PROJECT SUMMARY Adenovirus (AdV) and other human DNA viruses with limited genome size have maximized gene expression and coding potential through the use of alternative splicing and polyadenylation. The AdV double-stranded DNA genome has served as a powerful system for seminal discoveries in gene regulation and RNA processing. There are five early transcription units (E1A, E1B, E2, E3 and E4) which are transcribed by host RNA polymerase II and processed by cellular machinery. Each viral transcription unit is driven by a designated promoter and gives rise to multiple mRNAs generated by alternative splicing. The canonical proteins produced from each viral transcript have been assigned functions through the study of infection with viral mutants. The E2 region encodes three proteins involved in viral DNA replication. The E4 region encodes multifunctional proteins involved in regulation of transcription, splicing and translation of viral mRNAs, as well as antagonizing intrinsic cellular defenses. We combined short-read and direct long-read sequencing technologies to define viral transcripts during AdV infection, and thus generated the most complete annotation of the virus transcriptome. Our analysis identified a number of non-canonical chimeric proteins that fuse polypeptides from different early transcriptional units. Here we focus on one of these unexpected gene products that we have designated E4orf6/DBP since it spans the E2 and E4 regions. Our preliminary data show that this novel chimeric protein is robustly expressed, and we propose that it could provide a yet unidentified function during AdV infection. In Aim 1 we will evaluate how the transcript is generated, and the importance of its interacting partners. In Aim 2 we will determine functions of the fusion protein, and its impact on infection. We have generated or accumulated all reagents and technical expertise necessary to define the functions of the E4orf6/DBP chimeric protein. Our findings challenge the concept of discrete early transcriptional units with their own designated promoter and poly(A) site, and suggest that studying individual open reading frames may not reflect the true complexity of the viral transcriptome. Our studies aim to uncover molecular mechanisms that regulate virus-host interactions by providing a deeper understanding of non-canonical fusion proteins that expand the diversity of the viral proteome. The proposal will generate a model approach for interrogating functions of non-canonical chimeric proteins generated by viruses with complex viral proteomes.