Swine influenza-like viruses in turkeys: potential source of virus for humans?

Influenza A viruses (subtype H1N1), recently isolated from turkeys in different areas of the United States, were determined to be closely related to strains typically associated with pigs. This conclusion was based on comparisons of H1N1 isolates from pigs, humans, ducks, and turkeys with polyclonal and monoclonal antibodies, RNA-RNA competitive hybridization, and replication studies. One of the H1N1 isolates from turkeys caused influenza in a laboratory technician, who displayed fever, respiratory illness, virus shedding, and seroconversion. These results suggest that turkeys as well as pigs are involved in the maintenance of these viruses and their transmission to humans.

Reassortant virus derived from avian and human influenza A viruses is attenuated and immunogenic in monkeys.

An influenza A reassortant virus that contained the hemagglutinin and neuraminidase genes of a virulent human virus, A/Udorn/72 (H3N2), and the six other influenza A virus genome segments from an avirulent avian virus, A/Mallard/New York/6750/78 (H2N2), was evaluated for its level of replication is squirrel monkeys and hamsters. In monkeys, the reassortant virus was as attenuated and as restricted in its level of replication in the upper and lower respiratory tract as its avian influenza virus parent. Nonetheless, infection with the reassortant induced significant resistant to challenge with virulent human influenza virus. In hamsters, the reassortant virus replicated to a level intermediate between that of its parents. These findings suggest that the nonsurface antigen genes of the avian parental virus are the primary determinants of restriction of replication of the reassortant virus in monkeys. Attenuation of the reassortant virus for primates is achieved by inefficient functioning of the avian influenza genes in primate cells, while antigenic specificity of the human influenza virus is provided by the neuraminidase and hemagglutinin genes derived from the human virus. This approach could lead to the development of a live influenza A virus vaccine that is attenuated for man if the avian influenza genes are similarly restricted in human cells.

Mass mortality of harbor seals: pneumonia associated with influenza A virus.

More than 400 harbor seals, most of them immature, died along the New England coast between December 1979 and October 1980 of acute pneumonia associated with influenza virus, A/Seal/Mass/1/180 (H7N7). The virus has avian characteristics, replicates principally in mammals, and causes mild respiratory disease in experimentally infected seals. Concurrent infection with a previously undescribed mycoplasma or adverse environmental conditions may have triggered the epizootic. The similarities between this epizootic and other seal mortalities in the past suggest that these events may be linked by common biological and environmental factors.

Evaluation of Moloney murine sarcoma and leukemia virus complex as a model for airborne oncogenic virus biohazards: survival of airborne virus and exposure of mice.

Aerosols of the Moloney murine sarcoma virus (MuSV-M) and leukemia virus (MuLV-M) complex (MuSV-M/MuLV-M) were generated from refluxing atomizers and then aged in rotating drums at 21 degrees C holding temperature with relative humidities ranging from 25 to 76%. The MuSV-M and MuLV-M aerosolized from the same tumor extract preparation survived almost equally at the four humidity levels. Both viruses remained viable in the airborne state for at least 2 hours after aerosolization. When mice were exposed to airborne MuSV-M/MuLV-M, no macroscopic lesions were observed in lungs or other tissues examined during the 2-month postexposure period. On the basis of this study, MuSV-M was determined unsuitable as a "model system" in which a simple aerosol dose response could be used for biohazard evaluation of oncogenic virus aerosols.

The influence of calcium and reactive oxygen species on influenza virus-induced apoptosis.

In previous studies we observed that influenza A and B viruses induce apoptosis in Madin-Darby canine kidney (MDCK) cells and that this apoptosis is blocked by expression of bcl-2. Using a well-characterized, highly virulent, avian influenza virus, A/Turkey/Ontario/7732/66 (H5N9) (Ty/Ont), we sought to better understand this system. We investigated the influence of two cellular factors that are known to function in other models of apoptosis inhibited by bcl-2, calcium (Ca(2+)) and reactive oxygen species (ROS). Although Ca(2+) chelators generally inhibit apoptosis, treatment of MDCK cells with either an extracellular chelator, EGTA, or an intracellular chelator, BAPTAAM, induced apoptosis instead and enhanced Ty/Ont-induced apoptosis. Conversely, treatment with an ionophore, ionomycin, blocked the viral-induced apoptosis. In terms of ROS, neither treatment with antioxidants, N(2) flushing to induce hypoxia, nor nigericin (a compound which, like bcl-2, stabilizes the mitochondrial membrane potential against the effects of ROS and subsequent Ca(2+) dysregulation) were able to block Ty/Ont-induced apoptosis. Therefore, it is likely that ROS play little, if any, role in influenza-induced apoptosis in MDCK cells and the influence of Ca(2+) appears to be opposite to that in the majority of other more classical models of apoptosis.

Genetic relatedness of the nucleoprotein (NP) of recent swine, turkey, and human influenza A virus (H1N1) isolates.

The sequences of nucleoprotein (NP) genes of recent human and turkey isolates of influenza A viruses, which serologically could be correlated to contemporary swine viruses, were determined. These sequences were closely related to the NPs of these swine viruses and they formed a separate branch on the phylogenetic tree. While the early swine virus from 1931 resembled the avian strains in consensus amino acids of the NP and in its ability to rescue NP ts mutants of fowl plague virus in chicken embryo cells, the later strains on that branch were different: at 15 positions they have their own amino acids and they rescued the NP ts mutants only poorly. Of the NPs of the human New Jersey/76 isolates analysed, one clustered with the recent H1N1 swine viruses of the U.S.A., the other one with contemporary human strains. Since the NP is one of the main determinants of species specificity it is concluded that, although the H1N1 swine isolates from the U.S.A. form their own branch in the phylogenetic tree, they can be transmitted to humans and turkeys, but they do not spread further in these populations and so far have not contributed to human pandemics. It is not very likely that they will do so in future, since its branch in the phylogenetic tree develops further away from the human and avian branch.

Influenza viruses related to A/Shearwater/Australia/1/72 (Hav6 Nav5) in domestic and feral birds.

One influenza-A virus isolated from domestic turkeys in California in 1964 and two from migrating ducks in Delaware in 1973 were classified as Hav6 and Nav5, i.e., antigenically the same as A/Shearwater/Australia/1/72. The detection of antigenically related viruses in three different species over a ten-year period suggests a broad host range, contributing to continued circulation in nature. The turkeys suffered severe respiratory disease although infected migratory birds have not revealed signs of disease. These results suggest migratory birds, which travel over vast areas, as a source of influenza viruses infecting domestic species.

Inhibition of nitric oxide induction from avian macrophage cell lines by influenza virus.

The virulent avian influenza virus A/Ty/Ont/7732/66 (H5N9) (Ty/Ont) causes a rapid destruction of lymphoid cells in infected birds. Avian macrophage cell lines, HD11 and MQ-NCSU, support productive replication of Ty/Ont and other influenza viruses. Therefore, the ability of these cell lines to produce nitric oxide (NO), a potentially cytotoxic mediator, in response to infection with Ty/Ont was examined. Although treatment with bacterial lipopolysaccharides (LPS) resulted in high NO levels, infection of macrophages with Ty/Ont resulted in NO levels lower than NO levels in untreated cells. Furthermore, Ty/Ont was able to inhibit the positive response to LPS in cultures simultaneously treated with LPS and virus. However, inactivated influenza virus did not exhibit this inhibitory effect. Different strains of influenza virus varied in their ability to inhibit NO production by the macrophages; this may be related to the level of virus replication in these cells. These data suggest that the ability of the avian macrophage to activate the NO synthesis pathway is seriously impaired by infection with virulent influenza viruses such as Ty/Ont.

Specific antibody responses and generation of antigenic variants in chickens immunized against a virulent avian influenza virus.

To examine the specificity of the antibody response to the influenza hemagglutinin and the generation of antigenic variants, chickens were immunized against the highly virulent H5 virus A/Ty/Ont/7732/66 (H5N9) and then challenged with a lethal dose of the virus. The antibody responses of these chickens to the hemagglutinin (HA) were examined with an enzyme-linked immunosorbent assay (ELISA) in which their sera were titrated for the ability to block the binding of monoclonal antibodies (MAbs) to five distinct neutralizing epitopes on the viral HA. Based on the ELISA results, a majority (5/6) of the chickens produced antibodies to three of the five neutralizing epitopes on the viral HA. After challenge, two of six immunized chickens shed virus and died; antigenic comparisons of isolates from these two chickens indicated the presence of an antigenic variant; i.e., there was a change in one neutralizing epitope on the HA of virus shed by one chicken. None of the chickens had produced antibodies to this particular epitope on the viral HA. Inoculation of chickens with this variant resulted in 100% mortality, demonstrating that a change in this particular epitope did not alter the virulence of the virus. These studies indicate that chickens immunized against highly virulent influenza viruses may excrete virulent variants following challenge with live virus.

Influenza A outbreaks in Minnesota turkeys due to subtype H10N7 and possible transmission by waterfowl.

Avian influenza outbreaks in Minnesota involving the H10N7 subtype occurred on two turkey farms in 1979 and on a third in 1980. The H10N7 (Hav2 Neq1) subtype had not previously been detected in turkeys in Minnesota or reported in the United States. The clinical signs ranged from severe, with a mortality rate as high as 31%, to subclinical. Antigenically indistinguishable viruses were isolated from healthy mallards on a pond adjacent to the turkey farms, suggesting that the virus responsible for the outbreak may have been introduced by feral ducks.

Influenza viruses and paramyxoviruses in ducks in the Atlantic flyway, 1977-1983, including an H5N2 isolate related to the virulent chicken virus.

From 1977 to 1983, waterfowl migrating along the Atlantic flyway were annually monitored for orthomyxoviruses and paramyxoviruses in an area in central New York State. A total of 168 influenza isolates were obtained from 1,430 waterfowl. Twenty-four combinations of hemagglutinin and neuraminidase subtypes were detected, with as many as 12 found in a single year. One combination, an H5N2 isolate in 1982, was closely related to the virulent chicken virus that appeared in Pennsylvania in 1983. The prevalence of influenza varied greatly among the common waterfowl species: mallards 42%, black ducks 30%, blue-winged teal 11%, wood ducks 2%, and Canada geese 0%. A total of 89 paramyxoviruses were also from these waterfowl. In contrast to findings with influenza virus, the prevalence of paramyxoviruses did not differ significantly among the duck species. Serotype 1 (Newcastle disease virus) was predominant; three other serotypes were also identified. These findings indicated that ducks in the Atlantic flyway continually harbor influenza viruses and paramyxoviruses. The viruses may be a source of infection for other species.

Epizootiology of avian influenza--simultaneous monitoring of sentinel ducks and turkeys in Minnesota.

Isolation-reared mallards (Anas platyrhynchos) were placed on ponds in turkey-rearing areas in Minnesota, and their cloacae were periodically swabbed to attempt isolating virus from embryonated chicken eggs. Nearby turkeys were sampled by taking cloacal and tracheal swabs as well as blood samples. Hemagglutinating viruses were identified at the National Veterinary Services Laboratory, U.S. Department of Agriculture, Ames, Iowa. During this two-year study, the weekly influenza virus-isolation rate from ducks varied from 0 to 24.4%. A total of 213 influenza viruses were isolated from the ducks. Twenty-six influenza virus subtypes were detected. Ninety-seven flocks of turkeys were diagnosed as having influenza by virus isolation and/or serology. Eight influenza virus subtypes were involved in the turkey outbreaks, and seven of these were also detected in the ducks and/or other avian species. The weekly infection rate of the sentinel ducks correlated directly with observations of wild ducks at the monitoring sites. Influenza virus was isolated from water samples collected near the sentinel duck sites during the study.

Influenza virus surveillance in waterfowl in Pennsylvania after the H5N2 avian outbreak.

During the latter stages of the lethal H5N2 influenza eradication program in domestic poultry in Pennsylvania in 1983-84, surveillance of waterfowl was done to determine if these birds harbored influenza viruses that might subsequently appear in poultry. From late June to November 1984, 182 hemagglutinating viruses were isolated from 2043 wild birds, primarily ducks, in the same geographical area as the earlier lethal H5N2 avian influenza outbreak. The virus isolates from waterfowl included paramyxoviruses (PMV-1, -4, and -6) and influenza viruses of 13 antigenic combinations. There was only one H5N2 isolate from a duck. Although this virus was antigenically related to the lethal H5N2 virus, genetic and antigenic analysis indicated that it could be discriminated from the virulent family of H5N2 viruses, and it did not originate from chickens. Many of the influenza viruses obtained from wild ducks were capable of replicating in chickens after experimental inoculation but did not cause disease. These studies show that many influenza A virus strains circulating in waterfowl in the vicinity of domestic poultry in Pennsylvania did not originate from domestic poultry. These influenza viruses from wild ducks were capable of infecting poultry; however, transmission of these viruses to poultry apparently was avoided by good husbandry and control measures.

Micro neuraminidase-inhibition assay for classification of influenza A virus neuraminidases.

A neuraminidase-inhibition (NI) assay performed in microtiter plates is described. This micro-NI assay is a modification of the NI assay recommended by the World Health Organization. It reduces the quantity of reagents required and permits antigenic classification of many isolates simultaneously. To determine the accuracy and sensitivity of this micro-NI assay, 110 influenza A viruses, representing all subtypes, based upon the nine known neuraminidases (NAs), were classified by both the micro-NI and macro-NI assays in two separate laboratories. The NAs were identified accurately by the micro-NI assay. Virus mixtures were detected by both assays, although the macro-NI was clearly more sensitive. The micro-NI assay was also suitable for testing sera for the presence of antibodies to the NAs. Although the micro-NI assay did not provide the quantitation of the macro-NI assay, it did prove to be a rapid method for virus classification and antibody studies on influenza A viruses.

Kidney lesions associated with mortality in chickens inoculated with waterfowl influenza viruses.

Seventy-six type A influenza viruses recovered from waterfowl in Wisconsin, California, South Dakota, Florida, Texas, Alabama, and Nebraska were tested for virulence in chickens. The challenge to chickens was intravenous inoculation of first-, second-, or third-egg-passage virus. Each of the virus strains was tested separately in three or four chickens. Eighteen of the 76 viruses caused the death of one or more chickens following inoculation. Postmortem lesions were similar in all dead birds. In decreasing order of frequency, gross lesions included: swollen kidneys evident as accentuated lobular patterns, urates in the pericardial sac, and urates on the surface of the liver. Microscopic lesions present in kidneys were consistent with visceral gout. Mortality was associated with inoculations having higher concentrations of infectious virus. These results indicate that the influenza A viruses circulating in duck populations may include strains potentially pathogenic for chickens.

Isolation of feline leukemia virus from clinical specimens.

Specimens obtained from feline leukemia virus (FeLV)-positive cats were examined for infectious FeLV. Feline leukemia virus was detected by a focus-forming assay and confirmed by florescent antibody. Techniques of sample processing were evaluated and adjusted for optimum detection of FeLV. Low levels of FeLV were detected in 2 of 10 oral samples; however, the majority of these samples (17 of 27 tested) produced cytopathic effects in tissue culture which prevented Fe LV detection. Three of 24 urine samples and 1 of 20 rectal specimens were positive for FeLV. One milk sample contained high levels of FeLV.

Use of polymerase chain reaction to detect swine influenza virus in nasal swab specimens.

Rapid and accurate detection of a virus in a population is a critical factor in the eventual treatment and/or control of the virus. In this study, we examined use of the polymerase chain reaction (PCR) to detect swine influenza virus in nasal swab specimens from infected pigs. This approach was first standardized, using viral RNA purified by guanidinium/phenol-chloroform extraction and placed in the same transport medium as the swabs. By using highly conserved primers for the swine H1 hemagglutinin, we amplified a 591-base pair fragment that was analyzed by use of agarose gel electrophoresis, Southern blot, and DNA sequencing. To evaluate PCR as a potential diagnostic tool for detection of swine influenza virus infection, we obtained nasal swab specimens from experimentally infected pigs. Amplification by PCR and reamplification of extracted samples with internal primers yielded detectable bands for an amount of virus less than that required to infect embryonating chicken eggs. We also tested swab specimens from pigs involved in 3 separate, natural episodes of swine influenza. These swab specimens were extracted, amplified and reamplified, producing visible bands on the gel and in Southern blots. We performed Southern blot analyses on all PCR products, to confirm that they were from viral H1 RNA. We also cloned and sequenced a 591-base pair product from 1 specimen and found that it was 100% identical to the hemagglutinin gene sequence of A/Sw/Ind/1726/88. Results indicate that PCR can be used to detect swine influenza virus, even in nasal swab specimens, the specimen typically collected for diagnosis of virus infection.

Antigenic and genetic analysis of a recently isolated H1N1 swine influenza virus.

Hemagglutinins (HA) of H1N1 swine influenza viruses isolated in the United States have remained antigenically and genetically conserved for many years. In contrast to such conservation, the HA of A/Swine/Nebraska/1/92 (Sw/Neb) could readily be distinguished from those of contemporary porcine viruses. Twenty-eight amino acid mutations differentiated the HA of Sw/Neb and A/Swine/Indiana/1726/88, the most recent H1N1 swine influenza virus for which HA sequence data were available. Among these differences were mutations at potential asparagine-linked glycosylation sites and charge changes at many residues. The Sw/Neb virus also could be differentiated from other swine influenza viruses in hemagglutination-inhibition assays with monoclonal antibodies to recent H1 swine HA. Nonetheless, overall sequence analysis of the HA and the nucleoprotein genes of Sw/Neb indicated that this virus was more closely related genetically to classic H1N1 swine influenza viruses than to H1N1 avian or human viruses. Infection of swine with Sw/Neb under experimental conditions induced clinical signs and lesions typical of swine influenza. However, affected swine in the field had high, persistent fevers, but relatively mild signs of respiratory tract disease. This study indicated that an antigenically and genetically novel variant of swine influenza virus was detected in the United States.

Antigenic and phenotypic variants of a virulent avian influenza virus selected during replication in ducks.

To evaluate the replication of a highly virulent avian influenza A virus in a potential reservoir host, mallard ducks (Anas platyrhynchos) were inoculated with the virulent strain A/Ty/Ont/7732/66 (H5N9). Viruses recovered from the ducks were analyzed by hemagglutination inhibition (HI) and enzyme-linked immunosorbent assay (ELISA) and found to possess antigenically altered viral hemagglutinins. Plaque formation on the Madin-Darby Canine Kidney (MDCK) cell line and on primary chicken embryo cells was investigated, and isolates recovered from the ducks differed from the wild type by being unable to form plaques on MDCK cells without trypsin. This phenotype did not appear to be due to inefficient cleavage of the hemagglutinin by host cell proteases since hemagglutinin immunoprecipitated from cell lysates was cleaved. Although the plaquing phenotype suggested attenuation of the isolates from the ducks, they were not significantly altered in their virulence for chickens shown by infectivity studies in vivo. These results indicate that replication of influenza A/Ty/Ont/7732/66 virus in ducks can produce antigenic and phenotypic variants which are still highly virulent for domestic poultry.

Influenza A viruses isolated from migrating ducks in Oklahoma.

Nine type A influenza viruses were isolated from migrating and wintering ducks in Oklahoma in 1976-77. Antigenic classification of the viruses isolated revealed three different subtypes: Hav1 Nav2, Hws N1, and Hav6 N2. Transmission of influenza viruses from the wild ducks to sentinel birds (McGraw mallards) on the same lakes was not detected.