Search

Seed viruses used to produce inactivated H5N1 influenza vaccines are recombinant viruses with modified avirulent-type hemagglutinin (HA) and intact neuraminidase (NA) genes, both derived from an H5N1 isolate, and all remaining genes from the PR8 strain, which grows well in eggs. However, some reassortants grow suboptimally in eggs, imposing obstacles to timely, cost-efficient vaccine production. Here, we demonstrate that our PR8 strain supports better in ovo growth than the PR8 strain used for the WHO-recommended seed virus, NIBRG-14. Moreover, inclusion of an alternative NA protein further enhanced viral growth in eggs. These findings suggest that our H5N1 vaccine candidates would increase the availability of H5N1 vaccine doses at the onset of a new pandemic.

An H5N1 influenza A virus was isolated from duck meat processed for human consumption, imported to Japan from Shandong Province, China in 2003. This virus was antigenically different from other H5 viruses, including the Hong Kong H5N1 viruses isolated from humans in 1997 and 2003. Sequence analysis revealed that six genes (PB1, PA, HA, NA, M, and NS) of this virus showed >97% nucleotide identity with their counterparts from recent H5N1 viruses, but that the remaining two genes (PB2 and NP) were derived from other unknown viruses. This duck meat isolate was highly pathogenic to chickens upon intravenous or intranasal inoculation, replicated well in the lungs of mice and spread to the brain, but was not as pathogenic in mice as H5N1 human isolates (with a dose lethal to 50% of mice (MLD50)=5x10(6) 50% egg infectious doses [EID50]). However, viruses isolated from the brain of mice previously infected with the virus were substantially more pathogenic (MLD50=approximately 10(2) EID50) and possessed some amino acid substitutions relative to the original virus. These results show that poultry products contaminated with influenza viruses of high pathogenic potential to mammals are a threat to public health even in countries where the virus is not enzootic and represent a possible source of influenza outbreaks in poultry.

In Japan, between the end of December 2003 and March 2004, four outbreaks of acute, highly transmissible and lethal disease occurred in birds in three prefectures separated by 150-450 km, involving three chicken farms and a group of chickens raised as pets. The cause of each outbreak was an H5N1 influenza A virus-the first highly pathogenic virus to be isolated from the outbreaks in Japan since 1925. The H5N1 virus was also isolated from dead crows, apparently infected by contact with virus-contaminated material. These H5N1 viruses were antigenically similar to each other, but could be differentiated from other H5 viruses, including those isolated from Hong Kong in 1997 and 2003, by use of a panel of monoclonal antibodies in hemagglutination inhibition assays. Genetically, the H5N1 viruses in Japan were closely related to each other in all genes and were genetically closely related to a single isolate of genotype V that was isolated in 2003 in the Guandong Province of mainland China (A/chicken/Shantou/4231/2003). The virulence of the index isolate (A/chicken/Yamaguchi/7/2004) was studied in chickens and mice. Chickens intravenously or intranasally inoculated with the isolate died within 1 or 3 days of inoculation, respectively. In mice, although this virus replicated well in the lung without prior adaptation and spread to the brain, the dose lethal to 50% of the mice was 5 x 10(5) 50% egg infectious doses (EID50), which is less pathogenic than the Hong Kong 1997 H5N1 viruses isolated from humans. Our findings indicate that the H5N1 viruses associated with the influenza outbreaks in chickens in Japan were genotypically closely related to an H5N1 virus isolated from chicken in China in 2003 (genotype V), but were different from those prevalent in southeastern Asia in 2003-2004 (i.e., genotype Z) and that these highly pathogenic viruses can be transmitted to crows, which are highly susceptible to these viruses.

To evaluate the protective effect of epidermal powder immunization (EPI) against equine herpesvirus-1 (EHV-1) infection, we prepared a powder vaccine in which formalin-inactivated virions were embedded in water-soluble, sugar-based particles. A PowderJect device was used to immunize mice with the powder vaccine via their abdominal skin. We found that twice-immunized mice were protected against challenge with the wild-type virus. This protective effect was equivalent to or better than that observed in mice immunized with other types of vaccines, including a gene gun-mediated DNA vaccine containing the glycoprotein D (gD) gene or conventional inactivated virus vaccines introduced via intramuscular or intranasal injections. These findings indicate that the powder vaccine is a promising approach for the immunological control of EHV-1 infection, either alone or as a part of prime-boost vaccination strategies.

Many viral membrane glycoproteins are post-translationally processed by intracellular endoproteases such as subtilisin-like proteases. These proteases recognize a cleavage site sequence comprising basic amino acids positioned upstream of the cleavage site of the viral proteins. Here, we mutated the glycine residue immediately downstream of the cleavage site (P1) of hemagglutinin (HA) from a virulent avian influenza virus, A/turkey/Ontario/7732/66 (H5N9) (R-R-R-K-K-R/G), to examine the effect of this mutation on its clevability. Substitution of Gly with Ile, Leu, Val, or Pro, but not Ala, Asp, Phe, His, Ser, or Thr, resulted in substantial reduction of HA cleavage by endogenous endoproteases in CV-1 cells and by vaccinia-expressed PC6 and, albeit to a lesser extent, furin. We conclude that HA cleavage by subtilisin-like proteases is influenced by the downstream P1 amino acid in the absence of upstream cleavage site sequence alterations.

Avian influenza A viruses of the H5 and H7 subtypes periodically cause severe outbreaks of disease in poultry. The question we wished to address in this study is whether these highly pathogenic strains constitute unique lineages or whether they and related nonpathogenic viruses are derived from common ancestors in the wild bird reservoir. We therefore compared the nucleotide and amino acid sequences of the hemagglutinin (HA) genes of 15 H5 and 26 H7 influenza A viruses isolated over 91 years from a variety of host species in Eurasia, Africa, Australia, and North America. Phylogenetic analysis indicated that the HA genes of H5 and H7 viruses that cause severe disease in domestic birds do not form unique lineages but share common ancestors with nonpathogenic H5 and H7 viruses. These findings predict that highly pathogenic avian H5 and H7 influenza A viruses will continue to emerge from wild bird reservoirs. Another important question is whether H7 influenza viruses found in mammalian species are derived from avian strains. We included eight equine influenza viruses and one seal isolate in the phylogenetic analysis of H7 HA genes. We could show that the HA genes of both, the equine and the seal viruses, shared ancestors with avian H7 HA genes. This indicates that currently circulating H7 viruses with an avian HA gene may have the potential to adapt to mammalian species and to cause an influenza outbreak in the new host.

The emergence of virulent avian influenza viruses in poultry is unpredictable. To gain insight into the mechanism for this event, we sought to identify the molecular changes in virulent mutants that occur during replication in 14-day-old embryonated chicken eggs. After three passages in 14-day-old eggs, avirulent H5 viruses with the K/R-K-K/T-R sequence at the hemagglutinin (HA) cleavage site became virulent in chickens, concomitantly acquiring high HA cleavability, whereas those with the R-E-T-R sequence did not. None of the test viruses converted to a virulent phenotype when passaged in 10-day-old eggs. Nucleotide sequence analysis indicated that the virulent mutants either lost a glycosylation site near the HA cleavage site or acquired an additional arginine at the latter. Avirulent viruses that became virulent after passage in older eggs included an H5N2 avian strain with the R-K-T-R sequence that was isolated in 1993, indicating that viruses with this sequence motif, which are currently circulating in bird populations, should be considered potentially virulent. Failure to generate virulent mutants from viruses with R-E-T-R at the HA cleavage site underscores the pathogenic heterogeneity among avian influenza viruses.

Highly pathogenic influenza A viruses periodically infect both humans and nonhuman animals, including chickens. To gain insight into the origin of influenza outbreaks in poultry, we investigated two H5N2 viruses, A/chicken/Pennsylvania/13609/93 (Ck/PA/93) and A/chicken/Florida/25717/93 (Ck/FLA/93), that had been isolated in live-bird markets in Pennsylvania and Florida during surveillance studies in 1993. Phylogenetic analysis of the HA genes of these isolates, as well as H5N2 viruses isolated from ruddy turnstone surfbirds in 1991 (A/ruddy turnstone/Delaware/244/91 [RT/DE/91]) and other known H5 strains, indicated that Ck/PA/93 and Ck/FLA/93 share a common ancestor with RT/DE/91 and did not originate from A/chicken/Pennsylvania/1370/83 (Ck/PA/1370/83), which devastated chicken populations in 1983-1984. Both isolates were nonpathogenic in chickens by experimental infection and their HA protein (HA0) could not be cleaved into HA1 and HA2 without trypsin. The sequences at the HA cleavage sites of Ck/PA/93 and Ck/FLA/93 (R-K-T-R) appear to be intermediate between those of virulent and avirulent viruses, raising the possibility that a single mutation could promote virulence in chickens. We therefore recommend eradication of such viruses as soon as they appear.

A BamHI, EcoRI, and XhoI restriction endonuclease map of Marek's disease virus (MDV) serotype 2 (MDV2) DNA was constructed by double-digest analyses of 28 cloned BamHI and 11 cloned EcoRI fragments of MDV2 DNA, followed by hybridization tests of these cloned BamHI DNA fragments with electrophoretically separated digests of MDV2-infected cell DNA. On this map, MDV2 genome consisted of two segments which have unique regions inserted between two inverted repeat regions as observed in MDV serotype 1 and 3 genomes. Further, the DNA homology among three serotypes of MDV was examined by hybridization under less stringent conditions using cloned BamHI fragments of MDV2 DNA. Most of the MDV2 fragments located within the unique regions hybridized with MDV serotype 1 and 3 DNAs, indicating the presence of the collinear relationship among three serotypes. In addition, MDV2 DNA fragments which hybridized with the DNA fragments encoding MDV1 gp57-65 (or A antigen) or MDV1 gp100, gp60, gp49 (or B antigen) were identified and these fragments of serotypes 1 and 2 found to be collinear.

By transfection of a recombinant plasmid containing the feline immunodeficiency virus (FIV) long terminal repeat (LTR) linked to the chloramphenicol acetyltransferase (CAT) gene followed by infection of feline herpesvirus type 1 (FHV-1) into Crandell feline kidney cells and Felis catus whole fetus 4 cells, enhancement of CAT activity was demonstrated. Furthermore, individual feline T-lymphocytes were productively co-infected with both FIV and FHV-1 in vitro as determined by two-color immunofluorescence and electron microscopy analyses. These results revealed the transactivation of the FIV LTR by FHV-1 and the dual infection of T-lymphocytes with both viruses. The possibility that FHV-1 might be a cofactor which plays a role in the pathogenesis of FIV infection is discussed.