Epstein-Barr virus nuclear antigen 3A protein regulates CDKN2B transcription via interaction with MIZ-1.
Bottom Line: Focusing on the EBNA3A/MIZ-1 interaction we demonstrate that binding occurs in EBV-infected cells expressing both proteins at endogenous physiological levels and that in the presence of EBNA3A, a significant fraction of MIZ-1 translocates from the cytoplasm to the nucleus.Moreover, we show that a trimeric complex composed of a MIZ-1 recognition DNA element, MIZ-1 and EBNA3A can be formed, and that interaction of MIZ-1 with nucleophosmin (NPM), one of its coactivator, is prevented by EBNA3A.Finally, we show that, in the presence of EBNA3A, expression of the MIZ-1 target gene, CDKN2B, is downregulated and repressive H3K27 marks are established on its promoter region suggesting that EBNA3A directly counteracts the growth inhibitory action of MIZ-1.
Affiliation: CIRI, International Center for Infectiology Research, Oncogenic Herpesviruses team, Université de Lyon, Lyon 69364, France Université Lyon 1, Centre International de Recherche en Infectiologie, Lyon 69364, France CIRI, International Center for Infectiology Research, Cell Biology of Viral Infections team, Université de Lyon, Lyon 69364, France INSERM, U1111, Lyon 69364, France CNRS, UMR5308, Lyon 69364, France.Show MeSH
Related in: MedlinePlus
Mentions: Because MIZ-1 is implicated in the transcriptional regulation of the cyclin-dependent kinase inhibitor gene CDKN2B (encoding p15INK4B), we asked whether EBNA3A can modulate the expression of this gene. For this, we compared the level of transcript expressed in a B-LCL established with wt EBV (LCLwt) to that expressed in a LCL (LCLΔ3A) infected with an EBV mutant deficient for EBNA3A (E3AmtB mutant) (7). We first verified that the level of MIZ-1 protein was not affected by the presence or absence of EBNA3A in the LCLs (Supplementary Figure S3A). We then analyzed the levels of the CDKN2B transcript by standard and quantitative RT-PCR (Figure 5A and B). Interestingly, the level of the CDKN2B transcripts was drastically reduced in the presence of EBNA3A. We confirmed these observations in several LCLs immortalized independently with the E3AmtB mutant as well as LCLs obtained using a different donor and immortalized with a different type of EBNA3A-deficient mutant (mutant E3AmtA (7)). In all cases, we observed a similar decrease in CDKN2B RNA levels in the LCLs expressing EBNA3A compared to their counterpart infected with EBNA3A-deficient viruses (Supplementary Figure S3B). Finally, we analyzed the repression kinetics of CDKN2B. For this we made use of a LCL that expresses EBNA3A in a doxycycline (Dox) dependent manner (ΔE3A-LCLdoxE3A) (31). As can be seen in Figure 5C, the transcript levels of CDKN2B fell rapidly—within 24 h—upon EBNA3A induction. Interestingly, we also observed a downregulation in the expression of CDKN1C and CEBPD—two genes that have previously been reported to be directly regulated by Miz-1 (41,67,68)—in the presence of EBNA3A (Supplementary Fig S4). In the case of CDKN1A—also a well documented target gene of Miz-1—we only observed a marginal, but reproducible, increase (Supplementary Figure S4). Moreover, in a recent paper Yenamandra et al. (69), using these same LCLs, found a 8-fold decrease in the level of CDKN1A transcripts in the presence of EBNA3A, which supports a role for EBNA3A in CDKN1A regulation. From these results we conclude that EBNA3A can repress transcription of MIZ-1 target genes.
Affiliation: CIRI, International Center for Infectiology Research, Oncogenic Herpesviruses team, Université de Lyon, Lyon 69364, France Université Lyon 1, Centre International de Recherche en Infectiologie, Lyon 69364, France CIRI, International Center for Infectiology Research, Cell Biology of Viral Infections team, Université de Lyon, Lyon 69364, France INSERM, U1111, Lyon 69364, France CNRS, UMR5308, Lyon 69364, France.