Norum et al. (1980, 1982) studied 2 sisters, aged 30 and 25, with very low HDL and heart failure from coronary artery disease. Both had arcus cornealis, xanthelasmata and extensive infiltrative xanthoma of the neck and antecubital fossa, resembling somewhat the changes of pseudoxanthoma elasticum. The skin histology showed collections of lipid-laden histiocytes. Plasma cholesterol was 177 and 135 mg/dl; HDL cholesterol was 4 and 7 mg/dl. Only traces of apoprotein A-I were detected in whole plasma; in addition, apoprotein C-III was not detectable. The parents and children of the 2 women had low HDL cholesterol and apoA-I levels consistent with heterozygosity. Low levels of HDL cholesterol concentration have been associated with an increased frequency of coronary artery disease even when HDL is no less than 50% of normal (Miller and Miller, 1975). Heart failure without myocardial infarction is unusual in coronary atherosclerosis, especially in young women, suggesting small vessel disease. The patient of Gustafson et al. (1979), although clinically similar, differed by having high apoC-III rather than absent apoC-III.
Karathanasis et al. (1983) showed that the probands in the family of Norum et al. (1982) were both homozygous for a defect in the apoA-I locus, namely, an insertion in an intron. They could identify heterozygotes unequivocally. The parents had the same gene defect; they were not known to be related but both had ancestors of Scottish extraction who lived in the Appalachian mountain region of southeastern Kentucky. When McKusick saw the 2 sisters in 1983, he was impressed that the xanthomatosis of the neck and antecubital fossae simulated the changes of PXE (177850, 264800). The obligatory heterozygotes may be at increased risk of atherosclerosis. Norum and Alaupovic (1984) pointed out that although the only lesion demonstrated is the insertion in the apoA-I gene, the finding of reduced concentrations of both A-I and C-III in heterozygotes suggests that the apoC-III deficiency in the homozygotes is not secondary but due either to mutation also in the apoC-III gene or to an effect of the apoA-I gene on the cis apoC-III gene. Either hypothesis suggests linkage of the 2 loci. Norum (1983) suggested that the gene for apolipoprotein C-II may be in the same cluster on chromosome 11 because it, like C-III, was severely deficient in the 2 sisters. Karathanasis et al. (1983) studied the genomic sequences flanking the APOA1 gene and found that the APOC3 gene (see 107720) lies about 2.6 kb downstream of the 3-prime end of the APOA1 gene. They also showed that the 2 genes are 'convergently transcribed' and that the polymorphism reported by Rees et al. (1983) to be associated with hypertriglyceridemia may be due to a single basepair substitution in the 3-prime-noncoding region of apoC-III mRNA. Forte et al. (1984) cited evidence that the 6.5-kb insert in the APOA1 gene is deleted from its normal position in the promoter region for the closely linked APOC3 gene. Protter et al. (1984) isolated and characterized the APOC3 gene. The coding sequence was found to be interrupted by 3 introns. The authors compared it with the APOA1 gene and sequenced the DNA lying between the 2 genes. Karathanasis et al. (1986) studied the restriction pattern of the APOA4 gene in the sisters with combined apoA-I and apoC-III deficiency. Although apoA-IV had not been demonstrated in the plasma of these patients, the relatively high levels of plasma LCAT activity (40% of normal) and the possible involvement of apoA-IV in LCAT activation suggested that the APOA4 gene of these patients is functionally normal. Karathanasis et al. (1987) demonstrated that these patients had a rearrangement in the form of an inversion containing portions of the 3-prime ends of the APOA1 and APOC3 genes, including the DNA between these genes. The breakpoints were located within the fourth exon of the APOA1 gene and the first intron of the APOC3 gene. The fusion gene was expressed as a fusion mRNA.
