2 Thyroid hormone resistance

2 Thyroid hormone resistance

2 Thyroid hormone resistance V. K R I S H N A K. C H A T T E R J E E PAOLO BECK-PECCOZ A major characteristic of thyroid hormones (thyroxine (T4), t...

1MB Sizes 2 Downloads 32 Views

2 Thyroid hormone resistance V. K R I S H N A K. C H A T T E R J E E PAOLO


A major characteristic of thyroid hormones (thyroxine (T4), triiodothyronine (T3)) is the multiplicity of cellular functions they regulate in virtually every type of tissue. The diverse responses to thyroid hormone include regulation of growth, metabolic processes, muscular activity and myocardial contractility. Thyroid hormone is also required for developmental processes such as functional differentiation of the central nervous system and metamorphosis in amphibia (Chatterjee and Tata, 1992). The synthesis of thyroid hormones is controlled by thyroid-stimulating hormone (TSH) from the pituitary and, in turn, T4 and T3 regulate TSH production as part of a classical negative-feedback loop. The syndromes of resistance to thyroid hormone (RTH) are rare disorders characterized by reduced targettissue responsiveness to circulating free thyroid hormones. The hallmark of RTH is resistance to thyroid hormone action in the pituitary-thyroid axis, such that continued TSH production drives hypersecretion of T4 and T3 to establish a new equilibrium with high serum levels of free thyroid hormones together with a non-suppressed TSH. On this background, affected individuals exhibit a variable degree of resistance in peripheral tissues, so that two major forms of the disorder are recognized. Many individuals with abnormal thyroid function tests are asymptomatic, leading to a diagnosis of generalized resistance (GRTH). In other cases, thyroid dysfunction is associated with a variety of thyrotoxic features, suggesting a selective pituitary defect (PRTH). The pleiotropic effects of thyroid hormone on physiological processes are accompanied and mediated by its actions on key target genes in different tissues. Thus, the feedback effects of thyroid hormones on TSH production are mediated by inhibition of TSH-~- and -[3-subunit gene expression (Franklyn and Sheppard, 1988). Target genes that are induced by thyroid hormone include malic enzyme and sex hormone-binding globulin (SHBG) in the liver, myosin heavy chain and sodium-calcium ATPase in myocardium, myelin basic protein in brain and sodium-potassium ATPase in skeletal muscle. The transcriptional regulation of these genes by thyroid hormone is now known to be mediated by a nuclear receptor protein which is a member of the steroid receptor superfamily (Chin, 1991). The receptor contains a highly conserved central 'zinc finger' domain which mediates Baillibre's ClinicalEndocrinology and Metabolism-267 VoL 8, No. 2, April 1994 Copyright © 1994, by Bailli~re Tindall ISBN 0-7020-1854-6 All rights of reproduction in any form reserved



binding to specific regulatory DNA sequences or thyroid response elements (TREs), usually located in the promoter regions of target genes. Many TREs consist of a direct repeat of the hexanucleotide motif A G G T C A and the thyroid hormone receptor interacts with these binding sites as a heterodimer with the retinoid X receptor (RXR). Some target gene TREs consist of this motif arranged as an everted repeat and homodimeric receptor-DNA interactions have also been demonstrated in these cases (Lazar, 1993). The carboxy-terminal domain of the receptor mediates hormone binding, which in turn leads to activation or repression of target gene transcription. In humans, two highly homologous thyroid hormone receptors, denoted hTR-~ and hTR-[~, are encoded by separate genes on chromosomes 17 and 3 respectively. Alternate splicing of each gene generates two major receptor isoforms hTR-al and hTR-[31, which are ubiquitously but differentially expressed. In the rat and mouse, a third isoform (TR-[32) is also produced by alternate splicing of the ~3gene. This receptor variant is selectively expressed in the pituitary and hypothalamus (Lazar, 1993) and is thought to have a human counterpart with a similar tissue distribution. Following the cloning of thyroid hormone receptors, familial GRTH was shown to be tightly linked to the hTR-[3 gene locus (Usala et al, 1988). Since then a number of groups have identified mutations in the thyroid hormone receptor [3 gene in a large number of patients with this disorder (Refetoff et al, 1993). This chapter reviews the differential diagnosis of RTH, the features of generalized versus pituitary resistance and the genetic defects that have been described in both forms of the disorder. Current concepts relating to the pathogenesis of the disorder and the implications for optimal treatment of these cases are discussed. DIFFERENTIAL DIAGNOSIS

The characteristic biochemical abnormality in R T H is elevated serum thyroid hormone (T4 and T3) concentrations, together with measurable TSH levels. The advent of sensitive immunometric TSH assays has enabled a clear distinction between patients with suppressed versus non-suppressed TSH levels, allowing a diagnosis of 'inappropriate TSH secretion' to be made more easily. However, as shown in Table 1, a variety of different conditions may be associated with hyperthyroxinaemia and detectable serum TSH concentrations. An increase in serum binding proteins is a common cause of elevated total T4 levels and can be excluded by measuring free thyroid hormones (FT4, FT3) using appropriate methods (Ekins, 1990). Familial dysalbuminaemic hyperthyroxinaemia is a disorder associated with a qualitatively abnormal serum binding protein. The use of direct 'two-step' or 'one-step' labelled monoclonal antibody techniques to measure FT4 overcomes the interference of abnormal forms of albumin with 'analogue' assay methods (Wood et al, 1987). In addition, both total and free T3 levels are usually normal in patients with familial dysalbuminaemic hyperthyroxinaemia, which can be useful in differentiating this disorder from RTH. Other causes of hyperthyroxinaemia with low T3 levels include the neonatal



Table 1. Hyperthryoxinaemia and inappropriate

serum TSH levels. Raised serum binding proteins Familial dysalbuminaemia Anti-iodothyronine antibodies Heterophilic antibodies Anti-TSH antibodies Non-thyroidal illness Acute psychiatric disorders Neonatal period Drugs (amiodarone) Thyroxine replacement therapy TSH-secreting pituitary turnouts Resistance to thyroid hormones

period, systemic illness, acute psychiatric disorders and the effects of various drugs. In these situations, the differential diagnosis rests on recognition of the abnormal clinical context as well as documenting subsequent normalization of thyroid function test results at a later stage, or following recovery or drug withdrawal. Hyperthyroxinaemia with normal TSH levels is also a feature of L-T4 replacement therapy (Pearce and Himsworth, 1984), but the administration of higher doses of T4 or T3 will inhibit TSH secretion. The presence of circulating antibodies that interfere with the measurement of thyroid hormones or TSH may also produce biochemical abnormalities that mimic RTH. Approximately 8% of patients with thyroid disease may have circulating anti-iodothyronine antibodies, which may lead to spuriously high total and free thyroid hormone levels (Sakata et al, 1985). This interference can again be overcome by measurement of free thyroid hormones using two-step or one-step as opposed to analogue assays (BeckPeccoz et al, 1984; Crin6 et al, 1992). Circulating heterophilic antibodies directed against mouse immunoglobulin (mIg) interfere with monoclonal antibodies in TSH assays. The interfering antibodies can be sequestered by the addition of exogenous mIg to the assay buffer. In contrast, assay artefacts associated with endogenous anti-TSH antibodies are unavoidable but can be detected by showing a diminished recovery of TSH standard which has been incubated with the patient's serum. However, these antibodies usually lead to an underestimation of TSH levels and more infrequently overestimation (Beck-Peccoz et al, 1985). Hyperthyroxinaemia, non-suppressed TSH levels and thyrotoxic symptoms are often features of RTH, but are also associated with inappropriate TSH secretion from a pituitary tumour, sometimes leading to diagnostic difficulties. Table 2 compares the salient features of both disorders, and shows that there are no significant differences in age, sex, FT4, FT3 or TSH levels between the two groups of patients. The presence of a goitre or thyrotoxic symptoms has often led to inappropriate surgical or 131I thyroid ablation in either group. However, the following investigations can be helpful in differentiating the two conditions. The presence of abnormal thyroid function test results in other family members suggests a familial disorder such as RTH. Dynamic testing of the pituitary-thyroid axis can be


V. K. K. CHATTERJEEAND P. BECK-PECCOZ Table 2. Comparison of features in resistance to thyroid hormone (RTH) versus TSH-secreting tumours. Feature* Age (years) Sex ratio (F:M) Familial cases (%) Previous thyroid ablation (%) CT scan or MRI lesions (%) Mean FT4 level (9-18pmol/1)~Mean FT3 level (3-8 pmol/l)t Mean TSH level (0.4-4 mUff)t Normal a-subunit :TSH molar ratio (%) Elevated SHBG (%) Normal or exaggerated TSH response to TRH (%) Inhibition of TSH secretion following T3 (%):~



0.1-80 207:177 78 49 2.3 31.2 +_+7.7 _ 11.8 _+3.0 1.8 + 1.1 98

11-84 84:63 0 67 97.2 35,2 + 5.1 13.5 + 4.0 3.0 + 4.0 8.6

0 96

94.6 9.6



* Includes both GRTH and PRTH cases. t Normal ranges for hormone data in parentheses. $ Werner's test (80-100p~gT3 for 8-10 days). Completesuppression of TSH has not been observed in either group. useful in determining the source of inappropriate TSH secretion. The majority of patients with autonomous TSH-secreting tumours show no response to thyrotrophin-releasing hormone (TRH), whereas in R T H physiological responses are preserved with a normal or exaggerated rise in T S H levels (Weintraub et al, 1981; Faglia et al, 1987). Likewise, the administration of T3 leads to a suppression (albeit incomplete) of basal or TRH-stimulated TSH secretion in R T H , whereas T S H secretion from a tumour remains refractory to inhibition. Imaging of the pituitary fossa by computed tomography or magnetic resonance imaging usually confirms the presence of a tumour, but appearances can sometimes be equivocal, as with a partially empty sella documented in one case of R T H (Mariotti et al, 1987). In these instances, measurement of serum glycoprotein h o r m o n e a-subunit levels and an elevated a-subunit : TSH molar ratio are indicative of a TSH-secreting tumour (Beck-Peccoz et al, 1992). Lastly, we have observed that serum S H B G levels can be a useful discriminant, being almost invariably normal in patients with R T H but often elevated into the thyrotoxic range in those with TSH-secreting tumours (Table 2).

C L I N I C A L FEATURES Resistance to thyroid hormone was first described in 1967 in two siblings who were euthyroid despite high circulating thyroid hormone levels and who exhibited a number of other abnormalities including deaf-mutism, stippled femoral epiphyses with delayed bone maturation and short stature, as well as dysmorphic facies, winging of the scapulae and pectus carinatum (Refetoff et al, 1967). It is now clear that some of these features are unique to this



kindred, in which the disorder was recessively inherited. The majority of RTH cases that have been described since then are dominantly inherited with highly variable clinical features. Many patients with RTH are either asymptomatic or have non-specific symptoms and are noted to have a goitre, prompting thyroid function testing which makes the diagnosis. In this context, we have recently shown that the biological activity of circulating TSH is elevated in RTH and suggest that this factor may account for both thyroid hormone hypersecretion and goitre in individuals with normal immunoreactive TSH levels (Persani et al, 1994). Attempts to treat the biochemical abnormality with surgery or radio-iodine are often unsuccessful, with recrudescence of the goitre and thyroid dysfunction (Refetoff et al, 1983). In these individuals, classified as exhibiting generalized resistance (GRTH), the high thyroid hormone levels are thought to compensate for ubiquitous tissue resistance, producing a euthyroid state. In contrast, a number of individuals with the same biochemical abnormalities exhibit signs and symptoms associated with thyrotoxicosis; in adults these can include weight loss, tremor, palpitations, insomnia and heat intolerance; in children, failure to thrive, accelerated growth and hyperkinetic behaviour have also been noted. When this clinical entity was first described, patients were thought to have 'selective' pituitary resistance to thyroid hormone action (PRTH) with preservation of normal hormonal responses in peripheral tissues (Gershengorn and Weintraub, 1975). Finally, features of hypothyroidism such as growth retardation and a delayed bone age in children or hypercholesterolaemia in adults have also been recorded and may coexist with thyrotoxic symptoms in the same individual (Magner et al, 1986). However, we have compared the clinical and biochemical characteristics of individuals classified as having GRTH or PRTH and find that there is a wide overlap between these entities (Table 3). There were no differences in age, sex ratio, frequency of goitre or levels of FT4, FT3 or TSH in patients with the two types of disorder. Significantly, features such as tachycardia, hyperkinetic behaviour and emotional disturbance were documented in individuals with either GRTH or PRTH. Normal serum SHBG levels, indicating hepatic resistance to thyroid hormone action, were found in both groups. Another factor that can confound the assessment and classification of these patients is a temporal variation in clinical features. In two cases of PRTH, affected individuals exhibited thyrotoxic symptoms and signs that varied spontaneously over several years (Beck-Peccoz and Chatterjee, 1994). Overall, these observations indicate that whilst all patients with RTH exhibit abnormal thyroid function test results, the clinical presentation varies widely both between and within individuals. Nevertheless, the absence or presence of overt thyrotoxic features allows patients to be classified as having either GRTH or PRTH, and this clinical distinction will remain useful as a guide to the most appropriate therapeutic modality (see below). Two recent studies have documented neuropsychological abnormalities in a large number of patients with RTH. First, a history of attention-deficit hyperactivity disorder in childhood was elicited more frequently in patients


V. K. K . C H A T r E R J E E A N D P. B E C K - P E C C O Z

Table 3. Clinical and biochemical features of patients with generalized (GRTH) and pituitary (PRTH) resistance to thyroid hormones (literature and authors' unpublished cases). Feature Age (years) Sex ratio (F:M) Previous thyroid ablation (%) Goitre (%) Tachycardia (%) Emotional disturbance (%) Hyperkinetic behaviour (%) TSH (mU/1)* TSH B :I ratio~: FT4 (pmol/1)§ FF3 (pmol/1)[I SHBG (nmol/1)¶



0.1-75 161:151 45 95 75 60 68 1.6 +- 1.07 4.0 + 1.27 30.5 + 6.17 11.5+3.07 58.5 _+35.7?

1.3-80 46:26 69 96 94 84 88 2.2 +-1.3 4.7 +__2.0 32.8 +-11.1 12.3+2.7 37.1 +_20.7

*TSH levels in patients with GRTH and PRTH were not significantly different from those in unaffected relatives (1.8 + 1.1 mU/l: P > 0.4). ~No significant differences between values observed in patients with GRTH and those with PRTH. :~TSH B :I is the ratio of bioactive to immunoreactive TSH; the ratio was significantly higher in patients with GRTH and PRTH than in normal controls (1.3 +_0.2; P<0.01). §VF4 levels were significantly higher in patients with GRTH and PRTH than in unaffected relatives (11.9 +_2.0 pmol/1; P<0.001). lIFT3 levels were significantly higher in patients with GRTH and PRTH than in unaffected relatives (5.3 + 1.0 pmol/1; P<0.001). ¶ SHBG levels in patients with GRTH and PRTH were no different from values in unaffected relatives (39.1 + 22.0 nmol/1; P > 0.6). with G R T H c o m p a r e d with their unaffected relatives ( H a u s e r et al, 1993). A l t h o u g h this is an i m p o r t a n t association, it is unlikely that a substantial p r o p o r t i o n of unselected children with this disorder will also have R T H . A s e c o n d study s h o w e d that both children and adults with R T H exhibited p r o b l e m s with language d e v e l o p m e n t manifested by p o o r reading skills, dyslexia and p r o b l e m s with articulation (Mixson et al, 1992).

ASSESSMENT OF RESISTANCE TO HORMONE ACTION In addition to s y m p t o m s and signs, the m e a s u r e m e n t o f indices of thyroid h o r m o n e action is of potential use in evaluating the differing responses of various target organs and tissues to elevated thyroid h o r m o n e levels (Table 4). A l t h o u g h these m e a s u r e m e n t s are most useful in assessing the effects of m a r k e d thyroid h o r m o n e excess states such as overt h y p e r t h y r o i d i s m , they m a y be less discriminatory in individuals with borderline thyroid dysfunction, or with s o m e exceptions (e.g. b o n e age, systolic time intervals, basal metabolic rate, C P K ) in hypothyroidism. In our experience, m e a s u r e m e n t s such as systolic time intervals, basal metabolic rate, ankle jerk relaxation time, s e r u m S H B G (Beck-Peccoz et al, 1990), osteocalcin (Faber et al, 1990), soluble interleukin 2 receptor levels (Mariotti et al, 1992) and urinary



Table 4. Indices of thyroid hormone action. Tissue or organ


Pituitary Liver Bone

TSH Cholesterol, sex hormone-bindingglobulin, ferritin Osteocalcin, bone alkalinephosphatase, urinary pyridiniumcross-links, urinary hydroxyproline,bone age Carnitine, creatinine, Na÷-K+-ATPase activity, ankle jerk relaxation time Sleeping pulse rate, pulse wave arrival time (Qk,~),systolictime intervals (PEP/LVET) Red blood cell Na+ content, RBC Na+-K+-ATPase, thrombomodulin, soluble interleukin2 receptor Full-scale IQ, hyperactivityscore, verbal performance Basal metabolic rate, angiotensin-convertingenzyme, urea, creatinine phosphokinase

Musculoskeletal Heart Haemopoietic Brain Multiple

pyridinium cross-links (Harvey et al, 1991) are the most reliable indices. Some of these indices are in the hyperthyroid range, particularly in patients with PRTH, confirming that some peripheral tissues are not refractory to thyroid hormone action. Conversely, serum SHBG levels are invariably in the normal range in PRTH, indicating that this disorder is not associated with selective pituitary resistance (Beck-Peccoz et al, 1990). Finally, to improve the sensitivity and specificity of these indices, it has been suggested that patients with R T H be assessed following the administration of graded supraphysiological doses of T3 (50, 100 and 200 txg/day, each given for a period of 3 days) with comparison of any change in indices to values obtained in normal subjects (Sarne et al, 1988).

MOLECULAR GENETICS Although a small number of sporadic cases have been documented (Parrilla et al, 1991; V.K.K. Chatterjee and P. Beck-Peccoz, unpublished data), in the majority of families with R T H the disorder is dominantly inherited. Consonant with this mode of inheritance, many groups have reported that affected individuals are heterozygous for mutations in the TR-[3 gene. An analysis of the published literature (Refetoff et al, 1993), together with our own unpublished series, shows that 26 different mutations, including 20 point mutations, three in-frame deletions of single codons and three frameshift insertions have been documented to date. All the mutations localize to the hormone-binding domain of the receptor (Figure 1). Consequently, the ability of in-vitro synthesized mutant proteins to bind T3 is moderately or markedly reduced and their ability to activate or repress target gene expression is impaired (Chatterjee et al, 1991; Meier et al, 1992; Adams et al, 1994). However, in the recessively inherited form of R T H first described, affected individuals were found to be homozygous for a complete deletion of both alleles of the TR-[3 receptor gene (Takeda et al, 1992). Significantly, the obligate heterozygotes in this family, harbouring a deletion of one TR-[3 allele, were completely normal with no evidence of




234 264




o **o~o o



IDNA Bindin¢ Domain





o • o o****"

***, ,



I..I I - I - I ~ . , ~



Heptad repeats

** .

o *

Hormone Binding Domain

Figure 1. Schematic representation of the functional domains of hTR-[31, illustrating the clustering of most resistance mutations within two regions. The codon nomenclature is based on a predicted protein sequence containing 461 residues (Sakurai et al, 1990a). Asterisks (*) and open circles (o) denote point mutations described in patients with GRTH and PRTH respectively. In-frame codon deletions in patients with GRTH are indicated by triangles (A) and in those with PRTH by filled circles (.). Arrows indicate insertion mutations which shift the reading frame. Two regions involved in dimerization are shown: the Trap domain (residues 286--305) mediates receptor interaction with auxiliary proteins including RXR (O'Donnell and Koenig, 1990), and a series of hydrophobic heptad repeats (residues 334-428) has also been implicated in protein-protein interactions (Forman and Samuels, 1990). However, the overlap of one cluster of mutations (310-349) with the first and second heptad repeats suggests that they may not be critical for dimerization.

thyroid dysfunction (Table 5). This suggested that mere deficiency of functional [3-receptor as a consequence of a single deleted TR-[3 allele was insufficient to generate the resistance phenotype. Accordingly, we and others put forward the hypothesis that the mutant receptors in dominantly inherited RTH were not simply functionally impaired but also capable of inhibiting wild-type receptor action. Indeed, in-vitro experiments indicate that, when co-expressed, the mutant proteins are able to inhibit the function of their wild-type counterparts in a 'dominant negative' manner (Sakurai et al, 1990b; Chatterjee et al, 1991). Further clinical and genetic evidence to support this hypothesis is provided by a single case in which severe resistance was associated with marked developmental delay and growth retardation (Ono et al, 1991). This individual was homozygous for a mutation in both Table 5. Biochemical and clinical features in individuals homozygous or heterozygous for mutations ([3m) or deletions ([3°) in the TR-[3 gene with two, presumably normal, TR-~ alleles. Genotype a/a




t?,°/l?, °



13~/1~ ~


Failure to thrive Tachycardia Growth retardation Developmental delay

220 1.9

653 389

Clinical phenotype Deaf-mutism Normal Stippled epiphyses Dysmorphic Thyroid function* T4 (nmol/1) TSH (mU/1)

344 1,6

128 1.8

*Normal range: T4, 64-154 nmol/1; TSH, 0.5-4.0 mU/l.



alleles of the TR-[3 gene, and the extreme phenotype presumably reflects the inhibitory effects of two dominant negative mutant [3-receptos (Table 5) (Usala et al, 1991). In X-linked androgen insensitivity syndrome, deleterious mutations have been described throughout the androgen receptor (see Chapter 8), whilst recessively inherited vitamin D-resistant rickets is characterized by DNAbinding domain mutations in the majority of cases (see Chapter 4). In contrast, with two exceptions, all the mutations described hitherto in RTH cluster within two areas of the hormone-binding domain and lie outside regions that are important for DNA binding and dimerization (Figure 1) (Parrilla et al, 1991). Our recent analyses of 20 different mutant receptors indicate that, although they are functionally impaired, their ability to bind DNA, to form heterodimers with RXR and to exert a dominant negative effect on positively and negatively regulated target genes is preserved (Adams et al, 1994). Conversely, others have shown that the introduction of additional artificial mutations that abolish DNA binding or heterodimer formation abrogates the dominant negative potential of mutant receptors (Nagaya et al, 1992; Nagaya and Jameson, 1993). These observations suggest that receptor mutants with impaired transcriptional function, but normal DNA binding and dimerization properties, retain dominant negative potential leading to resistance to hormone action (Figure 2). As a corollary it is tempting to speculate that mutations elsewhere in the receptor elude discovery because they lack dominant negative activity and are therefore clinically and biochemically silent. However, such structure-function correlations might not be the sole determinants of the clustered distribution of receptor mutations in RTH. It has also been observed that mutations are non-uniformly distributed within the two major clusters, such that some




Figure 2. Possible mechanisms for dominant negative inhibition of wild-type (WT) receptor action. TREs in some target genes (left) interact with T R - R X R heterodimers and, here, functionally impaired mutant receptor-RXR complexes compete with wild-type receptorRXR complexes for DNA binding. TREs from other target genes (right) bind TR-TR homodimers in the absence of ligand which dissociate in the presence of hormone. Here, mutant receptor homodimers with impaired T3-binding properties fail to dissociate, occluding the binding site (Yen et al, 1992).


V. K, K. C H A T T E R J E E A N D P. B E C K - P E C C O Z

codon changes (e.g. arginine to tryptophan at codon 338; arginine to histidine at codon 438) are particularly frequent (Weiss et al, 1993a). These mutations represent transitions in CpG dinucleotides that are known to be frequent sites of point mutation in several other genes, suggesting that the concurrence of CpG dinucleotides within a cluster leads to the overrepresentation of certain codon changes. Based on the supposition that PRTH was associated with selective pituitary resistance, it had been hypothesized that this disorder could be associated with defects in the pituitary type II 5'-deiodinase enzyme or the TR-[32 receptor isoform (Franklyn, 1991). However, two recent case reports have documented TR-[3 mutations in PRTH (Mixson et al, 1993a; Sasaki et al, 1993), and we have extended these observations and identified further mutations in a number of other PRTH cases (Chatterjee et al, 1993). Some of the mutations observed in individuals with PRTH have also been reported in patients with G R T H from unrelated kindreds. Furthermore, we have found that, even within a family, the same receptor mutation may be associated with abnormal thyroid function and thyrotoxic features consistent with P R T H in some individuals but similar biochemical abnormalities and a lack of symptoms indicative of G R T H in other members. Overall, these findings suggest that GRTH and PRTH may represent the phenotypic spectrum of a single genetic entity. In a significant minority of individuals with biochemical evidence of RTH, we have been unable to find a mutation in coding exons of h T R - ~ . Possible explanations in these cases include a somatic TR-[31 mutation whose expression is limited so as to be undetectable in peripheral blood leukocyte DNA, or a mutation in TR-[32, the pituitary-specific receptor isoform. Homologous point mutations in hTR-aa generate mutant receptor proteins that are also capable of exerting dominant negative effects (Zavacki et al, 1993), raising the possibility that naturally occurring defects in the TR-a gene may also be associated with a resistance phenotype. Finally, the possibility of novel, non-receptor mechanisms by which thyroid hormone action could be disrupted to produce RTH should also be considered. PATHOGENESIS OF RESISTANCE

We suggest that the ability to exert a dominant negative effect within the pituitary-thyroid axis is a key property of mutant receptor proteins and generates the characteristically abnormal thyroid function test results that lead to the detection of RTH (see Figure 1). On this background, the variable clinical phenotype may be due to variable degrees of peripheral resistance in different patients, as well as variable resistance in different tissues within a single individual. The latter observation may be partly explicable on the basis of the differing tissue distributions of receptor isoforms. The liver and pituitary express predominantly TR-[31 and TR-[32 receptors respectively (Rodd et al, 1991; Lazar, 1993), whereas TR-c~I is the major species detected in myocardium (Falcone et al, 1992). Therefore, mutations in the TR-[3 gene are likely to be associated with pituitary and



132+~1 >> o : I ~



' cd/~l

.od Figure 3. Influence of the tissue distribution of thyroid hormone receptor isoforms on the clinical phenotype. The preponderance of TRq32 and TR-[31 isoforms in the pituitary generates resistance within the pituitary-thyroid axis, leading to elevated levels of serum thyroid hormones and inappropriate TSH secretion. Peripheral tissue responses to hyperthyroxinaemia are partly dependent on their receptor status. Predominance of TR-[31 in the liver is associated with resistance, whereas the relative abundance of TR-cq in myocardium is associated with retention of sensitivity to thyroid hormones. Both TR-~Xland TR-[31 are expressed in different regions of the brain, which may lead to mixed responses.

liver resistance, as exemplified by normal SHBG and non-suppressed TSH levels, whilst the tachycardia often seen in RTH may represent retention of cardiac sensitivity to thyroid hormone action mediated by a normal eL-receptor (Figure 3). Another factor that may regulate the degree of tissue resistance is the relative expression of mutant versus wild-type alleles of the TR-[3 gene. Although one study has suggested that both alleles are equally expressed (Hayashi et al, 1993), another showed marked differences in the relative levels of wild-type and mutant receptor messenger RNA in skin fibroblasts from two patients with RTH (Mixson et al, 1993b). In one of these individuals a temporal variation in expression of the mutant allele correlated with the degree of resistance in bone. We and others have also observed that the dominant negative potential of mutant receptors can differ depending on the nature and configuration of TREs (Meier et al, 1993; Zavacki et al, 1993; Adams et al, 1994), and suggest that this is a third variable that may influence target gene resistance to thyroid hormone action. Finally, factors not related to receptor mutation may also affect the clinical and biochemical phenotype. For example, a deleterious arginine to histidine mutation at codon 316 was associated with normal thyroid function in one kindred (Geffner et al, 1993), but in an unrelated family from our series the same mutation was associated with abnormal thyroid function (unpublished data), suggesting that other variables in the pituitary-thyroid axis can modulate mutant receptor action. Indirect evidence in favour of this notion is also provided by the observation that the unaffected first-degree



relatives in a kindred with RTH had normal but above average total T4 levels compared with those in unrelated controls (Weiss et al, 1993b). MANAGEMENT

One of the most important reasons for recognizing R T H is that its management differs from that of other common forms of thyroid dysfunction. In addition, the distinction between GRTH and PRTH on the basis of clinical criteria remains useful, since the management of the two states differs. In most individuals with GRTH, the receptor defect is compensated by high circulating thyroid hormone levels, leading to a euthyroid state not associated with abnormalities other than goitre. Certain circumstances, such as hypercholesterolaemia in adults or growth retardation in young children, may warrant the administration of supraphysiological doses of L-T4 to overcome the high degree of resistance in certain tissues. Although successful in some patients (Refetoff et al, 1993), such therapy needs careful monitoring of a number of other indices of peripheral thyroid hormone action, to avoid the adverse cardiac effects or excess catabolism associated with overtreatment. Misdiagnosis of RTH, followed by inappropriate thyroid ablation, invariably renders the resistant patient hypothyroid and is another context in which thyroxine replacement in supraphysiological dosage is indicated. In contrast, a general reduction in thyroid hormone levels may be of benefit in the management of patients with PRTH and thyrotoxic symptoms. However, the administration of conventional antithyroid drugs usually causes a further rise in serum TSH levels with consequent thyroid enlargement, and may also be associated with a theoretical risk of inducing autonomous pituitary TSH-secreting neoplasms. Accordingly, agents that inhibit pituitary TSH secretion, yet are devoid of peripheral thyromimetic effects, are used to reduce thyroid hormone levels. In a number of cases, the thyroid hormone analogue 3,3,5-tri-iodothyroacetic acid has been shown to be beneficial (Beck-Peccoz et al, 1983; Faglia et al, 1987; Salmela et al, 1988; Crin6 et al, 1992) but did not lead to clinical amelioration in one instance (Kunitake et al, 1989). Dextro-thyroxine (I)-T4) is another useful agent that has been effective in some individuals (Hamon et al, 1988; Dorey et al, 1990). If these agents fail, the dopaminergic agent bromocriptine (Dulgeroff et al, 1992) or the somatostatin analogue octreotide (Williams et al, 1986) may be administered. However, past experience indicates that TSH secretion escapes the inhibitory effects of bromocriptine (Beck-Peccoz et al, 1983; Dorey et al, 1990) as well as those of octreotide (Beck-Peccoz et al, 1989). In view of the spontaneous variation of symptoms in PRTH, we recommend periodic cessation of all therapy and re-evaluation of the clinical status of the patient. The treatment of PRTH in childhood again requires careful monitoring to ensure that any reduction in thyroid hormone levels is not associated with growth retardation or adverse neurological sequelae. Indeed, control of cardiac and sympathomimetic manifestations with [3-blockade may be the safest course in these circumstances. In this context,



a cardioselective agent may be preferable to propranolol as the latter may induce hypothyroidism in some tissues by inhibiting the conversion of T4 to T3. In both age groups, measures aimed at thyroid ablation are best used as a last resort or avoided altogether, as they are irreversible and may worsen the compensated hypothyroidism in some tissues, with potentially harmful consequences. Finally, the future development of thyroid hormone analogues with selective TR-[3 agonist activity or TR-oL-specific antagonists may represent a more rational basis for the treatment of these disorders.

FUTURE DIRECTIONS The elucidation of a genetic defect associated with RTH allows the disorder to be diagnosed definitively. Prospective studies of such genetically defined cases will enable the clinical features and their relationship to the associated receptor mutation to be delineated more precisely. The introduction of mutations into the mouse TR-[3 gene by homologous recombination will generate an animal model for the disorder that will facilitate biochemical and molecular biological studies to determine the pathogenesis of the variable phenotype.


The syndromes of resistance to thyroid hormone (RTH) are rare disorders characterized by elevated levels of circulating free thyroid hormones, inappropriate TSH secretion and variably reduced peripheral tissue responses to iodothyronine action. On the basis of clinical features, two major forms of RTH are recognized: generalized resistance (GRTH) in which patients are asymptomatic with few clinical signs, and pituitary resistance (PRTH) where patients present with features associated with thyrotoxicosis. However, a review of the literature and our own experience indicates that there is a wide overlap of clinical and biochemical features between individuals with GRTH or PRTH. Genetic analysis shows that both disorders are associated with a number of different mutations in the thyroid hormone receptor [3 (TR-[3) gene which localize to two regions in the hormone-binding domain. The mutant proteins are transcriptionally impaired but preserve the ability to bind DNA, dimerize and inhibit the function of their wild-type counterparts in a dominant negative manner. Dominant negative effects of mutant receptors within the pituitary-thyroid feedback axis generate abnormal thyroid function test results characteristic of RTH. The variable peripheral resistance may be related to differences in tissue distribution of TR-o~ versus TR-[3 receptor isoforms, variable dominant negative effects of mutant receptors on different target genes or other factors not related to the receptor mutation. Although GRTH and PRTH represent the variable phenotypic spectrum of a single genetic entity, this clinical distinction will remain useful as a guide to appropriate treatment.



Acknowledgements The authors' work is supported by grants from the Medical Research Council and Wellcome Trust (UK) to V.K.K.C. and from Murst and CNR (Rome, Italy) to P.B.-P. They are also greatly indebted to many physicians for the referral of patients, without whom these studies would not have been possible.

REFERENCES Adams M, Collingwood TN, Tone Y e t al (1944) Functional properties of thyroid hormone receptor mutations in resistance syndromes. Journal of Cellular Biochemistry 18B (supplement): K500 (abstract). Beck-Peccoz P & Chatterjee VKK (1994) The variable clinical phenotype in thyroid hormone resistance syndrome. Thyroid (in press). Beck-Peccoz P, Piscitelli G, Cattaneo MG & Faglia G (1983) Successful treatment of hyperthyroidism due to nonneoplastic pituitary TSH hypersecretion with 3,5,3'-triiodothyroacetic acid (TRIAC). Journal of Endocrinological Investigation 6: 217-223. Beck-Peccoz P, Romelli PB, Cattaneo MG et al (1984) Evaluation of free thyroxine methods in the presence of iodothyronin-binding autoantibodies. Journal of Clinical Endocrinology and Metabolism 58: 736--739. Beck-Peccoz P, Piscitelli G, Medri G et al (1985) Thyroid test strategy. Lancet i: 1456. Beck-Peccoz P, Mariotti S, Guillausseau PJ et al (1989) Treatment of hyperthyroidism due to inappropriate secretion of thyrotropin with the somatostatin analog SMS 201-995. Journal of Clinical Endocrinology and Metabolism 68: 208-214. Beck-Peccoz P, Roncoroni R, Mariotti S e t al (1990) Sex hormone-binding globulin measurement in patients with inappropriate secretion of thyrotropin (IST): evidence against selective pituitary thyroid hormone resistance in nonneoplastic IST. Journal of Clinical Endocrinology and Metabolism 71: 19-25. Beck-Peccoz P, Persani L & Faglia G (1992) Glycoprotein hormone c~-subunit in pituitary adenomas. Trends in Endocrinology and Metabolism 3: 41-45. Chatterjee VKK & Tata JR (1992) Thryoid hormone receptors and their role in development. Cancer Surveys 14: 147-167. Chatterjee VKK, Nagaya T, Madison LD et al (1991) Thyroid hormone resistance syndrome. Inhibition of normal receptor function by mutant thyroid hormone receptors. Journal of Clinical Investigation 87: 1777-1784. Chatterjee VKK, Adams M, Matthews C et al (1993) Thyroid hormone receptor mutations in both generalized and pituitary resistance syndromes. Journal of Endocrinological Investigation 16 (6) (supplement 2): 57 (abstract). Chin WW (1991) Nuclear thyroid hormone receptors. In Parker MG (ed.) Nuclear Hormone Receptors, pp 79-102. London: Academic Press. Crin6 A, Borrelli P, Salvatori R et al (1992) Anti-iodothyronine autoantibodies in a girl with hyperthyroidism due to pituitary resistance to thyroid hormones. Journal of Endocrinological Investigation 15: 113-120. Dorey F, Strauch G & Gayno JP (1990) Thyrotoxicosis due to pituitary resistance to thyroid hormones. Successful control with D-thyroxine; a study in three patients. ClinicalEndocrinology (Oxford) 32: 221-227. Dulgeroff AJ, Geffner ME, Koyal SN et al (1992) Bromocriptine and Triac therapy for hyperthyroidism due to pituitary resistance to thyroid hormone. Journal of Clinical Endocrinology and Metabolism 75: 1071-1075. Ekins RP (1990) Measurement of free hormones in blood. Endocrine Reviews 11: 5-46. Faber J, Perrild H & Johansen JS (1990) Bone Gla protein and sex hormone-binding globulin in nontoxic goiter: parameters for metabolic status at the tissue level. Journal of Clinical Endocrinology and Metabolism 70: 49-55. Faglia G, Beck-Peccoz P, Piscitelli G & Medri G (1987) Inappropriate secretion of thyrotropin by the pituitary. Hormone Research 26: 79-99.



Falcone M, Miyamoto T, Fierro-Renoy F et al (1992) Antipeptide polyclonal antibodies specifically recognize each human thyroid hormone receptor isoform. Endocrinology 131: 2419-2429. Forman BM & Samuels HH (1990) Interactions among a subfamily of nuclear receptors: the regulatory zipper model. Molecular Endocrinology 4: 1293-1301. Franklyn JA (1991) Syndromes of thyroid hormone resistance. Clinical Endocrinology (Oxford) 34: 237-245. Franklyn JA & Sheppard MC (1988) Thyrotrophin gene regulation. Journal of Endocrinology 117: 161-165. Geffner ME, Su F, Ross NS et al (1993) An arginine to histidine mutation in codon 311 of the e-erbAf3 gene results in a mutant thyroid hormone receptor that does not mediate a dominant negative phenotype. Journal of Clinical Investigation 91: 538-546. Gershengorn MC & Weintraub BD (1975) Thyrotropin-induced hyperthyroidism caused by selective pituitary resistance to thyroid hormone. A new syndrome of inappropriate secretion of TSH. Journal of Clinical Investigation 56: 633-642. Hamon P, Bovier-LaPierre M, Robert M e t al (1988) Hyperthyroidism due to selective pituitary resistance to thyroid hormones in 15-month-old boy: efficacy of D-thyroxine therapy. Journal of Clinical Endocrinology and Metabolism 67: 1089--1093. Harvey RD, McHardy KC, Reid IW et al (1991) Measurement of bone collagen degradation in hyperthyroidism and during thyroxine replacement using pyridinium cross-links as specific urinary markers. Journal of Clinical Endocrinology and Metabolism 72: 1189-1194. Hauser P, Zametkin AJ, Martinez P e t al (1993) Attention deficit-hyperactivity disorder in people with generalized resistance to thyroid hormone. New England Journal of Medicine 328: 997-1001. Hayashi Y, Janssen OE, Weiss RE et al (1993) The relative expression of mutant and normal thyroid hormone receptor genes in patients with generalized resistance to thyroid hormone determined by estimation of their specific messenger ribonucleic acid products. Journal of Clinical Endocrinology and Metabolism 76: 64-69. Kunitake JM, Hartman N, Henson LC et al (1989) 3,5,3'-Triiodothyroacetic acid therapy for thyroid hormone resistance. Journal of Clinical Endocrinology and Metabolism 69: 461-466. Lazar MA (1993) Thyroid hormone receptors: multiple forms, multiple possibilities. Endocrine Reviews 14: 184-193. Magner JA, Petrick P, Menezes-Ferreira M & Weintraub BD (1986) Familial generalized resistance of thyroid hormones: a report of three kindreds and correlation of patterns of affected tissues with the binding of p2sI]triiodothyronine to fibroblast nuclei. Journal of Endocrinological Investigation 9: 459-469. Mariotti S, Anelli S, Bartalena L et al (1987) Familial hyperthyroidism due to inappropriate TSH secretion associated with sellar abnormalities. Journal of Endocrinological Investigation 10 (supplement I): 20. Mariotti S, Caturegli G, Barbesino G e t al (1992) Thyroid function and thyroid autoimmunity independently modulate serum concentrations of soluble interleukin 2 receptor (slL-2R) in thyroid diseases. Clinical Endocrinology 37: 415-422. Meier CA, Dickstein BM, Ashizawa K et al (1992) Variable transcriptional activity and ligand binding of mutant 1313,5,3'-triiodothyronine receptors from four families with generalised resistance to thyroid hormone. Molecular Endocrinology 6: 248-258. Meier CA, Parkison C, Chen A e t al (1993) Interaction of human 13i thyroid hormone receptor and its mutants with DNA and retinoid X receptor. T3 response element-dependent dominant negative potency. Journal of Clinical Investigation 92: 1986-1993. Mixson AJ, Parrilla R, Ransom SC et al (1992) Correlation of language abnormalities with localization of mutations in the 13-thyroid hormone receptor in 13 kindreds with generalized resistance to thyroid hormone: identification of four new mutations. Journal of C1inical Endocrinology and Metabolism 75: 1039-1045. Mixson A J, Renault JC, Ramson S et al (1993a) Identification of a novel mutation in the gene encoding the [3-triiodothyronine receptor in a patient with apparent selective pituitary resistance to thyroid hormone. Clinical Endocrinology (Oxford) 38: 227-234. Mixson AJ, Hauser P, Tennyson G e t al (1993b) Different expression of mutant and normal beta T3 receptor alleles in kindreds with generalized resistance to thyroid hormone. Journal of Clinical Investigation 91: 2296-2300.



Nagaya T & Jameson JL (1993) Thyroid hormone receptor dimerization is required for dominant negative inhibition by mutations that cause thyroid hormone resistance. Journal of Biological Chemistry 268: 15766-15771. Nagaya T, Madison LD & Jameson JL (1992) Thyroid hormone receptor mutants that cause resistance to thyroid hormone: evidence for receptor competition for DNA sequences in target genes. Journal of Biological Chemistry 267:13014--13 019. O'Donnell AL & Koenig RJ (1990) Mutational analysis identifies a new functional domain of the thyroid hormone receptor. Molecular Endocrinology 4: 715-720. Ono S, Schwartz ID, Mueller OT et al (1991) Homozygosity for a dominant negative thyroid hormone receptor gene responsible for generalized thyroid hormone resistance. Journal of Clinical Endocrinology and Metabolism 73: 990--994. Parrilla R, Mixson A J, McPherson JA et al (1991) Characterization of seven novel mutations of the c-erbAf5 gene in unrelated kindreds with generalized thyroid hormone resistance. Evidence for two 'hot spot' regions of the ligand binding domain. Journal of Clinical Investigation 88: 2123-2130. Pearce CJ & Himsworth RL (1984) Total and free thyroid hormone concentrations in patient receiving maintenance replacement treatment with thryoxine. British Medical Journal 288: 693-695. Persani L, Asteria C, Tonacchera Met al (1994) Evidence for the secretion of thyrotropin with enhanced bioactivity in syndromes of thyroid hormone resistance. Journal of Clinical Endocrinology and Metabolism (in press). Refetoff S, De Wind LT & DeGroot LJ (1967) Familial syndrome combining deaf-mutism, stippled epiphyses, goiter and abnormally high PBI: possible target organ refractoriness to thyroid hormone. Journal of Clinical Endocrinology and Metabolism 27: 279-294. Refetoff S, Salazar A, Smith TJ & Scherberg NH (1983) The consequences of inappropriate treatment due to failure to recognise the syndrome of pituitary and peripheral tissue resistance to thyroid hormone. Metabolism 32: 822-834. Refetoff S, Weiss RE & Usala SJ (1993) The syndromes of resistance to thyroid hormone. Endocrine Reviews 14: 348--399. Rodd C, Schwartz HL, Strait KA & Oppenheimer JH (1991) Ontogeny of hepatic nuclear triiodothyronine receptor isoforms in the rat. Endocrinology 131: 2559-2564. Sakata S, Nakamura S & Miura K (1985) Autoantibodies against thyroid hormones or iodothyronin. Annals of Internal Medicine 103: 57%589. Sakurai A, Nakai A & DeGroot LJ (1990a) Structural analysis of the thyroid hormone receptor 13gene. Molecular and Cellular Endocrinology 71: 83-91. Sakurai A, Miyamoto T, Refetoff S & DeGroot LJ (1990b) Dominant negative transcriptional regulation by a mutant thyroid hormone receptor 13in a family with generalised resistance to thyroid hormone. Molecular Endocrinology 4: 1988-1994. Salmela PI, Wide L, Juustila H & Ruokonen A (1988) Effects of thyroid hormones (T4, T3), bromocriptine and TRIAC on inappropriate TSH hypersecretion. Clinical Endocrinology (Oxford) 28: 497-507. Sarne DH, Refetoff S, Rosenfield RL & Farriaux JP (1988) Sex hormone-binding globulin in the diagnosis of peripheral tissue resistance to thyroid hormone: the value of changes after short term triiodothyronine administration. Journal of Clinical Endocrinology and Metabolism 66: 740-746. Sasaki S, Nakamura H, Tagami T et al (1993) Pituitary resistance to thyroid hormone associated with a base mutation in the hormone binding domain of the human 3,5,3'triiodothyronine receptor beta. Journal of Clinical Endocrinology and Metabolism 76: 1254. Takeda K, Sakurai A, DeGroot LJ & Refetoff S (1992) Recessive inheritance of thyroid hormone resistance caused by a complete deletion of the protein-coding region of the thyroid hormone receptor-13 gene. Journal of Clinical Endocrinology and Metabolism 74: 49-55. Usala SJ, Bale AE, Gesundheit N e t al (1988) Tight linkage between the syndrome of generalized thyroid hormone resistance and the human c-erbA13 gene. Molecular Endocrinology 2: 1217-1220. Usala SJ, Menke JB, Watson TL et al (1991) A homozygous deletion in the c-erbA~3 thyroid hormone receptor gene in a patient with generalised thyroid hormone resistance: isolation and characterization of the mutant receptor. Molecular Endocrinology 5: 327-335.



Weintraub BD, Gershengorn MC, Kourides IA & Fein H (1981) Inappropriate secretion of thyroid-stimulating hormone. Annals of Internal Medicine 95: 339-351. Weiss RE, Weinberg M & Refetoff S (1993a) Identical mutations in unrelated families with generalised resistance to thyroid hormone occur in cytosine-guanine-rich areas of the thyroid hormone receptor gene; analysis of 15 families. Journal of ClinicalInvestigation 91: 2408-2415. Weiss RE, Marcocci C, Bruno-Bossio G & Refetoff S (1993b) Multiple genetic factors in the heterogeneity of thyroid hormone resistance. Journal of Clinical Endocrinology and Metabolism 76: 257-259. Williams G, Kraenzlin M, Sandler L e t al (1986) Hyperthyroidism due to non-tumoural inappropriate TSH secretion: effect of long-acting somatostatin analogue (SMS 201-995). Acta Endocrinologica (Copenhagen) 113: 42-46. Wood DF, Zalin AM, Ratcliffe WA & Sheppard MC (1987) Elevation of free thyroxine measurements with thyrotoxicosis. Quarterly Journal of Medicine 246z 863-870. Yen PM, Sugawara A, Refetoff S & Chin WW (1992) New insights on the mechanism(s) of the dominant negative effect of mutant thyroid hormone receptor in generalized resistance to thyroid hormone. Journal of Clinical Investigation 90: 1825-1831. Zavacki AM, Harney JW, Brent GA & Larsen PR (1993) Dominant negative inhibition by mutant thyroid hormone receptors is thyroid response element and receptor isoform specific. Molecular Endocrinology 7: 1319-1330.