A significant and global decrease in viral gene expression occurs with deletion of the carboxyl terminal domain of the human cytomegalovirus essential protein pUL34

The human cytomegalovirus UL34 gene encodes sequence-specific DNA-binding proteins and is essential for viral replication. Functionally, UL34 proteins (pUL34) repress expression of the viral immune evasion genes, US3 and US9. The amino terminal domain of pUL34 is sufficient for nuclear localization, DNA-binding and transcriptional repression. A virus lacking the entire carboxyl terminal of pUL34 was replication competent; however, deletion of the carboxyl terminal domain significantly impaired viral replication. There was a significant decrease in the transcription of viral genes at 4 hours post-infection following infection with the mutant virus which lacks the entire carboxyl terminal end of UL34 (UL34-COOH) relative to wild type virus, and an associated decrease in protein expression levels. Viral gene expression for the mutant virus was also decreased at 48 hpi but to a much smaller degree. RNA was harvested from human diploid fibroblasts at 4 and 48 hours post-infection with either HCMV AD169 (derived from the BAC pHB5) or from the virus containing a carboxyl terminal truncation of UL34. Cells were infected at a multiplicity of infection of 1 plaque forming unit/cell. The infections and RNA extractions were performed in duplicate experiments. RNA was purified using Direct-zol RNA purification columns (Zymo Research, Irvine, CA, USA) or Qiagen RNA easy kits. For RNA-seq analysis, strand-specific RNA library construction and sequencing were performed on duplicate biological samples for each virus and time point by Admera Health, South Plainfield, NJ and the team at Admera Health LLC (via Science Exchange). Over 30 million reads were obtained for each sample. RNA-seq results were analyzed using CLC Genomics Workbench (Qiagen); reads were aligned against a combined genome comprised of the human and the HCMV AD169 genomes. TPM (transcripts per kilobase million) were determined for each sample; average TPM and p-values were determined using EDGE analysis. Genes with fewer than 20 unique reads were discarded as background.