Since the Fourth International Symposium on Avian Influenza (AI) there has been considerable AI activity in the Eastern Hemisphere. The higher profile of AI resulting from the human infections with H5N1 and H9N2 viruses in Hong Kong, in 1997 and 1999, respectively, resulted in increased reporting and active surveillance.

There have been three reported incidents of high-pathogenicity (HP) AI: H5N2 in northeastern Italy in 1997 (eight outbreaks); H5N1 in Hong Kong in 1997 recurring in 2001 and 2002; H7N1 in northeastern Italy resulting in 413 outbreaks in 1999–00. The Italian HPAI outbreaks were preceded by 199 H7N1 low-pathogenicity (LP) AI outbreaks in 1999, and this virus continued to cause some problems after the eradication of HPAI.

During the second half of the 1990s outbreaks of LPAI due to H9N2 subtype have been reported in Germany, Italy, Ireland, South Africa, Hungary, Korea, China, Hong Kong, countries of the Middle East, Iran, and Pakistan. The continued presence of virus of this subtype in the Middle and Far East may mean it is becoming an established endemic disease in those regions. Other more restricted outbreaks in poultry have resulted in the isolation of LPAI viruses of H5, H6, H7, and H10 subtypes.

Between 1997 and 2001, there was one report of highly pathogenic avian influenza (HPAI) in the Western Hemisphere and Pacific Basin. In 1997, in New South Wales, Australia, an outbreak caused by avian influenza (AI) virus subtype H7N4 involved both chickens and emus. All other reports of infections in poultry and isolations from wild bird species in the region pertained to low pathogenicity (LP) AI virus. Animal Health Officials in Canada reported isolations of subtypes H1, H6, H7, and H10 from domestic poultry and subtypes H3 and H13 from imported and wild bird species. In Mexico, the H5N2 LPAI virus, the precursor of the HPAI outbreak in 1994–95, was isolated from poultry in each year from 1997 to 2001. Since 1997, Mexico has used approximately 708 million doses of a killed H5N2 vaccine and an additional 459 million doses of a recombinant fowlpox-H5 vaccine in their H5N2 control program. In Central America, avian influenza was diagnosed for the first time when H5N2 LPAI virus was isolated from chickens in Guatemala and El Salvador in 2000 and 2001, respectively. The H5N2 virus was genetically similar to the H5N2 virus found in Mexico. Surveillance activities in the United States resulted in the detection of AI virus or specific antibodies in domestic poultry from 24 states. Eleven of the fifteen hemagglutinin (H1, H2, H3, H4, H5, H6, H7, H9, H10, H11, and H13) and eight of the nine neuraminidase (N1, N2, N3, N4, N6, N7, N8, and N9) subtypes were identified. Two outbreaks of LPAI virus were reported in commercial table-egg producing chickens; one caused by H7N2 virus in Pennsylvania in 1996–98 and the other caused by H6N2 virus in California in 2000–01. In addition, isolations of H5 and H7 LPAI virus were recovered from the live-bird markets (LBMs) in the northeast United States.

In November of 1997 an outbreak of highly pathogenic avian influenza occurred near the town of Tamworth, in northern New South Wales, Australia. The viruses isolated from chickens on two commercial chicken farms were identified as H7N4 viruses, with hemagglutinin cleavage site amino acid sequences of RKRKRG and intravenous pathogenicity indices of 2.52 and 2.90, respectively. A virus with an identical nucleotide sequence, but with an intravenous pathogenicity index of 1.30, was also isolated from cloacal swabs collected from asymptomatic emus kept on a third property.

H7N2 low-pathogenicity (LP) avian influenza (AI) virus was isolated from chickens submitted to the Pennsylvania Animal Diagnostic Laboratory System on December 4 and 5, 2001. The cases were from two broiler breeder flocks in central Pennsylvania that had clinical signs of an acute, rapidly spreading respiratory disease. Seroconversion to AI virus was detected on follow-up sampling. Subsequently, H7N2 LPAI virus was isolated in five different broiler flock cases submitted between December 14, 2001, and January 3, 2002. Clinical signs and lesions in broilers, when present, were compatible with multicausal respiratory disease. With the exception of one broiler flock that was processed, birds from all of the virus positive flocks were euthanatized in-house within 11 days of the original case submission date. Increased surveillance of poultry flocks within 10-mile radius zones centered at the foci of the positive farms continued until March 1, 2002. No additional cases were detected.

Recently seven isolates of avian influenza virus (AIV) serotype H9N2 recovered from an outbreak of AI were analyzed on the basis of their biological and molecular characteristics. All the isolates belonged to the low-pathogenicity group of AIV. To further evaluate their pathogenic potential in association with other organisms, an isolate was inoculated experimentally in chickens using different routes and subsequently challenged with infectious bronchitis virus, Ornithobacterium rhinotracheale or Escherichia coli. The virus isolation and seromonitoring data revealed a significant role of Escherichia coli in aggravating the clinical condition of the birds earlier infected with AIV (H9N2). The AIV-antigen was detected in lung, trachea, kidney, and cloacal bursa among the infected birds, using immunofluorescent antibody technique. In another experiment, chickens that were immunosuppressed chemically showed high mortality when challenged with AIV H9N2. The results indicated that this low pathogenicity AIV (H9N2) isolate could produce severe infection depending on the type of secondary opportunistic pathogens present under field conditions. This may explain the severity of infection with the present H9N2 outbreak in the field. A prolonged antibacterial therapy in flocks infected with AIV H9N2 and use of oil-based vaccine at an early age in new flocks has helped to control this infection and the disease.

The nonpathogenic avian influenza (AI) outbreak in Pennsylvania began in December 1996 when there was a trace back from a New York live bird market to a dealer's flock. A total of 18 commercial layer flocks, two commercial layer pullet flocks, and a commercial meat turkey flock were diagnosed with nonpathogenic AI (H7N2) viral infection with an economic loss estimated at between $3 and $4 million. Clinical histories of flocks infected with the disease included respiratory disease, elevated morbidity and mortality throughout the house, egg production drops, depression, and lethargy. A unique gross lesion in the commercial layers was a severe, transmural oviduct edema with white to gray flocculent purulent material in the lumen. Layer flocks on two separate premises were quarantined but permitted to remain in the facilities until cessation of virus shed was determined through virus isolation. Several months later, clinical AI appeared again in these flocks. It is not known whether the recurrence of disease in these cases is due to persistence of the organism in the birds or the environment. In addition to serologic testing and virologic testing by chicken embryo inoculation, an antigen capture enzyme immunoassay was evaluated as a diagnostic tool for AI. Research projects related to disinfection, burial pits, and geographical system technology were developed because of questions raised concerning transmission, diagnosis, and control of nonpathogenic AI (H7N2).

An epidemic of avian influenza (AI) (H9N2) occurred in broiler chicken farms in Iran during 1998–01. Mortality between 20% and 60% was commonly observed on the affected farms. Mixed infections of the influenza virus with other respiratory pathogens, particularly infectious bronchitis virus and Mycoplasma gallisepticum, were thought to be responsible for such high mortality, which resulted in great economic losses. Clinical signs included swelling of the periorbital tissues and sinuses, typical respiratory discharge, and severe respiratory distress. Gross lesions included extensive hyperemia of the respiratory system followed by exudation and cast formation in the tracheal biforcation extending into the secondary bronchi. Light microscopy lesions were characterized by severe necrotizing tracheatis. Serological examination using H9N2 AI viral antigen produced inconsistent results. Ultrastructural findings showed typical viral replication through budding processes on cell membranes of the tracheal epithelium.

In 1997, a high-pathogenicity H5N1 avian influenza virus caused serious disease in both man and poultry in Hong Kong, China. Eighteen human cases of disease were recorded, six of which were fatal. This unique virus was eliminated through total depopulation of all poultry markets and chicken farms in December 1997. Other outbreaks of high-pathogenicity avian influenza (HPAI) caused by H5N1 viruses occurred in poultry in 2001 and 2002. These H5N1 viruses isolated had different internal gene constellations to those isolated in 1997. No new cases of infection or disease in man due to these or other H5N1 viruses have been reported. This paper provides an overview and chronology of the events in Hong Kong relating to avian influenza, covering the period from March 1997 to March 2002.

From 1997 to 2001, Italy has been affected by two epidemics of high-pathogenicity avian influenza. The first epidemic was caused by a virus of the H5N2 subtype and was limited to eight premises in backyard and semi-intensive flocks. The prompt identification of the disease was followed by the implementation of European Union (EU) directive 92/40/EEC and resulted in the eradication of infection without serious consequences to the poultry industry. The 1999–00 epidemic was caused by a virus of the H7N1 subtype that originated from the mutation of a low pathogenic virus and resulted instead in a devastating epidemic that affected industrially reared poultry, culminating in the infection of 413 flocks. The description of the epidemics and the result of the control policies are reported.

From the end of March to the beginning of December 1999, an epidemic of low-pathogenicity avian influenza (LPAI), caused by a H7N1 type A influenza virus, affected the intensively reared poultry population of Northeastern Italy. A total of 199 flocks were diagnosed with influenza infection. The highest number affected flocks were in meat turkeys (164), with only a limited number of turkey breeder, chicken (breeders, broilers, and table egg layers), and guinea fowl flocks infected. Following the circulation of the LPAI virus in a susceptible population for several months, a high-pathogenicity avian influenza (HPAI) virus emerged. Over 13 million birds on 413 different premises were affected by the HPAI virus, including turkey, chicken, guinea fowl, pheasant, Japanese quail, ostrich, and waterfowl flocks. In the present paper we report on the clinical, gross, histopathological, and immunohistochemical investigations performed on different avian species naturally infected by the LPAI virus and the HPAI virus.

There is increasing evidence that stable lineages of influenza viruses are being established in chickens. H9N2 viruses are established in chickens in Eurasia, and there are increasing reports of H3N2, H6N1, and H6N2 influenza viruses in chickens both in Asia and North America. Surveillance in a live poultry market in Nanchang, South Central China, reveals that influenza viruses were isolated form 1% of fecal samples taken from healthy poultry over the course of 16 months. The highest isolation rates were from chickens (1.3%) and ducks (1.2%), followed by quail (0.8%), then pigeon (0.5%). H3N6, H9N2, H2N9, and H4N6 viruses were isolated from multiple samples, while single isolates of H1N1, H3N2, and H3N3 viruses were made. Representatives of each virus subtype were experimentally inoculated into both quail and chickens. All the viruses replicated in the trachea of quail, but efficient replication in chickens was confined to 25% of the tested isolates. In quail, these viruses were shed primarily by the aerosol route, raising the possibility that quail may be the “route modulator” that changes the route of transmission of influenza viruses from fecal–oral to aerosol transmission. Thus, quail may play an important role in the natural history of influenza viruses. The pros and cons of the use of inactivated and recombinant fowl pox–influenza vaccines to control the spread of avian influenza are also evaluated.

Using reverse transcription/polymerase chain reaction (RT-PCR), we have screened more than 8500 wild birds in Northern Europe in 1999 and 2000 for the presence of influenza A virus. Although our primary focus was on ducks, geese, and shorebirds, we have also tested thousands of samples from other bird species. Approximately 1% of our samples were positive for influenza A virus by RT-PCR, and from half of these we were able to isolate influenza A virus in embryonated chicken eggs. A wide variety of isolates was obtained representing hemagglutinin (HA) subtypes 1 through 7, 10, 11, 13, an unidentifiable HA, and neuraminidase (NA) subtypes 1 through 8.

The mechanisms of perpetuation of influenza A viruses in aquatic birds, their main reservoir in nature, have not yet been completely clarified. One hypothesis is that they continue to circulate in waterfowl throughout the year, even though virus isolations during the winter months are rare. We analyzed influenza virus circulation in wild ducks in Italy during six winter seasons (1993–99), using virus isolations and serological analyses. It was apparent that influenza A viruses were constantly circulating in wild birds during all the seasons considered. Moreover, seroconversion rates (obtained from ducks recaptured during the same season) suggest a frequency of influenza infections higher than expected on the basis of the virus isolation rates.

Although wild ducks are known to be a major reservoir for avian influenza viruses (AIV), there are few recent published reports of surveillance directed at this group. Predominant AIV hemagglutinin (HA) subtypes reported in previous studies of ducks in North America include H3, H4, and H6, with the H5, H7, and H9 subtypes not well represented in these host populations. The objective of this study was to determine whether these subtype patterns have persisted. Each September from 1998 to 2000, cloacal swabs were collected from wild ducks banded in Roseau and Marshall counties, MN. Mallards (Anas platyrhynchos) were sampled all years, and northern pintails (A. acuta) were sampled only in 1999. Influenza viruses were isolated from 11%, 14%, and 8% of birds during 1998, 1999, and 2000, respectively. Prevalence, as expected, was highest in juveniles, ranging from 11% to 23% in mallards. Viruses representative of the HA subtypes 2, 3, 4, 5, 6, 7, 9, 10, 11, and 12 were isolated. Viruses in the H5, H7, and H9 subtypes, which are associated with high-pathogenicity influenza in poultry or recent infections in humans, were not uncommon, and each of these subtypes was isolated in 2 out of the 3 years of surveillance.

During 2000, 2001, and January 2002, avian influenza virus was isolated from chickens from 12 different locations in California. All the isolates were typed as H6N2 and determined to be of low pathogenicity for chickens. Nine of the isolates came from commercial layer flocks; one from a backyard flock; one from a mixed age flock, where ducks and squabs were also present; and one from a primary broiler breeder. Although a drop in egg production and increased mortality were among the disease signs reported in the layer flocks, the pathological changes observed in the early cases were primarily associated with mild respiratory infections. It was not until August 2001 that yolk peritonitis was observed; this has been a feature of all the remaining cases through 2001 and 2002.

All the isolates clustered as a unique group separate from other influenza viruses based upon sequence data of the H6, neuraminidase (N2), and matrix (MA) genes, indicating a common ancestor for these three gene segments. However, sequencing of the nonstructural (NS) gene indicates introductions from two separate origins. With the first isolate CK/CA/431/00 as the index case, the N2, MA, and NS genes are more closely related to North American isolates, as is the NP gene of CK/CA/650/00. In contrast, the H6 gene is more closely related to a Eurasian influenza isolate. Comparison of amino acid sequences of the N2 and MA genes of these isolates with available type A influenza viruses identified two unique changes in the MA gene and nine in the N2 gene, as well as four progressive changes. These results are discussed in relation to available clinical and epidemiological data.

Avian influenza viruses are major contributors to viral disease in poultry as well as humans. Outbreaks of high-pathogenicity avian influenza viruses cause high mortality in poultry, resulting in significant economic losses. The potential of avian influenza viruses to reassort with human strains resulted in global pandemics in 1957 and 1968, while the introduction of an entirely avian virus into humans claimed several lives in Hong Kong in 1997. Despite considerable research, the mechanisms that determine the pathogenic potential of a virus or its ability to cross the species barrier are poorly understood. Reverse genetics methods, i.e., methods that allow the generation of an influenza virus entirely from cloned cDNAs, have provided us with one means to address these issues. In addition, reverse genetics is an excellent tool for vaccine production and development. This technology should increase our preparedness for future influenza virus outbreaks.

Avian influenza is endemic in wild birds in North America, and the virus routinely has been transmitted from this reservoir to poultry. Influenza, once introduced into poultry, can become endemic within the poultry population. It may be successfully eradicated by human intervention, or the virus may fail to successfully spread on its own. In the last 5 yr, influenza virus has been isolated from poultry in the United States on numerous occasions, and, with the use of molecular epidemiology, the relationships of these different viruses can be determined. There are 15 different hemagglutinin subtypes of avian influenza viruses, but infections with virus of H5 and H7 subtypes are of the most concern because of the potential for these viruses to mutate to the highly pathogenic form of the virus. Most of the influenza isolations in the United States have been associated with the live-bird markets (LBMs) in the Northeast. This has included primarily H7N2 influenza viruses, but also H7N3, H5N2, and other subtypes. Most of the H7N2 viruses were part of a single lineage that was first observed in 1994, but new introductions of H7N2 and H7N3 were also observed. The predominant H7N2 LBM lineage of virus spread to large commercial poultry operations on at least three occasions since 1997, with the largest outbreak occurring in Virginia in 2002. The H5N2 viruses in the LBMs included viruses from domestic ducks, gamebirds, and environmental samples. Some H5N2 viruses isolated in different years and in different locations had a high degree of sequence relatedness, although the reservoir source, if it is endemic, has not been identified. Finally, an H1N2 virus, associated with a drop in egg production, was isolated from turkeys in Missouri in 1999. This virus was a complex reassortant with swine, human, and avian influenza genes that was similar to recent swine isolates from the Midwest. Additional serologic evidence suggests that flocks in other states were infected with a H1N2 virus.

Surveillance for H5 and H7 subtypes of avian influenza virus (AIV) in the live-bird markets (LBMs) of the northeastern United States has been in effect since 1986 when the markets were first recognized as a potential reservoir for AIV. Long-term maintenance of AIV in the LBM system has been documented. However, little is known about the influence of successive cycles of replication in unnatural avian hosts (gallinaceous birds) on the genetics of the virus, especially in the region of the hemagglutinin (HA) gene that can influence pathogenicity. Isolation of low-pathogenicity H5 AIVs from the LBMs has been sporadic; however, in 1994 a low-pathogenicity H7N2 virus was isolated that has persisted in the LBMs for more than 7 yr. Efforts to eliminate the H7 virus from the markets have been unsuccessful. During the 7-yr period, several molecular changes have occurred at the hemagglutinin cleavage site of the H7 virus. These changes include substitutions of proline for threonine and lysine for asparagine, respectively, at the −2 and −5 positions of the HA1 protein. In addition, there has been a 24 nucleotide base-pair deletion in the receptor binding region of the HA1. The accumulation of an additional basic amino acid at the cleavage site is a cause for concern to regulatory authorities, and, therefore, efforts to eliminate the virus from the LBM system have been intensified.

From February 2000 through September 2001, a limited number of H6N2 influenza viruses were isolated from chickens in California. This report describes the genetic characterization of nine of these H6N2 viruses. All of the viruses analyzed had phylogenetically similar hemagglutinin (HA) and neuraminidase molecules that suggested the viruses shared a recent common ancestor. The analysis of the HA sequence of these viruses with all available H6 viruses from different hosts and locations showed that these genes do not separate into well-defined North American and Eurasian lineages. The neuraminidase genes of the California viruses contain an 18 amino acid deletion, a possible adaptation to growth in chickens. Analysis of the remaining gene segments of the California viruses revealed that three distinct genotypes of H6N2 viruses were present.

The H5N1 virus (H5N1/97) that caused the bird flu incident in Hong Kong in 1997 has not been isolated since the poultry slaughter in late 1997. But the donor of its H5 hemagglutinin gene, Goose/Guangdong/1/96-like (Gs/Gd/96-like) virus, established a distinct lineage and continued to circulate in geese in the area. In 2000, a virus from the Goose/Guangdong/1/96 lineage was isolated for the first time from domestic ducks. Subsequently, it has undergone reassortment, and these novel reassortants now appear to have replaced Gs/Gd/96-like viruses from its reservoir in geese and from ducks. The internal gene constellation is also different from H5N1/97, but these variants have the potential for further reassortment events that may allow the interspecies transmission of the virus.

In the late 1990s, H5N1 and H9N2 avian influenza viruses caused respiratory infections in humans in Hong Kong. Exposure to domestic poultry in live-bird markets was significantly associated with human H5N1 disease. Seroepidemiologic studies conducted among contacts of H5N1-infected persons determined that human-to-human transmission of the avian H5N1 viruses occurred but was rare. The relatively high rates of H5 and H9 antibody seroprevalence among Hong Kong poultry workers in 1997 highlight the potential for avian viruses to transmit to humans, particularly those with occupational exposure. Such transmission increases the likelihood of reassortment between a currently circulating human virus and an avian virus and thus the creation of a strain with pandemic potential.

Wild waterfowl that were captured between 1915 and 1919 and preserved in 70% ethyl alcohol were tested for influenza A virus RNA. Most of the HA1 domain of the hemagglutinin (HA) gene segment was sequenced from one bird, captured in 1917, that was infected with a virus of the same HA subtype as the 1918 human pandemic virus. The 1917 HA sequence is closely related to modern avian HA sequences, suggesting little drift in avian sequences in 80 years and that the 1918 pandemic virus probably did not acquire its hemagglutinin directly from a bird. A 151-bp fragment of the nucleoprotein gene segment was sequenced from two pre-1918 birds and compared to avian and mammalian influenza strains. The 1917 avian NP sequences are also closely related to modern avian sequences and distinct from the mammalian clade in which the 1918 NP sequence is found.

Two candidate formalin-inactivated vaccines, made from high-growth reassortant viruses with the HA and NA genes from avian viruses in a background of genes derived from A/Puerto Rico/8/34 (PR8), were prepared against H5N1 and H9N2 subtypes (designated as H5N1/PR8 and H9N2/PR8, respectively). These viruses bear the genotypes, antigenicity, and attenuation in mouse models that are desirable in candidate vaccines. The pathogenicity of the newly generated avian-human reassortant vaccine viruses was also evaluated in chickens. Neither H5N1/PR8 nor H9N2/PR8 were highly pathogenic for chickens. No clinical signs, gross legions, or histological lesions were observed in chickens that were administered H5N1/PR8 either intranasally (i.n.) or intravenously (i.v.), and virus was not detected in oropharyngeal or cloacal swabs. When H9N2/PR8 was administered i.n., no clinical signs, gross lesions, or histological lesions were observed and no virus was detected in cloacal swabs. However, virus was isolated at low titer from oropharyngeal swabs of all eight chickens. Although no clinical signs were observed when H9N2/PR8 was administered i.v., mild tracheitis was seen in one of two chickens. Moderate amounts of antigen were observed in tracheal respiratory epithelium, and low titers of virus were recovered from oropharyngeal and cloacal swabs of some chickens. In summary, both reassortant vaccine viruses replicated poorly in chickens. These studies suggest that these candidate vaccine viruses carry a low risk of transmission to chickens.

Cynomolgus macaques (Macaca fascicularis) infected with influenza virus A/HongKong/156/97 (H5N1) developed acute respiratory distress syndrome (ARDS) with fever. Reverse transcriptase/polymerase chain reaction (RT/PCR) and virus isolation showed that the respiratory tract is the major target of the virus. The main lesion observed upon necropsy, performed 4 or 7 days postinfection, was a necrotizing bronchointerstitial pneumonia, similar to that found in primary influenza pneumonia in human beings. By immunohistochemistry, influenza virus antigen proved to be limited to pulmonary tissue and tonsils. The data indicate that ARDS and multiple organ dysfunction syndrome (MODS), observed in both humans and monkeys infected with this virus, are caused by diffuse alveolar damage from virus replication in the lungs alone.

The World Health Organization (WHO) Global Influenza Program makes annual recommendations on influenza vaccine formulation and related activities. This results in 230 million annual doses of vaccine produced for human use. The success of this program is based on the collection and genetic and antigenic analyses of influenza viruses collected by the WHO Global Influenza Surveillance Program. New programs focus on pandemic preparedness and include development and distribution of testing reagents for emerging or potentially emerging human influenza viruses. WHO Animal Influenza Network focuses on aspects of ecology and molecular biology of animal influenza viruses in the context of human health.

High-pathogenicity avian influenza (HPAI) viruses emerged from low-pathogenicity avian influenza (LPAI) viruses in Pennsylvania (1983–84), Mexico (1994–95), and Italy (1999–2000). Here we focus on the question of why the HPAI virus supersedes the LPAI virus, once it has appeared during the epidemic. To study this, we used an experimental model in chickens that enabled us to estimate the reproduction ratio (R₀). Using this model, we determined the R₀ of the A/Chicken/Pennsylvania/21525/83 (LPAI) and of the A/Chicken/Pennsylvania/1370/83 (HPAI). Comparing the R₀ of both viruses, we concluded that the R₀ of the HPAI virus is significantly higher than the R₀ of the LPAI.

Sequence analysis of the hemagglutinin (HA) gene of H5 and H7 viruses was used to determine phylogenetic relationships between high-pathogenicity avian influenza (HPAI) and low-pathogenicity avian influenza (LPAI) viruses from avian influenza (AI) outbreaks in Norfolk in 1979 and 1991 and Italy in 1999–2000. A common feature within these groups of viruses was the acquisition of additional glycosylation sites near the receptor binding site of the HA. Passage of H5 viruses through 14-day-old embryonated fowls' eggs readily selected viruses with additional glycosylation of HA1. Although additional glycosylation may not correlate with increased pathogenicity for fowl, it may predispose viruses to become highly pathogenic.

The introduction of an influenza A virus possessing a novel hemagglutinin (HA) into an immunologically naive human population has the potential to cause severe disease and death. Such was the case in 1997 in Hong Kong, where H5N1 influenza was transmitted to humans from infected poultry. Because H5N1 viruses are still isolated from domestic poultry in southern China, there needs to be continued surveillance of poultry and characterization of virus subtypes and variants. This study provides molecular characterization and evaluation of pathogenesis of a recent H5N1 virus isolated from duck meat that had been imported to South Korea from China. The HA gene of A/Duck/Anyang/AVL-1/01 (H5N1) isolate was found to be closely related to the Hong Kong/97 H5N1 viruses. This virus also contained multiple basic amino acids adjacent to the cleavage site between HA1 and HA2, characteristic of high-pathogenicity avian influenza viruses (HPAI). The pathogenesis of this virus was characterized in chickens, ducks, and mice. The DK/Anyang/AVL-1/01 isolate replicated well in all species and resulted in 100% and 22% lethality for chickens and mice, respectively. No clinical signs of disease were observed in DK/Anyang/AVL-1/01-inoculated ducks, but high titers of infectious virus could be detected in multiple tissues and oropharyngeal swabs. The presence of an H5N1 influenza virus in ducks bearing a HA gene that is highly similar to those of the pathogenic 1997 human/poultry H5N1 viruses raises the possibility of reintroduction of HPAI to chickens and humans.

Seventeen avian species and two mammalian species were intranasally inoculated with the zoonotic A/chicken/Hong Kong/220/97 (chicken/HK) (H5N1) avian influenza (AI) virus in order to ascertain a relative range of susceptible hosts and the pathobiology of the resultant disease. A direct association was demonstrated between viral replication and the severity of disease, with four general gradations being observed among these species. These gradations included the following: 1) widespread dissemination with rapid and high mortality, 2) neurological disease relative to viral neurotropism, 3) asymptomatic infection or only mild transient depression associated with minor viral replication, and 4) absence of disease relative to minimal to no viral replication. This investigation not only demonstrates that the chicken/HK virus could infect multiple avian species, but also that the virulence of the chicken/HK virus varied significantly among avian species, including those species that are members of the same order.

Apoptosis is essential in many physiological processes including wound healing and development of the immune response. Apoptosis also plays an important role in the pathogenesis of many infectious diseases including those caused by viruses. Influenza viruses induce apoptosis in cells that are permissive for viral replication and cells that do not support viral replication. The cellular pathways involved in influenza virus induced apoptosis are currently ill defined. Previous studies suggest that influenza virus infection increased the expression of the Fas antigen in HeLa cells, and that Fas antigen is partially involved in apoptosis. In these studies we examined the cellular pathways involved in avian influenza virus induced apoptosis in two cell lines that support productive viral replication: Madin–Darby canine kidney cells (MDCK) and mink lung epithelial (Mv1Lu) cells.

The Office International des Epizooties (OIE) has developed international standards to reduce the risk of the spread of high-pathogenicity avian influenza though international trade. These standards include providing a definition of high-pathogenicity avian influenza (HPAI), procedures for prompt reporting of HPAI outbreaks, requirements that must be met for a country or zone to be defined as free of HPAI, requirements that should be met to import live birds and avian products into a HPAI-free country or zone, and the general provisions that countries should meet to reduce the risk of spread of HPAI through trade. The goal of these standards is to facilitate trade while minimizing the risk of the introduction of HPAI.

The current definitions of high-pathogenicity avian influenza (HPAI), formulated over 10 years ago, were aimed at including viruses that were overtly virulent in in vivo tests and those that had the potential to become virulent. At that time the only virus known to have mutated to virulence was the one responsible for the 1983–84 Pennsylvania epizootic. The mechanism involved has not been seen in other viruses, but the definition set a precedent for statutory control of potentially pathogenic as well as overtly virulent viruses.

The accumulating evidence is that HPAI viruses arise from low-pathogenicity avian influenza (LPAI) H5 or H7 viruses infecting chickens and turkeys after spread from free-living birds. At present it can only be assumed that all H5 and H7 viruses have this potential and mutation to virulence is a random event. Therefore, the longer the presence and greater the spread in poultry the more likely it is that HPAI virus will emerge. The outbreaks in Pennsylvania, Mexico, and Italy are demonstrations of the consequences of failing to control the spread of LPAI viruses of H5 and H7 subtypes. It therefore seems desirable to control LPAI viruses of H5 and H7 subtype in poultry to limit the probability of a mutation to HPAI occurring. This in turn may require redefining statutory AI. There appear to be three options: 1) retain the current definition with a recommendation that countries impose restrictions to limit the spread of LPAI of H5 and H7 subtypes; 2) define statutory AI as an infection of birds/poultry with any AI virus of H5 or H7 subtype; 3) define statutory AI as any infection with AI virus of H5 or H7 subtype, but modify the control measures imposed for different categories of virus and/or different types of host.

During the past decade, several examples of the ability of H5 and H7 low-pathogenicity avian influenza (LPAI) viruses to mutate to high-pathogenicity (HP) viruses have been documented worldwide. During this time, the introduction and persistence of an H7N2 LPAI virus in the northeast live-bird marketing system in the United States has raised concern on how to prevent the possibility of such a mutation occurring in this country. The United States has periodically experienced trade restrictions based on the occasional introduction of H5 and H7 LPAI viruses into commercial poultry and based on AI-related changes in the import requirements for poultry and poultry products of several of our trading partners. Consequently, the U.S. Department of Agriculture (USDA) is exploring options for how our regulatory response to H5 and H7 LPAI viruses might be revised to better protect our domestic poultry flocks from HPAI and to ensure that any interruptions in trade are scientifically supportable. The options under consideration include mandatory and voluntary measures to improve the surveillance for and control of H5 and H7 LPAI virus infections.

New Zealand has never experienced an outbreak of avian influenza, and the Ministry of Agriculture and Forestry has long been wary of the possibility of introducing high-pathogenicity avian influenza (HPAI) viruses in imported goods. Besides the potential threat posed to poultry, there are concerns that introduced viruses might have negative effects on already endangered native avian species. Under the framework of the World Trade Organization, the sanitary and phytosanitary (SPS) agreement requires member countries to base their sanitary measures for imported animal products on the Office International des Epizooties (OIE) standard or on a scientific assessment of risk. This paper presents the New Zealand experience with assessing the risk of avian influenza viruses in imported chicken meat and considers how the assessment of risk has changed in recent years as a result of the advances in understanding of the disease. The currently accepted view that low-pathogenicity avian influenza (LPAI) viruses are widespread and that they mutate to virulence after introduction into poultry has important implications concerning the appropriate definition for avian influenza viruses of regulatory concern and has possible implications concerning the significance of viruses present in this country.

In 2001, all 109 retail live-bird markets (LBMs) in New York and New Jersey were surveyed for the presence of avian influenza virus (AIV) by a real time reverse transcriptase/polymer chain reaction assay (RRT/PCR) and results compared to virus isolation (VI) in embryonating chicken eggs. The RRT/PCR had a 91.9% sensitivity and 97.9% specificity in detecting presence of AIV at the market level. However, the sensitivity at the sample level is 65.87%. The RRT/PCR is a reliable method to identify AIV at the market level. In addition, a cross-sectional epidemiologic study of the LBMs showed that, during the past 12 months, markets that were open 7 days per week and those that also sold rabbits had the highest risk for being positive for AIV. Markets that were closed one or more days per week and those that performed daily cleaning and disinfecting had the lowest risk for being AIV positive.

The avian influenza high-pathogenicity virus was eradicated in poultry of Mexico in a relatively short period by the use of inactivated emulsified vaccine, enforcing biosecurity, and controlling movement of poultry and poultry products. Mexico maintains a permanent and reliable monitoring program for AI. H5N2 is the only avian influenza subtype identified. It is possible to control and eradicate the avian influenza low-pathogenicity virus mainly by controlled depopulation of positive poultry, reinforcing biosecurity, and the use of vaccines.

In 1999–2000, Italy was affected by the most severe avian influenza (AI) epidemic that has ever occurred in Europe. The epidemic was caused by a type A influenza virus of the H7N1 subtype, which originated from the mutation of a low-pathogenicity (LP) AI virus of the same subtype. From August to November 2000, 4 months after the eradication of the highly pathogenic (HP) AI virus, the LPAI strain re-emerged and infected 55 poultry farms mainly located in the southern area of Verona province (Veneto region). To supplement disease control measures already in force, an emergency vaccination program against the disease was implemented in the area. Vaccination was carried out using an inactivated heterologous vaccine (A/chicken/Pakistan/1995-H7N3). In order to establish whether LPAI infection was circulating in the area, regular serological testing of sentinel birds in vaccinated flocks and a discriminatory test able to distinguish the different types of antineuraminidase antibodies (anti-N1 and anti-N3) were performed. Shortly after the beginning of the vaccination campaign (December 2000 to March 2001), the H7N1 LPAI virus emerged again, infecting 23 farms. Among these, only one vaccinated flock was affected, and infection did not spread further to other vaccinated farms. The data reported in the present paper indicate that the combination of biosecurity measures, official control, and vaccination can be considered successful for the control of LPAI infections in densely populated poultry areas.

A Geographic Information System (GIS) is a very powerful and flexible software tool for effective management of spatially referenced data (e.g., geodata). Coupling database and GIS technology provides the tools for a detailed analysis of spatial patterns and distributions in veterinary applications. A specific veterinary GIS (VetGIS) toolbox was developed to perform the calculation of indices such as Lorenz curve, GINI index, and a kernel-based animal density estimation. This software was employed for the analysis and management of avian influenza in Italy during the 1999–2000 epidemic.

The survival or clearance of the avian influenza virus (AIV) of subtype H7N2 in its chicken host was evaluated using experimentally infected specific pathogen free (SPF) chickens of different age groups. Birds of different ages were successfully infected with infectious doses ranging between 10^4.7 and 10^5.7 ELD₅₀ per bird. In infected birds, the infective virus was undetectable usually by the third week following exposure. The infectivity or inactivation time of the H7N2 AIV in various environmental conditions was studied using chicken manure, heat, ethanol, pH, and disinfectants. The H7N2 AIV was effectively inactivated by field chicken manure in less than a week at an ambient temperature of 15–20°C. At a pH 2, heating at 56°C, and exposure to 70% ethanol or a specific disinfectant, the AIV infectivity was destroyed in less than 30 min.

An outbreak of H7N2 low-pathogenicity (LP) avian influenza (AI) occurred in a two-county area in Pennsylvania from December of 1996 through April of 1998. The outbreak resulted in infection of 2,623,116 commercial birds on 25 premises encompassing 47 flocks. Twenty-one (one premise with infection twice) of the twenty-five infected premises housed egg-laying chickens and one premise each had turkeys, layer pullets, quail, and a mixed backyard dealer flock. Despite close proximity of infected flocks to commercial broiler flocks, no infected broilers were identified. Experimentally, when market age broilers were placed on an influenza-infected premise they seroconverted and developed oviduct lesions. The outbreak was believed to have originated from two separate introductions into commercial layer flocks from premises and by individuals dealing in sales of live fowl in the metropolitan New York and New Jersey live-bird markets. Source flocks for these markets are primarily in the northeast and mid-Atlantic areas, including Pennsylvania. Mixed fowl sold include ducks, geese, guinea hens, quail, chukar partridges, and a variety of chickens grown on perhaps hundreds of small farms. Infections with the H7N2 AI virus were associated with variable morbidity and temporary decreases in egg production ranging from 1.6% to 29.1% in commercial egg-laying chickens. Egg production losses averaged 4.0 weeks duration. Mortality ranged from 1.5 to 18.3 times normal (mean of 4.3 times normal). Duration of mortality ranged from 2 to 13 weeks (average of 3.9 weeks) in flocks not depopulated. Lesions observed were primarily oviducts filled with a mucous and white gelatinous exudates and atypical egg yolk peritonitis. Quarantine of premises and complete depopulation were the early measures employed in control of this outbreak. Epidemiological studies suggested that depopulation furthered the spread of influenza to nearby flocks. Thereafter, later control measures included quarantine, strict biosecurity, and controlled marketing of products.

Retail live poultry markets (LPMs) may act as a reservoir of avian influenza viruses (AIV). In this study we test the hypothesis that a rest day in the LPMs where the stalls are completely emptied of poultry, cleansed, and restocked will reduce the isolation rates of avian influenza viruses. The isolation rate of H9N2 subtype viruses from chicken was significantly lower after the rest day than prior to it, indicating its impact in reducing transmission. In contrast, Newcastle disease virus (NDV) isolation rates appear unaffected by this intervention, possibly reflecting differences in herd immunity or virus transmission dynamics.

The minimum requirements for assessing the immunogenicity of an experimental avian influenza (AI) vaccine prepared from inactivated A/Turkey/Italy/2676/99 (H7N1) low-pathogenicity (LP) AI (LPAI) virus were determined in chickens of different ages. A correlation between the amount of hemagglutinin (HA) per dose of vaccine and the protection against clinical signs of disease and infection by A/Chicken/Italy/13474/99 highly pathogenic (HP) AI (HPAI) virus was established. Depending on the vaccination schedule, one or two administrations of 0.5 μg of hemagglutinin protected chickens against clinical signs and death and completely prevented virus shedding from birds challenged at different times after vaccination.

Current vaccines to prevent avian influenza rely upon labor-intensive parenteral injection. A more advantageous vaccine would be capable of administration by mass immunization methods such as spray or water vaccination. A recombinant vaccine (rNDV-AIV-H7) was constructed by using a lentogenic paramyxovirus type 1 vector (Newcastle disease virus [NDV] B1 strain) with insertion of the hemagglutinin (HA) gene from avian influenza virus (AIV) A/chicken/NY/13142-5/94 (H7N2). The recombinant virus had stable insertion and expression of the H7 AIV HA gene as evident by detection of HA expression via immunofluorescence in infected Vero cells. The rNDV-AIV-H7 replicated in 9–10 day embryonating chicken eggs and exhibited hemagglutinating activity from both NDV and AI proteins that was inhibited by antisera against both NDV and AIV H7. Groups of 2-week-old white Leghorn chickens were vaccinated with transfectant NDV vector (tNDV), rNDV-AIV-H7, or sterile allantoic fluid and were challenged 2 weeks later with viscerotropic velogenic NDV (vvNDV) or highly pathogenic (HP) AIV. The sham-vaccinated birds were not protected from vvNDV or HP AIV challenge. The transfectant NDV vaccine provided 70% protection for NDV challenge but did not protect against AIV challenge. The rNDV-AIV-H7 vaccine provided partial protection (40%) from vvNDV and HP AIV challenge. The serologic response was examined in chickens that received one or two immunizations of the rNDV-AIV-H7 vaccine. Based on hemagglutination inhibition and enzyme-linked immunosorbent assay (ELISA) tests, chickens that received a vaccine boost seroconverted to AIV H7, but the serologic response was weak in birds that received only one vaccination. This demonstrates the potential for NDV for use as a vaccine vector in expressing AIV proteins.

Previously, we have shown that intramuscular vaccination of chickens with the eukaryotic expression vector (EEV), expressing the influenza H5 hemagglutinin (H) protein, can stimulate a measurable and protective antibody response. Based on these results, we cloned other H genes from Eurasian H5, North American and Eurasian H7, and H15 influenza viruses into the EEV for use in vaccination of chickens to produce reference antibodies for diagnostic purposes, such as the hemagglutination inhibition (HI) test. Three-week-old specific pathogen free (SPF) chickens were vaccinated with 100 μg of EEV mixed with a cationic lipid by intramuscular injection. Then the birds were boostered twice at monthly intervals after the original vaccination. Measurable antibody titers were present for most birds after 1 month and generally increased after each boost. To examine the cross reactivity of the sera with other subtypes, HI test was conducted with antigens prepared from 15 subtypes of influenza virus. Subtype specificity of the antisera prepared by DNA vaccination were comparable or better than the antisera prepared by traditional method using whole virus vaccination. Preparation of reference antisera by DNA vaccination holds good promise because it is safe and allows for the production of H specific antibodies without producing antibodies specific to other influenza viral proteins.

Using a monoclonal antibody (MAb) specific for the H7 influenza surface glycoproteins, a serological enzyme-linked immunosorbent assay (ELISA) test has been developed. This MAb was made using the low-pathogenicity (LP) avian influenza (AI) strain (BS2676/99) isolated in Italy during a recent outbreak. The test is able to detect H7 antibodies in avian sera. The H7 ELISA has a 99% concordance of results with the classical hemagglutination inhibition (HI) test.

The development of a discriminatory test, based on the differentiation between N1 and N3 antibodies, to be used in the framework of a vaccination program, based on vaccination with a heterologous H7N3 inactivated vaccine against the Italian H7N1 field virus, is reported. The indirect immunofluorescence antibody (iIFA) assay was based on the expression of the N1 protein in a baculovirus system. HighFive® insect cells were transfected with the recombinant virus and used as an antigen in the iIFA test. Preliminary validation on 608 turkey sera yielded relative sensitivity and specificity of 98.1% and 95.7%, respectively, when compared to the HI test with an almost perfect agreement between the two methods (Kappa value = 0.93). It is concluded that the iIFA test is a valid tool for monitoring avian influenza infection in a vaccinated population.

As of October 2001, the potential for use of infectious agents, such as anthrax, as weapons has been firmly established. It has been suggested that attacks on a nations' agriculture might be a preferred form of terrorism or economic disruption that would not have the attendant stigma of infecting and causing disease in humans. Highly pathogenic avian influenza virus is on every top ten list available for potential agricultural bioweapon agents, generally following foot and mouth disease virus and Newcastle disease virus at or near the top of the list. Rapid detection techniques for bioweapon agents are a critical need for the first-responder community, on a par with vaccine and antiviral development in preventing spread of disease. There are several current approaches for rapid, early responder detection of biological agents including influenza A viruses. There are also several proposed novel approaches in development. The most promising existing approach is real-time fluorescent PCR analysis in a portable format using exquisitely sensitive and specific primers and probes. The potential for reliable and rapid early-responder detection approaches are described, as well as the most promising platforms for using real-time PCR for avian influenza, as well as other potential bioweapon agents.

Nucleic acid sequence-based amplification (NASBA) allows the rapid amplification of specific regions of nucleic acid obtained from a diverse range of sources. It is especially suitable for amplifying RNA sequences. A NASBA technique was developed that allows the detection of avian influenza A subtype H5 from allantoic fluid harvested from inoculated chick embryos. The amplified viral RNA is detected by electrochemiluminescence. The described NASBA technique is a specific, rapid, and sensitive method of detection of influenza A subtype H5 viruses. More importantly, it can be used to distinguish high- and low-pathogenicity strains of the H5 subtype.

A one-tube reverse transcriptase/polymerase chain reaction coupled with an enzyme-linked immunosorbent assay (RT-PCR-ELISA) was developed for the rapid detection of avian influenza virus (AIV) in clinical specimens. A total of 419 swab pools were analyzed from chickens experimentally infected with low-pathogenicity AIV, from wild aquatic birds, and from domestic ducks. The AIV was detected in 32 swab pools by RT-PCR-ELISA compared to 23 by virus isolation (VI) in embryonated specific pathogen free (SPF) chicken eggs. Thus, 39% more specimens were positive by RT-PCR-ELISA than by VI. Two of the twenty-three VI-positive specimens were negative when tested by RT-PCR-ELISA. The diagnostic sensitivity and specificity of the RT-PCR-ELISA was 91% and 97%, respectively, using VI in SPF eggs as the gold reference standard.

A real-time reverse transcriptase/polymerase chain reaction (RRT-PCR) assay was developed using hydrolysis probes for the detection of avian influenza virus (AIV) and the H5 and H7 subtypes. The AIV specific primers and probes were directed to regions of the AIV matrix gene that are conserved among most type A influenza viruses. The H5 and H7 primers and probes are directed to H5 and H7 hemagglutinin gene regions that are conserved among North American avian influenza viruses. The sensitivity and specificity of this RRT-PCR assay was compared to virus isolation (VI) in chicken embryos with 1550 clinical swab samples from 109 live-bird markets (LBMs) in New York and New Jersey. RRT-PCR detected influenza in samples from 61 of 65 (93.8%) of the LBMs that were the sources of VI positive samples. Of the 58 markets that were positive for H7 influenza by hemagglutination inhibition assay, RRT-PCR detected H7 influenza in 56 markets (96.5%). Too few H5 positive samples were obtained to validate the H5 RRT-PCR assay in this study. Although RRT-PCR was less sensitive than VI on an individual sample basis, this study demonstrated that the AIV and H7 RRT-PCR assays are good tools for the rapid screening of flocks and LBMs.

An outbreak of highly pathogenic avian influenza caused by multiple genotypes of H5N1 virus occurred in Hong Kong, commencing in January 2002. Infection in local chicken farms was preceded by the detection of virus in multiple retail markets and the main poultry wholesale market. The first case of this disease on a local farm was detected on February 1, 2002. By February 9, 2002, 15 farms were infected, and by late March a total of 22 infected farms had been identified. Three main clusters of infected farms were seen, suggesting multiple incursions of virus, and subsequent limited lateral spread to neighboring farms. Control of this disease has been effected through a combination of quarantine, tightening of biosecurity measures, and depopulation of infected and contact farms. About 950,000 birds have been destroyed. Vaccination using a killed H5 vaccine was introduced in April 2002 to farms in one zone where infection has persisted. None of the viruses isolated contained the internal genes found in the 1997 H5N1 virus.

A multiplex real-time reverse transcriptase-polymerase chain reaction (RRT-PCR) assay for the simultaneous detection of the H5 and H7 avian influenza hemagglutinin (HA) subtypes was developed with hydrolysis type probes labeled with the FAM (H5 probe) and ROX (H7 probe) reporter dyes. The sensitivity of the H5-H7 subtyping assay was determined, using in vitro transcribed RNA templates, to have a reproducible detection limit for H7 of approximately 10⁴ HA gene copies and approximately 10⁴–10⁵ HA gene copies of H5. A direct comparison of H5-H7 multiplex RRT-PCR with hemagglutination inhibition (HI) was performed with 83 AI RRT-PCR and virus isolation positive tracheal and cloacal swab samples obtained from various avian species and environmental swabs from live-bird markets in New York and New Jersey. Both multiplex RRT-PCR and HI agreed on the subtype determination of 79 (95.2%) of the 83 samples, of which 77 were positive for H7 and two were determined to be non-H5/non-H7 subtypes. No samples were determined to be the H5 subtype by either assay.

An avian influenza (AI) real time reverse transcriptase-polymerase chain reaction (RRT-PCR) test was previously shown to be a rapid and sensitive method to identify AI virus-infected birds in live-bird markets (LBMs). The test can also be used to identify avian influenza virus (AIV) from environmental samples. Consequently, the use of RRT-PCR was being considered as a component of the influenza eradication program in the LBMs to assure that each market was properly cleaned and disinfected before allowing the markets to be restocked. However, the RRT-PCR test cannot differentiate between live and inactivated virus, particularly in environmental samples where the RRT-PCR test potentially could amplify virus that had been inactivated by commonly used disinfectants, resulting in a false positive test result. To determine whether this is a valid concern, a study was conducted in three New Jersey LBMs that were previously shown to be positive for the H7N2 AIV. Environmental samples were collected from all three markets following thorough cleaning and disinfection with a phenolic disinfectant. Influenza virus RNA was detected in at least one environmental sample from two of the three markets when tested by RRT-PCR; however, all samples were negative by virus isolation using the standard egg inoculation procedure. As a result of these findings, laboratory experiments were designed to evaluate several commonly used disinfectants for their ability to inactivate influenza as well as disrupt the RNA so that it could not be detected by the RRT-PCR test. Five disinfectants were tested: phenolic disinfectants (Tek-trol and one-stroke environ), a quaternary ammonia compound (Lysol no-rinse sanitizer), a peroxygen compound (Virkon-S), and sodium hypochlorite (household bleach). All five disinfectants were effective at inactivating AIV at the recommended concentrations, but AIV RNA in samples inactivated with phenolic and quaternary ammonia compounds could still be detected by RRT-PCR. The peroxygen and chlorine compounds were effective at some concentrations for both inactivating virus and preventing amplification by RRT-PCR. Therefore, the RRT-PCR test can potentially be used to assure proper cleaning and disinfection when certain disinfectants are used.

Over the last 10 years, low-pathogenicity avian influenza (LPAI) viruses have been isolated from the live-bird markets (LBMs) of the Northeast. Despite educational efforts, surveillance, and increased state regulatory efforts, the number of positive markets has persisted and increased. In an effort to address the continued levels of LPAI in the retail LBM and address the question of persistence and circulation of the virus within the LBM system itself, these markets were closed for a continuous 3-day period. This effort was a cooperative effort between the State Departments of Agriculture and coordinated by the U.S. Department of Agriculture and led to the first successful system-wide closure of the retail LBMs in the Northeast.

An outbreak of low-pathogenicity H7N2 avian influenza virus (AIV) in the Shenandoah Valley of Virginia during the spring and summer of 2002 affected 197 farms and resulted in the destruction of over 4.7 million birds. The outbreak affected primarily turkey farms (28 breeders, 125 grow out) with some spillover into chicken farms (29 breeders, 13 grow out, 2 table-egg layers). Although no direct link was established, the strain of H7N2 AIV in this outbreak had a molecular fingerprint that was essentially identical to the H7N2 AIV strain that has circulated in the live bird markets of the northeastern United States for the last 8 yr. After an initial delay caused by lack of viable disposal options, depopulation and disposal, primarily in sanitary landfills, was carried out within 24 hr of detection of a positive flock. Increased surveillance efforts included once-a-week testing of the daily mortality of all poultry farms in the region, testing of all breeder farms every 2 wk, and testing of all flocks prior to movement for any reason. A statistical sampling of backyard flocks and wild birds found no evidence of the virus. The successful eradication of this outbreak was the result of the efforts of a highly effective task force of industry, state, and federal personnel.

A hemagglutinating virus was isolated from a dead turkey in a small mixed free-range flock in Southern Germany. It was identified as influenza virus type A of subtype H7N7. The pathogenicity was low. An intravenous pathogenicity index of 0.03 was recorded, and the nucleotide sequencing revealed the amino acid sequence NVPEIPKGR*GLFG at the cleavage site of the hemagglutinin. Antibodies as well as virus were detected in the affected flock. Further virus spreading to other flocks was prevented by stamping out policy. Serological monitoring of contact flocks revealed one small backyard flock of 18 hens, which was positive. This flock was also destroyed. The origin of the virus could not be identified.

Virus surveillance in free-flying, nonmigratory ducks living on the eastern shore of Maryland indicated that influenza A viruses were introduced into the area or that the prevalence of endemic infections increased between July 15 and August 27, 1998. Cloacal swabs collected between May 28 and July 15, 1998, were negative for influenza A virus recovery (0/233), whereas 13.9% (29/209) of swabs collected between August 27 and September 2, 1998, were positive for influenza A virus recovery. Five hemagglutinin subtypes (H2, H3, H6, H9, and H12), six neuraminidase subtypes (N1, N2, N4, N5, N6, and N8), and nine HA-NA combinations were identified among 29 influenza A isolates. Interestingly, 18 of the 29 isolates initially appeared to contain two or more HA and/or NA subtypes. The free-flying, nonmigratory ducks served as excellent sentinels for the early detection of type A influenza viruses in the southern half of the Atlantic Migratory Waterfowl Flyway during the earliest phase of the yearly southern migration.

Low-pathogenic avian influenza virus (AIV) continues to be isolated from the live bird markets (LBMs) in the Northeastern United States. Recent years have seen increasing numbers of these markets opening and an expansion of the type of animals they sell in conjunction with traditional live poultry. Specific-pathogen-free chickens were released into the livestock area of 13 New York City LBMs and then tested for evidence of AIV. We were able to recover virus or demonstrate seroconversion among the chickens introduced to four of the markets.

Chickens, quail, and other land-based birds are extensively farmed around the world. They have been recently implicated in zoonotic outbreaks of avian influenza in Hong Kong. The possibility that land-based birds could act as mixing vessels or disseminators of avian/mammalian reassortant influenza A viruses with pandemic potential has not been evaluated. In this report, we investigated whether chickens and Japanese quail are susceptible to a mammalian influenza virus (A/swine/Texas/4199-2/98 [H3N2]). This virus did not grow in chickens and replicated to low levels in Japanese quail but did not transmit. Replacing the H3 gene of this virus for one of the avian H9 viruses resulted in transmission of the avian/swine reassortant virus among quail but not among chickens. Our findings demonstrated that Japanese quail could provide an environment in which viruses like the A/swine/Texas/4199-2/98 [H3N2] virus could further reassort and generate influenza viruses with pandemic potential.

The RNA of the hemagglutinin (HA) gene of A/Chicken/Guangdong/SS/1994 (H9N2) was reverse transcription–polymerase chain reaction amplified, and the cDNA was cloned into a plasmid vector. The complete coding sequence of the HA gene was sequenced and included 1683 nucleotides, which encoded for a protein of 560 amino acids. The potential glycosylation sites related to HA protein function were highly conserved. The amino acid sequence of the HA proteolytic cleavage was G-S-S-R/G. This cleavage site sequence is compatible with a low-pathogenic avian influenza virus. Sequence comparison of this HA gene with other H9 influenza virus sequences in the GenBank database showed a 82%–97% nucleotide and amino acid sequence similarity.

The H5N1 viruses isolated from humans in Hong Kong directly infected both mice and ferrets without prior adaptation to either host. Two representative viruses, A/Hong Kong/483/97 (HK/483) and A/Hong Kong/486/97 (HK/486) were equally virulent in outbred ferrets but differed in their virulence in inbred mice. Both HK/483 and HK/486 replicated systemically in ferrets and showed neurologic manifestations. In contrast, intranasal infection of mice with HK/483, but not HK/486, resulted in viral spread to the brain, neurologic signs, and death. However, HK/486 was able to replicate in the brain and induce lethal disease following direct intracerebral inoculation.

H9N2 subtype avian influenza viruses have been identified in avian species worldwide, and infections in pigs were confirmed in Hong Kong in 1998. Subsequently, H9N2 viruses were isolated from two children in Hong Kong in 1999, and five human infections were reported from China, raising the possibility that H9N2 viruses pose a potential pandemic threat for humans. These events prompted us to develop a vaccine candidate to protect humans against this subtype of influenza A viruses. Reassortant H1N1 and H3N2 human influenza A viruses with the six internal gene segments of A/Ann Arbor/6/60 (H2N2)(AA) cold-adapted (ca) virus have been tested extensively in humans and have proved to be attenuated and safe as live virus vaccines. Using classical genetic reassortment, we generated a reassortant that contains the hemagglutinin and neuraminidase genes from A/chicken/Hong Kong/G9/97 (H9N2) and six internal gene segments from the AAca virus. The G9/AAca reassortant virus exhibits the ca phenotype and the temperature-sensitive phenotypes of the AAca virus and was attenuated in mice. The reassortant virus was immunogenic and protected mice from wild-type H9N2 virus challenge. The G9/AAca virus bears the in vitro and in vivo phenotypes specified by the AAca virus and will be evaluated as a potential vaccine candidate in humans.

Avian-like H5N1 influenza viruses isolated from humans in 1997 were shown to have two distinct pathogenic phenotypes in BALB/c mice, after intranasal inoculation and without prior adaptation to this host. To further understand the mechanisms of H5N1 pathogenicity, we investigated the consequences of the route of viral inoculation on morbidity and mortality, viral replication in pulmonary and systemic organs, and lymphocyte depletion. This study demonstrates the importance of extrapulmonary spread and replication, particularly in the brain, for the lethality of H5N1 viruses.

The outbreak of avian influenza H5N1 in Hong Kong in 1997 raised concerns about the potential for the H5 subtype to cause a human pandemic. In 2001 a new H5N1 virus, A/Duck Meat/Anyang/AVL-1/2001 (A/Dkmt), was isolated from imported duck meat in Korea. The pathogenesis of this virus was investigated in mice. A/Dkmt virus had low infectivity but was lethal for mice at high doses, and at lethal doses, the virus replicated in the brains of infected mice. A/Dkmt virus cross-reacted poorly with ferret antisera raised against human H5N1 viruses, but prior infection with A/Dkmt virus protected mice from death after secondary infection with human H5N1 virus.

Zanamivir has been shown to inhibit both human and avian influenza viral neuraminidases (NAs) and has been approved in several countries for the treatment and prophylaxis of influenza infection. Reliable monitoring of drug resistance is important for assessment of the impact of drug therapy on circulating virus populations. This study compares the current fluorometric (FL) method for evaluating zanamivir susceptibility with a recently developed chemiluminescent (CL) NA activity assay using viruses representative of all nine NA subtypes. The CL assay displayed signal/noise ratios that are 50–100 times greater than those associated with the FL assay. Human H3N2 strains appeared to exhibit greater NA activity relative to avian subtypes with the FL substrate but not with the CL substrate. Additionally, the CL assay remained linear over three orders of magnitude compared to only one order of magnitude for the FL assay. Four of the nine NA subtypes tested in this study displayed slightly higher inhibitor concentration that inhibits 50% of neuraminidase activity values by CL than by FL, while four displayed the opposite effect. Implications for the routine determination of resistance to NA inhibitors are discussed.

An immunohistochemical investigation was performed to assess tissue tropism and viral replication of Italian H7N1 isolates belonging to different lineages in developing chicken, turkey, Muscovy duck, and mallard duck embryos. Low-pathogenic avian influenza (LPAI) isolates were selected on the basis of the location in the phylogenetic tree; a progenitor strain, A/turkey/Italy/977/V99 (exhibiting no additional glycosylation sites, nAGS), strain A/turkey/Italy/2379/V99 (AGS in position 123), and strain A/turkey/Italy/3675/V99 (AGS in position 149) were selected. The latter two strains belonged to distinct lineages originating from the pool of progenitor strains. The highly pathogenic avian influenza (HPAI) isolate A/turkey/Italy/4580/V99 was also included in the test. All the embryos tested supported the growth of HPAI. The LPAI isolates replicated readily in the allantoic layer of the chorioallantoic membrane of all the species tested and did not replicate to detectable levels in the developing chicken, turkey, and Muscovy duck embryos. In contrast, they replicated to different extents in the respiratory tract of the developing mallard embryo. The findings indicate that the pathogenesis of LPAI infections in mallard embryos is different to that observed in other species and should be investigated further.

Infections of ostriches with avian influenza A viruses are generally associated with clinical disease, but the occasional high mortality in young birds does not appear to be related directly to virus pathotype. In this study we investigated the pathogenesis of two H7 viruses for 11-wk-old ostriches inoculated intranasally, and clinical symptoms, virus excretion, and immune response were studied. One of the viruses (A/Ostrich/Italy/1038/00) was highly pathogenic for chickens, whereas the other (A/Ostrich/South Africa/1609/91) was of low pathogenicity for chickens. Clinical signs in ostriches receiving virulent virus were slight depression and hemorrhagic diarrhea, while the group receiving avirulent virus was clinically normal except for green diarrhea. Both viruses were transmitted to in-contact sentinel birds housed with the infected groups 3 days postinfection. Postmortem examination of the birds infected (including the sentinel bird) with virus highly pathogenic for chickens were grossly normal except for localized pneumonic lesions. The results of the study are presented and discussed.

To study whether influenza virus receptors in chickens differ from those in other species, we compared the binding of lectins and influenza viruses with known receptor specificity to cell membranes and gangliosides from epithelial tissues of ducks, chickens, and African green monkeys. We found that chicken cells contained Neu5Acα(2-6)Gal–terminated receptors recognized by Sambucus nigra lectin and by human viruses. This finding explains how some recent H9N2 viruses replicate in chickens despite their human virus-like receptor specificity. Duck virus bound to gangliosides with short sugar chains that were abundant in duck intestine. Human and chicken viruses did not bind to these gangliosides and bound more strongly than duck virus to gangliosides with long sugar chains that were found in chicken intestinal and monkey lung tissues. Chicken and duck viruses also differed by their ability to recognize the structure of the third sugar moiety in Sia2-3Gal–terminated receptors. Chicken viruses preferentially bound to Neu5Acα(2-3)Galβ(1-4)GlcNAc–containing synthetic sialylglycopolymer, whereas duck viruses displayed a higher affinity for Neu5Acα(2-3)Galβ(1-3)GalNAc–containing polymer. Our data indicate that sialyloligosaccharide receptors in different avian species are not identical and provide a potential explanation for the differences between the hemagglutinin and neuraminidase proteins of duck and chicken viruses.

The morphology of plaques induced by Italian, H7N1, low-pathogenic avian influenza (LPAI) viruses belonging to different lineages was investigated in primary chicken, turkey, Muscovy duck, and mallard duck kidney cells and in MDCK cells in the absence of trypsin. LPAI isolates were selected on the basis of the location in the phylogenetic tree: 977/V99 (located at the root, no additional glycosylation site [nAGS]), 2379/V99 (AGS in position 123), and 3675/V99 (AGS in position 149). Different isolates did not induce plaques with a statistically significant different size in MDCK cells. However, in primary cells of different avian origin, the presence or absence of AGS significantly influenced plaque size. Generally speaking, 977/V99 was the least efficient at plaquing in all cells, while 2379/V99 (AGS in position 123) plaqued more efficiently in turkey cells and 3675/V99 (AGS in position 149) in chicken cells. The presence of either AGS induced statistically significant larger plaques in cells of waterfowl origin.

A comparative study of the hemagglutinin (HA) receptor binding site (RBS) of a number of H13 influenza viruses isolated from Laridae family of birds (gulls) and other influenza viruses obtained from the Anatidae family (ducks) was conducted. The affinity of all viruses to alpha N-acetylneuraminic acid (Neu5Acα), 3′sialyllactose (3′SL), and sialylglycopolymers bearing 3′-sialyl(N-acetyllactosamine) (3′SLN-PAA), [Neu5Acα(2-3)Galβ(1-4)][-Fucα(1-3)]GlcNAcβ (SLe^x-PAA), and [Neu5Acα(2-3)Galβ(1-3)][-Fucα(1-4)]GlcNAcβ (SLe^a-PAA), was determined. The last three polymer glycoconjugates were synthesized for determining the contribution of carbohydrate chains after the galactose link to the binding with the receptor. The difference in affinity between 3′SL and Neu5Acα in all studied H13 viruses is small, which indicates a less significant role of the galactose moiety in the binding to the receptor. The results of virus binding with polymer sialylglycoconjugates indicates that the method of linking, the third monosaccharide moiety, and the presence of an extra fucose substitute in this moiety may influence the binding considerably. For viruses isolated from ducks, the suitable polymer is SLe^a-PAA (i.e., a 1-3 linkage between galactose and glucosamine is optimal). This finding is in accord with the data that H13 viruses isolated from the gulls differ based on their ability to interact with polymer sialylglycoconjugates. The affinity to all three polymers is uniform, and the presence of GlcNAc-linked fucose does not prevent the binding. A comparative analysis of six sequenced HA H13 viruses and other subtype viruses showed presence of substantial differences in the composition of amino acids of this region in H13 viruses.

Low pathogenicity avian influenza virus (AIV) H7N2 has been isolated since 1994 from retail live-bird markets (LBMs) in the northeastern United States. This study examines the suppliers to the LBMs in New York and New Jersey. In 2001, 185 supplier premises in nine states were surveyed for the presence of AIV by virus isolation (VI) in embryonating chicken eggs. No H7 or H5 virus was isolated. In addition, 104 producer premises in two states were serologically negative for H7 and H5 AIV. Information on management practices was obtained via questionnaire for 191 premises in 12 states. The survey results suggest that current biosecurity practices at supplier premises could be improved, especially regarding movement of birds. The study supports the hypothesis that H7N2 AIV is primarily maintained within the LBMs and, if reintroduction from suppliers is occurring, it is likely retintroduced at a very low level or from suppliers not included in this study.

In March 1999 a syndrome characterized by depression, anorexia, fever, and respiratory and enteric signs appeared in many flocks of turkeys and, to a lesser extent, chickens in the densely populated poultry-rearing regions of Northeastern Italy. Initially the disease was characterized by sinusitis, tracheitis, peritonitis, and pancreatitis. The responsible agent was identified as low-pathogenicity (LP) avian influenza (AI) of H7N1 subtype. Concerning the light layers, the mortality was variable, from 1.7% to 9.5%, whereas egg production decreased by 10% to 40%. According to the epidemiologic data, chickens seemed to be less sensitive to the virus than were turkeys. Nine months later, the AI virus changed to a highly pathogenic (HP) AI virus and affected, besides turkeys, a great number of pullet and layer flocks, with high mortality (80%–100%) in a few days. However, the course of disease was more prolonged in pullets. Within 3½ mo, over 100 outbreaks were reported. Following the HPAI outbreaks, in late 2000 and early 2001, LPAI reemerged, but only one flock of layers was affected.

Eukaryotic expression plasmids encoding either the avian influenza hemagglutinin or matrix genes (pCMV-HA and pCMV-M, respectively) were constructed. The viral genes were derived from a low-pathogenicity H7N1 strain, A/Chicken/Italy/1067/99, isolated during the 1999–2001 epizootic in Italy. The plasmid was administered to 4-to-5-wk-old specific-pathogen-free chickens by several different injection methods. For the initial studies comparing methods of vaccine injection, results were compared based on hemagglutination inhibition (HI) response following immunization with pCMV-HA. Additional studies with coadministration of both pCMV-HA and pCMV-M was evaluated based on HI response and viral isolation after homologous challenge. Preliminary results indicate that a device intended to inject insulin in humans (Medijector) and the coadministration of both plasmids improved protection against H7 infection.

Using clinical materials from experimentally infected poultry, we established an effective method for the preparation of viral RNA directly from tissue samples and eggs. Furthermore, our type A–specific matrix reverse transcription–polymerase chain reaction (RT-PCR) test was improved, and an H7 subtype–specific nested RT-PCR, which includes the hemagglutinin cleavage site, was designed. Both RT-PCR systems proved to be as sensitive as virus isolation. In addition, the labeled H7 HA-nested PCR primers were suitable for sequencing of the PCR products. The RT-PCR amplification of viral RNA and sequencing of the PCR product allows for the sensitive and rapid differentiation between low-pathogenic and highly pathogenic avian influenza viruses.

The 1985 outbreak of high-pathogenicity avian influenza (HPAI) in Victoria, Australia, took 5 days to confirm by standard laboratory tests, during which time infected chickens continued excreting virus, thus creating the opportunity for transmission to other farms. An immunofluorescence test for the detection of viral antigen in tissue impression smears was evaluated as a rapid diagnostic test for HPAI virus infections of poultry. Several test configurations were compared for background reactions and strength of fluorescence, with the optimum combination found to be an influenza A group-specific monoclonal antibody, detected by an anti-mouse fluorescein isothiocyanate conjugate. Immunohistochemical examination of tissues from chickens experimentally infected with low-pathogenicity and HPAI viruses identified the pancreas as the organ most consistently containing high concentrations of HPAI viral antigen. This test has since been used in Australia in the rapid laboratory confirmation of three avian influenza outbreaks and in showing that numerous other suspect cases were not caused by avian influenza.

Determination of the avian influenza (AI) status of a flock has traditionally been done by detection of serum antibodies. However, for many diseases, detection of antibodies in egg yolk has been effective in monitoring the disease status of laying flocks. This study compared the utility of egg yolk vs. serum for determining AI status in laying hen flocks. Specific-pathogen-free white leghorn hens were inoculated via the respiratory tract with a low-pathogenic H7N2 AI virus or sterile allantoic fluid or subcutaneously with an inactivated oil emulsion vaccine produced from the same AI virus or normal allantoic fluid. Antibody levels were determined by the agar gel immunodiffusion (AGID) test, the hemagglutination-inhibition (HI) test, and the enzyme-linked immunosorbent assay (ELISA). Anti-influenza antibodies were detected in sera of all live virus–inoculated hens by day 7 postinoculation (PI) (AGID and ELISA tests), but detection of antibodies in egg yolk was delayed by a few days, with all being positive by day 14 PI. Sera from all vaccinated hens were positive by day 14 PI (AGID and ELISA tests), and egg yolk was positive by day 18 PI. The HI test was less sensitive than the ELISA and AGID tests in detecting anti-influenza antibodies in both sera and yolk. Serum and yolk from all control birds remained negative throughout the study. These studies show that currently used serologic tests can detect antibodies in serum and yolk samples from hens exposed to live AI virus or from those that have been vaccinated. Antibody is detected earlier in the serum than in the yolk and antibody is detected earlier from birds exposed to a live infection compared to birds vaccinated with an inactivated oil emulsion vaccine.

We have developed a reverse transcriptase polymerase chain reaction (RT-PCR)-based assay to detect influenza A in guano samples as part of our program to investigate ancient viral RNA from under Antarctic Adelie penguin (Pygoscelis adeliae) colonies. Of five extraction protocols tested (RNeasy, GTC TRIZOL, GTC Silica, Rnaid, and AGPC), AGPC proved to be the most consistent and sensitive to low concentrations of influenza, but further purification with commercial viral nucleic acid spin filter system was still required to remove remaining PCR inhibitors. RT-PCR was then performed on the eluent and 650 bases of the M1 gene were amplified. The assay was found to be able to detect as little as 100 μl of 0.1 hemagglutination units (HU)/ml influenza.

Ostriches were inoculated with a highly pathogenic avian influenza (HPAI) virus of ratite origin, A/emu/Texas/39924/93 (H5N2) clone c1B. The aim of this study was to evaluate the pathogenicity of this isolate for ostriches and to assess the ability of routine virologic and serologic tests to detect infection. Avian influenza virus (AIV) was isolated from tracheal swabs from 2 to 12 days postinfection and from cloacal swabs from 3 to 10 days postinfection. AIV was also isolated from a wide range of tissues. Birds seroconverted as early as 7 days postinfection. This study indicates that HPAI virus of ratite origin replicates extensively in infected ostriches without causing significant clinical disease or mortality.

The National Centre for Foreign Animal Disease (NCFAD) in Winnipeg, Manitoba, is the Canadian Food Inspection Agency's (CFIA) newest high biocontainment laboratory. One of the functions of the NCFAD is to serve as a national reference laboratory for avian influenza. Between 1997 and 2001, 15 avian influenza virus isolates were characterized. These isolates originated from domestic poultry, imported caged birds held in quarantine, and wild birds. Diagnostic specimens were submitted to the NCFAD by CFIA field veterinarians, provincial veterinary diagnostic laboratories, and veterinary colleges. Characterization of isolates included the determination of H and N subtypes: H1, H6, H7, and H10 subtypes were isolated from domestic poultry; H3, H4, and three H13 viruses were isolated from water fowl, and six H3 viruses were isolated from caged birds being held in import quarantine. Selected isolates were characterized with respect to their pathogenic potential by intravenous inoculation of 4-to-6-wk-old chickens. A molecular-based protocol was used to assess the pathogenicity of one H7 isolate. During this period, work was also carried out toward validating our molecular pathotyping protocol for avian influenza viruses with H5 and H7 hemagglutinin subtypes.

Between February 2000 and February 2002, the California Animal Health and Food Safety Laboratory System diagnosed 26 cases of low-pathogenic H6N2 avian influenza from 12 commercial egg-laying farms. The most common gross and histologic lesions observed in infected chickens were fibrinous yolk peritonitis, salpingitis, oophoritis, and nephritis. Edema of the mesentery of the oviduct and pale, swollen kidneys were also observed. Mortality in infected flocks ranged from 0.25% to 3%, and egg production dropped 7% to 40%.

See PDF for Participant List