The acylation of lysophosphatidylcholine by subcellular fractions of guinea-pig cerebral cortex

231

Biochimica et Biophysics Acta, 618 (1980) 231-241 @ Elsevier/North-Holland Biomedical Press

BBA 57561

THE ACYLATION OF LYSOPHOSPHATIDYLCHOLINE BY SUBCELLULAR FRACTIONS OF GUINEA-PIG CEREBRAL CORTEX

STEPHEN K. FISHER * and CHARLES E. ROWE Department 2TT (U.K.)

of Biochemistry,

(Received September l&h,

University of Birmingham,

P.O. Box 363, Birmingham B15

1979)

Key words: Acylation; Lysophosphatidylcholine; Synaptic plasma membrane; Endoplasmic reticulum; (Guinea-pig cerebral cortex)

Summary The acylation of lysophosphatidylcholine by isolated subcellular fractions of guinea-pig cerebral cortex has been determined. The microsomal fraction contained the highest acylation activity, in terms of both specific and total activity. In all particulate fractions, including synaptic plasma membrane and mitochondria, there was a high correlation (correlation coefficient r = 0.90; P < 0.001) between acylation and the activity of the microsomal enzyme, NADPH-cytochrome c reductase. No correlation existed between acylation and the activities of (Na’ + K’)-ATPase, acetylcholinesterase or succinate dehydrogenase. Acyl-CoA synthetase and lysophosphatidylcholine/acyltransferase, the individual enzymes responsible for acylation were enriched in the microsomal fraction. The activities of both enzymes in subcellular fractions correlated well with those of NADPH-cytochrome c reductase, with the exception that acylCoA synthetase activity in the mitochondrial fraction was largely independent of endoplasmic reticulum. Neither synaptic plasma membranes nor mitochondria appeared to possess significant amounts of acyltransferase activity. The results indicate that the acylation of lysophosphatidylcholine is confined to the endoplasmic reticulum, and that activity present in the synaptic plasma membrane or mitochondrial fraction is attributable to microsomal contamination.

* Present address: Neuroscience Laboratory, University of Michigan, 1103 E. Huron. Ann Arbor. 48109, U.S.A. Abbreviations: PPO. 2.5-diphenyloxazole; POPOP, 1,4-bis-(5-phenyloxazolyl-2)-benzene.

MI

232

Introduction The formation of choline phosphoglyceride * from the acylation of lysophosphatidylcholine has been demonstrated in a number of tissues, including brain [l-4]. In conjunction with phospholipase activity, acylation provides the basis for the deacylation-acylation cycle whereby the molecular species of choline phosphoglycerides may be interconverted through alteration of the fatty acids within the molecule [ 11. While in liver, the endoplasmic reticulum is undoubtedly the principal site of activity, it is less certain whether acylation may occur at additional subcellular sites, such as plasma membranes or mitochondria [ 5-91. In contrast, there is good evidence for the acylation of lysophosphatidylcholine by plasma membranes from acanthamoeba [lo], erythrocytes [ 11,121, lymphocytes [ 131 and reticulocytes [ 141. The subcellular distribution of acylation in brain has not been studied extensively. Of particular interest is the question of whether the acylation of lysophosphatidylcholine can occur in synaptic plasma membranes or mitochondria, in addition to the established microsomal location [3]. Recently, Corbin and Sun [ 151 described the properties of an acylation activity present in mouse brain synaptosomes, but the subsynaptosomal location of enzyme activity remains uncertain. In this study, we have investigated the distribution of acylation of lysophosphatidylcholine in subcellular fractions of guinea-pig cerebral cortex that had been assessed for cross-contamination by the use of marker enzymes. Evidence is presented that most, if not all, of the ability of subcellular fractions to acylate lysophosphatidylcholine is attributable to that part of the microsomal fraction associated with NADPH-cytochrome c reductase. A preliminary communication of part of this work has been presented

[=I. Materials

and Methods

Guinea-pigs, Dunkin-Hartley strain, were purchased from David Hall, Burtonon-Trent. [ 9,10(n)-3H] Oleic acid, [ l-14C] oleic acid and [ l-14C]palmitic acid were supplied by the Radiochemical Centre, Amersham. Bovine serum albumin (fatty acid free), lysophosphatidylcholine, ATP, CoA, oleic acid, palmitic acid, oleoyl- and palmitoyl-CoA were obtained from Sigma (London) Ltd. Solvents were of analytical grade. Preparation of [9,1 0(n)-3H]oleoyl-CoA. Radioactive oleoyl-CoA was routinely prepared by the procedure of Sanchez et al. [ 171. On extraction of an aqueous solution by the procedure of Dole [ 181 less than 5% of total radioactivity was soluble in n-heptane and when subjected to TLC [ 19,201, more than 94% of radioactivity cochromatographed with authentic oleoyl-CoA. The ratios of absorption of light at different wavelengths were 250 : 259 nm, 0.87; 280 : 259 nm, 0.25; 232 : 259 nm, 0.60, in close agreement to previously published values [ 211.

* In

this

monoac~l,

paper

the

term

choline

phosphoglyceride

monoalkenyl-sn-glycero-3-phosphocholine.

refers

to

1,2_diacvl-sn-glycero-3-phosphocholine

and

233

Preparation of subcellular fractions. All procedures were carried out at 4°C. Six adult male guinea-pigs were used for each experiment and were killed by stunning and exsanguination. Whole brains rostral to the colliculi were removed within 2 min and rinsed in 0.32 M sucrose to remove excess blood. The cerebral cortices were then separated from the underlying midbrain structures and a 10% (w/v) homogenate in 0.32 M sucrose prepared. Subcellular fractionation was then carried out by the method of Gray and Whittaker [22]. Synaptic plasma membranes were obtained from synaptosomes by the method of Sun et al. [23], following osmotic lysis of synaptosomes in alkaline buffer [24]. Resuspension

of subcellular

fractions

for determination

of enzyme

activities.

Homogenate, crude nuclear fraction, myelin and synaptic plasma membranes were diluted with a large excess of 0.32 M sucrose, and centrifuged at 105 000 X g for 60 min to obtain particulate material. The pelleted fractions were resuspended in distilled water to give concentrations of l-6 mg protein/ ml. Suspensions of osmotically ruptured synaptosomes were adjusted to pH 7.4 by the addition of 150 mM Tris-HCl buffer (pH 7.4). Mitochondrial and microsomal fractions were resuspended directly in water. Determinations were made on the supernatant (S,) fraction without prior removal of sucrose. Determination of the (lysophosphatidylcholine).

acylation

of

l-acyl-sn-glycero-3-phosphorylcholine

Overall acylation was determined in the presence of lysophosphatidylcholine and the substrates necessary for the generation of acyl-CoA. Subcellular fractions (0.5-2.5 mg protein) were incubated for 10 min at 37°C in an assay system containing (final concentrations): 15 mM Tris-HCl buffer (pH 7.4); 20 mM MgCl*; 10 mM disodium ATP (neutralized with NaOH); 100 PM CoA; 0.5 mM lysophosphatidylcholine and 0.5 mM r3H]or [‘4C]oleate or [14C]palmitate (spec. act. approx. l-2 Ci/mol) complexed to delipidated bovine serum albumin (5 mg/ml) [25], in a total volume of 1 ml. Reactions were terminated by the addition of 4 ml chloroform/methanol (1 : 2, v/v) and lipids extracted as described previously [26]. Lipid extracts were concentrated to dryness at 45-50°C under Nz and separated by TLC [27]. After brief exposure to iodine vapour, choline phosphoglyceride was located on TLC plates by means of a phosphatidylcholine standard. The corresponding area of gel was transferred to a vial containing 2 ml CH30H/H20 (1 : 1, v/v) and 10 ml Triton X-l 00 scintillation fluid (1 : 2, v/v). The scintillation fluid consisted of a solution of 0.6% (w/v) PPO and 0.015% (w/v) POPOP dissolved in xylene. Radioactivity was determined using a Philips scintillation counter. The overall recovery of radioactive choline phosphoglyceride, after extraction and TLC stages was better than 83%, as judged from the recovery of 0.1 PCi [14C]phosphatidylcholine added to the reaction mixture. Fatty acid:CoA ligase (acyl-CoA synthetase, EC 6.2.1.3). Activity was determined essentially as described by Lloyd-Davies and Brindley [28]. The components of the assay system and final concentrations were the same as those in the determination of acylation of lysophosphatidylcholine with the exception that lysophosphatidylcholine was omitted and the final volume was 0.5 ml. Subcellular fractions (50-400 pg of protein) were incubated for 5 min at 37°C. Reactions were terminated by the addition of 2 ml Dole reagent [18] and radioactive acyl-CoA extracted into the aqueous layer as described previously [29] with the exception that the aqueous layer was washed five times

234

with 1.2 ml n-heptane. Aliquots of the aqueous phase were then taken for radioactive determinations [ 281. That the radioactivity in the aqueous phase was acyl-CoA was verified by subjecting ahquots of the aqueous phase to the chromatographic procedures of Pullman [ 191 and Ullman and Radin [20]. 85-90% of recovered radioactivity co-chromatographed with authentic oleoylCoA, using these procedures. Acyl-CoA:lacyl-sn-glycero-3-phosphorylcholine (lysophatidylcholine): acyltransferase (EC 2.3.1.23). Activity was determined from the measurement of

formation of radioactive choline phosphoglyceride from [9,10(n)-3H]oleoylCoA and lysophosphatidylcholine. Subcellular fractions (50500 pg protein) were incubated for 10 min in an assay system containing (final concentrations); 15 mM Tris-HCl buffer (pH 7.4); 0.15 mM [9,10(n)-3H]oleoyl-CoA (spec. act. approx. 2 Ci/mol); 0.5 mM lysophosphatidylcholine and 5 mg/ml delipidated bovine serum albumin, in a total volume of 0.25 ml. Reactions were terminated by the addition of 2 ml chloroform/methanol (1 : 2, v/v) and carrier phosphatidylcholine (0.1 pmol) and oleic acid (2 pmol) added. Lipids were extracted and quantified as described for the acylation of lysophosphatidylcholine. Acyl-CoA hydrolase. Enzyme activity was routinely measured by a spectrophotometric procedure [30]. 50-400 pg of protein were incubated in an assay system containing (final concentrations): 15 mM Tris-HCl buffer (pH 7.4) and 0.2 mM 5,5’dithiobis(2nitrobenzoic acid). Reactions were initiated by the addition of an aqueous solution of 50 PM oleoyl- or palmitoyl-CoA and enzyme rates were measured for approx. 5 min after a lag period of l-2 min [ 301. A series of preliminary experiments determined that the assay conditions chosen in this study for the measurement of acylation of lysophosphatidylcholine, acyl-CoA synthetase, lysophosphatidylcholine/acyltransferase and acyl-CoA hydrolase were such that the activities in the subcellular fractions were optimal and proportional to both the duration of assay and amount of tissue protein present. All assays were performed in duplicate or triplicate. The following marker enzyme activities were determined by standard procedures: NADPH-cytochrome c reductase (EC 1.6.2.4) [31]; succinate dehydrogenase (EC 1.3.99.1) [32]; lactate dehydrogenase (EC 1.1.2.3) [33]. Acetylcholinesterase (EC 3.1.1.7) was measured at 37°C by the method of Ellman et al. [34]. A set of parallel assays were run which included 0.33 mM 284-C51 (Wellcome Research Laboratories U.K.) to assess pseudocholinesterase activity [ 351. Na’ + K’-activated Mg2’-ATPase ((Na’ + K’)-ATPase, EC 3.6.1.3) was determined by incubating 30-100 pg of tissue protein for 5 min under the assay conditions described by Sun and Samorajski [36]. Reactions were terminated by the addition of 0.5 ml 40% (w/v) trichloroacetic acid and 10% (w/v) sodium dodecyl sulphate. Inorganic phosphate released was measured by the method of Baginski et al [37]. The difference between the release of inorganic phosphate in the absence and presence of 0.25 M ouabain was used to determine the activity of (Na’ + K’)-ATPase. Protein was measured by the method of Geiger and Bessman [ 381. Chronological

order of enzyme

assay

Due to the large amount of information required, it was not possible to determine all the enzyme activities on the same day. Determinations of acyla-

235

tion of lysophosphatidylcholine, acyl-CoA choline/acyltransferase were carried out Suspensions of tissue fractions were then activities were determined l-8 days after Pilot studies indicated that the subcellular not significantly altered by such storage.

synthetase, and lysophosphatidylon the day of tissue preparation. stored at -20°C. All other enzyme the preparation of tissue fractions. distribution of marker enzymes was

Results and Discussion Acylu tion of lysophosphatidylcholine

by su bcellular fractions

Since interpretation of the present results depends upon knowledge of crosscontamination between fractions, isolated fractions were characterized by determination of five different marker enzyme activities. Four of the enzymes are usually considered to be membrane bound, and for this reason the specific activities of all enzymes were calculated relative to the specific activity of homogenate-particulate rather than total homogenate. The marker enzyme distribution showed that the characteristic separation of subcellular fractions had been achieved (Table I). Acylation of added lysophosphatidylcholine using either oleate or palmitate was highest in the microsomal fraction with respect to both specific and total activity. With the exception of the crude nuclear fraction, which showed a small enrichement of activity when oleate was used as substrate, no other subcellular fraction was enriched over homogenate. Acylation activity was virtually undetectable in the soluble fraction. The enrichment of acylation activity closely paralleled that of NADPH-cytochrome c reductase, an established marker for endoplasmic reticulum in guinea-pig forebrain [ 391. When all the particulate subcellular fractions were compared, there was a highly significant correlation between acylation and NADPH-cytochrome c reductase activities (r = 0.90; P < 0.001). In marked contrast there was no significant correlation between acylation of lysophosphatidylcholine and (Na’ + K’)ATPase, acetylcholinesterase or succinate dehydrogenase activities. It is concluded therefore, that acylation occurs predominantly in the endoplasmic reticulum and is not significantly associated with plasma membrane or mitochondrial structures. Lysophosphatidylcholine/acyltransferase, h ydrolase in su bcellular fractions

acyl-CoA

synthetase

and acyl-CoA

Acylation of lysophosphatidylcholine by brain, utilizing internally generated acyl-CoA, results from the action of at least two enzymes, acyl-CoA synthetase and lysophosphatidylcholine/acyltransferase. In addition, the activities of these two enzymes could conceivably be influenced by acyl-CoA hydrolase, regulating the concentration of acyl-CoA. While overall acylation occurs predominantly in the endoplasmic reticulum, it remained possible that other subcellular fractions possess one or more of these activities. Lysophosphatidylcholine/ acyltransferase, with oleoyl-CoA as acyl group donor, was concentrated in the microsomal fraction (Table II). Activity was also present in other fractions, but with the exception of the crude nuclear fraction, the relative specific activities were less than 1. In all subcellular fractions, there was a highly significant correlation between acyltransferase activity and NADHPH-cytochrome c reductase (r = 0.93 P < 0.001). During the course of this work, Pik and

DISTRIBUTION

OF ACYLATION

OF LYSOPHOSPHATIDYLCHDLINE,

MARKER

ENZYMES AND PROTEIN IN GUINEA-PIG CERE-

76.6 k 1.8

% Recovery

t 1.3 t 0.5 i 1.1 + 0.4 + 1.0 c 0.1

26.2 8.0 17.8 9.7 14.9 2.2 -

1 (crude nuclear) P2A (myelin) PzB, (lysed synaptosomes) P2C (mitochondria) P3 (microsomes) SPM (synaptic plasma membranes) S3 (supematant)

P

Protein (%f (14)

Fraction

86.2

i 3.9

0.00

70.8, 71.5

0.00,

0.73.0.94 0.45,0.57 0.61, 0.62 0.55.0.41 2.22, 2.28 0.53, 0.69

1.10 0.40 0.68 0.72 2.19 0.58 0.04

f 0.05 f 0.05 * 0.06 f 0.05 + 0.11 * 0.04 f 0.02

Palmitate (21

Oleate (9-l 1)

Acylation of added lysophosphatidylchorine

Relative specific activity

84.0

0.95 0.59 0.82 0.62 2.10 0.84 0.58 + 6.0

f 0.07 f 0.08 f 0.08 f 0.09 + 0.09 + 0.04 f 0.10

NADPHcytochrome reductase (5-13) c

84.8

0.72 0.86 1.22 0.29 2.24 3.72 0.22

f 6.8

t 0.05 k 0.11 + 0.08 f 0.03 f 0.09 + 0.24 + 0.06

Acetyl cholinesterase (5-6)

95.7

1.15 0.89 1.55 0.51 1.17 4.54 0.00

t 7.1

i 0.05 5 0.25 * 0.14 i 0.05 f 0.15 + 0.61 + 0.00

(Na+ + K+b ATPase (4)

80.9

0.78 0.08 0.92 4.23 0.13 0.13 0.00

+ 3.3

j_ 0.08 I 0.02 _+0.08 I 0.46 t 0.03 + 0.02 * 0.00

Succinate dehydrogenase (5-6)

79.9

0.94 0.71 0.97 1.20 1.34 0.46 2.83

+ 4.0

* 0.04 10.09 ? 0.09 + 0.09 i 0.04 * 0.09 f 0.13

Lactate dehydrogenase (5)

Relative specific activity (mean f S.E.) is defined as the specific activity of the fraction relative to that of homogenate-particulate. The numbers in parentheses refer to the number of separate experiments in which activities were determined. In the case of acylation of lysophosphatidyfcholiine with paImitate as substrata, the values from two separate experiments are shown. Actual mean specific activities (nmollmin Per mg Protein i: S.E.) in the homogenate-particulate were: AcyIation of lysophosphatidylcholine with oieate and p&&ate, 0.76 i 0.07 (xl), 0.69 (2) respectively: NADPH-cytochrame c reductase, 6.73 * 0.45 (13); acetyleholinesterase, 75.3 + 6.69 (6): (Na+ + K+>ATPase. 106 I 28.0 (4); succinate dehydrogenase, 19.5 i. 3.30 (5);Iactate dehydrogenase. 536 t 59 (5). Protein in homogenate-particulate was 78.8 + 2.7 (14) mg/g wet wt. Homogenate-particulate consisted of material that sedimented after centrifugation of the total homogenate at 105 000 X g for 60 min.

THE SUBCELLULAR BRAL CORTEX

TABLE I

237 TABLE II SUBCELLULAR DISTRIBUTIONS OF LYSOPHOSPHATIDYLCHOLINE:ACYLTRANSFERANSFERASE. OLEOYL- AND PALMITOYL-CoA SYNTHETASE, OLEOYL-CoA AND PALMITOYL-CoA HYDROLASE IN GUINEA-PIG CEREBRAL CORTEX Actual mean specific activities (nmol/min per mg protein + S.E.) in the homogenate-particulate were as follows: lysophosphatidyl/acyltransferase. 1.28 f 036 (5); oleoyl-CoA synthetase, 2.98 + 0.41 (5); palmitoyl-CoA synthetase. 2.88 t 0.64 (4); oleoyl-CoA hydrolase, 10.94 f 1.08 (5): palmitoyl-CoA hydrolase. 8.79 + 0.88 (3). The relative specific activity is defined as the specific activity of the fraction relative to that of homogenate-particulate. Relative specific activity Lysophosphatidylcholine/oleoyl-CoA acyltransferase

Oleoyl-CoA synthetase

Palmitoyl-CoA synthetase

Oleoyl-CoA hydrolase

Palmitoyl-CoA hydrolase

(5)

(4)

(4-5)

(3)

(5) Pl

P2A P2 B, P2C P3 SPM S3 % Recovery

1.10 0.19 0.68 0.63 1.90 0.69 0.14 81.7

+ 0.07 r 0.07 + 0.07 f 0.02 +_0.12 f 0.03 + 0.04 + 5.6

0.87 0.35 0.69 1.16 1.32 0.48 0.41 68.0

+ 0.08 + 0.03 + 0.08 + 0.09 +_0.11 + 0.08 + 0.08 r 7.0

0.76 0.33 0.67 1.16 1.46 0.47 0.27 66.8

+ 0.03 + 0.03 + 0.09 + 0.12 +_0.16 * 0.05 * 0.03 + 4.0

0.94 0.63 0.87 1.00 1.43 0.29 3.50 76.3

* + + f f c +

0.12 0.07 0.13 0.13 0.10 0.11 0.50

f 6.7

0.95 0.52 0.83 1.02 1.47 0.43 3.86 75.6

? * + + t f c

0.09 0.10 0.09 0.09 0.16 0.08 0.60

f 2.0

Thompson [ 401 also concluded that both lysophosphatidylcholine/acyltransferase and lysophosphatidylinositol/acyltransferase activities cofractionated with NADPH-cytochrome c reductase in subcellular and subsynaptosomal fractions isolated from rat brain. Thus both acylation and acyltransferase activities, in subcellular fractions other than microsomes, may be accounted for by contamination of these fractions with endoplasmic reticulum. Acyl-CoA synthetase activity, obtained using palmitate or oleate as substrates, was most enriched in the mitochondrial and microsomal fractions (Table II). The bimodal distribution and enrichment of enzyme activity in these fractions is similar to that obtained previously with other tissues [41]. In comparison, synaptic plasma membranes contained little acyl-CoA synthetase activity. AcylCoA synthetase activity was correlated with NADPH-cytochrome c reductase activity (r = 0.88; P < 0.001) in all particulate fractions other than the mitochondria, indicating that brain mitochondria contain acyl-CoA synthetase activity independent of the endoplasmic reticulum. Oleoyl-CoA- and palmitoyl-CoA hydrolase activities were concentrated in the soluble fraction but, in addition, considerable activity was detected in the microsomal fraction (Table II). The relative specific activities of oleoyl- and palmitoyl-CoA hydrolase in all the subcellular fractions correlated well with those of lactate dehydrogenase (r = 0.83 and 0.82, respectively; P < 0.001). Thus acyl-CoA hydrolase activity in fractions other than the soluble and nerve-ending fraction (which contains occluded cytoplasm) may reflect the adsorption of soluble fraction components to membrane fragments during homogenization and subsequent tissue fractionation. The codistribution of acyl-CoA hydrolase and lactate dehydrogenase supports the view that acyl-CoA hydrolase is predominantly, if not exclusively, a cytoplasmic enzyme 130,421. The specific activities of both

238

oleoyl-CoA- and palmitoyl-CoA hydrolase exceeded those of acyl-CoA synthetase and acyltransferase by at least &fold in all particulate fractions. However, hydrolase activity was essentially abolished by the inclusion of bovine serum albumin at a concentration (5 mg/ml) used routinely in the determination of acyl-CoA synthetase and acyltransferase activities. Thus, it is unlikely that the observed subcellular distribution of either acyl-CoA synthetase or acyltransferase was significantly altered by acyl-CoA hydrolase activity. Acyl-CoA synthetase and lysophosphatidylcholine plasma membranes and mitochondria

: acyltransferase

in synaptic

In order to investigate more closely the possibility that synaptic plasma membranes or mitochondria contain synthetase or acyltransferase activities independent of microsomal contamination, a series of experiments was conducted in which synaptic plasma membranes or mitochondria were mixed with increasing amounts of microsomes. Enzyme activity (ordinate) was then measured and plotted as a function of NADPH-cytochrome c reductase activity. If all activity present is attributable to microsomal contamination, the extrapolated line passes through the origin. If activity, independent of microsomes exists, then the extrapolated line bisects the ordinate above zero [39]. Although differences existed in the specific activities of acyl-CoA synthetase and lysophosphatidylcholine/acyltransferase obtained in the three separate experiments performed, the conclusion from each experiment was essentially the same. Thus, as expected, there was an acyl-CoA synthetase activity detected in the mitochondrial fraction that was independent of the presence of

2 NADPH-cytochrome

c

A

6

reductore

(n mol/min)

Fig. 1. The effect of additions of microsomes on acyl-CoA synthetase activity in synaptic plasma membranes (A) or mitochondria (B). O-, ~-0; oleoyl-CoA formation by synaptic plasma membranes and mitochondria respectively: l-, nB; palmitoyl-CoA formation by synaptic plasma membranes and mitochondria respectively. Acul-CoA synthetase activity was determined either in synaptic plasma membrane or mitochondrial fractions alone, or mixed with increasing amounts of microsomal protein. Results are plotted as a function of total NADPH-cytochrome c reductase activity present, calculated from the specific activities of NADPH-cytochrome c reductase in synaptic plasma membranes, mitochondria and microsomes. The first point on each graph represents the activity obtained with synaptic plasma membranes or mitochondria alone. Results from three separate experiments are shown. with slopes and intercepts calculated by linear regression analysis.

239

NADPH-cytochrome c reductase (Fig. 1). In contrast, the synaptic plasma membranes did not contain acyl-CoA synthetase activity other than that due to contaminating endoplasmic reticulum. In the case of the acyltransferase, there was no consistent indication that the mitochondrial or synaptic plasma membrane fractions possessed significant amounts of enzyme activity independent of microsomes (Fig. 2). Role of the endoplasmic

reticulum

in brain as a site for phospholipid

synthesis

Conclusions regarding the intracellular distribution of enzyme activities in this study clearly depend on the assumption that NADPH-cytochrome c reductase in brain is exclusively located in the endoplasmic reticulum, as is well documented [39,43,44]. It then follows that the endoplasmic reticulum contains most if not all of the acylation and acyltransferase activities. Synaptic plasma membranes, which contain little or no acyltransferase, also contain little or no acyl-CoA synthetase. They are thus deficient in the two enzymes necessary for the acylation of lysophosphatidylcholine by free fatty acids. Mitochondria contain acyl-CoA synthetase but no acyltransferase and therefore lack the ability to acylate lysophosphatidylcholine. Only the endoplasmic reticulum contains both the enzymes necessary for acylation. Since both lysophosphatidylcholine and unesterified fatty acids are transported into the brain [ 45,461, the acylation reaction by brain endoplasmic reticulum using substrates from plasma could be a source of cerebral choline phosphoglyceride. The endoplasmic reticulum in brain has also been cited as the major, if not sole, subcellular location of both de novo synthesis of choline phosphoglyceride and the Ca”-catalyzed base-exchange reaction [ 39,471. Isolated synaptosomes may also be capable of phospholipid synthesis since

0 NADPH-cytochrome

2 c

4

6

reduciare

(n mol/min)

Fig. 2. The effect of additions of microsomes on lysophosphatidylcholinelacyltransferase activity in synaptic plasma membranes (A) and mitochondria (B). OF, 0-O; formation of radioactive choline phosphoglyceride by synaptic plasma membrane and mitochondria respectively. See legend to Fig. 1. for experimental details. Results from three separate experiments are shown with slopes and intercepts calculated by linear regression analysis.

240

they contain occluded smooth endoplasmic reticulum [ 39,441. Thus, latent CDPcholine:1,2diacylglycerol cholinephosphotransferase (EC 2.7.8.2) activity in nerve endings becomes detectable when the nerve endings are osmotically lysed [48]. While both acylation and acyltransferase activities have also been measured in synaptosome preparations from mouse brain [ 151, the results from the present study indicate that this acylation activity is attributable to the endoplasmic reticulum rather than synaptic plasma membrane or intraterminal mitochondria. Choline phosphoglyceride that is synthesized in the endoplasmic reticulum by acylation, de novo synthesis or base-exchange may be transported to other subcellular sites such as synaptic plasma membranes or mitochondria by means of phospholipid transfer proteins, located in the soluble cytoplasm

[@I * Acknowledgments We are grateful to Miss Susan Hinks for skilled technical assistance and to Dr. C.W. Wharton for advice on statistical interpretations. This work was supported by a Project Grant from the Medical Research Council of Great Britain. References 1

Holub,

2

Webster,

B.J.

G.R.

and

and

Alpern,

R.J.

3

Murphy,

M.G.

and

Spence,

M.W.

4

Baker,

R.R.

5

Eibl.

6

Yousef,

H..

7

Waite,

8

Stein.

Y.,

9

Stahl,

W.L.

Kuksis,

and

Hill,

M..

and

Fisher,

S&on, and

Victoria,

E.J.

11

Stadler,

J. and

12

Mulder,

E.,

Lands,

and

Franke,

van

den

16,

J. Biol. (1969)

Eur.

J. and

Holub,

E. (1970)

0.

E.D.

(1968)

W.W.

J. Cell

van

B.J. Biol.

Res.

Biochim. and

7060-7065

J. Biochem.

Lipids

12,

140-144

9. 746-753

39,

185-192

Acta

16,

163,

459-471

5440

Biophys. Deenen,

9, 250-258

(1977)

Biophys.

J. Lipid

(1973)

J.W.O.

251-259

248,

Biochemistry

Biochim.

(1975)

35-42 29,

Chem.

W.E.M.

(1968)

l-125

J. 90,

J. Neurochem.

(1973)

Stein,

Berg,

Res.

Piekarski,

E.G.

Kom,

Lipid

Biochem.

(1977)

Blackwell,

C. and

Trams,

Adv. (1964)

W.

M.M.,

P. and

Widnell.

10

(1978)

Thompson,

E.E.

I.M.,

A.

Acta

311,

L.L.M.

205-213

(1965)

Biochim.

Biophys.

Acta

106.

127 13

Ferber,

14

Warrendorf,

E. and

15

Corbin, Fisher,

17

Sanchez,

S.K.

18

Dole,

and

Sun,

and

19

Pullman,

20

Ullman.

M.D.

21

Seubert,

W.

M.,

V.P.

M.E.

E.G.

(1973)

Gray, Sun,

24

Cotman.

C.W.

and

25

Webster,

G.R.

(1967)

Bligh,

27

Skipski,

E.G.

28

Lloyd-Davies,

29

Bar-Tane.

30

Anderson,

and

V.P.,

31

Sottocasa,

32

Porteous,

33

Wroblewski.

G.Y.

Prep.

Biochem.

Dyer,

W.J.

R.F.

K.A.

Brindley,

A.D. G.L.,

and

and

G. and

F. and

Clark, LaDue,

Trans.

(1973)

49,

919-925

77-82 6, 101

Biochem.

J. 132,

697-706

188-198 Res.

J. Anat. T.

13.

96,

(1971)

422-423

79-88

J. Neurochem.

Biochim.

Can.

18,

Biophys.

J. Biochem.

Barclay. D.N.

V.G.

M.

(1975)

B. (1971) (1971)

B.,

B. (1965) J.S.

335-349

7, 80-83

(1971)

Shapiro,

Erwin.

296,

1711-1718

Acta

249,

380-394

J. 102,373-380

and

Kuylenstierna, and

54,

(1962)

(1959)

Peterson,

30.

Sot.

D.N.

J. Lipid

Samorajski, D.A.

Acta

J. Biochem.

150-154

(1972)

V.P.

and

35,

Biochem.

N.S.

Matthews,

J., Rose.

J.W.

Biochem.

Brindley,

Invest.

Whittaker,

Sun.

Can.

(1978)

Biochem.

22

Biophys.

D. (1971)

J. Neurochem.

and

Anal.

Radin.

(1960)

Biochim.

(1978)

C.E. D.G.

J. Clin.

23

26

G.Y.

Rowe.

and

and

(1973)

Rubinstein.

Nicholls,

(1956)

A.Y.,

K.

and

D.R.

16

Resch.

E.M.

Biochem. Proc.

Biochem. Biochem. L. and

J. 122. 18,

911417 J. 90,

J. 152,

Bergstrand,

J. 96. Sot.

37.

Biochem.

J. Neurochem.

Emster,

(1955)

Physiol.

(1964)

374-378

39-49 353-362

1179-1186 A.

(1967)

J. Cell

159-171

Exp.

Biol.

Med.

90.

210-213

Biol.

32.

415438

118-

241 34

Ellman. 88-9

G.L.,

Courtney.

K.D..

Andres,

V.

and

Featherstone.

R.M.

35

Austin,

36

Sun,

37

Baginski,

38

Geiger.

L. and

A.Y.

Foa,

P.J. and

Miller. Pik,

E.K.

41

Groat.

42

Knauer.

43

Possmayer,

44

Possmayer,

J.R.

Dawson.

(1979) F.,

P.G.R.

45

Illingworth,

46

Dhopeshwarkar, 255.

and

B. (1967)

L.,

Biochem.

Pharmacol.

7,

O.W.

13,

49, 56.

(1976)

326---332

467-473

J. 126,

J. Biochem. W.C.

1365-1372

805-821 765-768 Adv.

Lipid

14,

75-126

J.B. G.,

(1973)

32,

J. 132. P.J.,

381-394

Won&

T.,

MacPherson.

C.F.C.

and

889-906

(1972)

Subramanian,

Biochem.

Stewart-DeHaan,

C..

Biochem. McConnell,

J. 130, D.H.

557-567

and

Mead,

J.F.

(1972)

Biochim.

Biophys.

572-579

47

Dawson,

48

Strosznajder,

R.M.C.

(1973) J.,

Sub-Cell.

Biochem.

Radominska-Pyrek.

A.

2, 69-89 and

Horrocks,

L.A.

48-56 Miller,

Res.

515-523

Mudd, Duwe,

17. Chem.

Biochkm.

Hulsmann.

J. 179,

Portman,

ixP

Biochem.

Can.

J. Neurochem.

G.A..

Clin.

Anal.

(1972)

(1978) and

J. 53.

J. Neurochem.

(1972)

B. and

Kleine,

D.R.

S.P.

H.R.

(1979)

Biochem.

Zak,

Biochem.

Meiners.

F.,

and

W.

Scholte,

T.E.

(1953) T. (1970)

R.M.C.

Thompson,

P.H.E.,

Hardin,

P.P.

Bessman,

and

and

W.K.

Samorajski,

ES.,

39

Acta

Berry,

and

40

49

(1961)

5

E.K.

and

Dawson,

R.M.C.

(1972)

Biochem.

J. 126,

823-835

(1979)

Biochim.

Biophys.

Acta

574,