In vitro 3,5,3'-triiodothyronine binding to rat cerebrocortical neuronal and glial nuclei suggests the presence of binding sites unavailable in vivo.

The maximal binding capacity (MBC) of the rat cerebrocortical nuclear T3 receptor, as determined by in vivo saturation techniques, is approximately half that measured in vitro on isolated nuclei or solubilized receptors. To investigate this disparity, the MBC values determined in vivo and in vitro for both rat cerebral cortex and liver were compared, taking into account nuclear receptor loss or inactivation and the presence of endogenous T3. By Scatchard analysis of T3 binding to isolated nuclei in vitro at 37 C, the uncorrected MBC values (mean +/- SEM; n = 3) for the cerebrocortical nuclear T3 receptor in euthyroid and hypothyroid rats were 0.80 +/- 0.14 and 0.66 +/- 0.07 ng T3/mg DNA, respectively, and were not significantly different. The Kd values were also not significantly different (5.6 +/- 0.3 and 5.2 +/- 0.9 X 10(-10) M, respectively). After corrections for incomplete dissociation and receptor inactivation under the in vitro conditions, the overall mean MBC increased by approximately 33% to 0.97 ng T3/mg DNA, or about 3.6 times the in vivo MBC. In addition, cerebrocortical nuclei prelabeled in vivo with +/- 131I]T3 at near-saturating levels and subsequently incubated with [125I]T3 in vitro at concentrations up to 10 times the Kd were shown to bind as much as 4 times more T3 in vitro relative to the amount of endogenous hormone which dissociated, thus exceeding the in vivo MBC by a factor of two. Parallel experiments with isolated liver nuclei did not show the existence of nuclear T3 receptors which were available only in vitro, even when the corrected MBC (0.77 ng T3/mg DNA) was compared with the MBC obtained by the in vivo saturation technique (0.76 ng T3/mg DNA). The experiments with liver nuclei were done at 25 C to reduce the rate of inactivation or loss of nuclear T3 receptors in this tissue. By fractionating isolated cerebrocortical nuclei into neuronal and glial subpopulations on discontinuous sucrose gradients, the high affinity, limited capacity nuclear T3 receptor could only be identified in the neuronal fraction. No consistent specific binding of T3 was observed in glial nuclei that were 80% pure, suggesting that either glial cells in the adult rat are not likely to be direct targets of thyroid hormone or that thyroid hormone may act via nonnuclear receptor-mediated pathways. We conclude that only neurons have specific high affinity, limited capacity nuclear T3 receptors and that as many as half of these receptors may not be accessible to plasma T3.(ABSTRACT TRUNCATED AT 400 WORDS)