Studies on the Cyclic AMP Response to Thyroid Stimulating Immunoglobulin (TSI) and Thyrotropin (TSH) in Human Thyroid Cell Monolayers zyxwvutsrqponmlkjihgfedcbaZYXWVUTSR B. Rapoport, S. Filetti, N. Takai, P. Seto, and G. Halverson zyxwvutsrqponmlkjihgfedcbaZYXWVUT Studies were conducted on the cultured human thyroid cell bioassay for thyroid stimulating immunoglobulin (TSI) and thyrotropin (TSH). In confirmation of the data of Kasagi, et al.,” incubation of human thyroid cells in Hank’s balanced salt solution deficient in NaCl increased the sensitivity of the c-AMP response to bovine TSH by approximately one order of magnitude. Half-maximal stimulation was attained at approximately 0.1 mU bTSH/ml. The effect of NaCl in the medium was greatest with stimulation by TSI > hTSH > bTSH. In contrast to incubations in NaCI( +) medium, with NaCI( -) medium most (70%~80%) of the c-AMP produced was released into the medium; this proportion remaining relatively constant over a wide range of bTSH and hTSH concentrations. At TSI concentrations higher than 3mg/ml efflux of c-AMP into the medium was significantly diminished. Stimulation by cholera toxin and prostaglandin E of thyroid cell c-AMP generation was not enhanced in NaCI( -) medium, in contrast to stimulation by TSH and TSI. The presence of lo-“M cycloheximide in NaCI( +) medium enhanced the c-AMP response to physiological concentrations of TSH. As with NaCIf- 1 medium, cycloheximide increased the sensitivity but not the maximum response of the c-AMP response to TSH. However no additivity was observed with NaCIf -1 medium and cycloheximide. Human thyroid cells obtained from patients with Graves’ disease are relatively insensitive to TSI stimulation. In NaCIf - ) medium, however, the sensitivity of these cells to TSI stimulation is sufftcient to enable them to be utilized in the TSI assay. The present state of the TSI assay is discussed. zyxwvutsrqponml R APID TECHNOLOGICAL progress is being made in the bioassay of thyrotropin (TSH) and thyroid stimulating immunoglobulin (TSI). Less than a decade ago the standard assay for these substances was the laborious, relatively imprecise and insensitive mouse bioassay.’ In recent years, there has been a proliferation of in vitro bioassays for TSH and TSI, providing, to different degrees, increased sensitivity, precision and convenience. The cytochemical assay is the most sensitive bioassay and is indeed even more sensitive than the TSH radioimmunoassay.* However this advantage is offset by its impracticality. Most of the new assays are based on the c-AMP response to agonist stimulation of thyroid tissue. At first, either thyroid slices’,4 or thyroid plasma membranes’-’ were used. Following the initial demonstration that dog thyroid cell monolayers offer the potential for a sensi- tive TSH bioassay,’ this system was subsequently adapted for this purpose.’ More recently, technical improvements in the culture of human thyroid cells have led to the establishment of human thyroid cell monolayer bioassays for TSI.‘&13 Despite these developments, however, a number of laboratories are attempting to improve further the sensitivity and convenience of the cultured human thyroid cell c-AMP assay. Recently, Kasagi et al.” made the important observation that the culture of human thyroid cell monolayers in medium lacking NaCl greatly improves the sensitivity of the assay to TSH and TSI stimulation.‘4*‘5 The present studies: (1) Confirm and provide additional information on the phenomenon described by Kasagi et al., (2) describe other means by which the sensitivity of the cultured cell assay may be improved, and (3) describe addi- zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCB Metabolism, Vol. 3 1, No. 1 1 (November), 1982 tional modifications of the cultured thyroid cell assay, combining features of our technique and those of Kasagi et al., that greatly improve the practicality of the assay. MATERIALS AND METHODS Cell zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCB Cultures Human thyroid cells were established in primary monolayer culture as previously described.” with the exception that I-3mm chunks of human thyroid tissue were cut with scissors, and the Stadie-Riggs slicing and Mcllwain chopping were omitted. In addition, collagenase 1 mg/ml is consistently used now. rather than Dispase. Cell dispersion time is now prolonged to 3-5 hr. The yield of viable cells is approximately l-2 x IO’ cells/g of tissue. Human thyroid cells, stored in liquid nitrogen, were thawed as previously described,” and plated in Falcon microtiter well plates (catalogue #3042) at 2-6 x IO’ cells per well. Crude IgG from serum was prepared by ammonium sulfate precipitation as pre- viously described” except that this material was dialyzed against NaCl-free Hank’s balanced salt solution, pH 7.4 (5 mM KCI, I .3 mM CaCI,, 0.4 mM MgSO,, 0.34 mM Na,HPO,, 0.44 mM KH*PO, and 0.1% glucose. This medium is referred to in the text as NaCI( -) medium. Incubations were in microtiter wells in NaCI( -) medium (0.2 ml) containing 20 mM Hepes, 2 mM 3-isobutyl- 1 -methylaxan- zyxwvutsrqponm From the Medical Service, Veterans Administration Hospital and the University of California Medical Center, San Francisco, California. Received for publication June 17. 1982. This work was supported by N.I.H. grant AM 19289, and the Medical Research Service of ihe Veterans’ Administration. Dr. Rapoport is rhe recipienl0fN.I.H. Career Development Award AM 00500. Address reprint requests lo B. Rapoport. Medical Service (I I IF) Veterans’ Administration Hospital, 4150 Clement Street, San Fran- cisco, California 94121. Q I982 by Grune & Sirarton. Inc. 0026&0495/82/31 I ILOOl5$OI .00/O 1159