Thyroid-Stimulating Hormone Receptor

Cloning and Expression

Nagayama et al. (1989) isolated a TSHR cDNA from a human thyroid cDNA library. The deduced 764-amino acid protein has a molecular mass of 86.8 kD and contains a signal peptide, 7 transmembrane regions, 5 potential glycosylation sites, and a short intracytoplasmic region. The TSHR cDNA encoded a functional receptor that activated adenylate cyclase in response to TSH.

Libert et al. (1989) used a dog Tshr cDNA to isolate a human TSHR cDNA from a thyroid cDNA library. The cDNA encodes a deduced 744-amino acid protein with 90.3% homology to the dog protein. Two major 4.6- and 4.4-kb mRNA transcripts were identified, suggesting alternative splicing.

By analyzing several TSHR cDNA clones, Misrahi et al. (1990) determined that the mature TSHR polypeptide contains 743 amino acids with a calculated molecular mass of 84.5 kD. The putative TSH receptor has a 394-residue extracellular domain, a 266-residue transmembrane domain, and an 83-residue intracellular domain. The authors observed a high degree of homology with the luteinizing hormone/choriogonadotropin receptor (LHCGR; 152790).

Kakinuma and Nagayama (2002) found that the TSHR gene can express at least 5 alternatively spliced forms.

Gene Function

The TSH receptor differs from the LHCG receptor by the presence of 2 unique insertions of 8 and 50 amino acids in the extracellular domain. Wadsworth et al. (1990) showed that the 8-amino acid tract near the amino terminus of the TSH receptor is an important site of interaction with both TSH and autoantibodies against the TSH receptor (thyroid-stimulating immunoglobulins, TSI). Either deletion or substitution of this region abolished the interaction, whereas a deletion of the 50-amino acid tract had no effect.

Contiguous to the 5-prime end of the thyroid transcription factor-1 (TTF1; 600635) element upstream and within the TSHR promoter is an element on the noncoding strand with single-strand binding activity that is important for regulation of TSHR expression. Ohmori et al. (1996) identified a cDNA encoding a single-strand binding protein (SSBP), referred to as SSBP1, that forms a specific complex with this element on the noncoding strand of TSHR. SSBP1 is a ubiquitous transcription factor that contributes to TSHR maximal expression, and mutation analyses showed that a GXXXXG motif is important for the binding and enhancer function of SSBP1. The authors concluded that the common transcription factors regulate TSHR and major histocompatibility gene expression. They also concluded that SSBP1 is a member of a family of SSBPs that interact with RNA and with the promoter of retroviruses, and are important in RNA processing. Members of this family also can interact with c-myc (190080), a gene linked to growth and DNA replication.

Biochemical Features

The high sequence homology with the LHCG receptor, which is composed of a single polypeptide chain, led many to suppose a similar structure for the TSH receptor. However, Loosfelt et al. (1992) presented evidence for a heterodimeric structure of TSHR. The extracellular (hormone-binding) alpha subunit had an apparent molecular mass of 53 kD, whereas the membrane-spanning beta subunit seemed heterogeneous and had an apparent molecular mass of 33 to 42 kD. Human thyroid membranes contained 2.5 to 3 times as many beta subunits as alpha subunits; however, the 2 subunits probably derive from a single gene since a single reading frame was demonstrated by cDNA cloning and sequencing. The exact site of cleavage that results in the 2 subunits was difficult to define.

The TSH receptor is the antigen targeted by autoantibodies in Graves disease (275000). By PCR amplification of specific cDNA, Feliciello et al. (1993) demonstrated that mature TSH receptor mRNA is expressed in the retroorbital tissue of both healthy subjects and patients with Graves disease. Of other tissues and cells tested, only thyroid tissue expressed the TSHR mRNA. The findings provided a link between orbital involvement and thyroid disease in Graves disease.

Graves et al. (1999) used epitope-mapped monoclonal and polyclonal antibodies to TSHR as immunoblot probes to detect and characterize the molecular species of the receptor present in normal human thyroid tissue. In reduced membrane fractions, both full-length (uncleaved) holoreceptor and cleavage-derived subunits of the holoreceptor were detected. Uncleaved holoreceptor species included a nonglycosylated form of apparent molecular mass 85 kD and 2 glycosylated forms of approximately 110 and 120 kD. The membranes also contained several forms of cleavage-derived TSHR alpha and beta subunits. Alpha subunits were detected by antibodies to epitopes localized within the N-terminal end of the TSHR ectodomain and migrated diffusely between 45 and 55 kD, reflecting a differentially glycosylated status. Several species of beta subunit were present, the most abundant having apparent molecular masses of 50, 40, and 30 kD. The authors concluded that posttranslational processing of the TSHR occurs in human thyroid tissue and involves multiple cleavage sites.

Lazar et al. (1999) studied the expression of 4 thyroid-specific genes (sodium-iodide symporter (NIS, or SLC5A5; 601843), thyroid peroxidase (TPO; 274500), thyroglobulin (TG; 188450), and TSHR) as well as the gene encoding glucose transporter-1 (GLUT1, or SLC2A1; 138140) in 90 human thyroid tissues. Messenger RNAs were extracted from 43 thyroid carcinomas (38 papillary and 5 follicular), 24 cold adenomas, 5 Graves thyroid tissues, 8 toxic adenomas, and 5 hyperplastic thyroid tissues; 5 normal thyroid tissues were used as reference. A kinetic quantitative PCR method, based on the fluorescent TaqMan methodology and real-time measurement of fluorescence, was used. NIS expression was decreased in 40 of 43 (93%) thyroid carcinomas and in 20 of 24 (83%) cold adenomas; it was increased in toxic adenomas and Graves thyroid tissues. TPO expression was decreased in thyroid carcinomas but was normal in cold adenomas; it was increased in toxic adenomas and Graves thyroid tissues. TG expression was decreased in thyroid carcinomas but was normal in the other tissues. TSHR expression was normal in most tissues studied and was decreased in only some thyroid carcinomas. In thyroid cancer tissues, a positive relationship was found between the individual levels of expression of NIS, TPO, TG, and TSHR. No relationship was found with the age of the patient. Higher tumor stages (stages greater than I vs stage I) were associated with lower expression of NIS and TPO. Expression of the GLUT1 gene was increased in 1 of 24 (4%) adenomas and in 8 of 43 (19%) thyroid carcinomas. In 6 thyroid carcinoma patients, 131-I uptake was studied in vivo. NIS expression was low in all samples, and 3 patients with normal GLUT1 expression had 131-I uptake in metastases, whereas the other 3 patients with increased GLUT1 gene expression had no detectable 131-I uptake. The authors concluded that (1) reduced NIS gene expression occurs in most hypofunctioning benign and malignant thyroid tumors; (2) there is differential regulation of the expression of thyroid-specific genes; and (3) an increased expression of GLUT1 in some malignant tumors may suggest a role for glucose-derivative tracers to detect in vivo thyroid cancer metastases by positron-emission tomography scanning.

Chia et al. (2007) studied the diagnostic value of circulating TSHR mRNA for preoperative detection of differentiated thyroid cancer (DTC) in patients with thyroid nodules. Based on cytology/pathology, 88 patients had DTC and 119 had benign thyroid disease. The TSHR mRNA levels in cancer patients were significantly higher than in benign disease (P less than 0.0001). At a cutoff value of 1.02 ng/g total RNA, the TSHR mRNA correctly classified 78.7% of patients preoperatively (sensitivity = 72.0%; specificity = 82.5%). Chia et al. (2007) concluded that TSHR mRNA measured with fine needle aspirations enhances the preoperative detection of cancer in patients with thyroid nodules, reducing unnecessary surgeries, and immediate postoperative levels can predict residual/metastatic disease.

Gene Structure

Kakinuma and Nagayama (2002) determined that the TSHR gene contains 13 exons.

Mapping

Akamizu et al. (1990) mapped the TSHR gene to human chromosome 14 by study of somatic cell hybrid DNAs. By in situ hybridization, Rousseau-Merck et al. (1990) and Libert et al. (1990) regionalized the gene to 14q31.

Akamizu et al. (1990) mapped the mouse Tshr gene to chromosome 12 using linkage studies in interspecies backcross mice. Wilkie et al. (1993) also localized the mouse Tshr gene to chromosome 12.

Molecular Genetics

Nonautoimmune Hyperthyroidism

Duprez et al. (1994) demonstrated heterozygous constitutively activating germline mutations in the TSHR gene (603372.0019; 603372.0020) in patients with hereditary nonautoimmune hyperthyroidism (609152). The functional in vitro characteristics of these 2 mutations were similar to those already described previously for autonomously functioning thyroid adenomas (Van Sande et al., 1995), and thus explained the development of thyroid hyperplasia and hyperthyroidism in the affected patients.

Paschke and Ludgate (1997) found reports of 4 infants with sporadic congenital hyperthyroidism occurring from a de novo germline mutation. In all cases, both parents were euthyroid. The authors noted that a number of gain-of-function mutations had been observed as somatic mutations in hyperfunctioning thyroid adenomas and in familial autosomal dominant hyperthyroidism. In their Figure 1, Paschke and Ludgate (1997) outlined the constitutively activating and inactivating mutations of the TSHR gene, as well as the location of somatic mutations found in thyroid carcinomas. At some locations, several different amino acid substitutions had been described. Most gain-of-function mutations were in exon 10.

Hypothyroidism, Congenital, Nongoitrous, 1

Alberti et al. (2002) sequenced the entire TSHR gene in a series of 10 unrelated patients with slight (6.6-14.9 mU/liter) to moderate (24-46 mU/liter) elevations of serum TSH, associated with normal free thyroid hormone concentrations, consistent with a diagnosis of thyrotropin resistance (CHNG1; 275200). Thyroid volume was normal in all patients, except 2 with modest hypoplasia. Autoimmune thyroid disease was excluded in all patients on the basis of clinical and biochemical parameters. Eight patients had at least 1 first-degree relative bearing the same biochemical picture. TSHR mutations were detected in 4 of 10 (40%) cases by analyzing DNA from peripheral leukocytes (see, e.g., 603372.0006; 603372.0029; 603372.0030; 603372.0031; 603372.0013). The authors concluded that partial resistance to TSH action is a frequent finding among patients with slight hyperthyrotropinemia of nonautoimmune origin, and that heterozygous germline mutations of TSHR may be associated with serum TSH values fluctuating above the upper limit of the normal range.

Calebiro et al. (2005) cotransfected COS-7 cells with wildtype TSHR and mutant receptors (C41S, 603372.0013; C600R, 603372.0029; L467P, 603372.0030) found in patients with autosomal dominant partial TSH resistance. Variable impairment of cAMP response to bovine TSH stimulation was observed, suggesting that inactive TSHR mutants may exert a dominant-negative effect on wildtype TSHR. By using chimeric constructs of wildtype or inactive TSHR mutants fused to different reporters, the authors documented an intracellular entrapment, mainly in the endoplasmic reticulum, and reduced maturation of wildtype TSHR in the presence of inactive TSHR mutants. Fluorescence resonance energy transfer and coimmunoprecipitation experiments supported the presence of oligomers formed by wildtype and mutant receptors in the endoplasmic reticulum. Calebiro et al. (2005) concluded that their findings provide an explanation for the dominant transmission of partial TSH resistance.

Familial Gestational Hyperthyroidism

Rodien et al. (1998) described a gain-of-function mutation of the TSHR gene (603372.0024) as the cause of familial gestational hyperthyroidism (603373). The mutation rendered the thyrotropin receptor hypersensitive to chorionic gonadotropin.

Hyperfunctioning Thyroid Adenoma and Thyroid Carcinoma with Thyrotoxicosis, Somatic

In 3 of 11 hyperfunctioning thyroid adenomas, Parma et al. (1993) identified somatic mutations in the TSHR gene (603372.0002; 603372.0003). These mutations were restricted to tumor tissue.

By direct sequencing, Fuhrer et al. (1997) screened a consecutive series of 31 toxic thyroid nodules (TTNs) for mutations in exons 9 and 10 of the TSHR gene and in exons 7 to 10 of the Gs-alpha protein gene (GNAS1; 139320). Somatic TSHR mutations were identified in 15 of the 31 (48%) TTNs. The TSHR mutations were localized in the third intracellular loop (asp619 to gly, 603372.0002; ala623 to val; and a 27-bp deletion resulting in deletion of 9 amino acids at codons 613 to 621), the sixth transmembrane segment (phe631 to leu, 603372.0004; thr632 to ile; and asp633 to glu), the second extracellular loop (ile568 to thr), and the third extracellular loop (val656 to phe). One mutation, ser281 to asn, was found in the part of the extracellular domain encoded by exon 9. All of the identified TSHR mutations resulted in constitutive activity. No mutations were found in exons 7 to 10 of GNAS1. The authors concluded that constitutively activating TSHR mutations occur in 48% of TTNs, representing the most frequent molecular mechanism known to cause TTNs.

Parma et al. (1997) investigated 33 different, autonomous hot nodules from 31 patients for the presence of somatic mutations in the TSHR and Gs-alpha genes. Twenty-seven mutations (82%) were found in the TSHR gene, affecting a total of 12 different residues or locations. All but 2 of the mutations studied had previously been identified as activating mutations. The authors identified the 2 novel mutations as a point mutation causing a leu629-to-phe substitution (L629F; 603372.0022), a deletion of 12 bases removing residues 658-661 (asn-ser-lys-ile) at the C-terminal portion of exoloop 3 (del658-661). Only 2 mutations (6%) were found in Gs-alpha genes. In 4 nodules, no mutation was detected. Five residues (ser281, ile486, ile568, phe631, and asp633) were found to be mutated in 3 or 4 different nodules, making them hotspots for activating mutations. The authors concluded that in a cohort of patients from a moderately iodine-deficient area, somatic mutations increasing the constitutive activity of TSHR are the major cause of autonomous thyroid adenomas.

Possible Association with Toxic Multinodular Goiter

Toxic multinodular goiter (TMNG) represents a frequent cause of endogenous hyperthyroidism, affecting 5 to 15% of such patients. To search for alterations of TSHR in autonomously functioning thyroid nodules (AFTN) and TMNG, Gabriel et al. (1999) used bidirectional, dye primer automated fluorescent DNA sequencing of the entire transmembrane domain and cytoplasmic tail of TSHR using DNA extracted from nodular regions of 24 patients with TMNG and 7 patients with AFTN. Eight of the 24 (33.3%) patients with TMNG were heterozygous for an asp727-to-glu polymorphism (D727E) in the cytoplasmic tail of TSHR. Three of the 24 (12.5%) patients with TMNG were heterozygous for a missense mutation, and 1 patient had multiple heterozygous mutations. Two patients had silent polymorphism of codons 460 and 618. The authors found no mutations in the transmembrane domain and cytoplasmic tail of TSHR in the 7 patients with AFTN, except for a silent polymorphism of codon 460 in 1. DNA fingerprinting of codon 727 using restriction enzyme NlaIII and genomic DNA confirmed the sequencing results in all cases, indicating that the sequence alterations were not somatic in nature. This technique was also used to examine peripheral blood genomic DNA from 52 normal individuals and 49 patients with Graves disease; 33.3% of TMNG (P of 0.019 vs normal subjects), 16.3% of Graves disease patients (p of 0.10 vs normal subjects), and 9.6% of normal individuals were heterozygous for the D727E polymorphism. Expression of the D727E variant in eukaryotic cells resulted in an exaggerated cAMP response to TSH stimulation compared with that of the wildtype TSHR. The authors concluded that the germline polymorphism D727E is associated with TMNG, and suggested that its presence is an important predisposing genetic factor in TMNG pathogenesis.

Muhlberg et al. (2000) compared the D727E frequencies of 128 European Caucasian patients with toxic nonautoimmune thyroid disease (83 with toxic adenoma, 31 with toxic multinodular goiter, and 14 with disseminated autonomy) with those of 99 healthy controls and 108 patients with Graves disease. They found no significant differences in codon 727 polymorphism frequencies between patients with autonomously functioning thyroid disorders (13.3%) and the healthy control group (16.2%). Moreover, the subtypes of toxic nonautoimmune thyroid disease were not related to significant differences in codon 727 polymorphism frequencies compared with the healthy control group. There was no significant difference between the polymorphism frequency among patients with Graves disease (21.3%) and that of healthy controls. The authors concluded that there was no association between the D727E polymorphism of the TSHR and toxic thyroid adenomas or toxic multinodular goiter in their study population.

Tonacchera et al. (2000) searched for inactivating TSHR or Gs-alpha mutations in areas of toxic or functionally autonomous multinodular goiters that appeared hyperfunctioning at thyroid scintiscan but did not clearly correspond to definite nodules at physical or ultrasonographic examination. Activating TSHR mutations were detected in 14 of these 20 hyperfunctioning areas, whereas no mutation was identified in nonfunctioning nodules or areas contained in the same gland. No Gs-alpha mutation was found. The authors concluded that activating TSHR mutations are present in the majority of nonadenomatous hyperfunctioning nodules scattered throughout the gland in patients with toxic or functionally autonomous multinodular goiter.

Possible Association with Susceptibility to Graves Disease

Although Heldin et al. (1991) and Bahn et al. (1994) suggested that substitutions in the TSHR gene (D36H; 603372.0001 and pro52-to-thr; P52T) were associated with Graves disease (275000) and Graves ophthalmopathy, respectively, Simanainen et al. (1999) reported that the D36H and P52T substitutions were polymorphic variants with a frequency of approximately 5% and 7.3%, respectively. Simanainen et al. (1999) found no association between these 2 polymorphisms and Graves disease. Similarly, Kotsa et al. (1997) found no association between the TSHR P52T polymorphism and Graves disease among 180 patients with Graves disease. The variant allele was present in 8.3% of patients and 7.3% of controls.

For additional discussion of a possible association between variation in the TSHR gene and Graves disease, see 275000.

Reviews

Vassart et al. (1991) reviewed the molecular genetics of the thyrotropin receptor.

TSHR Mutation Database

Trulzsch et al. (1999) described a database of TSHR mutations. The desirability of such a database came from the growing number of mutations identified and the variety of clinical phenotypes associated with the different mutations: somatic constitutively activating mutations in toxic thyroid nodules (e.g., 603372.0002); constitutively activating germline mutations as the cause of sporadic (e.g., 603372.0004) and familial (e.g., 603372.0019) nonautoimmune autosomal dominant hyperthyroidism (609152); and inactivating mutations associated with inherited TSH resistance (275200) (e.g., 603372.0005).

Animal Model

Using an adenovirus-mediated mouse model of Graves disease, Chen et al. (2003) demonstrated that goiter and hyperthyroidism occurred to a significantly greater extent when the adenovirus expressed the free alpha subunit as opposed to a genetically modified TSHR that cleaves minimally into subunits (p less than 0.005). Chen et al. (2003) concluded that shed alpha subunits induce or amplify the immune response leading to hyperthyroidism in Graves disease.

Abe et al. (2003) generated Tshr-null mice by replacing exon 1 of Tshr with a GFP cassette. They detected intense GFP fluorescence in thyroid follicles. Western blot analysis showed a 50% decrease in Tshr expression in heterozygotes and no expression in Tshr-null mice. Tshr-null mice were runted and hypothyroid, and they died by age 10 weeks with severe osteoporosis and significant reduction of calvarial thickness. Profound osteoporosis and focal osteosclerosis were observed in heterozygotes. Confocal microscopy demonstrated expression of Tshr in bone cells. They found 3-fold increased expression of Tnf (191160) in the bone marrow of Tshr-null mice. Neutralizing anti-Tnf antibody inhibited enhanced osteoclastogenesis in Tshr-null bone marrow cell cultures, suggesting that TNF is a proosteoclastic signal mediating the effects of TSHR deletion. Abe et al. (2003) found that TSH activation of Tshr resulted in attenuated osteoclast formation by inhibiting Jnk (see 601158) and Nfkb (see 164011) signaling, resorption, and survival. They showed that TSH regulated osteoblast differentiation through a Runx2 (600211)- and osterix (SP7; 606633)-independent mechanism that involved downregulation of the prodifferentiation factors Lrp5 (603506) and Flk1 (KDR; 191306). Abe et al. (2003) concluded that TSH acts as a single molecular switch in the independent control of both bone formation and resorption. Hase et al. (2006) found that the increased osteoclastogenesis in homozygous and heterozygous Tshr-null mice was rescued with graded reductions in the dosage of the Tnf gene.

Rubin et al. (2010) described the use of massively parallel sequencing to identify selective sweeps of favorable alleles and candidate mutations that have had a prominent role in the domestication of chickens and their subsequent specialization into broiler (meat-producing) and layer (egg-producing) chickens. Rubin et al. (2010) generated 44.5-fold coverage of the chicken genome using pools of genomic DNA representing 8 different populations of domestic chickens as well as red jungle fowl (Gallus gallus), the major wild ancestor. Rubin et al. (2010) reported more than 7,000,000 SNPs, almost 1,300 deletions, and a number of putative selective sweeps. One of the most striking selective sweeps found in all domestic chickens occurred at the locus for thyroid-stimulating hormone receptor (TSHR), which has a pivotal role in metabolic regulation and photoperiod control of reproduction in vertebrates. Several of the selective sweeps detected in broilers overlapped genes associated with growth, including growth hormone receptor (600946), appetite, and metabolic regulation. Rubin et al. (2010) found little evidence that selection for loss-of-function mutations had a prominent role in chicken domestication, but they detected 2 deletions in coding sequences, including one in SH3RF2 (613377), that the authors considered functionally important.