PAF-acetylhydrolase activity of Lp(a) before and during Cu(2+)-induced oxidative modification in vitro.

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 = 1.050-1.100 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 LDL5 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 +/- 9 mumol/l and Vmax = 94 +/- 5 nmol/mg protein per min, and isoform S2/S3 had apparent Km = 36 +/- 9 mumol/l and Vmax = 25 +/- 5 nmol/mg protein per min. Removal of apolipoprotein (a) (apo(a)) from Lp(a) by reductive cleavage with dithiothreitol, slightly affected the amount of PAF-AH existing on Lp(a) since, only 15 +/- 5% of the total enzyme activity dissociated from its particle after density gradient ultracentrifugation. During Cu(2+)-induced Lp(a) oxidation, the PAF-AH activity decreased from 10.90 +/- 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 Km of the enzyme remained essentially unchanged during oxidation, whereas Vmax was significantly decreased from 58.6 +/- 7.8 nmol/mg protein per min to 38.2 +/- 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 +/- 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 of PAF-AH among the other lipoproteins. Apo(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.