Identification of a Second egfr Gene in Xiphophorus Uncovers an Expansion of the Epidermal Growth Factor Receptor Family in Fish
http://www.100md.com
分子生物学进展 2004年第2期
Physiological Chemistry I, Theodor Boveri Institute for Biosciences (Biocenter), University of Würzburg, Würzburg, Germany
E-mail: claudia.wellbrock@icr.ac.uk.
Abstract
The epidermal growth factor receptor (EGFR) gives name to a family of receptors formed by four members in mammals (EGFR, ErbB2, ErbB3, and ErbB4). Members of this family can be activated to become potent oncogenes, and many human and animal tumors express qualitatively or quantitatively altered receptors from this group. We have isolated and characterized a second egfr gene in the melanoma model fish Xiphophorus. Both Xiphophorus egfra and egfrb duplicates are co-orthologs of the mammalian egfr gene. Database analysis showed that not only egfr but also erbB3 and erbB4 are present as duplicates in some fish species. They originated from ancient duplication events that might be consistent with the hypothesis of a fish-specific genome duplication. In Xiphophorus, the egfrb gene underwent a second duplication that generated the melanoma-inducing oncogene Xmrk. The study and comparison of some of the functional characteristics of both Xiphophorus EGF receptors, including expression profile, ligand-binding abilities, and intracellular signal transduction revealed that Xiphophorus Egfra not only shares common features with Egfrb and the human EGFR but also shows significant differences in its functional characteristics. The mechanism of maintenance of these duplicates remains to be clarified.
Key Words: EGF receptor ? gene duplication ? receptor tyrosine kinase ? Xiphophorus
Introduction
Receptor tyrosine kinases (RTKs) comprise a diverse group of enzymes with intrinsic tyrosine kinase activity and a similar topology (Ullrich and Schlessinger 1990). All RTKs contain an extracellular domain where ligand binding takes place. A single transmembrane stretch connects the extracellular region with the intracellular domains, which include a highly conserved catalytic domain and a C-terminal tail that binds different intracellular signal transducers. Ligand binding and subsequent dimerization stimulate the receptor enzymatic activity. This results in autophosphorylation of their cytoplasmic domains and initiates a signaling cascade of protein phosphorylations inside the cell, providing a control on fundamental cellular processes such as growth and differentiation (Schlessinger 2000).
On the basis of sequence similarity and structural characteristics, RTKs have been grouped into families. The epidermal growth factor receptor (EGFR) family, or subclass I, of RTKs is one of the best studied. In mammals, this family consists of four members: EGFR/ErbB1, HER2/ErbB2, HER3/ErbB3, and HER4/ErbB4. The extra- cellular domain of these receptors includes four sub-domains. Subdomains S1 and S2 (or II and IV) are cysteine-rich regions whose spacing and number of cysteines is conserved among the members of the family; subdomains L1 and L2 (or I and III) are involved in ligand binding. EGF and a large group of related growth factors bind and activate specifically distinct receptors of this family, leading to the formation of homodimers and heterodimers. This results in different combinations of intracellular signals diversifying the biological responses triggered by these receptors (Olayioye et al. 2000).
The EGFR was the first member characterized in this family. It has been isolated in several mammals (Ullrich et al. 1984; Petch et al. 1990; Luetteke et al. 1994) and in chicken (Lax et al. 1988). The Drosophila Der receptor (Livneh et al. 1985) and let-23 from Caenorhabditis elegans (Aroian et al. 1990) are the single known invertebrate subclass I RTKs in flies and nematodes.
In the fish Xiphophorus, one receptor that belongs to the EGFR family has been isolated so far. It was named Xmrk (Xiphophorus melanoma receptor kinase) because its overexpression together with activating mutations induces melanoma formation in this fish (Wittbrodt et al. 1989; Gomez et al. 2001). Despite the high similarity of Xmrk with the human EGFR (hEGFR), there was evidence that this might be a new member of the subclass I gene family not described in any other species so far. Two copies of the Xmrk gene, the oncogene and the corresponding proto-oncogene have been identified. The proto-oncogene INV-Xmrk is invariably (INV) present in all teleost fish species tested so far (Dimitrijevic et al. 1998), but its physiological role remains to be elucidated. The oncogene can be found only in some species of Xiphophorus and represents an additional copy of the proto-oncogene produced by a recent gene duplication event. It is under a different transcriptional control and, as a result of this, is overexpressed in certain hybrids, leading to neoplastic transformation of the pigment cells (Adam, Dimitrijevic, and Schartl 1993).
The human receptors EGFR and Her2 are implicated in the development of numerous human cancers and appear constitutively active in a large number of tumors (Tang 1998; Olayioye et al. 2000). The physiological role of the EGFR has been extensively demonstrated in mammals. Knock-out mice for this gene present strain-dependent phenotypes that range from placental defects to postnatal abnormalities in multiple organs. The broad range of affected organs shows that this receptor is involved in a wide variety of cellular activities (Miettinen et al. 1995; Sibilia and Wagner 1995; Threadgill et al. 1995).
Some fish species are now generally accepted, valuable model systems for studying the normal function of molecules and developmental process as well as their malfunction in disease conditions. In contrast to the numerous studies carried out in mammals, little is known about the characteristics and action of the EGFR in fish. Only some reports exist in goldfish and rainbow trout that show the effect of heterologous EGF as multifunctional effector in the ovary and its direct effect in oocyte apoptosis and DNA synthesis (Srivastava and van der Kraak 1995; Wood and van der Kraak 2002). The use of heterologous antibodies has revealed an immunoreactivity in medaka embryos (Boomsma, Scott, and Walters 2001), rainbow trout liver (Newsted and Giesy 2000), and goldfish ovarian follicles (MacDougall and van der Kraak 1998). However, with the exception of the Xiphophorus Xmrk genes, none of the genes coding for the putative receptors had been isolated so far.
We describe here the isolation and characterization of a Xiphophorus egfr gene (egfra) (GenBank accession number AY230135) and present evidence that both the proto-oncogene INV-Xmrk (now egfrb) and egfra have been generated by a fish-specific duplication of an ancestral egfr gene. We show that, in fact, three members of the egfr family have been duplicated in fish. This finding is in accordance with the hypothesis of an ancient whole-genome duplication in the ray-finned fish lineage. The identification of egfra uncovers the existence of three egfr-related genes (egfra, egfrb, and Xmrk) and, thus, hypothetically eight members of the egfr family in Xiphophorus. Further analysis revealed that the new Xiphophorus egfra gene codes for a functional tyrosine kinase. It not only shares common features with Egfrb and with the human EGFR but also shows significant differences in its functional characteristics.
Materials and Methods
Library Screening and Isolation of Clones
A gt10 cDNA library from the Xiphophorus xiphidium embryonic cell line A2 (1.2 x 1010 pfu/ml) was screened at low-stringency conditions (40% form-amide/42°C) according to standard methods (Sambrook, Fritsch, and Maniatis 1989). The cDNA from Xmrk was used as probe. A second screen was performed to isolate the complete cDNA of egfra. In this case, a 1.45-kb PCR fragment containing part of egfra extracellular domain was used as a probe at high stringency conditions for hybridization (50% formamide/42°C). DNA fragments used as probes were randomly labeled with [-32P]dCTP.
DNA fragments contained in isolated phages were subcloned in pBlueScript using standard methods. Unidirectional, sequential deletions of pBlueScript inserts were obtained by exonuclease III-S1 digestion using a kit from Promega. Automatic sequencing was performed using ThermoSequenase fluorescent-labeled primer cycle sequencing kit and ALF automated sequencers (Amersham Pharmacia Biotech). Manual sequencing was performed using Sequenase (USB) and [-35S]dATP.
Sequence Analysis
Multiple sequence alignments were generated using "PileUp" of the GCG Wisconsin package version 10.0 (Genetics Computer Group, Madison, Wis.). Protein phylogenies were done with PAUP* (D.L. Swofford, Sinauer Associates, Inc., Sunderland, Mass.) with maximum parsimony (bootstrap analysis, 100 replicates), neighbor-joining (bootstrap analysis, 1,000 replicates [Saitou and Nei 1987]), and maximum likelihood (quartet puzzling, 10,000 puzzling steps [Strimmer and von Haeseler 1996]). Neighbor-joining was performed either considering all sites or only the unsaturated fraction of sites using AsaturA, a program recently applied to fish phylogenies (Van de Peer et al. 2002). The pairwise number of synonymous versus nonsynonymous substitutions per site between two aligned sequences was estimated with "Diverge" of GCG using an unambiguous alignment of DNA sequences (1,767 nt in length). The estimation of the number of substitutions per site at third codon positions was corrected for multiple events according to (Tajima and Nei 1984). Gene structure was analyzed using programs available on the NIX server (http://menu.hgmp.mrc.ac.uk/menu-bin/Nix). Deduced sequence assemblies from Takifugu rubripes and Danio rerio are available from J.-N.V on request.
Northern Blot Analysis
Total RNA from adult X. xiphidium organs and cell lines was isolated with TRIZOL reagent (GIBCO-BRL, Grand Island, N.Y.), and 30 μg of total RNA per lane were electrophoresed in a denaturing formaldehyde gel and blotted onto a Hybond-N nylon membrane (Amersham Pharmacia Biotech). Prehybridizations and hybridizations were carried out in 50% formamide at 42°C. DNA fragments used as probes were labeled as described above. Filters were washed at a maximum stringency of 1 x SSC/1% SDS at 60°C to 65°C. To control equal loading and integrity of the RNA, filters were stained with 0.04% methylene blue.
Construction of Expression Plasmids
To obtain an expression construct for egfra, its entire coding sequence was removed from the plasmid pBlueScript-egfra by EcoRI-SpeI digestion and ligated to the EcoRI and NheI digested expression vector pRK5(-HindIII) (Wittbrodt et al. 1992). This resulted in the expression construct pRK5-egfra. The chimeric construct HER-XER was generated in several steps. First, a silent mutation was introduced by PCR in the sequence of egfra (C2186 to G replacement) to generate an NarI restriction site. This newly introduced NarI site is located at the end of the transmembrane domain, in the same position where hegfr contains an NarI site. A 2-kb NarI/SpeI fragment containing the intracellular domains of egfra (XER) was ligated to a 6.9-kb NarI/NheI fragment containing the expression vector pRK5, and the region encoding the extracellular and transmembrane domains of the hEGFR (HER). This last fragment was obtained from the HER-mrk plasmid (Wittbrodt et al. 1992), which encodes the extracytoplasmic domains of hEGFR fused to the intracellular domains of Xmrk. The ligation sites and the egfra fragment amplified by PCR were checked by sequencing.
Cell Culture and Thymidine Incorporation
The IL-3-dependent mouse pro–B-cell line Ba/F3 (Palacios and Steinmetz 1985) was cultured in RPMI 1640, with 5% FCS and 5% of supernatant from X63Ag8-653 BVP m-IL-3 expressing cells (Karasuyama and Melchers 1988). BaF XER, BaF HER-XER, and BaF INV transfectants expressing Egfra, HER-Egfra, and Egfrb, respectively, were generated by electroporation (Easyject [Eurogentec] 200 V, 1800 μF) of BaF/3 cells with the expression plasmids pRK5-egfra, HER-XER, and pRK5-INV (Dimitrijevic et al. 1998) and a hygromycin-resistant plasmid (tgCMV/HyTK). After 10 days in medium containing hygromycin B (1 mg/ml), resistant clones were selected and tested for receptor expression on a western blot. Before stimulation with EGF or IL-3, the cells were washed twice with PBS and starved for 6 h in RPMI. Starved cells were treated with IL-3 (5% sup.) or the specified growth factor for the indicated times and collected by centrifugation. For thymidine incorporation assays, 4 x 104 cells per well of a 96-well plate were treated with the indicated factors and after 20 h, 0.5 μCi of 3H-thymidine per well was added for 4 h, and cells were harvested onto glass fiber filters. Incorporated 3H-thymidine was quantified by liquid scintillation counting, and IL-3 induced incorporation was set 100%.
Antibodies
mAb-108.1 is a mouse monoclonal antibody directed against the extracellular domain of hEGFR. 5E.2 was used as antiphosphotyrosine antibody. Both were kindly provided by A. Ullrich. Anti-STAT5b (C-17), anti-STAT3 (C-20), and anti-ERK2 (C-14) were from Santa Cruz Biotechnology. Rabbit polyclonal anti-phospho Akt and anti-Akt were obtained from New England Biolabs. Anti-phospho MAPK clone D12 was from Nanotools (Germany). For western blot analysis, all antibodies were used in a dilution of 1:1000 in TBST/1.5% BSA.
Cell Lysis, Immunoprecipitations, and Western Blotting
After stimulation, cells were rinsed twice with cold PBS and lysed in lysis buffer (50 mM HEPES, pH 7.5, 150 mM NaCl, 1.5 mM MgCl2, 1 mM EGTA, 10% glycerol, 1% Triton X-100, 1 mM PMSF, 1 mM Na3VO4, 10 μg/ml leupeptin, and 10 μg/ml aprotinin) as described (Wellbrock et al. 1998). Lysates were directly analyzed or incubated with the indicated antibodies and protein A-Sepharose for at least 1 h at 4°C. Proteins were analyzed by western blotting as described (Wellbrock et al. 1998).
Results
Structure of egfra, an Additional egfr Gene from Xiphophorus
The egfra cDNA was isolated from a cDNA library derived from the embryonic cell line A2 of Xiphophorus xiphidium. Four positive clones were obtained, all of which contained a 4,126-bp EcoRI fragment. It included an open reading frame of 3,627 nt encoding a protein of 1,209 amino acids that was named Egfra (fig. 1).
FIG. 1. Sequence alignment of vertebrate EGF receptors. Accession numbers: XP_044653 (EGFR Homo sapiens), P13387 (Egfr Gallus gallus), AY154653 (Egfra Danio rerio), AAD10500 (Egfrb Xiphophorus xiphidium), and P13388 (Xmrk Xiphophorus maculatus). Conserved amino acid residues are boxed in black, and conservative substitutions are in gray. The start and the end of each conserved domain is indicated by an arrow
Comparisons of the Egfra-deduced amino acid sequence with sequence databases showed the highest similarities with the EGFR from chicken, different mammalian species, and Xmrk from Xiphophorus, with overall amino acid similarities ranging from 67% to 69%. Egfra contains all the features of an RTK from the EGFR family. The predicted extracellular domain of 519 amino acids includes two cysteine rich domains (S1 and S2), containing 22 and 20 cysteines each. The spacing and number of these cysteines is in agreement with the conserved cysteine pattern of the members of the EGFR family (Abe et al. 1998). The extracellular domain is followed by a segment of hydrophobic amino acids that can form a single membrane-spanning region. The first 44 amino acids of the putative cytoplasmic part of Egfra are highly similar to the cytoplasmic juxtamembrane of other members of the EGFR family. The tyrosine kinase domain is the most conserved region. All 12 subdomains and structural determinants that characterize a functional tyrosine kinase domain are present (Hanks and Hunter 1995). Among them are residues important for ATP binding (glycine rich motif, K755, and E772), the D847 residue in the catalytic loop, and the DFG motif in the activation domain. These motifs are in the Egfra sequence in equivalent positions to other RTKs. The carboxy-terminal domain is the most divergent part of the protein when compared with the hEGFR and Xmrk/Egfrb. It contains a number of tyrosines that can act as docking sites for intracellular substrates.
Expansion of the EGFR Family by Gene Duplication in Fish
To search for the presence of genes encoding receptors from the EGFR family in other fish species, the almost completely sequenced genome of the Japanese pufferfish Takifugu rubripes (Aparicio et al. 2002) http://fugu.hgmp.mrc.ac.uk/), as well as the less complete zebrafish genome database (http://www.ensembl.org/Danio_rerio/blastview), were analyzed. Two egfr genes, egfra and egfrb, were identified in the pufferfish, and egfra was found in the zebrafish as well. All types of phylogenetic analyses performed with nucleic (not shown) and amino acid sequences (fig. 2) were concordant and indicated that egfra and egfrb were the result of the duplication of an egfr ancestor within the (ray-finned) fish lineage before the divergence of Xiphophorus, pufferfish, and zebrafish more than 100 MYA (Carroll 1997; Lydeard and Roe 1997). This analysis also shows that in Xiphophorus, the formerly called INV-Xmrk and ONC-Xmrk genes (for the proto-oncogene and the oncogene, now egfrb and Xmrk, respectively) are like egfra co-orthologs of the unique egfr gene found in birds and mammals.
FIG. 2. Phylogeny of receptor tyrosine kinases from the EGFR family. Bootstrap values (%) using maximum parsimony (100 replicates, first values) and neighbor-joining (1,000 replicates, second and third values), as well as the reliability values (%) for maximum likelihood (quartet puzzling, 10,000 puzzling steps, fourth values) are given. Neighbor-joining was performed either considering all sites (second values) or considering only the unsaturated fraction of sites (third values) using AsaturA (Van de Peer et al. 2002). The tree (maximum parsimony) has been rooted on invertebrate sequences. Branches with less than 50% support, using maximum parsimony, have been collapsed. A 630–amino acid alignment of sequences starting in the C-terminal end of the first receptor L domain and ending about 20 amino acids upstream from the catalytic domain was used for phylogenetic analysis. Accession numbers: XP_044653 (EGFR Homo sapiens), AAA17899 (Egfr Mus musculus), AAF14008 (Egfr Rattus norvegicus), P13387 (Egfr Gallus gallus), AAD10500 (Egfrb Xiphophorus xiphidium), P13388 (Xmrk Xiphophorus maculatus), CAA27060 (ERBB2 Homo sapiens), BAA23127 (ErbB2 Canis familiaris), CAA27059 (ErbB2 Rattus norvegicus), AAA35790 (ERBB3 Homo sapiens), AAC28498 (ErbB3 Rattus norvegicus), AAC34391 (ErbB3a Takifugu rubripes), NP_005226 (ERBB4 Homo sapiens), AAD08899 (ErBb4 Rattus norvegicus), AAD31764 (ErbB4 Gallus gallus), AAD26132 (DER Drosophila melanogaster), CAC35008 (Egfr Anopheles gambiae), and S70712 (CER Caenorhabditis elegans). Danio rerio ErbB3b sequence was deduced from the unfinished high-throughput genomic sequence AL591365. Takifugu rubripes sequences were deduced from data provided by the Fugu Genome Consortium (http://fugu.hgmp.mrc.ac.uk/): Egfra (S002539), Egfrb (S003004), ErbB2 (S002501), ErbB3b (S002894), ErbB4a (S001137), and ErbB4b (S002275). Danio rerio sequences were deduced from data from the zebrafish genome database (http://www.ensembl.org/Danio_rerio/blastview): Egfra (z06s003948), Erbb2 (z06s013389), Erbb3a (z06s021951), and Erbb4b (z06s009145)
Genes orthologous to erbB2, erbB3, and erbB4 from higher vertebrates were identified in the genomes of both Japanese pufferfish and zebrafish. Interestingly, duplicates were identified not only for egfr but also for erbB3 (pufferfish and zebrafish) and erbB4 (pufferfish), indicating that at least three different genes encoding receptors from the EGFR family are duplicated in the genome of T. rubripes and probably of other fish species. Analysis of the ratio of synonymous to nonsynonymous substitutions strongly suggested that all fish duplicates evolved under purifying selection and therefore probably do not correspond to pseudogenes: the (average) ratios between fish-specific duplicates (for example egfra versus egfrb) were 5.7, 9.0, and 14.1 for egfr, erbB3, and erbB4, respectively, compared with 7.9, 8.9, and 26.4 between their orthologs in higher vertebrates. Similar ratios indicating purifying selection were also observed in comparison involving fish orthologs, for example Xiphophorus and pufferfish egfra genes.
There is accumulating evidence for an ancient whole-genome duplication (as well as for more recent local events) having taken place during the course of evolution of the (ray-finned) fish lineage after its separation from the sarcopterygian lineage, including coelacanth, lungfish, and vertebrates (Wittbrodt, Meyer, and Schartl 1998; Meyer and Schartl 1999). The relatively low level of nucleotide identity between fish egfr (average 59.6%), erbB3 (65.5%), and erbB4 paralogs (75.1%) was rather suggestive of ancient events of duplication. The duplications that led to the formation of fish egfr, erbB3, and erbB4 paralogs probably arose before the divergence between zebrafish and pufferfish/Xiphophorus more than 100 MYA. Taylor et al. (2001) estimated that the putative fish-specific genome duplication occurred approximately 350 MYA by plotting the number of nucleotide substitutions at third-codon positions against divergence dates for different taxa. Using the same method, we found that the number of substitutions per site at third-codon positions between the different fish egfr (average 1.36), erbB3 (average 1.46), and erbB4 (average 0.96) paralogs was similar to the average value (1.02, SD = 0.24) obtained by Taylor et al. (2001) between unlinked zebrafish duplicates generated by the putative fish-specific genome duplication. As a basis for comparison, the same (average) value between orthologs from mammals and chicken (divergence about 300 MYA [Kumar and Hedges 1998]) was slightly lower than between the corresponding fish paralogs: 0.97 for egfr and 0.84 for erbB4. We conclude that the fish egfr, erbB3, and erbB4 paralogs resulted from ancient duplication event(s) and might be remnants of the proposed fish-specific genome duplication.
egfra and egfrb Show Different Expression Patterns in Xiphophorus
The expression pattern of egfra in different tissues of adult Xiphophorus and two cell lines was determined. A single transcript of 5.0 kb was detected in all analyzed normal tissues, although at different levels (fig. 3A). The highest expression was detected in kidney, followed by skin and gills and the embryonic cell line A2. This pattern is similar to that described for mouse egfr. The mammalian gene is expressed in different organs but preferentially in tissues with regenerative ephithelium such as skin (Adamson 1990). When comparing the expression of egfra and egfrb in the same tissues (fig. 3B), it appeared that the expression level of egfrb was lower than that of egfra. Besides, these Xiphophorus genes are preferentially expressed in different tissues. Whereas egfra shows a prominent expression in kidney and skin, egfrb transcripts are more abundant in liver compared with other tissues.
FIG. 3. Northern blot analysis. Expression of Xiphophorus egfra (A), egfrb (B), and Xmrk (C) in different organs of adult Xiphophorus, in the embryonic cell line A2, and in the melanoma derived cell line PSM. A 4-kb EcoRI fragment containing the whole coding sequence of egfra was used as a probe for (A). (B) and (C) were probed with a 3.6-kb XbaI/DraI fragment corresponding to the complete coding sequence of Xmrk. (B) is a longer exposure of (C) for detection of the 5.8-kb egfrb transcript. (C) corresponds to a short exposure where only the abundant 4.7-kb transcript of Xmrk can be seen. Equivalent RNA blotting was verified by methylene blue staining of ribosomal RNA
Analysis of melanoma and a melanoma-derived cell line (PSM), in which the Xmrk gene is overexpressed (fig. 3B and C), revealed a significant level of expression of egfra. However, although the expression in melanoma is notable, it is not higher than in normal skin and is even weaker than in kidney (fig. 3A).
Egfra Can Be Stimulated by hEGF but Not by TGF- or mEGF
To determine if Egfra is able to cross-react with ligands of EGFR-family receptors and whether it acts as a growth factor receptor, we chose the IL-3 dependent cell line Ba/F3. Ba/F3 cells lack endogenous expression of all EGFR family members (Riese et al. 1995), which excludes cross-reaction of defined ligands of this receptor family with other receptors than the ectopically expressed one.
Stimulation of Ba/F3 cells stably expressing Egfra (BaF XER cells) with TGF-, as well as murine EGF (mEGF), had a very weak but not significant mitogenic effect measured as 3H-thymidine incorporation during S-phase (fig. 4A). The presence of 100 nM human EGF (hEGF), however, resulted in approximately 24% of the amount of 3H-thymidine incorporation reached after IL-3 stimulation, which was set 100%. The induced DNA synthesis occurred in a dose-dependent manner, since 500 nM EGF induced an even higher response of 31%. These values indicate that the Egfra can be stimulated by hEGF and that this stimulation results in efficient intracellular signaling in Ba/F3 cells. In contrast to Egfra, Egfrb showed no cross-reactivity with either of the factors when expressed in Ba/F3 cells (BaF INV cells [fig. 4A]).
FIG. 4. Egfra stimulation with different growth factors. (A) The growth response to TGF-, mEGF, or hEGF of Ba/F3 cells expressing Egfra (BaF XER) and Ba/F3 cells expressing Egfrb (BaF INV cells) was determined by 3H-thymidine incorporation 24 h after stimulation. 3H-thymidine incorporation in response to IL-3 stimulation was set 100%. Asterisks (*) indicate P < 0.0001 versus control. (B) The growth response of Ba/F3 cells expressing either the Egfra (BaF XER), the hEGFR (BaF HER cells), or the chimeric HER-XER receptor (BaF HER-XER) to hEGF was determined by 3H-thymidine incorporation 24 h after stimulation. 3H-thymidine incorporation in response to IL-3 stimulation was set 100%
The hEGFR was chosen as a reference for the stimulation by the different growth factors. Strikingly, although the receptor has been shown to become catalytically activated by 8 nM hEGF in Ba/F3 cells (Morcinek et al. 2002), it was not able to trigger efficient cell cycle progression in these cells even when stimulated with 100 nM hEGF (Morcinek et al. 2002 and fig. 4B). This opposite behavior of hEGFR compared with the Egfra pointed to differences in the intracellular part of the receptors in terms of substrate binding and signal transduction.
To test this hypothesis, we generated a chimeric receptor, HER-XER. This chimera consists of the extracellular ligand-binding domain of the hEGFR and the intracellular part of Egfra. Treatment of Ba/F3 cells expressing HER-XER (BaF HER XER cells) with 100 nM hEGF resulted in 43% 3H-thymidine incorporation compared with IL-3 stimulation (fig. 4B). Whereas the response of Egfra was lower with approximately 25%, the hEGFR again showed no significant mitogenic activity (fig. 4B).
The stronger hEGF response of the HER-XER chimera compared with that of Egfra showed that although hEGF is able to cross-react with Egfra, its affinity to the fish receptor is lower than to the human receptor. In addition, the fact that HER-XER and HER, which are differing only in their intracellular domain, showed a different capacity in stimulating mitogenic signaling in Ba/F3 cells suggests that the Egfra induces a different signal transduction than the hEGFR.
Xiphophorus Egfra Signaling Includes Activation of MAP Kinase, PI3-kinase Signaling, and Stimulation of STAT3 and STAT5
To investigate the signal transduction of Egfra, we analyzed the activation of different intracellular substrates known to be activated by either Xmrk or the hEGFR in Ba/F3 cells. For these studies we used the HER-XER chimera because its stimulation by hEGF is more efficient.
First, HER-XER was immunoprecipitated with a monoclonal antibody raised against the extracellular domain of hEGFR and analyzed with antiphosphotyrosine after stimulation with hEGF. This experiment showed that the receptor becomes tyrosine phosphorylated after hEGF stimulation (fig. 5A), which makes it able to interact with intracellular substrates.
FIG. 5. Egfra signaling. (A) BaF HER-XER cells were either stimulated with 8 nM hEGF for 10 min (+) or were left untreated (–). HER-XER was immunoprecipitated (Ip) from lysates of treated or untreated cells with mAb-108.1, and immune complexes were analyzed with antiphosphotyrosine (anti-ptyr). (B) Lysates of EGF stimulated (+) or untreated (–) BaF HER-XER cells were used for immunoprecipitation with either anti-STAT3 or anti-STAT5 and immune complexes were analyzed with anti-ptyr. (C) Ba/F3 wild-type (BaF WT) and BaF HER-XER cells were either left untreated (–) or stimulated with EGF for 10 min (+), and cellular lysates were analyzed subsequently with anti-phospho MAPK and anti-ERK2 or with anti-phospho Akt and anti-Akt on a western blot
Next we analyzed the activation of the signal transducers and activators of transcription STAT3 and STAT5. It is known that whereas Xmrk is able to activate both STAT proteins in Ba/F3 cells, the hEGFR is incapable of activating STAT5 (Morcinek et al. 2002). Immunoprecipitation of STAT3 as well as STAT5 before and after stimulation of HER-XER revealed that the receptor induces the tyrosine phosphorylation of both STAT proteins (fig. 5B). In the immunoprecipitates of STAT3, the receptor was coprecipitated and detectable as a 165-kDa tyrosine phosphorylated protein after stimulation with EGF.
The stimulation of HER-XER resulted in activation of the MAP kinase pathway as seen by the specific phosphorylation of the MAP kinases ERK1 and ERK2 (fig. 5C). The MAP kinase cascade was initiated by binding of GRB2 and Shc to HER-XER, both of which could be coimmunoprecipitated with the receptor (data not shown). Also an activating serine phosphorylation of Akt, a protein acting further downstream in the PI3-kinase signaling pathway, was significantly increased after stimulation of the HER-XER receptor in Ba/F3 cells (fig. 5C).
Discussion
We have isolated and characterized an additional epidermal growth factor receptor gene in the fish Xiphophorus. This gene, egfra, encodes a protein that contains all the structural features of subclass I receptor tyrosine kinases. Two other receptors closely related to the hEGFR had been already described in Xiphophorus. One of them, INV-Xmrk (Dimitrijevic et al. 1998), now renamed as Egfrb, is present in all fish of the genus Xiphophorus, as well as in more divergent teleost species. The egfrb gene represents the proto-oncogene of the melanoma-inducing gene ONC-Xmrk (now Xmrk) that exists only in some Xiphophorus species (Wittbrodt et al. 1989). The Xmrk (onco)gene arose more than 5 to 6 MYA by a local gene duplication event from egfrb, and both genes are linked and located on the sex chromosomes (Adam, Dimitrijevic, and Schartl 1993). Earlier linkage analysis had already revealed the presence of another locus encoding a third egfr-related gene in an autosomal region (Harless et al. 1990). Sequence comparison of egfra with the probe used to determine the linkage group (linkage group VI [Harless et al. 1990]) revealed that this egfr-related gene is egfra.
Egfra and egfrb, as well as the duplicated EGFR family members present in zebrafish and pufferfish, might be the result of whole-genome duplication in teleost fish. This origin was already proposed to explain the existence of seven Hox gene clusters in zebrafish (Amores et al. 1998) in contrast with the four clusters existing in tetrapods. More recently, additional Hox clusters have been described in medaka (Naruse et al. 2000), tilapia (Malaga-Trillo and Meyer 2001), and Fugu (Aparicio et al. 2002).
One of the characteristics of RTKs is the binding of a ligand in the extracellular domain, but no EGFR ligand has been isolated in fish so far. Nevertheless, a few reports exist where the application of different mammalian EGFR ligands in fish tissues in vitro provoked a physiological response (Srivastava and van der Kraak 1995; Wood and van der Kraak 2002). In the present work, we have shown that Ba/F3 cells containing Egfra are able to proliferate when hEGF is added, indicating that this ligand is able to bind and activate Egfra. However, this ligand could not stimulate proliferation when applied to a Ba/F3 cell line stably expressing Egfrb. This may indicate that the extracellular domains of both Xiphophorus EGF receptors differ in specific residues responsible for ligand binding and may point to the binding of different (and perhaps duplicated) ligands in Xiphophorus. The crystal structure of the hEGFR in complex with hEGF has been recently determined (Ogiso et al. 2002). The study has identified hEGF binding sites in domains I and III of the hEGFR. Two hEGFR residues involved in direct interactions with hEGF are conserved in Egfra but not in Egfrb. This could explain the different ligand-binding properties of the fish receptors. The side chain of Phe357 in the hEGFR hydrophobically interacts with Tyr13 of hEGF. In Egfra, a tyrosine is existing in an equivalent position (Tyr391). This would allow as well the interaction with Tyr13 through the aromatic side chain. However, the histidine (His383) in Egfrb at the equivalent position does not allow this contact. The importance of this ligand-receptor contact was shown by substitution of Tyr13 with polar residues that blocked this interaction and considerably decreased receptor binding activity (Tadaki and Niyogi 1993). In addition the Gln384 side chain of hEGFR hydrogen bonds with Gln43 and Arg45 of hEGF. This Gln is also present in Egfra (Gln418) but not in Egfrb, where a Met (Met410) at an equivalent position prevents the formation of this hydrogen bond. Whether these residues are the only ones responsible for the different binding properties between Egfra and Egfrb is difficult to say because other amino acid differences in the fish receptors could account for a different local structure in key positions.
The MAP kinase cascade was activated by the HER-XER chimera resulting in phosphorylation of ERK1 and ERK2. MAP kinase signaling was initiated by binding of the receptor to the adaptor proteins Shc and GRB2 (data not shown) for which the Egfra contains potential binding sites in its carboxy-terminus. Tyr1097 (YMNH) and Tyr1113 (YPNY) both represent the YXNX motif that has been shown to be selectively bound by Grb2 SH2 domain when tyrosine phosphorylated (Songyang et al. 1993). Tyr1097 in Egfra is placed in a homologous position to hEGFR Tyr1068 and Xmrk Tyr1033, both binding sites for the GRB2 SH2 domain (Batzer et al. 1994; Wellbrock and Schartl 1999). Egfra Tyr1113 is also contained in a YXNX-binding motif; however, no tyrosine is present in hEGFR or Xmrk in this position. The sequence amino-terminal to Tyr1181 (NPDY) of Egfra fits with the NPXY-binding motif for the PTB (phosphotyrosine-binding) domain of Shc (Batzer et al. 1995). Tyr1181 is placed in an equivalent site to Tyr1148 of the hEGFR, which binds Shc (Batzer et al. 1995). However, whether these binding sites in Egfra are used by the receptor to interact with GRB2 and Shc, respectively, remains to be determined.
Activation of the Egfra also led to phosphorylation and thus activation of Akt. This protein is a downstream effector of Pi3-kinase and has been shown to be involved in antiapoptotic signaling (Coffer, Jin, and Woodgett 1998). Both Egfra and Egfrb, as well as Xmrk and hEGFR, lack the YXXM motif for binding the SH2 domains of the PI3-kinase regulatory subunit p85. It has been shown that hEGFR and Xmrk can indirectly activate PI3-kinase by binding to other intracellular proteins that act as adaptors (Rodrigues et al. 2000; Wellbrock and Schartl 2000). Therefore, it is feasible that the Egfra makes use of a similar mechanism for activating PI3-kinase pathways.
Analysis of the activation of STAT proteins revealed that both STAT3 and STAT5 are activated by the intracelluar part of Egfra expressed in Ba/F3 cells. STAT3 can be also activated by hEGFR as well as Xmrk in this cell system; however, in contrast to Xmrk, hEGFR is incapable of activating STAT5 (Morcinek et al. 2002). STAT5 has been shown to be a crucial factor involved in melanoma development in Xiphophorus, and it is responsible for maintaining cell progression of Ba/F3 cells sustained by Xmrk signaling (Wellbrock et al. 1998; Morcinek et al. 2002). So, the fact that the Egfra activates STAT5 could account for the ability to induce proliferation in Ba/F3 cells. In this cellular context, Egfra and Egfrb/Xmrk behave in a similar way and very differently from hEGFR, which is unable to activate STAT5 and to induce mitogenic activity. These data indicate that Egfra and Egfrb have kept similar characteristics that are different from that of the hEGFR.
Only one member of the EGFR family has been identified in several invertebrates (Livneh et al. 1985; Aroian et al. 1990), whereas four members exist in mammals. Different evolutionary analyses (Stein and Staros 2000; Martin 2001) support the appearance of the EGFR family by two rounds of duplication of an ancestral gene. A first duplication would have generated an erbB1/erbB2 precursor and an erbB3/erbB4 precursor. A second duplication led to the existence of four receptors as found in mammals today. A fish-specific genome duplication could then have generated the additional receptors in fish, leading to more subclass I RTKs in these animals. On top of that, a lineage-specific gene duplication led to an additional egfrb copy in Xiphophorus that became the Xmrk oncogene.
Different fates have been proposed for duplicated genes (reviewed by Prince and Pickett [2002]). A classical model predicts that one of the copies can degenerate by accumulation of deleterious mutations. This is clearly not the case for egfra and egfrb, which both encode entirely functional tyrosine kinases. Another well-documented model proposes that each duplicated copy will retain only some of their original functions, complementing each other to fulfill the original ancestral gene activities (subfunctionalization). Less frequently, one of the duplicated copies may acquire a new function, whereas the other copy retains the original one (neofunctionalization).
The functional study of the Egfra and its comparison with Egfrb and hEGFR revealed that none of the above mentioned evolutionary fates can be assigned to these fish duplicates. Although Egfra and Egfrb present clear differences in their ligand-binding ability, their pattern of expression in adult fish is partially overlapping, and their signal transduction seems quite similar. This might provoke the question of a certain redundancy for Egfra and Egfrb. To explain preservation of redundant multidomain proteins, Gibson and Spring (1998) proposed that accumulation of mutations in proteins with a significant functional role will occur more slowly than expected, as damaging a single domain can give rise to a dominant-negative phenotype affecting a whole network of proteins. On the other hand, the significant contribution of preserved duplicated genes to genetic robustness against loss-of-function mutations has to be considered (Meyer 2003; Gu et al. 2003). However, some kind of asymmetry in the duplicates' function is necessary to preserve redundancy (Krakauer and Nowak 1999). In the case of the EGFR, its involvement in a wide variety of cellular processes during development and in adult organisms makes a dissection of its functions extremely difficult.
Subclass I RTKs are potent tumor genes after oncogenic activation. Some fish species are known to be especially susceptible to developing tumors after exposure to environmental and experimental carcinogens, and some fish display a considerable rate of spontaneous tumors. It is tempting to speculate that the enlarged gene families, many of which are involved in the regulation of cell growth and proliferation, confer more targets for oncogenic activation to fish genomes.
Acknowledgements
We thank Nicola Dimitrijevic for the isolation of a partial cDNA clone of egfra. This work is supported by grants to C.W. and M.S. from the Deutsche Forschungsgemeinschaft through the SFB487 ("Regulatorische Membranproteine") and the SFB465 ("Entwicklung und Manipulation pluripotenter Zellen") and the Fonds der Chemischen Industrie. J.-N.V. is supported by the BioFuture program of the German Bundesministerium für Bildung und Forschung. Takifugu data have been provided freely by the Fugu Genome Consortium for use in this publication only. We are grateful to Yves Van de Peer (Gent, Belgium) and Tancred Frickey (Tübingen, Germany) for their help concerning the AsaturA program.
Literature Cited
Abe, Y., M. Odaka, F. Inagaki, I. Lax, J. Schlessinger, and D. Kohda. 1998. Disulfide bond structure of human epidermal growth factor receptor. J. Biol. Chem. 273:11150-11157.
Adam, D., N. Dimitrijevic, and M. Schartl. 1993. Tumor suppression in Xiphophorus by an accidentally acquired promoter. Science 259:816-819.
Adamson, E. D. 1990. Developmental activities of the epidermal growth factor receptor. Curr. Top. Dev. Biol. 24:1-29.
Amores, A., A. Force, and Y. L. Yan, et al. (13 co-authors). 1998. Zebrafish hox clusters and vertebrate genome evolution. Science 282:1711-1714.
Aparicio, S., J. Chapman, and E. Stupka, et al. (41 co-authors). 2002. Whole-genome shotgun assembly and analysis of the genome of Fugu rubripes. Science 297:1301-1310.
Aroian, R. V., M. Koga, J. E. Mendel, Y. Ohshima, and P. W. Sternberg. 1990. The let-23 gene necessary for Caenorhabditis elegans vulval induction encodes a tyrosine kinase of the EGF receptor subfamily. Nature 348:693-699.
Batzer, A. G., P. Blaikie, K. Nelson, J. Schlessinger, and B. Margolis. 1995. The phosphotyrosine interaction domain of Shc binds an LXNPXY motif on the epidermal growth factor receptor. Mol. Cell Biol. 15:4403-4409.
Batzer, A. G., D. Rotin, J. M. Urena, E. Y. Skolnik, and J. Schlessinger. 1994. Hierarchy of binding sites for Grb2 and Shc on the epidermal growth factor receptor. Mol. Cell Biol. 14:5192-5201.
Boomsma, R. A., H. Scott, and K. Walters. 2001. Immunocytochemical localization of epidermal growth factor receptor in early embryos of the Japanese medaka fish (Oryzias latipes). Histochem. J. 33:37-42.
Carroll, R. L. 1997. Patterns and processes of vertebrate evolution. Cambridge University Press Cambridge, UK.
Coffer, P. J., J. Jin, and J. R. Woodgett. 1998. Protein kinase B (c-Akt): a multifunctional mediator of phosphatidylinositol 3-kinase activation. Biochem. J. 335:(pt 1): 1-13.
Dimitrijevic, N., C. Winkler, C. Wellbrock, A. Gomez, J. Duschl, J. Altschmied, and M. Schartl. 1998. Activation of the Xmrk proto-oncogene of Xiphophorus by overexpression and mutational alterations. Oncogene 16:1681-1690.
Gibson, T. J., and J. Spring. 1998. Genetic redundancy in vertebrates: polyploidy and persistence of genes encoding multidomain proteins. Trends Genet. 14:46-49.
Gomez, A., C. Wellbrock, H. Gutbrod, N. Dimitrijevic, and M. Schartl. 2001. Ligand-independent dimerization and activation of the oncogenic Xmrk receptor by two mutations in the extracellular domain. J. Biol. Chem. 276:3333-3340.
Gu, Z., L. M. Steinmetz, X. Gu, C. Scharfe, R. W. Davis, and W. H. Li. 2003. Role of duplicate genes in genetic robustness against null mutations. Nature 421:63-66.
Hanks, S. K., and T. Hunter. 1995. Protein kinases 6. The eukaryotic protein kinase superfamily: kinase (catalytic) domain structure and classification. FASEB J. 9:576-596.
Harless, J., R. Svensson, K. D. Kallman, D. C. Morizot, and R. S. Nairn. 1990. Assignment of an erbB-like DNA sequence to linkage group VI in fishes of the genus Xiphophorus (Poeciliidae). Cancer Genet. Cytogenet. 50:45-51.
Karasuyama, H., and F. Melchers. 1988. Establishment of mouse cell lines which constitutively secrete large quantities of interleukin 2, 3, 4 or 5, using modified cDNA expression vectors. Eur. J. Immunol. 18:97-104.
Krakauer, D. C., and M. A. Nowak. 1999. Evolutionary preservation of redundant duplicated genes. Semin. Cell Dev. Biol. 10:555-559.
Kumar, S., and S. B. Hedges. 1998. A molecular timescale for vertebrate evolution. Nature 392:917-920.
Lax, I., A. Johnson, R. Howk, J. Sap, F. Bellot, M. Winkler, A. Ullrich, B. Vennstrom, J. Schlessinger, and D. Givol. 1988. Chicken epidermal growth factor (EGF) receptor: cDNA cloning, expression in mouse cells, and differential binding of EGF and transforming growth factor alpha. Mol. Cell Biol. 8:1970-1978.
Livneh, E., L. Glazer, D. Segal, J. Schlessinger, and B. Z. Shilo. 1985. The Drosophila EGF receptor gene homolog: conservation of both hormone binding and kinase domains. Cell 40:599-607.
Luetteke, N. C., H. K. Phillips, T. H. Qiu, N. G. Copeland, H. S. Earp, N. A. Jenkins, and D. C. Lee. 1994. The mouse waved-2 phenotype results from a point mutation in the EGF receptor tyrosine kinase. Genes Dev. 8:399-413.
Lydeard, C., and K. J. Roe. 1997. The phylogenetic utility of the mitochondrial cytochrome b gene for inferring relationships among actinopterygian fishes. Pp. 285–303 in T. C. Kocher and C. A. Stepien, eds. Molecular systematics of fishes. Academic Press. San Diego, Calif.
MacDougall, T. M., and G. van der Kraak. 1998. Peptide growth factors modulate prostaglandin E and F production by goldfish ovarian follicles. Gen. Comp. Endocrinol. 110:46-57.
Malaga-Trillo, E., and A. Meyer. 2001. Genome duplications and accelerated evolution of Hox genes and cluster architecture in teleost fishes. Am. Zool. 41:676-686.
Martin, A. 2001. Is tetralogy true? Lack of support for the "one-to-four rule". Mol. Biol. Evol. 18:89-93.
Meyer, A. 2003. Molecular evolution: duplication, duplication. Nature 421:31-32.
Meyer, A., and M. Schartl. 1999. Gene and genome duplications in vertebrates: the one-to-four (-to-eight in fish) rule and the evolution of novel gene functions. Curr. Opin. Cell Biol. 11:699-704.
Miettinen, P. J., J. E. Berger, J. Meneses, Y. Phung, R. A. Pedersen, Z. Werb, and R. Derynck. 1995. Epithelial immaturity and multiorgan failure in mice lacking epidermal growth factor receptor. Nature 376:337-341.
Morcinek, J. C., C. Weisser, E. Geissinger, M. Schartl, and C. Wellbrock. 2002. Activation of STAT5 triggers proliferation and contributes to anti-apoptotic signalling mediated by the oncogenic Xmrk kinase. Oncogene 21:1668-1678.
Naruse, K., S. Fukamachi, and H. Mitani, et al. (20 co-authors). 2000. A detailed linkage map of medaka, Oryzias latipes: comparative genomics and genome evolution. Genetics 154:1773-1784.
Newsted, J. L., and J. P. Giesy. 2000. Epidermal growth factor receptor-protein kinase interactions in hepatic membranes of rainbow trout (Oncorhynchus mykiss). Fish Physiol. Biochem. 22:181-189.
Ogiso, H., R. Ishitani, and O. Nureki, et al. (11 co-authors). 2002. Crystal structure of the complex of human epidermal growth factor and receptor extracellular domains. Cell 110:775-787.
Olayioye, M. A., R. M. Neve, H. A. Lane, and N. E. Hynes. 2000. The ErbB signaling network: receptor heterodimerization in development and cancer. EMBO J. 19:3159-3167.
Palacios, R., and M. Steinmetz. 1985. Il-3-dependent mouse clones that express B-220 surface antigen, contain Ig genes in germ-line configuration, and generate B lymphocytes in vivo. Cell 41:727-734.
Petch, L. A., J. Harris, V. W. Raymond, A. Blasband, D. C. Lee, and H. S. Earp. 1990. A truncated, secreted form of the epidermal growth factor receptor is encoded by an alternatively spliced transcript in normal rat tissue. Mol. Cell Biol. 10:2973-2982.
Prince, V. E., and F. B. Pickett. 2002. Splitting pairs: the diverging fates of duplicated genes. Nat. Rev. Genet. 3:827-837.
Riese, D. J., T. M. van Raaij, G. D. Plowman, G. C. Andrews, and D. F. Stern. 1995. The cellular response to neuregulins is governed by complex interactions of the erbB receptor family. Mol. Cell Biol. 15:5770-5776.
Rodrigues, G. A., M. Falasca, Z. Zhang, S. H. Ong, and J. Schlessinger. 2000. A novel positive feedback loop mediated by the docking protein Gab1 and phosphatidylinositol 3-kinase in epidermal growth factor receptor signaling. Mol. Cell Biol. 20:1448-1459.
Saitou, N., and M. Nei. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual. 2 edition. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
Schlessinger, J. 2000. Cell signaling by receptor tyrosine kinases. Cell 103:211-225.
Sibilia, M., and E. F. Wagner. 1995. Strain-dependent epithelial defects in mice lacking the EGF receptor. Science 269:234-238.
Songyang, Z., S. E. Shoelson, and M. Chaudhuri, et al. (14 co-authors). 1993. SH2 domains recognize specific phosphopeptide sequences. Cell 72:767-778.
Srivastava, K. R., and G. van der Kraak. 1995. Multifactorial regulation of DNA synthesis in goldfish ovarian follicles. Gen. Comp. Endocrinol. 100:397-403.
Stein, R. A., and J. V. Staros. 2000. Evolutionary analysis of the ErbB receptor and ligand families. J Mol. Evol. 50:397-412.
Strimmer, K., and A. von Haeseler. 1996. Quartet puzzling: a quartet maximum-likelihood method for reconstructing tree topologies. Mol. Biol. Evol. 13:964-969.
Swofford, D. L. 1998. PAUP*: phylgenetic analysis using parsimony (*and other methods), version 4.0. Sinauer Associates, Sunderland, Mass.
Tadaki, D. K., and S. K. Niyogi. 1993. The functional importance of hydrophobicity of the tyrosine at position 13 of human epidermal growth factor in receptor binding. J Biol. Chem. 268:10114-10119.
Tajima, F., and M. Nei. 1984. Estimation of evolutionary distance between nucleotide sequences. Mol. Biol. Evol. 1:269-285.
Tang, C. K., and M. E. Lippman. 1998. EGF family receptors and their ligands in human cancer. Pp. 113–165 in B. W. O'Malley, ed. Hormones and signaling, Vol. I. Academic Press, San Diego, Calif.
Taylor, J. S., P. Y. Van de, I. Braasch, and A. Meyer. 2001. Comparative genomics provides evidence for an ancient genome duplication event in fish. Philos. Trans. R. Soc. Lond B Biol. Sci. 356:1661-1679.
Threadgill, D. W., A. A. Dlugosz, L. A. Hansen, T. Tennenbaum, U. Lichti, D. Yee, C. LaMantia, T. Mourton, K. Herrup, and R. C. Harris. 1995. Targeted disruption of mouse EGF receptor: effect of genetic background on mutant phenotype. Science 269:230-234.
Ullrich, A., L. Coussens, J. S. Hayflick, T. J. Dull, A. Gray, A. W. Tam, J. Lee, Y. Yarden, T. A. Libermann, and J. Schlessinger. 1984. Human epidermal growth factor receptor cDNA sequence and aberrant expression of the amplified gene in A431 epidermoid carcinoma cells. Nature 309:418-425.
Ullrich, A., and J. Schlessinger. 1990. Signal transduction by receptors with tyrosine kinase activity. Cell 61:203-212.
Van de Peer, Y., T. Frickey, J. S. Taylor, and A. Meyer. 2002. Dealing with saturation at the amino acid level: a case study based on anciently duplicated zebrafish genes. Gene 295:205-211.
Wellbrock, C., E. Geissinger, A. Gomez, P. Fischer, K. Friedrich, and M. Schartl. 1998. Signalling by the oncogenic receptor tyrosine kinase Xmrk leads to activation of STAT5 in Xiphophorus melanoma. Oncogene 16:3047-3056.
Wellbrock, C., and M. Schartl. 1999. Multiple binding sites in the growth factor receptor Xmrk mediate binding to p59fyn, GRB2 and Shc. Eur. J. Biochem. 260:275-283.
Wellbrock, C., and M. Schartl. 2000. Activation of phosphatidylinositol 3-kinase by a complex of p59fyn and the receptor tyrosine kinase Xmrk is involved in malignant transformation of pigment cells. Eur. J. Biochem. 267:3513-3522.
Wittbrodt, J., D. Adam, B. Malitschek, W. Maueler, F. Raulf, A. Telling, S. M. Robertson, and M. Schartl. 1989. Novel putative receptor tyrosine kinase encoded by the melanoma-inducing Tu locus in Xiphophorus. Nature 341:415-421.
Wittbrodt, J., R. Lammers, B. Malitschek, A. Ullrich, and M. Schartl. 1992. The Xmrk receptor tyrosine kinase is activated in Xiphophorus malignant melanoma. EMBO J. 11:4239-4246.
Wittbrodt, J., A. Meyer, and M. Schartl. 1998. More genes in fish? Bioessays 20:511-515.
Wood, A. W., and G. van der Kraak. 2002. Inhibition of apoptosis in vitellogenic ovarian follicles of rainbow trout (Oncorhynchus mykiss) by salmon gonadotropin, epidermal growth factor, and 17beta-estradiol. Mol. Reprod. Dev. 61:511-518.(Ana Gómez1, Jean-Nicolas )
E-mail: claudia.wellbrock@icr.ac.uk.
Abstract
The epidermal growth factor receptor (EGFR) gives name to a family of receptors formed by four members in mammals (EGFR, ErbB2, ErbB3, and ErbB4). Members of this family can be activated to become potent oncogenes, and many human and animal tumors express qualitatively or quantitatively altered receptors from this group. We have isolated and characterized a second egfr gene in the melanoma model fish Xiphophorus. Both Xiphophorus egfra and egfrb duplicates are co-orthologs of the mammalian egfr gene. Database analysis showed that not only egfr but also erbB3 and erbB4 are present as duplicates in some fish species. They originated from ancient duplication events that might be consistent with the hypothesis of a fish-specific genome duplication. In Xiphophorus, the egfrb gene underwent a second duplication that generated the melanoma-inducing oncogene Xmrk. The study and comparison of some of the functional characteristics of both Xiphophorus EGF receptors, including expression profile, ligand-binding abilities, and intracellular signal transduction revealed that Xiphophorus Egfra not only shares common features with Egfrb and the human EGFR but also shows significant differences in its functional characteristics. The mechanism of maintenance of these duplicates remains to be clarified.
Key Words: EGF receptor ? gene duplication ? receptor tyrosine kinase ? Xiphophorus
Introduction
Receptor tyrosine kinases (RTKs) comprise a diverse group of enzymes with intrinsic tyrosine kinase activity and a similar topology (Ullrich and Schlessinger 1990). All RTKs contain an extracellular domain where ligand binding takes place. A single transmembrane stretch connects the extracellular region with the intracellular domains, which include a highly conserved catalytic domain and a C-terminal tail that binds different intracellular signal transducers. Ligand binding and subsequent dimerization stimulate the receptor enzymatic activity. This results in autophosphorylation of their cytoplasmic domains and initiates a signaling cascade of protein phosphorylations inside the cell, providing a control on fundamental cellular processes such as growth and differentiation (Schlessinger 2000).
On the basis of sequence similarity and structural characteristics, RTKs have been grouped into families. The epidermal growth factor receptor (EGFR) family, or subclass I, of RTKs is one of the best studied. In mammals, this family consists of four members: EGFR/ErbB1, HER2/ErbB2, HER3/ErbB3, and HER4/ErbB4. The extra- cellular domain of these receptors includes four sub-domains. Subdomains S1 and S2 (or II and IV) are cysteine-rich regions whose spacing and number of cysteines is conserved among the members of the family; subdomains L1 and L2 (or I and III) are involved in ligand binding. EGF and a large group of related growth factors bind and activate specifically distinct receptors of this family, leading to the formation of homodimers and heterodimers. This results in different combinations of intracellular signals diversifying the biological responses triggered by these receptors (Olayioye et al. 2000).
The EGFR was the first member characterized in this family. It has been isolated in several mammals (Ullrich et al. 1984; Petch et al. 1990; Luetteke et al. 1994) and in chicken (Lax et al. 1988). The Drosophila Der receptor (Livneh et al. 1985) and let-23 from Caenorhabditis elegans (Aroian et al. 1990) are the single known invertebrate subclass I RTKs in flies and nematodes.
In the fish Xiphophorus, one receptor that belongs to the EGFR family has been isolated so far. It was named Xmrk (Xiphophorus melanoma receptor kinase) because its overexpression together with activating mutations induces melanoma formation in this fish (Wittbrodt et al. 1989; Gomez et al. 2001). Despite the high similarity of Xmrk with the human EGFR (hEGFR), there was evidence that this might be a new member of the subclass I gene family not described in any other species so far. Two copies of the Xmrk gene, the oncogene and the corresponding proto-oncogene have been identified. The proto-oncogene INV-Xmrk is invariably (INV) present in all teleost fish species tested so far (Dimitrijevic et al. 1998), but its physiological role remains to be elucidated. The oncogene can be found only in some species of Xiphophorus and represents an additional copy of the proto-oncogene produced by a recent gene duplication event. It is under a different transcriptional control and, as a result of this, is overexpressed in certain hybrids, leading to neoplastic transformation of the pigment cells (Adam, Dimitrijevic, and Schartl 1993).
The human receptors EGFR and Her2 are implicated in the development of numerous human cancers and appear constitutively active in a large number of tumors (Tang 1998; Olayioye et al. 2000). The physiological role of the EGFR has been extensively demonstrated in mammals. Knock-out mice for this gene present strain-dependent phenotypes that range from placental defects to postnatal abnormalities in multiple organs. The broad range of affected organs shows that this receptor is involved in a wide variety of cellular activities (Miettinen et al. 1995; Sibilia and Wagner 1995; Threadgill et al. 1995).
Some fish species are now generally accepted, valuable model systems for studying the normal function of molecules and developmental process as well as their malfunction in disease conditions. In contrast to the numerous studies carried out in mammals, little is known about the characteristics and action of the EGFR in fish. Only some reports exist in goldfish and rainbow trout that show the effect of heterologous EGF as multifunctional effector in the ovary and its direct effect in oocyte apoptosis and DNA synthesis (Srivastava and van der Kraak 1995; Wood and van der Kraak 2002). The use of heterologous antibodies has revealed an immunoreactivity in medaka embryos (Boomsma, Scott, and Walters 2001), rainbow trout liver (Newsted and Giesy 2000), and goldfish ovarian follicles (MacDougall and van der Kraak 1998). However, with the exception of the Xiphophorus Xmrk genes, none of the genes coding for the putative receptors had been isolated so far.
We describe here the isolation and characterization of a Xiphophorus egfr gene (egfra) (GenBank accession number AY230135) and present evidence that both the proto-oncogene INV-Xmrk (now egfrb) and egfra have been generated by a fish-specific duplication of an ancestral egfr gene. We show that, in fact, three members of the egfr family have been duplicated in fish. This finding is in accordance with the hypothesis of an ancient whole-genome duplication in the ray-finned fish lineage. The identification of egfra uncovers the existence of three egfr-related genes (egfra, egfrb, and Xmrk) and, thus, hypothetically eight members of the egfr family in Xiphophorus. Further analysis revealed that the new Xiphophorus egfra gene codes for a functional tyrosine kinase. It not only shares common features with Egfrb and with the human EGFR but also shows significant differences in its functional characteristics.
Materials and Methods
Library Screening and Isolation of Clones
A gt10 cDNA library from the Xiphophorus xiphidium embryonic cell line A2 (1.2 x 1010 pfu/ml) was screened at low-stringency conditions (40% form-amide/42°C) according to standard methods (Sambrook, Fritsch, and Maniatis 1989). The cDNA from Xmrk was used as probe. A second screen was performed to isolate the complete cDNA of egfra. In this case, a 1.45-kb PCR fragment containing part of egfra extracellular domain was used as a probe at high stringency conditions for hybridization (50% formamide/42°C). DNA fragments used as probes were randomly labeled with [-32P]dCTP.
DNA fragments contained in isolated phages were subcloned in pBlueScript using standard methods. Unidirectional, sequential deletions of pBlueScript inserts were obtained by exonuclease III-S1 digestion using a kit from Promega. Automatic sequencing was performed using ThermoSequenase fluorescent-labeled primer cycle sequencing kit and ALF automated sequencers (Amersham Pharmacia Biotech). Manual sequencing was performed using Sequenase (USB) and [-35S]dATP.
Sequence Analysis
Multiple sequence alignments were generated using "PileUp" of the GCG Wisconsin package version 10.0 (Genetics Computer Group, Madison, Wis.). Protein phylogenies were done with PAUP* (D.L. Swofford, Sinauer Associates, Inc., Sunderland, Mass.) with maximum parsimony (bootstrap analysis, 100 replicates), neighbor-joining (bootstrap analysis, 1,000 replicates [Saitou and Nei 1987]), and maximum likelihood (quartet puzzling, 10,000 puzzling steps [Strimmer and von Haeseler 1996]). Neighbor-joining was performed either considering all sites or only the unsaturated fraction of sites using AsaturA, a program recently applied to fish phylogenies (Van de Peer et al. 2002). The pairwise number of synonymous versus nonsynonymous substitutions per site between two aligned sequences was estimated with "Diverge" of GCG using an unambiguous alignment of DNA sequences (1,767 nt in length). The estimation of the number of substitutions per site at third codon positions was corrected for multiple events according to (Tajima and Nei 1984). Gene structure was analyzed using programs available on the NIX server (http://menu.hgmp.mrc.ac.uk/menu-bin/Nix). Deduced sequence assemblies from Takifugu rubripes and Danio rerio are available from J.-N.V on request.
Northern Blot Analysis
Total RNA from adult X. xiphidium organs and cell lines was isolated with TRIZOL reagent (GIBCO-BRL, Grand Island, N.Y.), and 30 μg of total RNA per lane were electrophoresed in a denaturing formaldehyde gel and blotted onto a Hybond-N nylon membrane (Amersham Pharmacia Biotech). Prehybridizations and hybridizations were carried out in 50% formamide at 42°C. DNA fragments used as probes were labeled as described above. Filters were washed at a maximum stringency of 1 x SSC/1% SDS at 60°C to 65°C. To control equal loading and integrity of the RNA, filters were stained with 0.04% methylene blue.
Construction of Expression Plasmids
To obtain an expression construct for egfra, its entire coding sequence was removed from the plasmid pBlueScript-egfra by EcoRI-SpeI digestion and ligated to the EcoRI and NheI digested expression vector pRK5(-HindIII) (Wittbrodt et al. 1992). This resulted in the expression construct pRK5-egfra. The chimeric construct HER-XER was generated in several steps. First, a silent mutation was introduced by PCR in the sequence of egfra (C2186 to G replacement) to generate an NarI restriction site. This newly introduced NarI site is located at the end of the transmembrane domain, in the same position where hegfr contains an NarI site. A 2-kb NarI/SpeI fragment containing the intracellular domains of egfra (XER) was ligated to a 6.9-kb NarI/NheI fragment containing the expression vector pRK5, and the region encoding the extracellular and transmembrane domains of the hEGFR (HER). This last fragment was obtained from the HER-mrk plasmid (Wittbrodt et al. 1992), which encodes the extracytoplasmic domains of hEGFR fused to the intracellular domains of Xmrk. The ligation sites and the egfra fragment amplified by PCR were checked by sequencing.
Cell Culture and Thymidine Incorporation
The IL-3-dependent mouse pro–B-cell line Ba/F3 (Palacios and Steinmetz 1985) was cultured in RPMI 1640, with 5% FCS and 5% of supernatant from X63Ag8-653 BVP m-IL-3 expressing cells (Karasuyama and Melchers 1988). BaF XER, BaF HER-XER, and BaF INV transfectants expressing Egfra, HER-Egfra, and Egfrb, respectively, were generated by electroporation (Easyject [Eurogentec] 200 V, 1800 μF) of BaF/3 cells with the expression plasmids pRK5-egfra, HER-XER, and pRK5-INV (Dimitrijevic et al. 1998) and a hygromycin-resistant plasmid (tgCMV/HyTK). After 10 days in medium containing hygromycin B (1 mg/ml), resistant clones were selected and tested for receptor expression on a western blot. Before stimulation with EGF or IL-3, the cells were washed twice with PBS and starved for 6 h in RPMI. Starved cells were treated with IL-3 (5% sup.) or the specified growth factor for the indicated times and collected by centrifugation. For thymidine incorporation assays, 4 x 104 cells per well of a 96-well plate were treated with the indicated factors and after 20 h, 0.5 μCi of 3H-thymidine per well was added for 4 h, and cells were harvested onto glass fiber filters. Incorporated 3H-thymidine was quantified by liquid scintillation counting, and IL-3 induced incorporation was set 100%.
Antibodies
mAb-108.1 is a mouse monoclonal antibody directed against the extracellular domain of hEGFR. 5E.2 was used as antiphosphotyrosine antibody. Both were kindly provided by A. Ullrich. Anti-STAT5b (C-17), anti-STAT3 (C-20), and anti-ERK2 (C-14) were from Santa Cruz Biotechnology. Rabbit polyclonal anti-phospho Akt and anti-Akt were obtained from New England Biolabs. Anti-phospho MAPK clone D12 was from Nanotools (Germany). For western blot analysis, all antibodies were used in a dilution of 1:1000 in TBST/1.5% BSA.
Cell Lysis, Immunoprecipitations, and Western Blotting
After stimulation, cells were rinsed twice with cold PBS and lysed in lysis buffer (50 mM HEPES, pH 7.5, 150 mM NaCl, 1.5 mM MgCl2, 1 mM EGTA, 10% glycerol, 1% Triton X-100, 1 mM PMSF, 1 mM Na3VO4, 10 μg/ml leupeptin, and 10 μg/ml aprotinin) as described (Wellbrock et al. 1998). Lysates were directly analyzed or incubated with the indicated antibodies and protein A-Sepharose for at least 1 h at 4°C. Proteins were analyzed by western blotting as described (Wellbrock et al. 1998).
Results
Structure of egfra, an Additional egfr Gene from Xiphophorus
The egfra cDNA was isolated from a cDNA library derived from the embryonic cell line A2 of Xiphophorus xiphidium. Four positive clones were obtained, all of which contained a 4,126-bp EcoRI fragment. It included an open reading frame of 3,627 nt encoding a protein of 1,209 amino acids that was named Egfra (fig. 1).
FIG. 1. Sequence alignment of vertebrate EGF receptors. Accession numbers: XP_044653 (EGFR Homo sapiens), P13387 (Egfr Gallus gallus), AY154653 (Egfra Danio rerio), AAD10500 (Egfrb Xiphophorus xiphidium), and P13388 (Xmrk Xiphophorus maculatus). Conserved amino acid residues are boxed in black, and conservative substitutions are in gray. The start and the end of each conserved domain is indicated by an arrow
Comparisons of the Egfra-deduced amino acid sequence with sequence databases showed the highest similarities with the EGFR from chicken, different mammalian species, and Xmrk from Xiphophorus, with overall amino acid similarities ranging from 67% to 69%. Egfra contains all the features of an RTK from the EGFR family. The predicted extracellular domain of 519 amino acids includes two cysteine rich domains (S1 and S2), containing 22 and 20 cysteines each. The spacing and number of these cysteines is in agreement with the conserved cysteine pattern of the members of the EGFR family (Abe et al. 1998). The extracellular domain is followed by a segment of hydrophobic amino acids that can form a single membrane-spanning region. The first 44 amino acids of the putative cytoplasmic part of Egfra are highly similar to the cytoplasmic juxtamembrane of other members of the EGFR family. The tyrosine kinase domain is the most conserved region. All 12 subdomains and structural determinants that characterize a functional tyrosine kinase domain are present (Hanks and Hunter 1995). Among them are residues important for ATP binding (glycine rich motif, K755, and E772), the D847 residue in the catalytic loop, and the DFG motif in the activation domain. These motifs are in the Egfra sequence in equivalent positions to other RTKs. The carboxy-terminal domain is the most divergent part of the protein when compared with the hEGFR and Xmrk/Egfrb. It contains a number of tyrosines that can act as docking sites for intracellular substrates.
Expansion of the EGFR Family by Gene Duplication in Fish
To search for the presence of genes encoding receptors from the EGFR family in other fish species, the almost completely sequenced genome of the Japanese pufferfish Takifugu rubripes (Aparicio et al. 2002) http://fugu.hgmp.mrc.ac.uk/), as well as the less complete zebrafish genome database (http://www.ensembl.org/Danio_rerio/blastview), were analyzed. Two egfr genes, egfra and egfrb, were identified in the pufferfish, and egfra was found in the zebrafish as well. All types of phylogenetic analyses performed with nucleic (not shown) and amino acid sequences (fig. 2) were concordant and indicated that egfra and egfrb were the result of the duplication of an egfr ancestor within the (ray-finned) fish lineage before the divergence of Xiphophorus, pufferfish, and zebrafish more than 100 MYA (Carroll 1997; Lydeard and Roe 1997). This analysis also shows that in Xiphophorus, the formerly called INV-Xmrk and ONC-Xmrk genes (for the proto-oncogene and the oncogene, now egfrb and Xmrk, respectively) are like egfra co-orthologs of the unique egfr gene found in birds and mammals.
FIG. 2. Phylogeny of receptor tyrosine kinases from the EGFR family. Bootstrap values (%) using maximum parsimony (100 replicates, first values) and neighbor-joining (1,000 replicates, second and third values), as well as the reliability values (%) for maximum likelihood (quartet puzzling, 10,000 puzzling steps, fourth values) are given. Neighbor-joining was performed either considering all sites (second values) or considering only the unsaturated fraction of sites (third values) using AsaturA (Van de Peer et al. 2002). The tree (maximum parsimony) has been rooted on invertebrate sequences. Branches with less than 50% support, using maximum parsimony, have been collapsed. A 630–amino acid alignment of sequences starting in the C-terminal end of the first receptor L domain and ending about 20 amino acids upstream from the catalytic domain was used for phylogenetic analysis. Accession numbers: XP_044653 (EGFR Homo sapiens), AAA17899 (Egfr Mus musculus), AAF14008 (Egfr Rattus norvegicus), P13387 (Egfr Gallus gallus), AAD10500 (Egfrb Xiphophorus xiphidium), P13388 (Xmrk Xiphophorus maculatus), CAA27060 (ERBB2 Homo sapiens), BAA23127 (ErbB2 Canis familiaris), CAA27059 (ErbB2 Rattus norvegicus), AAA35790 (ERBB3 Homo sapiens), AAC28498 (ErbB3 Rattus norvegicus), AAC34391 (ErbB3a Takifugu rubripes), NP_005226 (ERBB4 Homo sapiens), AAD08899 (ErBb4 Rattus norvegicus), AAD31764 (ErbB4 Gallus gallus), AAD26132 (DER Drosophila melanogaster), CAC35008 (Egfr Anopheles gambiae), and S70712 (CER Caenorhabditis elegans). Danio rerio ErbB3b sequence was deduced from the unfinished high-throughput genomic sequence AL591365. Takifugu rubripes sequences were deduced from data provided by the Fugu Genome Consortium (http://fugu.hgmp.mrc.ac.uk/): Egfra (S002539), Egfrb (S003004), ErbB2 (S002501), ErbB3b (S002894), ErbB4a (S001137), and ErbB4b (S002275). Danio rerio sequences were deduced from data from the zebrafish genome database (http://www.ensembl.org/Danio_rerio/blastview): Egfra (z06s003948), Erbb2 (z06s013389), Erbb3a (z06s021951), and Erbb4b (z06s009145)
Genes orthologous to erbB2, erbB3, and erbB4 from higher vertebrates were identified in the genomes of both Japanese pufferfish and zebrafish. Interestingly, duplicates were identified not only for egfr but also for erbB3 (pufferfish and zebrafish) and erbB4 (pufferfish), indicating that at least three different genes encoding receptors from the EGFR family are duplicated in the genome of T. rubripes and probably of other fish species. Analysis of the ratio of synonymous to nonsynonymous substitutions strongly suggested that all fish duplicates evolved under purifying selection and therefore probably do not correspond to pseudogenes: the (average) ratios between fish-specific duplicates (for example egfra versus egfrb) were 5.7, 9.0, and 14.1 for egfr, erbB3, and erbB4, respectively, compared with 7.9, 8.9, and 26.4 between their orthologs in higher vertebrates. Similar ratios indicating purifying selection were also observed in comparison involving fish orthologs, for example Xiphophorus and pufferfish egfra genes.
There is accumulating evidence for an ancient whole-genome duplication (as well as for more recent local events) having taken place during the course of evolution of the (ray-finned) fish lineage after its separation from the sarcopterygian lineage, including coelacanth, lungfish, and vertebrates (Wittbrodt, Meyer, and Schartl 1998; Meyer and Schartl 1999). The relatively low level of nucleotide identity between fish egfr (average 59.6%), erbB3 (65.5%), and erbB4 paralogs (75.1%) was rather suggestive of ancient events of duplication. The duplications that led to the formation of fish egfr, erbB3, and erbB4 paralogs probably arose before the divergence between zebrafish and pufferfish/Xiphophorus more than 100 MYA. Taylor et al. (2001) estimated that the putative fish-specific genome duplication occurred approximately 350 MYA by plotting the number of nucleotide substitutions at third-codon positions against divergence dates for different taxa. Using the same method, we found that the number of substitutions per site at third-codon positions between the different fish egfr (average 1.36), erbB3 (average 1.46), and erbB4 (average 0.96) paralogs was similar to the average value (1.02, SD = 0.24) obtained by Taylor et al. (2001) between unlinked zebrafish duplicates generated by the putative fish-specific genome duplication. As a basis for comparison, the same (average) value between orthologs from mammals and chicken (divergence about 300 MYA [Kumar and Hedges 1998]) was slightly lower than between the corresponding fish paralogs: 0.97 for egfr and 0.84 for erbB4. We conclude that the fish egfr, erbB3, and erbB4 paralogs resulted from ancient duplication event(s) and might be remnants of the proposed fish-specific genome duplication.
egfra and egfrb Show Different Expression Patterns in Xiphophorus
The expression pattern of egfra in different tissues of adult Xiphophorus and two cell lines was determined. A single transcript of 5.0 kb was detected in all analyzed normal tissues, although at different levels (fig. 3A). The highest expression was detected in kidney, followed by skin and gills and the embryonic cell line A2. This pattern is similar to that described for mouse egfr. The mammalian gene is expressed in different organs but preferentially in tissues with regenerative ephithelium such as skin (Adamson 1990). When comparing the expression of egfra and egfrb in the same tissues (fig. 3B), it appeared that the expression level of egfrb was lower than that of egfra. Besides, these Xiphophorus genes are preferentially expressed in different tissues. Whereas egfra shows a prominent expression in kidney and skin, egfrb transcripts are more abundant in liver compared with other tissues.
FIG. 3. Northern blot analysis. Expression of Xiphophorus egfra (A), egfrb (B), and Xmrk (C) in different organs of adult Xiphophorus, in the embryonic cell line A2, and in the melanoma derived cell line PSM. A 4-kb EcoRI fragment containing the whole coding sequence of egfra was used as a probe for (A). (B) and (C) were probed with a 3.6-kb XbaI/DraI fragment corresponding to the complete coding sequence of Xmrk. (B) is a longer exposure of (C) for detection of the 5.8-kb egfrb transcript. (C) corresponds to a short exposure where only the abundant 4.7-kb transcript of Xmrk can be seen. Equivalent RNA blotting was verified by methylene blue staining of ribosomal RNA
Analysis of melanoma and a melanoma-derived cell line (PSM), in which the Xmrk gene is overexpressed (fig. 3B and C), revealed a significant level of expression of egfra. However, although the expression in melanoma is notable, it is not higher than in normal skin and is even weaker than in kidney (fig. 3A).
Egfra Can Be Stimulated by hEGF but Not by TGF- or mEGF
To determine if Egfra is able to cross-react with ligands of EGFR-family receptors and whether it acts as a growth factor receptor, we chose the IL-3 dependent cell line Ba/F3. Ba/F3 cells lack endogenous expression of all EGFR family members (Riese et al. 1995), which excludes cross-reaction of defined ligands of this receptor family with other receptors than the ectopically expressed one.
Stimulation of Ba/F3 cells stably expressing Egfra (BaF XER cells) with TGF-, as well as murine EGF (mEGF), had a very weak but not significant mitogenic effect measured as 3H-thymidine incorporation during S-phase (fig. 4A). The presence of 100 nM human EGF (hEGF), however, resulted in approximately 24% of the amount of 3H-thymidine incorporation reached after IL-3 stimulation, which was set 100%. The induced DNA synthesis occurred in a dose-dependent manner, since 500 nM EGF induced an even higher response of 31%. These values indicate that the Egfra can be stimulated by hEGF and that this stimulation results in efficient intracellular signaling in Ba/F3 cells. In contrast to Egfra, Egfrb showed no cross-reactivity with either of the factors when expressed in Ba/F3 cells (BaF INV cells [fig. 4A]).
FIG. 4. Egfra stimulation with different growth factors. (A) The growth response to TGF-, mEGF, or hEGF of Ba/F3 cells expressing Egfra (BaF XER) and Ba/F3 cells expressing Egfrb (BaF INV cells) was determined by 3H-thymidine incorporation 24 h after stimulation. 3H-thymidine incorporation in response to IL-3 stimulation was set 100%. Asterisks (*) indicate P < 0.0001 versus control. (B) The growth response of Ba/F3 cells expressing either the Egfra (BaF XER), the hEGFR (BaF HER cells), or the chimeric HER-XER receptor (BaF HER-XER) to hEGF was determined by 3H-thymidine incorporation 24 h after stimulation. 3H-thymidine incorporation in response to IL-3 stimulation was set 100%
The hEGFR was chosen as a reference for the stimulation by the different growth factors. Strikingly, although the receptor has been shown to become catalytically activated by 8 nM hEGF in Ba/F3 cells (Morcinek et al. 2002), it was not able to trigger efficient cell cycle progression in these cells even when stimulated with 100 nM hEGF (Morcinek et al. 2002 and fig. 4B). This opposite behavior of hEGFR compared with the Egfra pointed to differences in the intracellular part of the receptors in terms of substrate binding and signal transduction.
To test this hypothesis, we generated a chimeric receptor, HER-XER. This chimera consists of the extracellular ligand-binding domain of the hEGFR and the intracellular part of Egfra. Treatment of Ba/F3 cells expressing HER-XER (BaF HER XER cells) with 100 nM hEGF resulted in 43% 3H-thymidine incorporation compared with IL-3 stimulation (fig. 4B). Whereas the response of Egfra was lower with approximately 25%, the hEGFR again showed no significant mitogenic activity (fig. 4B).
The stronger hEGF response of the HER-XER chimera compared with that of Egfra showed that although hEGF is able to cross-react with Egfra, its affinity to the fish receptor is lower than to the human receptor. In addition, the fact that HER-XER and HER, which are differing only in their intracellular domain, showed a different capacity in stimulating mitogenic signaling in Ba/F3 cells suggests that the Egfra induces a different signal transduction than the hEGFR.
Xiphophorus Egfra Signaling Includes Activation of MAP Kinase, PI3-kinase Signaling, and Stimulation of STAT3 and STAT5
To investigate the signal transduction of Egfra, we analyzed the activation of different intracellular substrates known to be activated by either Xmrk or the hEGFR in Ba/F3 cells. For these studies we used the HER-XER chimera because its stimulation by hEGF is more efficient.
First, HER-XER was immunoprecipitated with a monoclonal antibody raised against the extracellular domain of hEGFR and analyzed with antiphosphotyrosine after stimulation with hEGF. This experiment showed that the receptor becomes tyrosine phosphorylated after hEGF stimulation (fig. 5A), which makes it able to interact with intracellular substrates.
FIG. 5. Egfra signaling. (A) BaF HER-XER cells were either stimulated with 8 nM hEGF for 10 min (+) or were left untreated (–). HER-XER was immunoprecipitated (Ip) from lysates of treated or untreated cells with mAb-108.1, and immune complexes were analyzed with antiphosphotyrosine (anti-ptyr). (B) Lysates of EGF stimulated (+) or untreated (–) BaF HER-XER cells were used for immunoprecipitation with either anti-STAT3 or anti-STAT5 and immune complexes were analyzed with anti-ptyr. (C) Ba/F3 wild-type (BaF WT) and BaF HER-XER cells were either left untreated (–) or stimulated with EGF for 10 min (+), and cellular lysates were analyzed subsequently with anti-phospho MAPK and anti-ERK2 or with anti-phospho Akt and anti-Akt on a western blot
Next we analyzed the activation of the signal transducers and activators of transcription STAT3 and STAT5. It is known that whereas Xmrk is able to activate both STAT proteins in Ba/F3 cells, the hEGFR is incapable of activating STAT5 (Morcinek et al. 2002). Immunoprecipitation of STAT3 as well as STAT5 before and after stimulation of HER-XER revealed that the receptor induces the tyrosine phosphorylation of both STAT proteins (fig. 5B). In the immunoprecipitates of STAT3, the receptor was coprecipitated and detectable as a 165-kDa tyrosine phosphorylated protein after stimulation with EGF.
The stimulation of HER-XER resulted in activation of the MAP kinase pathway as seen by the specific phosphorylation of the MAP kinases ERK1 and ERK2 (fig. 5C). The MAP kinase cascade was initiated by binding of GRB2 and Shc to HER-XER, both of which could be coimmunoprecipitated with the receptor (data not shown). Also an activating serine phosphorylation of Akt, a protein acting further downstream in the PI3-kinase signaling pathway, was significantly increased after stimulation of the HER-XER receptor in Ba/F3 cells (fig. 5C).
Discussion
We have isolated and characterized an additional epidermal growth factor receptor gene in the fish Xiphophorus. This gene, egfra, encodes a protein that contains all the structural features of subclass I receptor tyrosine kinases. Two other receptors closely related to the hEGFR had been already described in Xiphophorus. One of them, INV-Xmrk (Dimitrijevic et al. 1998), now renamed as Egfrb, is present in all fish of the genus Xiphophorus, as well as in more divergent teleost species. The egfrb gene represents the proto-oncogene of the melanoma-inducing gene ONC-Xmrk (now Xmrk) that exists only in some Xiphophorus species (Wittbrodt et al. 1989). The Xmrk (onco)gene arose more than 5 to 6 MYA by a local gene duplication event from egfrb, and both genes are linked and located on the sex chromosomes (Adam, Dimitrijevic, and Schartl 1993). Earlier linkage analysis had already revealed the presence of another locus encoding a third egfr-related gene in an autosomal region (Harless et al. 1990). Sequence comparison of egfra with the probe used to determine the linkage group (linkage group VI [Harless et al. 1990]) revealed that this egfr-related gene is egfra.
Egfra and egfrb, as well as the duplicated EGFR family members present in zebrafish and pufferfish, might be the result of whole-genome duplication in teleost fish. This origin was already proposed to explain the existence of seven Hox gene clusters in zebrafish (Amores et al. 1998) in contrast with the four clusters existing in tetrapods. More recently, additional Hox clusters have been described in medaka (Naruse et al. 2000), tilapia (Malaga-Trillo and Meyer 2001), and Fugu (Aparicio et al. 2002).
One of the characteristics of RTKs is the binding of a ligand in the extracellular domain, but no EGFR ligand has been isolated in fish so far. Nevertheless, a few reports exist where the application of different mammalian EGFR ligands in fish tissues in vitro provoked a physiological response (Srivastava and van der Kraak 1995; Wood and van der Kraak 2002). In the present work, we have shown that Ba/F3 cells containing Egfra are able to proliferate when hEGF is added, indicating that this ligand is able to bind and activate Egfra. However, this ligand could not stimulate proliferation when applied to a Ba/F3 cell line stably expressing Egfrb. This may indicate that the extracellular domains of both Xiphophorus EGF receptors differ in specific residues responsible for ligand binding and may point to the binding of different (and perhaps duplicated) ligands in Xiphophorus. The crystal structure of the hEGFR in complex with hEGF has been recently determined (Ogiso et al. 2002). The study has identified hEGF binding sites in domains I and III of the hEGFR. Two hEGFR residues involved in direct interactions with hEGF are conserved in Egfra but not in Egfrb. This could explain the different ligand-binding properties of the fish receptors. The side chain of Phe357 in the hEGFR hydrophobically interacts with Tyr13 of hEGF. In Egfra, a tyrosine is existing in an equivalent position (Tyr391). This would allow as well the interaction with Tyr13 through the aromatic side chain. However, the histidine (His383) in Egfrb at the equivalent position does not allow this contact. The importance of this ligand-receptor contact was shown by substitution of Tyr13 with polar residues that blocked this interaction and considerably decreased receptor binding activity (Tadaki and Niyogi 1993). In addition the Gln384 side chain of hEGFR hydrogen bonds with Gln43 and Arg45 of hEGF. This Gln is also present in Egfra (Gln418) but not in Egfrb, where a Met (Met410) at an equivalent position prevents the formation of this hydrogen bond. Whether these residues are the only ones responsible for the different binding properties between Egfra and Egfrb is difficult to say because other amino acid differences in the fish receptors could account for a different local structure in key positions.
The MAP kinase cascade was activated by the HER-XER chimera resulting in phosphorylation of ERK1 and ERK2. MAP kinase signaling was initiated by binding of the receptor to the adaptor proteins Shc and GRB2 (data not shown) for which the Egfra contains potential binding sites in its carboxy-terminus. Tyr1097 (YMNH) and Tyr1113 (YPNY) both represent the YXNX motif that has been shown to be selectively bound by Grb2 SH2 domain when tyrosine phosphorylated (Songyang et al. 1993). Tyr1097 in Egfra is placed in a homologous position to hEGFR Tyr1068 and Xmrk Tyr1033, both binding sites for the GRB2 SH2 domain (Batzer et al. 1994; Wellbrock and Schartl 1999). Egfra Tyr1113 is also contained in a YXNX-binding motif; however, no tyrosine is present in hEGFR or Xmrk in this position. The sequence amino-terminal to Tyr1181 (NPDY) of Egfra fits with the NPXY-binding motif for the PTB (phosphotyrosine-binding) domain of Shc (Batzer et al. 1995). Tyr1181 is placed in an equivalent site to Tyr1148 of the hEGFR, which binds Shc (Batzer et al. 1995). However, whether these binding sites in Egfra are used by the receptor to interact with GRB2 and Shc, respectively, remains to be determined.
Activation of the Egfra also led to phosphorylation and thus activation of Akt. This protein is a downstream effector of Pi3-kinase and has been shown to be involved in antiapoptotic signaling (Coffer, Jin, and Woodgett 1998). Both Egfra and Egfrb, as well as Xmrk and hEGFR, lack the YXXM motif for binding the SH2 domains of the PI3-kinase regulatory subunit p85. It has been shown that hEGFR and Xmrk can indirectly activate PI3-kinase by binding to other intracellular proteins that act as adaptors (Rodrigues et al. 2000; Wellbrock and Schartl 2000). Therefore, it is feasible that the Egfra makes use of a similar mechanism for activating PI3-kinase pathways.
Analysis of the activation of STAT proteins revealed that both STAT3 and STAT5 are activated by the intracelluar part of Egfra expressed in Ba/F3 cells. STAT3 can be also activated by hEGFR as well as Xmrk in this cell system; however, in contrast to Xmrk, hEGFR is incapable of activating STAT5 (Morcinek et al. 2002). STAT5 has been shown to be a crucial factor involved in melanoma development in Xiphophorus, and it is responsible for maintaining cell progression of Ba/F3 cells sustained by Xmrk signaling (Wellbrock et al. 1998; Morcinek et al. 2002). So, the fact that the Egfra activates STAT5 could account for the ability to induce proliferation in Ba/F3 cells. In this cellular context, Egfra and Egfrb/Xmrk behave in a similar way and very differently from hEGFR, which is unable to activate STAT5 and to induce mitogenic activity. These data indicate that Egfra and Egfrb have kept similar characteristics that are different from that of the hEGFR.
Only one member of the EGFR family has been identified in several invertebrates (Livneh et al. 1985; Aroian et al. 1990), whereas four members exist in mammals. Different evolutionary analyses (Stein and Staros 2000; Martin 2001) support the appearance of the EGFR family by two rounds of duplication of an ancestral gene. A first duplication would have generated an erbB1/erbB2 precursor and an erbB3/erbB4 precursor. A second duplication led to the existence of four receptors as found in mammals today. A fish-specific genome duplication could then have generated the additional receptors in fish, leading to more subclass I RTKs in these animals. On top of that, a lineage-specific gene duplication led to an additional egfrb copy in Xiphophorus that became the Xmrk oncogene.
Different fates have been proposed for duplicated genes (reviewed by Prince and Pickett [2002]). A classical model predicts that one of the copies can degenerate by accumulation of deleterious mutations. This is clearly not the case for egfra and egfrb, which both encode entirely functional tyrosine kinases. Another well-documented model proposes that each duplicated copy will retain only some of their original functions, complementing each other to fulfill the original ancestral gene activities (subfunctionalization). Less frequently, one of the duplicated copies may acquire a new function, whereas the other copy retains the original one (neofunctionalization).
The functional study of the Egfra and its comparison with Egfrb and hEGFR revealed that none of the above mentioned evolutionary fates can be assigned to these fish duplicates. Although Egfra and Egfrb present clear differences in their ligand-binding ability, their pattern of expression in adult fish is partially overlapping, and their signal transduction seems quite similar. This might provoke the question of a certain redundancy for Egfra and Egfrb. To explain preservation of redundant multidomain proteins, Gibson and Spring (1998) proposed that accumulation of mutations in proteins with a significant functional role will occur more slowly than expected, as damaging a single domain can give rise to a dominant-negative phenotype affecting a whole network of proteins. On the other hand, the significant contribution of preserved duplicated genes to genetic robustness against loss-of-function mutations has to be considered (Meyer 2003; Gu et al. 2003). However, some kind of asymmetry in the duplicates' function is necessary to preserve redundancy (Krakauer and Nowak 1999). In the case of the EGFR, its involvement in a wide variety of cellular processes during development and in adult organisms makes a dissection of its functions extremely difficult.
Subclass I RTKs are potent tumor genes after oncogenic activation. Some fish species are known to be especially susceptible to developing tumors after exposure to environmental and experimental carcinogens, and some fish display a considerable rate of spontaneous tumors. It is tempting to speculate that the enlarged gene families, many of which are involved in the regulation of cell growth and proliferation, confer more targets for oncogenic activation to fish genomes.
Acknowledgements
We thank Nicola Dimitrijevic for the isolation of a partial cDNA clone of egfra. This work is supported by grants to C.W. and M.S. from the Deutsche Forschungsgemeinschaft through the SFB487 ("Regulatorische Membranproteine") and the SFB465 ("Entwicklung und Manipulation pluripotenter Zellen") and the Fonds der Chemischen Industrie. J.-N.V. is supported by the BioFuture program of the German Bundesministerium für Bildung und Forschung. Takifugu data have been provided freely by the Fugu Genome Consortium for use in this publication only. We are grateful to Yves Van de Peer (Gent, Belgium) and Tancred Frickey (Tübingen, Germany) for their help concerning the AsaturA program.
Literature Cited
Abe, Y., M. Odaka, F. Inagaki, I. Lax, J. Schlessinger, and D. Kohda. 1998. Disulfide bond structure of human epidermal growth factor receptor. J. Biol. Chem. 273:11150-11157.
Adam, D., N. Dimitrijevic, and M. Schartl. 1993. Tumor suppression in Xiphophorus by an accidentally acquired promoter. Science 259:816-819.
Adamson, E. D. 1990. Developmental activities of the epidermal growth factor receptor. Curr. Top. Dev. Biol. 24:1-29.
Amores, A., A. Force, and Y. L. Yan, et al. (13 co-authors). 1998. Zebrafish hox clusters and vertebrate genome evolution. Science 282:1711-1714.
Aparicio, S., J. Chapman, and E. Stupka, et al. (41 co-authors). 2002. Whole-genome shotgun assembly and analysis of the genome of Fugu rubripes. Science 297:1301-1310.
Aroian, R. V., M. Koga, J. E. Mendel, Y. Ohshima, and P. W. Sternberg. 1990. The let-23 gene necessary for Caenorhabditis elegans vulval induction encodes a tyrosine kinase of the EGF receptor subfamily. Nature 348:693-699.
Batzer, A. G., P. Blaikie, K. Nelson, J. Schlessinger, and B. Margolis. 1995. The phosphotyrosine interaction domain of Shc binds an LXNPXY motif on the epidermal growth factor receptor. Mol. Cell Biol. 15:4403-4409.
Batzer, A. G., D. Rotin, J. M. Urena, E. Y. Skolnik, and J. Schlessinger. 1994. Hierarchy of binding sites for Grb2 and Shc on the epidermal growth factor receptor. Mol. Cell Biol. 14:5192-5201.
Boomsma, R. A., H. Scott, and K. Walters. 2001. Immunocytochemical localization of epidermal growth factor receptor in early embryos of the Japanese medaka fish (Oryzias latipes). Histochem. J. 33:37-42.
Carroll, R. L. 1997. Patterns and processes of vertebrate evolution. Cambridge University Press Cambridge, UK.
Coffer, P. J., J. Jin, and J. R. Woodgett. 1998. Protein kinase B (c-Akt): a multifunctional mediator of phosphatidylinositol 3-kinase activation. Biochem. J. 335:(pt 1): 1-13.
Dimitrijevic, N., C. Winkler, C. Wellbrock, A. Gomez, J. Duschl, J. Altschmied, and M. Schartl. 1998. Activation of the Xmrk proto-oncogene of Xiphophorus by overexpression and mutational alterations. Oncogene 16:1681-1690.
Gibson, T. J., and J. Spring. 1998. Genetic redundancy in vertebrates: polyploidy and persistence of genes encoding multidomain proteins. Trends Genet. 14:46-49.
Gomez, A., C. Wellbrock, H. Gutbrod, N. Dimitrijevic, and M. Schartl. 2001. Ligand-independent dimerization and activation of the oncogenic Xmrk receptor by two mutations in the extracellular domain. J. Biol. Chem. 276:3333-3340.
Gu, Z., L. M. Steinmetz, X. Gu, C. Scharfe, R. W. Davis, and W. H. Li. 2003. Role of duplicate genes in genetic robustness against null mutations. Nature 421:63-66.
Hanks, S. K., and T. Hunter. 1995. Protein kinases 6. The eukaryotic protein kinase superfamily: kinase (catalytic) domain structure and classification. FASEB J. 9:576-596.
Harless, J., R. Svensson, K. D. Kallman, D. C. Morizot, and R. S. Nairn. 1990. Assignment of an erbB-like DNA sequence to linkage group VI in fishes of the genus Xiphophorus (Poeciliidae). Cancer Genet. Cytogenet. 50:45-51.
Karasuyama, H., and F. Melchers. 1988. Establishment of mouse cell lines which constitutively secrete large quantities of interleukin 2, 3, 4 or 5, using modified cDNA expression vectors. Eur. J. Immunol. 18:97-104.
Krakauer, D. C., and M. A. Nowak. 1999. Evolutionary preservation of redundant duplicated genes. Semin. Cell Dev. Biol. 10:555-559.
Kumar, S., and S. B. Hedges. 1998. A molecular timescale for vertebrate evolution. Nature 392:917-920.
Lax, I., A. Johnson, R. Howk, J. Sap, F. Bellot, M. Winkler, A. Ullrich, B. Vennstrom, J. Schlessinger, and D. Givol. 1988. Chicken epidermal growth factor (EGF) receptor: cDNA cloning, expression in mouse cells, and differential binding of EGF and transforming growth factor alpha. Mol. Cell Biol. 8:1970-1978.
Livneh, E., L. Glazer, D. Segal, J. Schlessinger, and B. Z. Shilo. 1985. The Drosophila EGF receptor gene homolog: conservation of both hormone binding and kinase domains. Cell 40:599-607.
Luetteke, N. C., H. K. Phillips, T. H. Qiu, N. G. Copeland, H. S. Earp, N. A. Jenkins, and D. C. Lee. 1994. The mouse waved-2 phenotype results from a point mutation in the EGF receptor tyrosine kinase. Genes Dev. 8:399-413.
Lydeard, C., and K. J. Roe. 1997. The phylogenetic utility of the mitochondrial cytochrome b gene for inferring relationships among actinopterygian fishes. Pp. 285–303 in T. C. Kocher and C. A. Stepien, eds. Molecular systematics of fishes. Academic Press. San Diego, Calif.
MacDougall, T. M., and G. van der Kraak. 1998. Peptide growth factors modulate prostaglandin E and F production by goldfish ovarian follicles. Gen. Comp. Endocrinol. 110:46-57.
Malaga-Trillo, E., and A. Meyer. 2001. Genome duplications and accelerated evolution of Hox genes and cluster architecture in teleost fishes. Am. Zool. 41:676-686.
Martin, A. 2001. Is tetralogy true? Lack of support for the "one-to-four rule". Mol. Biol. Evol. 18:89-93.
Meyer, A. 2003. Molecular evolution: duplication, duplication. Nature 421:31-32.
Meyer, A., and M. Schartl. 1999. Gene and genome duplications in vertebrates: the one-to-four (-to-eight in fish) rule and the evolution of novel gene functions. Curr. Opin. Cell Biol. 11:699-704.
Miettinen, P. J., J. E. Berger, J. Meneses, Y. Phung, R. A. Pedersen, Z. Werb, and R. Derynck. 1995. Epithelial immaturity and multiorgan failure in mice lacking epidermal growth factor receptor. Nature 376:337-341.
Morcinek, J. C., C. Weisser, E. Geissinger, M. Schartl, and C. Wellbrock. 2002. Activation of STAT5 triggers proliferation and contributes to anti-apoptotic signalling mediated by the oncogenic Xmrk kinase. Oncogene 21:1668-1678.
Naruse, K., S. Fukamachi, and H. Mitani, et al. (20 co-authors). 2000. A detailed linkage map of medaka, Oryzias latipes: comparative genomics and genome evolution. Genetics 154:1773-1784.
Newsted, J. L., and J. P. Giesy. 2000. Epidermal growth factor receptor-protein kinase interactions in hepatic membranes of rainbow trout (Oncorhynchus mykiss). Fish Physiol. Biochem. 22:181-189.
Ogiso, H., R. Ishitani, and O. Nureki, et al. (11 co-authors). 2002. Crystal structure of the complex of human epidermal growth factor and receptor extracellular domains. Cell 110:775-787.
Olayioye, M. A., R. M. Neve, H. A. Lane, and N. E. Hynes. 2000. The ErbB signaling network: receptor heterodimerization in development and cancer. EMBO J. 19:3159-3167.
Palacios, R., and M. Steinmetz. 1985. Il-3-dependent mouse clones that express B-220 surface antigen, contain Ig genes in germ-line configuration, and generate B lymphocytes in vivo. Cell 41:727-734.
Petch, L. A., J. Harris, V. W. Raymond, A. Blasband, D. C. Lee, and H. S. Earp. 1990. A truncated, secreted form of the epidermal growth factor receptor is encoded by an alternatively spliced transcript in normal rat tissue. Mol. Cell Biol. 10:2973-2982.
Prince, V. E., and F. B. Pickett. 2002. Splitting pairs: the diverging fates of duplicated genes. Nat. Rev. Genet. 3:827-837.
Riese, D. J., T. M. van Raaij, G. D. Plowman, G. C. Andrews, and D. F. Stern. 1995. The cellular response to neuregulins is governed by complex interactions of the erbB receptor family. Mol. Cell Biol. 15:5770-5776.
Rodrigues, G. A., M. Falasca, Z. Zhang, S. H. Ong, and J. Schlessinger. 2000. A novel positive feedback loop mediated by the docking protein Gab1 and phosphatidylinositol 3-kinase in epidermal growth factor receptor signaling. Mol. Cell Biol. 20:1448-1459.
Saitou, N., and M. Nei. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual. 2 edition. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
Schlessinger, J. 2000. Cell signaling by receptor tyrosine kinases. Cell 103:211-225.
Sibilia, M., and E. F. Wagner. 1995. Strain-dependent epithelial defects in mice lacking the EGF receptor. Science 269:234-238.
Songyang, Z., S. E. Shoelson, and M. Chaudhuri, et al. (14 co-authors). 1993. SH2 domains recognize specific phosphopeptide sequences. Cell 72:767-778.
Srivastava, K. R., and G. van der Kraak. 1995. Multifactorial regulation of DNA synthesis in goldfish ovarian follicles. Gen. Comp. Endocrinol. 100:397-403.
Stein, R. A., and J. V. Staros. 2000. Evolutionary analysis of the ErbB receptor and ligand families. J Mol. Evol. 50:397-412.
Strimmer, K., and A. von Haeseler. 1996. Quartet puzzling: a quartet maximum-likelihood method for reconstructing tree topologies. Mol. Biol. Evol. 13:964-969.
Swofford, D. L. 1998. PAUP*: phylgenetic analysis using parsimony (*and other methods), version 4.0. Sinauer Associates, Sunderland, Mass.
Tadaki, D. K., and S. K. Niyogi. 1993. The functional importance of hydrophobicity of the tyrosine at position 13 of human epidermal growth factor in receptor binding. J Biol. Chem. 268:10114-10119.
Tajima, F., and M. Nei. 1984. Estimation of evolutionary distance between nucleotide sequences. Mol. Biol. Evol. 1:269-285.
Tang, C. K., and M. E. Lippman. 1998. EGF family receptors and their ligands in human cancer. Pp. 113–165 in B. W. O'Malley, ed. Hormones and signaling, Vol. I. Academic Press, San Diego, Calif.
Taylor, J. S., P. Y. Van de, I. Braasch, and A. Meyer. 2001. Comparative genomics provides evidence for an ancient genome duplication event in fish. Philos. Trans. R. Soc. Lond B Biol. Sci. 356:1661-1679.
Threadgill, D. W., A. A. Dlugosz, L. A. Hansen, T. Tennenbaum, U. Lichti, D. Yee, C. LaMantia, T. Mourton, K. Herrup, and R. C. Harris. 1995. Targeted disruption of mouse EGF receptor: effect of genetic background on mutant phenotype. Science 269:230-234.
Ullrich, A., L. Coussens, J. S. Hayflick, T. J. Dull, A. Gray, A. W. Tam, J. Lee, Y. Yarden, T. A. Libermann, and J. Schlessinger. 1984. Human epidermal growth factor receptor cDNA sequence and aberrant expression of the amplified gene in A431 epidermoid carcinoma cells. Nature 309:418-425.
Ullrich, A., and J. Schlessinger. 1990. Signal transduction by receptors with tyrosine kinase activity. Cell 61:203-212.
Van de Peer, Y., T. Frickey, J. S. Taylor, and A. Meyer. 2002. Dealing with saturation at the amino acid level: a case study based on anciently duplicated zebrafish genes. Gene 295:205-211.
Wellbrock, C., E. Geissinger, A. Gomez, P. Fischer, K. Friedrich, and M. Schartl. 1998. Signalling by the oncogenic receptor tyrosine kinase Xmrk leads to activation of STAT5 in Xiphophorus melanoma. Oncogene 16:3047-3056.
Wellbrock, C., and M. Schartl. 1999. Multiple binding sites in the growth factor receptor Xmrk mediate binding to p59fyn, GRB2 and Shc. Eur. J. Biochem. 260:275-283.
Wellbrock, C., and M. Schartl. 2000. Activation of phosphatidylinositol 3-kinase by a complex of p59fyn and the receptor tyrosine kinase Xmrk is involved in malignant transformation of pigment cells. Eur. J. Biochem. 267:3513-3522.
Wittbrodt, J., D. Adam, B. Malitschek, W. Maueler, F. Raulf, A. Telling, S. M. Robertson, and M. Schartl. 1989. Novel putative receptor tyrosine kinase encoded by the melanoma-inducing Tu locus in Xiphophorus. Nature 341:415-421.
Wittbrodt, J., R. Lammers, B. Malitschek, A. Ullrich, and M. Schartl. 1992. The Xmrk receptor tyrosine kinase is activated in Xiphophorus malignant melanoma. EMBO J. 11:4239-4246.
Wittbrodt, J., A. Meyer, and M. Schartl. 1998. More genes in fish? Bioessays 20:511-515.
Wood, A. W., and G. van der Kraak. 2002. Inhibition of apoptosis in vitellogenic ovarian follicles of rainbow trout (Oncorhynchus mykiss) by salmon gonadotropin, epidermal growth factor, and 17beta-estradiol. Mol. Reprod. Dev. 61:511-518.(Ana Gómez1, Jean-Nicolas )