycoprostanic acids, prostaglandins, and other eicosanoids, and xenobiotics (for reviews, see Refs. 12 and 13). For most of these substrates it is not known by which oxidase(s) they are oxidized. Therefore, we determined the substrate specificity of each of the three acyl-CoA oxidases, using commercially available CoA esters as well as CoA esters that were synthesized in our laboratory.

’ G. Vanhove, P. P. Van Veldhoven, and G. P. Mannaerts, unpublished results.

20065

20066 Peroxisomal Acyl-CoA Oxidases

EXPERIMENTAL PROCEDURES

Materials-Carbonyldiimidazole was obtained from Fluka, Buchs, Switzerland. Glutaryl-CoA, isovaleryl-CoA, isobutyryl-CoA, lignoceroyl-CoA, and bovine serum albumin (Cohn fraction V) were pur- chased from Sigma. The albumin was defatted according to Chen (14). Prostaglandin E, (97%) and 4,4'-dithiodipyridine were from Janssen Chimica, Beerse, Belgium; peroxidase, FAD, and P-cyclodextrin from Boehringer Mannheim, Germany; hexanoyl-CoA, decanoyl-CoA, lauryl-CoA, palmitoyl-CoA, CoA, and Sephadex LH-20 from Pharmacia Belga, Brussels, Belgium; 2-methylhexanoic acid (99%), hexadecanedioic acid (98%), and 6-phenylhexanoic acid (99%) were from Aldrich Europe, Brussels, Belgium. Supelclean solid-phase extraction tubes (SPE, LC-18, 1 g) were from Supelco, Bellefonte, PA.

Synthesis of CoA Esters-Pristanoyl-CoA (3), 2-methylpalmitoyl- CoA (3), trihydroxycoprostanoyl-CoA (15), and the CoA ester of prostaglandin E, (16) were synthesized as described before. The other CoA derivatives (2-methylhexanoyl-CoA, hexadecanedioyl-CoA, and 6-phenylhexanoyl-CoA) were made by reacting their 1-acylimidazole derivative with CoA (17). For the preparation of the 1-acylimidazoles a 1.35-fold excess (on a molar base) of carbonyldiimidazole over fatty acid was used except for hexadecanedioic acid where the carbonyldiimidazole was reduced to 0.5 mol/mol of fatty acid.

After evaporation of the organic solvents, 6-phenylhexanoyl- and hexadecanedioyl-CoA were precipitated with HC10, at pH 1. The precipitates were washed once with 0.7% (w/v) HCIOI, dissolved in 50 mM Tris-HC1 buffer, pH 7.0, and applied to preactivated SPE-C,, columns. After washing the columns with water, the CoA esters were eluted by means of an ethanol gradient in water. 2-Methylhexanoyl- CoA, which did not precipitate at acidic pH, was purified as follows. After evaporation of the organic solvents, the aqueous phase was brought to pH 4 with sodium acetate and applied to a SPE-C, column. After washing the column with the water, the CoA ester was eluted by means of an ethanol gradient in water.

The CoA esters were dissolved in water, and their purity was determined by thin layer chromatography on Silica Gel 60-G in n- butyl alcohol/acetic acid/water (50:20:30, v/v). Only one spot was visible with R, values of 0.55,0.57, and 0.49 for hexadecanedioyl-CoA, 6-phenylhexanoyl-CoA, and 2-methylhexanoyl-CoA, respectively. The aqueous solutions were standardized by measuring their absorbance at 260 nm in 10 mM potassium-phosphate buffer, pH 7.0 (c = 15,400). No free CoASH could be detected in the solutions by following their reaction with 4,4'-dithiodipyridine a t 324 nm.

Partial Purification of Hepatic Acyl-CoA Oxidases and Chromato- graphic Separation-Acyl-CoA oxidases were partially purified from livers of control and clofibrate-treated rats by a combination of heat treatment in the presence of FAD and ammonium sulfate fractionation as described before (3).

For anion-exchange chromatography, the partially purified oxidase preparation was dialyzed overnight against 20 mM Tris-HC1 buffer, p H 8.6, containing 20% (w/v) glycerol. After removal of any partic- ulate material by centrifugation, an aliquot (1 ml) was injected onto a Protein Pak Glass DEAE-5PW column (80 X 7.5 mm, 10-pm particle size, O.l-pm pore size, Nihon Waters Ltd., Tokyo, Japan) equilibrated with dialysis buffer and eluted at 1 ml/min. Bound proteins were eluted by means of a linear pH/salt gradient (0-100% buffer containing 20 mM Tris-HC1, pH 7.8. 0.25 M NaCl, 20% (w/v) glycerol) over 50 min, followed by a wash with dialysis buffer containing 0.5 M NaC1. Fractions of 2 ml were collected in tubes containing 20 p1 of 1 mM FAD and analyzed for oxidase activity. Oxidase measurements were performed by following the substrate-dependent peroxide production by means of the peroxidase-catalyzed dimeriza- tion of homovanillic acid, essentially as described before (3), except that Triton X-100 was omitted and that the concentrations of the CoA esters and albumin were varied as indicated in the legends to the figures.

RESULTS

Substrate Specificities of the Peroxisomal Acyl-CoA Oxi- dases-Fig. 1 shows the separation of a partially purified preparation of rat liver acyl-CoA oxidases on a DEAE anion- exchange column. In agreement with an earlier report from our laboratory, the inducible acyl-CoA oxidase, which is often called palmitoyl-CoA oxidase, eluted first from the column, followed (in a second peak) by trihydroxycoprostanoyl-CoA

5- L q L ~ .,,1;/ m

0, 4 0 2 u

.L "u"l"u P _uw______

1 10 20 30 1 10 20 30 1 10 20 30 FRACTION NUMBER

FIG. 1. Substrate specificity of hepatic peroxisomal acyl- CoA oxidases. A partially purified peroxisomal oxidase preparation obtained by heat treatment and ammonium sulfate fractionation was separated on a DEAE anion-exchange column. Recoveries (uersus the original whole liver homogenate) of enzyme activities in the partially purified preparation were 30.4% (palmitoyl-CoA oxidase activity), 35.6% (2-methylpalmitoyl-CoA oxidase activity), and 37.8% (trihydroxycoprostanoyl-CoA oxidase activity). Enrichments (uersus the homogenate) were 23.6-fold (palmitoyl-CoA oxidase activity), 24.5- fold (2-methylpalmitoyl-CoA oxidase activity), and 27.6-fold (trihydroxycoprostanoyl-CoA oxidase activity). The column eluate fractions were analyzed for oxidase activity with the following substrates (substrate concentrations and recoveries after chromatography are indicated in parentheses): A, glutaryl-CoA (250 p ~ , 81%); B, hexanoyl-CoA (250 p ~ , 56%); C, prostaglandin E&oA (50 PM, 104%); D, 6-phenylhexanoyl-CoA (50 p ~ , 65%); E, hexadecanedioyl- CoA (50 pM, 93%); F, palmitoyl-CoA (100 pM, 88%); G, lignoceroyl- CoA (75 pM, 103%); H, 2-methylpalmitoyl-CoA (75 pM, 101%); I, pristanoyl-CoA (75 p ~ , 103%); J , 2-methylhexanoyl-CoA (175 p ~ , 53%); K, trihydroxycoprostanoyl-CoA (75 p ~ , 53%). Panel L shows the absorbance and the concentration of the NaCl gradient. In choos- ing the substrate concentrations for the determination of the oxidase activities, a compromise was made between enzymatic activity (for substrate dependences of the oxidase activities, see Figs. 2-19) on the one hand and cost and availability of the CoA esters on the other hand. Oxidase activities were measured in the presence of 9 p~ albumin except for trihydroxycoprostanoyl-CoA which was measured in the presence of 36 p~ albumin. When lignoceroyl-CoA was used, the assays contained 525 p~ p-cyclodextrin in addition to the albumin.

oxidase and (in a third peak) by the non-inducible acyl-CoA oxidase, which we call pristanoyl-CoA oxidase (3). The eluate fractions were analyzed for oxidase activity with different CoA esters as the substrates. The figure shows that the CoA esters of short chain mono- and dicarboxylic fatty acids such as hexanoyl-CoA and glutaryl-CoA were oxidized exclusively by palmitoyl-CoA oxidase as was the CoA ester of prostaglandin E*. Elongation of hexanoyl-CoA by means of a 6-phenyl

Peroxisomal Acyl-CoA Oxidases 20067

substitution resulted in a substrate that was oxidized still predominantly by palmitoyl-CoA oxidase but that was also recognized, albeit poorly, by pristanoyl-CoA oxidase. The CoA esters of long chain mono- and dicarboxylic fatty acids such as palmitoyl-CoA and hexadecanedioyl-CoA were oxidized by palmitoyl-CoA oxidase but, in addition, a substantial part (-30%) of the total eluate oxidase activity was associated with the fractions containing pristanoyl-CoA oxidase.' As the long chain acyl-CoAs, the very long chain lignoceroyl-CoA was oxidized both by palmitoyl-CoA oxidase and pristanoyl- CoA oxidase. However, the latter enzyme was now the more active one, catalyzing -65% of the total eluate activity. The CoA esters of the long chain 2-methyl branched fatty acids, 2-methylpalmitoyl-CoA and pristanoyl-CoA, were oxidized almost completely by pristanoyl-CoA oxidase. A small portion (10% or less) of the overall eluate activity was catalyzed by trihydroxycoprostanoyl-CoA oxidase? This is perhaps not surprising since the bile acid intermediates di- and trihydroxycoprostanic acids possess a branched 2,6-dimethylhexanoic acid side chain. 2-Methylhexanoyl-CoA was oxidized for even more than 50% by trihydroxycoprostanoyl-CoA oxidase, the remainder of the total eluate activity being associated with the pristanoyl-CoA oxidase peak. Trihydroxycoprostanoyl- CoA was oxidized exclusively by trihydroxycoprostanoyl-CoA oxidase. No oxidase activity was found with isovaleryl-CoA and isobutyryl-CoA, catabolites of leucine and valine, respectively, in either the partially purified enzyme preparation that was loaded on the column or the column eluate.

Identical substrate specificities were observed with purified palmitoyl-CoA oxidase and pristanoyl-CoA oxidase. (Trihy- droxycoprostanoyl-CoA oxidase has not been purified yet.) When a partially purified oxidase preparation from a clofibrate-treated rat was separated on a DEAE column, the same substrate specificities were again detected. As expected, the activities eluting with the first peak (inducible acyl-CoA oxidase) were severalfold increased (data not shown).

Substrate Dependence of Peroxisomal Acyl-CoA Oxidases: Influence of Albumin-The influence of albumin and the substrate dependence in the absence and presence of albumin were studied for each of the oxidases with the acyl-CoA esters used in Fig. 1. The experiments which are presented in detail further in this section, revealed two major points that are worth mentioning at the beginning.

1) In the absence and presence of albumin, the substrate dependences of palmitoyl-CoA oxidase and pristanoyl-CoA oxidase were almost identical for each of their common substrates (palmitoyl-CoA, hexadecanedioyl-CoA, lignoceroyl- CoA): with a particular substrate each enzyme reached its half-maximal activity at a similar substrate concentration; the same was true for the substrate concentrations at which maximal activities were reached. Since in the partially purified enzyme preparation that was loaded on the DEAE col-

Oxidase activities were also measured with octanoyl-CoA (200 pM), decanoyl-CoA (175 pM), and lauryl-CoA (150 p M ) in fractions 12 and 21, containing the highest palmitoyl-CoA oxidase activity and pristanoyl-CoA oxidase activity, respectively. The ratios of activities found in fraction 12 (palmitoyl-CoA oxidase) over those found in fraction 21 (pristanoyl-CoA oxidase) were 3.7, 2.1, and 3.9 for octanoyl-CoA, decanoyl-CoA, and lauryl-CoA, respectively, as com- pared to 2.8 for palmitoyl-CoA. These results indicate that also the CoA esters of medium chain fatty acids are oxidized partly by pristanoyl-CoA oxidase.

In an earlier study we did not find activity of trihydroxycoprostanoyl-CoA oxidase toward 2-methylpalmitoyl-CoA or pristanoyl- CoA (3). This is related to the assay conditions. In these earlier experiments 2-methylpalmitoyl-CoA oxidase and pristanoyl-CoA oxidase were measured in the absence of albumin, a condition in which trihydroxycoprostanoyl-CoA oxidase is hardly active (see further).

umn, recoveries of palmitoyl-CoA oxidase and pristanoyl-CoA oxidase were comparable as were their enrichments, the similar substrate dependences allow for an extrapolation of the data of Fig. 1, which were obtained at saturating or near- saturating substrate concentrations, to the in vivo situation where much lower substrate concentrations most probably prevail. This extrapolation suggests that, at least in liver, pristanoyl-CoA oxidase oxidizes a substantial part of the long chain acyl-CoAs and especially of the very long chain acyl- CoAs.

2) Trihydroxycoprostanoyl-CoA oxidase was hardly active in the absence of albumin, which, when used in the assay, bound most of the trihydroxycoprostanoyl-CoA. At equal unbound trihydroxycoprostanoyl-CoA concentrations, trihydroxycoprostanoyl-CoA oxidase was many times more active in the presence than in the absence of albumin, suggesting that the enzyme may prefer bound trihydroxycoprostanoyl- CoA over unbound trihydroxycoprostanoyl-CoA. This may not be a unique feature of trihydroxycoprostanoyl-CoA oxidase, since also retinoic acid appears to be a better substrate for its catabolizing enzymes, when bound to cellular retinoic acid-binding protein (19).

Figs. 2-19 show the substrate dependences in the absence and presence of albumin (upper panels) and the influence of increasing albumin concentrations at a fixed substrate concentration (lower panels) for palmitoyl-CoA oxidase (Figs. 2- 8), pristanoyl-CoA oxidase (Figs. 9-15), and trihydroxycoprostanoyl-CoA oxidase (Figs. 16-19). The experiments were carried out on pooled DEAE column eluate fractions containing the highest activity of the corresponding enzyme (Fig. 1, fractions 11-13 for palmitoyl-CoA oxidase, fractions 21-23 for pristanoyl-CoA oxidase, fractions 14-16 for trihydroxycoprostanoyl-CoA oxidase).

The figures reveal that for each enzyme and for each substrate enzymatic activities depend not only on the substrate concentration but also on the albumin concentration and on the substrate/albumin molar ratios and that conditions in which optimum activities are reached differ from enzyme to enzyme and from substrate to substrate. In vir- tually every instance, albumin was stimulatory, moderately or strongly, when used at the appropriate concentration.

3 r o o k / L - > + H pM SUBSTRATE

8 2001 \

. 1 . 3 . 5 X (w/v) ALBUMIN

FIG. 2. Palmitoyl-CoA dependence of palmitoyl-CoA oxidase: influence of albumin. Palmitoyl-CoA oxidase was measured in fractions 11-13 from the DEAE column (Fig. 1) with palmitoyl- CoA as the substrate. Upper panel, palmitoyl-CoA dependence in the absence (A) and presence (a) of 9 pM (0.06%, w/v) albumin. Lower panel, albumin dependence in the presence of 100 p~ palmitoyl-CoA.


150L 50 . I . 3 .5

X (w/v) ALBUMIN

FIG. 3. Hexadecanedioyl-CoA dependence of palmitoyl- CoA oxidase: influence of albumin. Palmitoyl-CoA oxidase was measured in fractions 11-13 from the DEAE column (Fig. 1) with hexadecanedioyl-CoA as the substrate. Upperpanel, hexadecanedioyl- CoA deDendence in the absence (A) and uresence (0) of 9 p~ (0.06%, . . . w/v) albumin. Lower panel, albumin dependence'in 50 pM hexadecanedioyl-CoA.

50t / //

50 150 250 uM SUBSTRATE lool

75 t

the presence of

X (w/v) ALBUMIN

FIG. 4. Lignoceroyl-CoA dependence of palmitoyl-CoA oxidase: influence of albumin. Palmitoyl-CoA oxidase was measured in fractions 11-13 from the DEAE column (Fig. 1) with lignoceroyl- CoA as the substrate. Upper panel, lignoceroyl-CoA dependence in the absence (A) and presence (0) of 9 p~ (0.06%, w/v) albumin. Lower panel, albumin dependence in the presence of 75 p~ lignoceroyl-CoA. All assays (upper and lower panels) also contained 525 p~ 6-cyclodextrin. Assay mixtures without albumin became cloudy at lignoceroyl-CoA concentrations above 100 p ~ .

Theoretically, the stimulatory effect of albumin could be due to several factors or to a combination of these factors: stabi- lization or protection from inactivation of the enzyme, release of substrate inhibition by binding excess substrate, enhance- ment of product dissociation from the enzyme by product binding, and preference by the enzyme of bound over unbound substrate.

Palmitoyl-CoA oxidase displayed marked substrate inhibi-

6o t 40b

.I .3 . 5 X (w/v) ALBUMIN

FIG. 5. Prostaglandin EZ-CoA dependence of palmitoyl-CoA oxidase: influence of albumin. Palmitoyl-CoA oxidase was measured in fractions 11-13 from the DEAE column (Fig. 1) with prostaglandin E&oA as the substrate. Upper panel, prostaglandin E2-CoA dependence in the absence (A) and presence (0) of 9 NM (0.06%, w/ v) albumin. Lower panel, albumin dependence in the presence of 50 pM prostaglandin E2-CoA.

c ." W 4J 0 L a cn E c

\ E

0 E c

-4 - - > I-

w I- u U W ffl U 0 n X 0 U 0 u -I > u

CI

H

a

40 BO 120 160 U M SUBSTRATE

.I . 3 .5 X (w/v) ALBUMIN

FIG. 6. 6-Phenylhexanoyl-CoA dependence of palmitoyl- CoA oxidase: influence of albumin. Palmitoyl-CoA oxidase was measured in fractions 11-13 from the DEAE column (Fig. 1) with 6- phenylhexanoyl-CoA as the substrate. Upper panel, 6-phenylhexanoyl-CoA dependence in the absence (A) and presence (0) of 9 p~ (0.06%, w/v) albumin. Lower panel, albumin dependence in the presence of 75 p~ 6-phenylhexanoyl-CoA.

tion at higher concentrations of palmitoyl-CoA (Fig. 2) and lignoceroyl-CoA (Fig. 4). The stimulatory effect of albumin at these elevated substrate concentrations may be partly the result of binding of the excess CoA esters. This cannot be the only explanation, however. Since albumin contains six to seven binding sites for palmitoyl-CoA (18), its binding capac-


"0

&""-A"""""-. I I

> I-

100 200 H JJM SUBSTRATE

# 60

> t- H, > c H .r4 t - a l v u

L a 0 w a

n~ am ffl

H . X t 0 -ri

E

0" v o I E 1 c > - U

a \

a

FIG. I. mM SUBSTRATE

, Hexanoyl-CoA dependence of palmitoyl-CoA oxidase: influence of albumin. Palmitoyl-CoA oxidase was measured in fractions 11-13 from the DEAE column (Fig. 1) with hexanoyl- CoA as the substrate in the absence (A) and presence (0) of 9 p~ (0.06%, w/v) albumin. The stimulatory effect of albumin did not vary over a concentration range of 0.01-0.5% (w/v); hexanoyl-CoA concentration, 100 pM (data not shown).

Y I I I 1 2 3

mM SUBSTRATE FIG. 8. Glutaryl-CoA dependence of palmitoyl-CoA oxi-

dase: influence of albumin. Palmitoyl-CoA oxidase was measured in fractions 11-13 from the DEAE column (Fig. 1) with glutaryl-CoA as the substrate in the absence (A) and presence (0) of 9 p~ (0.06%, w/v) albumin. The stimulatory effect of albumin did not vary over a concentration range of 0.01-0.5% (w/v); glutaryl-CoA concentration, 100 pM (data not shown).

ity, when used at a concentration of 9 pM (Fig. 2, upperpanel), does not seem to suffice to explain the stimulatory effect observed at palmitoyl-CoA concentrations higher than 100 p ~ . At low concentrations of long and very long chain acyl- CoAs albumin became inhibitory, most likely by binding the CoA esters, which would result in a lowering of the unbound substrate concentration. With the more polar substrates, which did not induce substrate inhibition, the stimulatory effect of albumin was less dramatic and also present at low substrate concentrations (Figs. 5-8). The reason for this stimulation remains unclear.

With pristanoyl-CoA oxidase, long (straight and branched) and very long chain acyl-CoA esters caused much less substrate inhibition at elevated concentrations (Figs. 9-13) and the stimulatory effect of albumin was also less substantial than that seen with palmitoyl-CoA oxidase. At lower concentrations of these CoA esters albumin became again inhibitory. As was the case for palmitoyl-CoA oxidase, it cannot be deduced from our experiments why albumin stimulated pristanoyl-CoA oxidase with the more polar substrates, which did not cause substrate inhibition, and why this stimulation was

% (w/v) ALBUMIN

FIG. 9. Pristanoyl-CoA dependence of pristanoyl-CoA oxidase: influence of albumin. Pristanoyl-CoA oxidase was measured in fractions 21-23 from the DEAE column (Fig. 1) with pristanoyl- CoA as the substrate. Upperpanel, pristanoyl-CoA dependence in the absence (A) and presence (0) of 9 p~ (0.06%, w/v) albumin. Lower panel, albumin dependence in the presence of 75 p~ pristanoyl-CoA.

300 c

> I-

20 40 60 BO H > p M SUBSTRATE

300L

u 4

.I . 3 .5 % (w/v) ALBUMIN

FIG. 10. 2-Methylpalmitoyl-CoA dependence of pristanoyl- CoA oxidase: influence of albumin. Pristanoyl-CoA oxidase was measured in fractions 21-23 from the DEAE column (Fig. 1) with 2- methylpalmitoyl-CoA as the substrate. Upper panel, 2-methylpalmitoyl-CoA dependence in the absence (A) and presence (0) of 9 p~ (0.06%, w/v) albumin. Lowerpanel, albumin dependence in the presence of 75 p~ 2-methylpalmitoyl-CoA.

particularly marked at low substrate concentrations (see Fig. 14).

For palmitoyl-CoA oxidase as well as pristanoyl-CoA oxidase, apparent K,,, values increased with decreasing chain length and/or increasing polarity of the substrates. For each of the substrates that are shared by palmitoyl-CoA oxidase and pristanoyl-CoA oxidase (palmitoyl-CoA, hexadecane-


.A c W 4 0 L a 0 E

c E .A

> 0

c - > I-

> I- U 4

W cn 4

H

H

z X 0 U U 0

J > U a

105 "g /L

140 - yM SUBSTRATE

35 -

.I % (w/v) ALBUMIN

.3 .5

FIG. 11. Palmitoyl-CoA dependence of pristanoyl-CoA oxidase: influence of albumin. Pristanoyl-CoA oxidase was measured in fractions 21-23 from the DEAE column (Fig. 1) with palmitoyl- CoA as the substrate. Upper panel, palmitoyl-CoA dependence in the absence (A) and presence (0) of 9 p~ (0.06%, w/v) albumin. Lower panel, albumin dependence in the presence of 100 p~ palmitoyl-CoA.

- c W 4 0 L

"

a 0) E

c

\ E

0 E c

."

- > L > I- U 4

W rn 0 H X 0 4 0 u I -1 > U

H

a

a

~"""""4

20

25 75 125 yM SUBSTRATE

Bot

;"\. 2 .I % (w/v) ALBUMIN . 3 .5

FIG. 12. Hexadecanedioyl-CoA dependence of pristanoyl- CoA oxidase: influence of albumin. Pristanoyl-CoA oxidase was measured in fractions 21-23 from the DEAE column (Fig. 1) with hexadecanedioyl-CoA as the substrate. Upperpanel, hexadecanedioyl- CoA dependence in the absence (A) and presence (0) of 9 pM (0.06%, w/v) albumin. Lower panel, albumin dependence in the presence of 50 p~ hexadecanedioyl-CoA.

dioyl-CoA, lignoceroyl-CoA), substrate dependences in the absence and presence of albumin were comparable for palmitoyl-CoA oxidase and pristanoyl-CoA oxidase (compare Figs. 2 and 11, Figs. 3 and 12, Figs. 4 and 13) indicating that the affinity of the enzymes for a particular common substrate is similar.

A surprising finding was that trihydroxycoprostanoyl-CoA

FIG. 13. Lignoceroyl-CoA dependence of pristanoyl-CoA oxidase: influence of albumin. Pristanoyl-CoA oxidase was measured in fractions 21-23 from the DEAE column (Fig. 1) with lignoceroyl-CoA as the substrate. Upper panel, lignoceroyl-CoA dependence in the absence (A) and presence (0) of 9 pM (0.06%, w/v) albumin. Lower panel, albumin dependence in the presence of 75 FM lignoceroyl-CoA. All assays (upper and lower panels) also contained 525 PM (3-cyclodextrin. Assay mixtures without albumin became cloudy at lignoceroyl-CoA concentrations above 100 p ~ .

I ?

10

0.5 1 .o 1.5 mM SUBSTRATE

301 2ob '1 -


FIG. 14. 2-Methylhexanoyl-CoA dependence of pristanoyl- CoA oxidase: influence of albumin. Pristanoyl-CoA oxidase was measured in fractions 21-23 from the DEAE column (Fig. 1) with 2- methylhexanoyl-CoA as the substrate. Upper panel, 2-methylhexanoyl-CoA dependence in the absence (A) and presence (0) of 9 p~ (0.06%, w/v) albumin. Lower panel, albumin dependence in the presence of 225 pM 2-methylhexanoyl-CoA.

oxidase is hardly active in the absence of albumin (Figs. 16- 19). Although substrate inhibition was observed with some of the substrates, the magnitude of the stimulatory effect makes it doubtful that a mere release of substrate inhibition was


> I-

40 BO 1; CI w p M SUBSTRATF H I

I I 1 I I I

% (w/v) ALBUMIN .I . 3 .5

FIG. 15. 6-Phenylhexanoyl-CoA dependence of pristanoyl- CoA oxidase: influence of albumin. Pristanoyl-CoA oxidase was measured in fractions 21-23 from the DEAE column (Fig. 1) with 6- phenylhexanoyl-CoA as the substrate. Upper panel, 6-phenylhexanoyl-CoA dependence in the absence (A) and presence (0) of 9 pM (0.06%, w/v) albumin. Lower panel, albumin dependence in the presence of 75 p~ 6-phenylhexanoyl-CoA.

> 50 100 150 200 H p M SUBSTRATE I- > ,

I I 1 I 1


FIG. 16. Trihydroxycoprostanoyl-CoA dependence of trihydroxycoprostanoyl-CoA oxidase: influence of albumin. Trihydroxycoprostanoyl-CoA oxidase was measured in fractions 14- 16 from the DEAE column (Fig. 1) with trihydroxycoprostanoyl-CoA as the substrate. Upper panel, trihydroxycoprostanoyl-CoA dependence in the absence (A) and presence of 9 p~ (0.06%, w/v) albumin (0) or 36 pM (0.24%, w/v) albumin (0). Lower panel, albumin dependence in the presence of 75 p~ trihydroxycoprostanoyl-CoA.

(the only) cause of the stimulation. The phenomenon was investigated in more detail with trihydroxycoprostanoyl-CoA, the natural substrate for trihydroxycoprostanoyl-CoA oxidase. The enzyme was not irreversibly inactivated during the assay in the absence of albumin, since addition of albumin to

* O r

I- H mM SUBSTRATE

4 !:Il , , , , V

> u 5 10

a


FIG. 17. 2-Methylhexanoyl-CoA dependence of trihydroxycoprostanoyl-CoA oxidase: influence of albumin. Trihydrox- ycoprostanoyl-CoA oxidase was measured in fractions 14-16 from the DEAE column (Fig. 1) with 2-methylhexanoyl-CoA as the substrate. Upper panel, 2-methylhexanoyl-CoA dependence in the absence (A) and presence (0) of 9 p~ (0.06%, w/v) albumin. Lowerpanel, albumin dependence in the presence of 225 p~ 2-methylhexanoyl-CoA.

FIG. 18. Pristanoyl-CoA dependence of trihydroxycoprostanoyl-CoA oxidase: influence of albumin. Trihydroxycopros- tanoyl-CoA oxidase was measured in fractions 14-16 from the DEAE column (Fig. 1) with pristanoyl-CoA as the substrate. Upper panel, pristanoyl-CoA dependence in the absence (A) and presence (0) of 9 p~ (0.06%, w/v) albumin. Lower panel: albumin dependence in the presence of 75 p~ pristanoyl-CoA.

an assay mixture containing enzyme and substrate and prein- cubated in the absence of albumin, restored full activity (data not shown). In another experiment, we investigated whether albumin binds trihydroxycoprostanoyl-CoA. Fig. 20 shows that at trihydroxycoprostanoyl-CoA concentrations below 70 PM more than 90% of the trihydroxycoprostanoyl-CoA was


X (w/v) ALBUMIN

FIG. 19. 2-Methylpalmitoyl-CoA dependence of trihydroxycoprostanoyl-CoA oxidase: influence of albumin. Trihydrox- ycoprostanoyl-CoA oxidase was measured in fractions 14-16 from the DEAE column (Fig. 1) with 2-methylpalmitoyl-CoA as the substrate. Upper panel, 2-methylpalmitoyl-CoA dependence in the absence (A) and presence (0) of 9 p M (0.06%, w/v) albumin. Lowerpanel, albumin dependence in the presence of 7 5 p M 2-methylpalmitoyl-CoA.

TOTAL CONCENTRATION (pM)

FIG. 20. Trihydroxycoprostanoyl-CoA binding to albumin. [26-14C]Trihydroxycoprostanoyl-CoA (specific radioactivity, 5.7 pCi/ pmol), synthesized as described before (15), was added to 40 mM potassium phosphate buffer, pH 8.3, containing 0.24% (w/v) albumin (conditions used in the trihydroxycoprostanoyl-CoA oxidase assay) at the concentrations indicated in the figure. After 30 min of equili- bration, an aliquot of the mixtures was counted for radioactivity (total concentration), and unbound and bound trihydroxycoprostanoyl-CoA were separated by ultrafiltration (Centrifree microparti- tion system, Amicon, Danvers, MA) of the remainder of the mixtures. An aliquot of the ultrafiltrates was counted for radioactivity (unbound concentration).

bound at the albumin concentration used in the assay. Com- parison of Figs. 16 and 20 reveals that at equal unbound trihydroxycoprostanoyl-CoA concentrations trihydroxycoprostanoyl-CoA oxidase was many times more active in the presence than in the absence of albumin. This suggests that trihydroxycoprostanoyl-CoA oxidase prefers bound trihy-

droxycoprostanoyl-CoA over unbound trihydroxycoprostanoyl-CoA as the substrate. Whether the same holds for 2- methylhexanoyl-CoA (Fig. 17) remains to be investigated, since we could not accurately determine the albumin binding due to the unavailability of the radioactive CoA ester. Also unexplained remains the stimulatory effect of albumin on the oxidation of pristanoyl-CoA (Fig. 18) and 2-methylpalmitoyl- CoA (Fig. 19). The fact that elevated amounts of albumin depressed enzymatic activity at a fixed concentration of these substrates (Figs. 18 and 19, lower panels) does not agree with the hypothesis that the enzyme would also accept these substrates in their albumin-bound form.

Although the effects of albumin can be explained only partly, our experiments illustrate that, for each substrate, assay conditions must be carefully chosen and that small alterations in the assay conditions can sometimes result in dramatic changes in enzymatic activity.

DISCUSSION

As mentioned in the Introduction and as confirmed in this report, peroxisomes are capable of @-oxidizing a wide variety of acyl-CoA derivatives. The substrate spectrum of peroxisomes partly overlaps that of the mitochondria, which are also capable of oxidizing a broad range of CoA esters. Al- though mitochondria and peroxisomes share a number of substrates, the contribution of each organelle to the oxidation of these common substrates is not equally important in the intact cell. Available evidence indicates that in the intact cell or in the intact organism mitochondria oxidize the major portion of the long chain acyl-CoAs (20-23),4 whereas peroxisomes are responsible for the oxidation of the major portion of the CoA esters of very long chain fatty acids (24-27), dicarboxylic fatty acids (28-30), 2-methyl branched fatty acids (9), prostaglandins (16, 31), and possibly other eicosanoids (32). The CoA esters of the bile acid intermediates di- and trihydroxycoprostanic acids are oxidized solely by peroxisomes; they are not substrates for mitochondria (33-35).

Our results show that among the physiological substrates for peroxisomes (i.e. those substrates that are oxidized exclusively or predominantly by peroxisomes in the intact cell) the CoA esters of prostaglandins and those of the bile acid intermediates are oxidized exclusively by palmitoyl-CoA oxidase and trihydroxycoprostanoyl-CoA oxidase, respectively. The CoA ester of the isoprenoid-derived 2-methyl branched pristanic acid is oxidized almost completely by pristanoyl-CoA oxidase. Trihydroxycoprostanoyl-CoA oxidase also shows some activity with this substrate but it is low. Because of this low activity and because trihydroxycoprostanoyl-CoA oxidase is absent from extrahepatic tissues (3), it is doubtful that this enzyme plays a significant role in the degradation of long chain isoprenoid-derived fatty acids in uiuo. Short chain 2- methyl branched fatty acids, which may originate from the breakdown of long chain isoprenoid-derived fatty acids, are perhaps better substrates for trihydroxycoprostanoyl-CoA oxidase than for pristanoyl-CoA oxidase. Since the apparent K,,, values of the peroxisomal oxidases appear to be inversely related to the chain length of the substrates, it is again doubtful that short chain branched acyl-CoAs (as well as short chain straight acyl-CoAs and short chain dicarboxylyl- CoAs such as glutaryl-CoA) are degraded to any large extent by peroxisomes in uiuo. This contention agrees with the general belief that peroxisomal @-oxidation does not go to completion but that it rather acts as a chain-shortening

It should be noted that this statement is true only for animal cells. In plants and eukaryotic microorganisms, peroxisomes (or glyoxysomes) are the only site of @-oxidation.


system (see Refs. 12 and 13). As long chain acyl-CoAs, long chain dicarboxylyl-CoAs appear to be oxidized partly by palmitoyl-CoA oxidase and partly by pristanoyl-CoA oxidase, the former enzyme being the more active one. For the oxidation of very long chain acyl-CoAs, pristanoyl-CoA oxidase seems to be more important, at least in liver, than palmitoyl-CoA oxidase.

Our observations raise a number of questions with regard to the molecular defects in certain peroxisomal diseases. Is it possible, if the same substrate specificities occur in the human, that some cases of X-linked adrenoleukodystrophy are the result of a pristanoyl-CoA oxidase deficiency? X-Linked adrenoleukodystrophy, a disease that affects mainly the cen- tral nervous system and the adrenals, is characterized by an accumulation of very long chain fatty acids in tissues and body fluids and is believed to be caused by a deficiency of a peroxisomal very long chain acyl-CoA synthetase (36). Bio- chemically, X-linked adrenoleukodystrophy is diagnosed by measuring very long chain fatty acid levels. In view of our results, it might be of importance to also measure pristanic acid levels.

On the other hand, the few cases of isolated palmitoyl-CoA oxidase deficiency that have been described also present with elevated very long chain fatty acid levels (37, 38). This suggests that pristanoyl-CoA oxidase alone does not suffice to handle the organism’s very long chain fatty acid load or that the latter enzyme is less active toward very long chain fatty acids in the human. Clearly, the determination of the activities of palmitoyl-CoA oxidase and pristanoyl-CoA oxidase in different human tissues and the determination of the enzymes’ substrate specificities are needed before these questions can be answered.

In our experiments glutaryl-CoA appeared to be oxidized by palmitoyl-CoA oxidase and not by a separate enzyme (39). I n agreement with this idea, the glutaryl-CoA oxidase activity copurified with palmitoyl-CoA oxidase during the further purification of the enzyme to homogeneity,’ confirming earlier findings by Poosch and Yamazaki (40). Our findings are difficult to reconcile with the recent description of a glutaryl- CoA oxidase deficiency in a patient with glutaric aciduria, who had a seemingly normal palmitoyl-CoA oxidase activity (41). However, we cannot exclude the possibility of the exist- ence of a separate glutaryl-CoA oxidase that was lost or inactivated in the early purification steps of our enzyme preparation.

A surprising finding of the present study is the strong albumin dependence of the partially purified trihydroxycoprostanoyl-CoA oxidase. One possible explanation that needs further investigation is that the enzyme prefers bound substrate. Is it possible that in the intact cell sterol carrier protein-2, which is highly concentrated in the peroxisomal matrix (42, 43), plays a role in binding the CoA esters of the bile acid intermediates? This question may be related to the unexplained findings in a patient whom we recently described (44). This patient was suspected to suffer from a trihydroxycoprostanoyl-CoA synthetase or a trihydroxycoprostanoyl- CoA oxidase deficiency on clinical-chemical grounds (elevated plasma levels of di- and trihydroxycoprostanic acids, normal plasma levels of very long chain fatty acids). Trihydroxyco- prostanoyl-CoA synthetase and trihydroxycoprostanoyl-CoA oxidase were normal as was the oxidation of [26-14C]trihydroxycoprostanoyl-CoA in liver tissue from the patient (44). These findings suggest that some other hitherto unknown factor, possibly a binding or transport protein, was deficient.

We could not detect oxidase activity with isovaleryl-CoA

’ P. P. Van Veldhoven and G. P. Mannaerts, unpublished results.

and isobutyryl-CoA, catabolites of leucine and valine, respectively. This suggests that, in contrast with certain peroxisomes from plants (45, 46), mammalian peroxisomes do not contain oxidases involved in the catabolism of branched amino acids.

Finally, some general conclusions can be drawn with regard to the chemical characteristics of the acyl-CoA oxidase substrates. Palmitoyl-CoA oxidase appears to oxidize only the CoA esters of straight chain fatty acids and of other molecules that possess an unbranched carboxyl side chain (e.g. 6-phenylhexanoic acid, prostaglandin EJ. The apparent K , of the enzyme increases with decreasing chain length and/or increasing polarity (e.g. prostaglandin E*) of the substrate. Modification of the wend and w-substitutions do not seem to affect the enzyme’s activity in a dramatic way. Pristanoyl- CoA oxidase oxidizes the CoA esters of straight as well as 2- methyl branched fatty acids. Shorter and/or more polar molecules are poor (branched molecules) or no (straight molecules) substrates for the enzyme. Trihydroxycoprostanoyl- CoA oxidase recognizes only 2-methyl branched substrates. Besides oxidizing the CoA esters of the bile acid intermediates di- and trihydroxycoprostanic acids, the enzyme is capable of oxidizing short as well as long chain 2-methyl branched fatty acids. Pristanoyl-CoA oxidase recognizes both optical isomers of its 2-methyl branched substrates, since excess enzyme completely desaturates limiting amounts of the CoA esters of synthetically prepared 2-methylpalmitic and pristanic acids.‘ In this regard, it resembles trihydroxycoprostanoyl-CoA oxidase, which also does not discriminate between the R and S isomers of trihydroxycoprostanoyl-CoA (33).

Acknowledgments-We are grateful to Luc Govaert, Patricia Van Rompuy, and Chan ta l Brees for expert technical assistance and t o Mouche Bareau for dedicated secretarial help.

REFERENCES 1. Lazarow, P. B., and de Duve, C. (1976) Proc. Natl. Acad. Sci. U. S. A. 73,

2043-2046 2. Schepers, L., Van Veldhoven, P. P., Casteels, M., Eyssen, H. J., and

Mannaerts, G. P. (1990) J. Biol. Chem. 265,5242-5246 3. Van Veldhoven, P. P., Vanhove, G., Vanhoutte, F., Dacremont, G., Parm-

entier, G., Eyssen, H. J., and Mannaerts, G. P. (1991) J. Biol. Chem.

4. Osumi, T., Hashimoto, T. and Ui, N. (1980) J . Biochem. (Tokyo) 87,1735- 266 , 24676-24683

5. Inestrosa, C. N., Bronfman, M., and Leighton, F. (1980) Biochem. Biophys. 1746

6. Furata, S., Hashimoto, T., Miura, S., Mori, M., and Tatihana, M. (1982) Res. Commun. 9 5 , 7-12

Biochem. Biophys. Res. Commun. 105,639-646 7. Miura, S., Mori, M., Takiguchi, M., Tatihana, M., Furata, S., Miyazawa,

8. Miyazawa, S., Hayashi, H., Hijikata, M., Ishii, N., Furata, S., Kagamiyama, S., and Hashimoto, T. (1984) J. Biol. Chem. 259,6397-6402

H., Osumi, T., and Hashimoto, T. (1987) J . Biol. Chem. 262,8131-8137 9. Vanhove, G., Van Veldhoven, P. P., Vanhoutte, F., Parmentier, G., Eyssen,

10. Casteels, M., Schepers, L., Van Veldhoven, P. P., Eyssen, H. J., and H. J., and Mannaerts, G. P. (1991) J. Biol. Chem. 266,24670-24675

11. Wanders, R. J. A,, ten Brink, H. J., van Roermund, C. W. T., Schutgens, Mannaerts, G. P. (1990) J. Lipid Res. 3 1 , 1865-1872

R. B. H., Tager, J. M. and Jakobs, C. (1990) Biochem. Biophys. Res. Commun. 172,490-495

12. Mannaerts, G. P., and Van Veldhoven, P. P. (1990) Biochem. SOC. Transact. 18,87-89

13. Mannaerts, G. P., and Van Veldhoven, P. P. (1991) in Inborn Errors of Metabolism (Schauh, J., Van Hoof, F., and Vis, H. L., eds), Nestle Nutrition Workshop Series, Vol. 24, pp. 1-14, Raven Press, New York

14. Chen, R. F. (1967) J. Biol. Chem. 242 , 173-181 15. Casteels, M., Schepers, L., Van Eldere, J., Eyssen, H. J., and Mannaerts,

16. Schepers, L., Casteels, M., Vamecq, J., Parmentier, G., Van Veldhoven, P.

17. Kawaguchi, A,, Yoshimura, T., and Okuda, S. (1981) J. Biochem. (Tokyo)

18. Richards. E. W.. Hamm. M. W.. Fletcher. J. E.. and Otto. D. A. (1990)

G. P. (1988) J. Biol. Chem. 263,4654-4661

P., and Mannaerts, G. P. (1988) J. Biol. Chem. 263,2724-2731

89,337-339

Biochim. Biophys. Acta 1044,361-367 19. Fiorella. P. D.. and NaDoli. J. L. (1991) J. Biol. Chem. 266. 16572-16579 20. Mannaerts, G.’ P., DebLer, L. J., Thomas, J., and De Schepper, P. J. (1979)

21. Thomas, J., Debeer, L. J., De Schepper, P. J., and Mannaerts, G. P. (1980)

22. Foerster, E.-C., Fihrenkemper, T., Rahe, U., Graf, P., and Sies, H. (1981)

23. Skorin, C., Necochea, C., Johow, V., Soto, U., Grau, A. M., Bremer, J., and

J. B i d . Chem. 254,4585-4595

Biochem. J. 190, 485-494

Biochem. J . 196, 705-712

Leighton, F. (1992) Biochem. J . 281 , 561-567

20074 Peroxisomal Acyl-CoA Oxidases 24. Singh, I., Moser, H. W., Moser, A. B., and Kishimoto, Y. (1981) Biochem. 35. Kase, B. F., Prydz, K., Bjorkhem, I., and Pedersen, J. I. (1986) Biochim.

Biophys. Res. Commun. 102, 1223-1229 25. Sing, I., Moser, A. B., Goldfischer, S., and Moser, H. W. (1984) Proc. Natl. 36. Maser &'w and Maser, A. B. (lgS9) in The Metabolic Basis InherLted

Bioph s Acta 877,37-42

Acad. Sci. U. S. A. 81,4203-4207 Dislase (dr iver C. R. Beaudet A. L. SI W S , and Valfe, D., eds) Vol. I1 15111-1532 McGraw-kill, N'ew qork' '

27. Jakobs, B. S., and Wanders, R. J. A. (1991) Biochem. Biophys. Res. Com- and Singh, Y. (1990) J. Lipid Res. 3 1, 583-595

A,, Schram, A,, Tager, J. M., and Saudubray, J. M. (1988) Am. J. Hum. B. H., Wanders, R. J. A,, van Roermund, C. W. T., van Wijlani, d. J:

28. Cerdan, S., Kunnecke, B., Dolle, A,, and Seelig, J. (1988) J. Biol. Chem. 38. Wanders R. J. A. van Roermund, C. W. T., Schut ens R. B H , Barth, mun. 178,842-847 Genet. 42,422-434

263, 11664-11674 P. G., keymans: H. S. A., van den Bosch, H., and%agkr, J. M . (1990) J. 29. Suzuki, H., Yamada, J., Watanabe, T., and Suga, T. (1989) Biochim. Inherited Metab. Dis. 13,4-36

Biophys. Acta 490,25-30 39. Vamecq, J., and Van Hoof, F. (1984) Biochem. J. 221, 203-211 30. Leighton, F., Bergseth, J., Rortveit, T., Christiansen, E. N., and Bremer, J. 40. Poosch, M. S., and Yamazaki, R. K. (1989) Biochim. Biophys. Acta 1006,

(1989) J. Biol. Chem. 264, 10347-10350 41. Bennett M. J., Pallitt, R. J., Goodman, S. I., Hale, D. E., and Vamecq, J. 291-298

31. Diczfalusy, U., Kase, B. F., Alexson, S. E. H., and Bjorkhem, I. (1991) J. Clin. Inuest. 88,978-984

(1991) J. Inherited Metab. Dis. 14, 165-173 42. Tsuneoka, M., Yamamato, A,, Fujiki, Y., and Tashiro, Y. (1988) J. Biochem.

32. Jedlitschky, G., Huber, M., Volkl, A,, Muller, M., Leier, I., Muller, J., Lehmann, w.-D,, ~ ~ h i ~ i , H, D., and ~ ~ ~ ~ l ~ ~ , D, (1991) J . ~ i ~ l . chern, 43. Keller, G. A., Scallen, T. J., Clarke, D., Maher, p. A., Krisans, s. K., and

(Tokyo) 104,560-564

266,24763-24772 Singer, S. J. (1989) J. Cell B~ol. 108, 1353-1361 33. Kase, F., Bjorkhem, I., and Pedersen, J. I. (1983) J. Lipid Res. 24, 1560- 44' Wanders R' J' A' Casteel'' M'' Mannaerts' G' "' van Roermund' c' w'

1567 T Schtgens k. B. H., Kozich, V., Zeman, J., and Hyanek, J. (1991)

34. Prydz, K., Kase, B. F., Bjorkhem, I., and Pedersen, J. I. (1986) J. Lipid 45. Gerbling H., and Gerhardt B. (1988) Plant Physiol. 88, 13-15 d n . Chirn. Adta 202, 123-132

Res. 27,622-628 46. Gerbling: H., and Gerhardt: B. (1989) Plant Physiol. 91, 1387-1392

26. Laze, 0.) Contreras, M., Yoshida, Y., Singh, A. K., Stanley, w., Weise, M.. 37. Poll-Th; V.T., Reels, $., ogier, H., scotto, J., Vamecq, J., Schut ens R

THE JOURNAL OF CHEMISTRY Vol. 267, No 28, ,5, PP .20065 ... · THE JOURNAL OF BIOLOGICAL CHEMISTRY Q 1992 by The American Society for Biochemistry and Molecular Biology, Inc. Vol.

Documents