'Brain-specific' transcription and evolution of the identifier sequence.

A recent model for the transcriptional control of gene expression in neural cells involves a dispersed repetitive DNA sequence termed the identifier (ID) sequence. However, the model is based on circumstantial evidence from studies on rat brain gene expression. Furthermore, available data are complicated by observations from several laboratories which suggest that the ID sequence is a family of mobile genetic elements. Although this does not preclude a role for some family members in regulating gene expression, the contention that these sequences are transcribed tissue-specifically is not proof of such a role. We have now measured the genomic copy number and tissue pattern of transcription of ID sequences in the rat, mouse and hamster, and have found that ID-homologous, BC1-like RNAs are restricted to brain in all three species, but that ID-homologous transcripts occur in total cellular RNAs of brain, liver and kidney of all three organisms. The genomic copy number of the ID sequences varies over two orders of magnitude between these species. Our data suggest that most ID sequences in these genomes are dispersed at random with respect to transcription units. A cis-acting, transcriptional-level controlling role for the ID therefore seems unlikely.