The large tegument proteins of herpesviruses encode conserved cysteine proteases of unknown function. Here we show that BPLF1, the Epstein-Barr-virus-encoded member of this protease family, is a deneddylase that regulates virus production by modulating the activity of cullin-RING ligases (CRLs). BPLF1 hydrolyses NEDD8 conjugates in vitro, acts as a deneddylase in vivo, binds to cullins and stabilizes CRL substrates. Expression of BPLF1 alone or in the context of the productive virus cycle induces accumulation of the licensing factor CDT1 and deregulates S-phase DNA synthesis. Inhibition of BPLF1 during the productive virus cycle prevents cellular DNA re-replication and inhibits virus replication. Viral DNA synthesis is restored by overexpression of CDT1. Homologues encoded by other herpesviruses share the deneddylase activity. Thus, these enzymes are likely to have a key function in the virus life cycle by inducing a replication-permissive S-phase-like cellular environment.

The N-terminal domains of the herpesvirus large tegument proteins encode a conserved cysteine protease with ubiquitin- and NEDD8-specific deconjugase activity. The proteins are expressed during the productive virus cycle and are incorporated into infectious virus particles, being delivered to the target cells upon primary infection. Members of this viral enzyme family were shown to regulate different aspects of the virus life cycle and the innate anti-viral response. However, only few substrates have been identified and the mechanisms of these effects remain largely unknown. In order to gain insights on the substrates and signaling pathways targeted by the viral enzymes, we have used co-immunoprecipitation and mass spectrometry to identify cellular proteins that interact with the Epstein-Barr virus encoded homologue BPLF1. Several members of the 14-3-3-family of scaffold proteins were found amongst the top hits of the BPLF1 interactome, suggesting that, through this interaction, BPLF1 may regulate a variety of cellular signaling pathways. Analysis of the shared protein-interaction network revealed that BPLF1 promotes the assembly of a tri-molecular complex including, in addition to 14-3-3, the ubiquitin ligase TRIM25 that participates in the innate immune response via ubiquitination of cytosolic pattern recognition receptor, RIG-I. The involvement of BPLF1 in the regulation of this signaling pathway was confirmed by inhibition of the type-I IFN responses in cells transfected with a catalytically active BPLF1 N-terminal domain or expressing the endogenous protein upon reactivation of the productive virus cycle. We found that the active viral enzyme promotes the dimerization and autoubiquitination of TRIM25. Upon triggering of the IFN response, RIG-I is recruited to the complex but ubiquitination is severely impaired, which functionally inactivates the RIG-I signalosome. The capacity to bind to and functionally inactivate the RIG-I signalosome is shared by the homologues encoded by other human herpesviruses.

The large tegument proteins of herpesviruses contain N-terminal cysteine proteases with potent ubiquitin and NEDD8-specific deconjugase activities, but the function of the enzymes during virus replication remains largely unknown. Using as model BPLF1, the homologue encoded by Epstein-Barr virus (EBV), we found that induction of the productive virus cycle does not affect the total level of ubiquitin-conjugation but is accompanied by a BPLF1-dependent decrease of NEDD8-adducts and accumulation of free NEDD8. Expression of BPLF1 promotes cullin degradation and the stabilization of cullin-RING ligases (CRLs) substrates in the nucleus, while cytoplasmic CRLs and their substrates are not affected. The inactivation of nuclear CRLs is reversed by the N-terminus of CAND1, which inhibits the binding of BPLF1 to cullins and prevents efficient viral DNA replication. Targeting of the deneddylase activity to the nucleus is dependent on processing of the catalytic N-terminus by caspase-1. Inhibition of caspase-1 severely impairs viral DNA synthesis and the release of infectious virus, pointing a previously unrecognized role of the cellular response to danger signals triggered by EBV reactivation in promoting virus replication.

BackgroundIn water-soluble proteins it is energetically favorable to bury hydrophobic residues and to expose polar and charged residues. In contrast to water soluble proteins, transmembrane proteins face three distinct environments; a hydrophobic lipid environment inside the membrane, a hydrophilic water environment outside the membrane and an interface region rich in phospholipid head-groups. Therefore, it is energetically favorable for transmembrane proteins to expose different types of residues in the different regions.ResultsInvestigations of a set of structurally determined transmembrane proteins showed that the composition of solvent exposed residues differs significantly inside and outside the membrane. In contrast, residues buried within the interior of a protein show a much smaller difference. However, in all regions exposed residues are less conserved than buried residues. Further, we found that current state-of-the-art predictors for surface area are optimized for one of the regions and perform badly in the other regions. To circumvent this limitation we developed a new predictor, MPRAP, that performs well in all regions. In addition, MPRAP performs better on complete membrane proteins than a combination of specialized predictors and acceptably on water-soluble proteins. A web-server of MPRAP is available at http://mprap.cbr.su.se/ConclusionBy including complete a-helical transmembrane proteins in the training MPRAP is able to predict surface accessibility accurately both inside and outside the membrane. This predictor can aid in the prediction of 3D-structure, and in the identification of erroneous protein structures.

The conserved N-terminal domains of the major tegument proteins of herpes viridae encode cysteine proteases with potent ubiquitin and NEDD8-specific deconjugase activity. Here we show that the Epstein-Barr virus-encoded member of this enzyme family, BPLF1, is targeted to cullin-RING ubiquitin ligases (CRLs) via the interaction of the conserved helix-2 with helix-23 of the C-terminal domain (CTD) of cullins, at a site involved in electrostatic interaction with helix-8 of the CRL regulator CAND1. Mutation of the solvent-exposed Asp86 and Asp90 of helix-2 to Arg does not affect the enzymatic activity of BPLF1 but abolishes cullin binding and prevents CRL inactivation. The binding of the catalytically active BPLF1 to cullins inhibits the recruitment of CAND1 to the deneddylated CRLs and promotes the selective degradation of cullins by the proteasome. Cullin proteolysis is rescued by the overexpression of CAND1 or its CTD-binding N-terminal domain. These findings illustrate a new strategy for viral modulation of CRL activity where the combined effects of cullin deneddylation and their targeting for proteasomal degradation drive stable inactivation of the ligases.