A nonsense mutation in the apolipoprotein A-I gene is associated with high-density lipoprotein deficiency and periorbital xanthelasmas

Inborn errors of apolipoprotein A-I metabolism: implications for disease, research and development

PURPOSE OF REVIEW

We review current knowledge regarding naturally occurring mutations in the human apolipoprotein A-I (APOA1) gene with a focus on their clinical complications as well as their exploitation for the elucidation of structure-function-(disease) relationships and therapy.

RECENT FINDINGS

Bi-allelic loss-of-function mutations in APOA1 cause HDL deficiency and, in the majority of patients, premature atherosclerotic cardiovascular disease (ASCVD) and corneal opacities. Heterozygous HDL-cholesterol decreasing mutations in APOA1 were associated with increased risk of ASCVD in several but not all studies. Some missense mutations in APOA1 cause familial amyloidosis. Structure-function-reationships underlying the formation of amyloid as well as the manifestion of amyloidosis in specific tissues are better understood. Lessons may also be learnt from the progress in the treatment of amyloidoses induced by transthyretin variants. Infusion of reconstituted HDL (rHDL) containing apoA-I (Milano) did not cause regression of atherosclerosis in coronary arteries of patients with acute coronary syndrome. However, animal experiments indicate that rHDL with apoA-I (Milano) or apoA-I mimetic peptides may be useful for the treatment of heart failure of inflammatory bowel disease.

SUMMARY

Specific mutations in APOA1 are the cause of premature ASCVD or familial amyloidosis. Synthetic mimetics of apoA-I (mutants) may be useful for the treatment of several diseases beyond ASCVD.

[Inborn errors of high-density lipoprotein metabolism]

Both low and very high levels of high-density lipoprotein cholesterol (HDL-C) increase the risk of atherosclerotic cardiovascular disease (ASCVD) and shorten life expectancy. Low and high levels of HDL‑C are often caused by underlying diseases, lifestyle or medication, which should primarily be excluded. Much less frequently, monogenic diseases due to mutations in the APOA1, ABCA1 and LCAT genes are the cause of very low or unmeasurable HDL‑C levels or in the CETP, LIPC and SCARB1 genes for very high HDL‑C values. Genetic and detailed biochemical diagnostics should be considered, especially in cases of absolute HDL deficiency, early onset ASCVD or the presence of clinical symptoms or laboratory values characteristic for deficiencies of apolipoprotein A‑I (ApoA-I), lecithin cholesterol acyltransferase (LCAT) or Tangier disease. These included corneal opacities, xanthomas, large tonsils, hepatomegaly, peripheral neuropathy, proteinuria, anemia or thrombocytopenia. Sequencing of the APOA1 gene should also be considered in familial amyloidosis. There is no specific treatment for monogenic HDL diseases. Cholesterol and blood pressure lowering are indicated for the prevention of cardiovascular and renal complications.

Genetic control of high density lipoprotein-cholesterol in AcB/BcA recombinant congenic strains of mice

Epidemiological studies show that high HDL-cholesterol (HDLc) decreases the risk of cardiovascular disease. To map genes controlling lipid metabolism, particularly HDLc levels, we screened the plasma lipids of 36 AcB/BcA RC mouse strains subjected to either a normal or a high-fat/cholesterol diet. Strains BcA68 and AcB65 showed deviant HDLc plasma levels compared with the parental A/J and C57BL/6J strains; they were thus selected to generate informative F2 crosses. Linkage analyses in the AcB65 strain identified a locus on chromosome 4 (Hdlq78) responsible for high post-high fat diet HDLc levels. This locus has been previously associated at genome-wide significance to two regions in the human genome. A second linkage analysis in strain BcA68 identified linkage in the vicinity of a gene cluster known to control HDLc levels. Sequence analysis of these candidates identified a de novo, loss-of-function mutation in the ApoA1 gene of BcA68 that prematurely truncates the ApoA1 protein. The possibility of dissecting the specific effects of this new ApoA1 deficiency in the context of isogenic controls makes the BcA68 mouse a valuable new tool.

Complete Apo AI deficiency in an Iraqi Mandaean family: case studies and review of the literature

Complete apo A1 deficiency is a rare genetic disorder that has been associated with premature atherosclerosis. We describe a family of Iraqi Mandaean background with complete apo A1 deficiency caused by a new nonsense mutation in the APOA1 gene. Interestingly, there were marked differences in the clinical presentation of the two homozygotes in this family. A 35-year-old woman presented with xanthelasmas and xanthomas but showed only minimal changes on cardiovascular examinations and no clinical symptoms. However, her 37-year-old brother was diagnosed with myocardial infarction at age 35. In addition, both the homozygotes had elevated C-reactive protein levels. The C-reactive protein levels increased three-fold during pregnancy, then decreased postpartum and further decreased with statin treatment. Cholesterol ester transfer protein mass was close to the upper reference range, whereas the activity was low, likely because of the lack of the substrate. Here, we characterize the phenotype and genotype of the first Middle Eastern family with apo A1 deficiency and compare and contrast the findings in the two homozygous siblings and review the previously reported cases of apo A1 deficiency.

A novel nonsense apolipoprotein A-I mutation (apoA-IE136X) causes low HDL cholesterol in French Canadians

The molecular causes of severe high-density lipoprotein cholesterol (HDL-C) deficiency was examined in a group of 54 unrelated French Canadian subjects. The lecithin:cholesterol acyl transferase (LCAT) and apolipoprotein (apo) A-I gene were analyzed in all probands by direct DNA sequencing. While no LCAT mutation was detected, a novel nonsense apoA-I mutation (E136X) was found in 3/54 probands. Genetic analysis of two kindreds showed a strong co-segregation of the apoA-I locus with the low HDL-C trait. The E136X mutation was detected in families by MaeI restriction digestion. E136X carriers (n=17) had marked HDL-C deficiency; among the nine carriers > or = 35 years old, five men had developed premature coronary artery disease (CAD). A peptide of apparent molecular weight of 14 kDa was identified in fresh plasma, the HDL fractions and lipoprotein deficient plasma from the three probands but not in normal controls (n=3), suggesting that the mutant apoA-I peptide is secreted and binds lipids. The mutation was not observed in an additional 210 chromosomes from unrelated subjects of French Canadian descent, < 60 years of age, with CAD and low HDL-C levels. We conclude that apoA-I (E136X) is a cause of HDL-C deficiency in the French Canadian population and is associated with premature CAD.

Differential diagnosis of familial high density lipoprotein deficiency syndromes

Monogenic high density lipoprotein (HDL) deficiency, because of defects in the genes of apolipoprotein A-I (apoA-I), adenosine triphosphate binding cassette transporter A1 (ABCA1) or lecithin:cholesterol acyltransferase (LCAT), can be assumed in patients with HDL cholesterol levels below the fifth percentile within a given population. As in a first step underlying diseases should be excluded. Patients with a virtual absence of HDL must undergo careful physical examination to unravel the clinical hallmarks of certain HDL deficiency syndromes. In addition, family studies should be initiated, to demonstrate the vertical transmission of the low HDL cholesterol phenotype. Definitive diagnosis requires specialized biochemical tests and the demonstration of a functionally-relevant mutation in one of the three discussed candidate genes. As yet no routinely used drug is able to increase HDL cholesterol levels in patients with familial low HDL cholesterol so that prevention of cardiovascular disease in these patients must be focused on the avoidance and treatment of additional risk factors.

Recurrent mutations of the apolipoprotein A-I gene in three kindreds with severe HDL deficiency

Two siblings with high density lipoprotein (HDL) deficiency and no plasma apolipoprotein A-I (Apo A-I) were found to be homozygous for a cytosine deletion in exon 3 of Apo A-I gene (c.85 del C, Q5FsX11). This mutation causes a frameshift leading to a premature stop codon and abolishes the synthesis of Apo A-I. Although both siblings had corneal opacifications and planar xanthomas, only one of them had premature coronary artery disease, probably as the result of mildly elevated LDL levels. In two other unrelated subjects HDL deficiency was due to heterozygosity for a nucleotide substitution in exon 4 of Apo A-I gene (c.494 T>G, L141R). Both Apo A-I mutations were reported previously in an Italian kindred which included compound heterozygotes and simple heterozygotes. We investigated all carriers of these mutations in the three kindreds and in the one previously reported. Plasma Apo A-I and HDL-C levels were lower in the mutation carriers than in non-carrier family members. These levels, however, were lower in L141R carriers than in carriers of c.85 del C. Haplotype analysis performed using several polymorphisms suggested that both the c.85 del C and L141R are likely to be recurrent mutations.

Apolipoprotein A-I deficiency with accumulated risk for CHD but no symptoms of CHD

We evaluated a 69-year-old Japanese woman with apolipoprotein (apo) A-I deficiency, high levels of low-density lipoprotein (LDL)-cholesterol, hypertension and impaired glucose tolerance. The patient had corneal opacity, but neither xanthomas, xanthelasma, nor tonsillar hypertrophy. She was not symptomatic for coronary heart disease (CHD), and had normal electrocardiograms at rest and exercise using a cycle ergometer. She had severely reduced levels of high-density lipoprotein (HDL)-cholesterol (0.10-0.18 mmol/l) and no apo A-I (<0.6 mg/dl). LDL-cholesterol and apo B as well as apo E were increased even under treatment with 10 mg pravastatin per day. Gel filtration chromatography revealed that in addition to VLDL and LDL fractions, she had apo A-II rich and apo E rich fractions, which were present in the HDL fraction separated by ultracentrifugation. A cytosine deletion was identified by genomic DNA sequencing of the apo A-I gene of the patient at the third base of codon 184 in the fourth exon, which led to a frame shift mutation and early termination at codon 200. This patient is the oldest among those with apo A-I deficiency reported in the literature, and she had no symptoms of CHD despite the accumulated risk for the disease.

Quantitative trait locus mapping of genes that regulate HDL cholesterol in SM/J and NZB/B1NJ inbred mice

To investigate the quantitative trait loci (QTL) regulating plasma cholesterol, the female progeny of an (SMxNZB/ B1NJ)xNZB/B1NJ backcross were fed an atherogenic diet. After 18 wk, plasma total cholesterol and high-density lipoprotein cholesterol (HDL-C) was measured. HDL-C concentrations were greater in NZB than in SM mice. For standard chow-fed mice, QTL were found near D5Mit370 and D18Mit34. For mice fed an atherogenic diet, a QTL was found near D5Mit239. The QTL for chow-fed and atherogenic-fed mice on chromosome 5 seem to be two different loci. We used a multitrait analysis to rule out pleiotropy in favor of a two-QTL hypothesis. Furthermore, the HDL-C in these strains was induced by the high-fat diet. For inducible HDL-C, one significant locus was found near D15Mit39. The gene for an HDL receptor, Srb1, maps close to the HDL-C QTL at D5Mit370, but the concentrations of Srb1 mRNA and SR-B1 protein and the gene sequence of NZB/B1NJ and SM/J did not support Srb1 as a candidate gene. With these QTL, we have identified chromosomal regions that affect lipoprotein profiles in these strains.

The effects of altered apolipoprotein A-I structure on plasma HDL concentration

Approximately 46 human apolipoprotein A-I (apoA-I) coding sequence mutations have been reported to date. Roughly half of these mutations are associated with lower than average plasma concentrations of high-density lipoprotein (HDL) apoA-I. Mutations associated with low HDL apoA-I concentrations fall into two main categories: those which poorly activate the enzyme lecithin:cholesterol acyltransferase (LCAT) and those associated with amyloidosis. These phenotypically distinct groups of mutations are uniquely localized in different regions of the apoprotein sequence. Mutations associated with abnormal LCAT activation are located within repeats 5, 6, and 7, corresponding to amino acids 121 to 186, while many of the mutations found in amyloid deposits are clustered at the amino terminus of the protein, namely residues 1 to 90. These observations strongly support the idea that the tertiary structure of apoA-I determines its intravascular fate and ultimately the steady state concentration of plasma HDL.

Enhanced fractional catabolic rate of apo A-I and apo A-II in heterozygous subjects for apo A-I(Zaragoza) (L144R)

We have recently reported a new apolipoprotein (apo) A-I variant (apo A-I(Zaragoza) L144R) in a Spanish family with HDL-C levels below the 5th percentile for age and sex and low apo A-I concentrations. All the apo A-I(Zaragoza) subjects were heterozygous and none of them showed evidence of coronary artery disease (CAD). Mean plasma HDL-C, apo A-I, and apo A-II levels were lower in apo A-I(Zaragoza) carriers as compared to control subjects (40, 60, and 50%, respectively). Lipid composition analysis revealed that apo A-I(Zaragoza) carriers had HDL particles with a higher percentage of HDL triglyceride and a lower percentage of HDL esterified cholesterol as compared to those of control subjects. Lecithin:cholesterol acyltransferase (LCAT) activity and cholesterol esterification rate of apo A-I(Zaragoza) carriers were normal. Apo A-I and apo A-II metabolic studies were performed on two heterozygous apo A-I(Zaragoza) carriers and on six control subjects. We used a primed constant infusion of [5,5,5-2H3]leucine and HDL apo A-I and apo A-II tracer/tracee ratios were determined by gas chromatography mass spectrometry and fitted to a monoexponential equation using SAAM II software. Both subjects carrying apo A-I(Zaragoza) variant showed mean apo A-I fractional catabolic rate (FCR) values more than two-fold higher than mean FCR values of their controls (0.470+/-0.0792 vs. 0.207+/-0.0635 x day(-1), respectively). Apo A-I secretion rate (SR) of apo A-I(Zaragoza) subjects was slightly increased compared with controls (17.32+/-0.226 vs. 12.76+/-3.918 mg x kg(-l) x day(-1), respectively). Apo A-II FCR was also markedly elevated in both subjects with apo A-I(Zaragoza) when compared with controls (0.366+/-0.1450 vs. 0.171+/-0.0333 x day(-1), respectively) and apo A-II SR was normal (2.31+/-0.517 vs. 2.1+/-0.684 mg x kg(-l) x day(-1), respectively). Our results show that the apo A-I(Zaragoza) variant results in heterozygosis in abnormal HDL particle composition and in enhanced catabolism of apo A-I and apo A-II without affecting significantly the secretion rates of these apolipoproteins and the LCAT activation.

Cholesterol efflux from normal and Tangier disease fibroblasts into normal, high-density lipoprotein-deficient, and apolipoprotein E-deficient plasmas

Tangier disease (TD) fibroblasts have defective cholesterol release in the presence of lipid-free apolipoproteins. We compared normolipidemic probands and patients with apolipoprotein A-I (apoA-I) deficiency, apoE deficiency, or TD in terms of the plasma capacity to induce the efflux of [3H]-cholesterol from normal and TD fibroblasts and to esterify this cell-derived cholesterol. Compared with normal fibroblasts, TD fibroblasts released a significantly smaller fraction of [3H]-cholesterol into normal, high-density lipoprotein (HDL)-deficient, and apoE-deficient plasmas. Supplementation of apoE-deficient plasma with exogenous apoE normalized the cholesterol efflux from normal cells but did not fully restore the reduced cholesterol efflux from TD fibroblasts. Compared with control plasma, HDL- and apoE-deficient plasmas had a significantly reduced activity to esterify cell-derived cholesterol. Cholesterol derived from TD fibroblasts was less available for esterification in either patient or normal plasmas than cholesterol derived from normal cells. The esterification defect of TD cell-derived cholesterol was more pronounced in patient plasmas than in control plasma. We conclude that (1) apoA-I and, to a lesser degree, apoE are important determinants of the cholesterol efflux and esterification capacity of plasma, (2) TD fibroblasts have a reduced capacity to release cholesterol into the plasma, and (3) TD cell-derived cholesterol is less available for esterification in plasma than cholesterol from normal fibroblasts. The absence of distinct apoA-I- or apoE-containing subclasses aggravates the defective efflux and esterification of cholesterol derived from TD cells.

Evidence of linkage of familial hypoalphalipoproteinemia to a novel locus on chromosome 11q23

Coronary heart disease (CHD) accounts for half of the 1 million deaths annually ascribed to cardiovascular disease and for almost all of the 1.5 million acute myocardial infarctions. Within families affected by early and apparently heritable CHD, dyslipidemias have a much higher prevalence than in the general population; 20%-30% of early familial CHD has been ascribed to primary hypoalphalipoproteinemia (low HDL-C). This study assesses the evidence for linkage of low HDL-C to chromosomal region 11q23 in 105 large Utah pedigrees ascertained with closely related clusters of early CHD and expanded on the basis of dyslipidemia. Linkage analysis was performed by use of 22 STRP markers in a 55-cM region of chromosome 11. Two-point analysis based on a general, dominant-phenotype model yielded LODs of 2.9 for full pedigrees and 3.5 for 167 four-generation split pedigrees. To define a localization region, model optimization was performed using the heterogeneity, multipoint LOD score (mpHLOD). This linkage defines a region on 11q23.3 that is approximately 10 cM distal to-and apparently distinct from-the ApoAI/CIII/AIV gene cluster and thus represents a putative novel localization for the low HDL-C phenotype.

Lipid-free apolipoprotein (apo) A-I is converted into alpha-migrating high density lipoproteins by lipoprotein-depleted plasma of normolipidemic donors and apo A-I-deficient patients but not of Tangier disease patients

Plasma of patients with Tangier disease (TD) is devoid of alpha-LpA-I (apolipoprotein A-I-containing lipoprotein), which in normolipidemic plasma constitutes the majority of high density lipoprotein (HDL). The residual amounts of apolipoprotein A-I (apo A-I) in TD plasma have electrophoretic prebeta1-LpA-I mobility. We have previously demonstrated that TD plasma does not convert prebeta1-LpA-I into alpha-LpA-I. In this study we found that plasmas of normolipidemic controls, apo A-I-deficient patients and patients with fish-eye disease, but not plasmas of six TD patients, convert biotinylated lipid-free apo A-I into alpha-LpA-I. Supplementation of plasma with free oleic acid or fatty acid free albumin neither inhibited conversion activity in normal plasmas nor reconstituted it in TD plasma. In normal plasma the conversion activity was assessed in HDL and in the lipoprotein-free fraction. The latter fraction, however, generated larger particles only in the presence of exogenous phospholipid vesicles. To obtain particles with alpha-mobility, these vesicles had to contain phosphatidylinositol and/or cholesterol. Lipoprotein-depleted TD plasma did not convert lipid-free apo A-I into alpha-LpA-I even in the presence of exogenous vesicles with phospholipids or cholesterol. Taken together we conclude that disturbed transfer of glycerophospholipds onto apo A-I or prebeta1-LpA-I prevents maturation of HDL and thereby possibly causes deficiency of HDL cholesterol in patients with TD. Moreover, the lack of alpha-LpA-I in TD plasma together with its failure to convert exogenous apo A-I into an alpha-migrating particle provide specific tests for the diagnosis of TD.

A thyroid hormone binding motif is evolutionarily conserved in apolipoproteins

High-density lipoproteins (HDL) are the major lipoprotein (Lp) plasma carriers of thyroid hormones, binding mediated by a specific interaction with their apolipoproteins (A-I, A-II, A-IV, C-I, C-II, C-III and E). The single binding site of these apolipoproteins is encoded by exon 3 (exon 2 for apoA-IV) of the respective gene and has amino acid sequence homology with regions of the three major thyroid hormone plasma transport proteins (TBG, TTR, and albumin) known to contain the corresponding hormone binding site(s). Within the hormone domain, we identified a 5-residue hydrophobic motif "Y, L/I/M, X, X, V/L/I" that is extremely well conserved in apolipoproteins. The exon-3 coded region of human apo E contains a hydrophobic pocket which is formed by Trp 34 and a number of neighboring leucines (residues no. 28, 30, 37 or 43). The location of this pocket overlaps strikingly that of the region (aa 26-40) where the said homology is maximal and where the motif YLRVW, (aa 36-40) lies. This hydrophobic pocket should represent the thyroid hormone site of apo E and, because of the said homology, should exist in the other HDL apolipoproteins. Because TBG and/or TTR are not present in all animal species, but Lp are, and because fish HDL bind thyroid hormones, I postulated that thyroid hormone binding to HDL apolipoproteins is conserved through the phylum. To this end, I evaluated the conservation of the 5-residue motif in all the apolipoprotein sequences known (PIR data bank no. 42) and tested the thyroid hormone binding properties of two animal apolipoproteins that were available (bovine apo A-I and rabbit apo E). I found that the conservation does exist and that the binding properties of the two animal apolipoproteins match those of the respective human counterpart. In addition, I found that the 5-residue motif is spared by naturally occurring mutations, which is not the case for other domains. I therefore conclude that the interaction of thyroid hormones with Lp represents the first plasma transport for these hormones that appeared in the animal world and that preservation of the structure of the hormone domain appears to be more important than preservation of other domains.

In vivo metabolism of apo A-I and apo A-II in subjects with apo A-I(Lys107-->0) associated with reduced HDL cholesterol and Lp(AI w AII) deficiency

Apolipoprotein A-I (apo A-I) and apolipoprotein A-II (apo A-II) represent 80 90% of the protein content of high density lipoproteins (HDL). Previously we have identified a Finnish family with an apo A-I variant (Lys107-->0) associated with reduced plasma HDL cholesterol level and decreased lipoprotein (Lp)(AI w AII) concentration compared to unaffected family members. To determine the in vivo metabolism of apo A-I and apo A-II in the carriers of apo A-I (Lys107-->0) variant we radioiodinated normal apo A-I with 125I and apo A-II with 131I and compared the kinetic data of two heterozygous apo A-I(Lysl07-->0) patients (HDL cholesterol leves 0.31 and 0.69 mmol/l) to that of eight normolipidemic, healthy control subjects. Plasma radioactivity curves of 125I-labelled normal apo A-I of the patients demonstrated accelerated clearance of apo A-I compared to control subjects. In the two patients the fractional catabolic rates (FCR) of apo A-I were 0.347/day and 0.213/day, respectively, while the mean FCR of apo A-I of the control subjects was 0.151 +/- 0.041/day. Similarly, the plasma decay curves of the 131I-labelled apo A-II showed more rapid clearance of apo A-II in the two patients than in control subjects. The FCR of apo A-II in the two patients were 0.470/day and 0.234/day, while the mean FCR of apo A-II in control subjects was 0.154 +/- 0.029/day. The calculated production rates of apo A-I were similar in patients and in control subjects, and the production rates of apo A-II were significantly higher in patients than in control subjects. Our results show that the Lp(AI w AII) deficiency in patients with the apo A-I(Lys107-->0) is associated with increased fractional catabolic rates of normal apo A-I and apo A-II, while the production rates of these apolipoproteins are normal (apo A-I) or slightly increased (apo A-II).