Enhanced macrophage uptake of lipoprotein(a) after Ca2+-induced aggregate-formation.

We tested the hypothesis that aggregated lipoprotein(a) [Lp(a)] is avidly taken up by macrophages. Lp(a) was isolated by sequential centrifugations and gel chromatography from a patient with high plasma levels of Lp(a) who was being treated with low density lipoprotein (LDL)-apheresis. Aggregated Lp(a) was prepared by mixing native Lp(a) with 2.5 mmol/L CaCl2, and 54% of the 125I-Lp(a) aggregated after interacting with CaCl2. The binding and degradation of aggregated Lp(a) in macrophages were 4.6- and 4.7-fold higher than those of native Lp(a), respectively. An excess amount of LDL did not inhibit either increase. Cholesterol esterification in macrophages was markedly stimulated by aggregated Lp(a), and macrophages were transformed into foam cells. Cytochalasin B, a phagocytosis inhibitor, strongly inhibited the degradation and cholesterol esterification (78 and 83%, respectively). These findings suggested that aggregation may be partially involved in Lp(a) accumulation, thereby contributing to the acceleration of atherosclerosis.

Corticotropin increases the receptor-specific uptake of native low-density lipoprotein (LDL)--but not of oxidized LDL and native or oxidized lipoprotein(a) [Lp(a)]--in HEPG2 cells: no evidence for Lp(a) catabolism via the LDL-receptor.

To understand the interaction of corticotropin (ACTH) and lipid catabolism, we analyzed the influence of ACTH on receptor-mediated lipoprotein uptake and compared the uptake and degradation of human native (N-LDL) and oxidized (Ox-LDL) low-density lipoprotein and native (N-Lp(a)) and oxidized (Ox-Lp(a)) lipoprotein(a) by human hepatoma (HepG2) cells. The receptor affinity of N-LDL, Ox-LDL, N-Lp(a), and Ox-Lp(a) was comparable (Kd, 33, 13, 24, and 13 micrograms/mL medium), whereas the maximum degradative capacity was 10.5-fold higher in N-LDL (Vmax, 1,978 ng/mg cell protein) compared with Ox-LDL (189 ng/mg). In N-LDL, it was 4.5-fold higher than in N-Lp(a) (442 ng/mg) and eightfold higher than in Ox-Lp(a) (246 ng/mg) (P < .05). Addition of ACTH to the cell cultures increased receptor-specific degradation of N-LDL by 44% (2,866 v 1,978 ng/mg, P < .05), whereas changes in Ox-LDL, N-Lp(a), and Ox-Lp(a) showed no significant increase. No differences in uptake specificity were observed with or without ACTH. In addition, a 12-hour preincubation of liver cells with LDL increased Lp(a) uptake by 40% to 50% with (411 v 620 ng/mg) and without (393 v 558 ng/mg) ACTH administration. These data indicate that ACTH elevates receptor-specific uptake of N-LDL, but only to a low extent versus Ox-LDL, N-Lp(a), or Ox-Lp(a). These results support the hypothesis that catabolism of oxidized lipoproteins and Lp(a) through the LDL receptor pathway is only a minor route of lipid metabolism, whereas LDL receptor activity itself can be stimulated by ACTH.

Metabolism of Lp(a): assembly and excretion.

Lp(a) is one of the most atherogenic lipoproteins, and we know much more about the pathophysiology of Lp(a) than about its physiological function and metabolism. From our previous investigations and the new results reported here, we propose the following model of Lp(a) metabolism: apo(a) is biosynthesized in liver cells and the size of the isoform determines its rate of synthesis and excretion. Specific kringle-4 domains in apo(a), mainly T-6 and T-7, bind in a first step to circulating LDL, followed by the stabilization of the newly formed Lp(a) complex by a disulfide bridge. Circulating Lp(a) interacts specifically with kidney cells, or possibly other tissues, causing cleavage of 2/3-3/4 of the N-terminal part of apo(a) by a collagenase-type protease. Part of the apo(a) fragments is found in the urine, but there are indications that they in fact represent the biologically active form of apo(a). The core portion of Lp(a) in turn is cleared by the LDL-receptor or another specific binding system of the liver. Strategies for reducing plasma Lp(a) levels with medication should aim at interfering with the assembly of Lp(a) on one hand and the stimulation of apo(a) fragmentation on the other hand.

Functional and metabolic differences between elastase-generated fragments of human lipoprotein[a] and apolipoprotein[a].

We have previously shown that a functional free apolipoprotein[a] (apo[a]) can be isolated from its parent lipoprotein[a] (Lp[a]) by a mild reductive procedure. To shed further light on the properties of Lp[a] and apo[a] we subjected them to a limited proteolysis by porcine pancreatic elastase. This enzyme cleaved both at the Ile3520-Leu3521 bond in the linker between kringles IV-4 and IV-5 of apo[a] generating two fragments F1 and F2. In contrast to F1, which represented the N-terminal portion of apo[a] and was functionally inert, F2, representing the C-terminal domain, bound to lysine-Sepharose, fibrinogen, and fibronectin and formed a miniLp[a] particle when incubated with LDL. The proteolytic pattern by pancreatic elastase was also exhibited by human leukocyte elastase. F1, injected intravenously into normal mice, was rapidly cleared (Ty2, 2.9 h) and after 1 h fragments in the size range of 100-33 kDa were observed in the urine. In turn, F2 had a longer residence time (Ty2, 5 h) and was excreted in the urine only after 5 h as fragments of 70-45 kDa. Fragments in the same size range as found after F1 injection were also present in the urine after injection of apo[a] or Lp[a]. Moreover, apo[a] fragments of the size seen in mouse urine were spontaneously present in normal human urine and appeared derived from larger apo[a] fragments in the plasma. Our results indicate that enzymes of the elastase family cleave human apo[a] in vitro into two main fragments that differ in structural and functional properties and metabolic behavior. The comparable size of apo[a] fragments observed in the urine of humans and injected mice invites the speculation that enzymes of the elastase family may play a role in the biology of Lp[a] in vivo.

In vivo kinetics of lipoprotein(a) in homozygous Watanabe heritable hyperlipidaemic rabbits.

In vivo kinetics of lipoprotein(a) [Lp(a)] were investigated in homozygous Watanabe heritable hyperlipidaemic (WHHL) rabbits (an animal model of familial hypercholesterolemia (FH)), and in normolipidemic Japanese White rabbits (controls). 125I-labelled Lp(a) and 131I-labelled LDL were simultaneously injected intravenously. Blood samples were then taken periodically. Kinetic parameters were calculated from the plasma radioactivity decay curves. The fractional catabolic rates (FCRs) of both Lp(a) and LDL (1.355 +/- 0.189 pools per day and 1.278 +/- 0.397 pools per day, respectively) in the WHHL rabbits were significantly (P < 0.005) smaller than those in the control rabbits (2.008 +/- 0.083 pools per day and 2.855 +/- 0.759 pools per day, respectively). In WHHL rabbits, the FCRs of Lp(a) and LDL were similar. However, in control rabbits, the FCR of Lp(a) was significantly (P < 0.01) smaller than that of LDL. In WHHL rabbit organs, the mean ratio of 125I-Lp(a): 131I-LDL, 48 h after injection, normalized to the corresponding isotope ratio in plasma, were 1.525, 1.020, 1.819 and 1.967, in liver, kidney, spleen and bile, respectively. These values were significantly higher than the corresponding values in control rabbits (0.590, 0.677, 0.862 and 0.766, respectively). Our data strongly suggest that Lp(a) clearance is not entirely dependent upon LDL receptors and may be mediated by some other mechanisms.