The Highly Pathogenic Avian Influenza Virus

David M. Robinson, D.V.M., Ph.D. (1)
Robert M. DeBell, Ph.D. (2)
(1) Battelle Memorial Institute and
(2) Titan Corporation

One of the co-authors (RMD) of this article published a review of influenza in the October 2004 issue of this publication. The October review touched on the history of influenza, its symptoms, and its potential as an agent of bioterrorism. In this article we will expand on some of these points and emphasize recent efforts to accelerate the production of vaccines protecting against novel strains of highly pathogenic avian influenza (HPAI).

The influenza viruses are termed orthomyxoviruses, and there are two major groups: the type A and B viruses are grouped together in a single (cont. p. 10 - Avian Influenza) (Avian Influenza - from p. 1)

group and are the major cause of human disease, while the group C viruses constitute the second group. Some authors classify each of the virus types separately listing each as a genus. However, the type A and B viruses are structurally closely related in that they contain eight separate genomic segments of negative strand RNA while the group C viruses contain seven segments. The most important type from the standpoint of human disease are the Influenza A viruses that have caused seasonal epidemics and global pandemics and infect, in addition to humans, a number of species including wild and domestic birds, pigs, horses, dogs, and cats (1). The remainder of this paper will consider only the type A influenza viruses.

Influenza A viruses are subtyped based on their hemagglutinin (H) and neuraminidase (N) glycoproteins (2). These surface glycoproteins are associated with attachment and entry via H into the cell and release via N from the cell to spread to other cells. At least 15 distinct antigenic types of H glycoproteins (H1 – H15) have been identified and 9 types of N glycoproteins (N1 – N9). The principal protective antigen is the hemagglutinin, but immunity to the neuraminidase does provide some protection, as well. Historically, viruses producing pandemics have been typed as H1N1, the 1918 Spanish flu; H2N2, the 1957 Asian flu; and H1N1, the 1977 Russian flu. In the recent past, H1N1 and H3N2 viruses have concurrently circulated in humans and produced the morbidity and mortality commonly observed during the winter flu season.

As mentioned above, Influenza A viruses can infect a broad range of domestic and wild animals and fowl, as well as humans. This allows a phenomenon of “antigenic shift” to occur. Antigenic shift is thought to occur when two subtypes of Influenza A virus infect a host animal or bird at the same time, and through genetic reassortment between the eight genomic segments in each virus subtype, a novel virus emerges. This virus can have either a new type hemagglutinin or a new type neuraminidase or both. Since the human population is not immune to the virus, antigenic shifts in hemagglutinin are the genesis of new virus types that produce pandemics. These antigenic shifts are thought to occur very rarely with only six established for the hemagglutinin gene since 1889 (1). On the other hand, “antigenic drifts” are much more subtle changes produced by mutations in the genes of the virus with resulting modifications to the surface glycoproteins. These changes affect the antigenic expression of the subtype rendering a vaccine that was protective against the original virus to be less protective against the mutated virus.

While influenza viruses containing all 15 serotypes of hemagglutinin and all 9 serotypes of neuraminidase circulate in wild aquatic fowl, only subtypes H5 and H7 contain strains that can be called highly pathogenic to fowl. HPAI viruses can produce severe respiratory disease and mortality approaching 100% in chickens. Early isolates from 2003 were reported to kill wild fowl, but more recent isolates from 2004 do not kill domestic ducks. In fact, the disease was limited to the gastrointestinal tract, and virus was shed in the feces (3). The isolates from 2004 also appeared to be more stable to environmental conditions. This environmental stability is one of the basic criteria for a microbe to be considered an effective agent as a biological weapon and potentially provide a bioterrorism threat.

These subtypes of influenza that are normally restricted to avian species appear to be expanding their host range with differences in the resulting pathology. In Thailand, two tigers and two leopards contracted H5N1 avian influenza as a result of being fed chicken carcasses from birds that had presumably died of the disease (4). While all of the animals had severe pulmonary lesions, one tiger and one leopard also showed lesions typical of encephalitis on histological examination. Encephalitis is characteristic of the pathology observed in laboratory mice. Additionally, domestic cats can be infected with an H5N1 strain isolated in Vietnam in 2004 with resulting alveolar lesions, virus shedding, lateral transmission and, in some animals, death (5).

The disease caused by H5N1 highly pathogenic viruses in humans has been characterized primarily by pulmonary distress with accompanying fever, headache, myalgia, and malaise. Gastrointestinal involvement, while not a prominent sign, has been reported. However, a fatal case from February 2004 presenting with severe diarrhea, convulsions and coma but no respiratory symptoms was reported from Vietnam (6). Virus was isolated from cerebral fluid. The patient’s sister had died of a similar syndrome two weeks earlier, and the clinical diagnosis in both siblings was acute encephalitis. A second case, with symptoms restricted to diarrhea, nausea and vomiting has been reported from Thailand (7). The occurrence of atypical influenza with encephalitis as a primary lesion suggests that the virus is becoming adapted to grow in neural tissue, while the occurrence of gastrointestinal symptoms suggests that the HA is changing to preferentially attach to receptors on the intestinal cells.

In the accompanying table we have collated the published information related to human outbreaks of avian influenza through 2004. From these data, the incidence of human disease resulting from avian strains appears to be increasing. The H5N1 type is likely endemic in migratory fowl and ducks in much of Asia at the present time; however, avian strains have caused outbreaks in poultry in the past without accompanying human disease. It would also appear that the incidence of human disease caused by avian strains of influenza is increasing and that the H5N1 strain is more pathogenic for humans than the H7 or H9 strains.

Year Location Subtype Infected Deaths

1997 Hong Kong H5N1 18 6

1999 Hong Kong H9N2 2 0

2002 USA H7N2 1 0

2003 China H5N1 2 1

2003 Netherlands H7N7 80 1

2003 Hong Kong H9N2 1 0

2004 Canada H7N3 2 0

2004 Vietnam H5N1 27 20

2004 Thailand H5N1 17 12

This increase in the occurrence of outbreaks of disease caused by the H5N1 type has caused concern in the public health communities around the world. WHO has reported that the death rate among 69 confirmed cases is 68% (47 deaths). The point has been made that the H5N1 virus only needs to gain the capacity to efficiently spread from person to person to attain full pandemic potential. Probable spread from person to person was reported in 2005 (8).

While quarantine, isolation, and travel restriction will be important components of any control measures, vaccination will be crucial to controlling the spread of the disease. The bulk of current vaccines for influenza are produced by inoculating embryonated hens' eggs with the virus. Following an appropriate incubation period, the allantoic fluid is harvested, and the virus is purified and inactivated to produce the vaccine. The main components of the process are the viral seed strains, the growth substrate, the purification process, and the formulation of the final product.

To accelerate the production of appropriate vaccine seed strains, researchers have developed the plasmid based reverse genetics procedure which uses cDNA to generate strains that grow well in eggs, produce the HA antigen of the highly pathogenic strains, and are nonpathogenic in common laboratory hosts (9). This procedure has been used to generate vaccine strains containing H5 antigens. These seeds have been amplified, tested, and used to prepare small lots of vaccine for clinical trials. They will be available for production of H5 vaccine. Other groups are investigating the use of recombinant technology to express the H5 antigen in baculovirus and other systems for use to produce the active pharmaceutical ingredient in the vaccine (10).

Producing the current vaccine from embryonated hens' eggs is a laborious procedure and enhances the opportunity for contamination both because hens’ eggs are not sterile and because of the requirement for handling large quantities of material under aseptic conditions. Many organizations are studying the use of cell cultures for the propagation of the influenza virus (10). The major advantage of cells is the ability to rapidly scale-up the cells from frozen, validated, seed stocks and, as a result, to more quickly produce vaccine. The purification process is simplified because cell cultures have less extraneous material as compared to allantoic fluid.

Adjuvants will allow the maximum use of the immunogen that is produced by reducing the quantity of antigen necessary in each dose of the vaccine. Aluminum hydroxide is the only adjuvant currently recognized by the U.S. FDA for use in vaccines; however, other proprietary adjuvants, such as Chiron’s MF59, are widely used outside the U.S (11).A promising new group of adjuvants are the CpG compounds being studied for their ability to improve the response of the elderly to influenza vaccines (12).

In conclusion, the occurrence of human disease caused by the HPAI, subtype H5N1, is increasing in Asia. The virus has produced high mortality rates and has spread to other species. Following infection the virus is shed in high titers by domestic ducks which show no signs of the disease, and the virus is wide-spread in migratory water fowl in Asia. All of these factors point to the possibility of a major pandemic caused by this strain either by natural means or initiated by a bioterrorist event. In either case, the control measures will be the same. An organized collaborative program is in progress which will result in more knowledge about influenza, its virulence factors, and its potential for use in a bioterrorist event. This program will result in the ability to immunize more people with the current vaccine production capability, more rapid development and production of vaccines, and improved efficacy of antiviral drugs. This is a unique opportunity for the medical community to prepare for a major epidemic before it occurs.

Literature cited.

1. Flint, S.J., L.W. Enquist, V.R. Racaniello, and A.M. Skalka. In: Principles of Virology: Molecular Biology, Pathogenesis, and Control of Animal Viruses, 2nd Ed. ASM Press, Herndon, VA; 2003. p.814-816.

2. Centers for Disease Control and Prevention. Information About Influenza Pandemics. Available from http://www.cdc.gov/flu/avian/gen-info/pandemics.htm

3. Normile, D. Ducks May Magnify Threat of Avian Flu Virus. Science 2004;306:953.

4. Keawcharoejn, J., K. Oraveerakul, T. Kuiken, et al. Avian Influenza in Tigers and Leopards. Emerg Infect Dis 2004;10:2189-2191.

5. Kuiken, T., G. Rimmelzwaan, D. van Riel, et al. Avian H5N1 Influenza in Cats. Sciencexpress 2004; 10.1126/science.1102287.

6. de Jong, M., B.V. Cam, P.T. Qui, et al. Fatal Avian Influenza A (H5N1) in a Child Presenting with Diarrhea Followed by Coma. N Engl J Med 2005;352:686-691.

7. Apisarnthanarak, A., R. Kitphati, K. Thongphubeth, et al. Atypical Avian Influenza (H5N1). Emerg Infect Dis 2004;10:1321-1324.

8. Reuters Health Information. Bird Flu Clusters May Signal Virus Change: WHO. Available at http://www.medscape.com/viewarticle/501439.

9. Nicolson, C., D. Major, J. Wood, et al. Generation of influenza vaccine viruses on Vero cells by reverse genetics: an H5N1 candidate vaccine strain produced under a quality system. Vaccine 2005;23:2943-2952.

10. Cox, M. Cell-based protein vaccines for influenza. Curr Opin Mol Ther 2005;7:24-29.

11. Podda, A. The adjuvanted influenza vaccines with novel adjuvants: experience with the MF59-adjuvanted vaccine. Vaccine 2001;19:2673-2680.

12. Cooper, C.L., H.L. Davis, M.L. Morris, et al. Safety and immunogenicity of CPG 7909 injection as an adjuvant to Fluarix influenza vaccine. Vaccine 2004;22:3136-3143.