Material and methods

One hundred and thirty unrelated subjects (53 females and 77 males) with hypercholesterolaemia (plasma total cholesterol above 7.0 mmol/l) were studied. Two pedigrees (10 family members) of carriers of the R3500Q mutation were available for analysis as well. All subjects were recruited after their informed consent.

Blood samples were drawn in 5.0 ml EDTA anticoagulated Vacutainer tubes (Sherwood Medical, Germany) after an overnight fast. Plasma lipid and apolipoprotein parameters (total cholesterol, HDL cholesterol, triglycerides, ApoA1, and ApoB-100) were analysed on an automated analyser (Cobas-Mira Plus, Hoffman-LaRoche Diagnostics Inc, Germany), and the concentrations of LDL cholesterol were determined using Friedewald's formula. Laboratory variables for the carriers of the R3500Q mutation were compared with those for non-carriers using Student's t test. Two tailed test was applied and p values <0.05 were considered significant for rejection of the null hypothesis.

DNA was extracted by standard techniques27 from 10 ml whole venous blood collected in EDTA anticoagulated Vacutainer tubes (Sherwood Medical, Germany) and stored at 4°C until analysis. Detection of the R3500Q mutation was done by competitive allele specific PCR.28 The allele frequency was estimated using the “gene counting” method. Screening of theApoB region (nucleotides 10551-10895) coding for a part of the ApoB-100 receptor binding domain was done by single strand conformation analysis (SSCA). Amplicons were obtained by PCR with oligonucleotides under conditions used by Tybjaerg-Hansenet al 29 and were separated using a procedure described elsewhere.30 Five polymorphisms within the region of ApoB were analysed. Ins/Del in the promoter region (sp24/27),XbaI (codon 2488),MspI (codon 3611),EcoRI (codon 4154), and 3′VNTRs were analysed as described previously,31-33 and their cosegregation with the R3500Q mutation was followed in the available pedigrees.

Results

Four of the 130 unrelated hypercholesterolaemic subjects were found to be carriers of the R3500Q mutation (0.031, 95% CI 0.01-0.082). Three additional carriers were identified among the 10 studied members of the two pedigrees.

The comparison of laboratory parameters of the seven R3500Q mutation carriers identified in this study (four probands and three family members) with the group of 126 hypercholesterolaemic non-carriers showed that the carrier group as a whole was not significantly different from the non-carrier group (table 1). The individual values of total cholesterol, LDL cholesterol, and ApoB-100 of the R3500Q mutation carriers ranged from 6.11 to 10.90 mmol/l, from 4.44 to 9.22 mmol/l, and from 1.41 to 2.03 g/l, respectively, showing broad interpersonal and intrafamilial variability. Similar findings have been observed in samples of hypercholesterolaemic subjects from other European populations and are attributed to differences in environmental factors, diet, and gene-gene interactions. It is worth noting that some of the mutation carriers have phenotypes indistinguishable from those of healthy, normolipaemic, non-carriers of the same gender and age by their forties, meaning that presymptomatic identification of the R3500Q mutation in this subgroup, although highly desirable, would require wide scale screening.

Mutations in the receptor binding domain resulting in mild to moderate hypercholesterolaemia have been reported in some other populations with a much lower frequency than the R3500Q mutation.2 3 Our screening of the ApoB region (nucleotides 10 551-10 895), coding for a part of the APOB-100 receptor binding domain, showed two distinct mobility patterns (results not shown). Four of the samples coinciding with the heterozygous samples for the R3500Q mutation, as shown by competitive allele specific PCR analysis, showed additional sharp and clearly separated bands in comparison with the remaining 126 identical patterns. Thus, other mutations in the screened region, apart from R3500Q, were excluded in this sample of Bulgarian hypercholesterolaemic subjects by SSCA. Therefore, we conclude that other mutations in this region are rare causative factors for hypercholesterolaemia in Bulgaria.

Analysis of the five polymorphisms within the region ofApoB-100 in the two pedigrees showed unequivocal linkage of the R3500Q mutation with a haplotype Ins,XbaI−, MspI+,EcoRI−, 49 3′VNTR (I, X−, M+, R−, 49 3′VNTR). The results from one of the two pedigrees are shown in fig 1. The association of the mutation with the same haplotype could not be ruled out by the observed genotypes of the other two unrelated carriers with hypercholesterolaemia, who did not have families available for analysis. The association of the R3500Q mutation in the Bulgarian population with the same rare haplotype, which is reported to be associated with the mutation almost exclusively across different ethnic groups,7 supports the hypothesis of a common origin of the mutation.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Pedigree of a proband (I.1) carrying the R3500Q mutation. All R3500Q mutations are associated with the haplotype I, X−, M+, R−, 49 3′VNTR (shown on a dark background). The individual laboratory variables of the family members are shown below the pedigree.

Discussion

In our sample of unrelated hypercholesterolaemic subjects from Bulgaria, the R3500Q mutation accounts for 0.99-8.17% (95% CI) of the cases with hypercholesterolaemia (defined as total cholesterol >7.0 mmol/l), and represents the most common single gene defect resulting in hypercholesterolaemia identified so far in Bulgaria. Population studies and estimations based on extrapolation from the prevalence of the R3500Q mutation among hypercholesterolaemic subjects show a nearly 18-fold difference in its frequency in different populations.5 6 Recently, Myant et al 7 estimated that the mutation originated some 6000-7000 years ago. Miserez and Muller22 confirmed the mutation's age and hypothesised that the founder mutation arose in the Celtic population, settled in central Europe, and spread across Europe. Data on mutation frequencies reviewed in this study, together with the mutation's high prevalence in populations from eastern Europe,11 12 19 lead us to assume that clear distribution gradients can be tracked from central Europe in all directions except to the south east, for which continuous data are available for Hungary only (fig 2). On the other hand, as for other fringes of Europe,23-25 the mutation is extremely rare at its most south easterly edge (Turkey26). Very recently, commenting on genetic diversity in Europe as a result of its geographical versus linguistic separation, Rosseret al 34 suggested that “populations such as the Hungarians and Turks are unlikely to be separated from surrounding populations by genetic barriers”. As to the R3500Q mutation, these theories raise the question of how far the mutation spread has reached in the hitherto unstudied populations in the south east between Hungary and Turkey. Taking into account the frequency of people with total cholesterol above 7.0 mmol/l (6.34%) among an unselected group of 4800 Bulgarians (N Vassilevski, Countrywide Integrated Noncommunicable Diseases Intervention - CINDI - Program, personal communication), and the frequency of the R3500Q mutation in our group, one can estimate a prevalence of 1:451 in adult Bulgarians. This places the Bulgarian population among the European populations with a medium prevalence of the mutation. Our findings also suggest that along with its spread in other European areas, the mutation's expansion has moved to the south east and involved the Balkans. Given that Bulgaria is situated in the heart of the Balkans, occurrence of the R3500Q mutation might be expected in other neighbouring populations; this requires further studies on other populations from the region.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Estimated prevalence of the R3500Q mutation in European countries. Prevalences are ranked in descending order.

In conclusion, this is the first study showing occurrence of the R3500Q mutation in south eastern Europe. The mutation is linked to the same rare haplotype as in the other populations; this supports the hypothesis of its common origin and suggests that along with its spread from central Europe to other European areas, the mutation's expansion has moved to the south east.