Cooperation of the Ebola Virus Proteins VP40 and GP with BST2 To Activate NF-κB Independently of Virus-Like Particle Trapping.

BST2 is a host protein with dual functions in response to viral infections: it traps newly assembled enveloped virions at the plasma membrane in infected cells, and it induces NF-κB activity, especially in the context of retroviral assembly. In this study, we examined whether Ebola virus proteins affect BST2-mediated induction of NF-κB. We found that the Ebola virus matrix protein, VP40, and envelope glycoprotein, GP, each cooperate with BST2 to induce NF-κB activity, with maximal activity when all three proteins are expressed. Unlike human immunodeficiency virus type 1 Vpu protein, which antagonizes both virion entrapment and the activation of NF-κB by BST2, Ebola virus GP does not inhibit NF-κB signaling even while it antagonizes the entrapment of virus-like particles. GP from Reston ebolavirus, a nonpathogenic species in humans, showed a phenotype similar to that of GP from Zaire ebolavirus, a highly pathogenic species, in terms of both the activation of NF-κB and the antagonism of virion entrapment. Although Ebola virus VP40 and GP both activate NF-κB independently of BST2, VP40 is the more potent activator. Activation of NF-κB by the Ebola virus proteins either alone or together with BST2 requires the canonical NF-κB signaling pathway. Mechanistically, the maximal NF-κB activation by GP, VP40, and BST2 together requires the ectodomain cysteines needed for BST2 dimerization, the putative BST2 tetramerization residue L70, and Y6 of a potential hemi-ITAM motif in BST2's cytoplasmic domain. BST2 with a glycosylphosphatidylinositol (GPI) anchor signal deletion, which is not expressed at the plasma membrane and is unable to entrap virions, activated NF-κB in concert with the Ebola virus proteins at least as effectively as wild-type BST2. Signaling by the GPI anchor mutant also depended on Y6 of BST2. Overall, our data show that activation of NF-κB by BST2 is independent of virion entrapment in the case of Ebola virus. Nonetheless, BST2 may induce or amplify proinflammatory signaling during Ebola virus infection, potentially contributing to the dysregulated cytokine response that is a hallmark of Ebola virus disease. Understanding how the host responds to viral infections informs the development of therapeutics and vaccines. We asked how proinflammatory signaling by the host protein BST2/tetherin, which is mediated by the transcription factor NF-κB, responds to Ebola virus proteins. Although the Ebola virus envelope glycoprotein (GP) antagonizes the trapping of newly formed virions at the plasma membrane by BST2, we found that it does not inhibit BST2's ability to induce NF-κB activity. This distinguishes GP from the HIV-1 protein Vpu, the prototype BST2 antagonist, which inhibits both virion entrapment and the induction of NF-κB activity. Ebola virus GP, the Ebola virus matrix protein VP40, and BST2 are at least additive with respect to the induction of NF-κB activity. The effects of these proteins converge on an intracellular signaling pathway that depends on a protein modification termed neddylation. Better mechanistic understanding of these phenomena could provide targets for therapies that modulate the inflammatory response during Ebola virus disease.