Lipoprotein(a) [Lp(a)] is a quantitative genetic trait that in the general
population is largely controlled by 1 major locus-the locus for the apolipo
protein(a) [apo(a)] gene. Sibpair studies in families including familial de
fective apolipoprotein B or familial hypercholesterolemia (FH) heterozygote
s have demonstrated that, in addition, mutations in apolipoprotein B and in
the LDL receptor (LDL-R) gene may affect Lp(a) plasma concentrations, but
this issue is controversial. Here, we have further investigated the influen
ce of mutations in the LDL-R gene on Lp(a) levels by inclusion of FH homozy
gotes. Sixty-nine members of 22 families with FH were analyzed for mutation
s in the LDL-R as well as for apo(a) genotypes, apo(a) isoforms, and Lp(a)
plasma levels. Twenty-six individuals were found to be homozygous for FH, a
nd 43 were heterozygous for FH. As in our previous analysis, FH heterozygot
es had significantly higher Lp(a) than did non-FH individuals from the same
population. FH homozygotes with 2 nonfunctional LDL-R alleles had almost 2
-fold higher Lp(a) levels than did FH heterozygotes. This increase was not
explained by differences in apo(a) allele frequencies. Phenotyping of apo(a
) and quantitative analysis of isoforms in family members allowed the assig
nment of Lp(a) levels to both isoforms in apo(a) heterozygous individuals.
Thus, Lp(a) levels associated with apo(a) alleles that were identical by de
scent could be compared. In the resulting 40 allele pairs, significantly hi
gher Lp(a) levels were detected in association with apo(a) alleles from ind
ividuals with 2 defective LDL-R alleles compared with those with only 1 def
ective allele. This difference of Lp(a) levels between allele pairs was pre
sent across the whole size range of apo(a) alleles. Hence, mutations in the
LDL-R demonstrate a clear gene-dosage effect on Lp(a) plasma concentration
s.