Biology and Pathogenesis of Rhabdo- and Filoviruses

The following sections are included:

• FOREWORD
• CONTENTS

Enveloped viruses with a negative-sense, single-stranded monopartite RNA genome have been classified into the order Mononegavirales. Five families of viruses that constitute the order are: Rhabdoviridae, Filoviridae, Paramyxoviridae, Bornaviridae and Nyamiviridae. Members of these families possess a helical nucleocapsid core containing the viral genome and a host-derived lipid envelope containing viral proteins. This introductory chapter provides a brief overview of the Rhabdoviridae and the Filoviridae, the two families of viruses that are the subject of this book. Many members of these two families are highly significant human and animal pathogens. The rationale and goal of the book is to provide the reader with the most recent information on the structure, genome organization and replication strategy, epidemiology, evolution and emergence, host response to infection, viral countermeasures as well as vaccines and antivirals against these pathogens. More detailed descriptions of these topics are presented in individual chapters of this book.

Rhabdoviruses are rod-shaped particles. The surface glycoprotein exists as trimers anchored in the viral envelope. The matrix protein forms a two dimensional mesh underneath the envelope with one side interacting with the cytoplasmic tail of glycoprotein trimers, and the other side winding the packaged nucleocapsid into a bullet-shaped superhelix. The phosphoprotein (the P protein) and the large subunit of the viral polymerase (the L protein) are located inside the nucleocapsid helix. The glycoprotein recognizes the host cell receptor to enter the cell through endocytosis. Inside the endosome, a conformational change of the glycoprotein is induced by acidic pH to project the membrane fusion loop into the target cellular membrane, which leads to fusion of the viral membrane with the endosomal membrane. After membrane fusion, the nucleocapsid is released into the cytoplasm together with the P and L proteins. The viral genomic RNA remains sequestered in the nucleocapsid. The viral polymerase consisting of the P and L proteins specifically recognizes the nucleocapsid and unveils the sequestered genomic RNA for viral RNA synthesis. The progeny genome is encapsidated concomitantly during viral replication. The linear nucleocapsid is assembled with the nucleocapsid protein subunits (the N protein), which are aligned in parallel. The viral genomic RNA resides in the V-shaped cavity formed by the N-terminal and C-terminal domains of the N protein. The N protein subunits form a network of interactions across neighboring molecules to stabilize the nucleocapsid. The nucleocapsid is transported to the plasma membrane where the matrix protein orchestrates the assembly of the virus particle.

Vesicular stomatitis virus (VSV), like other Rhabdoviruses, uses the endocytic pathway to gain entry into the cell. VSV is an enveloped virus studded with the spike glycoprotein (VSV G) which mediates receptor recognition at the cell surface and fusion with endosomal membranes in a pH-dependent manner. After internalization via the clathrin-dependent pathway, the virus reaches the endosomes where fusion of the viral envelope with endosomal membranes releases the nucleocapsid into the cytoplasm allowing infection to proceed. Based primarily on the time-course of infection and the pH requirements of virus fusion, it has been proposed that fusion with the early endosomal membrane releases the nucleocapsid directly into the cytosol. However, studies of the molecular mechanisms that control viral RNA release into the cytoplasm have led to a different model. These studies have allowed the identification of endosomal factors that regulate the process. These include the endosomal phospholipid lysobisphosphatidic acid, the ESCRT-associated protein ALIX and proteins of ESCRT complexes, suggesting that RNA may be released in a two-step process, first into intralumenal vesicles of endosomes and then into the cytoplasm. In this chapter we will discuss these models.

Rhabdoviruses are unique among enveloped animal viruses in that they encode a single, uncleaved glycoprotein (G) that functions for both receptor binding and membrane fusion during virus entry. G also contributes to efficient virus assembly and release by budding of nascent virions from the plasma membrane. The G proteins of vesicular stomatitis virus (VSV) and rabies virus (RABV) are the best studied and VSV G has been used as a model membrane protein for studying the host cell machinery responsible for transport through the secretory pathway, which is due in part to the rapid transport kinetics of G from the endoplasmic reticulum through the Golgi and to the cell surface. It is assumed that the G proteins of other viruses from the Rhabdoviridae family have properties similar to VSV and RABV G. Indeed, sequence alignments of the G proteins from VSV, RABV, bovine ephemeral fever virus (BEFV) and infectious haematopoietic necrosis virus (IHNV) demonstrate that the locations of neutralizing epitopes and cysteine residues in the ectodomains of these proteins are similar, which supports the idea that the basic elements of the folded structure of rhabdovirus G proteins are preserved. Despite intensive interest in understanding the mechanism of rhabdovirus G-induced membrane fusion and the conformational changes induced during the membrane fusion reaction, only recently has the structure of one of these (VSV G) in the pre- and post-fusion conformations been determined. Because of the medical significance of RABV, and the extensive body of literature on the structure and function of VSV G, this chapter will primarily focus on these two representative members of the Rhabdoviridae family, although where appropriate, information on other rhabdovirus G proteins will also be presented.

Vesicular stomatitis virus (VSV) is a humble livestock pathogen that causes little or no disease in humans, but it is by far the best-characterized non-segmented negative-strand (NNS) RNA virus when it comes to transcription and replication functions. It has the simplest genome in its class, grows readily in a wide variety of cell types, and its robust virion-associated polymerase activity has long ago given it a leg up as a model system for NNS viral RNA synthesis. The basic features of VSV transcription and replication (gene order, polymerase components, polymerase products, dual mode of synthesis) were uncovered in the first 20 years or so following the landmark discovery of the VSV virion-associated polymerase by Baltimore and colleagues. The advent of reverse genetic systems for VSV in the early 1990's ushered in a detailed characterization of cis-acting RNA template sequences, a task now largely complete. Reverse genetic systems have also enabled structure-function analysis of the viral proteins that carry out VSV RNA synthesis. More recently, components of the transcription and replication machinery, or parts thereof, have been crystallized and the atomic structures have provided tantalizing insights into mechanisms of viral RNA synthesis. But many questions remain unanswered. This review focuses on recent progress and gaps in our understanding.

Vesicular stomatitis virus (VSV), the prototypic rhabdovirus, has been used as an excellent paradigm for understanding the mechanisms of virus replication, pathogenesis, host response to virus infection and also for myriads of studies on cellular and molecular biology. Biochemical studies as well as high-throughput genomics, proteomics, and chemical approaches have revealed a plethora of cellular factors and pathways that regulate replication of VSV. These factors include those that support virus replication and also those that restrict its replication. This chapter discusses the role(s) of many of these host cell factors and pathways involved in VSV replication. Although mechanistic understanding of the roles of some of these factors in VSV replication has been obtained, the roles of many others need to be investigated for a better understanding of the virus-host cell interactions.

The extent of induction of cytopathic effects (CPE) among rhabdoviruses ranges from the extensive CPE induced by vesiculoviruses, which are among the most cytopathic mammalian viruses, to the virtual absence of CPE among many lyssaviruses. Most rhabdovirus cytopathogenesis is due to activation of the two principal pathways responsible for apoptosis, the death receptor and mitochondrial pathways. There are two major viral factors that induce apoptosis. One is inhibition of host gene expression by viral matrix (M) proteins that is characteristic of most vesiculoviruses and novirhabdoviruses. The other is the overexpression of viral pathogen-associated molecular patterns (PAMPs), which occurs to differing extents with most rhabdoviruses. However, there is a striking diversity in the biochemical and morphological changes that occur in rhabdovirus-infected cells as a result of variations in the activation of these pathways. This is due to cellular factors including the regulators of apoptosis present at the time of infection versus those that are induced by infection, and by viral factors such as the level of expression of inducers of apoptosis and the expression of viral proteins that suppress apoptosis. Indeed, the extent of this variation is a major theme of this chapter.

Rhabdoviruses (vesicular stomatitis virus [VSV] and rabies virus [RABV]) and filoviruses (Ebola [EBOV] and Marburg [MARV] viruses) are enveloped, negative-sense RNA viruses that can cause severe disease in both animals and humans. Assembly and budding of rhabdoviruses and filoviruses are critically dependent on both viral and host proteins/pathways. The matrix proteins of rhabdoviruses and filoviruses direct the budding process by coordinating and condensing viral components at specialized regions of the plasma membrane and by recruiting specific host proteins associated with the vacuolar protein sorting (vps), ubiquitination, and cytoskeletal pathways. The highly conserved Late (L) budding domains present within the matrix (M) protein of rhabdoviruses and VP40 matrix protein of filoviruses hijack host proteins to facilitate the last step of virus-cell separation. Our continued understanding of the fundamental mechanisms of rhabdoviruses and filovirus budding may lead to the development of new strategies to inhibit virus egress, transmission, and spread. This chapter summarizes recent key findings and current knowledge of these complex virus-host interactions involved in assembly/budding of rhabdoviruses and filoviruses.

Vaccines based on live-attenuated viruses often induce life-long immunity due to the potent stimulation of innate and adaptive immune responses. Recombinant viruses expressing protective antigens from foreign pathogens can also serve as potent vaccines protecting against the foreign pathogens. In this chapter we focus on the development of a rhabdovirus, vesicular stomatitis virus (VSV), as a vaccine vector. In more than fifty pre-clinical studies, vaccine vectors based on VSV have proven effective in animal models of HIV/AIDS, and in multiple other viral and bacterial disease models. VSV-based vaccines are often effective in a single dose. They can also be effective as post-exposure vaccines against deadly viruses such as Ebola, Marburg and influenza, and as therapeutic vaccines eliminating tumors in a rabbit model for cancer caused by papilloma virus. Here we review the history of VSV vaccine vector development, including the detailed process of attenuation as well as other steps that were required to obtain FDA approval for the recently completed Phase I clinical trial of a VSV-based vector.

Oncolytic rhabdoviruses including vesicular stomatitis virus (VSV) and Maraba virus are powerful self-replicating cancer biotherapeutics. These rhabdoviruses are naturally oncolytic due to their basic biology and epidemiology: interferon (IFN) sensitivity, rapid viral growth kinetics, and an immunologically naïve human population. Over the past two decades, scientists have derived novel approaches to improve their therapeutic applications in terms of both efficacy and safety, including mutagenesis, genetic engineering to express therapeutic payload, and combination therapies. With the first oncolytic rhabdovirus, VSV-hIFNβ, currently being evaluated in human clinical trials, we are at a historical time in the development of oncolytic rhabdovirus therapeutics.

The ability to completely map all synaptic connections (i.e. the connectome) of the brain would provide neuroscientists the possibility to test predictive hypotheses about the brain organization/behavioral capability, compare variations between two brains and find common denominators between species. Yet, the techniques to take on such a challenging task in the mammalian brain with ~10¹¹-10¹⁵ synapses depending on the species, are just evolving. Large research and funding efforts devoted to the unraveling of the synaptic connectivity of individual neurons, specific neuron types, and of neural circuits to link their function to the underlying structure are underway. Modern methodological approaches from a variety of disciplines are applied to tackle the immense complexity of the task. Amongst those approaches, recombinant neurotropic viruses of the Rhabdoviridae family (in particular rabies virus and vesicular stomatitis virus) have become important tools. Advances in molecular biology, mouse genetics, and in virology have increased the repertoire of tools available for defining the initial targets for infection as well as permitting improved tracing of synaptically connected neurons. At the same time, new constructs enhance our ability to manipulate circuits and to monitor their activity. Recombinant neurotropic virus-based strategies, together with optogenetic, electrophysiological, imaging methods, and behavioral tasks enable previously unimaginable experiments. In this chapter, we will summarize the most recent advances in the use of recombinant rhabdoviruses for understanding the function and structure of neuronal circuitry.

Our understanding of viral evolution comes from two main sources: phylogenetic analyses and experimental evolution. The latter is largely based on highly controlled, in vitro work, but recently in vivo models have been developed. Among the Rhabdoviruses, vesicular stomatitis virus (VSV) has been one of the most useful models, and has served to test ideas of general evolutionary biology such as the punctuated equilibrium, Red Queen hypothesis, the competitive exclusion principle and Muller's ratchet. In more recent years VSV has served to demonstrate the quasispecies nature of RNA viruses, an evolutionary model that predicts group selection mediated by resistance to the deleterious effects of mutations when sloppy polymerases produce high mutation rates. Experimental evolution and phylogeny have helped us understand the forces that operate in nature in viral populations, including the role of hosts and host diversity.

The family Rhabdoviridae is comprised of over 65 diverse species, including viruses of plants and animals. With the notable exception of the lyssaviruses, the majority of the rhabdoviruses rely on arthropod vectors for transmission. This diversity lends itself to the emergence and re-emergence of zoonotic and epizootic viral diseases. In this chapter we will focus on emergent and re-emergent rhabdoviruses with clinical importance, as well as animal viruses with potential economic relevance. We will also consider a few of the burgeoning numbers of newly identified rhabdoviruses, their potential to become more widespread, and touch on the challenges involved in defining disease association. A number of rhabdoviruses, from diverse genera within the family Rhabdoviridae, can be considered new or emerging. Their possible clinical relevance, together with the potential for rapid spread and global impact makes these viruses an interesting and important group for future surveillance and study.

Rabies virus (RABV), a negative sense, single-strand RNA virus of the Rhabdoviridae family causes one of the most lethal zoonotic diseases. It spreads through the neural pathways from the site of infection to the central nervous system and is most often fatal following the initial presentation of symptoms. These symptoms include the classical hydrophobia and increased aggression or hyper-activity, which is associated with the “furious” form of rabies, but can also present as flaccid paralysis associated with the “dumb” form. Depending upon whether the virus is a field/“street” isolate, laboratory adapted/“fixed” isolate or a vaccine strain, infection can result in different outcomes for infected cells. However, natural cases and in vivo experimental models of peripheral infection follow a similar lifecycle. Beginning with initial replication at the site of infection, the virus then travels through peripheral nerves through retrograde rapid axonal transport to the spinal column and central nervous system. Following replication within the CNS, there is centrifugal spread of the virus through nerves to other tissues, including the salivary glands. The major determinant of RABV tropism is the viral membrane glycoprotein (G), which binds to cellular receptors that are generally restricted to neurons. In addition, we will discuss the role of the other viral proteins in RABV replication and pathogenesis.

Virus-host relationships are profoundly shaped by the host innate immune system, particularly the type I interferon (IFN) system. In the past years, tremendous progress was achieved in elucidating the molecular basis of virus sensing and the mechanisms of host innate immune system activation. In parallel, viral countermeasures allowing viruses to withstand the potent cellular defense were revealed. In this chapter, the activation of mammalian cells by pathogen associated molecular patterns (PAMPs) and the viral countermeasures are described for the prototypic rhabdoviruses, rabies virus (RABV) and vesicular stomatitis virus (VSV). In spite of a similar organization of RNA genomes and virus structure, the cell biology of these viruses is highly divergent in important aspects, including their interplay with the intrinsic antiviral host defense. The insect-transmitted VSV causes a rapid general host cell shut down to prevent cellular responses, involving multiple activities of the matrix (M) protein. In contrast, the neurotropic RABV, which is directly transmitted between mammals, relies on a stealth strategy. In this virus, the phosphoprotein (P) has evolved multiple functions to interfere with specific steps of the activation and function of the IFN system, as well as with the function of antiviral proteins.

Rabies is the most fatal infectious disease in the world, causing more than 55,000 deaths annually. Vaccination can prevent the disease either in pre- or post-exposure settings. However, there is no effective treatment available once symptoms develop. Inactivated cell culture rabies virus (RABV) vaccines are now extensively used and have been demonstrated to be safe and effective for humans, however, the high cost and complicated vaccination regimen impede their usage in many less-developed countries. Therefore, there is a need for development of more affordable and efficacious rabies vaccines and control strategies. Most importantly, human rabies can be prevented by controlling rabies in domestic (particularly dogs) and wild animals. Inactivated rabies vaccines are used widely for domestic animals and live-attenuated RABVs or live-recombinant vaccinia viruses have been developed and licensed as oral vaccines for wild animals. However, the inactivated vaccines are expensive to utilize in developing countries and the live-attenuated or live-recombinant vaccines possess safety concerns. There is a large population of stray dogs that are not suitable for any of the above-mentioned vaccines. Therefore it is urgent to explore novel vaccines for stray dogs or wild animals. This chapter is intended to summarize current rabies vaccines used in humans and animals or vaccines in the development stage including DNA, live-recombinant, live-attenuated, and replication-deficient rabies vaccines for both human and animals.

Filoviruses are unusual in being filamentous animal viruses. Like other mononegaviruses, filoviruses have helical nucleocapsids (NCs), which are the most complex amongst this virus order, containing 5 proteins: L, NP, VP35, VP30 and VP24 along with the 18.9 kb monopartite, negative-sense RNA genome. The filovirus NC is also unique in that it forms a double-layered helix, with an inner layer containing RNA and nucleoprotein that is stabilized by an outer layer of VP24-VP35 bridges. These NCs bud from the cell surface during morphogenesis as they gain an envelope containing two integral plasma membrane proteins, the matrix protein (VP40) and the glycoprotein (GP). The putative receptor-binding site is occluded by the mucin-like domains within the native GP molecule, while a major neutralizing epitope is exposed on the surface proximal to the viral envelope. The VP40 forms a regular lattice within the envelope, although its contacts with the NC are irregular. Non-equivalent molecular contacts occur between VP40 and the NC proteins and are important for morphogenesis. Unusual amongst viruses, filoviruses can package multiple copies of the RNA genome to produce polyploid virus particles with an extreme degree of length polymorphism. Up to ~50% of Ebola virus particles are polyploid and virions with 10 or more genome copies have been observed. Overall, filoviruses have a well-ordered, symmetrical NC that is packaged within a flexible, tubular membrane envelope, giving rise to several types of virions, including single-genome filamentous particles, multiple-genome continuous or linked virions. Comma-shaped, checkmark-shaped, and torus-shaped particles are also produced, in which the NC is bent back on itself within the envelope. These various morphologies may play roles in the ability of many filoviruses to replicate to very high titers and to infect a wide range of different hosts and tissues. In addition, molecular structural information may reveal targets for therapeutic approaches.

Filoviruses, including Ebola and Marburg, are among the deadliest viruses known and are capable of causing severe hemorrhagic fever outbreaks with case fatality rates as high as 90%, depending on the virus. Almost all filovirus outbreaks have so far originated and occurred in Africa, with a frequency that has been increasing since the first outbreak was recorded in 1967. Mounting ecological and epidemiological evidence now implicates several species of bats as at least one reservoir for filoviruses. Studies suggest that contact with these animals, as well as certain other susceptible hosts, including primates, plays a significant role in transmitting virus to humans. Indeed, it is thought that human filovirus outbreaks originate from multiple spillover events from a widely distributed reservoir or host, an assertion that is supported by serological evidence indicating a potentially large geographic distribution of these viruses. In humans, most filoviruses cause severe hemorrhagic fever that often results in death precipitated by multi-organ failure and shock. Early infection of phagocytic cells permits rapid and systemic virus replication that elicits the formation of a cytokine storm and a critically dysfunctional immune system. The defective immune response, in conjunction with massive virus replication, results in the coagulation and vascular abnormalities that are hallmarks of filovirus hemorrhagic fevers, as well as the extreme organ and tissue damage that, in many cases, overwhelms the patient. Despite the increasing frequency of filovirus outbreaks, and the risk of importation or bioterrorism, there are still no licensed vaccines or therapeutics to treat filovirus infections.

Filoviruses cause highly lethal zoonotic infections in humans and non-human primates. The filamentous nucleocapsid core of filovirus virions is enveloped by a host-derived lipid bilayer, in which are embedded trimers of the viral glycoprotein (GP). GP is both sufficient and necessary for filovirus entry into susceptible target cells. Intense research in the last couple of decades has greatly increased our understanding of how these viruses exploit cellular pathways to enter cells and initiate infection. Virions are taken up by a macropinocytosis-like process and trafficked to endo/lysosomes where endosomal cysteine proteases prime the viral GP to bind to the essential endosomal receptor, Niemann-Pick C1 (NPC1). After engaging NPC1, viral GP is proposed to undergo extensive conformational changes that lead to fusion of the viral envelope with a cellular membrane, thereby releasing the viral genome into the cytoplasm. This chapter reviews our current understanding of the filovirus entry mechanism, while emphasizing important unanswered questions in the field.

Filoviruses belong to the group of nonsegmented, negative-sense (NNS) RNA viruses and are members of the order Mononegavirales along with the rhabdo-, paramyxo-, nyami-, and bornaviruses. Mononegaviruses share a general mechanism to replicate and transcribe their genomes, reflected in functionally homologous proteins and a similar genome structure. Although basic mechanisms are similar between rhabdo- and filoviral RNA synthesis, both virus families have developed distinct features. In this chapter, we describe the concepts underlying filovirus replication and transcription. We introduce the technologies that have enabled scientists to decipher the viral replication cycle and discuss recent insights in the regulation of filoviral RNA synthesis. We also highlight differences in RNA synthesis between filoviruses and two widely studied rhabdoviruses, vesicular stomatitis virus (VSV) of the Vesiculovirus genus and rabies virus (RABV) of the Lyssavirus genus.

The filoviruses, which include ebolaviruses (EBOVs) and marburgviruses (MARVs), are noteworthy because they cause severe, often deadly, hemorrhagic fever in humans. A central feature of filovirus biology is the ability to antagonize innate antiviral defenses of the host. This capacity likely contributes to uncontrolled virus replication and the severity of filoviral disease. This chapter reviews our current understanding of the ways in which filoviruses counter host innate antiviral defenses. Among the mechanisms discussed are VP35 proteins which inhibit several features of the innate immune response to infection, including RIG-I-like receptor signaling pathways that typically trigger interferon (IFN) production. The viruses also block the ability of infected cells to respond to IFNs. The EBOV VP24 protein inhibits nuclear translocation of the STAT1 transcription factor that is critical for IFN-induced gene expression. In MARVs, the VP40 protein blocks IFN-induced activation of the kinase Jak1, which normally phosphorylates and activates STAT1. In addition, these viruses are able to counteract the antiviral function of the IFN-induced protein tetherin, to modulate viral polymerase expression in response to cell stress and to avoid activating apoptosis. Understanding these immune evasion functions provides insights into the severity of filovirus disease and suggests new therapeutic approaches.

For more than 35 years the filoviruses, Ebola virus and Marburg virus, have caused sporadic outbreaks of hemorrhagic fever resulting in severe and often fatal disease in humans and nonhuman primates. Pathogenic Ebola and Marburg viruses are endemic in resource-poor regions in Central and West Africa and are also of concern as they have the potential for deliberate misuse as bioweapons. Although no vaccines or antivirals for filoviruses are currently licensed for human use, remarkable progress has been made in developing candidate preventive vaccines and antivirals against Ebola and Marburg viruses in nonhuman primate models. Two efficacious non-replicating vaccines tested in nonhuman primates consist of DNA or virus-like particles, while most of the effective vaccines are based on viral vectors including: recombinant adenoviruses, alphaviruses, paramyxoviruses, and rhabdoviruses. Successful post-exposure treatments against filoviruses to date consist of recombinant vesicular stomatitis virus vaccine vectors, antibody cocktail treatments, targeting virus genes or genomes through RNA interference, and use of a nucleoside analogue. While these vaccines and antivirals have shown great promise in nonhuman primate models, there are questions that remain in regard to providing broad-spectrum immunity and/or treatments for the different species of Ebola and strains of Marburg viruses.

The following section is included:

• Index