Research Structure Aim

"Prevention and Control of Avian Influenza in the U.S." Research Structure

SPECIFIC AIMS

Specific Aim 1 . TOP To determine the molecular basis for adaptation of influenza A viruses from wild aquatic birds to land-based poultry. The critical first event in the development of HPAI is efficient replication of a LPAI. We will study the molecular basis that allows the transmission and perpetuation of AI from wild waterfowl to land-based poultry. We will study the molecular basis of the events that allow the transmission of AI from wild waterfowl and the ensuing adaptation to and perpetuation in land-based poultry. We will use a combination of forward and reverse genetics in combination with classical virologic and pathogenesis studies, as follows:

Objective 1a : TOP To define the role of intermediate hosts, particularly, quail and swine, in the emergence of influenza A viruses in chickens and turkeys. Rationale: Influenza A viruses from wild aquatic birds have been identified as the source of influenza viruses isolated from birds of the Galliformes order (e.g., chickens). Little is known about the relative frequency and the range of initial efficiency with which aquatic bird influenza can replicate in chickens. Our recent studies indicate that quail, which are available in LBMs, are more susceptible than chickens to most influenza A viruses from feral waterfowl. We postulate that waterfowl influenza viruses that would not otherwise replicate in chickens may do so after replication in quail. In other words, quail are an intermediate host in the emergence of AI in chickens. This study will examine the reservoir of influenza A in wild waterfowl from the Western Hemisphere with regards to their replication efficiency in chickens and quail. We will determine whether the increased susceptibility of quail to influenza A viruses from feral waterfowl facilitates the transmission of viruses from aquatic birds to chickens. Specifically, we propose to determine whether H5, H6, and H7 subtypes of influenza viruses present in wild aquatic birds (ducks and shorebirds) can replicate in chickens and quail and whether quail enhance transmission to other bird species, particularly chickens.

Outbreaks of LPAI and HPAI caused by H5 and H7 strains have inflicted important economic losses to the US poultry industry. There is evidence that LPAI H6 and H7 viruses continue to circulate in LBMs in the US Although H7 viruses have been circulating in the Northeastern LBM system for several years, the molecular features that allowed their adaptation to chickens remain poorly characterized. H6 infections in poultry in California have been associated with respiratory syndrome and drop in egg production. Again, the molecular features that have allowed H6 viruses to become established in chickens are largely unknown. Similar arguments can be made about the outbreaks of H5 in Pennsylvania in 1983-84. Although the determinants of high virulence were identified the key molecular events that allowed the virus to cross to chickens in the first place are not known. We hypothesize that intermediate hosts, which are themselves more susceptible than chickens to waterfowl influenza viruses, support the multiplication and selection of AI virus strains that can infect chickens more readily than the ancestral viruses (61).

Transmission of avian influenza viruses to pigs and the ensuing roles that pigs may play in pandemic influenza virus creation (serving as hosts for adaptation of avian viruses to replication in mammals and for avian and human virus reassortment) have received extensive research attention (79, 80, 112, 113). However, virus transmission between birds and pigs is a bi-directional process (4, 6, 60, 64, 120). Pig-to-bird transmission poses economically important disease concerns for domestic turkeys and also raises the possibility for widespread distribution of influenza viruses across the human population via the migration of free-ranging waterfowl (46, 55, 68, 69). The turkey industry in the U.S. includes production of over 270 million birds and a gross income of over 2.8 billion dollars per year (5). Six LPAI outbreaks in turkeys in the U.S. in 1978-2002 led to $179 million in losses (5). Beyond such direct monetary costs, a potential mechanism for enhancing the infectivity of HPAI viruses for human beings is via initial replication in pigs. Interestingly, turkeys could act also as amplifiers of influenza viruses that can transmit to humans as suggested by Hinshaw et al. (39). More recently, Suarez et al. have shown the presence in turkey breeder hens of H1N2 influenza viruses with swine, human, and avian lineage genes (96). Nonetheless, there has been very little research into the viral factors that control virus movement �backwards� from pigs to birds. We seek to address this gap in our knowledge of influenza virus host range determinants.

Experimental strategy 1 , transmission of influenza viruses in quail and chickens: If quail (or pigs) can act as amplifiers of duck influenza viruses and promote their transmission to other species, we should be able to mimic this phenomenon experimentally. Alternatively, if a virus from the aquatic bird reservoir can replicate and transmit in chickens, we should be able to establish this observation in the laboratory. We will consider the influenza A/Mallard/Alberta/24/01 (H7N3) virus as an example. Our initial preliminary experiments with this virus indicate that it replicates in quail but not in chickens. However, this virus transmits poorly in quail. Control viruses that replicate and transmit efficiently in chicken and quail have already been isolated and we will be used. The experiments will be performed essentially as recently described (61, 71).

We will determine whether these viruses can spread to other quail and/or chickens without adaptation. The extent of fecal and aerosol transmission will be evaluated. If the virus is not able to transmit, we will test whether the virus can be adapted to quail and transmit among quail after adaptation by performing sequential rounds of re-infection in quail with viruses isolated from lung homogenates. Once this has been accomplished, we will investigate whether it can transmit from quail to chickens and from chicken to chicken. If it does not transmit from chicken to chicken, it will be adapted to chicken as described above for quail and then tested for transmission among chickens, from chicken to quail, and among quail. In the next section, we describe how the molecular changes on the duck; quail-adapted and chicken-adapted viruses will be used to identify the molecular changes associated with interspecies transmission. The same strategy will be followed with the H6 (A/Mallard/Alberta/206/96 (H6N8)) and H5 viruses (A/Mallard/Alberta/271/88 (H5N3)). Detailed explanations of the experimental procedures to accomplish these objectives can be found in the cited literature. Proposal page limitations left us no choice other than to refer the reader to them.

Anticipated results and significance. Influenza viruses A/Mallard/Alberta/24/01 (H7N3), A/Mallard/Alberta/206/96 (H6N8), and A/Mallard/Alberta/271/88 (H5N3) transmit inefficiently (H7) or not al all (H5 and H6) among quail (Makarova et al., unpublished results, (61)). We will determine whether quail adaptation endows these viruses with the ability to replicate in chickens; i.e. an extended host range. If this is the case, it provides evidence that quail can serve as amplifiers and modifiers of influenza viruses that can then cross to other species. Our study will provide experimental evidence of the rapid adaptation of these viruses to quail and the associated expansion of their host range to chickens. We will also determine their pathogenicity for both, quail or chickens. Although virulence gains of the quail-adapted virus are influenced by multiple factors, this study may provide the first direct evidence of virulence gains under controlled experimental conditions. However, it is important to emphasize that the purpose of these studies is the analysis of the process by which duck viruses gain the ability to replicate and transmit in land-based poultry. Further genomic characterization (Objective 1b) will indicate the genetic changes that may be associated with the host "speciation".

Pitfalls: We anticipate that adaptation of these viruses to reach efficient replication and transmission in quail and chicken will require no more than 10 sequential passages. Our preliminary results showed replication of AI viruses from Mallard ducks in quail, suggesting that adaptation to achieve transmission in quail and chickens may not be difficult. The fact that H5, H6, and H7 subtype viruses have already evolved features that allow the infection of poultry suggests that these natural events can be mimicked in a controlled laboratory setting with little difficulty. We are also aware that multiple avian species may contribute to the viral adaptation in chickens and the selection of a single species (quail) may be an oversimplification. However, our recent studies showed that quail is an adequate representative of intermediate host species in this process and inclusion of additional species is not warranted. Similar studies can be performed in other land-based birds to define their role as potential intermediate hosts.

Experimental strategy 2 , transmission and adaptation of swine influenza viruses in swine, duck and turkeys. The ability of swine H3N2, H1N1, and H1N2 viruses will be tested for replication and transmission in turkey poults and mallard ducklings; and, to thereby establish definable infectivity and replication phenotypes with which genetic differences can be associated. We will adapt the swine viruses to infection in birds by repeated in vivo passage in poults and ducklings; and, to define sequence changes that may reflect adaptation of swine viruses to avian species by comparing sequences of the passaged viruses to the parental isolates and to define the receptor preferences of the parental, passage-derived, and reverse genetics-created (see below) viruses. As a control each virus will be evaluated for replication and transmission using 4 one-week-old piglets or germfree conventional hysterectomy-derived piglets.

The level of replication of virus in infected and sentinel birds will be evaluated by virus titration of swabs (virus isolation and RT/PCR) and serology. In all the experiments described above, the animals will be observed daily for clinical signs. Alternatively, animals will be bled and euthanized at 3, 6, 10 and 14 days post challenge. Different parts of the respiratory and gastrointestinal tracts will be examined for gross lesions and swabbed and tissues will be collected for histopathology. Sera will be tested for antibodies using the homologous virus in the HI test. Sections from tissues will be examined microscopically for lesions.

Anticipated results, pitfalls and significance: Our studies are aimed at characterizing swine viruses that differ in their ability to replicate and transmit in turkeys and ducks and to determine the molecular features that account for the differences in phenotypes. We expect to identify one isolate of each subtype that does not replicate and does not transmit in turkeys. We also expect to identify one isolate that replicates and transmits readily in turkeys. Alternatively, we expect other viruses to fulfill these requirements. For example, if all swine viruses tested have the capacity to replicate and transmit in turkeys, it will be indicative that these features were acquired early on the emergence of these viruses. Since some of these viruses have genes derived from human, swine and avian lineages, we could use a recent or an old human H3N2 or H1N1 viruses, which are expected to not replicate in turkeys. Although we cannot completely rule it out, it will be very unlikely to find that all influenza viruses of mammalian origin replicate and transmit in turkeys. As mentioned above, the molecular features that determine the capacity of swine viruses to go to turkeys are not defined, although recurrent outbreaks of swine influenza have caused important economic losses to the turkey industry.

Objective 1b: TOP To determine the molecular basis of transmissibility of influenza viruses in land-based avian intermediate hosts. Understanding interspecies transmission through reverse genetics. Rationale: Current plasmid-based reverse genetics methods have furthered our understanding of the biology of influenza viruses, including the mechanisms of interspecies transmission. Due to space limitations, the strategy below is presented only in the context of the quail and chickens studies. A similar rationale is used and similar studies will be performed to determine the molecular basis for transmissibility of swine viruses in turkeys and ducks.

Background on reverse genetics: The genome of negative-strand RNA viruses cannot be used as template for protein synthesis. Instead, the vRNA must be packaged into ribonucleoprotein particles (vRNPs) along with its own RNA-dependent RNA polymerase to start the first round of viral mRNA synthesis (40). We have recently succeeded in rescuing infectious particles of influenza A virus entirely from plasmid DNA (66) by using 17 plasmids. Eight plasmids in the set encode full-length cDNA copies of the 8 genes of influenza A virus flanked by the human RNA polymerase I (pol I) promoter and the mouse pol I termination signal. In this set, recognition of the pol I promoter by the RNA pol I complex results in the synthesis of influenza vRNA segments with exact 5' and 3' termini. The second group of plasmids corresponds to protein expression units encoding nine of the influenza virus proteins (PB1, PB2, PA, HA, NP, NA, M1, M2, and NS2) under the control of an RNA polymerase II promoter (pol II). Thus, transfection of all 17 plasmids in the set into human embryonic kidney cells (293T) results in the production of infectious influenza virus particles that can be further propagated in MDCK cells or chicken embryos. Hoffman et al. developed an improved version of this system consisting of only 8 plasmids (40) using pol I and pol II promoters running in opposite directions, which allows mRNA and vRNA synthesis for each gene from a single template. We have used the 8-plasmid system to analyze the biological effects of the interaction between two of the viral polymerases-PA and PB1 and rescued three avian influenza viruses (RG) with growth characteristics in eggs similar to the wild type viruses (58, 61, 71, 72). We have also developed a universal primer set for the full-length amplification of all influenza viruses. The primers rely on the fact that the 15 and 21 terminal segment-specific nucleotides of the genomic viral RNA is conserved among all influenza A viruses and are unique for each segment (41). This PCR-primer set has proven useful and reliable for cloning into reverse genetics vectors, sequencing, and subtyping.

Experimental strategy: Our strategy involves several steps, which are explained in detail below.
Cloning of H5, H6, and H7 influenza viruses by reverse genetics: The original wild type H5, H6, and H7 viruses described above will be cloned and rescued using our reverse genetics system. In addition, H6 viruses isolated in California from chickens (A/Chicken/California/1316/2001 H6N1) (114) will be similarly cloned and rescued and used in the �gain-of-function� and �loss-of-function� experiments as explained below. One important consideration for our studies is to test whether the viruses generated by reverse genetics maintain the growth characteristics of the parent viruses. Our extensive experience with this system makes us confident that we will obtain these viruses by reverse genetics with biological characteristics similar to the parental viruses.

To define which genes are responsible for replication and transmission in quail and chickens, we will perform "directed" reassortment, using our plasmid-based reverse genetic system. In this approach, one or more of the plasmids in the 8-plasmid set is replaced by a plasmid(s) encoding the homologous gene(s) from a different virus strain. The reassortant viruses will be analyzed for replication and transmission in quail and chickens as explained above.

Anticipated results: Defining the role of the eight genes in replication and transmission. Since replication and transmission of influenza A viruses in any given host depends on a combination of factors, we could anticipate that the HA alone will not be by itself the only determining factor in interspecies transmission. By following a "checkerboard" approach, we will engineer single gene reassortant viruses containing one gene from one virus group and the rest from a second virus group. For example, we will use the background of the virus that does not replicate or does not transmit in quail or chickens and will replace its HA gene with homologous gene from the virus that replicates and/or transmits in quail or chickens. With these observations and the available sequence information, we will determine which amino acids allow viruses carrying these HA genes to replicate and/or transmit in quail and chickens. We propose making reassortant viruses producing "gain-of-function" and "loss-of-function" viruses to define the role of one or more genes in replication and transmission. We would like to emphasize that the internal genes of influenza viruses have already been implicated in host range, although the molecular basis for such role is poorly understood. Since the internal genes are always co-evolving, a second issue that emerges is compatibility among the internal genes with our "checkerboard" approach. One could argue that if replication and transmission are polygenic traits then the single gene reassortant strategy will not work. Our approach of using very closely related viruses minimizes this issue, allowing us to study the contribution of each gene for transmission and replication. Alternatively, we will reduce incompatibility by constructing reassortants involving a functional gene cluster: for example the three P genes from a donor virus will be transferred as a group to a recipient virus. The same principle can be applied to the membrane proteins. Taken together, these studies should reveal the contribution of each gene or functional gene cluster to efficient replication in land-based poultry.

Pitfalls: Reverse transcription and PCR can introduce spurious mutations into influenza virus genes. Two or three rounds of RT-PCR and sequencing should provide a very high level of confidence about the exact sequence of each gene. In addition, generating viruses via reverse genetics will minimize spurious mutations that usually arise through multiple passages of reassortant viruses generated by conventional techniques.

Significance: The approach presented here will identify genes and their encoded amino acids that are necessary for alteration of host range of a group of different influenza virus subtypes (H5, H6, and H7). Although the receptor specificity of the HA protein is known to play a major role, particularly in the differences between avian and mammalian viruses, host range is polygenic in nature. Our studies will provide clues to the importance of specific genes in the replication and transmission of influenza viruses among land-based avian species.

Objective 2: TOP To determine the molecular parameters for detecting and diagnosing avian influenza viruses in chickens and turkeys affected by co-infections by other common respiratory and immunosuppressive viruses. Rationale: Low path AI viruses replicate in respiratory tissues of the chicken and turkey, as do infectious bronchitis virus (IBV) (primarily chickens only) and Newcastle disease virus (NDV). IBV is known to interfere with the replication of RNA viruses, specifically paramyxoviruses NDV and avian pneumovirus (9, 10, 34, 73). During the 2004 LP H7N2 outbreak on Delmarva, concern was expressed about the ability to detect AIV in chicks infected with vaccinal and field strains of IBV and NDV. Furthermore, the role of immunosuppressive viruses, such as infectious bursal disease (IBDV), chicken infectious anemia (CIAV) or hemorraghic enteritis (HE) viruses, on factors influencing the diagnosis of AIV in poultry has not been well studied. A limited amount of work on interactions of other viruses or bacteria on AIV pathogenicity has been done (7).

Experimental strategy: Effects of Concurrent Respiratory Infections on Diagnosis and Detection of AIV. Two-weeks old commercial-type broiler eggs will be vaccinated with a full dose of a commercial combination NDV and IBV live vaccine. On days 0, 3 and 7 after vaccination, chicks will be inoculated with a selected Delmarva/2004 LP H7N2 AI isolate. Tracheal, cloacal and blood samples will be collected at several intervals after AI inoculation for virus isolation and serology. Histopathological examinations will be performed on different tissues and clinical response, morbidity and mortality will be recorded daily. Serology will be performed. AI viruses recovered from the trachea and cloaca will be characterized by sequencing analysis and compared to the input virus.

Effects of Concurrent Immunosuppression. IBDV and CIAV antibody negative, female SPF leghorns will be exposed to IBDV, CIAV, or a combination of both. Seven days post-inoculation the birds will be infected with a selected Delmarva/2004 LP H7N2 AI isolate. Clinical and AIV serological responses after AI inoculation. Virus persistence will be assessed by virus isolation from trachea and cloaca swabs. Immunosuppressive effects on AI mutation and/or stability will be characterized by sequencing analysis and compared to the input virus. Histopathological examination will be performed.

Similar studies will be designed in turkeys to determine the interaction of NDV vaccination on the ability to detect AI and the effect of HEV infection on the pathogenesis of AI.

Summary . TOP These studies exploit the power of reverse genetics to dissect the molecular events that take place for avian influenza viruses to become endemic in land-based poultry particularly chickens and quail. We will elucidate the role of quail, a common bird species in LBMs, and pigs, which are considered important intermediate hosts of AI, in the natural history and emergence of influenza viruses that can infect chickens and turkeys. Most importantly, we will generate the necessary scientific evidence from which to establish agricultural and production management practices aimed at preventing the emergence of influenza A viruses in poultry. Currently, Hong Kong prohibits the sale of live quail in LBMs on the grounds that quail is a suspected amplifier of influenza viruses. Likewise, efforts are being made to avoid contact between swine and avian species. Similar restrictions have not been implemented in the US LBMs, where aquatic and land-based bird species are housed together. However, it is not clear whether the host range research performed with Eurasian waterfowl strains of influenza is applicable to understand the threat posed by North American lineages to US poultry. The proposed research using North American influenza viruses will provide the scientific basis to justify the need for future control measures. The use of state-of-the-art molecular biology, reverse genetics, and classical virology techniques will reveal molecular markers associated with interspecies transmission of influenza A viruses in quail and chickens. Until now, many land-based bird species have been overlooked as a potentially important intermediate host in influenza emergence. Similarly, the effect of other viral or bacterial diseases have not been carefully considered for the perpetuation of AI in LBMs or as factors promoting the emergence of AI in land-based poultry. Our studies will determine the importance of intermediate hosts, specifically quail and swine, and the role of concomitant viral infections associated with the establishment of influenza viruses in chickens and turkeys.

Specific Aim 2. TOP To determine the dynamics and evolution of influenza A viruses in the LBM systems, wild birds and gamebirds across the United States, to characterize the risks factors that contribute to the perpetuation of viruses in these populations, and to bring forward educational programs aimed at preventing, containing and controlling the spread of avian influenza in these systems as well as in commercial poultry.

Risk factors, education, and biosecurity programs. Rationale: The live bird market (LBM) systems of the Northeastern US (NE) and in Hong Kong, China are reputed to be mixing vessels for avian influenza viruses (AIV) from commercial and non-commercial birds of various species. The largely uncharacterized links between the LBM system and commercial poultry have led to AI outbreaks in both. The LB marketing systems have demonstrated that they embody the conditions under which dangerous viruses can emerge. However, the specific practices that drive the emergence of viruses, which can cross species barriers, and can spread readily in commercial poultry are not well established. Wild birds such as waterfowl and shore birds are major reservoirs for diverse populations of type A influenza viruses. Abundant epidemiological and virological data provide convincing evidence that influenza A viruses circulating in these wild bird reservoirs serve as parent strains for the high pathogenic and low pathogenic H5 and H7 influenza A viruses responsible for AI outbreaks in commercial poultry. Since the 1960s HPAI outbreaks have been documented in United Kingdom, Italy, Netherlands, Australia, Mexico, United States, Canada, Hong Kong, China and Chile. In this aim, we will investigate the risk factors associated with the emergence and perpetuation of influenza A viruses in bird species in the LBM systems in several areas across the country. We will also characterize AIV circulating in wild waterfowl and their captive raised counterparts. We will characterize the viruses antigenically, molecularly and in vivo in order to determine their phylogenetic relationships, to understand the movement of viruses across different bird species and raised under different conditions, the level of reassortment taking place and whether novel variants are generated. Once we have established the risk factors associated with the perpetuation of avian influenza in different settings, we will develop training, education, and biosecurity programs.

Objective 3. TOP Characterization of the risk factors associated with LBM systems in California, New York and Minnesota. Experimental strategy. Questionnaires will be administered to producers, distributors, dealers, transporters and marketers gathering data regarding three categories of risk factors for AI infection including health care, biosecurity and management practices. Supply farms and LBM will be enrolled in the study. Questionnaires will be personally administered and every effort will be made to do it in the optimal language for the producer. In addition, a standard form for the capture of GIS data for all participating flocks will be developed to be used in the mapping of bird movements and later modeling of outbreaks.

AI in LBMs. The birds in the LBM will be sampled in two ways: 1) every other day for two weeks, birds will be swabbed choanally and cloacally (birds will be selected using a stratified random method). All sample sizes will be individually calculated for each species in the market based on average population in the market with 95% confidence and 80% power estimating a range of prevalences of 1%-10% (±2%) in birds entering the market. 2) 50 birds entering the market will be swabbed, and banded before they are placed in the market. The random sampling portion of this study will be repeated. In addition to the sample techniques described, arrangements will be made to place 5-8 week old AI-negative white leghorn and Japanese quail. The sentinels will be obtained at 3-8 weeks of age placed in the market and sampled every third day for fifteen days. All of the sentinels will be banded and each sample identified by bird. Depending on the market and conditions of density for the market, the sentinels will be placed in cages in all rows, or in selected floor pens. In the laboratory, each sample will be processed for virus isolation and/or detection by RRT-PCR as described (93). Each isolate will be sent to the National Veterinary Services Laboratory for antigenic typing. A 400-500 bp region from each gene from each isolate will be sequenced. All virus sequences will be made available to other investigators through submission to the Los Alamos repository. Selected isolates will be inoculated into guinea fowl, Japanese quail, chickens and turkeys for a limited pathogenesis study and to determine species tropism.

Data analysis : The data will be analyzed collectively with a simulation model. For the simulation model to be useful, it needs to be based on real information concerning the frequency and range of contacts among all poultry operations including supply flocks, distribution centers, and markets. These contacts include birds, vehicles and humans, bird movements, and their geographic range, including frequency of visits by a veterinarian or technician and the distance traveled by delivery/pickup vehicles, which characterize the activity to and from a poultry operation. For example, data collected in an FMD modeling study allowed us to estimate disease transmission potential among livestock premises, either directly from movement of animals, or indirectly via vehicles or persons (8). As data from these studies becomes available, the assumptions for each operation will be altered to more accurately reflect the conditions associated with the LBM system.

Objective 4 . TOP Establish and maintain a coordinated, systematic type A influenza virus surveillance network in wild bird populations for the United States. The surveillance network will consist of multi-participants using standardized protocols for sampling and analysis (88, 90). Two or more study sites will initially be established on each of the four major migratory waterfowl flyways. One site will be in the northern reaches and the other site in the southern portion of each flyway within the boarders of the U.S. A unique element of this project is a focus on non-migratory wild bird populations residing in areas along migratory waterfowl flyways. Non-migratory wild bird study sites will be selected as close as possible to the respective migratory bird study sites and, whenever possible, in close proximity to areas of significant live bird market suppliers and/or commercial poultry production. Migratory birds will be sampled at the earliest opportunity upon first arrival at designated study site, midway through local migration period and at the end of migration season for each respective study site. Non-migratory wild birds will be samples as early as possible in the year and periodically during the year (as can be arranged), but at least one time before arrival of migrants, in the middle of migratory season, and after the migrants leave. All surveillance activities will include collection of ground-truth data using GPS data collection systems with database synchronization capabilities to the GIS Decision Support System. Antigenic and genetic characterization AI viruses isolated from wild birds will be performed. In addition, limited pathogenesis studies will be performed in vivo on selected isolates in order to characterize their ability to cross species barriers.

Objective 5. TOP Building effective education and biosecurity programs. The risk assessments performed in Objectives 3 and 4 will be used by researchers to address the specific needs associated with the development of a comprehensive education, outreach, and biosecurity programs for both live bird market systems, for the producers of wild captive waterfowl, and for the producers of commercial poultry. The education and outreach efforts will be designed to target different stages of understanding, address specific audience needs, and will use a targeted incremental approach that emphasizes genuine understanding gained over time. Educational materials on avian influenza control including stand-up presentations, printed and on-line fact sheets, and CDs will be generated and delivered to targeted audiences including integrated and semi-integrated poultry meat and egg industry, allied industry representatives, independent poultry producers, poultry dealers and distributors, live bird market owners and workers, gamebird producers, poultry and waterfowl fanciers and 4-H club members. A comprehensive educational program for gamebird producers will be offered to all states. In the first year, workshops will be held in Texas, Ohio, Minnesota, and Maryland. Local veterinarians hosting the program will not only receive the materials needed to put a game bird program together but a principal investigator will also travel to each state as an expert speaker on game bird diseases for the program. Once the program has been delivered in a state, the local veterinarians will have received the training required to conduct additional training sessions using the same template. The educational program delivered to this audience will serve to establish local contacts with the producers of gamebirds and to establish a means through which real epidemiologic data can be collected for this underserved and relatively unknown part of the commercial poultry industry. Quality assurance and regional biosecurity programs will be developed to specifically assist each of these groups. These materials will be integrated with those of other participating regions so that both general concepts and region-specific concepts will be covered. Post-intervention data on producers and retailers will be collected using written surveys and individual interviews. Data from both sources will be organized and analyzed using appropriate statistical methods. Worldwide web outreach: We will develop and maintain a project-specific website, linkable from well-established, germane URLs. The strategy for developing this website will be to provide utilitarian, non-biased information, using an intuitive web interface. Textual content will be developed at the 8th grade reading level to support outreach for both lay and scientific communities.

Objective 6. TOP Develop methods of poultry pathogen inactivation and create a network of local expertise for the disposal of catastrophic mortality. Various methods for the inactivation of AI on poultry equipment and in poultry houses will be evaluated experimentally. In addition, a highly successful education and training program will be delivered in the major meat bird producing regions of the nation. This program teaches participants a highly effective method of in house euthanasia and composting. These methods were used successfully in the recent Delmarva AI outbreaks.

Pitfalls . There are several roadblocks that must be overcome in order to implement biosecurity programs that include both commercial and niche market poultry producers. One of those critical factors is the distrust that these groups feel for each other. However, once a common goal, i.e., AI control, is established, then it is much easier to work out the details of a plan. In the early development of a plan, it must be clear to all parties that they cannot economically live with AI. Then, it is important to gain agreements that there will be no punishment for people that conduct surveillance and find infections. Honesty among producers of all types is essential in AI control.

Summary. TOP The success of any major program aimed at controlling any given infectious disease is measured by the prevalence of the agent at the beginning and at the end of the program. This will be true for all except the surveillance conducted in wild free-ranging waterfowl in which no interventions will be attempted nor will any changes be expected. Working closely with the live bird marketing system and producers that supply those markets as well as the producers of wild captive birds, we will determine the risks factors that promote the perpetuation of AI in land-based poultry (chickens). We will determine the incidence of AI in the LBM system and devise strategies to control the spread of and, perhaps, eliminate AI in these systems. Our phylogenetic studies will provide evidence on the overall movement and spread of viruses in different states and across the country. By the end of the 3-year period of this proposal, the education programs (traditional- and electronic-based) that we will be developed and implemented will definitely raise public awareness among market owners, suppliers, wholesalers, dealers of commercial poultry and the general public of the importance of AI for poultry and public health. Without scientific evidence it is difficult to change cultural, management, traditional, and market practices. Our studies will change the way the general public perceives AI and will help in the implementation of bioregulatory strategies that will limit the emergence of novel and more virulent influenza A virus strains.

Specific Aim 3. TOP To develop critical diagnostic tests and vaccines for avian influenza control. This objective seeks to exploit the expertise and knowledge from the previous specific aims to address two key high priority needs: 1) Development of highly sensitive and rapid diagnostic tests for avian influenza during surveillance and outbreaks capable of discriminating between non-vaccinated and vaccinated birds as well as infected birds. 2) Development of alternative strategies for vaccination against avian influenza exploiting the latest molecular approaches available including reverse genetics and recombinant expression vectors.

Objective 7. TOP Diagnosis of AI. Rationale and Experimental Strategy: A crucial element in the control of AI outbreaks is the early identification of flocks infected with avian influenza. The earlier an infected flock is identified and quarantined, the less chance of the virus spreading from that flock will occur and the outbreak can be contained faster and at lower cost. Although we have tests for the direct detection of AI as well as serologic tests, for the eradication phase of a control effort, the direct detection methods are the most important for controlling the disease. Three direct detection techniques are commonly used for the detection of avian influenza virus, including virus isolation in eggs, RRT-PCR (93), or antigen capture tests. Each of these tests has their advantages of speed and sensitivity. Virus isolation is sensitive, but time consuming. The RRT-PCR test, which is being used as part of the National Animal Health Network System for AI surveillance, is rapid and sensitive, but requires expensive equipment and must be performed in the laboratory. Antigen-capture tests, such as the Directigen test, are rapid and can be performed with a minimum of equipment, but suffer from poor sensitivity and relatively high cost per test. To fill this gap in the diagnostic toolkit available to the poultry health professional, we propose to develop a diagnostic test that incorporates all four features (rapid, sensitive, inexpensive, and can be performed penside). A simple, sensitive and specific penside assay will expedite response time in the event of an outbreak and will limit the opportunity for viral spread. Deployment of a test that quickly detects the presence of influenza virus in a sample will also streamline virus characterization in the lab.

One new technology to be evaluated is the optical interferometric biosensor, which has been previously been used for the rapid identification of pathogenic bacteria on chicken carcasses in slaughterhouses (Xu, J. and Gottfried, D.S., unpublished and (82). With the use of an inexpensive reader, samples can be collected and tested on site with high sensitivity. The biosensor project will develop a test that can identify specific hemagglutinin subtypes, with a concentration on H5 and H7 viruses. Once the test has been shown to work with experimental samples in the laboratory, clinical samples from experimentally birds will be tested in comparison with other diagnostic tests (virus isolation, RRT-PCR or commercial antigen capture tests.) As part of the collective project, without direct cost to the grant, ongoing efforts to improve the sensitivity, throughput or functionality of existing tests, such as the RRT-PCR test, will be conducted by members of the group and the results will be shared.
Other diagnostics tests will be developed during the course of the project to meet specific needs of the poultry industry. They include the development of serologic tests for different subtypes of the neuraminidase (NA) protein to be used with the DIVA (Differentiating Infected from Vaccinated Animals) vaccination strategy (17). Currently, only killed whole virus adjuvanted vaccines have been used for vaccination in the U.S., and it is not possible to differentiate birds that have been vaccinated from those that are naturally infected with our currently available serologic tests. This inability to distinguish vaccinated from naturally infected birds has been a major impediment for the use of vaccination, since our trading partners want assurance that poultry or poultry products have never been infected with avian influenza. The NA DIVA approach vaccinates poultry with a killed virus that has an HA protein of the same subtype as the field strain, but the vaccine has a NA protein different from the field strain. This provides a way to tell if a bird was vaccinated or naturally exposed to avian influenza, since vaccinated birds would have antibodies to one NA subtype and infected (and even vaccinated and infected) would have antibodies to the NA subtype of the field strain. In this project we will develop both IFA and ELISA tests using neuraminidase proteins produced in a baculovirus expression system to allow a fast, sensitive and cost effective way of testing samples for antibodies to different NA subtypes. The DIVA approach can then be used to provide assurance to our trading partners that even vaccinated birds have not been exposed to specific avian influenza viruses.

A DNA microarray test will be developed that can quickly and accurately characterize avian influenza viruses on a large number of samples (50, 57, 81) and an Antibody-based Virus Concentrator (AVC). Past research has shown that DNA microarrays can differentiate between human hemagglutinin subtypes, but this project will expand that approach to include all hemagglutinin and neuraminidase subtypes (Keeler, C. et al, unpublished). Additional work will determine if the microarray can differentiate the source of internal genes, for example human vs. avian vs. swine matrix genes. AVC-like methods have been developed for food safety-related samples that can process large sample volumes and that produce a rapid and sensitive signal (118) . We will adapt this technique to a simple and easy operated syringe system that will contain small glass immuno-concentrator beads carrying the AI-specific polyclonal antibodies attached through a cleavable spacer arm. Aspiration of a processed sample will allow for capture of the virus particles by the virus-specific antibody. An enzyme-linked detection antibody (also virus-specific) will then be aspirated through the syringe allowing for colorimetric detection of the captured virus. The virus can then be recovered for further molecular analysis. The AVC system will provide a rapid virus detection method allowing an easy and systematic recovery of AI in a small sample volume, devoid of any interfering factors. Preliminary experiments have delivered proof of concept for the basic immunochemistry underlying this technique. Due to its simplicity, the AVC will be useful in the field as well as in the diagnostic and research laboratory levels. In addition, the following methodology will be developed for detection of AI in tracheal, cloacal and environmental samples and their sensitivity and specificity evaluated: microsphere-based multiplex immunoassays (M-BMI) (16, 26, 121). M-BMI is an emerging technology with tremendous potential in the clinical and epidemiological management of infectious diseases.

Anticipated results: The implementation of these objectives will require active collaboration between the molecular biology and immunology groups on one hand, and the infectious diseases/epizootiology groups working in the LBM and commercial poultry facilities. As soon as the tests are validated for influenza virus experimental diagnostic use, we will either engage our technology transfer offices in getting private sector involvement to develop these tests into licensed commercial products, or directly transfer the technologies to the appropriate diagnostic laboratories. To maximize the efficient utilization of resources, a decision will be made each year of the grant to select the most promising technologies for further development with enhanced resources.

Objective 8. Vaccines. TOP
Evaluation of Avian Influenza Low Pathogenic H5 and H7 Subtype Isolates for Suitability as Vaccine Seed Strains for Emergency Vaccine Stockpiles AI vaccination strategies. Rationale and Experimental Strategy: There is very little doubt that AI transmission is effectively prevented in sero-positive flocks (reviewed in (36). Inactivated oil-emulsion AI vaccines are very effective in protecting poultry against disease and reducing the spread of HPAI. The choice of avian influenza vaccine seed viruses is critical to the success of both vaccine production and vaccine efficacy when applied in the field. Due to the continued threat of exposure of American poultry to AI the USDA's Animal and Plant Health Inspection Service (APHIS) has established a program to develop a stockpile of H5 and H7 inactivated vaccine for emergency use in an outbreak situation. Currently, two H5 and two H7 low pathogenic avian influenza isolates have been selected as candidate strains for these vaccines. The isolates were selected in part because they have neuraminidase subtypes that are not usually found in gallinaceous poultry, thus facilitating the implementation of DIVA strategies. The ability of each virus candidate to replicate to adequate titers in embryonated eggs, immunogenicity and stability of the inactivated vaccines, and genetic and antigenic relatedness to currently circulating strains will be evaluated.

Evaluation of Antigenic Drift in Mexican Avian Influenza Viruses Using DNA Vaccines. Rationale: The reduction in AI viral shedding is correlated to how close the vaccine is to the challenge strain. This has been particularly shown using the fowlpox vectored H5 vaccine in which the amount of virus shedding from the oropharynx was significantly correlated to the genetic similarity of the vaccine to the HPAI challenge strain (102). The identification and characterization of the antigenic sites on the HA gene can aid in the prediction of which vaccine seed strain to use or when to change the vaccine seed strains for optimal protection from vaccination.

Experimental Strategy: Specific changes to the Mexican lineage H5 HA gene will be made by site-directed mutagenesis in a eukaryotic expression vector for the purposes of mapping the antigenic regions. The modified plasmids will be used to vaccinate chickens to generate a serologic immune response to each plasmid. The different sera samples will be tested with one-way and two way cross HI titers to assess how the specific changes made to the hemagglutinin gene will affect the binding of the antibody produced from the vaccine. Using a mapping approach, amino acid changes that are important for immunity will be identified. Other representative H5 viruses will also be used for comparison purposes. These studies highly valuable information that will help in the identification of future candidate vaccine strains.

Evaluation of Recombinant anti-AI vaccines generated by Reverse Genetics. Rationale: Major scientific breakthroughs in the field of reverse genetics of negative strand viruses and DNA immunization may provide the ideal candidates for the future implementation of vaccination alternatives to help control AI disease and spread of HPAI and LPAI. More importantly, some of these new technologies may provide the long-sought mucosal immunity against AI. We will explore six alternative strategies for vaccination against HPAI (and LPAI) using 1) a replication deficient recombinant adenovirus vector (84), 2) a recombinant live lentogenic Newcastle Disease Virus vector (NDV) (45), 3) a recombinant Marek's disease virus (MDV) vector (75) 4) a recombinant Semliki Forest Virus (SFV) suicide replicon (78), 5) a recombinant Infectious Bronchitis Virus (IBV) vector (Youn, et al. unpublished), and 6) a whole avian recombinant reverse genetics oil-inactivated influenza A virus vaccine (Song and Perez, unpublished). The choice of these live vectors vs. inactivated is justified on the basis of their proven performance in poultry (NDV, MDV, or IBV) or their greater acceptability from the perspective of safety for poultry (replication deficient adenovirus, SFV, and inactivated influenza). We will perform comparative studies to reduce the number of approaches as soon as possible, maximizing available resources.

Experimental strategy: The HA genes of representative H5 and H7 HA genes of HPAI will be cloned into recombinant vectors NDV, MD, and SV5 vectors. Simultaneously, the H5 and H7 genes from LPAI (or from HPAI with their polybasic amino acid region removed) will be rescued using an influenza virus reverse genetics system. Several NA subtypes will be tested in the recombinant influenza viruses to determine which combination offers the best protection in challenge studies and allows the differentiation of vaccinated versus non vaccinated birds (DIVA approach). Vaccine studies will be performed in white leghorn SPF chickens. Shedding and transmission among vaccinated and non-vaccinated controls will be evaluated. Finally, once we have determined the best vaccination strategy and system under controlled laboratory conditions, we will develop plans to test it in a real LBM setting.

Anticipated results: This objective will represent a collaborative effort among several groups involved in the research and development of vaccines. We expect significant differences in the rates of infection with field AI strains in sentinel and vaccinated animals: this will be indicative of protection. We also anticipate that the different strategies described above will show protection against HPAI and provide alternatives to the use of traditional oil-inactivated AI vaccines. The impact of vaccination on AI shedding will be revealed by virus isolation rates from swabs from sentinel and vaccinated birds. These strategies will allow the adequate differentiation of vaccinated versus non-vaccinated birds; which could stimulate the routine use of vaccines to control the spread LPAI. With the testing of one of these vaccine strategies (or a combination) in a real LBM system, we expect to demonstrate that such strategy will effectively control AI spread, decrease viral load and not interfere with the epidemiological monitoring of AI in the market.

Pitfalls : The diagnostic methods that will be developed and tested in this aim are already being used in the diagnosis of human infectious diseases. Thus, the adaptation of these techniques to detection of AI in poultry should not be difficult. Likewise, the vaccine methods that will be utilized have been demonstrated to protect against other diseases in poultry (including HPAI). In fact, some of these vaccine vectors are widely used in the poultry industry (NDV and MDV). However, more studies are needed to provide a more comprehensive understanding of their effectiveness against AI.

Summary . TOP Influenza A viruses continue to pose a major threat to the poultry industry and public health. In fact, the Office International des Epizooties (OIE) has recently (May 2003) revised its international standard on avian influenza concerned with the risk that a spread of avian influenza may present to the international community not only from economic but also from public health points of view. Uncontrolled avian influenza in the US could paralyze the industry and compromise our position as the leading exporter of poultry in the world. As part of the preparedness for AI outbreaks, it will be important to maintain an active network that effectively monitors influenza activity in poultry across the country, using the most versatile and fastest methods of detection available. Likewise, access to reliable vaccine strategies that can be quickly implemented to control the spread of AI during outbreaks is another important national priority. Our research on better and faster diagnostic methods will result in the replacement of virus isolation in eggs and AGID, both tedious and heavily time-consuming, as the standard methods for AI diagnosis. The technologies that will be developed during this project are already changing the way in which some human infectious diseases are diagnosed. We intend to do the same with the detection of AI in poultry. We will demonstrate the efficacy of our newly developed vaccines in experimental settings as well as under LBM conditions. Our vaccines studies will contribute to a better understanding of the possible application of vaccination to control the emergence and spread of potentially HPAI viruses.




> Next
LITERATURE SEARCHED