Human respiratory syncytial virus (RSV) is an enveloped RNA-containing virus of Family Paramyxoviridae, Order Mononegavirales, the nonsegmented negative strand RNA viruses. RSV is the most important viral agent of pediatric respiratory tract disease worldwide but lacks an approved vaccine or effective antiviral therapy. We previously showed that the RSV genome is a single negative strand of RNA of 15,222 nucleotides that encodes 10 major mRNAs and 11 proteins, and in previous studies identified the functions of a number of these proteins. We have continued to study the M2-1 protein, which was previously shown to be a novel transcription anti-termination protein. In one set of studies, which employed a small cDNA-encoded mini-replicon complemented by plasmid-expressed viral nucleocapsid and polymerase proteins, we studied the ability of the polymerase to correctly recognize wild type and mutant transcription gene-end signals in the presence and absence of M2-1. This showed that M2-1 facilitates read-through of gene-end signals, which in nature gives rise to the synthesis of polycistronic transcripts that constitute a small fraction of the total viral mRNA. This readthrough effect of M2-1 was particularly pronounced on suboptimal gene-end signals, indicating a sequence-specificity to this effect. We also examined the effect of the M2-1 protein on movement of the polymerase from one gene to the next across the intergenic regions. This showed that the presence of M2-1, which is essential for efficient transcription within genes, has no effect on the ability of the polymerase to traverse long intergenic regions between genes. It might be that M2-1 cycles off and on the transcribing complex at the gene junctions. We also have studied the glycoproteins involved in viral attachment and penetration of host cells using mutant recombinant RSV. We previously showed that the putative attachment protein G and the small hydrophobic protein SH could be deleted from virus without much effect on growth in certain cells in vitro. By the process of elimination, this implies that the F protein can act as an auxiliary attachment protein, at least in vitro. Virus containing only F as a viral surface protein appeared to enter cells by two pathways, one that involves cell surface glycosaminoglycans and a second that remains uncharacterized. Although virus lacking G can replicate efficiently in vitro, it was found to replicate very inefficiently in the respiratory tract of mice. This supports the idea that G is the major attachment protein in vivo, even if it is dispensable in vitro. In 1987, we showed that the external domain of the RSV G protein was very divergent among RSV strains, but exhibited a high degree of conservation in a very circumscribed domain that had characteristics of being a cysteine-bound knob. This domain was considered to be a likely candidate to be involved in receptor binding in vivo. However, we have now found that deletion of this domain in recombinant virus resulted in a mutant that replicated with wild type efficiency in vitro and in vivo. Thus, the conservation of this domain presumably reflects some function other than attachment, perhaps one involving the CX3C chemokine-like motif that is contained within this segment. We also initiated study of RSV infection in primary human airway cells that are grown on a filter membrane at an air-liquid interface and form a differentiated pseudostratified polarized mucocilliary tissue that is nearly indistinguishable morphologically and functionally from ex vivo airway epithelium. Infection of these cultures with a recombinant RSV that expresses the green fluorescent protein showed that RSV infection, replication and virus release was restricted to ciliated cells at the apical face. This will help refocus research to identify cellular proteins involved in RSV attachment. Remarkably, infected ciliated cells exhibited no obvious cytopathology over a period of several weeks, in contrast to the rapid and extensive cell destruction by an influenza virus control, and the virus appeared to spread by ciliary beat. Thus, while RSV is highly cytopathic in monolayer cultures of immortalized cell lines, it is much less cytopathic in highly organized, polarized multicellular tissue that resemble the epithelium of the human airway. This suggests that host immunity, rather than direct viral cytopathogenesis, is responsible for the damage that occurs to the ciliated epithelial cells in the airways during RSV infection in vivo. The observation that RSV is specific for ciliated cells will help guide the search for cellular proteins involved in RSV infection, and this in vitro model will be useful for studying the response of the epithelium to viral infection.
Showing the most recent 10 out of 19 publications
