Hypercholesterolemia, Familial, 1

A number sign (#) is used with this entry because familial hypercholesterolemia-1 (FHCL1) can be caused by heterozygous, compound heterozygous, or homozygous mutation in the low density lipoprotein receptor gene (LDLR; 606945) on chromosome 19p13.

Description

Familial hypercholesterolemia is an autosomal dominant disorder characterized by elevation of serum cholesterol bound to low density lipoprotein (LDL), which promotes deposition of cholesterol in the skin (xanthelasma), tendons (xanthomas), and coronary arteries (atherosclerosis). The disorder occurs in 2 clinical forms: homozygous and heterozygous (Hobbs et al., 1992).

The FHCL1 phenotype can be modified by mutation in other genes. For example, in individuals with the LDLR mutation IVS14+1G-A (606945.0063), the phenotype can be altered by a SNP in the APOA2 gene (107670.0002), a SNP in the EPHX2 gene (132811.0001), or a SNP in the GHR gene (600946.0028).

Genetic Heterogeneity of Familial Hypercholesterolemia

Other forms of monogenic familial hypercholesterolemia include FHCL2 (144010), caused by mutation in the APOB gene (107730); FHCL3 (603776), caused by mutation in the PCSK9 gene (607786); and FHCL4 (603813), caused by mutation in the LDLRAP1 gene (605747).

Clinical Features

Individuals with heterozygous familial hypercholesterolemia develop tendinous xanthomas, corneal arcus, and coronary artery disease; the last usually becomes evident in the fourth or fifth decade. Homozygous individuals have a more severe clinical picture with earlier presentation, usually in the first 2 decades of life (Hobbs et al., 1992).

The ranges of serum cholesterol and LDL-cholesterol are, in mg per dl, 250-450 and 200-400 in heterozygotes, greater than 500 and greater than 450 in homozygous affecteds, and 150-250 and 75-175 in homozygous unaffecteds, with some positive correlation with age (Khachadurian, 1964; Kwiterovich et al., 1974).

In homozygous familial hypercholesterolemia, the aortic root is prone to develop atherosclerotic plaque at an early age. Such plaques can accumulate in unusual sites, such as the ascending aorta and around the coronary ostia. Summers et al. (1998) evaluated the aortic root using MRI imaging in a blinded, prospective study of 17 homozygous FH patients and 12 healthy controls. When patient age and body mass index were taken into account, 53% of patients with homozygous FH had increased aortic wall thickness compared to controls; this was thought to result from a combination of medial hyperplasia and plaque formation. Supravalvular aortic stenosis was seen in 41% of patients.

Houlston et al. (1988) studied the relationship of lipoprotein(a) (152200) levels and coronary heart disease in patients with familial hypercholesterolemia. Individuals with coronary artery disease had a significantly higher mean lipoprotein(a) concentration than those without coronary heart disease, suggesting that lipoprotein(a) measurements may help predict the risk of coronary heart disease in individuals with familial hypercholesterolemia.

Deramo et al. (2003) investigated the relationship between nonarteritic ischemic optic neuropathy (NAION; 258660) and serum lipid levels in 37 consecutive patients diagnosed with NAION at or below age 50 years and 74 age- and gender-matched controls. They found that hypercholesterolemia was a risk factor in these patients and suggested that NAION might be the first manifestation of a previously unrecognized lipid disorder. The patients had experienced a focal, microvascular central nervous system ischemic event at a relatively young age. Deramo et al. (2003) suggested that aggressive treatment of lipid abnormalities might be warranted in these patients.

Pathogenesis

By studies of cultured fibroblasts from homozygotes, Goldstein and Brown (1973) and Brown and Goldstein (1974) showed that the basic defect concerns the cell membrane receptor for LDL. Normally, LDL is bound at the cell membrane and taken into the cell ending up in lysosomes where the protein is degraded and the cholesterol is made available for repression of microsomal enzyme 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA) reductase, the rate-limiting step in cholesterol synthesis. In familial hypercholesterolemia, there is a binding defect due to a dysfunctional receptor. At the same time, a reciprocal stimulation of cholesterol ester synthesis takes place. Harders-Spengel et al. (1982) presented evidence that the receptor defect is present on liver membranes.

To determine the influences of intrauterine and genetic factors on atherogenic lipid profiles in later life, Ijzerman et al. (2001) investigated 53 dizygotic and 61 monozygotic adolescent twin pairs. They found an association between low birth weight and high levels of total cholesterol, LDL cholesterol, and apolipoprotein B that persisted in the intrapair analysis in dizygotic twin pairs but was reversed within monozygotic twin pairs. Furthermore, they found that the association between low birth weight and low levels of HDL cholesterol tended to persist in the intrapair analysis in both dizygotic and monozygotic twins. These data suggested that genetic factors may account for the association of low birth weight with high levels of total cholesterol, LDL cholesterol, and apolipoprotein B, whereas intrauterine factors possibly play a role in the association of low birth weight with low levels of HDL cholesterol.

Garcia-Otin et al. (2007) determined serum noncholesterol sterols in normolipidemic control subjects and in well-phenotyped patients with dyslipidemias, including autosomal dominant hypercholesterolemia (ADH) with and without known genetic defects and familial combined hyperlipidemia (FCHL; 144250). Intestinal cholesterol absorption was highest in ADH without known defect, followed by ADH with known defect, then controls, and then FCHL. Garcia-Otin et al. (2007) concluded that intestinal cholesterol absorption might play a role in the lipid abnormalities of patients with autosomal dominant hypercholesterolemia without identified genetic defects. They suggested that serum noncholesterol sterols are a useful tool for the differential diagnosis of genetic hypercholesterolemias.

Diagnosis

Humphries et al. (1985) found a RFLP of the LDL receptor gene using the restriction enzyme PvuII. About 30% of persons are heterozygous for the polymorphism which is, therefore, useful in family studies and early diagnosis of FHC. Schuster et al. (1989) also used RFLPs of the LDLR gene in the diagnosis of FH.

Bhatnagar et al. (2000) reported a case-finding experience in the UK among relatives of patients with familial hypercholesterolemia by a nurse-led genetic register. By performing cholesterol tests on 200 relatives, 121 new patients with familial hypercholesterolemia were discovered. The newly diagnosed patients were younger than the probands and were generally detected before they had clinically overt atherosclerosis. A case was made for organizing a genetic register approach, linking lipid clinics nationally.

Umans-Eckenhausen et al. (2001) found that in the Netherlands targeted family screening with DNA analysis proved to be highly effective in identifying patients with hypercholesterolemia. Most of the identified patients sought treatment and were successfully started on cholesterol-lowering treatment to lower the risk of premature cardiovascular disease.

Newson and Humphries (2005) discussed cascade testing in familial hypercholesterolemia. They questioned whether and how family members should be contacted for testing. The implications of the test results for life planning, employment, or ability to obtain life insurance are concerns. The pros and cons of cascade testing were reviewed by de Wert (2005).

Prenatal Diagnosis

Vergotine et al. (2001) demonstrated the feasibility of prenatal diagnosis of homozygous familial hypercholesterolemia in the Afrikaner population.

Clinical Management

Starzl et al. (1984) performed both heart transplant and liver transplant in a 6.75-year-old girl with homozygous familial hypercholesterolemia.

Tonstad et al. (1996) conducted a double-blind placebo-controlled trial over 1 year using 8 grams of cholestyramine in prepubertal children (aged 6-11 years) with familial hypercholesterolemia. After 1 year of a low-fat, low-cholesterol diet, children with a family history of premature cardiovascular disease had LDL cholesterol levels at or greater than 4.9 mmol/liter, while children without such a family history had LDL cholesterol levels at or greater than 4.1 mmol/liter. The LDL cholesterol levels in the test group lowered by 16.9% (95% confidence interval), compared with a 1.4% increase in the placebo group. Growth velocity was not adversely affected in the treatment group, although folate and 25-hydroxyvitamin D deficiency were noted among a small number of treated children. Additionally, a boy who had an appendectomy 3 months before the study required surgery for intestinal obstruction after he had taken the first 2 cholestyramine doses. Given the number of gastrointestinal side effects, Tonstad et al. (1996) recommended caution in starting cholestyramine after abdominal surgery in children.

Cuchel et al. (2007) treated 6 patients with homozygous familial hypercholesterolemia with an inhibitor of microsomal triglyceride transfer protein (157147). A reduction of LDL cholesterol levels was observed, owing to reduced production of apolipoprotein B. However, the therapy was associated with elevated liver aminotransferase levels and hepatic fat accumulation.

Statin Therapy

The 'statin' drugs are potent competitive inhibitors of 3-hydroxy-3-methylglutaryl coenzyme-A reductase and have proven useful in the treatment of hypercholesterolemia (Betteridge et al., 1978; Goldstein and Brown, 1987; Hoeg and Brewer, 1987). Brorholt-Petersen et al. (2001) tested the hypothesis that the cholesterol lowering effect of statin therapy is a function of the particular type of LDLR mutation. They studied the response to treatment with fluvastatin in 28 patients with heterozygous FH as a result of a receptor-negative mutation (trp23 to ter; 606945.0060) and in 30 patients with a receptor-binding defective mutation (trp66 to gly; 606945.0003). They found no statistically significant differences. A tabulation of the results of this and earlier studies suggested that differences in treatment response as an apparent function of LDLR gene mutation type occur mainly in populations with recent genetic admixture. The authors suggested that in such populations persons with the same mutation in the LDLR gene are also more likely to share other but undetermined genetic variations affecting the pharmacology of statins.

Chaves et al. (2001) examined the presence of mutations in the LDLR gene among subjects clinically diagnosed with FH and analyzed whether the molecular diagnosis helped to predict the response to simvastatin treatment in their FH population. They conducted a randomized clinical trial with simvastatin in 42 genetically diagnosed subjects with FH, with 22 classified as carriers of null mutations and 20 with defective mutations. A mutation causing FH was identified in 46 probands (84%). In 41 of them (89%), a total of 28 point mutations were detected, 13 of which had not been previously described. FH with null mutations showed a poor response to simvastatin treatment. The mean percentage reduction of plasma total and LDL cholesterol levels in these subjects was significantly lower than in subjects with defective mutations. Subjects with FH with null mutations (class I) showed lower plasma HDL cholesterol values and a poor LDL cholesterol response to simvastatin treatment.

Hedman et al. (2005) studied the efficacy and safety for up to 2 years of pravastatin treatment in 19 girls and 11 boys with autosomal dominant familial hypercholesterolemia. Pravastatin was started at 10 mg/d, with a forced titration by 10 mg at 2, 4, 6, and 12 months until the target cholesterol level of less than 194 mg/dl was reached. By 2, 4, 6, 12, and 24 months of treatment, the total cholesterol levels had, respectively, decreased by 19, 20, 23, 27 and 26%, and the LDL cholesterol levels had decreased by 25, 27, 29, 33, and 32% compared with the dietary baseline values. The side effects were mild, and no clinically significant elevations in alanine aminotransferase, creatine kinase, or creatinine were seen. The authors concluded that the efficacy in children with slight or moderate hypercholesterolemia was satisfactory, but in children with severe hypercholesterolemia it was insufficient.

Gene Therapy

Wilson et al. (1992) presented a detailed clinical protocol for the ex vivo gene therapy of familial hypercholesterolemia. The approach, which they proposed to use to treat homozygous FH patients with symptomatic coronary artery disease who have a relatively poor prognosis but can tolerate a noncardiac surgical procedure with acceptable risks, involves recovery of hepatocytes from the patient and reimplanting them after genetic correction by a retrovirus-mediated gene transfer. Not only were the technical details of vectors and viruses, transduction and delivery of hepatocytes, evaluation of engraftment and rejection, etc., discussed, but also assessment of risks versus benefits and informed consent for both adult and child patients.

Grossman et al. (1994) reported a 29-year-old woman with FH caused by mutation in the LDLR gene (606945.0003) who underwent hepatocyte-directed ex vivo gene therapy with LDLR-expressing retroviruses. She tolerated the procedures well, liver biopsy after 4 months showed engraftment of the transgene, and there was no clinical or pathologic evidence for autoimmune hepatitis. The patient showed an improvement in serum lipids up to 18 months after the treatment.

Mapping

Three independent linkage studies, by Ott et al. (1974), Berg and Heiberg (1976), and Elston et al. (1976), strongly suggested loose linkage between familial hypercholesterolemia and the third component of complement; C3 (120700) has been mapped to chromosome 19 by somatic cell hybridization. Donald et al. (1984) presented further data on HC-C3 linkage, bringing the combined male-female lod score to a maximum of 3.79 at theta 0.25. C3 and FHC are about 20 cM apart; APOE (107741) and C3 are about 15 cM apart. FHC is not closely linked to APOE, suggesting that these 2 loci are on opposite sides of C3. The LDLR gene was regionalized to 19p13.1-p13.3 by in situ hybridization (Lindgren et al., 1985). Judging by the sequence of loci suggested by linkage data (pter--FHC--C3--APOE/APOC2), the location of FHC (LDLR) is probably 19p13.2-p13.12 and of C3, 19p13.2-p13.11.

Leppert et al. (1986) found tight linkage between a RFLP of the LDL receptor gene and dominantly inherited hypercholesterolemia; specifically, no exception to cosegregation was found between high-LDL cholesterol phenotype and a unique allele at the LDLR locus. The maximum lod score was 7.52 at theta = 0.

In 3 adult Dutch, Swedish, and Australian twin samples totaling 410 dizygotic twin pairs, Beekman et al. (2003) found consistent evidence for linkage between chromosome 19 and LDL cholesterol levels, with maximum lod scores of 4.5, 1.7, and 2.1, respectively. No linkage was observed in an adolescent Dutch twin sample of 83 dizygotic twin pairs. Combined analysis of the adult samples increased the maximum lod to 5.7 at 60 cM from pter. Beekman et al. (2003) concluded that there is strong evidence for the presence of a QTL on chromosome 19 with a major effect on LDL cholesterol levels.

Molecular Genetics

Horsthemke et al. (1987) analyzed DNA from 70 patients in the UK with heterozygous familial hypercholesterolemia. In most, the restriction fragment pattern of the LDLR gene was indistinguishable from the normal; however, 3 patients were found to have a deletion of about 1 kb in the central portion of the gene. In 2 patients, the deletion included all or part of exon 5 (606945.0027); in the third, the deletion included exon 7 (606945.0033). Including a previously described patient with a deletion in the 3-prime part of the gene, these results indicated that 4 of 70 patients, or 6%, have deletions.

Hobbs et al. (1990) reviewed the many mutations found in the LDLR gene as the cause of familial hypercholesterolemia.

Varret et al. (2008) reviewed 17 published studies of autosomal dominant hypercholesterolemia and evaluated the contribution of mutations in the LDLR, APOB, and PCSK9 genes. They noted that the proportion of subjects without an identified mutation ranged from 12% to 72%, suggesting the existence of further genetic heterogeneity.

In a patient diagnosed with probable heterozygous FH, Bourbon et al. (2007) analyzed the LDLR gene and identified a novel variant initially assumed to be a silent polymorphism (R385R; 606945.0065); however, analysis of mRNA from the patient's cells showed that the mutation introduces a new splice site predicted to cause premature termination of the protein. The R385R mutation was also found in a Chinese homozygous FH patient.

Defesche et al. (2008) analyzed the LDLR gene in 1,350 patients clinically diagnosed with familial hypercholesterolemia who were negative for functional DNA variation in the LDLR, APOB (107730), and PCSK9 (607786) genes. The authors examined the effects of 128 seemingly neutral exonic and intronic DNA variants and identified 2 synonymous variants in LDLR, R385R and G186G (606945.0066), that clearly affected splice sites and segregated with hypercholesterolemia in all families examined. The R385R variant was found in 2 probands of Chinese origin, whereas G186G was found in 35 Dutch probands, 2 of whom were homozygous for the variant and had a more severe phenotype, with myocardial infarction occurring in both before the age of 20 years.

Kulseth et al. (2010) performed RNA analysis in 30 unrelated patients with clinically defined hypercholesterolemia but without any LDLR mutations detected by standard DNA analysis; sequencing of RT-PCR products from an index patient revealed an insertion of 81 bp from the 5-prime end of intron 14 of LDLR, and DNA sequencing of exons 13 and 14 detected an splice site mutation in intron 14 (606945.0067). Twelve of 23 family members tested were heterozygous for the mutation, and carriers had significantly increased total cholesterol levels compared to noncarriers. Kulseth et al. (2010) analyzed an additional 550 index patients and identified the same splice site mutation in 3 more probands. In 1 proband's family, the mutation was found in 6 of 7 tested family members, who all had LDL cholesterol levels above the 97th percentile.

Do et al. (2015) sequenced the protein-coding regions of 9,793 genomes from patients with myocardial infarction (MI) at an early age (50 years or younger in males and 60 years or younger in females) along with MI-free controls. They identified 2 genes in which rare coding-sequence mutations were more frequent in MI cases versus controls at exomewide significance: LDLR (606945) and APOA5 (606368). Carriers of rare nonsynonymous mutations in LDLR were at 4.2-fold increased risk for MI, while carriers of null alleles in LDLR were at even higher risk (13-fold difference). Approximately 2% of early MI cases harbor a rare, damaging mutation in LDLR; this estimate is similar to one made by Goldstein et al. (1973) using an analysis of total cholesterol. Among controls, about 1 in 217 carried an LDLR coding-sequence mutation and had plasma LDL cholesterol greater than 190 mg/dl. Carriers of rare nonsynonymous mutations in APOA5 were at 2.2-fold increased risk for MI. When compared with noncarriers, LDLR mutation carriers had higher plasma LDL cholesterol, whereas APOA5 mutation carriers had higher plasma triglycerides (see 145750). Evidence has connected MI risk with coding-sequence mutations at 2 genes functionally related to APOA5, namely lipoprotein lipase (LPL; 609708) and apolipoprotein C-III (APOC3; 107720). Do et al. (2015) concluded that LDL cholesterol as well as disordered metabolism of triglyceride-rich lipoproteins contributes to myocardial infarction risk.

Associations Pending Confirmation

A SNP in the promoter region of the G-substrate gene (GSBS; 604088.0001) correlated with elevated plasma total cholesterol levels.

A SNP in intron 17 of the ITIH4 gene (600564.0001) was associated with hypercholesterolemia susceptibility in a Japanese population.

Genotype/Phenotype Correlations

Goldstein et al. (1977) found that both receptor-absent and receptor-defective mutants occur and they concluded that some of the 'homozygotes' are in fact genetic compounds. An internalization mutant of the LDL receptor binds LDL but is unable to facilitate passage of LDL to the inside of the cell. A patient was found to be a genetic compound, having inherited the internalization mutant from the father and the binding mutant from the mother. From the fact that an individual was shown by family studies to be a genetic compound and that complementation did not occur, Goldstein et al. (1977) concluded that the gene for binding of LDL and the gene for internalization of LDL are allelic mutations at the structural locus for the LDL receptor. Miyake et al. (1981) found homozygosity for the internalization defect.

The LDL receptor is synthesized as a 120-kD glycoprotein precursor that undergoes change to a 160-kD mature glycoprotein through the covalent addition of a 40-kD protein. Tolleshaug et al. (1982) reported a heterozygous child who inherited one allele from his mother which produced an abnormal 120-kD protein that could not be further processed, and one allele from his father which produced an elongated 170-kD precursor that underwent an increase in molecular weight to form an abnormally large receptor of 210 kD.

Levy et al. (1986) reported 2 brothers with a unique genetic compound form of 'homozygous' hypercholesterolemia in which the mother had typical FHC and the father and 3 of his close relatives had what they termed the HMWR (high molecular weight receptor) trait. In these persons 2 types of functional LDL receptors were found in cultured skin fibroblasts: one with molecular weight of 140,000 and one with molecular weight of 176,000. Curiously and puzzlingly, the compound heterozygotes and the regular heterozygotes for the HMWR showed increased cholesterol synthesis, which the authors suggested may play a significant role in the pathology of the disease.

Funahashi et al. (1988) studied 16 Japanese kindreds with homozygous FHC. Ten had a receptor-negative form of the disease; 5 had a receptor-defective form; and 1 represented an internalization defect. The receptor-defective group, in which residual amounts of functional receptors were produced, showed a lower tendency to coronary artery disease than the receptor-negative group.

Modifiers

Feussner et al. (1996) described a 20-year-old man with a combination of heterozygous FH caused by splice mutation (606945.0054) and type III hyperlipoproteinemia (107741). He presented with multiple xanthomas of the elbows, interphalangeal joints and interdigital webs of the hands. Active lipid-lowering therapy caused regression of the xanthomas and significant decrease of cholesterol and triglycerides. Flat xanthomas of the interdigital webs were described in 3 of 4 formerly reported patients with a combination of these disorders of lipoprotein metabolism. Feussner et al. (1996) proposed that the presence of these xanthomas should suggest compound heterozygosity (actually double heterozygosity) for FH and type III hyperlipoproteinemia.

Sass et al. (1995) described a 4-generation French-Canadian kindred with familial hypercholesterolemia in which 2 of the 8 heterozygotes for a 5-kb deletion (exons 2 and 3) in the LDLR gene were found to have normal LDL-cholesterol levels. Analyses showed that it was unlikely that variation in the genes encoding apolipoprotein B (107730), HMG-CoA reductase (HMGCR; 142910), apoAI-CIII-AIV (see APOA1; 107680), or lipoprotein lipase was responsible for the cholesterol-lowering effect. Expression of the LDL receptor, as assessed in vitro with measurements of activity and mRNA levels, was similar in normolipidemic and hyperlipidemic subjects carrying the deletion. Analysis of the apoE isoforms (107741), on the other hand, revealed that most of the E2 allele carriers in this family, including the 2 normolipidemic 5-kb deletion carriers, had LDL cholesterol levels substantially lower than subjects with the other apoE isoforms. Thus, this kindred provided evidence for the existence of a gene or genes, including the apoE2 allele, with profound effects on LDL-cholesterol levels.

Vergopoulos et al. (1997) presented findings suggesting the existence of a xanthomatosis susceptibility gene in a consanguineous Syrian kindred containing 6 individuals with homozygous FH (see 602247). Half of the homozygotes had giant xanthomas, while half did not, even though their LDL-cholesterol concentrations were elevated to similar degrees (more than 14 mmol/l). Heterozygous FH individuals in this family were also clearly distinguishable with respect to xanthoma size. By DNA analysis they identified a hitherto undescribed mutation in the LDLR gene in this family: a T-to-C transition at nucleotide 1999 in codon 646 of exon 14, resulting in an arginine for cysteine substitution. Segregation analysis suggested that a separate susceptibility gene may explain the formation of giant xanthomas.

In a 13-year-old girl with severe hypercholesterolemia, Ekstrom et al. (1999) demonstrated compound heterozygosity for a cys240-to-phe mutation (606945.0059) and a tyr167-to-ter mutation (606945.0045) in the LDLR gene. Her 2 heterozygous sibs also carried the C240F mutation, but only one of them was hypercholesterolemic. The authors concluded that there may be cholesterol-lowering mechanisms that are activated by mutations in other genes.

Knoblauch et al. (2000) studied an Arab family that carried the tyr807-to-cys substitution (606945.0019). In this family, some heterozygous persons had normal LDL levels, while some homozygous individuals had LDL levels similar to those persons with heterozygous FH. The authors presented evidence for the existence of a cholesterol-lowering gene on 13q (604595).

Takada et al. (2002) demonstrated that a SNP of the promoter of the APOA2 gene, -265T-C (107670.0002), influenced the level of total cholesterol and low density lipoprotein (LDL) cholesterol in members with the IVS14+1G-A mutation (606945.0063) in the LDLR gene causing hypercholesterolemia. Strikingly lower total cholesterol and LDL cholesterol values were observed among most of the LDLR mutation carriers who were simultaneously homozygous for the -265C allele of the APOA2 gene.

In the same large family reported by Takada et al. (2002), Takada et al. (2003) found that a SNP in the GHR gene, resulting in a L526I (600946.0028) substitution, influenced plasma levels of high density lipoprotein (HDL) cholesterol in affected family members with the IVS14+1G-A mutation. The lowest levels of plasma HDL were observed among leu/leu homozygotes, highest levels among ile/ile homozygotes, and intermediate levels among leu/ile heterozygotes. No such effect was observed among noncarriers of the LDLR mutation.

In the pedigree reported by Takada et al. (2002), Sato et al. (2004) demonstrated a significant modification of the phenotype of familial hypercholesterolemia resulting from the IVS14+1G-A mutation by the arg287 variation in the EPHX2 gene (132811.0001).

Population Genetics

In most populations the frequency of the homozygote is 1 in a million (probably a minimal estimate, being a prevalence figure rather than incidence at birth) and the frequency of heterozygotes not less than 1 in 500. Thus, heterozygous familial hypercholesterolemia is the most frequent mendelian disorder, being more frequent than either cystic fibrosis or sickle cell anemia which, in different populations, are often given that distinction. Among survivors of myocardial infarction, the frequency of heterozygotes is about 1 in 20.

Seftel et al. (1980) pointed to a high frequency of hypercholesterolemic homozygotes in South Africa. In a 7-year period, 34 homozygotes were seen in one clinic in Johannesburg. All were Afrikaners and most lived in Transvaal Province. The authors calculated the frequency of heterozygotes and homozygotes to be 1 in 100 and 1 in 30,000, respectively. The oldest of their patients was a 46-year-old woman. Of the 34, six were age 30 or older. The authors concluded that the high frequency of the gene is attributable to founder effect, as in the case of porphyria variegata (176200), lipoid proteinosis (247100), and sclerosteosis (269500). Torrington and Botha (1981) found that 20 of 26 families with FHC (77%) belonged to the Gereformeerde Kerk, whereas according to the 1970 census only 5% of the Afrikaans-speaking white population of South Africa belonged to this religious group. Again, the data were consistent with a founder effect. Using the LDLR activity of lymphocytes, Steyn et al. (1989) calculated the prevalence of heterozygous FHC in the permanent residents of a predominantly Afrikaans-speaking community in South Africa to be 1 in 71--the highest prevalence reported to date.

In the Saguenay-Lac-Saint-Jean region of Quebec Province, De Braekeleer (1991) estimated the prevalence of familial hypercholesterolemia as 1/122, compared to the usually used frequency of 1/500 for European populations.

Defesche and Kastelein (1998) stated that more than 350 different mutations had been found in patients with familial hypercholesterolemia. They tabulated the preferential geographic distribution that has been demonstrated for some of the LDL receptor mutations. For example, in the West of Scotland about half of the index cases of FH were found to have the cys163-to-tyr mutation (606945.0058). Defesche and Kastelein (1998) commented on the geographic associations of LDL receptor mutations within the Netherlands.

Deletion of gly197 (606945.0005) is the most prevalent LDL receptor mutation causing familial hypercholesterolemia in Ashkenazi Jewish individuals. Studying index cases from Israel, South Africa, Russia, the Netherlands, and the United States, Durst et al. (2001) found that all traced their ancestry to Lithuania. A highly conserved haplotype was identified in chromosomes carrying this deletion, suggesting a common founder. When 2 methods were used for analysis of linkage disequilibrium between flanking polymorphic markers and the disease locus and for the study of the decay of LD over time, the estimated age of the deletion was found to be 20 +/- 7 generations, so that the most recent common ancestor of the mutation-bearing chromosomes would date to the 14th century. This corresponds with the founding of the Jewish community of Lithuania (1338 A.D.), as well as with the great demographic expansion of Ashkenazi Jewish individuals in eastern Europe, which followed this settlement. Durst et al. (2001) could find no evidence supporting a selective evolutionary metabolic advantage. Therefore, the founder effect in a rapidly expanding population from a limited number of families remains a simple, parsimonious hypothesis explaining the spread of this mutation in Ashkenazi Jewish individuals.

Animal Model

Kingsley and Krieger (1984) identified 4 different types of mutant Chinese hamster ovary cells with defective LDL receptor function. One locus, called ldlA, apparently represents the structural gene for LDL receptor, whereas the others--ldlB, ldlC, and ldlD--appear to have defects involved in either regulation, synthesis, transport, recycling, or turnover of LDL receptors.

The Watanabe heritable hyperlipidemic (WHHL) rabbit has a genetic deficiency of LDL receptors and is therefore a superb experimental model (Hornick et al., 1983). Kita et al. (1987) found that probucol prevented the progression of atherosclerosis in the Watanabe rabbit by limiting oxidative LDL modification and foam cell transformation of macrophages. Probucol was originally developed as an antioxidant. Yamamoto et al. (1986) showed that the defect in the Watanabe heritable hyperlipidemic rabbit is a mutant receptor for LDL that is not transported to the cell surface at a normal rate. Cloning and sequencing of complementary cDNAs from normal and Watanabe rabbits showed that the defect arises from an in-frame deletion of 12 nucleotides that eliminates 4 amino acids from the cysteine-rich ligand binding domain of the LDL receptor. Yamamoto et al. (1986) detected a similar mutation by S1 nuclease mapping of LDL receptor mRNA from a patient with familial hypercholesterolemia whose receptor also failed to be transported to the cell surface. These findings suggested to Yamamoto et al. (1986) that animal cells may have fail-safe mechanisms that prevent surface expression of improperly folded proteins with unpaired or improperly bonded cysteine residues.

Scanu et al. (1988) investigated hypercholesterolemia due to deficiency of the LDL receptor in a family of rhesus monkeys. Hummel et al. (1990) used PCR to analyze the mutation carried by members of a family of rhesus monkeys with spontaneous hypercholesterolemia and low density lipoprotein receptor deficiency. Affected monkeys are heterozygous for a nonsense mutation in exon 6, changing codon 284 from TGG to TAG. The G-to-A transition creates a new SpeI restriction site. LDLR RNA is reduced by about 50% on quantitative analysis of RNA obtained at liver biopsy in affected animals.

Hofmann et al. (1988) found that overexpression of LDL receptors caused elimination of both apoE and apoB, the 2 ligands, from the plasma in transgenic mice derived from fertilized eggs injected with the LDLR gene under control of the mouse metallothionein-I promoter. They speculated that overexpression of other receptors, such as those for insulin (147670) or transferrin (190000), might have pathologic effects leading to a 'ligand steal' syndrome.

Chowdhury et al. (1991) used the Watanabe rabbit for the development of liver-directed gene therapy based on transplantation of autologous hepatocytes that had been genetically corrected ex vivo with recombinant retroviruses. Animals transplanted with LDLR-transduced autologous hepatocytes demonstrated a 30 to 50% decrease in total serum cholesterol that persisted for the duration of the experiment (122 days). Recombinant-derived LDLR RNA was harvested from tissues with no diminution for up to 6.5 months after transplantation. Ishibashi et al. (1993) developed a new animal model for homozygous FH through targeted disruption of the LDLR gene in mice. Homozygous LDL receptor-deficient mice showed delayed clearance of VLDL, intermediate density lipoproteins (IDL), and LDL from plasma. As a result, total plasma cholesterol level rose from 108 mg/dl in wildtype mice to 236 mg/dl in homozygous deficient mice. Adult mice did not exhibit gross evidence of xanthomatosis, however, and the extent of aortic atherosclerosis was minimal. On the other hand, Ishibashi et al. (1994) showed that in mice homozygous for the targeted disruption of the LDLR gene who were fed a diet high in cholesterol, total plasma cholesterol rose from 246 to more than 1,500 mg/dl. In wildtype littermates fed the same diet, total plasma cholesterol remained less than 160 mg/dl. After 7 months, the homozygous deficient mice developed massive xanthomatous infiltration of the skin and subcutaneous tissue. The aorta and coronary ostia exhibited gross atheromata, and the aortic valve leaflets were thickened by cholesterol-laden macrophages.

Mice homozygous for targeted replacement with human APOE2 (107741.0001), regardless of age or gender, develop type III hyperlipoproteinemia, whereas homozygosity for APOE2 results in HLP in no more than 10% in humans, predominantly in adult males. By generating mice homozygous for human APOE2 and heterozygous for human LDLR and mouse Ldlr, Knouff et al. (2001) achieved increased stability of mRNA in liver associated with a truncation of the 3-prime-UTR of LDLR. Plasma lipoprotein levels were normal in the LDLR heterozygotes. Knouff et al. (2001) concluded that moderate and controlled overexpression of LDLR completely ameliorates the type III HLP phenotype of APOE2 homozygous mice.

Hasty et al. (2001) generated mice deficient in both the low density lipoprotein receptor and leptin (ob/ob). These doubly mutant mice exhibited striking elevations in both total plasma cholesterol and triglyceride levels and had extensive atherosclerotic lesions throughout the aorta by 6 months of age. Although fasting, diet restriction, and low-level leptin treatment significantly lowered total plasma triglyceride levels, they caused only slight changes in total plasma cholesterol levels. Hepatic cholesterol and triglyceride contents as well as mRNA levels of cholesterologenic and lipogenic enzymes suggested that leptin deficiency increased production of hepatic triglycerides, but not cholesterol, in the ob/ob mice regardless of their Ldlr genotype. These data provided evidence that the hypertriglyceridemia and hypercholesterolemia in the doubly mutant mice were caused by distinct mechanisms, suggesting that leptin might have some impact on plasma cholesterol metabolism, possibly through an LDLR-independent pathway.

History

Much of the early nosologic work that established the hyperlipoproteinemia phenotype (Fredrickson et al., 1967) and suggested familial occurrence (Hould et al., 1969; Schrott et al., 1972; Kwiterovich et al., 1974) was done before the extensive genetic heterogeneity of the phenotype was defined.