| |||||||||||||||||||||||||||||||
Hyperemesis gravidarum is characterized by excessive vomiting in early pregnancy, leading to the loss of 5 percent or more of body weight.4 It is usually self-limited and therefore of little clinical consequence.5,7 Some women with the disorder have high serum thyroid hormone concentrations, and a few have sufficient clinical manifestations of hyperthyroidism to warrant short-term treatment with antithyroid drugs.8,9 Many but not all women with hyperemesis gravidarum and hyperthyroidism have high serum chorionic gonadotropin concentrations, raising the possibility that other factors contribute to the hyperthyroidism.3,8,9
We describe a woman and her mother who had recurrent gestational hyperthyroidism and normal serum chorionic gonadotropin concentrations. Both women were heterozygous for a missense mutation in the extracellular domain of the thyrotropin receptor. The mutant receptor was more sensitive than the wild-type receptor to chorionic gonadotropin, thus accounting for the occurrence of hyperthyroidism despite the presence of normal chorionic gonadotropin concentrations.
Case Report
The proband was a 27-year-old woman who was 10 weeks' pregnant when referred for the evaluation and treatment of hyperthyroidism. This was her third pregnancy, the first and second having resulted in early miscarriage accompanied by severe nausea and vomiting. She again reported severe nausea and vomiting and had recently lost 5 kg in weight. Physical examination revealed persistent tachycardia (heart rate, 120 beats per minute), excessive sweating, and tremor of the hands. There was a small, diffuse goiter and no ophthalmopathy. Laboratory studies revealed the following values: serum thyrotropin concentration, <0.07 µU per milliliter (normal, 0.2 to 6); serum free thyroxine concentration, 4.7 ng per deciliter (60 pmol per liter; normal, 0.8 to 1.9 ng per deciliter [11 to 24 pmol per liter]); and serum triiodothyronine concentration, 605 ng per deciliter (9.77 nmol per liter; normal, 65 to 170 ng per deciliter [1.05 to 2.75 nmol per liter]). No antibodies to thyroid peroxidase or the thyrotropin receptor were detected in serum.
The patient was treated with 450 mg of propylthiouracil per day for eight weeks, and the dose was then tapered to 150 mg per day. Her condition improved rapidly, and she remained clinically and biochemically euthyroid for the rest of her pregnancy. She delivered a normal girl at 38 weeks of gestation, at which time the propylthiouracil was discontinued. The patient did not return for a follow-up examination post partum.
Eighteen months later, after seven weeks of amenorrhea, she was referred for a recurrence of hyperthyroidism associated with hyperemesis gravidarum. The findings on examination were similar to those during the previous pregnancy. The patient denied having any symptoms during the intervening period. The results of laboratory studies were as follows: serum thyrotropin concentration, <0.07 µU per milliliter; serum free thyroxine concentration, 4.5 ng per deciliter (58 pmol per liter); and serum chorionic gonadotropin, 70 U per milliliter (normal range for the first trimester of pregnancy, 38 to 173). Serum thyroid-stimulating activity was undetectable.
The patient was treated with 250 mg of propylthiouracil per day for four weeks, and because her serum thyrotropin concentrations remained low, the dose was tapered to between 50 and 150 mg per day for the rest of the pregnancy (Figure 1). She delivered a normal boy at 38 weeks of gestation, at which time the propylthiouracil was again discontinued. She was clinically and biochemically euthyroid when last examined four months after delivery.
|
The study was approved by the local ethics committee, and both women gave informed consent.
Methods
Hormone Assays
Serum free thyroxine was measured by radioimmunoassay (Sorin Biomedica, Anthony, France), and serum thyrotropin and chorionic gonadotropin were measured by chemiluminescence assays (ACS180, Chiron, Cergy Pontoise, France). Thyroid-stimulating activity in the serum was assayed in cultures of Chinese-hamster–ovary cells, stably transfected with the human thyrotropin receptor.10
Sequencing of the Thyrotropin-Receptor Gene
To determine the sequence of the thyrotropin-receptor gene, DNA was extracted from peripheral-blood leukocytes, and the sequences of all exons of the thyrotropin-receptor gene and the intron–exon junctions were determined as described previously.11
Functional Characterization of the Mutant
The mutation identified in the thyrotropin-receptor gene of the patient and her mother was introduced in wild-type thyrotropin-receptor complementary DNA (cDNA) by site-directed mutagenesis based on the polymerase chain reaction (PCR). In brief, an Xho I–Afl III restriction fragment encompassing the mutated site was ligated to wild-type thyrotropin-receptor cDNA that had been inserted in the expression vector pSVL (Pharmacia, Roosendaal, the Netherlands). The sequence of the segment containing the mutation was verified on both strands.
Plasmids encoding wild-type or mutant thyrotropin receptors and wild-type luteinizing hormone or follicle-stimulating hormone receptors were transfected into COS-7 cells by the DEAE–dextran method.11 Cell-surface expression of the wild-type and mutant receptors was assessed by a FACScan flow cytometer (Becton Dickinson, San Diego, Calif.) with a mouse monoclonal antibody (BA-8) that recognizes a conformational epitope of the extracellular domain of the human thyrotropin receptor.12
The production of cyclic AMP (cAMP) was measured in the cells at base line and after incubation for 60 minutes11 with bovine thyrotropin (Sigma, Bornem, Belgium), chorionic gonadotropin from the urine of pregnant women (Intervet, Chorulon, Turnhout, Belgium), and recombinant follicle-stimulating hormone (Organon, Brussels, Belgium).
Results
Sequence Determination
Direct sequencing of PCR products amplified from genomic DNA from the patient identified the substitution of guanine for adenine at codon 183 in exon 7 in one allele of the thyrotropin-receptor gene. This substitution results in the replacement of a lysine residue with an arginine (K183R) at position 183 of the receptor, which is in the middle of its extracellular N-terminal domain. The patient's mother was heterozygous for the same mutation. The mutation nullifies an Alu I restriction site, a characteristic that allows easy screening of the general population. None of 100 unrelated normal subjects carried the mutation, ruling out the possibility that it is a common polymorphism.
Functional Characterization of the Mutant Receptor
When transfected into COS-7 cells, the mutant cDNA was expressed at the cell surface at a concentration similar to that of the wild-type receptor (data not shown). Basal production of cAMP in cells expressing the mutant receptor was similar to that in cells expressing the wild-type receptor, as were the sensitivity and maximal responsiveness of the cells to bovine thyrotropin (Figure 2A). In contrast, chorionic gonadotropin in concentrations of up to 1000 U per milliliter caused a dose-dependent increase in cAMP production by cells containing the mutant receptor, but only a minimal increase in cells containing the wild-type receptor (Figure 2B). Comparison of the stimulation of cAMP production by chorionic gonadotropin in cells containing the wild-type luteinizing hormone receptor or the mutant thyrotropin receptor indicated that the mutant was about 1000 times less responsive (Figure 2C).
|
The results of assays of direct binding of radiolabeled chorionic gonadotropin or the displacement of radiolabeled thyrotropin by chorionic gonadotropin from the mutant receptors were negative (data not shown).
Discussion
The phenotype and genotype of our patient and her mother, together with the functional characteristics of the mutant thyrotropin receptor, led to a syndrome of hereditary gestational hyperthyroidism caused by hypersensitivity of the thyrotropin receptor to chorionic gonadotropin. The mutation is remarkable in that it broadens the ligand specificity of a G-protein–coupled receptor without altering sensitivity to the native ligand, thyrotropin.
The mechanism responsible for gestational hyperthyroidism in these women differs from that of the hyperthyroidism associated with molar pregnancies and, at least in some women, hyperemesis gravidarum. In women with the latter two conditions, hyperthyroidism results from the activation of the thyrotropin receptor by excessive quantities of chorionic gonadotropin or by chorionic gonadotropin molecules with increased thyrotropin-like activity.13,14,15,16 These two situations may be viewed as exaggerated forms of the thyroid stimulation that occurs at the time of maximal chorionic gonadotropin production in many normal pregnant women.1 In contrast, in our patient and her mother, both of whom were heterozygous for the K183R mutation, normal serum chorionic gonadotropin concentrations that would not stimulate the wild-type thyrotropin receptor excessively caused hyperthyroidism that was severe enough to necessitate antithyroid-drug therapy during pregnancy.
The relation between hyperemesis gravidarum and gestational hyperthyroidism is not clear. In some studies, high serum free thyroid hormone concentrations were found in 30 to 70 percent of women with hyperemesis gravidarum,5,7 but in other studies, few women had high values.17 Whether hyperemesis results from hyperthyroidism, from the hyperestrogenic state associated with hyperstimulation by chorionic gonadotropin, or from other mechanisms is not known.9,18,19,20 The finding of severe recurrent hyperemesis in the presence of normal serum chorionic gonadotropin concentrations in our patient suggests that hyperemesis is related to hyperthyroidism.
Considering its functional consequences, the mutation identified — the substitution of arginine for lysine at position 183 — is surprisingly conservative. This position is in a region of the receptor that constitutes the putative surface of interaction with thyrotropin.21,22 An arginine at position 183 may increase the stability of the illegitimate complex between chorionic gonadotropin and the thyrotropin receptor enough to cause signal transduction by the increased serum chorionic gonadotropin concentrations present in pregnant women.23 However, despite the fact that luteinizing hormone activates the wild-type thyrotropin receptor more easily than does chorionic gonadotropin,24 the serum luteinizing hormone concentrations after menopause would remain too low to cause activation of the mutant receptor. Indeed, the mother of our patient remained euthyroid after menopause. This finding is compatible with a relatively small gain of function in the mutant thyrotropin receptor in response to stimulation by chorionic gonadotropin, in agreement with our inability to detect binding of chorionic gonadotropin.
Unlike other mammals, primates rely on chorionic gonadotropin for the maintenance of the corpus luteum in early pregnancy.25 Our data, together with the fact that thyrotropin secretion is often partially suppressed during the period when chorionic gonadotropin concentrations are maximal, suggest that evolution has led to the selection of physiologic mechanisms that operate close to the border of hyperthyroidism during normal pregnancy.
Supported by funds from the European Union Program for Training and Mobility (to Dr. Rodien); by the Belgian Program of University Poles of Attraction, Service for Sciences, Technology and Culture; and by grants from the Fonds de la Recherche Scientifique Médicale and the Fonds National de la Recherche Scientifique.
We are indebted to Dr. T. Minegishi (Maebashi, Japan) for the cDNA encoding the receptors for luteinizing hormone, chorionic gonadotropin, and follicle-stimulating hormone; to Dr. C. Gervy for the bovine thyrotropin assays; to Ms. Muriel Nguyen and Mr. Claude Massart for expert technical assistance; to Professor Samuel Refetoff for his critical reading of the manuscript; and to Organon for the generous gift of recombinant follicle-stimulating hormone.
Source Information
From the Institut de Recherche Interdisciplinaire (P.R., J.P., J.V., S.C., G.V., L.D.) and Service de Génétique Médicale (J.P., G.V.), Faculté de Médecine, Université Libre de Bruxelles, Brussels, Belgium; the Service d'Endocrinologie Maladies Métaboliques, Hôpital Cochin, Paris (P.R., C.B., J.-P.L.); and the Centre National pour la Recherche Scientifique, Unité Propre de Recherche 1524, Paris (M.-L.R.S.).
Address reprint requests to Dr. Vassart at IRIBHN, Campus Erasme, 808 Route de Lennik, B-1070 Brussels, Belgium.
References
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
