help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2003-0913
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Safi, R.
Right arrow Articles by Laudet, V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Safi, R.
Right arrow Articles by Laudet, V.
Endocrinology Vol. 145, No. 2 760-772
Copyright © 2004 by The Endocrine Society

The Axolotl (Ambystoma mexicanum), a Neotenic Amphibian, Expresses Functional Thyroid Hormone Receptors

Rachid Safi, Stéphanie Bertrand, Oriane Marchand, Marilyne Duffraisse, Amaury de Luze, Jean-Marc Vanacker, Marie Maraninchi, Alain Margotat, Barbara Demeneix and Vincent Laudet

Centre National de la Recherche Scientifique, Unité Mixte de Recherche (UMR) 5665, Laboratoire de Biologie Moléculaire et Cellulaire de la Cellule, Ecole Normale Supérieure de Lyon (R.S., S.B., O.M., M.D., J.-M.V., V.L.), 69364, Lyon, France; Centre National de la Recherche Scientifique, UMR 5166, Evolution des Régulations Endocriniennes, Muséum National d’Histoire Naturelle (A.d.L., B.D.), 75321 Paris, France; and Institut National de la Santé et de la Recherche Médicale, Unité-476, Faculté de Médecine (M.M., A.M.), 13385 Marseille, France

Address all correspondence and requests for reprints to: Dr. Vincent Laudet, Centre National de la Recherche Scientifique, Unité Mixte de Recherche 5665, Laboratoire de Biologie Moléculaire et Cellulaire, Ecole Normale Supérieure de Lyon, 46 allée d’ Italie, 69364 Lyon Cedex 07, France. E-mail: vincent.laudet{at}ens-lyon.fr.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Neotenic amphibians such as the axolotl (Ambystoma mexicanum) are often unable to undergo metamorphosis under natural conditions. It is thought that neoteny represents a deviation from the standard course of amphibian ontogeny, affecting the thyroid axis at different levels from the central nervous system to peripheral organs. Thyroid hormone receptors (TRs) that bind the thyroid hormone (TH) T3 have been described in axolotl. However, the full sequences of TR were needed to better characterize the TH response and to be able to assess their functional capacity at the molecular level. We report that each of the {alpha} and ß axolotl TRs bind both DNA and TH, and they activate transcription in response to TH in a mammalian cell-based transient transfection assay. Moreover, both TRs are expressed in axolotl tissues. Interestingly, each TR gene generates alternatively spliced isoforms, harboring partial or total deletions of the ligand-binding domain, which are expressed in vivo. Further, we found that in the axolotl, TH regulates the expression of stromelysin 3 and collagenase 3, which are TH target genes in Xenopus. Taken together, these results suggest that axolotl TRs are functional and that the molecular basis of neoteny in the axolotl is not linked to a major defect in TH response in peripheral tissues.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
AMPHIBIAN METAMORPHOSIS is a spectacular postembryonic developmental process that is governed by thyroid hormones (TH). In the classical model Xenopus laevis as well as in other species, a peak of the circulating concentration of TH coincides with the climax of metamorphosis (1 ; reviewed in Refs. 2 and 3). Furthermore, the administration of TH in anuran tadpoles, even before the normal period of metamorphosis, triggers this process (4). On the contrary, blockage of endogenous TH production by goitrogens such as propylthiouracil or potassium perchlorate blocks metamorphic induction and produces giant tadpoles.

There is, however, a wide variety in the extent and importance of metamorphosis across different amphibian classes. In both urodeles and anourans, alternative strategies, such as direct development and neoteny, can be found (reviewed in Refs.5, 6, 7). Neoteny, the ability to reproduce while conserving larval characters, can be either facultative, such as in many species of the genus Ambystoma, or obligatory, such as in Necturus, Proteus, or Siren. Salamanders of the genus Ambystoma, one of which is the axolotl, Ambystoma mexicanum, are a complex monophyletic group that lives in North America from northern Mexico to southern Canada (8, 9). Natural populations of Ambystoma can be purely neotenic, transforming, or polymorphic (i.e. some specimens undergo metamorphosis, whereas others reproduce as larvae), and it has been suggested that this represents an ecological adaptation useful for a better exploitation of the available resources (5). Phylogenetical analyses have shown that metamorphic failure evolved independently several times during Ambystoma evolution, suggesting that there is an increased ability to select this life history trait in all species of this genus (8). In addition, the phylogenetic data suggest that different defects may be responsible for the appearance of metamorphic failure in different Ambystoma species. In the laboratory, most Ambystoma species are able to undergo metamorphosis when treated with TH, whereas the concentration of TH necessary to trigger metamorphosis is highly variable (10). The most widely studied model system for neoteny, the laboratory axolotl, is at least in part derived from wild A. mexicanum and A. tigrinum.

A number of studies have focused on the molecular basis of metamorphic failure in the axolotl. The fact that the tissues of axolotl can respond to TH suggests that metamorphic failure in this animal resides mainly in abnormal TH production and a low rate of secretion of TSH (11). Several researchers also suggested that in contrast to Xenopus, the axolotl pituitary responds only very weakly to TRH (12, 13). Although it is likely that metamorphic failure in the axolotl can be attributed in large part to the low circulating levels of TH, hormone deficiency may not be the only factor involved. This is why other researchers used biochemical methods to assess the presence of TH receptors (TRs) as well as deiodinase activity in axolotl tissues (14). Further, using heterologous probes it has been shown that the TRß gene can be up-regulated by TH treatment in axolotl (15). However, the full sequences of axolotl TRs, their expression levels, their response to TH, and their functional capacity have yet to be analyzed.

Most, if not all, of the biological effects of TH, i.e. the precursor, T4 (L-T4 or 3,3',5,5'-tetraiodo-L-thyronine), and the more active compound, T3 (3,3',5-triiodo-L-thyronine), are mediated through two TRs, TR{alpha} and TRß (NR1A1 and NR1A2, respectively, in the nomenclature for nuclear receptors) (16, 17, 18). TRs are members of a large superfamily of nuclear receptors that act as ligand-dependent transcription factors (reviewed in Ref.19). This family also includes vitamin D3, steroid, and retinoid receptors. Nuclear receptors share conserved domains that are critical for their function: a short DNA-binding domain encompassing two zinc finger motifs (DBD) and a C-terminal hydrophobic ligand-binding domain (LBD) (19). The receptors bind specific DNA sequences termed hormone response elements (HREs), that contain repeats of a consensus half-site sequence (5'-AGGTCA-3'). TRs, like many members of this superfamily, form heterodimers with retinoid X receptor (RXR; NR2B), a receptor specific for 9-cis-retinoic acid, and bind to TH response element (TRE). These sequences usually consist of two half-sites arranged as a direct repeat separated by four nucleotides (DR4) (20), but sometimes may also include a palindromic element (HREpal) (21 ; for a review, see Ref.22). In the absence of hormone, it has been shown that the RXR/TR heterodimer binds to corepressors to inhibit transcription (23, 24). In the presence of TH, the receptor undergoes a conformational change that induces the release of the corepressors and the recruitment of coactivators (reviewed in Ref.25). This change results in a potent activation of target gene transcription.

To increase our knowledge of TR function in axolotl, we decided to study the TH response and TR genes of the laboratory axolotl in greater detail. We had previously cloned PCR products corresponding to partial sequences of two TR genes from the axolotl. Here, we obtained full-length clones and show that these two TR genes encode functional proteins that are able to bind DNA and TH like their Xenopus homologs. We also show that when transiently transfected in mammalian cells, these receptors are able to activate transcription in the presence of TH. Interestingly, we found that several isoforms of both TR{alpha} and TRß are likely to be produced by alternative splicing. Finally, we found that the TR{alpha} and TRß genes are expressed in both larval and adult axolotl and that the genes encoding stromelysin 3 and collagenase 3, known to be TH responsive in Xenopus, are also regulated by TH in axolotl.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Animals
The laboratory has the official approval of the Centre National de la Recherche Scientifique for axolotl care, and all experiments were performed according to their guidelines. Laboratory axolotl were maintained in 10-liter tanks containing dechlorinated and filtered water at 18–20 C. Water was changed daily, and the animals were fed worms. The animals were treated with 5 x 10-8 M T3 for 2, 5, or 10 d, and the water with hormone was changed daily. Four or five control animals and eight or nine treated animals were used at each point. Animals were anesthetized with MS-222 (3-aminobenzoic acid ethyl ester, Sigma-Aldrich Corp., St. Louis, MO), which was added to the water, and were subsequently killed by decapitation.

Isolation of axolotl TR cDNAs
We previously isolated RT-PCR fragments from axolotl corresponding to the central region of the TR{alpha} and TRß genes (26). The fragments were used to screen a cDNA library constructed in {lambda}ZAP GT11 from axolotl blastoma cells, a gift from Carl Seguin. Three clones were isolated for each gene of 4 x 106 plaque-forming units. Hybridization was performed overnight at 42 C in 50% formamide, 5x standard saline citrate, 5x Denhardt’s solution, 0.5% sodium dodecyl sulfate, and 0.1 mg/ml denatured salmon sperm DNA. Membranes were washed twice in 2x standard saline citrate/0.1% sodium dodecyl sulfate at 37 C for 15 min.

Each cDNA was sequenced with a 373A ABI automated DNA sequencer (PE Applied Biosystems, Foster City, CA) with the PRISM Ready Reaction Dye Deoxy Terminator cycle sequencing kit (PerkinElmer, Norwalk, CT). The three clones for both TRs were shown to be identical by restriction digestion and partial sequencing. The sequence of each cDNA was compared with the sequence of the RT-PCR fragments to ensure that we obtained full-length clones for both axolotl TR genes. Each sequence was subsequently compared with reference sequences using the FASTA program.

The accession numbers of the axolotl TRs are as follows: TR{alpha},: AY174871; TRß, AY174872; TR{alpha} small isoform, AY280713; and TRß small isoform, AY280714.

In vitro translation and T3 binding assays
The TRs and RXR, cloned in pSG5 (Stratagene, La Jolla, CA), were transcribed and translated using the TNT T7-coupled reticulocyte lysate system (Promega Corp., Madison, WI) according to the manufacturer’s instructions in the presence or absence of [35S]methionine (1000 Ci/mmol; Amersham Pharmacia Biotech, Arlington Heights, IL). The 35S-labeled TRs were analyzed by SDS-PAGE in 8–18% gradient gels, followed by autoradiography. The level of protein-incorporated [35S]methionine was measured using the trichloroacetic acid precipitation method according to the recommended protocol (Promega Corp.). An estimate neosynthesized [35S]TR was calculated from [35S]Met incorporation taking into account the presence of methionine within the rabbit reticulocyte lysate (5 µM[SCAP];, Promega’s technical recommendation). The T3 binding affinity of the receptors was determined comparatively to that of human TRß1 in saturation experiments using [125I]T3 (3 µCi/µg; Amersham Pharmacia Biotech). In vitro translated receptors (1.5–3 µl) were incubated for 3 h at 20 C in 200 µl binding buffer [20 mM Tris-HCl, 1 mM MgCl2, 2 mM EDTA, 100 mM KCl, 5 mM dithiothreitol, and 10% (vol/vol) glycerol, pH 7.95] containing 0.01–0.5 nM [125I]T3. Nonspecifically bound T3 was always determined in parallel incubation with 0.1 µM unlabeled T3 and subtracted from total bound T3. Scatchard analyses of the binding data were performed as previously described (27).

Electrophoretic mobility shift assays
Full-length axolotl TR{alpha} and TRß as well as Xenopus TRs and Xenopus RXR{alpha} proteins were translated in vitro using the TNT kit (Promega Corp.). A DR4 probe (xDR4) from the Xenopus TRß promoter encompassing the transcription start site (positions -12 to 18) (27) was labeled with T4 polynucleotide kinase in the presence of [{gamma}-32P]ATP. Binding reactions were performed for 10 min using the radiolabeled DNA probe (1 ng, 40,000 cpm) and wheat-germ lysate in 10% glycerol, 10 mM HEPES, 30 mM KCl, 4 mM spermidine, 0.1 mM EDTA, 0.25 mM dithiothreitol, 1 mM Na2PO4, and poly(dI-dC) (1.5 µg). Unlabeled competitor oligonucleotides were included at the indicated molar excess in the binding reactions. Samples were loaded on a 5% nondenaturing polyacrylamide gel and electrophoresed for 2 h at 180 V.

Transient transfection assays
The axolotl TR{alpha} and TRß cDNAs were subcloned into BamHI/XhoI sites from a modified pSG5 vector. Ros 17/2.8 (rat osteosarcoma) or HepG2 (human hepatoma) cells were plated in DMEM (Bio Media, Paris, France) supplemented with 10% fetal calf serum at a density of 105 cells/dish. Transient transfections were performed using ExGen (Euromedex, Souffelweiersheim, France) with DNA mixture containing 1 µg total DNA, including 0.5 µg luciferase reporter construct (DR4-pGL2, Promega), 30 ng normalized vector, 0.1 µg receptor expression plasmid, and 0.37 µg pBSKSII (Stratagene, La Jolla, VA). Cytomegalovirus ß-galactosidase (Stratagene), containing the bacterial lacZ gene expressed from the simian virus 40 early promoter, was also introduced into the cells and used as an internal control to account for transfection efficiency. Cells were washed after 6 h, and fresh medium was added together with 10-12–10-6 M T3. Cells were lysed 48 h after transfection and assayed for luciferase and ß-galactosidase activities. Luciferase activity was divided by ß-galactosidase activity to normalize for transfection efficiency. Transfections were performed in triplicate.

RT-PCR isolation in axolotl of homologs of Xenopus TR target genes
Given the nearly ubiquitous expression pattern of the TRs in vertebrates (28, 29), we used total RNA extracted from various axolotl adult tissues (muscle, gills, liver, etc.) and from TH-treated as well as untreated larvae. RNA was extracted from frozen tissues using TRIzol reagent (Life Technologies, Inc., Gaithersburg, MD). The RNA samples were treated with RQ1 deoxyribonuclease [Promega; 2 U for 5–50 µg RNA in 10 mM Tris-HCl (pH 7.5) and 10 mM MgCl2, 37 C, 30 min], extracted using phenol/chloroform/isoamylic alcohol (25:24:1) and chloroform/isoamylic alcohol (24:1), and finally precipitated with ethanol.

One to 5 µg total RNA were reverse transcribed using random primers or specific primers and Moloney murine leukemia virus reverse transcriptase in 20 µl reaction mixture according to the manufacturer’s instructions (Life Technologies, Inc., Moloney murine leukemia virus reverse transcriptase kit). The resulting cDNA was amplified by PCR in a 50-µl volume with 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2 (PerkinElmer), 0.25 mM of each deoxy-XTP, 2.5 U Taq Gold DNA polymerase (PerkinElmer), and 300 ng of each primer.

Degenerate primers were designed using an alignment of published Xenopus nucleotide sequences and homologs from other vertebrate species according to the method described previously (30). The primers are described in Table 1Go. Most of these primers are degenerated and were used in a "touch-down" PCR assay (31) during which the hybridization temperature is reduced by 5 C every five cycles to allow hybridization of at least one of the primers of the degenerated mix. Denaturation and elongation temperatures remain unchanged. The complete PCR cycle is 94 C for 1 min, hybridization from 53–37 C for 1 min, and 72 C for 1 min for 40 cycles. To avoid contamination, all of the RT-PCR reactions as well as RNA extraction were conducted in a special nucleic acid-free room under positive air pressure using specific reagents (32, 33). For each pair of primers used, we prepared a control reaction without DNA to ensure that the PCR reactions were not contaminated by any mammalian genes routinely used in our laboratory. PCR products were cloned into the PCR2.1-TOPO vector (Invitrogen, Carlsbad, CA). Each plasmid was sequenced with a 373A ABI automated DNA sequencer with the PRISM Ready Reaction Dye Deoxy Terminator cycle sequencing kit (PerkinElmer). Each amplified sequence was analyzed phylogenetically to verify that the sequence was from axolotl and did not result from a Xenopus contamination. The accession numbers of the axolotl genes are as follows: collagenase 3, AY139972; fibronectin, AY139973; and TIF2, AY139974


View this table:
[in this window]
[in a new window]
  		TABLE 1. Oligonucleotides used in this study

 
Semiquantitative RT-PCR assay
To study the expression of the various isoforms of axolotl TR{alpha} and TRß and of the homologs of TH target genes, we designed specific primers (Table 1Go). Other primers were used as a positive control for the PCR reactions. In parallel to each PCR, a PCR reaction was performed to amplify the 28S rRNA to normalize the quantity of cDNA used for each sample. RTs were carried out with a pool of RNA for each point using the same amount of RNA for each animal. We previously determined the number of cycles to use to remain in the linear range. We used a standard PCR protocol consisting of 30 cycles for TR{alpha}, TRß, and collagenase 3; 25 cycles for stromelysin 3, fibronectin, and TIF2; and 18 cycles for 28S (denaturation at 94 C for 45 sec; hybridization at 55 C for 45 sec; elongation at 72 C for 45 sec). PCR products were separated on agarose gels and stained with ethidium bromide. The gels were then transferred to a Hybond N nylon membrane (Amersham Pharmacia Biotech) overnight in a solution of 0.5 M NaOH, and 1.5 M NaCl. DNA was fixed on the membrane by UV exposure (0.360 J/cm2). Oligonucleotides specific for each PCR fragment were labeled by [{gamma}-32P]ATP using T4 polynucleotide kinase (Life Technologies, Inc.) and the conditions and buffers recommended by the enzyme supplier. The labeled oligonucleotides were purified on microspin G-50 columns (Amersham Pharmacia Biotech). After prehybridization (1 h at 37 C) with prehybridization buffer, hybridization was carried out overnight at 37 C with 500 ng labeled oligonucleotide. Subsequently, the membranes were rinsed several times for 10 min at 37 C and exposed with a phosphorimager film for about 12 h. The pixel quantification of each band was performed using IQMac version 1.2 software from Molecular Dynamics (Sunnyvale, CA).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Isolation of full-length cDNAs encoding axolotl TRs
A cDNA library derived from axolotl limb blastoma cells was screened with the previously described RT-PCR fragments corresponding to A. mexicanum TR{alpha} and TRß (26). For both the TR{alpha} and TRß probes, we isolated cDNAs that were subsequently sequenced. We found that the sequences of these cDNAs were 100% identical with the sequences of the RT-PCR fragments that were used as probes during the screening procedure, suggesting that we isolated bona fide axolotl TRs. The two cDNAs encoding full-size TR{alpha} and TRß proteins are, respectively, 1315 and 1428 bp long and contain in-frame start codons (at position 89 for the TR{alpha} cDNA and 307 for the TRß cDNA) and stop codons (at position 1313 for the TR{alpha} cDNA and 1426 for the TRß cDNA), resulting in proteins with lengths of 408 and 373 amino acids, respectively. The two cDNAs encode full-size TR{alpha} and TRß gene products because they both contain in-frame stop codons 5' to the first methionine.

The sequences of these clones were aligned with known TRs from vertebrates (Fig. 1Go). As expected, the axolotl TR sequences show strong sequence identities with known TR{alpha} and TRß proteins. Overall, the axolotl TR{alpha} exhibits 86.8% and 88.8% amino acid identities with human and Xenopus TR{alpha}, respectively, whereas axolotl TRß is 75.7% and 91.4% identical to the human and Xenopus TRß. The difference in amino acid identity between axolotl TRß and the human sequence is largely due to the divergence of the human TRß in the A/B region. When this region is excluded, identity levels are 93.4% with the human sequence and 91.5% with the Xenopus protein. When specific domains are considered, we observed very strong conservation in the C domain (93–100% between axolotl TR{alpha} and its vertebrate homologs, 93–98% for TRß) and in the E domain (88–90% between axolotl TR{alpha} and its vertebrate homologs, 92–93% for TRß). Note that in each clone both the P box, which directly contacts the DNA, and the D box, which is another important feature of the DBD, are highly conserved. Among the 3 {alpha}-helixes that were found in the 3-dimensional structure of the DBD, helix H1, which contains the P box and establishes direct base-specific contact with the hormone response element, is fully conserved. In contrast, helix H2 and H3 in the axolotl TRß contain some substitutions compared with either the human or Xenopus sequences. These domains are important for protein-protein interactions and, in the case of H3, for heterodimerization with RXR and nonspecific contacts with DNA. In the LBD of both TR{alpha} and TRß we also find very strong conservation in the 12 {alpha}-helixes that form the core of the 3-dimensional structure (34), suggesting that the TRs from axolotl have all the features to bind TH and to trans-activate target genes in response to T3. The helixes H2, H3, H5–6, H9, H11, and H12 are very highly conserved. Interestingly, all of the amino acids that are in direct contact with the ligand in the 3-dimensional structure of the rat TR{alpha} LBD (namely N228, R277, R311, R315, and S326, which is an N for TRß receptors; see Fig. 1Go) are fully conserved in the axolotl TRs, as are the regions containing the autonomous trans-activation domain AF2-AD. The A/B region is very poorly conserved when mammalian proteins are compared with those from axolotl, but a significant degree of sequence identity exists within this region between the amphibian TR{alpha} and TRß proteins. In both Xenopus and axolotl, the A/B region of the TRß protein is very short, suggesting that we have isolated a specific isoform of axolotl TRß that probably corresponds to the Xenopus TRß isoform, the expression of which is directly regulated by T3 (35). Although the D domain contains more substitutions than the other domains, it is quite conserved between TR{alpha} and TRß.



View larger version (41K):
[in this window]
[in a new window]
  		FIG. 1. Amino acid sequence alignment of TR{alpha} and TRß from human, chicken, Xenopus, and axolotl. The various structural domains and their boundaries are indicated. For the LBD (also called the E domain) the 12 helixes found in the three-dimensional structure of the mammalian TRs are indicated. The accession numbers are as follow: human TR{alpha}, AAA66021; chicken TR{alpha}, CAA68792; Xenopus TR{alpha}, AAA49970; axolotl TR{alpha}, AY174871; human TRß, CAA28412; chicken TRß, CAA35544; Xenopus TRß, AAA49657; and axolotl TRß, AY174872.

 
Functional characterization of the axolotl TRs
To test the abilities of these cDNAs to encode functional TRs, we carried out a series of in vitro assays. First, we used electromobility shift assays to test whether the axolotl TRs synthesized in vitro bind to DNA. We used the DR4 element from Xenopus (xDR4) as the labeled probe and a number of other TREs as competitors. The xDR4 was isolated from the promoter of the Xenopus TRß gene and has been characterized as the TRE mediating TRß autoinduction during Xenopus metamorphosis (reviewed in Ref.3) (Fig. 2AGo). As shown in Fig. 2BGo, neither axolotl TR{alpha} nor TRß can bind to this element alone (lanes 2 and 11), but they form a strong heterodimer, which binds to the xDR4 probe in presence of the Xenopus RXR{alpha} protein (lanes 3 and 12). To ensure the specificity of the binding of the TR-RXR heterodimers to the xDR4 probe, we performed competition experiments with four TREs: the xDR4 element itself, a consensus DR4 sequence, a palindromic element (HREpal) (15) and an unrelated (NS) oligonucleotide (Fig. 2AGo). We also used more complex elements, such as the one upstream the myelin basic protein gene (data not shown). For both TR{alpha} and TRß, the binding can be competed out by the xDR4 sequence (Fig. 2BGo, lanes 4, 5 and 13, 14), but not by an unrelated oligonucleotide (lanes 9 and 19). These competition experiments also reveal that axolotl TR{alpha} and TRß proteins can bind to a perfect consensus DR4 element (lanes 6, 7 and 15, 16) as well as the HREpal (lanes 8, 9 and 17, 18). All of these results are consistent with observations made with Xenopus TR{alpha} and TRß proteins in similar experiments (data not shown). Taken together, these data suggest that the DNA-binding characteristics of axolotl and Xenopus TRs are very similar, if not identical.



View larger version (47K):
[in this window]
[in a new window]
  		FIG. 2. A, Sequences of the probes used. Consensus half-site are underlined. B, EMSA experiments. The labeled probe is the DR4 of the Xenopus TRß promoter (xDR4) in competition with the others at the indicated molar excess. The experiments were performed with in vitro-synthesized TR{alpha} (lanes 1–10) or TRß (lanes 11–19). For each lane, in vitro-synthesized Xenopus RXR{alpha} was added to allow dimerization with axolotl TR, and the arrow shows the heterodimer RXR/TR binding the X-DR4-labeled probe. NS, Nonspecific probe. This experiment was performed three times with identical results.

 
We next examined the trans-activating properties of the axolotl TRs using transient transfection assays. These assays were performed in osteosarcoma Ros 17.2/8 cells, because these cells contain a high level of RXR, which is sufficient to provide a strong heterodimerization partner for transfected TRs. In addition, these cells contain very low levels of TR that avoid interference between endogenous and transfected receptors (36). The results show that axolotl TR{alpha} and TRß proteins are both able to activate transcription of the xDR4-luciferase reporter construct in a T3-dependent manner (Fig. 3AGo). In addition, we performed dose-response experiments by transient transfection of HepG2 cells with the TR expression vectors and the xDR4 reporter construct. This experiment clearly shows that the transcriptional activity of the receptors can be detected at 0.1 nM and is maximal at 1 µM. All of the receptors activate transcription within the same range, except axolotl TRß, which seems to be more potent than the others. This difference should nevertheless be interpreted with caution (see Discussion).



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 3. A, Transcriptional activity of axolotl TR in response to T3. The data are presented as luciferase activity for axolotl TR{alpha} and TRß in response to 10-7 M T3. The pSG5, pSG5-TR{alpha}, and pSG5-TRß were transfected together with xDR4 LUC reporter construct in Ros 17.2/8 cells, and T3 was added to the cell culture. B, Dose-response experiment performed by transient transfection of HepG2 cells. The luciferase activity was plotted against the log of the T3 concentration. The activity of the axolotl TRs was compared with that of the Xenopus receptors as well as to human TRß activity. The transfections were performed in triplicate, and the whole set of transfections was performed twice with identical results.

 
All of these data suggest that the axolotl TRs are, like their Xenopus homologs, able to bind to T3. To directly test this hypothesis, we measured the binding affinity of the in vitro translated receptors to labeled T3. Scatchard plots show that the binding affinity of the axolotl TRs is comparable to the affinity of Xenopus TRs to T3 (Kd values for axolotl TR{alpha}, 60.6 pM; for axolotl TRß, 15.1 pM; Fig. 4Go). All of these data suggest that axolotl TRs are functional because they can bind to DNA, heterodimerize with RXR, bind to T3, and activate transcription in mammalian cells in response to this hormone.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 4. T3 binding of axolotl TRs compared with their Xenopus homologs. Each Scatchard plot of a representative experiment is presented with the data obtained using human TRß1 as a control. In each case the Ka value is compared with that for human TRß1. {circ} and {bullet}, Amphibian and human data, respectively. B, Bound T3; B/F, bound to free T3 ratio. Each experiment was performed at least four times.

 
Presence of different isoforms of TR{alpha} and TRß
We studied the expression of TR{alpha} and TRß genes by RT-PCR with 5' and 3' primers specific for the C and E domains, respectively. We noticed, in addition to the expected bands, shorter fragments for both TR{alpha} and TRß. These bands were cloned and sequenced. For TR{alpha}, we found that the shorter band corresponds to a deletion of 147 nucleotides (data not shown). Strikingly, the boundaries of this deletion, which we call TR{Delta}{alpha}, corresponds to the boundaries of the human TR{alpha} exon 7, suggesting that this exon may be left out in some mRNA transcripts (19) (see Fig. 5Go). In this case, the resulting protein contains an in-frame deletion of 49 amino acids that corresponds to the region of helixes H3 and H4 and is unlikely to bind to T3 because this region encompasses some of the residues that are in direct contact with the ligand. For TRß, the cloned, shorter RT-PCR fragment includes a deletion of 309 nucleotides, the boundaries of which coincide with the human TRß exon 6. As for TR{alpha}, this strongly suggests that the shorter fragment corresponds to an alternatively spliced isoform of the axolotl TRß, which we call TR{Delta}ß (Fig. 5Go). Because the deletion of this exon in axolotl TRß changes the reading frame of the resulting transcript, and a stop codon is present two amino acids downstream from the alternative spliced boundary, this alternatively spliced variant generates a truncated protein. This 90-amino acid-long product will contain only the A/B and C regions, and no D and E domains. This product will thus be unable to bind T3 and is unlikely to bind to DNA because it lacks the helix H3 of the DBD, which is important for nonspecific interaction with the sugar-phosphate backbone of the DNA molecule.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 5. A, Structure of the axolotl TR{Delta}{alpha} isoform. The deletion of 49 amino acids removes part of the LBD. B, Structure of the axolotl TR{Delta}ß isoforms. The deletion of 309 nucleotides results in a change in the reading frame of the transcript. Thus, this isoform does not contain a LBD. For A and B, the primers used to test the expressions of the different isoforms are indicated by arrows. C, Genomic organization of the human TR{alpha} and TRß genes. The axolotl TR{Delta}{alpha} isoform results from a deletion of an exon homologous to exon 7 in the human TR{alpha} gene. The axolotl TR{Delta}ß isoform corresponds to the deletion of an exon homologous to exon 6 of the human TRß gene. The functional domains are indicated above the genomic structure. Exons are represented in proportion to their size in base pairs, whereas intron lengths are arbitrary. D, Detection of the TR{Delta} isoforms in axolotl tissues. RT-PCR amplification of all the TR isoforms with cDNA coming from whole axolotl larvae. WM, Molecular weight marker. The sizes of the expected PCR products are indicated. The experiments were repeated twice with similar results.

 
To check that these isoforms do not represent RT-PCR artifacts, RNA from whole axolotl larvae were analyzed by RT-PCR with primers specific for each isoform (Fig. 5Go, A and D, and Table 1Go). TR{alpha}- and TRß-specific primers were used as positive controls. Interestingly, we observed a band at the expected size for full-size TR{alpha} and TRß transcripts. We also found strong expression of the TR{Delta}{alpha} isoform, whereas TR{Delta}ß was expressed at a much lower level. These results suggest that both isoforms are expressed and should correspond to bona fide transcripts, which may play a role in regulating the T3 response during axolotl postembryonic development.

We then tested whether the expression of the bona fide TR transcripts as well as the transcripts encoding the isoforms was regulated by T3 using a semiquantitative RT-PCR assay. Surprisingly, the results (data not shown) reveal no significant variation in TR{alpha}, TRß, TR{Delta}{alpha}, or TR{Delta}ß expression under T3 treatment in the head or the tail of the axolotl.

TH response in axolotl
As axolotl tissues expressed functional TR mRNAs, we next studied the influence of T3 on the regulation of other genes that are known to be important in Xenopus metamorphosis. In Xenopus, it has been proposed that TH treatment induces the expression of early response genes, such as TRß itself, as well as transcription factors, such as BTEB or TH-bZIP, and matrix metalloproteases, such as stromelysin 3 (37, 38, 39). It has also been proposed that TRß activates the transcription of late genes, such as collagenase 3 and fibronectin. Using RT-PCR and degenerate primers, we thus cloned axolotl homologs of matrix metalloproteases (stromelysin 3 and collagenase 3) and other matrix-associated genes, such as fibronectin. In addition, we isolated an axolotl homolog of the transcriptional coactivator TIF2, which plays an important role in mediating TR transcriptional activity in mammals (see Ref.19 for references). The RT-PCR fragments were cloned and sequenced, and their authenticity was assessed by careful phylogenetic analyses. We found that the position of each of the sequences in the tree is consistent with the phylogenetic position of its respective species of origin (data not shown). Using these sequences we designed specific primers (Table 1Go) that are able to amplify the axolotl sequence in semiquantitative RT-PCR experiments.

Using samples of head and tail tissues of T3-treated animals, we observed that the expression of both stromelysin 3 and collagenase 3 is induced by T3 treatment in the tail and the head (Fig. 6Go). Interestingly, the time course of activation of these two genes varies according to the region of the animal used. In the head, stromelysin3 is rapidly induced 2 d after T3 treatment, whereas collagenase 3 induction requires a longer exposure (strong activation after 10 d). In contrast, in the tail, both genes are rapidly induced. Fibronectin and TIF2 expression were not affected by T3 treatment (not shown). All of these data suggest a complex response mechanism induced by T3 treatment in axolotl larvae. This concept is consistent with our functional results showing that axolotl TRs are able to activate transcription in response to T3 in transient transfection assays.



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 6. Induction of target gene expression after T3 treatment. A, Stromelysin 3 mRNA in the head. B, Collagenase 3 mRNA in the head. C, Stromelysin 3 mRNA in the tail. D, Expression of collagenase 3 mRNA in the tail. The histograms represent the average radioactivity quantified for each gene normalized with the 28S rRNA for each time of treatment as fold expression compared with the untreated animals. For each time point, RT-PCR was performed with a mixture of RNA from different animals (same amount for each animal). RT-PCR was performed twice independently. In the head, as in the tail, stromelysin 3 and collagenase 3 expression is induced by T3.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Axolotl TRs are functional
Our analysis clearly shows, in accordance with previous studies (14), that TRs are present, functional, and expressed in axolotl tissues. It was important to characterize in more detail axolotl TRs to know whether their ability to respond to TH is comparable to that of Xenopus TRs. At the level of sequence, the extent of conservation between Xenopus and axolotl TR proteins is impressive. The DNA-binding C domains of Xenopus and axolotl TR{alpha} are 100% identical, whereas for TRß only three substitutions are obvious in helix 2, which plays no direct role in DNA binding or heterodimerization. Helix 3, which contains the T and A boxes important for DNA binding affinity and response element selection, is also well conserved. The few sequence differences in this helix between axolotl and Xenopus TRs are conservative substitutions (e.g. Q to E at position 192, or K to R at position 196), which do not alter the potential to form an helix. In the LBD, several lines of evidence also suggest that this domain is functional: 1) all of the amino acids that are in direct contact with the ligand in the three-dimensional structure are conserved; 2) none of the mutations that are known to occur in the human TRß gene in TH resistance syndromes can be found in the axolotl TRs (40); 3) all regions known to be important for the various functions of these receptors are well conserved (helix 9 for dimerization, helixes 11 and 12 for trans-activation function, and helixes 3 and 4 for coactivator binding) (see Ref.19 for references); and 4) all of the substitutions observed in the axolotl genes are in loops and not in helixes, suggesting that substitutions accumulate in nonessential regions. Thus we found no substitutions suggesting any important functional differences between Xenopus and axolotl TRs.

The functional characterization of the axolotl TRs clearly shows that they bind DNA and heterodimerize with Xenopus RXR in a manner indistinguishable from the Xenopus TR orthologs. We found that TR{alpha} and TRß are able to bind to a natural response element, namely the DR4 element found upstream of the Xenopus TRß gene. In addition, using competition experiments, we found that axolotl TRs can recognize palindromic elements, suggesting that in axolotl, as in Xenopus, TRs can bind to a wide variety of response elements. The trans-activation assays that we performed in mammalian cells also suggest that the axolotl TRs are functional because we observed a very strong rise in transcriptional activity of the reporter gene in response to TH. The conditions we used resulted in very low activity of the reporter gene in the absence of exogenously transfected TRs and did not allow us to test whether the axolotl TR exhibits repressive activity in the absence of TH, as has been reported for TRs from other species. The activation seen with the axolotl TRs in dose-response experiments is comparable to results obtained with the respective Xenopus receptors, except for TRß, which appears to be a little more efficient in axolotl than in Xenopus. However, these data should be interpreted with caution. Indeed, the trans-activation experiments were performed in a heterologous cell system containing mammalian coactivators. The detailed comparison of the trans-activation potentials of Xenopus and axolotl TRs should be made in Xenopus and axolotl cells to avoid possible differences in affinity between the various coactivators and the TRs. Study of the direct binding of TH to the receptor suggests that, at least at the level of hormone-receptor interaction, Xenopus and axolotl TRs are extremely similar, as the affinities of the receptors for the hormone are very similar in both species.

All of these observations fit very well with the finding that TH induces metamorphosis in axolotl (Ref.41 and references therein). The study of several target genes, such as TRß, collagenase 3, stromelysin 3, and fibronectin, revealed that a clear TH response can be observed in axolotl, as observed in Xenopus (see below).

New isoforms of TR{alpha} and TRß
We identified isoforms for both axolotl TR{alpha} and TRß. For TR{alpha}, we found an in-frame deletion of the exon homologous to exon 7 of the human TR{alpha} gene, which results in a deletion of 49 amino acids localized in the LBD. Surprisingly, we isolated the same isoform in the newt Pleurodeles waltii (Safi, R., and V. Laudet, unpublished observations) as well as in a teleost fish, the tilapia (Oreochromis niloticus) (42). We thus think that this isoform corresponds to an alternatively spliced variant with a large deletion inside the LBD. This deletion corresponds to the region of helixes H3 and H4, which are important for the ligand-binding pocket as well as for the interaction with coactivators. We have also functionally characterized this isoform and found that it exhibits a very weak DNA-binding ability on the XDR4 element (data not shown). We interpret this result as an inability of this protein to interact with RXR because the large deletion in the middle of the LBD disrupts the RXR heterodimerization interface that is present in the LBD, mainly in helix H9. Consistent with this idea, this isoform is transcriptionally silent and does not exhibit a clear dominant negative activity when tested in transient transfection assays (Safi, R., unpublished observations). As it is expressed in vivo and found in various unrelated animal species (amphibians and teleost fishes), we propose that it has a biological function that remains to be defined. For TRß, we found another alternative splice variant resulting in a deletion of the homolog of exon 6, which generates a 90-amino acid product that is unable to bind both DNA and TH and is transcriptionally inactive per se and in a dominant negative activity assay (data not shown). This isoform is expressed in vivo at lower levels than the TR{Delta}{alpha} isoform, and its expression is not changed during TH- induced metamorphosis.

Nearly all known TR genes generate isoforms that are unable to bind T3 and are transcriptionally silent. This is the case for the TR{alpha}2 isoform in mammals, which is a C-terminal, alternatively spliced variant of the bona fide TR{alpha}1 and has a dominant negative activity in transient transfection assays (43, 44, 45). In addition, the human and mouse TR{alpha} genes encode isoforms generated from an internal promoter located in exon 7 that represent a truncated version of the LBD (46). Careful genetic dissection of the functions of the various isoforms encoded by the TR{alpha} locus have revealed that these short isoforms play an important role in buffering the actions of both TR{alpha} and TRß (29, 47, 48). This is also the case in chickens, where several transcriptionally inactive, N terminally truncated, TR proteins are generated by alternative in-frame ATG usage (49). The physiological role of these TH-independent isoforms is still unclear. Perhaps it is physiologically relevant for animals to be able to generate different TR proteins, which are unable to bind T3 and/or to activate target genes at certain times during embryogenesis or postembryonic development.

TH response in axolotl
In Xenopus, numerous TH-dependent genes were isolated from the resorbing tail using a PCR-based subtractive hybridization procedure (50). Half of the 17 up-regulated genes were shown to respond rapidly to and be directly regulated by the hormone, because their induction is not abrogated by cycloheximide treatment (51). Four of these early response genes are transcription factors (TRß, xBTEB, TH/bZIP, and FRA-2), which can, in turn, induce a cascade of TH-dependent gene transcription (39, 52). Among the other early-induced genes are the enzyme type III deiodinase and a matrix metalloprotease, stromelysin 3 (52). Another class of genes, the delayed response genes, is induced about 24 h after TH treatment. These delayed response genes include extracellular matrix components, such as fibronectin and integrin {alpha}1, and other proteolytic enzymes, such as FAP{alpha}, peptidase E, or collagenase 3 (52). Study of the expression of these genes during metamorphosis has led to a model in which a unique hormone (TH) induces distinct, and even opposite, processes through independent regulatory programs (Fig. 7Go) (37, 38). It has been proposed that when TH surges at the beginning of metamorphosis, it binds to TR{alpha}, which is expressed early in postembryonic development and is ubiquitously expressed in tadpoles. Liganded TR{alpha} probably induces early genes such as TRß, which subsequently up-regulates its own transcription by an autoregulatory loop and thus mediates the response to TH. TRß as well as the other transcription factors subsequently induces different cascades of gene activation. By studying the in situ localization of the transcripts of all of these genes in the tail as well as the head of tadpoles Berry et al. (37, 38) suggested that the early response transcription factors xBTEB, TH/bZIP, and FRA-2 induce the growth of adult tissues such as cartilage, nervous tissues, adult epidermis, and adult muscles (Fig. 7Go). In contrast, TRß, by regulating the delayed response genes such as collagenase 3 and fibronectin, is implicated in the resorption of larval tissues such as cartilage, muscle, and epidermal lamellae in the tail. Stromelysin 3, which is also an early response gene, is implicated in the resorption of larval structures in the tail and other tissues.



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 7. Hypothesis for a gene expression regulation cascade during X. laevis metamorphosis. TH interacts with the receptor TR{alpha}, which activates the expression of the early genes, among them TRß. Subsequently, the hormone interacts with TRß, activating the expression of the delayed genes that are responsible for the resorption of larval tissues. The transcription factors, except TRß, among the early induced genes play an important role in the growth of adult tissues. BTEB, Basic transcription element-binding protein; FAP, fibroblast-activated protein; Fra-2, Fos-related activator 2; TH/bZip, TH-activated protein with bZip domain. The collagenase 3 and stromelysin 3 genes that we found regulated by TH in axolotl in this study are italicized.

 
We have isolated axolotl homologs of some of the genes implicated in the TH cascade in Xenopus, such as TRs, collagenase 3, stromelysin 3, and fibronectin. Despite extensive efforts, we failed to isolate homologs of the early transcription factor genes such as BTEB, FRA-2, or TH/bZIP. We found that only two of the axolotl genes isolated, collagenase 3 and stromelysin 3, are regulated by TH, even if in some cases (collagenase 3 in the head) this induction requires a relatively long delay. At present, the reason and significance of these differences between axolotl and Xenopus remain unclear, and more studies are needed to compare the genetic programs induced by TH in the two species. It is striking that we did not find any obvious induction of TR genes after TH treatment in either the tail or the head of axolotl. This result differs from those obtained by other researchers (53) and may be due to age- or tissue-specific factors, as gills were used by Yaoita and Brown (53A ) to measure TR induction. A definitive answer to this question requires a more comprehensive analysis of TR expression in vivo in axolotl.

Interestingly, in the course of this study we observed in axolotl a strong TH-dependent up-regulation of collagenase 3. In Xenopus, collagenase 3 is expressed in fibroblasts, which either line or surround the notochord or are located beneath the epidermal lamellae (11, 12). During tail resorption, these fibroblasts invade the neighboring tissues and are crucial in the tail resorption process. It is striking to find induction of collagenase 3 in axolotls, because the tail of the axolotl, like that in other urodeles, does not regress during metamorphosis. These data suggest that the axolotl tail is TH responsive, and that collagenase 3 induction itself is not sufficient to induce a tail resorption program. Alternatively, it is possible that in axolotl, a block of collagenase 3 action occurs downstream of their activation by TH. In contrast, the fact that we found TH induction of stromelysin 3 is consistent with previous observations in Xenopus, where stromelysin 3 is induced in the dorsal and ventral fins. In axolotl, the first morphological event after TH treatment is the resorption of the ventral and dorsal fins. Thus, it is tempting to predict that stromelysin 3 is induced in the fins of the axolotl. It will be interesting to determine in which cell types the induction of these various genes occurs and if other matrix metalloproteinases, such as gelatinase A or MT3-MMP, that have been described as being involved in tail resorption are also regulated (53).

Axolotl: a promising model for postembryonic development
Considerable interest is directed to understanding evolution of developmental processes and their link to morphological or life history traits. Axolotl, which presents two interesting and opposite events in metamorphosis (a suppression of metamorphosis through the process of neoteny and its induction by thyroid hormone), is a promising model for the comparative study of postembryonic development at the molecular level in vertebrates (8, 54). Genetic analysis suggests that metamorphic failure in laboratory axolotl is associated with a major locus that is not TR{alpha} or TRß (55). Interestingly, a more recent paper suggests that TR{alpha} may contribute to variations in metamorphic timing in wild-caught axolotls, suggesting the possibility of a minor effect of TRs on metamorphic failure (55). In addition, the genetic data indicate that the mechanistic basis of metamorphic failure is different between wild-caught and laboratory axolotl, suggesting that it may have more than one mechanistic basis (56, 57). Given this complex pattern for metamorphic failure for Ambystoma, it is likely that several factors have to be taken into account to fully explain the common occurrence of neoteny in these animals. In addition, the axolotl is a promising model to study how the action of a hormone, namely TH, can evolve in different species. For example, the isolation of TH-regulated genes from axolotl tails using a subtractive screening procedure and analysis of the regulation of these genes compared with the regulation of their homologs in Xenopus would allow insight into why the urodeles tail is refractory to the action of TH. Our results, although still preliminary, suggest that the axolotl tail can respond to TH, but that the main differences occur at the level of target genes.


   Acknowledgments
 
We thank Hector Escriva, Michael Schubert, and Randal Voss for critical reading of the manuscript; Julien Fellah, Véronique Laurens, and Yannick Andreol for providing animals; Jean-Paul Chaumel for taking care of the animals; and Carl Seguin for the generous gift of the axolotl cDNA library.


   Footnotes
 
This work was supported by Ministere de l’Education Nationale de la Recherche et de la Technologie, Centre National de la Recherche Scientifique, Association pour la Recherche Contre le Cancer, Fondation pour la Recherche Médicale, La Ligue Regionale Contre le Cancer du Rhône et de la Loire and HRA-PHARMA.

R.S. is the recipient of a fellowship from the Ligue Nationale contre le Cancer and the Fondation pour la Recherche Médicale.

O.M. is the recipient of a fellowship from Ministere de l’Education Nationale de la Recherche et de la Technologie and ARC.

Abbreviations: DBD, DNA-binding domain; HRE, hormone response element; LBD, ligand-binding domain; RXR, retinoid X receptor; TH, thyroid hormone; TR, thyroid hormone receptor; TRE, thyroid hormone response element.

Received July 21, 2003.

Accepted for publication October 1, 2003.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Leloup J, Buscaglia M 1977 La triiodothyronine hormone de la métamorphose des amphibiens. C R Acad Sci Ser D 284:2261–2263
  2. Tata JR 1997 Hormonal signalling and amphibian metamorphosis. Adv Dev Biol 5:237–274
  3. Shi YB 1999 Amphibian metamorphosis: from morphology to molecular biology. New York: Wiley & Sons
  4. Gudernacht JF 1912 Feeding experiments on tadpoles. I. The influence of specific organs given as food on growth and differentiation. A contribution to the knowledge of organs with internal secretion. Wilhelm Rou Arch Entwicklungsmech Org 35:457–483
  5. Hanken J 1999 Larvae in amphibian development and evolution. In: Hall BK, Wake MH, eds. The origin and evolution of larval forms. New York: Academic Press; 61–108
  6. Rose CS 1999 Larvae in amphibian development and evolution. In: Hall BK, Wake MH, eds. The origin and evolution of larval forms. New York: Academic Press; 167–217
  7. Rosenkilde P, Ussing AP 1996 What mechanisms control neoteny and regulate induced metamorphosis in urodeles? Int J Dev Biol 40:665–673[Medline]
  8. Shaffer HB 1993 Phylogenetics of model organisms: the laboratory axolotl Ambystoma mexicanum. Systematic Biology 42:508–552[CrossRef]
  9. Shaffer HB 1996 The polytypic species revisited: genetic differentiation and molecular phylogenetics of the tiger salamander Ambystoma tigrinum (Amphibia: Caudata) comple. Evolution 50:417–433
  10. Voss SR 1995 Genetic basis of paedomorphosis in the axolotl Ambystoma mexicanum: a test of single-gene hypothesis. J Hered 86:441–447[Abstract/Free Full Text]
  11. Taurog A 1974 Effect of TSH and long-acting thyroid stimulator on thyroid 131I-metabolism and metamorphosis of the Mexican axolotl (Ambystoma mexicanum). Gen Comp Endocrinol 24:257–266[CrossRef][Medline]
  12. Taurog A, Oliver C, Eskay RL, Porter JC, McKenzie JM 1974 The role of TRH in the neoteny of the Mexican axolotl (Ambystoma mexicanum). Gen Comp Endocrinol 1974 24:267–279[CrossRef][Medline]
  13. Darras VM, Kuhn ER 1983 Effects of TRH bovine TSH and pituitary extracts on thyroidal T4 release in Ambystoma mexicanum. Gen Comp Endocrinol 51:286–291[CrossRef][Medline]
  14. Galton VA 1992 Thyroid hormone receptors and iodothyronine deiodinases in the developing Mexican axolotl Ambystoma mexicanum. Gen Comp Endocrinol 85:62–70[CrossRef][Medline]
  15. Yaoita Y, Shi YB, Brown DD 1990 Xenopus laevis alpha and beta thyroid hormone receptors. Proc Natl Acad Sci USA 87:7090–7094[Abstract/Free Full Text]
  16. Nuclear Receptor Nomenclature Committee 1999 A unified nomenclature system for the nuclear receptor superfamily. Cell 97:161–163[CrossRef][Medline]
  17. Sap JA, Munoz K, Damm Y, Goldberg J, Ghysdael A, Leutz H, Beug Vennstrom B 1986 The c-erb-A protein is a high-affinity receptor for thyroid hormone. Nature 324:635–640[CrossRef][Medline]
  18. Weinberger C, Thompson CC, Ong ES, Lebo R, Gruo DJ, Evans RM 1986 The c-erb-A gene encodes a thyroid hormone receptor. Nature 324:641–646[CrossRef][Medline]
  19. Laudet V, Gronemeyer H 2002 The nuclear receptors facts book. New York: Academic Press
  20. Umesono K, Murakami KK, Thompson CC, Evans RM 1991 Direct repeats as selective response elements for the thyroid hormone retinoic acid and vitamin D3 receptors. Cell 65:1255–1266[CrossRef][Medline]
  21. Glass CK, Holloway JM, Devary OV, Rosenfeld MG 1988 The thyroid hormone receptor binds with opposite transcriptional effects to a common sequence motif in thyroid hormone and estrogen response elements. Cell 54:313–323[CrossRef][Medline]
  22. Desvergne B 1994 How do thyroid hormone receptors bind to structurally diverse response elements? Mol Cell Endocrinol 100:125–131[CrossRef][Medline]
  23. Horlein AJ, Naar AM, Heinzel T, Torchia J, Glos B, Kurokawa R, Ryan A, Kamei Y, Soderstrom M, Glass CK, Rosenfeld MG 1995 Ligand-independent expression by the thyroid hormone receptor mediated by a nuclear receptor co-repressor. Nature 377:397–404[CrossRef][Medline]
  24. Chen JD, Evans RM 1995 A transcriptional co-repressor that interacts with nuclear hormone receptors. Nature 377:454–457[CrossRef][Medline]
  25. McKenna NJ, Lanz RB, O’Malley BW 1999 Nuclear receptor coregulators: cellular and molecular biology. Endocr Rev 20:321–44[Abstract/Free Full Text]
  26. Safi R, Begue A, Hänni C, Stéhelin D, Tata JR, Laudet V 1997 Thyroid hormone receptor genes of neotenic amphibians. J Mol Evol 44:595–604[CrossRef][Medline]
  27. Anselmet A, Gharbi-Chihi J, Torresani J 1984 Nuclear triiodothyronine receptor in differentiating preadipocytes cloned from obese and lean mice. Endocrinology 114:450–456[Abstract]
  28. Forrest D, Sjoberg M, Vennstrom B 1990 Contrasting developmental and tissue-specific expression of {alpha} and ß thyroid hormone receptor genes. EMBO J 9:1519–1528[Medline]
  29. Fraichard A, Chassande O, Plateroti M, Rou JP, Trouillas J, Dehay C, Legrand C, Gauthier K, Kedinger M, Malaval L, Rousset B, Samarut J 1997 The T3R {alpha} gene encoding a thyroid hormone receptor is essential for post-natal development and thyroid hormone production. EMBO J 16:4412–4420[CrossRef][Medline]
  30. Escriva H, Laudet V 1999 Evolutionary biology of the nuclear receptor superfamily. In: Picard D, ed. Nuclear receptors: a practical approach. Oxford: Oxford University Press; 1–28
  31. Don RH, Cox PT, Wainwright BJ, Baker K, Mattick JS 1991 ‘Touchdown’ PCR to circumvent spurious priming during gene amplification. Nucleic Acids Res 19:4008[Free Full Text]
  32. Safi R, Deprez A, Laudet V 1997 Thyroid hormone receptors in perennibranchiate amphibians. Int J Dev Biol 1997 41:533–535[Medline]
  33. Hänni C, Laudet V, Stéhelin D, Taberlet P 1994 Tracking the origins of the cave bear (Ursus spelaeus) by mitochondrial DNA sequencing. Proc Natl Acad Sci USA 91:12336–12340[Abstract/Free Full Text]
  34. Wagner RL, Apriletti JW, McGrath ME, West BL, Bater JD, Fletterick RJ 1995 A structural role for hormone in the thyroid hormone receptor. Nature 378:690–697[CrossRef][Medline]
  35. Kanamori A, Brown DD 1992 The regulation of thyroid hormone receptor ß genes by thyroid hormone in Xenopus laevis. J Biol Chem 267:739–745[Abstract/Free Full Text]
  36. Vanacker JM, Bonnelye E, Delmarre C, Laudet V 1998 Activation of the thyroid hormone receptor alpha gene promoter by the orphan nuclear receptor ERR{alpha}. Oncogene 17:2429–35[CrossRef][Medline]
  37. Berry DL, Schwartzman RA, Brown DD 1998 The expression pattern of thyroid hormone response genes in the tadpole tail identifies multiple resorption programs. Dev Biol 1998 203:12–23[CrossRef][Medline]
  38. Berry DL, Rose CS, Remo BF, Brown DD 1998 The expression pattern of thyroid hormone response genes in remodeling tadpole tissues defines distinct growth and resorption gene expression programs. Dev Biol 203:24–35[CrossRef][Medline]
  39. Hoopfer ED, Huang L, Denver RJ 2002 Basic transcription element binding protein is a thyroid hormone-regulated transcription factor expressed during metamorphosis in Xenopus laevis. Dev Growth Differ 44:365–381[CrossRef][Medline]
  40. Weiss RE, Refetoff S 2000 Resistance to thyroid hormone. Rev Endocr Metab Disord 1:97–108[CrossRef][Medline]
  41. Brown DD 1997 The role of thyroid hormone in zebrafish and axolotl development. Proc Natl Acad Sci USA 94:13011–13016[Abstract/Free Full Text]
  42. Marchand O, Safi R, Escriva H, Van Rompaey E, Prunet P, Laudet V 2001 Molecular cloning and characterization of thyroid hormone receptors in teleost fish. J Mol Endocrinol 26:51–65[Abstract]
  43. Lazar MA, Hodin RA, Chin WW 1989 Human carboyl-terminal variant of {alpha}-type c-erbA inhibits trans-activation by thyroid hormone receptors without binding thyroid hormone. Proc Natl Acad Sci USA 86:7771–7774[Abstract/Free Full Text]
  44. Koenig RJ, Lazar MA, Hodin RA, Brent GA, Larsen PR, Chin WW, Moore DD 1989 Inhibition of thyroid hormone action by a non-hormone binding c-erbA protein generated by alternative mRNA splicing. Nature 337:659–661[CrossRef][Medline]
  45. Katz D, Lazar MA 1993 Dominant negative activity of an endogenous thyroid hormone receptor variant ({alpha}2) is due to competition for binding sites on target genes. J Biol Chem 268:20904–20910[Abstract/Free Full Text]
  46. Chassande O, Fraichard A, Gauthier K, Flamant F, Legrand C, Savatier P, Laudet V, Samarut J 1997 Identification of transcripts initiated from an internal promoter in the c-erbA {alpha} locus that encode inhibitors of retinoic acid receptor-alpha and triiodothyronine receptor activities. Mol Endocrinol 11:1278–1290[Abstract/Free Full Text]
  47. Macchia PE, Takeuchi Y, Kawai T, Cua K, Gauthier K, Chassande O, Seo H, Hayashi Y, Samarut J, Murata Y, Weiss RE, Refetoff S 2001 Increased sensitivity to thyroid hormone in mice with complete deficiency of thyroid hormone receptor {alpha}. Proc Natl Acad Sci USA 98:349–354[Abstract/Free Full Text]
  48. Flamant F, Samarut J 2003 Thyroid hormone receptors: lessons from knockout and knock-in mutant mice. Trends Endocrinol Metab 14:85–90[CrossRef][Medline]
  49. Bigler J, Eisenman RN 1988 c-erbA encodes multiple proteins in chicken erythroid cells. Mol Cell Biol 8:4155–4161[Abstract/Free Full Text]
  50. Wang Z, Brown DD 1991 A gene expression screen. Proc Natl Acad Sci USA 88:11505–11509[Abstract/Free Full Text]
  51. Wang Z, Brown DD 1993 Thyroid hormone-induced gene expression program for amphibian tail resorption. J Biol Chem 268:16270–16278[Abstract/Free Full Text]
  52. Brown DD, Wang Z, Furlow JD, Kanamori A, Schwartzman RA, Remo BF, Pinder A 1996 The thyroid hormone-induced tail resorption program during Xenopus laevis metamorphosis. Proc Natl Acad Sci USA 93:1924–1929[Abstract/Free Full Text]
  53. Jung JC, Leco KJ, Edwards DR, Fini ME 2002 Matrix metalloproteinases mediate the dismantling of mesenchymal structures in the tadpole tail during thyroid hormone-induced tail resorption. Dev Dyn 223:402–413[CrossRef][Medline]
  54. Yaoita Y, Brown DD 1990 A correlation of thyroid hormone receptor gene expression with amphibian metamorphosis. Genes Dev 4:1917–1924[Abstract/Free Full Text]
  55. Shaffer HB, Voss SR 1996 Phylogenetic and mechanistic analysis of a developmentally integrated character comple: alternate life history modes in ambystomatid salamanders. Am Zool 36:24–35
  56. Voss SR, Shaffer HB, Taylor J, Safi R, Laudet V 2000 Candidate gene analysis of thyroid hormone receptors in metamorphosing vs nonmetamorphosing salamanders. Heredity 2000 85:107–114
  57. Voss SR, Shaffer HB 1997 Adaptive evolution via a major gene effect: paedomorphosis in the Mexican axolotl. Proc Natl Acad Sci USA 94:14185–14189[Abstract/Free Full Text]
  58. Voss SR, Shaffer HB 2000 Evolutionary genetics of metamorphic failure using wild-caught vs laboratory axolotls (Ambystoma mexicanum). Mol Ecol 9:1401–1407[CrossRef][Medline]