Studies of Nuclear 3,5,3′-Triiodothyronine Binding in Primary Cultures of Rat Brain*

Abstract

Primary cultures of enzymatically dispersed cells from 17-day-old fetal rat cerebral hemispheres were used to detect the presence of nuclear T3 receptors. Cells grown in Minimum Essential Medium supplemented with 10% fetal bovine serum were grown in parallel with cytosine-arabinoside (ARA-C)-treated counterparts which had been exposed to the antimetabolite for 18 h on culture days 3 and 5 or 4 and 6. Five days after the second ARA-C treatment, phase contrast photomicrographs showed substantial loss of the proliferating basal cells, corresponding to an 85% decrease in cell number. Immunocytochemical studies using antiglial fibrillary acidic protein (anti-GFAP) and antineurofilament (anti-NF) antisera demonstrated loss of GFAP-positive cells (astrocytes) and preservation of NF-positive cells (neurons), the latter considered to be a nondividing population under the culture conditions. Nuclei obtained from the brain cell cultures at this time by Triton washing bound T3 with properties similar to those observed in vivo. Scathcard analysis showed a single, high affinity, limited capacity nuclear T3 receptor with a maximal binding capacity (MBC) of 0.53 .+-. 0.10 ng T3/mg DNA and a Kd of 0.19 .+-. 0.02 nM (.+-. SE; n = 4). ARA-C treatment resulted in a mean decreased in DNA per culture dish of 54%, with an accompanying 2-fold enrichment of the MBC, and no change in the Kd. In untreated cultures grown for 20 days, DNA per dish increased until day 14 and subsequently remained stable at approximately 100 .mu.g/dish. The MBC also increased from days 0 to 7, and remained stable until day 14. On day 20, the MBC had declined by approximately 60% to 0.21 ng T3/mg DNA, at which time the neuron population was decreased. The extracted nuclear receptor from brain cell cultures had a sedimentation coefficient of 3.6S. The relative binding affinities of the nuclear receptor for T3 and several analogs were identical to those found in vivo, making significant contamination of the nuclei with cytosolic or serum binding proteins very unlikely. Nuclei isolated from long term, neuron-free glial cell cultures failed to show any consistent high affinity saturable T3 binding. We conclude that 1) primary brain cell cultures of dispersed fetal rat cerebral hemispheres contain nuclear T3 receptors similiar in quantity, affinity, and specificity to those found in vivo; 2) the ARA-C-susceptible dividing cells in these cultures lack detectable nuclear T3 receptors and appear to be of glial origin, consistent with our earlier in vivo results; and 3) most, if not all, nuclear T3 receptors are in neurons, and the number of receptors per neuronal nucleus may increase over the first week in culture, approaching the quantity seen in pituitary. The assignment of nuclear receptors to neuronal cells is in agreement with our earlier conclusions from studies of adult rats and illustrates the potential of the primary cell cultures for the study of T3 effects on neural tissue.