Characterization of Cytoplasmic T₃Binding Sites by Adsorption to Hydroxyapatite: Effects of Drug Inhibitors of T₃and Relationship to Glutathione-S-Transferases

To facilitate studies of thyroid hormone (T3) binding to cytoplasmic proteins, we prepared monkey (M. fascicularis) liver cytosol (100,000g supernatant) and examined T3 binding using hydroxyapatite (HAP) separation. HAP adsorbs cytoplasmic and nuclear binding sites but not serum T4 binding proteins. Cytosol was incubated with [I-125]T3 for 30 min at 4-degrees-C and separated by adding an equal volume of HAP (15 g/100 mL). After a further incubation of 10 min, the HAP pellet was washed three times in buffer containing Triton X-100, 0.5%. With this method, a single class of T3 binding site was observed with K(d) 15.8 +/- 1.2 nM, concentration 0.62 +/- 0.17 pmol/mg protein (n = 3, xBAR +/- SD). We used this assay to assess potential drug inhibitors of cytoplasmic binding and to evaluate the proposal that glutathione-S-transferases (GST) and cytoplasmic T3 binding proteins are identical. Displacement of [I-125]T3 by unlabeled iodothyronines relative to T3 (100) was T4 58, Triac 7, rT3 7, Tetrac less-than-or-equal-to 1. This hierarchy indicates that this binding site is distinct from nuclear or serum binding sites. T3 binding was displaceable by nonsteroidal anti-inflammatory drugs (NSAID) and nonbile acid cholephils (NBAC). Half-inhibitory concentrations (mu-M, xBAR +/- SD, n greater-than-or-equal-to 3) were diclofenac 4.9 +/- 1.3, mefenamic acid 13.6 +/- 0.6, bromosulphthalein 45 +/- 3, iopanoic acid almost-equal-to 200. Amiodarone and furosemide were inactive up to 100-mu-M. No displacement was observed with cortisol or the bile acid taurocholate, up to 100-mu-M. Dithiothreitol, 5 mM, did not change binding affinity or capacity. These results suggest that T3 cytosolic binding protein may also bind NSAID and NBAC. Lack of furosemide and amiodarone competition with T3 indicates that the T3 binding site is distinct from that in serum and nucleus. Lack of effect by cortisol, taurocholate, and DTT does not support the suggestion that the binding protein and GST are the same. We conclude that this technique identifies a distinct cytoplasmic T3 binding site and that NSAID and NBAC may influence cytoplasmic binding of T3 in vivo. The magnitude of this effect will depend on their free cytoplasmic concentrations.