Cloning of Epidermal Growth Factor (EGF) and EGF ... - CiteSeerX

Download PDF
3 downloads 0 Views 3MB Size Report
Comment
Apr 28, 2004 - Ovary: Evidence for EGF as a Potential Paracrine Factor from the Oocyte to. Regulate Activin/Follistatin System in the Follicle Cells1. Yajun Wang and Wei .... near the 5-end of the 5-RACE product and used to amplify the full-.
BIOLOGY OF REPRODUCTION 71, 749–760 (2004) Published online before print 28 April 2004. DOI 10.1095/biolreprod.104.028399

Cloning of Epidermal Growth Factor (EGF) and EGF Receptor from the Zebrafish Ovary: Evidence for EGF as a Potential Paracrine Factor from the Oocyte to Regulate Activin/Follistatin System in the Follicle Cells1 Yajun Wang and Wei Ge2 Department of Biology, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, China ABSTRACT

the prototype of the receptor tyrosine kinase family [7]. The binding of EGFR by EGF activates its intrinsic receptor tyrosine kinase and rapidly elicits the downstream signaling cascades [8]. Interestingly, transforming growth factor a (TGFa), which shares only about 30% structural homology with EGF, also signals via EGFR with binding affinity similar to that for EGF [9]. EGF plays important roles in cell proliferation and differentiation in a variety of tissues, including the ovary. As a local paracrine/autocrine regulator, EGF production has been demonstrated in the ovarian follicles of both mammals [10–12] and birds [13]. In mammals, EGF stimulates DNA synthesis [14, 15] and granulosa cell proliferation [16] in the ovarian follicles. It also modulates ovarian steroidogenesis and granulosa cell differentiation by increasing progesterone production and inhibiting FSH-induced biosynthesis of estradiol and receptors of FSH and LH in cultured granulosa cells [17–23]. In addition, EGF enhances oocyte maturation in several mammalian species [24–26]. In the chicken, EGF stimulates granulosa and theca-interstitial cell proliferation and inhibits LH-stimulated estradiol and progesterone production [27–30]. In nonmammalian vertebrates, particularly lower vertebrates, studies on EGF and its roles in the ovary are rather limited, partly because the molecule has not been cloned in any nonmammalian species. In fish ovary, EGF may play roles in controlling follicle survival and steroidogenesis as well as DNA synthesis in the vitellogenic follicles [31–34]. Recently, we have demonstrated that EGF and TGFa are equipotent in increasing the rate of zebrafish oocyte maturation and their effects are likely mediated by activin [35], another important growth factor expressed in the ovary of all major vertebrate groups, including mammals, birds, and fish [36–41]. Activin is a member of the transforming growth factor b (TGFb) superfamily, and it is a homo- or heterodimer consisting of two distinct but related b subunits, bA and bB [42]. Lines of evidence suggest that activin is extensively involved in follicular growth, granulosa cell differentiation, oocyte maturation, and steroidogenesis in vertebrate ovaries [36, 37, 43–45]. The activities of activin are modulated by follistatin, a specific activin-binding protein that neutralizes the biological effects of activin [46, 47]. In contrast with EGF, activin stimulates FSH receptor biosynthesis in the ovary [48] and enhances FSH-induced aromatase and LH receptor expression [49–51]. Because EGF and activin are both important components of the local regulatory network in the ovary, it is of great interest to understand the relationship between EGF and activin/follistatin systems. The present study was therefore undertaken to clone zebrafish EGF and EGFR in the ovary and characterize their spatiotemporal patterns of expression during follicle and ovarian development. Using a primary

In the present study, we cloned full-length cDNAs for epidermal growth factor (EGF), EGF receptor (EGFR), and three truncated forms of EGFR (EGFR15, 12, and 8) from the zebrafish ovary. Zebrafish EGF was predominantly expressed in the ovary and testis, while EGFR and its truncated forms were highly expressed in all tissues examined except the liver. In the ovary, the expression of EGF seemed to be more abundant in the follicles of early stages, while EGFR had much higher expression levels at later stages. Interestingly, although EGF was expressed in both the follicle cells and oocytes, its expression level was significantly higher in the oocytes. However, the expression of EGFR was mainly restricted to the follicle cells with little expression in the oocytes. The unique spatial patterns of EGF and EGFR expression within the follicle suggest that EGF may serve as a messenger from the oocyte to signal the follicle cells. EGF strongly stimulated the expression of both activin bA and bB, while it suppressed basal and hCG-induced follistatin expression in cultured follicle cells. These results, together with the evidence that EGF was predominantly expressed in the oocytes whereas EGFR was expressed in the follicle cells, strongly suggest that EGF is likely a potential paracrine/juxtacrine factor from the oocytes to regulate the function of the follicle cells.

activin, follistatin, growth factors, oocyte development, ovary

INTRODUCTION

The vertebrate ovary is an extremely dynamic organ, the development and function of which are mainly controlled by gonadotropins, follicle-stimulating hormone (FSH), and luteinizing hormone (LH) from the pituitary [1]. However, evidence has been increasing in recent years that various growth factors produced locally in the ovary play important roles in mediating or modulating gonadotropin actions. One of the extensively studied growth factors is epidermal growth factor (EGF) [2]. EGF was first identified in the submaxillary gland of the mouse [3]. Unlike most other growth factors, mature EGF (53 amino acids) is released from a large precursor anchored in the plasma membrane [4, 5]. EGF acts by binding to a specific EGF receptor (EGFR) on the cell surface with high affinity [6]. EGFR is a glycosylated transmembrane protein with a molecular mass of 170 kDa, which represents Substantially supported by grants (CUHK4176/99M, CUHK4150/01M, and CUHK4258/02M) from the Research Grants Council of the Hong Kong Special Administrative Region to W.G. 2 Correspondence. FAX: 852 2603 5646; e-mail: [email protected] 1

Received: 13 February 2004. First decision: 9 March 2004. Accepted: 19 April 2004. Q 2004 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org

749

750

WANG AND GE

culture of zebrafish ovarian follicle cells, we further evaluated the effects of recombinant human EGF on the expression of activin bA and bB subunits as well as follistatin. Because our previous studies demonstrated that pituitary gonadotropin(s) has (have) profound effects on the differential expression of activin bA, bB, and follistatin in the zebrafish ovary [52–54], we also analyzed the interaction between EGF and gonadotropin on the expression of activin subunits and follistatin. MATERIALS AND METHODS

Chemicals All chemicals were obtained from Sigma (St. Louis, MO) and restriction enzymes were from Promega (Madison, WI) unless otherwise stated. Human chorionic gonadotropin (hCG) was purchased from Sigma, and recombinant human EGF was obtained from Promega. EGF and hCG were first dissolved in water and then diluted to the desired concentrations with the medium before use.

Animals Zebrafish, Danio rerio, were purchased from local pet stores and maintained in flow-through aquaria (36 L) at 258C on a 14L:10D photoperiod. The fish were fed twice a day with commercial tropical fish food with supplement of live brine shrimp once or twice a week. All experiments were performed under license from the Government of the Hong Kong SAR and endorsed by the Animal Experimentation Ethics Committee of the Chinese University of Hong Kong.

Cloning and Sequencing of Full-Length cDNAs for Zebrafish EGF and EGFR Eight pairs of degenerate primers (sequences not shown) were designed based on the conserved regions of EGF precursors from human [5], rat [55], and mouse [4]. A cDNA fragment of ;1500 base pairs (bp) was amplified from the zebrafish ovary by reverse transcription-polymerase chain reaction (RT-PCR) with the primers that code for the amino acid sequences EDVNEC and CQCLKG, respectively. The fragment was then cloned into pBluescript II KS (1) (Stratagene, La Jolla, CA) through T/ A cloning for sequencing. Based on the sequence of the cloned cDNA fragment, gene-specific primers were designed for 59-RACE (rapid amplification of cDNA ends) to amplify the 59-region of the cDNA using the SMART-RACE cDNA Amplification Kit (Clontech, Palo Alto, CA). The amplified 59-cDNA fragments were cloned into pBluescript II KS (1) through T/A cloning and then sequenced. New primers were then designed near the 59-end of the 59-RACE product and used to amplify the fulllength cDNA using 39-RACE. Sequencing of the full-length cDNA was performed on a series of overlapping subclones generated by exonuclease III and mung bean nuclease deletion. Both strands of the cDNA were sequenced using the dRhodamine Terminator Cycle Sequencing Ready Reaction Kit (Perkin-Elmer, Foster City, CA) followed by analysis on the ABI PRISM 3100 Genetic Analyzer (Perkin-Elmer). The full-length cDNA of EGFR was also cloned from the zebrafish ovary with similar strategy and then was sequenced, but the primers used were designed based on a cDNA fragment (fk96gll.y1) from Washington University’s GSC zebrafish EST database.

Primary Follicle Cell Culture The primary culture of zebrafish ovarian follicle cells was performed according to our previous report [54]. Briefly, the follicles from about 20 female zebrafish were isolated and washed with Medium 199 (Gibco Invitrogen, Carlsbad, CA). The follicles were then cultured in Medium 199 supplemented with 10% fetal calf serum (Hyclone, Logan, UT) at 288C in 5% CO2 for 6 days for the follicle cells to proliferate. The proliferated follicle cells were harvested by trypsinization and plated in 24-well plates for 24 h before treatments.

Total RNA Isolation Total RNA was isolated from various tissues, ovarian follicles, and cultured follicle cells with Tri Reagent (Molecular Research Center, Cincinnati, OH) according to the protocol of the manufacturer and our pre-

vious report [53]. The isolation of RNA from the follicle layers and oocytes was performed according to the protocol we described previously [41].

Northern Blot Hybridization Northern blot hybridization was performed according to our previous report [56]. Total RNA (20 mg) from various tissues or isolated ovarian follicles at different stages (previtellogenic, early vitellogenic, midvitellognenic, and full-grown immature follicles) was electrophoresed in a 1.1% denaturing agarose gel containing 2.2 M formaldehyde, transferred to positively charged nylon membrane (Roche Applied Science, Mannheim, Germany) and ultraviolet (UV) cross-linked using the GS Gene Linker (Bio-Rad, Hercules, CA). The blots were hybridized with digoxigenin-labeled antisense RNA probes prepared from the cloned cDNA fragments of EGF (304 bp, 1118–1422) and EGFR (462 bp, 1770–2232) by in vitro transcription and detected with the Chemiluminescent Detection Kit according to manufacturer’s instruction (Roche Applied Science).

Validation of Semiquantitative RT-PCR Assays for Zebrafish EGF and EGFR Single-stranded cDNA was synthesized at 428C for 2 h in a total volume of 10 ml consisting of 3 mg total RNA, 13 Single Strand Buffer (Gibco Invitrogen), 10 mM dithiothreitol, 0.5 mM each dNTP, 0.5 mg oligo-dT, and 100 U SuperScript II (Gibco Invitrogen). To optimize the cycle number used for semiquantitative PCR analysis for each gene, the RT product (0.6 ml) from the cultured follicle cells or isolated ovarian follicles was used as the template for PCR amplification. The primers used for EGF, EGFR, activin bA, activin bB, follistatin, and b-actin are listed in Table 1. PCR was carried out in a volume of 30 ml (13 PCR buffer, 0.2 mM each dNTP, 2.5 mM MgCl2, 0.2 mM each primer, 0.6 U of Taq polymerase) on the Thermal Cycler 9600 (Eppendorf, Hamburg, Germany) for various cycles with the profile of 30 sec at 948C, 30 sec at 628C for EGF and EGFR, and 60 sec at 728C. The PCR products from different cycles of amplification were visualized on a UV-transilluminator after electrophoresis on 1.8% agarose gel containing ethidium bromide, and the signal intensity was quantitated with the Gel-Doc 1000 system and Molecular Analyst Software (Bio-Rad). The cycle numbers that generate halfmaximal amplification were used for subsequent semiquantitative analysis of gene expression, and they are 28 cycles for EGF and 25 cycles for EGFR. The identity of PCR products was confirmed by cloning the PCR products into pBluescript II KS (1) and sequencing. To further validate the semiquantitative RT-PCR assays, PCR amplification (30 ml) was performed on 3 ml of serially diluted plasmids containing EGF or EGFR cDNA to evaluate the correlation between the input of template and the output of PCR amplification. The semiquantitative RT-PCR assays for activin bA, bB, follistatin, and b-actin had been optimized in our previous studies [52, 53].

Data Analysis The mRNA level of each gene was first calculated as the ratio to that of b-actin, which was amplified as the internal control, and then expressed as the percentage of the control group. The data were analyzed by oneway analysis of variance followed by Dunnett test using GraphPad Prism 4.0 for Macintosh OS X (GraphPad Software, San Diego, CA). We performed all the experiments at least twice to confirm the results using different batches of animals.

RESULTS

Cloning and Sequence Characterization of Zebrafish EGF

A full-length cDNA coding for the precursor of EGF was cloned from the zebrafish ovary by 59- and 39-RACE after cloning a cDNA fragment with degenerate primers. The zebrafish EGF cDNA is 3662 bp in length, with a potential polyadenylation site (ATTAAA) 23 bp upstream of the poly(A) tail. The deduced EGF precursor contains 1114 amino acid residues (AY332224), which shares about 41% sequence homology with that of humans [5], rats [55], and mice [4] (Fig. 1). The protein can be divided into three regions based on hydropathy analysis: a long N-terminal extracellular region (966 amino acids), a membrane-span-

751

EGF AND EGFR IN THE ZEBRAFISH OVARY TABLE 1. Primers used in the RT-PCR assays* Gene

Primer

Size (bp)

Accession no.

EGF

Sense: 59-ATGGAATCACGGCTGCTCTCTGGGA-39 Antisense: 59-GCATTCACAAACCCACCGGCCTGGA-39 Sense: 59-ACCTGCAACGGCCCTGGACCCGA-39 Antisense: 59-AGCCCCGGAGCCCAGCACTTTGA-39 Sense: 59-ACCTGCAACGGCCCTGGACCCGA-39 Antisense: 59-TCAGGCCCAGTACATCTGCA-39 Sense: 59-CTCAGCACCACATTCAGAG-39 Antisense: 59-ATATGGCCCTAGATTGACTC-39 Sense: 59-ATGGCAGGACCAACTGA-39 Antisense: 59-AGCAATCGCATTACGTTTG-39 Sense: 59-TGCTGCAAGCGACAATTTTA-39 Antisense: 59-CATTCGTTTCGGGACTCAAG-39 Sense: 59-CAACTTAGATGGACACGCTG-39 Antisense: 59-GTGGATGTCGAGGTCTTGTG-39 Sense: 59-CCAGACTGCTCTAACGTCAC-39 Antisense: 59-CCTGCTTCATGGCACACTC-39 Sense: 59-CCCCTTGTTCACAATAACCT-39 Antisense: 59-TCTGTGGCTTTGGGATTCA-39

304

AY332224

462

AY332223

802

AY533379

486

AY533378

1058

AY533377

400

AF475092

382

X76051

541

AF084948

382

AF057040

EGFR EGFR15 EGFR12 EGFR8 Activin bA Activin bB Follistatin b-actin

* All the primers were synthesized by Integrated DNA Technologies, Inc. (Coralville, IA).

FIG. 1. Comparison of the deduced amino acid sequence of zebrafish EGF precursor with that of mouse (NP034243), rat (P07522), and human (NP001954). Seven EGF-like motifs and mature EGF peptide are boxed with the sequence identities indicated by the values in parentheses. The fully conserved cysteine residues in mature EGF and EGF-like motifs are shaded. The potential transmembrane domain (TMD) is also boxed. The asterisks (*) indicate the amino acids fully conserved among the species compared, and the gaps are shown as dashes (-).

752

WANG AND GE

FIG. 2. Sequence alignment of full-length zebrafish EGFR (EGFR28) with its truncated forms, EGFR15, EGFR12, and EGFR8. The overlined regions are potential ligandbinding domains (LBD). The unique C-terminal amino acid sequences of EGFR12 and EGFR8 are boxed. The structure and exon composition of EGFR and its truncated forms are shown at the bottom with exons marked by numbers in boxes. TMD, Transmembrane domain.

ning hydrophobic domain (21 amino acids), and a short Cterminal cytoplasmic tail (127 amino acids). The extracellular region contains seven EGF-like motifs in addition to the mature EGF domain, which is typical of EGF precursor, and they share different degrees of homology (29–66%) with the corresponding mammalian counterparts [4, 5, 55]. The number and locations of cysteine residues critical for the formation of intramolecular disulfide bonds in each EGF-like motif are strictly conserved between zebrafish and mammals. The mature EGF peptide (53 amino acids) is located near the transmembrane domain, and it shares significant homology with EGF from humans (60%), rats (53%), and mice (55%) [4, 5, 55]. The transmembrane domain is rich in hydrophobic amino acids, typical of membrane-anchored proteins. Compared with the extracellular region, the cytoplasmic tail of EGF precursor shows little homology between zebrafish and mammals (Fig. 1). What is intriguing about the cloned zebrafish EGF precursor is lack of a hydrophobic N-terminus that may serve as the signal peptide, resulting in the slightly shorter N-terminus compared with mammalian molecules. Because the 59-untranslated region contains multiple stop codons in all three reading frames, it seems unlikely that the cDNA clone isolated was not complete. We have performed several inde-

pendent 59-RACEs and all produced the same sequence (data not shown). Cloning and Characterization of Zebrafish EGFR and Its Truncated Forms

We also cloned zebrafish EGFR from the ovary using a similar approach. The full-length zebrafish EGFR cDNA is 4042 bp in length and codes for 1191 amino acids (AY332223). Zebrafish EGFR shares high homology (;63%) with EGFR of humans [57], rats [58], mice [59], and chickens [60], with all the important functional domains well conserved. The receptor can also be divided into three regions based on hydropathy analysis: an N-terminal extracellular region (644 amino acids) for ligand-binding, a single hydrophobic transmembrane domain (23 amino acids), and a long cytoplasmic region (524 amino acids). The extracellular region contains two putative ligand-binding domains (LBD). The intracellular region contains a highly conserved tyrosine kinase domain (257 amino acids) with ;92% homology across vertebrates (Fig. 2). While this manuscript was being prepared, Goishi et al. also published a cDNA sequence of zebrafish EGFR [61]. Comparison of the two sequences shows high identity with 18 nucleotide

EGF AND EGFR IN THE ZEBRAFISH OVARY

753

differences and 4 amino acid changes (data not shown). Because both studies used RACE to clone the cDNA, it remains unclear whether these discrepancies resulted from PCR amplification or they represent different isoforms of the receptor. In addition to the full-length EGFR, we also isolated three cDNAs that code for truncated forms of EGFR. All these truncated forms of EGFR encode the extracellular EGF-binding region. Comparison with the genomic structure of the mammalian EGFR gene suggests that the three truncated forms of zebrafish EGFR are likely generated through differential splicing of mRNA transcript, containing the first 15, 12, and 8 exons, respectively, and they are therefore named EGFR15, EGFR12, and EGFR8 in the present study. EGFR15, 12, and 8 exhibit identical nucleotide and amino acid sequences with the full-length EGFR (or EGFR28) at the 59-end or N-terminus; however, each truncated form has a novel nucleotide sequence at the 39-end, which is likely the alternative exon spliced from the intron 15, 12, or 8. EGFR12 and 8 carry a unique C-terminal sequence of 6 and 56 amino acids, respectively, resulting from the extension of translation into their unique 39 regions (Fig. 2). All three truncated forms of EGFR contain one or more consensus polyadenylation sites (AATAAA) upstream of their poly(A) tails (data not shown). Tissue Distribution of EGF and EGFR Expression

In mammals, EGF is predominantly expressed in the submaxillary gland of the male mouse and kidney of both sexes, while the gonads have much lower transcriptional levels [62, 63]. Due to the lack of sequence information, the distribution of EGF in most vertebrate groups remains largely unknown. In this experiment, we examined EGF expression in different zebrafish tissues using RT-PCR. When PCR was performed for 33 cycles, the expression of EGF could be demonstrated in all tissues examined, including the brain, gill, intestine, kidney, liver, muscle, ovary, and testis. In contrast with the spatial expression pattern reported in mammals, the strongest expression of zebrafish EGF was detected in the ovary, testis, and liver, while the gill and intestine generated the weakest signal (Fig. 3). As the control, the RNA samples without RT produced no signal for all tissues tested. PCR performed at lower cycle number (28 cycles) further confirmed the strongest amplification signal in the ovary, testis, and liver, while no or very weak signal was detected in other tissues. We used bactin as an internal control for RNA loading and RT efficiency, and it did not show significant variation among the tissues investigated. In comparison with EGF, EGFR exhibited more ubiquitous expression in the tissues investigated except the liver, which showed a relatively low expression level of EGFR in contrast with its high expression level of EGF. We used both Northern blot analysis and RT-PCR to examine the expression of EGFR in different tissues to ensure that the RT-PCR assay was reliable. The PCR results using primers that only amplify the full-length receptor were consistent with those demonstrated by Northern blot analysis, which revealed one major mRNA species with one or two minor ones. Interestingly, similar spatial patterns of expression were also demonstrated for the truncated forms of EGFR (EGFR15, 12, and 8), which were detected with primers (Table 1) that specifically amplify each isoform (Fig. 3).

FIG. 3. Tissue distribution of zebrafish EGF and EGFR. Upper panel: RTPCR detection of EGF, full-length EGFR and its truncated forms (EGFR15, EGFR12, and EGFR8). Numbers in parentheses indicate the cycle numbers of PCR amplification. 1, RT positive; 2, RT negative; M, marker. Lower panel: Northern blot analysis of EGFR. B, Brain; G, gill; I, intestine; K, kidney; L, liver; M, muscle; O, ovary; T, testis.

Temporal Expression Profiles of EGF and EGFR in the Ovary During Sexual Maturation

Because EGF is abundantly expressed in the ovary, we further examined the expression profiles of EGF and EGFR in the zebrafish ovary during ovarian development. The experiments started with sexually immature young zebrafish. Because the exact age of the fish was not known, the time of the first sampling was designated Day 0. The ovaries were collected from 12 female zebrafish each time on Days 0, 3, 6, and 10. One ovary from each fish was fixed in Bouin solution for histological examination, and the other one was used for RNA extraction. The semiquantitative RTPCR assays for EGF and EGFR expression were validated as described in the Materials and Methods (Fig. 4). Histological analysis showed that, on Day 0, all ovaries collected had oocytes at primary growth stage (PG) only. On Day 3, some oocytes had started to accumulate cortical vesicles, representing the first cohort of follicles entering vitellogenesis; however, no oocyte contained yolk granules (previtellogenic stage, PV). On Day 6, these growing oocytes had accumulated substantial amounts of yolk granules (midvitellogenic stage, MV), and they became nearly fullgrown on Day 10 (late vitellogenic stage, LV) (histological data not shown). During this 10-day period of ovarian development, the expression of EGF in the whole ovary showed a slight decline, with the highest level detected on Days 0 and 3 and the lowest on Day 10 (Fig. 5A). In contrast with EGF, EGFR expression was very low at early stages, but it dramatically increased during the ovarian de-

754

WANG AND GE

FIG. 4. Validation of the RT-PCR assays for EGF and EGFR. A, B), Kinetics of PCR amplification for EGF and EGFR. The cycle number that generates half maximal reaction was used to analyze the expression level of each gene. C, D) PCR amplification of serially diluted EGF and EGFR cDNA using the cycle numbers obtained from A and B. Each value represents the mean 6 SEM of three PCR reactions, and the electrophoretic image is shown at the bottom of each graph.

velopment with the highest level reached on Day 6, when the first wave of follicles was at the midvitellogenic stage (Fig. 5B). Stage-Dependent Expression of EGF and EGFR in the Ovarian Follicles from Sexually Mature Zebrafish

To investigate if the expression profiles of EGF and EGFR observed in the ovary during sexual maturation are related to the first wave of developing follicles, we further examined the relative levels of EGF and EGFR expression in the ovarian follicles of different developmental stages isolated from sexually mature zebrafish, including the previtellogenic (PV, ;0.25 mm in diameter), early vitellogenic (EV, ;0.35 mm), midvitellogenic (MV, ;0.45 mm), and full-grown immature follicles (FG, ;0.65 mm). Northern blot analysis demonstrated a significant change in EGF expression during follicle development. The highest expression of EGF was detected in the previtellogenic follicles, and the level decreased with follicle development (Fig. 6A). To confirm the relative mRNA abundance of EGF in the developing follicles, we further analyzed its expression with semiquantitative RT-PCR. Consistent with Northern blot analysis, RT-PCR also revealed expression of EGF at all stages of follicle development; however, its expression level decreased with the advancement of follicle development from the previtellogenic to full-grown stage (Fig. 6B). In contrast with EGF, the expression of EGFR showed an opposite trend of variation; it increased during the follicle development as demonstrated by both Northern hybridization and RT-PCR (Fig. 6, C and D). The low expression

FIG. 5. Temporal expression profiles of EGF and EGFR in the whole ovary during sexual maturation. The expression levels are first normalized to b-actin and then expressed as the percentage of PG (Day 0). The data are combined from two independent experiments using two batches of zebrafish. Each value represents the mean 6 SEM of RT-PCR reactions on 12 individual zebrafish sampled at each stage (n 5 12). The electrophoretic image shown at the bottom is from one representative experiment (n 5 6). PG, Primary growth; PV, previtellogenic; MV, midvitellogenic; LV, late vitellogenic stage. * P , 0.05; ** P , 0.01 vs. PG.

level of EGFR detected at the full-grown stage by Northern analysis may be partly due to the low RNA loading. Localization of EGF and EGFR Expression in the Follicle Both EGF and EGFR were expressed in cultured zebrafish follicle cells (Fig. 7A). However, analysis of their relative abundance of expression in the full-grown oocytes and the surrounding follicle layers showed that EGF and EGFR had distinct spatial patterns of expression within the follicle. EGF could be easily detected in both the isolated follicle layers and the oocytes; however, its expression level was significantly higher in the oocytes than that in the follicle layers (Fig. 7B). In contrast, EGFR was highly expressed by the follicle layers, but its signal was barely detectable in the oocytes (Fig. 7C). Effects of Recombinant Human EGF on Basal and Gonadotropin-Regulated Expression of Zebrafish Activin Subunits and Follistatin in the Follicle Cells Our previous studies have demonstrated that EGF and TGFa significantly enhance the final oocyte maturation in

EGF AND EGFR IN THE ZEBRAFISH OVARY

the zebrafish, and their effects are likely mediated by increasing activin activity in the follicles [35]. To further understand the functional relationship between EGF and activin system in the zebrafish ovary, we investigated the effects of human EGF on the expression of activin bA and bB subunits as well as follistatin in cultured zebrafish follicle cells. EGF (20 nM) significantly stimulated basal activin bA and bB expression, but suppressed follistatin expression in a time-dependent manner. The effects reached peak levels at 2 h of the treatment, and longer treatment (4 and 8 h) led to reduced responses (Fig. 8). When applied for 2 h in the presence or absence of hCG (5 IU/ml), EGF exhibited dose-dependent stimulatory effects on basal expression of both activin bA and bB, and its effect on activin bA was much more powerful than that of hCG (5 IU/ml) even at doses as low as 0.2–2 nM. When applied together, however, EGF and hCG did not exhibit evident additive effects on activin bA expression at any doses used (Fig. 9A). As for activin bB, we have recently demonstrated that hCG and goldfish pituitary extract significantly suppress its expression in cultured zebrafish follicle cells [53]. In the presence of hCG, which lowered basal activin bB expression, EGF still significantly enhanced bB expression and hCG did not seem to suppress the magnitude of the response to EGF (Fig. 9B). Interestingly, although EGF has been reported to increase follistatin mRNA level significantly in cultured porcine granulosa cells [64], it suppressed both basal and hCG-stimulated follistatin expression in the zebrafish ovary. The inhibition of basal follistatin expression by EGF was not as prominent as that of hCG-induced expression because of the extremely low basal expression of the gene in cultured follicle cells (Fig. 9C). DISCUSSION In the present study, we cloned a full-length cDNA of EGF from the zebrafish ovary. To our knowledge, this represents the first in nonmammalian vertebrates. Although the amino acid sequence of zebrafish EGF precursor is slightly shorter than its mammalian counterparts, the arrangement of seven EGF-like motifs and mature EGF peptide is similar to those in mammals. The overall sequence homology of EGF precursor between zebrafish and mammals is not high [4, 5, 55]; however, the homology is higher over the seven EGF-like motifs and mature EGF peptide with the first EGF-like motif near the N-terminus showing the highest homology (66%) with that of mammalian molecules. The cysteines in each EGF-like motif and mature EGF are fully conserved. The biological functions of the seven EGFlike peptides in EGF precursor remain unknown [65]. Like other EGF family members, such as TGFa, amphiregulin (AR), and heparin-binding EGF-like growth factor (HBEGF) [66], zebrafish EGF precursor also contains a hydrophobic transmembrane domain, suggesting that the protein is probably anchored on the plasma membrane before EGF is released by cleavage. In addition to EGF, we also cloned a full-length cDNA coding for EGFR from the zebrafish ovary together with several truncated forms of EGFR. In comparison with EGF, zebrafish EGFR shares relatively higher homology with human and chicken EGFR [57, 60]. Interestingly, all the truncated forms of EGFR (EGFR15, 12, and 8) code for the extracellular ligand-binding domains, and they are likely released as free proteins. The functions of these truncated forms of EGFR remain unknown. Because various truncated forms of EGFR containing the extracellular domains

755

FIG. 6. Stage-dependent expression of EGF (A, B) and EGFR (C, D) in isolated follicles. A, C) Northern blot analysis. B, D) Semiquantitative RTPCR assay. The expression levels are first normalized to b-actin and then expressed as the percentage of PV. Each value represents the mean 6 SEM of three or four independent RT-PCR reactions from one representative experiment. PV, Previtellogenic; EV, early vitellogenic; MV, midvitellogenic; FG, full-grown immature follicles; * P , 0.05; ** P , 0.01 vs. PV.

have also been reported in chickens, rodents, and humans [67], the finding of similar molecules in fish strongly suggests that they may play critical biological roles in vertebrates. One possibility is that the secreted extracellular fragments of EGFR may serve as EGF-binding proteins to fine tune the diverse biological activities of EGF or EGF-related ligands in various tissues, as suggested by their wide spatial distribution in zebrafish body. In contrast with EGF in mammals, where it is abundantly expressed in the submaxillary gland of male mouse and the kidney with much less expression in the gonads [62, 63], EGF in the zebrafish exhibited distinct expression patterns among the tissues examined. Interestingly, the ovary and testis were the predominant sites of EGF expression in the zebrafish, whereas the kidney had a much lower mRNA level, strongly implicating EGF in the regulation of reproduction in fish. In contrast, EGFR and its truncated forms exhibited more ubiquitous expression in different tissues except the liver, suggesting diverse roles for EGF in multiple sites in addition to the gonads. In mammals, although there have been studies demonstrating the expression of EGF and EGFR in the ovary [10– 12, 68, 69], the results vary among different reports about the spatial and temporal patterns of EGF expression in the ovarian follicles. For example, Qu et al. demonstrated weak signal of EGF in the oocytes of primordial and primary

756

WANG AND GE

FIG. 7. Localization of EGF and EGFR expression within the full-grown follicle. A) RT-PCR demonstration of EGF and EGFR expression in cultured zebrafish follicle cells. RT1, PCR with reverse transcription; RT2, PCR without reverse transcription. B) Expression of EGF in the follicle layers and oocytes. C) Expression of EGFR in the follicle layers and oocytes. The expression levels are first normalized to b-actin and then expressed as the percentage of that in the follicle layers. Each value represents the mean 6 SEM of four independent samples and RT-PCR reactions. * P , 0.05; *** P , 0.001 vs. follicle layer.

follicles and the theca cells as well in humans [12]. However, Maruo et al. [10] reported no immunostaining for EGF and EGFR in the primordial follicles. The staining for EGF and EGFR appeared first in the oocytes of preantral follicles followed by the granulosa and theca interna cells of the antral follicles. The signal in the oocytes increased as the follicles reached the pre-ovulatory stage [10]. In the zebrafish, the present study demonstrated that EGF was highly expressed in the previtellogenic ovary of sexually immature fish and its expression showed a declining trend with the

FIG. 8. Time course of EGF (20 nM) effects on the expression of activin bA, bB, and follistatin in cultured follicle cells. Twenty-four hours after subculturing, the medium was changed and the cells were further incubated for 8 h, during which EGF was added at different times (2, 4, and 8 h) before RNA extraction at the end of incubation. The control cells were also incubated for the same period of time (8 h) but received no EGF treatment. The expression levels are first normalized to b-actin and then expressed as the percentage of respective control. Each value represents the mean 6 SEM of three RT-PCR reactions. * P , 0.05; ** P , 0.01 vs. control.

development of the ovary. This temporal pattern of expression during ovarian development was further supported by the stage-dependent expression pattern of EGF in the follicles isolated from sexually mature fish. EGF had relatively higher expression levels in the previtellogenic and early vitellogenic follicles than the later stages. In contrast, the expression of EGFR did not follow the pattern of EGF during ovarian and follicle development; it was very low in the ovary of immature fish but increased significantly with the onset of vitellogenesis in the ovary and remained high in the late stages of ovarian development. This was again consistent with its expression pattern in isolated follicles. The discrepancy between the temporal patterns of EGF and EGFR expression seems perplexing; however, it is not too surprising in view of the fact that EGFR is also activated by several other EGF-like ligands, such as TGFa. The expression of EGFR in the zebrafish ovary agrees with a previous report in the goldfish that a single class of high-affinity binding sites for EGF exists in the ovarian membrane preparations [33]. One of the exciting findings of the present study was the localization of EGF and EGFR expression in different functional compartments of the follicle, i.e. the oocyte and the follicle layer surrounding the oocyte. Although EGF was expressed in both follicle cells and oocytes, it is noteworthy that its expression level in the oocytes was significantly higher. In contrast, EGFR was exclusively present in the follicle cells, with little signal detected in the oocytes. This information provides important clues to the potential role of EGF in the intrafollicular communication between the oocyte and follicle cells and has led us to hypothesize that the oocyte-derived EGF may act as a paracrine or juxtacrine factor to signal the surrounding follicle cells that abundantly express its receptors (Fig. 10). This hypothesis needs to be further tested at the protein level and confirmed by functional studies in the future. Interestingly, a similar phenomenon has also been reported in other model organisms, including invertebrates, suggesting fundamental roles for such a mechanism in controlling follicle and oocyte development. In the chicken, EGF has been localized to the germinal disc (GD) of the oocyte right under the oolemma by immunocytochemical staining [70]. Further evidence indi-

EGF AND EGFR IN THE ZEBRAFISH OVARY

757

FIG. 10. Hypothetical model for paracrine action of EGF from the oocyte on the follicle cells to control the expression of activin bA (ActbA), bB (ActbB), and follistatin (FS). Other signaling pathways are based on our previous studies [35, 41, 44, 52–54], and the pathways in dotted lines need to be experimentally elucidated. LH (hCG) stimulates ActbA and FS expression through the cAMP-PKA pathway, but suppresses ActbB through a cAMP-dependent but PKA-independent pathway. However, EGF from the follicle cells and oocyte increases the expression of both ActbA and ActbB, but inhibits that of FS. The activin from the follicle cells may act directly on the oocyte because its entire signaling system, including receptors and intracellular signaling molecules (Smads), is expressed in the oocyte. AC, Adenylate cyclase; DHP, 17a,20b-dihydroxy4-pregnen-3-one; PDE, phosphodiesterase; PKA, protein kinase A; 1, stimulation; 2, inhibition.

FIG. 9. Dose response of EGF effects (2 h) on basal and hCG-regulated expression of activin bA, bB, and follistatin in cultured follicle cells. The expression levels are first normalized to b-actin and then expressed as the percentage of control or hCG-alone group. Each value represents the mean 6 SEM of three RT-PCR reactions. * P , 0.05; ** P , 0.01 vs. control; 11, P , 0.01 vs. hCG-alone group.

cated that the GD of chicken oocyte seemed to secrete EGF that stimulated proliferation of the overlying granulosa cells in a paracrine manner [71]. In Drosophila, an EGF motifcontaining TGFa-like ligand, Gurken (Grk), is also a membrane-bound protein exclusively expressed in the oocytes [72]. After cleavage, the released Grk signals the surrounding follicle cells by binding and activating Drosophila orthologue of EGFR [73], and one of the responses to the oocyte-derived Grk is the guided migration of special follicle cells (border cells) during oogenesis, an event essential for female fertility [74]. In mammals, there have also been reports that EGF/TGFa and EGFR are expressed in both the somatic follicle cells and oocytes, but the results from

different studies in different species are somehow confusing and inconsistent [10, 12, 68, 69, 75, 76]. A recent study demonstrated that co-incubation with denuded bovine oocytes suppressed FSH-induced secretion of steroids and inhibin family members from the granulosa cells, and these effects could be mimicked by TGFa, suggesting a TGFamediated paracrine regulation of the granulosa cells by the oocytes [77]. The unique spatiotemporal patterns of EGF and EGFR expression in the zebrafish ovary immediately raise a question about the role of EGF in controlling follicle development and function as well as its relationship with endocrine hormones such as gonadotropins and other ovarian factors in fish. In the goldfish, EGF has been reported to stimulate follicle cell proliferation, and its effect was more effective in the vitellogenic follicles than the late full-grown immature follicles [31]. A similar effect of EGF has also been reported in cultured chicken granulosa and theca-interstitial cells from the follicles of early stages [27, 28]. These results suggest that EGF expressed in the early follicles may be related to the active proliferation of the follicle cells. In

758

WANG AND GE

the late stages of follicle development, EGF has been demonstrated to promote oocyte maturation in the goldfish [33] and zebrafish [35], and its effect appears to involve de novo mRNA transcription and translation [35]. EGF also seems to be a powerful inhibitor of apoptosis in the follicles or follicle cells in both fish [32] and mammals [20, 78, 79]. We have recently demonstrated that human EGF and TGFa both promote zebrafish oocyte maturation in vitro and their effects can be completely blocked by follistatin, a potent binding protein for activin, suggesting a role for the ovarian activin system in the regulatory pathway of EGF [35]. This hypothesis is supported by the evidence from the present study that EGF significantly stimulated activin bA and bB but suppressed basal and hCG-stimulated follistatin expression in cultured zebrafish follicle cells. The stimulation of activin bA by EGF was more powerful than that by hCG. Consistent with our previous finding that hCG inhibits activin bB expression in the zebrafish ovary [53], hCG significantly reduced basal expression of activin bB in the present study; however, hCG did not seem to affect the stimulatory effect of EGF on activin bB expression, although the overall level was lower in the presence of hCG because of the reduced basal expression. The inhibition of hCG-induced follistatin expression by EGF in the zebrafish is similar to a report in the rat that EGF inhibited FSH-stimulated follistatin expression in cultured granulosa cells [80], but it is different from other studies in mammals. EGF stimulates follistatin expression in cultured porcine granulosa cells [64] and rat renal mesangial cell line [81]. It has been demonstrated in mammals that, when applied together, EGF is capable of decreasing the gonadotropin-induced cAMP level in cultured rat granulosa cells or luteal membrane [22, 82]. Therefore, the possibility exists that the inhibition of hCG-induced follistatin expression by EGF in the zebrafish is due to a decrease of hCGinduced intracellular cAMP level, which we have shown to enhance follistatin expression in cultured zebrafish ovarian follicle cells [52]. Recently, we have demonstrated that activin bA and bB are both predominantly expressed in the somatic follicle cells, whereas the entire activin signaling system, including its receptors (types I and II), and intracellular signaling molecules (Smad2, 3, 4, and 7) is abundantly expressed in the oocytes. These results suggest a messenger role for activin from the follicle cells to signal the oocyte [41]. The expression of EGF in the oocyte and its potent effects on the expression of activin subunits and follistatin in the follicle cells may represent a reciprocal intrafollicular communication mechanism by which the oocyte influences the functionality of the surrounding follicle cells. It would be of particular interest to examine if activin has any effect on the expression of EGF in the oocyte. In summary, we have cloned EGF, EGFR, and several truncated forms of EGFR from the zebrafish ovary. Zebrafish EGF exhibited distinct tissue distribution compared with that in mammals in that its expression was the highest in the ovary and testis, which strongly implicates EGF in the regulation of fish reproduction. EGF and EGFR also showed distinct patterns of expression during ovarian development and among developing follicles. Within the follicle, EGF was found to be expressed abundantly in the oocyte whereas EGFR was exclusively expressed in the follicle cells, suggesting a potential mechanism for EGF to signal the follicle cells by the oocyte. To provide evidence for our previous hypothesis that the ovarian activin system is likely a downstream mediator of EGF signaling in the

zebrafish ovary, the present study further examined the effects and interactive effects of EGF and hCG on the expression of ovarian activin/follistatin system in the follicle cells. Recombinant human EGF induced activin bA and bB expression but suppressed basal and hCG-induced follistatin expression in cultured zebrafish ovarian follicle cells (Fig. 10). REFERENCES 1. McGee EA, Hsueh AJ. Initial and cyclic recruitment of ovarian follicles. Endocr Rev 2000; 21:200–214. 2. Hsueh AJ, Billig H, Tsafriri A. Ovarian follicle atresia: a hormonally controlled apoptotic process. Endocr Rev 1994; 15:707–724. 3. Cohen S. Isolation of a mouse submaxillary gland protein accelerating incisor eruption and eyelid opening in the new-born animal. J Biol Chem 1962; 237:1555–1562. 4. Scott J, Urdea M, Quiroga M, Sanchez-Pescador R, Fong N, Selby M, Rutter WJ, Bell GI. Structure of a mouse submaxillary messenger RNA encoding epidermal growth factor and seven related proteins. Science 1983; 221:236–240. 5. Gray A, Dull TJ, Ullrich A. Nucleotide sequence of epidermal growth factor cDNA predicts a 128,000-molecular weight protein precursor. Nature 1983; 303:722–725. 6. Carpenter G. Epidermal growth factor: biology and receptor metabolism. J Cell Sci Suppl 1985; 3:1–9. 7. Boulougouris P, Elder J. Epidermal growth factor receptor structure, regulation, mitogenic signalling and effects of activation. Anticancer Res 2001; 21:2769–2775. 8. Wells A. EGF receptor. Int J Biochem Cell Biol 1999; 31:637–643. 9. Derynck R. Transforming growth factor-a. Mol Reprod Dev 1990; 27:3–9. 10. Maruo T, Ladines-Llave CA, Samoto T, Matsuo H, Manalo AS, Ito H, Mochizuki M. Expression of epidermal growth factor and its receptor in the human ovary during follicular growth and regression. Endocrinology 1993; 132:924–931. 11. Roy SK, Greenwald GS. Immunohistochemical localization of epidermal growth factor-like activity in the hamster ovary with a polyclonal antibody. Endocrinology 1990; 126:1309–1317. 12. Qu J, Nisolle M, Donnez J. Expression of transforming growth factora, epidermal growth factor, and epidermal growth factor receptor in follicles of human ovarian tissue before and after cryopreservation. Fertil Steril 2000; 74:113–121. 13. Onagbesan OM, Gullick W, Woolveridge I, Peddie MJ. Immunohistochemical localization of epidermal growth factor receptors, epidermal-growth-factor-like and transforming-growth-factor-a-like peptides in chicken ovarian follicles. J Reprod Fertil 1994; 102:147–153. 14. Roy SK, Greenwald GS. In vitro effects of epidermal growth factor, insulin-like growth factor-I, fibroblast growth factor, and follicle-stimulating hormone on hamster follicular deoxyribonucleic acid synthesis and steroidogenesis. Biol Reprod 1991; 44:889–896. 15. Roy SK, Harris SG. Antisense epidermal growth factor oligodeoxynucleotides inhibit follicle-stimulating hormone-induced in vitro DNA and progesterone synthesis in hamster preantral follicles. Mol Endocrinol 1994; 8:1175–1181. 16. Hammond JM, English HF. Regulation of deoxyribonucleic acid synthesis in cultured porcine granulosa cells by growth factors and hormones. Endocrinology 1987; 120:1039–1046. 17. Bendell JJ, Dorrington JH. Epidermal growth factor influences growth and differentiation of rat granulosa cells. Endocrinology 1990; 127: 533–540. 18. Hsueh AJ, Welsh TH, Jones PB. Inhibition of ovarian and testicular steroidogenesis by epidermal growth factor. Endocrinology 1981; 108: 2002–2004. 19. Mondschein JS, Schomberg DW. Growth factors modulate gonadotropin receptor induction in granulosa cell cultures. Science 1981; 211: 1179–1180. 20. Tilly JL, LaPolt PS, Hsueh AJ. Hormonal regulation of follicle-stimulating hormone receptor messenger ribonucleic acid levels in cultured rat granulosa cells. Endocrinology 1992; 130:1296–1302. 21. Feng P, Catt KJ, Knecht M. Transforming growth factor b regulates the inhibitory actions of epidermal growth factor during granulosa cell differentiation. J Biol Chem 1986; 261:14167–14170. 22. Knecht M, Catt KJ. Modulation of cAMP-mediated differentiation in ovarian granulosa cells by epidermal growth factor and platelet-derived growth factor. J Biol Chem 1983; 258:2789–2794.

EGF AND EGFR IN THE ZEBRAFISH OVARY 23. Jones PB, Welsh TH Jr, Hsueh AJ. Regulation of ovarian progestin production by epidermal growth factor in cultured rat granulosa cells. J Biol Chem 1982; 257:11268–11273. 24. Ding J, Foxcroft GR. Epidermal growth factor enhances oocyte maturation in pigs. Mol Reprod Dev 1994; 39:30–40. 25. Coskun S, Lin YC. Mechanism of action of epidermal growth factorinduced porcine oocyte maturation. Mol Reprod Dev 1995; 42:311– 317. 26. Dekel N, Sherizly I. Epidermal growth factor induces maturation of rat follicle-enclosed oocytes. Endocrinology 1985; 116:406–409. 27. Onagbesan OM, Peddie MJ, Williams J. Regulation of cell proliferation and estrogen synthesis by ovine LH, IGF-I, and EGF in theca interstitial cells of the domestic hen cultured in defined media. Gen Comp Endocrinol 1994; 94:261–272. 28. Peddie MJ, Onagbesan OM, Williams J. Chicken granulosa cell proliferation and progesterone production in culture: effects of EGF and theca secretions. Gen Comp Endocrinol 1994; 94:341–356. 29. Li Z, Johnson AL. Regulation of P450 cholesterol side-chain cleavage messenger ribonucleic acid expression and progesterone production in hen granulosa cells. Biol Reprod 1993; 49:463–469. 30. Pulley DD, Marrone BL. Inhibitory action of epidermal growth factor on progesterone biosynthesis in hen granulosa cells during short term culture: two sites of action. Endocrinology 1986; 118:2284–2291. 31. Kumar Srivastava R, Van Der Kraak G. Multifactorial regulation of DNA synthesis in goldfish ovarian follicles. Gen Comp Endocrinol 1995; 100:397–403. 32. Janz DM, Van Der Kraak G. Suppression of apoptosis by gonadotropin, 17b-estradiol, and epidermal growth factor in rainbow trout preovulatory ovarian follicles. Gen Comp Endocrinol 1997; 105:186– 193. 33. Pati D, Balshaw K, Grinwich DL, Hollenberg MD, Habibi HR. Epidermal growth factor receptor binding and biological activity in the ovary of goldfish, Carassius auratus. Am J Physiol 1996; 270:R1065– 1072. 34. MacDougall TM, Van Der Kraak G. Peptide growth factors modulate prostaglandin E and F production by goldfish ovarian follicles. Gen Comp Endocrinol 1998; 110:46–57. 35. Pang Y, Ge W. Epidermal growth factor and TGFa promote zebrafish oocyte maturation in vitro: potential role of the ovarian activin regulatory system. Endocrinology 2002; 143:47–54. 36. Knight PG, Glister C. Potential local regulatory functions of inhibins, activins and follistatin in the ovary. Reproduction 2001; 121:503–512. 37. Ethier JF, Findlay JK. Roles of activin and its signal transduction mechanisms in reproductive tissues. Reproduction 2001; 121:667– 675. 38. Davis AJ, Johnson PA. Expression pattern of messenger ribonucleic acid for follistatin and the inhibin/activin subunits during follicular and testicular development in Gallus domesticus. Biol Reprod 1998; 59:271–277. 39. Peng C, Mukai ST. Activins and their receptors in female reproduction. Biochem Cell Biol 2000; 78:261–279. 40. Ge W. Roles of the activin regulatory system in fish reproduction. Can J Physiol Pharmacol 2000; 78:1077–1085. 41. Wang Y, Ge W. Spatial expression patterns of activin and its signaling system in the zebrafish ovarian follicle: evidence for paracrine action of activin on the oocytes. Biol Reprod 2003; 69:1998–2006. 42. Ying SY. Inhibins, activins, and follistatins: gonadal proteins modulating the secretion of follicle-stimulating hormone. Endocr Rev 1988; 9:267–293. 43. Pang Y, Ge W. Activin stimulation of zebrafish oocyte maturation in vitro and its potential role in mediating gonadotropin-induced oocyte maturation. Biol Reprod 1999; 61:987–992. 44. Pang Y, Ge W. Gonadotropin and activin enhance maturational competence of oocytes in the zebrafish (Danio rerio). Biol Reprod 2002; 66:259–265. 45. Wu T, Patel H, Mukai S, Melino C, Garg R, Ni X, Chang J, Peng C. Activin, inhibin, and follistatin in zebrafish ovary: expression and role in oocyte maturation. Biol Reprod 2000; 62:1585–1592. 46. Nakamura T, Takio K, Eto Y, Shibai H, Titani K, Sugino H. Activinbinding protein from rat ovary is follistatin. Science 1990; 247:836– 838. 47. Phillips DJ, de Kretser DM. Follistatin: a multifunctional regulatory protein. Front Neuroendocrinol 1998; 19:287–322. 48. Tano M, Minegishi T, Nakamura K, Karino S, Ibuki Y, Miyamoto K. Transcriptional and posttranscriptional regulation of FSH receptor in rat granulosa cells by cyclic AMP and activin. J Endocrinol 1997; 153:465–473.

759

49. Miro F, Hillier SG. Relative effects of activin and inhibin on steroid hormone synthesis in primate granulosa cells. J Clin Endocrinol Metab 1992; 75:1556–1561. 50. Xiao S, Robertson DM, Findlay JK. Effects of activin and folliclestimulating hormone (FSH)-suppressing protein/follistatin on FSH receptors and differentiation of cultured rat granulosa cells. Endocrinology 1992; 131:1009–1016. 51. Tsuchiya M, Minegishi T, Kishi H, Tano M, Kameda T, Hirakawa T, Ibuki Y, Mizutani T, Miyamoto K. Control of the expression of luteinizing hormone receptor by local factors in rat granulosa cells. Arch Biochem Biophys 1999; 367:185–192. 52. Wang Y, Ge W. Gonadotropin regulation of follistatin expression in the cultured ovarian follicle cells of zebrafish, Danio rerio. Gen Comp Endocrinol 2003; 134:308–315. 53. Wang Y, Ge W. Involvement of cyclic adenosine 39,59-monophosphate in the differential regulation of activin bA and bB expression by gonadotropin in the zebrafish ovarian follicle cells. Endocrinology 2003; 144:491–499. 54. Pang Y, Ge W. Gonadotropin regulation of activin bA and activin type IIA receptor expression in the ovarian follicle cells of the zebrafish, Danio rerio. Mol Cell Endocrinol 2002; 188:195–205. 55. Saggi SJ, Safirstein R, Price PM. Cloning and sequencing of the rat preproepidermal growth factor cDNA: comparison with mouse and human sequences. DNA Cell Biol 1992; 11:481–487. 56. Wang Y, Ge W. Cloning of zebrafish ovarian carbonyl reductase-like 20 b-hydroxysteroid dehydrogenase and characterization of its spatial and temporal expression. Gen Comp Endocrinol 2002; 127:209–216. 57. Ullrich A, Coussens L, Hayflick JS, Dull TJ, Gray A, Tam AW, Lee J, Yarden Y, Libermann TA, Schlessinger J, Downward J, Mayes ELV, Whittle N, Waterfield MD, Seeburg PH. Human epidermal growth factor receptor cDNA sequence and aberrant expression of the amplified gene in A431 epidermoid carcinoma cells. Nature 1984; 309:418– 425. 58. Petch LA, Harris J, Raymond VW, Blasband A, Lee DC, Earp HS. A truncated, secreted form of the epidermal growth factor receptor is encoded by an alternatively spliced transcript in normal rat tissue. Mol Cell Biol 1990; 10:2973–2982. 59. Avivi A, Lax I, Ullrich A, Schlessinger J, Givol D, Morse B. Comparison of EGF receptor sequences as a guide to study the ligand binding site. Oncogene 1991; 6:673–676. 60. Lax I, Johnson A, Howk R, Sap J, Bellot F, Winkler M, Ullrich A, Vennstrom B, Schlessinger J, Givol D. Chicken epidermal growth factor (EGF) receptor: cDNA cloning, expression in mouse cells, and differential binding of EGF and transforming growth factor a. Mol Cell Biol 1988; 8:1970–1978. 61. Goishi K, Lee P, Davidson AJ, Nishi E, Zon LI, Klagsbrun M. Inhibition of zebrafish epidermal growth factor receptor activity results in cardiovascular defects. Mech Dev 2003; 120:811–822. 62. Rall LB, Scott J, Bell GI, Crawford RJ, Penschow JD, Niall HD, Coghlan JP. Mouse prepro-epidermal growth factor synthesis by the kidney and other tissues. Nature 1985; 313:228–231. 63. Scott J, Patterson S, Rall L, Bell GI, Crawford R, Penschow J, Niall H, Coghlan J. The structure and biosynthesis of epidermal growth factor precursor. J Cell Sci Suppl 1985; 3:19–28. 64. Lindsell CE, Misra V, Murphy BD. Regulation of follistatin messenger ribonucleic acid in porcine granulosa cells by epidermal growth factor and the protein kinase-C pathway. Endocrinology 1993; 132:1630– 1636. 65. Van Zoelen EJ, Stortelers C, Lenferink AE, Van de Poll ML. The EGF domain: requirements for binding to receptors of the ErbB family. Vitam Horm 2000; 59:99–131. 66. Massague J, Pandiella A. Membrane-anchored growth factors. Annu Rev Biochem 1993; 62:515–541. 67. Reiter JL, Threadgill DW, Eley GD, Strunk KE, Danielsen AJ, Sinclair CS, Pearsall RS, Green PJ, Yee D, Lampland AL, Balasubramaniam S, Crossley TD, Magnuson TR, James CD, Maihle NJ. Comparative genomic sequence analysis and isolation of human and mouse alternative EGFR transcripts encoding truncated receptor isoforms. Genomics 2001; 71:1–20. 68. Singh B, Rutledge JM, Armstrong DT. Epidermal growth factor and its receptor gene expression and peptide localization in porcine ovarian follicles. Mol Reprod Dev 1995; 40:391–399. 69. Reeka N, Berg FD, Brucker C. Presence of transforming growth factor a and epidermal growth factor in human ovarian tissue and follicular fluid. Hum Reprod 1998; 13:2199–2205. 70. Volentine KK, Yao HH, Bahr JM. Epidermal growth factor in the

760

71. 72. 73.

74. 75.

76. 77.

WANG AND GE

germinal disc and its potential role in follicular development in the chicken. Biol Reprod 1998; 59:522–526. Yao HHC, Bahr JM. Germinal disc-dervied epidermal growth factor; a paracrine factor to stimulate proliferation of granulosa cells. Biol Reprod 2001; 64:390–395. Neuman-Silberberg FS, Schupbach T. The Drosophila dorsoventral patterning gene gurken produces a dorsally localized RNA and encodes a TGF a-like protein. Cell 1993; 75:165–174. Ghiglione C, Bach EA, Paraiso Y. Carraway KL 3rd, Noselli S, Perrimon N. Mechanism of activation of the Drosophila EGF Receptor by the TGFa ligand Gurken during oogenesis. Development 2002; 129:175–186. Duchek P, Rorth P. Guidance of cell migration by EGF receptor signaling during Drosophila oogenesis. Science 2001; 291:131–133. Bennett RA, Osathanondh R, Yeh J. Immunohistochemical localization of transforming growth factor-a, epidermal growth factor (EGF), and EGF receptor in the human fetal ovary. J Clin Endocrinol Metab 1996; 81:3073–3076. Chegini N, Williams RS. Immunocytochemical localization of transforming growth factors (TGFs) TGF-a and TGF-b in human ovarian tissues. J Clin Endocrinol Metab 1992; 74:973–980. Glister C, Groome NP, Knight PG. Oocyte-mediated suppression of follicle-stimulating hormone- and insulin-like growth factor-induced

78.

79.

80. 81. 82.

secretion of steroids and inhibin-related proteins by bovine granulosa cells in vitro: possible role of transforming growth factor a. Biol Reprod 2003; 68:758–765. Tilly JL, Billig H, Kowalski KI, Hsueh AJ. Epidermal growth factor and basic fibroblast growth factor suppress the spontaneous onset of apoptosis in cultured rat ovarian granulosa cells and follicles by a tyrosine kinase-dependent mechanism. Mol Endocrinol 1992; 6:1942– 1950. Luciano AM, Pappalardo A, Ray C, Peluso JJ. Epidermal growth factor inhibits large granulosa cell apoptosis by stimulating progesterone synthesis and regulating the distribution of intracellular free calcium. Biol Reprod 1994; 51:646–654. Michel U, McMaster JW, Findlay JK. Regulation of steady-state follistatin mRNA levels in rat granulosa cells in vitro. J Mol Endocrinol 1992; 9:147–156. Michel U, Farnworth P. Follistatin steady state messenger ribonucleic acid levels in confluent cultures of a rat renal mesangial cell line are regulated by multiple factors. Endocrinology 1992; 130:3684–3693. Lamm ML, Rajagopalan-Gupta RM, Hunzicker-Dunn M. Epidermal growth factor-induced heterologous desensitization of the luteinizing hormone/choriogonadotropin receptor in a cell-free membrane preparation is associated with the tyrosine phosphorylation of the epidermal growth factor receptor. Endocrinology 1999; 140:29–36.