The nanoscale molecular assembly of mammalian viruses during their infectious life cycle remains poorly understood. Their small dimensions, generally bellow the 300nm diffraction limit of light microscopes, has limited most imaging studies to electron microscopy. The recent development of super-resolution (SR) light microscopy now allows the visualisation of viral structures at resolutions of tens of nanometers. In addition, these techniques provide the added benefit of molecular specific labelling and the capacity to investigate viral structural dynamics using live-cell microscopy. However, there is a lack of robust analytical tools that allow for precise mapping of viral structure within the setting of infection. Here we present an open-source analytical framework that combines super-resolution imaging and naïve single-particle analysis to generate unbiased molecular models. This tool, VirusMapper, is a high-throughput, user-friendly, ImageJ-based software package allowing for automatic statistical mapping of conserved multi-molecular structures, such as viral substructures or intact viruses. We demonstrate the usability of VirusMapper by applying it to SIM and STED images of vaccinia virus in isolation and when engaged with host cells. VirusMapper allows for the generation of accurate, high-content, molecular specific virion models and detection of nanoscale changes in viral architecture.

The use of deep neural networks (DNNs) for analysis of complex biomedical images shows great promise but is hampered by a lack of large verified data sets for rapid network evolution. Here, we present a novel strategy, termed “mimicry embedding,” for rapid application of neural network architecture-based analysis of pathogen imaging data sets. Embedding of a novel host-pathogen data set, such that it mimics a verified data set, enables efficient deep learning using high expressive capacity architectures and seamless architecture switching. We applied this strategy across various microbiological phenotypes, from superresolved viruses to in vitro and in vivo parasitic infections. We demonstrate that mimicry embedding enables efficient and accurate analysis of two- and three-dimensional microscopy data sets. The results suggest that transfer learning from pretrained network data may be a powerful general strategy for analysis of heterogeneous pathogen fluorescence imaging data sets. IMPORTANCE In biology, the use of deep neural networks (DNNs) for analysis of pathogen infection is hampered by a lack of large verified data sets needed for rapid network evolution. Artificial neural networks detect handwritten digits with high precision thanks to large data sets, such as MNIST, that allow nearly unlimited training. Here, we developed a novel strategy we call mimicry embedding, which allows artificial intelligence (AI)-based analysis of variable pathogen-host data sets. We show that deep learning can be used to detect and classify single pathogens based on small differences.

All poxviruses contain a set of proteinaceous structures termed lateral bodies (LB) that deliver viral effector proteins into the host cytosol during virus entry. To date, the spatial proteotype of LBs remains unknown. Using the prototypic poxvirus, vaccinia virus (VACV), we employed a quantitative comparative mass spectrometry strategy to determine the poxvirus LB proteome. We identified a large population of candidate cellular proteins, the majority being mitochondrial, and 15 candidate viral LB proteins. Strikingly, one-third of these are VACV redox proteins whose LB residency could be confirmed using super-resolution microscopy. We show that VACV infection exerts an anti-oxidative effect on host cells and that artificial induction of oxidative stress impacts early and late gene expression as well as virion production. Using targeted repression and/or deletion viruses we found that deletion of individual LB-redox proteins was insufficient for host redox modulation suggesting there may be functional redundancy. In addition to defining the spatial proteotype of VACV LBs, these findings implicate poxvirus redox proteins as potential modulators of host oxidative anti-viral responses and provide a solid starting point for future investigations into the role of LB resident proteins in host immunomodulation.

Poxvirus egress is a complex process whereby cytoplasmic single membrane–bound virions are wrapped in a cell-derived double membrane. These triple-membrane particles, termed intracellular enveloped virions (IEVs), are released from infected cells by fusion. Whereas the wrapping double membrane is thought to be derived from virus-modified trans-Golgi or early endosomal cisternae, the cellular factors that regulate virus wrapping remain largely undefined. To identify cell factors required for this process the prototypic poxvirus, vaccinia virus (VACV), was subjected to an RNAi screen directed against cellular membrane-trafficking proteins. Focusing on the endosomal sorting complexes required for transport (ESCRT), we demonstrate that ESCRT-III and VPS4 are required for packaging of virus into multivesicular bodies (MVBs). EM-based characterization of MVB-IEVs showed that they account for half of IEV production indicating that MVBs are a second major source of VACV wrapping membrane. These data support a model whereby, in addition to cisternae-based wrapping, VACV hijacks ESCRT-mediated MVB formation to facilitate virus egress and spread.

Like many viruses, herpes simplex virus 1 (HSV1) expresses an endoribonuclease, the virion host shutoff (vhs) protein, which regulates the RNA environment of the infected cell and facilitates the classical cascade of virus protein translation. It does this by causing the degradation of some mRNA molecules and the nuclear retention of others.

Enveloped viruses exploit cellular trafficking pathways for their morphogenesis, providing potential scope for the development of new antiviral therapies. We have previously shown that herpes simplex virus 1 (HSV1) utilizes recycling endocytic membranes as the source of its envelope, in a process involving four Rab GTPases. To identify novel factors involved in HSV1 envelopment, we have screened a small interfering RNA (siRNA) library targeting over 80 human trafficking proteins, including coat proteins, adaptor proteins, fusion factors, fission factors, and Rab effectors. The depletion of 11 factors reduced virus yields by 20- to 100-fold, including three early secretory pathway proteins, four late secretory pathway proteins, and four endocytic pathway proteins, three of which are membrane fission factors. Five of the 11 targets were chosen for further analysis in virus infection, where it was found that the absence of only 1, the fission factor CHMP4C, but not the CHMP4A or CHMP4B paralogues, reduced virus production at the final stage of morphogenesis. Ultrastructural and confocal microscopy of CHMP4C-depleted, HSV1-infected cells showed an accumulation of endocytic membranes; extensive tubulation of recycling, transferrin receptor-positive endosomes indicative of aberrant fission; and a failure in virus envelopment. No effect on the late endocytic pathway was detected, while exogenous CHMP4C was shown to localize to recycling endosomes. Taken together, these data reveal a novel role for the CHMP4C fission factor in the integrity of the recycling endosomal network, which has been unveiled through the dependence of HSV1 on these membranes for the acquisition of their envelopes. IMPORTANCE Cellular transport pathways play a fundamental role in secretion and membrane biogenesis. Enveloped viruses exploit these pathways to direct their membrane proteins to sites of envelopment and, as such, are powerful tools for unraveling subtle activities of trafficking factors, potentially pinpointing therapeutic targets. Using the sensitive biological readout of virus production, over 80 trafficking factors involved in diverse and poorly defined cellular processes have been screened for involvement in the complex process of HSV1 envelopment. Out of 11 potential targets, CHMP4C, a key component in the cell cycle abscission checkpoint, stood out as being required for the process of virus wrapping in endocytic tubules, where it localized. In the absence of CHMP4C, recycling endocytic membranes failed to undergo scission in infected cells, causing transient tubulation and accumulation of membranes and unwrapped virus. These data reveal a new role for this important cellular factor in the biogenesis of recycling endocytic membranes.

The use of deep neural networks (DNNs) for analysis of complex biomedical images shows great promise but is hampered by a lack of large verified datasets for rapid network evolution. Here we present a novel "mimicry embedding" strategy for rapid application of neural network architecture-based analysis of biomedical imaging datasets. Embedding of a novel biological dataset, such that it mimicks a verified dataset, enables efficient deep learning and seamless architecture switching. We apply this strategy across various microbiological phenotypes; from super-resolved viruses to in vivo parasitic infections. We demonstrate that mimicry embedding enables efficient and accurate analysis of three-dimensional microscopy datasets. The results suggest that transfer learning from pre-trained network data may be a powerful general strategy for analysis of heterogenous biomedical imaging datasets. Deep learning | capsule networks | transfer learning | super-resolution microscopy | vaccinia virus | Toxoplasma gondii | zebrafishCorrespondence: jason.mercer@ucl.ac.uk, artur.yakimovich@ucl.ac.uk

Modulation of the host cell cycle is a common strategy used by viruses to create a pro-replicative environment. To facilitate viral genome replication, vaccinia virus (VACV) has been reported to alter cell cycle regulation and trigger the host cell DNA damage response. However, the cellular factors and viral effectors that mediate these changes remain unknown. Here, we set out to investigate the effect of VACV infection on cell proliferation and host cell cycle progression. Using a subset of VACV mutants, we characterise the stage of infection required for inhibition of cell proliferation and define the viral effectors required to dysregulate the host cell cycle. Consistent with previous studies, we show that VACV inhibits and subsequently shifts the host cell cycle. We demonstrate that these two phenomena are independent of one another, with viral early genes being responsible for cell cycle inhibition, and post-replicative viral gene(s) responsible for the cell cycle shift. Extending previous findings, we show that the viral kinase F10 is required to activate the DNA damage checkpoint and that the viral B1 kinase and/or B12 pseudokinase mediate degradation of checkpoint effectors p53 and p21 during infection. We conclude that VACV modulates host cell proliferation and host cell cycle progression through temporal expression of multiple VACV effector proteins. (209/200.)