Article

• The EMBO Journal (1999) 18, 623 - 631
• doi:10.1093/emboj/18.3.623

Different functions for the thyroid hormone receptors TR and TR in the control of thyroid hormone production and post-natal development

Karine Gauthier^1,5, Olivier Chassande^1,5, Michela Plateroti¹, Jean-Paul Roux², Claude Legrand¹, Bertrand Pain¹, Bernard Rousset³, Roy Weiss⁴, Jacqueline Trouillas³ and Jacques Samarut¹

1. CNRS UMR 49-INRA LA 913, Ecole Normale Supérieure, 46 allée d'Italie, 69364 Lyon cedex, France
2. INSERM U403, Hopital Edouard Herriot, Lyon, France
3. INSERM U369, Faculté de Medecine, RTH Laennec, Lyon, France
4. Thyroid Unit, MC 3090, University of Chicago, Chicago IL 60637, USA
5. K.Gauthier and O.Chassande contributed equally to this work

Correspondence to:

Olivier Chassande, E-mail: ochassan@ens-lyon.fr

Received 26 October 1998; Accepted 25 November 1998

• Keywords:

  • bone,
  • intestine,
  • knockout,
  • thyroid hormone receptor,
  • TSH

Introduction

Thyroid hormones are known to induce metamorphosis in amphibians by mediating remodelling of specific tissues and organs (for review see Tata, 1993; Kaltenbach, 1996). Triiodothyronin (T3) is essential for post-natal development of mammals, as hypothyroidism leads to growth retardation and impaired neurogenesis (Legrand, 1986). T3 is also involved in many aspects of adult life. In humans, hypothyroidism affects the function of many organs and results in alterations of thermogenesis and behaviour (Legrand, 1986).

The functions of T3 are mediated by three nuclear thyroid receptors, TR1, TR1 and TR2, encoded by two genes, TR and TR, respectively (Sap et al., 1986; Weinberger et al., 1986). The TR locus generates the 1 and 2 receptors by using two different promoters and alternative splicing. The TR locus generates the 1 receptor and three non-T3 binding proteins: TR2, which results from alternative splicing of the TR primary transcript (Koenig et al., 1989; Lazar et al., 1989); and TR1 and TR2 transcripts, which are generated from an internal promoter located in intron 7 of the TR locus (Chassande et al., 1997). The thyroid hormone receptors belong to the family of nuclear receptors which includes retinoic acid receptors, 9-cis retinoic acid receptors, vitamin D3 receptors and peroxisome proliferator-activated receptors (Laudet et al., 1992). All these receptors contain a DNA-binding domain and a ligand-binding domain; they mediate ligand-dependent transcriptional control of target genes (Mangelsdorf et al., 1995). The 1 and 2 receptors only differ in their N-terminal region. The and receptors display remarkably conserved sequences in their DNA-binding, ligand-binding and ligand-dependent transactivation domains, but their sequences differ completely in the N-terminal region. The consequences of these structural homologies is that all three receptors bind the same ligand and the same motifs on DNA. Despite these similarities, recent in vivo investigations of the respective functions of these receptors have suggested that each of them is likely to mediate a limited number of the physiological activities of T3.

The TRs not only mediate the action of T3 but also play a role in maintaining the concentration of ligand at a constant level. This is achieved by feedback regulation involving pituitary thyroid stimulating hormone (TSH). T3 negatively controls the synthesis and secretion of TSH and thyrotropin releasing hormone (TRH) at the level of pituitary and hypothalamus, respectively, through a negative feedback loop (Lezoualc'h et al., 1992; Scanlon and Toft, 1996). The disruption of the TR gene leads to hyperthyroxinemia (Forrest et al., 1996; Weiss et al., 1998). In contrast, mice with targeted disruption of the TR1 and TR1 genes (TR1^-/-) exhibit a moderate hypothyroxinemia and mild hypothyroidism (Wikstrom et al., 1998), and mice lacking both TR1 and TR2 display hypothyroxinemia which increases upon aging and severe hypothyroidism (Fraichard et al., 1997). These data suggest that the products of both the and genes are involved, although in different ways, in the control of TSH production.

Other physiological functions are differentially affected by the deletion of either of the receptors. The inactivation of the TR gene results in impairment of the auditory function, but no alteration in development, metabolism or neurological functions has been described in these animals (Forrest et al., 1996). TR1^-/- mice show an abnormal heart rate and lower body temperature (Wikstrom et al., 1998). Mice which lack both the TR1 and the TR2 isoforms (TR^-/-) exhibit growth retardation, a lower body temperature, a delayed maturation of bone and intestine and they die shortly after weaning (Fraichard et al., 1997). From these studies, it is clear that and receptors are differentially involved in the control of developmental, endocrine and metabolic processes.

The difference in the pattern of expression of the three receptors may partly account for these functional differences. The pleiotropic effect of the mutation in TR is consistent with the wide distribution of TR transcripts, whereas the more limited alterations generated by the ablation of the TR gene are in agreement with the more restricted pattern of expression observed for TR transcripts. A few organisms and tissues such as early chick embryos (Forrest et al., 1991; Flamant and Samarut, 1998) and rat cerebellum (Bradley et al., 1989) express only isoforms but most tissues express several isoforms with varying ratios. Redundancy may occur in tissues where both and receptors are expressed.

To investigate the respective roles of the TR and TR genes, we have generated mice with a targeted disruption of the TR gene and, using the previously described TR^-/- mice, produced double mutant mice. We have compared the phenotypes of wild-type, TR^-/-, TR^-/- and TR^-/-TR^-/- mice, and investigated the pituitary and thyroid functions of these mice in order to analyse the respective parts played by the and receptors in the control of the production of thyroid hormones. We have also compared the effects of the single and double mutations on the maturation of bone and intestine, in order to analyse potential functional redundancies between and receptors in these organs.

Production of mice with inactivated TR

To inactivate the TR gene, a recombination cassette containing both a lacZ coding sequence and a Neo^R gene driven by the -actin promoter was introduced downstream of exon 3 (Figure 1A). The presence of two polyadenylation sites downstream of the lacZ and Neo^R genes, respectively, was designed to prevent further transcription of both TR1 and TR2 mRNAs. Moreover, the region of the gene which encodes the DNA-binding domain of the receptors was removed in the replacement vector, to prevent the production of a chimeric protein that could potentially compete with other receptors for the binding to DNA. Two independent clones of recombinant embryonic stem (ES) cell clones were isolated and injected into blastocysts. Heterozygous mice were derived in an inbred 129SV background. Homozygous animals were obtained by intercrossing heterozygous animals (Figure 1B). As previously described (Forrest et al., 1996), homozygous animals were fertile and displayed hyperthyroxinemia.

Figure 1.

Disruption of the TR gene by homologous recombination. (A) Diagram of the TR gene. Large arrows represent the two alternative transcriptional initiation sites used to generate the 1 and 2 transcripts. Arrowheads represent the position of the oligonucleotides used to identify the wild-type and mutant alleles by PCR. (B) Genotypes of the progenies of the TR^+/- intercrosses. The upper band ('-') is the mutant 1.2 kb fragment amplified with oligonucleotides i3 and laczAS. The lower band ('+') is the wild-type 210 bp fragment amplified using the e5 and e5A primers. (C) Expression of the TR and TR transcripts in wild-type and mutant mice. The TR and TR transcripts (lines) were amplified in lung and intestine from TR^-/-, TR^-/-, double mutant and wild-type mice (columns).

View full figure (44 KB)

Production of mice with both inactivated TR and TR

TR^+/+TR^-/- mice were first crossed with TR^+/-TR^+/+ mice to generate double heterozygous animals, which were fertile and intercrossed to yield viable and fertile TR^+/-TR^-/- mice. These latter mice were further intercrossed to generate double homozygous animals. TR^-/-TR^-/- mice were born alive with no obvious morphological or physiological alterations. Among 148 TR^-/- pups, 45 (30.4%), 69 (46.6%) and 34 (23%) were TR^+/+, TR^+/- and TR^-/-, respectively, The newborn mice were able to move and suckle normally, compared with TR^-/- or wild-type mice. These data show that the absence of both TR and TR receptors does not impair the embryonic development of the mouse.

RT–PCR experiments were designed to monitor the expression of the transcripts produced from the TR and TR loci. Figure 1C shows that no transcript containing sequences downstream of the lacZNeo^R cassette was detected in intestine or lung from TR^-/- or TR^-/-TR^-/- mice.

Altered growth rate and post-natal death in TR^-/-TR^-/- mice

There was no obvious difference in the growth rates of wild-type and TR^-/- animals, respectively (data not shown), in agreement with previous observations (Forrest et al., 1996). As previously described for the TR^-/- mice, the newborn double mutant animals exhibited an altered growth rate. Until the age of 15 days, the difference in body weight between TR^-/- and double mutant mice was moderate (Figure 2). From this age on, the growth of double mutants was stopped and after 3 weeks a decrease in body weight was observed, leading to death by the age of 5 weeks. Growth of the TR^+/-TR^-/- mice was slower than that of TR^-/- mice.

Figure 2.

Growth curves of the seven pups' progeny resulting from a TR^+/-TR^-/- mice intercross. Squares, TR^-/- animals; triangles, TR^-/-TR^+/- animals; circles, TR^-/-TR^-/- animals.

View full figure (68 KB)

As for the TR^-/- mice, morphological examination of skeleton, liver, heart, lung and kidneys of 3-week-old TR^-/-TR^-/- mice showed that these organs did not reveal overt abnormalities despite a significant reduction in size compared with wild-type or TR^-/- animals (data not shown).

Hyperproduction of TSH and thyroid hormones in TR^-/-TR^-/- mice

We showed previously that from the age of 3 weeks TR^-/- mice were progressively becoming hypothyroid. By the age of 5 weeks, plasma thyroid hormone concentrations reached values representing 30% of hormone levels in normal mice (Fraichard et al., 1997). On the contrary, elevated plasma concentrations of TSH, T3 and thyroxin (T4) have been reported for TR^-/- mice (Forrest et al., 1996). We measured the plasma concentrations of these hormones in 3-week-old wild-type, TR^-/-, TR^-/- and TR^-/- TR^-/- mice (Figure 3). In these experiments, comparison of the concentrations of TSH and thyroid hormones in TR^-/- mice versus wild-type mice did not have any statistical significance, as assessed by an unpaired t-test (p >0.37 for the three parameters). This was due to the large variance and the low number of animals tested. Comparisons of all parameters in TR^-/- and TR^-/-TR^-/- versus wild-type on the one hand, TR^-/- and TR^-/-versus TR^-/-TR^-/- on the other, were meaningful as assessed by p values <0.001 (see Figure 3). Comparison of TR^-/- versus TR^-/- was valid for T3 and T4 but not for TSH. In agreement with the observations from others (Forrest et al., 1996; Weiss et al., 1998), TSH, T4 and T3 concentrations in TR^-/- mice were increased 3- to 6-fold compared with normal controls (Figure 3). In double mutant mice, both TSH and thyroid hormone concentration were surprisingly high. Plasma T4 and T3 levels were 10-fold higher than those of normal mice, whereas plasma TSH concentration was increased by >100-fold (Figure 3).

Figure 3.

T3, T4 and TSH are markedly increased in mice lacking TR receptors. Thyroid hormones (total T4, total T3) and TSH concentrations in serum from 3-week-old wild-type (wt; black bars), TR^-/- (hatched bars), TR^-/- (cross-hatched bars) or double mutant mice (white bars). The number of animals used for each measurement is indicated in parentheses on the top of each bar. Statistical analysis was performed using an unpaired t-test to assess the significance of the differences observed between the values obtained in the different genotypes.

View full figure (83 KB)

Histological features of the thyroid in TR^-/-, TR^-/- and double mutants

At the autopsy of 3-week-old mice, gross examination revealed an enlargement of the thyroid gland in TR^-/- and TR^-/-TR^-/- mice compared with wild-type mice, whereas in TR^-/- mice the thyroid gland was much smaller and could not be distinguished from the muscular and the adipose surrounding tissues. Histological comparison of the largest sections of the tracheal block confirmed the diffuse hyperplasia of the thyroid gland in the TR^-/- (Figure 4, compare C1 with A1) and in the TR^-/-TR^-/- mice (compare D1 with A1 in Figure 4), in contrast to the hypoplasia observed in the thyroid gland of the TR^-/- mice (Figure 4, compare B1 with A1). On the high magnification, the follicles in TR^-/- mice were small and disorganized (Figure 4, compare B2 with A2). TR^-/- and double mutants exhibited an increase in the number of follicles but the thyroid glands of these animals differed in size and content of the follicles, aspects of colloid and development of vascularization (Figure 4, compare C2 and D2 with A2). In agreement with the observations of Forrest et al. (1996), Figure 4C2 clearly shows an increase in size of the follicles in TR^-/- compared with wild-type mice (Figure 4A2). In TR^-/-TR^-/- mice (Figure 4D2), the follicles were smaller and contained a less abundant colloid with vacuoles near the epithelium. The capillaries filled with red cells seemed more numerous in TR^-/-TR^-/- than in TR^-/- mice. Active colloid resorption and increased vascularity observed in the thyroid gland of double mutant animals are consistent with hyperactivity.

Figure 4.

Histology of the thyroid gland in TR mutants. Transverse sections through the thyroid trachea block of a 3-week-old control mouse (A), and TR^-/- (B), TR^-/- (C) and TR^-/-TR^-/- (D) mutant mice. The TR^-/- (C1) and TR^-/-TR^-/- (D1) exhibited a bilateral thyroid enlargement and the TR^-/- (B1) exhibited a moderate but clear hypoplasia, compared with controls (A1). Scale bars in A1–D1, 200 m. pt: parathyroid gland. Higher magnification showed that enlargement was due to increased number and size of follicles (f) in TR^-/- mice (C2), compared with controls (A2). In TR^-/-TR^-/- mutants (D2), small follicles exhibited signs of hyperactivity: colloid vacuoles (v), cuboid epithelium and hypervascularization (bv). Scale bars in A2–D2, 20 m.

View full figure (97 KB)

Delayed maturation of bone and intestine in TR^-/-TR^-/- mice

The overall morphology of the skeleton was not disturbed in TR^-/- or in double mutant animals. Histological examination of the tibia of TR^-/- mice did not reveal any difference compared with the wild-type animals (Figure 5A and B). In contrast, the tibia of TR^-/-TR^-/- mice exhibited impaired development of epiphyseal bone centres characterized by a hypertrophied cartilage associated with a low ossification (Figure 5D). This phenotype is identical to the one observed in TR^-/- mice (Figure 5C; Fraichard et al., 1997). This study was carried out with 2-week-old animals. At this age, TR^-/- and wild-type animals have similar thyroid hormone plasma concentrations (data not shown), avoiding any interference of hypothyroidy with the phenotype.

Figure 5.

Bone development is impaired in double mutant as in TR^-/- mice. Longitudinal bone sections (7 m) of wild-type (A), TR^-/- (B), TR^-/- (C) and TR^-/-TR^-/- (D) mice. Bar, 200 m. m, metaphysis; d, diaphysis. The horizontal arrow indicates the growth plate.

View full figure (68 KB)

Morphological appearance of the small intestine in the four groups of 2-week-old animals was also studied. As previously described, TR^-/- animals showed reduced length of the intestine as well as a greater fragility compared with the wild-types. This was also observed in the double mutant mice. On the contrary, we did not observe any difference between wild-type and TR^-/- animals. Histological analysis was then conducted at the proximal and distal levels of the small intestine and in the proximal colon. In accordance with previous observations, the colon was not affected by the receptor deficiency. Here we show histological staining at the distal ileum level (Figure 6). The intestinal size and the length of the crypt-villus unit was reduced in the TR^-/- compared with wild-type mice (Figure 6, C and D versus A and B). On the contrary, TR^-/- animals did not show such intestinal alterations (Figure 6E and F). The intestines of the double knockout mice displayed a histological impairment similar to TR^-/- mice at the proximal jejunum level (data not shown; Fraichard et al., 1997). Strikingly, a much more dramatic alteration was observed in the distal ileum of double mutant mice compared with TR^-/- mice (Figure 6G–I). This severe phenotype was characterized by very few and short villi-lined, flat epithelial cells (Figure 6I). The thickness of the external muscle coats in double mutants was similar to TR^-/- DI (Figure 6H versus D).

Figure 6.

Small intestinal phenotypic maturation is more severely impaired in double mutant than in TR^-/- mice. Periodic acid–Schiff (PAS) staining of transverse section of wild-type (A and B), TR^-/- (C and D), TR^-/- (E and F) and TR^-/-TR^-/- (G and H) ileum. At low magnification (A, C, E and G), bar = 250 m. v, villi; c, crypt. At higher magnification (B, D, F and H), bar = 100 m. cml, circular muscle layer; lml, longitudinal muscle layer. At very high magnification (I), bar = 40 m.

View full figure (103 KB)

Unliganded TR is not responsible for the phenotype resulting from the disruption of the TR gene

We have previously described a severe phenotype in mice lacking the TR1 and TR2 proteins (Fraichard et al., 1997). TR^-/- animals displayed progressive hypothyroidism together with multiple disorders leading to death shortly after the weaning period. We postulated that this phenotype could be the result of hypothyroidism reinforced by the dominant negative activity of unliganded TR in these mice. The ablation of the TR gene does not correct the severe phenotype observed in TR^-/- animals, showing that the phenotype generated by the disruption of TR is not the result of a dominant-negative activity of the TR aporeceptor.

TR and TR cooperate to repress TSH production

The hyperthyroxinemia observed in TR^-/- animals (our data; Forrest et al., 1996) indicates that, in normal mice, liganded TR receptors repress TSH synthesis, and this very probably mainly results from direct interaction of TR with negative response elements present in the TSH gene promoter (Wondisford et al., 1989; Wood et al., 1989), although TRH disregulation may also occur (Feng et al., 1994; Yamada et al., 1997). These data suggest that TR alone is not able to fully control TSH expression. However, the comparison between TR^-/- and TR^-/- TR^-/- mice presenting extremely high circulating TSH levels, shows that TR is also capable of repressing TSH expression. This confirms recent observations suggesting that TR1 could regulate the production of TSH and thyroid hormones in the absence of TR receptors (Weiss et al., 1997). We propose that in normal mice, TR and TR cooperate to repress TSH transcription, TR being a more potent repressor.

In apparent contradiction with this model, we have previously shown that the absence of TR receptors in wild-type mice leads to hypothyroxinemia (Fraichard et al., 1997), and others have described limited hypothyroxinemia in TR1^-/- mice (Wikstrom et al., 1998). These data suggested that the TR proteins might balance the action of TR receptors and positively control the production of thyroid hormones. This contradiction is easily resolved if we assume that the absence of the weak repressor, TR, eliminates the competition between the two types of receptors and enables the stronger repressor, TR, to fully exert its repression potential. This assumption is supported by data showing that TR proteins are more efficient than TR1 in binding the negative thyroid hormone response element of the TSH promoter and repressing its activity (McCabe et al., 1998).

The TR and TR proteins have redundant functions in some but not all tissues

The extended pattern of expression of the TR gene may account for the pleiotropic phenotype of TR^-/- mice. In contrast, TR^-/- mice display a mild and tissue-restricted phenotype (Forrest et al., 1996; our data), consistent with their more restricted pattern of expression. In some tissues, however, the TR and TR genes are coexpressed (Bradley et al., 1989, 1992; Forrest et al., 1990, 1991; Macchia et al., 1990; Strait et al., 1990, 1991). This raises the possibility of a functional redundancy between TR and TR proteins. Our data show that this assumption is verified in some, but not all tissues.

As bones from young TR^-/- and TR^-/-TR^-/- mice look identical, and are normal in TR^-/- animals, it seems unlikely that TR plays any function in long bone formation although it is expressed in chondrocytes (Abu et al., 1997). We conclude that TR, but not TR, is essential for the normal post-natal maturation of bone, and that TR cannot substitute for TR in this process.

In contrast, histological analysis reveals clear differences between TR^-/-TR^-/- and TR^-/- in the distal ileum, but not in the jejunum or in the colon. In the ileum, both TR and TR are coexpressed in the epithelium (Plateroti,M., Chassande,O., Fraichard,A., Gauthier,K., Freund,J.-N., Samarut,J. and Kedinger,M., submitted). Our data suggest that these genes are redundant in distal ileum, although TR is not able to fully compensate for the absence of TR. They also suggest that TR is responsible for the partial recovery of the intestinal phenotype after treatment of TR^-/- mice by T3 injections (Plateroti,M., Chassande,O., Fraichard,A., Gauthier,K., Freund,J.-N., Samarut,J. and Kedinger,M., submitted). We assume that redundancy occurs for other functions and enables the rescue of TR^-/- mice by T3 (Fraichard et al., 1997).

The thyrocytes express TR and TR isoforms (Selmi-Ruby and Rousset, 1996), but the functions of these proteins in the thyroid gland remain poorly understood. TSH exerts a tight control on the development and metabolism of the thyroid, as an increase in TSH concentration triggers hypertrophy and hyperplasia of the thyroid gland and results in enhanced secretion of thyroid hormones. Therefore, it is likely that the morphological and histological features of the thyroid gland in the TR^-/- mice is the consequence of the increase of TSH concentration. Further increase in TSH concentration would be expected to result in enhanced hypertrophy, as it is observed in pathological or experimental hypothyroidism or in TR^-/-TR1^-/- mice (D.Forrest, personal communication). However, in our double mutant animals, which display a very high concentration of TSH, the hypertrophy of the gland is blunted. Since the disruption of the TR gene in TR^-/- animals leads to hypotrophy of the thyroid gland, it is possible that their absence in double mutant mice prevents a larger hypertrophy in response to TSH. From these data we assume that TR plays an important role in the development of the thyroid gland. In conclusion, this work shows that the currently known T3 receptors are dispensable for the embryonic development of the mouse and that the products of the TR and TR genes display tissue-specific functions with limited redundancy.

Further investigations using targeted disruption of the TR and TR genes should unravel the respective contribution of each receptor isoform to peripheric functions, and the molecular basis of the tissue-specific functions.

ES cell selection and generation of mutant mice

The TR targeting vector was constructed using pGNA as starting plasmid (gift from Dr P.Brulet). The 3' arm homologous to TR was cloned from a EMBL4 library (gift from Dr J.P.Magaud). The 5' arm was amplified by PCR from mouse 129 DNA. The thymidine kinase gene was inserted as previously described (Fraichard et al., 1997). The targeting vector was digested with SpeI. 'ENS' ES cells were established from 129sv blastocysts in our laboratory by B.Pain and D.Aubert according to the technique described previously (Robertson, 1987). These cells were karyotyped as male and were proven to contribute with a very high efficiency to the germline when injected into blastocysts. We routinely maintain 'ENS' cells on MEF feeders in Glasgow minimal essential medium supplemented with 10% fetal bovine serum and 1000 U/ml LIF. 'ENS' ES cells were electroporated with 40 g linearized plasmid, then selected with G418 (Gibco-BRL) and Gancyclovir and, after 10 days, clones were picked for screening. Positive clones were subsequently tested for correct karyotype and absence of mycoplasma and subsequently injected into blastocysts from C57Bl6 recipient mice. The genotype of the progenies of the mice carrying a TR mutation was analysed using a PCR-based screening procedure. DNA (20–100 ng) extracted from tails was subjected to amplification using a mixture of four oligonucleotides: i3 (GGAGTCCTCACTAGAGTCACC) and lacZAS (CCTCTTCGCTATTACGCCAGCTGG) allowed the amplification of the mutant allele. e5 (TGGTGCTGGATGACAGCAAG) and e5A (CAGGAATTTCCGCTTCTGCTT) allowed the amplification of the fragment from the wild-type allele. The amplification was carried out on a thermocycler (Perkin Elmer) using the following procedure: 94°C, 2 min, then (94°C, 15 s, 70°C, 15 s, 72°C, 60 s) for 10 cycles then (94°C, 15 s, 60°C, 15 s, 72°C, 60 s) for 30 cycles.

RNA analysis by RT–PCR

Oligonucleotides were from MWG ScienceTech (France). Reverse transcription was performed as follows: 1 g of total RNA and 0.5 g of random primers were mixed in 10 l of water, heated to 68°C for 5 min and cooled to 37°C. Then, 10 l of polymerization mix [100 mM Tris–HCl pH 8.3, 150 mM KCl, 20 mM dithiothreitol, 6 mM MgCl₂, 0.5 mM each dNTP, 200 U reverse transcriptase (Promega)] were added and the mixture was incubated at 37°C for 30 min. PCR was performed using 0.8 units Eurobiotaq thermostable DNA polymerase (Eurobio, France) in a 50 l reaction mix. One microlitre of reverse transcription mix was used in each reaction. The detection of TR transcripts was carried out using oligonucleotides E6 (GGTGCTGGATGACAGCAAGA) and E7A (GCATTCACGATGGGTGCTTGT) were used as primers. Oligonucleotides VIS (GGAGATGATTCGCTCACTGCAG) and 15A (CAGCCTGCAGCAGAGCCACTTCCGT) were used as primers to detect TR transcripts. The amplification was performed using the following parameters: 94°C, 2 min, (94°C, 15 s, 58°C, 15 s, 72°C, 60 s) for 35 cycles.

Intestine and bone analysis

Intestine analysis. Jejunum and ileum were dissected from 2-week-old pups, fixed in Bouin Holland's reagent and embedded in paraffin for serial sections and morphological observations. Sections of 5 m were stained according to the standard PAS protocol (Segal and Petras, 1992).

Bone analysis. Leg tibia were dissected out from 2-week-old pups, fixed in 70% alcohol and embedded. Undecalcified sections (7 m) were obtained and stained with Goldner's trichrome, according to the bone histomorphometric protocol (Meunier, 1983). Longitudinal medial sections were analysed with a true colour image-processing workstation Visiolab 1000 (Biocom, France).

Hormone assays and thyroid gland histology

Three-week-old mice were anaesthetized with ketamine (Panpharma, France) + Hypnovel (Roche, France) and bled intracardiacally. Sampling for all experiments was performed from 12 a.m. to 3 p.m. to minimize day–night variations. Serum concentrations of total T3 and total T4 were determined by specific radioimmunoassay, using sodium anilino-naphtalene sulfonate and acidic buffer, as binding inhibitors of serum proteins (Rousset et al., 1984). TSH was measured by specific radioimmunoassay, as described previously (Weiss et al., 1997). Tracheal blocks of mice were dissected out, immediately fixed in Bouin Holland, embedded in paraffin and cut in totality by serial sections of 5 m. Sections were stained according to the standard Masson's Trichrome protocol. Thyroid areas were compared on the largest sections.

We are grateful to Dr F.Flamant for advice, helpful criticism and repeated encouragement. We would like to warmly thank Dr Denise Aubert who manages the transgenic facility at ENS de Lyon. We thank D.Belgarbi and C.Morin for animal breeding. We are grateful to Pascale Chevalier for the histology of the thyroid gland, to Dr H.Bornet for the measurements of the thyroid hormones in plasma, to Christiane Arnold for the histological analysis of the intestine and to Sandrine Mouradian for technical assistance in the construction of homologous recombination vectors. We thank Luc Malaval, Patrick Delmas and Michele Kedinger for their contribution to this work. This work was supported by grants of the Association pour la Recherche contre le Cancer and Region Rhone Alpes.

Abu EO, Bord S, Horner A, Chatterjee VKK and Compston JE (1997) The expression of thyroid hormone receptors in human bone. Bone, 21, 137–142. | Article | PubMed | ISI | ChemPort |

Bradley DJT, Young WD and Weinberger C (1989) Differential expression of and thyroid hormone receptor genes in rat brain and pituitary. Proc Natl Acad Sci USA, 86, 7250–7254. | PubMed | ChemPort |

Bradley DJT, Howard C and Scott Young W,III (1992) Spatial and temporal expression of - and -thyroid hormone receptor mRNAs, including the 2 subtype, in the developing mammalian nervous system. J Neurosci, 12, 2288–2302. | PubMed | ISI | ChemPort |

Chassande O, Fraichard A, Gauthier K, Flamant F, Legrand C, Savatier P, Laudet V and Samarut J (1997) Identification of transcripts initiated from an internal promoter in the c-erbA locus that encode inhibitors of retinoic acid receptor- and triiodothyronine receptor activities. Mol Endocrinol, 11, 1278–1290. | Article | PubMed | ISI | ChemPort |

Feng P, Li QL, Satoh T and Wilber JF (1994) Ligand (T3) dependent and independent effects of thyroid hormone receptors upon human TRH gene transcription in neuroblastoma cells. Biochem Biophys Res Commun, 200, 171–177. | Article | PubMed | ISI | ChemPort |

Flamant F and Samarut J (1998) Involvement of thyroid hormone and its receptor in avian neurulation. Dev Biol, 197, 1–11. | Article | PubMed | ISI | ChemPort |

Forrest DH, Sjöberg M and Vennström B (1990) Contrasting developmental and tissue-specific expression of and thyroid hormone receptor genes. EMBO J, 9, 1519–1528. | PubMed | ISI | ChemPort |

Forrest DH, Hallböök F, Persson H and Vennström B (1991) Distinct functions for thyroid hormone receptors and in brain development indicated by differential expression of receptor genes. EMBO J, 10, 269–275. | PubMed | ISI | ChemPort |

Forrest D, Hanebuth E, Smeyne RJ, Everds N, Stewart CL, Wehner JM and Curran T (1996) Recessive resistance to thyroid hormone in mice lacking thyroid hormone receptor : evidence for tissue-specific modulation of receptor function. EMBO J, 15, 3006–3015. | PubMed | ISI | ChemPort |

Fraichard A et al. (1997) The TR gene encoding a thyroid hormone receptor is essential for post-natal development and thyroid hormone production. EMBO J, 16, 4412–4420. | Article | PubMed | ISI | ChemPort |

Kaltenbach JC (1996) In Gilbert,L.I., Tate,J.B. and Atkinson,B.G. (eds), Endocrinology of Amphibian Metamorphosis. Academic Press, New York, NY, 403–431.

Koenig RJ, Lazar AA, Hodin RA, Brent GA, Larsen PR, Chin WW and 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. | Article | PubMed | ISI | ChemPort |

Laudet V, Hänni C, Coll J, Catzeflis F and Stehelin D (1992) Evolution of the nuclear receptor gene superfamily. EMBO J, 11, 1003–1013. | PubMed | ISI | ChemPort |

Lazar MA, Hodin RA and Chin WW (1989) Human carboxyl-terminal variant of -type c-erbA inhibits trans-activation by thyroid hormone receptors without binding thyroid hormone. Proc Natl Acad Sci USA, 86, 7771–7774. | PubMed | ChemPort |

Legrand J (1986) In Hennemann,G. (ed.), Thyroid Hormone Effects on Growth and Development Marcel Dekker, Rotterdam, The Netherlands, Vol. 78, 503–534.

Lezoualc'h F, Hassan AH, Giraud P, Loeffler JP, Lee SL and Demeneix BA (1992) Assignment of the -thyroid hormone receptor to 3,5,3'- triiodothyronine-dependent inhibition of transcription from the thyrotropin-releasing hormone promoter in chick hypothalamic neurons. Mol Endocrinol, 6, 1797–1804. | Article | PubMed | ChemPort |

Macchia E et al. (1990) Characterization of site-specific polyclonal antibodies to c-erbA peptides recognizing human thyroid hormone receptors 1, 2 and and native, 3, 5, 3'-triiodothyronine receptor and study of tissue distribution of the antigen. Endocrinology, 126, 3232–3239. | PubMed | ISI | ChemPort |

Mangelsdorf DJ et al. (1995) The nuclear receptor superfamily: the second decade. Cell, 83, 835–839. | Article | PubMed | ISI | ChemPort |

McCabe CJ, Yarwood NJ, Gurr JA, Gittoes NJ, Sheppard MC and Franklyn JA (1998) Differential regulation of the human thyrotropin -subunit promoter by thyroid hormone receptors 1 and 1. Thyroid, 8, 601–608. | PubMed | ISI | ChemPort |

Meunier PJ (1983) A Yearly Survey of Developments in the Field of Bone and Mineral Metabolism W.A.Peck, Amsterdam, The Netherlands.

Robertson EJ (1987) Embryo-derived stem cell lines. In Robertson,E.J. (ed.), Teratocarcinomas and Embryonic Stem Cells. A Practical Approach. IRL Press, Oxford, 84–96.

Rousset B, Cure M, Jordan D, Kervran A, Bornet H and Mornex R (1984) Metabolic alterations induced by chronic heat exposure in the rat: involvement of thyroid function. Eur J Physiol, 401, 64–70. | ChemPort |

Sap J, Munoz A, Damm K, Goldberg Y, Ghysdael J, Leutz A, Beug H and Vennström B (1986) The c-erbA protein is a high-affinity receptor for thyroid hormone. Nature, 324, 635–640. | Article | PubMed | ISI | ChemPort |

Scanlon MF and Toft AD (1996) In Braverman,L.E. and Utiger,R.D. (eds), Regulation of Thyrotropin Secretion Lippincott-Raven, New York, NY, 220–240.

Segal GH and Petras RE (1992) Histology for Pathologists. Raven Press, New York, NY.

Selmi-Ruby S and Rousset B (1996) Analysis of the functional state of T3 nuclear receptors expressed in thyroid cells. Mol Cell Endocrinol, 119, 95–104. | Article | PubMed | ChemPort |

Strait KAS, Harold L, Perez-Castillo A and Oppenheimer JH (1990) Relationship of c-erbA mRNA content to tissue triiodothyronine nuclear binding capacity and function in developing and adult rats. J Biol Chem, 265, 10514–10521. | PubMed | ISI | ChemPort |

Strait KAS, Harold L, Seybold VS, Ling NC and Oppenheimer JH (1991) Immunofluorescence localization of thyroid hormone receptor protein 1 and variant 2 in selected tissues: cerebellar Purkinje cells as a model for 1 receptor-mediated developmental effects of thyroid hormone in brain. Proc Natl Acad Sci USA, 88, 3887–3891. | PubMed | ChemPort |

Tata JR (1993) Gene expression during metamorphosis: an ideal model for post-embryonic development. BioEssays, 15, 239–248. | PubMed | ISI | ChemPort |

Weinberger C, Thompson CC, Ong ES, Lebo R, Gruol DJ and Evans RM (1986) The c-erbA gene encodes a thyroid hormone receptor. Nature, 324, 641–646. | Article | PubMed | ISI | ChemPort |

Weiss RE, Forrest D, Pohlenz J, Cua K, Curran T and Refetoff S (1997) Thyrotropin regulation by thyroid hormone in thyroid hormone receptor beta-deficient mice. Endocrinology, 138, 3624–3629. | Article | PubMed | ISI | ChemPort |

Weiss RE, Murata Y, Cua K, Hayashi Y, Seo H and Refetoff S (1998) Thyroid hormone action on liver, heart and energy expenditure in thyroid hormone receptor deficient mice. Endocrinology, 139, 4945–4952. | Article | PubMed | ISI | ChemPort |

Wikstrom L, Johansson C, Salto C, Barlow C, Campos Barros A, Baas F, Forrest D, Thoren P and Vennström B (1998) Abnormal heart rate and body temperature in mice lacking thyroid hormone receptor 1. EMBO J, 17, 455–461. | Article | PubMed | ISI | ChemPort |

Wondisford FE, Farr EA, Radovick S, Steinfelder HJ, Moates JM, McClaskey JH and Weintraub BD (1989) Thyroid hormone inhibition of human thyrotropin -subunit gene expression is mediated by a cis-acting element located in the first exon. J Biol Chem, 264, 14601–14604. | PubMed | ISI | ChemPort |

Wood WM, Kao MY, Gordon DF and Ridgway EC (1989) Thyroid hormone regulates the mouse thyrotropin -subunit gene promoter in transfected primary thyrotropes. J Biol Chem, 264, 14840–14847. | PubMed | ISI | ChemPort |

Yamada M et al. (1997) Tertiary hypothyroidism and hyperglycemia in mice with targeted disruption of the thyrotropin-releasing hormone gene. Proc Natl Acad Sci USA, 94, 10862–10867. | Article | PubMed | ChemPort |

Main navigation

• Search EMBO Journal

  Search

• Advanced search

• The EMBO Journal

  • Home
  • Aims and scope
  • Current issue
  • Advance online publication
  • Web focuses
  • Archive:
  • Browse by issue
  • Browse by subject
  • Browse by category
  • Free online sample issue
  • Press releases

• Authors and Referees

  • Editorial process
  • Guide for authors
  • Submit an article
  • Guide for referees
  • Editors and Editorial Board
  • Contact editorial office

• Customer Service

  • Subscribe
  • Order sample copy
  • Purchase articles
  • Reprints and permissions
  • Contact NPG
  • Advertising