Low Density Lipoprotein-activated Lysolecithin Acylation by Human Plasma Lecithin-Cholesterol Acyltransferase

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 255, No. 19, Issue of October 10, pp ,1980 Printed in U.S.A. Low Density Lipoprotein-activated Lysolecithin Acylation by Human Plasma Lecithin-Cholesterol Acyltransferase IDENTITY OF LYSOLECITHIN ACYLTRANSFERASE AND LECITHIN-CHOLESTEROL ACYLTRANSFERASE* (Received for publication, April 4, 1980, and in revised form, June 23, 1980) Papasani V. SubbaiahS, John J. Albersg, Ching Hong Chen, and John D. Bagdade From the Department of Medicine, University of Washington, Providence Medical Center, and Harborview Medical Center, Seattle, Washington There is in normal plasma an enzyme activity which formation, as does lecithin-cholesterol acyltransferase, we converts labeled lysolecithin to lecithin by an energy- have undertaken systematic studies to determine if this enindependent low density lipoprotein-activated path- zyme activity is due to lecithin-cholesterol acyltransferase way. Studies were undertaken to compare the identity itself. Our results indicate that the acylation of both lysoleciof this enzyme with lecithin-cholesterol acyltransfer- thin and cholesterol are carried out by the same protein in ase. During purification of the enzyme by ultracentrif- human plasma; whereas the acylation of cholesterol is actiugation and by chromatography on high density lipo- vated by apolipoprotein A-I, the acylation of lysolecithin is protein affinity column, DEAE-Sepharose column, and activated by LDL. hydroxylapatite column, both the lysolecithin acyltransferase activity andthe lecithin-cholesterol acyl MATERIALS AND METHODS transferase activity were found in the same fractions and were enriched to the samextent at each step. The Substrates-Egg lecithin, soy lecithin, and human serum albumin final purified preparation which had 16,000- to 24,000- (essentially fatty acid-free) were obtained from Sigma Chemical Co. fold higher specific activities than starting plasma gave [4-4C]Cholesterol (specific activity, 54.0 mci/mmol) was purchased a single protein band on polyacrylamide gel electro- from New England Nuclear Co. Lecithin-cholesterol liposomes were phoresis and this single band contained both the activ- prepared as described before (5). [ H]Lysolecithin was prepared by ities. Also, the effects of ph, heat, and chemical inhibi- the action of snake venom phospholipase A on [ H]glycerol-labeled tors on the enzyme activities were similar. Plasma from lecithin, which in turn was prepared from germinating soybeans (6) patients with familial lecithin-cholesterol acyltransfer- by the following modifications: 50 g of soybeans (W. Atlee Burpee Co., Riverside, Calif.) were washed in running tap water in the ase deficiency also lacked lysolecithin acyltransferase presence of small amounts of bleach for 30 min. They were then activity. These results indicate that a single enzyme carries out both lecithin-cholesterol acyltransferase and lysolecithin acyltransferase activities. The purified enzyme required apolipoprotein A-I for lecithin-cholesterol acyltransferase activity, but required low density lipoprotein for lysolecithin acyltransferase activity. Large amounts of lysolecithin are produced continuously in normal human plasma by the actions of lecithin-cholesterol acyltransferase (EC ) (1) and phospholipase A (EC ) (2). However, the further metabolism of this potentially cytolytic product in the plasma is not well studied. We have recently demonstrated a lysolecithin acyltransferase activity in normal human plasma which converts lysolecithin to lecithin by an energy-independent pathway (3). This enzyme is associated with the high density lioprotein and is activated by low density lipoprotein (4). Because the enzyme associates with HDL and transfers a fatty acid independent of acyl-coa * This research was supported in part by Grants HL , HL , and HL of National Institutes of Health and by Contract NIHV A from Lipid Metabolism Branch, National Heart, Lung, and Blood Institute. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ To whom correspondence should be addressed at the University of Washington, Providence Medical Center. f Established Investigator of the American Heart Association. The abbreviations used are: HDL, high density lipoproteins; LDL, low density lipoproteins; apo A-I, apolipoprotein A-I allowed to absorb 5 mci of [2- H]glycerol (ICN Pharmaceuticals, Imine, Calif.) (specific activity, 250 mci/mmol) dissolved in 5 ml of boiled water. The seeds were then incubated in the dark at 25 C for 3 days. The germinated seeds were homogenized in a blender with 100 ml of cold water (4 C) and the lipids were extracted by the procedure of Bligh and Dyer (7). The chloroform extract was evaporated to dryness, redissolved in 30 ml of chloroform, and loaded onto a neutral alumina column (250 g) which was prepared in chloroform. The neutral lipids were eluted with 1.5 liters of chloroform followed by the elution of lecithin with 3 liters of ch1oroform:methanol (1:1, v/ v) mixture. Fractions of 250 ml were collected and those containing lecithin were pooled and concentrated. The lecithin was further purified on preparative thin layer chromatographic plates (0.5 mm of Silica Gel G; E. Merck, Darmstadt, West Germany) using the solvent system of ch1oroform:methanol:water (65:25:4 by volume). The bands corresponding to lecithin were eluted using Bligh and Dyer procedure (7). About 175 pci of [ H]glycerol-labeled chromatographically pure lecithin with a specific activity of 1.9 pci/pmol were obtained by this procedure. The lecithin preparation was dissolved in 20 ml of ethyl ether and incubated at room temperature for 24 h with 5 mg of crude Crotalus adamanteus venom (Sigma Chemical Co.) dissolved in 0.1 ml of 10 mm CaC12 in 0.1 M Tris-C1, ph 7.4. The lysolecithin produced was purified on Silica Gel G thin layer plates. Alkaline hydrolysis of the lysolecithin prepared revealed that over 97% of the label was in the glycerol portion. The labeled lysolecithin was diluted with unlabeled soy lysolecithin, prepared by the action of snake venom phos- pholipase A on unlabeled soy lecithin, to give the desired specific activity. [A~yl-~~C]lysolecithin was prepared in a similar manner from the I4C-labeled lecithin. The latter was prepared by incubating the germinating soybeans in presence of sodium [l-i4c]acetate(new England Nuclear, Boston, Mass.) (specific activity, 59 mci/mmol). More than 95% of the label was found to be in the fatty acids after the alkaline hydrolysis of [14C]lysolecithin. Apo A-Z and LDL-Apolipoprotein A-I was prepared as described by Cheung and Albers (8). For LDL preparation, outdated plasma

2 Lysolecithin Lecithin-Cholesterol Acylation by Acyltransferase obtained from the Puget Sound Blood Center was fist centrifuged at a density of1.019 g/ml (KBr) at 100,000 X g for 18 h and the infranatant was brought to density with KBr and centrifuged at the same speed for 20 h. The floated LDL was removed by tube slicing and recentrifuged at the same density to remove any heavier density components. The LDL preparation was then dialyzed extensively against 0.15 M NaCI-1 mm EDTA, ph 7.4, and concentrated by ultrafitration using Centriflo membrane filters (CF-50A, Amicon Corp., Lexington, Mass.) to give a protein concentration of approximately 4 mg/ml. The LDL preparations contained small amounts of apolipoprotein A-I (10 to 20 pg of A-I/mg of B protein) as determined by the immunodiffusion method (8). However, this LDL preparation did not activate esterification of cholesterol by purified lecithin-cholesterol acyltransferase but actually inhibited the apo A-I-activated fractional esterification rate by 60 to 8076, probably due to dilution of the labeled cholesterol in the substrate. Lecithin-Cholesterol Acyltransferase Assay-This activity was determined from the conversion of [4-4C]cholesterol to labeled cholesterol ester in presence of egg lecithin, apo A-I, and serum albumin as described by Albers et al. (5). The reaction mixture contained 15 nmol of unlabeled cholesterol, 5 nmol of [ 4C]cholesterol, 80 nmol of lecithin, and 20 pg of apo A-I, 0.5% of human serum albumin, and 5 mm mercaptoethanol in a final volume of 2.0 ml. Solutions of unlabeled cholesterol (0.237 mg), egg lecithin (2.48 mg), and [4-4C]cholesterol (0.072 mg, 10 pci) were mixed and evaporated to dryness in a glass vial and redissolved in 1 ml of absolute ethanol. Of this substrate solution, 25 pl were used for each assay. The reactions were carried out in screw-cap culture tubes (16 X 125 mm) in which the reagents were added in the following order: 1.25 ml of Tris buffer (10 mm Tris-C1, 1 mm EDTA, 140 m~ NaC1, ph 7.4), 25 p1of apo A-I solution (0.8 mg/ml in Tris buffer), 0.5 ml of 2% human serum albumin in Tris buffer, and 25 p1 of the substrate solution. The reagents were mixed immediately in a vortex mixer, kept at 4 C for 10 min, and 0.1 ml of 0.1 M mercaptoethanol in Tris buffer was added, followed by 0.1 ml of enzyme preparation. The final reaction mixture (2.0 ml) was incubated at 37 C for 1 h and the reactions were stopped by the addition of 2 ml of ethanol and extracted with 5 ml of hexane containing 50 pg each of unlabeled cholesterol and cholesterol linoleate. The extraction was repeated with 3 ml of hexane and the pooled extracts were evaporated to dryness, redissolved in 0.5 ml of chloroform, and applied to a 0.5-mm layer of Silica Gel H (E. Merck). The cholesterol and cholesteryl ester were separated with the solvent system, petroleum etherethyl acetate (85:15, v/v) and the radioactivity in each lipid region was determined in a Beckman LS 7000 liquid scintillation counter. Control reactions containing no enzyme were run simultaneously to correct for nonenzymatic reaction. The enzyme activity is expressed as nanomoles of labeled cholesterol esterified/h/ mg of protein. Lysolecithin Acyltransferase Assay-This activity was determined from the conversion of [3H]glycerol-labeled lysolecithin to lecithin in presence of LDL. [ H]Lysolecithin(0.2 pmol, 70,000 dpm) was dispersed in 0.1 ml of 0.1 M Tris-C1, ph 7.4, containing 0.4 mm dithiothreitol, by vortexing for 15 s. LDL preparation (d to 1.063) from normal human plasma containing 0.4 mg of protein in 0.1 ml was then added, followedby the enzyme (0.1 ml to 0.2 ml). The incubations were carried out in a final volume of 0.4 ml at 37 C for 2 h in a water bath with mechanical shaking. The reaction was stopped by the addition of 1.0 ml of methanol and the lipids were extracted by the procedure of Bligh and Dyer (7). Lecithin and lysolecithin were separated on Silica Gel G thin layer plates with the solvent system, chloroform:methanol:water (65:25:4,by volume) and the separated spots visualized by exposure to iodine vapors were counted in a Beckman LS 7000 liquid scintillation counter with 10 ml of toluenebased scintillation fluid containing 0.5% 2, 5-diphenyloxazole, 0.005% 1,4-bis[2-(5-phenyloxazolyl) ]benzene, 7.5% methanol, and 3% acetic acid. Control reactions (complete reaction mixture without enzyme) were run simultaneously to correct for any nonenzymatic reaction. The enzyme activity is expressed as nanomoles of [3H]lysolecithin acylated to lecithin/h/mg of protein. Purification of the Enzyme-Freshly prepared normal human plasma (in EDTA) was used as the enzyme source. The purification of the enzyme having both activities was done essentially as described for lecithin-cholesterol acyl transferase (5). Other Methods-Protein was determined by the procedure of Lowry et al. (9) using bovine serum albumin as standard. Electrophoresis on7.5% polyacrylamide gels, ph 8.6,was performed by the method of Davis (lo), in the absence of urea. RESULTS Co-purification of Lysolecithin Acyltransferase and Lecithin-Cholesterol Acyltransferase Actiuities-Initial experiments with purified lecithin-cholesterol acyltransferase preparation also showed the presence of lysolecithin acyl transferase activity. Therefore, the procedure described for the purification of lecithin-cholesterol acyltransferase (5) was followed and at each step the fractions were assayed for both lysolecithin acyltransferase and lecithin-cholesterol acyltransferase activities. The purification procedure involved ultracentrifugation to obtain a d 1.21 to 1.25 fraction, and serial chromatography on an HDL affinity column, a DEAE-Sepharose column, and a hydroxylapatite column. The elution profiles ( Froctton number FIG. 1. Co-purification of lecithin-cholesterol acyltransferase and lysolecithinacyltransferase activities during HDL affinitycolumnchromatography. Normal human plasma was obtained from freshly drawn blood (in EDTA) from healthy male volunteers. The plasma was fist subjected to ultracentrifugation at a KBr density of 1.25 g/ml for 26 h at 100,000 X g. The lipoproteins were separated by tube-slicing and diluted to a density of 1.21 g/ml and subjected to centrifugation again at 100,000 X g for 44 h. The middle clear layer from each tube was taken by tube slicing, dialyzed against 10 mm Tris, 1 mm EDTA, and 150 mm NaC1, ph 7.4, and then subjected to chromatography on HDLa-Sepharose column (5 X 20 cm), as described by Albers et al. (5). The enzyme activities were eluted with 0.5 mm sodium taurocholate in water. Each fraction was then dialyzed against the Tris-EDTA buffer and 0.1 ml from each was tested for both activities as described in the text. W, lysolecithin acyltransferase activity; M, lecithin-cholesterol acyltransferase activity; E- - I, absorbance at 280 nm. Enzyme activities expressed as nanomoles/h/ml of each fraction Fractlon number FIG. 2. Chromatography of plasma lecithin-cholesterol acyltransferase and lysolecithin acyltransferase activities on DEAE-Sepharose column. The pooled eluate from HDL affinity column was dialyzed against a buffer containing 1 m~ Tris-C1.5 mm EDTA, and 25 m~ NaCI, ph 7.4, and loaded onto a DEAE-Sepharose column (1.2 X 20 cm). The enzyme activities were eluted with a NaCl gradient of 25 to 200 m~ NaCl in Tris-EDTA, ph 7.4. Each fraction was dialyzed against the 10 m~ Tris, 1 m~ EDTA, and 140 m~ NaC1, ph 7.4, and 0.1-ml aliquots were assayed for enzyme activities. O 0, lysolecithin acyltransferase activity; W, lecithin-cholesterol acyltransferase activity; U - U, absorbance at 280 nm;, NaCl gradient (mm).

3 Lysolecithin Acylation by Lecithin-Cholestwol Acyltransferase of the two activities at each chromatographic step of one isolation are shown in Figs. 1 to 3. Bothactivities were recovered in thesamefractions in eachchromatographic procedure for six different preparations. The second peak of activity shown in the DEAE-Sepharose chromatography step was not observed in all preparations (Fig. 2). There was no difference in thesubsequentchromatographicbehavior of enz-yme activities in this peak, when compared with those in the main peak. The proteins which were not bound to the columns did not contain either activity. In Table I the specific activities of the pools from each step for lysolecithin acyltransferase and lecithin-cholesterol acyltransferase activities are compared. Both the activities were IO 26 I FractNon number FIG.3. Hydroxylapatite column chromatography of pooled active fractions from DEAE-Sepharose column.all fractions of DEAE-Sepharose column containing lecithin-cholesterol acyltransferase and lysolecithin acyltransferaseactivities werepooled and dialyzed against 150 mm NaCI, 4 m~ sodium phosphate, ph 6.9, and loaded onto the hydroxylapatite column (2.5 X 10 cm) prepared in the same buffer. The enzyme activities were eluted witha linear gradient of 15 mm to 60 m~ sodium phosphate, ph 6.9, in 150 mm NaC1. Each fraction was dialyzed against 10 mm Tris, 1 mm EDTA, and 140 mm NaCI, ph 7.4, and assayed for both the enzyme activities. M, lysolecithin acyltransferase activity; M, lecithin-cholesterol acyltransferase activity; E- - -0, absorbance at 280 nm; - - -, PO, gradient (mm). TABLE I Co-purification of lecithin cholesterol acyltransferase a n d Lysolecithin Acyltransferase actioities of normal human plasma Normal human plasma was obtained from freshly drawn blood (in EDTA) from healthy male volunteers. The plasma was subjected to ultracentrifugation a t a KBr density of 1.25 g/ml for 26 h at 100,OOO X g.the lipoproteins were separated by tube slicing and diluted toa density of 1.21 g/ml and subjected to centrifugation for 44 h a t 100,OOO X g. The middle clear layer from each tube was then subjected to various chromatographic procedures asdescribed by Albers et al. (5). At eachstep,the pooled fractionscontaining lecithin-cholesterol acyltransferase activity were dialyzed against a buffer containing 1 mm Tris, 5 mm EDTA, and 25 mm NaCI, ph 7.4. and tested for both enzyme activities as described in the text. Lysolecithin acyltransferase activity Fraction Specific activity" Whole plasma d 1.21 to 1.25 fraction HDL affinity column eluate DEAE-Sepharose column eluate Hydroxylapatite column eluate ,941 Purificatlon -fold Lecithin-Cholesterol acyltransferaseratio of activ~ty specific activipurifispecific tles cation Activity*.fold ,882 1, lfi Nanomoles of lysolecithin acylated/h/mg of protein. acylated/h/mg of protein. * Nanomoles of cholesterol FIG. 4. Polyacrylamide gel electrophoresis of the purified enzyme preparation. The activefractions from hydroxylapatite chromatography were pooled andconcentrated by ultrafiltration through dialysis bags under vacuum. The concentrated sample (50 pg of protein) was then subjected to electrophoresis on 7.5$ polyacrylamide gels ( I O ) containing 0.1 mm rnercaptoethanol, butno denaturing agent. The electrophoresis was done at 4 C for 2 h and the gels were immediately sliced into 3-mm sections at 4OC. Each slice was cut longitudinally and one-half used for lecithin-cholesterol acyltransferase assay. while the other half was used for lysolecithin acyltransferase assay. The slice was dropped into the complete reactionmixture and was crushed with the help of a glass rod and incubatedat 37 C for 2h. Duplicate gel was stained with 0.04q Coomassie brilliant blue. Slice 1 is the top of the separating gel and lecithin-cholesterol acyltransthe tracking dye is in slice 20.M, lysolecithin acyltransferase activity. ferase activity; M, enriched to asimilar extent at each step except after the DEAE-Sepharosechromatographywherethe lysolecithin acyl transferase activity was found to be consistently lower than lecithin-cholesterol acyltransferase activity. This was apparently due tosome inhibition of lysolecithin acyltransferase activity and not due toloss of enzyme protein because at the next step (hydroxylapatite column) the lysolecithin acyltransferase activity was comparable to thatof lecithin-cholesterol acyltransferase. Similar results were obtained with five other preparations. The ratio of activities did not change appreciably from the whole plasma during the purification, except at the DEAE-Sepharose step. These results thus indicate that both activities are co-purified even up to an enrichment of 16,000- to 24,000-fold. The recovery of activities was 5 to 10% from the starting plasma in various preparations. Assay of Activities in Polyacrylamide Gels-The pooled eluate from thehydroxylapatite column was subjectedto electrophoresis on 7.5%polyacrylamide gels at 4'C in presence of 0.1 mm mercaptoethanol and in the absence of denaturing agent (10). As reportedearlier (5) thepreparation gave a single protein band when stained with 0.04% Coomassie brilliant blue (Fig. 4). A duplicate gel was sliced into 3-mmslices immediately after electrophoresis and eachslice then cut into two halves longitudinally. One-half of each slice was used for assay of lecithin-cholesterol acyltransferase and the other half for lysolecithin acyltransferase activity. The slices were directly added to complete reaction mixtures containinglabeled substrate and co-factors and crushed with a glass rod in the reaction mixture. Incubations were carried out at 37 C for 2 h. Both activities were found in the same slice which also corresponded to the stained protein band (Fig.4). Similar results were obtained with three other preparations. These results thus show that both the enzyme activities are due to a single protein. Effects o f p H a n d H e aon t the Two Activities-In order to further substantiate thatlysolecithin acvltransferase and lec-

4 9278 Lysolecithin Acylation by Lecithin- Cholesterol Acyltransferase ithin-cholesterol acyl transferase activities are mediated by the same protein, we studied the effects of several physical and chemical agents on the enzyme activities of the purified preparation. The effect of ph on both the reactions was the same (Fig. 5), the optimum activities occurring at ph 8.0 in Britton and Robinson type universal buffer (11). This ph value for lecithin-cholesterol acyltransferase activity corresponds to the value reported by Aron et al. (12) for their purified preparation. With the whole plasma and partly purified preparation (d 1.21 to 1.25 plasma fraction) the ph optimum for both activities was approximately 7.4 (results not shown). When the purified enzyme preparation was kept for 20 min at various temperatures (Fig. 6, left), there was no loss of either activity up to 30 C but both activities were rapidly lost above 37 C. No significant difference was noted in the heat TABLE I1 Assay of lecithin synthesis in lecithin-cholesterol acyltransferasedeficient plasma For Experiment I a sample of plasma prepared in acid-citrate dextrose from a patient with familial lecithin-cholesterol acyltransferase deficiency was obtained from Dm. John Glomset and Kaare Norum. The sample was kept at 4 C until the assay which was performed within 4 days of plasma preparation. Control plasma from a normal female donor which was prepared in a similar manner was assayed simultaneously for lysolecithin acyltransferase activity using 0.2-ml aliquots. In Experiment I1 control and lecithin-cholesterol acyltransferase-deficient plasma samples were obtained from Dr. J. Frohlich and assays were carried out using 0.1-ml aliquots of plasma. The reaction mixtures contained plasma from the patient or control, 0.17 pmol of ["H]lysolecithin, 10 pmol of Tris-CI, ph 7.4, and, where indicated, LDL (1.019 to density) from pooled normal plasma at a concentration of 0.4 mg of protein/reaction mixture. Incubations (in duplicate) were carried out for 2 h at 37 C. lability of the two enzyme activities at any temperature. Their Addition Lecithin synthesis nmol/h/rnl plasma Experiment I: Plasma (0.2 ml) Control 6.85 Control LDL Patient (M. R.) 0.05 Patient (M. R) LDL 0.36 Control Patient's plasma ( ml) Experiment 11: Plasma donor F. D. (control) 8.89 D. H. (patient) 0.35 S. F. (patient) '0 i0 60 8'0 90 IO0 ph FIG. 5. Effect of ph on lecithin-cholesterol-acyltransferase and lysolecithin acyltransferase. The pooled fractions from the HDL affinity column were dialyzed against 10 mm Tris, 1 m~ EDTA, and 140 mm NaC1, ph 7.4, and 0.1 ml of the dialyzed sample was used for an enzyme assay. The assay of both enzyme activities was as described in the text, except that the buffer in the reaction mixture was replaced by the universal buffer (Britton and Robinson type) (11) at the indicated ph. W, lysolecithin acyltransferase; M, lecithin-cholesterol acyltransferase. M IO IO Temp "C (20 mln) Tlme of heatlng at 45"(n#m) FIG. 6. Effect of heat on acylation of lysolecithin and cholesterol by purified enzyme preparation. Left, effect of heating the purified enzyme preparation at various temperatures. Aliquots of the eluate from the hydroxylapatite column were dialyzed against 10 mm Tris, 1 mm EDTA, and 140 mm NaCI, ph 7.4, kept at the indicated temperatures for 20 min, cooled to room temperature, and immediately assayed for lecithin-cholesterol acyltransferase and lysolecithin acyltransferase activities at 37'C. The activity of the enzyme sample kept at 4 C was taken as 100% and the other values were calculated as percentages of this value. Right, effect of heating the purified enzyme at 45 C for various periods of time. A sample of enzyme preparation was preincubated at 45'C in a water bath. At the indicated time period, an aliquot was withdrawn, cooled to room temperature, and assayed for the enzyme activities at 37 C. The control preparation preincubated at 4OC was taken as 100% enzyme activity and all other values were expressed as percentages of this value. W, lysolecithin acyltransferase; , lecithin-cholesterol acyltransferase. inactivation profiies were also similar when the purified enzyme preparation was heated at 45 C for various periods of time (Fig. 6, right). The effect of some chemical inhibitors on both activities was also tested.p-hydroxymercuric benzoate (0.5 m ~), phenyl methyl sulfonyl fluoride (2.5 m ~), sodium deoxycholate (0.1 mm), and Zn'+ (10 mm) inhibited both activities by more than 90% (results not shown). Assay of Lysolecithin Acyltransferase in Lecithin-Cholesterol Acyltransferase-deficient Plasma-In further experiments to prove that the acylation of lysolecithin and cholesterol is carried out by the same enzyme, we estimated the lysolecithin acyltransferase activity in plasma from three patients with familial lecithin-cholesterol acyltransferase defi- ciency. In one experiment the lysolecithin acyltransferase activity was assayed in the whole plasma from a lecithincholesterol acyltransferase-deficient patient in the presence and absence of added LDL from normal human plasma. As seen in Table 11, there was no labeled lecithin formation in the patient's plasma even in presence of normal LDL. Addi- tion of the patient's plasma to the normal plasma did not result in any decrease of lysolecithin acylation, indicating that the lack of activity is not due to the presence of an inhibitor in the plasma of the patient. In another experiment lysolecithin acylation was assayed in plasma from one normal control and from two lecithin-cholesterol acyltransferase-deficient offspring of the control. These patients also lacked lysolecithin acyltransferase activity, further supporting the fact that lysolecithin acyltransferase activity is due to the same enzyme which carries out the esterification of cholesterol. Co-factor Requirements for Lysolecithin Acylation-Our previous results showed that lysolecithin acylation in normal human plasma required LDL for maximal activity (4). In further experiments to corroborate this observation, we found that the purified enzyme preparation also requires LDL for lysolecithin acylation (Table 111). Apo A-I, which is obligatory

5 Lysolecithin Acylation by Lecithin-Cholesterol Acyltransferase TABLE Co-factor requirement for lecithin synthesis by purified enzyme preparation Purified enzyme preparation containing 5 pg of protein (lecithincholesterol acyltransferase activity, 5.6 nmol of cholesterol esterified/ h) was used per each assay. LDL preparation (0.5 mg of protein/ assay) was prepared from normal pooled human plasma as described under Materials and Methods. Apolipoprotein A-I and rabbit antihuman A-I sera were prepared as described by Cheung and Albers (8). In experiments using - I&, - the LDL and lysolecithin were preincubated for 15 min at 37 C followed by incubation of this mixture with anti-a-i IgG or nonimmune IgG for 1 h at 37 C. The enzyme was added last and the incubations continued for 2 h at 37 C. Addition Lecithin synthesis Enzyme only (5 pg protein) + Apo A-I (20 pg) + LDL (0.5 mg) + LDL + apo A-I + LDL + IgG from rabbit anti-a-i sera + LDL + IgG from nonimmune rabbit sera LDL only + Apo A-I III nmol/h R This amount of IgG from anti-a-i sera inhibited 75% of the lecithin-cholesterol acyltransferase activity. TABLE Mechanism of lysolecithin acylation [ HIGlycerol-labeled lysolecithin and [ Clacyl-labeled lysolecithin were prepared from germinating soybeans as described in the text. The two preparations of lysolecithin were mixed to give a C: H isotope ratio of 1:5, and 0.2 pmol of the lysolecithin mixture used for each assay. The reaction mixture also contained 0.4 mg of LDL where indicated and either 1.0 mg of protein from d 1.21 to 1.25 fraction or 21 pg of protein from HDL affinity column eluate. Incubations were csrried out for 2 h at 37 C and the radioactivity in lecithin and lysolecithin determined in a Beckman LS-7C00 liquid scintillation system. Lysoleci- C:, H ratio Enzyme scmrce LDL thin acylated Lysolecithin Lecithin B:A (A) (B) nmol/z h d 1.21 to 1.25 fraction No d 1.21 to 1.25 fraction Yes HDL affinity column No 0.13s eluate HDL affinity column Yes eluate for the esterification of cholesterol and for phospholipase activity, could not stimulate lysolecithin acylation. Addition of LDL alone was sufficient for the lecithin formation, and addition of apo A-I with LDL did not stimulate the activity further. Treatment of LDL with 0.1 mg of IgG from rabbit anti-a-i sera did not have a significant effect on the LDLactivated lysolecithin acylation although it inhibited 75% of the A-I-activated cholesterol esterification (Table III). Mechanism of Lysolecithin Acylation-There are two known pathways in the biological systems for the acylation of lysolecithin to lecithin (13). One pathway involves the formation of an acyl-coa and the esterification of lysolecithin with the activated fatty acid, while the other pathway is energy-independent and involves a transesterification between two lysolecithin molecules, producing one molecule of lecithin and one molecule of glyceryl phosphorylcholine. The acylation of lysolecithin in human plasma does not require ATP and CoA (3). On the basis of this observation and the preliminary results from snake venom treatment of the lecithin synthesized from [acyl- %]lysolecithin, we postulated IV that the lecithin is formed by transesterification between two lysolecithin molecules. However, a more direct method of determining the reaction mechanism is now used, utilizing a doubly labeled lysolecithin as substrate. The lysolecithin contained 14C in the fatty acid side chain and 3H in the glycerol portion. A transesterification between two such lysolecithin molecules should result in a doubling of C: H ratio in the synthesized lecithin when compared to the ratio in lysolecithin. However, the results presented in Table IV show that the ratio of Y?H in lecithin is the same as in the substrate lysolecithin, indicating that the acyl group for lysolecithin acylation is coming from endogenous sources. Although it is not possible to identify the endogenous source of acyl group from the results, it is likely that one of the LDL lipids is acting as acyl donor in the reaction, because the amount of lipid present in the enzyme preparations used here is negligible and because there is little synthesis of lecithin in the absence of LDL. DISCUSSION The enzymatic acylation of lysolecithin has been shown in several tissues (13) and has been implicated in membrane phospholipid turnover (13, 14), intestinal absorption of phospholipids (15), and in increased synthesis of membrane phospholipids in response to external stimuli (16, 17). We have recently demonstrated acylation of lysolecithin in normal human plasma (3, 4). Because of the possible importance of the reaction in the metabolism of lysolecithin which is continuously produced in the plasma by the actions of lecithincholesterol acyl transferase and phospholipase A, the characterization of the enzyme catalyzing this reaction is of great interest. The results presented here indicate that the same enzyme which is responsible for the production of most of the lysolecithin in the plasma can carry out the acylation of lysolecithin to lecithin. The identical profiles of lysolecithin acyltransferase and lecithin-cholesterol acyl transferase activities during purification by different chromatographic procedures, as well as the presence of both activities in the same protein band on the polyacrylamide electrophoretic gels, indicates that a single protein carries out both the reactions. Further evidence that both activities are carried out by the lecithin-cholesterol acyltransferase protein is obtained by the results on the effects of ph, heat, and inhibitors on the purified enzyme preparation, and from the absence of lysolecithin acyltransferase activity in patients with familial lecithin-cholesterol acyltransferase deficiency. Purified lecithin-cholesterol acyltransferase has been shown to have phospholipase A activity, especially in the absence of unesterified cholesterol (12, 18). The present research shows for the fast time that the purified lecithin-cholesterol acyltransferase also has lysolecithin acyltransferase activity. The requirement of apoprotein A-I for the lecithin-cholesterol acyltransferase and phospholipase activities of the purified enzyme preparation is well established (5, 12, 18, 19) although the mechanism of the apoprotein-dependent activity or its physiological importance is not clear. Unlike this reaction, however, the acylation of lysolecithin requires the presence of LDL. LDL alone stimulated lysolecithin acylation by the purified enzyme, and addition of apo A-I to the LDL-containing reaction mixtures did not have any further effect on lysolecithin acylation. Pretreatment of the LDL with IgG from rabbit anti-human A-I serum did not have any inhibitory effect on this reaction. Thus it appears that the enzyme preferentially esterifies cholesterol in presence of HDL, but esterifies lysolecithin in presence of LDL. The precise role of LDL in lysolecithin acylation is not

6 9280 Lysolecithin Acylation clear, but its affinity to lysolecithin has been shown to be higher than that of serum albumin (20) and therefore it may be providing a binding surface for the substrate. also It appears likely that the LDL acts as an acyl donor for the lysolecithin acylation reaction, because in a reaction mixture with purified enzyme, there is no other source of acyl groups except for lysolecithin itself. Although our preliminary studies suggested that a transesterification between two lysolecithin molecules may be taking place (3), the results presented here using doubly labeled lysolecithin show that the acyl group may be coming from LDL. There are several potential acyl donors in LDL phospholipids, cholesteryl esters, triglycerides, and mono- and diglycerides. However, since lecithin is the substrate for lecithin-cholesterol acyltransferase and phospholipase reactions, it may provide the acyl group for lysolecithin acylation. Thus it is tempting to speculate that the formation of an acyl-enzyme intermediate with the 2-acyl group of lecithin is a common step for the three reactions catalyzed by the enzyme and they differ only in the second stage of the reaction, namely the transfer of the acyl group to an acyl group acceptor. The acyl group is transferred to the cholesterol in lecithin-cholesterol acyltransferase reaction, while it is transferred to water in phospholipase reaction and to lysolecithin in lysolecithin acyltransferase reaction. Such a mechanism is also supported by the fact that the acyl acceptor specificity of lecithin-cholesterol acyltransferase is not strictly toward cholesterol, but other sterols and even fatty alcohols can accept the acyl group (21). The formation of an acyl intermediate is also supported by the inhibition of enzyme activity by serine esterase inhibitors like diisopropyl fluorophosphate and diethyl p-nitrophenyl phosphate. It should be pointed out that the concentration of lysolecithin used for the lysolecithin acyltransferase assay (0.5 mm) is inhibitory for the esterification of cholesterol. Thus, in accordance with the results of Fielding et al. (22) we found that lysolecithin at concentrations as low as 3 ~ Linhibited M the lecithin-cholesterol acyltransferase activity by over 50% in the absence of serum albumin.' This inhibitory action of lysolecithin can at least partly be explained by the postulate that lysolecithin com- ' J. J. Albers, J. Lin, C. H. Chen, and P. V. Subbaiah, unpublished observations. by Lecithin-Cholesterol Acyltransferase petes with cholesterol for the acyl group during the transfer reaction. Acknowledgments-We wish to thank Dr. John Glomset of the University of Washington, Dr. Kaare Norum of the University of Oslo, Norway, and Dr. J. Frohlich of Vancouver General Hospital, Vancouver, B. C., Canada, for providing the samples of lecithincholesterol acyltransferase-deficient plasma. REFERENCES 1. Glomset, J. A. (1968) J. Lipid Res. 9, Vogel, W. C., and Bierman, E. L. (1967) J. Lipid Res. 9, Subbaiah, P. V., and Bagdade, J. D. (1978) Life Sci. 22, Subbaiah, P. V., and Bagdade, J. D. (1979) Biochim. Biophys. Acta 573, Albers, J. J., Lin, J., and Roberts, G. P. (1979) Artery 5, Parthasarathy, S., and Ganguly, J. (1973) Biochim Biophys. Acta 296, Bligh, E. G., and Dyer, W. J. (1959) Can. J. Biochem. Physiol. 37, Cheung, M. C., and Albers, J. J. (1977) J. Clin. Znuest. 60, Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, Davis, B. J. (1964) Ann. N. Y. Acad. Sci. 121, Johnson, W. C., and Lindsey, A. J. (1939) Analyst 64, Aron, L., Jones, S., and Fielding, C. J. (1978) J. Biol. Chem. 253, Van Den Bosch, J., Van Golde, L. M. G., and Van Deenen, L. L. M. (1972) Ergeb. Physiol. Biol. Chem. Exp. Pharmakol. 66, Lands, W. E. M., and Crawford, C. G. (1976) in The Enzymes of Biological Membranes (Martonosi, A., ed) Vol. 2, pp. 3-85, Plenum Press, New York 15. Subbaiah, P. V., Sastry, P. S., and Ganguly, J. (1970) Biochem. J Elsbach, P., and Levy, S. (1969) J. Clin. Knuest. 47, Ferber, E., Redly, E. C., and Resch, K. (1976) Biochim. Biophys. Acta 448, Piran, U., and Nishida, T. (1976) J. Biochem. (Tokyo) 80, Albers. J. J. (1978) Scand. J. Clin. Lab. Knuest. 38, Suppl. 150, Portman, 0. W., and Illingworth, D. R. (1974) Scand. J. Clin. Lab. Knuest. 33, Suppl. 137, Kitabatake, K., Piran, U., Kamio, Y., Doi, Y., and Nishida, T. (1979) Biochirn. Biophys. Acta 573, Fielding, C. J., Shore, V. G., and Fielding, P. E. (1972) Biochim. Biophys. Acta 270,