Human Reproduction and Embryology

European Society of
Human Reproduction and
Embryology








COURSE 9


Genetics of female reproduction anomalies


Special Interest Group Reproductive Genetics



18 June 2006
Prague - Czech Republic

---

---

Contents

Program
page 2

Submitted contributions


X-chromosomal disorders impairing folliculogenesis –
J.L. Simpson (USA)
page 3


Molecular genetics of POF2 locus in Xq13.3-22
(DACH2; DIAPH2) - S. Bione (I)
page 9

Autosomal gene mutations impairing female reproduction (FIGa,
FSHR, MATER, NANOS, NOBOX, OBOX) - A. Rajkovic (USA)
page 13

FOXL2 expression in folliculogenesis - L. Crisponi (I)
page 18

Mullerian aplasia -genetic aspects - K. Aittomaki (FIN)
page 25

Oocyte derived growth factors and ovarian function (TGFß
superfamily: INHa ; BMP15, GDF9 ) - S. Galloway (NZ)
page 30

McKusick-Kaufmann syndrome - A. Slavotinek (USA)
page 36

Polycystic Ovarian Syndrome - J.L. Simpson (USA)
page 41



1

---

Course 9 - Pre-congress course organised by the
Special Interest Group Reproductive Genetics

“Genetics of female reproduction anomalies”

PROGRAM

Course co-ordinators: P.H. Vogt (D) and J. P.M. Geraedts (NL)

Course description: The aim of this course is to provide the attendants with recent
information in an important field of reproductive genetics. The study of genetic
causes of reproductive abnormalities is not only necessary to understand some aspects
of infertility. After the course the participants are also supposed to be able to
understand some basic mechanisms that are important for the normal development
and functioning of the female reproductive organs.Molecular developments have led
to important new findings in this area of reproductive genetics. All lecturers in this
course have recently published original work in this area and are leaders or members
of groups that are leading in this field.

09.00 - 09.30
X-chromosomal disorders impairing folliculogenesis –

J.L. Simpson (USA)
09.30 - 09.45
Discussion
09.45 - 10.15
Molecular genetics of POF1 locus in Xq26-28 (FMR-1, FMR-2) –

A. Smits (NL)

10.15 - 10.30
Discussion
10.30 - 11.00
Coffee break
11.00 - 11.30
Molecular genetics of POF2 locus in Xq13.3-22 (DACH2;
DIAPH2)
-
S. Bione (I)
11.30 - 11.45
Discussion
11.45 - 12.15
Autosomal gene mutations impairing female reproduction (FIGa,

FSHR, MATER, NANOS, NOBOX, OBOX) - A. Rajkovic (USA)
12.15 - 12.30
Discussion
12.30 - 13.00
FOXL2 expression in folliculogenesis - L. Crisponi (I)
13.00 - 13.15
Discussion
13.15 - 14.00
Lunch
14.00 - 14.30
Mullerian aplasia -genetic aspects - K. Aittomaki (FIN)
14.30 - 14.45
Discussion
14.45 - 15.15
Oocyte derived growth factors and ovarian function (TGFß

superfamily: INHa ; BMP15, GDF9 ) - S. Galloway (NZ)

15.15 - 15.30
Discussion
15.30 - 16.00
Coffee break
16.00 - 16.30
McKusick-Kaufmann syndrome - A. Slavotinek (USA)
16.30 - 16.45
Discussion
16.45 - 17.15
Polycystic Ovarian Syndrome - J.L. Simpson (USA)
17.15 - 17.30
Discussion and conclusions
17.30
Business meeting of the SIG Reproductive Genetics
2

---

X-Chromosome and Disorders of Folliculogenesis or Ovarian Failures

Joe Leigh Simpson, M.D.
Ernst W. Bertner Chairman and Professor
Baylor College of Medicine
Department of Obstetrics and Gynecology
1709 Dryden Rd., Suite 1100
Houston, Texas 77030

Learning Objectives:

1. Describe the scientific basis for assuming that ovarian failure in 45,X women
is caused by increase oocyte attention.
2. State which regions on the X chromosome that are relatively more pivotal for
ovarian differentiation.
3. List several candidate genes on the X that may be causative for ovarian
failure.

Failure of germ cell development or folliculogenesis leads to ovarian failure. Complete
and premature ovarian failure (POF) may be different manifestations of the same
underlying pathogenic and etiologic processes. Chromosomal abnormalities, mutations
of autosomal or X-linked genes, and polygenic/multifactorial factors all play a role. In
this contribution, we shall enumerate disorders of ovarian failure caused by perturbations
of the X chromosome. For citations alluded to, see reviews cited in the Suggested
Reading section.

I. Ovarian differentiation requires only one x (constitutive)
In the absence of the Y chromosome, the indifferent embryonic gonad always develops
into an ovary. Germ cells exist in 45,X human fetuses (Jirasek, 1976). Oocytes initially
exist even in 46,XY phenotypes females, such as infants with XY gonadal dysgenesis
(Cussen and MacMahon, 1979) or the genito-palato-cardiac syndrome (Greenberg et al.,
1987). Oocyte development in the presence of a Y chromosome is well documented in
other mammalian species. Thus, the pathogenesis of germ cell failure in humans can be
deduced to be increased germ cell attrition. If two intact X chromosomes are lacking,
ovarian follicles in 45,X individuals usually degenerate by birth. Genes on the second X
chromosome are thus necessary for ovarian maintenance, rather than ovarian
differentiation per se.

Presumably specific gene product(s) is responsible for primary ovarian differentiation,
but identifying them has proved elusive. An erstwhile candidate was AHC (Adrenal
Hypoplasia Congenital), the gene that encodes DAX-1. Duplication of Xp21 resultes in
46,XY embryos differentiating into females (Bardoni et al., 1994); thus, it was reasoned
that this region could play a primary role in ovarian differentiation in 46,XX individuals.
The region contained AHC, a locus that includes or is identical to DAX 1 (Dosage
Sensitive Sex Reversal/Adrenal hypoplasia critical region X); the mouse homolog is
Ahch. Ahch is upregulated in the XX mouse ovary, as predicted if Ahch (DAX1) were to


3

---

play a pivotal role in primary ovarian differentiation. Transgenic XY mice that
overexpress Ahch develop as females. However, XX mice lacking Ahch (knockout)
show normal ovarian differentiation, ovulation and fertility (Yu et al., 1998), and XY
mice mutant for Ahch show testicular germ cell defects. Thus, Ahch cannot be
responsible for primary ovarian differentiation in mice, nor presumably could DAX 1
(AHC) in humans.

II. Localizing ovarian maintenance genes to specific regions of the X
The X clearly is pivotal for normal ovarian developmental and folliculogenesis.
Determining the specific region is a first step in understanding normal ovarian
differentiation and in producing gene products of therapeutic benefit. Until the 1990s,
phenotypic-karyotypic correlations to deduce location of gonadal and somatic
determinants relied solely on metaphase analysis. Deductions made in this fashion are
limited. Prometaphase kayotypes allow 1,200 band analysis (traditional GTG-banding
400 to 500), but each band still contains considerable DNA. More refined analysis can
be achieved using polymorphic DNA markers that allow precise resolution far beyond
the capacity of light microscopy. This can be followed by sequencing and other
molecular approaches to identify pivotal genes.

Progress has been surprisingly slow, especially compared to that achieved in
delineating the regions of the Y necessary for testicular differentiation (SRY) or
spermatogenesis (DAZ). Several impediments help explain this relative lack of progress.
First, the incidence of X-deletions and translocations is low. Thus, the ideal approach of
analyzing cases ascertained by population-based methods is impractical. Selection bias
also exists. Most del(Xp) or del(Xq) individuals have been identified only because they
manifested clinical abnormalities, exceptions being rare familial cases or cases detected
in fetuses at the time of prenatal genetic diagnosis Less severely affected individuals
may escape detection.

Utilizing X-autosome translocations for analysis is popular but potentially hazardous
because of vicissitudes of X-inactivation. Autosomal regions themselves are not devoid
of significance for gonadal differentiation.

These pitfalls notwithstanding, Figure 1 shows clinical characteristics associated with
terminal deletions. Regions of relatively greater significance can be deduced.

Deletions of X Short Arm
Deletions of the short arm of the X chromosome show variable phenotype, depending
upon amount of Xp persisting. The most common breakpoint for terminal deletions is
Xp11.2611.4. Approximately half of 46,X,del(Xp)(p11) individuals show primary
amenorrhea and gonadal dysgenesis. The other cases menstruated and usually showed
breast development. In a 1989 tabulation by the author, 12 of 27 reported
del(X)(p11.2611.4) individuals, menstruated spontaneously; however, menstruation was
rarely normal (Simpson, 1987). More recent compilations have not altered these general
conclusions (Simpson, 1997b;Simpson, 1998;Simpson, 1997a). Deletions characterized
by progressively more distal breakpoints have also been reported: Xp21, 22.1, 22.3.


4

---

Menstruation occurs more frequently.

X long arm deletions originating at Xq13 are almost always associated with primary
amenorrhea, lack of breast development, and complete ovarian failure (Simpson, 1997a).
Xq13 thus seems to be an important region for ovarian maintenance. This region has
been said to contain the POF2 locus, even while poorly defined. The key loci could lie in
proximal Xq21, but not more distal given that del(X)(q21) to (q24) individuals
menstruate far more often than del(X)(q13). Del(X)(q21) women who menstruate could
have retained a region that contained an ovarian maintenance gene, whereas del(X)(q13
or 21) women with primary amenorrhea might have lost such a locus.

In more distal Xq deletions, the more common phenotype is premature ovarian failure
(Krauss et al., 1987;Fitch et al., 1982;Simpson, 1997b;Simpson, 1998). While distal Xq
seems less important for ovarian maintenance than proximal Xq, the former must still
have regions important for ovarian maintenance.

Although demarcation into discrete regions is not possible, it is
heuristically useful to stratify terminal deletions into those occurring in these regions: Xq
136 21, Xq22-25, Xq 26-28?

Familial Xq terminal or interstitial deletions have been characterized by various break
points between Xq25 to Xq28. Break points near or in Xq27 seem most common. The
locus for fragile X (FRAXA) also lies in this region, and this will be discussed separately
(Section IV).

III. Candidate genes on the X


Although chromosomal regions on Xp (and Xq) are presumed to contain genes pivotal to
ovarian germ cell function and folliculogenesis, identifying the actual gene and gene
products remain frustratingly unclear. A host of candidate genes are being studied.

X Short Arm
USP9X (ubiquitin specific protease 9): This gene maps to Xp11.4 (Jones et al., 1996).
The Drosophila orthologue of USP9X is required for eye development and oogenesis.
The role USP9X plays in human gonadal development is still unclear. An attractive
feature is its location in a region known to have ovarian determinants.

BMP15: Bone Morphogenetic Protein 15: Bone morphogenetic protein 15 (BMP) is a
member of the transforming growth factor-beta (TGFβ) superfamily. These genes direct
many developmental pathways though binding and activating transmembrane
serine/threonine kinase receptors. BMP is involved in folliculogenesis and embryonic
development, being expressed in gonads. The BMP 15 gene is located on Xp11.2 and
has two exons. Animal studies initially suggested that perturbations of BMP 15 could be
important in ovarian development. Heterozygous Inverdale sheep carrying a mutation in
the BMP 15 gene show an increased ovulation rate, with twin and triplet births. Primary
ovarian failure occurs in homozygotes (Galloway et al., 2000). Bmp15 knockout mutant
female mice are subfertile, showing decreased ovulation rates, reduced litter size and


5

---

decreased number of litters per lifetime (Dube et al., 1998).

In the humans, (Di Pasquale et al., 2004) reported a heterozygous Y235C missense
mutation in the second exon of the BMP15 gene in each of two sisters having ovarian
failure. The proband had streak gonads and elevated FSH (80miU/L); the younger sib
had only one episode of vaginal spotting and at age 18 years had an FSH of 67 mlU/L.
The mother was homozygously normal (Y235) at this allele; the C235 allele was
inherited from the father. The authors presented in vitro evidence for a dominant
negative mechanism.

Consistent with heterozygous dominant negative effect is that an autosomal TGF family
member (GDF-9) has also been implemented in ovarian failure and in these cases the
mutation was also heterozygous (Dixit et al., 2004). Given that proteins of TGF family
members (BMP-15, GDF-9) may form heterodimers, a single mutation could plausibly
generate a dysfunctional gene product.




Region Localized by Quantitative Linkage Analysis: Genome wide linkage analysis has
been applied to localize regions influencing age of menopause. One such study has
shown association between region Xp11 and age of menopause.

ZFX (Zinc finger X): Mapped to Xp22.1-21.3 Zfx, is a candidate gene for short stature
and ovarian failure (Zinn and Ross, 2001). It has attracted attention on the basis of
homology to Zfy, once the prime candidate gene for male sex determination. Mice null
for Zfx are small, less viable, less fertile and characterized by diminished germ cell
number in ovaries and testes (Luoh et al., 1997). Their external and internal genitalia are
otherwise normal.

X Long Arm
XIST: Xq13 contains the X-inactivation center and XIST. Loss of germ cells may or may
not be the result of perturbation of XIST, despite years of speculation that disturbances of
X-inactivation per se lead to ovarian failure. The concept of a variably-defined “critical
region” inviolate for ovarian development receives less attention than in the past.
However, Xq13 is clearly a region rich in pivotal genes.
A Drosphila homologue (dac) exists and is expressed in multiple tissues.

DACH2: Localized on Xq21, this gene has been implicated in POF on the basis of a
breakpoint in an X-autosome translocation (Prueitt et al., 2000). DACH2 consists of 13
exons, and does not undergo X-inactivation. Bione et al. (2004) studied 257 Italian
patients, of whom 19 had primary amenorrhea and 238 secondary amenorrhea either
before age 40 (N=212 having traditionally defined POF) or between aged 40 and 45 years
(N=26, having “early menopause”). A total of 5 missense mutations (2.7%) were found
in evolutionarily conserved amnio acids. These same alleles were also found in controls,
but less frequently (0.7%).

DIAPH2: Localized to Xq22, human DIAPH2 is the homologue of Drosophila
melanogaster diaphanous (dia). Drosophila dia is a member of a gene family whose gene


6

---

products help establish cell polarity, govern cytokinesis, and reorganize the actin
cytoskeleton. In both males and females perturbations of dia cause sterility in flies
(Castrillon and Wasserman, 1994). In a human Xq21/autosome translocation, disruption
of the last intron of DIAPH2 was found. (Bione et al., 1998;Bione and Toniolo,
2000;Prueitt et al., 2000)

FMR1: On Xq27 lies the FMR1 locus, perturbation of which causes the fragile X
syndrome. About 20% of females with a FMR1 premutation (≥ 60 CGG repeats) develop
premature ovarian failure, although paradoxically those with the full mutation do not.
Fragile X syndrome is discussed below. This fragile X locus (FMR1) cannot precisely
correspond to that which when deleted causes ovarian failure in del(Xq) (2.7 or 2.8).

IV. Fragile X syndrome (FMR1)
A relationship exists between fragile X syndrome, caused by perturbation FMR1, and
ovarian failure. Fragile X syndrome is a common form of X-linked mental retardation
that is caused by expansion of the FMRI gene, located on Xq27. If more than 200 CGG
repeats exist, transcriptional silencing of a RNA-binding protein occurs. In normal
males, the normal number of CGG repeats is less than 55. Males or heterozygous
females having 55 to 199 repeats are said to have a premutation (Sullivan et al., 2005).
During female (but not male) meiosis, the number of triplet repeats may increase
(expand). A phenotypically normal woman with a FRAXA premutation may thus have
an affected son if the number of CGG repeats on the oocyte of the X she transmits to her
male offspring expands during meiosis to greater than 200. Affected males show mental
retardation, characteristic facial features, and large testes. Expansion does not occur if
there are fewer than 55 CGG repeats. Females may also show mental retardation, but
phenotype is less severe than in males.

Approximately 20 to 25% of females with the FRAXA premutation show premature
ovarian failure (POF), defined as menopause prior to age 40 years. Schwartz et al.
(1994) found oligomenorrhea in 38% of premutation carriers versus 6% of controls.
Analyzing 1,268 controls, 50 familial POF cases and 244 sporadic POF cases,
Allingham-Hawkins et al. (1999), reported that 63 of 395 premutation carriers (16%)
underwent menopause prior to 40 years of age, versus 0.4% in controls. Sullivan et al.
(2005) found 12.9% of premutation carriers have POF versus 1.3% of controls.
Surprisingly, FSH was increased in premutation carriers aged 30 to 39 years, but not in
carriers of other ages. The number of repeats correlates with risk of POF, but only within
a specified range. The risk is only slightly increased risk existed at 40 to 79 repeats, but
much higher at 80 to