| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Clinical Studies |
Department Internal Medicine I (M.B., U.L.), University of Ulm, D-89070 Ulm, Germany; and Department of Medicine (D.B.R.), University of Birmingham, Queen Elizabeth Hospital, B15 2TH Edgbaston, Birmingham, United Kingdom
Address all correspondence and requests for reprints to: Prof. Dr. med. Ulrich Loos, Abteilung Innere Medizin I, Medizinische Klinik und Poliklinik, Universität Ulm, Robert-Koch-Str. 8, D-89070 Ulm, Germany.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the present study, we describe the analysis of the TRß1 gene from a severely retarded RTH patient, designated EZ, and additionally, the performance of a prenatal genetic TRß1 analysis. In the TRß gene of patient EZ, we have found a point mutation producing a premature stop codon, giving rise to a TRß with a remarkably large carboxy-terminal deletion (16). This mutant was shown to lack totally the property of T3 binding. Additionally, using three different consensus TREs, the TRß variant was shown to display abnormal DNA-binding properties, as well as both dominant negative activity and the ability to silence basal gene transcription.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This investigation was based on two principal subjects. The first subject, EZ, was a 15-yr-old female who displayed a severe form of RTH. She was diagnosed as RTH, directly after birth, by routine postnatal hypothyroidism screening. At the age of 2 yr, she was treated with T4 using a regime that rendered her serum TSH concentration within the normal range. At birth, her weight was 3250 g and her length was 50 cm. Mental and physical development were extremely retarded. She had to attend a special school for mentally hindered children. Puberty was somewhat early, with menarche starting at age 11 yr, followed by regular menstruations. Our first clinical examination was performed at the age of 15 yr. Her height (132,7 cm), BW (46,3 kg), and IQ (<50) were evidence of the patient’s physical underdevelopment and severe mental retardation. Bone age was accelerated, being equivalent to that of an average 17-yr-old. Her laboratory data were typical for RTH, i.e. high serum thyroid hormone levels in the presence of nonsuppressed TSH: T4, 25 µg/100 mL (5, 2–12, 6 normal range); FT4E, 10,28 (1, 53–4, 12); T3, 369 ng/100 mL (67–162); FT3E, 152 (15–67); TSH-basal, 3,7 µU/mL (0, 2–2, 9); IgF1, 91,6 ng/mL (182–780); IgFBP3, 1,4 µg/mL (2, 3, 4, 5, 4); alkaline phosphatase, 111 U/L (50–162). The second subject was the unborn child of the parents of EZ. In addition, a limited screen of both parents was carried out.
TRß-gene analysis
The TRß gene in the affected sibling, EZ. Genomic DNA was isolated from whole blood, as previously described (17). TRß exons 4–10 of patient EZ were amplified by PCR using intronic primers (18, 19), and the resultant products cloned into pUC18 (Pharmacia, Freiburg, Germany). The DNA of 8–10 clones per exon was sequenced automatically using an ABI sequencer (model 373A; Applied Biosystems, Warrington, UK).
Prenatal screening for RTH. RNA was isolated from amniocytes taken during the 16th week of pregnancy, which were then cultured for 2 weeks. Full-length TRß complementary DNA (cDNA) was synthesized via RT-PCR. This was used as a template for a series of asymmetrical amplification. The resulting single-stranded PCR products were sequenced directly, as previously described (17).
Parental screening for RTH. Allele-specific PCR (17) was used for screening genomic DNA of the parents for the TRß gene.
Plasmids
A mutant TRß transcription vector specific for the 28-amino acid deletion, termed phTRß-EZ, was generated by cloning the DraIII/BsmI fragment from a pUC18-mutant TRß clone into wild-type TRß transcription vector phTRß-WT (provided by Drs. A. Sakurai and L. J. DeGroot, University of Chicago, Chicago, IL) (20), from which the corresponding wild-type fragment had been excised. For transfection studies, an EcoRI fragment of the TRß-deletion variant phTRß-EZ, harboring the complete mutant TRß cDNA, was cloned into pCDNA3 (Invitrogen, Leek, Netherlands) to produce an expression vector, pCTRß-EZ. The same procedure was used to produce a wild-type expression vector termed pCTRß-WT. The cDNA of mouse RXRß in pBS (provided by Drs. K. Hamada and K. Ozato, NIH, Bethesda, MD) (21) was subcloned in pCDNA3 giving the expression vector pCRXRß. This construct was used for in vitro translations of RXRß. Luciferase-reporter vectors, pTF2- and pTDR4-LUC, were the gift of Drs. P. M. Yen and W. W. Chin (Brigham and Women’s Hospital, Boston, MA) (22). These contained the inverted repeat TRE F2 from the chicken lysozyme gene (23) and an idealized direct half-site repeat TRE, DR4, (24), respectively, coupled to the thymidine kinase (TK)-promotor of pT109LUC (ATCC, Rockville, MD). pTPAL-LUC was generated by cloning two copies of an idealized palindromic TRE (PAL) of the rat GH gene (25) in front of the TK promotor of pT109LUC. Nucleotide sequences of DR4, F2, and PAL are shown in the electrophoretic mobility shift assay (EMSA) section. Control vector pSV-ßGAL was obtained from Promega (Heidelberg, Germany).
In vitro expression of nuclear receptors
A coupled transcription/translation rabbit reticulocyte lysate system (TnT Lysate, Promega) was used to express TRß or RXRß cDNAs from phTRß-WT, phTRß-EZ, and pCRXRß, respectively. 35S-methionine labeled receptors were quantitated by scintillation counting after precipitating them with trichloroacetic acid and visualized by SDS-PAGE.
T3-binding analysis
T3-binding assays (three experiments performed in duplicate) were carried out with aliquots of TRß-translation products, as previously described (17).
EMSA
EMSAs were performed by using double-stranded synthetic oligonucleotides representing the following consensus TREs (capital letters) with BamHI-restriction sites (italics) on either end: 1) DR4 (5'-gatcctacttatAGGTCAcatgAGGTCAagttacg-3'); 2) F2 (5'-gatccacttatTGACCCc-agctgAGGTCAagttacg-3'); and 3) PAL (5'-gatcctcAGGTCATGACCT-gag-3').
Each binding reaction initially contained the following: 2–4 µL in vitro translated 35S-labeled TRß and/or RXRß, brought to a constant amount of reticulocyte lysate by adding untranslated lysate where necessary; DNA-binding buffer (10 mmol/L Tris-HCl, pH 7.5, 50 mmol/L NaCl, 0.5 mmol/L dithiothreitol, 0.05% NP-40, 10% glycerol) to give a final vol of 20 µL. This also contained 50 µg/mL poly (dI-dC) (Pharmacia), 1 µmol/L 9cis-retinoic acid (9c-RA), kindly provided by Drs. U. Fischer, F. Schneider, and P. Weber (Hoffmann-La Roche, Basel, Suisse) and varying concentrations of T3. After 20 min preincubation, 32P-labeled oligonucleotide (2 x 105 cpm), produced by Klenow-fill in of BamHI overhangs, was added, and incubation was continued for 40 min at 22 C. Protein-DNA complexes were resolved in 5% nondenaturing polyacrylamide gels of 0.5 x TBE buffer (45 mmol/L Tris-borate, 1 mmol/L ethylenediaminetetraacetic acid) at constant voltage of 300 V for 2 h. After fixation in acetic acid/isopropanol/water (7:10:83, vol/vol/vol), gels were dried and autoradiographed.
Cotransfection assay
COS-7 and CV-1 cells were grown in 6-well plates in DMEM (Gibco Life Technologies, Eggenstein, Germany) supplemented with 10% FCS (Gibco Life Technologies). Twenty-four hours before transfection, 1 x 105 cells per well were seeded in T3-free medium (26). Each transfection, performed by using the calcium phosphate coprecipitation method (27), contained a total of 0.1–1 µg pCTRß expression vector, 2 µg pTDR4-, pTF2-, or pTPAL-LUC reporter gene vectors, respectively, and 1 µg pSV-ßGAL control vector. The total amount of DNA in each transfection was kept constant by adding appropriate amounts of pCDNA3. After transfection, cells were incubated, either with graded concentrations of T3 or without T3, for 24 h. Luciferase activity of 1/10 cell extract was measured in a luminometer of Berthold (Lumat LB9501, Wildbad, Germany). ß-galactosidase activity of cell extract was determined and used to standardize luciferase activity. Cell extract preparations and luciferase and ß-galactosidase assays were carried out according to the protocols of Promega. Basal TK-promotor activity was determined by cotransfection of cells using pCDNA3 expression vector and the respective TRE-containing LUC-reporter plasmid under the same conditions used in transfections with TRß-expression plasmids. The luciferase values obtained were considered to be the basal TK-promotor activity (Y), and activation (net fold; higher values) or repression (lesser values) were multiples of this basal value. Therefore, fold activation or repression = [(observed LUC activity - Y)/Y] x 100. LUC values for basal TK-promotor activity were uninfluenced by the various T3 concentrations used, so that the mean activity was obtained from measurements over the whole of the T3 concentration range used. Experiments were performed two to three times in triplicate or quadruplicate, and the data represent mean values ± SEM.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Exons 4–10 of the TRß-EZ gene were sequenced. This revealed a
nucleotide exchange in exon 10 (C to A at cDNA-nucleotide position
1587), which resulted in the replacement of cysteine (TGC) codon 434 by
an opal-stop codon TGA (Fig. 1
). Of the twelve clones
analyzed, an equal ratio of mutant to wild-type sequence was found.
This premature stop codon resulted in the deletion of the last 28 amino
acids of the wild-type protein. The point mutation occurred within the
downstream hot-spot region of the ligand-binding domain, and the amino
acids deleted were from a major part of a postulated ligand-binding
subdomain, L2 (28) of the TRß. The deletion, however, was located
outside of the predicted dimerization subdomain.Sequencing of the fetal
TRß cDNA revealed the presence of only the wild-type allele. Thus,
this gene defect was excluded in the unborn sibling. The molecular
evidence of normal fetus TRß cDNA in the 16th week of pregnancy was
confirmed by the birth of a child with normal thyroid status and whose
subsequent development also was normal. This is the first report of a
prenatal diagnosis based on TRß-gene sequencing. Allele-specific
amplification from genomic DNA of EZ’s parents clearly showed that
they too lacked the gene defect exhibited by EZ, which was in accord
with the fact they clinically were normal.
|
Scatchard analysis of in vitro translated TRs showed a
normal affinity constant of 7 x 109
mol/L-1 for the wild-type form, whereas T3
binding for the mutant was undetectable (data not shown). TRß-WT, in
the absence of T3, bound to three TREs (DR4, F2, PAL) as a
homodimer (Fig. 2). In the presence of 10-7
mol/L T3, TRß-WT homodimer dissociated from these TREs,
but there was an accompanying increase in monomer binding. In contrast,
homodimer binding of TRß-EZ was markedly weaker than that of TRß-WT
for DR4 and F2, and TRß-EZ homodimer binding on these two TREs could
not be disrupted by T3. In contrast, homodimer binding of
TRß-EZ to PAL was much stronger, and TRß-EZ homodimer binding to
PAL was not disrupted by T3. Although the electrophoretic
mobilities of TRß-WT and TRß-EZ were slightly different, it was not
possible to identify dimer formation between them. This type of
heterodimer formation may occur because, when using DR4, the monomer
band of TRß-EZ, which was visible when incubated separately with
oligonucleotide, was absent when coincubated with TRß-WT. In
addition, as T3 increased, monomer TRß-WT formation was
less pronounced when TRß-WT was coincubated with TRß-EZ, compared
with its incubation alone with DR4 oligonucleotide. Heterodimerization
of TRß-EZ with RXRß on DR4, F2, and PAL was not affected and EZ:RXR
heterodimers could not be dissociated by either T3
(10-7 mol/L) or a mixture of both T3
(10-7 mol/L) and 9c-RA (10-6 mol/L) (data not
shown).
|
The transcriptional properties of TRß-EZ on different TREs were
analyzed by expression in COS-7 and CV-1 cells. In the absence of
T3, both TRß-EZ and TRß-WT strongly repressed the basal
TK-promotor activity (Fig. 3). In contrast to TRß-WT,
TRß-EZ repressed basal TK-promotor activity, even in the presence of
a supramaximal concentration of T3 (5 µmol/L). This was
seen with all three TREs in both cell lines. In the case of PAL, the
repression was less pronounced, and in the absence of T3,
it was only half that of TRß-WT. The mutant TRß-EZ displayed a
strong dominant negative effect on wild-type activity (Fig. 4). High cotransfection ratios (5:1 and 10:1, TRß-EZ
to TRß-WT) resulted in pronounced increases in dominant negative
activity, in that wild-type transcriptional activity was reduced to
almost zero or basal TK-promotor activity was repressed. At these
cotransfection ratios, addition of 1 µmol/L or 5 µmol/L
T3 increased the transcriptional activity of the TRß
mixture only marginally, but interestingly, at a EZ:WT ratio of 1:1 in
COS-7 cells, the dominant negative activity of TRß-EZ acting via the
response elements, F2 and PAL, seen at 5 nmol/L T3 was
markedly reduced by increasing the T3 concentration to 1
µmol/L. This effect was more pronounced with F2 than with PAL. This
attenuation of dominant negative action is the more noteworthy, because
TRß-EZ, when transfected alone, was unable to act as a transactivator
via F2 and PAL, even in the presence of 5 µmol/L T3.
Attenuation of TRß-EZ’s dominant negative action was not seen using
DR4 in COS-7 cells, nor was it seen with all three TREs in CV-1
cells.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The clinical consequences of the presence of this TRß are both
unusual and interesting, in that the affected individual displayed both
hypothyroid and hyperthyroid features. The first notable feature is the
extreme degree to which the proband is affected. Both mentally and
physically, the patient is severely retarded, having an IQ less than 50
and being of short stature. Although mental and physical retardation
have been described before (1, 2, 3, 4), the degree to which our patient is
affected is similar to that described in congenital cretinism. In
contrast to this picture of retardation, bone age was advanced, and the
onset of puberty was relatively early (menarche at 11 yr). We are not
aware that the combination of short stature and advanced bone age has
been described in RTH previously. The patient was somewhat overweight
in relation to her height, and although this may be caused in part by
relatively high food intake, it is also a typical feature of
hypothyroidism. However, her serum cholesterol was low, which suggests
that the high circulating levels of thyroid hormone were overcoming the
effects of the variant receptor and inducing normal levels of those
lipolytic enzymes known to be regulated by thyroid hormone. Finally, in
the absence of cardiac failure, tachycardia is a recognized feature in
hyperthyroidism, indicating her relatively normal response of the heart
to the high circulating levels of T3 and T4.
This combination of hypo- and hyperthyroid features may reflect the
fact that transmission of thyroid hormone action is mediated by two
different TRs (
and ß). It is noteworthy that the TR has been
claimed to mediate action in the heart (29), and this organ exhibits
hyperthyroid features. Alternatively, the degree to which the dominant
negative properties of TRß-EZ may be overcome by the high circulating
T3 and T4 levels may be related to the
structure of the TRE-regulating individual genes or the presence or
absence of the various organ- or cell-specific cofactors that normally
interact with TRs.
Because of the severity of RTH in the patient described and the weight of evidence that RTH is the result of TRß-gene defects, genetic analysis of an unborn child was performed using RT-PCR and amniocyte mRNA as template to produce TRß cDNA. DNA sequencing of the RT-PCR product resulted in the exclusion of any TRß-gene defect in the fetus. The analysis was validated with the birth of a healthy infant who is developing normally. Our demonstration that prenatal genetic diagnosis in the case of extreme RTH is an option, opens up the possibility, in high risk families, for therapeutic intervention either in the prenatal period or immediately at birth. The prenatal approach, which has already been performed in the case of severe androgen and insulin resistance (30, 31), may be superior to postnatal treatment only (32).
The truncation of the carboxy-terminal end of the TRß resulted in the apparent total loss of T3-binding activity and was shown to have profound consequences for the receptor’s DNA binding and transcriptional properties. On the molecular level, the mutant EZ showed some features common with other RTH-associated mutant TRs but also some unusual characteristics. Compared with the large number of TRß mutants, the EZ mutant belongs to a small subgroup of mutant TRs that are virtually deficient in T3 binding (12, 20, 33, 34, 35, 36). Typically, most TRß mutants show only reduced T3 binding. They can be transcriptionally active in the presence of high T3 levels, and their dominant negative actions at low T3 concentrations can be attenuated by saturating concentrations of T3 (12). In contrast, the non-T3-binding variants have been shown to lack transactivating potential (12, 33, 35, 36) and, in the case of kindred S and Mf-1, the mutants even silence basal gene transcription over a broad T3-concentration range (14, 15). Secondly, their dominant negative activity could not be attenuated by supramaximal T3 levels (12, 15, 33, 35, 37, 38). We have demonstrated that the EZ mutant is a potent transcriptional silencer in the absence and in the presence of very high concentrations of T3, as is the case with the S or Mf-1 mutant (14, 15). Interestingly, a feature that makes the EZ mutant distinct from other non-T3-binding mutants is our demonstration of the partial negation by high concentrations of T3 of its dominant negative action on TRß-WT acting via response elements, F2 and PAL. This is the more remarkable because TRß-EZ did not transactivate from these TREs, even at supramaximal T3 levels, but even repressed basal transcription. The attenuation of dominant negative action occurred in vitro when the transfection ratio of EZ:WT was 1:1, the ratio that occurs in vivo. Attenuation was not observed with DR4, which is consistent with earlier studies showing the TRE- and receptor isoform-dependence of the dominant negativity (38, 39, 40, 41). Of interest also is the fact that the attenuation of dominant negativity was seen in one cell line only. This may result from the relative abundance in cells of a spectrum of often cited, but poorly defined, cell-specific TR-auxiliary proteins. Another unusual feature of TRß-EZ is the poor correlation between strength of TRß-dimer formation with different TREs and dominant negative action. In contrast, Meier et al. (40), Zhu et al. (41), and Wong et al. (42) showed the affinity binding of mutant TRß homodimers and mutant:wild-type heterodimer for a given TRE correlated with dominant negativity. They suggested that competition between transcriptional defective and transcriptional potent homo- and heterodimers for TREs is the basis of dominant negativity, a hypothesis our findings do not support. This is not to say that competition between mutant and wild-type homo- and heterodimer plays no part in dominant negativity, but that the exact nature of the interaction of the mutant TRß and other nuclear heterodimerization partners with the individual TRE is crucial. From the data presented here it can be seen that the T3-binding subdomain missing from this truncated receptor either plays a direct part in this interaction or has a profound influence on the architecture of the protein so that the interaction is dramatically altered.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received July 30, 1996.
Revised October 31, 1996.
Accepted December 9, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2 and peroxisome proliferator-activated
receptor: evidence for different mechanisms of inhibition. Mol Cell
Endocrinol. 107:55–66.[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  S. A. Phillips, P. Rotman-Pikielny, J. Lazar, S. Ando, P. Hauser, M. C. Skarulis, F. Brucker-Davis, and P. M. Yen Extreme Thyroid Hormone Resistance in a Patient with a Novel Truncated TR Mutant J. Clin. Endocrinol. Metab., November 1, 2001; 86(11): 5142 - 5147. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Asteria, O. Rajanayagam, T. N. Collingwood, L. Persani, R. Romoli, D. Mannavola, P. Zamperini, F. Buzi, F. Ciralli, V. K. K. Chatterjee, et al. Prenatal Diagnosis of Thyroid Hormone Resistance J. Clin. Endocrinol. Metab., February 1, 1999; 84(2): 405 - 410. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
