EFFECTS OF DELETION OF THE CARBOXYL-TERMINAL DOMAIN OF APOA-I OR OF ITS SUBSTITUTION WITH HELICES OF APOA-II ON IN-VITRO AND IN-VIVO LIPOPROTEIN ASSOCIATION

In the present study, the lipoprotein association of apoA-I, an apoA-I (Delta Ala(190)-Gln(243)) deletion mutant and an apoA-I (Asp(1)-Leu(1 89))/apoA-II (Ser(12)-Gln(77)) chimera were compared. At equilibrium, 80% of the I-125-labeled apolipoproteins associated with lipoproteins in rabbit or human plasma but with very different distribution profile s. High density lipoprotein (HDL),(2,3)-associated fractions were 0.60 for apoA-I, 0.30 for the chimera, and 0.15 for the deletion mutant, a nd corresponding very high density lipoprotein-associated fractions we re 0.20, 0.50, and 0.65. Clearance curves after intravenous bolus inje ction of I-125-labeled apolipoproteins (3 mu g/kg) in normolipemic rab bits could be adequately fitted with a sum of three exponential terms, yielding overall plasma clearance rates of 0.028 +/- 0.0012 ml . min( -1) for apoA-I (mean +/- S.E.; n = 6), 0.10 +/- 0.008 ml . min(-1) for the chimera (p < 0.001 versus apoA-I) and 0.38 +/- 0.022 ml . min(-1) for the deletion mutant (p < 0.001 versus apoA-I and versus the chime ra). Fractions that were initially cleared with a t1/2 of 3 min, most probably representing free apolipoproteins, were 0.30 +/- 0.04, 0.50 /- 0.06 (p = 0.02 versus apoA-I), and 0.64 +/- 0.07 (p = 0.002 versus apoA-I), respectively. At 20 min after the bolus, the fractions of inj ected material associated with HDL(2,3) were 0.55 +/- 0.06, 0.25 +/- 0 .03 (p = 0.001 versus apoA-I), and 0.09 +/- 0.01 (p < 0.001 versus apo A-I and versus the chimera), respectively, whereas the fractions assoc iated with very high density lipoprotein were 0.15 +/- 0.006, 0.25 +/- 0.03 (p = 0.008 versus apoA-I), and 0.27 +/- 0.03 (p = 0.003 versus a poA-I), respectively. The ability of the different apolipoproteins to bind to HDL, particles and displace apoA-I in vitro were compared, The molar ratios at which 50% of I-125-labeled apoA-I was displaced from the surface of HDL, particles were 1:1 for apoA-I, 3:1 for the chimera and 12:1 for the deletion mutant, indicating 3- and 12-fold reduction s of the affinities for HDL, of the chimera and the deletion mutant, r espectively. These data suggest that the carboxyl-terminal pair of hel ices of apoA-I are involved in the initial rapid binding of apoA-I to the lipid surface of HDL. Although the lipid affinity of apoA-II is hi gher than that of apoA-I, substitution of the carboxyl-terminal helice s of apoA-I with those of apoA-II only partially restores its lipoprot ein association. Thus, this substitution may affect cooperative intera ctions with the middle amphipathic helices of apoA-I that are critical for its specific distribution over the different HDL species.