Search

To provide information on the mechanism of perpetuation of influenza viruses among waterfowl reservoirs in nature, virological surveillance was carried out in Alaska during their breeding season in summer from 1991 to 1994. Influenza viruses were isolated mainly from fecal samples of dabbling ducks in their nesting places in central Alaska. The numbers of subtypes of 108 influenza virus isolates were 1 H2N3, 37 H3N8, 55 H4N6, 1 H7N3, 1 H8N2, 1 H10N2, 11 H10N7, and H10N9. Influenza viruses were also isolated from water samples of the lakes where they nest. Even in September of 1994 when the most ducks had left for migration to south, viruses were still isolated from the lake water. Phylogenetic analysis of the NP genes of the representative isolates showed that they belong to the North American lineage of avian influenza viruses, suggesting that the majority of the waterfowls breeding in central Alaska migrate to North America and not to Asia. The present results support the notion that influenza viruses have been maintained in waterfowl population by water-borne transmission and revealed the mechanism of year-by-year perpetuation of the viruses in the lakes where they breed.

Summary A severe influenza outbreak occurred in horses in South Africa in 1986. The causative agent was identified as an influenza virus [A/Equine/Johannesburg/86 (H3N8)]. Antigenic analyses of the hemagglutinin (HA) with ferret antisera and monoclonal antibodies showed that the Eq/Johannesburg/86 virus is similar to recent equine H3 viruses. The nucleotide sequence analysis on the HA genes of Eq/Johannesburg/86 and other equine H3 influenza viruses, together with the epidemiological data, clearly demonstrated that the Eq/Johannesburg/86 virus was derived from a virus that had been circulating in horses in the United States in 1986–87. The epidemiological information suggests that the unusually severe influenza outbreak in South Africa may be due to the lack of immunity to these viruses in the horse population.

Summary The sites of replication of influenza A viruses in ferrets and pigs were studied. The majority of the swine, equine, and avian influenza A viruses tested were recovered from the intestinal tract of ferrets as well as from the respiratory tract; most of the human influenza viruses studied were recovered only from the respiratory tract. In contrast with ferrets, only Hong Kong/1/68 (H 3 N 2) influenza virus was recovered from the intestinal tract of pigs. Despite the large biological variability found in ferrets and in pigs, the results do establish that the majority of influenza viruses have the potential to replicate in the intestinal tissues of some mammals. Additionally, the study suggests that there are differences among the influenza A viruses in tissue tropism in different mammals. Both viral and host genetic factors determine the tissue tropism of influenza viruses in mammals.

Summary Swine are an animal reservoir for influenza viruses capable of causing disease in humans. A serological survey in 1988–1989 demonstrates that subtype H1 influenza viruses continue to circulate at high frequency among swine in the north-central U.S.A. (average 51% incidence). Subtype H3 viruses antigenically similar to current human H3 viruses are circulating at low frequency (average 1.1%), particularly in the southeast U.S.A.

Human influenza viruses are more efficiently isolated by inoculating patient samples into the amniotic rather than the allantoic cavity of embryonated chicken eggs. This type of cultivation selects virus variants with mutations around the hemagglutinin (HA) receptor binding site. To understand the molecular basis of these phenomena, we investigated the abundances of sialic acid (SA) linked to galactose (Gal) by the alpha-2,3 linkage (SA alpha2,3Gal) and SA alpha2,6Gal in egg amniotic and allantoic cells and in Madin-Darby canine kidney (MDCK) cells. Using SA-Gal linkage-specific lectins (Maackia amurensis agglutinin specific for SA alpha2,6Gal and Sambucus nigra agglutinin specific for SA alpha2,3Gal), we found SA alpha2,3Gal in both allantoic and amniotic cells and SA alpha2,6Gal in only the amniotic cells. MDCK cells contained both linkages. To investigate how this difference in abundances of SA alpha2,3Gal and SA alpha2,6Gal in allantoic and amniotic cells affects the appearance of host cell variants in eggs, we determined the receptor specificities and HA amino acid sequences of two different patient viruses which were isolated and passaged in the amnion or in the allantois and which were compared with MDCK cell-grown viruses. We found that the viruses maintained high SA alpha2,6Gal specificities when grown in MDCK cells or following up to two amniotic passages; however, further passages in either the amnion or allantois resulted in the acquisition of, or a complete shift to, SA alpha2,3Gal specificity, depending on the virus strain. This change in receptor specificity was accompanied by the appearance of variants in the population with Leu-to-Gln mutations at position 226 in their HA. These findings suggest that lack of SA alpha2,6Gal linkages in the allantois of chicken eggs is a selective pressure for the appearance of host cell variants with altered receptor specificities and amino acid changes at position 226.

Among the proprotein-processing subtilisin-related endoproteases, furin has been a leading candidate for the enzyme that activates the hemagglutinin (HA) of virulent avian influenza viruses. In the present study, we examined the cleavage activity of two other recently isolated ubiquitous subtilisin-related proteases, PACE4 and PC6, using wild-type HA of A/turkey/Ireland/1378/83 (H5N8) and a series of its mutant HAs. Vaccinia virus-expressed wild-type HA was not cleaved in human colon adenocarcinoma LoVo cells, which lack active furin. This processing defect was corrected by the expression of furin and PC6 but not of PACE4 and a control wild-type vaccinia virus. PC6 showed a sequence specificity similar to that with the endogenous proteases in cultured cells. When LoVo cells were infected with a virulent avian virus, A/turkey/Ontario/7732/66 (H5N9), only noninfectious virions were produced because of the lack of HA cleavage. However, when the cells were coinfected with vaccinia virus that expressed either furin or PC6, the avian virus underwent multiple cycles of replication, indicating that both furin and PC6 specifically cleave the virulent virus HA at the authentic site. These data suggest that PC6, as well as furin, can activate virulent avian influenza viruses in vivo, implying the presence of multiple HA cleavage enzymes in animals.

In this study, we investigated the role of the conserved neuraminidase (NA) cytoplasmic tail residues in influenza virus replication. Mutants of influenza A virus (A/WSN/33 [H1N1]) with deletions of the NA cytoplasmic tail region were generated by reverse genetics. The resulting viruses, designated NOTAIL, contain only the initiating methionine of the conserved six amino-terminal residues. The mutant viruses grew much less readily and produced smaller plaques than did the wild-type virus. Despite similar levels of NA cell surface expression by the NOTAIL mutants and wild-type virus, incorporation of mutant NA molecules into virions was decreased by 86%. This reduction resulted in less NA activity per virion, leading to the formation of large aggregates of progeny mutant virions on the surface of infected cells. A NOTAIL virus containing an additional mutation (Ser-12 to Pro) in the transmembrane domain incorporated three times more NA molecules into virions than did the NOTAIL parent but approximately half of the amount incorporated by the wild-type virus. However, aggregation of the progeny virions still occurred at the cell surface. All NOTAIL viruses were attenuated in mice. We conclude that the cytoplasmic tail of NA is not absolutely essential for virus replication but exerts important effects on the incorporation of NA into virions and thus on the aggregation and virulence of progeny virus. In addition, the relative abundance of long filamentous particles formed by the NOTAIL mutants, compared with the largely spherical wild-type particles, indicates a role for the NA cytoplasmic tail in virion morphogenesis.

The reverse genetics system has made it possible to modify the influenza virus genome. By this method, we were able to assess influenza virus as a vaccine vector for protecting BALB/c mice against otherwise lethal lymphocytic choriomeningitis virus (LCMV) infection. A single dose of influenza virus [A/WSN/33 (H1N1)] bearing a cytotoxic T-lymphocyte-specific epitope of the LCMV nucleoprotein (residues 116 to 127) in the neuraminidase stalk protected mice against LCMV challenge for at least 4 months. The immunity was mediated by cytotoxic T lymphocytes and was haplotype specific, indicating that the observed protective response was solely a consequence of prior priming with the H-2d LCMV nucleoprotein epitope expressed in the recombinant influenza virus. We also found that as many as 58 amino acids could be inserted into the neuraminidase stalk without loss of viral function. These findings demonstrate the potential of influenza virus as a vaccine vector, with the neuraminidase stalk as a repository for foreign epitopes.

To obtain direct evidence for a relationship between hemagglutinin (HA) cleavability and the virulence of avian influenza A viruses, we generated a series of HA cleavage mutants from a virulent virus, A/turkey/Ontario/7732/66 (H5N9), by reverse genetics. A transfectant virus containing the wild-type HA with R-R-R-K-K-R at the cleavage site, which was readily cleaved by endogenous proteases in chicken embryo fibroblasts (CEF), was highly virulent in intramuscularly or intranasally/orally inoculated chickens. By contrast, a mutant containing the HA with an avirulent-like sequence (R-E-T-R) at the cleavage site, which was not cleaved by the proteases in CEF, was avirulent in chickens, indicating that a genetic alteration confined to the HA cleavage site can affect cleavability and virulence. Mutant viruses with HA cleavage site sequences of T-R-R-K-K-R or T-T-R-K-K-R were as virulent as viruses with the wild-type HA, whereas a mutant with a two-amino-acid deletion but retention of four consecutive basic residues (R-K-K-R) was as avirulent as a virus with the avirulent-type HA. Interestingly, although a mutant containing an HA with R-R-R-K-T-R, which has reduced cleavability in CEF, was as virulent as viruses with high HA cleavability when given intramuscularly, it was less virulent when given intranasally/orally. We conclude that the degree of HA cleavability in CEF predicts the virulence of avian influenza viruses.

To locate antigenic epitopes on the N8 neuraminidase (NA), we generated a panel of 97 monoclonal antibodies (MAbs), 66 of which inhibited NA activity (NI antibodies). Three groups of NI MAbs were identified from their different reactivities with escape mutants. Group 1 antibodies recognized the peptide loop containing residues 344 to 346, which appears to be an immunodominant region on the rim of the enzyme center of the N8 NA. Group 2 antibodies recognized a novel epitope containing residues 150, 199, 367, 399, and 400 (N2 numbering). From the location of these residues on the three-dimensional structure of the N8 NA, the epitope appears to be located at the interface of two adjacent monomers in the tetrameric NA, one contributing residues 150 and 199 and the other contributing residues 367 and 399 to 400. The available evidence indicates that the MAbs of this group react with the NA only after it is fully assembled. The third group of antibodies recognized the peptide loops containing residues 367 and 399 to 400. All of the amino acid substitutions in N8 escape mutants which affect the NI activity of antibodies were located in the peptide loops known to form epitopes in the N2 and N9 subtypes, indicating that antigenic regions in the NA head inducing NI antibodies appear to be similar among different subtypes of influenza A viruses. The MAbs used in this study will be valuable in studying the role of each N8 NA epitope in host immune defense systems and in the kinetics analysis of the biosynthesis of the enzyme.

Cleavage of the hemagglutinin (HA) in tissue culture systems has been correlated with virulence of avian influenza viruses. To examine the structural requirements for cleavage of the HA, the HA gene from a virulent H5 influenza virus was expressed in mammalian cells (CV-1), and the cleavage site of the HA was explored by using site-specific mutagenesis. The expressed HA protein exhibited normal cleavage, transport to the cell membrane, and ability to adsorb and to fuse erythrocytes at pH 5. Site-specific mutagenesis of the HA directly established that (i) most of the basic amino acids at this site are critical for cleavage activation; (ii) besides the connecting peptide sequence, at least one other structural feature of the HA is required for enzyme recognition; and (iii) the length of the connecting peptide can abrogate the structural feature(s).

Central nervous system lesions of chickens inoculated with three highly pathogenic avian influenza virus strains, A/chicken/Victoria/1/85 (H7N7), A/turkey/England/50-92/91 (H5N1), and A/tern/South Africa/61 (H5N3), were examined histologically and immunohistochemically. The chickens either died within 7 days of inoculation or were killed 2 weeks after inoculation. No significant differences were observed in the lesions induced by these three viruses. The lesions were divided into two types, disseminated foci of microgliosis and necrosis, and ventriculitis. The former lesions were associated with infection of the vascular endothelium and dissemination of the virus to the peripheral parenchymal cells of the chickens that died within 3 days of inoculation. The ventriculitis lesions, however, were observed mainly in the chickens that died between 4 and 7 days after inoculation. These findings suggest that viral infection of the vascular endothelium and subsequent involvement of ependymal cells play important roles in the pathogenesis of the central nervous system lesions.

Influenza virus nucleoprotein (NP) shuttles between the nucleus and the cytoplasm. A nuclear localization signal (NLS) has been identified in NP at amino acids 327 to 345 (J. Davey et al., Cell 40:667-675, 1985). However, some NP mutants that lack this region still localize to the nucleus, suggesting an additional NLS in NP. We therefore investigated the nucleocytoplasmic transport of NP from influenza virus A/WSN/33 (H1N1). NP deletion constructs lacking the 38 N-terminal amino acids, as well as those lacking the 38 N-terminal amino acids and the previously identified NLS, localized to both the cytoplasm and the nucleus. Nuclear localization of a protein containing amino acids 1 to 38 of NP fused to LacZ proved that these 38 amino acids function as an NLS. Within this region, we identified two basic amino acids, Lys7 and Arg8, that are crucial for NP nuclear import. After being imported into the nucleus, the wild-type NP and the NP-LacZ fusion construct containing amino acids 1 to 38 of NP were both transported back to the cytoplasm, where they accumulated. These data indicate that NP has intrinsic structural features that allow nuclear import, nuclear export, and cytoplasmic accumulation in the absence of any other viral proteins. Further, the information required for nuclear import and export is located in the 38 N-terminal amino acids of NP, although other NP nuclear export signals may exist. Treatment of cells with a protein kinase C inhibitor increased the amounts of nuclear NP, whereas treatment of cells with a phosphorylation stimulator increased the amounts of cytoplasmic NP. These findings suggest a role of phosphorylation in nucleocytoplasmic transport of NP.

As the initial step in generating a live attenuated influenza A vaccine, we attempted to substitute an unrelated amino acid sequence (FLAG) for a portion of the neuraminidase (NA) molecule in influenza virus A/WSN/33 (H1N1), using a recently developed technique (reverse genetics [W. Luytjes, M. Krystal, M. Enami, J. D. Parvin, and P. Palese, Cell 59:1107-1113, 1989]). This technique allowed us to rescue the NA molecules containing the FLAG sequence (Asp-Tyr-Lys-Asp-Asp-Asp-Asp-Lys) at the bottom portion of the boxlike head of the molecule immediately above the stalk region (amino acid residues 63 to 70 [WSN NA numbering]). An anti-FLAG monoclonal antibody immunoprecipitated the NA molecules with the FLAG sequence, demonstrating that the foreign epitope was exposed on the virion surface. The dose of FLAG-containing transfectant virus required to kill 50% of mice was 100-fold higher than the required dose of parent virus. The FLAG sequence was stably maintained in the NA molecule during passage of the virus in tissue culture and in mice. These findings demonstrate that live influenza A vaccine strains with stable attenuating mutations in the coding region of the viral genes can be generated by reverse genetics.

In this review we examine the hypothesis that aquatic birds are the primordial source of all influenza viruses in other species and study the ecological features that permit the perpetuation of influenza viruses in aquatic avian species. Phylogenetic analysis of the nucleotide sequence of influenza A virus RNA segments coding for the spike proteins (HA, NA, and M2) and the internal proteins (PB2, PB1, PA, NP, M, and NS) from a wide range of hosts, geographical regions, and influenza A virus subtypes support the following conclusions. (i) Two partly overlapping reservoirs of influenza A viruses exist in migrating waterfowl and shorebirds throughout the world. These species harbor influenza viruses of all the known HA and NA subtypes. (ii) Influenza viruses have evolved into a number of host-specific lineages that are exemplified by the NP gene and include equine Prague/56, recent equine strains, classical swine and human strains, H13 gull strains, and all other avian strains. Other genes show similar patterns, but with extensive evidence of genetic reassortment. Geographical as well as host-specific lineages are evident. (iii) All of the influenza A viruses of mammalian sources originated from the avian gene pool, and it is possible that influenza B viruses also arose from the same source. (iv) The different virus lineages are predominantly host specific, but there are periodic exchanges of influenza virus genes or whole viruses between species, giving rise to pandemics of disease in humans, lower animals, and birds. (v) The influenza viruses currently circulating in humans and pigs in North America originated by transmission of all genes from the avian reservoir prior to the 1918 Spanish influenza pandemic; some of the genes have subsequently been replaced by others from the influenza gene pool in birds. (vi) The influenza virus gene pool in aquatic birds of the world is probably perpetuated by low-level transmission within that species throughout the year. (vii) There is evidence that most new human pandemic strains and variants have originated in southern China. (viii) There is speculation that pigs may serve as the intermediate host in genetic exchange between influenza viruses in avian and humans, but experimental evidence is lacking. (ix) Once the ecological properties of influenza viruses are understood, it may be possible to interdict the introduction of new influenza viruses into humans.

To investigate the biologic importance of the neuraminidase (NA) stalk of influenza A virus, we generated mutant viruses of A/WSN/33 (H1N1) with stalks of various lengths (0 to 52 amino acids), by using the recently developed reverse genetics system. These mutant viruses, including one that lacked the entire stalk, replicated in tissue culture to the level of the parent virus, whose NA stalk contains 24 amino acid residues. In eggs, however, the length of the stalk was correlated with the efficiency of virus replication: the longer the stalk, the better the replication. This finding indicates that the length of the NA stalk affects the host range of influenza A viruses. The NA stalkless mutant was highly attenuated in mice; none of the animals died even after intranasal inoculation of 10(6) PFU of the virus (the dose of the parent virus required to kill 50% of mice was 10(2.5) PFU). Moreover, the stalkless mutant replicated only in the respiratory organs, whereas the parent virus caused systemic infection in mice. Thus, attenuation of the virus with the deletion of the entire NA stalk raises the possibility of its use as live vaccines.

Equine H7N7 influenza A viruses, representing a broad range of isolates, were lethal in mice without adaptation. After repeated passages, A/Equine/London/1416/73 acquired neurotropism upon intranasal infection. Thus, mice infected with equine influenza A viruses provide a model system for the study of highly virulent mammalian influenza viruses.

A phylogenetic analysis of 52 published and 37 new nucleoprotein (NP) gene sequences addressed the evolution and origin of human and swine influenza A viruses. H1N1 human and classical swine viruses (i.e., those related to Swine/Iowa/15/30) share a single common ancestor, which was estimated to have occurred in 1912 to 1913. From this common ancestor, human and classical swine virus NP genes have evolved at similar rates that are higher than in avian virus NP genes (3.31 to 3.41 versus 1.90 nucleotide changes per year). At the protein level, human virus NPs have evolved twice as fast as classical swine virus NPs (0.66 versus 0.34 amino acid change per year). Despite evidence of frequent interspecies transmission of human and classical swine viruses, our analysis indicates that these viruses have evolved independently since well before the first isolates in the early 1930s. Although our analysis cannot reveal the original host, the ancestor virus was avianlike, showing only five amino acid differences from the root of the avian virus NP lineage. The common pattern of relationship and origin for the NP and other genes of H1N1 human and classical swine viruses suggests that the common ancestor was an avian virus and not a reassortant derived from previous human or swine influenza A viruses. The new avianlike H1N1 swine viruses in Europe may provide a model for the evolution of newly introduced avian viruses into the swine host reservoir. The NPs of these viruses are evolving more rapidly than those of human or classical swine viruses (4.50 nucleotide changes and 0.74 amino acid change per year), and when these rates are applied to pre-1930s human and classical swine virus NPs, the predicted date of a common ancestor is 1918 rather than 1912 to 1913. Thus, our NP phylogeny is consistent with historical records and the proposal that a short time before 1918, a new H1N1 avianlike virus entered human or swine hosts (O. T. Gorman, R. O. Donis, Y. Kawaoka, and R. G. Webster, J. Virol. 64:4893-4902, 1990). This virus provided the ancestors of all known human influenza A virus genes, except for HA, NA, and PB1, which have since been reassorted from avian viruses. We propose that during 1918 a virulent strain of this new avianlike virus caused a severe human influenza pandemic and that the pandemic virus was introduced into North American swine populations, constituting the origin of classical swine virus.

Nucleotide sequences of 24 nucleoprotein (NP) genes isolated from a wide range of hosts, geographic regions, and influenza A virus serotypes and 18 published NP gene sequences were analyzed to determine evolutionary relationships. The phylogeny of NP genes was determined by a maximum-parsimony analysis of nucleotide sequences. Phylogenetic analysis showed that NP genes have evolved into five host-specific lineages, including (i) Equine/Prague/56 (EQPR56), (ii) recent equine strains, (iii) classic swine (H1N1 swine, e.g., A/Swine/Iowa/15/30) and human strains, (iv) gull H13 viruses, and (v) avian strains (including North American, Australian, and Old World subgroups). These NP lineages match the five RNA hybridization groups identified by W. J. Bean (Virology 133:438-442, 1984). Maximum nucleotide differences among the NPs was 18.5%, but maximum amino acid differences reached only 10.8%, reflecting the conservative nature of the NP protein. Evolutionary rates varied among lineages; the human lineage showed the highest rate (2.54 nucleotide changes per year), followed by the Old World avian lineage (2.17 changes per year) and the recent equine lineage (1.22 changes per year). The per-nucleotide rates of human and avian NP gene evolution (1.62 x 10(-3) to 1.39 x 10(-3) changes per year) are lower than that reported for human NS genes (2.0 x 10(-3) changes per year; D. A. Buonagurio, S. Nakada, J. D. Parvin, M. Krystal, P. Palese, and W. M. Fitch, Science 232:980-982, 1986). Of the five NP lineages, the human lineage showed the greatest evolution at the amino acid level; over a period of 50 years, human NPs have accumulated 39 amino acid changes. In contrast, the avian lineage showed remarkable conservatism; over the same period, avian NP proteins changed by 0 to 10 amino acids. The specificity of the H13 NP in gulls and its distinct evolutionary separation from the classic avian lineage suggests that H13 NPs may have a large degree of adaptation to gulls. The presence of avian and human NPs in some swine isolates demonstrates the susceptibility of swine to different virus strains and supports the hypothesis that swine may serve as intermediates for the introduction of avian influenza virus genes into the human virus gene pool. EQPR56 is relatively distantly related to all other NP lineages, which suggests that this NP is rooted closest to the ancestor of all contemporary NPs. On the basis of estimation of evolutionary rates from nucleotide branch distances, current NP lineages are at least 100 years old, and the EQPR56 NP is much older.(ABSTRACT TRUNCATED AT 400 WORDS)

In 1991 and 1992, H1N2 influenza A viruses were isolated from the lungs of pigs with overt signs of respiratory disease at farms in the Chiba and Kanagawa prefectures of Japan. To determine the genetic origin of these isolates, we phylogenetically analyzed partial nucleotide sequences of their genes. The results indicate that influenza viruses possessing the N2 of human influenza virus and seven other gene segments of classical H1N1 swine influenza virus, which were first isolated in 1980, have become established in Japanese pigs.

In October of 1993, there was decreased egg production and increased mortality among Mexican chickens, in association with serologic evidence of an H5N2 influenza virus. First isolated from chickens in May of 1994, after spreading widely in the country, the virus caused only a mild respiratory syndrome in specific pathogen-free chickens. Because eradication of the virus by destruction of infected birds posed major obstacles to the poultry industry in Mexico, we were able to conduct a "field experiment" to determine the fate of an avirulent virus after repeated cycles of replication in millions of chickens. By the end of 1994, the virus had mutated to contain a highly cleavable hemagglutinin (HA), but remained only mildly pathogenic in chickens. Within months, however, it had become lethal in poultry. Nucleotide sequence analysis of the HA cleavage site of the original avirulent strain revealed R-E-T-R, typical of avirulent viruses and unlike the K-K-K-R sequence characterizing viruses responsible for the 1983 outbreak in poultry in the United States. Both mildly and highly pathogenic isolates contained insertions and a substitution of basic residues in the HA connecting peptide, R-K-R-K-T-R, which made the HA highly cleavable in trypsin-free chicken embryo fibroblasts. Phylogenetic analysis of the HA of H5 avian influenza viruses, including the Mexican isolates, indicated that the epidemic virus had originated from the introduction of a single virus of the North American lineage into Mexican chickens. This sequence of events demonstrates, apparently for the first time, the stepwise acquisition of virulence by an avian influenza virus in nature.

Genetic reassortment between influenza A viruses in humans and in animals and birds has been implicated in the appearance of new pandemics of human influenza. To determine whether such reassortment has occurred in the United States, the authors compared the genetic origins of gene segments of 73 swine influenza virus isolates (1976-1990), representing 11 states, and 11 turkey virus isolates (1980-1989), representing eight states. The host origin of gene segments encoding the internal proteins of H1N1 swine and turkey influenza viruses was identified by developing a dot-blot assay. All gene segments of swine influenza viruses were characteristic of influenza virus genes from that species, indicating that pigs may not be frequent participants in interspecies genetic exchange and reassortment of influenza viruses in the United States. In contrast, 73% of the turkey influenza virus isolates contained genes of swine origin. One turkey isolate was a reassortant having three genes characteristic of avian influenza virus and three of swine origin. These findings document a high degree of genetic exchange and reassortment of influenza A viruses in domestic turkeys in the United States. The molecular biologic techniques used by the authors should aid future epidemiologic studies of influenza pandemics.

The N1 and N9 neuraminidase (NA) subtypes of influenza A viruses exhibit significant hemadsorption activity that localizes to a site distinct from that of the enzymatic active site. To determine the conservation of hemadsorption activity among different NAs, we have examined most of the NA subtypes from avian, swine, equine, and human virus isolates. All subtypes of avian virus NAs examined and one equine virus N8 NA possessed high levels of hemadsorption activity. A swine virus N1 NA exhibited only weak hemadsorption activity, while in human virus N1 and N2 NAs, the activity was detected at a much lower level than in avian virus NAs. NAs which possessed hemadsorption activity for chicken erythrocytes (RBCs) were similarly able to adsorb human RBCs. However, none of the hemadsorption-positive NAs could bind equine, swine, or bovine RBCs, suggesting that RBCs from these species lack molecules, recognized by the NA hemadsorption site, present on human and chicken RBCs. Mutagenesis of the putative hemadsorption site of A/duck/Hong Kong/7/75 N2 NA abolished the high level of hemadsorption activity exhibited by the wild-type protein but also resulted in a 50% reduction of the NA enzymatic activity. A transfectant virus, generated by reverse genetics, containing this mutated NA replicated 10-fold less efficiently in chicken embryo fibroblast cultures than did a transfectant virus expressing the wild-type NA. However, both viruses replicated equally well in Peking ducks. Although conservation of NA hemadsorption activity among avian virus NAs suggests the maintenance of a required function of NA, loss of the activity does not preclude the replication of the virus in an avian host.

Coinfection of wild birds by influenza A viruses is thought to be an important mechanism for the diversification of viral phenotypes by generation of reassortants. However, it is not known whether coinfection is a random event or follows discernible patterns with biological significance. In the present study, conducted with viruses collected throughout 15 years from a wild-duck population in Alberta, Canada, we identified three discrete distributions of coinfections. In about one-third of the events, which involved subtypes of viruses that appear to be maintained in this duck reservoir, coinfection occurred at rates either close to or significantly lower than one would predict from rates of single-virus infection. Apparently, the better adapted an influenza A virus is to an avian population, the greater is its ability to prevent coinfections. Conversely, poorly adapted, nonmaintained viruses were significantly overrepresented as coinfectants. Rarely encountered subtypes appear to represent viruses whose chances of successfully infiltrating avian reservoirs are increased by coinfection. Mallards (Anas platyrhynchos) and pintails (A. acuta) were significantly more likely to be infected by a single influenza A virus than were the other species sampled, but no species was significantly more likely to be coinfected. These observations provide the first evidence of nonrandom coinfection of wild birds by influenza A viruses, suggesting that reassortment of these viruses in a natural population does not occur randomly. These results suggest that even though infections may occur in a species, all subtypes are not maintained by all avian species. They also suggest that specific influenza A virus subtypes are differentially adapted to different avian hosts and that the fact that a particular subtype is isolated from a particular avian species does not mean that the virus is maintained by that species.

Avian influenza A viruses from Asia are recognized as the source of genes that reassorted with human viral genes to generate the Asian/57 (H2N2) and Hong Kong/68 (H3N2) pandemic strains earlier in this century. Here we report the genetic analysis of avian influenza A H1N1 viruses recently isolated from pigs in southern China, a host suspected to generate new pandemic strains through gene reassortment events. Each of the eight gene segments was of avian origin. Phylogenetic analysis indicates that these genes form an Asian sublineage of the Eurasian avian lineage, suggesting that these viruses are an independent introduction into pigs in Asia. The presence of avian influenza viruses in pigs in China places them in an optimal position for transmission to humans and may serve as an early warning of the emergence of the next human influenza virus pandemic.

We established a reverse genetics system for the M gene of influenza A virus, using amantadine resistance as a selection criterion. Transfection of an artificial M ribonucleoprotein complex of A/Puerto Rico/8/34 (H1N1), a naturally occurring amantadine-resistant virus, and superinfection with amantadine-sensitive A/equine/Miami/1/63 (H3N8), followed by cultivation in the presence of the drug, led to the generation of a transfectant virus with the A/Puerto Rico/8/34 (H1N1) M gene. With this system, we attempted to generate a virus containing a deletion in an M-gene product (M2 protein). Viruses lacking the carboxyl-terminal Glu of M2, but not those lacking 5 or 10 carboxyl-terminal residues, were rescued in the presence of amantadine. These findings indicate that carboxyl-terminal residues of the M2 protein play an important role in influenza virus replication. The M-gene-based reverse genetics system will allow the study of different M-gene mutations to achieve a balance between attenuation and virus replication, thus facilitating the production of live vaccine strains.

The virulence of avian influenza viruses correlates with the sensitivity of their hemagglutinin (HA) to cellular proteases. Furin, a proprotein-processing subtilisin-related endoprotease, is a leading candidate for the enzyme that cleaves the HA of virulent avian viruses. We therefore compared the specificity of furin with those of proteases in a variety of cultured cells and in a rat Golgi fraction, using the HA cleavage mutants of a virulent avian influenza virus, A/Turkey/Ireland/1378/85 (H5N8). The results indicated similar sequence specificities among the endoproteases when purified furin was used. In experiments with the vaccinia virus expression system, overexpressed furin cleaved mutant HAs that were not recognized by the endogenous proteases, resulting in an apparent broader specificity of furin. These findings authenticate the proposed role of furin as an HA-activating protease in vivo and caution against the use of expression vectors to study protease sequence specificity.

Live attenuated influenza A virus vaccines are currently produced by the transfer of attenuating genes from a donor virus to new epidemic variants of influenza A virus, with the selection of reassortant viruses that possess the protective antigens (i.e., the two surface glycoproteins) of the epidemic virus and the attenuating genes from the donor virus. The previously studied attenuated donor viruses were produced by conventional methods such as passage of virus at low temperature or chemical mutagenesis. The present paper describes a new strategy for the generation of a donor virus bearing an attenuating, non-surface-glycoprotein gene. This strategy involves the introduction of attenuating mutations into the cDNA copy of the PB2 polymerase gene by site-directed mutagenesis, transfection of in vitro RNA transcripts of PB2 cDNA, and recovery of the transfected PB2 gene into an infectious virus. An avian-human influenza A virus PB2 single-gene reassortant virus (with an avian influenza A virus PB2 gene) that replicates efficiently in avian tissue but poorly in mammalian cells was used as a helper virus to rescue a transfected synthetic RNA derived from a human influenza A virus PB2 gene. The desired human influenza A virus mutant PB2 transfectant was favored in this situation because the avian influenza A virus PB2 gene restricts viral replication in mammalian cells in culture, the system used for rescue, thereby providing strong selection for the virus bearing the human influenza A virus PB2 gene. We validated the feasibility of this approach by rescuing the PB2 gene of the wild-type influenza A/Ann Arbor/6/60 virus and a mutant derivative that had a single amino acid substitution introduced at position 265 by site-directed mutagenesis. Previously, this amino acid substitution had been shown to specify both a temperature-sensitive (ts) and an attenuation (att) phenotype. The rescued mutant 265 PB2 transfectant virus exhibited the ts and att phenotypes, which confirms that these phenotypes were specified by this single amino acid substitution. The transfectant virus was immunogenic and protected hamsters from subsequent challenge with wild-type virus. The cDNA copy of this influenza A/Ann Arbor/6/60 virus mutant 265 PB2 gene will be used as a substrate for the introduction of additional attenuating mutations by site-directed mutagenesis.

The significance of the conserved cytoplasmic tail sequence of influenza A virus neuraminidase (NA) was analyzed by the recently developed reverse genetics technique (W. Luytjes, M. Krystal, M. Enami, J. D. Parvin, and P. Palese, Cell 59:1107-1113, 1989). A chimeric influenza virus A/WSN/33 NA containing the influenza B virus cytoplasmic tail rescued influenza A virus infectivity. The transfectant virus had less NA incorporated into virions than A/WSN/33, indicating that the cytoplasmic tail of influenza virus NA plays a role in incorporation of NA into virions. However, these results also suggest that the influenza A virus and influenza B virus cytoplasmic tail sequences share common features that lead to the production of infectious virus. Transfectant virus was obtained with all cytoplasmic tail mutants generated by site-directed mutagenesis of the influenza A virus tail, except for the mutant resulting from substitution of the conserved proline residue, presumably because of its contribution to the secondary structure of the tail. No virus was rescued when the cytoplasmic tail was deleted, indicating that the cytoplasmic tail is essential for production of the virus. The virulence of the transfectant viruses in mice was directly proportional to the amount of NA incorporated. The importance of the NA cytoplasmic tail in virus assembly and virulence has implications for use in developing antiviral strategies.

Phylogenetic analysis of 42 membrane protein (M) genes of influenza A viruses from a variety of hosts and geographic locations showed that these genes have evolved into at least four major host-related lineages: (i) A/Equine/prague/56, which has the most divergent M gene; (ii) a lineage containing only H13 gull viruses; (iii) a lineage containing both human and classical swine viruses; and (iv) an avian lineage subdivided into North American avian viruses (including recent equine viruses) and Old World avian viruses (including avianlike swine strains). The M gene evolutionary tree differs from those published for other influenza virus genes (e.g., PB1, PB2, PA, and NP) but shows the most similarity to the NP gene phylogeny. Separate analyses of the M1 and M2 genes and their products revealed very different patterns of evolution. Compared with other influenza virus genes (e.g., PB2 and NP), the M1 and M2 genes are evolving relatively slowly, especially the M1 gene. The M1 and M2 gene products, which are encoded in different but partially overlapping reading frames, revealed that the M1 protein is evolving very slowly in all lineages, whereas the M2 protein shows significant evolution in human and swine lineages but virtually none in avian lineages. The evolutionary rates of the M1 proteins were much lower than those of M2 proteins and other internal proteins of influenza viruses (e.g., PB2 and NP), while M2 proteins showed less rapid evolution compared with other surface proteins (e.g., H3HA). Our results also indicate that for influenza A viruses, the evolution of one protein of a bicistronic gene can affect the evolution of the other protein.(ABSTRACT TRUNCATED AT 400 WORDS)

Phylogenetic analysis of 20 influenza A virus PB2 genes showed that PB2 genes have evolved into the following four major lineages: (i) equine/Prague/56 (EQPR56); (ii and iii) two distinct avian PB2 lineages, one containing FPV/34 and H13 gull virus strains and the other containing North American avian and recent equine strains; and (iv) human virus strains joined with classic swine virus strains (i.e., H1N1 swine virus strains related to swine/Iowa/15/30). The human virus lineage showed the greatest divergence from its root relative to other lineages. The estimated nucleotide evolutionary rate for the human PB2 lineage was 1.82 x 10(-3) changes per nucleotide per year, which is within the range of published estimates for NP and NS genes of human influenza A viruses. At the amino acid level, PB2s of human viruses have accumulated 34 amino acid changes over the past 55 years. In contrast, the avian PB2 lineages showed much less evolution, e.g., recent avian PB2s showed as few as three amino acid changes relative to the avian root. The completion of evolutionary analyses of the PB1, PB2, PA and NP genes of the ribonucleoprotein (RNP) complex permits comparison of evolutionary pathways. Different patterns of evolution among the RNP genes indicate that the genes of the complex are not coevolving as a unit. Evolution of the PB1 and PB2 genes is less correlated with host-specific factors, and their proteins appear to be evolving more slowly than NP and PA. This suggests that protein functional constraints are limiting the evolutionary divergence of PB1 and PB2 genes. The parallel host-specific evolutionary pathways of the NP and PA genes suggest that these proteins are coevolving in response to host-specific factors. PB2s of human influenza A viruses share a common ancestor with classic swine virus PB2s, and the pattern of evolution suggests that the ancestor was an avian virus PB2. This same pattern of evolution appears in the other genes of the RNP complex. Antigenic studies of HA and NA proteins and sequence comparisons of NS and M genes also suggest a close ancestry for these genes in human and classic swine viruses. From our review of the evolutionary patterns of influenza A virus genes, we propose the following hypothesis: the common ancestor to current strains of human and classic swine influenza viruses predated the 1918 human pandemic virus and was recently derived from the avian host reservoir.

The ability of many viruses to replicate in host cells depends on cleavage of certain viral glycoproteins, including hemagglutinin (HA). By generating site-specific mutant HAs of two highly virulent influenza viruses, we established that the relationship between carbohydrate in the stalk and the length of the connecting peptide is a critical determinant of cleavability. HAs that lacked an oligosaccharide side chain in the stalk were cleaved regardless of the number of basic amino acids at the cleavage site, whereas those with the oligosaccharide side chain resisted cleavage unless additional basic amino acids were inserted. This finding suggests that the oligosaccharide side chain interferes with HA cleavage if the number of basic amino acids at the cleavage site is not adequate to nullify this effect. Similar interplay could influence cleavage of other viral glycoproteins, such as those of human and simian immunodeficiency viruses and paramyxoviruses.

The nucleotide and amino acid sequences of 40 influenza virus hemagglutinin genes of the H3 serotype from mammalian and avian species and 9 genes of the H4 serotype were compared, and their evolutionary relationships were evaluated. From these relationships, the differences in the mutational characteristics of the viral hemagglutinin in different hosts were examined and the RNA sequence changes that occurred during the generation of the progenitor of the 1968 human pandemic strain were examined. Three major lineages were defined: one containing only equine virus isolates; one containing only avian virus isolates; and one containing avian, swine, and human virus isolates. The human pandemic strain of 1968 was derived from an avian virus most similar to those isolated from ducks in Asia, and the transfer of this virus to humans probably occurred in 1965. Since then, the human viruses have diverged from this progenitor, with the accumulation of approximately 7.9 nucleotide and 3.4 amino acid substitutions per year. Reconstruction of the sequence of the hypothetical ancestral strain at the avian-human transition indicated that only 6 amino acids in the mature hemagglutinin molecule were changed during the transition between an avian virus strain and a human pandemic strain. All of these changes are located in regions of the molecule known to affect receptor binding and antigenicity. Unlike the human H3 influenza virus strains, the equine virus isolates have no close relatives in other species and appear to have diverged from the avian viruses much earlier than did the human virus strains. Mutations were estimated to have accumulated in the equine virus lineage at approximately 3.1 nucleotides and 0.8 amino acids per year. Four swine virus isolates in the analysis each appeared to have been introduced into pigs independently, with two derived from human viruses and two from avian viruses. A comparison of the coding and noncoding mutations in the mammalian and avian lineages showed a significantly lower ratio of coding to total nucleotide changes in the avian viruses. Additionally, the avian virus lineages of both the H3 and H4 serotypes, but not the mammalian virus lineages, showed significantly greater conservation of amino acid sequence in the internal branches of the phylogenetic tree than in the terminal branches. The small number of amino acid differences between the avian viruses and the progenitor of the 1968 pandemic strain and the great phenotypic stability of the avian viruses suggest that strains similar to the progenitor strain will continue to circulate in birds and will be available for reintroduction into humans.

The cleavability of the hemagglutinin (HA) molecule is related to the virulence of avian influenza A viruses, but its influence on human influenza virus strains is unknown. Two structural features are involved in the cleavage of avian influenza A virus HAs: a series of basic amino acids at the cleavage site and an oligosaccharide side chain in the near vicinity. The importance of these properties in the cleavability of a human influenza A virus (A/Aichi/2/68) HA was investigated by using mutants that contained or lacked an oligosaccharide side chain and had either four or six basic amino acids. All mutants except the one that contains a single mutation at the glycosylation site were cleaved, although not completely, demonstrating that a series of basic amino acids confers susceptibility to cellular cleavage enzymes among human influenza virus HAs. The mutants containing six basic amino acids at the cleavage site showed limited polykaryon formation upon exposure to low pH, indicating that cleavage was adequate to impart fusion activity to the HA. Deletion of the potential glycosylation site had no effect on the cleavability of these mutants; hence, the oligosaccharide side chain appears to have no role in human influenza virus HA cleavage. The inability to induce high cleavability in a human influenza A virus HA by insertion of a series of basic amino acids at the cleavage site indicates that other, as yet unidentified structural features are needed to enhance the susceptibility of these HAs to cellular proteases.

Normal horse and guinea pig sera contain the glycoprotein inhibitor alpha 2-macroglobulin, which inhibits the infectivity and hemagglutinating activity of influenza A viruses of the H2 and H3 subtypes. In the current study, the presence of inhibitors of influenza A virus in pig and rabbit sera was investigated. Variants of influenza virus type A/Los Angeles/2/87(H3N2) that were resistant to horse, pig, or rabbit serum were isolated. Analysis of the variant viruses with anti-hemagglutinin (HA) monoclonal antibodies revealed that antigenic changes occurred with the development of serum inhibitor resistance. Characterization of the inhibitors in pig and rabbit sera by using periodate and receptor-destroying enzyme demonstrated that carbohydrate is an important constituent of the active portion of both inhibitor molecules and that sialic acid is involved in the interaction of the inhibitors with influenza virus HA. Nucleotide sequence analysis of the HA molecule revealed that the serum-resistant variants each acquired a different set of amino acid alterations. The multiply resistant variants maintained the original amino acid changes and acquired additional changes. Sequence modifications in the HA involved the conserved amino acids within the receptor binding site (RBS) at position 137 and the second-shell RBS residues at positions 155 and 186. Amino acid changes also occurred within antigenic site A (position 145) and directly behind the receptor binding pocket (position 220). Amino acid alterations resulted in the acquisition of a potential glycosylation site at position 128 and the loss of potential glycosylation sites at positions 246 and 248. The localization of the amino acid changes in HA1 to the region of the RBS supports the concept of serum inhibitors as receptor analogs. The unique set of mutations acquired by the serum inhibitor-resistant variants strongly suggests that horse, pig, and rabbit sera each contain distinct glycoprotein inhibitors of influenza A virus.

We determined the origin and evolutionary pathways of the PB1 genes of influenza A viruses responsible for the 1957 and 1968 human pandemics and obtained information on the variable or conserved region of the PB1 protein. The evolutionary tree constructed from nucleotide sequences suggested the following: (i) the PB1 gene of the 1957 human pandemic strain, A/Singapore/1/57 (H2N2), was probably introduced from avian species and was maintained in humans until 1968; (ii) in the 1968 pandemic strain, A/NT/60/68 (H3N2), the PB1 gene was not derived from the previously circulating virus in humans but probably from another avian virus; and (iii) a current human H3N2 virus inherited the PB1 gene from an A/NT/60/68-like virus. Nucleotide sequence analysis also showed that the avian PB1 gene was introduced into pigs. Hence, transmission of the PB1 gene from avian to mammalian species is a relatively frequent event. Comparative analysis of deduced amino acid sequences disclosed highly conserved regions in PB1 proteins, which may be key structures required for PB1 activities.

The influenza A virus [A/Chicken/Pennsylvania/1370/83 (H5N2)] that caused up to 80% mortality among chickens provided a model system for testing the efficacy of chemotherapeutic agents against highly virulent influenza virus. Amantadine and rimantadine administered in drinking water were efficacious both prophylactically and therapeutically. However, under conditions simulating natural transmission of virus, amantadine- and rimantadine-resistant viruses arose and were transmitted to other birds in contact with the infected chickens, causing mortality. Simultaneous administration of inactivated H5N2 vaccine and amantadine provided protection. Thus, chemotherapy may be useful in the treatment of a highly pathogenic influenza virus outbreak in humans or other animals when used in combination with vaccine.

In April 1983, an influenza virus of low virulence appeared in chickens in Pennsylvania. Subsequently, in October 1983, the virus became virulent and caused high mortality in poultry. The causative agent has been identified as an influenza virus of the H5N2 serotype. The hemagglutinin is antigenically closely related to tern/South Africa/61 (H5N3) and the neuraminidase is similar to that from human H2N2 strains (e.g., A/Japan/305/57) and from some avian influenza virus strains (e.g., A/turkey/Mass/66 [H6N2]). Comparison of the genome RNAs of chicken/Penn with other influenza virus isolates by RNA-RNA hybridization indicated that all of the genes of this virus were closely related to those of various other influenza virus isolates from wild birds. Chickens infected with the virulent strain shed high concentrations of virus in their feces (10(7) 50% egg infective dose per g), and the virus was isolated from the albumin and yolk of eggs layed just before death. Virus was also isolated from house flies in chicken houses. Serological and virological studies showed that humans are not susceptible to infection with the virus, but can serve as short-term mechanical carriers. Analysis of the RNA of the viruses isolated in April and October by gel migration and RNA-RNA hybridization suggested that these strains were very closely related. Oligonucleotide mapping of the individual genes of virulent and avirulent strains showed a limited number of changes in the genome RNAs, but no consistent differences between the virulent and avirulent strains that could be correlated with pathogenicity were found. Polyacrylamide gel analysis of the early (avirulent) isolates demonstrated the presence of low-molecular-weight RNA bands which is indicative of defective-interfering particles. These RNAs were not present in the virulent isolates. Experimental infection of chickens with mixtures of the avirulent and virulent strains demonstrated that the avirulent virus interferes with the pathogenicity of the virulent virus. The results suggest that the original avirulent virus was probably derived from influenza viruses from wild birds and that the virulent strain was derived from the avirulent strain by selective adaptation rather than by recombination or the introduction of a new virus into the population. This adaptation may have involved the loss of defective RNAs, as well as mutations, and thus provides a possible model for a role of defective-interfering particles in nature.

A reverse transcriptase PCR (RT-PCR) was used for rapid determination of the hemagglutinin (HA) cleavage site sequence, a marker for the virulence potential of avian influenza viruses. When applied to specimens from chickens experimentally infected with either a virulent or an avirulent virus, RT-PCR uniformly detected the HA gene, even in specimens that were negative for virus by standard testing in eggs. This technique, combined with sequencing of the HA cleavage site, offers a rapid and sensitive way to assess the virulence potential of avian influenza viruses. Early detection of field isolates with virulence-associated structural motifs at the HA cleavage site would allow better control of influenza among large poultry populations.

Directigen FLU-A, a new enzyme immunoassay membrane test, rapidly detects influenza A virus antigen in specimens from patients. Nasopharyngeal washes and pharyngeal gargles were used to determine the effectiveness of the assay as applied to different types of routinely collected clinical samples. All specimens had been previously shown to contain influenza A virus by virus isolation in tissue culture. Directigen FLU-A was 90% sensitive (95% confidence interval, 56 to 99.7%) with nasopharyngeal washes but only 39% sensitive (95% confidence interval, 17 to 64%) with pharyngeal gargles (P = 0.018) when used with samples containing similar amounts of infectious virus (50% tissue culture infective dose, 1.0 to 4.5). The intensity of the positive reaction with Directigen FLU-A did not correlate with the amount of virus in the specimens. Directigen FLU-A was found to detect cell-associated antigen more readily than free virus; only 20 infected cells were required to identify cell-associated influenza A virus antigen, whereas the limit of detection for free virus was 1.63 x 10(3) infectious virus particles. These findings suggest that Directigen FLU-A detects the cell-associated antigen present in clinical specimens rather than free virus. In addition, Directigen FLU-A detected avian and swine influenza A viruses in both cloacal swabs (75% sensitivity) and swine lung homogenates (86% sensitivity), indicating its potential usefulness in the surveillance of nonhuman influenza A viruses.

The A/Los Angeles/2/87 (H3N2) (A/LA/2/87) virus is sensitive to inhibitors of hemagglutination present in certain human sera. It was found that the effect of these inhibitors could be removed by treating sera with high-concentration receptor-destroying enzyme or trypsin-periodate or by using inhibitor-resistant viruses in the hemagglutination inhibition (HAI) test. Inhibitor-resistant viruses were not effective for detecting rises in antibody titers in the sera of volunteers infected with the A/LA/2/87 wild-type virus, while rises in antibody titer were readily detected in sera treated with trypsin-periodate and tested against A/LA/2/87 wild-type virus in an HAI test. It is therefore suggested that chemical or enzymatic methods be used to inactivate serum inhibitors and that standard virus be used in the HAI test for the currently circulating H3N2 viruses.