Strategies for inducing protection against avian influenza A virus subtypes with DNA vaccines.

The cross-species transfer of a H5N1 influenza virus from birds to humans, and the systemic spread of this virus in mice, has accelerated the efforts to devise protective strategies against lethal influenza viruses. DNA vaccination with the highly conserved nucleoprotein gene appears to provide cross protection against influenza A viruses in murine models. Whether such vaccines would protect human hosts against different influenza A viruses, including strains with pandemic potential, is unclear. Our aim in this study is to evaluate the ability of a combination DNA vaccine consisting of two plasmids encoding the HA genes from two different subtypes and a DNA vaccine encoding the viral nucleoprotein gene from a H5 virus to induce protection against highly lethal infection caused by H5 and H7 influenza viruses in chickens. Chickens given a single dose of plasmids expressing H5 and H7 hemagglutinins protected the birds from infection by either subtype. However, birds immunized with nucleoprotein DNA and challenged with either A/Ck/Vic/1/85(H7N7) or A/Ty/Ir/1/83 (H5N8) showed definite signs of infection, suggesting inadequate immunity against viral infection. Fifty percent of the nucleoprotein DNA immunized birds survived infection by influenza A/Ty/Ir/1/83 (H5N8) virus (virus of same subtype) while 42% survived infection by influenza A/Ck/Vic/1/85/(H7N7) virus (virus of a different subtype). These studies demonstrate that immunization with DNA encoding a type-specific gene may not be effective against either homologous or heterologous strains of virus, particularly if the challenge virus causes a highly lethal infection. However, the combination of HA subtype vaccines are effective against lethal infection caused by viruses expressing any of the HA subtypes used in the combination preparation.

Growth and immunogenicity of influenza viruses cultivated in Vero or MDCK cells and in embryonated chicken eggs.

Vero cells, MDCK cells and embryonated chicken eggs (eggs) were used to evaluate influenza virus growth characteristics and immunogenicity induced by inactivated influenza B vaccines. Both cell lines produced comparable quantities of total viral and haemagglutinin (HA) proteins. Sequence analysis indicated genetic identity of the HA of Vero- and MDCK-grown virus counterparts with maintenance of antigenic characteristics of viruses derived from humans. The egg-grown influenza B/Memphis/1/93 variant differed from cell-grown counterparts at amino acid position 198 (Pro-Thr) and lost a glycosylation site. The level of neuraminidase (NA) activity was the highest in egg-grown virus, while MDCK- and Vero cell-grown viruses possessed 70% and 90% less NA activity respectively when fetuin was used as a substrate. Although each of the vaccines induced high and comparable levels of serum antibodies, mammalian cell-derived vaccines induced antibodies that were more cross reactive, and those antibodies induced by egg-derived vaccine were more specific to the homologous antigen. ELISPOT analysis indicated that the mammalian cell-grown vaccines induced high frequencies of IgG-producing cells directed against both cell- and egg-grown antigens, while egg-grown vaccine induced high frequencies of IgG and IgM-producing cells reacting with homologous antigen and low levels of IgG-producing cells reactive with cell-grown virus antigen. Taken together, our results suggest that mammalian cells are a viable option for the production of influenza virus vaccines.

Amino acid residues contributing to the substrate specificity of the influenza A virus neuraminidase.

Influenza A viruses possess two glycoprotein spikes on the virion surface: hemagglutinin (HA), which binds to oligosaccharides containing terminal sialic acid, and neuraminidase (NA), which removes terminal sialic acid from oligosaccharides. Hence, the interplay between these receptor-binding and receptor-destroying functions assumes major importance in viral replication. In contrast to the well-characterized role of HA in host range restriction of influenza viruses, there is only limited information on the role of NA substrate specificity in viral replication among different animal species. We therefore investigated the substrate specificities of NA for linkages between N-acetyl sialic acid and galactose (NeuAcalpha2-3Gal and NeuAcalpha2-6Gal) and for different molecular species of sialic acids (N-acetyl and N-glycolyl sialic acids) in influenza A viruses isolated from human, avian, and pig hosts. Substrate specificity assays showed that all viruses had similar specificities for NeuAcalpha2-3Gal, while the activities for NeuAcalpha2-6Gal ranged from marginal, as represented by avian and early N2 human viruses, to high (although only one-third the activity for NeuAcalpha2-3Gal), as represented by swine and more recent N2 human viruses. Using site-specific mutagenesis, we identified in the earliest human virus with a detectable increase in NeuAcalpha2-6Gal specificity a change at position 275 (from isoleucine to valine) that enhanced the specificity for this substrate. Valine at position 275 was maintained in all later human viruses as well as swine viruses. A similar examination of N-glycolylneuraminic acid (NeuGc) specificity showed that avian viruses and most human viruses had low to moderate activity for this substrate, with the exception of most human viruses isolated between 1967 and 1969, whose NeuGc specificity was as high as that of swine viruses. The amino acid at position 431 was found to determine the level of NeuGc specificity of NA: lysine conferred high NeuGc specificity, while proline, glutamine, and glutamic acid were associated with lower NeuGc specificity. Both residues 275 and 431 lie close to the enzymatic active site but are not directly involved in the reaction mechanism. This finding suggests that the adaptation of NA to different substrates occurs by a mechanism of amino acid substitutions that subtly alter the conformation of NA in and around the active site to facilitate the binding of different species of sialic acid.

DNA vaccine encoding hemagglutinin provides protective immunity against H5N1 influenza virus infection in mice.

In Hong Kong in 1997, a highly lethal H5N1 avian influenza virus was apparently transmitted directly from chickens to humans with no intermediate mammalian host and caused 18 confirmed infections and six deaths. Strategies must be developed to deal with this virus if it should reappear, and prospective vaccines must be developed to anticipate a future pandemic. We have determined that unadapted H5N1 viruses are pathogenic in mice, which provides a well-defined mammalian system for immunological studies of lethal avian influenza virus infection. We report that a DNA vaccine encoding hemagglutinin from the index human influenza isolate A/HK/156/97 provides immunity against H5N1 infection of mice. This immunity was induced against both the homologous A/HK/156/97 (H5N1) virus, which has no glycosylation site at residue 154, and chicken isolate A/Ck/HK/258/97 (H5N1), which does have a glycosylation site at residue 154. The mouse model system should allow rapid evaluation of the vaccine's protective efficacy in a mammalian host. In our previous study using an avian model, DNA encoding hemagglutinin conferred protection against challenge with antigenic variants that differed from the primary antigen by 11 to 13% in the HA1 region. However, in our current study we found that a DNA vaccine encoding the hemagglutinin from A/Ty/Ir/1/83 (H5N8), which differs from A/HK/156/97 (H5N1) by 12% in HA1, prevented death but not H5N1 infection in mice. Therefore, a DNA vaccine made with a heterologous H5 strain did not prevent infection by H5N1 avian influenza viruses in mice but was useful in preventing death.

Immunogenicity and protective efficacy in mice of influenza B virus vaccines grown in mammalian cells or embryonated chicken eggs.

The immunogenicity and protective efficacy of formalin-inactivated influenza B/Memphis/1/93 virus vaccines propagated exclusively in Vero cells, MDCK cells, or embryonated chicken eggs (hereafter referred to as eggs) were investigated. Mammalian cell-grown viruses differ from the egg-grown variant at amino acid position 198 (Pro/Thr) in the hemagglutinin gene. The level of neuraminidase activity was highest in egg-grown virus, while MDCK and Vero cell-derived viruses possessed 70 and 90% less activity, respectively. After boosting, each of the vaccines induced high levels of hemagglutinin-inhibiting, neuraminidase-inhibiting, and neutralizing antibodies that provided complete protection from MDCK-grown virus challenge. Mammalian cell-derived virus vaccines induced serum antibodies that were more cross-reactive, while those induced by egg-grown virus vaccines were more specific to the homologous antigen. Enzyme-linked immunospot analysis indicated that cell-grown virus vaccines induced high frequencies of immunoglobulin G (IgG)-producing cells directed against both cell- and egg-grown virus antigens, whereas egg-grown virus vaccine induced higher frequencies of IgG- and IgM-producing cells reacting with homologous antigen and low levels of IgG-producing cells reactive with cell-grown viruses. These studies indicate that influenza B virus variants selected in different host systems can elicit different immune responses, but these alterations had no detectable influence on the protective efficacy of the vaccines with the immunization protocol used in this study.

Characterization of avian H5N1 influenza viruses from poultry in Hong Kong.

The transmission of avian H5N1 influenza viruses to 18 humans in Hong Kong in 1997 with six deaths established that avian influenza viruses can transmit to and cause lethal infection in humans. This report characterizes the antigenic and biological properties of the H5N1 influenza viruses isolated from chickens, ducks, and geese from farms and poultry markets in Hong Kong during 1997 and compares them with those of virus isolated from the index human case. Each of the H5N1 viruses from Hong Kong poultry markets that were tested were lethal in chickens, possessed polybasic amino acids at the carboxy-terminus of HA1, and by definition were highly pathogenic in poultry. The available nonpathogenic H5 influenza viruses and the pathogenic H5N1 virus from Hong Kong were analyzed with monoclonal antibodies prepared to A/chicken/Pennsylvania/1370/83 (H5N2). The analysis revealed limited antigenic drift in 15 years and established that monoclonal antibodies are useful reagents for identification and antigenic analysis of avian strains that may transmit to humans in the future. One of the monoclonal antibodies permitted separation of the H5N1 influenza viruses from poultry into two groups that correlated with the presence or absence of a carbohydrate at residue 158 adjacent to the receptor binding site on HA. The H5N1 viruses examined replicated in geese, pigs, rats, and mice, but to only a very limited extent in ducks. It is noteworthy that all infected geese shed virus and that the H5N1 viruses caused disease signs and death in a portion (3 of 16) of the geese, with evidence of systemic spread to the brain. The tropism for geese is unusual and may provide insight into the origin of these viruses. In mice, the H5N1 virus caused lethal pneumonia and spread systemically to the brain. Mice would thus provide an ideal model system for studying immune responses and pathogenesis. Transmission experiments in chickens revealed that the H5N1 viruses are spread by fecal-oral transmission rather than by aerosol, and that the viruses are inactivated by drying of feces at ambient temperature. However, infectivity is maintained for at least 4 days in wet feces at 25 degreesC. There were differences in the morphology of the H5N1 viruses isolated from birds and humans. The perpetuation of H5N1 influenza viruses in the poultry markets in Hong Kong and the transmission of these viruses to humans emphasize the importance of these markets in the epidemiology of influenza. The poultry markets are of critical importance in the perpetuation and transmission of influenza viruses to other avian species and to mammals, including humans.

Cross-protection among lethal H5N2 influenza viruses induced by DNA vaccine to the hemagglutinin.

Inoculation of mice with hemagglutinin (HA)-expressing DNA affords reliable protection against lethal influenza virus infection, while in chickens the same strategy has yielded variable results. Here we show that gene gun delivery of DNA encoding an H5 HA protein confers complete immune protection to chickens challenged with lethal H5 viruses. In tests of the influence of promoter selection on vaccine efficacy, close correlations were obtained between immune responses and the dose of DNA administered, whether a cytomegalovirus (CMV) immediate-early promoter or a chicken beta-actin promoter was used. Perhaps most important, the HA-DNA vaccine conferred 95% cross-protection against challenge with lethal antigenic variants that differed from the primary antigen by 11 to 13% (HA1 amino acid sequence homology). Overall, the high levels of protection seen with gene gun delivery of HA-DNA were as good as, if not better than, those achieved with a conventional whole-virus vaccine, with fewer instances of morbidity and death. The absence of detectable antibody titers after primary immunization, together with the rapid appearance of high titers immediately after challenge, implicates efficient B-cell priming as the principal mechanism of DNA-mediated immune protection. Our results suggest that the efficacy of HA-DNA influenza virus vaccine in mice extends to chickens and probably to other avian species as well. Indeed, the H5 preparation we describe offers an attractive means to protect the domestic poultry industry in the United States from lethal H5N2 viruses, which continue to circulate in Mexico.

Selection of a single amino acid substitution in the hemagglutinin molecule by chicken eggs can render influenza A virus (H3) candidate vaccine ineffective.

This study investigated whether a single amino acid change in the hemagglutinin (HA) molecule influenced the efficacy of formalin-inactivated influenza A (H3N1) vaccine candidates derived from high-growth reassortants between the standard donor of high-yield genes (A/PR/8/34 [H1N1]) and host cell variants generated from the same clinical isolate (A/Memphis/7/90 [H3N2]) by passage in embryonated chicken eggs. Two clones of the isolate generated by growth in eggs differed from the parent virus (represented by an MDCK cell-grown counterpart) solely by the presence of Lys (instead of Glu) at position 156 or Ile (instead of Ser) at position 186 in the HA1 subunit. The protective efficacy of egg-grown HA Lys-156 and HA Ile-186 reassortant variants was compared with that of the MDCK cell-grown reassortant vaccine. Classically, antibody titers in serum have been used to demonstrate vaccine efficacy. Here, parameters of B-cell responsiveness were monitored, including the kinetics, character, and localization of the primary antibody-forming cell (AFC) response and the development of B-cell memory in lymphoid tissues associated with the priming site (spleen) and responsive to pulmonary challenge with infectious virus (upper and lower respiratory tract lymph nodes). We show that the egg-grown HA Lys-156 variant induced an AFC profile vastly different from that elicited by the other two reassortant vaccines. The vaccine was poorly immunogenic; it induced antibodies that were cross-reactive prior to challenge but which, postchallenge with a lethal dose of the MDCK cell-grown reassortant virus, were targeted primarily to the HA Lys-156 variant, were of the immunoglobulin M isotype, were nonprotective, and were derived from the spleen. In contrast, the egg-grown HA Ile-186 variant was remarkably like the MDCK cell-grown virus in that protective immunoglobulin G antibodies were unaffected by the Ile-186 substitution but poorly recognized HA with Lys-156. Furthermore, memory AFC responsiveness was localized to regional lymphoid tissue in the upper respiratory tract, where challenge HA was found. Thus, it is recommended that in the selection of vaccine candidates, virus populations with the egg-adapted HA Lys-156 substitution be eliminated and that, instead, egg-grown isolates which minimally contain Ile-186 be used as logical alternatives to MDCK cell-grown viruses.

A type-specific avian influenza virus subunit vaccine for turkeys: induction of protective immunity to challenge infection.

The fraction NP/HA (nucleoprotein/haemagglutinin) obtained from n-octyl-beta-D-glucopyranoside-treated influenza A H5N2 virus was highly enriched for NP with residual haemagglutinin. This preparation was incorporated in ISCOMs. This potent 'immunostimulating complex' induced the production of high antibody titres in turkeys. The NP/HA ISCOMs preparation was found to protect turkeys from both homologous and heterologous challenge infection as shown by reduced viral titres in the lung and trachea of vaccinated turkeys. Clearance of the virus from trachea and lungs was seen at late stages of infection. The vaccine also induced a cellular immune response as measured by T-cell proliferation and a delayed-type hypersensitivity response. The results reported in this study demonstrate that the NP/HA ISCOM vaccine is capable of inducing type-specific immunity and that it has potential utility as a vaccine in turkeys.

Effect of avian influenza virus infection on the phagocytic function of systemic phagocytes and pulmonary macrophages of turkeys.

The effects of avian influenza virus (AIV) infection on systemic phagocytes and pulmonary macrophages of turkeys were studied. There was a significant increase (P < 0.0001) in oxidative burst in systemic phagocytes of AIV-inoculated turkeys on 2, 4, 6, and 8 days postinoculation (PI), as measured by chemiluminescence. There was also a significant increase (P < 0.02) in oxidative burst in pulmonary macrophages on day 4 PI. The chemiluminescence response was depressed on 6, 8, and 10 days PI in AIV-inoculated turkeys compared with controls. The increase in oxidative response in both systemic phagocytes and pulmonary macrophages correlated with the peak virus titer in the lungs and trachea of AIV-inoculated inoculated turkeys. Bacterial killing by pulmonary macrophages from AIV-inoculated turkeys was reduced on days 6 and 10 PI compared with uninoculated controls. Histopathological changes in trachea were more pronounced on day 6 PI in AIV-inoculated turkeys; no significant changes were detected in the lungs. These data indicate that compromised functional capacity of pulmonary macrophages predisposes turkeys to secondary bacterial infections.

Antigen-capture enzyme immunoassay for detection of avian influenza virus in turkeys.

A double-antibody sandwich ELISA (DAS-ELISA) was developed for detection of avian influenza virus (AIV) antigen. A monoclonal antibody to the viral nucleoprotein (NP) was used to coat the ELISA plates. A direct DAS-ELISA and an indirect DAS-ELISA were evaluated. In the direct DAS-ELISA, monoclonal antibody to the AIV NP conjugated with horseradish peroxidase was used. The direct DAS-ELISA was evaluated for its sensitivity to detect purified NP; this procedure detected as little as 0.1 ng. In the indirect DAS-ELISA, rabbit NP antibody and horseradish peroxidase-conjugated goat anti-rabbit immunoglobin were used as primary and secondary antibodies, respectively. The indirect DAS-ELISA was evaluated for its ability to detect the AIV antigen in tracheal and cloacal specimens from turkeys inoculated with AIV. Results of indirect DAS-ELISA were compared with those of conventional virus isolation. Percentage agreement between indirect DAS-ELISA and virus isolation in AIV-positive samples was found to be 76.1% and, in AIV-negative samples, it was found to be 82.1%. These results indicate that the DAS-ELISA might be a viable alternative to virus isolation because of its rapidity, compared with virus isolation.