Familial Hypercholesterolemia Associated with Severe Hypoalphalipoproteinemia in a Moroccan Family

c Indian Academy of Sciences

RESEARCH NOTE

Familial hypercholesterolemia associated with severe hypoalphalipoproteinemia in a Moroccan family

KARIMA AIT CHIHAB1,2, RACHID CHATER1,2, ANA CENARRO3, ANASS KETTANI2, SERGIO CASTILLO3, ∗ MOHAMED LOUTFI1, JOSEP RIBALTA 4, AHMED ADLOUNI2, MIGUEL POCOVI3 and MARIAME EL MESSAL1

1Laboratoire de Biochimie et Biologiè Molecularie, Groupe de Géné tique et Biologie Moléculaire, Faculté des Sciences Ain chock. B. P. 5366, Casablanca, Morocco 2Laboratoire de Recherche sur les Lipoprotéines et l’Athérosc-lérose, Faculté des Sciences Ben M’Sik. B. P. 7955, Casablanca, Morocco 3Laboratorio de Investigacion Molecular, Hospital Universitario Miguel Servet. Po Isbel la catolica, 1–3, Zaragaza, Spain 4Unitat de Recerca de Lipids i Anteriosclerosi, Facultat de Medicina, Universitat Rovira i virgili, Sant Lloren 21. 43201 Reus, Spain

Introduction screened N370S and L444P, the most frequent mutations in the β-glucocerebrosidase gene (GBA) that have been asso- Familial hypercholesterolemia (FH) is an autosomal dom- ciated with HALP (Pocovi et al. 1998). This study revealed inant genetic disorder characterized by elevated levels of the IVS3-23C→A mutation in LCAT gene, although it did not low-density-lipoprotein cholesterol (LDL-C), tendon xan- appear to cosegregate with HALP phenotype in this family. thomas and increased risk of premature coronary heart disease (CHD). The FH phenotype results from defects in the Materials and methods LDL receptor gene (LDLR), and also defects in other genes like apolipoprotein B (apoB) (familial defective apo B) or Subjects proprotein convertase subtilisin/kexin type 9 (PCSK9)(Soria At the first consultation, the proband, 15 years old, pre- et al. 1989; Abifadel et al. 2003). High-density-lipoprotein sented extra vascular lipid deposits and lipid profile charac- cholesterol (HDL-C) levels are significantly reduced in many teristics of FH homozygotes. However, no evident signs of FH families. However, the metabolic basis of this hypoal- atherosclerosis had been revealed. Because LDL apheresis is phalipoproteinemia (HALP) has not been clearly understood. not available in Morocco, a high dose of statin was prescribed It has been reported that FH heterozygotes with HALP are to the patient. Proband’s relatives were recruited and eleven prone to develop more severe premature artery disease (de were available for clinical examination and blood analyses. Sauvage Nolting et al. 2003). Indeed, the latest guidelines All subjects gave their informed consent prior to their inclu- for the diagnosis and management of FH consider levels of sion in the study. At the age of 22 years, the proband died by HDL-C less than 40 mg/dl as one of the major cardiovascular myocardial infarction. risk factors in the FH population (Civeira 2004). In this report, we describe a Moroccan FH family with as- Lipid analysis sociated HALP. After screening of the LDLR gene, we iden- We analysed serum TC, TG and HDL-C by enzymatic meth- tified a novel frameshift mutation in exon 5 of the LDLR ods and apo A-I and apo B by an immunoturbidimetric gene (756del7). To elucidate the inheritance of the HALP method (Brustolin et al. 1991). We calculated LDL-C by in this family, we analysed some other genes involved in Friedewald formula (Friedewald et al. 1972). HDL metabolism, such as apoAI, lecithin:cholesterol acyltransferase (LCAT) and lipoprotein lipase (LPL). We also Genetic analyses DNA isolation: Genomic DNA from white blood cells was *For correspondence. E-mail: [email protected]. isolated using a salting-out procedure (Miller et al. 1988). Keywords. familial hypercholesterolemia; hypoalphalipoproteinemia; LDLR gene; LCAT gene; SSCP; DNA sequencing.

Journal of Genetics, Vol. 86, No. 2, August 2007 159 Karima Ait Chihab et al.

LDLR gene: We analysed LDLR gene by Southern blot (El LCAT gene: We analysed the promoter region, exons and Messal et al. 2003) and PCR–SSCP. For SSCP analysis, PCR exon–intron junctions of LCAT gene by PCR–SSCP and se- products were added to 95% formamide buffer, denatured quenced the PCR fragment (exon 4) as described previously at 95◦C for 5 min, immediately chilled on ice and elec- (Recalde et al. 2002). Mutation confirmation was carried out trophoresed at 1050 V, at 25◦Cor15◦C in a MDE gel with by PCR-restriction enzyme digestion with MspIofexon4 or without 5% glycerol, respectively, in 0.6 × TBE buffer and subsequent 3% Nusieve agarose gel electrophoresis. for 15 h, on an automated DNA sequencer equipped with a water jacket (ALF-ExpressTM, Pharmacia Biotech). We GBA gene: We screened the GBA gene for the presence of sequenced a PCR fragment of LDLR gene (exon 5) using the most frequent mutations, N370S and L444P,byPCRand Big Dye Terminator Cycle Sequencing Kit (Perkin Elmer) restriction enzyme digestion with XhoIandNciI, respectively and a CEQ8000 DNA automated sequencer (Beckman Coul- (Beutler et al. 1990; Tsuji et al. 1987). ter). For mutation confirmation, PCR products were elec- trophoresed at 96 V at room temperature, for 3 h, in 2% LPL gene: We analysed the promoter region, exons and agarose gel in 1× TAE. exon–intron junctions of LPL gene by direct sequencing in an ABIPRISM 3100 Genetic Analyser. Sequence of primers used were reported previously by Abifadel et al. (2004). Apo B gene: We screened for R3500Q and R3531C muta- apoB et al tions in gene as described by Rabès . (1997). Results

Apo E gene: We determined the apoE genotype by PCR– Family analysis restriction analysis with HhaI as described by Hixon and The biochemical and clinical features of the proband and his Vernier (1990). relatives are presented in table 1. The proband (II-5) and his sisters II-7 and II-8 (ﬁgure 1) showed biochemical and clin- Apo AI gene: We analysed the promoter region, exons and ical features of homozygous FH. Among the other recruited exon–intron junctions of apoAI gene by PCR–SSCP (Re- relatives, only the proband’s sisters II-2 and II-6 would be calde et al. 2002). heterozygous FH patients (table 1), according to proband’s

Figure 1. Pedigree of the analysed individuals indicating LDL-C levels, FH genotype, HDL-C levels and HALP genotype. Proband (deceased) is indicated by an arrow; wt, wild type allele; mut, mutant allele. LDL-C and HDL-C levels were obtained before any medical treatment.

160 Journal of Genetics, Vol. 86, No. 2, August 2007 FH associated with severe hypoalphalipoproteinemia

Table 1. Clinical and biochemical features of the proband and his relatives.

TC HDL-C LDL-C TG Apo AI Apo B FH Subjects Sex Age (mg/dl) (mg/dl) (mg/dl) (mg/dl) (mg/dl) (mg/dl) EVLD CAD diagnosis I-1 Female 46 166 40 115 53 ND ND No No Non FH I-2 Female 40 220 23 173 116 89 117 No No Non FH (Mother) I-3 Male 50 279 24 221 174 84 167 No No Non FH (Father) I-4 Female 28 184 47 124 66 ND ND No No Non FH II-2 Female 22 438 38 368 190 143 222 No No Heterozygous FH II-3 Female 21 173 24 133 83 71 99 No No Non FH II-4 Male 31 101 29 61 57 ND ND No No Non FH II-5 Male 15 837 3 821 67 ND ND PCX, Deceased Homozygous (Proband) TX, LX, by FH 20 (462) (15) (416) (156) (51) (264) AC myocardial infarction II-6 Female 17 248 30 194 126 83 128 No No Heterozygous FH II-7 Female 15 736 16 693 136 61 394 PCX, Ischemia Homozygous TX, AC FH II-8 Female 10 722 14 682 141 52 409 PCX, No Homozygous TX, AC FH III-1 Female 10 213 35 152 131 101 111 No No Non FH TC, total cholesterol; HDL-C, HDL cholesterol; LDL-C, LDL cholesterol; TG, triglycerides; ND, not determined; EVLD, extra vascular lipid deposits; PCX, planar cutaneous xanthomas; TX, tendon xanthomas; LX, left xanthelasma; AC, arcus cornealis; CAD, coronary artery disease. Lipid values are obtained before any medical treatment. Lipid values of the proband after high dose of lipid lowering treatment are given in parentheses. The clinical diagnosis of the proband’s relatives is based on ”proband’s relative diagnosis” (Civeira 2004). relative diagnostic criteria (Civeira 2004). Except II-7, none putative heterozygotes by PCR and electrophoresis showed of the examined subjects showed signs of ischaemia in elec- two bands: normal allele (180 bp) and mutant allele (173 trocardiograms or cardiac scans. Surprisingly, the three ho- bp). The genotype of the studied subjects is indicated in fig- mozygous FH patients had a very low concentration of HDL- ure 1. In the same way, we analysed 107 healthy subjects for C and apo AI, never previously reported in FH subjects. the presence of the identified mutation. None of them were Also, except I-1 and I-4, the other relatives showed a de- carrier of the 756del7 mutation in the LDLR gene. crease in HDL-C levels, but not to the extent observed in We screened the proband and his relatives for R3500Q homozygous subjects. After an irregular observance of high and R3531C mutations in the apoB gene, but none of them dose lipid lowering treatment, the proband died by myocar- had either of these two FDB-causing mutations. We also dial infarction. analysed the apoE genotype, in order to assess its contribu- tion to FH phenotype. All subjects of the studied family were FH molecular analysis homozygous for the E3 allele. Southern blot analysis did not reveal any major defects in LDLR gene. However, LDLR gene PCR-SSCP analysis of HALP molecular analysis all studied subjects revealed two abnormal SSCP profiles in exon 5, when compared with the healthy subject pattern: an The entire apoAI, LCAT and LPL genes were screened, and heterozygous pattern for the parents, II-2, II-3, II-6, and III- the two most frequent mutations in GBA gene, N370S and 1, and an homozygous profile for the proband, II-7, and II-8. L444P, were analysed. PCR–SSCP of LCAT gene of the Concerning the subjects I-1, I-4, and II-4, they presented a proband and his parents showed a heterozygous pattern in normal SSCP profile. DNA sequencing of PCR products of the father, when compared with the healthy control subject the proband, II-7, and II-8 subjects showed a novel deletion and the heterozygous IVS3-23C→A positive control sam- of 7 bp at nucleotide 756 of cDNA, 756del7, in homozygous ple. DNA sequencing confirmed the presence of the IVS3- state. This frameshift mutation creates a stop codon at posi- 23C→A mutation, located 22 bases upstream of the acceptor tion 241, and would give rise to a truncated protein (data not splicing site of intron 3 of LCAT gene. The screening of shown). The confirmation of the heterozygous status in the IVS3-23C→A in the rest of relatives, by restriction analysis

Journal of Genetics, Vol. 86, No. 2, August 2007 161 Karima Ait Chihab et al. with MspI, showed heterozygosity in II-2, II-3, II-4 and III-1 fer protein (Guérin et al. 1994) or overexpression of SR-BI, subjects, and homozygosity in I-4 subject. a hepatic receptor for HDL (Wang et al. 1998). Variations The cosegregation analysis (figure 1) showed a non- in any gene involved in HDL metabolism, like mutations in cosegregation of the IVS3-23C→A mutation with the HALP LPL gene, could also be related to HALP associated with phenotype: the proband II-5, II-7, and II-8 subjects, with se- FH (Pimstone et al. 1995). In our studied family, no muta- vere HALP, and I-2 and II-6 subjects, with moderate HALP, tions were found in the LPL gene or in apoAI.Also,N370S did not carry the LCAT mutation. However, the subject I-4, and L444P, the most frequent mutations in GBA gene, as- with normal HDL-C levels, was homozygous for this muta- sociated with HALP, were not found. However, we identi- tion. fied the IVS3-23 C→A mutation in LCAT gene, previously described in Spanish heterozygous HALP subjects (Recalde Discussion et al. 2002). By introducing a new acceptor splicing dinu- cleotide AG, this mutation could partially induce an aberrant In this study, we describe a Moroccan family with a processing of LCAT mRNA and be responsible for the HALP combined phenotype of FH and HALP. The novel LDLR phenotype (Recalde et al. 2002). However, the noncosegre- gene mutation identified in this family, 756del7, induces a gation of IVS3-23 C→A mutation and the reduced HDL-C frameshift with creation of a stop codon at position 241 and levels observed in our Moroccan family is not in accordance it would cause a class 1 LDL receptor defect. This 756del7 with a causal effect of this mutation. mutation is likely to be the FH causing mutation, as nei- It has been reported that FH heterozygotes with reduced ther any other LDLR gene mutation nor the two most fre- HDL-C levels are prone to develop more severe premature quent FDB-causing mutations, R3500Q and R3531C,have artery disease when compared with FH heterozygotes with been identified in this family. Moreover, the 756del7 mu- normal HDL-C levels (de Sauvage Nolting et al. 2003). In tation has not been detected in a population of 107 healthy our studied family, all FH heterozygotes are asymptomatic controls, suggesting that it is not a polymorphism, but a dis- for myocardial ischaemia in electrocardiograms and cardiac ease causing mutation. The homozygous carriers of 756del7 scans. However, before drawing any conclusions about the mutation are severely affected: the proband died prematurely increased susceptibility to atherosclerosis due to the combi- by CHD, his two sisters are affected by severe hypercholes- nation of HALP and FH phenotypes in the studied family, terolemia and one of them shows ischemia complications. more cardiovascular analyses need to be performed. Four out of six heterozygotes have LDL-C levels in nor- In conclusion, the 756del7 novel mutation in LDLR gene mal range. Several examples of FH with low LDL-C have appears to be the FH causing mutation in the studied fam- been reported and the lack of correlation between phenotype ily. While the correlation between genotype and phenotype and genotype has been explained by environmental (Sun et is clear in homozygotes, it is less evident in heterozygotes al. 1994), and/or genetic factors (Sass et al. 1995). For the for FH. These results show the improved diagnostic preci- family reported here, environmental factors could explain in sion obtained by introducing genetic diagnosis in FH fami- part the interindividual phenotype variation (data not shown). lies with identified mutation (Civeira 2004), even when the The presence of homozygous apoE3 genotype in all the het- mutation spectrum of the LDLR gene in Moroccan popula- erozygotes of 756del7 mutation would exclude this gene in tion seems to be heterogeneous (El Messal et al. 2003, 2006; FH phenotype suppression, but does not exclude the probable Chater et al. 2006). Finally, further studies are needed to presence of other genetic factors. elucidate the molecular basis of the HALP observed in the All carriers of the 756del7 mutation present severe studied family and the role of the reduced HDL-C levels in HALP, particularly pronounced in the homozygotes. Signifi- the development of atherosclerosis in a FH context. cantly reduced plasma levels of HDL-C have been reported in many FH families but the mechanism of this HALP is poorly Acknowledgements understood. Kinetic studies suggest that the catabolism of We are indebted to all members of the studied family for their coop- HDL-apo AI is increased in both homozygous and heterozy- eration. Our thanks to Dr Assali (cardiologist, Larache, Morocco) gous FH patients, which is thought to be unrelated to the for his collaboration. We also thank the group of Prof. C. Boileau magnitude of hypercholesterolemia (Frénais et al. 1999). In (INSERM U383) for their help with the LPL gene study. This work / our FH HALP subjects, an interaction between LDL and was supported in part by “Action integree Maroco-Espagnole no. HDL metabolism is likely to occur, since we observed an 54/PR/99” and a grant of “Fondo de Investigaciones Sanitarias FIS ff important di erence between homozygous and heterozygous PI031106, Spain”. FH patients regarding their HDL-C levels, and also found that the decrease of LDL-C in the proband after lipid low- References ering therapy was accompanied by a large increase of HDL- Abifadel M., Jambart S., Rabes J. P., Varret M., Allard D., Derre C (table 1). Other explanations have been put forward for A. et al. 2004 Identification of the first Lebanese mutation in the reduced HDL-C levels in lacking functional LDLR back- the LPL gene and description of a rapid detection method. Clin. ground, such an enhanced activity of cholesteryl ester trans- Genet. 65, 158–161.

162 Journal of Genetics, Vol. 86, No. 2, August 2007 FH associated with severe hypoalphalipoproteinemia

Abifadel M., Varret M., Rabes J. P., Allard D., Ouguerram K., Dev- Miller S. A., Dykes D. D. and Polesky H. F. 1988 A simple salting illers M. et al. 2003 Mutations in PCSK9 cause autosomal dom- out procedure for extracting DNA from nucleated cells. Nuclic inant hypercholesterolemia. Nat. Genet. 34, 154–156. Acids Res. 16, 1215. Beutler E., Gelbart T. and West C. 1990 The facile detection of Pimstone S. N., Gagne S. E., Gagne C., Lupien P. J., Gaudet D., the nt1226 mutation of glucocerebrosidase by mismatched PCR. Williams R. R. et al. 1995 Mutations in the gene for lipopro- Clin. Chim. Acta 194, 161–166. tein lipase: a cause for low HDL cholesterol levels in individu- Brustolin D., Maierna M., Aguzzi F., Zoppi F., Tarenghi G. and als heterozygous for familial hypercholesterolemia. Arterioscler. Berti G. 1991 Immunoturbidimetric method for routine determi- Thromb. Vasc. Biol. 15, 1704–1712. nations of apolipoproteins A-I and B. Clin. Chem. 37, 742–747. Pocovi M., Cenarro A., Civeira F., Torralba A., Perez-Calvo J. I., Chater R., Rabes J. P., Ait Chihab K., Varret M., Loutfi M., Ee Jahiri Mozas P. et al. 1998 Beta-glucocerebrosidase gene locus as a link Y. et al. 2006 Mutational heterogeneity in low-density lipopro- for Gaucher’s disease and familial hypo-alpha-lipoproteinaemia. tein receptor gene related to familial hypercholestrolemia in Mo- Lancet 351, 1903–1904. rocco clin. clin. Acta. 373. 62–69. Rabès J. P., Varret M., Saint-Jore B., Erlich D., Jondeau G., Civeira F. 2004 International panel on management of familial hy- Giraudet P. et al. 1997 Familial ligand-defective apolipopro- percholesterolemia: guidelines for the diagnosis and manage- tein B-100: simultaneous detection of the Arg3500→Gln and ment of heterozygous familial hypercholesterolemia. Atheroscle- Arg3531→Cys mutations in a French population. Hum. Mut. 10, rosis 173, 55–68. 160–163. de Sauvage Nolting P. R. W., Defesche J. C., Buirma R. J. A., Hut- Recalde D., Cenarro A., Garcia-Otin A. L., Gomez-Coronado ten B. A., Lansberg P. J. and Kastelein J. J. P. 2003 Prevalence D., Civeira F., Pocovi M. 2002 Analysis of apolipoprotein and significance of cardiovascular risk factors in a large cohort of A-I, lecithin:cholesterol acyltransferase and glucocerebrosidase patients with familial hypercholesterolemia. J. Intern. Med. 253, genes in hypoalphalipoproteinemia. Atherosclerosis 63, 49–58. 161–168. Sass C., Giroux L. M., Ma Y., Roy M., Lavigne J., Lussier- El Messal M., Ait Chihab K., Chater R., Vallve J. C., Bennis F., Cacan S. et al. 1995 Evidence for a cholesterol-lowering gene in Hafidi A. et al. 2003 Familial hypercholesterolemia in Morocco: a French-Canadian kindred with familial hypercholesterolemia. first report of mutations in the LDL receptor gene. J. Hum. Genet. Hum. Genet. 96, 21–26. 48, 199–203. El Messal M., Ait Chihab K., Chater R., Kettani A., Hafidi A., Soria L. F., Ludwig E. H., Clarke H. R. G., Vega G. L., Grundy Adlouni A. 2006 Which management of familial hypercholes- S. M., McCarty B. J. et al. 1989 Association between a spe- terolemia in Morocco? Acta. Cardiologica 61(5), 551–555 cific apoB mutation and familial defective apo B 100. Proc. Natl. Frénais R., Ouguerram K., Maugeais C., Marchini J. S., Benlian Acad. Sci. USA 86, 587–591. P., Bard J. M. et al. 1999 Apolipoprotein A-I kinetics in het- Sun X. M., Patel D. D., Webb J. C., Knight B. L., Fan L. M., Cai erozygous familial hypercholesterolemia: a stable isotope study. H. J., Soutar A. K. et al. 1994 Familial hypercholesterolemia in J. Lipid Res. 40, 1506–1511. China: identification of mutations in the LDL-receptor gene that Friedewald W. T., Levy R. I. and Fredrickson D. S. 1972 Estima- result in a receptor-negative phenotype. Arterioscler. Thromb. tion of the concentration of low-density lipoprotein cholesterol Vasc. Biol. 14, 85–94. in plasma, without use of the preparative ultracentrifuge. Clin. Tsuji S., Choudary P. V., Martin B. M., Stubblefield B. K., Mayor Chem. 18, 499–502. J. A., Barranger J. A. and Ginns E. I. 1987 A mutation in the Guérin M., Dolphin P. J. and Chapman M. J. 1994 Preferential human glucocerebrosidase gene in neuronopathic Gaucher’s dis- cholesteryl ester acceptors among the LDL subspecies of sub- ease. New Engl. J. Med. 316, 570–575. jects with familial hypercholesterolemia. Arterioscler. Thromb. Wang N., Arai T., Ji Y., Rinninger F. and Tall A. R. 1998 Liver- Vasc. Biol. 14, 679–685. specific overexpression of scavenger receptor BI decreases lev- Hixon J. E. and Vernier D. I. 1990 Restriction isotyping of human els of very low density lipoprotein ApoB, low density lipoprotein apolipoprotein E by gene amplification and cleavage with HhaI. ApoB, and high density lipoprotein in transgenic mice. J. Biol. J. Lipid Res. 31, 545–548. Chem. 273, 32920–32926.

Received 15 July 2005

Journal of Genetics, Vol. 86, No. 2, August 2007 163