Mutation analysis of two candidate genes for premature ovarian failure, DACH2 and POF1B

S. Bione1, F. Rizzolio2, C. Sala2, R. Ricotti1, M. Goegan2, M.C. Manzini1, R. Battaglia1, A. Marozzi3, W. Vegetti4, L. Dalprà5, P.G. Crosignani4, E. Ginelli3, R. Nappi6, S. Bernabini7, V. Bruni8, F. Torricelli7, O. Zuffardi9 and D. Toniolo1,2,10

1Institute of Molecular Genetics-CNR, 27100 Pavia, 2Dibit-San Raffaele Scientific Institute, 20132 Milano, Departments of 3 Biology and Genetics for Medical Sciences and 4Obstetrics and Gynecology, University of Milano, 20133 Milano, 5San Gerardo Hospital, University of Milano Bicocca, 20126 Milano, Departments of 6 Obstetrics and Gynecology and 9 Pathology and Medical Genetics, University of Pavia, 27100 Pavia, 7 Cytogenetics and Genetics Unit, Azienda Ospedaliera Careggi, 50139 Firenze and 8 Department of Gynecology and Human Reproduction, University of Firenze, Firenze, Italy

10 To whom correspondence should be addressed at: Dibit-San Raffaele Scientific Institute, Via Olgettina 58, 20132 Milano, Italy. Email: toniolo.daniela{at}hsr.it


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: Balanced X;autosome translocations interrupting the ‘critical region’ of the long arm of the human X chromosome are often associated with premature ovarian failure (POF). However, the mechanisms leading to X-linked ovarian dysfunction are largely unknown, as the majority of the X chromosome breakpoints have been mapped to gene-free genomic regions. A few genes have been found to be interrupted, but their role has never been clarified. METHODS AND RESULTS: By fine mapping of the X chromosome breakpoint of an X;autosome balanced translocation, we identified a new interrupted gene, POF1B. We performed a mutation analysis of POF1B and of another gene previously identified, DACH2, localized ~700 kb distal in Xq21, in a cohort of >200 Italian POF patients. Rare mutations were found in patients in both genes. CONCLUSIONS: Our findings could not demonstrate any involvement of POF1B, but suggest that rare mutations in the DACH2 gene may have a role in the POF phenotype.

Key words: DACH2/POF1B/premature ovarian failure/susceptibility gene


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Premature ovarian failure (POF; OMIM 311360) is a disorder characterized by a lack of ovulation and elevated serum gonadotropin level before 40 years of age. POF has a frequency of ~1% among females, but with the recent increase in womens reproductive age it has become a relevant cause of infertility, accounting for ~10% of all female sterility (Luborsky et al., 2003Go). Environmental factors such as infections, stress and antitumour therapies have been implicated in the aetiology of the disorder. The frequent finding of POF familial cases, together with the relative frequency of chromosomal abnormalities, suggests a major genetic component. However, the limited extent of the known POF pedigrees and the observed heterogeneity in genetic segregation, ranging from autosomal recessive to autosomal dominant or sex-limited X-linked inheritance, hampered the definition of a precise pattern of inheritance and the identification of contributing loci.

Rare mutations in a few genes responsible for autosomal recessive (Aittomaki et al., 1995Go) autosomal dominant (Di Pasquale et al., 2004Go) or syndromic forms (Crisponi et al., 2001Go; Fogli et al., 2003Go) of the disease were identified. Recently, two further genes have been implicated in the POF phenotype. One is the premutation allele of the FMR1 gene at the FRAXA locus in Xq27, shown to act as a risk factor for POF with an estimated relative risk of ~20 (Murray, 2000Go). The other is the G769A mutation in the inhibin {alpha} (INHA) gene, which was found to be significantly associated with POF in two independent studies (Shelling et al., 2000Go; Marozzi et al., 2002Go). Taken together these data indicate that POF may be rarely inherited as a mendelian disorder, and more often is due to mutations in susceptibility genes contributing to POF in a multifactorial fashion.

The involvement of the X chromosome in POF arose from the frequent observation of sex chromosome anomalies in POF patients. Turner syndrome (TS), partial X chromosome monosomies and X;autosome balanced translocations have been observed in association with POF, and their description led to the cytogenetic definition of a ‘critical region’ for normal ovarian function on the long arm of the X chromosome, corresponding to the Xq13.3-q26/27 interval (Therman et al., 1990Go). Molecular definition of the critical region by fine mapping of X;autosome balanced translocations confirmed their distribution along a large genomic region, suggesting the presence of several loci involved in the disorder (Sala et al., 1997Go; Prueitt et al., 2000Go; Mumm et al., 2001Go). The finding of chromosomal rearrangements in normal women interspersed with those associated with POF showed that the rearrangements per se could not be responsible for the disorder, but alternative explanations, such as interruption of genes in the critical region, position effects on the expression of X-linked or autosomal genes, or increased apoptosis due to meiotic mispairing of the X chromosomes during oocyte maturation, could not be excluded.

Transcriptional characterization of breakpoint regions led to the identification of three genes interrupted by translocations, the DACH2 gene in Xq21.3 (Prueitt et al., 2002Go), the DIAPH2 gene in proximal Xq22 (Bione et al., 1998Go) and the XPNPEP2 gene in Xq25 (Prueitt et al., 2000Go). Most of the other breakpoints described in POF patients were mapped to genomic regions free of transcribed sequences (Mumm et al., 2001Go; Prueitt et al., 2002Go), suggesting that different genetic mechanisms may be responsible for X-linked POF. However, the role of the interrupted genes was never clarified.

Here we report the mutation analysis of two genes, DACH2 (Mumm et al., 2001Go; Prueitt et al., 2002Go) and POF1B, a fourth gene found to be interrupted by an X;1 balanced translocation in a patient presenting secondary amenorrhea at the age of 17 years (Riva et al., 1996Go). The two genes are localized 700 kb apart in Xq21, a gene-poor region where several balanced translocation breakpoints were mapped (Sala et al., 1997Go; Prueitt et al., 2000Go; Mumm et al., 2001Go). Searching for mutations in a large cohort of >200 Italian POF patients with normal karyotype revealed rare coding variants in both genes. Analysis of the mutations found in patients in a panel of >900 selected female controls failed to show association of POF1B, but suggested that rare variants in DACH2 might be related to the POF phenotype.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Patients information
A collection of 275 genomic DNAs of POF-affected patients from Northern and Central Italy was established. Patients presented primary amenorrhea (n=19) or secondary amenorrhea (n=230), with onset between 12 and 40 years of age. In addition, a group of women (n=26) entering menopause between 40 and 45 years of age was studied. This group of patients was defined as early menopause (EM), and was introduced in the study because EM is sometimes found in POF families (Tibiletti et al., 1999Go). In all cases, patients had amenorrhea from at least 6 months, and elevated serum gonadotropin levels (FSH >40 IU/l; LH >15 IU/l). All patients presented normal karyotype as determined by standard banding techniques on at least 30 metaphases with the exception of one patient, who carried a Robertsonian translocation [45,XX,t(13;14)], not involving loss of DNA. The premutation allele at the FRAXA locus was investigated in all cases: 10 patients (3.6%) carried the premutation allele. All the patients included in the study gave informed consent to the molecular analysis of their DNA. Anonymous geographically matched controls (>1100) were selected among women who had entered menopause at physiological age (>48 years). The number of controls was sufficient to verify the presence of alleles with frequency >0.05.

Statistical analysis
Case–control association was calculated for each coding single nucleotide polymorphisms (cSNP) using a two-tailed Fisher's exact test. Differences in allele frequencies were considered significant at P<0.05

Fluorescence in-situ hybridization mapping
Fluorescence in-situ hybridization (FISH) was performed on metaphase preparations, with PAC or BAC DNA labelled by nick-translation, with biotin-16 dUTP (Roche Diagnostics Gmbh, Mannheim, Germany) as described previously (Rossi et al., 1994Go). Genomic clones used as probes were obtained by PCR screening of the RPCI1 and RPCI5 human PAC libraries with primer pairs corresponding to Sequence Tag Sites (STSs).

Pulsed-field gel electrophoresis analysis and Southern blot
Genomic DNA from lymphoblastoid cell lines of patient LA1 and of female and male controls was digested with restriction enzymes as suggested by the supplier (Promega, Madison, WI, USA) and fractionated by pulsed-field gel electrophoresis (PFGE) in 1.5% agarose gel, at 170 V for 20 h at 14°C with pulse intervals of 5 s. Southern blots and hybridizations were performed using standard procedures.

Reverse transcriptase–PCR
Total RNA from lymphoblastoid and hybrid cell lines was extracted by standard techniques. Reverse transcription was performed from 1 µg of total RNA as described previously (Bione et al., 1998Go). PCRs were performed in 50 µl reactions with 0.2 mM dNTPs, 0.5 µM of each primer, 1.5 mM Mg2 + and 1.25 U Taq Polymerase (Promega) in the buffer supplied. Primer sequences and amplification condition for XIST and MIC2 (Tribioli et al., 1994Go) and for POF1B, CHM and ZNF6 can be found on-line at www.sanraffaele.org/research/toniolo.

DNA methylation analysis
Genomic DNA from male and female peripheral blood leukocytes and ectopic brain samples was extracted by standard procedures and digested with restriction enzymes as per supplier's instructions (Promega). Undigested samples were incubated in the same buffer and conditions but without enzyme. Digested and undigested DNAs were purified by ethanol precipitation and amplified with primers DACH-12F and DACH-20R in the same conditions used for denaturing high-performance liquid chromatography (DHPLC).

DHPLC mutation analysis
PCR fragments corresponding to all exons of the POF1B and DACH2 genes were amplified from primers pairs listed in Supplementary data (www.sanraffaele.org/research/toniolo). All PCR amplifications were carried out from 50 ng of genomic DNA in 50 µl volume with 0.2 mM dNTPs, 0.5 µM of each primer, 1.5 mM Mg2 + and 1.25 U Taq Polymerase (Promega) for 40 cycles. Cycles were for 30 s at 94°C, 30 s at suitable annealing temperature (see Supplementary data) and 30 s at 72°C, with 7 min of final elongation at 72°C. DHPLC analysis was performed on an automated HPLC (WAVE system; Transgenomic, Omaha, NE, USA). PCR products were subjected to denaturation and reannealing from 95°C to 65°C for 30 min prior to DHPLC analysis. Fragments were fractionated on a linear gradient of acetonitrile at a flow rate of 0.9 ml/min (see Supplementary data for temperature and gradient conditions (www.sanraffaele.org/research/toniolo).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The POF1B gene in Xq21 is interrupted by a POF-associated breakpoint
A total of 17 cases of POF-affected women carrying X;autosome balanced translocations were collected and finely mapped (Sala et al., 1997Go). One of the patients analysed, LA1, carried an X;1 balanced translocation and presented with secondary amenorrhea at the age of 17 years (Riva et al., 1996Go). The precise position of the X chromosome breakpoint was determined by FISH hybridization using PAC clones as probes. Clone dJ75N13 spanned the breakpoint (Figure 1A). The proximal overlapping clone dJ1090F18, on the derivative X (not shown), narrowed the breakpoint position to a 50 kb critical interval in Xq21.2 (Ensembl position: 83354549–83404549). At the time of the analysis, some predicted exons were present in the region and were experimentally verified by reverse transcriptase–PCR (RT–PCR): the presence of a new gene, called POF1B (NM_024921) was confirmed (not shown). The gene is composed of 17 exons spanning a region of ~100 kb, between the ZNF6 gene (5 kb proximal) and the CHM gene (~500 kb distal). The gene has no homology to known genes that could suggest a putative function, and is found only in vertebrates. To define the precise position of the breakpoint with respect to the POF1B gene, genomic DNA of patient LA1 and control DNAs were digested with the restriction enzyme BamHI, fractionated by PFGE and hybridized to genomic fragments corresponding to the exons in the breakpoint region. Hybridization to the exon 3 probe revealed the expected 35 kb band in all samples and an abnormal BamHI fragment of ~45 kb in patient LA1 (Figure 1B). The hybridization of the exon 4 probe showed the normal fragment of 35 kb as well as a second anomalous fragment of ~9 kb in the LA1 DNA. From the size of the two anomalous bands and the position of the BamHI sites, we mapped the LA1 breakpoint in the distal 4 kb of the third intron of the POF1B gene.



View larger version (65K):
[in this window]
[in a new window]
 
Figure 1. Mapping of the breakpoint in patient LA1. (a) FISH on a metaphase preparation of patient LA1. The probe used was PAC dJ75N13. Normal and derivative chromosomes are indicated by arrows. Chromosomes were stained with 4,6-diamino-2-phenylindole. (b) Southern blot analysis of genomic DNAs digested with BamHI restriction enzyme and fractionated by PFGE. XX and XY indicate genomic DNAs extracted from lymphoblastoid cell lines of female and male, respectively. LA1 is genomic DNA extracted from lymphoblastoid cell line of the patient. Probes were PCR fragments corresponding to the exons indicated and obtained by amplification of genomic DNA with the same primers used for DHPLC analysis.

 
The POF1B gene escapes X-inactivation
Genes involved in POF may be responsible for TS ovarian dysgenesis and may escape X-inactivation (Zinn and Ross, 1998Go). We studied the X-inactivation status of POF1B by RT–PCR amplification of total RNA from five human–hamster hybrid cell lines containing only the inactive (Xi) human X chromosome (Tribioli et al., 1994Go). As shown in Figure 2A, RT–PCR amplification of the POF1B transcript gave the expected 181 bp fragment in total human RNA from XX, XY, from the Xa-containing cell line and from the five different cell lines, containing only the inactive human X chromosome. In the same cell hybrids, the inactivation of the two genes flanking the POF1B gene was also determined. As previously reported, the ZNF6 gene was inactivated (Lloyd et al., 1991Go), while the CHM gene was inactivated only in three out of five of the cell lines containing only the inactive X chromosome (Carrel et al., 1999Go). Escape from inactivation of the POF1B gene was confirmed by RT–PCR amplification of total RNA from the lymphoblastoid cell line of the patient LA1 and of normal controls. Using primers in exons 2 and 6, flanking the breakpoint, a common fragment of 420 bp was obtained in all normal controls as well as in the LA1 patient (Figure 2B), indicating that the product of the normal X chromosome, which is inactivated, was present.



View larger version (25K):
[in this window]
[in a new window]
 
Figure 2. Analysis of the inactivation status of the POF1B and DACH2 genes. (a) RT–PCR amplification of the genes indicated from total RNA of lymphoblastoid cell lines from normal female (XX), normal male (XY), human–hamster hybrids containing only the active (Xa) or the inactive (Xi) X chromosome and from Chinese hamster ovary (CHO) cells. (b) RT–PCR products synthesized from total RNA of HeLa cell line (H), and lymphoblastoid cell lines from normal female (XX), normal male (XY) and patient LA1. Amplification was from primers MCM-B3F and MCM-8. The RT reactions were carried out with (+) or without (–) reverse transcriptase. ‘C’ indicates PCRs without template cDNA. Molecular weights are indicated on the right. (c) Analysis of DNA methylation of the CpG island of the DACH2 gene. Genomic DNA from female (XX) and male (XY) peripheral leukocytes (blood) and from ectopic brain samples were digested (+) or not (–) with the restriction enzymes indicated at the top and amplified with primers DACH-12F and DACH-20R surrounding the CpG island of the gene. At the bottom is a representation of the genomic region corresponding to the DACH2 CpG island in which the position of primers used and of restriction sites are indicated. The start codon position of the DACH2 gene is indicated by ATG.

 
Since genes escaping inactivation may have a Y chromosome homologue, the presence of a homologue of POF1B was excluded by Southern blot of XX, XY and Y only hybrid DNAs digested with four different restriction enzymes (EcoRI, PstI, HindIII and PvuII) and hybridized to the POF1B cDNA (from exon 5 to exon 13). The pattern of bands was identical in all DNAs, and the ratio between the amounts of DNA in each band did not change between XX and XY DNAs (not shown).

The DACH2 gene is inactivated
The pattern of inactivation of the DACH2 gene, localized 700 kb distal from POF1B and interrupted by a POF-associated breakpoint (Prueitt et al., 2002Go), was also investigated.

Since the DACH2 gene is not expressed in lymphoblastoid cell lines, its inactivation status was established by analysis of the methylation pattern of a CpG island at its 5' end (Figure 2C). Genomic DNA extracted from male and female peripheral blood leukocytes was digested with the restriction enzymes HpaII and BssHII, sensitive to DNA methylation. A fragment of 378 bp surrounding the start codon of the gene and containing two BssHII and five HpaII recognition sites was amplified from digested and non-digested DNAs. A PCR product was detected from undigested as well as from digested female genomic DNA, whereas it was not detectable in male DNA digested with either enzymes (Figure 2C). To verify that methylation of the DACH2 CpG island was not due to its lack of expression in white blood cells, we performed HpaII restriction analysis in DNA extracted from female brain samples in which the gene was expressed (not shown). The presence of a PCR product from amplification of HpaII-digested female DNA confirmed the inactivation status (Figure 2C).

Mutation analysis of the POF1B gene
The contribution of the POF1B to the aetiology of POF was evaluated by mutation analysis. In all patients, we analysed by DHPLC exons and splice junctions aiming at identification of causative mutations in coding regions and in splice junctions. Synonymous mutations and intronic variants of undetermined role were not considered. The 17 exons of the POF1B gene were analysed in 223 POF patients presenting primary amenorrhea (n=16) secondary amenorrhea (n=183) or EM (n=24), and revealed the presence of 30 SNPs. Fifty-two patients of the collection were not analysed as they were not available when the POF1B analysis was carried out, but they could not change the results obtained with the first group of 223. Five variants were found in the coding portion of the gene (cSNPs), and in all cases the nucleotide replacement resulted in a missense mutation of the POF1B protein. The T1191A/L349M substitution found in exon 10 is a common variant, with a frequency of the minor allele of 0.24. The genotypes at this locus were determined in 95 patients and in 95 controls, and did not show differences between cases and controls, excluding its association with the disease (data not shown). The remaining four variants (C239S, R329Q, Q434K and C444Y) were found in a total of six patients with menopause ranging in age from 22 to 40 years (Table I). The presence of these variants was assessed in a panel of 900 controls. The R329Q and the Q434K variants were found three times and twice, respectively, in the control group, whereas the C239S variant was found more frequently, as it appeared 18 times in the 900 control DNAs. The remaining 210 controls were not analysed as they were not enough to change the statistical significance of results. The C444Y variant was found twice among patients and was not found in controls. By itself, it showed a significant association to the disease (P=0.0393), but when corrections for multiple testing were introduced, the association was lost (P=0.157). The remaining three variants were not associated.


View this table:
[in this window]
[in a new window]
 
Table I. Analysis of rare cSNPs in the POF1B gene

 
Mutation analysis of the DACH2 gene
Analysis of the 13 exons of the DACH2 gene was performed in a panel of 257 Italian POF patients with primary amenorrhea (n=19), secondary amenorrhea (n=212) or EM (n=26). A total of 15 SNPs were identified, five of which were in the coding portion of the gene. All five cSNPs resulted in amino acid substitutions (P36L, R37L, G59D, F316S and R412K). The presence of the five variants was verified in a panel of 1110 female controls (Table II). Three of the variants (P36L, G59D and R412K) were found also in the control group. Each of the variants considered by itself did not show significant association with the POF condition (P > 0.1880). However, these missense mutations in the coding region of the DACH2 gene were more frequent in POF patients than in controls (P=0.0125).


View this table:
[in this window]
[in a new window]
 
Table II. Mutation analysis of the DACH2 gene

 
The five missense mutations effected evolutionary conserved amino acids. The 599 amino acid protein encoded by the DACH2 gene is a transcriptional cofactor characterized by the presence of three conserved domains. The DD1 domain at the N-terminus (amino acids 66–162) and the DD2 domain at the C-terminus (amino acids 452–543) are highly conserved in all members of the DACH protein family and appear to be involved in DNA binding and in the interaction with the EYA proteins, respectively (Davis et al., 2001Go; Kim et al., 2002Go). A third domain, which we named DD3 (Figure 3), is present in the central portion of the protein (amino acids 314–412) and is shared by all members of the DACH1 and DACH2 subfamilies, but is absent in the Drosophila dac protein. The functional role of the DD3 domain is not known. The P36, R37 and G59 occurred in the N-terminal portion of the protein close to the DD1 (Figure 3A). The three amino acid residues appeared to be specific to the DACH2 subfamily, as they resulted conserved in the DACH2 orthologue of Mus musculus, Gallus gallus and Danio rerio, but are not present in the DACH1 protein subfamily (not shown). The F316 and the R412 residues are inside the DD3 domain and are conserved in all known DACH1 and DACH2 orthologues and in the DACH protein of Xenopus laevis and Oryzias latipes (Figure 3B).



View larger version (88K):
[in this window]
[in a new window]
 
Figure 3. Conservation of DACH2 residues and position of the mutations found in POF patients. (a) Alignment of the N-terminal portion of the human DACH2 protein (NP_444511) and its mouse (Mm Dach2; NP_291083), chicken (Gg DACH2; AAF22354) and Danio rerio (Dr. dacha; AAL79823) orthologues. (b) Alignment of the DD3 domain of the DACH1 and DACH2 protein families. Besides the DACH2 proteins, the human (Hs. DACH1; NP_542937), chicken (Gg. DACH1; AAL76234), mouse (Mm. Dach1; NP_031852), rat (Rn. Dach1; XP_224440), Xenopus (Xl. Dach; CAD88222), medaka (Ol. dach; CAC48006), tetraodon (Tn. dach; CAG07301), D.rerio (Dr. dachc; NP_694489) and rat partial Dach2 (Rn. Dach2; XP_228459) were aligned. In both panels, asterisks indicate residues identical in all sequences, whereas double-dots and dots indicate conservative or semi-conservative substitutions, respectively, as calculated by the ClustalW multiple alignment program (http://www.ebi.ac.uk/clustalw/). The position of the missense mutations is indicated by arrowheads.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
In this paper we report the genetic analysis of two X-linked genes interrupted by X;autosome balanced translocations in POF patients, POF1B and DACH2. The two genes mapped about 700 kb apart in Xq21, in a gene-poor region where many X chromosome breakpoints have been mapped in our laboratory (Sala et al., 1997Go) and by others (Prueitt et al., 2000Go; Mumm et al., 2001Go). Two of them interrupted POF1B and DACH2, three interrupted the CHM gene (Cremers et al., 1990Go; van Bokhoven et al., 1994Go; Lorda-Sanchez et al., 2000Go) and several were mapped outside of genes (Rizzolio and Toniolo, unpublished results). Several mutations have been described in the CHM gene, but they were never associated with POF in choroideremia families (van den Hurk et al., 1997Go). The role of the other two genes and the effect of the chromosomal breakpoints was never clarified.

POF1B, a gene of unknown function, was found interrupted by the breakpoint of a de-novo X;1 balanced translocation in a patient presenting secondary amenorrhea at the age of 17 years (Riva et al., 1996Go). The POF1B gene is a good candidate for POF and for TS-associated ovarian dysgenesis, as it escapes X inactivation and has no Y homologue. POF1B maps to the long arm of the X chromosome, the region that first diverged between the X and the Y, between 200 and 300 million years ago (Lahn and Page, 1999Go): conservation of two active copies of the gene for this long evolutionary time is not common, and strengthens the hypothesis that a double dose of POF1B in females is functionally relevant. The pattern of expression of Pof1b in the ovary was studied in the mouse (not shown), where it was absent from adult ovary but was found between days E16.5 and P5, suggesting that the gene may be critical for one of the steps required for early ovary development leading to the final number of germ cells.

To clarify the contribution of POF1B in the aetiology of the disorder, we performed a case–control study on >200 POF patients from Northern and Central Italy and >900 geographically matched controls, selected among women who had entered menopause at the physiological age. In all patients we analysed exons and splice junctions, with the aim of identifying causative mutations in coding regions and in splice junctions. Rare variants affecting amino acid composition were found, but the genetic analysis failed to demonstrate association. Thus, mutations in POF1B are not involved in POF or their contribution cannot be demonstrated using the very heterogeneous group of patients available.

The DACH2 gene was subjected to X inactivation, as were the two genes previously found interrupted in POF patients, DIAPH2 and the XPNPEP2 (Bione et al., 1998Go; Rizzolio and Toniolo, unpublished results). These genes may be required in double dose throughout the life of the oocytes, when the presence of two active X chromosomes is a rule. Missense mutations were found in DACH2; they were in highly conserved residues and in most cases (P36L, R37L, F316S) they caused non-conservative amino acid substitutions. In Drosophila, dac is a transcriptional cofactor known to participate in a nuclear complex consisting of sine oculis (so), eyes absent (eya) and dachshund (dac) (Chen et al., 1997Go; Heanue et al., 1999Go). Highly conserved vertebrate homologues have been identified that are coexpressed in multiple organs including eye, inner ear, pituitary gland, muscle and kidney, and seem to act synergically during organogenesis (Davis et al., 2001Go; Li et al., 2002Go; 2003Go). The Drosophila eya gene was reported to have an essential role in fruit fly oogenesis for the correct differentiation of somatic follicle cells into polar cells (Bai and Montell, 2002Go). No evidence of the involvement of dac was reported, but the high level of functional conservation lead us to suggest that a complex involving DACH2 and one of the mammalian EYA and SIX genes may be important for development of mammalian gonads, and that alteration of the human DACH2 protein may act as a risk-factor for POF by altering the correct process of ovarian follicle differentiation.

The missense mutations found in patients in the DACH2 gene were more frequent in patients (2.7%) compared with the control group (0.7%). The difference was statistically sign0ificant (two-sided Fisher's exact test: P=0.0125). Replication in a similarly large independent population may definitively confirm our data. Since none of the mutations by itself was significantly enriched in the patients, our results suggest that a number of different private mutations that have arisen recently in DACH2 may be involved in POF. Reduced fitness due to low fertility may well explain this result, as well as the observed incomplete penetrance of some of the mutations found. Both factors could be responsible for the genetic heterogeneity of the disorder. In conclusion, while DACH2 mutations may be too rare to be useful in diagnostic for POF, they should be considered in the analysis with other risk factors such as FRAXA premutation, or the many that are expected but as yet unidentified.

The DACH2 gene is localized 700 kb distal from POF1B, and several X breakpoints in POF patients were mapped between the two genes. It is thus possible that the X chromosome breakpoints outside of the gene may exert a position effect on the expression of DACH2 gene. We did not test directly the CHM gene, but the data available tentatively exclude its involvement in POF (van den Hurk et al., 1997Go). Mutations in DACH2 were found in patients entering amenorrhea at very variable ages, ranging from 17 to 42 years. The phenotype of the patients carrying balanced translocations was more severe, as most of them presented with primary amenorrhea or with secondary amenorrhea at a very young age and after few and very irregular cycles. Based on this observation, we propose that POF associated with balanced translocation in the Xq critical region may be caused by a position effect of the chromosomal rearrangements on X-linked as well as autosomal genes. They may possibly also alter the expression of POF1B. Further characterization of the genomic region and of the genes involved and the analysis of a larger group of patients may definitively demonstrate this point, but the data presented here begin to shed new light into the very complex role of the genes in the critical region for ovarian failure.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
We thank M.Rocchi for the gift of the Y only hybrid DNA. We thank S.Giglio, P.Vineis and L.Davico for the collaboration in the collection of control DNAs. We thank Dr. L. Magrassi for the gift of brain ectopic samples. This work was supported by Telethon Italy, by the EU grant QLG1-CT-1999-00791, MIUR-FIRB RBNE0189HM_005, and MIUR-Genomica Funzionale CNR02.00157.ST97 and by the Fondazione Cariplo.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Aittomaki K, Lucena JL, Pakarinen P, Sistonen P, Tapanainen J, Gromoll J, Kaskikari R, Sankila EM, Lehvaslaiho H, Engel AR et al. (1995) Mutation in the follicle-stimulating hormone receptor gene causes hereditary hypergonadotropic ovarian failure. Cell 82, 959–968.[ISI][Medline]

Bai J and Montell D (2002) Eyes absent, a key repressor of polar cell fate during Drosophila oogenesis. Development 129, 5377–5388.[Abstract/Free Full Text]

Bione S, Sala C, Manzini C, Arrigo G, Zuffardi O, Banfi S, Borsani G, Jonveaux P, Philippe C, Zuccotti M et al. (1998) A human homologue of the Drosophila melanogaster diaphanous gene is disrupted in a patient with premature ovarian failure: evidence for conserved function in oogenesis and implications for human sterility. Am J Hum Genet 62, 533–541.[CrossRef][ISI][Medline]

Carrel L, Cottle AA, Goglin KC and Willard HF (1999) A first-generation X-inactivation profile of the human X chromosome. Proc Natl Acad Sci USA 96, 14440–14444.[Abstract/Free Full Text]

Chen R, Amoui M, Zhang Z and Mardon G (1997) Dachshund and eyes absent proteins form a complex and function synergistically to induce ectopic eye development in Drosophila. Cell 91, 893–903.[ISI][Medline]

Cremers FP, van de Pol DJ, van Kerkhoff LP, Wieringa B and Ropers HH (1990) Cloning of a gene that is rearranged in patients with choroideraemia. Nature 347, 674–677.[CrossRef][ISI][Medline]

Crisponi L, Deiana M, Loi A, Chiappe F, Uda M, Amati P, Bisceglia L, Zelante L, Nagaraja R, Porcu S et al. (2001) The putative forkhead transcription factor FOXL2 is mutated in blepharophimosis/ptosis/epicanthus inversus syndrome. Nat Genet 27, 159–166.[CrossRef][ISI][Medline]

Davis RJ, Shen W, Sandler YI, Heanue TA and Mardon G (2001) Characterization of mouse Dach2, a homologue of Drosophila dachshund. Mech Dev 102, 169–179.[CrossRef][ISI][Medline]

Di Pasquale E, Beck-Peccoz P and Persani L (2004) Hypergonadotropic ovarian failure associated with an inherited mutation of human bone morphogenetic protein-15 (BMP15) gene. Am J Hum Genet 75, 106–111.[CrossRef][ISI][Medline]

Fogli A, Rodriguez D, Eymard-Pierre E, Bouhour F, Labauge P, Meaney BF, Zeesman S, Kaneski CR, Schiffmann R and Boespflug-Tanguy O (2003) Ovarian failure related to eukaryotic initiation factor 2B mutations. Am J Hum Genet 72, 1544–1550.[CrossRef][ISI][Medline]

Heanue TA, Reshef R, Davis RJ, Mardon G, Oliver G, Tomarev S, Lassar AB and Tabin CJ (1999) Synergistic regulation of vertebrate muscle development by Dach2, Eya2, and Six1, homologs of genes required for Drosophila eye formation. Genes Dev 13, 3231–3243.[Abstract/Free Full Text]

Kim SS, Zhang RG, Braunstein SE, Joachimiak A, Cvekl A and Hegde RS (2002) Structure of the retinal determination protein Dachshund reveals a DNA binding motif. Structure (Camb) 10, 787–795.[Medline]

Lahn BT and Page DC (1999) Four evolutionary strata on the human X chromosome. Science 286, 964–967.[Abstract/Free Full Text]

Li X, Perissi V, Liu F, Rose DW and Rosenfeld MG (2002) Tissue-specific regulation of retinal and pituitary precursor cell proliferation. Science 297, 1180–1183.[Abstract/Free Full Text]

Li X, Oghi KA, Zhang J, Krones A, Bush KT, Glass CK, Nigam SK, Aggarwal AK, Maas R, Rose DW et al. (2003) Eya protein phosphatase activity regulates Six1-Dach-Eya transcriptional effects in mammalian organogenesis. Nature 426, 247–254.[CrossRef][ISI][Medline]

Lloyd SL, Sargent CA, Chalmers J, Lim E, Habeebu SS and Affara NA (1991) An X-linked zinc finger gene mapping to Xq21.1-q21.3 closely related to ZFX and ZFY: possible origins from a common ancestral gene. Nucleic Acids Res 19, 4835–4841.[Abstract]

Lorda-Sanchez IJ, Ibanez AJ, Sanz RJ, Trujillo MJ, Anabitarte ME, Querejeta ME, Rodriguez de Alba M, Gimenez A, Infantes F, Ramos C et al. (2000) Choroideremia, sensorineural deafness, and primary ovarian failure in a woman with a balanced X-4 translocation. Ophthalmic Genet 21, 185–189.[CrossRef][Medline]

Luborsky JL, Meyer P, Sowers MF, Gold EB and Santoro N (2003) Premature menopause in a multi-ethnic population study of the menopause transition. Hum Reprod 18, 199–206.[Abstract/Free Full Text]

Marozzi A, Porta C, Vegetti W, Crosignani PG, Tibiletti MG, Dalpra L and Ginelli E (2002) Mutation analysis of the inhibin alpha gene in a cohort of Italian women affected by ovarian failure. Hum Reprod 17, 1741–1745.[Abstract/Free Full Text]

Mumm S, Herrera L, Waeltz PW, Scardovi A, Nagaraja R, Esposito T, Schlessinger D, Rocchi M and Forabosco A (2001) X/autosomal translocations in the Xq critical region associated with premature ovarian failure fall within and outside genes. Genomics 76, 30–36.[CrossRef][ISI][Medline]

Murray A (2000) Premature ovarian failure and the FMR1 gene. Semin Reprod Med 18, 59–66.[CrossRef][ISI][Medline]

Prueitt RL, Ross JL and Zinn AR (2000) Physical mapping of nine Xq translocation breakpoints and identification of XPNPEP2 as a premature ovarian failure candidate gene. Cytogenet Cell Genet 89, 44–50.[ISI][Medline]

Prueitt RL, Chen H, Barnes RI and Zinn AR (2002) Most X;autosome translocations associated with premature ovarian failure do not interrupt X-linked genes. Cytogenet Genome Res 97, 32–38.[CrossRef][ISI][Medline]

Riva P, Magnani I, Fuhrmann Conti AM, Gelli D, Sala C, Toniolo D and Larizza L (1996) FISH characterization of the Xq21 breakpoint in a translocation carrier with premature ovarian failure. Clin Genet 50, 267–269.[ISI][Medline]

Rossi E, Faiella A, Zeviani M, Labeit S, Floridia G, Brunelli S, Cammarata M, Boncinelli E and Zuffardi O (1994) Order of six loci at 2q24-q31 and orientation of the HOXD locus. Genomics 24, 34–40.[CrossRef][ISI][Medline]

Sala C, Arrigo G, Torri G, Martinazzi F, Riva P, Larizza L, Philippe C, Jonveaux P, Sloan F, Labella T et al. (1997) Eleven X chromosome breakpoints associated with premature ovarian failure (POF) map to a 15-Mb YAC contig spanning Xq21. Genomics 40, 123–131.[CrossRef][ISI][Medline]

Shelling AN, Burton KA, Chand AL, van Ee CC, France JT, Farquhar CM, Milsom SR, Love DR, Gersak K, Aittomaki K et al. (2000) Inhibin: a candidate gene for premature ovarian failure. Hum Reprod 15, 2644–2649.[Abstract/Free Full Text]

Therman E, Laxova R and Susman B (1990) The critical region on the human Xq. Hum Genet 85, 455–461.[ISI][Medline]

Tibiletti MG, Testa G, Vegetti W, Alagna F, Taborelli M, Dalpra L, Bolis PF and Crosignani PG (1999) The idiopathic forms of premature menopause and early menopause show the same genetic pattern. Hum Reprod 14, 2731–2734.[Abstract/Free Full Text]

Tribioli C, Mancini M, Plassart E, Bione S, Rivella S, Sala C, Torri G and Toniolo D (1994) Isolation of new genes in distal Xq28: transcriptional map and identification of a human homologue of the ARD1 N-acetyl transferase of Saccharomyces cerevisiae. Hum Mol Genet 3, 1061–1067.[Abstract]

van Bokhoven H, van den Hurk JA, Bogerd L, Philippe C, Gilgenkrantz S, de Jong P, Ropers HH and Cremers FP (1994) Cloning and characterization of the human choroideremia gene. Hum Mol Genet 3, 1041–1046.[Abstract]

van den Hurk JA, Schwartz M, van Bokhoven H, van de Pol TJ, Bogerd L, Pinckers AJ, Bleeker-Wagemakers EM, Pawlowitzki IH, Ruther K, Ropers HH et al. (1997) Molecular basis of choroideremia (CHM): mutations involving the Rab escort protein-1 (REP-1) gene. Hum Mutat 9, 110–117.[CrossRef][ISI][Medline]

Zinn AR and Ross JL (1998) Turner syndrome and haploinsufficiency. Curr Opin Genet Dev 8, 322–327.[CrossRef][ISI][Medline]