Laboratory Studies of Human Respiratory Syncytial Virus and Other Pneumoviruses

National Institute Of Allergy And Infectious Diseases

Investigators

Linked publications, trials & patents

Abstract

Human respiratory syncytial virus (RSV) and human metapneumovirus (HMPV) are cytoplasmic enveloped RNA viruses of the pneumovirus family. Their genomes are single strands of negative-sense RNA of 15.2 kb (RSV) or 13.3 kb (HMPV) that encode 10 mRNAs and 11 unique proteins (RSV) or 8 mRNAs and 9 unique proteins (HMPV). Each virus encodes a nucleoprotein N, phosphoprotein P, matrix protein M, small hydrophobic protein SH, major glycoprotein G, fusion glycoprotein F, polymerase factors M2-1 and M2-2, and the polymerase protein L. In addition, RSV encodes 2 nonstructural proteins NS1 and NS2. Murine pneumonia virus (MPV, previously known as pneumonia virus of mice, PVM) is a close relative of RSV whose natural host is the mouse and which provides a convenient permissive animal model. We recently evaluated the strategy of codon-pair deoptimization (CPD) as a means to attenuate RSV. In this strategy, one or more ORFs are recoded to introduce under-represented codon-pairs, with no changes in amino acid coding, codon usage, or nucleotide composition. This can cause suboptimal translation, among other effects, resulting in attenuation. Recoding potentially can involve hundreds or thousands of changes and thus should be refractory to de-attenuation. We made 4 CPD derivatives of RSV: Min A (CPD NS1, NS2, N, P, M, and SH), Min B (CPD G and F), Min L (CPD L), and Min FLC (CPD all ORFs except M2-1 and M2-2). In vitro, the CPD RSVs replicated more slowly and to lower titers than wt RSV even at the normally permissive temperature of 32C, with the following order of growth efficiency: WT>Min L>Min A>Min FLC>Min B. At higher incubation temperatures, all 4 of the CPD viruses were temperature-sensitive (ts), with the order of increasing sensitivity Min A<Min B<Min L<Min FLC. The CPD RSVs exhibited a range of attenuation phenotypes in mice and African Green Monkeys (AGMs). Min FLC was the most restricted, followed by Min L. Min L was similar to wt RSV in immunogenicity despite its growth restriction, suggesting that it was a promising vaccine candidate. Min L FLC was significantly less immunogenic, consistent with its high level of restriction, raising the possibility that it might be over-attenuated. During the present review period, Min FLC and Min L were evaluated for genetic and phenotypic stability by in vitro stress tests, in which 10 replicate cultures were infected and passaged serially in parallel at progressively increasing temperatures, with 2 additional control flasks passaged in parallel at 32C. Each passage level was evaluated for virus replication, and virus from each level was analyzed to determine the consensus sequence for the full genome. Min FLC (containing 2,692 silent mutations) was genetically stable, showing that it is a highly attenuated, stable RSV vaccine candidate. In contrast, Min L (1,378 silent mutations) quickly lost substantial attenuation. Numerous mutations were detected by sequence analysis. A number of these occurred in the CPD L ORF, but many also occurred in other ORFs that had not been subjected to CPD. Different lineages of virus passaged in parallel acquired different groups of mutations, with some overlap. Mutations were more common for proteins of the nucleocapsid/polymerase complex, suggesting that changes in these proteins that function in trans with the L polymerase might help compensate for reduced expression from the CPD L ORF. At least some of these mutations were de-attenuating, because the passaged virus exhibited a partial loss of ts phenotype. A number of individual mutations from the N, P, M2-1, and L ORFs identified from the stress tests were introduced individually and in combinations into Min L by reverse genetics. Analysis of the resulting viruses confirmed that a number of these mutations indeed were de-attenuating, and conferred a reduction in temperature sensitivity as well as an increase in expression of the CPD L ORF. In particular, 2 competing mutations in M2-1 substantially reversed the restricted transcription of the CPD L ORF. Molecular modeling based on the published crystal structure of the M2-1 tetramer suggested that each of the 2 major mutations in the M2-1 protein was predicted to increase the stability of the tetramer. A derivative of Min L was made containing 4 mutations that had been identified in the N, P, L, and M2-1 ORFs by the stress test analysis. Unexpectedly, this virus exhibited the paradoxical effect of increased rather than decreased attenuation in hamsters and mice. Paradoxically, it exhibited increased immunogenicity, and was stable during a further stress test. Thus, this provides an improved vaccine candidate. We performed (in collaboration with NCATS) a genome-wide high-throughput siRNA screen of more than 20,000 cellular genes to identify cellular proteins involved in RSV infection, using human airway epithelial A549 cells infected with RSV expressing green fluorescent protein. Hits were validated with 6 siRNAs, and computer analysis was used to identify and discount off-target effects. From the top 65 hits, we are systematically evaluating genes of interest. One hit was actin-related protein 2 (ARP2). ARP2 knockdown did not reduce RSV entry or gene expression during the first 24 h of infection, but decreased viral gene expression thereafter, an effect that appeared to be due to inhibition of viral spread to neighboring cells. Consistent with reduced spread, there was a 10-fold reduction in the release of infectious progeny virions in ARP2-depleted cells. In addition, RSV infection induced filopodia formation and increased cell motility in A549 cells. Filopodia appeared to shuttle RSV to nearby uninfected cells, facilitating virus spread. Knockdown studies indicated that ARP2 was involved in filopodia formation and motility. Expression of the RSV F protein alone by a plasmid or heterologous viral vector induced filopodia, indicating a new role for the RSV F protein, driving filopodia induction and virus spread. Thus, this study identified roles for ARP2 and filopodia in RSV production and cell-to-cell spread. We are evaluating additional genes of interest.