Mol. Hum. Reprod. Advance Access originally published online on March 25, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Molecular Human Reproduction, Vol. 10, No. 6, pp. 373-381, 2004
© European Society of Human Reproduction and Embryology 2004
Increased expression of the FIGLA transcription factor is associated with primordial follicle formation in the human fetal ovary
MRC Human Reproductive Sciences Unit, Centre for Reproductive Biology, University of Edinburgh Chancellor’s Building, 49 Little France Crescent, Edinburgh EH16 4SB, UK
1 To whom correspondence should be addressed. e-mail: r.bayne{at}hrsu.mrc.ac.uk
 |
ABSTRACT |
|---|
 Top ABSTRACT IntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
The process of primordial follicle formation is central to the determination of a woman’s reproductive lifespan, and in humans occurs towards the end of mid-gestation. Gene knockout analysis in the mouse has shown that Figla, a transcription factor specifically expressed in germ cells, is essential for oocytes to survive and form primordial follicles. Our objective was to investigate whether a human homologue present in the genome database plays a similar role in human ovary development. Standard and real-time RT–PCR demonstrated that the human FIGLA gene is expressed in the fetal ovary but not by a range of other tissues, and that expression increases across mid-gestation, rising some 40-fold by the time of primordial follicle formation. The entire coding sequence was cloned and new exonic sequences identified. Electrophoretic mobility shift assays with in vitro-expressed human FIGLA protein showed that, as in the mouse, FIGLA can heterodimerize with E12 protein and bind to the E-box of the human ZP2 promoter. Similar mobility shifts were identified in human fetal ovary extracts. These results suggest that FIGLA is involved in continued oocyte survival as primordial follicles form in the human as in the rodent ovary.
Key words: FIGLA transcription factor/folliculogenesis/human fetal ovary/oocyte/primordial follicle
|
Introduction |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
The determination of primordial follicle formation and the associated maintenance of primordial follicle arrest are central to reproductive lifespan, yet little is known about the growth and survival factors involved, and their interactions. The number of follicles in the mammalian ovary is determined in the immediate postnatal period in the rodent and during mid-gestation in humans (Baker, 1963). Primordial germ cells migrate to the fetal gonad and proliferate by mitosis in both sexes (Byskov, 1986). In the male, sexual differentiation is achieved by sex-specific expression of genes such as SRY, leading to the development of the spermatic cords, which are recognizable in the human from 6 weeks of fetal life (Tilmann and Capel, 1999, 2002). A different pattern of development is followed in the female: germ cells continue to proliferate by mitosis until relatively late in gestation, with the definitive structure of the ovary—the primordial follicle—only forming from
18 weeks gestation. Germ cells have by then entered meiosis, arrested in prophase of meiosis I, and become surrounded by pre-granulosa cells. Those germ cells which do not form follicles will fail to survive (Byskov, 1986; McLaren, 1991). Oocytes will remain in meiosis I until reactivated in the final stages of development leading up to ovulation, possibly decades later. Primordial follicle formation involves an interaction between two cell types: oocytes and somatic pre-granulosa cells. As in later stages of follicle development, both cell types are believed to play an active role in the process (Kierszenbaum and Tres, 2001; Eppig et al., 2002; Kezele et al., 2002; Matzuk et al., 2002). A number of factors have been demonstrated to be important for cell–cell interactions and cell survival at various stages of the process of transition from primordial germ cell to primordial follicle. These include kit ligand and its receptor c-KIT (Manova et al., 1993; Richards, 2001; Robinson et al., 2001; Klinger and De Felici, 2002), Wnt-4 (Vainio et al, 1999) and evidence is emerging of the involvement of others such as the neurotrophins (Dissen et al., 1995; Ojeda et al., 2000; Anderson et al., 2002; Spears et al., 2003). In the mouse, a number of genes have been identified whose expression is restricted to the oocyte (Amleh and Dean, 2002; Epifano and Dean, 2002; Canning et al., 2003). Figla (Factor in the germline alpha), a basic helix-loop-helix (bHLH) transcription factor, was originally identified through its involvement in the expression of the genes encoding zona pellucida proteins ZP1, ZP2 and ZP3 (Liang et al., 1997). However, knockout of the Figla gene yielded female mice in which embryonic gonadogenesis appeared normal, but around birth a massive depletion of oocytes occurred and primordial follicles failed to form (Soyal et al., 2000). Expression of Figla in the mouse ovary increases markedly during development with a peak immediately after birth and only a low level of expression continuing thereafter, associated with the expression of zona pellucida genes (Soyal et al., 2000). These results are consistent with a crucial role for Figla in controlling genes required for primordial follicle formation and/or oocyte survival at that time.
bHLH transcription factors belong to one of several classes (Massari and Murre, 2000). The Class I bHLH proteins include HEB, E2-2 and E2a proteins E12 and E47. These fairly ubiquitous bHLH proteins form functional transcription factors by dimerization, usually with tissue-specific Class II bHLH proteins which show preference for which Class I protein(s) they bind (Murre et al., 1989). These heterodimers then bind to DNA at the ‘E-box’. The canonical sequence of the E-box is CANNTG. This sequence will occur fairly frequently at random in DNA but functional E-boxes are usually found in proximal promoter regions within the first 200 bp of the transcription start site. Activation of mouse Zp genes by FIGLA is achieved by heterodimerization with E12 and subsequent binding to the E-box in each of the three Zp gene promoters (Liang et al., 1997).
We have investigated the expression of FIGLA in the developing human ovary and in mature oocytes. The structure of the gene has been determined, and found to differ from that predicted from computer analysis (Huntriss et al., 2002). Recombinant human FIGLA protein is shown to heterodimerize with E12 and bind the E-box of the human ZP2 promoter. Similar oocyte-specific E12-containing complexes, capable of binding the ZP2 promoter in vitro, are demonstrated to be present in human fetal ovary specimens at the developmental stage at which primordial follicles begin to appear.
|
Materials and methods |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Tissues
Human fetal ovaries and other organs were obtained following medical termination of pregnancy. Women gave written consent according to national guidelines (Polkinghorne, 1989) and the study was approved by the Lothian Paediatrics/Reproductive Medicine Research Ethics Sub-Committee. Termination of pregnancy was induced by treatment with mifepristone (200 mg orally) followed by misoprostol (Pharmacia, UK) 800 mg 3-hourly per vaginam. None of the terminations were for reasons of fetal abnormality, and all fetuses appeared morphologically normal. Gestational age was determined by ultrasound examination prior to termination and confirmed by subsequent direct measurement of foot length. Organs were dissected free and snap-frozen on dry ice or stored in RNA Later (Ambion, UK) prior to protein or RNA extraction. Human adult oocytes at germinal vesicle stage, which had been stripped of their attached cumulus granulosa cells prior to ISCI but which were deemed too immature for clinical use, were snap-frozen in liquid nitrogen.
Isolation of RNA and synthesis of cDNA
Total RNA was extracted using either TRI-Reagent (Sigma, UK) according to the manufacturers’ instructions for organs or the Qiagen Rneasy Mini kit and a Qiashredder (Qiagen Ltd, UK) for single oocytes. Due to the tiny amounts of material, adult oocyte RNA was precipitated using Pellet Paint (Merck Biosciences Ltd, UK) and resuspended in a volume of 8 µl. First strand cDNA synthesis was performed on the entire RNA preparation using a First Strand cDNA Synthesis kit (Amersham Biosciences, UK) with p(dN)6 primers in a final volume of 15 µl (Young et al., 1998). In order to remove any contaminating genomic DNA from the larger yields of RNA from organs, 3 µg of total RNA was DNase-treated using Amplification grade DNase-1 (Invitrogen, UK) according to the manufacturer’s instructions. The RNA was primed for reverse transcription with oligo(dT) (Roche Diagnostics Ltd, UK) at 65°C for 10 min. The entire reaction was added to a total volume of 57 µl containing dNTP to 1 mmol/l, dithiothreitol (DTT) to 10 mmol/l, 1xExpand Reverse Transcriptase (RT) buffer and 120 IU RNasin Ribonuclease Inhibitor (Promega Ltd, UK). One-third (19 µl) of this reaction was added to 1 µl water (RT– reaction), which acted as a negative control to establish the efficacy of the DNase treatment. One hundred units (IU) of Expand Reverse Transcriptase (Roche Diagnostics Ltd) was added to the remaining 38 µl (RT+ reaction) and both reactions were incubated for 1 h at 42°C. Reactions were stored at –70°C until required.
RT–PCR
Target-specific PCR was performed using 1 µl of the RT+ or RT– cDNA synthesis reactions, 2 µl of oocyte cDNA or H2O as template in a reaction volume of 25 µl containing 1.5 IU of Larova PCR-Flex DNA polymerase (VHBio, UK), 2.5 µl 10xreaction buffer, MgCl2 to a concentration of 1.5 mmol/l, dNTP to 200 µmol/l, appropriate forward and reverse primers to 500 nmol/l. PCR primer sequences and specific reaction conditions were as indicated in Table I. General reaction conditions were 95°C for 2 min followed by 35 cycles of 94°C for 20 s/30 s at the appropriate annealing temperature/72°C for the appropriate extension time and a final cycle of 72°C for 10 min. The identity of all PCR products was confirmed by direct sequencing using an Applied Biosystems Prism 310 Genetic Analyser.
|
Real-time quantitative PCR
Quantitative real-time RT–PCR was performed using the Lightcycler (Roche Diagnostics Ltd) as described previously (Hartley et al., 2002). GAPD levels were used for normalization because they are not expected to vary across the gestational range studied. Although no published data are available on fetal ovaries to confirm this assumption, we have found that the majority of genes analysed by real-time PCR relative to GAPD do not vary significantly across mid-gestation. In addition, when cDNA arrays are probed with first-strand cDNA from early and late mid-gestation ovaries, GAPD levels show very similar relative levels to those of several other housekeeping genes, including ribosomal proteins. Reverse-transcribed RNA samples (n = 3–5 for each gestation) were diluted in water as indicated below. One microlitre of diluted first-strand cDNA was added to a final volume of 10 µl containing 2.5 mmol/l MgCl2 and 0.5 µmol/l each of forward and reverse primer in 1xLightcycler Fast Start DNA MasterSYBR Green 1 Master Mix (Roche Diagnostics Ltd). Amplification was continued for 45 cycles with signal acquisition at 84°C after each round of extension. Following amplification, continuous melt curve analysis was performed to ensure product accuracy and samples were analysed by agarose gel electrophoresis (data not shown) to confirm product size. Primers for GAPD are given in Table I, hFigaF2 and hFigaR2 were used for FIGLA.
Standard curves for GAPD and FIGLA were derived by making a series of dilutions (1/10, 1/20, 1/25, 1/40, 1/50, 1/80 1/100, 1/1000) of first-strand cDNA from a 19 week ovary. The number of cycles needed to yield a fluorescent signal above background (the cross-over point, Cp) at each dilution was plotted against the log of relative concentration using LightCycler Software (Molecular Dynamics Ltd, UK). The dilutions yielded a straight line for each product, confirming that Cp is a good indicator of target concentration across at least 2 orders of magnitude. The slopes of these curves are a measure of the efficiency of the PCR, which gave an amplification rate of 1.94-fold/cycle for GAPD and 1.79-fold/cycle for FIGLA. For quantification, ovarian cDNA was used at 1/25 dilution. For each experiment, both GAPD and FIGLA amplification reactions were performed in duplicate for every cDNA sample used. Calculations for FIGLA mRNA concentration were made relative to GAPD from the same sample to allow comparisons between ovaries. Allowance for differences in amplification rate for GAPD and FIGLA was achieved by determining the actual amount of amplification required to yield a signal for each target.
Construction of pCRFIGLA and pCRE12 mammalian expression vectors
Primers (FigaF1/FigaR1 and hE12F1/hE12R1, Table I) were designed to cover the entire predicted FIGLA or E12 coding regions as well as translation initiation sequences and a few base pairs downstream of the predicted STOP codons. These primers were used to PCR amplify the FIGLA or E12 gene from first strand fetal ovary cDNA and the products were then cloned into pCR3.1 using the Eukaryotic Bidirectional TA cloning Kit (Invitrogen, UK). After transformation into TOP10F’ E.coli, colonies containing FIGLA or E12 sequences in the correct orientation for expression were identified by restriction analysis of plasmid minipreps. The sequence of the entire insert from a number of such plasmids was determined by sequence analysis in an ABI Prism 310 Genetic Analyser using vector primers and the FigaF2 and FigaR2 or hE12F2 and hE12R2 internal primers. The 697 bp FIGLA PCR product was also sequenced directly.
In vitro production of FIGLA and E12 proteins
FIGLA and E12 proteins were generated in vitro using coupled transcription and translation with the TNT T7 Coupled Wheat Germ Extract System (Promega Ltd, UK) and pCRFIGLA, pCRE12-102 or pCRE12-104. Plasmids were linearized at the 3'-end of the gene insert with XhoI (pCRFIGLA) or XbaI (pCRE12 constructs) to allow transcription from the T7 promoter contained in the vector. Blank extracts in which no DNA was added were prepared as negative controls and to keep volumes of extracts between assays constant. Although extracts for gel mobility shift assays were unlabelled, control reactions for each protein were labelled with L-[35S]methionine (Amersham Biosciences, UK) according to the manufacturer’s protocol and 5 µl samples analysed by sodium dodecyl sulphate–polyacrylamide gel electrophoresis on a 10% resolving gel. Gels were fixed in 50% methanol, 7% acetic acid for 15 min and dried at 80°C under vacuum in a gel-drier (BioRad Laboratories, UK) before direct autoradiography onto Kodak XAR film for 1 h. In each case, single bands of the expected size were produced (data not shown).
Preparation of ovary and testis protein extracts
Ovary and testis whole cell extracts from 18 week gestation fetuses were prepared by homogenization on ice in non-denaturing extraction buffer [10 mmol/l HEPES, pH 7.9; 400 mmol/l NaCl; 0.5 mmol/l DTT; 0.1 mmol/l EDTA; 5% glycerol; 0.625% Nonidet P-40 containing 1 mini-Complete protease inhibitor cocktail tablet (Roche Diagnostics Ltd) per 10 ml buffer] followed by brief centrifugation to remove insoluble cell debris. Nuclear extracts of fetal ovary and testis were prepared by homogenizing frozen half-organs of 18 weeks gestation in 400 µl ice-cold Buffer A (10 mmol/l HEPES, pH 7.9; 10 mmol/l KCl; 0.1 mmol/l EDTA; 0.1 mmol/l EGTA; 1 mmol/l DTT and containing 1 mini-Complete protease inhibitor cocktail tablet per 10 ml buffer), incubating on ice for 15 min and adding 25 µl of 10% NP-40 in Buffer A, vortexing quickly for 10 s and centrifuging at 12 000 g for 45 s. Cytosolic supernatants were removed and the clear glassy nuclear pellets were resuspended in 50 µl Buffer B (20 mmol/l HEPES, pH 7.9; 400 mmol/l NaCl; 1 mmol/l EDTA; 1 mmol/l EGTA; 1 mmol/l DTT and containing 1 mini-Complete protease inhibitor cocktail tablet per 10 ml buffer) without pipetting, by shaking on ice for 15 min and then centrifuged at 12 000 g for 5 min at 4°C to remove debris before snap-freezing. Protein concentrations were determined with BCA Protein assay reagent (Pierce, USA).
Electrophoretic mobility shift assays
Single-stranded oligonucleotides (MWG Biotech UK Ltd, UK) corresponding to each strand of the sequence centring around the E-box of the human ZP2 promoter (Table I) were annealed to generate double-stranded (ds) oligonucleotides which had a single G overhang at each 5' end. Identical ds-oligonucleotides in which the E-box sequence CANNTG had been mutated to AGATCT were also generated (Table I). This enabled radiolabelled probes to be synthesized by fill-in reactions using the Klenow fragment of DNA Polymerase 1 (Roche Diagnostics Ltd) and [
-32P]dCTP (Amersham Biosciences, UK). All oligonucleotides were labelled to equivalent specific activities and 105 c.p.m. of labelled oligonucleotide was used per binding assay. Binding reactions were performed in 25 µl volumes in binding buffer [50 mmol/l NaCl; 10 mmol/l Tris–HCl, pH 7.5; 5 mmol/l MgCl2; 0.5 mmol/l DTT; 1 mmol/l EDTA; 5% (v/v) glycerol; 0.01% Nonidet P-40] with 2 µg poly(dI.dC)(dI.dC) (Amersham Biosciences) and 5 µl of each in vitro TNT extract to be tested (made up to 10 µl with blank extract where necessary) or with 10 µg total ovary/testis protein or 1 µg ovary nuclear extract made up to 10 µl in extract buffer. Unlabelled competitor ds-oligonucleotides were also added at this stage where appropriate. Reactions were preincubated on ice for 10 min before addition of labelled oligonucleotide probe and incubation for a further 30 min on ice. For antibody super-shift assays, 2 µg Transcruz anti E2a.E12 (V-18) antibody (sc-349X; Santa Cruz Biotech. Inc., USA) was added after the 10 min preincubation step and allowed to bind to the protein complexes on ice for 15 min before continuing as above. After the addition of Loading Dye [final concentration 25 mmol/l Tris–HCl, pH 7.5; 4% (v/v) glycerol; 0.02% Bromophenol Blue], binding reactions were run on a 5 or 6% acrylamide/2.5% glycerol/0.5xTBE gel which had been pre-run at 150 V for 1 h prior to loading. The gel was run in 0.5xTBE at 300 V until the marker dye had migrated about three-quarters of the way down the gel. The gel was transferred onto 3 mm paper, dried at 80°C in a gel drier (BioRad Laboratories, UK) and exposed for autoradiography to Kodak XAR film at –70°C with intensifying screens for 16 h.
|
Results |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Identification and structure of the human FIGLA gene
The coding sequence of the mouse Figla gene was used to perform a BLAST search of the human public domain databases in order to identify a human homologue. A single genomic BAC clone (RP11-50401, Accession No. AC007395) from chromosome 2p12 was identified. This had been annotated as having a sequence with similarity to mouse Figla and the predicted coding region (Accession No. XM_292886) was obtained. Primers (hFigaF2 and hFigaR2, Table I) covering regions of high similarity to the mouse sequence were designed for RT–PCR. A product of the predicted size (300 bp) was obtained from human fetal ovary specimens over the gestational range of 14–19 weeks (Figure 1a). FIGLA expression was also demonstrated in individual adult germinal vesicle stage human oocytes (Figure 1b). These products were sequenced and shown to encode the human homologue of FIGLA. In order to investigate whether FIGLA expression in the human fetus is restricted to the germ-line, RNA was prepared from a number of fetal tissues of 17 weeks gestation (brain, heart, kidney, liver, lung, placenta, spleen, uterus, testis and ovary) and subjected to RT–PCR as before (Figure 1c). FIGLA expression was only detected in fetal ovary.
|
Experiments to investigate the function of FIGLA in the human fetal ovary require protein expression. The entire predicted FIGLA coding region was therefore PCR-amplified, cloned and sequenced (Figure 2a). The 5' end of the sequence matched the predicted coding sequence (Accession No. XM_292886) but differences were found towards the 3' end. However, the entire sequence of the cloned products was in agreement with that of a near full-length cDNA isolated from a human adult brain medulla library (IMAGE Clone 5744748, Accession No. BM560361) in the human EST database. We identified this cDNA as the sole match to human FIGLA sequences in a BLAST search of public domain human EST databases. This suggested that the sequence discrepancy we observed was not the result of a PCR mistake or generation of a chimeric clone. Screening for the new FIGLA sequences in the published genomic sequence (Accession No. NT_022184.12, Build 33) indicated that sequences matching the new 3' end were present in the genomic clone and that their boundaries matched consensus splice sites (Figure 2c). This suggests that the FIGLA gene is composed of five rather than four exons (Figure 2b). Exons 1 and 2 remain as published (Huntriss et al., 2002), as do sequences present in exon 3, but the use of an alternative splice donor site 71 bp downstream of the published one extends exon 3 by this length (Figure 2c). An entirely new exon 4 is found between the predicted exons 3 and 4. Exon 5 is essentially the same as the previously predicted exon 4 except that the addition of the extra exonic sequences upstream shifts the open reading frame so that an alternative STOP codon, 10 bp upstream of the previously predicted one, is utilized (Figure 2b, c).
|
The question remained as to whether this new sequence represents the only FIGLA transcript or whether alternative splicing occurs and the original predicted sequence is also represented. Primers hFigaF2 and hFigaR1 were used to perform RT–PCR on fetal ovaries and adult oocytes. The predicted size of the product based on the new FIGLA sequence is 537 bp whereas that of the sequence predicted in the database is 431 bp (Figure 2b). In both fetal ovaries and adult oocytes, only the 537 bp product was obtained (Figure 2d). This result indicates that the new FIGLA coding sequence presented here is the sole transcript.
In silico translation of the new human FIGLA sequence predicts a 219 amino acid protein of mol. wt 24 000. The amino acid sequence of this protein was compared to the sequences of FIGLA homologues from mouse (NP_036143), rat (XP_232137) and the Japanese rice fish, medaka (Oryzias latipes, AAD38902) (Figure 3). The bHLH domain is highly conserved between all species (human:mouse, 96%, human:rat, 94% and human:medaka, 54% identity across 52 amino acids). Conservation in the mammalian homologues remains high over the majority of the protein length (identity between human:mouse/rat is 68/67% respectively) although the carboxy-terminus of the human protein shows little homology to the rodent proteins. However, a TRS motif near the carboxy-terminus is present in all three mammalian homologues.
|
Expression of FIGLA increases across mid-gestation
FIGLA mRNA expression was quantified in human fetal ovary across the gestational range 14–19 weeks by real-time quantitative RT–PCR (Figure 4). Data were analysed by analysis of variance, with Student–Neuman–Keuls P-test. At 14 weeks gestation, FIGLA expression was low (0.1% of GAPD levels). Between 14 and 19 weeks there was a highly significant rise in expression (P = 0.0004). This rise first became statistically significant at 17 weeks (P < 0.01 versus 14 weeks), with further significant rises at each week interval from 17 weeks onwards (P < 0.05). By 19 weeks, expression levels reached
4.0% of GAPD levels, an increase of 40-fold.
|
0.05).Isolation and cloning of human E12 coding sequences into a eukaryotic expression vector
Mouse FIGLA functions as a heterodimer with another bHLH protein, E12 (Liang et al., 1997). It therefore seemed likely that expression of E12 would also be necessary to investigate human FIGLA function. Screening of public databases for human E12 mRNA sequences identified a single entry which contained the entire E12 coding sequence (Accession No. M31222). Two pairs of PCR primers (Table I) were designed to detect this sequence in fetal ovary. The first pair, hE12F1 and hE12R1, amplified the entire open reading frame. The second pair, hE12F2 and hE12R2, span the E12 bHLH domain, unique to E12 and not present in the alternatively spliced form E47, thus allowing a specific assay for E12 expression. RT–PCR with the latter set of primers indicated that E12 is expressed in human fetal ovary throughout the 14–19 weeks gestation period (Figure 5a). Three clones encompassing the entire open reading frame of human E12 in pCR3.1 were obtained, two in the sense orientation and one in the antisense orientation. However, sequence analysis of these clones demonstrated that 150 bp of sequence encoding two exons (3 and 4) of the published full-length E12 sequence were missing from one sense clone (pCRE12-102) and that a short 35 bp stretch of sequence just upstream of the bHLH domain was absent from the other two clones (pCRE12-104 is in the sense orientation) (Figure 5b). The end-point of this small 35 bp deletion indicates that it could be the result of the use of an alternative splice acceptor site. Both alternative splice events have been detected in EST databases. Therefore, two new splice variants of E12 have been identified, although whether these are ovary specific is not known. RT–PCR using primers within the two deleted regions indicates that a full-length version of the E12 coding sequence is transcribed in the fetal ovary (data not shown). Nevertheless, further attempts to generate a full-length E12 clone were unsuccessful. Translation of pCRE12-102 will generate a near full-length E12 protein with its bHLH domain intact. It is this region which interacts with FIGLA and binds DNA, and therefore we predicted that the protein may have sufficient function for in vitro assays. The deletion of 35 bp just upstream of the bHLH domain in pCRE12-104 causes a frame-shift which introduces a premature stop codon a short distance downstream (Figure 5b). This protein therefore lacks the bHLH domain, preventing its dimerization or DNA binding, and thus provided a negative control for subsequent experiments.
|
Human FIGLA binds to the human ZP2 promoter E-box
Although the role of Figla during mouse primordial follicle formation has been established (Soyal et al., 2000), the only target genes identified to date for this transcription factor are those encoding zona pellucida proteins ZP1, ZP2 and ZP3 (Liang et al., 1997), none of which is required until later in folliculogenesis. Nevertheless, the ability of human FIGLA to bind the human ZP2 promoter under the less stringent requirements of in vitro assays would provide a good demonstration of functional activity in the fetal ovary. We assessed whether human E12 and FIGLA can bind to the human ZP2 promoter E-box in vitro by electrophoretic mobility shift assays (EMSA). FIGLA and E12 proteins were generated in vitro. Addition of FIGLA to the human ZP2 E-box-containing probe resulted in a retarded band (Figure 6a). A second specific, more retarded band (Figure 6a) was observed only when both FIGLA and E12-102 extracts were present in the binding reaction. Neither E12-102 nor E12-104 alone produced any specific mobility shifts (Figure 6a), nor did E12-104 in combination with FIGLA, except for the higher mobility FIGLA-specific band (data not shown). A 500-fold excess of cold ZP2 E-box oligonucleotide competed out the ZP2 E-box band-shifts (Figure 6b, lane 2). To confirm the specificity of FIGLA/E12 binding to the E-box, a reaction containing a 500-fold excess of cold ZP2 oligonucleotide containing mutant E-box sequences had no effect (Figure 6b, lane 3). Similarly, the ZP2 mutant E-box, when used as the probe, was unable to form complexes with FIGLA and/or E12-102 (Figure 6b, lane 4).
|
An antibody specific to E2a (E12 and E47) was next used to super-shift FIGLA/E12 complexed with the ZP2 E-box oligonucleotide, using the in vitro-generated proteins. Similar experiments were performed using human fetal ovary or testis extracts to investigate the presence of endogenous FIGLA and E12 proteins in these human tissues. When the antibody was added to a binding reaction containing both FIGLA and E12-102 TNT extracts, the intensity of the lower mobility ZP2 E-box band-shift was reduced in intensity and a new, super-shifted band was observed even further up the gel (Figure 6c, compare lanes 1 and 2). In the presence of 500-fold excess cold specific competitor oligonucleotide, neither the original band-shift nor the super-shifted band was observed (Figure 6c, lane 3). These results confirmed that the protein complex which produces the lower-mobility shift contains E12 protein and that the antibody is capable of generating a super-shift under the conditions used. A number of bands with reduced mobility compared to free probe were observed after incubation of ZP2 E-box oligonucleotide with extracts of fetal ovary or testis (Figure 6c, lanes 4, 7, 9). In addition, a weak band of lower mobility than the FIGLA/E12 TNT band-shift was found which was clearer in nuclear than in whole cell ovary extracts, and not present in testis extracts (Figure 6c, lanes 4, 7, 9). This band was super- shifted with the anti-E2a antibody (Figure 6c, lanes 5, 8) and disappeared when 500-fold excess cold probe was added (Figure 6c, lane 6).
|
Discussion |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
The definitive structure of the mammalian ovary, the oocyte-containing follicle, is formed in humans during the mid-trimester of pregnancy. The identification of genes whose function is critical to this developmental stage is of importance both for our understanding of the regulation of follicle formation and for the investigation of conditions in which follicle formation is abnormal and results in premature ovarian failure. One such candidate gene is the human homologue of Figla (Soyal et al., 2000). Expression of FIGLA mRNA in adult human oocytes has been demonstrated previously (Huntriss et al., 2002) but no other data regarding expression or potential function in the human are available.
The human homologue of mouse Figla was identified by a BLAST search of public domain databases. RT–PCR demonstrated that FIGLA expression is restricted to the ovary, where it is expressed during both fetal development and in adult oocytes.
Comparison of the corrected human FIGLA protein sequence with its mouse, rat and medaka homologues indicates a high level of conservation across the bHLH domain for all species. Outside this domain, the medaka sequence diverges significantly from the mammalian proteins but homology between the mammalian species remains high at 80%, throughout the amino terminal of the proteins. The mouse and rat sequences remain conserved right to the carboxy terminus but the human protein diverges and has a longer C-terminus. However, within this divergent carboxy end there is a TRS motif which is conserved between all three mammalian proteins. The context of the serine residue within this motif indicates a high likelihood that it could be phosphorylated (Blom et al., 1999), allowing speculation that post-translational modification of FIGLA by phosphorylation may have functional relevance.
Expression of the FIGLA transcript was demonstrated, by quantitative RT–PCR, to increase in the ovary across mid-gestation, reaching levels some 40-fold higher at 19 weeks compared to 14 weeks gestation. This rise is particularly marked between 17 and 19 weeks, the precise time at which primordial follicles are first formed. A similar rise in Figla transcript levels is associated with the timing of primordial follicle formation in the mouse (Soyal et al., 2000) and gene knockout studies indicate that Figla is required for primordial follicle formation and/or oocyte survival at that time. Our results are consistent with a similar role in human ovary development. If the increased expression of FIGLA in the fetal ovary is confined to more mature oocytes, either entering or in the process of primordial follicle formation, the increase in expression within those cells may be enormously higher than the 40-fold increase detected, since, even at 19 weeks gestation, the majority of oocytes are not yet integrated within primordial follicles. Characterization of FIGLA expression during more advanced stages of primordial follicle formation in the human is not feasible as it is not possible to obtain ovaries from later gestations.
FIGLA function in the mouse is associated with its heterodimerization with another bHLH protein, E12 (Liang et al., 1997). We demonstrated E12 expression in the human fetal ovary by RT–PCR and cloned the open reading frame into an expression vector. Sequence analysis of our clones indicated that a number of alternative splice products are expressed in the fetal ovary. Full-length E12 transcripts do appear to be present in the fetal ovary based on RT–PCR using primers in the exons deleted in the cloned splice variants, but the relevance of different splice variants to FIGLA function is not clear at this time. Although we were unable to derive a clone containing the entire open reading frame with all published E12 exons, a clone lacking exons 3 and 4 appears to be functional in vitro, based on its ability to complex with FIGLA and allow binding to the ZP2 E-box.
The FIGLA–ZP2 E-box interaction is unlikely to be relevant physiologically in the fetal ovary, but under the reduced stringency required in vitro for EMSA, the ability of FIGLA to bind the ZP promoters allowed us to test the functionality of FIGLA in fetal ovary extracts with the only FIGLA target identified thus far. EMSA assays with human E12 and FIGLA proteins expressed in vitro demonstrated that both proteins are required to form a complex on a ZP2 E-box oligonucleotide and that the binding site requires an intact E-box. A faster-migrating (and thus smaller) complex did appear to form when FIGLA protein alone was incubated with the promoter probe containing an intact E-box, suggesting that FIGLA may bind to the E-box by itself in the absence of E12. As with the FIGLA/E12 complex, this complex was competed out with cold specific competitor DNA but not E-box mutants. However, whether FIGLA alone can form a functional transcription complex with the E-box is not known.
The FIGLA/E12 complex on the ZP2 E-box was super-shifted with an anti-E2a antibody which recognizes E12. This same antibody was also used to investigate potential E12-containing complexes in fetal ovary extracts, with fetal testis extracts acting as a negative control as FIGLA is not expressed in that tissue. An ovary-specific complex with the ZP2 promoter E-box was super-shifted with the anti-E2a antibody and competed out with cold oligonucleotide competitor DNA. This complex did not have the same mobility as that formed with in vitro-expressed E12 and FIGLA, but this may be explained by the expression of a slightly truncated E12 protein in in vitro extracts and/or the absence of post-translational modifications of these proteins in the in vitro assay that would normally occur in vivo. In light of the work on the Zp2 promoter in the mouse (Liang et al., 1997) and the ovary specificity observed here, it seems likely that this ovarian complex will also contain FIGLA although confirmation will require a FIGLA-specific antibody.
Although ectopic expression of luciferase driven by a mouse Zp2 promoter was up-regulated by co-transfection with Figla and E12 in mouse 10T1/2 cells, transcription of the Zp2 gene from the endogenous Zp2 promoter was not similarly increased in these cells, suggesting that other germ cell-specific factors are also required for expression of Zp genes in vivo (Liang et al., 1997). In addition to the explanations presented earlier for the difference in mobility of the complexes formed on the human ZP2 promoter E-box from ovary extracts and from in vitro-expressed FIGLA and E12, it is also possible that other proteins, either co-factors or inhibitors, may be present in the ovarian complexes which would affect their mobility. FIGLA up-regulates different genes at different times—expression of the ZP genes occurs after primordial follicle formation whereas genes which are involved in the processes leading up to follicle formation must be expressed earlier. In order for FIGLA to induce transcription of these different classes of genes at the correct time, other factors specific to the particular stage must be involved. Of note in this context, expression of E2A protein (i.e. E12 and E47) has been shown to be inhibited by stem cell factor (kit ligand) in B-cell precursors (Riley et al., 2002) and AKT kinase can regulate the assembly and activity of bHLH–co-activator complexes (Vojtek et al., 2003). Both of these signalling systems have been implicated in follicle formation and development (Manova et al., 1993; Kissel et al., 2000).
In conclusion, we have demonstrated the expression of the transcription factor FIGLA in the human fetal ovary and in mature adult oocytes, but not in a range of other tissues. Expression of FIGLA rises across mid-gestation. The corrected gene structure of FIGLA is presented: comparison with sequence in other species identifies a potential phosphorylation site which may be of functional significance. Human FIGLA has been shown to dimerize with the transcription factor E12 and bind to the E-box of the ZP2 promoter, as has an extract of human fetal ovary, indicating the presence of FIGLA protein in the human fetal ovary. These results are consistent with a role for FIGLA in oocyte survival as primordial follicles form in the developing human ovary, as has been demonstrated in the rodent.
|
Acknowledgements |
|---|
We are grateful to the staff of the Bruntsfield Suite and the Assisted Conception Unit, Royal Infirmary of Edinburgh, for assistance with the provision of the clinical specimens.
|
REFERENCES |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Amleh A and Dean J (2002) Mouse genetics provides insight into folliculogenesis, fertilization and early embryonic development. Hum Reprod Update 8,395–403.
Anderson RA, Robinson LL, Brooks J and Spears N (2002) Neurotropins and their receptors are expressed in the human fetal ovary. J Clin Endocrinol Metab 87,890–897.
Baker TG (1963) A quantitative and cytological study of germ cells in human ovaries. Proc R Soc Lond B 158,417–433.[Medline]
Blom N, Gammeltoft S and Brunak S (1999) Sequence and structure-based prediction of eukaryotic protein phosphorylation sites. J Mol Biol 294,1351–1362.[CrossRef][ISI][Medline]
Byskov AG (1986) Differentiation of mammalian embryonic gonad. Physiol Rev 66,71–117.
Canning J, Takai Y and Tilly JL (2003) Evidence for genetic modifiers of ovarian follicular endowment and development from studies of five inbred mouse strains. Endocrinology 144,9–12.[Abstract]
Dissen GA, Hirshfield AN, Malamed S and Ojeda SR (1995) Expression of neurotrophins and their receptors in the mammalian ovary is developmentally regulated: changes at the time of folliculogenesis. Endocrinology 136,4681–4692.[Abstract]
Epifano O and Dean J (2002) Genetic control of early folliculogenesis in mice. Trends Endocrinol Metab 13,169–173.[CrossRef][ISI][Medline]
Eppig JJ, Wigglesworth K and Pendola FL (2002) The mammalian oocyte orchestrates the rate of ovarian follicular development. Proc Natl Acad Sci USA 99,2890–2894.
Hartley PS, Bayne RA, Robinson LL, Fulton N and Anderson RA (2002) Developmental changes in expression of myeloid cell leukemia-1 in human germ cells during oogenesis and early folliculogenesis. J Clin Endocrinol Metab 87,3417–3427.
Huntriss J, Gosden R, Hinkins M, Oliver B, Miller D, Rutherford AJ and Picton HM (2002) Isolation, characterization and expression of the human Factor In the Germline alpha (FIGLA) gene in ovarian follicles and oocytes. Mol Hum Reprod 8,1087–1095.
Kezele P, Nilsson E and Skinner MK (2002) Cell-cell interactions in primordial follicle assembly and development. Front Biosci 7,d1990–1996.[ISI][Medline]
Kierszenbaum AL and Tres LL (2001) Primordial germ cell-somatic cell partnership: a balancing cell signaling act. Mol Reprod Dev 60,277–280.[CrossRef][ISI][Medline]
Kissel H, Timokhina I, Hardy MP, Rothschild G, Tajima Y, Soares V, Angeles M, Whitlow SR, Manova K and Besmer P (2000) Point mutation in kit receptor tyrosine kinase reveals essential roles for kit signaling in spermatogenesis and oogenesis without affecting other kit responses. EMBO J 19,1312–1326.[CrossRef][ISI][Medline]
Klinger FG and De Felici M (2002) In vitro development of growing oocytes from fetal mouse oocytes: stage-specific regulation by stem cell factor and granulosa cells. Dev Biol 244,85–95.[CrossRef][ISI][Medline]
Liang L, Soyal SM and Dean J (1997) FIGalpha, a germ cell specific transcription factor involved in the coordinate expression of the zona pellucida genes. Development 124,4939–4947.[Abstract]
Manova K, Huang EJ, Angeles M, De Leon V, Sanchez S, Pronovost SM, Besmer P and Bachvarova RF (1993) The expression pattern of the c-kit ligand in gonads of mice supports a role for the c-kit receptor in oocyte growth and in proliferation of spermatogonia. Dev Biol 157,85–99.[CrossRef][ISI][Medline]
Massari ME and Murre C (2000) Helix-loop-helix proteins: regulators of transcription in eucaryotic organisms. Mol Cell Biol 20,429–440.
Matzuk MM, Burns KH, Viveiros MM and Eppig JJ (2002) Intercellular communication in the mammalian ovary: oocytes carry the conversation. Science 296,2178–2180.
McLaren A (1991) Development of the mammalian gonad: the fate of the supporting cell lineage. Bioessays 13,151–156.[CrossRef][ISI][Medline]
Murre C, McCaw PS, Vaessin H, Caudy M, Jan LY, Jan YN, Cabrera CV, Buskin JN, Hauschka SD, Lassar AB et al (1989) Interactions between heterologous helix-loop-helix proteins generate complexes that bind specifically to a common DNA sequence. Cell 58,537–544.[CrossRef][ISI][Medline]
Ojeda SR, Romero C, Tapia V and Dissen GA (2000) Neurotrophic and cell-cell dependent control of early follicular development. Mol Cell Endocrinol 163,67–71.[CrossRef][ISI][Medline]
Polkinghorne J (1989) Review of the Guidance on the Research Use of Fetuses and Fetal Material. HMSO, London.
Richards JS (2001) Perspective: the ovarian follicle—a perspective in 2001. Endocrinology 142,2184–2193.
Riley RL, Knowles J and King AM (2002) Levels of E2A protein expression in B cell precursors are stage-dependent and inhibited by stem cell factor (c-kit ligand). Exp Hematol 30,1412–1418.[CrossRef][ISI][Medline]
Robinson LL, Gaskell TL, Saunders PT and Anderson RA (2001) Germ cell specific expression of c-kit in the human fetal gonad. Mol Hum Reprod 7,845–852.
Soyal SM, Amleh A and Dean J (2000) FIGalpha, a germ cell-specific transcription factor required for ovarian follicle formation. Development 127,4645–4654.[Abstract]
Spears N, Molinek MD, Robinson LL, Fulton N, Cameron H, Shimoda K, Telfer EE, Anderson RA and Price DJ (2003) The role of neurotrophin receptors in female germ-cell survival in mouse and human. Development 130,5481–5491.
Tilmann C and Capel B (1999) Mesonephric cell migration induces testis cord formation and Sertoli cell differentiation in the mammalian gonad. Development 126,2883–2890.[Abstract]
Tilmann C and Capel B (2002) Cellular and molecular pathways regulating mammalian sex determination. Recent Prog Horm Res 57,1–18.
Vainio S, Heikkila M, Kispert A, Chin N and McMahon AP (1999) Female development in mammals is regulated by Wnt-4 signalling. Nature 397,405–409.[CrossRef][Medline]
Vojtek AB, Taylor J, DeRuiter SL, Yu JY, Figueroa C, Kwok RP and Turner DL (2003) Akt regulates basic helix-loop-helix transcription factor–coactivator complex formation and activity during neuronal differentiation. Mol Cell Biol 23,4417–4427.
Young LE, Butterwith SC and Wilmut I (1998) A novel method for quantifying mRNA levels in single embryos. Theriogenology 49,192.
Submitted on December 18, 2003; resubmitted on February 16, 2004; accepted on March 7, 2004.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. E. Dickinson, M. Myers, and W. C. Duncan Novel Regulated Expression of the SLIT/ROBO Pathway in the Ovary: Possible Role during Luteolysis in Women Endocrinology, October 1, 2008; 149(10): 5024 - 5034. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Walton, R. A. L. Bayne, I. Wallace, D. T. Baird, and R. A. Anderson Direct Effect of Progestogen on Gene Expression in the Testis during Gonadotropin Withdrawal and Early Suppression of Spermatogenesis J. Clin. Endocrinol. Metab., July 1, 2006; 91(7): 2526 - 2533. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Guigon and S. Magre Contribution of Germ Cells to the Differentiation and Maturation of the Ovary: Insights from Models of Germ Cell Depletion Biol Reprod, March 1, 2006; 74(3): 450 - 458. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Pangas and A. Rajkovic Transcriptional regulation of early oogenesis: in search of masters Hum. Reprod. Update, January 1, 2006; 12(1): 65 - 76. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. K. Skinner Regulation of primordial follicle assembly and development Hum. Reprod. Update, September 1, 2005; 11(5): 461 - 471. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||