Infective Virus Substructure from Vesicular Stomatitis Virus

Treatment of suspensions of vesicular stomatitis virus with Tween-ether results in a rapid and considerable loss of infectivity (ca. 4 logs in 2 min), but the residual infectivity is comparatively stable to further treatment with ether. The infectivity remaining after the short exposure to Tween-ether is not due to virus for the following reasons. (i) It is much less infective for tissue cultures than for mice, whereas the intact virion is equally infective for both hosts. (ii) The residual infectivity is much less stable than virus infectivity in both sucrose and tartrate gradients. (iii) Virus immune serum does not neutralize its activity. (iv) The infectivity is associated with material which sediments further in sucrose gradients and has a greater buoyant density in tartrate gradients than the virion. Experiments with (32)P-labeled virion showed that the infective substructure contains ribonucleic acid with the same sedimentation characteristics as that extracted from the virion. Electron microscopy shows that the infective component has the same overall bullet-like structure as the virion but lacks the outer envelope and fringe structure.     

Proteins of Vesicular Stomatitis Virus and of Phenotypically Mixed Vesicular Stomatitis Virus-Simian Virus 5 Virions

The identity of the glycoprotein of vesicular stomatitis virus (VSV) as the spike protein has been confirmed by the removal of the spikes with a protease from Streptomyces griseus, leaving bullet-shaped particles bounded by a smooth membrane. This treatment removes the glycoprotein but does not affect the other virion proteins, apparently because they are protected from the enzyme by the lipids in the viral membrane. The proteins of phenotypically mixed, bullet-shaped virions produced by cells mixedly infected with VSV and the parainfluenza virus simian virus 5 (SV5) have been analyzed by polyacrylamide gel electrophoresis. These virions contain all the VSV proteins plus the two SV5 spike proteins, both of which are glycoproteins. The finding of the SV5 spike glycoproteins on virions with the typical morphology of VSV indicates that there is not a stringent requirement that only the VSV glycoprotein can be used to form the bullet-shaped virion. On the other hand, the SV5 nucleocapsid protein and the major non-spike protein of the SV5 envelope were not detected in the phenotypically mixed virions, and this suggests that a specific interaction between the VSV nucleocapsid and regions of the cell membrane which contain the nonglycosylated VSV envelope protein is necessary for assembly of the bullet-shaped virion.     

Carbohydrate Composition of Vesicular Stomatitis Virus

Analysis by gas-liquid chromatography of the trimethylsilylated sugar residues of purified vesicular stomatitis virus grown in L cells or chick embryo cells revealed the presence in the whole virion of four hexoses (glucose, galactose, mannose, and fucose), two hexosamines (glucosamine and galactosamine), and 34 to 40% neuraminic acid. The isolated viral glycoprotein was devoid of galactosamine and fucose, both of which sugars were present in whole virions presumably as part of the membrane glycolipids.     

Entry of Vesicular Stomatitis Virus into L Cells

Early stages of the entry of vesicular stomatitis (VS) virus into L cells were followed by electron microscopy with the aid of ferritin antibody labeling. Cells which were infected at 0 C and incubated for 10 min at 37 C were reacted first with antiviral-antiferritin hybrid antibody and then with ferritin or fluorescein-labeled apoferritin. Extensive ferritin labeling of the cell surface was detected by both electron and fluorescence microscopy. The labeled regions of the cell surface were continuous with and indistinguishable from the rest of the host cell membrane, suggesting incorporation of viral antigens into the cell surface during viral penetration. Fusion of parental viral membrane with host cell membrane was further demonstrated by examining the localization of (3)H-labeled viral structural proteins in cells infected at 0 C and incubated for short periods at 37 C. Viral nucleoprotein was found in a soluble fraction of the cells which was derived primarily from the cytoplasm, whereas a particulate fraction from the cells was enriched in viral envelope proteins. Cytoplasmic membrane was isolated from these cells, and this membrane contained viral envelope proteins. These results suggest that penetration by VS virus occurs by fusion of the viral and cellular membranes followed by release of nucleo-protein into the cytoplasm.     

Comparison of Vesicular Stomatitis Virus Intracellular and Virion Ribonucleoproteins

Vesicular stomatitis virus ribonucleoproteins (RNP) obtained by a detergent treatment of purified virus (vRNP) or from infected HeLa cell cytoplasm (icRNP) were examined by sedimentation in sucrose or Renografin gradients in the presence or absence of EDTA. It was shown that vRNP and icRNP sediment at the same rate in sucrose and Renografin in the absence of EDTA; however, icRNP sedimented more slowly in the presence of EDTA than did vRNP. Polyacrylamide gel electrophoresis of the proteins of vRNA and icRNP recovered from EDTA-containing gradients demonstrated that both RNP structures contained L, N, and NS proteins in the same proportion. Electron microscopy of both RNP structures, in the absence of EDTA, demonstrated that both exist as helical structures ~20 by 700 nm. However, in the presence of EDTA the icRNP was completely uncoiled with a mean length of 4,095 nm, whereas vRNP was hardly affected. The addition of excess Mg²⁺ or Mn²⁺ to uncoiled icRNP preparations partially restored the coiled configuration. These observations suggest that the change in sedimentation of icRNP in the presence of EDTA is due to a change from a coiled to an uncoiled conformation, that icRNP and vRNP are not structurally identical, and that icRNP must undergo a conformational change during maturation of VSV from the 20-by-700-nm intracellular form to the 50-by-175-nm form found in intact virus. The icRNP containing L, N, and NS proteins (icRNP^L,N,NS) and icRNP containing only N protein (icRNP^N), prepared by centrifugation of icRNP^L,N,NS in CsCl to remove L and NS, were compared by cosedimentation in sucrose gradients. There was a decrease in sedimentation rate of icRNP^N due to loss of L and NS. This sedimentation difference was also apparent in the presence of EDTA; however, both icRNP^L,N,NS and icRNP^N sedimented at a much slower rate in the presence of EDTA, and by electron microscopy both were completely uncoiled. These observations suggest that N protein alone is responsible for the 20-by-700-nm coiled structure and that the divalent cation interactions disrupted by EDTA are N-N or N-RNA interactions. These results are discussed with regard to vesicular stomatitis virus maturation.     

Association of Ribosomes with Intracellular Vesicular Stomatitis Virus Particles

Ribosomes are observed intimately associated with nucleocapsids of vesicular, stomatitis virus, especially those that line structures that are either cytoplasmic vesicles or invaginations of the plasma membrane.     

Characteristics of Oncolytic Vesicular Stomatitis Virus Displaying Tumor-Targeting Ligands

We sought proof of principle that tumor-targeting ligands can be displayed on the surface of vesicular stomatitis virus (VSV) by engineering its glycoprotein. Here, we successfully rescued VSVs displaying tumor vasculature-targeting ligands. By using a rational approach, we investigated various feasible insertion sites on the G protein of VSV (VSV-G) for display of tumor vasculature-targeting ligands, cyclic RGD (cRGD) and echistatin. We found seven sites on VSV-G that tolerated insertion of the 9-residue cRGD peptide, two of which could tolerate insertion of the 49-amino acid echistatin domain. All of the ligand-displaying viruses replicated as well as the parental virus. In vitro studies demonstrated that the VSV-echistatin viruses specifically bound to targeted integrins. Since the low-density lipoprotein receptor (LDLR) was recently identified as a major receptor for VSV, we investigated the entry of ligand-displaying viruses after masking LDLR. The experiment showed that the modified viruses can enter the cell independently of LDLR, whereas entry of unmodified virus is significantly blocked by a specific monoclonal antibody against LDLR. Both parental and ligand-displaying viruses displayed equal oncolytic efficacies in a syngeneic mouse myeloma model. We further demonstrated that single-chain antibody fragments against tumor-specific antigens can be inserted at the N terminus of the G protein and that corresponding replication-competent VSVs can be rescued efficiently. Overall, we demonstrated that functional tumor-targeting ligands can be displayed on replication-competent VSVs without perturbing viral growth and oncolytic efficacy. This study provides a rational foundation for the future development of fully retargeted oncolytic VSVs.     

Attenuation of Vesicular Stomatitis Virus Encephalitis through MicroRNA Targeting

Vesicular stomatitis virus (VSV) has long been regarded as a promising recombinant vaccine platform and oncolytic agent but has not yet been tested in humans because it causes encephalomyelitis in rodents and primates. Recent studies have shown that specific tropisms of several viruses could be eliminated by engineering microRNA target sequences into their genomes, thereby inhibiting spread in tissues expressing cognate microRNAs. We therefore sought to determine whether microRNA targets could be engineered into VSV to ameliorate its neuropathogenicity. Using a panel of recombinant VSVs incorporating microRNA target sequences corresponding to neuron-specific or control microRNAs (in forward and reverse orientations), we tested viral replication kinetics in cell lines treated with microRNA mimics, neurotoxicity after direct intracerebral inoculation in mice, and antitumor efficacy. Compared to picornaviruses and adenoviruses, the engineered VSVs were relatively resistant to microRNA-mediated inhibition, but neurotoxicity could nevertheless be ameliorated significantly using this approach, without compromise to antitumor efficacy. Neurotoxicity was most profoundly reduced in a virus carrying four tandem copies of a neuronal mir125 target sequence inserted in the 3′-untranslated region of the viral polymerase (L) gene.Vesicular stomatitis virus (VSV) is a nonsegmented, negative-strand rhabdovirus widely used as a vaccine platform as well as an anticancer therapeutic. While VSV is predominantly a pathogen of livestock (34), it has a very broad species tropism. The cellular tropism of VSV is determined predominantly at postentry steps, since the G glycoprotein of the virus mediates entry into most tissues in nearly all animal species (10).Though viral entry can take place in nearly all cell types, in vivo models of VSV infection have revealed that the virus is highly sensitive to the innate immune response, limiting its pathogenesis (4). VSV is intensively responsive to type I interferon (IFN), as the double-stranded RNA (dsRNA)-dependent PKR (2), the downstream effector of pattern recognition receptors MyD88 (32), and other molecules mediate shutdown of viral translation and allow the adaptive immune response to clear the virus. The vulnerability of the virus to the type I IFN response, typically defective in many cancers, has been exploited to generate tumor-selective replication (49), such that the virus is now poised to enter phase I trials. However, the virus remains potently neurotoxic, causing lethal encephalitis not only in rodent models (7, 22, 53) but also in nonhuman primates (25).VSV very often infiltrates the central nervous system (CNS) through infection of the olfactory nerves (41). When administered intranasally, the virus replicates rapidly in the nasal epithelium and is transmitted to olfactory neurons, from which it then moves retrograde axonally to the brain and replicates robustly, causing neuropathogenesis. While intranasal inoculation does cause neuropathy in mice, neurotoxicity following viral administration also occurs when the virus is delivered intravascularly (47), intraperitoneally (42), and (not surprisingly) intracranially (13). Previously, other groups have modified the VSV genome to be more sensitive to cellular IFNs (49) and have actually encoded IFN in the virus (36). However, the former can result in attenuation of the virus, such that it has reduced anticancer potential, while the latter still results in lethal encephalitis (unpublished results). In order to mitigate the effects of VSV infection on the brain without perturbing the potent oncolytic activity of the virus, we utilized a microRNA (miRNA) targeting paradigm, whereby viral replication is restricted in the brain without altering the tropism of the virus for other tissues.To redirect the tissue tropism of anticancer therapeutics, we (26) and others (11, 14, 55) have previously exploited the tissue-specific expression of cellular miRNAs. miRNAs are ∼22-nucleotide (nt) regulatory RNAs that regulate a diverse and expansive array of cellular activities. Through recognition of sequence-complementary target elements, miRNAs can either translationally suppress or catalytically degrade both cellular (6) and viral (50) RNAs. We have determined that cellular miRNAs can potentially regulate numerous steps of a virus life cycle and that this regulation of the virus by endogenous miRNAs can then abrogate toxicities of replication-competent viruses (27; E. J. Kelly et al., unpublished data).miRNAs are known to be highly upregulated in many different tissues, including (but not limited to) muscle (40), lung (44), liver (15, 44), spleen (44, 46), and kidney (51). In addition, the brain has a number of upregulated miRNAs, with each different subtype of cell having a unique miRNA profile. miR-125 is highly upregulated in all cells in the brain (neurons, astrocytes, and glia cells), while miR-124 is found predominantly in neuronal cells (48). Glial cells and glioblastomas are thought to have decreased expression of miR-128 compared to neurons (17), while miR-134 is particularly abundant in dendrites of neurons in the hippocampus (43). In addition to these miRNAs, the tumor suppressor miRNA let-7 and miRs 9, 26, and 29 (51) are also found to be enriched in the brain, with expression varying not only between different cell types and regions of the brain but also temporally (48).MicroRNAs have previously been exploited to modulate the tissue tropism of nonreplicating lentiviral vectors (8, 9), as well as curbing known toxicities of replication-competent picornaviruses (5, 26), adenoviruses (11), herpes simplex virus 1 (33), and influenza A virus (39). In addition, a recombinant VSV encoding a tumor suppressor target was found to be responsive to sequence-complementary miRNAs in vitro, possibly by affecting expression of the matrix (M) protein (14), and evidence from Dicer-deficient mice suggests that endogenously expressed microRNA targets within the P and L genes of VSV could restrict enhanced pathogenicity of the virus (37). However, in vivo protection from neuropathogenesis by this means has not been demonstrated for VSV.Here we evaluate the efficiencies of different brain-specific miRNAs for shutting down gene expression and extensively characterize the ability of miRNA targeting to attenuate the neurotoxicity of vesicular stomatitis virus in vivo. We constructed and evaluated recombinant VSVs with miRNA target (miRT) insertions at different regions of the viral genome, with special focus upon those affecting viral L expression. In addition, we looked at the regulatory efficiency of different brain-specific miRNAs and the impact of miRT orientation on VSV replication and determined the impact of the virus on oncolytic activity in vivo.     

Incorporation of L Cell Sterols into Vesicular Stomatitis Virus

The incorporation of host cell sterol into vesicular stomatitis virus can be effectively studied in an L cell system. The end product of de novo sterol synthesis in the L cell is desmosterol, and as the concentration of cholesterol in the medium is increased the cells incorporate the exogenous cholesterol and the synthesis of desmosterol decreases. L cells which contained desmosterol as their sole sterol produced virus whose sterol content was similarly composed of only desmosterol. Virus grown in L cells which had a constantly changing sterol ratio also contained a mixture of cholesterol and desmosterol, but the virus was found to be more enriched in cholesterol than in the L cells in which it was grown. Viral stability, growth, and plaquing efficiency were tested and found not to be affected by the alteration of its sterol composition, i.e., by partially or completely replacing cholesterol with desmosterol.     

Adenylate-Rich Sequences in Vesicular Stomatitis Virus Messenger Ribonucleic Acid

Vesicular stomatitis virus (VSV)-specific messenger ribonucleic acid (mRNA) species contain sequences of adenylate-rich RNA which are more heterogeneous in their migration through sodium dodecyl sulfate-polyacrylamide gels than the corresponding fractions from HeLa cell mRNA. VSV virion RNA contains no adenylaterich sequences. The possible role of such sequences in the mRNA species of a cytoplasmically replicating virus is discussed.     

Comparison of Membrane Protein Glycopeptides of Sindbis Virus and Vesicular Stomatitis Virus

A comparison has been made of the membrane glycoproteins and glycopeptides from two enveloped viruses, Sindbis virus and vesicular stomatitis virus (VSV). Glycopeptides isolated from Sindbis virus and VSV grown in the same host appear to differ principally in the number of sialic acid residues per glycopeptide; when sialic acid is removed by mild acid treatment, the glycopeptides of the two viral proteins are indistinguishable by exclusion chromatography. Preliminary evidence argues that the carbohydrate moiety covalently bound to different virus-specified membrane proteins may be specified principally by the host.