Dissection of the C-terminal region of E1A redefines the roles of CtBP and other cellular targets in oncogenic transformation.

Human adenovirus E1A makes extensive connections with the cellular protein interaction network. By doing so, E1A can manipulate many cellular programs, including cell cycle progression. Through these reprogramming events, E1A functions as a growth-promoting oncogene and has been used extensively to investigate mechanisms contributing to oncogenesis. Nevertheless, it remains unclear how the C-terminal region of E1A contributes to oncogenic transformation. Although this region is required for transformation in cooperation with E1B, it paradoxically suppresses transformation in cooperation with activated Ras. Previous analysis has suggested that the interaction of E1A with CtBP plays a pivotal role in both activities. However, some C-terminal mutants of E1A retain CtBP binding and yet exhibit defects in transformation, suggesting that other targets of this region are also necessary. To explore the roles of these additional factors, we performed an extensive mutational analysis of the C terminus of E1A. We identified key residues that are specifically required for binding all known targets of the C terminus of E1A. We further tested each mutant for the ability to both localize to the nucleus and transform primary rat cells in cooperation with E1B-55K or Ras. Interaction of E1A with importin α3/Qip1, dual-specificity tyrosine-regulated kinase 1A (DYRK1A), HAN11, and CtBP influenced transformation with E1B-55K. Interestingly, the interaction of E1A with DYRK1A and HAN11 appeared to play a role in suppression of transformation by activated Ras whereas interaction with CtBP was not necessary. This unexpected result suggests a need for revision of current models and provides new insight into transformation by the C terminus of E1A.

Adenovirus evasion of interferon-mediated innate immunity by direct antagonism of a cellular histone posttranslational modification.

Overcoming the cellular type I interferon (IFN) host defense response is critical for a virus to ensure successful infection. Investigating the effects of human adenovirus (HAdV) infection on global cellular histone posttranslational modification (hPTM), we discovered that virus infection-induced activation of IFN signaling triggers a global increase in the monoubiquitination of histone 2B (H2B) at lysine 120, which is a mark for transcriptionally active chromatin. This hPTM, catalyzed by the hBre1/RNF20 complex, is necessary for activation of the cellular IFN-stimulated gene (ISG) expression program in response to viruses. To establish effective infection, the HAdV E1A protein binds to and dissociates the hBre1 complex to block IFN-induced H2B monoubiquitination and associated ISG expression. Together, these data uncover a key role for H2B monoubiquitination in the type I IFN response and a viral mechanism of antagonizing this hPTM to evade the IFN response.

Identification of a molecular recognition feature in the E1A oncoprotein that binds the SUMO conjugase UBC9 and likely interferes with polySUMOylation.

Hub proteins have central roles in regulating cellular processes. By targeting a single cellular hub, a viral oncogene may gain control over an entire module in the cellular interaction network that is potentially comprised of hundreds of proteins. The adenovirus E1A oncoprotein is a viral hub that interacts with many cellular hub proteins by short linear motifs/molecular recognition features (MoRFs). These interactions transform the architecture of the cellular protein interaction network and virtually reprogram the cell. To identify additional MoRFs within E1A, we screened portions of E1A for their ability to activate yeast pseudohyphal growth or differentiation. This identified a novel functional region within E1A conserved region 2 comprised of the sequence EVIDLT. This MoRF is necessary and sufficient to bind the N-terminal region of the SUMO conjugase UBC9, which also interacts with SUMO noncovalently and is involved in polySUMOylation. Our results suggest that E1A interferes with polySUMOylation, but not with monoSUMOylation. These data provide the first insight into the consequences of the interaction of E1A with UBC9, which was initially described in 1996. We further demonstrate that polySUMOylation regulates pseudohyphal growth and promyelocytic leukemia body reorganization by E1A. In conclusion, the interaction of the E1A oncogene with UBC9 mimics the normal binding between SUMO and UBC9 and represents a novel mechanism to modulate polySUMOylation.