Introduction

Familial Mediterranean fever (FMF) is an autoinflammatory, autosomal recessive disorder, prevalent in the Mediterranean basin and characterized by recurrent attacks of fever and serositis. The MEditerranean FeVer gene (MEFV) encodes a protein, called pyrin/marenostrin, which is involved in the regulation of inflammation and apoptosis.1 Mutations in MEFV (more than 50 thus far) have been associated with FMF. In Israel, FMF most commonly affects Jews of North African and Iraqi extraction (Sepharadi Jews), in whom disease frequency is 1:200 and 1:1000, respectively,2 whereas in Ashkenazi Jews, originating from Central Europe, the disorder is rare with a frequency of 1:160 000 only.3 However, direct assessment of mutation frequency, as reported in several studies, revealed quite surprising results of extremely higher-than-expected rates.4, 5, 6 Furthermore, the mutation rates in these two ethnic groups, so vastly separated by FMF disease frequency, were quite comparable, 1 in 5 vs 1 in 7–8 in Sepharadi and Ashkenazi Jews, respectively. These findings suggest that only direct testing of MEFV mutation frequency is appropriate for the assessment of its role.

This study aims to define the frequency of MEFV mutations in the Israeli-Jewish population with low risk for FMF, specifically those originated from the Jewish diaspora of Buchara, Georgia, Yemen, Turkey and Bulgaria. Knowledge of the prevalence of MEFV mutation carriage may help in elucidating their biological advantage and their role in the modulation of non-FMF inflammatory disorders.

Materials and methods

The study group included 163 Jewish individuals, born to parents, both originating from the same country, Buchara, Turkey, Georgia, Yemen or Bulgaria. They were enrolled, provided they stated no personal or family history of FMF. Participants were interviewed, and a questionnaire focusing on manifestations indicative of inflammatory diseases was completed. The study was approved by the institutional review board, and all participants signed an informed consent.

DNA was analyzed for the three most common Jewish MEFV mutations (M694V, V726A, E148Q), using PCR amplification and restriction enzyme digestion.7

Statistical analysis was performed for categorical variables using χ2-test or Fisher's exact test, according to the size of the cells examined; and for continuous variables using the Student's t-test. All tests of significance were two-tailed; P-values <0.05 were considered statistically significant.

Results

A total of 163 individuals, originating from ethnicities with low rates of FMF, were tested for mutations in the MEFV gene. Baseline demographic data are summarized in Table 1. There were 27 carriers of MEFV mutations, 2 of whom, with no manifestations of FMF, carried two mutations each. The mean age and gender distribution were equal among carriers and non-carriers of an MEFV mutation (58.3±17 vs 61.9±19 years and male/female ratio of 16/11 and 72/64, respectively). Mutation carriage rate and distribution of the three studied mutations by origin are detailed in Table 2. The E148Q mutation was the most prevalent, found in 19/326 chromosomes (6%).

Table 1 Demographic characteristics of study participants by origin
Table 2 MEFV mutation by ethnic origin

The rate of FMF-like conditions, which may indicate a hitherto unrecognized FMF, amyloidosis, splenomegaly (indicative of subclinical inflammation) and a long list of inflammatory diseases, more commonly encountered in FMF, among mutation carriers and non-carriers, are listed in Table 3. There was a statistically significant excess of the combined frequency of FMF-like manifestations (acute arthritis, pericarditis, erysipelas episodes and Behçet's-like disease) in MEFV mutation carriers (P=0.001 for the combination). No difference was noted in the frequency of non-inflammatory diseases aside from hypertension, which was unexplainably overrepresented among non-carriers (67/136 vs 7/27 in carriers, P=0.03, respectively).

Table 3 Frequency of FMF-related conditions

Discussion

The rate of MEFV mutation carriage in Israeli Jewish individuals from ethnicities with a low risk for FMF may be as high as 20%, reaching the frequencies described in North-African Jews, in which overt FMF is prevalent. The combination, in these ethnicities, of a high mutation frequency, on the one hand, with a low disease prevalence, on the other, indicates low mutation penetrance. Indeed, E148Q and V726A, which are considered ‘weak’ mutations,8 were the most commonly found in most ethnic groups except for Georgian individuals, in whom M694V, which is associated with a more severe disease phenotype, as well as amyloidosis,9 was commonly detected (M694V/E148Q ratio of 4/5).

The finding of a high mutation frequency in Jewish communities immigrated to Israel from Eastern Europe, supports the view that this population has actually emerged originally from an area endemic to a hitherto unrecognized pathogen, from which the mutations protect, comparable with Jewish subgroups with high risk for FMF.

On the basis of our experience at the National Center for FMF, which holds in its registry data on most FMF patients in Israel, the frequency of FMF in Georgian Jews is very low (1:5000). This suggests that Georgian Jews carry a modifier gene or system, or are exposed to an environmental moderating factor, which inhibits the clinical expression of the most penetrant MEFV mutation, M694V.

Finally, our findings reiterate the higher frequency of phenotype 3 (genetic positive, clinically negative individuals) in Jewish individuals and the occurrence of FMF-like symptoms and inflammatory diseases in MEFV mutation carriers. The reduced rate of hypertension in mutation carriers needs to be ascertained and further investigated.