Two Novel Molecular Defects in the LCAT Gene Are Associated With Fish Eye Disease
Abstract A 53-year-old man with a severely reduced HDL cholesterol level, dense corneal opacities, normal renal function, and premature coronary artery disease was investigated together with 16 members of his family. The proband was diagnosed with fish eye disease. As in previously reported patients with fish eye disease, the endogenous plasma cholesterol esterification rate was near normal, yet lecithin:cholesterol acyltransferase (LCAT) activity was almost absent when measured with exogenous HDL analogues used as substrate. Direct sequencing of the LCAT gene revealed two novel missense mutations in exon 1 and exon 4, resulting in the substitution of Pro10 with Gln (P10Q) and Arg135 with Gln (R135Q), respectively. Both missense mutations were located on different alleles. Genetic analysis by polymerase chain reaction revealed 4 carriers of the P10Q and 3 carriers of the R135Q defect. Functional assessment of both missense mutations revealed that when exogenous HDL analogues were used as substrate, the specific activity of rLCATP10Q was 18% of wild type (WT); however, when LDL was used as substrate, the activity was 146% of WT. By contrast, rLCATR135Q was inactive against both substrates. Thus, we conclude that the LCATR135D mutation is causative for complete LCAT deficiency and that the clinical phenotype of fish eye disease seen in this patient is due to the Pro10 mutation. The presence of premature coronary artery disease in the absence of other risk factors in this new case of fish eye disease raises questions regarding the risk of atherosclerosis, which has previously been reported to be nonexistent.
- Received July 10, 1995.
- Accepted October 16, 1995.
Numerous epidemiological studies have shown an inverse correlation of HDL cholesterol levels with the incidence and severity of CAD. To explain the role of HDL in lipid metabolism and atherogenesis, the model of “reverse cholesterol transport” is now widely accepted. In this model, the HDL particle is thought to mediate cholesterol transport from the periphery to the liver. The exact mechanism by which the HDL particle exerts this protective role in atherogenesis remains to be elucidated. Interestingly, a few rare hereditary disorders of HDL metabolism have been reported with no apparent risk for CAD despite severe HDL deficiency. FED, first described by Carlson and Philipson,1 is an example of such a disorder. Although HDL cholesterol is often reduced to below the detection level in these patients, corneal opacities are reported to be the only clinical hallmark. It is known that both FED and familial complete LCAT deficiency are caused by defects in the LCAT gene.2 3 4 5 6 7 8 9 10 11 12 13 However, the clinical presentation of the latter disorder includes hemolysis and renal failure in addition to the HDL deficiency. Because of the limited number of families studied, the risk for premature atherosclerosis is not well documented in either disorder.
In the present study, we describe compound heterozygosity for two novel mutations in the LCAT gene in a 53-year-old man with FED who suffers from severe premature CAD.
The 53-year-old proband was originally investigated at age 38 for unstable angina pectoris. He was subsequently referred to our clinic because of an extremely low HDL cholesterol level. Angiography at the initial presentation revealed 95% stenosis of the right coronary artery and moderate, grade II aortic regurgitation, for which bypass surgery was performed together with replacement of the aortic valve. He developed recurrent chest pain 5 years later. On recatheterization, the aortic valve appeared severely insufficient and had to be replaced. The bypass graft was patent, but the left anterior descending artery showed a 50% stenosis. Since that time, he had no further cardiac complaints. There were no other risk factors for cardiovascular disease, with the exception of cigarette smoking from age 18 until age 30. The family history was negative for premature cardiovascular disease.
On physical examination, the patient appeared well. There was severe corneal clouding, which had begun around puberty and had become more marked over the years. Loss of visual acuity was not apparent. Laboratory investigation revealed no glomerular disease or anemia.
Blood Samples and DNA Isolation
Blood from the proband and his family members was collected in EDTA tubes after an overnight fast and placed immediately on wet ice. Plasma was separated from cells by centrifugation (1200g for 15 minutes at 4°C), frozen in liquid nitrogen, and stored at −70°C before shipment to Vancouver, where apos and various LCAT parameters were determined. Genomic DNA was extracted from white cells as described previously.14 15
Lipoprotein and Apolipoprotein Analysis
Cholesterol and triglyceride concentrations in plasma were measured by enzymatic reagents (Boehringer-Mannheim; Miles Laboratories) on a Multistat III centrifugal analyzer. Plasma HDL cholesterol was determined by the polyethylene glycol 6000 method.16 Total cholesterol and FC levels in the various lipoprotein fractions were determined by commercially available enzymatic methods (Boehringer-Mannheim). Plasma apo A-I and apo B-100 were measured by nephelometry. LDL protein was determined by the method of Lowry et al.17 VLDL+IDL, LDL, HDL2, and HDL3 were separated by density gradient ultracentrifugation.18 Quantification of LpA-I particles in plasma was performed by electroimmunodiffusion in agarose gel with Hydragel LpA-I particle kits (Sebia). Lp A-I represents the amount of plasma apo A-I that is not present in the LpA-I:A-II particles, whereas LpA-I:A-II is calculated by subtraction of the LpA-I from the total amount of plasma apo A-I and LpA-I/A-II represents the amount of plasma apo A-I that is not present in the LpA-I particles.19
LCAT Activity, LCAT Concentration, and Measurement of the CER
LCAT activity was determined as described previously.20 LCAT activity represents the ability of plasma to esterify cholesterol in an exogenously presented proteoliposome substrate whereas CER reflects the esterification of cholesterol within the endogenous lipoproteins of the plasma. LCAT mass was measured by radioimmunoassay by Dr J.J. Albers (University of Washington School of Medicine, Seattle).21 CER was determined by measuring the rate of esterification of [3H]cholesterol22 and was measured in both plasma and LDL/VLDL–depleted plasma. LDL/VLDL–depleted plasma was prepared by precipitation of Apo B-containing lipoproteins with phosphotungstate-MgCl2.23 24 CER-plasma and CER-HDL were calculated from FERs and plasma and HDL-FC levels, respectively, as previously described.25
Amplification of LCAT Fragments by PCR
The 3′-primers used in PCR reactions were biotinylated at the 5′-end with biotin phosphoramidite (Glen Research Corp). Three DNA fragments, encompassing exon 1/2, exon 3/4/5, and exon 6 of the LCAT gene, were amplified by PCR from genomic DNA of the proband and a control subject by use of a DNA thermal cycler (Perkin Elmer Cetus). The amplification reactions were carried out in 10 mmol/L Tris-HCl (pH 9.0), 50 mmol/L KCl, 0.1% wt/vol gelatin, 1.5 mmol/L MgCl2, 1% Triton X-100, 0.2 mg/mL bovine serum albumin containing 0.5 to 1.0 μg genomic DNA, with final concentrations of 200 μmol/L dNTPs and 0.2 to 0.3 μmol/L primers in a total volume of 50 μL. After initial denaturation (10 minutes, 95°C), 0.3 to 1.0 U thermostable DNA polymerase (Supertaq; HT Biotechnology Ltd) was added, followed by 30 cycles of 95°C (1 minute), 65°C (1 minute), and 72°C (1 minute) with a final extension step of 10 minutes at 72°C.
PCR products were precipitated with NH4Cl and gel purified with Geneclean (Bio 101 Inc). Three to four micrograms of purified PCR product was incubated for 30 minutes at 37°C with 30 μL streptavidin-coated beads (Dynal AS) in a total volume of 100 μL saline Tris-EDTA (1 mol/L NaCl, 5 mmol/L Tris, pH 7.4; 0.5 mmol/L Tris-EDTA). The supernatant was discarded and a magnet particle concentrator (MPC-E; Dynal AS) was used to separate phases. After three washing steps with 100 μL saline Tris-EDTA, the beads were resuspended in 10 μL 0.1 mol/L NaOH and incubated for 10 minutes at room temperature to denature the DNA. Single-stranded DNA in the supernatant was recovered by neutralization, precipitated, and dissolved in deionized water for subsequent sequencing. The beads were rinsed with 60 μL 0.1 mol/L NaOH followed by three washing steps with 10 mmol/L Tris (pH 7.4) and 1 mmol/L EDTA and resuspended in dH2O. Single-stranded DNA [35S]dATP sequencing was performed by the dideoxy chain termination method by use of nested primers and Sequenase version 2.0 (United States Biochemical Corp) on both single-stranded templates.
Restriction Analysis of PCR-Amplified Genomic DNA
A 63-bp mutagenic primer with a 5′ GC-clamp (5′- CCGCCGCGCCCCGCGCCCGTCCCGCCGC-CCCCGCCCCC-TGGCTCCTCAATGTGCTCTTCCCGC-3′) was used to amplify a 180-bp DNA fragment of the proband and his kindred. This primer creates a Bbv I cutting site when the C-to-A937 nucleotide substitution in exon 1 is present. PCR products were digested with Bbv I, subjected to electrophoresis on a 3% agarose gel, and stained with ethidium bromide.
A 180-bp portion of the LCAT gene of the proband and family members, spanning the site of the missense mutation in exon 4, was amplified by PCR. The G2218-to-A sequence change in exon 4 eliminates an Aci I cutting site in exon 4. PCR products were digested with Aci I subjected to electrophoresis on a 3% agarose gel and stained with ethidium bromide. The enzymes were used according to the instructions of the manufacturer (New England Biolabs).
Subcloning of PCR Products
PCR products encompassing exon 1 to 5 of the LCAT gene of the proband were ligated into a pGEM-T vector (pGEM-T Vector System). Plasmid clones containing the DNA fragment were analyzed for the presence of nucleotide changes in exon 1 (C937 to A) and exon 4 (G2218 to A); double-stranded DNA plasmid sequencing was performed by the dideoxy chain termination method as described above.
Mutagenesis of the LCAT cDNA
Both missense mutations were introduced into full-length LCAT cDNA by use of unique restriction sites within the LCAT cDNA.
Exon 1 (C to A937)
A 364-bp DNA fragment containing the mutant gene region in exon 1 was amplified by PCR with the genomic DNA of the proband. Gel-purified PCR product was digested with Nco I and Bpu1102I according to the manufacturer’s instructions (New England Biolabs). A 100-bp DNA fragment containing the mutation was isolated from a 4% agarose gel (Boehringer-Mannheim) with Mermaid (Bio 101 Inc). A four-end ligation was performed to insert this fragment as a cassette in the WT LCAT cDNA in pNUT. Double-stranded plasmid DNA sequencing was used to identify a clone that contained the desired mutation at codon 10 but no other mutations (P10Q).
Exon 4 (G to A2218)
Kpn I and BssHII (New England Biolabs) were used to digest a 428-bp PCR-amplified DNA fragment of the proband containing the mutant gene region in exon 4. A 57-bp DNA fragment encompassing the region of the mutation was isolated from a 2% agarose gel (50% low–melting point agarose; Boehringer-Mannheim) with Mermaid (Bio 101, Inc). After cloning, double-stranded plasmid DNA sequencing was performed for identification of a clone containing the mutation at codon 135 (R135Q).
Stable Transfection of BHK Cells
The pNUT-LCAT constructs, ie, pNUT-LCAT-P10Q and pNUT-LCAT-R135Q, were used to establish stable cell lines of BHK cells as previously described.26 The pNUT vector contains a mutant form of the dihydrofolate reductase gene that permits selection of cells containing the plasmid DNA by their survival in high concentrations of methotrexate. BHK cells were maintained in DMEM (Gibco-BRL) supplemented with 10% heat-inactivated fetal bovine serum. Stable transfection of BHK cells was performed by calcium phosphate coprecipitation as previously described.26
Analysis of BHK Clones
Clones expressing rLCAT were identified by LCAT enzyme activity and solid-phase LCAT immunoassay as previously described.27 One cell line of each clone was selected for further analysis. The specific enzyme activities of the WT and mutant rLCATs were determined with HDL analogues and LDL used as substrates. LDL was prepared by preparative ultracentrifugation from blood collected from normal volunteers after 12 hours of fasting. The LDL fraction was dialyzed extensively at 4°C against 0.01 mol/L Tris-HCl (pH 7.4) containing 0.15 mol/L NaCl and 0.005 mol/L EDTA. After heat inactivation at 56°C for 30 minutes to eliminate endogenous LCAT activity, the concentration of unesterified cholesterol was determined enzymatically by a reagent kit (Boehringer-Mannheim).
The intron-exon boundaries and the coding regions of the LCAT gene were sequenced in the proband and a normal control subject. Direct sequencing revealed heterozygosity for two novel missense mutations (Fig 1A⇓ and 1B⇓): (1) a C-to-A sequence change in exon 1, resulting in the substitution of a proline (CCG) for a glutamine (CAG) at position 10 of the mature protein, and (2) a G-to-A base substitution in exon 4, introducing an arginine (CGG) for a glutamine (CAG) at position 135. To determine whether these mutations were either in cis or trans conformation, an LCAT fragment encompassing both mutations (exon 1 to exon 5) was amplified and cloned. It was shown by sequence analysis that clones either contained the C-to-A point mutation in exon 1 or the G-to-A point mutation in exon 4 (results not shown), thereby demonstrating that the mutations were present on different alleles. The mutation in exon 1 neither created nor eliminated a restriction site for a known endonuclease. To screen the genomic DNA of the family members for the presence of this mutation, a mutagenic PCR was devised that introduced a Bvb I cutting site when the mutation was present (Fig 2A⇓). Analysis of family members revealed four additional heterozygotes for this mutation. The G-to-A point mutation in exon 4 eliminates an Aci I restriction site. A 180-bp fragment of the LCAT gene, encompassing the site of the mutation, was amplified by PCR. Digestion of normal LCAT sequence resulted in two DNA fragments of 45 and 135 bp, respectively, of which only the 135-bp fragment is clearly visible when subjected to electrophoresis with a 3% agarose gel (Fig 2B⇓, lane 2). The undigested PCR product in lane 3 of Fig 2B⇓ is indicative for heterozygosity due to the loss of a cutting site on one allele. When the DNA of the family members was screened for the presence of this mutation, three more carriers of this mutation were identified (Fig 3⇓). Restriction analysis confirmed the proband’s compound heterozygosity for both point mutations in exon 1 and 4. The complete family tree of the studied kindred is shown in Fig 3⇓, which illustrates the pattern of inheritance for each mutation.
Lipoprotein and Apo Analysis
Plasma lipoprotein concentrations of the proband and all family members are shown in Table 1⇓. The carrier status of family members was determined by use of PCR-based DNA analysis, as described above. The plasma cholesterol level in the proband (II-1) was in the normal range, but he showed mildly elevated triglyceride levels and LDL cholesterol levels. HDL cholesterol level was reduced to <10% of the levels exhibited by other members of his family. Heterozygotes had significantly reduced HDL cholesterol levels (P<.002 by Student’s t test) compared with unaffected family members. Apo and HDL subfraction data are summarized in Table 2⇓. The proband was characterized by severely reduced plasma apo A-I concentrations and elevated apo B-100 levels. Plasma concentrations of apo A-I in heterozygotes were significantly decreased to ≈70% of apo A-I levels exhibited by unaffected family members (P<.003 by Student’s t test). The HDL deficiency in the index patient was due to a specific near-total loss of LpA-I/A-II particles, whereas LpA-I levels were half of normal. The decrease in HDL cholesterol in heterozygotes, however, was reflected by a significant decrease of both LpA-I and LpA-I/A-II particles (P<.005 and P<.019 respectively, versus normal levels).
VLDL+IDL, LDL, HDL2, and HDL3 lipoprotein fractions were separated by density gradient ultracentrifugation (Fig 4⇓) and the FC and CE contents determined in all family members (Table 3⇓). Fig 4⇓ illustrates the HDL deficiency in the proband and an apparent selective loss of HDL2 in two heterozygotes for the P10Q and the R135D defect (subjects I-1 and III-1, respectively). Lipid analysis of each class of lipoprotein indicated no significant differences when VLDL+IDL and LDL lipoprotein fractions of heterozygotes were compared with those of normal subjects. By contrast, VLDL+IDL FC and LDL FC were increased in the proband compared with control subjects. Both HDL2 FC and HDL3 FC were reduced to ≈35% of normal in the proband. HDL2 FC was significantly decreased in heterozygotes (P<.07 by rank sum two-sample test), whereas HDL3 FC was normal compared with unaffected family members. Furthermore, the FC/CE ratio in the HDL2 fraction was significantly higher in heterozygotes compared with unaffected siblings (P<.03 by rank sum two-sample test).
LCAT Activity and Endogenous Esterification Rate
Data on LCAT activity and related parameters are summarized in Table 4⇓. LCAT activity was measured in both whole plasma and VLDL- and LDL-depleted plasma with HDL analogues used as substrate. LCAT activity in the plasma of the proband was decreased to <8% of the activity of unaffected family members, whereas CER-plasma was only slightly reduced. Cholesterol esterification associated with HDL particles in the proband was only one third of that in control subjects. Heterozygotes showed ≈60% of normal LCAT activity in plasma and 70% of normal activity in plasma depleted of apo B–containing lipoproteins. The decrease of cholesterol esterification in heterozygotes in both whole and VLDL/LDL–depleted plasma (P<.01 and P<.007, respectively, by Student’s t test) was mainly caused by the low CER of subjects III-4 and III-7, who interestingly had low triglyceride levels and plasma FC but high HDL2. Except for subject I-1, the proband and all heterozygous carriers presented with lower LCAT concentration and a tendency toward lower specific activity compared with unaffected family members.
In Vitro Expression of rLCAT
To assess the functional significance of the amino acid substitutions Pro10 with Gln (P10Q) and Arg135 with Gln (R135Q) in the LCAT protein, both mutant enzymes were expressed as recombinant proteins in BHK cells. The ability of these mutant proteins to esterify cholesterol in HDL analogues and LDL is shown in Fig 5⇓. When HDL analogues were used as substrate, the specific activity of LCATP10Q was 18±0.5% of WT; however, P10Q showed higher specific activity when LDL was used as substrate (146±19.6% of WT). LCATR135Q showed low activity against HDL analogues (1.2±0.5% of WT) as well as against LDL (3.6±0.7% of WT).
Two Novel Mutations Associated With FED
In the present study, we describe a 53-year-old man in whom severe HDL deficiency, dense corneal opacities, decreased LCAT activity, and near-normal CER were pathognomonic for FED. The presence of premature atherosclerosis in this subject at age 38 while additional risk factors were absent is of particular interest.
The proband was shown to be compound heterozygous for two novel missense mutations in exon 1 and exon 4 of the LCAT gene, resulting in the substitution of Pro10 with Gln (P10Q) and Arg135 with Gln (R135Q), respectively. On screening of the family’s DNA for the presence of both mutations, four carriers of the P10Q defect and three carriers of the R135Q defect were identified. Their carrier status was in keeping with reduced LCAT activity, indicating the sensitivity and specificity of this assay.
To properly classify LCAT gene defects that may be causative for either FED or LCAT deficiency, the effects of each mutation on LCAT activity must be defined by in vitro analysis and compared with in vivo data. This is essential in cases of compound heterozygosity such as reported in the present study. In vitro expression of both mutations present in this family was performed to study the functional effects of these nucleotide changes on the LCAT protein. Mutated LCAT cDNAs were ligated into the pNUT expression vector and used for stable transfection of BHK cells. rLCATP10Q was secreted at normal levels compared with WT rLCAT and was shown to be only partially active when HDL analogues were used as substrate (18% of WT), whereas no decrease in activity against LDL cholesterol was observed (146% of WT). We obtained almost identical results previously after expression of the Thr123-Ile defect in COS-I cells.27 On the other hand, although sufficient rLCATR135D protein was secreted, the mutant protein had negligible catalytic activity against both substrates, which was characteristic of classic LCAT deficiency. Therefore, we conclude that compound heterozygosity for P10Q and R135Q underlies HDL deficiency in the proband, whereas the P10Q defect dominates the biochemical phenotype and therefore is responsible for the clinical expression of FED. Furthermore, the in vivo generation of cholesteryl esters in the proband’s plasma is limited to the LDL/VLDL fraction and is maintained by the apparent enhanced activity of LCATP10Q on LDL. In contrast, this activity of LCATP10Q in heterozygotes does not seem to contribute substantially to cholesterol esterification, as shown by identical CER values in both P10Q and R135Q carriers. This indicates the absolute LCAT concentration or activity is not the rate-limiting step in the production of cholesteryl esters in plasma.
Different nucleotide changes in the same codons were described earlier by Skretting and Prydz2 and Assmann et al.13 A homozygous C-to-T substitution in codon 10 altering Pro10 to Leu (P10L) was identified in the original Swedish FED patients, whereas a C-to-T mutation in codon 135 resulting in the exchange of an Arg135 for Trp (R135W) in combination with a frameshift in exon 1 was identified in a Canadian patient who suffered from classic LCAT deficiency. Although these amino acid substitutions at positions 10 and 135, respectively, differ from those in our proband, the effects on the LCAT protein are identical. Both R135Q and R135W defects (unpublished data, 1994) result in a catalytically inactive enzyme, whereas LCATP10Q and LCATP10L (unpublished data, 1994) are causative for an FED phenotype, since both retain the ability to esterify cholesterol in LDL.
Severe Premature Atherosclerosis in the Index Patient
The major clinical finding of the present study is the clear presence of premature atherosclerosis in the proband despite the absence of other risk factors. In addition, we observed premature CAD in two male probands in another Dutch FED family.28 Although this observation must not be overinterpreted, it causes us to question the earlier assumption that the FED phenotype is not associated with increased risk of CAD.13 It is difficult to evaluate this issue, since relatively few patients with FED have been described and no rigorous statistical analysis is possible. It is vital, however, that some consensus is reached with respect to risk for newly discovered probands and their family members.
We have reviewed the published clinical data on patients with proven FED (Table 5⇓). The two original Swedish probands, both female (kindred I1 ), did not present with premature CAD. The index patient was referred to the clinic because of hypertriglyceridemia. Atherosclerosis in this kindred developed with age: the older sister had a myocardial infarction at age 77; the father, who suffered from the same eye disease, died at age 76 of a myocardial infarction. An unrelated Swedish female proband was alive and well at age 70 (kindred II29 ). Her referral basis was ophthalmology. A third case of FED was presented by Frohlich et al30 : a 16-year-old boy was noted to have low HDL cholesterol and corneal opacities (kindred III). This patient was identified by molecular diagnosis as suffering from FED (unpublished data, 1994). Although there was some history of cardiovascular disease in his mother’s family, many relatives lived into their 90s. His father, an apparent heterozygote, underwent coronary bypass surgery at age 65. Funke and coworkers5 (kindred IV) were the first to report premature atherosclerosis in FED: the index patient had suffered since the age of 50 from angiographically assessed two-vessel CAD with a 60% stenosis of the anterior interventricular artery and a 50% stenosis of the posterior interventricular artery, whereas his brother was reported to have experienced angina pectoris since the age of 60. The family history, however, was negative with regard to any increased risk for CAD. Two unrelated Dutch probands (kindred V5 31 ) were brought to our attention by an ophthalmologist. Although these patients were originally referred as healthy and without any signs or symptoms of atherosclerosis, the elder brother suffered from angina that required bypass surgery at age 51 (unpublished data, 1994). Recently, Klein et al3 presented a 66-year-old German patient with FED (kindred VI). His excellent condition to date could be related to his remarkable lifestyle: physically very active, lifelong low-fat diet, nonsmoker. The seventh kindred with FED was reported by Clerc et al32 : a 53-year-old man and two of his sisters, 49 and 50 years old, respectively, had good general health and showed no signs of atherosclerosis. However, the proband of this family and his two sisters were homozygous for the Leu300 deletion.4 This mutation is associated with a biochemical FED phenotype that differs from other FED patients in that the near-normal cholesterol esterification is not accompanied by specific loss of activity on HDL. This might have implications for the risk to develop atherosclerosis. As mentioned above, we have recently described a large Dutch kindred (kindred VIII28 ) in which the two male homozygous individuals had proven CAD at age 43 and 54, respectively.
This analysis of literature and an evaluation of the current condition of FED patients further indicates the potential for the early onset of atherosclerosis. We believe that the apparent longevity of the original Swedish patients (kindreds I and II) and the young age of the Canadian proband (kindred III) may have provided an early bias in the assessment of risk for CAD. It is now clear that several of the male patients with FED do indeed suffer from CAD, and this is exemplified by severe premature atherosclerosis in the proband reported in the present study (Table 5⇑).
It is important to recognize the potential for selection bias in all kindreds described here, since the reason for referral may have an impact on the relative number of patients who appear to develop premature CAD. For example, it is obvious that patients referred for diagnosis by cardiologists are far more likely to have CAD than those referred by ophthalmologists or general practitioners who note the corneal opacity or HDL deficiency. Therefore, it seems likely that several families with FED remain undiagnosed or misdiagnosed. Thus, any accurate prediction of FED-induced risk of CAD must be based on a representative sample of patients that includes those who are asymptomatic with respect to CAD. Since HDL deficiency is characteristic of all cases of FED, large-scale screening of lipid profiles will likely identify new families with FED who have not been investigated as a consequence of the presence of disease. Overall, no definitive conclusion regarding FED and risk of CAD can be made until more families with FED are identified.
If the cause of CAD in the family in the present study is related to FED, the mechanism involved cannot be determined from this study. It may not simply be related to changes in HDL concentration and structure but could be due to increased LDL cholesterol and apo B levels. In contrast, it is tempting to speculate that inactive or partially active LCAT increases the risk for atherosclerosis, supporting the protective action of an intact LCAT protein against CAD, by its role in reverse cholesterol transport. Although total plasma cholesterol esterification is near normal in FED, the lack of normal maturation of HDL particles (as described below) could affect resistance against atherosclerosis.
Structural Changes in the HDL Pool of Homozygotes and Heterozygotes
The changes in lipids and apoprotein levels in the plasma of the proband were not markedly different from those reported earlier for other patients with FED. Of particular interest, however, is our observation that the decreased HDL level is due to a near-absolute deficiency of LpA-I/A-II particles. This phenomenon has been observed in other patients with FED33 (eg, kindreds III, VI, and VII; Table 5⇑), but to a less marked degree. The pathophysiological basis of the loss of apo A-II–containing particles appears to be a marked hypercatabolism of apo A-II. Since this effect is also seen in LCAT deficiency, Rader et al33 concluded that LCAT activity is required to maintain the maturation and accumulation of apo A-II–containing HDL particles.
The effects of heterozygosity for either allele on the biochemical phenotype in this family are also interesting. The small number of patients carrying each mutation makes it difficult to compare the different heterozygotes; the total heterozygous group is different from those family members who are genetically unaffected. We observed a significant decrease in both HDL cholesterol and apo A-I levels in heterozygotes compared with unaffected individuals. In addition, we noted a 46% decrease in HDL2 cholesterol levels but only a 23% decrease in HDL3 cholesterol levels. This reduction in HDL cholesterol level was associated with a highly significant decrease of LpA-I levels (P<.005) and, to a lesser extent, LpA-I/A-II levels (P<.019) in heterozygotes. Clearly, the heterozygous state for FED in this family affects both the total amount and subclass distribution of the HDL pool. Since a significant decrease of both HDL2 and LpA-I levels would be associated with increased risk of CAD in the general population, we believe that the relative risk for CAD in individuals who are heterozygous for defects of the LCAT gene requires further evaluation.
In the current study, we report a patient with FED who was shown to be a compound heterozygote for two novel point mutations in the LCAT gene resulting in the FED phenotype. Furthermore, we indicate that the incidence of atherosclerosis in patients with FED may be higher than previously assumed and also demonstrate that heterozygotes for this genetic defect may also be at increased risk.
Selected Abbreviations and Acronyms
|BHK||=||baby hamster kidney|
|CAD||=||coronary artery disease|
|CER||=||cholesterol esterification rate|
|CER-HDL||=||cholesterol esterification rate of HDL|
|CER-plasma||=||cholesterol esterification rate of plasma|
|FC||=||free (unesterified) cholesterol|
|FED||=||fish eye disease|
|PCR||=||polymerase chain reaction|
|P10Q||=||clone containing a mutation at codon 10 (substitution of Pro10 with Gln)|
|R135Q||=||clone containing a mutation at codon 135 (substitution of Arg135 with Gln)|
This research was supported by grants from the Dutch Heart Foundation (No. 89201) and the Medical Research Council of Canada. J.J.P. Kastelein is clinical investigator of the Dutch Heart Foundation. We wish to thank Dr S. Graafsma (Maria Hospital, Tilburg, Netherlands) for referring the proband to our lipid clinic and Dr Albers for performing LCAT mass determinations. Furthermore, we are obliged to Lida Adler for her excellent assistance with the LCAT activity determinations. We are indebted to all family members who kindly participated in this investigation.
Carlson LA, Philipson B. Fish-eye disease: a new familial condition with massive corneal opacities and dyslipoproteinemia. Lancet. 1979;2:921-923.
Klein HG, Lohse P, Pritchard PH, Bojanovski D, Schmidt H, Brewer HB Jr. Two different allelic mutations in the lecithin-cholesterol acyltransferase gene associated with the fish eye syndrome: lecithin-cholesterol acyltransferase (Thr-123-Ile) and lecithin-cholesterol acyltransferase (Thr-347-Met). J Clin Invest. 1992;89:499-506.
Klein HG, Santamarina-Fojo S, Duverger N, Clerc M, Dumon MF, Albers JJ, Marcovina S, Brewer HB Jr. Fish eye syndrome: a molecular defect in the lecithin-cholesterol acyltransferase (LCAT) gene associated with normal alpha-LCAT-specific activity—implications for classification and prognosis. J Clin Invest. 1993;92:479-485.
Funke H, von Eckardstein A, Pritchard PH, Albers JJ, Kastelein JJP, Droste C, Assmann G. A molecular defect causing fish eye disease: an amino acid exchange in lecithin-cholesterol acyltransferase (LCAT) leads to the selective loss of alpha-LCAT activity. Proc Natl Acad Sci U S A. 1991;88:4855-4859.
Von Eckardstein A, Funke H, Pritchard PH, Albers JJ, Kastelein JJP, Droste C, Assmann G. Fish-eye disease: characterization of the molecular defect. Circulation. 1990;82:425. Abstract.
Funke H, von Eckardstein A, Pritchard PH, Hornby AE, Wiebusch H, Motti C, Hayden MR, Dachet C, Jacotot B, Gerdes U, Faergeman O, Albers JJ, Colleoni N, Catapano A, Frohlich J, Assmann G. Genetic and phenotypic heterogeneity in familial lecithin:cholesterol acyltransferase (LCAT) deficiency: six newly identified defective alleles further contribute to the structural heterogeneity in this disease. J Clin Invest. 1993;91:677-683.
Klein HG, Lohse P, Duverger N, Albers JJ, Rader DJ, Zech LA, Santamarina-Fojo S, Brewer HB Jr. Two different allelic mutations in the lecithin:cholesterol acyltransferase (LCAT) gene resulting in classic LCAT deficiency: LCAT (tyr83 to stop) and LCAT (tyr156 to asn). J Lipid Res. 1993;34:49-58.
Bujo H, Kusunoki J, Ogasawara M, Yamamoto T, Ohta Y, Shimada T, Saito Y, Yoshida S. Molecular defect in familial lecithin:cholesterol acyltransferase (LCAT) deficiency: a single nucleotide insertion in LCAT gene causes a complete deficient type of the disease. Biochem Biophys Res Commun. 1991;181:933-940.
Demacker PNM, Hijmans AG, Vos-Janssen HE, Van ‘t Laar A, Jansen AP. A study of the use of polyethylene glycol in estimating cholesterol in high density lipoprotein. Clin Chem. 1980;26:1775-1779.
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951;193:265-275.
Demacker PNM, van Sommeren-Zondag DF, Stalenhoef AFH, Stuyt PMJ, van ‘t Laar A. Ultracentrifugation in swinging-bucket and fixed-angle rotors evaluated for isolation and determination of high-density lipoprotein subfraction HDL2 and HDL3. Clin Chem. 1983;29:656-663.
Parra HJ, Mezdour H, Ghalim N, Bard JM, Fruchart JC. Differential electroimmunoassay of human LpA-I lipoprotein particles on ready-to-use plates. Clin Chem. 1990;36:1431-1435.
Albers JJ, Adolphson JC, Chen CH. Radioimmunoassay of human lecithin-cholesterol acyltransferase. J Clin Invest. 1981;67:141-148.
Dobiasova M, Schutzova M. Cold labelled substrate and estimation of cholesterol esterification rate in lecithin cholesterol acyltransferase radioassay. Physiol Bohemoslov. 1986;35:319-327.
Warnick GR, Nguyen T, Albers AA. Comparison of improved precipitation methods for quantification of high-density lipoprotein cholesterol. Clin Chem. 1985;31:217-222.
Burnstein M, Scholnick HR, Morfin R. Rapid method for the isolation of lipoproteins from human serum by precipitation with polyanions. J Lipid Res. 1970;11:583-595.
Hill JS, O K, Wang X, Paranjape S, Dimitrijevich D, Lacko AG, Pritchard PH. Expression and characterization of recombinant human lecithin: cholesterol acyltransferase. J Lipid Res. 1993;34:1245-1251.
O K, Hill JS, Wang X, Pritchard PH. Recombinant lecithin:cholesterol acyltransferase containing a Thr123 to Ile mutation esterifies cholesterol in low density lipoprotein but not in high density lipoprotein. J Lipid Res. 1993;34:81-88.
Kuivenhoven JA, Wiebusch H, van Voorst tot Voorst E, Hill JS, Pritchard PH, Funke H, Kastelein JJP. Fish-eye disease and LCAT deficiency, extremes of one phenotype: a novel LCAT-mutation in a large family of Dutch descent. Atherosclerosis.. 1994;109:230. Abstract.
Rader DJ, Ikewaki K, Duverger N, Schmidt H, Pritchard PH, Frohlich JJ, Clerc M, Dumon MF, Fairwell T, Zech L, Santamarina-Fojo S, Brewer HB Jr. Markedly accelerated catabolism of apolipoprotein A-II (ApoA-II) and high density lipoproteins containing ApoA-II in classic lecithin:cholesterol acyltransferase deficiency and fish-eye disease. J Clin Invest. 1994;93:321-330.