DNA sequence variations of the entire anti-Müllerian hormone (AMH) gene promoter and AMH protein expression in patients with the Mayer–Rokitanski–Küster–Hauser syndrome

P. Oppelt1,*, P.L. Strissel1,*, A. Kellermann1, S. Seeber2, A. Humeny2, M.W. Beckmann1 and R. Strick1,3

1 Department of Gynaecology and Obstetrics and 2 Department of Biochemistry at the University of Erlangen-Nuremberg, Erlangen, Germany

3 To whom correspondence should be addressed at: University of Erlangen-Nuremberg, Department of Gynaecology and Obstetrics, Universitätsstrasse 21–23, D-91054 Erlangen, Germany. Email: reiner.strick{at}gyn.imed.uni-erlangen.de


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: The etiology of the Mayer–Rokitanski–Küster–Hauser (MRKH) syndrome, where congenitally the Müllerian ducts fail to develop into the uterus, cervix and upper vagina, along with other malformations, is unresolved. Anti-Müllerian hormone (AMH) signal transduction inducing the degradation of Müllerian ducts in males is implicated in the MRKH syndrome. This study examined if DNA sequence variations are responsible for the activation of AMH and aberrant hormone levels in MRKH patients. METHODS: The entire AMH promoter and 3' regulatory elements of the constitutively expressed splicing factor SF3a2 were sequenced in 30 MRKH patients and genotyped in 48 control individuals using matrix-assisted laser desorption/ionization-time-of-flight mass spectronomy. Ovarian AMH promoter function was correlated with protein expression in plasma and peritoneal fluid of MRKH patients. RESULTS: Of six identified AMH promoter variations, two at positions –639 (SP1-binding site) and –210 [steroidogenic factor (SF)1-binding site] were homozygote in 73% of patients, and 69% of control individuals, destroying the SP1-binding site. AMH protein levels in plasma and peritoneal fluid from patients were equivalent to control individuals, however in three patients plasma levels were abnormally high. CONCLUSIONS: AMH is an important indicator for ovarian function. AMH promoter sequence variations or the previously proposed SF3a2-AMH fusion co-transcripts cannot be responsible for aberrant AMH expression leading to Müllerian duct degradation.

Key words: anti-Müllerian hormone/DNA polymorphism/ionization-time-of-flight/matrix-assisted laser desorption/Mayer–Rokitanski–Küster–Hauser syndrome/promoter elements


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The incidence of congenital uterine anomalies in the general female population is ~5 per 1000, and of infertile women more frequent at 35–63 per 1000 (Raga et al., 1997Go; Nahum, 1998Go). Another, but more rare, anomaly represents congenital uterine and vaginal aplasia, which is observed in patients with the Mayer–Rokitanski–Küster–Hauser (MRKH) syndrome.

In general, most MRKH cases are isolated among families and karyotypic analyses have demonstrated normal 46,XX profiles (Simpson, 1999Go). The MRKH syndrome has an incidence of 1 per 4000–5000 female live births, where during fetal development the Müllerian ducts fail to develop into the uterus, cervix and upper vagina (Hauser and Schreiner, 1961Go; Basile and De Michele, 2001Go). Most of the MRKH patients have normal morphological ovaries and fallopian tubes and due to steroid hormone production develop normal secondary sex characteristics. Primary amenorrhea is the most common symptom leading to diagnosis. The presence of only a shallow vaginal pouch can be corrected with reconstructive surgery (Vecchietti, 1979Go). MRKH is classified clinically into the following subtypes: (i) typical patients have normal developed fallopian tubes, ovaries and a renal system, (ii) atypical patients have renal, ovarian or other reproductive organ abnormalities, and (iii) the most rare and severe form is known as MURCS (Müllerian–renal–cervicothoracic somite abnormalities) where abnormalities of the renal system or cervical-thoracic (partial or complete absence of the spinal column, ribs, or arms; asymmetric ribs or arms; improperly developed spinal column, ribs, or arms) or somite dysplasia occurs (Duncan et al., 1979Go).

The AMH (anti-Müllerian hormone, also called Müllerian-inhibiting-substance = MIS or -factor = MIF) gene, on chromosome 19p13.3, codes for a dimeric glycoprotein of 560 amino acids with a 24–25 amino acid leader, and a C-terminus sharing homology to the transforming growth factor (TGF)-{beta} family. In human testis, AMH protein is secreted from Sertoli cells from the start of sex differentiation and remains at high levels until puberty (Jamin et al., 2002Go; MacLaughlin and Donahoe, 2002Go; Visser, 2003Go). Specifically, AMH in the 6th gestational week initiates Müllerian duct regression until the 9th week by inducing ductal epithelial regression through a paracrine mechanism, which then initiates apoptosis (Taguchi et al., 1984Go; Visser, 2003Go). In contrast, in XX females, where no AMH is expressed during fetal development, the paramesonephric or Müllerian duct grows into the oviducts, uterus, cervix, and the upper 75% of the vagina. AMH is produced at low levels in ovarian granulosa cells during the perinatal period through menopause. Female adult AMH serum levels are normally below <75 pmol/l and decrease over time, while AMH male serum levels range from 270–640 pmol/l during fetal to pre-pubertal periods (Rey et al., 1999Go). In addition, abnormally low AMH serum levels can be a marker for boys with cryptorchidism and children with intersex conditions (Lee et al., 1997Go), whereas high AMH serum levels can be a marker of sertoli- and granulosa-cell tumors (Rey et al., 2000Go). However, the AMH blood protein levels in MRKH or in other patients with uterine malformations have not been previously determined.

The AMH gene is located downstream of the splicing factor 3a subunit 2 (SF3a2) gene, also called SAP62, a house keeping gene and a member of the U2snRNP complex (Bennett and Reed, 1993Go). The 3' untranslated region of SF3a2 is located only 739 bp 5' from the transcriptional start site of AMH (Dresser et al., 1995Go). The occurrence of read through transcription of SF3a2 into AMH, has been hypothesized due to DNA sequence variations resulting in a failure of SF3a2 polyadenylation (Dresser et al., 1995Go). If this were to occur, following RNA and protein processing AMH would be constitutively expressed along with SF3a2 and result in higher than normal AMH blood levels.

The spatial and temporal transcription of AMH depends upon a series of essential transcription factors binding to the promoter, especially the steroidogenic factor (SF)-1 at position –92 and –218, SOX9 at –141 and GATA4 at –75 and –168 (Morikawa et al., 2000Go; Rey et al., 2003Go l; Figure 1). SF1 and GATA4 DNA binding sites are essential for AMH expression in vitro (Arango et al., 1999Go). Transgenic mice studies involving AMH promoter mutations, demonstrated that the SF1 binding site is a quantitative regulator of AMH, but is not essential for Müllerian duct regression (Arango et al., 1999Go). These same mice studies showed that the SOX9 binding site at AMH position –141 was essential for Müllerian duct regression. Interestingly, human XY male patients with SOX9 gene mutations develop campomelic dysplasia where most exhibit sex reversal (Foster et al., 1994Go).



View larger version (13K):
[in this window]
[in a new window]
 
Figure 1. Schematic drawing including exon 8 of the SF3a2 gene, the anti-Müllerian hormone (AMH) promoter and first exon of the AMH gene. The 1066 bp PCR amplified product represents the DNA sequenced and analysed from 30 Mayer–Rokitanski–Küster–Hauser (MRKH) syndrome patients. Top forward (TF) and bottom reverse (BR) represent the location of the DNA primers in exons. In addition, the distances from the primer to both the transcriptional termination signal (TGA) and the ATG sites are noted as 104 bp and 115 bp. The dashed line represents exon 8 of SF3a2 including the location of the TGA and polyadenylation site (ATTAAA). The dotted line in the middle represents the AMH promoter region showing from right to left the AMH transcription initiation site (Inr) and different transcription factor and other additional regulatory DNA binding sites (black arrows) reported in the literature (see Introduction) as well as those sites determined using sequence analysis in this report. The positive and negative numbers which span the 1066 bp region correlate with the location of the Inr (+1), and the transcription and regulatory DNA elements in relation to the ATG site 5' of the promoter.

 
Following AMH gene activation and protein expression, AMH binds to the serine/threonine kinase AMH-receptor 2 (AMHR2), a homologue to TGF{beta}-related proteins (Imbeaud et al., 1995Go; MacLaughlin and Donahoe, 2002Go). AMHR2 phosphorylates the AMH receptor 1 (AMHR1), serine/threonine kinases, which are also called activin-like receptor kinase 2 (ALK2) and ALK3 (Jamin et al., 2002Go). ALK2 and 3 most likely form a heteromeric complex, but a third factor ALK6 has also been implicated in the AMH signalling (Visser, 2003Go). The AMHR1 phosphorylation induces a TGF-{beta}-like signal cascade of gene regulation involving protein–protein interactions and phosphorylation events including the Smad proteins. Interestingly, DNA sequence variations like deletions and single base point mutations in AMH or AMHR2 coding regions resulting in a loss of function were found in patients with persistent Müllerian duct syndrome and represent phenotypically internal pseudohermaphrodites, having both male and female reproductive organs (Imbeaud et al., 1995Go). AMHR2 mutant male mice also presented as internal pseudohermaphrodites and the phenotype of AMH/AMHR2 double-knockout mutant males was indistinguishable from that of either single mutant (Behringer et al., 1994Go). To date, no significant AMH or AMHR2 mutations in exons and introns have been identified in MRKH patients and control individuals, suggesting that these two genes of the AMH signalling pathway are not involved in the MRKH syndrome (Resendes et al., 2001Go; Zenteno et al., 2004Go). In contrast to mutation analyses in the coding regions of AMH and AMHR2, differences in gene expression levels could be involved in the activation of the AMH signal transduction pathway. Interestingly, in vivo studies of female transgenic mice constitutively over-expressing AMH led to a phenotype with partial similarities to the human MRKH syndrome including a blind vagina, no uterus or oviducts and ovarian degeneration post birth (Behringer et al., 1990Go).

Several examples in the literature describe DNA promoter sequence variations stimulating gene expression. For example, a DNA polymorphism at position + 331G/A in the progesterone receptor promoter creates a new transcription start site for the B isoform and appears to be associated with an increased risk for endometrial cancer (De Vivo et al., 2002Go). DNA polymorphisms within the interleukin-10 (IL-10) gene promoter associate with high IL-10 expression associated with systemic lupus erythematosus (Lazarus et al., 1997Go). The present study hypothesized that DNA polymorphisms could induce aberrant AMH expression during fetal development leading to Müllerian duct regression and the MRKH syndrome. Thirty MRKH patients including three MURCS phenotypes were analyzed for DNA sequence variations in the entire AMH gene promoter and downstream regulatory elements of the SF3a2 gene, which could be involved in aberrant AMH gene activation and protein production as detected by blood plasma levels.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Patient groups
The 30 MRKH patients (aged 17–49 years, mean 27.6 years) were either from the Department of Gynecology at the University of Erlangen-Nuremberg or from other German Institutes. All patient handling and patient blood samples were in accordance with the Ethics Committee review and approval at the University of Erlangen-Nuremberg. All patients or parents of patients gave written informed consent. Patients were previously diagnosed with the MRKH syndrome and were contacted to visit their gynecologist and donate a blood sample for this study. Clinical examination of the internal organs was determined either by laparoscopy (n=24) or by ultrasound for patients 155, 1232, 1336, 1337, 1504, 1899. All patient clinical MRKH descriptions are shown in Table I.


View this table:
[in this window]
[in a new window]
 
Table I. Mayer–Rokitanski–Küster–Hauser (MRKH) syndrome patient characteristics

 
Female adult control individuals from the University of Erlangen-Nuremberg for both DNA and blood plasma were pre-selected to have normal ovulatory cycles, normal female reproductive tracts, children and no prior benign or cancer diseases. Plasma control samples were collected from: (i) male adult individuals, and (ii) cord blood from male and female newborns (Department of Obstetrics) at the University of Erlangen-Nuremberg. An MRKH webpage from the Department of Gynecology at the University of Erlangen-Nuremberg is available at: http://www.mrkh-syndrom.de/index.php

DNA, plasma and peritoneal fluid collection and processing
All blood samples were collected in EDTA or CPDA tubes for preservation of the lymphocyte and plasma blood fractions. After an initial centrifugation at 1300 g, plasma was aliquoted and stored at –80°C. DNA from 8 ml of blood was extracted using a genomic DNA purification kit (Puregene) with modifications. Briefly, after initial centrifugation, the white blood cell layer was removed and added to red blood cell lysis buffer, pH 7.3, containing 0.15 M NH4Cl, 0.01 M K2CO3 and 0.1 mM Na-EDTA. After a 10 min incubation the cells were centrifuged at 2000 g and incubated with 3 ml of cell lysis buffer containing 20 mM Tris, 15 mM Na-EDTA, 1% sodium dodecyl sulphate and treated with RNase A and proteinase K. The proteins were precipitated with 1 ml of protein precipitation solution (Puregene) and the DNA precipitated by the addition of isopropanol, washed with 70% ethanol, dried, solubilized with a Tris–EDTA buffer, pH 7.5, quantitated using a spectrophotometer and stored at –80°C. Using this methodology an average of 70–100 µg of DNA per patient was obtained. Several peritoneal fluids were collected from MRKH patients during a laparoscopic exam.

PCR amplification of the entire AMH promoter and DNA sequencing
The complete gene sequence was extracted from the gene data bank at NCBI for chromosome 19 (NT_011255) and for AMH (NT_011255) and SF3a2 (NM_007165). A standard 100 µl PCR reaction was performed to amplify the entire AMH promoter (1066 bp DNA fragment) spanning from the last exon of SF3a2 to the first exon of AMH: 200 ng of total genomic DNA was incubated with 100 ng of primers SAP-TF (5'-GTCCACCCTCAGCCTCCGG GAGTTCACC) and MIS-BR (5'-GATGCCTGGAGGCCAGTCCAAGTCTTCTCG), 200 µM dNTP, PCR buffer containing MgCl2 and 2.5 U Taq DNA polymerase for 30 cycles (3 min 95°C initial, 90 s 95°C, 90 s 57°C, 3 min 72°C). The MRKH patient genomic AMH promoter was 5' and 3' DNA sequenced [in part by GATC-Biotech (Konstanz, Germany) and at the Department of Biochemistry, University of Erlangen (Germany)] using the primers SAP-TF and MIS-BR and in addition MISMP-Seq: 5'-AGGCCATCTCCAAGGTACTG with dye termination (Pharmacia Amersham) and then analysed on an ABI377. All DNA alignments and analyses were performed using MacVector 7.11 (Accelrys, Inc.).

The AMH specific enzyme-linked immunosorbent assay (Beckman-Coulter) uses monoclonal antibodies for detecting AMH in serum or plasma and was performed according to the manufacturer's instructions in the ultrasensitive mode with a detection limit of 0.7 pM. No cross-reactivity with TGF-{beta} was noted.

Matrix-assisted laser desorption/ionization-time-of-flight mass spectronomy (MALDI-TOF-MS based genotyping)
Aliquots of AMH promoter PCR products for –639C/T using –639TF: CAGCAAGCCCAGCGCCAGGTGCTCTTGCCT and –639BR: CAGGCCATCTCCAAGGTACAGCCATCT and for –210 G/A using –210TF: CAGCGCTGTCTAGTTTGGTTGCCTGGCCGTCA and –210BR: GCCTGCCTTAAGTGAGCCGAGTGGAAGGTG were purified using magnetic beads (genopure dsTM) as specified by the supplier (Bruker Daltonik, Bremen, Germany). Purified PCR DNA was used for the primer extension reactions after addition of 10 µl of extension mix, containing 12 pmol extension primer (–639C/T forward TGATGTCCGCAGCGC or –210G/A reverse CCTGATGTGTCAACATGC), 1U Thermosequenase, reaction buffer (Pharmacia, Germany), 2 nmol ddCTP, ddGTP and dTTP. Extension reactions were performed for 45 cycles (2 min 96°C initial, 30 s 96°C, 30 s 53°C, 3 min 72°C). To confirm the MALDI-TOF-MS based genotyping, samples without DNA were used as negative and samples with known genotypes as positive controls. Primer extension products were purified with magnetic beads (genopure oligoTM) and then aliquots spotted onto matrix crystals of 3-hydroxypicolinic acid on an anchor targetTM (Bruker Daltonik) and air dried. Mass determinations were performed on an AutoflexTM MALDI-TOF mass spectrometer (Bruker Daltonik) equipped with a nitrogen laser ({lambda}=337 nm) and delayed extraction. Laser-desorbed positive ions were evaluated following acceleration by 20 kV in the linear mode. External calibration was performed using a standard oligonucleotide mixture. Generally, 20 individual spectra were averaged to produce a mass spectrum. Genotypes were determined by analyzing the signals observed in the spectra: the forward extension primer for nucleotide position –639 C/T had a theoretical molecular mass of 4570 Da and was designated to anneal to DNA adjacent to the polymorphic C/T side. In case of the C-allele, the primer was terminated directly by a ddCTP, resulting in a mass of 4843 Da. When the allelic C to T exchange was present a dTTP was incorporated while the following ddGTP terminated the primer extension reaction yielding an extended primer of 5187 Da. The reverse extension primer used to analyze the –210 G/A single nucleotide polymorphism possessed a molecular mass of 5475 Da. When the G-allele was present, the primer was extended by ddC resulting in a mass of 5748 Da, while it was elongated by dTTP and terminated by ddG yielding a mass of 6092 Da in case of the A-allele.

Statistical methods
The SPSS 11.1 (SPSS Inc., Chicago, IL) package was used for all statistical analysis. The Hardy–Weinberg agreement was proven by a {chi}2 test.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
MRKH clinical information including the MURCS phenotype
All 30 MRKH patients were evaluated clinically to have an aplasia of the vagina and uterus and are described in Table I. Seventeen patients were typical, 10 patients atypical and three patients with the MURCS MRKH class types. Uncommon clinical patient characteristics are the following: patient 1402 lacked fallopian tubes, patients 331, 751 and 1239 had unilateral fallopian tubes, patients 1665 and 1832 had rudimentary or streak ovaries and patients 751 and 1239 had only one ovary. Several patients had benign tumors, for example, two patients, 1617 and 1664 had ovarian cysts and patient 1104 a uterus myomatosus at the remaining uterus horn. Twenty-three patients underwent reconstructive surgery for a neovagina, two patients had a vaginal expansion and the remaining five patients presently have no vaginal reconstruction.

AMH and SF3a2 DNA sequencing of 30 MRKH patients and MALDI-TOF MS of 48 controls for detection of DNA sequence variations
From 30 MRKH patients, genomic sequencing of the entire AMH promoter including parts of the last SF3a2 exon and first exon of AMH resulted in several genotype distributions (Table II). Two main DNA polymorphisms were detected in MRKH patients: 22 (73%) were both homozygous for a DNA base change from C to T at position –639 (–639 C/T) and homozygous for another DNA base change from G to A at position –210 (–210 G/A). Six patients (20%) were both heterozygous for –639 C/T and homozygous for –210 G/A (Table II and IIITable II and III).


View this table:
[in this window]
[in a new window]
 
Table II. SF3A2 exon 8 and anti-Müllerian hormone (AMH) promoter DNA polymorphisms in MRKH patients

 

View this table:
[in this window]
[in a new window]
 
Table III. Comparison of the different genotypes at position –639 and –210 of the AMH promoter in MRKH patients and control individuals

 
Genotyping the control population for the two most common AMH promoter DNA sequence variations at positions –639 C/T and –210 G/A (Figure 2; Table III) showed a similar distribution of the polymorphism as found in the 30 MRKH patients. Double homozygote DNA polymorphisms at positions –639 C/T and –210 G/A were detected in 33 (69%) control individuals, and 11 (23%) individuals demonstrated heterozygote phenotypes at positions –639 C/T (Table III). The wild-type genotype matching the NCBI data base at position –639 and –210 was only found in one control individual. According to the Hardy–Weinberg agreement using the {chi}2-test no significant differences between MRKH patients and the control population were found, thus supporting a normal functioning AMH promoter for all individuals tested.



View larger version (28K):
[in this window]
[in a new window]
 
Figure 2. (A) Schematic primer extension assay for matrix-assisted laser desorption/ionization-time-of-flight mass spectronomy (MALDI-TOF-MS) based genotyping of the C and T DNA nucleotides at position –639 of the AMH promoter. The double-stranded DNA molecule above depicts the genomic sequence of nucleotides –660 to –627 including the polymorphic position indicated by bold letters in a grey shadowed box. The single strand primer extension products below with dideoxynucleotides specific for the DNA polymorphism indicate each allele with C or T. (B) MALDI-TOF-MS genotyping C and T nucleotides at position –639. The MALDI-TOF-MS spectra showing the relative intensity (rel. int.=y-axis) and size in Daltons (m/z = x-axis) of C-allele (4843 Da), T-allele (5187 Da) and a water control. The remaining unextended primers (4570 Da) were used for internal mass calibration.

 
Additional AMH promoter sequence polymorphisms were also detected in a few MRKH patients. These include positions –769 C/G, –718 G/T, –527 G/A and –510 G/A (Table II). Comparing all observed DNA base changes with known regulatory promoter elements, the –639 C/T polymorphism located within a GC-box, an SP1 transcription factor binding site, and the –210 G/A polymorphism mapped in a SF1 binding domain (Figure 1). A further DNA sequence analysis of the AMH promoter demonstrated other possible binding sites for regulatory elements (Figure 1). Putative transcription factor DNA binding sites are three GC boxes at –111, –370 and –635 with 81% homology to the consensus motif. An activating transcription factor (ATF) binding site at –655 (71%), a nuclear factor (NF){kappa}B-binding site at –352 (80%) and a CAAT-box at –131 (90%). Putative inducible DNA element binding sites are a phorbol ester binding site (TRE) at –358 (70%) and a serum response element (SRE) at –201 (70%). In addition, one partial estrogen response element (ERE) was found at –376. All of the above regulatory elements showed no DNA base changes in MRKH patients.

Interestingly, in seven of 30 MRKH patients DNA polymorphisms were also detected in the coding region of the last exon of SF3a2 resulting in specific amino acid exchanges (Table II). For example, the codon at position (+1321)–(+1323) ATG (Met) of SF3a2 either changed from an A to C at + 1321 (Met to Leu) for patients 155, 288, 331 and 390; a T to C at position + 1322 (Met to Pro) for patient 390, or a G to C at position + 1323 (Met to Ile) for patients 205 and 368. These base changes did not associate with any specific MRKH phenotype. In addition, no DNA sequence variations were detected in the transcriptional stop site or the poly A regulatory signal element of SF3a2, thus by DNA sequence analysis the possibility for read through transcripts containing both SF3a2 and AMH cannot be a possibility in MRKH patients.

AMH protein levels in blood and peritoneal fluid of MRKH patients and controls
In a series of control individuals AMH blood plasma levels were determined in adult males, females and cord blood samples from newborn babies (Figure 3). AMH concentration in 10 adult males (age 22–39 years, mean 28.7 years) averaged 97.85 pM (range 41.6–206.7 pM). In cord blood from females, AMH averaged 9.95 pM and in boys 267.98 pM, a difference of ~27-fold. These AMH protein blood levels are in accordance with reports from normal individuals in the literature (Rey et al., 1999Go). AMH protein plasma analysis of the 30 MRKH patients in this study (age 17–49 years, mean 27.6 years) resulted in a mean concentration of 44.44 pM (range 15.2–138.9 pM), compared with 45.08 pM (range 14.9–64.15 pM) in 10 normal ovulatory control women (age 18–42 years, mean 27.5 years) (Figure 3). However, MRKH patients 205, 331 and 1740 had relatively high AMH levels of 87.8, 138.9 and 121.5 pM, respectively.



View larger version (17K):
[in this window]
[in a new window]
 
Figure 3. AMH blood plasma protein expression of MRKH patients and control cohorts. The graph shows the enzyme-linked immunosorbent assay results in pmol/l of AMH from different patient and control groups, from left to right: MRKH patients (n=30), Co/F = Control female individuals (n=10), Co/M = control male individuals (n=10), CB/F = cord blood from female newborns (n=5) and CB/M = cord blood from male newborns (n=5).

 
The AMH concentration in peritoneal fluids from several MRKH patients (n=3) were similar to both normal peritoneal fluid values and plasma values (data not shown). This observation demonstrates that AMH protein production at the site of the ovaries is detected at the region of secretion. The findings of not aberrantly high AMH blood or peritoneal fluid protein levels support our genomic DNA sequence findings of a normal AMH promoter and SF3a2 3' regulatory elements. Because AMH protein is exclusively produced from ovarian granulosa cells in females following birth until menopause the finding of normal AMH blood levels is a direct and specific tissue indicator for healthy functioning ovaries.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The MRKH syndrome occurs congenitally, but because of normal pubertal development most patients are not diagnosed until adolescence. Some theories for the origin of MRKH syndrome have been previously proposed. Küster (1910)Go explained the MRKH syndrome as the regression of a Müllerian duct. Schmid-Tannwald and Hauser (1977)Go proposed the regression of Müllerian duct due to a temporary secretion of AMH during the first fetal weeks. Morphologically, according to Ludwig (1998)Go the MRKH syndrome results from a non-fusion of the Müllerian duct with the Wolffian duct, where the rudimentary development of the vagina is possibly caused by a deficiency of gestagen and/or estrogen receptors (ER). According to Ludwig (1998)Go the non-fusion of the two ducts could explain why a very small part of the cornu uteri proceeds only as far as the attachment point. Interestingly, in most MRKH patients the fallopian tubes persist, however rare cases exist with abnormalities. In this study, we observed that patients 331, 751 and 1239 have a unilateral and patient 1402 a bilateral absence of fallopian tubes. In addition, patients 751 and 1239 have an ipsilateral absence of the ovaries (Table I).

LH, FSH, prolactin, E2 and progesterone levels have previously been determined to be normal in MRKH patients (Carranza-Lira et al., 1999Go). In the present study AMH has been measured and determined as generally normal in blood (Figure 3) and peritoneal fluid of MRKH patients. This finding reflects normally regulated AMH gene promoters, transcription and AMH protein production in adult MRKH patients. Normal female blood AMH levels decrease over time and associate with the number of remaining antral follicles (Fanchin et al., 2003Go), therefore AMH represents a specific marker for ovarian egg age and could be helpful for predicting successful IVF or surrogacy pregnancies. AMH serum concentrations of non-ovulatory women in the reproductive years were increased 3-fold when compared to normal-ovulatory women of the same age (Laven et al., 2004Go). In this study, MRKH patients 205 had a 2-fold, 331 and 1740 an ~3-fold higher AMH concentration (Figure 3), suggesting a non-ovulating state, whereas all other MRKH patients most likely have normal ovulation according to the AMH and other hormones, but no menstrual cycle. In addition, MRKH patients 751 and 1239 with unilateral ovarian agenesis and patients 1665 and 1832 with rudimentary and streak ovaries, respectively, support that these abnormal ovaries are sufficient for AMH production. Taken together the findings of normal hormone levels in MRKH individuals supports normal pituitary gland and ovarian function with ovulation.

Maternalships of MRKH patients through 34 successful surrogacy pregnancies have been described and resulted in 17 male and 17 normal female babies (Petrozza et al., 1997Go). Besides normal hormone levels and egg development among MRKH women, these maternalships strongly support a multifactorial inheritance that cannot be explained by autosomal or X-linked dominant genes. Further support for a multifactorial inheritance comes from a study of monozygotic twins where one of the twins had a uterus and vagina aplasia and the other twin had severe skeletal defects (Steinkampf et al., 2003Go), however another report described only one discordant monozygotic twin with the MRKH phenotype (Lischke et al., 1973Go). In contrast, Shokeir, (1978)Go described the MRKH syndrome inherited as sex-linked dominant, while other studies demonstrated a more recessive inheritance (Jones and Mermut, 1972Go). In the present investigation none of the MRKH patients had affected siblings with MRKH, indicating no clear inheritance pattern.

In this investigation we determined that the AMH promoter and SF3a2 downstream regulatory elements of MRKH patients are similar to control individuals indicating that aberrant developmental AMH gene activation or constitutive expression of AMH cannot be due to DNA sequence variations (Tables II and III). However, some DNA base changes were observed in the AMH promoter and in the last exon of SF3a2. For example, in the AMH promoter the DNA base change at –639 C/T alters the GC-Box 5'-GGGCGG to 5'-GGGCAG destroying the SP1 transcription factor binding site, thus only two partial GC-boxes remain at positions –111 and –370. Variation in the number of promoter SP1 sites appears to be common in the population and could cause differences in gene transcription levels. All other DNA polymorphisms did not occur in any known transcription factor binding sites. Interestingly, the –639 C/T and the –210 G/A polymorphisms represent the prominent AMH genotypes in the general population (Table II). Therefore, the DNA sequence in the NCBI databank depicts the rare polymorphic DNA sequence. In addition, one AMH promoter polymorphism at –380 found in the Japanese population (Haga et al., 2002Go) was neither detected in the MRKH patients nor in the control population, probably representing a race specific polymorphism. The finding of single nucleotide base changes in SF3a2 exon 8 alters the amino acids in the codon, however, would not influence activation of AMH.

In contrast to the bovine, rat and mouse AMH gene, the human AMH gene promoter only partly has a consensus to the TATAA-box element (CTTAA) at position –27. In vitro studies have demonstrated that human AMH transcription initiation is directed by an initiator region at –10 to + 10 bound by transcription factor (TF) II-I (Morikawa et al., 2000Go). Other regulatory elements also mapped in the AMH promoter including possible protein DNA binding sites for NFKB, ATF, nuclear factor (NF)1/CTF, SP1, activator protein (AP)1, serum response factor, as well as for ER (Figure 1). All of these DNA elements were normal in MRKH patients. In previous studies, DNA sequence variations were screened for in a variety of genes to determine an association with MRKH. AMH and AMHR2 gene coding regions have been analysed for polymorphisms in over 37 MRKH patients (Resendes et al., 2001Go; Zenteno et al., 2004Go). The identified polymorphisms in AMH and AMHR2 genes stated above were different between both patient groups in nucleotide position, but were not significant when compared to the control population. The present and above mutation analyses of the AMH promoter and gene strongly support that there is no link between the MRKH syndrome and AMH at the genomic level. Analysis of the Wilms Tumor (WT) 1 gene and cystic fibrosis transmembrane conductance regulator gene also demonstrated no significant mutations in MRKH patients (Van Lingen et al., 1998Go; Timmreck et al., 2003Go). In addition, a DNA polymorphism in the galactose-1-phosphate-uridyl transferase (GALT) gene appears to associate with MRKH (Cramer et al., 1996Go), but is not proven as yet (Klipstein et al., 2003Go; Zenteno et al., 2004Go).

Other candidate genes than AMH or AMHR2 in the AMH signaling pathway may play a role in the molecular etiology of the MRKH syndrome. Mutations in the receptor AMHR1 genes (ALK2 and 3), especially the kinase domain could lead to constitutive self-phosphorylation activity. However, a deletion of AMHR1 cannot be responsible for the MRKH syndrome, due to knock-out AMHR1 mice experiments, which was embryonic lethal (Goumans and Mummery, 2000Go). Interestingly, ALK6 knock-out mice showed Müllerian duct regression and developed bone defects (Clarke et al., 2001Go).

Alternative mechanisms for the MRKH syndrome could be due to maternal AMH or environmental sources, which have to cross the placental barrier. For example, higher than normal maternal AMH protein levels during pregnancy could be responsible for the degradation of female fetal Müllerian ducts during development. Other substances influencing the steroid hormone synthesis have been implicated in AMH expression and sex reversal phenotypes. E2 has been shown to induce AMH expression in cell lines (Chen et al., 2003Go). It could be possible that high E2 or high amounts of estrogen-like bioflavonoids found in foods and dietary supplements, like genistein (Strick et al., 2000Go) during early pregnancy could induce AMH and result in Müllerian duct regression. Interestingly, one putative ER binding element (75% match to the consensus) was identified in the AMH promoter (Figure 1).

Finally, it will be important to compare the genetic etiologies in other syndromes as they become known, which share similar phenotypes with MRKH. For example, the rare Winter's syndrome, Klippel–Feil syndrome, and the vaginal aplasia syndrome. The Klippel–Feil syndrome, a rare skeletal disorder that affects both males and females, can show a familial dominant or recessive inheritance pattern (Clarke et al., 1995Go) and where some females with normal secondary sex characteristics also have abnormalities of the fallopian tubes, uterus and/or vagina (Park et al., 1971Go). Other syndromes include the urogenital adysplasia syndrome, showing an autosomal dominant pattern of inheritance with incomplete penetrance and is also associated with Müllerian defects. Thus, determining the molecular cause for the MRKH syndrome may also help to discern if the other syndromes described above have different molecular etiologies, or more importantly, share overlaps with some of the involved genes.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors wish to dedicate this study to one of the pioneers of the MRKH syndrome: Professor G.A.Hauser, Lucerne, Switzerland. The authors are especially grateful to the patients who participated in this study and to the Department of OB/GYN, Erlangen. We thank Professor C.-M.Becker (Department of Biochemistry) for his support of this study, Mrs Oeser and Stiegler (OB/GYN) and Mrs Wenzeler (Department of Biochemistry) for expert technical assistant. This work was supported by the ELAN Fonds (03.03.10.1) to P.O., R.S. and P.L.S.


    Notes
 
* These authors contributed equally to this work Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Arango NA, Lovell-Badge R and Behringer RR (1999) Targeted mutagenesis of the endogenous mouse Mis gene promoter: in vivo definition of genetic pathways of vertebrate sexual development. Cell 99, 409–419.[ISI][Medline]

Basile C and De Michele V (2001) Renal abnormalities in Mayer-Rokitanski-Kuster-Hauser syndrome. J Nephrol 14, 316–318.[CrossRef][ISI][Medline]

Behringer RR, Finegold MJ and Cate RL (1994) Mullerian-inhibiting substance function during mammalian sexual development. Cell 79, 415–425.[ISI][Medline]

Behringer RR, Cate RL, Froelick GJ, Palmiter RD and Brinster RL (1990) Abnormal sexual development in transgenic mice chronically expressing mullerian inhibiting substance. Nature 345, 167–170.[CrossRef][ISI][Medline]

Bennett M and Reed R (1993) Correspondence between a mammalian spliceosome component and an essential yeast splicing factor. Science 262, 105–108.[ISI][Medline]

Carranza-Lira S, Forbin K and Martinez-Chequer JC (1999) Rokitansky syndrome and MURCS association-clinical features and basis for diagnosis. Int J Fertil Womens Med 44, 250–255.[ISI][Medline]

Chen G, Shinka T, Kinoshita K, Yan HT, Iwamoto T and Nakahori Y (2003) Roles of estrogen receptor alpha (ER alpha) in the regulation of the human Mullerian inhibitory substance (MIS) promoter. J Med Invest 50, 192–198.[Medline]

Clarke RA, Singh S, McKenzie H, Kearsley JH and Yip MY (1995) Familial Klippel-Feil syndrome and paracentric inversion inv(8)(q22.2q23.3). Am J Hum Genet 57, 1364–1370.[Medline]

Clarke TR, Hoshiya Y, Yi SE, Liu X, Lyons KM and Donahoe PK (2001) Mullerian inhibiting substance signaling uses a bone morphogenetic protein (BMP)-like pathway mediated by ALK2 and induces SMAD6 expression. Mol Endocrinol 15, 946–959.[Abstract/Free Full Text]

Cramer DW, Goldstein DP, Fraer C and Reichardt JK (1996) Vaginal agenesis (Mayer-Rokitansky-Kuster-Hauser syndrome) associated with the N314D mutation of galactose-1-phosphate uridyl transferase (GALT). Mol Hum Reprod 2, 145–148.[Abstract]

De Vivo I, Huggins GS, Hankinson SE, Lescault PJ, Boezen M, Colditz GA and Hunter DJ (2002) A functional polymorphism in the promoter of the progesterone receptor gene associated with endometrial cancer risk. Proc Natl Acad Sci USA 99, 12263–12268.[Abstract/Free Full Text]

Dresser DW, Hacker A, Lovell-Badge R and Guerrier D (1995) The genes for a spliceosome protein (SAP62) and the anti-Mullerian hormone (AMH) are contiguous. Hum Mol Genet 4, 1613–1618.[Abstract]

Duncan PA, Shapiro LR, Stangel JJ, Klein RM and Addonizio JC (1979) The MURCS association: Mullerian duct aplasia, renal aplasia, and cervicothoracic somite dysplasia. J Pediat 95, 399–402.[ISI][Medline]

Fanchin R, Schonauer LM, Righini C, Guibourdenche J, Frydman R and Taieb J (2003) Serum anti-Mullerian hormone is more strongly related to ovarian follicular status than serum inhibin B, estradiol, FSH and LH on day 3. Hum Reprod 18, 323–327.[Abstract/Free Full Text]

Foster JW, Dominguez-Steglich MA, Guioli S et al. (1994) Campomelic dysplasia and auto-somal sex reversal caused by mutations in an SRY-related gene. Nature 372, 525–530.[ISI][Medline]

Goumans MJ and Mummery C (2000) Functional analysis of the TGFbeta receptor/Smad pathway through gene ablation in mice. Int J Dev Biol 44, 253–265.[ISI][Medline]

Haga H, Yamada R, Ohnishi Y, Nakamura Y and Tanaka T (2002) Gene-based SNP discovery as part of the Japanese Millennium Genome Project: identification of 190,562 genetic variations in the human genome. Single-nucleotide polymorphism. J Hum Genet 47, 605–610.[CrossRef][ISI][Medline]

Hauser GA and Schreiner WE (1961) Das Mayer-Rokitanski-Küster-Syndrome. Schweiz Med Wochenschrift 61, 1–11.

Imbeaud S, Faure E, Lamarre I et al. (1995) Insensitivity to anti-Mullerian hormone due to a mutation in the human anti-Mullerian hormone receptor. Nature Genet 11, 382–388.[ISI][Medline]

Jamin SP, Arango NA, Mishina Y and Behringer RR (2002) Genetic studies of MIS signalling in sexual development. Novartis Found Symp 244, 157–164.[ISI][Medline]

Jones HW Jr and Mermut S (1972) Familial occurrence of congenital absence of the vagina. Am J Obstet Gynecol 114, 1100–1101.[ISI][Medline]

Klipstein S, Bhagavath B, Topipat C, Sasur L, Reindollar RH and Gray MR (2003) The N314D polymorphism of the GALT gene is not associated with congenital absence of the uterus and vagina. Mol Hum Reprod 9, 171–174.[Abstract/Free Full Text]

Küster H (1910) Uterus bipartitus solidus rudimentarius cum vagina solida. Z Geburtshilfe Gynakol 67, 692–718.

Laven JS, Mulders AG, Visser JA, Themmen AP, De Jong FH and Fauser BC (2004) Anti-Mullerian hormone serum concentrations in normoovulatory and anovulatory women of reproductive age. J Clin Endocrinol Metab 89, 318–323.[Abstract/Free Full Text]

Lazarus M, Hajeer AH, Turner D, Sinnott P, Worthington J, Ollier WE and Hutchinson IV (1997) Genetic variation in the interleukin 10 gene promoter and systemic lupus erythematosus. J Rheumatol 24, 2314–2317.[ISI][Medline]

Lee MM, Donahoe PK, Silverman BL, Hasegawa T, Hasegawa Y, Gustafson ML, Chang Y and MacLaughlin DT (1997) Measurements of Serum Mullerian Inhibiting Substance in the Evaluation of children with nonpalpable gonads. N Engl J Med 336, 1480–1486.[Abstract/Free Full Text]

Lischke JH, Curtis CH and Lamb EJ (1973) Discordance of vaginal agenesis in monozygotic twins. Obstet Gynecol 41, 920–924.[ISI][Medline]

Ludwig KS (1998) The Mayer-Rokitansky-Küster syndrome. An analysis of its morphology and embryology. Arch Gynecol Obstet 262, 1–26.[CrossRef][ISI][Medline]

MacLaughlin DT and Donahoe PK (2002) Mullerian inhibiting substance: an update. Adv Exp Med Biol 511, 25–38.[ISI][Medline]

Morikawa N, Clarke TR, Novina CD, Watanabe K, Haqq C, Weiss M, Roy AL and Donahoe PK (2000) Human Mullerian-inhibiting substance promoter contains a functional TFII-I-binding initiator. Biol Reprod 63, 1075–1083.[Abstract/Free Full Text]

Nahum GG (1998) Uterine anomalies. How common are they, and what is their distribution among subtypes? J Reprod Med 43, 877–887.[ISI][Medline]

Park IJ, Jones HW, Nager GT, Chen SCA and Hussels IE (1971) A new syndrome in two unrelated females: Klippel-Feil deformity, conductive deafness and absent vagina. Birth Defects Orig Art Ser VIII(6), 311–317.

Petrozza JC, Gray MR, Davis AJ and Reindollar RH (1997) Congenital absence of the uterus and vagina is not commonly transmitted as a dominant genetic trait: outcomes of surrogate pregnancies. Fertil Steril 67, 387–389.[CrossRef][ISI][Medline]

Raga F, Bauset C, Remohi J, Bonilla-Musoles F, Simon C and Pellicer A (1997) Reproductive impact of congenital Mullerian anomalies. Hum Reprod 12, 2277–2281.[Abstract]

Resendes BL, Sohn SH, Stelling JR, Tineo R, Davis AJ, Gray MR and Reindollar RH (2001) Role for anti-Mullerian hormone in congenital absence of the uterus and vagina. Am J Med Genet 98, 129–136.[CrossRef][ISI][Medline]

Rey RA, Belville C, Nihoul-Fekete C et al. (1999) Evaluation of gonadal function in 107 intersex patients by means of serum antimullerian hormone measurement. J Clin Endocrinol Metab 84, 627–631.[Abstract/Free Full Text]

Rey R, Sabourin JC, Venara M, Long WQ, Jaubert F, Zeller WP, Duvillard P, Chemes H and Bidart JM (2000) Anti-Mullerian hormone is a specific marker of sertoli- and granulosa-cell origin in gonadal tumors. Hum Pathol 31, 1202–1208.[CrossRef][ISI][Medline]

Rey R, Lukas-Croisier C, Lasala C and Bedecarras P (2003) AMH/MIS: what we know already about the gene, the protein and its regulation. Mol Cell Endocrinol 211, 21–31.[CrossRef][ISI][Medline]

Schmid-Tannwald I and Hauser GA (1977) [Atypical forms of the Mayer-Rokitansky-Kuster-syndrom]. Geburtshilfe Frauenheilkd 37, 386–392.[ISI][Medline]

Shokeir MH (1978) Aplasia of the Mullerian system: evidence for probable sex-limited autosomal dominant inheritance. Birth Defects Orig Artic Ser 14, 147–165.[Medline]

Simpson JL (1999) Genetics of the female reproductive ducts. Am J Med Genet 89, 224–239.[CrossRef][ISI][Medline]

Steinkampf MP, Dharia SP and Dickerson RD (2003) Monozygotic twins discordant for vaginal agenesis and bilateral tibial longitudinal deficiency. Fertil Steril 80, 643–645.[CrossRef][ISI][Medline]

Strick R, Strissel PL, Borgers S, Smith SL and Rowley JD (2000) Dietary bioflavonoids induce cleavage in the MLL gene and may contribute to infant leukemia. Proc Natl Acad Sci USA 97, 4790–4795.[Abstract/Free Full Text]

Taguchi O, Cunha GR, Lawrence WD and Robboy SJ (1984) Timing and irreversibility of Mullerian duct inhibition in the embryonic reproductive tract of the human male. Dev Biol 106, 394–398.[ISI][Medline]

Timmreck LS, Gray MR, Handelin B, Allito B, Rohlfs E, Davis AJ, Gidwani G and Reindollar RH (2003) Analysis of cystic fibrosis transmembrane conductance regulator gene mutations in patients with congenital absence of the uterus and vagina. Am J Med Genet 120, 72–76.[Medline]

Van Lingen BL, Reindollar RH, Davis AJ and Gray MR (1998) Further evidence that the WT1 gene does not have a role in the development of the derivatives of the mullerian duct. Am J Obstet Gynecol 179, 597–603.[ISI][Medline]

Vecchietti G (1979) [The neovagina in the Robitansky-Kuster-Hauser syndrome]. Rev Med Suisse Romande 99, 593–601.[Medline]

Visser JA (2003) AMH signaling: from receptor to target gene. Mol Cell Endocrinol 211, 65–73.[CrossRef][ISI][Medline]

Zenteno JC, Carranza-Lira S and Kofman-Alfaro S (2004) Molecular analysis of the anti-Mullerian hormone, the anti-Mullerian hormone receptor, and galactose-1-phosphate uridyl transferase genes in patients with the Mayer-Rokitansky-Kuster-Hauser syndrome. Arch Gynecol Obstet 269, 270–273.[Medline]

Submitted on May 4, 2004; resubmitted on July 8, 2004; accepted on September 10, 2004.