Lipoprotein (a) and apoliprotein (a) isoform size in thyroid disease: the quest for the golden fleece.

Abstract

345 LIPOPROTEIN(A) [Lp(a)] concentration in plasma varies considerably between individuals because it is determined by a series of autosomal alleles at a single locus, encoding its specific protein marker apo-lipoprotein [apo(a)] (a) (1). Apo(a) is highly polymorphic, ranging in molecular weight from 300,000 to 700,000 (2). The polymorphism of apo(a) is complex with as many as 11 isoforms of the molecule having been detected (3). Individuals exhibiting only one isoform are considered homozygotes and those with two heterozygotes. The levels of Lp(a) are controlled more frequently by synthesis than by clearance rate and the observed variations in interindividual plasma Lp(a) levels are because of the differences in the size of the apo(a) glycoprotein (4). It is noteworthy that Lp(a) possesses a striking homology to plasminogen, a protein responsible for the lysis of blood clots; it competes for its binding sites on the endothelium and thus may promote atherosclerosis by interfering with thrombolysis (5). Various prospective epidemiologic studies have demonstrated an association between plasma Lp(a) concentration and coronary heart disease (CHD), whereas others have failed to implicate Lp(a) as a risk factor (6–8). Nevertheless, a strong correlation between Lp(a) and CHD in the presence of increased low-density lipoprotein (LDL) levels has been detected (9). Considering the physiologic role of Lp(a), it should be remembered that apo(a), rich in kringle domains and LDL and rich in cholesterol, are completely different functional systems (2,10). According to a provocative hypothesis, Lp(a) may bridge the two systems, thereby accomplishing functionality (11). Thus, Lp(a) interacting with fibrin, may deliver cholesterol to places of injury or it may be involved in platelet function. Hence, in overt hypothyroidism the usually elevated levels of LDL may be accompanied by elevated levels of Lp(a) suggesting increased atherogenesis (12). However, the influence of thyroid hormones on serum Lp(a) concentration is controversial and because of methodologic difficulties in measuring Lp(a) and to various dose schedules and duration of levothyroxine treatment, remains a matter of dispute (13,14). In this issue of Thyroid, Milionis et al., (15) report an interesting approach to the challenging problem of Lp(a) in patients with subclinical hypothyroidism (SH). They evaluated the serum Lp(a) levels and the apo(a) phenotypes in patients with SH. They found increased Lp(a) levels together with total cholesterol, LDL cholesterol and apo(B) in 69 patients with SH by comparison with 83 euthyroid controls. In contrast, no significant difference in the frequencies of apo(a) phenotypes could be detected between the two study groups. Lp(a) levels were higher in SH patients with high molecular weight (HMW) apo(a) isoform size than in the control group. Replacement treatment with levothyroxine induced an overall decrease of Lp(a), which was more evident in those patients with low molecular weight (LMW) apo(a) isoforms associated with increased baseline Lp(a) levels. However, there was no significant effect in patients with low to moderate Lp(a) levels suggesting that levothyroxine treatment may be beneficial in patients with SH with raised Lp(a) levels and LMW. Taken together, the results may indicate that the increased Lp(a) levels are not the result of genetic predisposition and that thyroid function may be important for Lp(a) metabolism. The findings are at variance with previous data in overt hypothyroidism showing increased small Lp(a) phenotype in patients by comparison with controls, whereas no effect of levothyroxine treatment on Lp(a) levels was registered (16). The discrepancies may be due more to the small number of patients studied and to the polymorphism of apo (a) and less to ethnic differences. There are inconsistent results in the few studies that have investigated Lp(a) levels in SH. Increased Lp(a) levels related to TSH, were mostly found in patients with SH but elevated Lp(a) levels, independent of thyrotropin (TSH) values, have also been reported (17,18). In the light of more recent data, Lp(a) levels were found increased in 25% of patients with SH and no effect of levothyroxine treatment over a 6-month period was registered (19). Furthermore, in the same study a significant association was found between elevated Lp(a) levels and a positive family history for CHD that would seem to indicate a genetic influence (19). In contrast, in another recent study from the Mayo Clinic, thyroid hormone replacement induced a significant decrease of Lp(a) levels resulting