Apolipoprotein(a) [apo(a)] exhibits a genetic size polymorphism explaining about 40% of the variability in lipoprotein(a) [Lp(a)] concentration in Tyroleans. Lp(a) concentrations and apo(a) phenotypes were determined in 7 ethnic groups (Tyrolean, Icelandic, Hungarian, Malay, Chinese, Indian, Black Sudanese) and the effects of the apo(a) size polymorphism on Lp(a) levels were estimated in each group. Average Lp(a) concentrations were highly significantly different among these populations, with the Chinese (7.0 mg/dl) having the lowest and the Sudanese (46 mg/dl) the highest levels. Apo(a) phenotype and derived apo(a) allele frequencies were also significantly different among the populations. Apo(a) isoform effects on Lp(a) levels were not significantly different among populations. Lp(a) levels were however roughly twice as high in the same phenotypes in the Indians, and several times as high in the Sudanese, compared with Caucasians. The size variation of apo(a) explains from 0.77 (Malays) to only 0.19 (Sudanese) of the total variability in Lp(a) levels. Together these data show (I) that there is considerable heterogeneity of the Lp(a) polymorphism among populations, (II) that differences in apo(a) allele frequencies alone do not explain the differences in Lp(a) levels among populations and (III) that in some populations, e.g. Sudanese Blacks, Lp(a) levels are mainly determined by factors that are different from the apo(a) size polymorphism.

Elevated concentrations of lipoprotein(a) (Lp[a]) in plasma are associated with premature coronary heart disease (CHD). Lp(a) levels are largely determined by alleles at the hypervariable apolipoprotein(a) (apo[a]) gene locus, but other genetic and environmental factors as well as diseases also affect plasma Lp(a) concentrations. It is therefore unclear whether Lp(a) is a primary genetic risk factor or whether Lp(a) levels are elevated secondary to disease in CHD patients. We have analyzed apo(a) phenotypes that represent a stable genetic trait in subjects with CHD and control subjects from different populations representing a variety of ethnic groups (Tyrol, Germany, Wales, Israel, Singapore Chinese, and Singapore Indian). Despite differences in sampling design and disease definition in this multipopulation case-control study, those apo(a) isoforms associated with high Lp(a) plasma concentrations (B, S1, and S2) were more frequent in the CHD patients in each ethnic group. These differences were significant in three of the studied populations and highly significant (p < 0.001) in the pooled (total) group. Lp(a) concentrations were also measured in all groups except Germans and were found to be consistently higher in cases than in control subjects in each ethnic group. For all but one population (Israeli) the differences were significant. The effects of the apo(a) size polymorphism on Lp(a) levels were similar in CHD patients and control subjects from different populations. The data demonstrate that alleles at the apo(a) locus determine the risk for CHD through their effects on Lp(a) concentrations across multiple populations with large differences in CHD frequency and risk factor profiles.

Lipoprotein(a) [Lp(a)] is a quantitative trait in human plasma. Lp(a) consists of a low-density lipoprotein and the plasminogen-related apolipoprotein(a) [apo(a)]. The apo(a) gene determines a size polymorphism of the protein, which is related to Lp(a) levels in plasma. In an attempt to gain a deeper insight into the genetic architecture of this risk factor for coronary heart disease, we have investigated the basis of the apo(a) size polymorphism by pulsed field gel electrophoresis of genomic DNA employing various restriction enzymes (SwaI, KpnI, KspI, SfiI, NotI) and an apo(a) kringle-IV-specific probe. All enzymes detected the same size polymorphism in the kringle IV repeat domain of apo(a). With KpnI, 26 different alleles were identified among 156 unrelated subjects; these alleles ranged in size from 32 kb to 189 kb and differed by increments of 5.6 kb, corresponding to one kringle IV unit. There was a perfect match between the size of the apo(a) DNA phenotypes and the size of apo(a) isoforms in plasma. The apo(a) DNA polymorphism was further used to estimate the magnitude of the apo(a) gene effect on Lp(a) levels by a sib-pair comparison approach based on 253 sib-pairs from 64 families. Intra-class correlation of log-transformed Lp(a) levels was high in sib-pairs sharing both parental alleles (r = 0.91), significant in those with one common allele (r = 0.31), and absent in those with no parental allele in common (r = 0.12). The data show that the intra-individual variability in Lp(a) levels is almost entirely explained by variation at the apo(a) locus but that only a fraction (46%) is explained by the DNA size polymorphism. This suggests further heterogeneity relating to Lp(a) levels in the apo(a) gene.

Patients with terminal renal insufficiency suffer from an increased incidence of atherosclerotic diseases. Elevated plasma concentrations of lipoprotein(a) ILp(a)I have been established as a genetically controlled risk factor for these diseases. Variable alleles at the apo(a) gene locus determine to a large extent the Lp(a) concentration in the general population. In addition, other genetic and nongenetic factors also contribute to the plasma concentrations of Lp(a).We therefore investigated Apo(a) phenotypes and Lp(a) plasma concentrations in a large group of patients with endstage renal disease (ESRD) and in a control group. Lp(a) concentrations were significantly elevated in ESRD patients (20.1±20.3 mg/dl) as compared with the controls (12.1±15.5 mg/dl, P < 0.001). However, no difference was found in apo(a) isoform frequency between the ESRD group and the controls. Interestingly, only patients with large size apo(a) isoforms exhibited two-to fourfold elevated levels of Lp(a), whereas the small-size isoforms had similar concentrations in ESRD patients and controls. Beside elevated Lp(a) concentrations, ESRD patients had lower levels of plasma cholesterol and apolipoprotein B.These results show that elevated Lp(a) plasma levels might significantly contribute to the risk for atherosclerotic diseases in ESRD. They further indicate that nongenetic factors related to renal insufficiency or other genes beside the apo(a) structural gene locus must be responsible for the high Lp(a) levels. (J. Clin. Invest. 1993. 91:397-401.) Key words: lipoprotein(a) -apolipoprotein(a) phenotypes * terminal renal insufficiency -end-stage renal disease * atherosclerosis