Single-nucleotide polymorphisms in the promoter region influence the expression of the human folliclestimulating

Transcription

1 REPRODUCTIVE BIOLOGY Single-nucleotide polymorphisms in the promoter region influence the expression of the human folliclestimulating hormone receptor Alain Wunsch, M.Sc., a Yuni Ahda, Ph.D., a,b Ferya Banaz-Yaşar, M.Sc., a,c Barbara Sonntag, M.D., d Eberhard Nieschlag, M.D., a Manuela Simoni, M.D., Ph.D., a and Jörg Gromoll, Ph.D. a a Institute of Reproductive Medicine, University Hospital, Münster, Germany; b University of Indonesia, Jakarta, Indonesia; c Institute of Anatomy of the University of Essen, Essen, Germany; and d Department of Obstetrics and Gynecology, University Hospital, Münster, Germany Objective: To characterize novel single-nucleotide polymorphisms (SNPs) in the human FSH receptor (FSHR) promoter region. Design: Retrospective and basic research study. Setting: University hospital. Patients: Women (202 from Germany and 55 from Indonesia) with male or tubal factor infertility undergoing controlled ovarian stimulation for IVF treatment. Interventions: None. Main Outcome Measure(s): Frequency, distribution, and correlation with clinical data of the SNPs. Dual luciferase assays and electrophoretic mobility shift assays (EMSA). Result(s): We identified two SNPs and three mutations in the promoter region of the human FSHR which could be allocated to positions 29, 37, 114, 123, and 138 upstream of the translational initiation codon. One SNP showed a high incidence ( 29: 44%, n 202), but no correlation with basal FSH serum levels or ovarian response with the SNP at position 29 was found. Luciferase reporter assays, using pgl3 vector constructs, showed that mutations at positions 37 and 138 lead to significantly higher promoter activity. EMSA indicate that putative binding sites for transcription factors are affected by the SNPs. Conclusions: The newly identified SNPs do not seem to influence clinical parameters substantially, but modulate expression of the FSHR via changes in transcription factor binding sites. (Fertil Steril 2005;84: by American Society for Reproductive Medicine.) Key Words: FSH receptor, promoter region, single-nucleotide polymorphisms The action of follicle-stimulating hormone (FSH) is mediated through the FSH receptor (FSHR), a member of the G protein coupled receptors (GPCR) family. The FSHR and its ligand play an important role in the regulation of gametogenesis. The receptor is specifically expressed in Sertoli and granulosa cells and mainly activates the adenylate cyclase/camp signal transduction pathway (1). It can be divided into three regions: the extracellular domain, a transmembrane region and the intracellular domain (2, 3). The gene consists of nine introns and ten exons and has been assigned to chromosome 2p21. Received November 26, 2004; revised and accepted February 4, A.W. was supported by a fellowship of the Ernst Schering Research Foundation. Y.A. was recipient of a fellowship of the German Academic Exchange Service (DAAD). This study was supported by the Deutsche Forschungsgemeinschaft (SI-526/1-1). Reprint requests: Eberhard Nieschlag, M.D., Institute of Reproductive Medicine of the University of Münster, D Münster, Germany (FAX: ; nieschl@uni-muenster.de). Mutational screening of the FSH receptor coding region showed the presence of two common polymorphisms within exon 10 at amino acid positions 307 and 680. Further studies showed that the polymorphism in exon 10 of the FSH receptor gene influences the receptor protein and seems to affect ovarian sensitivity to FSH induction in women undergoing assisted reproduction treatment (4 6), but Laven et al. (7) could not confirm any altered ovarian sensitivity to exogenous FSH during ovulation induction in anovulatory patients. Besides the impact of single-nucleotide polymorphisms (SNPs) on the protein level, polymorphisms in the promoter region could also account for different FSH sensitivities by influencing FSHR mrna expression. Unlike the FSHR gene structure, which has been extensively studied, information about the human FSHR gene promoter is still limited. The FSHR promoter belongs to the TATA-less promoter group and has several transcriptional start sites. There are two transcriptional start sites in the rat 446 Fertility and Sterility Vol. 84, No. 2, August /05/$30.00 Copyright 2005 American Society for Reproductive Medicine, Published by Elsevier Inc. doi: /j.fertnstert

2 promoter at positions 80 and 98 relative to translational start site (8), and studies in the human FSH receptor promoter revealed five transcriptional start sites at positions 184, 114, 99, 83, and 79, with the main transcriptional start site found at position 99 (2). Deletion analysis in the 5= flanking region of the human FSHR showed that the region between 1 and 225 is needed for full promoter activity (2). Studies on rat, mouse, and sheep FSHR genes identified several important regulatory regions in the core promoter, e.g., E box element, AP-1 site, InR element, E2F element, GATA sites, CACC box, and two steroidogenic factor-1 (SF-1) like binding sites (SLBS) (8 13). However, these sites have not been found in the core promoter region of the human FSH receptor so far, except for the E box element that seems to be conserved in most animal species. Block-replacement mutants analysis in the important regulatory sites of rat and sheep FSHR, such as E box, GATA sites, and CACC box, indicated decreasing activities of the promoter at variable levels. Promoter activity decreases to the lowest level when the E box element has been mutated (8, 9, 11). Electrophoretic mobility shift assays (EMSA) revealed several transcriptional factors binding to the regulator elements, such as USF1/2 (E box), GATA-1 (GATA site), and two Krupple transcription factors (CACC box) (8, 14). Because most of the regulatory sites that were found in the rat, mouse, and sheep FSHR promoters do not exist in the human FSHR promoter, it is assumed that the human FSHR promoter has a different mechanism for transcriptional regulation. To gain more information about the human FSHR promoter, we screened the promoter region of the FSHR for single-nucleotide polymorphisms. We performed a clinical and functional characterization of the newly identified SNPs to investigate whether they may influence expression of the FSHR. Because some studies suggest that ethnic differences could be involved in the effect of exon 10 SNPs (6, 7, 15, 16) our study group consisted of German and Indonesian women. MATERIALS AND METHODS Study Population The study population consisted of 257 women (202 from Germany and 55 from Indonesia) undergoing controlled ovarian stimulation for IVF treatment according to standard stimulation protocols. Only subjects with male or tubal factor infertility were included. All procedures were carried out with the informed consent of the patients and approved by the local ethics committee. DNA Isolation Genomic DNA was extracted from blood samples of the subjects. Extraction was performed using the Blood and Cell Culture DNA kit (Qiagen, Düsseldorf, Germany) according to the manufacturer s instructions. Hormonal Data Basal FSH levels (day 3 of the menstrual cycle) were obtained in one of the previous cycles before ovarian stimulation. Peak E 2 levels on the day of hcg administration were determined as indicator of ovarian response. Serum levels of FSH and E 2 were measured by standard specific immunoassay on the respective cycle days (Vitros EC; Ortho-Clinical Diagnostics, Schwabach, Germany). Detection of Polymorphisms in the Promoter Initially, the FSHR promoter region was amplified from genomic DNA of women by polymerase chain reaction (PCR) using the forward primer 5=-TATTCCAGACATGC- CTAATGG-3= and the reverse primer 5=-CCAGCAAA- GAGACCAGGAGC-3=. Single-nucleotide polymorphisms were detected using single-stranded conformation polymorphism (SSCP) gel electrophoresis as described before (17). Amplicons from patients displaying aberrant migration patterns on the SSCP were sequenced. Polymorphisms at position 29 (G A) in DNA samples of women undergoing IVF treatment were further investigated using allelic discrimination analysis on an ABI PRISM 7000 detection system (Applied Biosystems, Darmstadt, Germany). Each PCR reaction contained 2 L DEPC treated water, 0.25 L of each probe, 4.5 L of each primer (5 pmol/ml), and 1 L genomic DNA. The following probes were used (lowercase letters mark the SNP position): 29-G: 5=-TGCAAATG- CAGgAAG-3= (FAM fluorescence); 29-A: 5=-TGCAAATG- CAGaAAG-3= (VIC fluorescence). Primer sequences used were as follows: 29fwd: 5=-AGCTTCTGAGATCTGTG- GAGGTTT-3=; 29rev: 5=-AATTATGCATCCATCCACCT- GATT-3=. Detection of Polymorphisms in Exon 10 Analysis of the SNP at position 680 in exon 10 was done in analogy to the detection of the 29 SNP with the difference that the following probes and primers were used: 680-G: 5=-AGAGTCACCAgTGGTT-3= (FAM fluorescence); 680-A: 5=-AGTCACCAaTGGTTC-3= (VIC fluorescence); 680fwd: 5=-AAGGAATGGCCACTGCTCTTC-3=; 680rev: 5=-GGGCTAAATGACTTAGAGGGACAA-3= (18). Cells Cells of the murine Sertoli cell line SK-11 (19) were grown at 33 C (5% CO 2 ) in Dulbecco s modified Eagle medium (DMEM) containing 10% fetal calf serum (FCS) (Invitrogen-Gibco, Karlsruhe, Germany). COS-7 cells were grown in DMEM 10% FCS at 37 C. The porcine JC-410 cell line (20) was grown in Medium 199 (Invitrogen) 10% FCS. Cells of the human COV 434 granulosa cell line (21) were grown in DMEM 10% FCS at 37 C. Primary granulosa cells were obtained from women undergoing IVF treatment at the time of oocyte pick-up. Primary Sertoli cells from 10-day-old Sprague Dawley rats were obtained as described before (22). Fertility and Sterility 447

3 Luciferase Reporter Assay Genomic DNA samples from patients with identified SNPs were amplified using the forward primer 5=-TATTCCAGA- CATGCCTAATGG-3= and the reverse primer 5=-AATTAT- GCATCCATCCACCTG-3=, resulting in a 263-bp PCR product comprising the different polymorphisms. In total, six different amplicons corresponding to the genetic variant at positions 29, 37, 114, 123, 138, and wt were cloned into pcr 2.1 TOPO vector (Invitrogen) and sequence fidelity was verified by DNA sequencing. The promoter inserts were excised by restriction digestion using KpnI and XhoI and the resulting fragments were cloned into the pgl3- Basic vector (Promega, Mannheim, Germany) containing a luciferase reporter gene. To analyze the effect of SNPs on the expression, dual-luciferase reporter assays (Promega) were performed according to the manufacturer s instructions. Briefly, cell lysates were obtained 48 h after lipofectamine (Invitrogen) transfection of the pgl3 vector constructs into the various cell lines, using a passive lysis buffer. 100 L luciferase assay reagent was added to the lysate (20 L) and after incubation with the Stop&Glo reagent luciferase activity was determined using a luminometer (Berthold LB952, Bad Wildbad, Germany). An empty pgl3 vector was used as a negative control. The activity of the cotransfected renilla luciferase was used to normalize the values (firefly luciferase values were divided by renilla luciferase values). Electrophoretic Mobility Shift Assays Electrophoretic mobilty shift assays (EMSA) were performed to analyze the binding of nuclear proteins from different cell types to oligonucleotides comprising the different SNPs. To prepare nuclear extracts, cells were harvested using ice-cold PBS (1 ). After centrifugation, the pellet was resuspended in hypotonic buffer (10 mmol/l HEPES ph 7.9, 1.5 mmol/l MgCl 2, 10 mmol/l KCl, 0.2 mmol/l phenylmethylsulfonylfluoride [PMSF], and 0.5 DTT) and incubated on ice for 10 min. The cell suspension was precipitated by centrifugation (12,000 rpm; 10 s), resuspended in high-salt buffer (20 mmol/l HEPES ph 7.9, 1.5 mmol/l MgCl 2, 420 mmol/l NaCl, 0.2 mmol/l EDTA, 25% glycerol, 0.2 mmol/l PMSF, 0.5 mmol/l DTT) and then incubated on ice for 20 min. Cell debris was precipitated (12,000 rpm; 2 min) and the supernatant stored at 70 C. Nuclear extract protein concentrations were determined using a BCA-protein assay (Bio-Rad Laboratories). DNA oligonucleotides used in EMSAs were as follows: 29 wt: 5=-CAA ATG CAG GAA GAA ATC AGG TGG-3= 29 SNP: 5=-CAA ATG CAG aaa GAA ATC AGG TGG-3= 37 wt: 5=-GTT TTT CTC TGC AAA TGC AGG AAG-3= 37 SNP: 5=-GTT TTT CTC TGC gaa TGC AGG AAG-3= 114 wt: 5=-TCA CAT GAC CCT ACC AGT TCT CAA GTC-3= 114 SNP: 5=-TCA CAT GAC CCc ACC AGT TCT CAA GTC-3= 123 wt: 5=-CTT GGT GGG TCA CAT GAC CCT ACC-3= 123 SNP: 5=-CTT GGT GGG TCg CAT GAC CCT ACC-3= 138 wt: 5=-AAA AAG CAT CCC TTG GTG GGT CAC-3= 138 SNP: 5=-AAA AAG CtT CCC TTG GTG GGT CAC-3=. Lowercase letters indicate SNP position. For each oligonucleotide a complementary oligonucleotide was used. Oligonucleotides were labelled with ( - 32 P)ATP (Amersham, Freiburg, Germany) by T4 polynucleotide kinase (New England Biolabs, Frankfurt a.m, Germany). Unincorporated nucleotides were removed using MicroSpin columns (Amersham). Labelled complementary oligonucleotides were annealed (3 min, 50 C) to form double-stranded probes. To allow DNA-protein binding, nuclear extracts (10 g) and oligonucleotides ( 50,000 cpm) were incubated for 30 min at 25 C (binding buffer: 20 mmol/l HEPES, 5 mmol/l MgCl 2, 80 mmol/l KCl, 0.5 mmol/l DTT, 0.25 mmol/l EDTA, 10% glycerol, 300 g/ml BSA) after addition of poly (di:dc) (0.5 g/ L). Supershift experiments were performed using an antiets-1 antibody (sc-350x; Santa Cruz, Heidelberg, Germany). As a positive control we used oligonucleotides comprising an Ets-1 consensus binding site (GGAA) (annealed with complementary oligonucleotides) Ets-1 5=-GAT CTC GAG CAG GAA GTT CGA-3=, and as negative control the mutated Ets-1 Mut 5=-GAT CTC GAG CAa GAA GTT CGA DNA-protein complexes were separated from nonbound oligonucleotides by electrophoresis (5% polyacrylamide gel, 90 min, 35 ma). The gel was dried and DNA-protein complexes were detected using autoradiography. Statistics Statistical analyses (Kruskal-Wallis test and one-way ANOVA) were performed using SigmaStat statistical software (SPSS, Chicago, IL). RESULTS Identification of Single-Nucleotide Polymorphisms To identify SNPs in the promoter region of the human FSHR gene, SSCP analyses of DNA samples extracted from blood of 46 women of a Westphalian population were performed. Five mutations were detected in the core promoter region, downstream of a variable (11 17 bases) poly-a stretch starting at position 141. They could be allocated to the positions 29, 37, 114, 123, and 138 upstream of the translational initiation codon (Fig. 1). Incidence of the SNPs The initial screening of 46 women revealed an incidence of over 10% for two of the identified mutations ( 29: 43%; and 114: 13%); thus these are considered as SNPs. The incidence of the other three heterozygous mutations ( 37, 123, and 138) lies below 1%, and they are thus considered as simple point mutations. Accurate determination of the extent of the poly-a stretch turned out to be incompatible with routine analysis and this part was therefore not integrated in the further investigations. The most frequent SNP ( 29) was integrated into further analysis of patients under- 448 Wunsch et al. SNPs in the human FSHR promoter region Vol. 84, No. 2, August 2005

4 FIGURE 1 Core promoter region of the human follicle stimulating hormone receptor (FSHR). White arrows indicate positions of single-nucleotide polymorphisms. Numbers indicate base positions with respect to the translation initiation site (dark arrow). 123 is located in the E-box. Italicized letters indicate the translated region. going IVF stimulation protocols, so that in total 202 samples from Westphalian women were screened by Taqman analysis. The distribution is as follows: G/G (55.4%), G/A (37.6%), and A/A (6.9%) (Fig. 2); the population is in Hardy-Weinberg equilibrium. The DNA samples of 46 female patients were analyzed for the SNP at position 114: 87% (n 40) were homozygous T/T and 13% (n 6) homozygous C/C (no heterozygous patients). Ethnic Differences To investigate whether there are ethnic differences concerning the SNPs in the promoter region, DNA samples of 55 Indonesian women were analyzed. The investigation was restricted to the SNP at position 29 because, as the most common SNP, it was considered most relevant and, considering the small patient population available, an informative result could not be expected for the other SNPs. Interestingly, a different distribution pattern was found compared with the Caucasian population. Being in perfect Hardy- Weinberg equilibrium, the Indonesian women showed the following (mendelian) repartition: G/G (n 16; 29%), G/A (n 27; 49%), and A/A (n 12; 22%) (Fig. 2), in contrast to the nonmendelian distribution observed before. Correlation With Clinical Parameters A correlation analysis was performed to detect whether the SNP at position 29, as the most frequent one, may affect clinical parameters in patients undergoing ovarian hyperstimulation for IVF treatment. No significant correlation of the SNP in the promoter region with either basal FSH or peak estradiol levels could be found (Table 1). Two SNPs in exon 10 (Thr307Ala and Asn680Ser) of the FSHR gene have been described to influence basal FSH levels (5). Both SNPs are linked, so that analysis can be limited to one position. A first step in the investigation of a possible combined effect of the SNP in the promoter region and the coding region was genetic linkage analysis: The SNP 29 was not linked to the SNP Asn680Ser in exon 10 of the FSHR (Table 2). FIGURE 2 Distribution of the single-nucleotide polymorphism (SNP) at position 29. German (n 202) and Indonesian (n 55) women were analyzed. Fertility and Sterility 449

5 TABLE 1 Basal FSH levels and E 2 levels subdivided according to the SNP-29 (n 202). G/G G/A A/A FSH (U/L) a 7.0 ( ) 7.1 ( ) 5.8 ( ) E 2 (pmol/l) a 5,770 (120 22,317) 6,077 (445 27,929) 5,947 (1,023 15,764) a Data are shown as median and range (P not significant, Kruskal Wallis test). For haplotype analysis, patients were divided into 9 subgroups (all possible combinations of the 29 and Asn680Ser alleles) to see if the SNP at position 29 influences the known effect of exon 10 SNPs on basal FSH levels. Although the different subgroups were too small to find a statistically significant effect, we could observe the trend that FSH levels in patients with homozygous 29 A/A were lower than in the other patients (Table 2). None of the analyzed patients showed any phenotypical effect of the mutations or SNPs. Promoter Constructs In vitro experiments were performed to study the potential influence of the promoter SNPs on FSHR expression. Therefore, promoter constructs comprising the different SNPs were transfected into three different cell lines: the murine Sertoli cell line SK-11, the porcine granulosa cell line JC410, and COS-7 cells. Those with mutations at positions 37 and 138 showed a significantly stronger induction of luciferase expression than the wildtype promoter (Fig. 3). SNPs 29 and 114, however, had no significant effect on luciferase expression. We used constructs with a mutation in the E-box (position 123, see Fig. 1) as a control for our experiments; as expected (9), the point mutation caused reduced luciferase expression. Thus, dual luciferase reporter assays revealed that expression of the FSHR is influenced by polymorphisms in the promoter region. Although we used target (SK-11 and JC410) and nontarget cells (COS-7) for the FSHR, we could not observe any cell-specific effects. EMSA To investigate the possible effect of the SNPs on the binding of transcription factors, electrophoretic mobility shift assays (EMSA) were performed. We compared DNA-protein binding patterns of oligonucleotides containing the different SNPs to wildtype oligonucleotides. Different nuclear extracts (NE) isolated from COS-7, SK-11 cells, JC-410 cells, COV434 cells, primary Sertoli cells, and granulosa cells were incubated with the labeled oligonucleotides. Minor differences between wt and polymorphic oligonucleotides could be observed in some combinations of 37 and 123 oligonucleotides with SK-11, JC410, and primary granulosa cell nuclear extracts (not shown), but the most striking discrepancy in binding patterns appeared with 29 and 138 oligonucleotides. Besides one band appearing weaker, the change from G to A at position 29 caused a new DNA-protein complex in COS-7 cells (Fig. 4, upper arrow). The SNP at position 138 leads to stronger binding of proteins (Fig. 5) in primary granulosa cells, possibly suggesting the presence of enhancers upregulating FSHR expression. TABLE 2 Association of SNP-29 with SNP Asn680Ser and FSH levels (IU/ml). G/G a G/A a A/A a Total Asn/Asn b n (%) 29 (14.4%) 25 (12.4%) 7 (3.5%) FSH c 5.8 ( ) 6.0 ( ) 6.1 ( ) 61 (30.2%) Asn/Ser b n (%) 51 (25.2%) 33 (16.3%) 3 (1.5%) FSH c 7.1 ( ) 7.8 ( ) 5.5 ( ) 87 (43.1%) Ser/Ser b n (%) 32 (15.8%) 18 (8.9%) 4 (2.0%) FSH c 7.3 ( ) 7.9 ( ) 5.2 ( ) 54 (26.7%) Total n (%) 112 (55.4%) 76 (37.6%) 14 (6.9%) 202 (100%) a SNP at position 29. b SNP in exon 10. c Data are shown as median and range (P not significant, Kruskal Wallis test). 450 Wunsch et al. SNPs in the human FSHR promoter region Vol. 84, No. 2, August 2005

6 FIGURE 3 Promoter activity in JC-410/COS-7/SK-11 cells transfected with pgl3 promoter/reporter gene constructs. JC-410/COS-7/SK-11 cells were transfected with luciferase-plasmids containing a wild-type promoter or a promoter comprising one single-nucleotide polymorphism (SNP). Bars show relative luciferase activity SD (n 3). *Significant difference to the respective wt promoter (P.001, one-way ANOVA). c-ets-1 We used the TRANSFAC database to predict putative binding sites for transcription factors within the polymorphic region of the FSHR promoter region. It was observed that the SNP (G/A) at position 29 is located in a consensus sequence (GGAA) for the c-ets-1 transcription factor. To ascertain whether this transcription factor is responsible for the DNA-protein complexes observed in the EMSAs with the 29 oligonucleotides, EMSA-supershift experiments using an antibody against c-ets-1 were performed. However, the results revealed no supershift complexes (not shown), indicating that c-ets-1 does not bind to the oligonucleotides; therefore this factor probably does not regulate promoter activity of the FSHR gene. DISCUSSION The human FSHR core promoter region differs strongly from the rat, ovine, or murine promoters, which have been thoroughly investigated. It lacks conventional TATA and CAAT sequences and is devoid of CpG dinucleotide methylation sites known to influence the expression of the rat FSHR (23). In contrast to the many elements (E2F element, GATA site, Ap-1 site, SF-1 like binding sites) discovered in rat, ovine, and murine promoters (9 11, 13), no functional elements, besides the conserved E box and InR1 elements, have been described in the human FSHR so far. The human promoter seems to have evolved further from, e.g., the rodents promoter sequences and it can be speculated that it has transcriptional mechanisms other than its nonhuman homologues. A systematic study of human promoter sequences has shown that in 35% of the genes SNPs can be found, and it is suggested that a third of promoter variants may alter gene expression to a functionally relevant extent (24). In this study we gained new insights into the FSHR promoter by characterizing SNPs identified, for the first time, in the human gene. Interestingly, all SNPs were allocated to the proximal part (up to position 138) of the core promoter region ( 250 bp), and not in the distal part, which is known to be needed for full promoter activity (2). More upstream parts were not investigated because of their unclear contribution to promoter activity. As the polymorphisms were discovered in genomic DNA extracted from patients undergoing controlled ovarian hyperstimulation for IVF treatment, we chose the most common SNP ( 29, G A) to perform a clinical characterization. As the FSH levels are known to be influenced by SNPs in exon 10 of the FSHR (4 6), we retrospectively analyzed FSH levels on day 3 of the menstrual cycle and peak E 2 levels at the time of ovulation induction. Furthermore, we investigated the correlation between the promoter SNP 29 and Asn680Ser in the coding region. No significant effect on the parameters analyzed Fertility and Sterility 451

7 FIGURE 4 Competition electrophoretic mobility shift assays (EMSA) with oligonucleotides comprising the single-nucleotide polymorphism (SNP) at position 29 and nuclear extracts from COS-7 cells (C). In lanes marked wt, wild-type oligonucleotides were used; in lanes marked 29, mutated oligonucleotides were used. For competition analysis, unlabeled oligonucleotides were added as cold competitor in 100 excess. Arrows indicate DNA-protein complexes. One representative experiment (out of three) is shown. Isolated statistical significance is not necessarily a guarantee for association; neither does a lack of formal significance exclude the possibility of an association. We analyzed DNA samples of Indonesian women to investigate ethnic differences, and the determination of SNP 29 revealed a mendelian distribution pattern strikingly different from that of German women, suggesting a putative ethnic variability in the incidence of the SNP as described for other polymorphisms (18). Functional characterization was performed using promoter constructs and synthetic oligonucleotides. Luciferase experiments showed that, besides the expected effect of the mutated E box (9), obviously only the mutations at positions 37 and 138 influence the expression of the FSHR. It is noteworthy that these two base changes increase the expression, in contrast to the majority of mutations found in promoter regions. Although the FSHR is, in vivo, only expressed in Sertoli and granulosa cells, the observed effects showed no cell specificity and were present in all investigated cell lines. The increased induction of luciferase expression could be due either to silencer proteins that no longer bind because of the point mutation or to a new binding site being created where transcription factors can enhance expression. Considering the stronger binding of proteins to 138 SNP oligonucleotides, the latter option must be favored. Interestingly, the increased activity of the FIGURE 5 Competition electrophoretic mobility shift assays (EMSA) with oligonucleotides comprising the single-nucleotide polymorphism (SNP) at position 138 and nuclear extracts from human primary granulosa cells (GC). In lanes marked wt, wildtype oligonucleotides were used; in lanes marked 138, mutated oligonucleotides were used. For competition analysis, unlabeled oligonucleotides were added as cold competitor in 100 excess. Arrows indicate DNA-protein complexes. One representative experiment (out of three) is shown. could be detected, indicating that this SNP has little impact on FSHR expression and gonadal function. Although we observed a trend that patients homozygous A/A had slightly lower FSH levels, the subgroups created for haplotype analyses were too small, owing to the low incidence of A/A, to find significant differences. According to Ioannidis et al. (25) studies analyzing genetic associations have to be considered with prudence, as they may require cautious replication. 452 Wunsch et al. SNPs in the human FSHR promoter region Vol. 84, No. 2, August 2005

8 mutated 37 promoter construct is not reflected by a differential binding pattern in EMSA experiments. Database searches revealed no clear binding site around position 37, so that a new transcription factor might be involved or the SNP may allow a better binding of the transcription initiation complex. We excluded c-ets-1 as a transcription factor responsible for some of the observed effects, but which factors indeed influence expression needs further investigations. FSH stimulation could possibly alter the influence of the mutations on the promoter activity. This eventuality should be clarified in further experiments. In conclusion, here we described and characterized novel SNPs discovered in the promoter region of the human FSHR gene. The most common of the newly identified polymorphisms ( 29) in the core promoter region of the FSHR does not seem to influence clinical parameters. Only SNPs with a very low incidence affect the FSHR expression in vitro. In the future, it could be useful to screen for these mutations because they might help to explain some cases of idiopathic infertility. In our study, the patients were selected for male or tubal factor infertility and therefore showed no characteristic phenotype. The decrease of FSHR expression caused by the E box mutation ( 123) could account for an ovarian resistance to FSH. Considering that the mutations at positions 38 and 137 possibly upregulate the FSHR expression, one could speculate that they might be the underlying reason for some cases of ovarian hyperstimulation syndrome. Acknowledgments: The authors thank Nicole Terwort and Elisabeth Pekel for their technical assistance. The porcine granulosa cell line JC-410 was a generous gift of J. Chedrese (Saskatchewan, Canada). The language editing of S. Nieschlag is gratefully acknowledged. REFERENCES 1. Simoni M, Gromoll J, Nieschlag E. The follicle-stimulating hormone receptor: biochemistry, molecular biology, physiology, and pathophysiology. Endocr Rev 1997;18: Gromoll J, Dankbar B, Gudermann T. Characterization of the 5= flanking region of the human follicle-stimulating hormone receptor gene. Mol Cell Endocrinol 1994;102: Gromoll J, Pekel E, Nieschlag E. The structure and organization of the human follicle-stimulating hormone receptor (FSHR) gene. Genomics 1996;35: De Castro F, Ruiz R, Montoro L, Perez-Hernandez D, Sanchez-Casas PE, Real LM, et al. Role of follicle-stimulating hormone receptor Ser680Asn polymorphism in the efficacy of follicle-stimulating hormone. Fertil Steril 2003;80: Perez-Mayorga M, Gromoll J, Behre HM, Gassner C, Nieschlag E, Simoni M. Ovarian response to follicle-stimulating hormone (FSH) stimulation depends on the FSH receptor genotype. J Clin Endocrinol Metab 2000;85: Sudo S, Kudo M, Wada S, Sato O, Hsueh AJ, Fujimoto S. Genetic and functional analyses of polymorphisms in the human FSH receptor gene. Mol Hum Reprod 2002;8: Laven JS, Mulders AG, Suryandari DA, Gromoll J, Nieschlag E, Fauser B, et al. Follicle-stimulating hormone receptor polymorphisms in women with normogonadotropic anovulatory infertility. Fertil Steril 2003;80: Heckert LL, Daggett MA, Chen J. Multiple promoter elements contribute to activity of the follicle-stimulating hormone receptor (FSHR) gene in testicular Sertoli cells. Mol Endocrinol 1998;12: Goetz TL, Lloyd TL, Griswold MD. Role of E box and initiator region in the expression of the rat follicle-stimulating hormone receptor. J Biol Chem 1996;271: Kim JS, Griswold MD. E2F and GATA-1 are required for the Sertoli cell-specific promoter activity of the follicle-stimulating hormone receptor gene. J Androl 2001;22: Xing W, Sairam MR. Characterization of regulatory elements of ovine follicle-stimulating hormone (FSH) receptor gene: the role of E-box in the regulation of ovine FSH receptor expression. Biol Reprod 2001;64: Xing W, Sairam MR. Role of CACC-box in the regulation of ovine follicle-stimulating hormone receptor expression. Biol Reprod 2001;65: Levallet J, Koskimies P, Rahman N, Huhtaniemi I. The promoter of murine follicle-stimulating hormone receptor, functional characterization and regulation by transcription factor steroidogenic factor 1. Mol Endocrinol 2001;15: Xing W, Sairam MR. Cross talk of two Krupple transcription factors regulates expression of the ovine FSH receptor gene. Biochem Biophys Res Commun 2002;295: Simoni M, Gromoll J, Hoppner W, Kamischke A, Krafft T, Stahle D, et al. Mutational analysis of the follicle-stimulating hormone (FSH) receptor in normal and infertile men: identification and characterization of two discrete FSH receptor isoforms. J Clin Endocrinol Metab 1999; 84: Conway GS, Conway E, Walker C, Hoppner W, Gromoll J, Simoni M. Mutation screening and isoform prevalence of the follicle stimulating hormone receptor gene in women with premature ovarian failure, resistant ovary syndrome and polycystic ovary syndrome. Clin Endocrinol (Oxf) 1999;51: Gromoll J, Brocker M, Derwahl M, Hoppner W. Detection of mutations in glycoprotein hormone receptors. Methods 2000;21: Simoni M, Nieschlag E, Gromoll J. Isoforms and single nucleotide polymorphisms of the FSH receptor gene: implications for human reproduction. Hum Reprod Update 2002;8: Walther N, Jansen M, Ergun S, Kascheike B, Ivell R. Sertoli cell lines established from H-2Kb-tsA58 transgenic mice differentially regulate the expression of cell-specific genes. Exp Cell Res 1996;225: Chedrese PJ, Rodway MR, Swan CL, Gillio-Meina C. Establishment of a stable steroidogenic porcine granulosa cell line. J Mol Endocrinol 1998;20: Zhang H, Vollmer M, De Geyter M, Litzistorf Y, Ladewig A, Durrenberger M, et al. Characterization of an immortalized human granulosa cell line (COV434). Mol Hum Reprod 2000;6: Schlatt S, de Kretser DM, Loveland KL. Discriminative analysis of rat Sertoli and peritubular cells and their proliferation in vitro: evidence for follicle-stimulating hormone-mediated contact inhibition of Sertoli cell mitosis. Biol Reprod 1996;55: Griswold MD, Kim JS. Site-specific methylation of the promoter alters deoxyribonucleic acid-protein interactions and prevents folliclestimulating hormone receptor gene transcription. Biol Reprod 2001;64: Hoogendoorn B, Coleman SL, Guy CA, Smith K, Bowen T, Buckland PR, et al. Functional analysis of human promoter polymorphisms. Hum Mol Genet 2003;12: Ioannidis JP, Ntzani EE, Trikalinos TA, Contopoulos-Ioannidis DG. Replication validity of genetic association studies. Nat Genet 2001;29: Fertility and Sterility 453