Nature Genetics
27, 132 - 134 (2001)
doi:10.1038/84735
A fork in the road to fertilityRobyn L Prueitt
& Andrew R ZinnMcDermott Center for Human Growth and Development, The University of Texas Southwestern Medical School, 5323 Harry Hines Boulevard, Dallas, Texas 75390, USA.
andrew.zinn@utsouthwestern.edu Haploinsufficiency of FOXL2, a new forkhead transcription factor, causes blepharophimosis/ptosis/epicanthus inversus syndrome (BPES), a rare developmental disorder affecting the eyelid and sometimes the ovary. A new study implicates FOXL2 as the first human gene required for the maintenance of ovarian follicles. The discovery of FOXL2 may provide insight into the causes of idiopathic premature ovarian failure, a disease that burdens many infertile couples.Premature ovarian failure (POF), defined as the cessation of menses accompanied by elevated gonadotropins before age 40, affects an estimated 1% of women. POF can result from decreased initial follicle number, increased follicular loss, or resistance to gonadotropins, and the primary defect can lie in either the oocyte or the surrounding follicle cells. Most cases of POF are idiopathic and presumed to be genetic. The most common identifiable etiology involves an X-chromosome abnormality. Other causes of POF include autoimmunity, irradiation and chemotherapy. Galactosemia and mutations in the follicle-stimulating hormone (FSH) receptor gene (FSHR) are rare recessive causes. So far, the only autosomal dominant disorder associated with POF is the blepharophimosis/ptosis/epicanthus inversus syndrome (BPES)—and, on page 159 of this issue, Laura Crisponi and colleagues1 report that mutations in a new forkhead gene are the culprit.
Insight into BPES
The disorder BPES, which is characterized by a distinctive eyelid abnormality, occurs in two forms. In type I (but not type II), the eyelid abnormality is associated with POF, which may present as either primary or secondary amenorrhea. Although the pathogenesis of POF in BPES has not been studied in detail, gonadotropin resistance followed by progressive follicular depletion has been described2. (Males with BPES are fertile.)
Cytogenetic studies1,
3,
4,
5 indicated that the causative gene maps to 3q23, an assignment confirmed for both types of BPES by genetic linkage studies. A gene that encodes a cellular retinol-binding protein, RBP1, was known to map to this region and therefore was an attractive candidate because mice lacking the retinoic acid receptor have abnormal eyelids. However, no RBP1 mutations were identified in BPES patients3.
The positional cloning race began in earnest after three balanced translocation breakpoints associated with BPES were mapped to a small interval between markers WI-15160 and D3S3586, contained on a single YAC clone (Fig. 1). The transcription map of this region proved to be complex. De Baere et al.3 first reported the fine-mapping of a BPES-associated t(3;4) breakpoint, but had the misfortune of identifying two irrelevant genes: one encoding a second retinol-binding protein and one encoding a  -coatamer protein. They next identified a novel gene, BPESC1 (BPES candidate 1), interrupted by the translocation4. However, BPESC1 is expressed in only adult testis, and no disease-causing mutations were found in other BPES patients. Praphanphoj et al.5 also mapped a t(3;21) breakpoint to the same PAC clone as the t(3;4) breakpoint, but did not report any new candidate genes.
 | | |
In the words of Yogi Berra, "It ain't over 'til it's over." In the current study, Crisponi et al.1 mapped a t(3;7) breakpoint and identified two more genes in the WI-15160−D3S3586 interval. The translocation interrupts C3orf5, a large ubiquitously expressed gene of unknown function, but no C3orf5 mutations were present in other BPES subjects. The BPES hunt ended at last when the second gene proved to be the `FOX'.
Forkhead family grows
FOXL2 is a member of the winged helix/forkhead family of transcription factors. These proteins contain a characteristic DNA-binding domain of 100 amino acids, originally identified in Drosophila melanogaster forkhead and the rat hepatocyte nuclear factor 3. Forkhead proteins are found in all eukaryotes and serve important functions in the establishment of the body axis and the development of tissues from all three germ layers in animals.
Now, more than 20 human forkhead genes are known, and several have been implicated in tumorigenesis. In addition to BPES, four hereditary disorders have been shown recently to be caused by mutations in forkhead genes (Table 1). The phenotypes are pleiotropic and include circulatory, immune, skeletal and craniofacial defects. Notably, two of the disorders include eye abnormalities6,
7. Given the conserved developmental roles of Drosophila homologs, it's a safe bet that mutations in other human forkhead genes will cause genetic disorders.
| | |
Mouse Foxl2 is expressed almost exclusively in the ovary follicle cells and the developing eyelid1, consistent with the BPES phenotype, although pituitary expression hints at pleiotropic growth effects seen in some patients. The FSHR also acts in follicle cells; FSHR mutations cause arrest of primary follicle development8. Mice lacking Foxl2 should provide a means to better understand the pathogenesis of ovarian failure in type I BPES and identify relevant FOXL2 transcriptional targets. One intriguing possibility is that FOXL2 regulates transforming growth factor -related signaling pathways involved in both eyelid development and gonadal function, for example, activin/inhibin9.
The data of Crisponi et al.1 rule out the possibility that BPES type I is a contiguous gene syndrome involving separate genes that affect the developing eyelids and ovaries. Four type I BPES families segregate mutations that result in protein truncation 3' to the forkhead domain, whereas two apparently unrelated type II BPES families have an expansion of a polyalanine repeat 3' to the forkhead domain. This same expansion was found in a male with sporadic BPES and probably represents a recurrent mutation.
The authors suggest that the type I BPES truncations are null mutations that cause FOXL2 haploinsufficiency, whereas the type II BPES polyalanine expansion is a hypomorphic allele. As the truncations in type I BPES are downstream of the forkhead domain, it is possible that they act in a dominant-negative fashion by producing proteins that bind to DNA but fail to activate transcription. The presence of ovarian failure in a young girl with an interstitial deletion of 3q21−23 (ref. 10) supports the haploinsufficiency model for type I BPES. Although the reported mutations cause BPES, other FOXL2 changes might result in ovarian failure without eyelid abnormalities—so it will be important to screen for FOXL2 mutations in women with idiopathic POF.
Ironically, of the three translocation breakpoints mapped to the BPES region, the one studied by Crisponi et al.1 is furthest (about 180 kb) from FOXL2. All three breakpoints appear to be 5' to the gene; BPES in these cases is presumably due to the disruption of a distant enhancer or to some other position effect. Such effects, which often confound positional cloning, may be especially prevalent in human genetic diseases involving haploinsufficiency of developmental transcription factors7,
11. The moral of this story is (with apologies to Mr. Berra): "If you come to a forkhead in the genome, study it!"
REFERENCES
- Crisponi, L. et al. Nature Genet. 27, 159−166 (2001). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Fraser, I.S., Shearman, R.P., Smith, A. & Russell, P. Fertil. Steril. 50, 747−51 (1988). | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- De Baere, E. et al. Genomics 57, 70−78 (1999). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- De Baere, E. et al. Genomics 68, 296−304 (2000). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Praphanphoj, V. et al. Genomics 65, 67−69 (2000). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Nishimura, D.Y. et al. Nature Genet. 19, 140−147 (1998). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Fang, J. et al. Am. J. Hum. Genet. 67, 1382−1388 (2000). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Aittomaki, K. et al. Cell 82, 959−968 (1995). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Vassalli, A., Matzuk, M.M., Gardner, H.A., Lee, K.F. & Jaenisch, R. Genes Dev. 8, 414−427 (1994). | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Ogata, T. et al. Horm. Res. 50, 190−192 (1998). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Kleinjan, D.J. & van Heyningen, V. Hum. Mol. Genet. 7, 1611−1618 (1998). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Clifton-Bligh, R.J. et al. Nature Genet. 19, 399−401 (1998). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Mears, A.J. et al. Am. J. Hum. Genet. 63, 1316−1328 (1998). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Chatila, T.A. et al. J. Clin. Invest. 106, R75−R81 (2000). | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Bennett, C.L. et al. Nature Genet. 27, 20−21 (2001). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Wildin, R.S. et al. Nature Genet. 27, 18−20 (2001). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
|
'-COP encodes a 
