Introduction:
Low-density lipoprotein cholesterol (LDL-C) is a causal risk factor for coronary artery disease (CAD). Rare mutations in at least 6 genes lead to Mendelian forms of high or reduced LDL-C; three (
APOB, LDLR, PCSK9
) act in a dominant pattern whereas three (
ABCG5, ABCG8, LDLRAP1
) in a recessive pattern. We address to what extent rare mutations in Mendelian LDL-C genes contribute to early CAD risk in the population.
Methods:
We sequenced the exons of the 6 genes in 9,329 early CAD cases (myocardial infarction, angiographic CAD, or coronary revascularization in men≤50 and women≤60) and 10,245 controls from 9 studies using targeted and whole exome next-generation sequencing. We tested 3 sets: ‘Null alleles’ (nonsense, splice-site, or frameshift); ‘Deleterious (7/7)’ (null and missense annotated as damaging by 7 algorithms); and ‘Deleterious (6/7)’ (null and missense annotated as damaging by at least 6 algorithms). Given the rarity of deleterious mutations, we aggregated these mutations in each gene and tested for an excess or deficit in cases
vs
. controls.
Results:
Counts of mutations are provided in Table. Null mutations in
LDLR
, carried by 1:500 participants, confered a 8-fold increase in CAD risk (P=8х10
-7
) whereas heterozygosity for a null mutation in
ABCG5
(1:650 frequency) was associated with a 3-fold increased risk (P=5х10
-3
). ‘Deleterious (7/7)’ mutations in
LDLR
, carried by 1:100 participants, confered a 4-fold increased risk (P=8х10
-17
) whereas heterozygosity for a ‘Deleterious (7/7)’ mutation in
ABCG5
(1:250 frequency) was associated with a 2-fold increased risk (P=2х10
-3
). Heterozygous null allele carriers at
LDLR
and
ABCG5
had increased LDL-C (P<0.001).
Conclusions:
Of early CAD cases, 2-3% carry a rare, deleterious mutation at
LDLR
or
ABCG5
associated with increased risk. Although previously reported to cause recessive sitosterolemia, we find that heterozygosity for a null allele at
ABCG5
is associated with markedly higher early CAD risk.