| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Endocrinologie Moléculaire de la Reproduction, UMR CNRS 6026, Campus de Beaulieu, 35042 Rennes cedex, France
b Laboratoire de Biologie Moléculaire et Cellulaire, UMR CNRS 5665, Ecole Normale Supérieure de Lyon, 69364 Lyon cedex 07, France
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ERß, and increasing evidence that the ERß subtype has more than one form. However, there is little information on the characteristics and functional significance of these ERs in adults and during development. Here, we report the cloning and characterization of three functional ER forms, zfER, zfERß1, and zfERß2, in the zebrafish. The percentages of identity between these receptors suggest the existence of three distinct genes. Each cDNA encoded a protein that specifically bound estradiol with a dissociation constant ranging from 0.4 nM (zfERß2) to 0.75 nM (zfER and zfERß1). In transiently transfected cells, all three forms were able to induce, in a dose-dependent manner, the expression of a reporter gene driven by a consensus estrogen responsive element; zfERß2 was slightly more sensitive than zfER and zfERß1. Tissue distribution pattern, analyzed by reverse transcription polymerase chain reaction, showed that the three zfER mRNAs largely overlap and are predominantly expressed in brain, pituitary, liver, and gonads. In situ hybridization was performed to study in more detail the distribution of the three zfER mRNAs in the brain of adult females. The zfER mRNAs exhibit distinct but partially overlapping patterns of expression in two neuroendocrine regions, the preoptic area and the mediobasal hypothalamus. The characterization of these zfERs provides a new perspective for understanding the mechanisms underlying estradiol actions in a vertebrate species commonly used for developmental studies.
central nervous system, estradiol, estradiol receptor, hypothalamus, mechanisms of hormone action, neuroendocrinology, steroid hormones, steroid hormone receptors
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ERß, have been cloned from different mammalian and nonmammalian vertebrates (e.g., human [1, 2], rat [3, 4], quail [5], eel [6], catfish [7], seabream [8]). Like the other members of the nuclear receptor superfamily, all these proteins share a common structure-function organization in different functional modules [9]. The N-terminal A/B domain regulates gene transcription in a promoter- and cell-specific manner through its AF1 activation function. The highly conserved C domain contains two zinc fingers that enable the receptor to bind DNA on specific palindromic sequences, the estrogen responsive elements (EREs). The intermediary D domain is poorly conserved but appears necessary to the maintenance of the ER tridimensional structure [10, 11]. In the C-terminal region, the E/F domain is involved in the binding of specific ligands and plays a key role in the ligand-dependent transactivation function (AF2).
The two ER subtypes in mammals, ER and ERß, present a similar overall structure, but in vitro they have different functional characteristics [12–16], notably in their ligand affinities and transactivation capacities. These receptors exhibit a wide distribution in reproductive and nonreproductive adult rat tissues but present different expression patterns as shown by reverse transcription polymerase chain reaction (RT-PCR) [12] or as demonstrated by in situ hybridization and immunohistochemistry in the brain [17, 18] and in gonads [19]. However, in several diencephalic regions, a strong overlapping of ER and ERß mRNAs was reported [17]. Within these regions, double-labeling experiments demonstrated that the two ER subtypes are expressed in different cells but also coexist in the same neurons [18]. Furthermore, during mouse embryogenesis, ER and ERß are characterized by a differential spatiotemporal expression [20]. Altogether, these results suggest specific functions for each ER subtype in adult and during development, but at the present stage the respective physiological roles of each ER protein remain unclear.
Very recently, two ER forms, phylogenetically related to the tetrapod ERß subtype, have been isolated in two fish species [21, 22]. These fish ERßs exhibit clear structural differences throughout their entire coding region, indicating the existence of two distinct ERß forms originating from different genes in these species. However, information on the hormonal specificity, binding characteristics, and transactivating capacities of these new teleost ERßs, notably in comparison with ER, has been scarce, and their precise distribution within the central nervous system is unknown. For this reason, we decided to study the estradiol binding affinities, the hormonal specificity, the transcriptional activities, the expression pattern in different tissues, and the brain distribution of three ERs (zfER, zfERß1, and zfERß2) within a teleost model now widely used for development studies, the zebrafish.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, ERß1, ERß2 cDNA
One microgram of total RNA from estradiol-treated adult (males and females) zebrafish liver was reverse transcribed with oligodT by expand reverse transcriptase (Boehringer Mannheim, Mannheim, Germany). The entire coding region of zfER cDNA (GenBank AB037185) has been amplified by PCR with oligonucleotides P1 (5'-CTCTCACCCATGTACCCTAAGGA-3') and P2 (5'-CTCTCACCCATGTACCCTAAGGA-3'). The zfERß1 cDNA has also been amplified by PCR with primers ß1P1 (5'-CACATCACCTGCTGCCTGCTAA-3') and ß1P2 (5'-ACTGATGGATGGATGAATGAAATGCC-3'), chosen on both sides of the zfERß coding region (GenBank AJ275911). PCR products were visualized by electrophoresis on a 1% agarose gel, subcloned in Topo-pCDNA3 expression vector (Invitrogen, San Diego, CA), and fully sequenced.
GenBank accession BF717864 was defined for a partial sequence (401 base pairs [bp]) of a putative ER obtained by systematic sequencing of the zebrafish genome [23]. We obtained this clone from I.M.A.G.E. Consortium (ID 2601181) and fully sequenced it. This clone corresponds to full length zfERß2 cDNA. The entire coding region has been amplified by PCR using primers ß2P1 (5'-GATAGTAGTGTGGTATAC-3') and ß2P2 (5'-TTCGTCTTGTGAAATCCTACT-3') and then subcloned in Topo-pCDNA3 expression vector. All cloned fragments were sequenced by the PRISM Ready Reaction Big Dye Terminator cycle sequencing protocol (PE Biosystems, Courtaboeuf, France).
Sequence Analysis
Multiple alignments and identity analysis were performed and a phylogenetic tree was generated with Clustal W [24]. The phylogenetic tree was constructed using the following ER and estrogen receptor-related receptor (ERR) amino acid sequences available from GenBank with their hypervariable A/B domains deleted. Sequence alignment was performed according to the Lipman-Pearson algorithm using the Dayhoff matrix for amino acid residue homology. The sequence sources and GenBank numbers are as follows: Atlantic croaker ER (acER, AAG16713), Atlantic croaker ERß (acERß, AAG16711), Atlantic croaker ER
(acER, AAG16712), catfish ER (cfER, AAC69548), catfish ERß (cfERß, AAF63157), chicken ER (chickER, CAA27433), chicken ERß (chickERß, BAA88667), Japanese eel ER (eER, BAA19851), gilthead seabream ER (gsER, CAB51479), gilthead seabream ERß (gsERß, AAD31033), goldfish ERß (gfERß, AAD26921), goldfish ERß2 (gfERß2, AAF35170), hamster ER (hmER, AAD53956), human ER (hER, X03635), human ERß (hERß, AF051427), human ERR1 (hERR1, P11474), human ERRß (hERRß, NP004443), human ERRß2 (hERRß2, AAC99409), human ERR (hERR, NP001429), medaka ER (mdER, P50241), mouse ER (mER, M38651), mouse ERß (mERß, CAA03949), quail ERß (qERß, AAC36463), rainbow trout ERß (rtERß, CAC06714), rainbow trout ER long (rtER L, CAB45139), rainbow trout ER short (rtER s, CAB45140), rat ER (rER, CAA68287), rat ERß (rERß, AAC52602), tilapia ER1 (tER1, AAD00245), tilapia ER2 (tER2, AAD00246), Xenopus ER (xER, P81559), zebra finch (finchER,AAB81108), zebrafish ER (zfER, AB037185), zebrafish ERß1 (zfERß1, AJ414566), zebrafish ERß2 (zfERß2, AJ414567).
In Vitro Transcription/Translation and Binding Analysis
The in vitro translation reactions were performed using 1 µg of each expression vector and T7 RNA polymerase in a rabbit reticulocyte lysate to synthesize in vitro zfER proteins. Reactions were performed at 30°C for 90 min as recommended by the supplier (Quick TNT; Promega, Madison, WI). 35S-Methionine-labeled proteins were visualized by autoradiography after SDS-PAGE.
For estradiol-binding assay, 2–4 µl of unlabeled lysate containing each zfER was diluted in a total volume of 50 µl of TEG (10 mM Tris-HCl, pH 7.4, 1.5 mM EDTA, 10% glycerol) and incubated for 16 h at 4°C, with 0.155–5 nM [2.4.6.7-3H]estradiol-17ß (E2, specific radioactivity 99 Ci/mmol; Amersham, Uppsala, Sweden). Nonspecific binding was determined in the presence of a 150-fold excess of unlabeled E2. Bound and unbound fractions were separated by charcoal/dextran treatment. After counting, specific binding was obtained by subtracting nonspecific binding from total binding. The dissociation constant (Kd) was determined by linear Scatchard transformation [25] and was the mean of three independent experiments, each performed in duplicate.
Cell Culture and Transfection Assays
Chinese hamster ovary (CHO) cells were maintained in Dulbecco modified Eagle medium with F12 (DMEM-F12; Sigma, St. Louis, MO) supplemented with 5% fetal calf serum (FCS; Life Technologies, Carlsbad, CA), 100 U/ml penicillin, 100 µg/ml streptomycin, and 25 µg/ml amphotericin (Sigma). Cells were cultured at 37°C in 5% CO2 atmosphere. Cells were seeded in 24-well plates with phenol red-free DMEM-F12 containing 2% of charcoal/dextran FCS 24 h before transfection. Transfections were performed at 50–60% confluence by a CaPO4/DNA coprecipitation method 1 h after changing the medium. Transfected DNA contained 5 ng of expression vector (with or without the coding region of each zfER), 50 ng of ERE-TK-Luc reporter, 150 ng of internal ß-galactosidase control vector (PCH110), and 795 ng of DNA carrier pBluescript plasmid. Cells were shocked overnight at 37°C in 2% CO2 and washed twice with a saline phosphate buffer solution, and medium was replaced with fresh phenol red-free DMEM-F12 (without FCS) containing ethanol (0.1%) with or without hormones. For the dose effect analyses, cells were treated with E2 at 10-12 M, 5 x 10-12 M, 10-11 M, 5 x 10-11 M, 10-10 M, 10-9 M, 10-8 M, and 10-7 M. For steroid specificity analysis, cells were treated with 10-8 M E2, 10-6 M ICI 164384 (ICI), 10-6 M 4-hydroxytamoxifen (OHT), 10-8 M E2 with 10-6 M ICI, 10-8 M E2 with 10-6 M OHT, 10-8 M diethylstilbestrol (DES), 10-8 M estrone (E1), 10-8 M estriol (E3), 10-8 M testosterone, and 10-8 M progesterone. After 36 h of treatment, cells were lysed, and 10% of the cellular extract was used to measure luciferase activity (luciferase assay system; Promega). ß-Galactosidase activity was used to normalize transfection efficiency in all experiments, which were performed in triplicate. Results were expressed in fold induction versus promoter activity and corresponded to the mean of three independent experiments. Differences between the control vector and each expression vector were analyzed using an ANOVA with the Fisher probable least squares difference test.
Qualitative RT-PCR
To analyze the distribution of mRNA zfER, zfERß1, and zfERß2, tissues from five to nine preovulatory females (brain, pituitary, liver, ovary, intestine, eyes) and five mature males (testis) were collected. Total RNA was extracted and reverse transcribed with random hexamers (Boehringer Mannheim). Complementary DNA fragments corresponding to the C-terminal region of each zfER were amplified by PCR (30 cycles) with the following primers: P3 (5'-CAGTAAATCAGGAGCGTCGCT-3') and P4 (5'-ATGTGTATATGGTGGCAGTGC-3'), ß1P3 (5'-ATGTGGAGAGTCGCGGGAA-3') and ß1P4 (5'-ACTGATGGATGGATGAATGAAATGCC-3'), and ß2P3 (5'-AGACCGGCCTTAGCTTCCAG-3') and ß2P4 (5'-GTCCATCCTCCCGAAACTACAGA-3'). The acidic ribosomal phosphoprotein PO mRNA [26, 27] was also amplified by PCR (27 cycles) as an internal control with the primers PO-1 (5'-AAYGTGGGCTCCAAGCACATG-3') and PO-2 (5'-GAGATGTTCAGCATGTTCAGC-3'). In each case, PCRs were also done without cDNA, as a negative control. Amplified PCR products were visualized on a 1.5% agarose gel stained by ethidium bromide.
Slot-Blot Assay and In Situ Hybridization
The specificity of each riboprobe used in the in situ hybridization protocol was first verified by slot-blot assay [28]. Expression vectors containing the full coding cDNA regions of zfER, zfERß1, and zfERß2 were used to synthetize in vitro the corresponding mRNAs with T7 RNA polymerase. After DNase I treatment and purification, 0.8 ng, 3.1 ng, and 12.5 ng of each mRNA were spotted onto a nylon Biodyne A membrane (Pall Gelman Sciences, Ann Arbor, MI). Membranes were hybridized with each [-32P]dCTP radiolabeled cDNA probe corresponding to the same sequence used for riboprobes. After highly stringent washing, autoradiographs were produced from the blots.
For radioactive in situ hybridization, three riboprobes were generated. The zfER riboprobe was synthetized from a fragment of 782 bp cloned into pBluescript and corresponding to the D/E domain of the zbER. The zfERß1 riboprobe was obtained from a fragment of 851 bp inserted into pBluescript and overlapping the D/E domain of the zbERß1. For the zfERß2 riboprobe, a fragment corresponding to 268 bp of the A/B domain and 140 bp of the 5' untranslated region was introduced into pCDNA3. For zfER and zfERß1, antisense and sense riboprobes were synthesized in vitro with T7 RNA polymerase with plasmids linearized with XhoI and with T3 RNA polymerase with plasmids linearized with BamHI. For zfERß2, antisense and sense riboprobes were generated with T7 RNA polymerase using plasmids linearized with XhoI. The template DNA was removed by DNase I treatment, and the radiolabeled riboprobes (35S-UTP) were separated from unreacted components on a Sephadex G50 column with ethanol precipitation. Labeled riboprobes were then resuspended in 0.1 M dithiothreitol.
In situ hybridization experiments were performed on five brains from adult female zebrafish. Animals anesthetized on ice were cut at the back of the head, and the skull was opened to expose the brain. Whole animals were then placed in a fixative solution (4% paraformaldehyde in 1 M phosphate buffer, pH 6.8) overnight at 4°C. Removed brains were postfixed for 24 h at 4°C in fresh fixative solution, rinsed in buffer, and dehydrated in a series of alcohols. Tissues were then embedded in paraffin, and cross sections (6 µm) were collected on Tespa-treated slides and stored at 4°C in boxes containing silica gel until in situ hybridization was performed, according to the protocol previously published [29]. After hybridization, sections were dehydrated and dipped into an Ilford K5 photographic emulsion. Slides were developed after 2 wk of exposure at 4°C, counterstained with toluidine blue (0.02%), and mounted with Depex. Slides were observed with an Olympus Provis photomicroscope under dark-field and bright-field illumination. Identification of brain nuclei was according to the zebrafish brain atlas [30].
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We characterized three cDNA clones. The predicted amino acid sequences of zfER, zfERß1, and zfERß2 cDNAs consisted of 569, 592, and 553 amino acid residues with estimated molecular masses of 63, 66, and 63kDa, respectively (Fig. 1). Our cDNA clone zfERß1 was obtained with primers corresponding to the sequence of GenBank zfERß (AJ275911); however, the alignment of the two sequences showed that GenBank zfERß possesses 24 additional nucleotides corresponding to 8 additional amino acid residues at the end of the C domain (data not shown). This last sequence could be an isoform of zfERß1, but these additional amino acid residues could not be properly aligned with the other ER sequences. Consequently, we submitted a new GenBank accession number for the sequence of zfERß1 (AJ414566).
|
Identity was 40.5% between zfER and zfERß1, 39.4% between zfER and zfERß2, and 51.5% between zfERß1 and zfERß2. However, the overall amino acid sequences for these receptors were clearly different (Fig. 2), indicating that each zfER protein was not an alternatively spliced isoform but was generated by a distinct gene. As shown in Figure 2, zfER was characterized by 47.1% identity with hER and only 40.9% identity with hERß, which suggests that zfER derives from an ER ancestral subtype rather than from an ERß subtype. This result was confirmed by phylogenetic analysis (Fig. 3), which clearly indicated that zfER belongs to the ER subgroup. Zebrafish ERß1 and zfERß2 had 46.8% and 51.5% identity, respectively, with hERß but only 37.8% and 39% identity with hER (Fig. 2). These sequence identities and the phylogenetic tree (Fig. 3) show that zfERß1 and zfERß2 are members of the ERß subgroup. In two other teleost species, two ERß members were recently reported, gfERß1 and gfERß2 in goldfish and acERß and acER in the Atlantic croaker [21, 22].
|
|
To identify zfER domains, each predicted amino acid sequence was aligned with human ERs (Fig. 2). The A/B domain is not a significantly conserved region, with only 8.4%–26.6% identity between zfERs and hERs (Fig. 2). Nevertheless, in human and zebrafish, one serine-proline motif was conserved in both ER and ERß (Fig. 1) and could be a potential phosphorylation site for the mitogen-activated protein kinase (MAPK) pathway [13, 31, 32]. As expected, the DNA binding domain (C domain) was highly conserved, with 82.1%–91.7% identity between zfER and hERs (Fig. 2). Each ER notably contained the eight cysteine residues involved in conformation of two zinc fingers (Fig. 1). Although the D domain appears poorly conserved (no significant identities; Fig. 2), alignment shows one conserved arginine residue surrounded by basic residues (Fig. 1). The C-terminal region (E/F domain) of zfERs has 55.2%–61.8% identity with human ERs. Five conserved motifs involved in E2 binding [33, 34] were found in each zfER (Fig. 1).
Estrogen Binding and Transcriptional Properties
As shown in Figure 4A, each expression vector containing the coding regions of zfER, zfERß1, and zfERß2 produced in vitro a single major protein with a molecular mass of approximately 65 kDa, as determined by SDS-PAGE. Estrogen binding affinity of each zfER protein, as determined by Scatchard analyses, showed a single population of binding sites with Kd values of 0.74 ± 0.2 nM for zfER (Fig. 4B), 0.75 ± 0.17 nM for zfERß1 (Fig. 4C), and 0.42 ± 0.06 nM for zfERß2 (Fig. 4D). These results indicate that all zfERs were able to bind E2 and that zfERß2 has a 1.8-fold higher affinity for E2 than do the two other forms.
|
The transcriptional capacities of zfER proteins were studied by cotransfection assays in CHO cells using zfER expression vectors and a luciferase reporter gene, ERE-TK-Luc. Dose response experiments have been carried out for each form. As shown in Figure 5, zfER and zfERß1 exhibited significant transcriptional activity at a dose of 5 x 10-11 M E2, whereas zfERß2 appeared to be slightly more active. Zebrafish ERß2 showed clear transcriptional activity at 10-11 M E2, in agreement with the Scatchard analyses showing that this receptor was characterized by a higher affinity for E2.
|
To investigate the ligand selectivity of each receptor, cotransfected cells were treated for 36 h with different steroids or ethanol. As expected, ICI or OHT did not activate zfERs, and a 100-fold excess of these compounds completely suppressed E2 stimulation of the reporter gene mediated by each zfER (Fig. 6). All three zfERs were activated by DES and by estradiol metabolites such as E1 and E3. In contrast, other steroids such as testosterone and progesterone were unable to induce expression of the reporter gene mediated by these zfERs (Fig. 6). As a control, the empty expression vector did not reveal any significant effects on the reporter gene activity after hormonal treatment (data not shown).
|
Tissue Distribution and Brain Localization of zfER, zfERß1, and zfERß2 mRNA-Expressing Cells
To study the tissue distribution of mRNA zfER, zfERß1, and zfERß2, a qualitative RT-PCR assay was performed with primers specific for each form (Fig. 7). Amplification products of the expected size were obtained for each form in reproductive tissues such as brain, pituitary, liver, ovary, and testis but also in the nonreproductive tissues such as intestine and eyes. These amplification products correspond to zfER mRNAs but not their genomic DNA. Based on the structure of the rainbow trout ER gene [35], all primers of zfERs have been chosen within the putative exon 9 and the putative exon 10. In each case, PCR was also performed with genomic DNA, and only fragments between 1 and 2 kilobases were observed (data not shown).
|
To first verify that each zfER riboprobe did not detect other mRNAs as a result of sequence similarities, in vitro synthesized zfER mRNAs were hybridized with each zfER cDNA probes in slot-blot assays. As shown in Figure 8, zfER, zfERß1, and zfERß2 probes recognized only their corresponding mRNA and did not show cross hybridization.
|
In the first in situ hybridization experiment, parallel tissue sections were treated alternately with sense and antisense probes. Hybridization with the sense probes resulted in a very weak and homogeneous background without preferential accumulation of grain clusters on particular cell bodies (data not shown). After 2 wk of autoradiography, positive hybridization signals were observed on brain sections with the three zfER antisense riboprobes. Signals were found consistently in the same brain regions in every fish: the preoptic area, hypothalamus, and posterior tuberculum. However, the signal intensity was always weaker in the two latter regions.
In the preoptic area, the three forms of zfER mRNA were detected in the anterior part of the parvocellular preoptic nucleus, but their respective distributions were not strictly identical, as observed on adjacent sections (Fig. 9). ER mRNA-expressing cells were exclusively localized in the most ventral zone of this nucleus (Fig. 9, A and B). A strong positive signal was observed with zfERß2 antisense riboprobe in that same ventral position, but zfERß2 signal also extended more dorsally with a lower intensity (Fig. 9, E and F). As shown in Figure 9, C and D, zfERß1 mRNA had a distribution corresponding to both the ventral and dorsal parts of the parvocellular preoptic nucleus; however, signal intensity was homogeneous over the whole area.
|
More caudally, zfER and zfERß2 mRNAs were detected in the ventral periventricular zone of the mediobasal hypothalamus (Fig. 10, A, B, E, and F), whereas zfERß1 mRNA could not be observed in that region (Fig. 10, C and D).
|
More dorsally, zfER and zfERß1 mRNA were present in the posterior tuberal nucleus (Fig. 10, A–D). However, their distributions were not strictly the same within this nucleus; the zfERß1 mRNA-expressing population was more restricted than that expressing zfER mRNA (Fig. 10, A–D). No hybridization signal was detected with the zfERß2 riboprobe in this nucleus (Fig. 10, E and F). The signal observed in the hypothalamus and the posterior tuberal nucleus was consistently weaker than that in the preoptic area.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ERß subtypes have been identified in fishes, as in mammals, the presence of multiple forms of ERß in some fishes has been recently reported [21, 22]. To understand the potential function of each receptor, we decided first to characterize these three ERs in the zebrafish, a well-studied model of vertebrate early development, organogenesis, toxicology, genetics, and phylogenetics.
Our phylogenetic analysis clearly revealed that zfER belongs to the ER subgroup, whereas zfERß1 and zfERß2 are members of the ERß subgroup. Moreover, the overall amino acid sequence identity between both zfERß forms and mammalian ERß is about 50%, whereas identity between both zfERß forms and mammalian ER is below 40%. According to a recent study, zfERß underwent duplication before the divergence of teleosts, whereas no duplication of the ER gene has been detected [36]. In agreement with this previous study and the present results, two ERs in the ERß subgroup have been identified in the goldfish [21] and in Atlantic croaker [22]. The absence of two clearly separated branches demonstrating the existence of two ERß forms in teleosts could be the result of an unsignificant bootstrap value of the branch containing the eel ER and the Atlantic croaker ERß. The proximity on the tree of ERß from these two phylogenetically distant species is rather suprising but suggests that the teleost ERß family is more complex than that of mammals, raising the question of the specific function of this additional ERß.
Alignment and identity results showed that the A/B domain of each zfER is hypervariable. In mammals, this region has been identified as the site for the functional differences between ER and ERß subgroups. Unlike hERß, the A/B domain of hER possesses a clear ligand-independent AF1 function that depends on the promoter and cell context [37]. Accordingly, the N-terminal region of ER fused to the heterologous GAL4-DBD shows a clearly autonomous and ligand-independent activity, whereas a GAL4-DBD fusion protein containing the N-terminal region of ERß shows no transcriptional activity in Hela or HepG2 cells [37]. Furthermore, the hERß deleted from the amino-terminal region was characterized by higher activity than the intact hERß [38]. These results suggest that the A/B domain of ERß could possess a repressor function that would be a target for corepressors, but the exact A/B domain function of ERß is undefined. The A/B domain of both zfERßs and of other teleost ERßs are clearly longer than those of mammalian ERß, suggesting that ERß may function differently in mammalian and nonmammalian species. However, sequence analysis of all ERs in this region revealed a conserved serine-proline motif that corresponds to serine 118 in hER. This amino acid residue is a direct substrate for Ras-Raf-MAPK, and phosphorylation of serine 118 was required for the full activity of hER [31, 32]. Likewise, other experiments carried out on mice [13] revealed that mERß transcriptional activity can be stimulated by MAPK and requires an intact N-terminal region and notably serine 60. The growth factors are also able to activate MAPK [39], and a clear cross-talk between growth factors and ER has been demonstrated in mammals [40, 41]. Consequently, it will be interesting to investigate whether this serine residue of zfERs can be phosphorylated by growth factor-activated pathways.
The C domain (DNA binding) is the most highly conserved region. In transfection experiments, zfER, zfERß1, and zfERß2 induced expression of a luciferase reporter gene monitored by one ERE. This finding suggests that all three zfERs could bind the same ERE, as demonstrated in rat for ER and ERß. However, Vanacker et al. [42] recently reported that ER transactivates a target gene through binding on a SF-1 response element site, whereas ERß was unable to do so. Thus, although the C domain is conserved between each ER form, one cannot exclude the possibility that different DNA cis-elements could mediate specific function of each zfER subtype in vivo.
The intermediary D domain is a less conserved region; however, a basic arginine residue seems to be conserved among all ERs. Moreover, in all cases, this residue is surrounded by some other basic residues. These charges could participate in ER secondary structure and in the stabilization of DNA binding [10, 11]. Furthermore, Wang et al. [43] recently reported that lysine residues within the D domain of ER can be directly acetylated by the nuclear receptor cointegrator P300. Substitution of these residues with charged or polar residues modifies ER transactivation and hormonal sensitivity.
In the carboxy-terminal region, the E/F domain is highly conserved and is known to contain a ligand-dependent transactivation site, AF2 [44]. The AF2 function involves coactivators that interact with helix H12 and are recruited in an estrogen-dependent manner. This transactivation function is similar for ER and ERß [45] and probably exists for each zfER. The tertiary structure of the carboxy-terminal region, which also contains helices H3, H6, H8, H11, and H12 surrounding the hydrophobe cavity [34], is relatively invariant among mammalian ERs and explains why ER and ERß bind estrogen with approximately the same affinity [12, 46]. Those particular tertiary structures apparently are also present in the E/F domain of each zfER form.
As demonstrated by binding assays, the zfER proteins were characterized by a high affinity for E2. The calculated Kd for zebrafish ERs is close to the ER Kd in mammals (0.2–0.5 nM). In a recent study comparing point mutants of human and rainbow trout ER, two residues within the C-terminal domain of hER (L349 and M528) are implicated in ligand-binding stability at different temperature [47]. Unlike rainbow trout ER, these residues are conserved in zfERs, which partly explains their higher affinity for E2 compared with other fish ERs [22, 48]. Transfection experiments carried out in CHO cells showed that all zfERs are able to induce a reporter gene driven by an ERE in a strictly estrogen-dependent manner. Dose-effect experiments revealed that the zfERß2 is high sensitive to E2 and is able to induce a reporter gene at a minimal concentration of 10-11 M E2. This high sensitivity for E2 could be due to a higher affinity for E2, as demonstrated by our Scatchard analysis, and/or a better recruitment of coactivators. The transcriptional activities of the zfERßs were approximately equal to that of zfER. However, transitory expression experiments in mammals have revealed that ERß has weaker transcriptional activity than does ER [13, 45], suggesting that teleost ERßs could be more potent transactivators.
As demonstrated by our qualitative RT-PCR experiments, all zfER mRNAs are detected in different tissues tested, such as brain, pituitary, liver, gonads, intestine, and eyes. To investigate whether these mRNAs are expressed in the same cell populations, in situ hybridization was carried out on an heterogenous E2 target tissue, the brain. A differential but partly overlapping distribution was found in the preoptic region, mediobasal hypothalamus, and posterior tuberculum, highlighting the fact that the E2 targets within the teleost brain mainly correspond to neuroendocrine regions. Concerning the distribution of ER, these results confirm those obtained in rainbow trout at both the mRNA and protein levels [49–51]. However, in a recent study in trout the rtER gene generated two classes of transcripts, resulting in short (deleted from the A domain) and long rtER proteins [52], but only the long isoform was expressed in the brain and pituitary [53]. The same situation probably exists in zebrafish, but further investigation is needed. There has been very little published information on the distribution of ERßs in the fish brain. In a previous study in Atlantic croaker, different ER mRNAs had distinct distributions in the brain, with the acER form localized in the preoptic area, acERß expressed in cells closed to the third ventricule, and acER preferentially situated more laterally in a structure identified as the suprachiasmatic nucleus [22]. Results in rodent brain have also shown a differential localization of ER and ERß mRNA in many cerebral regions, with some structures expressing exclusively one mRNA subtype and others expressing both [54].
Although each ER subtype exhibits a particular pattern of expression, there is considerable overlap in the distribution of those subtypes. This overlap is particularly obvious in the ventral part of the parvicellular area of the preoptic nucleus, where the three ER mRNA subtypes are strongly expressed. Evidence for colocalization of ER and ERß mRNA has been highlighted in the rat brain. A combination of in situ hybridization and immunohistochemistry revealed that numerous cell populations expressed both transcripts, with a high rate of coexpression in the preoptic area, bed nucleus of the stria terminalis, and amygdala [18, 54]. In zebrafish, the possible expression of two or three zfER forms in a single cell raises the question of the mode of regulation of E2 target genes. Each form could regulate different genes implicated in different physiological processes. These forms may also participate in different activation pathways for the same target gene. As demonstrated recently, ER and ERß interact differently with nuclear receptor cofactors or with the basal transcription machinery [15, 16]. Additionally, the zfER proteins could interact to form heterodimers, making more complex the potential mechanisms underlying estrogen physiological actions. In mammals, ER and ERß preferentially form heterodimer units with a transcriptional activity similar to that of the ER homodimer [55–57], suggesting that in vivo estrogen actions could depend on the ratio of each ER subtype in a given cell.
Currently, there is no information regarding the respective roles of these forms in the zebrafish brain during development or in the adult. In the mouse, invalidation of the ER subtype results in an increase in the number of dopaminergic neurons in the sexual dimorphic anterior ventral periventricular nucleus of males, leading to a feminization of this nucleus [58]. In contrast, ßERKO male mice exhibit a regional neuronal hypocellularity in different brain regions, including the medial preoptic area and median amygdala, whereas there is a strong proliferation of astroglial cells [59]. These data indicate that ER and ERß are probably involved in neuronal apoptosis and survival during development.
This study provides more detailed information in a model vertebrate species, the zebrafish, on three functional ER forms, zfER, zfERß1, and zfERß2, which in vitro are able to bind E2 and to transactivate a reporter gene. The three corresponding mRNAs are expressed in different reproductive and nonreproductive tissues and are characterized by a partial overlapping in the adult brain. These results tend to suggest that the mechanisms underlying the physiological actions exerted by estradiol in teleosts are more complex than originally believed and may notably be governed by the specific equipment of the cell in ER subtypes. These results pave the way for further investigations regarding the specific functions and mechanisms of actions of these three ERs in adult and developing zebrafish.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This work was supported by grants from the CNRS, the European Union, and the Fondation Langlois. A.M. is the recipient of a fellowship from the French Ministry of Research and Education. 
2 Correspondence. FAX: 33 2 23 23 67 94; farzad.pakdel{at}univ-rennes1.fr
Accepted: January 9, 2002.
Received: September 25, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ß are unstructured in vitro and show different TBP binding properties. J Biol Chem 2001; 276:45939-45944This article has been cited by other articles:
|
|
 |
|
  X. Liu, P. Zhu, K. W.Y. Sham, J. M.L. Yuen, C. Xie, Y. Zhang, Y. Liu, S. Li, X. Huang, C. H.K. Cheng, et al. Identification of a Membrane Estrogen Receptor in Zebrafish with Homology to Mammalian GPER and Its High Expression in Early Germ Cells of the Testis Biol Reprod, June 1, 2009; 80(6): 1253 - 1261. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Mouriec, M.-M. Gueguen, C. Manuel, F. Percevault, M.-L. Thieulant, F. Pakdel, and O. Kah Androgens Upregulate cyp19a1b (Aromatase B) Gene Expression in the Brain of Zebrafish (Danio rerio) Through Estrogen Receptors Biol Reprod, May 1, 2009; 80(5): 889 - 896. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Zhang, W. Zhang, H. Yang, W. Zhou, C. Hu, and L. Zhang Two cytochrome P450 aromatase genes in the hermaphrodite ricefield eel Monopterus albus: mRNA expression during ovarian development and sex change J. Endocrinol., November 1, 2008; 199(2): 317 - 331. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Lange, Y. Katsu, R. Ichikawa, G. C. Paull, L. L. Chidgey, T. S. Coe, T. Iguchi, and C. R. Tyler Altered Sexual Development in Roach (Rutilus rutilus) Exposed to Environmental Concentrations of the Pharmaceutical 17{alpha}-Ethinylestradiol and Associated Expression Dynamics of Aromatases and Estrogen Receptors Toxicol. Sci., November 1, 2008; 106(1): 113 - 123. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Cohen, M. Shmoish, L. Levi, U. Cheruti, B. Levavi-Sivan, and E. Lubzens Alterations in Micro-Ribonucleic Acid Expression Profiles Reveal a Novel Pathway for Estrogen Regulation Endocrinology, April 1, 2008; 149(4): 1687 - 1696. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. M. Schaaf, D. Champagne, I. H. C. van Laanen, D. C. W. A. van Wijk, A. H. Meijer, O. C. Meijer, H. P. Spaink, and M. K. Richardson Discovery of a Functional Glucocorticoid Receptor {beta}-Isoform in Zebrafish Endocrinology, April 1, 2008; 149(4): 1591 - 1599. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Cheshenko, F. Brion, Y. Le Page, N. Hinfray, F. Pakdel, O. Kah, H. Segner, and R. I.L. Eggen Expression of Zebra Fish Aromatase cyp19a and cyp19b Genes in Response to the Ligands of Estrogen Receptor and Aryl Hydrocarbon Receptor Toxicol. Sci., April 1, 2007; 96(2): 255 - 267. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. F Orlando, Y. Katsu, S. Miyagawa, and T. Iguchi Cloning and differential expression of estrogen receptor and aromatase genes in the self-fertilizing hermaphrodite and male mangrove rivulus, Kryptolebias marmoratus. J. Mol. Endocrinol., October 1, 2006; 37(2): 353 - 365. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I M McGonnell and R C Fowkes Fishing for gene function - endocrine modelling in the zebrafish. J. Endocrinol., June 1, 2006; 189(3): 425 - 439. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Liu, H. Wang, and Z. Gong Tandem-Repeated Zebrafish zp3 Genes Possess Oocyte-Specific Promoters and Are Insensitive to Estrogen Induction Biol Reprod, June 1, 2006; 74(6): 1016 - 1025. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Y. Kunz, H. F. Galicia, and K. Fent Comparison of In Vitro and In Vivo Estrogenic Activity of UV Filters in Fish Toxicol. Sci., April 1, 2006; 90(2): 349 - 361. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. D. Costache, P. K. Pullela, P. Kasha, H. Tomasiewicz, and D. S. Sem Homology-Modeled Ligand-Binding Domains of Zebrafish Estrogen Receptors {alpha}, {beta}1, and {beta}2: From in Silico to in Vivo Studies of Estrogen Interactions in Danio rerio as a Model System Mol. Endocrinol., December 1, 2005; 19(12): 2979 - 2990. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A.L. Filby and C.R. Tyler Molecular Characterization of Estrogen Receptors 1, 2a, and 2b and Their Tissue and Ontogenic Expression Profiles in Fathead Minnow (Pimephales promelas) Biol Reprod, October 1, 2005; 73(4): 648 - 662. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.B Hawkins, J Godwin, D Crews, and P Thomas The distributions of the duplicate oestrogen receptors ER-{beta}a and ER-{beta}b in the forebrain of the Atlantic croaker (Micropogonias undulatus): evidence for subfunctionalization after gene duplication Proc R Soc B, March 22, 2005; 272(1563): 633 - 641. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. B. Hawkins and P. Thomas The Unusual Binding Properties of the Third Distinct Teleost Estrogen Receptor Subtype ER{beta}a Are Accompanied by Highly Conserved Amino Acid Changes in the Ligand Binding Domain Endocrinology, June 1, 2004; 145(6): 2968 - 2977. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. P. Daston, J. C. Cook, and R. J. Kavlock Uncertainties for Endocrine Disrupters: Our View on Progress Toxicol. Sci., August 1, 2003; 74(2): 245 - 252. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
