Résumé : Circulating TSH bioactivity may vary in several clinical and experimental conditions. Since the reliability of the current methods for the measurement of TSH bioactivity is limited, a new bioassay based on cAMP accumulation in Chinese Hamster Ovary cells transfected with recombinant human TSH receptor (CHO-R) was set up. The sensitivity was 0.3 +/- 0.1, 0.4 +/- 0.1 and 0.01 +/- 0.01 micrograms/L for TSH IRP 80/558, recombinant human TSH and bovine TSH, respectively. Standard curves were parallel, and the intra- and inter-assay coefficients of variation were 13 +/- 1.1% and 22 +/- 1.9%, respectively. LH, FSH, CG and TSH subunits did not stimulate cAMP accumulation up to high concentrations. Circulating TSH was partially purified by immunoaffinity separation and concentrated before being bioassayed. However, plain sera with high TSH levels, such as those from primary hypothyroid patients (PH), could be directly tested in CHO-R bioassay, provided that sera were added at concentrations lower than 10%. TSH from 6 normal subjects had biological to immunological ratio (B/I) ranging from 0.6 to 2.1 (mean +/- SD = 1.4 +/- 0.5). TSH from 6 patients with PH showed bioactivity significantly lower than in normals (B/I = 0.6 +/- 0.3; p < 0.001; range = 0.3-1.1). TSH from 5 patients with central hypothyroidism of hypothalamic origin (CH) had undetectable basal bioactivity (B/I < 0.2), which normalized in only one patient after acute TRH and in all patients after chronic TRH administration. In conclusion, CHO-R cells provide an excellent tool for evaluating TSH bioactivity, owing to high sensitivity, specificity, reproducibility and feasibility of the assay.(ABSTRACT TRUNCATED AT 250 WORDS)