Comparison of binding of bovine and human thyroid-stimulating hormone to receptor sites on human thyroid membranes

The binding of 125I-labelled bovine TSH (bTSH) to a wide range of human thyroid membrane preparations was compared with that of 125I-labelled human TSH (hTSH). Much higher binding percentages were obtained with the 125I-labelled bTSH. This was because the receptors had a higher binding affinity for bTSH than for hTSH. No differences in tracer purity, nor differences in optimal conditions for the binding of bTSH or hTSH, nor tracer degradation contributed significantly to the better binding of 125I-labelled bTSH. Good correlation was found between binding percentages for 125I-labelled bTSH and 125I-labelled hTSH over the range of thyroid specimens. Useful information on human TSH receptors is, therefore, obtainable from binding studies with 125I-labelled bTSH. The TSH displacement curves yielded linear Scatchard plots whenever the tracer and displacing hormones were of the same species. The data were, therefore, consistent with a simple binding reaction between TSH and a single set of independent receptor sites.     

ANALYSIS OF TSH RECEPTORS AND MICROSOMAL ANTIGEN IN DIFFERENT HUMAN THYROID TISSUE SPECIMENS

Affinity labelling with a 125I-labelled photoactive derivative of TSH (HSAB-TSH) was used to analyse TSH receptor size in the following specimens of human thyroid tissue: (1) cold nodules; (2) autonomous nodules; (3) papillary carcinoma; (4) medullary carcinoma; (5) metastasis of papillary carcinoma to lymph node; (6) anaplastic carcinoma, and (7) Graves' thyroid. In addition, a sample of histologically normal thyroid tissue surrounding specimens 1-4 was analysed in each case. Thyroid microsomes were also prepared from the tissue samples, solubilized using 1% deoxycholate and labelled with 125I. The preparations were immunoprecipitated using microsomal autoantibodies and protein A and analysed by SDS-PAGE and autoradiography. These studies indicated that no differences in the characteristics of the TSH receptor or of microsomal antigen were observed in the tissue samples 1-3 and 7. Neither protein was detected in tissue specimens 4-6.     

The interaction of TSH receptor autoantibodies with 125I-labelled TSH receptor.

Detergent-solubilized porcine TSH receptor (TSHR) has been labeled with 125I using a monoclonal antibody to the C-terminal domain of the receptor. The ability of sera containing TSHR autoantibody to immunoprecipitate the labeled receptor was then investigated. Sera negative for TSHR autoantibody (as judged by assays based on inhibition of labeled TSH binding to detergent-solubilized porcine TSHR) immunoprecipitated about 4% of the labeled receptor, whereas sera with high levels of receptor autoantibody immunoprecipitated more than 25% of the labeled receptor. The ability to immunoprecipitate labeled TSHR correlated well with ability of the sera to inhibit labeled TSH binding to the receptor (r = 0.92; n = 63), and this is consistent with TSHR autoantibodies in these samples being directed principally to a region of the receptor closely related to the TSH binding site. Preincubation of labeled TSHR with unlabeled TSH before reaction with test sera inhibited the immunoprecipitation reaction, providing further evidence for a close relationship between the TSHR autoantibody binding site(s) and the TSH binding site. This was the case whether the sera had TSH agonist (i.e., thyroid stimulating) or TSH antagonist (i.e., blocking) activities, thus, providing no clear evidence for different regions of the TSHR being involved in forming the binding site(s) for TSHR autoantibodies with stimulating and with blocking activities. The ability of TSHR autoantibodies to stimulate cyclic AMP production in isolated porcine thyroid cells was compared with their ability to immunoprecipitate labeled porcine TSHR. A significant correlation was observed (r = 0.58; n = 50; P < 0.001) and the correlation was improved when stimulation of cyclic AMP production was compared with inhibition of labeled TSH binding to porcine TSHR (r = 0.76). Overall, our results indicate that TSHR autoantibodies bind principally to a region on the TSHR closely related to the TSH binding site, and this seems to be the case whether the autoantibodies act as TSH agonists or antagonists.     

AN IMPROVED RADIORECEPTOR ASSAY FOR TSH RECEPTOR ANTIBODIES

An improved receptor assay for TSH receptor antibodies is described in which detergent-solubilized TSH receptors and ¹²⁵I-labelled TSH are used. Normal human serum and immunoglobulin concentrates from normal serum showed little effect on the interaction between labelled TSH and detergent-solubilized receptors whereas immunoglobulin concentrates from some Graves' sera caused marked inhibition of TSH binding. Precipitation with polyethylene glycol was the most convenient way of preparing immunoglobulin concentrates and using this technique the assay could be completed in a few hours. The coefficient of inter assay variation at 51% inhibition of labelled TSH binding was 3.7% and analysis of serum samples from patients with Graves' disease (n= 42), Hashimoto's disease (n= 26), multinodular goitre (n= 9), rheumatoid arthritis (n= 10) and normal donors (n= 35) suggested that the assay could detect TSH receptor antibodies in about 80% of patients (treated and untreated) with Graves' disease.     

A functional site on the human TSH receptor: a potential therapeutic target in Graves' disease

OBJECTIVE: Identifying sites on the TSH-receptor that are involved in the pathological stimulation of the thyroid by autoantibodies in Graves' disease would aid the development of new therapies. We tested a series of monoclonal antibodies that recognize the native receptor for their ability to inhibit stimulation of the receptor in vitro. PATIENTS AND METHODS: Heterologous cells expressing the recombinant human TSH-receptor were stimulated with TSH or serum samples from 13 Graves' disease patients or the MRC Long-Acting Thyroid Stimulator standard B (LATS-B) and their cAMP responses measured. The effect on this stimulation of various doses of purified monoclonal antibodies with defined epitopes was determined. RESULTS: Antibodies against one epitope (residues 381-384) inhibited TSH-stimulated cyclic adenosine monophosphate (cAMP) production (1 microg/ml causing 50% inhibition of the response to 100 microU/ml TSH) and also inhibited cAMP production induced by sera from approximately 40% (6/14) of Graves' disease patients, including the MRC LATS-B standard. CONCLUSIONS: Residues 381-384 of the human TSH-receptor are important in the physiological and pathological stimulation of the thyroid. This opens the possibility of more specific therapy of some Graves' disease patients by agents directed against this epitope.     

Coexistence of autoantibody to human thyrotropin (TSH) and autoanti-idiotypic antibody to antihuman TSH antibody in a case with simple goiter

A 70-yr-old woman with simple goiter showed normal serum levels of T4, T3, free T4, TSH receptor antibody (TRAb) and increased TBG. Discrepancy in serum hTSH level was observed by different assay methods. Coexistence of both autoantibodies for hTSH and for anti-hTSH antibody were demonstrated by the reaction of the patient's antibody with both 125I-hTSH and 125I-anti-hTSH (monoclonal antibody; mAb). These two autoantibodies belong to the polyclonal immunoglobulin G (IgG). The autoantibody for hTSH recognized only beta-subunit of hTSH. Neither stimulating type of TRAb in Graves' disease nor blocking type of TRAb in primary hypothyroidism interfered with the binding of the patient's antibody to 125I-hTSH or 125I-anti-hTSH. Anti-idiotypic antibody (anti-ID antibody) for anti-hTSH antibody was purified by anti-hTSH antibody affinity chromatography. The binding reaction of 125I-anti-hTSH (mAb) with this anti-ID antibody could be inhibited by the unlabeled hTSH. This anti-ID antibody might represent the internal image of the nonbiological active site of TSH molecule, because of absence of thyroid stimulating activity. Goiter in this patient may have occurred by the unbound TSH with IgG (free TSH) and the bound TSH with IgG, because TSH levels in both the whole serum and the IgG free serum (the unbound TSH with IgG) were decreased significantly by T4 treatment. Coexistence of these antibodies may participate in the autoimmune mechanism of an idiotype-anti-idiotype network.     

TSH RECEPTOR BINDING AND THYROID STIMULATION BY SERA FROM PATIENTS WITH GRAVES'' DISEASE

An investigation of the ability of TSH receptor antibodies to bind to the TSH receptor and stimulate thyroid function is described. Binding studies were carried out using 125I-labelled TSH and detergent solubilised porcine TSH receptors, and the parameter of thyroid stimulation employed was 125I-organification in isolated porcine thyroid cells. Different Graves' sera were found to show different dose-response relationships and three types of antibody activity were evident in the six samples studied. In two samples receptor binding and thyroid stimulating activities were approximately parallel. In two receptor binding was detectable at lower doses than stimulation while in the last two stimulation was detectable at lower doses than binding. It is proposed that these characteristics are due to different populations of receptor antibodies which exhibit different degrees of TSH agonism. Furthermore one of the reasons for the relatively poor correlation between TSH receptor binding and thyroid stimulation observed in series of individual Graves' sera may be the presence of different populations of antibodies showing different dose-response relationships with regard to binding and stimulation.     

A RECEPTOR ASSAY FOR THE MEASUREMENT OF TSH RECEPTOR ANTIBODIES IN UNEXTRACTED SERUM

A receptor assay for TSH receptor antibodies is described in which unextracted serum, detergent solubilised TSH receptors and 125I-labelled TSH are used. The assay was rapid and reproducible with coefficients of inter-assay variation of 12.3%, 2.1 and 2.6% at mean inhibition of TSH binding values of 11, 53, and 79 respectively. Assay sensitivity could be increased by reducing the volume of receptors used but some increase in the scatter of values obtained with individual normal sera was also observed. Comparison of human and porcine TSH receptor preparations indicated that porcine tissue gave greater sensitivity. Analysis of different groups of patients and normal subjects (n = 21) showed the absence of detectable TSH receptor antibody activity in 16 patients with rheumatoid arthritis, 10 with multinodular goitre and 12 with Hashimoto's disease. However the antibody was readily detectable in 28 out of 28 Graves' patients (treated and untreated) who were hyperthyroid at the time of assay.     

Receptors for insulin and insulin-like growth factor in cultured arterial smooth muscle cells depend on their growth state

The binding of 125I-labelled insulin and 125I-labelled insulin-like growth factor (IGF) to cultured arterial smooth muscle cells from rats was studied during various growth states of the cells. The level of binding of 125I-labelled insulin to the cells was low in growing cells and high in stationary cells. The level of 125I-labelled IGF binding to the cells was high in growing cells and low in stationary cells. In addition, the effect of unlabelled IGF and insulin on the binding of both 125I-labelled hormones to the cells was examined during various growth states. In growing cells insulin displaced 125I-labelled insulin from its binding sites; IGF competed weakly with 125I-labelled insulin for the binding sites. In parallel, IGF displaced 125I-labelled IGF binding whereas insulin competed weakly with 125I-labelled IGF for the binding sites. In stationary cells both hormones displaced 125I-labelled IGF binding. Insulin-like growth factor also displaced 125I-labelled insulin binding whereas insulin could not significantly displace 125I-labelled insulin from the binding sites. Insulin only competed with 125I-labelled insulin for the binding sites after removal of the fetal calf serum from the culture medium.     

Thyrotropin (TSH) receptor modulation by specific TSH receptor antibodies in Graves' disease

In Graves' disease (GD), an antireceptor autoantibody disease, individual variability in the pathogenic interaction between TSH receptors and autoantibodies has been reported. This variability can be due to allotypic (person to person) variability in the receptors or differences in autoantibody amount or specificity. This fundamental issue was investigated by evaluating immunoglobulin G (Ig)-induced TSH receptor modulation in thyroid tissue from 19 patients with GD. TSH receptor modulation by Graves' Ig was defined as the appearance of 1 class of high affinity binding sites, instead of the usual 2 classes of binding sites. Ig-induced modulation of receptors occurred in 9 of 19 (47%) experiments with autologous (patient's own) tissues and correlated with the presence of TSH receptor antibodies, measured as TSH binding inhibitor Igs. Of these 9 receptor-modulating Graves' Ig preparations, 7 (78%) also had a receptor-modulating effect in other patient's (homologous) thyroid tissue. Nine of the 10 Graves' Ig preparations that were negative for TSH receptor-modulating activity in autologous thyroid tissue were tested with other patients' thyroid tissues; 7 (78%) were negative, and all were TSH binding inhibitor Ig negative. We conclude that variability in the occurrence of TSH receptor modulation was associated with the presence or absence of TSH-binding inhibitor Ig. No evidence for allotypic differences in TSH receptors in GD was found.     

Iodination of thyroid-stimulating hormone for receptor-binding studies with human thyroid membranes: effects of specific activity and method of iodination

Three different methods were compared for 125I-labelling of thyroid-stimulating hormone (TSH) for use in receptor-binding studies with human thyroid membranes: these were the chloramine-T, Bolton-Hunter and lactoperoxidase methods. Chloramine-T proved to be an inferior method to the other two. Iodinations to different specific activity (0.7--9.4 Bq/pg) were also compared: too high a specific activity led to reduced binding and a dramatic shift in the pH optimum for the TSH-receptor interaction. A specific acitivity of 2.5 Bq/pg should not be exceeded if binding of 125I-labelled TSH is to be representative of the binding of the natural hormone. Under these conditions, pH 7.5 was optimal for binding of TSH to its receptor. Repurification of the labelled TSH by receptor adsorption also proved to be essential.     

Affinity purification and diagnostic use of TSH receptor autoantibodies from human serum

Purification of TSH receptor autoantibodies (TRAb) from the serum of patients with Graves' disease (GD) might help to elucidate the nature of these disease causing autoantibodies. We describe here for the first time the successful affinity purification of human TRAb.Affinity purification was performed in a four step procedure with human recombinant TSH receptor (TSH-R) expressed in K562 cells. Purification from six different serum pools from patients with GD and two individual sera (one with only thyroid stimulating antibodies (TSAb) one with only thyroid blocking antibodies (TBAb)) resulted in a purity of 39.2+/-3.8 IU/mg TRAb or 25.7+/-2.1 microg IgG/IU (about 3.5-13.7 microg TRAb/ml serum). The average enrichment based on the respective original serum was 3420-fold (range 1200-10,000). The kDa of the purified TRAb were in the range of 0.7-2.6 x 10(-10)M. All purified TRAb (except from the TBAb serum which showed blocking activity) showed a more than 1000-fold stronger stimulation in the TSAb bioassay based on the IgG content than the original serum, and similar stimulation based on international units (IU/l) TRAb. When labelled purified TRAb were used in a competitive assay as tracer instead of bovine TSH, their binding to the human recombinant TSH-R on tubes was displaced by 99 of 100 GD sera (selected for TBII activity). Correlation to the standard TSH tracer was r=0.92. Interestingly, the use of TRAb tracer derived from a patient with TSAb and a patient with TBAb gave virtually identical results (r=0.93) with these patients, suggesting similar if not identical binding sites for both TRAb subtypes.In conclusion, this is the first report on the purification of human TRAb from the serum of patients with GD. The purified TRAb are of low concentration with high affinity, strong TBII and TSAb activity. Further characterisation may allow new insights in TRAb epitope localisation, the pathology of GD and the differences between TSAb and TBAb. Also, their use as tracer in a competitive assay is the first report on a completely homogenous assay with high sensitivity for TSH-R autoantibodies.     

The thyrotrophin receptor in human thyroid plasma membranes: effect of serum immunoglobulins.

A homologous receptor assay system using human thyroid membranes and 125I-labelled human TSH (hTSH) was used to study the effect of serum and serum fractions on the binding of [125I]hTSH to the membranes. Scatchard analysis showed a single population of binding sites for TSH. Gamma globulin fractions prepared from sera of patients with Graves' disease were able to displace [125I]hTSH from the membrane to a greater extent than normal gamma globulin in 21 out of 45 cases. Increased displacement activity was seen in patients with thyroiditis, hot nodules and euthyroid eye disease but not in patients with toxic multinodular goitres. Further fractionation of the gamma globulin fraction showed that the stimulatory activity was not confined to the IgG fraction. Scatchard plots showed gamma globulin fractions decreased the number of receptor sites available for TSH binding but did not alter the affinity of the receptor for TSH. IgG fractions showed different slopes and intercepts and appeared to decrease the affinity of the receptor for TSH. LATS activity in human serum may be explained on the basis of these observations on the properties of the TSH receptor.     

Studies of the TSH radioreceptor assay.

We have examined several variables in the reagents and procedures used in the TSH radioreceptor assay, the binding of iodinated TSH to its thyroidal receptor. We found that iodinated bovine TSH (S.A. 30 U/mg) was more effectively bound to receptor than iodinated human TSH (S.A. 7.3 U/mg). Iodination of TSH was the Bolton-Hunter acylation method apparently prevented binding to TSH receptor. Surgically removed human thyroid tissue specifically bound 10.3 +/- 1.0 (mean +/- SEM) of added [125I]TSH, but post-mortem human thyroid bound only 3.9 +/- 0.4% of [125I]TSH (p less than 0.001). Maximal binding of [125I]TSH was found at pH 5.8. Many tissue preparations contained activity, possibly due to proteases, which inactivated TSH, and inclusion of a protease inhibitor, aprotinin, significantly increased specific binding.     

The inhibition by calmodulin of thyroid-stimulating hormone binding to epididymal fat, testis and thyroid membranes in the guinea-pig

Calmodulin inhibited 125I-labelled TSH binding to the membranes of various target tissues for TSH (thyroid, epididymal fat and testis) of the guinea-pig. This inhibition was abolished by adding EGTA (1 mmol/l). Calmodulin did not inhibit the binding of 125I-labelled epidermal growth factor (EGF) to these membranes. It is suggested that the inhibitory effect of calmodulin on the binding of TSH to the receptor is specific and that this mechanism is due to the direct binding of calmodulin to receptor membranes. The ability of calmodulin to bind to the membranes was calcium-sensitive while that of TSH was not. The binding of 125I-labelled calmodulin to these membranes increased significantly when the endogenous calmodulin in the membranes was removed by EGTA. It was not inhibited by a pure preparation of TSH, but it was inhibited by contaminated calmodulin in a crude TSH preparation. On the other hand, 125I-labelled TSH binding to these membranes did not change after the removal of endogenous calmodulin. In conclusion, exogenous calmodulin has an inhibitory effect on the binding of TSH but not of EGF to the membranes of guinea-pig thyroid, epididymal fat and testis.     

Thyrotrophin receptors in normal and neoplastic (primary and metastatic) canine thyroid tissue.

Thyrotrophin (TSH) is the conditional growth factor of thyroid epithelial cells. Abnormalities in TSH-receptor binding such as a low receptor number or low binding affinity may be a marker of thyroid carcinoma or metastases, or may exhibit a relationship with the functional variability of such tissues. The dog was used as a model to characterize TSH-receptor binding in normal thyroid tissues, naturally occurring thyroid neoplasms and distant metastases. In normal dog thyroid tissues, specific 125I-labelled TSH binding ranged from 2.7 to 15.5%, and low cross-reactivity with bovine LH (0.023%) was observed. One class of TSH-binding sites was found in eight normal thyroid tissues and 22 thyroid carcinomas; two normal thyroid tissues and one tumour exhibited two classes of binding sites. The concentration of binding sites was lower in the five carcinomas with reduced pertechnetate uptake (0.09 pmol/mg protein) than in the five thyroid neoplasms with increased uptake (0.19 pmol/mg) (P = 0.055). Compared with the original carcinoma tissues, TSH binding revealed a reduced binding affinity in eight out of eleven metastases. Two metastases showed a complete absence of TSH binding, suggesting that they were not dependent on TSH for growth. We conclude that one class of TSH-binding site is predominant in normal dog thyroid tissues and dog thyroid carcinomas. TSH could therefore contribute, at least in theory, to further growth of primary dog thyroid carcinomas. Secondly, assays measuring TSH binding may not be able to discriminate between malignant and benign dog thyroid tumours. TSH receptor number or affinity may be related to the functional variability of thyroid neoplasms.(ABSTRACT TRUNCATED AT 250 WORDS)     

The interaction between the TSH receptor and Graves' sera with TSH agonist or antagonist properties

The A subunit of the TSH receptor was prepared by reduction of human thyroid membranes with dithiothreitol, and partially purified by gel filtration. The ability of Graves' sera to inhibit TSH binding to the TSH receptor and to stimulate cyclic AMP release from isolated thyroid cells was abolished by incubation with crude and partially purified preparations of the A subunit.     

Recombinant human thyrotropin (TSH) receptor in a radioreceptor assay for the measurement of TSH receptor autoantibodies

We studied the suitability of using the recombinant human TSH receptor expressed in Chinese hamster ovary cells in a TSH binding inhibition (TBI) assay for autoantibodies against the TSH receptor. Purified immunoglobulin G (IgG) containing potent thyroid-stimulating immunoglobulin bioactivity competed for radiolabeled TSH binding to recombinant TSH receptor in parallel to inhibition by unlabeled TSH. Using polyethylene glycol-prepared IgG, this assay discriminated very well between sera from normal individuals [TBI, 0.98 +/- 0.04 (+/- SD); range, 0.92-1.08; n = 35] and patients with untreated Graves' disease (TBI, 0.49 +/- 0.23; range, 0.06-0.98; n = 93). Only four of the sera from the untreated Graves' patients were TBI negative (greater than or equal to 0.92), providing a sensitivity of 96%. In sera from patients with Graves' disease receiving antithyroid drug therapy, TBI was 0.63 +/- 0.18 (range, 0.19-0.97; n = 75). In this group of treated patients, 2 of 75 were TBI negative. Four of 12 patients with Hashimoto's thyroiditis (33%) were positive for TBI activity. All 18 patients with nonautoimmune thyroid diseases (toxic nodular goiter, single toxic adenoma, subacute thyroiditis, or thyroid cancer) were TBI negative. Correlation of the TBI values with thyroid-stimulating immunoglobulin bioactivity revealed a generally positive correlation (r = 0.31; P less than 0.05); however, there were many discrepancies among individual sera. TSI bioactivity was undetectable in all 4 patients with Hashimoto's thyroiditis who were TBI positive. These data indicate that the recombinant TSH receptor in Chinese hamster ovary cells is a suitable system for detecting TBI activity in the sera of patients with autoimmune thyroid disease.     

Absence of specific cell-surface binding of tissue plasminogen activator in uterine cells

Tissue plasminogen activator (tPA), an arginine-specific serine protease, is an oestrogen-regulated protein in uterine and breast cancer tissue. It contains a domain which shares homology with epidermal growth factor (EGF). The aim of the present study was to determine whether specific tPA receptors or EGF receptors mediate the binding of tPA to cells and whether tPA possesses intrinsic mitogenic activity. The binding of 125I-labelled tPA to rat uterine and liver membranes was shown to be non-specific and could not be displaced by unlabelled tPA or EGF. Furthermore, acid washing of cell membranes did not unmask specific tPA-binding sites. In contrast, 125I-labelled EGF binding to both rat uterine and liver membranes was displaced in a dose-dependent manner by unlabelled EGF, and Scatchard analysis of the binding data revealed dissociation constant (Kd) values of 2.4 and 0.71 nM respectively. Unlabelled tPA (up to 20,000-fold excess) did not displace 125I-labelled EGF binding to these membranes. A study of the binding of 125I-labelled tPA and 125I-labelled EGF to endometrial carcinoma cells (Ishikawa), cervical carcinoma cells (HOG-1) and vulval carcinoma cells (A431) showed that up to a 100-fold excess of EGF or a 1000-fold excess of tPA did not displace 125I-labelled tPA binding to these cells. In contrast, 125I-labelled EGF binding was displaced by unlabelled EGF (Kd values for Ishikawa and HOG-1 cells were 2.72 and 1.92 nM respectively) but not by unlabelled tPA (1000-fold excess).(ABSTRACT TRUNCATED AT 250 WORDS)     

ANTI-TSH ANTIBODY WITH HIGH SPECIFICITY TO HUMAN TSH IN SERA FROM A PATIENT WITH GRAVES'' DISEASE: ITS ISOLATION FROM, AND INTERACTION WITH, TSH RECEPTOR ANTIBODIES

A patient with thyrotoxic Graves' disease had an apparent measurable level of serum TSH (2.5 microU/ml) by double-antibody radioimmunoassay (RIA). The serum IgG bound with both [125I]human(h)TSH and [125I]bovine(b)TSH. The [125I]hTSH binding was more effectively displaced by human than bovine TSH, whereas [125I]bTSH binding was displaced exclusively by bTSH. Scatchard analyses revealed that [125I]hTSH binding showed two components, whereas [125I]bTSH binding had only one component. Serum TSH determined by RIA became undetectable 21 months after antithyroid drug treatment with a parallel decrease of [125I]hTSH binding IgG activity. Four thyrotrophin binding inhibitory immunoglobulins (TBII) from other patients did not interfere with the binding of the patient's serum to [125I]h- or bTSH. Furthermore, the in-vitro thyroid stimulating activities of three thyroid stimulating antibodies (TSAb) were not affected by the addition of this patient's IgG. On the other hand, this patient's Ig (3 mg/ml) abolished the in-vitro thyroid stimulation by bTSH (100 microU/ml), but did not affect that by hTSH (100 microU/ml). The anti-hTSH antibody, TSH receptor antibody and anti-bTSH antibody in the serum, which contains TSAb as well as anti-TSH antibodies, could be partially purified by hTSH-agarose and subsequently by guinea pig fat cell membrane affinity absorptions. However, the anti-hTSH antibody fraction obtained had both hTSH binding activity and thyroid stimulating activity, and this fraction did not show any inhibitory effect on the in-vitro thyroid stimulation of autologous TSH receptor antibody or hTSH. The possible significance of anti-TSH antibodies is discussed.