Postnatal Development and Isolation of Peroxisomes from Brain

Journal of Neurochemistry Raven Press, Ltd., New York 0 1991 International Society for Neurochemistry

Postnatal Development and Isolation of Peroxisomes from Brain

Oscar Lazo, *Avtar K. Singh, and Inderjit Singh

Departments of Pediatrics and *Pathology, Medical University of South Carolina and Veterans Administration Hospital, Charleston, South Carolina, U S A .

Abstract: We analyzed the postnatal peroxisome development in rat brain by measuring the enzyme activities of catalase and acyl-CoA oxidase and @-oxidation of [ l-'4C]lignoceric acid. These enzyme activities were higher between 10 and 16 days of postnatal life and then decreased. We developed and compared two different methods for isolation of enriched peroxisomes from 10-day-old rat brain by using a combi- nation of differential and density gradient centrifugation techniques. Peroxisomes in Percoll (self-generating gradient) banded at a density of 1.036 a 0.0 12 g/ml and in Nycodenz continuous gradient at I. 125 k 0.0 14 dml. Acyl-CoA oxidase, Damino acid oxidase, L-pipecolic acid oxidase, and dihydroxyacetone phosphate acyltransferase activities and activ-

ities for the oxidation of very long chain fatty acid (lignoceric acid) were almost exclusively associated with catalase activity (a marker enzyme for peroxisomes) in the gradient. The postnatal increase in peroxisomal activity with the onset of myelination and the presence of enzyme for the biosynthesis of plasmalogens and oxidation of very long chain fatty acid (both predominant constituents of myelin) suggest that brain peroxisomes may play an important role in the assembly and turnover of myelin. Key Words: Brain-Peroxisomes- Postnatal development-Fatty acid P-oxidation-Hydrogen peroxide-producing oxidases. Lazo 0. et al. Postnatal development and isolation of peroxisomes from brain. J. Neuro- chem. 56, 1343-1353 (1991).

Recently, a number of demyelinating diseases associated with peroxisomal dysfunction have been identified (Schutgens et al., 1986~; Moser, 1986; Singh et al., 1988) and the abnormalities in peroxisomal function may be due to a specific enzyme deficiency [e.g., X-linked adrenoleukodystrophy (X-ALD)] (Hashmi et al., 1986; Lazo et al., 1988, 1989; Wanders et al., 1988) or to defects in the biogenesis of peroxisomes (e.g., Zellweger syndrome) (Goldfischer et al., 1973; Lazarow and Moser, 1989). Peroxisomes appear to have a ubiquitous distribution in mammalian cells and are composed of a granular matrix delimited by a single membrane (de Duve and Baudhuin, 1966). Al- though peroxisomes were initially believed to play a minor role in mammalian metabolism, it is now ob- vious that they catalyze essential reactions associated with the biogenesis of ether phospholipids, glyoxylate, and bile acids and the catabolism of very long chain (VLC) fatty acids (>C22), prostaglandins, pipecoIic acid, and phytanic acid (Tolbert, 198 1 ; Lazarow and Moser, 1989). Most of this knowledge about peroxisomal me-

tabolism has been obtained from extraneural tissue. Peroxisomes in brain are smaller in size (0.1-0.2 pm diameter) as compared to liver (0.3-0.9 pm diameter) (Holtzman, 1982). With cytochemical methods for the detection of catalase and D-aminO oxidase, two peroxisomal enzymes in liver, peroxisomes have been identified in neurons and oligodendrocytes and are particularly abundant in oligodendrocytes prior to myelination (Arnold and Holtzman, 1978). Although the role of peroxisomes in the biogenesis of ether phospholipids and the catabolism of VLC fatty acids (>C22) is well established, the lack of a suitable procedure for the isolation of peroxisomes from brain tissue has al- lowed only preliminary studies (Gaunt and de Duve, 1976; Hajra and Bishop, 1982; Singh and Singh, 1986). Therefore, because of the limited studies in brain, very little is known about the physiological role of peroxisomes in this tissue. In an effort to develop a procedure for the isolation of peroxisomes from brain, we first studied the postnatal development of peroxisornes by monitoring peroxisomal enzyme activities and then

Received April 26, 1990; final revised manuscript received October 16, 1990; accepted October 16, 1990.

Address correspondence and reprint requests to Dr. I. Sin& at Departments of Pediatrics and Cell Biology, Medical University of South Carolina, 171 Ashley Avenue, Charleston, SC 29425, U.S.A.

Abbreviations used: DAB, diaminobenzidiazine; DHAP, dihydroxyacetone phosphate; VLC, very long chain; X-ALD, X-linked adrenoleukodystrophy.

1343

1344 0. LAZO ET AL.

compared the properties of peroxisomes isolated by two different procedures from brains of 1 Oday-old rats, an age at which rat brain has the highest number of peroxisomes.

MATERIALS AND METHODS

Materials [ l-14C]Palmitic acid (58.7 mCi/mmol) and KI4CN (52.0

mCi/mmol) were purchased from New England Nuclear (Boston, MA, U.S.A.). Malate, FAD, L-carnitine, NAD, NADPH, pyruvic acid, lactate dehydrogenase, glycerol-3- phosphate dehydrogenase, palmitoyl-CoA, pnitrophenol-2- acetamide-2-deoxy-~-~-glucopyranoside, cytochrome c, lau- royl-CoA, and a-cyclodextrin were purchased from Sigma Chemical (St. Louis, MO, U.S.A.). ATP and CoASH were obtained from P-L Biochemicals (Milwaukee, WI, U.S.A.). ~-[U-'~C]Glycer01-3-phosphate (27 mCi/mmol) was obtained from ICN Radiochemicals (Irvine, CA, U.S.A.). Nycodenz was obtained from Accurate Chemical and Scientific (West- bury, NY, U.S.A.). Percoll was from Pharmacia Fine Chem- icals (Uppsala, Sweden). [ l-'4C]Lignoceric acid was synthe- sized by treatment of n-tricosanoyl bromide with K14CN as described previously by Hoshi and Kishimoto (1973). [U- ''C]Dihydroxyacetone phosphate ( [U-I4C]DHAP) was prepared as described previously (Schutgens et al. (1986b).

Postnatal developmental procedures Brains from Sprague-Dawley rats (obtained from Charles

River Laboratories, Wilmington, MA, U.S.A.) 3-36 days of age were homogenized in five volumes (wt/vol) of 0.25 M sucrose, 1 mM EDTA, 3 mM imidazole buffer, pH 7.4, with two strokes at approximately 1,000 rpm and then diluted 10 times (wt/vol) with buffer. The homogenate was centrifuged at 105,000 g for 45 rnin to separate the particulate fraction from the supernatant. The enzyme activities for catalase (Baudhuin et al., 1964), acyl-CoA oxidase (Small et al., 1985), and the enzyme system for the catabolism of [ l-14C]lignoceric acid (Singh et al., 1984) were examined in the particulate fraction as described previously.

Isolation of peroxisomal enriched fraction Method A. Rat brains at the time of highest peroxisomal

activity (10 days old) were homogenized in three volumes (wt/vol) of 0.25 M sucrose, 1 mM EDTA, 3 mM imidazole buffer, pH 7.4 (homogenization buffer), with two upand- down strokes at 1,000 rpm and the homogenate was diluted with homogenization buffer to a final volume of 10 times (wt/vol). The homogenate was centrifuged at 6,000 rpm for 10 rnin in JA20 Beckman rotor (4,354 g for 10 min) and the pellet (v fraction) (Leighton et al., 1968) was washed two times with homogenization buffer. The supernatant was laid on a solution containing 16 ml of 0.85 M sucrose, 1 mM EDTA, 3 mM imidazole, pH 7.4, and centrifuged at 20,000 rpm for 1 h in Beckman Ti65 rotor (34,809 g for 1 h). The supernatants (0.25 M and 0.85 M sucrose phases) (S, supernatant) were removed from the pellet and this pellet was resuspended in homogenization buffer and centrifuged at 18,000 rpm for 20 rnin in a JA20 Beckman rotor (39,191 g for 20 min). This new pellet was washed two times with homogenization buffer and the supernatant (S,) was combined with SI and was called I) fraction (Leighton et al., 1968). The pellet, lambda fraction (X), was resuspended in homogenization buffer, and 5 ml were fractionated in a 0-30% (wt/ vol) continuous Nycodenz gradient as described previously

by Lazo et al. (1989). The tubes (39 ml) were sealed and centrifuged at 18,000 rpm for I h in a JV-20 Beckman rotor (39,191 g for 1 h) with slow acceleration and deceleration. The fractions were collected from the bottom and analyzed for marker enzyme activities. The density of the fractions was measured by refractometry (refractometer Atago 500).

Method B. Brains from 10-day-old rats were homogenized in three volumes (wt/vol) of 0.85 M sucrose, in 1 mM EDTA, 3 mM imidazole buffer, pH 7.4, with two strokes at 1,000 rpm and the homogenate was diluted five times with 0.85 M sucrose, 1 mM EDTA, 3 mM imidazole buffer, pH 7.4. This homogenate was overlaid with a solution containing 0.25 M sucrose, 1 mM EDTA, and 3 mM imidazole, pH 7.4, and centrifuged at 27,000 rpm for 45 rnin (107,960 g for 45 min) in a SW28 Beckman rotor. The majority of the myelin was collected at the interface. The upper layer (0.25 M sucrose) and myelin at the interface were removed (My, myelin fraction). The lower layer (0.85 A4 sucrose) and the pellet were combined, diluted to a final concentration of0.25 Msucrose, 1 mM EDTA, 3 mM imidazole, pH 7.4, and homogenized with two up-and-down strokes. The nuclear (N), heavy mitochondria (M), light mitochondria (L), and microsomal (P) fractions were prepared by differential centrifugation at 2,000 rpm for 8 min (483 g for 8 rnin), 10,000 rpm for 11.2 min (12,096 g for 11.2 rnin), and 18,000 rpm for 24 rnin (39,191 g for 24 min) in a Beckman JA-20 rotor and 43,000 rpm for 60 rnin (190,479 g for 60 min) in a Beckman Ti-70 rotor, respectively (de Duve et al., 1955). Each fraction was washed two times by resuspending in 0.25 M sucrose solution. The peroxisomes from the L fraction, which sediment between 10,000 rpm and 18,000 rpm (12,096 g and 39,191 g), were purified further by a self-generating Percoll gradient in a Beckman JV-20 vertical rotor. The tube was first layered with 4 ml of 2.5 A4 sucrose followed by 28 ml of 13% Percoll in 0.25 Msucrose, 1 mUEDTA, 3 mMimidazole, pH 7.4, and 5 ml of L fraction. The tubes were sealed and centrifuged at 20,000 rprn for 30 rnin in a Beckman JV-20 rotor (48,384 g for 30 min) with slow acceleration and deceleration. The fractions were collected from the bottom and analyzed for marker enzyme activities. The density of the fractions from the Percoll gradient was determined by the procedure of Rome et al. (1979).

Marker enzyme assays Enzyme activities for catalase as a marker for peroxisomes

(Baudhuin et al., 1964), NADPH-cytochrome c reductase for microsomes (Beaufay et a]., 1974), cytochrome c oxidase for mitochondria (Cooperstein and Lazarow, 195 l), N-acetyl-P- glucosaminidase for lysosomes (Sellinger et al., 1959), 2',3'- cyclic AMP phosphodiesterase for myelin, and oligodendrocyte plasma membranes (Drummond et al., 197 1) were measured according to procedures described previously. The proteins from the development study and fractions from the differential centrifugation were determined as described by Lowry et al. (1951) and the protein in the Nycodenz and Percoll gradient fractions was determined as described by Bradford (1976).

Other enzymatic assays Acyl-CoA:DHAP acyltransferase (DHAP acyltransferase)

was determined by the method of Schutgens et al. (19864. Acyl-CoA oxidase was measured by the procedure of Small et al. (1985). This method (Small et al., 1985) was also used to measure Damino acid oxidase and L-pipecolic acid oxidase

J. Neurochem., Vol. 56, No. 4, 1991

BRA IN PEROXISOMES 1345

"i

by using 30 mM of D-alanine and L-pipecolic acid as sub- strates, respectively.

Enzyme assay for oxidation of l-14C-labeled fatty acids to acetate (water-soluble products) was measured as previously described by Singh et al. (1 984). In brief, the reaction mixture (0.5 ml) contained l-I4C-labeled fatty acids (4 nmol of lignoceric acid or 6 nmol of palmitic acid), 20 mM MOPS-HCl buffer, pH 7.8, 30 mM KCl, 5 mM MgClz, 8.5 mM ATP, 0.25 mM NAD, 0.17 mM FAD, 2.5 mM L-carnitine, 0.08 mM CoA, and 1 mg of a-cyclodextrin. The reaction was started by the addition of the enzyme fraction and was stopped with 1.25 ml of 1 M potassium hydroxide in methanol and the denatured protein was removed by centrifugation. The supernatant was incubated at 60°C for 1 h, then neutralized with acid and partitioned according to the procedure of Folch et al. (1957). The amount of radioactivity in the upper phase is an index of the amount of l-14C-labeled fatty acid oxidized to acetate. For solubilization of the fatty acid substrate with a-cyclodextrin, the fatty acid (20 X lo6 dpm) was first dried in a tube under nitrogen and resuspended in 3.5 ml(20 mg/ ml) of a-cyclodextrin by sonication for 1 h.

Morphological examination of subcellular fractions Mitochondria1 and peroxisomal enriched fractions from

the density gradients were palleted by centrifugation and the pellets were fixed with 3% glutaraldehyde in 0.1 M sodium cacodylate buffer, pH 7.4, overnight at 4°C. Samples were postfixed with 1% osmium tetroxide in cacodylate buffer and then dehydrated in ethyl alcohol and propylene oxide. Sam- ples were then infiltrated with Embed 8 12 and polymerized at 60°C for 36 h. Thin sections were stained with uranyl acetate and lead acetate and examined by electron microscopy. The peroxisomal fraction from the Nycodenz density gradient was previously stained for catalase with the alkaline diaminobenzidiazine (DAB) technique of Angermuller and Fahimi (198 1) using 30 mM of D-danine as substrate.

RESULTS

Postnatal developmental pattern of peroxisomes Postnatal expression of peroxisomal enzymes (cat-

alase, acyl-CoA oxidase, and the enzyme system for oxidation of lignoceric acid) in the brain particulate fraction of rats between the ages of 3 and 36 days was examined (Fig. 1). The activity of catalase was signif- icantly higher than the peroxisomal @-oxidation enzymes at 3 days of age when expressed as units per gram of tissue or units per milligram of protein, suggesting that postnatal expression of catalase is earlier than the peroxisomal Boxidation enzyme system. The postnatal developments of other peroxisomal enzyme activities were similar irrespective of the units of expression. Enzyme activities for the peroxisomal p- oxidation system (acyl-CoA oxidase and oxidation of lignoceric acid) were low in 3-day-old brain followed by an increase with a peak between 10 and 16 days that decreased thereafter (Fig. 1). The distribution of catalase activity in the particulate and soluble fractions of brain from animals in this age group (3-36 days) was 85 f 7 and 15 k 4%, respectively. These results are in agreement with previous histochemical studies

C ATALASE ACYL-COA OXIDASE [l-'V1 LIGNOCERIC

ACID 8-OXIDATION

mU )i b r i m

UI Ir . ln

4.0,

0 20 40

DAYS AFTER BIRTH

FIG. 1. Peroxisornal enzyme activities in developing rat brain. The activities of catalase, acyl-CoA oxidase. and [I -14C]lignoceric acid oxidation were measured in the particulate fraction of brain from 3 to 36 days after birth. The catalase, acyl-CoA oxidase, and [ l - 14C]lignoceric acid oxidation activities in the particulate fractions were 85 _t 7,83 t 8, and 91 t 10% of the total activity, respectively. The results are expressed as averages f SD from three to six rats for each group.

showing numerous peroxisomes in the first 2 weeks of postnatal life (Arnold and Holtzman, 1978). Table 1 shows the enzymatic activities of marker enzymes for peroxisomes (catalase, acyl-CoA oxidase, DHAP acyltransferase, Damino acid oxidase, and pipec colic acid oxidase), mitochondria (cytochrome c oxidase), microsomes (NADPH cytochrome c reductase), myelin and oligodendrocyte plasma membrane (2',3'-cyclic AMP phosphodiesterase), and for lysosomes (N-acetyl- P-glucosaminidase) in homogenates of 10-day-old brain.

Subcellular fractionation of brain and purification of peroxisomes

Ten-day-old rats were used, because of high peroxisomal content and lower amount of myelin, to develop two different procedures for the isolation of peroxisomes. It was quickly realized that procedures used for isolation of peroxisomes from liver or other tissues may not be directly applicable to brain because of such a high content of myelin membranes. Therefore, the most of the myelin was removed by centrifugation of the tissue homogenate on a dense sucrose solution prior to density gradient centrifugation as described in Ma- terials and Methods.

J. Nemxltem.. Vol. 56, No. 4, 1991

1346 0. LAZO ET AL.

TABLE 1. Specific activities in homogenate of 10-day-old rat brain

Enzyme X+ SD n

Catalase (mU/mg of protein) Cytochrome c oxidase (mU/mg of protein) NADPH cytochrome c reductax (mU/mg of protein) Acyl-CoA oxidase (mU/mg of protein) 2’,3’-Cyclic AMP phosphodiesterase (nmol Pi/min/mg of protein) N-Acetyl-@-glucosaminidase (mU/mg of protein) DHAP acyltransferase (nmol/h/mg of protein) [ I-14C]Lignoceric acid P-oxidation (pmol/h/mg of protein) [ 1 -I4C]Palmitic acid @-oxidation (pmol/h/mg of protein) D-Amino acid oxidase (mU/mg of protein) L-Pipecolic acid oxidase (mU/mg of protein)

9.5 k 0.3 19.1 t 4.7 10.3 f 2.0 80.6 k 13.6 21.2 f 3.1

6.9 f 0.1 18.3 -t 3.9

209.5 f 5 1.2 65 I .3 f 63.0

0.27 0.25

5 5 5 5 5 5 3 7 4 1 1

The results are expressed as the means + standard deviations. n is the number ofdifferent experiments for the enzyme activities.

Method A. The distributions of marker enzymes for different organelles in various subcellular fractions prepared by a modification of the procedure of Leigh- ton et al. (1968) are shown in Fig. 2. The v fraction was rich in mitochondria and the I) fraction was rich in myelin and microsomes, whereas the X fraction had a relatively higher concentration of peroxisomal and lysosomal marker enzymes and a relatively lower concentration of microsomes, mitochondria, and myelin. The X fraction was subfractionated further by isopycnic density gradient centrifugation on a linear density gradient of Nycodenz (Fig. 3). As judged by the activities of different marker enzymes, the peroxisomes ( 1.125 * 0.014 g/ml density) were well resolved from mitochondria (1.189 _t 0.017 g/ml). However, the micro-

C A T A L A S E

0 @t?I CYTOCHROME c OXIDASE

0 tl- NADPH

CYTOCHROME c REDUC TASE

:!&-, 0

2’.3’CYCLlC AMP PHOSPHODI ESTE RASE

0 LtEI 0 100

N ACETYL 6 GLUCOSAMINIDASE

* , PIPECOLIC ACID OXIDASE

ACYL CoA OXIDASE

D-AMINO ACID OXIDASE

L 4 “ /

0 100

P R O T E I N (%) FIG. 2. Subcellular fractionation of brain (10 days old) by differential centrifugation (Method A). Fractionation was performed as described by Leighton et al. (1 968) except for the modifications described in the text. Results were averaged from four experiments and are presented as averages t SD. Recoveries of enzymatic activities ranged between 86 and 104%.

some myelin and oligodendrocyte plasma membrane equilibrated at densities of 1.134 f 0.0 15 and 1.120 k 0.0 15 g/ml, respectively, which are close to the peroxisomal density. The pattern of L-pipecolic acid oxidase, acyl-CoA oxidase, and D-amino acid oxidase in the gradient was similar to that for catalase activity, suggesting that all these enzymes were peroxisomal enzymes (Fig. 3). Comparison of marker enzymes in the

NADPH CYTOCHROME c

1 REDUClASt

2 ’ 3 CYCLIC AM? 1 PHOSPHODIESTIR&S€

0 100 0 100

V O L U M E (%I FIG. 3. Subfractionation of X fraction from brain in Nycodenz continuous density gradient (Method A). The X fraction obtained by differential centrifugation (Fig. 2) was subfractionated in a Nycodenz continuous gradient as described in the text. An average of four experiments (except enzyme activities for L-pipecolic acid oxidase, acyl-CoA oxidase, and D-amino acid oxidase) are shown. Recov- eries of the enzyme activities ranged between 85 and 103%. The density profile is also shown.


BRAIN PEROXISOMES 1347

TABLE 2. Percent of recovery and speciJic activities of marker enzymes in X fraction and in mitochondria and peroxisomal enriched,fiactions isolated from Nycodenz gradient

Mitochondria1 Peroxisomal X Fraction peak fraction peak fraction --

Percent Specific Percent Specific Percent Specific recovery‘ activity recovery activity recovery activity n

Protein Catalase Acyl-CoA oxidase Cytochrome c oxidase NADPH cytochrome c reductase N- Acety l-Sglucosaminidase 2’,3’-Cyclic AMP phosphodiesterase DHAP acyltransferase D-Amino acid oxidase L-Pipecolic acid oxidase

8.9 f 1 . 1 13.1 f 0.6

14.3 17.6 f 2.2 2.6 f 0.5

16.9 f 2.1 9.3 f 1.6

15.2 f 2.4 14.4 13.9

14.0 f 0.6 129.6

37.6 f 4.7 3.0 f 0.5

13.1 f 1.6 22.1 f 3.8 31.3 f 4.8

436.3 391.3

8.0 f 1.5 2.6 f 0.4

3.6 18.2 f 2.7 0.9 f 0.3

17.2 f 3.1 3.6 f 0.8 3.1 f 0.4

3.8 3.7

4.5 f 0.7 58.2

85.9 f 10.3 0.3 f 0.1

33.7 f 8.6 5.8 f 1.3

12.1 f 1.6 207.3 181.2

5.5 f 1.3 11.2 f 3.1

1 1 . 1 2.8 f 0.4

15.6 f 3.2 3.1 f 0.7

12.2 ? 2.8 15.8 f 2.5

12.6 13.6

28.4 f 3.5 4 26 I .7 2

19.2 f 3.8 4 8.5 f 2.1 4 7.4 f 1.7 4

49.1 f 10.1 4 89.7 f 13.2 3

999.6 2 986.3 2

The results are expressed as means f standard deviations. n is the number of different experiments for the enzyme activities. The enzyme

a Percent recovery of enzyme activity in X fraction from homogenate. activities are expressed as in Table 1.

Percent recovery of the enzyme activity in the peak fraction of mitochondria and peroxisomes as compared with the X fraction.

A, peroxisomal, and mitochondria1 fractions showed that peroxisomes were purified 3.0-4.9 times compared to homogenate (Tables 1 and 2). The major contamination in the peroxisomal peaks was of 2’,3’-cyclic AMP phosphodiesterase (myelin and plasma membrane). Electron microscopic analysis showed that the peroxisomal peak fraction from the Nycodenz gradient, when stained with DAB, contained DAB-positive organelles with very heterogeneous sizes (0.12-0.47 pm diameter) and shape. This fraction contained other membranous components as well as some mitochondria, microsomes and myelin membranes (Fig. 6A).

Method B. The distribution of different marker enzymes in myelin (My), nuclear (N), heavy mitochondrial (M), light mitochondrial (L), microsomal (P), and cytosolic (S) fractions showed that the peroxisomal enzyme activities were enriched approximately two to three times in the L fraction as compared to the homogenate (Fig. 4 and Table 3). Therefore, the L fraction was used for further purification of peroxisomes by density gradient centrifugation in a Percoll gradient. The distribution of marker enzymes showed that the peroxisomal (density of 1.036 f 0.0 12 g/ml) and mitochondrial (1.07 l f 0.0 15 g/ml) peaks were well resolved from each other in the Percoll gradient (Fig. 5). However, both mitochondrial and peroxisomal fractions had some contamination of NADPH cytochrome c reductase and 2’,3‘-cyclic AMP phosphodiesterase. The percent recovery of enzyme activities in the L fraction and peroxisomal and mitochondrial enriched peak fractions is shown in Table 3. The relative specific activities of DHAP acyltransferase, catalase, cytochrome c oxidase, NADPH cytochrome c reductase, 2’,3‘-cyclic AMP phosphodiesterase, and N-acetylglu- cosaminidase in peroxisomes prepared by Method R were 6.7, 5.2 f 1.0, 1.1 +. 0.2, 1.2 k 0.2, 2.0 -t 0.3, and 1.2 f 0.2, respectively, compared to homogenate.

These results show the peroxisomal fraction prepared by Method B was only enriched in peroxisomes and myelin by six- and twofold, respectively. Electron microscopic examination revealed that the peroxisomal fraction was contaminated with myelin membrane as suggested by marker enzyme activities (Fig. 6B). Be- cause myelin membrane also contains NADPH cytochrome c reductase activity (Koul et al., 1980), the observed activity in the peroxisomal enriched fraction may be due the presence of myelin membranes in this preparation. Overall, peroxisomes isolated by Method

CATALASE CVlOCHROMEr N.ACETYL B OXIDASE GLUCOSAMINIDASE

I I h I 4 I IT i

A C Y L G A N A D P H “C .LIGNOC ERIC ACID = OXIDASE CYTOCHROME c OXIDATION

v) ;;h L l Y 0 h; = D H A P I’.TCVCLIC AMP “C.PALMITIC ACID c ACYL TRANSFERASE PHOSPHODIESTCRASE OXIDATION 4

0 100 0 100 0 100

P R O T E I N (%)

FIG. 4. Subcellular fractionation of brain (10 days old) by differential centrifugation (Method 8). Fractionation was performed as described by de Duve et al. (1955) except that the first fraction (My) was obtained by a discontinuous sucrose gradient. Results are averaged from four experiments and are presented as means f SD. Recoveries ranged from 86 to 104%.

J . Neurochem., Val. 56, No. 4, 1991

1348 0. LAZO ET AL.

TABLE 3. Percent of recovery and specijic activities of marker enzymes in L fraction and in mitochondria and peroxisomal enriched fractions isolated from Percoil gradient

L Fraction Mitochondria1 peak

fraction Peroxisomal peak fraction

Enzymes Percent Specific

recovery" activity n Percent Specific

recoveryb activity

Protein Catalase Acyl-CoA oxidase Cytochrome c oxidase NADPH cytochrome c reductase N- Acetyl-P-glucosaminidase 2',3'-Cyclic AMP phosphodiesterase DHAP acyltransferase

9.0 f 2.4 19.0 f 3.4 18.1 f 3.2 22.2 f 3.9

5.6 f 0.6 18.5 f 5.4 14.1 f 2.5 19.8 f 4.8

20.0 f 3.6 162.1 f 29.3 47.1 f 8.3

6.4 f 0.7 14.2 f 3.9 33.2 f 5.9 40.3 f 9.4

12.1 f 2.2 5.5 f 1.1 7.3 f 1.2

36.9 f 3.5 6.1 f 1.1

24.6 f 4.2 10.9 f 1.9

9.1 f 1.8 97.7 f 11.0

143.0 k 12.3 3.2 f 0.5

28.6 f 4.9 29.8 i 4.9

3.6 11.8

Percent recoveryb

11.2 t 3.1 27.6 rtr 5.4 20.8 f 3.8

5.2 f 0.9 21.9 f 3.1

6.7 t 1.2 4.3 f 2.6

34.3

Specific activity

48.8 f 10.2 300.8 f 47.3 21.9 f 3.7 12.9 f 2.1 8.4 f 1.5

42.4 & 7.7 122.3

n

4 4 3 4 4 4 4 2

-

~

The results are expressed as means f standard deviations. n is the number of different experiments for the enzyme activities. The enzyme

a Percent recovery of enzyme activity in an L fraction from homogenate. activities are expressed as in Table I .

Percent recovery of enzyme activity in peak fractions of mitochondria and peroxisomes from L fraction.

B are relatively pure as compared to ones prepared by Method A.

The density of peroxisomes ( 1.125 g/ml) isolated in Nycodenz gradient from young rat brain is lighter than that of peroxisomes from adult rat brain ( 1.18 g/ml) (Gaunt and de Duve, 1976) and heart muscle (1.215

CYTOCHROME c OXIDASL:

'! 6 n . .

1.31,

'. .-- - - .......-

2'.3'CYCLIC A M P PHOSPHODIEST E RASE

= "C LiCNOCERiC ACID I4C P A L M I T I C ACID N ACETYL 6 OXIDATION OXIDATION GLUCOSAMINIDASE

2 D H A P N A D P H PROTEIN ACYL TRANSFERASE CYTOCHROME c

6 / 1 R E D U C T A S E

V O L U M E (%I FIG. 5. Subfractionation of L fraction from brain in self-generating Percoll density gradient. L fractions were obtained by differential centrifugation (Fig. 5). Results shown are an average of four gradients except that for oxidation of [l-'4C]lignoceric and [I- ''C]palmitic acids and DHAP acyltransferase, which were from two experiments. Recoveries of the enzyme activities ranged between 87 and 106%. The profile of density is also shown.

g/ml) (Coonock and Perry, 1984) isolated on a sucrose gradient and from cultured skin fibroblasts isolated on a Metrizamide gradient (1.174 g/ml) (Santos et al., 1985) and a Nycodenz gradient (Santos et al., 1988; Lazo et al., 1988). The density of peroxisomes in the Percoll gradient ( 1.036 f 0.0 12 g/ml) was different than that observed in the Nycodenz gradient (Santos et al., 1988; Lazo et al., 1988), because peroxisomal membranes are permeable to Nycodenz whereas Percoll (a large polymer) cannot enter this organelle, resulting in a lower density. The density of peroxisomes isolated by Percoll gradients from brain was also lower than that of those isolated from brown adipose tissue (1.07 g/ml) (Norman and Flatmark, 1982). The low density of brain peroxisomes may be due to differences in composition, or their sticking to myelin may make them lighter because of the very high lipid content in myelin membrane. Electron microscopic analysis of enriched peroxisomal and mitochondrial peak fractions from the Percoll gradient showed that the peroxisomal fraction contained a heterogeneous population of peroxisomes (0.10-0.52 pm diameter) with some mitochondria and large membranous vesicles, presumably fragmented myelin membranes (Fig. 6B). The mitochondrial peak fraction showed condensed and swollen organelles and large membranous vesicles, probably from myelin membranes (Fig. 6C).

Oxidation of lignoceric and palmitic acid in mitochondria1 and peroxisomal enriched fractions from Percoll gradient

The oxidation of lignoceric acid in the Percoll gradient followed the pattern of the distribution of peroxisomal marker enzymes (catalase, acyl-CoA oxidase, and DHAP acyltransferase) (Fig. 5 ) . The specific activity for &oxidation of lignoceric acid in the peroxi- soma1 peak fraction was 4.2-fold higher than the one observed in the mitochondria1 peak fraction with spe-

J. Neurochem., Vol. 56. No. 4. 1991


FIG. 6. Morphological appearance of subcellular fraction from brain when examined under transmission electron microscopy. A The peroxisomal enriched fraction from Nycodenz density gradients were stained with alkaline DAB for cytochemical localization of catalase and then stained with uranyl acetate and lead citrate. This field shows several DAB-positive vesicles of heterogeneous size. The arrow shows peroxisomes (p). Several other membranous components are present, presumably from myelin membranes. X24,OOO. B A peroxisomal enriched fraction from Percoll density gradient is a very heterogeneous membranous fraction with large vesicles, presumably from myelin membranes. The size of peroxisomes is heterogeneous (p). X40.000. C Mitochondria1 enriched fractions from Percoll density gradient. This field shows condensed mitochondria (arrow) and swollen mitochondria (double arrow). Large membranous vesicles are probably from the myelin membranes. X40.000.

J. Neurochem.. Vol. 56, No. 4. 1991

1350 0. LAZO ET AL.

TABLE 4. Percent recovery and specijk activity of fatty acid oxidation enzyme system in Lfraction and in mitochondrial and microperoxisomal peak fraction from Percoll gradient

Mitochondria1 peak Peroxisomal peak L Fraction fraction fraction

Percent Specific Percent Specific Percent Specific recovery" activity n recovery activity recoveryb activity

[ 1 -'4C]Lignoceric acid oxidation 15.8 f 4.2 368 f 78 3 6.0' 181 23.4 765 [ l-14C]Palmitic acid oxidation 13.7 2 3.1 991 f 19 3 31.3 2,565 9.6 848

The results are expressed as means +. standard deviations. n is the number of different experiments for the enzyme activities. The specific

a Percent recovery of enzyme activity in an L fraction from homogenate.

'The results are the average of two gradients.

activities are expressed as pmol/h/mg of protein.

Percent recovery of enzyme activity in peak fraction of mitochondria and peroxisomes from L fraction.

cific activities of 765 and 18 1 pmol/h/mg of protein, respectively (Table 4). In contrast, the specific activity for the oxidation of palmitic acid was 2.9-fold higher in the mitochondrial peak fraction as compared to the peroxisomal peak with specific activities of 2.56 and 0.85 nmol/h/mg of protein, respectively (Table 4). The percent recovery of the enzyme activity for the oxidation of palmitic and lignoceric acid in the L fraction was similar whereas the recovery of oxidation activity in the mitochondrial peak fraction was 6.0 and 3 1.3% and in the peroxisomal enriched fraction 23.4 and 9.6% for lignoceric and palmitic acids, respectively (Table 4). The recovery of DHAP acyltransferase in the mitochondrial and peroxisomal enriched fractions prepared by Percoll gradient was 3.6 and 34.3%, respectively (Table 3).

DISCUSSION

The postnatal expressions of peroxisomal specific enzymes (catalase, acyl-CoA oxidase, and the enzyme system for &oxidation of lignoceric acid) in rat brain particulate fractions are in agreement with previous histochemical studies showing catalase-positive particles (peroxisomes) in oligodendrocytes during the first 2 postnatal weeks which decrease with further growth (Arnold and Holtzman, 1978). This increase in the number of peroxisomes in oligodendrocytes (Arnold and Holtzman, 1978) and the postnatal increase in the activities of peroxisomal enzymes associated with synthesis of myelin lipids (Fig. 1; Hajra et al., 1987) with the onset of myelination strongly support the role of peroxisomes in the biogenesis of myelin. Myelin is a major membrane in brain and is very rich in plasmalogens (20-30% phospholipids) as compared to other membranes. The major difficulty in purification of brain peroxisomes compared to peroxisomes from liver, kidney, and cultured skin fibroblasts was the presence of large amounts of myelin. During initial homogenization of brain, myelin membranes are bro- ken down into fragments of different size and density

which end up in various subcellular fractions. The peroxisomal enriched fraction prepared after removal of the most of the myelin prior to the gradient in both Method A and Method B surprisingly contained an appreciable amount of myelin marker activity, 1.1 zk 0.2 and 2.0 zk 0.3% of the homogenate 2',3'-cyclic AMP phosphodiesterase activity prepared by Methods A and B, respectively. The percent activity of subcellular organelles shows that peroxisomes prepared by Method B were relatively pure as compared to ones prepared by Method A. While this manuscript was in preparation, an article from another laboratory re- garding the isolation of peroxisomes from brain ap- peared (Singh et al., 1989). Comparison of our procedure and theirs showed three significant differences. We used 10-day-old Sprague-Dawley rats, the age at which the peroxisomal population in the oligodendrocytes is highest, whereas they used mature rats (Porton strain), an age at which oligodendrocytes have fewer peroxisomes (Arnold and Holtzman, 1978). Second, they used a gradient similar to the one described by Ghosh and Hajra ( 1986) with only one concentration of Nycodenz. In this gradient, peroxisomes settle down at the bottom of the tube whereas mitochondria, microsomes, and other membranes clump together at the top of the gradient. Their procedure also differs from ours in that it does not include a step for removal of myelin before fractionation of different subcellular organelles. Subcellular organelles prepared by this procedure are generally heavily contaminated with myelin. It was difficult to compare the purity of their peroxi- soma1 preparation with ours because different enzyme markers were used except for NADPH cytochrome c reductase as a microsomal marker. NADPH cytochrome c reductase (1 69 mU/min/mg of protein) was 7.7 times higher in their peroxisomal fraction (Table I of Singh et al., 1989) than the peroxisomal fraction (2 1.9 mU/min/mg of protein) prepared by our Method B (Table 3).

The parallel distribution of activity for oxidation of lignoceric acid with three peroxisomal marker enzymes



(catalase, acyl-CoA oxidase, and DHAP acyltransferase) in the gradient for purification of different subcellular organelles clearly demonstrates that in brain lignoceric acid is oxidized mainly in peroxisomes whereas palmitic acid is oxidized mainly in mitochondria (Fig. 5). These results are consistent with our previous observation in peroxisomes from liver (Singh et al., 1984), cultured skin fibroblasts (Lazo et al., 1988, 1989), and in crude subcellular fractions from brain (Singh and Singh, 1986) and also with the observations from other laboratories (Singh et al., 1987; Wanders et al., 1987). The lack of oxidation of lignoceric acid in mitochondria from liver and cultured skin fibroblasts was suggested to be due to the absence of lignoceroyl- CoA ligase in mitochondria (Singh et al., 1987; Lazo et al., 1990). Contrary to this conclusion, Poulos and co-workers have recently reported that unlike mitochondria from liver and cultured skin fibroblasts, the mitochondria from brain oxidize a significant amount of lignoceric acid (Singh et al., 1989). The specific activities for oxidation of lignoceric acid in mitochondrial and peroxisomal enriched fractions were 53 1 and 585 pmol/h/mg protein, respectively (Table 1 of Singh et al., 1989), whereas we observed a four times higher specific activity in peroxisomal (764.8 pmol/h/mg of protein) than in mitochondrial (18 1.2 pmol/h/mg of protein) enriched fractions (Table 4). At this time, it is difficult to understand these differences between liver and brain mitochondria except that their preparations of brain subcellular organelles were not as enriched as ours. Because mitochondria can oxidize lignoceric acid efficiently if supplemented with acyl-CoA ligase, it is possible that the mitochondrial preparation was contaminated with acyl-CoA ligase activities from other cellular membranes (microsomes, peroxisomes) (Lazo et a]., 1989) and/or myelin. Consistent with this hy- pothesis, the oxidation of lignoceric acid in the mitochondrial-microsomal region (fractions 1-4) of their gradient followed the pattern of the microsomal marker more than that of the mitochondrial marker (Fig. 2 of Singh et al., 1989). Although their mitochondrial preparation prepared by Percoll gradients was relatively free of microsomes and peroxisomes, contamination by myelin cannot be ruled out. Myelin is a major membrane in the brain with acyl-CoA ligase activity (Vas- wani and Leeden, 1987) and mitochondria in the presence of myelin may oxidize lignoceric acid. However, the final conclusion about the possible participation of mitochondria in the oxidation of lignoceric acid in brain should be provided by the identification of the presence of lignoceroyl-CoA ligase in this organelle.

Histochemical studies for catalase and Damino acid oxidase suggested regional differences in the distribution of these two peroxisomal marker enzymes in brain (Arnold and Holtzman, 1978; Holtzman, 1982). Cat- alase-rich peroxisomes were observed in cerebral hemisphere whereas D-amino acid oxidase rich peroxisomes were observed in cerebellum. By using su-

crose gradient centrifugation Gaunt and de Duve (1 976) were able to resolve cytoplasmic particles containing D-amino acid oxidase and the ones containing catalase. The majority of catalase (65-75) had latent activity whereas D-amino acid oxidase activity showed no such latency. In our study, the patterns of enzyme activities for catalase and D-amino acid oxidase in the Nycodenz gradient were similar, suggesting that even though their distribution in brain may be different, the density of peroxisomes containing these activities is similar in the Nycodenz gradient (Fig. 3). The observed regional, cellular, and developmental differences in the abundance of peroxisomes and the possibility of peroxisomes with different enzyme activities allows spec- ulation that peroxisomes in the nervous system may have special functions in different cell types of different regions during postnatal development. Pipecolic acid is an intermediate in the metabolism of lysine (Roth- stein and Miller, 1954). In rat liver, most lysine deg- radation does not involve the pipecolic acid pathway (Higashino et al., 1965), whereas in brain, lysine is pre- dominantly catabolized via the pipecolic acid pathway (Chang, 1978). The subcellular localization of L-pipecolic acid oxidation was shown to be species depen- dent. In rats and rabbits, this reaction was observed in mitochondria whereas in primates it occurs in peroxisomes (Mihalic et al., 1989; Mihalic and Rhead, 1989). However, Kramer et al. (1989) showed that rat liver and kidney oxidize L-pipecolic acid by a H202 producing reaction. This activity was enhanced in liver from animals treated with clofibrate and thyroxine, known peroxisomal proliferators. Our results show that L-pipecolic acid can be degraded in brain peroxisomal enriched fraction by a H202-producing enzyme activity.

The importance of peroxisomes in brain is high- lighted further by the discovery of a number of peroxisomal disorders associated with dysmyelination and demyelination (Schutgens et al., 1986~; Moser, 1986; Singh et al., 1988). These disorders are divided into two main groups, one with multiple enzyme deficiencies (e.g., Zellweger syndrome) and the other with specific peroxisomal enzyme deficiencies (e.g., X-ALD). Zellweger patients have a deficiency of plasmalogens and the enzyme necessary for their synthesis (DHAP acyltransferase) and the accumulation of VLC fatty acids and pipecolic acid due to the absence of peroxisomes (Lazarow and Moser, 1989). On the other hand, in X-ALD, peroxisomes are ultrastructurally normal but the oxidation of the VLC fatty acid lignoceric acid was found to be impaired, because of a single peroxisomal enzyme deficiency, lignoceroyl-CoA ligase (Hashmi et al., 1986; Lazo et al., 1988, 1989; Wanders et al., 1988). In X-ALD brain, the accumulation of VLC fatty acids (c22-c26) as a constituent of cholesterol esters compared to the accumulation of cholesterol esters with fatty acids of long chain length (c16-Cl8) observed in other demyelinating diseases suggests that accumulation of cholesterol esters containing VLC fatty

J. Neurochem.. Vol. 56, No. 4, 1991

1352 0. LAZO ET AL.

acids in X-ALD may not be a result of demyelination but rather due to an abnormality in their oxidation in peroxisomes. This is also supported by the recent observation of the accumulation of VLC fatty acids as a constituent of phosphatidylcholine in the tissue around the actively demyelinating area in X-ALD brain prior to the accumulation of VLC in cholesterol esters (Theda et al., 1990).

Acknowledgment: This work was supported by grants from the National Institutes of Health (NS-22576) and from the March of Dimes Birth Defects Foundation ( 1 - 1079). We thank Ms. J an Ashcraft and Ms. Terry Heuer for technical support and Ms. Fran Shuler for typing the manuscript.

REFERENCES

Angermuller S. and Fahimi H. D. (1981) Selective cytochemical localization of peroxidase, cytochrome oxidase and catalase in rat liver with 3,Y-diaminobenzidine. Histochemistry 71, 33-44.

Arnold G. and Holtzman E. ( 1 978) Microperoxisomes in the central nervous system of the post-natal rat. Brain Res. 55, 1-17.

Baudhuin P., Beaufay Y . , Rahman-Li. Y., Sellinger 0. Z., Wattiaux R., Jacques P., and de Duve C. (1964) Tissue fractionation studies. Biochem. J. 92, 179-184.

Beaufay H., Amar-Costesec A., Feytmants E., Thines-Sempoux D., Wibo M., Robbi M., and Berthet J. (1974) Analytical study of microsomes and isolated subcellular membranes from rat liver. J. Cell Biol. 61, 188-200.

Bradford M. M. (1976) A rapid and sensitive method for the quan- titation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 12, 248-254.

Chang Y. F. (1978) Lysine metabolism in the rat brain: the pipecolic acid-forming pathway. J. Neurochem. 30, 347-358.

Coonock M. J. and Perry S. R. (1984) Detection of acyl-CoA 0- oxidation enzymes in peroxisomes (microperoxisomes) of mouse heart. Biochern. Int. 6, 545-55 1.

Cooperstein S. J. and Lazarow P. B. (1951) A microspectrophoto- metric method for the determination of cytochrome oxidase. J. Biol. Chem. 189,665-670.

de Duve C. and Baudhuin P. (1966) Peroxisomes (microbodies and related particles). Physiol. Rev. 46, 323-357.

de Duve C., Pressman B. C., Gianetto R., Wattiaux R., and Appel- mans F. (1955) Tissue fractionation studies. Biochem. J . 60,

Drummond G. I., Emg D. Y. , and Mclntosh C. A. (1971) Ribonu- cleoside 2’,3’-cyclic phosphate diesterase activity and cerebroside levels in vertebrate and invertebrate nerve. Brain Res. 28, 153- 163.

Folch J., Lees M., and Stanley H. S. (1957) A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226,497-509.

Gaunt G. L. and de Duve C. (1976) Subcellular distribution of D- amino acid oxidase and catalase in rat brain. J. Neurochem. 26, 749-759.

Ghosh M. K. and Hajra A. K. (1986) A rapid method for the isolation of peroxisomes from rat liver. Anal. Biochem. 159, 169-174.

Goldfischer S., Moore C. L., Johnson A. B., Spiro A. J., Volsamis M. P., Wisniewski H. J., Richt R. H., Norton W. T., Rapin I . . and Gartner L. M. (1973) Peroxisomal and mitochondrial defects in the cerebro-hepato-renal syndrome. Science 182, 62-64.

Hajra A. K. and Bishop J. E. (1982) Glycerolipid biosynthesis in peroxisomes via the acyl-dihydroxyacetone pathway. Ann. N Y Acad. Sci. 386, 170-182.

Hajra A. K., Bishop J . E., and Webber K. 0. (1987) Ontogeny of

604-6 17.

glycerolipid-synthesizing enzymes in brain microbcdies. (Abstr.) J. Neurochem. 48 (Suppl.), S39.

Hashmi M., Stanley W., and Singh I. (1986) Very long chain acyl- CoA ligases: enzyme defect in childhood adrenoleukodystrophy. FEBS. Lett. 196, 247-250.

Hess R., Stabubli W., and Riess W. (1965) Nature of hepatomegalic effect produced by ethyl-chlorophenoxy-isobutyrate in rat liver. Nature 208, 856-858.

Higashino K., Tsukuda K., and Lieberman I. (1965) Saccaropine, a product of lysine breakdown by mammalian liver. Biochem. Biophys. Res. Commun. 20,285-290.

Holtzman E. ( 1982) Peroxisomes in nervous tissue. Ann. NY Acad. Sci. 386, 523-525.

Hoshi M. and Kishimoto Y. (1973) Synthesis of cerebronic acid from lignoceric acid by rat brain preparation. J. Biol. Chem. 248, 4 1 23-4 130.

Koul O., Chou K. H., and Jungalwala F. B. (1980) UDP-galactose- ceramide galactosyltransferase in rat brain myelin subfractions during development. Biochem. J. 186,959-969.

Kramer R., Kremser K., and Schon H. (1989) Peroxisomal oxidation of pipecolic acid in the rat. J. Clin. Chem. 27, 319-321.

Lazarow P. B. and Moser H. (1989) Disorders of peroxisome biogenesis, in The Metabolic Basis of Inherited Disease (Scriver C. R.. Beaudet A. L., Sly W. S., and Valle D., eds), pp. 1479- 1509. McGraw-Hill, New York.

Lazo O., Contreras M.. Hashmi M., Stanley W., and Singh I. (1988) Peroxisomal lignoceroyl-CoA ligase deficiency in childhood adrenoleukodystrophy and adrenomyeloneuropathy. Proc. Natl. Acad. Sci. USA 85,7647-765 1.

Lazo O., Contreras M., Bhushan A., Stanley W., and Singh I. (1989) Adrenoleukodystrophy: impaired oxidation of fatty acids due to peroxisomal IignoceroylCoA ligase deficiency. Arch. Biochem. Biouhvs. 270. 722-728.

Lazo 0.: Contreras M., Yoshida Y., Singh A. K., Stanley W., Weise M., and Singh 1. (1990) Cellular oxidation of lignoceric acid is regulated by the subcellular localization of lignoceroyl-CoA ligases. J. Lipid Rex 585, 2595-2608.

Leighton F., Poole B., Beaufay H., Baudhuin P., Coffay J. W., Fowler S., and de Duve C. (1968) The large-scale separation of peroxisomes, mitochondria, and lysosomes from the livers of rats in- jected with Triton WR-1339. J. Cell Biol. 37, 482-512.

Lowry 0. H., Rosebrough N. J., Farr A. L., and Randall R. J. (I95 1) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193,265-275.

Mihalic S. J. and Rhead W. J. (1989) L-Pipecolic acid oxidation in the rat cynomolgus monkey. J. Biol. Chem. 264,2509-2517.

Mihalic S., Moser H. W., Watkins P. A., Danks D. M., Poulos A., and Rhead W. J. ( 1 989) Peroxisomal L-pipecolic acid oxidation is deficient in liver from Zellweger syndrome patients. Pediatr. Rex 25, 548-552.

Moser H. W. (1986) Peroxisomal disorders. J. Pediatr. 108, 89-91. Norman P. T. and Ratmark T. (1982) Microperoxisomes and mi-

tochondria of brown adipose tissue. Hydrodynamic parameters, isolation and capacity of long-chain fatty acid oxidation. Biochim. Biophys. ilcta 112,621-627.

Novikoff A. B.. Novikoff P. M., Davis C., and Quintana N. (1973) Studies on microperoxisomes. V. Are microperoxisomes ubiquitous in mammalian cell? J. Histochem. Cytochem. 21, 737- 742.

Rome L. H., Garvin A. J., Allieta M. M., and Neufeld E. F. (1979) Two species of lysosomal organelles in cultured human fibroblasts. J. CellBiol. 17, 143-153.

Rothstein M. and Miller L. L. (1954) The conversion of lysine to pipecolic acid in rat. J. Biol. Chem. 211, 851-858.

Santos M. J., Ojeda J. M., Gamdo J., and Leighton F. (1985) Per- oxisomal organization in normal and cerebro-hepato-renal (Zellweger) syndrome fibroblast. Proc. Natl. Acad. Sci. USA 82, 6556-6560.

Santos M. J., Imanaka T., Shio H., and Lazarow P. B. (1988) Per- oxisomal integral proteins in control and Zellweger fibroblasts. J. Biol. Chem. 263, 10502-10509.



Schutgens R. B. H., Heymans H. S. A., Wanders R. J. A., Bosch H. V. D., and Tager J. M. (1986~) Peroxisomal disorders: a newly recognized group of genetic diseases. Eur. J. Pediatr. 144,

Schutgens R. B. H., Romeyn G. J., Ofman R., van den Bosch H., Tager J. M., and Wanders R. J. A. (1986b) Acyl-CoA dihydroxyacetone phosphate acyltransferase in human skin fibroblast; study of its properties using a new assay method. Biochim. Biu-

Sellinger 0. Z., Beaufay H., Jacques P., Doyen A., and de Duve C. (1959) Tissue fractionation studies. Biochem. J. 74,450-456.

Singh R. P. and Sin& I. (1986) Peroxisomal oxidation of fatty acid in brain. Neurochem. Res. 11, 281-289.

Sin& I., Moser A. N., Goldfischer S., and Moser H. W. (1984) Lig- noceric acid is oxidized in the peroxisome: implications for the Zellweger cerebrohepato-renal syndrome and adrenoleukodystrophy. Proc. Nutl. Acad. Sci. USA 81, 4203-4207.

Singh H., Denvas N., and Poulus A. (1987) Very long chain fatty acid P-oxidation by rat liver mitochondria and peroxisomes. Arch. Biochem. Biophys. 259,382-390.

Sin& I., Johnson G. H., and Brown F. R. (1988) Peroxisomal disorders. Am. J. Dis. Child. 142, 1297-1301.

Sin& H., Usher S., and Poulos A. (1989) Mitochondria1 and per-

430-440.

phys. ActU 879, 286-29 1.

oxisomal P-oxidation of stearic and lignoceric acids by rat brain. J. Neurochem. 53, 171 1-1718.

Small G. M., Burdett K., and Connock M. J. (1985) Sensitive spec- trophotemetric assay for peroxisomal acyl-CoA oxidase. J. Biochem. 227,205-2 10.

Theda C., Moser A., and Moser H. W. (1990) Abnormality of brain phosphatidylcholine fatty acids in peroxisomal diseases. Trans. Am. SOC. Neurochem. 21, 110.

Tolbert N. E. (1981) Metabolic pathways in peroxisomes and gly- oxisomes. Annu. Rev. Biochem. 50, 133-157.

Vaswani K. K. and Ledeen R. W. (1987) Long chain acyl-coenzyme A synthetase in rat brain myelin. J. Neurusci. Res. 17, 65-70.

Wanders R. J. A,, Van Roermund C. W. T., Van Wigland M. J. A,, Schutgens R. B. H., Schram A. W., Van den Bosch H., and Tager J. M. (1987) Studies on peroxisomal oxidation of palmitate and lignocerate in rat liver. Biochim. Biophys. Acia 919,2 1-25.

Wanders R. J. A., Van Roermund C. W. T., Van Wijland M. J. A,, Schutgens R. B. H., Van Den Bosch H., Scham A. W.. and Tager J. M. (1988) Direct demonstration that the deficient oxidation of very-long-chain fatty acids in X-linked adrenoleukodystrophy is due to an impaired ability of peroxisomes to activate very long chain fatty acids. Biochem. Biuphyx. Res. Commun. 153,618-624.

J Neurochem.. liol 56. N o 4, IY91

Postnatal Development and Isolation of Peroxisomes from Brain

Documents