| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Institute of Endocrine Sciences, Inc. (C.A., L.P., R.R., D.M., P.B.P.), University of Milan, Ospedale Maggiore IRCCS, Istituto Clinico Humanitas, Istituto Auxologico Italiano, IRCCS, and Department of Obstretics and Gynecology (P.Z.), Ospedale S. Paolo, Milan, Italy; Department of Medicine (O.R., T.N.C., V.K.K.C.), University of Cambridge, U.K., and Departments of Pediatrics (F.B.) and Neonatal Intensive Care (F.C.), University of Brescia, Italy
Address all correspondence and request for reprints to: Paolo Beck-Peccoz, M.D., Istituto Clinico Humanitas, Via Manzoni, 56, 20089-Rozzano-Milano, Italy; E-mail: paolo.beck-peccoz{at}humanitas.it
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
With regard to the treatment of RTH, it has been reported that the administration of TRIAC, a thyroid hormone analog, to patients presenting with symptoms and signs of hyperthyroidism is useful in inhibiting the secretion and biological activity of TSH with minimal thyromimetic effects at the level of peripheral tissues (2, 9, 13, 14). As a consequence, TRIAC is generally able to restore euthyroidism and to reduce goiter size in PRTH patients. Recently, it has been shown that TRIAC binds TRß1 with higher affinity than T3, selectively augmenting the function of this receptor (15). Furthermore, cotransfection experiments indicate that TRIAC is more effective than T3 in enhancing the function of mutant TRß1 or overcoming its dominant negative effect, thus providing further experimental support for its use in RTH patients (15).
Heretofore, no data have been available concerning the use of TRIAC during pregnancy in RTH patients. It is not known whether this drug, which crosses the placental barrier (16, 17), can adversely affect the development of a normal fetus or can be beneficial for the growth of an affected RTH fetus.
In this paper, we report the studies carried out in a PRTH patient harboring a novel TRß gene mutation who was successfully treated with TRIAC until the onset of pregnancy, after which therapy was discontinued to avoid potential side effects. However, the recurrence of hyperthyroidism in the patient after TRIAC withdrawal necessitated rapid reintroduction of the treatment and prompted prenatal diagnosis and monitoring of fetal thyroid status during pregnancy.
| Case Report |
|---|
|
|
|---|
Due to the presenting thyrotoxic features, the patient was successfully treated with TRIAC at a dose of 2.1 mg/day for 3 yr, up to the onset of pregnancy. At that time the therapy was discontinued, as TRIAC, which crosses the placental barrier, could have adversely affected the development of a normal fetus. However, after TRIAC withdrawal the patient developed recurrent hyperthyroidism, prompting a rapid fetal genotyping to determine whether it was also affected, enabling reinstitution of TRIAC therapy to restore maternal euthyroidism and for potentially beneficial effects in an affected RTH fetus.
A written informed consent to all procedures done was been obtained by the propositus, and the study was approved by our institutional ethical committee.
| Materials and Methods |
|---|
|
|
|---|
Maternal genomic DNA was extracted from peripheral blood leukocytes using standard techniques. Exon 9 of the TRß gene was amplified by PCR using flanking intronic primers (sense primer: 5'-AGT GAA TTC ACA GAA GGT TAT TCC TAT TGC-3'; antisense primer: 5'-GAT CTG CAG GCT CTT TGG ATG CCC ACT AAC-3'). PCR was performed using 1 µg genomic DNA and 10 pmol of each primer in 50 mL reaction volume containing 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 2 mM MgCl2, 500 mM dNTPs, and 0.5 U Taq polymerase (Perkin Elmer Cetus, Norwalk, CT). The reaction conditions were denaturation at 94 C for 3 min, then annealing at 55 C for 30 s, followed by 30 cycles of extension at 72 C for 30 s, denaturation at 94 C for 30 s, and annealing at 60 C for 60 s.
The PCR product was sequenced directly using an internal intronic primer (5'-GTA TGT TGT TCC TGA CTG GC-3') by dideoxy-nucleotide method (Thermosequenase-Amersham Kit Milan, Italy), according to the manufacturers instructions.
The presence of the mutation was verified by at least two independent PCR and sequencing reactions.
The prenatal diagnosis of RTH was performed at 17 weeks gestation on fetal DNA extracted by standard methods from chorionic villi. As the presence of the T337A mutation abolishes a restriction site for Mae III, normally present on the wild-type allele, PCR amplicon encompassing exon 9 of TRß was digested with this enzyme and analysed by agarose gel electrophoresis. The result was also confirmed by direct sequencing. Maternal and paternal DNA were used as controls.
Mutagenesis and plasmids
The T337A mutation was generated by site directed mutagenesis of the wild-type hTRß1 cDNA in M13 mp18 as previously described (18) and confirmed by sequencing of individual phage plaques. Both wild-type and mutant receptor cDNAs were subcloned into a vector downstream of the Rous sarcoma virus (RSV) enhancer and promoter for transient transfection assays. The reporter plasmid MAL-TKLUC contains a positively regulated thyroid response element (TRE) from the malic enzyme gene inserted upstream of the viral thymidine kinase gene promoter and luciferase cDNA. The internal control plasmid BOS-ßgal contains the elongation factor-1, a promoter driving expression of the ß-galactosidase gene (19).
Cell culture and transfection assay
JEG-3 cells were cultured in Optimem (Gibco BRL, Paisley, Scotland) containing 2% (vol/vol) fetal calf serum and 1% (vol/vol) Penicillin, Streptomycin, and Fungizone (Gibco BRL, Paisley, Scotland), and 18 h before transfection the medium was changed to Optimem with 2% resin-stripped fetal calf serum. Twenty-four well plates of cells were transfected by a 5 h exposure to calcium phosphate. Cells were exposed to 2 µg MAL-TK-LUC reporter plasmid, together with 100 ng wild-type or mutant receptor expression vector and 100 ng BOS-ßgal. After a further 36 h, with T3 as appropriate, the cells were lysed and extracts assayed for luciferase and ß-galactosidase activity. Luciferase activity was measured using an Autolumat LB 953 luminometer (Berthold, Stevenage, UK) and values normalized with ß-galactosidase activity to correct for transfection efficiency.
T3 binding assay
For the measurement of receptor hormone binding affinity, wild-type and mutant receptor proteins were synthesized from cDNA templates in pGEM7Z using the Promega Corp. TNT coupled transcription and translation system (Promega Corp., Southampton, UK). The ligand binding affinities of in vitro translated wild-type and mutant receptors was measured with 125I-T3 in a filter binding assay as previously described (20). Scatchard analyses were performed to generate affinity constants (Ka), and the results are the mean of three or more separate determinations.
Cordocentesis and fetal ultrasonography
Fetal blood was sampled from the placental insertion of the cord by a 22-gauge heparinized needle guided by ultrasonography (Ansaldo 450, equipped with Convex 5 MHz sector transducer) and under aseptic conditions using the technique first described by Daffos et al. (21). This method allows access to the fetal circulation and permits detailed study of fetal biochemistry and metabolism. The reliability of fetal sampling was assessed by analysis of red cell volume with a Coulter counter and confirmed by Kleihauer smears.
Ultrasound examinations were performed using an Ansaldo AU 4 sonograph with Convex 3.5 MHz probe.
Immunoradiometric Assay and measurement of TSH bioactivity
Circulating levels of fetal and maternal TSH, FT4 and FT3 were measured by means of ultrasensitive time-resolved and specific immunofluorimetric assay, using Delfia technology (Delfia® Pharmacia SpA, Milan, Italy).
Maternal and fetal circulating TSH was immunoconcentrated and bioassayed using Chinese Hamster Ovary cells expressing human TSH receptor (CHO-R), as previously described (22).
| Results |
|---|
|
|
|---|
Direct sequencing of the TRß gene in the mother indicated heterozygosity for a G to A substitution at nucleotide 1296 in exon 9. This corresponds to a threonine to alanine substitution at codon 337 in the predicted protein sequence.
The ligand-binding affinity of the mutant receptor protein was impaired. Its affinity constant (Ka) was 60% (1.32 x 1010 M-1) of the wild-type receptor, which bound T3 with an affinity of 2.2 x 1010 M-1.
The function of the T337A mutant receptor was tested by assaying its
ability to activate a reporter gene (MAL-TK-LUC) containing a positive
TRE in a hormone-dependent manner (Fig. 1
).
In comparison to the wild-type receptor, the T337A mutant exhibited
impaired function with a right-shifted activation profile, but at
saturating T3 concentrations (1000 nM) it
attained maximal activation comparable to the wild-type receptor.
|
|
As shown on the agarose gel depicted in Fig. 3
, digestion of maternal and fetal DNA with
Mae III produced two fragments of 229 and 184 bp from the wild-type
allele together with a 413 bp fragment from the mutant allele resistant
to digestion with the above mentioned enzyme. In contrast, paternal DNA
digestion showed only bands of 229 and 184 bp, corresponding to two
normal alleles.
|
Fetal biochemical and ultrasound studies during TRIAC therapy
As the mother exhibited a recurrence of thyrotoxic features and
the fetus was also affected by RTH, TRIAC therapy was reestablished at
20 weeks gestation in a dose of 2.1 mg/day (Table 1
). Four weeks later ultrasonography
showed a healthy fetus, normal for gestational age, without goiter. No
investigations were done at that time.
|
The maternal TSH ranged from 0.41 to 0.7 mU/L with serum FT4 levels of 10.3 to 12 pmol/L between 24 and 33 weeks of pregnancy.
The measurement of fetal TSH biological activity on CHO-R cells showed a very low ratio between the bioactivity and the immunoreactivity of circulating TSH (B/I: 1.1 ± 0.4) as compared with that found in age-matched normal fetuses (B/I: 12.7 ± 1.2).
Because of the unexpectedly high fetal TSH levels, the maternal TRIAC dose was increased to 3.5 mg/day to reduce the TSH concentrations and goiter size in the fetus. As the very raised TSH values and fetal goiter strongly suggested the presence of concomitant intrauterine primary hypothyroidism, another cordocentesis was performed at 33 weeks gestation, which showed TSH concentrations reduced by 50%, to 144 mU/L. Concomitant ultrasonography revealed a clear reduction in the size of the goiter (thyroid circumference: 55 mm; normal value 48.665.8 at 33 weeks gestation). However, during this cordocentesis, several acute life-threatening fetal complications occurred, necessitating a prompt cesarean section at 33 weeks gestation.
At birth, the female newborn exhibited a small goiter and was
critically ill with multiple-organ failure encompassing brain, heart,
and lung. The Apgar score was very low (1 at 1 min and 3 at 5 min), and
the most important clinical problem was the respiratory insufficiency
due to respiratory distress syndrome, which required continuous
positive airway pressure ventilation. The biochemical picture in the
newborn was characterized by persistently elevated TSH concentrations
(107.5 mU/L) together with free thyroid hormone levels in the
hypothyroid range (FT4: 3.2 pmol/L; FT3: 3.5
pmol/L) (Table 1
).
Consequently, L-T4 treatment at a dose of 10 µg/kg/day was started.
After two months of treatment, the neonate showed high FT4
and FT3 levels in the presence of normal TSH values (2.0
mU/L) and was clinically euthyroid (Table 1
). L-T4 therapy was
therefore discontinued, and a month later the biochemical picture was
totally compatible with RTH. At present, the 2-yr-old baby is
euthyroid, without goiter, and has a normal stature and normal mental
development.
| Discussion |
|---|
|
|
|---|

310 to 349) in the ligand binding domain of
TRß. Functional studies indicate that this mutant binds T3 with a
lower affinity than wild-type receptor, is transcriptionally impaired
with a right-shifted activation profile, and inhibits wild-type
receptor function in a dominant negative-manner. Thus, properties of
the T337A mutant are analogous to those of the others RTH mutants
described hitherto, indicating that it is casually linked to the RTH
phenotype in this family. However, although both the mother and child
harbored the same receptor defect, they exhibited divergent clinical
features, with hyperthyroidism in the mother and euthyroidism in the
neonate, underscoring the variable clinical phenotype in this
disorder. Although RTH has been well characterized clinically and biochemically in the adult population, no cases of RTH have been studied during fetal life, and there are very few examples of neonatal diagnosis of the disorder reported. In the few neonates studied heretofore, the biochemical picture seen at birth has appeared similar to that found in adulthood. Thus, RTH newborns can present with either borderline elevated (11) or high TSH concentrations (12) associated with increased total T4 levels. However, it should be noted that this biochemical finding is not specific for RTH, as it is also seen in other clinical conditions, such as inherited TBG excess or familial dysalbuminemic hyperthyroxinemia. Accordingly, further investigations, such as the measurement of serum TBG or free T4 concentrations are required to differentiate RTH from other conditions. In another RTH case, TSH concentrations measured at birth were in the normal range (11), and total T4 was not measured, highlighting the need to measure both these parameters to diagnose the disorder. On the basis of these observations it is clear that the current screening programs for the congenital hypothyroidism are not appropriate for early detection of RTH, as they usually rely on the measurement of only TSH on dried blood spot samples. On the other hand, the low frequency of the disorder and the rarity of clinical complications in RTH cases in the neonatal period suggest that there is no indication for prenatal or early diagnosis of the disease except in the case of a homozygous (25) or a compound heterozygous TRß defect. In our case the need to treat maternal hyperthyroidism prompted prenatal diagnosis of the disease.
As the fetus showed growth retardation and goiter from 29 weeks gestation, fetal thyroid function and growth were monitored by measuring TSH and free thyroid hormone concentrations by cordocentesis and ultrasound examinations, respectively. Testing performed during maternal TRIAC administration showed very high circulating TSH levels (287 mU/L), compatible with primary hypothyroidism (PH). Markedly reduced TSH bioactivity (B/I:1.1 ± 0.4 vs. 12.7 ± 1.2, P < 0.001) was also observed, consistent with this diagnosis (26). It is unlikely that these features were caused by TRIAC therapy as, in congenital hypothyroid fetuses, we demonstrated that TRIAC administration to the mother was capable to reduce the TSH hypersecretion and the size of fetal goiter (17). The transplacental passage of maternal TSH receptor blocking antibodies can be also excluded the presence of fetal goiter.
On the basis of these findings, the possible coexistence of RTH and PH in the fetus was suspected. Such an association could have had dramatic consequences on fetal development, expecially at a neurological level, and thus required a close monitoring with further cordocentesis. However, at postnatal follow-up the biochemical picture altered to that typical of RTH alone. Accordingly, high intrauterine TSH levels with reduced bioactivity may be a feature of RTH per se. The TSH hypersecretion in RTH fetuses may be explained by impaired feedback action of thyroid hormone due to the receptor defect, resulting in low thyroid hormone circulating levels, as in fetuses with congenital hypothyroidism (27) or maternal hypothyroidism induced by PTU administration (28, 29). The decreased TSH bioactivity may be due to impaired thyroid hormone dependent regulation of genes coding for glycosyltransferases involved in the posttranslational modification of pituitary TSH during fetal life (30, 31). In addition, maternal TRIAC administration may also have played a role in reducing TSH bioactivity by modulating the glycosylation of terminal sugar residues (32).
Our case also illustrates the effectiveness of TRIAC administration in both the mother and the fetus. Thus, this agent was able to both restore euthyroidism by attenuating maternal thyrotoxic symptoms and also inhibit fetal TSH secretion by 50%, consequently reducing the fetal goiter markedly. Moreover, TRIAC did not alter fetal heart rate, which was normal before and during treatment.
A surprising finding was the transient hypothyroidism observed in our patient at birth. This may have been related to the prematurity (33 weeks gestation) of the infant. The neonatal hypothyroxinemia might be related to a "sick euthyroid" state associated with impaired oxygenation and metabolism caused by the respiratory distress syndrome. The prompt initiation of L-T4 treatment postnatally was aimed at restoring euthyroidism rapidly to avert adverse sequelae including mental retardation (33, 34). After later L-T4 withdrawal, the patient remained euthyroid, confirming that the hypothyroidism was transient, and the biochemical picture became consistent with RTH alone.
In summary, we have described a novel mutation in the TRß gene associated with the divergent phenotypes of RTH with hyperthyroid features in the mother and intrauterine hypothyroidism in the fetus, and we have confirmed the feasibility of prenatal diagnosis of RTH on chorionic villus sampling using molecular genetic techniques, according to previous data (35).
Finally, our observations suggest a biochemical phenotype of transient hypothyroidism in RTH during fetal development. Although further studies in a larger series of RTH cases in pregnancy are necessary to determine whether high circulating TSH levels with markedly reduced bioactivity are common findings in RTH fetuses, prenatal diagnosis and antenatal TRIAC therapy in this case may have prevented fetal thyrotrope hyperplasia, reduced the fetal goiter, and assured the euthyroid state in the mother during pregnancy.
| Acknowledgments |
|---|
Received July 15, 1998.
Revised October 5, 1998.
Accepted October 22, 1998.
| References |
|---|
|
|
|---|
gene by
thyroid hormone: receptor interaction adjacent to the TATA box. Proc
Natl Acad Sci USA. 86:91149118.
-mannosidase-II messenger
ribonucleic acid levels increase with different kinetics in thyrotrophs
of hypothyroid mice. Endocrinology. 135:19801985.[Abstract]
This article has been cited by other articles:
![]() |
B. A. Bauer, P. L. Elkin, D. Erickson, G. G. Klee, and M. D. Brennan Symptomatic Hyperthyroidism in a Patient Taking the Dietary Supplement Tiratricol Mayo Clin. Proc., June 1, 2002; 77(6): 587 - 590. [Abstract] [PDF] |
||||
![]() |
G. J. Kahaly, C. H. Matthews, S. Mohr-Kahaly, C. A. Richards, and V. K. K. Chatterjee Cardiac Involvement in Thyroid Hormone Resistance J. Clin. Endocrinol. Metab., January 1, 2002; 87(1): 204 - 212. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Anselmo, T. Kay, K. Dennis, R. Szmulewitz, S. Refetoff, and R. E. Weiss Resistance to Thyroid Hormone Does Not Abrogate the Transient Thyrotoxicosis Associated with Gestation: Report of a Case J. Clin. Endocrinol. Metab., September 1, 2001; 86(9): 4273 - 4275. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. E. Weiss and S. Refetoff Editorial: Treatment of Resistance to Thyroid Hormone--Primum Non Nocere J. Clin. Endocrinol. Metab., February 1, 1999; 84(2): 401 - 404. [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 |