Analysis of poxvirus host cell entry by electron microscopy

Vaccinia virus (VACV) is the prototypic poxvirus and is complex in both composition and life cycle. In contrast to most viruses, poxviruses produce two types of infectious particles with different roles in virus dissemination: Mature virions (MVs) mediate host-to-host transmission, whereas extracellular virions (EVs) accomplish virus dissemination in an infected organism. While MVs are enveloped by a single lipid membrane, EVs exhibit an unusual composition -- they consist of an MV-like particle that is wrapped with a second membrane.
Like all viruses, poxviruses depend on the host cell machinery to reach their final site of replication in the cytosol. Host cell entry of MVs is known to involve macropinocytosis and fusion of viral membranes with cellular membranes of endocytic compartments. Due to their topology, EVs require an unusual entry strategy and the outermost membrane is thought to be lost by non-fusogenic disruption, whereas the inner membrane is removed by fusion. Hence, EV and MV entry both result in the release of viral cores containing the DNA genome into the host cell cytosol.
One aim of this thesis was to understand the cellular entry process of VACV EVs and MVs, in detail. In the case of EVs, we wanted to investigate whether this process involves endocytosis, where and how the EV membrane is lost, and which cellular factors and processes were exploited in the course of this process. Furthermore, we intended to study the processes that occur after release of viral cores into the cytosol and allow early gene expression.
The results of our investigations on the cellular entry mechanism of VACV EVs are presented in chapter ÒHost cell entry of VACV EVsÓ. EV entry into HeLa cells was studied using fluorescence microscopy and electron microscopy in combination with flow cytometry. We show that complete EVs are internalized by macropinocytosis and that this step is actively triggered by EVs. Acidification of macropinosomes induces disruption of EV membranes and this exposes the underlying MV-like particle. The latter presumably fuses with cellular membranes and releases the viral core into the host cell cytosol.
A further issue of this thesis was to support a high-throughput RNAi silencing screen, which identifies host cell factors required during vaccinia virus infection, with electron microscopy analysis. Validation and analysis of clustered hits revealed previously unknown processes during virus entry including a new mechanism for genome uncoating and was confirmed by EM. Viral core proteins were found to be ubiquitinated already during virus assembly. After entering the cytosol of an uninfected cell, the viral DNA was released from the core through the activity of the cell's proteasomes. Next, a Cullin3-based ubiquitin ligase-mediated round of ubiquitination and proteasome action was needed to initiate viral DNA replication.
Our efforts to clarify the mechanisms underlying core activation after fusion are presented in chapter ÒVACV LBs deliver effector proteinsÓ and the main research aspect of this thesis. Morphological changes in viral structures were observed by conventional and cryo-electron microscopy to occur immediately after viral fusion and may involve the reduction of disulfide bonds in proteins that comprise the viral core wall. Lateral bodies, proteinaceous structures that are packaged into VACV MVs and EVs in addition to the cores, dissociate from viral cores upon fusion. Both processes, core activation and dissociation of lateral bodies, require cellular factors or components. They cannot be recapitulated by removal of membranes and reduction in vitro.
We identified F18 as a major structural protein of the lateral bodies, which are left back at the membrane after fusion. We found that the decomposition of lateral bodies by the host cell proteasome releases at least one immediate effector protein into the host cell, the dual specific viral phosphatase VH1. Furthermore, we identified a number of disulfide-bonded core proteins and found that some of them are reduced during activation. This possibly contributes to subtle changes that finally allow early gene transcription in the core. We thus have compelling reasons to propose a new viral delivery mechanism for immediate effector proteins that is very likely used to release additional, so far unknown, effector proteins.