|
In order to test this hypothesis we considered it important to
identify specific brain genes which could be shown to be
responsive to thyroid hormone. Although thyroid hormone affects
multiple cell types, the hypothyroid state exerts an especially
striking effect on cerebellar Purkinje cells, resulting in
profound shortening of the dendrites. To study this phenomenon we
compared the initial expression of three candidate genes in
hypothyroid and euthyroid pups: Pcp-2, a protein of unknown
function unique to the Purkinje cells, and calbindin and the IP3
receptor, genes which are predominantly but not exclusively
expressed in Purkinje cells. These genes were generously supplied
to us by Dr. Harry Orr of the Human Genetics Institute of the
University of Minnesota. Lastly, we compared the effect of T3 on
the transcription of these genes to the response of the myelin
basic protein gene which is expressed in oligodendrocytes and had
previously been shown by Farsetti et al. to be regulated by T3.
Pups were rendered hypothyroid by the administration of
methimazole to the dam and euthyroid by the administration of one
microgram of T3. In each case, the initial expression of the gene
was very low. After a lag time of 2-3 days, T3 appeared to
accelerate the expression of each gene. However, regardless of
the specific treatment, the mRNA level eventually rose to
comparable levels. The potential developmental importance of the
initial lag is readily apparent. The first three weeks of
postnatal life are characterized by the appearance of a series of
rapid, complex but highly integrated and coordinated biological
changes. A delay in gene expression during this period could
exert profound influence on the ultimate neurological status of
the animal.
We focused our initial attention on Pcp2. We first transfected the cDNA supplied into CHO cells. Although addition of T3 to the medium resulted in a statistically valid increase in expression, the fold- increases were minimal. Regardless of these disappointing findings, we analyzed the upstream region of Pcp-2 and selected regions which contain DNA sequences suggestive of TREs. Three such sequences were identified and designated as A1, A2, and B1, the latter being situated in an intron. Of interest were the results of experiments in which the TRE candidates were inserted upstream of a MMTV promoter. These studies showed that A1 and B1 could under appropriate circumstances act as TRE since in the presence of T3 there was a substantial fold increase in reporter expression, comparable to that produced by the insertion of TREpal. However B1 was not effective. Gel shift experiments suggested that the increased expression of the CAT reporter in transfection experiments was associated with the binding of the receptor to the DNA fragment. This was supported by the finding that addition of TRb1 translation product plus brain extract (for RXR) generated two shifted bands (? dimers). Further, addition of TREpal displaced the shifted band by competing for the receptor. Similarly, A1 but not A2 displaced the bands by competition. These results indicated that A1 but not A2 functions as a TRE. These studies suggest that the regulation of brain gene expression during development may, at least in some instances, involve interactions between T3 receptors and specific TRE's situated on target genes. |
