paf-acetylhydrolase activity on lp(a) before and during

Atherosclerosis 125 (1996) 121 134

atherosclerosis

PAF-acetylhydrolase activity on Lp(a) before and during Cu2 + -induced oxidative modification in vitro

Sonia-Athena P. Karabina”, Moses C. Elisaf”, John Goudevenosc, Kostas C. Siamopoulos b, Dimitris Sideris”, Alexandros D. Tselepis”%*

“Laboratory of’ Biochemistry, Department of Chemistry, University of loannina, 451 10 loannina, Greece bDepartment of Internal Medicine, Medical School, University of loannina, 451 10 loannina, Greece

‘Department of Cardiology, Medical School, University of Ioannina, 451 10 Ioannina, Greece

Received 10 July 1995; revised 23 February 1996; accepted 29 March 1996

Abstract

In human plasma with no detectable lipoprotein (a) (Lp(a)) levels, platelet-activating factor acetylhydrolase (PAF-AH) is associated with low density lipoprotein (LDL) and high density lipoprotein (HDL) with a distribution of 70 and 30%, respectively. We used a density gradient ultracentrifugation procedure to study the distribution of PAF-AH among lipoproteins in plasma containing Lp(a). Lp(a) was migrated as a broad band in the density region of d = l.OSO- 1.100 g/ml, independently of its isoform size. In plasma with Lp(a) levels 30-40 mg/dl or 80-100 mg/dl the PAF-AH activity migrated in this density region was 4 or 9% higher as compared to plasma having Lp(a) levels < 8 mg/dl (P < 0.05 or P < 0.02, respectively). Enrichment of plasma with the dense LDL, subfraction, significantly increased the enzyme activity distributed in this density region. The physicochemical properties of the Lp(a)-associated PAF-AH activity were similar to those reported for the LDL-associated enzyme. However, the kinetic constants in small Lp(a) isoforms were significantly higher compared to large ones. Isoform F had apparent Km = 117 f 9 pmol/l and V,,, = 94 + 5 nmol/mg protein per min, and isoform S2jS3 had apparent K, = 36 + 9 pmol/l and V,., = 25 If: 5 nmol/mg protein per min. Removal of apolipoprotein (a) (ape(a)) from Lp(a) by reductive cleavage with dithiothreitol, slightly affected the amount of PAF-AH existing on Lp(a) since, only 15 i 5% of the total enzyme activity dissociated from its particle after density gradient ultracentrifugation. During Cu2 +-induced Lp(a) oxidation, the PAF-AH activity decreased from 10.90 k 2.30 nmol/mg per min to 2.57 + 0.56 nmol/mg per min 4 h after the initiation of the oxidation (P < 0.001). The apparent K, of the enzyme remained essentially unchanged during oxidation, whereas V,,, was significantly decreased from 58.6 + 7.8 nmol/mg protein per min to 38.2 f 8.7 nmol/mg protein per min (P < 0.03). An extensive hydrolysis of the endogenous phosphatidylcholine (PC) to lysophosphatidylcholine (Lyso-PC) was observed during Lp(a) oxidation, since the Lyso-PC/sphingomyelin molar ratio at the end of oxidation (0.55 rf: 0.09) was significantly higher than that before oxidation (0.19 & 0.01, P < 0.001). Our results show that the existence of Lp(a) in plasma alters the distribution

* Corresponding author, Tel.: + 30 651 98365; fax: + 30 651 47832

0021-9150/96/$15.00 0 1996 Elsevier Science Ireland Ltd. All rights reserved

PII SO02 l-9 150(96)05872-g

122 S.A.P. Karabina et al. / Atherosclerosis 125 (1996) 121-134

of PAF-AH among the other lipoproteins. Ape(a) seems to affect the association of the enzyme with Lp(a) but does not bind itself to PAF-AH. During Lp(a) oxidation, the PAF-AH activity decreases whereas an extensive hydrolysis of the endogenous PC to Lyso-PC is observed which is possibly due to the PAF-AH activity.

Keywords: PAF-acetylhydrolase; Lipoprotein (a); Ape(a) isoform; Lysophosphatidylcholine; Oxidation

1. Introduction

Lipoprotein (a), (Lp(a)) is considered to be an independent risk factor for atherosclerosis in both Caucasian and Oriental populations with plasma Lp(a) levels > 30 mg/dl [l-5]. As suggested by clinical studies, there is a positive correlation between elevated serum Lp(a) levels and coronary heart disease [l-3]. Additionally, serum Lp(a) levels are correlated best among all lipid parameters with the number and size of atherosclerotic plaques in patients with cervical atherosclerosis [4,5]. Lp(a) resembles low density lipoprotein (LDL) in lipid composition and also contains apolipoprotein B- 100 (apoB- 100). However, Lp(a) can be distinguished from LDL by the presence of apolipoprotein (a) (ape(a)), a high molecular mass glycoprotein that shows remarkable size polymor- phism and is attached to apoB-100 through disul- phide linkage [6-81. Ape(a) is highly homologous to the plasma protease zymogen plasminogen [9]. Thus, it has been suggested that Lp(a) may pro- mote thrombogenesis by inhibiting fibrinolysis. Several studies have shown that Lp(a) inhibits plasminogen activation, since it interferes with the binding of plasminogen to immobilized fibrin and to the plasminogen receptor [lo]. Apart from its role in thrombogenesis, Lp(a) has atherogenic properties, since it has been found in the intima of atherosclerotic lesions [l l] and like LDL it can be oxidatively modified when incubated with mononuclear cells [12] or Cu2 + in vitro [13]. Oxidative modification of Lp(a) causes marked changes in its structure and biological properties. Relative to native Lp(a), oxidized Lp(a) (Ox- Lp(a)) particles show complete loss of its antioxi- dants and polyunsaturated fatty acids [13], protein fragmentation and an increased negative charge [14]. Oxidatively modified Lp(a) is recog- nized and taken up by macrophages via the scavenger receptor [ 151.

Similar changes have been reported to take place during LDL oxidative modification in vitro [16]. Furthermore, during LDL oxidation an extensive hydrolysis of its phosphatidylcholine (PC) content to lysophosphatidylcholine (Lyso-PC) is observed and it is mediated through an LDL-associated phospholipase AZ-like activity [ 17,181. Several studies have shown that this phospholipase AZ-like activity is rather due to the LDL-associated PAF-acetylhydrolase (PAF-AH) (EC 3. 11. 48) [19,20]. PAF-AH exists in many tissues as well as in plasma and was initially described as the enzyme that hydrolyzes and inactivates platelet-activating factor (PAF), a potent media- tor in allergic and inflammatory diseases [21,22]. However, recent studies have shown that several oxidized phospholipids resembling PAF in structure and biological activity are also substrates for PAF-AH [23]. It has been shown that in human plasma PAF-AH is associated with lipoproteins and in donors with no detectable levels of Lp(a), it is associated with LDL and HDL with a distribution of 70 and 30%, respectively [24].

In the present study we investigated the distribution of PAF-AH among the lipoproteins in human plasma with Lp(a) levels > 30 mg/dl. Furthermore, we isolated Lp(a) and studied the physicochemical and kinetic properties of its endogenous PAF-AH, the possible influence of ape(a) on the enzyme activity, as well as the effect of the Cu* + -induced Lp(a) oxidative modification on PAF-AH activity. The endogenous PC hydrolysis and the possible involvement of the Lp(a)-associated PAF-AH were also studied.

2. Materials and methods

2.1. Chemicals

Bovine serum albumin (BSA), trypsin,

S.A.P. Karahirzcr et al. i Atherosclero.\i.s l-1.5 (1996) 121 134 123

iodoacetic acid (IAA), dithiothreitol (DTT), phenylmethylsulfonyl fluoride (PMSF), diisopropylfluorophosphate (DFP) and standard phospholipids were obtained from Sigma Chemical Co., CV 3988 was obtained from Takeda Chemi- cal Industries Ltd., organic solvents were from Lab-Scan (Ireland) and chemiluminescence reagent (Renaissance) from DuPont-New England Nuclear. Sheep polyclonal anti-ape(a) and anti- apoB-100 antibodies were kindly provided by Dr Jiielle Thillet (INSERM U321, Paris, France). PAF (hexadecyl), obtained from Bachem (Switzerland), was dissolved in ethanol (SO%, v/v) at a final concentration of 5 mM. This solution was mixed with 1-0-hexadecyl-2-[3H-acetyl]-sn- glycero-3-phosphocholine (10 Ci/mmol; DuPont- New England Nuclear) in various proportions, dried under a stream of nitrogen and redissolved in a solution containing BSA/saline (2.5 mg/ml), to obtain [“H]PAF solutions with concentrations ranging from Z!-25 mM.

2.2. Fractionution of plasma lipoproteins

Subjects were healthy normolipidemic volun- teers with plasma Lp(a) levels lower than 8 mgjdl (controls) or higher than 30 mg/dl (isoforms F. Sl, S2, S2,‘S3). Venous blood samples were collected from overnight fasting donors in tubes containing 0.01% Na,EDTA. Plasma lipoproteins were fractionated by density gradient ultracentrifugation in a Beckman L7-65 ultracentrifuge at 40000 rpm, 15°C for 24 h, using a type SW 41 rotor [13,25]. Briefly, 3 ml of plasma from each donor individually containing 1..23 g of dry solid KBr were successively overlayered by 3 ml of a KBr solution of d = 1.080 g/ml, 3 ml of d = 1.050 g/ml and 2 ml of distilled water. All density solutions and plasma contained 0.01% NazEDTA and 5 mg/ml gentamicin sulfate at pH 7.4. After ultracentrifugation 28 fractions of 400 ill were collected by successive aspiration from the menis- cus downwards. All fractions were analyzed for their cholesterol and Lp(a) content. PAF-AH activity was determined in all gradient fractions as described below. In some experiments before ultracentrifugation plasma was enriched with the dense LDL subfraction (LDL,), (d = 1.050 to

1.063 g,‘ml) isolated by isopycnic gradient ultracentrifugation from normolipidemic human plasma with Lp(a) levels lower than 8 mg/dl, [26J.

Eighteen-to-thirty-six millilitres of pooled plasma with Lp(a) levels higher than 30 mg/dl, were subjected to density gradient ultracentrifugation as described above. Using this method, Lp(a) migrated between LDL and HDL at the density of d = 1.050- 1.100 g/ml. This density region containing Lp(a) was carefully collected from all ultracentrifugal tubes. pooled and further purified from contaminating lipoproteins by gel filtration chromatography on Sepharose 6B column (70 x 1.5 cm) [27]. Chromatography was carried out at a flow rate of 18 ml/h using the following elution buffer: 10 mM Tris-HCl, 1 M NaCl, 0.02% NaN,, 1 mM Na,EDTA, pH 7.4. Fractions of 2.1 ml were collected. From the first UV-absorbing peak the first 10 eluting fractions (those containing pure Lp(a)) were retained and subjected to ultracentrifugation in a Beckman L7-65 ultracentrifuge at 40 000 rpm 15°C for 10 h, using a type NVT 65 rotor. The concentrated Lp(a) was recovered at the bottom of the tube and dialyzed for 24 h against 2 changes of a 200-fold volume of 10 mM PBS, pH 7.4 or 10 mM Tris-HCl, pH 7.4. The Lp(a) preparation was filter-sterilized (0.45 pm. Millipore, USA) and was used immediately or stored in the dark at 4°C. for up to 2 weeks in the presence of 0.01% Na,EDTA. In some experiments Lp(a) was isolated from each donor individually. The purity of Lp(a) in all preparations was evaluated by non-denaturing polyacrylamide gel electrophoresis (PAGE) in polyacrylamide gradient gels 2- 16% [28] and by 3.75% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) followed by immunoblotting using sheep polyclonal anti-ape(a) and anti-apoB- 100 antibodies and revealed by chemiluminescence. The purity of Lp(a) was also evaluated by gradient SDS-PAGE in 5 - 19% SDS-gradient polyacrylamide gels [29]. The Lp(a) preparations were essentially free of any other lipoprotein or albumin. Dialyzed Lp(a) was then submitted to lipid and protein analysis and was further characterized by agarose gel elec-

124 S.A.P. Karabina et al. I Atherosclerosis 125 (1996) 121-134

trophoresis (Hydragel Lipo and Lp(a) kit, Sebia) as well as by Lp(a) determination.

2.4. Determination of ape(a) isoforms

Ape(a) isofotm analysis was performed by electrophoresis in 1.5% SDS-agarose gels followed by immunoblotting and revealed by chemiluminescence according to the method of Kamboh et al. [30] as modified by Doucet et al. [31]. A standard (Immuno-France) was used, containing four different isoforms corresponding to the F, Sl, S2, and S3 isoforms in the nomenclature of Utermann et al. [32].

2.5. Reductive cleavage of Lp(a) and ultracentrifugation of reduced Lp(a)

Lp(a) containing 1.0 mg of total protein was dialyzed at 4°C for 24 h against 2 changes of a 200-fold volume of 20 mM Tris (pH 7.4), 0.15 M NaCl, 1 mM EDTA, 0.02% NaN,. Then it was incubated with 10 mM dithiothreitol (DTT) at 37°C for 3 h [33] and dialyzed overnight at 4°C against 10 mM Tris, 1 mM EDTA, pH 7.4. In control experiments, Lp(a) was incubated as above in the absence of DTT, whereas in other experiments the same treatment with or without DTT was performed using LDL isolated by se- quential ultracentrifugation from normolipidemic human plasma with Lp(a) levels lower than 8 mg/dl [20]. Dialyzed Lp(a) or LDL was submitted to density gradient ultracentrifugation as described above. After ultracentrifugation, 28 fractions of 400 ~1 were collected as described and tested for PAF-AH activity. The enzyme activity was also determined in 25 ~1 of Lp(a) and LDL before and after DTT treatment. The resultant lipoprotein after DTT treatment, and the two last bottom fractions (27 and 28) were characterized by agarose gel electrophoresis and by 3.75% SDS- PAGE in the absence of DTT followed by immunoblotting using sheep polyclonal anti-ape(a) and anti-apoB-100 antibodies and revealed by chemiluminescence. Furthermore, the two bottom fractions were submitted to protein and lipid analysis.

2.6. Oxidation of Lp(a)

A volume of 1.3 ml of Lp(a) containing 180 pg protein/ml PBS was oxidized in the presence of CuSO,, 16 PM final concentration [ 131. The kinetics of the oxidation was determined by moni- toring the increase in the 234-nm absorbance on a Perkin-Elmer L15 spectrophotometer, every 10 min for 4 h [34]. The lag time, the maximal rate of conjugated dienes formation, and the total amount of dienes formed were calculated as previously described [35]. Aliquots of 100 ~1 were taken before oxidation, at the end of propagation phase, as well as 4 h from the onset of the oxidation for the determination of PAF-AH activity and electrophoretic behaviour on agarose gels, expressed as relative to native electrophoretic mobility (R.E.M). In some experiments Lp(a) preparations in PBS, containing 0.01% Na,EDTA, were incubated with 20 mM diisopropylfluorophosphate (DFP) for 30 min at 37°C a procedure that completely inactivates PAF-AH [21]. Then it was dialyzed against PBS for 24 h at 4°C to remove both Na,EDTA and the excess of DFP and then oxidized as above. In selected experiments, Lp(a) in PBS containing 0.01% Na,EDTA was heated at 60°C for 1 h, then dialyzed against PBS for 24 h at 4°C and oxidized as above. Finally, in some experiments the oxidation was terminated at the end of the propagation phase for the determination of the enzyme kinetic constants.

2.7. PAP-acetylhydrolase assay and characteriza- tion

The Lp(a)-associated PAF-AH activity was measured by the trichloroacetic acid (TCA) precipitation procedure [36] with some modifications [26]. For the routine assay 10 pug of Lp(a) protein or 25 ~1 of the gradient fractions were mixed with 10 mM PBS (pH 7.4) in a final volume of 90 ~1, and the reaction was performed for 10 min at 37°C by adding 10 ~1 of [3H]PAF (final concentration, 50 PM). PAF-AH activity was expressed as nmol PAF degraded per min per mg of protein or mg of lipoprotein mass or ml of plasma. In some experiments PAF-AH activity was mea-

S.A.P. Karabina et al. 1 Athero.rclerosis 125 (1996) 121~ 134 125

sured in the presence of 10 mM Ca2 + or 10 mM Na,EDTA in 10 mM Tris-HCl. In other experiments Lp(a) was heated at 60°C for 1 h or pre-incubated at 37°C for 30 min in the presence of 10 mM CV 3988, prior to the addition of [3H]PAF. In selected experiments, Lp(a) in PBS containing 0.01% Na,EDTA, was treated with 20 mM DFP for 30 min at 37°C or with 4 mM phenylmethyl- sulphonyl fluoride (PMSF) for 1 h at 37°C and dialyzed overnight against PBS before PAF-AH measurement. Finally, in some experiments, Lp(a) containing 0.26 nmol/min of PAF-AH activity, was treated with 5 mg/ml trypsin, 2 mM iodoacetic acid (IAA) or 1 mM dithiothreitol (DTT) as described [26].

2.8. Kinetic studies of PAF-AH

The kinetic properties of the Lp(a)-associated PAF-AH were evaluated in each isoform as well as in pooled Lp(a) before oxidation and at the end of the propagation phase, using 2.55400 ,uM [3H]PAF final concentration. The apparent K, and V,,, values of the enzyme were calculated using the Lineweaver-Biirk representation of the data.

2.9. Lipid extraction and separation

Total lipids of Lp(a) corresponding to 180 ,ug protein were extracted before as well as at the end of the oxidation according to Bligh and Dyer [37]. Phospholipids were separated by thin layer chromatography (TLC) and quantified by phosphorus analysis as described [34].

2.10. Analytical methods

Total cholesterol, free cholesterol and triglycerides were analyzed by commercially available enzymic reagents (Boehringer-Mannheim). Cholesteryl ester mass was calculated as 1.67 x (free cholesterol mass). Protein was determined by the method of Lowry et al. [38]. Lp(a) was determined by an enzyme immunoassay method, Macra Lp(a) (Terumo Med. Corp., Elkton, MD, USA).

2. Il. Statistical analysis

Results are expressed as mean + S.D. Mean values were compared by Student’s t-test, with significance defined at a value of P < 0.05.

3. Results

3.1. Distribution of PAF-AH activity among plasma lipoproteins

Plasma lipoproteins were fractionated by density gradient ultracentrifugation and representative curves of the lipoprotein profiles obtained are shown in Fig. 1A and Fig. 1C. In controls (plasma with Lp(a) levels lower than 8 mg/dl), Lp(a) was not detectable in any gradient fraction or it was migrated as a narrow band in fractions 12-15 (d = 1.064-1.095 g/ml). In plasma with Lp(a) levels higher than 30 mg/dl, Lp(a) was migrated as a broad band in fractions 9-17 (d = 1.050- 1.100 g/ml). There were small differences in the density region where each Lp(a) isoform peak appeared but all were distributed within the density region of 1.050- 1.100 g/ml (data not shown). All gradient fractions were tested for PAF-AH activity and Fig. 1B and Fig. 1D show representative profiles of the enzymic activity obtained. In all subjects, PAF-AH activity was mainly associated with the dense portions of both LDL and HDL, an observation which is in accordance with our recently published results [26]. In controls, the enzyme activity migrated in fractions 9- 17 was 50.0 + 1.0% of the total PAF-AH activity present in all gradient fractions. Fractions l-8 (VLDL plus the major part of LDL) contained 18.9 + 1.6%, whereas fractions 18-28 (the major part of HDL f proteins) contained 3 1.1 i 1.2% of total PAF-AH activity (Fig. 1B). In plasma with Lp(a) levels 30-40 mg/dl, the enzyme activity migrated in fractions 9- 17 was 54 + 1.6% (4% higher as compared to controls, P < 0.05, n = 3). Fractions 1-8 and 18-28 contained 20 f 1.8% and 26 + 1.3% of PAF-AH activity, respectively. Finally, in plasma with Lp(a) levels 80- 100 mg,idl, the PAF-AH activity migrated in fractions 9- 17 was 59 + I .5% (9% higher as compared to

126 S.A.P. Karabina et al. 1 Atherosclerosis 125 (1996) 121-134

0.30 0.16

0.14 3 ,* 1 to.12 g

3 0.25 ZE L 9 0.20 r “a 0.10 .$j z 0.15

0.0s &

u 3 0.10

0.06 g

0.01 =

0.02 s

0.00 l

0 0.05

0.00 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

1 2 3 4 5 6 7 8 9 10111213141516171819202122232425262728

0.30 0.16 3

6 0.25 0.14 P

0.12 'L 0.20

0.10 2

0.15 0.08 %i

0.10 0.06 E

0.04 3

0.05 0.02

3

0.00 o.cll l

0 I 2 3 4 5 6 7 8 9 10111213141516171819202122232425262728

1 2 3 4 5 6 7 8 9 10 I1 12 13 14 l5 16 17 18 19 20 21 22 23 24 25 26 27 28

Gradient Fraction Number

Fig. 1. Distribution of PAF-AH activity on human plasma lipoproteins separated by density gradient ultracentrifugation. Density gradient uhracentrifugation was performed using a type SW 41 rotor at 40000 rpm, 15°C for 24 h as described in Section 2. PAF-AH activity was determined by the TCA precipitation method and Lp(a) was measured by an enzyme immunoassay method. A$. Representative curves of lipoprotein profile corresponding to plasma with Lp(a) levels 8 mg/dl and SO mg/dl (isoform Sl), respectively. B,D. PAF-AH activity in gradient fractions corresponding to plasma with Lp(a) levels 8 mg/dl and 80 mg/dl, respectively.

S.A.P. Karabina et al. 1 Atherosclerosis 125 (1996) 121-134 121

controls, P < 0.02, y1 = 3). Fractions l-8 and 18-28 contained 19 + 1.4% and 22 + 0.9% of the enzyme activity, respectively (Fig. 1D). No statistically significant differences were found in the distribution of PAF-AH activity among plasma having similar Lp(a) levels but different isoforms. To investigate whether the PAF-AH activity migrated in fractions 9- 17 can be influenced by the enzyme activity associated with the dense LDL subfraction LDL,, we performed density gradient ultracentrifugation of plasma with Lp(a) levels 80 mg/dl (isoform Sl) enriched with two different amounts of LDL, (total PAF-AH activity 6 and 12 nmol/min). We found that the enzyme activity was significantly increased by 6 and 8% respectively, as compared to that before- the enrichment (65 _+ 2.0% and 67 _+ 3.2’/0 respectively compared to 59 & 1.5O/u, P < 0.05, n = 3).

3.2. Properties of’ PAF-AH associated with Lp(a)

To study the physicochemical and kinetic properties of the Lp(a)-associated PAF-AH we isolated and purified Lp(a) from pooled plasma containing the isoforms F, Sl and S2. Lp(a) was initially submitted to compositional analysis and,

Table 1 Chemical composition of human Lp(a) isolated from pooled plasma, expressed as ‘%I of total mass

Parameter I-p(a) __--

Protein, ‘%I Free cholesterol, %) Cholesteryl esters %I Triglycerides, ‘I/u Phospholipids. ‘%I PE. ‘%I PC, % Sph, ‘%, Lyso-PC, ‘%I Total lipoprotein mass, mg/ml

36.8 k 2.9 8.1 ) 2.9

34.0 + 3.3 6.1 f 1.9

15.1 + 1.8 4.9 f 1.4

59.3 + 8.1 29.8 + 1.6 4.3 + 1.6 0.6 _+ 0.1

Each phospholiprd class is expressed as ‘% of total phospholipids. Values represent the mean + S.D. from 5 different Lp(d)

preparations. Total lipoprotein mass corresponds to the sum of all lipid and protein components of Lp(a). PE, phos- phatidylethanolamine: PC, phosphatidylcholine; Sph, sphingomyelin; Lyso-PC, lysophosphatidylcholine.

Table 2

Behaviour of the Lp(a)-associated PAF-AH activity to various treatments

Treatments

Control Ca’+, 10 mM EDTA, IO mM DFP, 20 mM, 30 min, 37°C PMSF, 4 mM, 1 h, 37°C CV 3988, 10 mM, 30 min, 37°C Trypsin, 5 mgiml, I h, 37°C DTT. 1 mM, 15 min, 37°C IAA, 2 mM, 15 min, 37°C Heating, 1 h, 60°C

PAF-AH activity, nmol,‘mg protein per min

- 10.90 * 2.30 10.30 * 1.47 il.64 & 1.13 0.10 + o.io* 1.60 i 0.78* 0.07 If- 0.01* 9.51 + 0.13

Il.97 * 0.42 12.16 f 0.15 10.12 + 0.17

The Lp(a)-associated PAF-AH activity was determined by the TCA precipitation method as described in Section 2. In several experiments Lp(a) was submitted to various treatments, as shown in the table. DFP, diisopropylfluorophosphate; PMSF. phenylmethylsulfonylfluoride; DTT, dithiothreitol; IAA. iodoacetic acid. Values represent the mean k S.D. from 3 experiments in duplicate. *P < 0.0001 compared to control.

as shown in Table 1, its chemical composition was similar to that previously reported [39]. Its lipoprotein mass was found to be 0.6 + 0.1 mg/ml. PAF-AH activity was initially determined using various Lp(a) final protein concentrations ranging from 20-200 pg/ml. We, found that enzyme activity was linearly related to protein concentrations less than 150 pgjml (data not shown). Thus, all the following assays were performed using 100 bug/ml final protein concentration. The PAF-AH activity in these Lp(a) preparations was 10.90 + 2.30 nmol/mg protein per min (Table 2). To eval- uate whether the properties of the Lp(a)-associated PAF-AH resemble those reported for the enzyme associated with the other lipoproteins, we studied the effect of Ca’+ or EDTA on the enzyme activity. We next established the susceptibility of the enzyme to the proteolytic activity of trypsin, as well as to treatment with the serine esterase inhibitors DFP and PMSF, the specific PAF antagonist CV 3988 [40], and the sulphydryl reagents, IAA and DTT. As shown in Table 2, the enzyme activity was neither influenced by the presence of Ca2 + or EDTA nor by treatment with trypsin, IAA or DTT. By contrast, PAF-AH ac-

128

Table 3

S.A.P. Karabina et al. / Atherosclerosis 125 (1996) 121-134

Specific activity and kinetic constants of PAF-AH associated with different Lp(a) isoforms

Lp(a) isoform K,,, wol/l V,,,, nmol/mg protein per min Specific activity, nmol/mg PAF-AH activity % of total lipoprotein mass per min plasma activity

F 117 f 9* 94 + 5* 14.0 * 1.2 (9.5 _+ 1.4)*” 29-39 Sl 98 f 8 85 k 3 11.0 + 0.8 (9.9 f 0.7) 31-38 s2 50 + 6 38 _+ 4 7.1 + 0.6 (6.5 & 0.4) 22-35 S2jS3 36 + 9 25 & 5 4.8 5 0.5 (3.2 + 0.4) IO-18

Ape(a) isoform analysis was performed by electrophoresis in 1.5% SDS-agarose gels followed by immunoblotting and revealed by chemiluminescence. PAF-AH activity was determined by the TCA precipitation method. The apparent K, and V,,, values were calculated using the Lineweaver-Btirk representation of the data. Total plasma activity was calculated adding the activities of the 28 gradient fractions and ranged from 19 nmol/ml plasma per min to 33 nmol/ml plasma per min. Values represent the mean _+ S.D. from 3 experiments in duplicate. “Values in parenthesis denote specific activity in nmol/ml plasma per min. *Statistically significant difference among F and Sl, Sl and S2, S2 and S2/S3, P < 0.05.

tivity was inhibited by DFP, PMSF or CV 3988 (P < 0.0001) (n = 3). Heating the Lp(a) at 60°C for 1 h, treatment sufficient to inactivate the 1ecithin:cholesterol acyltransferase (LCAT) activity [17], did not affect the PAF-AH activity. Fi- nally, we determined the Michaelis-Menten kinetics of the enzyme and using the Lineweaver- Btirk representation of the data we calculated the apparent K, = 72.1 + 14.2 pmol/l and V,,, = 58.6 + 7.8 nmol/mg protein per min.

3.3. PAF-AH activity on different isoforms of -Wa)

Lp(a) isoforms F, Sl, S2 and S2/S3 were isolated from individual donors and the PAF-AH specific activity, as well as the kinetic constants were determined. As shown in Table 3, the apparent K, and V,,, values decreased significantly as the ape(a) isoform size increased. Thus, in isoform F, the K, and V,, values were 117 + 9 pmol/l and 94 & 5 nmol/mg protein per min, respectively, whereas in isoform S2/S3 the K, value was 36 f 9 ,umol/l and V,,, 25 f 5 nmol/mg protein per min. Isoforms Sl and S2 exhibited K, and V,,, values between those of isoforms F and S2/S3. The same phenomenon was observed for the enzyme specific activity expressed either per mg of lipoprotein mass or per ml of plasma. The PAF-AH activity associated with each Lp(a) isoform expressed as percentage

of total plasma activity was influenced from the plasma Lp(a) levels and was ranged from 10 to 39% (Table 3).

3.4. PAF-AH activity after reductive cleavage of W-4

Incubation of Lp(a) isolated from pooled plasma with 10 mM DTT, a procedure that completely dissociates ape(a) from Lp(a) [41], did not affect the enzyme activity (total activity, 21.7 f 6.9 nmol/min before treatment and 23.8 +_ 6.8 nmol/min afterwards). However, when reduced Lp(a) was submitted to density gradient ultracentrifugation, 85 + 5% of total enzyme activity (20.2 + 1.2 nmol/min) was found associated with the resultant lipoprotein migrated at d = 1.038- 1.089 g/ml. This lipoprotein contained apoB-100 but not ape(a) and migrated with p-electrophoretic mobility on agarose gel electrophoresis. The remaining 15 + 5% of total enzyme activity (3.6 + 1.2 nmol/min) was found in the two bottom fractions 27,28 (d > 1.210 g/ml). These fractions contained ape(a) but not apoB- 100 and were almost lipid free, since only 0.72% of phospholipids and 0.38% of total cholesterol were associated with these fractions. After ultracentrifugation of unreduced Lp(a) (incubated at 37°C for 3 h in the absence of DTT), 97 + 3% of total enzyme activity was associated with the lipoprotein band migrating at d = 1.050-1.100 g/ml

S.A.P. Kurahina et al. 1 Atherosclerosis 125 (1996) 121~ 134 129

whereas the bottom fractions 27, 28 did not contain any enzyme activity or apolipoproteins or lipids. Similar results were obtained when LDL treated with DTT or without treatment was submitted to ultracentrifugation, with the exception that the band was migrating at lower density, d = 1.019-1.064 g/ml.

3.5. PAF-AH activity during Lp(a) oxidation

We next studied the behaviour of the Lp(a)-associated PAF-AH activity during oxidation. Lp(a) isolated from pooled plasma was oxidized in the presence of 16 ,uM Cu2 + under continuous moni- toring of the absorbance at 234 nm. The lag time was 69 f 23 min, the rate of oxidation 5.0 + 1.3 nmol/mg protein per min, and the total amount of the dienes 256 k 52 nmol/mg protein. The R.E.M values at the end of propagation phase were 1.19 + 0.01 while 4 h after the onset of the oxidation (decomposition phase) 1.35 + 0.05. As shown in Fig. 2, PAF-AH activity was substan- tially decreased during Lp(a) oxidation from 10.90 + 2.30 nmol/mg protein per min to 4.60 nmol/mg protein per min at the end of the propagation phase (P < O.OOl), and to 2.57 f 0.56 nmol/mg protein/min 4 h after the initiation of the oxidation (decomposition phase), P < 0.001 (n = 7). CL? + itself did not affect the enzyme

Lag Propagation Decomposition

Oxidation phase

Fig. 2. Decrease of the endogenous PAF-AH activity during Lp(a) oxidative modification. Oxidation was performed in the presence of 16 ,uM Cu*+ as described in Section 2. PAF-AH activity was measured before oxidation, at the end of the propagation phase and 4 h after the initiation of the oxidation (decomposition phase). Values represent the mean f S.D. from 7 experimenls in duplicate. *P < 0.001,

0.8

n Control 0.7 n Heam

* *

.o 0.6

jj 0.5

f

Lt 0.4

B I 0.3 P

2 0.2

0.1

0.0 0 4

Time of oxidation (hours)

Fig. 3. Lyso-PC/Sph molar ratio before and at the end of Lp(a) oxidation. Untreated Lp(a) (control), Lp(a) heated at 60°C for 1 h or treated by 20 mM DFP at 37°C for 30 min, was oxidized in the presence of 16 PM CL? + as described in Section 2. Lysophosphatidylcholine (Lyso-PC) and sphingomyelin (Sph) were quantified by phosphorus analysis. Val- ues represent the mean + S.D. from at least 3 experiments in duplicate. *P < 0.001, compared to control 0 h, **P < 0.05 compared to control 4 h.

activity before oxidation (data not shown). The apparent Km value measured at the end of the propagation phase remained essentially unchanged as compared to that before oxidation, K, = 61.4 + 7.4 pmol/l, whereas the V,,, value was significantly decreased, V,,, = 38.2 + 8.7 nmol/mg protein per min vs. 58.6 k 7.8 nmol/mg protein per min, P < 0.03 (n = 3).

3.6. Lyso-PC production during Lp(ai oxidation

To investigate whether the PC hydrolysis occurring during the LDL oxidation [17] is also observed during the Lp(a) oxidation, we measured the Lyso-PC levels before and at the end of oxidation (4 h from the onset of the oxidation at the decomposition phase). Since the sphingomyelin (Sph) content remains constant during oxidation [42], Lyso-PC was expressed in terms of Lyso-PC/Sph molar ratio. As shown in Fig. 3, the Lyso-PC/Sph molar ratio at the end of oxidation (0.55 f 0.09) was significantly higher than that before oxidation (0.19 & 0.01) (P < 0.001, rj = 7). The same result was obtained even when Lp(a) was heated at 60°C for 1 h before oxidation (Lyso-PC/Sph ratio, 0.19 f 0.01 before oxida-


tion, and 0.53 + 0.09 afterwards, P < 0.001, n

= 3). By contrast, treatment of Lp(a) with DFP before oxidation inhibits the Lyso-PC production (Lyso-PC/Sph ratio at the end of oxidation, 0.22 + 0.10 vs. 0.55 + 0.09 of untreated Lp(a), P < 0.001, n = 3).

4. Discussion

The results of our study showed that, upon fractionation of lipoproteins from human plasma with Lp(a) levels higher than 30 mg/dl, an increased amount of PAF-AH activity proportional to the Lp(a) levels, was associated with the gradient fractions 9- 17, d = 1.050- 1.100 g/ml that contained Lp(a). These results suggest that the existence of Lp(a) in plasma alters the distribution of PAF-AH among LDL and HDL (70 and 30%, respectively) occurring in normolipidemic plasma with no detectable Lp(a) levels [24]. Since plasma Lp(a) levels are related to the ape(a) isoform size [32], we isolated Lp(a) of different isoforms and measured the PAF-AH specific activity. We found that the enzyme specific activity is inversely related to the Lp(a) size isoform, the small isoforms having higher specific activity compared to large ones. Thus, the Lp(a) isoform could influence the PAF-AH activity distributed in this density region where Lp(a) migrates, in addition to the Lp(a) concentration. However, in our gradient experiments there was no difference in the distribution of PAF-AH among plasma having similar Lp(a) levels but different isoforms. Recently we showed that the enzyme distribution among the LDL and HDL subfractions in plasma with Lp(a) levels < 10 mg/dl is heterogenous and the majority of the enzyme activity is associated with the dense LDL and HDL subfractions [26,34]. Thus, another factor that could influence the amount of PAF-AH distributed in the density region of d = 1.050- 1.100 g/ml is the dense LDL subfraction LDL, (d = 1.050-1.063 g/ml) which contains 60% of the total LDL-associated enzyme activity [26]. There- fore, we enriched plasma containing Lp(a) with LDL, and studied the PAF-AH distribution. The enzyme activity in fractions 9- 17 was significantly increased indicating that the level of the dense

LDL subfraction is another factor that could influence the enzyme activity distributed in the density region containing Lp(a). Overall, the distribution of PAF-AH activity among lipoproteins in plasma containing Lp(a), studied by density gradient ultracentrifugation, depends on the Lp(a) levels as well as on the dense LDL levels. Taking into account the overlaps between the Lp(a) density region and the LDL as well as the HDL regions, and the differences in the enzyme activity among the LDL and HDL subfractions, we con- clude that using a density gradient ultracentrifugation method we cannot estimate the portion of the total plasma PAF-AH activity associated with Lp(a). However, when we isolated Lp(a) from donors with known Lp(a) levels, and measured the enzyme activity, we found that a relatively high portion of total plasma PAF-AH activity ranged from 10 to 39% was associated with Lp(a).

In an effort to compare the Lp(a)-associated PAF-AH with that associated with the other plasma lipoproteins, we isolated Lp(a) from pooled plasma and studied several physicochemical properties, as well as the kinetic constants of the enzyme. We found that the physicochemical properties of the enzyme were similar to those reported for the other lipoproteins [22], since its activity is Ca 2+-independent and is inhibited by the serine esterase inhibitors DFP and PMSF, as well as the specific PAF antagonist CV 3988. Furthermore, the PAF-AH activity was not markedly influenced by treatment with trypsin, DTT or IAA. Recent studies have shown that human plasma LCAT exhibits a PAF hydrolyzing activity [43]. To exclude the possibility that LCAT participates in the Lp(a)-associated PAF-hydrolyzing activity, we heated Lp(a) at 60°C for 1 h, a procedure sufficient to inactivate LCAT, and we found that this treatment did not affect PAF hydrolysis, an evidence that this activity can not be attributed to LCAT.

We next asked whether the ape(a) content of Lp(a) affects the enzyme activity as well as the association of the enzyme with the Lp(a) particle. We found that, the apparent K, and V,,, values measured in four different Lp(a) isoforms were significantly decreased as the ape(a) isoform size increased. The K, and V,,, values of the small

S.A.P. Karahina et al. ! Atherosclewsi~ 125 (1996;) 121 134 131

isoforms F and Sl resemble those reported for the LDL, subfraction, but are much higher as compared to the other LDL subfractions [26]. It is important to note that these Lp(a) isoforms also resemble LDL, in their densities [44]. Since PAF- AH was not purified from Lp(a), V,,, may reflect the amount of the enzyme associated with each Lp(a) isoform. In this context, we can hypothesize that the small Lp(a) isoforms contain more enzyme as compared to the large ones. However, the enzyme in these Lp(a) particles displayed a markedly lower affinity for the substrate (higher K, values) an evidence that the accumulation of large amounts of PAF-AH on these particles decreases the enzyme affinity for the substrate. These results suggest that the size of the ape(a) molecule influences the association (binding) of PAF-AH to the Lp(a) particles. However. removal of ape(a) from Lp(a) by reductive cleavage and density gradient ultracentrifugation, slightly affected the amount of PAF-AH on Lp(a) since, only 15 i- 5% of the total enzyme activity dissociated from Lp(a). Based on this observation, we may suggest that the majority of the Lp(a)-associated PAF-AH binds on apoB-100 and not on ape(a) and it is expectable since the enzyme is a highly hydrophobic molecule whereas ape(a) is a hydrophilic protein which does not have any affinity for lipids or lipophilic molecules. No data exist for the association and physicochemical interactions between PAF-AH with Lp(a), i.e. which factors influence the binding of the enzyme on Lp(a), which are the interactions between PAF-AH and apolipoproteins and most impor- tantly how and where (in plasma, extracellularly or intracellularly) occurs the assembly between PAF-AH and Lp(a) in vivo. It is well established that during differentiation into macrophages monocytes release the plasma type of PAF-AH and are considered to be a major source of the plasma enzyme [45,46]. However these cells do not synthesize Lp(a) [47]. Based on the observation that small Lp(a) isoforms carry higher amounts of PAF-AH compared to large ones, we can hypothesiz,e that ape(a), due to its hydrophilic nature, opposes the binding of the enzyme on Lp(a) an evidence that PAF-AH binds Lp(a) after the linkage of ape(a) to the apoB-100 moiety of

LDL that leads to the formation of Lp(a). It is profound that more studies are required to shed light on the binding of PAF-AH on Lp(a) and whether the differences in the enzyme kinetic constants observed among the different Lp(a) isoforms are due exclusively to the existed differences in ape(a) size or whether there are other factors contributing to this phenomenon.

Several studies suggest that PAF-AH plays an important role during the LDL oxidative modification. It has been shown that the PC hydrolysis and Lyso-PC production occurring during LDL oxidation [17,18] is mediated through the LDL- associated PAF-AH [19,20]. The Lyso-PC content of oxidized LDL is thought to be one of the molecules responsible for the various biological effects reported for oxidized LDL [48,49]. Various studies have shown that like LDL, Lp(a) can be oxidatively modified in vitro [12,13], though to the best of our knowledge there is no study describing the possible Lyso-PC production during Lp(a) oxidation. Our results showed that during Lp(a) oxidation, a relatively high amount of Lyso-PC is produced, and thus Lp(a) resembles LDL in this property. This phenomenon can not be attributed to the LCAT activity, since it is not influenced by heating Lp(a) at 60°C for 1 h. By contrast, Lyso- PC production is completely inhibited by DFP, indicating that it might be mainly mediated through the Lp(a)-associated PAF-AH (as it is observed for LDL). Additionally, we studied the PAF-AH activity as a function of the degree of the Lp(a) oxidation. During oxidation the enzyme activity is progressively decreased and 4 h from the onset of the oxidation (decomposition phase) only 23.6”/0 of the enzyme activity was retained. Furthermore, during oxidation the apparent K,, of the enzyme remained constant, whereas V,,,;,, was significantly decreased. We can therefore pos- tulate that the affinity of the enzyme for PAF is conserved during oxidation, an evidence that there is not substrate dilution by the oxidized phospholipids formed during oxidation, whereas the decrease in V,;,, reflects the decrease of the amount of the active enzyme present in Ox-Lp(a). This non-competitive inhibition of the enzyme activity agrees with that observed by other investi- gators for the LDL-associated enzyme during oxidation in vitro [50].


In conclusion, our results show that the distribution of PAF-AH on Lp(a) studied by density gradient ultracentrifugation of plasma lipoproteins is affected by the Lp(a) as well as the dense LDL levels. Lp(a) contains a relatively high amount of PAF-AH activity, which is inversely related to the ape(a) isoform size. The PAF-AH kinetic constants in small isoforms are significantly higher compared to large ones sug- gesting that ape(a) affects the association of the enzyme with Lp(a) but does not bind itself to PAF-AH, since only 15 +_ 5% of the enzyme dissociates from Lp(a) upon ape(a) removal. It is of special interest that except PAF-AH, this lipoprotein carries a relatively high amount of tissue factor pathway inhibitor (TFPI) [51]. Thus, Lp(a) in its native state may exert both anti-inflammatory and anti-coagulant effects, un- like its thrombogenic properties, which are due to the high homology of its ape(a) content to plasminogen [9]. However, when Lp(a) enters the arterial intima, it binds to proteoglycans tightly and during its long stay it can be oxidatively modified [ 1 1 - 131, losing progressively its anti-inflammatory properties because of the decrease of the PAF-AH activity. Furthermore, Ox-Lp(a) becomes potentially a bioactive and cytotoxic particle, since the hydrolysis of the endogenous oxidized PC produces a relatively high amount of Lyso-PC, a reaction that is possibly mediated through the Lp(a)-associated PAF-AH.

Acknowledgements

We gratefully acknowledge Dr. Jtielle Thillet INSERM U321, Paris, France for valuable sug- gestions and assistance in the experiments for reductive cleavage of Lp(a) and for determination of the Lp(a) isoforms. We also thank Dr. J. Theodorou and Dr. Ch. Pappas for providing the blood samples and Mr. C. Tzallas for measuring Lp(a). This study was partially supported by grants from the Greek Ministry of Research and Technology (PENED 91, 369 pro- gramme).

References

[I] Armstrong VW, Cremer P, Eberle E, Manke A, Schulze F, Wieland H, Kreuzer H, Seidel M. The association between serum Lp(a) concentrations and angiographically assessed coronary atherosclerosis. Dependence on serum LDL levels. Atherosclerosis 1986;62:249.

[2] Dahlen GH, Guyton J, Attar J, Farmer J, Kautz J, Gotto AM Jr. Association of levels of lipoprotein Lp(a), plasma lipids and other lipoproteins with coronary artery disease documented by angiography. Circulation 1986;74:758.

[3] Schaefer EJ, Lamon-Fava S, Jenner JL, McNamara JR, Ordovas JM, Davis CE, Abolafia JM, Lippel K, Levy RI. Lipoprotein (a) levels and risk of coronary heart disease in men: The Lipid Research Clinics Coronary Primary Prevention Trial. J Am Med Assoc 1994;271:999.

[4] Koltringer P, Jiirgens G. A dominant role of lipoprotein (a) in the investigation and evaluation of parameters indicating the development of cervical atherosclerosis. Atherosclerosis 1985;58:187.

[5] Zenker G, Kiiltringer P, Bone G, Niederkom K, Pfeiffer Kp, Jiirgens G. Lipoprotein (a) as a strong indicator for cerebra-vascular disease. Stroke 1986;17:942.

[6] Gaubatz JW, Heidelman C, Gotto AM Jr, Morrisett JD, Dahlen GH. Human plasma lipoprotein (a) structural properties. J Biol Chem 1983;285:4582.

[7] Utermann G, Weber W. Protein composition of Lp(a) lipoprotein from human plasma. FEBS Lett 1983;154:357.

[8] Fless GM, Rolih CA, Scanu AM. Heterogeneity of human plasma lipoprotein (a). J Biol Chem 1984;259: 11470.

[9] Scanu AM, Fless GM. Lipoprotein (a): heterogeneity and biological relevance. J Clin Invest 1990;85:1709.

[lo] Edelberg JM, Pizzo SV. Lipoprotein (a): the link between impaired fibrinolysis and atherosclerosis, Fibrinolysis 1991;5:135.

[ll] Beisiegel U. Lipoprotein (a) in the arterial wall. Curr Opin Lipidol 1991;2:317.

[12] Naruszewicz M, Selinger E, Davignon J. Oxidative modification of Lp(a) and the effect of p-carotene. Metabolism 1992;41: 1.

[13] Sattler W, Kostner GM, Waeg G, Esterbauer H. Oxida- tion of lipoprotein Lp(a). A comparison with low-density lipoproteins. Biochim Biophys Acta 1991;1081:65.

[14] Jurgens G, Ashy A, Esterbauer H. Detection of new epitopes formed upon oxidation of low-density lipoprotein, lipoprotein (a) and very-low density lipoprotein. Use of an anti-serum against 4-hydroxynoneonal- modified low-density lipoproteins. Biochem J 1990;265:605.

[15] De Rijke YB, Jiirgens G, Hessels EMAJ, Hermann A, van Berkel TJC. In vivo fate and scavenger receptor recognition of oxidized lipoprotein (a) isoforms in rats. J Lipid Res 1992;33:1315.

[16] Witztum JL, Steinberg D. Role of oxidized low density lipoprotein in atherogenesis. J Clin Invest 1991;88:1785.

[17] Steinbrecher UrsP, Parthasarathy S, Leake DS, Witztum JL, Steinberg D. Modification of low density lipoprotein by endothelial cells involves lipid peroxidation and degra-

S.A.P. Kurabinu et ul. /) Atherosclerosis 125 11996) I21 -134 133

[I81

s91

P-01

1211

PI

~231

v41

P51

WI

[271

WI

[291

[301

[311

dation of low density lipoprotein phospholipides. Proc Nat1 Acad Sci 1984;81:3883. Parthasarathy S, Steinbrecher UrsP, Barnett J, Witztum JL, Steinberg D. Essential role of phospholipase A2 activity in endothelial cell-induced modification of low density lipoprotein. Proc Nat1 Acad Sci 1985;82:3000. Steinbrecher UP, Pritchard PH. Hydrolysis of phosphatidylcholine during LDL oxidation is mediated by platelet-activaiing factor acetylhydrolase. J Lipid Res 1989;30:305. Liapikos ThA, Antonopoulou S, Karabina S-AP, Tsoukatos DC, Demopoulos CA, Tselepis AD. Platelet- activating factor formation during oxidative modification of low-density lipoprotein when PAF-acetylhydrolase has been inactivat,ad. Biochim Biophys Acta 1994:1212:353. Stafforini DM, McIntyre TM, Carter ME, Prescott SM. Human plasma platelet-activating factor acetylhydrolase. Association with lipoprotein particles and role in the degradation of platelet-Activating factor. J Biol Chem 1987;262:4215. Stafforini DM, Prescott SM, Zimmerman GA, McIntyre TM. Platelet-activating factor acetylhydrolase activity in human tissues and blood cells. Lipids 1991;26:979. Stremler KE. Stafforini DM, Prescott SM. McIntyre TM. Human plasma platelet-activating factor acetylhydrolase. Oxidatively fragmented phospholipids as substrates. J Biol Chem 1991;266:11095. Stafforini DM, Prescott SM. McIntyre TM. Human plasma platelet-activating factor acetylhydrolase. Purifica- tion and properties. J Biol Chem 1987;262:4223. Knipping G, Birchbauer A, Steyrer E, Groener J, Zechner R, Kostner GM. Studies on the substrate specificity of human and pig 1ecithin:cholesterol acyltransferase: role of low-density lipoproteins. Biochemistry 1986;25:5242. Tselepis AD. Dentan C, Karabina S-AP, Chapman MJ, Ninio E. PAFdegrading acetylhydrolase is preferentially associated wtth dense LDL and VHDL-I in human plasma: catalytic characteristics and relation to the monocyte-derived enzyme. Arterioscler Thromb Vast Biol 1995;15:1764. Labeur C, Michiels G, Bury J, Usher. DC. Rosseneu M. Lipoprotein (a) quantified by an enzyme-linked immuno- sorbent assay! with monoclonal antibodies. Clin Chem 1989;35:1380. Nichols AL, Krduss RM, Musliner TA. Non-denaturing polyacrylamide gradient gel electrophoresis. In: Segrest JP. Albers JJ editors. Methods in Enzymology. Academic Press. London, 1986;128:417. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680. Kamhoh Ml. Ferrell RE, Kottke BA. Expressed hyper- variable polyaorphism of apolipoprotein (a). Am J Hum Genet 1991;49:1063. Doucet C, Huby T, Chapman J, Thillet J. Lipoprotein (a) in the chimpanzee: relationship of ape(a) phenotype to elevated plasma Lp(a) levels. J Lipid Res 1994:35:263.

[321

[33]

[341

[351

WI

[371

[381

[391

[401

[411

[42]

[431

[441

[451

1461

Utermann G, Menzel HJ. Kraft HG, Duba HC, Kemmler HG. Seitz C. Lp(a) glycoprotein phenotypes. Inheritance and relation to Lp(a)-lipoprotein concentrations in plasma. J Clin Invest 1987;80:458. Armstrong VW, Walli AK, Seidel D. Isolation, character- ization and uptake in human fibroblasts of an ape(a)-free lipoprotein obtained on reduction of lipoprotein(a). J Clin Invest 1985;26:1314. Karabina SAP. Liapikos ThA, Grekas G, Goudevenos J. Tselepis AD. Distribution of PAF-acetylhydrolase activity in human plasma low-density lipoprotein subfractions. Biochim Biphys Acta 1994: 12 I3:34. De Graaf J, Hak-Lemmers HLM, Hectors MPC, Demacker PNM, Hendriks JCM, Stalenhoef AFH. En- hanced susceptibility to in vitro oxidation of the dense low density lipoprotein subfraction in healthy subjects. Arterioscl Thromb 1991:11:298. Pinckard RN, Ludwig JC. Determination of PAF-2 acyl- hydrolase activity. Fed Proc 1986:45:856 (abs). Bligh EG. Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959;37:91 I. Lowry OH. Rosebrough NJ, Farr AL. Randall RJ. Protein measurement with the folin phenol reagent. J Biol Chem 195 1:193:265. Morrisett JD. Guyton JR, Gaubatz JW. Gotto AMJr. Lipoprotein (a): structure, metabolism and epidemiology. In: Gotto r\MJr, editor. Plasma lipoproteins. Elsevier, Amsterdam, 1987; 129. Terashita Z-I. Tsushima S. Yoshioka Y. Nomura H, Inada Y. Nishikawa K. CV:3988: a specitic antagonist of platelet-activating factor (PAF). Life Science I983:32: 1975. Scanu AM. Pfaftinger D, Fless GM, Makino K. Eisen- bart J. Hinman J. Attenuation of immunologic reactivity of lipoprotein (a) by thiols and cysteine-containing com- pounds Structural implications, Arterioscler Thromb 1992; 121424. Estcrbauer H, Gebicki J, Puhl H, Jurgens G. The role of lipid peroxidation and antioxodants in oxidative modih- cation of LDL. Free Rad Biol Med 1992;13:341. Liu M. Subbaiah P. Hydrolysis and transesterification of platelet-activating factor by lecithin-cholesterol acyltransferase. Proc Natl Acad Sci USA 1994;91:6035. Kraft HG. Sandholzer C, Menzel HJ. Uterman G. Apolipoprotein (a) alleles determine lipoprotein (a) particle density and concentration in plasma. .Arterioscler Thromb. 1992:12:302. Stafforini DM. Elstad MR, McIntyre TM. Zimmerman GA, Prescott SM. Human macrophages secrete platelet- activating factor acetylhydrolasc. J Biol Chem 1990:265:9682. Tjoelker LW. Wilder C. Eberhardt C. Stafforini DM. Dietsch G. Schimpf B, Hooper S. Trong HL. Cousens LS. Zimmerman GA. Yamada Y, McIntyre TM. Prescott SM. Gray PW. Anti-inflammatory properties of a platelet-activating factor acetylhydrolase. Nature 1995:374:549.

134 S.A.P. Karabina et al. / Atherosclerosis 125 (1996) 121-134

[47] Wang-Iverson P, Gibson JC, Brown WV. Plasma apolipoprotein secretion by monocyte-derived macrophages. Biochim Biophys Acta 1985;834:256.

[48] Quinn MT, Parthasarathy S, Steinberg D. Lysophos- phatidylcholine: a chemotactic factor for human monocytes and its potential role in atherogenesis. Proc Nat1 Acad Sci USA 1988;85:2805.

[49] Mangin EL, Kugiyama K, Nguy JH, Kerns SA, Henry PD. Effects of lysolipids and oxidatively modified low density lipoprotein on endothelium-dependent relaxation of rabbit aorta. Circ Res 1993:72:161.

[SO] Dentan C, Lesnik P, Chapman MJ, Ninio E. PAF- acether-degrading acetylhydrolase in plasma LDL is inac- tivated by copper- and cell-mediated oxidation. Arterioscler Thromb 1994;14:353.

[51] Lesnik P, Vonika A, Geuerin M, Moreau M, Chapman M J. Anti-coagulant activity of tissue factor pathway inhibitor in human plasma is preferentially associated with dense subspecies of LDL and HDL and with Lp(a). Arterioscler Thromb 1993;13:1066.

paf-acetylhydrolase activity on lp(a) before and during

Documents