Bridge Technology with TSH Receptor Chimera for Sensitive Direct Detection of TSH Receptor Antibodies Causing Graves' Disease: Analytical and Clinical Evaluation.

Graves' disease is caused by stimulating autoantibodies against the thyrotropin receptor inducing uncontrolled overproduction of thyroid hormones. A Bridge Assay is presented for direct detection of these thyroid-stimulating immunoglobulins using thyrotropin receptor chimeras. A capture receptor, formed by replacing aa residues 261-370 of the human thyrotropin receptor with residues 261-329 from rat lutropin/choriogonadotropin receptor and fixed to microtiter plates, binds one arm of the autoantibody. The second arm bridges to the signal receptor constructed from thyrotropin receptor (aa 21-261) and secretory alkaline phosphatase (aa 1-519) inducing chemiluminescence. The working range of the assay is from 0.3 IU/l to 50 IU/l with a cutoff of 0.54 IU/l and functional sensitivity of 0.3 IU/l. Sensitivity and specificity are 99.8 and 99.1%, respectively, with a diagnostic accuracy of 0.998. The low grey zone is from 0.3-0.54 IU/l. The stimulatory character of the assayed antibodies is shown through a good correlation (r=0.7079, p<10(-7)) to serum T4 levels of untreated patients. In Graves' disease, titers are increased in associated eye disease. In 3 hypothyroid patients with sera positive in the thyrotropin receptor competition assay and in the blocking bioassay, antibodies are not detected by the Bridge Assay, while the monoclonal blocking antibody K1-70 was detected. In Hashimoto disease thyrotropin receptor autoantibodies are detected in some patients, but not in goiter. This Bridge Assay delivers good diagnostic accuracy for identification of Graves' disease patients. Its high sensitivity may facilitate early detection of onset, remission, or recurrence of Graves' disease enabling timely adaption of the treatment.Human genes: TSHR, Homo sapiens, acc. no. M31774.1.

TTG-2, a new gene encoding a cysteine-rich protein with the LIM motif, is overexpressed in acute T-cell leukaemia with the t(11;14)(p13;q11).

We have cloned 70 kb of DNA from chromosome 11p13 at the site of a recurrent translocation in T-cell leukaemia (T-ALL): t(11;14)(p13;q11). The translocation involves the TCR-delta gene on 14q11 and a new site on 11p13. Two new and 10 previously identified translocations all mapped within 25 kb on 11p13, the 11p13 T-cell translocation cluster (11p13 ttc). A search for expressed sequences surrounding the breakpoint cluster region on 11p13 identified a gene telomeric of all breakpoints which is overexpressed in three T-ALL samples with a t(11;14). The gene T-cell translocation gene (TTG-2) encodes a small cysteine-rich protein. Forty-eight per cent of the amino acids are identical with another translocation-deregulated gene, TTG-1 (T-cell translocation gene 1 or rhombotin) in 11p15. There are two copies of a cysteine-rich motif in both proteins. Two tandem copies of the same cysteine-rich motif are also present in the recently described lin-11, isl-1 and mec-3 gene products, and one motif is found in the CRIP protein. Therefore the proteins encoded by these two translocation-deregulated genes belong to this new class of cysteine-rich proteins with the 'LIM' motif, which are important in normal development.

The TTG-2/RBTN2 T cell oncogene encodes two alternative transcripts from two promoters: the distal promoter is removed by most 11p13 translocations in acute T cell leukaemia's (T-ALL).

The TTG-2 gene has been identified at the site of chromosomal translocations in acute T-cell leukemia's (T-ALL). These breakpoints map to a region between 2 and 30 kb upstream of TTG-2 in chromosome 11p13. To establish the role of these translocation breakpoints in the deregulation of TTG-2 in T-ALL we have determined the complete structure of this gene. Isolation of new TTG-2 cDNA clones from fetal liver identified an alternative transcript (TTG-2a) containing two new noncoding 5' exons. Analysis of exon/intron boundaries, identified 6 exons spread over 35 kb in 11p13. The gene encodes two alternative transcripts initiating from two promoters. TTG-2a, from promoter 1 (P1) and TTG-2b, from promoter 2 (P2) differ in the length of the 5' untranslated region, but encode the same protein. A high level of TTG-2a was present in fetal liver and spleen, whereas in adult kidney a low level of TTG-2a and a high level of TTG-2b was found. The transcription start site for TTG-2a was identified by RNase protection experiments and it displayed sequence homology to an initiator element (inr). P1 lacks a TATA box, but binding sites for SP1 and GATA-1 are present. This new genomic organisation revealed that all known chromosomal translocations map upstream of P2, removing P1 and putative upstream regulatory sequences leaving P2 intact. These results show that chromosomal translocations disrupt the TTG-2 gene itself, further confirming its role in the development of T-ALL.

A heritable point mutation in an extracellular domain of the TSH receptor involved in the interaction with Graves' immunoglobulins.

The TSH receptor (TSHR) is the central antigen in Graves' disease. Variant receptor proteins, arising from mutations in the TSHR gene, may play a role in the pathogenesis of the disease. Therefore, we analysed the TSHR from a 38-year-old patient affected with autoimmune hyperthyroidism, diffuse goitre and ophthalmopathy. Reverse transcription PCR and DNA amplification followed by DNA sequencing revealed a point mutation (C-->A) at cDNA position 253 in one of two alleles. This leads to the replacement of a proline (CCC) by threonine (ACC) at amino acid position 52 of the predicted receptor protein. Secondary structure predictions indicated a major change of protein structure as a result of the mutation. By using allele-specific PCR, we were able to show that this mutation is heritable. Screening of 50 random individuals revealed that four of them also carried this mutation in the heterozygous state. This study shows the presence of different forms of the TSHR gene in the population. The mutation, which is in a portion of the receptor apparently involved in binding of Graves' immunoglobulins, is discussed as to its possible pathophysiological role in autoimmune hyperthyroidism.

Modulation of iodide uptake by dialkyl phthalate plasticisers in FRTL-5 rat thyroid follicular cells.

Plasticisers imparting flexibility to plastics are man-made chemicals abundantly present in the environment. Effects of six different dialkyl phthalates were studied in vitro in the rat thyroid cell line FRTL-5 on their ability to modulate basal iodide uptake mediated by the sodium/iodide symporter (NIS). The present study shows that diisodecyl phthalate (DIDP), dioctyl phthalate (DOP), diisononyl phthalate (DINP) and bis (2-ethylhexyl) phthalate (DEHP) significantly enhance iodide uptake when concentrations in the magnitude between 10(-4) M and 10(-3) M were applied. In this range, these phthalates do not assess toxicity on the cells. Specific inhibiton of NIS demonstrated that enhancement of iodide uptake is due to NIS. In contrast, benzyl butyl phthalate (BBP) also augments iodide uptake at 1mM but this concentration has just exceeded the toxicity threshold and dibutyl phthalate (DBP), the most toxic compound did not modulate iodide uptake at any concentration applied. As we can deduce from our results, plasticisers are capable of significantly modulating NIS mediated iodide uptake activity.

A point mutation (Ala229 to Thr) in the hinge domain of the c-erbA beta thyroid hormone receptor gene in a family with generalized thyroid hormone resistance.

Various point mutations in the c-erbA thyroid hormone receptor (TR) beta gene of unrelated kindreds have been reported to be responsible for different phenotypes of generalized thyroid hormone resistance. We now report a new point mutation, Td, in one of two TR beta alleles of three affected members of one family, designated family T. In contrast to the previously described point mutations, all located in the T3-binding domain of the TR beta gene, mutation Td was identified in the carboxy-terminal part of the hinge domain. Direct sequencing of the polymerase chain reaction-amplified whole coding region of the patients' fibroblast TR beta genes displayed a single guanine to adenine transition at cDNA nucleotide position 985. This altered alanine (GCC) to threonine (ACC) in codon 229. Garnier prediction of the consequence of the mutation indicated an altered secondary structure. The G----A nucleotide substitution was not present in 80 random TR beta alleles, suggesting that this point mutation is responsible for generalized thyroid hormone resistance in family T. The in vitro expressed mutant TR beta was shown to bind with high affinity to various thyroid hormone response elements. However, the affinity of the TR beta to bind to T3 was reduced 3-fold, indicating that the hinge domain of the TR beta is important for full ligand-binding activity. Moreover, it seems that multiple subdomains of the TR beta interact cooperatively to achieve optimal T3 activity.

Deoxyribonucleic acid binding and transcriptional silencing by a truncated c-erbA beta 1 thyroid hormone receptor identified in a severely retarded patient with resistance to thyroid hormone.

We describe the analysis of a thyroid hormone receptor (TR) beta causing resistance to thyroid hormone, the patient exhibiting hypothyroid symptoms (severe mental retardation, hypoactivity, obesity) and hyperthyroid symptoms (tachycardia, low serum cholesterol) and, additionally, relative early puberty, advanced bone age, and short stature. The patient was heterozygous, with a point mutation producing a premature stop-codon in TR beta-gene exon 10, resulting in a 28-amino acid carboxy-terminal deletion in the cognate TR beta (TR beta-EZ). T3 binding was abolished. Homodimer binding of TR beta-EZ to DR4- and F2-T3 response elements (TREs) was weaker, and to a palindromic TRE (PAL) was stronger than that of wild-type TR beta (TR beta-WT) in the absence of T3. T3 dissociated TR beta-WT, but not TR beta-EZ homodimer, from DR4, F2, and Pal. Heterodimerization of TR beta-EZ with retinoid x receptor beta was seen. TR beta-EZ repressed basal thymidine kinase-promotor activity, coupled to DR4, F2, or PAL. Silencing of basal gene transcription via PAL was weaker, and via DR4 and F2 was more pronounced, compared with TR beta-WT. TR beta-EZ had a strong dominant negative effect on TR beta-WT, attenuated in a TRE- and cell-specific manner by high T3 concentrations. Finally, the degree of TR beta-EZ homodimer-binding affinity to DNA did not correlate with the degree of transcriptional dominant negative activity.

Thyroid transcription factor 1 and Pax8 synergistically activate the promoter of the human thyroglobulin gene.

Thyroglobulin (Tg) is an essential thyroid-specific protein, which serves as the matrix for thyroid hormone biosynthesis. To obtain new insights in the regulation of Tg gene expression, we investigated the interaction of the human Tg promoter with the thyroid-specific transcription factors TTF-1 and Pax8. A reporter gene, containing a 202 bp fragment from the human Tg 5'-flanking region including the promoter sequence and the transcriptional start site, and expression vectors containing the cDNAs for human TTF-1 and Pax8 were used in cotransfection experiments, in the non-thyroidal cell lines COS-7 and HeLa. Pax8 increased the specific transcriptional activity of the Tg promoter about threefold, whereas cotransfection with the homeodomain-containing protein TTF-1 stimulated promoter activity from six- to tenfold. The simultaneous expression of both factors stimulated the Tg promoter activity in a multiplicative manner up to 25-fold. TTF-1 binding sites could be localized precisely by lectron mobility shift assay. The two binding elements corresponded to sites A and C in the rat Tg promoter. Site-directed mutagenesis of three nucleotides in each binding element inhibited binding of TTF-1 to the two oligonucleotides. In cotransfection experiments, the mutant site C decreased TTF-1 transactivation to 26% of the wild-type, whereas an additional mutation in the site A reduced this value to almost zero, thus proving the physiological relevance of these sites. The present results demonstrate that the activity of the human Tg promoter is closely dependent on the function of TTF-1 and Pax8, opening the field for further investigations of pathological alterations of Tg gene expression.

Isolation of radiochemically pure 125I-labeled human thyrotropin receptor and its use for the detection of pathological autoantibodies in sera from Graves' patients.

We report a method for the purification and radioactive labeling of human TSH receptor (TSHR). The method is based on the construction of a fusion TSHR (TSHR-Xa-BIO) which consists of the N-terminal 725 amino acids of human TSHR linked to the 4-amino acid Xa protease cleavage site and the 87-amino acid C-terminal domain of the biotin carboxyl carrier protein subunit of Escherichia coli acetyl-CoA carboxylase (the C-terminal domain directs the efficient posttranslational biotinylation of the protein). TSHR-Xa-BIO was produced in HeLa cells using recombinant vaccinia virus. The expressed protein was fully functional and was biotinylated with an efficiency of about 90%. Streptavidin-agarose-immobilized TSHR-Xa-BIO was labeled with 125I using the chloramine T oxidation procedure and specifically eluted from the solid phase after cleavage with protease Xa. Isolated native radiochemically pure 125I-labeled TSHR specifically interacted with pathological autoantibodies in the sera of patients with Graves' disease, and thus could be useful for the detection of these autoantibodies by immunoprecipitation analysis.

Expression of S14-mRNA and its translational product in differentiating 3T3-L1 cells.

During adipogenic conversion of 3T3-L1 cells S14-mRNA increased from undetectable levels in preadipocytes to high levels in differentiated adipocytes (adipocyte-like cells). In vitro translation of hybrid-selected S14-mRNA revealed an identical protein in 3T3-L1 adipocytes and in rat liver with regard to migration properties in two-dimensional gel electrophoresis. No translation product was found in 3T3-L1 preadipocytes. In conclusion, protein-S14 is very similar or even identical in rat lipogenic tissues and in 3T3-L1 adipocytes. Therefore, this cell line can be employed to elucidate further physiological aspects of protein-S14.

The promoter of the human sodium/iodide-symporter gene responds to retinoic acid.

It was shown previously that hNIS mRNA expression is stimulated by retinoic acid (RA) in human follicular thyroid carcinoma cell lines FTC-133 and FTC-238, and patients with thyroid carcinomas lacking iodide uptake respond to RA treatment with increased radioiodide transport. Here, in transient transfection experiments using FTC-238 cells, hNIS promoter/luciferase reporter constructs showed an up to 2.5-fold increase in transcriptional activity after incubation with 1 microM RA. Stimulation by 10 nM T3 was up to 2.4-fold. Deletion or block mutation of a putative nuclear receptor recognition site, 'DR10', abolished RA and T3 responses. Four copies of the DR10 cloned 5' to the thymidine kinase promoter gave a 2.6-fold and a 1.4-fold increase in transcriptional activity after RA and T3 stimulation, respectively. In electrophoretic mobility shifts, a wildtype DR10 oligonucleotide, but not block mutants of either DR10 halfsite, interacted with nuclear receptors. Thus, RA redifferentiation of advanced thyroid carcinomas may reinduce iodide uptake by stimulating hNIS expression and thereby make tumours accessible for radioiodide therapy again.

Relevance of the hydrolysis and protein binding of melphalan to the treatment of multiple myeloma.

Experiments to determine the hydrolysis and protein binding of melphalan (L-phenylalanine mustard, L-PAM) were carried out in vitro for therapeutic concentration of the drug: the decrease in L-PAM concentration in plasma and whole blood during 24 h incubation at 37 degrees C was only 5% due to hydrolysis. Serum protein binding was about 90%, whereby 60% and 20% of this binding was due to interactions with albumin and acid alpha 1-glycoprotein, respectively. Immunoglobulins did not participate in the binding of L-PAM. The covalently bound part of L-PAM in serum was 30% in the concentration range of 1-30 micrograms/ml. The binding of dihydroxymelphalan (DOH) in serum did not exceed 20%. Glucocorticoids used in combination with L-PAM for treating multiple myeloma did not influence its protein binding. Our study with 35 sera from 15 patients with multiple myeloma shows that high levels of paraproteins do not increase but may decrease the binding of L-PAM, resulting in an elevated concentration of free drug.

Pharmacokinetics of melphalan in isolated limb perfusion.

The pharmacokinetics of melphalan was studied by sampling of tissue and plasma in 72 rats that underwent isolated hyperthermic limb perfusion under different conditions. A miniaturized extracorporeal circulation system for small animals was used for perfusion of the rat hindlimb. Melphalan levels (L-phenylalanine mustard, L-PAM) were determined by high-performance liquid chromatography (HPLC). The temperature of the perfusate plasma and tissue, pH, administration method, and flow rate were modified and compared with regard to their influence on pharmacokinetic parameters. The highest tissue penetration of melphalan was observed under the following conditions: (a) pH range of the perfusate plasma between 7.3 and 7.7 (physiological environment), (b) temperature range of the perfusate from 40 degrees to 41.5 degrees C (destruction of cellular carrier systems at higher temperatures and increased inactivation by hydrolysis of melphalan above 41.5 degrees C), (c) application of melphalan as a single dose into the reservoir of the extracorporeal circuit (optimal tissue penetration), and (d) reduced perfusate flow (prolonged contact time between perfusate and tissue).

A sensitive high-performance liquid chromatographic assay for melphalan and its hydrolysis products in blood and plasma.

A sensitive high-performance liquid chromatographic assay has been developed for the measurement of the alkylating cytostatic drug melphalan (4-[bis(2-chloroethyl)amino]-L-phenyl-alanine, or L-phenylalanine-mustard, L-PAM) and its two hydrolysis products, monohydroxy melphalan (MOH) and dihydroxy melphalan (DOH). A reversed-phase phenyl column and a mobile phase consisting of acetonitrile/citrate buffer made possible an isocratic separation and quantification. N,N-[bis(2-hydroxy-ethyl)]toluidine has been synthesized as an internal standard structurally related to DOH. A new, accurate "kinetic" calibration procedure enabled us to determine even the concentration of the unstable MOH. The lower limit of quantification was 30 ng/ml for L-PAM and 20 ng/ml for both DOH and MOH with fluorescence detection. The use of this method is illustrated by some pharmacokinetic data in systemic and locoregional melphalan therapy.

Studies on the pharmacokinetics of chlorambucil and prednimustine in patients using a new high-performance liquid chromatographic assay.

Following the oral administration of either chlorambucil/prednisolone or prednimustine to patients, the plasma levels of free chlorambucil and phenylacetic acid mustard, the beta-oxidation product of chlorambucil, were measured using a new high-performance liquid chromatographic (HPLC) assay. This assay permitted the simultaneous detection of the analyzed compounds with a lower limit of detection of 30 ng/ml. The pharmacokinetics of chlorambucil and phenylacetic acid mustard were found to be entirely different when prednimustine was administered as opposed to its components chlorambucil and prednisolone together. After the ingestion of the conjugate, the plasma concentration-time curves of chlorambucil and phenylacetic acid mustard showed a "delayed" pattern compared with those obtained after the administration of the components. The mean area under the concentration-time curves (AUCs) of prednimustine-derived chlorambucil and phenylacetic acid mustard were 25% and 40%, respectively, of the areas obtained after a stoichiometrically equivalent dose of chlorambucil. Free plasma prednimustine could not be detected at any time. This different pharmacokinetic behavior might offer an explanation for the superior therapeutic effects of prednimustine demonstrated by clinical studies.

Once-daily oral gatifloxacin vs three-times-daily co-amoxiclav in the treatment of patients with community-acquired pneumonia.

A double-blind, double-dummy, multicentre, multinational, parallel-group study was designed to establish proof of equivalence between oral gatifloxacin and oral co-amoxiclav in the treatment of 462 patients with mild-to-moderate community-acquired pneumonia. Eligible patients were randomised equally to either gatifloxacin 400 mg once-daily plus matching placebo for 5-10 days, or amoxycillin 500 mg + clavulanic acid 125 mg three-times-daily for 5-10 days. The primary efficacy endpoint was clinical response (clinical cure plus improvement) at the end of treatment. Overall, a successful clinical response was achieved in 86.8% of gatifloxacin-treated patients, compared with 81.6% of those receiving co-amoxiclav, while corresponding rates of bacteriological efficacy (eradication plus presumed eradication) were 83.1% and 78.7%, respectively. The safety and tolerability profile of gatifloxacin was comparable to that of co-amoxiclav, with adverse gastrointestinal events, e.g., diarrhoea and nausea, being the most common treatment-related adverse events in both groups. The study showed no evidence of gatifloxacin-induced phototoxicity, musculoskeletal disorders, or hepatic and renal problems. Overall, this study showed that gatifloxacin was equivalent clinically to a standard course of co-amoxiclav in patients with community-acquired pneumonia, and that gatifloxacin was safe and well-tolerated.

Expression of a biotinylated human thyrotropin receptor in HeLa cells using recombinant vaccinia virus and its application for the detection of Graves' autoantibodies.

We have prepared a biotinylated thyrotropin receptor (TSHR-BIO), and characterized its activity in cells and when bound to solid phase (streptavidin agarose). TSHR-BIO consists of the N-terminal 725 amino acids of the human thyrotropin (TSH) receptor linked to the 87-amino acid C-terminal domain of the biotin carboxyl carrier protein subunit of Escherichia coli acetyl-CoA carboxylase. The C-terminal domain directs the efficient post-translational biotinylation of the protein. TSHR-BIO was expressed using a vaccinia virus expression system. HeLa cells infected with recombinant virus produced large amounts of TSH receptor of approximately 120,000 molecules per cell. Vaccinia virus produced TSHR-BIO was fully functional interacting with TSH (Kd of 2.3+/-0.1 x 10(-10) M) and coupling to cyclic adenosine monophosphate (cAMP) second messenger system. The expressed protein was biotinylated with high efficiency; more than 90% of TSHR-BIO was bound to streptavidin. We have shown the application of streptavidin agarose immobilized TSHR-BIO for the detection of thyroid-binding inhibiting immunoglobulines in unfractionated sera. There was a good positive correlation between the results obtained in this assay and the commercially available TRAK assay performed with solubilized porcine TSH receptor (r = 0.71; p < 0.001, in 45 sera of patients with Graves' disease and 17 normal sera).

Differential effects of dopamine on glucoregulatory hormones in rats.

The effect of dopamine at different doses on serum concentrations of insulin, glucose and corticosterone and on plasma glucagon concentration was investigated in rats. Dopamine was given intravenously over 6 h with infusion rates of 2.5, 7.5, 15, and 60 micrograms/kg.min and in combination with phentolamine. Serum insulin concentration was unchanged at low doses of dopamine. It was significantly increased from 6.0 +/- 0.7 ng/ml to 13.7 +/- 2.3 ng/ml (P less than 0.01) when 7.5 micrograms/kg.min of dopamine were used, whereas it was significantly depressed to 3.96 +/- 0.89 and to 4.0 +/- 0.34 ng/ml (P less than 0.01), respectively, at the high doses of dopamine. This latter effect could be reversed to 6.7 +/- 1.19 ng/ml and inverted to 9.2 +/- 1.7 ng/ml (P less than 0.01) by simultaneously applied phentolamine at appropriate dosages. Serum glucose levels were markedly elevated from 154 +/- 7 to 234 +/- 42 mg/dl (P less than 0.01) by the higher doses of dopamine. A significant alteration of glucagon plasma concentrations from 18.9 +/- 2.8 to 42.3 +/- 14 pg/ml (P less than 0.01) was elicited only by 7.5 micrograms/kg.min of dopamine. The data clearly demonstrate that exogenous dopamine acts differently on glucose homeostasis according to the dosage. The study provides strong evidence that dopamine decreases insulin levels via alpha-adrenergic receptor stimulation. This effect may contribute to the deterioration of glucose homeostasis with high doses of dopamine.

Periodically hyperthyroid phenotype in thyroid hormone resistance is associated with mutation D322N in the thyroid hormone receptor beta gene: transcriptional properties of the mutant and the role of retinoid X receptor.

We report a point mutation in the ligand-binding domain of the TR beta 1 gene in an affected patient and his daughter. The phenotype was borderline hyperthyroid with periodic aggravation of symptoms. In the cognate variant TR beta (TR beta-CN) amino acid codon 322 was exchanged from aspartic acid to asparagine. TR beta-CN revealed strongly decreased T3-binding activity. At low T3 levels TR beta-CN transactivated a palindromic thyroid hormone response element (TRE-PAL) only to a limited extend, whereas full activity was retained at a high T3 concentration. At low T3 levels, TR beta-CN exerted a dominant negative effect on wild-type TR beta, whereas this effect was diminished in the presence of high T3 concentrations. TR beta-CN could not be activated by retinoid X receptor (RXR) beta in the presence of T3, whereas addition of 9cis-retinoic acid (9c-RA) resulted in the transactivation of TRE-PAL through RXR beta independently of the presence of TR beta-CN. In conclusion, the time dependent variable THR phenotype of patient CN might be influenced by the differential expression of RXRs and the T3 and 9c-RA hormonal status.

Detection of functionally different types of pathological autoantibodies against thyrotropin receptor in Graves' patients sera by luminescent immunoprecipitation analysis.

We describe a new method for the detection of different types of pathological autoantibodies against TSH receptor (TSHR) in Graves' patients sera by luminescent immunoprecipitation analysis. For this purpose three different chimeras composed of human TSHR and rat luteotropin/choriogonadotropin receptor (LH-CGR) were constructed, as was described previously (Tahara K, Ishikawa N, Yamamoto K, Hirai A, Ito K, Tamura Y, Yoshida S, Saito Y, Kohn LD. 1997 Thyroid 7:867-877). They were used in the immunoprecipitation reactions: (i) the wild type TSHR (for the detection of total TSHR autoantibodies), (ii) TSHR/LH-CGR chimera wherein TSHR amino acid residues 8-165 (epitopes for thyroid stimulating antibodies) are replaced by comparable LH-CGR residues, (iii) TSHR/LH-CGR chimera wherein TSHR amino acids 261-370 (epitopes for thyroid blocking antibodies) are replaced by comparable LH-CGR residues, and (iv) TSHR/LH-CGR chimera wherein TSHR amino acids 8-165 and 261-370 are replaced by comparable LH-CGR residues (for the detection of neutral TSHR autoantibodies). DNA encoding the N-terminal 725 (of 764) amino acids of wild type TSHR (or TSHR/LH-CGR chimera) was fused to the cDNA for the 550-amino acid firefly luciferase. The hybrid proteins were produced in HeLa cells using recombinant vaccinia viruses. All fusion proteins retained the enzymatic activity of firefly luciferase and TSHR-LUC interacted with TSH with the same affinity as wild type receptor. The luciferase tagged TSHR and TSHR/LH-CGR chimeras were used for the detection of different types of TSHR autoantibodies (i.e. total, neutral, thyroid stimulating and thyroid blocking) in 63 Graves' disease and 62 normal sera by immunoprecipitation analysis. The data demonstrated positive correlation between results of immunoprecipitation assay and results obtained using cAMP bioassay or assay for TSH binding inhibitory immunoglobulins in test sera.