Conservation of DNA sequence in the predicted major late promoter regions of selected mastadenoviruses.

The major late promoter (MLP) of the subgroup C human adenoviruses is a preeminent model for the study of the mechanisms of basal and activated transcription, both in vivo and in vitro. However, while the structure and function of the human virus MLP has been the subject of extensive investigation, the conservation of the various promoter elements among the adenoviruses from different species has not been examined. Conservation of specific elements would strongly suggest the importance and universality of their function. To address this issue, sequences were obtained from cloned DNAs of several representative Mastadenoviridae, mouse adenovirus type 1 (MAV-1), Tupaia adenovirus type 1 (TAV-1), and two bovine adenoviruses of two distinct subgroups, BAV-3 and BAV-7. The results of the sequencing studies showed that the TATA box and an upstream inverted CAAT box are conserved in all species and that the binding site for transcription factor USF is present in all except MAV-1, in which a sequence similar to an Sp1-binding site is present at a similar position. The initiator element (INR) sequence is not well conserved, and only one or other of the two downstream activating elements, DE1 and DE2, is predicted to be present in the nonprimate virus MLP regions. Ribonuclease protection assays on RNA isolated from MAV-1-infected cells late in infection indicated that the predicted MLP is functional, and transcription initiation and splice donor sites were identified. The human virus MLP is embedded in the essential DNA polymerase sequence on the opposite DNA strand. The primary amino acid sequences of the C-terminal regions of the predicted DNA polymerases show strong conservation of sequence motifs observed in replicative polymerases ranging from prokaryotes to mammals, and additional regions of strong conservation among the adenovirus polymerases. Pairwise comparisons between the newly sequenced regions of the polymerases and previously published sequences show that BAV-7 is most dissimilar to all others, while TAV-1 has a greater similarity to the primate sequences than to the others. The sequence data from both strands were also used to construct phylogenetic trees, based on BAV-7 as the outgroup. The trees constructed from the two sets of sequences are broadly similar, showing close relationships between primate viruses, but differing in the order of divergence of TAV-1 and MAV-1 branches.