Human parainfluenza virus (HPIV) serotypes 1, 2, and 3 are significant causes of severe respiratory tract disease in infants and young children worldwide. The HPIVs are enveloped, non-segmented, negative strand RNA viruses of the family Paramyxoviridae. These serotypes are immunologically distinct in that primary infection does not result in cross-neutralization or cross-protection. The HPIV genome encodes three nucleocapsid-associated proteins, namely the nucleoprotein N, phosphoprotein P, and large polymerase protein L, and three envelope-associated proteins, namely the internal matrix protein M and the fusion F and hemagglutinin-neuraminidase HN transmembrane surface glycoproteins. F and HN are the two viral neutralization antigens and the major protective antigens. In addition, the P gene encodes various accessory protein(s)from one or more additional ORFs: C (HPIV1), V (HPIV2), and C, D, and possibly V (HPIV3). HPIV1, 2, and 3 have been identified as the etiologic agents responsible for nearly 20% of hospitalizations of infants and young children for respiratory tract disease. Licensed vaccines or specific antiviral drugs are currently not available for any of the HPIVs. The goal of this project is to investigate viral molecular biology, pathogenesis, and immunobiology, and to use the resulting information and reagents to develop live attenuated vaccines. Candidate vaccine viruses are developed using reverse genetic systems that we developed previously. This provides the means to produce well-defined live vaccines. This year's report will focus on basic research studies. The C proteins of HPIV1 are a nested set of proteins arising from translational initiation at different start codons in the C ORF in the P gene. We previously demonstrated multiple inhibitory effects on host innate immune responses, and in particular on type I interferon (IFN) responses. IFNs play a crucial role in the host antiviral response. Whereas the C proteins of wild-type (WT) HPIV1 inhibit both IFN induction and signaling, a HPIV1 mutant encoding a single amino acid substitution (F170S) in the C proteins is unable to block either response. We previously investigated the inhibition of IFN induction and showed that this potent effect is indirect: the C proteins down-regulate viral RNA synthesis so as to prevent the formation of viral dsRNA that otherwise triggers IFN induction well as activation of dsRNA-regulated protein kinase PKR. In the present report, we examined signaling downstream of the IFN receptor in Vero cells to define at what stage WT HPIV1 can block, and F170S HPIV1 fails to block, IFN signaling. WT HPIV1 inhibited phosphorylation of both Stat1 and Stat2, and this activity was only slightly reduced for F170S HPIV1. Degradation of Stat1 or Stat2 was not observed. The HPIV1 C proteins were found to accumulate in the perinuclear space, often forming large granules, and co-localized with Stat1 and the mannose 6-phosphate receptor (M6PR) that is a marker for late endosomes. Upon stimulation with IFN-β, both the WT and F170S C proteins remained in the perinuclear space, but only the WT C proteins prevented Stat1 translocation to the nucleus. In addition, WT HPIV1 C proteins, but not F170S C proteins, co-immunoprecipitated Stat1 (and pStat1). Our findings suggest that the WT HPIV1 C proteins form a stable complex with Stat1 in perinuclear granules that co-localize with M6PR, and that this direct interaction between the WT HPIV1 C proteins and Stat1 is the basis for the ability of HPIV1 to inhibit IFN signaling. The F170S mutation in HPIV1 C did not prevent perinuclear co-localization with Stat1, but apparently weakened this interaction such that, upon IFN stimulation, Stat1 was translocated to the nucleus to induce an antiviral response. These results are of particular relevance because the F170S mutation is represented in an HPIV1 vaccine candidate presently in clinical trials. In another study, we compared HPIV1, 2, and 3 for the kinetics of replication and cytokine production in an in vitro model of human airway epithelium (HAE). This model consists of primary HAE cells from donors that are grown in vitro at an air-liquid interface and differentiate into a pseudostratified, polarized, mucociliary tissue that is functionally and physically a close facsimile of authentic HAE. This comparison was of interest because HPIV1, 2, and 3 differ in their frequency of infection of humans and in their disease spectrum: HPIV3 is a common cause of bronchiolitis and pneumonia, whereas HPIV1 and 2 are frequent causes of upper respiratory tract illness and croup. To assess how HPIV1, 2, and 3 differ with regard to replication and induction of type I IFNs and other relevant cytokines, we infected HAE cultures from the same tissue donors and examined replication kinetics and cytokine secretion. HPIV1 replicated to high titer, but the secretion of cytokines was late and low compared to the other viruses. HPIV2 replicated less efficiently but induced an early cytokine peak. HPIV3 replicated to high titer and induced strong rises in cytokine secretion that occurred somewhat later than HPIV2. The T cell chemoattractants CXCL10 and CXCL11 were the most abundant chemokines induced. Differences in replication and cytokine secretion might explain some of the differences in HPIV serotype-specific pathogenesis and epidemiology. We also used this HAE model to investigate the pathogenicity of parainfluenza 5 (PIV5) in human cells. PIV5 is well known for causing non-cytopathic infections in immortalized cultures of epithelial-type cells. It has been suggested that this feature is relevant to infection by PIV5 in vivo. Specifically, it has been suggested that PIV5 causes noncytopathic infections that are long term and make it possible for the virus to persist in an infected host. The natural history of PIV5 is poorly understood. It has been shown to be an important cause of respiratory disease in dogs (kennel cough), but it also has been isolated from monkey tissue and has been suggested as a possible infectious agent of humans and possibly a human pathogen. Therefore, it was of interest to investigate infection of human HAE. We infected parallel cultures of HAE with PIV5, HPIV3, and respiratory syncytial virus (RSV) and measured outcomes of cytopathology. Infection by PIV5 and HPIV3 was dependent on sialic acid residues, as would be expected. Only PIV5-infected cells formed syncytia. In contrast, RSV and HPIV3 do not form syncytia in these cultures, which is consistent with the limited pathology data available from infected humans. PIV5 infection resulted in a more rapid loss of infected cells by shedding from the infected culture into the apical overlay. These studies revealed striking and unexpected differences in the cytopathology of PIV5 versus HPIV3 or RSV. This showed that cell lines are poor predictors of cytopathology in primary differentiated epithelium. These results also suggest that, in authentic tissue, PIV5 is not a non-cytopathic virus. This study raises the possibility that the long standing speculation that PIV5 forms long-lasting, noncytopathic infections may be a feature that is specific to immortalized cell lines and may not be relevant in vivo. We also have been evaluating HPIV1, 2, and 3 as vectors for expressing the protective F and G proteins of RSV. This strategy involves using an attenuated version of HPIV1, 2, or 3 to express RSV F or G from an added gene. This provides a bivalent vaccine against both the PIV vector and RSV. This avoids the poor growth and physical lability of RSV, and may be particularly important for providing RSV vaccines to resource-challenged settings worldwide. We are investigating the effects of insert position, codon-optimization, and other features on vaccine performance. These studies will be reported next year.
Showing the most recent 10 out of 17 publications
