Differences in the biological phenotype of low-yielding (L) and high-yielding (H) variants of swine influenza virus A/NJ/11/76 are associated with their different receptor-binding activity.

Low- (L) and high-yielding (H) variants of A/sw/NJ/11/76 influenza virus were compared for their growth properties in embryonated chicken eggs and MDCK cells and for their binding affinity for the membrane fractions prepared from cells of the chicken embryo allantoic membrane. MDCK, and swine tracheal cells, as well as for soluble sialic acid containing macromolecules and monovalent sialosides. We have shown, that during infection in MDCK cells and in eggs, the progeny of the L variant remain predominantly cell associated, in contrast to those of H. As a result, accumulation of the L mutant in allantoic or culture fluid is significantly slowed in comparison with the H variant. Visualization of the infectious foci formed by the viruses in MDCK cell monolayers and on the allantoic membrane revealed that L spreads predominantly from cell to cell, while the spread of H involves release of the virus progeny into solution and its rapid distribution over the cell monolayer via convectional flow of the liquid. In the binding assays, L displayed significantly higher binding affinity than H for cellular membranes, gangliosides, and sialylglycoproteins, however, the affinity of the variants for the monovalent sialic acid compounds was comparable. Unlike H. L bound strongly to dextran sulfate. The data obtained suggest that all distinctions of the L and H biological phenotypes reported previously [Kilbourne, E.D., Taylor, A. H. Whitaker, C.W., Sahai, R., and Caton, A (1988) Hemagglutinin polymorphism as the basis for low-and high-yield phenotypes of swine influenza virus. Proc. Natl. Acad. Sci. USA 85, 7782-7785] could be rationally explained by a more avid binding of the L variant to the surface of target cells, and that this effect is mainly due to enhanced electrostatic interactions.

Supplementation of conventional influenza A vaccine with purified viral neuraminidase results in a balanced and broadened immune response.

Influenza virus neuraminidase was chromatographically extracted from A/Johannesburg/33/94 (H3N2) and used to supplement conventional monovalent H3JHN2JH inactivated influenza vaccine. Immunization of mice with this preparation resulted in high titers of antibodies to both hemagglutinin (HA) and neuraminidase (NA) equivalent for each antigen to titers in animals immunized with either antigen alone. Homotypic infection was suppressed and greater reduction in viral replication was observed following heterotypic infectious challenge than was observed following the non-supplemented vaccine. There was no evidence of suppression of the immune response to the HA despite the presence of high amounts of NA in the vaccine. Supplementation of conventional inactivated influenza vaccine with NA takes advantage of the equivalent immunogenicity of dissociated HA and NA, to produce a more balanced immune response to both surface antigens, without the antigenic competition tht occurs after immunization with conventional vaccine or infection. These studies in a mouse model system suggest that supplementation of current inactivated influenza vaccines offers the prospect of improved immunization of humans against influenza.

Perspectives on pandemics: a research agenda.

During the 20th century, indisputable pandemics of influenza occurred in 1918, 1957, and 1968. The pandemics of 1957 (A/H2N2) and 1968 (A/H3N2) were associated with major antigenic changes in the virus, probably reflecting introduction by recombination of animal virus genes. The 1918 epidemic is beyond the reach of modern virology but, based on seroarcheology, appears to have been caused by a virus very similar to present swine (A/H1N1) influenza viruses. Changes in both principal antigens of the A/H1N1 subtype in 1947 resulted in total vaccine failure and pandemic spread of virus. On the basis of three periods of prevalence in the 20th century, A/H1N1 may be the "default" human virus, although the 39-year persistence of A/H3N2 to the present challenges this view. Only H1, H2, and H3 and N1 and N2 antigens have been found in human influenza viruses, but virologic history is too brief to preclude the contribution of other antigens to future pandemics.

In pursuit of influenza: Fort Monmouth to Valhalla (and back).

In reviewing 50 years of personal research on influenza, I have journeyed, literally and figuratively, from an army camp epidemic in Fort Monmouth NJ in 1947 to a (literal and figurative) Valhalla, where I now conduct my research. Having entered the field as a physician, I have always sought practical applications of my work, yet in every instance, such applications have led me to seek further answers in basic research as new questions arose. I entered the area of influenza virus genetics by the back door through an interest in the effects of corticosteroid hormones on viral replication, used the genetic approach in analyzing the morphological variation of the virus and, in so doing, exploited the finding of a linkage of high-yield growth to spherical morphology. Today, all influenza vaccine viruses are high-yield genetic reassortants. Subsequent study of reassortant viruses facilitated the identification and isolation of the two major antigens of the virus in antigenic hybrids and showed their differing functions in the induction of immunity. In turn, a new approach to influenza vaccination has been discovered and is presently under clinical investigation.

Immunization with dissociated neuraminidase, matrix, and nucleoproteins from influenza A virus eliminates cognate help and antigenic competition.

When hemagglutinin (HA) and neuraminidase (NA) are presented together on an intact influenza virus particle, the antigens are competitive, with HA dominant over NA in T- and B-cell priming. Immunization with mixtures of purified HA and NA eliminates antigenic competition between HA and NA, as well as between N1-N2 NA mixtures. Evidence that vaccine preparations contain influenza virus matrix (M1) and nucleoprotein (NP) prompted the investigation of possible competing effects of these proteins on the anti-HA and anti-NA immune response. However, in BALB/c mice immunized with mixtures of purified NA, M1, and NP no antigenic competition was demonstrated in either the primary or the secondary response. When mice were immunized with intact virus or by infection, a lesser antibody response to M1 and NP was observed. Furthermore, as measured by mean pulmonary virus titers after infection, no additional protective effects were conferred on mice immunized with M1 and NP either alone or in conjunction with other antigens. These studies of influenza virus antigen mixtures have implications for vaccination against influenza and other vaccines consisting of combinations of antigens.

Neuraminidase-specific antibody responses to inactivated influenza virus vaccine in young and elderly adults.

Little information is available on the potential role of antibody to influenza virus neuraminidase (NA) in vaccine-induced immunity. In the present study, serologic responses to the N1Texas/91 and N2Beijing/92 NA components of trivalent inactivated influenza virus vaccine were measured by NA inhibition (NI) and enzyme-linked immunosorbent assay (ELISA), and the results for adults aged 18 to 45 (young) or > or = 65 (elderly) years were compared. The two age groups had comparable rates (32 to 50%) of NI response. In contrast, ELISA immunoglobulin G (IgG) antibody responses to N1 and N2 NAs occurred in 70 to 71 and 67 to 83%, respectively, of young subjects but in only 3 to 18 and 18 to 35%, respectively, of elderly subjects. prevaccination mean ELISA IgG and IgA NA antibody titers were generally lower for the young adults than they were for the elderly, whereas the corresponding NI titers were comparable. In young adults, plaque size-reducing NA antibody increases were positively associated with ELISA but not with NI antibody increases. There were no apparent age-related differences in the immunoglobulin isotype distribution of the anti-NA response, with IgG being the dominant class and IgG1 the dominant subclass of serum antibody. Anti-hemagglutinin antibody responses to H1Texas/91 and H3Beijing/92 were greater in magnitude and frequency than the corresponding NA-specific responses to N1Texas/91 and N2Beijing/92 when measured by hemagglutination inhibition and NI, respectively, but not when measured by ELISA. The discordance between NI and ELISA for measurement of NA-specific vaccine responses may reflect the relative insensitivity of NI in discriminating differences when initial antibody titers are low.

The emergence of "emerging diseases": a lesson in holistic epidemiology.

The term "emerging diseases" is a loosely defined category of entities comprising resurgent or recurrent old diseases (usually caused by "new" or mutated previously known agents), diseases truly new to man, but caused by preexisting ("old") zoonotic agents, and syndromes newly defined by the discovery of new agents through advances in biotechnology. Identification and solution of these problems depends, first, on recognition of their differences, and then upon tailoring appropriate strategies for their control. Thus, new influenza viruses appear each year to challenge immunity to their antecedents, but evoke the unchanged and centuries old symptom complex of influenza. Tuberculosis, is resurgent because of mycobacterial mutation to antibiotic resistance, immunosuppression by AIDS, and laxity in public health surveillance. Parvovirus B19 and herpesvirus 6 were revealed as cryptic infectors of white blood cells in studies of hepatitis B and AIDS, but since have been shown to be important causes of childhood rashes, aplastic anemia, and neurologic disease. The encroachment of human habitation on wilderness perimeters (ecosystem change) has increased contact with vectors of zoonotic viruses and bacteria, as evidenced by Lyme disease, Ebola virus infection, and the hemorrhagic fevers. The term "holistic epidemiology" embraces all these problems, from the molecular to the macroenvironmental level. Humans, parasites, and their environment will continue their ancient, fluctuating, dynamic relationship in the future, and new diseases will continue to emerge.

Preparation and characterization of a purified influenza virus neuraminidase vaccine.

Influenza virus neuraminidase (NA) has been shown to induce protective but infection-permissive immunity in experimental animals. Challenge infection following such immunization is attended by decreased viral replication and disease manifestations but is sufficient to provide antigenic stimulation and definitive immunity to the virus. The present report describes the preparation and characterization of a purified NA vaccine (NAV) used in Phase 1 (immunogenicity and toxicity) trials in humans. In essence, virion NA was isolated from detergent-disrupted virus by affinity chromatography on oxamic acid-agarose, treated with formalin and tested for its enzymatic activity and for its immunogenicity in Balb/c mice and New Zealand rabbits. The preparation was essentially free of viral hemagglutinin but contained residual NP and M1 proteins. Both dispersed and aggregated NA tetrameric heads were seen in electron micrographs. Enzymatic activity was preserved, and minimal immunogenic doses in mice and rabbits, respectively, were 3.7 and 0.027 micrograms per kg.

Purified influenza A virus N2 neuraminidase vaccine is immunogenic and non-toxic in humans.

The immunogenicity and toxicity of a purified influenza virus (N2) neuraminidase vaccine (NAV) were investigated in 88 human subjects aged 18-40, and compared to response to a conventional trivalent influenza vaccine, Fluogen (Parke-Davis). NAV doses ranged from 2.6 to 69.9 micrograms and were given intramuscularly. Serologic neuraminidase-inhibiting (NI) and neuraminidase-specific ELISA responses in this N2-primed population were roughly proportional to the dose administered. Maximal response was seen in 14-21 days and NI antibody titers persisted unabated for the 6-month post-vaccination follow-up period. All doses were well tolerated with respect to local and systemic reactions. NI tests performed with the putative (1975) priming N2 antigen demonstrated anamnestic response but did not reveal responses not already shown with the homologous (1992) antigen. Response to this purified, non-adjuvanted preparation encourages continuing investigation of the induction of infection-permissive immunity with influenza virus neuraminidase.

Immunogenicity of influenza A virus N2 neuraminidase produced in insect larvae by baculovirus recombinants.

Influenza A virus neuraminidase (NA) from A/Udorn/72 (H3N2) was expressed by recombinant baculovirus-infected insects. The recombinant NA was enzymatically active. Enzyme activity was neutralized by polyclonal antisera raised against virion-extracted NA. NA produced in whole insects by a baculovirus expression system is antigenically indistinguishable from virion NA by polyclonal antisera in functional assays (NI) and in ELISA, and is highly immunogenic without adjuvant. It is equivalent in immunogenicity to NA purified from influenza virus. Our results indicate that baculovirus-produced NA could be used as a source for large quantities of purified N2-NA for vaccine use.

Nonimmunoselected intrastrain genetic variation detected in pairs of high-yielding influenza A (H3N2) vaccine and parental viruses.

Seven influenza A (H3N2) high-yielding vaccine candidate strains were examined. Antigenic analysis revealed that 5 of the strains could be distinguished antigenically from their corresponding wild type parent viruses. Comparative sequence data for the HA1 domains of the HA (hemagglutinin) genes for these 5 high-yielding viruses and the corresponding wild type parents demonstrated one to three amino acid substitutions within each virus pair, with at least one amino acid change being located in a previously defined antigenic site. Comparison of the HA sequences of the 2 antigenically indistinguishable virus pairs revealed no amino acid differences in 1 and one amino acid change in the other. Examination of 1 additional wild type virus, A/Guangdong/39/89, and its three high-yielding derivatives obtained either by serial egg passage or by reassortment revealed an additive effect of the HA and M genes in creating the high-yielding phenotype.

Immunization with purified N1 and N2 influenza virus neuraminidases demonstrates cross-reactivity without antigenic competition.

Based on the absence of serologic cross-reactivity, the neuraminidases (NAs) of influenza A viruses are divided into antigenically discrete subtypes, analogous to the hemagglutinin (HA) major antigens with which they share the virion surface. An innovative approach to influenza vaccination takes advantage of the infection-permissive nature of immunization with NA as the minor surface antigen. However, evidence that HA dominates immune response when HA and NA are presented together in the intact virion prompted investigation of possible competing effects during immunization of NA subtype mixtures ultimately required for human vaccination. Immunization of BALB/c mice with purified N1- and N2-subtype NAs demonstrated no antigenic competition in primary or secondary response. However, when homotypic or heterotypic infection followed immunization, cross-reactive antibodies between N1 and N2 were found and "reverse antigen competition" occurred with initial NA priming suppressing response to HA following infection with virus containing homologous NA. These studies of antigen mixtures have implications for the use of combined and chimeric vaccines for diseases other than influenza.

Dissociation of influenza virus hemagglutinin and neuraminidase eliminates their intravirionic antigenic competition.

When presented together on the intact influenza virus particle, the external hemagglutinin (HA) and neuraminidase (NA) antigens are competitive, with HA dominant over NA in both T- and B-cell priming (B. E. Johansson, T. M. Moran, and E. D. Kilbourne, Proc. Natl. Acad. Sci. USA 84:6869-6873, 1987). Dissociation and purification of HA and NA from virus and their injection separately or in combination into BALB/c mice eliminates their antigenic competition as measured by antibody response, confirming that it is their structural association that leads to what we have termed intravirionic antigenic competition. We discuss this phenomenon with respect to previously described intermolecular antigenic competition and with regard to its probable mechanism. Our findings are relevant to contemporary interest in viral vaccine vectors and multicomponent vaccines.

Influenza A virus haemagglutinin polymorphism: pleiotropic antigenic variants of A/Shanghai/11/87 (H3N2) virus selected as high yield reassortants.

Genetic reassortment of the A/Shanghai/11/87 (H3N2) variant of influenza A virus with A/PR8/34 (H1N1) virus [the standard donor of high yield (hy) genes for influenza vaccine viruses] resulted in the isolation of two reassortants with differing H3 haemagglutinin (HA) phenotypes, X-99 and X-99a. The two HA phenotypes were derived from individual subpopulations of the H3N2 wild-type virus during the reassortment event. The HA mutants and their respectively derived reassortants (identical in RNA genotype) differed in antigenicity, replication characteristics, yield in chick embryos and haemagglutinin gene sequence. Despite antigenic differences in reactions to polyclonal rabbit antisera of 60%, both X-99 and X-99a, the hy reassortants, were equally immunogenic and protective in BALB/c mice to challenge by parental wild-type virus. Differences in HA phenotype were related to a Ser to Ile change at amino acid position 186. These findings emphasize the polymorphism of influenza virus strains as well as the need for caution in selection of vaccine strains from among antigenically distinct viral subpopulations.

Infection-permissive immunization with influenza virus neuraminidase prevents weight loss in infected mice.

In studies of infection of young Balb/c mice with a mouse virulent strain of X-31 (H3N2) influenza A virus we have shown a profound virus dose-related effect of infection on body weight. Most of this effect is prevented by prior administration of either inactivated whole virus vaccine, which prevents infection, or purified influenza virus neuraminidase, which is infection-permissive, but reduces pulmonary virus replication by 1.5 to 3 orders of magnitude. These studies support the concept of infection-permissive immunization and suggest that levels of virus replication previously shown to be antigenic can be sustained without significant systemic effects.

Influenza vaccine strain selection: equivalence of two antigenically distinct haemagglutinin variants of 1989 H3N2 influenza A virus in protection of mice.

Precise antigenic analysis with haemagglutinin-inhibition (HI) tests of 1989 H3N2 influenza A viruses with polyclonal ferret, rabbit and mouse antisera has shown, first, significant differences among 1989 wild-type isolates, second, antigenic differences between two high-yield vaccine candidate reassortant viruses, third, significant antigenic differences of one reassortant (X-105) from the wild-type virus (A/Guangdong A/39) from which it was derived, and fourth, dependence of antigenic characterization of viruses upon the host species used in immunization. Nevertheless, the two reassortant viruses (only 43% similar by HI test) were equally protective in preventing homovariant or heterovariant infection in either previously unimmunized or infection-primed mice. These results not only confirm the known antigenic heterogeneity of influenza A viruses, but raise questions about the adequacy of current methods of antigenic characterization of influenza viruses and the basis for decisions on vaccine strain selection.

Pandora's box and the history of the respiratory viruses: a case study of serendipity in research.

As fastidious human parasites, the respiratory viruses other than influenza viruses have been among the last of the human viruses to be isolated. Their recognition has been dependent upon the evolving technology of cell culture and equally upon a series of fortuitous observations by astute investigators. Adenoviruses were discovered independently by two different groups of scientists, one utilizing explantation of ostensibly normal human tissues and the other recovering virus directly from epidemics of acute disease. In other studies, a technique developed for detection of influenza virus in cell culture led to the discovery of other hemagglutinating viruses, now known as parainfluenza viruses. From such shreds of laboratory evidence, the structural and molecular characteristics of these diverse viruses have been defined, and each in turn has been retrospectively linked to a legacy of previously described clinical syndromes and epidemic patterns. Thus, neither technology or human imagination alone is sufficient for scientific advance, but, when combined, are the essence of scientific discovery.

Comparison of intranasal and aerosol infection of mice in assessment of immunity to influenza virus infection.

A comparison was made of intranasal and aerosol routes of infection with X-31 influenza A virus in Balb/c mice. Mice were first infected with 100 MID50 by either route then challenged 42 days later with the same virus given by the same or alternative route. Three days following each infection, pulmonary virus was measured by inoculation of chick embryos. Mice initially infected under ether anesthesia by intranasal inoculation experienced higher initial mortality but proved most resistant to subsequent challenge by either method. In contrast, mice first infected by aerosol were least resistant to intranasal challenge, as indicated by increased rate of infection and pulmonary virus titers, but, like mice previously infected intranasally, were not reinfected by the aerosol route. Thus, intranasal infection appears to be more effective both in inducing and challenging immunity from infection. These results should be considered in the design of experiments utilizing influenza virus infection of mice as a model system.