Proteasome-independent disruption of PML oncogenic domains (PODs), but not covalent modification by SUMO-1, is required for human cytomegalovirus immediate-early protein IE1 to inhibit PML-mediated transcriptional repression

Human cytomegalovirus (HCMV) major immediate-early protein IE1 is an abundant 72-kDa nuclear phosphoprotein that is thought to play an important role in efficient triggering of the lytic cycle, especially at low multiplicity of infection. The best-known properties of IE1 at present are its transient targeting to punctate promyelocytic leukemia protein (PML)-associated nuclear bodies (PML oncogenic domains [PODs] or nuclear domain 10 [ND10]), with associated displacement of the cellular PML tumor suppressor protein into a diffuse nucleoplasmic form and its association with metaphase chromosomes. Recent studies have shown that the targeting of PML (and associated proteins such as hDaxx) to PODs is dependent on modification of PML by ubiquitin-like protein SUMO-1. In this study, we provide direct evidence that IE1 is also covalently modified by SUMO-1 in both infected and cotransfected cells, as well as in in vitro assays, with up to 30% of the protein representing the covalently conjugated 90-kDa form in stable U373/IE1 cell lines. Lysine 450 was mapped as the major SUMO-1 conjugation site, but a point mutation of this lysine residue in IE1 did not interfere with its targeting to and disruption of the PODs. Surprisingly, unlike PML or IE2, IE1 did not interact with either Ubc9 or SUMO-1 in yeast two-hybrid assays, suggesting that some additional unknown intranuclear cofactors must play a role in IE1 sumoylation. Interestingly, stable expression of either exogenous PML or exogenous Flag-SUMO-1 in U373 cell lines greatly enhanced both the levels and rate of in vivo IE1 sumoylation during HCMV infection. Unlike the disruption of PODs by the herpes simplex virus type 1 IE110(ICP0) protein, the disruption of PODs by HCMV IE1 proved not to involve proteasome-dependent degradation of PML. We also demonstrate here that the 560-amino-acid PML1 isoform functions as a transcriptional repressor when fused to the GAL4 DNA-binding domain and that wild-type IE1 inhibits the repressor function of PML1 in transient cotransfection assays. Furthermore, both IE1(1-346) and IE1(L174P) mutants, which are defective in displacing PML from PODs, failed to inhibit the repression activity of PML1, whereas the sumoylation-negative IE1(K450R) mutant derepressed as efficiently as wild-type IE1. Taken together, our results suggest that proteasome-independent disruption of PODs, but not IE1 sumoylation, is required for efficient IE1 inhibition of PML-mediated transcriptional repression.