| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Professor, Chief Section of Reproductive Endocrinology, Infertility, & Genetics Department of Obstetrics & Gynecology Developmental Neurobiology Program The Institute of Molecular Medicine and Genetics The Neuroscience Program The Medical College of Georgia Augusta, Georgia 30912-3360
Address all correspondence and requests for reprints to: Lawrence C. Layman, Professor, Chief, Section of Reproductive Endocrinology, Infertility, & Genetics, Department of Obstetrics & Gynecology, Developmental Neurobiology Program, The Institute of Molecular Medicine and Genetics, The Neuroscience Program, The Medical College of Georgia, 1120 15th Street, Augusta, Georgia 30912-3360. E-mail: Llayman{at}mail.mcg.edu
Somewhat surprisingly, the identification of ovarian-determinant genes, particularly on the X-chromosome, has only minimally progressed until recently, despite the completion of the human genome mapping project. At least five autosomal loci have been confirmed to play a role in ovarian function but, by and large, mutations in these genes comprise only a small fraction of premature ovarian failure (POF) patients (Table 1
). These include the FSH receptor gene, mutations of which cause autosomal recessive hypergonadotropic hypogonadism in women who predominantly present with primary amenorrhea, with or without breast development (1). Variable expressivity is also exemplified by the findings of only primary or primordial follicles in some affected women, whereas others display follicle development to the corpus luteum stage. Several 46,XX females have been described with homozygous inactivating mutations of the LH receptor (homozygous mutations in genetic males cause a range of undermasculinization with an absent Mullerian system, and heterozygous activating mutations cause familial male precocious puberty). These genetic females usually present with normal breast development and primary amenorrhea or oligomenorrhea (2). Their endocrine profile is somewhat different than the typical patient with POF who has elevations of both gonadotropins, with FSH higher than LH. Females with LH receptor mutations may have a normal to elevated serum LH level and normal levels of FSH, or at least an elevated LH to FSH ratio.
|
Galactosemia has long been reported to be associated with POF, occurring in about two thirds of women rather than in men. Although galactosemia may be caused by mutations in any of three genes involved in galactose metabolism, galactose-1-phosphate uridyl transferase is the one most associated with ovarian failure (5). Unfortunately, treatment with a low galactose-containing diet may not protect the affected women from the development of ovarian failure. A polymorphism in the 
-subunit of inhibin has been reported, but there is not convincing evidence yet that this substitution actually affects protein function (6, 7). Ovarian failure may also be present as part of a syndrome, and three genes have been described to have mutations in women with ovarioleukodystrophy, an autosomal recessive disorder with substantial neurologic deficits and ovarian failure (8).
It has long been recognized that privation of one X-chromosome in patients with gonadal dysgenesis (Turner syndrome) comprises a significant proportion (up to two thirds) of women presenting with primary amenorrhea due to ovarian failure (9). This 45,X cell line with or without mosaicism (which should be assessed in a karyotype counting 
50 cells) is found much less frequently in women with secondary amenorrhea and ovarian failure (10). However, most of the patients encountered by clinicians with secondary amenorrhea do not have a 45,X cell line, but rather are 46,XX.
Several regions on the X-chromosome were thought to contain ovarian determinant genes, as large deletions or translocations were identified. With the report of a family with multiple women with POF in their 20s and 30s who had an interstitial deletion of Xq26q28 (11), it was certain that a gene or genes necessary for ovarian function had to be present in this "POF1" region. The POF1 region was further narrowed down to Xq27.2/Xq27.3-qter when a family with POF had a small deletion within this region (12). However, when two putative POF genes within the POF1 region, POF1B and DACH2, were studied for mutations, some rare missense mutations were identified, but they were found in controls or the frequency was not different from controls (13). In addition, no in vitro analysis was performed, so this awaits confirmation.
Additionally, when a translocation disrupting the diaphanous (DIAPH2) gene on Xq22 was described, it was thought that this gene certainly had to be a full-fledged ovarian determinant gene. It had homology to its ortholog in Drosophila and encodes a protein expressed in the gonads (14). However, there are no point mutations in the DIAPH2 gene from the POF2 region of X, questioning its importance in normal ovarian function. So the presumptive critical regions of the X-chromosome did not yield a single ovarian determinant gene!
One region on the X-chromosome that has assumed importance to POF is fragile X syndrome, an X-linked dominant disease with reduced penetrance for which the gene has been known for some time. Fragile X syndrome is caused by expansions in the triplet repeat sequence CGG in the 5' untranslated region of the FMR1 gene localized to Xq27.3 in the POF1 region. Normal individuals carry approximately five to 50 copies of the triplet repeat, whereas female carriers demonstrate an expansion from 55–200 repeats, generally referred to as the premutation allele. During female meiosis, further expansion may occur, producing repeats greater than 200, which causes the full phenotype of fragile X syndrome in males, namely developmental delay, abnormal facies, and macroorchidism, and perhaps fragile X-associated tremor/ataxia syndrome. It is now generally accepted that women with the premutation allele have an increased risk of POF—approximately 3–4% without a family history of ovarian failure but 10–15% if there is (15). Similarly, in known fragile X premutation allele carriers, about 15% will have POF. FMR1 is expressed in germ cells, and it is only the premutation allele that seems to be associated with POF (not the full mutation).
The authors of the report in this issue (16) describe the second ovarian determinant gene on the other arm (Xp11.2) of the X-chromosome, extending their prior observation that heterozygous mutations in the bone morphogenic protein 15 (BMP15) gene are associated with POF. BMP15 and its homologous (50% identify) growth-differentiation factor 9 (GDF9) are oocyte-specific factors and members of the TGF-ß family. GDF9 appears essential for function of the oocyte in the early follicular and periovulatory phases. Homozygous GDF9 female knockout mice are infertile due to a block in the primary follicle stage, the inability to form a theca, and impaired meiotic competence (17). No definitive human mutations in GDF9 have been identified, although some rare variants have been associated with POF (18). Homozygous BMP15 knockout mice are subfertile with a phenotype less severe than GDF9 knockout mice. They have ovarian morphology similar to wild-type mice but have decreased ovulation and fertilization rates.
In their previous study, Di Pasquale et al. (19) demonstrated a heterozygous missense BMP15 mutation (Y235C) in two sisters with POF, one with primary amenorrhea and the other with a single episode of vaginal bleeding. This mutation was inherited from the hemizygous father, who apparently had no phenotypic abnormalities. In this report, breast development was not commented upon for either sister, but it was stated that at the time of laparoscopic appendectomy, ovaries were "streaks" in the older sister, suggesting that she never initiated thelarche. Because the younger sister had one menstrual period, it is more likely that she had some breast development, but she had also taken estrogen and progesterone so these important details are not clear. If these observations are correct, this BMP15 mutation demonstrates variable expressivity ranging from a complete absence of sexual development to full breast development and even menarche, although short-lived. In future studies, these important clinical observations must be gathered to provide as accurate genotype-phenotype correlations as possible.
In their prior study, the authors demonstrated in vitro evidence supporting the pathogenesis of the Y235C allele as a dominant negative BMP15 mutation (19). The mutant, by the creation of a cysteine, demonstrated altered processing of the protein, as well as impaired granulosa cell proliferation and DNA synthesis compared with wild type. The incorporation of tritiated thymidine was increased by the administration of wild type, but not mutant or wild type plus mutant together. These findings, taken together with the phenotype of the affected women with POF, strongly support a role for BMP15 in normal functioning of the human ovary.
In the present study, Di Pasquale et al. (16) extend their observations to 166 unrelated Caucasian (79 American/87 European) patients with 46,XX, idiopathic POF, 25 with the more severe phenotype of primary amenorrhea, and 141 with secondary amenorrhea. Of the 25 women with primary amenorrhea, 10 were familial, whereas 67 of the 141 women with secondary amenorrhea had familial POF. The pedigree structure of these women was not stated, but it would be interesting to know, because some pedigrees could be consistent with autosomal dominant or autosomal recessive, and not X-linked inheritance (and so would not be expected to harbor X-chromosome BMP15 mutations). It would also be very interesting to know what percent of each group manifested the premutation allele for the FMR1 gene, but this is not described. The authors used a screening technique of denaturing HPLC, which generally detects more than 90% of available mutations (20). Denaturing HPLC was used to screen the two-exon gene in five overlapping PCR fragments, and variant fragments were subjected to DNA sequencing. Appropriately, 95 women with natural menopause (control A) were used as controls, as well as 86 women and 30 men from the general population (control B).
The authors found a 3-bp insertion (262insLeu) that is likely to be a polymorphism, as it was identified in two of 166 POF women and in five of 95 menopausal women. However, they also identified two different missense mutations, the R68W in one of 166 with POF and A180T in five of 166 POF women. These two missense mutations were only seen in women with secondary amenorrhea and not in controls. The previously identified Y235C was not seen in this study, so it comprised one of 166 patients. The missense mutations occurred in the prohormone region of the gene, like the Y235C mutation, and could be similarly involved in processing of BMP15; however, no in vitro analysis of the mutants was performed. They do state that the three BMP15 missense mutants were significantly (P = 0.004) more prevalent in POF women (7 of 332 POF alleles) vs. controls (0 of 392 alleles).
Where do the findings of this study leave us with regard to BMP15 as an important ovarian determinant gene in humans? As is true in many genetic disorders affecting reproduction, they are likely to be rare as identified causes in POF women. Nevertheless, the identification of uncommon human mutants provides support for their function in normal ovarian physiology, which may have profound implications for understanding ovarian biology and potentially for treatment. Determining the functional significance of missense mutations in human disease remains a very problematic dilemma. Segregation within the family (that is, affected members have the mutation and unaffected members do not), location of the mutation within a residue conserved by multiple species, and functional studies constitute supportive evidence for function (21). The authors have met these criteria for the previously described Y235C mutation, but do not provide convincing evidence yet for the other two missense mutations. They find association of these alleles with POF, but this must be viewed with caution. Just because the allele was not found in controls to date does not exclude that it is a rare polymorphism. The patients represent potentially heterogeneous groups (European and American patients), and it is not clear what is the ethnicity of the controls. These variants could simply represent a stratification artifact, meaning they are a variant common in a certain ethnic group, rather than being causative for POF. We anxiously await the in vitro analysis of these mutants from the authors in future publications and remain cautiously optimistic that BMP15 mutations cause POF in humans.
Acknowledgments
I appreciate the careful review of this manuscript by Paul G. McDonough and Sandra S. P. T. Tho, Department of Obstetrics/Gynecology, Medical College of Georgia.
Footnotes
This work was supported by National Institutes of Health Grants HD33004 and HD040287.
Abbreviations: BMP15, Bone morphogenic protein 15; GDF9, growth-differentiation factor 9; POF, premature ovarian failure.
Received March 10, 2006.
Accepted March 14, 2006.
References
This article has been cited by other articles:
 |
|
 |
|
  C.B. Tempfer, M. Simoni, B. Destenaves, and B.C.J.M. Fauser Functional genetic polymorphisms and female reproductive disorders: Part II--endometriosis Hum. Reprod. Update, January 1, 2009; 15(1): 97 - 118. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B.C.J.M. Fauser, K. Diedrich, P. Devroey, and on behalf of the Evian Annual Reproduction (EVAR) Predictors of ovarian response: progress towards individualized treatment in ovulation induction and ovarian stimulation Hum. Reprod. Update, January 1, 2008; 14(1): 1 - 14. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
