Premature ovarian failure and FRAXA premutation: Positive correlation in a Brazilian survey

INTRODUCTION

Fragile X syndrome (FRAXA) is the most common form of inherited mental retardation (MR), with an estimated prevalence of 1:4,000 in males and 1:6,000 in females [Turner et al., 1996] and a carrier frequency of 1:259 [Rousseau et al., 1995]. It is an X-linked disorder inherited as a dominant trait with incomplete penetrance. The molecular mechanism of the disease involves an expansion of a trinucleotide repeat—(CGG)_n—located at the 5′ untranslated region of the FMR-1 gene, which leads to transcriptional silencing by hypermethylation of the repeat itself and an adjacent CpG island [Fu et al., 1991; Oberlé et al., 1991; Pieretti et al., 1991; Verkerk et al., 1991]. The inactivation of the gene results in the symptoms of the syndrome as a consequence of absence of the coded protein, named FMRP [Verheij et al., 1993].

Usually normal individuals harbor 6–40 CGG alleles whereas affected ones show >200 [Fu et al., 1991; Oberlé et al., 1991]. Premutated alleles range from 61 to 200 repeats and can generate a full mutation when passed from a female to her offspring [Rousseau et al., 1991; Grasso et al., 1996]. Such females are called premutation carriers and they are phenotypically normal.

Premature ovarian failure (POF) is a highly heterogeneous disorder characterized by a cessation of ovarian functioning in women before the age of 40 years [Allingham-Hawkins et al., 1999]. Most POF cases are idiopathic or sporadic and this phenomenon occurs in the normal population with an estimated frequency of 1% [Coulam et al., 1986]. Moreover, the occurrence of familial cases of POF, showing a dominant pattern of inheritance as well as an association with X-chromosome abnormalities [Sala et al., 1997; Vegetti et al., 1998] has been described.

The identification of critical regions for normal ovarian function at Xq13.3-q21.1 [Powell et al., 1994] and Xq 26-28 [Skibsted et al., 1984] suggests that disruption of genes in these regions could be associated with POF.

In the last years several studies have suggested a direct correlation between premutation status at the FRAXA locus and an increased incidence of ovarian failure [Cronister et al., 1991; Schwartz et al., 1994; Turner et al., 1994; Conway et al., 1995, 1998; Partington et al., 1996; Vianna-Morgante et al., 1996, 1999; Murray et al., 1998; Allingham-Hawkins et al., 1999; Braat et al., 1999; Holden et al., 1999; Syrrou et al., 1999; Uzielli et al., 1999; Marozzi et al., 2000; Hundscheid et al., 2001]. Recently, Murray et al. [1999] showed that microdeletions in FMR-2, a gene located 600Kb distal to FMR-1 and associated with mild to moderate mental retardation without dysmorphic features, could also be associated with POF.

Preliminary data reported by an international collaborative project, concerning the analysis of 760 women from fragile X syndrome families worldwide, indicate that 16% of premutation carriers have experienced menopause prior to age 40 years, which is a significantly higher proportion than that of the normal population [Allingham-Hawkins et al., 1999]. In 1997, the Human Genetics Service of Universidade do Estado do Rio de Janeiro (UERJ) initiated a study on fragile X families, identified after characterization of a fully mutated patient, aiming to establish an association between POF and premutation carrier status. Our findings are reported below.

MATERIALS AND METHODS

RESULTS

We have contacted, interviewed and analyzed 58 women from 24 fragile X syndrome families identified after molecular confirmation of the full mutation in a family member referred for FRAXA testing. Southern blot analysis revealed 56.9% premutation carriers (33/58), 10.4% full mutation carriers mutated (6/58), and 32.7% normal women (19/58). Among the premutation carriers, 33.3% (11/33) experienced POF (including one case of somatic mosaicism), a fraction different from the normal individuals identified in these families, who did not experienced ovarian failure (0/19). Only one of the fully mutated women we identified shows a fragile X phenotype, including MR, facial dysmorphism, tactile defensiveness and hyperactivity. She was 16-years old at the time of analysis and had not yet experienced POF. The range of menopause in the premutation carrier group was 27–39 ± 3.9 years while the range of normal women when evaluation was performed was 25–40 ± 4.8 years.

DISCUSSION

We have performed a genetic investigation in a survey of 58 women from 24 fragile X families. For the purposes of this study, we have adopted a strict definition of POF as complete cessation of periods prior to age of 40 years as published elsewhere [Allingham-Hawkins et al., 1999]. We have identified 11 that have experienced POF among 33 premutation carriers (33.3%), which is a significantly higher proportion than expected in the general population. Recent findings demonstrate an incidence of 18.8% of POF in premutated women [Uzielli et al., 1999] and, according to the data published by the International collaborative group, the estimated frequency of premutation carriers who experienced POF is 16% [Allingham-Hawkins et al., 1999].

The age of onset of menopause observed in our sample was variable as described earlier [Vegetti et al., 1998] and could suggest expression variability or the anticipation phenomenon. However, it differs significantly from the normal age estimated for menopause, which is around 50 years [OMIM, 2001].

One of the POF/premutated women had a somatic mosaism as revealed by Southern analysis. It has been reported that those with somatic mosaism may show differences in the expansion of the trinucleotide repeat in different tissues [Kambouris et al., 1996; Orrico et al., 1998; Schmucker and Seidel, 1999]. As our DNA analysis was restricted to peripheral blood cells we could not exclude the possibility that the full mutation was absent in ovarian tissues. A more detailed study of this case is required to determine the extension of the mosaicism.

Our observations are concordant with the hypothesis that fragile X syndrome carriers are at high risk of premature ovarian failure as proposed by other authors [Cronister et al., 1991; Schwartz et al., 1994; Turner et al., 1994; Conway et al., 1995, 1998; Partington et al., 1996; Vianna-Morgante et al., 1996, 1999; Murray et al., 1998; Allingham-Hawkins et al., 1999; Braat et al., 1999; Holden et al., 1999; Syrrou et al., 1999; Uzielli et al., 1999; Marozzi et al., 2000; Hundscheid et al., 2001]. There is evidence that X chromosome abnormalities could be associated with ovarian dysfunction and the identification of two critical regions for normal ovarian function at Xq13.3-q21.1 and Xq26-28 [Skibsted et al., 1984; Powell et al., 1994; Sala et al., 1997] support this idea.

Recently, a relationship between ovary gene function and FRAXA status was proposed [Syrrou et al., 1999; Davison et al., 2000] and only the premutation appears to exert influence in ovary function [Uzielli et al., 1999]. Early signs of menopause were identified in 9 from 10 premutated women analyzed in a pilot study in which menstrual cycles and endocrine profile were evaluated [Braat et al., 1999]. Furthermore, it has been proposed that premutation at the FRAXA locus could affect nearby genes involved in ovary function [Vianna-Morgante et al., 1996]. In this case, POF could be an extreme effect in a spectrum of ovarian abnormalities associated with fragile X premutation [Vianna-Morgante et al., 1999]. The recent association of FRAXE microdeletions with POF occurrence [Murray et al., 1999] reinforces the importance of Xq26-28 as a critical region for normal ovary function

The mechanism by which premutated status at the FRAXA locus could interfere in normal ovarian function is still unclear. It has been proposed that premutated alleles might prevent the normal number of oocytes from being produced [Schwartz et al., 1994; Conway et al., 1995].

Since the estimated frequency of carriers is 1/259 [Rousseau et al., 1995] the recognition of the association between POF and premutation status suggests it is important to identify these females among POF patients, with the aim of performing adequate genetic counseling [Conway et al., 1998]. Our data provide more evidence that POF and premutation status at FRAXA locus are associated, corroborating the idea that females carrying alleles in the premutation range are at high risk of experiencing POF.