We use recombinant methods to develop attenuated strains of human respiratory syncytial virus (HRSV) as potential intranasal pediatric vaccines. Five recombinantly-derived viruses have been evaluated to date in clinical studies. The most promising virus, called rA2cp2484041030delSH, was safe and immunogenic in young infants. However, there was evidence of the loss of the 248 or 1030 attenuating point mutation (Tyr-1321 or Gln-831-Leu in the L protein) in some isolates. We evaluated a strategy to stabilize these mutations. We constructed a panel of viruses for each mutation representing all 20 possible amino acid assignments. These were analyzed for the temperature-sensitive (ts) and attenuation (att) phenotypes, thus identifying the various assignments as att, intermediate, and wt. The goal is to examine all possible codon choices for each amino acid and identify an att assignment that differs from all possible wt-like assignments by as many nt as possible, on the premise that an att-to-wt reversion involving a change of 1 nt has a frequency of approximately 10(-4) (the viral mutation frequency) whereas the frequencies associated with changes of 2 or 3 nt would be 10(-8) and 10(-12), respectively. In the case of the 248 mutation, viruses representing 18 of the 20 possible amino acid assignments were recoverable, the other possible assignments being lethal. We also evaluated several small deletions or insertions involving this locus, but these viruses could not be recovered. The original leucine mutation was found to be the most att of all of the recoverable assignments, with phenylalanine being the only other substantially att assignment. The paucity of highly att assignments limited the possibility of increasing genetic stability. Indeed, it was not possible to find a leucine or phenylalanine codon requiring more than a single nucleotide change to yield a """"""""non-att"""""""" codon, as is necessary for the stabilization strategy. Nonetheless, serial passage of the six possible leucine codons in vitro at increasing temperatures revealed differences, with slower reversion to non-att phenotypes observed for a subset of codons. Thus, it should be possible to modestly increase the phenotypic stability of the rA2cp248/404/1030delSH vaccine virus by codon modification at the 248 mutation. Analysis of the 1030 mutation is in progress. We also have been preparing additional att versions of HRSV under conditions suitable for making clinical material for evaluation in humans. These involve deletion of the type I interferon (IFN) antagonist NS1 protein or the M2-2 protein that is involved in regulating viral RNA synthesis. A clinical lot of the delM2-2 virus has been prepared and will be evaluated clinically. A seed for the delNS1 virus has been prepared and is presently being amplified to make clinical lot material. HRSV readily infects and reinfects during infancy and throughout life, despite maternal antibodies and immunity from prior infection and without the need for significant antigenic change. HRSV has two neutralization antigens, the F and G glycoproteins. G is expressed in both membrane-bound (mG) and secreted (sG) forms. We found that wt HRSV, expressing both mG and sG, was less sensitive to G-specific neutralizing antibodies in vitro compared to recombinant mG-HRSV, expressing only mG, whereas susceptibility to F-specific antibodies was equivalent. This difference disappeared when the virus preparations were purified to remove sG. Thus, sG appears to function as an antigen decoy that sops up neutralizing antibodies and spares infectious virus. We then passively transferred antibodies into mice and evaluated their effect on the pulmonary replication of wt HRSV versus mG- HRSV. As was observed in vitro, wt HRSV was less sensitive than mG-HRSV to G-specific and RSV-specific antibodies;however, a similar difference was also observed with F-specific antibodies. This confirmed that sG helps wt HRSV evade the antibody-dependent restriction of replication but indicated that, in mice, it is not acting primarily as a decoy for G-specific antibodies. Rather, we found that the greater sensitivity of mG-HRSV versus wt HRSV to passively transferred HRSV antibodies required the presence of inflammatory cells in the lung and was Fc gamma receptor dependent. Thus, sG helps HRSV escape the antibody-dependent restriction of replication via effects as an antigen decoy for G-specific antibodies and by reducing antibody-dependent-cell-mediated-cytotoxicity that is not limited to G-specific antibodies. The lack of a decoy effect in mice might be a consequence of the semi-permissiveness of this animal model for HRSV infection and concomitant poor expression of sG. If so, the effects observed in this study would be heightened in the fully permissive human host. Dendritic cells (DC) are potent antigen presenting cells that play a major role in initiating and modulating the immune response. We compared the effects of HRSV on human monocyte-derived DC in a side-by-side comparison with HMPV and HPIV3 using GFP-expressing viruses. We found that all three viruses infected DCs poorly (whereas a Newcastle disease virus-GFP control infected very efficiently) with only a few percent of cells being GFP+ and the remainder having low, abortive levels of viral RNA synthesis. Low infectivity and low intracellular antigen synthesis of these viruses likely reduces activation of CD8+ T cells important for host defense. The three viruses induced low-to-moderate maturation of DC and cytokine/chemokine responses, which also might reduce the immunological footprint of theses viruses. Infection at the individual cell level tended to be relatively benign, such that GFP+ cells were neither more nor less able to mature compared to GFP- bystanders. However, HPIV3 infection did down-regulate expression of CD38, an effect that was at the RNA level. We are presently evaluating the ability of DC inoculated with HRSV, HMPV, HPIV3, or influenza A virus to activate autologous CD4+ T cells. Pneumonia virus of mice (PVM), a murine relative of HRSV, provides a convenient model for studying severe disease caused by an HRSV-like virus. Using a reverse genetics system for PVM that we previously developed, we deleted the nonstructural NS1 and NS2 protein genes individually and in combination. Deletion of NS1 resulted in a virus that was substantially attenuated in mice, identifying NS1 as a virulence factor whose mechanistic basis is unknown. Deletion of NS2 was highly attenuating, an effect that was associated with the loss of the ability of PVM to suppress the host IFN response. Wild type PVM replicated efficiently and increased over a 6-day period, resulting in a strong up-regulation of IFN and a wide array of representative pro-inflammatory cytokines/chemokines and T cell-related cytokines and the appearance of overt disease on day 6. This also was observed for the delNS1 mutant, although disease was substantially less. In contrast, virus lacking NS2 was modestly attenuated for replication on day 3 concurrent with an early up-regulation of pulmonary IFN and CXCL10 (IP-10). By day 6, the viral titer was declining, the expression of IFN and CXCL10 was returning to baseline, and no other cytokines or chemokines were induced. These results provide evidence that severe PMV disease is associated with high and seemingly poorly-controlled virus replication driving the expression of high levels of pulmonary type I IFN and an array of cytokines/chemokines. In contrast, in the absence of NS2, there was an early innate response involving moderate levels of IFN and CXCL10 that restricted virus replication early and prevented disease. This illustrated protective versus pathogenic host responses to this HRSV-like virus in a permissive host.
Showing the most recent 10 out of 53 publications
