Avian Influenza virus and Specific Bacterial Synergism

Waisees Yeung(Yang Dewei) (1) Liu Fuan(2) Chen Bowen(3)

1 State Key Laboratory for Biocontrol and Biopharmaceutical Center, Zhongshan
University, Guangzhou 510275

2 College of Veterinary Medicine, South China Agricultural University

3 Sichuan Import/export Inspection and Quarantine Bureau

Abstract: This paper reports molecular virology and microbiological studies on the avian influenza outbreaks that occasioned high poultry mortality in Hong Kong during the period of 1997 through 2003. It was found that certain avian influenza strains could act synergistically with substances secreted by some bacteria, so that non-pathogenic mild avian influenza virus strains could evince the high pathogenicity of virulent strains.

Keywords: avian influenza virus¡]AIV¡^; virus synergism; antibiotic-tolerant bacteria

Influenza virus is an eight-segment single stranded RNA virus belonging to the Orthomyxoviridae. Each segment of the genome can encode a kind of protein¡Asome being functional proteins¡Asome structural proteins. The structural proteins consist of nucleocapsid protein (NP), matrix protein (M), hemagglutinin (HA) and neuraminidase (NA). On the basis of the antigenicity of the nucleocapsid protein and matrix protein¡Athe influenza viruses are classified in to A, B, and C serotypes. Humans are afflicted by all three types but generally only Type A can additionally infect other mammals and birds. Type A influenza viruses can be further subclassified into H subtypes based on the hemagglutinin spike antigen and N subtypes on the neuraminidase spike antigen. There are 17 H subtypes namely H1, H2, H3,.......H17, among which H1, H2, H3 and the H5 discovered 1997 in Hong Kong can occur in humans, and 11 N subtypes namely N1, N2, N3.........N11. Due to various combinations of H and N subtypes, we could theoretically have 17 x 11 or 187 subtypes, but in reality there have not been so many subtypes identified. By and large, all the Influenza Virus A subtypes H and N have been isolated from birds, but that is not true for each species. It is generally acknowledged that the range of influenza virus subtypes in the Anseriformes (ducks, geese etc.) is the most extensive. Influenza virus infection in birds is called avian influenza, and similarly you have swine Influenza, equine influenza and so on. It was customary in the past to assign the name of the animal from which the virus was isolated to the H and N subtype nomenclature¡Afor example¡AHav2Neq2¡Ameaning the second H subtype isolated from birds and the second N subtype isolated from horses. Since these subtypes could occur in various kinds of mammals and birds, a system was adopted subsequently in 1998 to unify the nomenclature of influenza viruses without designating the host origin. Thus, the original Hav2Neq2 became H10N8.

The first report of highly pathogenic influenza in chickens was from Italy in 1878, and the disease was called fowl plague. It was not until 1955 that the etiology was identified as an influenza virus¡Acarrying the type A influenza virus nucleocapsid protein. In order to differentiate the disease caused by it from another clinically similar disease originating from Asia but with a paramyxovirus as etiology, the former was named European Fowl Pest or Fowl Plague and the latter Asiatic Fowl Pest, Pseudo Fowl Plague or Newcastle Disease. Waterfowls infected with various kinds of influenza viruses¡Agenerally show no clinical symptoms; however, in chickens some virulent strains (highly pathogenic virus strain) may give rise to very high mortality¡Awhereas some (low pathogenic strains) do not or seldom give rise to death. Therefore¡Ahighly pathogenic avian influenza actually indicates the disease in chickens. Lest people equate highly pathogenic avian influenza to avian influenza¡Athe OIE showed reluctance in dropping the name "Fowl Plague" until lately. More than a thousand strains of influenza viruses from birds have been isolated, and the virus and/or antibodies detected in at least 50 species; up to now all the surface antigens of type A influenza virus have been found in birds. Guo Yuanji et al. have reported wide distribution of influenza virus in wild birds of China (1, 2, 3, 17).

The first case of human infected with H5N1 avian influenza virus was reported in 1997 from Hong Kong and up till February 2003, H5N1 avian influenza outbreaks continue to cause massive mortality in chicken farms, live poultry retail stalls and wholesale establishments. However, all the isolates were identified in the laboratory as low pathogenic avian influenza virus. Then why did these low pathogenic avian influenza virus lead to such high mortality¡H In this study, synergistic pathogenicity was especially looked at in an attempt to elucidate this phenomenon.

1. Material and Methods
1.1 Material
Avian Influenza virus H5N1 subtype was collected from Hong Kong, the sites including chicken farms, sewage canals near the farms, Hong Kong Mipu Bird Reservation, chicken retail stands, live chicken wholesale establishments, zoo and hospitals. PCR primers for amplification of various poultry viral gene fragments were designed and kept in this laboratory. Molecular biology reagents, bacteria antibiotic sensitivity test reagents, experimental animals, sterile bench and other virological and bacteriological instruments were conventional utilities of this laboratory. Bacteria collecting gadget is a self-made product, and the microbe/cell co-cultivation device is a patented invention of this laboratory.

1.1.2 WCK (Waisees Canine Kidney) cell line, a canine kidney cell line free of latent virus established by this laboratory.

1.2 Methods

1.2.1 Virus/bacteria synergism: Primer design, virus purification, virus nucleic acid extraction, PCR analysis and indirect ELISA, bacteria identification and antibiotic sensitivity test were done as reported previously (6-17). Raw material identified by PCR to contain avian influenza virus was further used in the following experiments.

1.2.1.1 Method used to assess the effect of concurrent bacterial infection on virus pathogenicity: Bacteria in the raw sample were isolated and cloned, then each representative bacterial clones was co-cultivated in cell culture with the avian influenza virus originating from the same raw sample, the cell culture procedure being as reported in (15). The bacteria was inoculated into a culture chamber so that the bacteria were separated by a 0.22 micron pore size millipore membrane from the animal cells and preventing their getting into direct contact with each other, the detailed procedure being:

(1)¢ÈEach bacterial sample was streaked onto nutrient agar plate, incubated at 37 „aC for 12 hours, after which isolated colonies were picked for identification.

(2¡^WCK cell line was seeded into 42 Koch's flasks using 1640 cell culture medium, and when the monolayer showed 80% confluence, 20 avian influenza virus isolates were separately inoculated into the culture flasks.

Group A: Ten specially fashioned bacterial culture chambers inoculated separately with each of ten bacterial clones, was placed in the culture medium of 10 flasks containing WCK cell monolayers, allowed to continue incubating at 37 C before removing the chambers, after which the WCK cells were further incubated for 36 hours.

Group B¡GTen avian influenza virus isolates were separately inoculated into 10 cell culture flasks¡Aand allowed to continue incubation at 37 „aC for 48 hours.

Group C¡GTen bacterial culture chambers each inoculated with a bacterial clone were transferred into 10 separate cell culture flasks, allowed to incubate at 37 „aC for 12 hours before removing the chambers, then continuing incubation for another 36 hours.

Group D: Ten cell culture flask without any pathogen inoculated was kept in incubation to serve as controls.
Each group of cell culture was examined for the appearance of any cytopathic effect (CPE).

1.2.2 Bacteria drug sensitivity test was done as previously reported (18).

1.3 Assessment
Based on the appearance of CPE in cell culture inoculated with low pathogenic avian influenza virus, the following conclusion could be arrived at.

?.. Should only cell cultures inoculated with bacteria secreting substilin-like protease show CPE, whereas those inoculated with bacteria secreting trypsin-like protease did not show CPE, it would indicate that the high mortality in the chicken outbreak under study was caused by concurrent infection of substilin-like secreting bacteria.¡F

B. Should only cell cultures inoculated with bacteria secreting trypsin-like protease show CPE, whereas those inoculated with bacteria secreting substilin-like protease did not show CPE, it would indicate that the high mortality in the chicken outbreak under study was caused by concurrent infection of trypsin-like secreting bacteria.

C. Should only cell cultures inoculated with bacteria secreting substilin-like and trypsin-like protease show CPE, whereas those without bacterial inoculation did not show CPE, it would indicate that the high mortality in the chicken outbreak under study could have been caused by concurrent infection of substilin-like or trypsin-like secreting bacteria.

Should the cell cultures that were only inoculated with influenza virus show CPE¡Ait would indicate that the high mortality in the chicken outbreak under study was caused by highly pathogenic influenza virus.

2. Results

2.1 The bacterial isolates being tested could be categorized into 3 major groups¡G¡]1¡^bacteria secreting trypsin-like protease¡]2¡^bacteria secreting substilin-like protease¡]3¡^unclassified bacteria secreting pathogenicity enhancing substances¡]table 1).

Table 1

2.2 Most of the bacterial strains whether secreting trypsin-like or substilin-like proteases were found sensitive to antibiotics (table 2).

Table 2

2.3 Virus-bacteria synergism findings

Group A All 10 WCK cell culture flasks, in the presence culture chambers containing either trypsin-like secreting or substilin-like secreting bacteria, developed CPE.

Group B The 10 WCK cell culture flasks, which had only been inoculated with avian influenza virus, did not show CPE.

Group C The 10 WCK cell culture flasks, in which only bacteria culture chambers containing purified isolates had been placed, did not exhibit CPE.

Group D The 10 WCK cell culture flasks, in which neither virus nor bacteria was inoculated, did not show any CPE.

3. Discussion

The present protocol adopted by OIE for differentiating highly pathogenic avian influenza viruses from those of low pathogenicity include the following¡G

(1) When 0.2 ml of a 1:10 dilution of infected chicken embryo allantoic fluid is injected intravenously into eight 6-week-old SPF chickens¡Aand mortality reaches up to 75%, the virus is considered highly pathogenic.

(2) When the virus is inoculated onto chicken fibroblast culture and CPE is produced¡Athe virus strain is considered highly pathogenic. However, if the cells need to be pretreated with trypsin before CPE can be generated, the strain is one of low pathogenicity.

(3) H5 and H7 isolates that are deemed of low pathogenicity by the above laboratory tests, should be further subjected to nucleotide sequencing of the HA gene cleavage site, and the deduced amino acid sequence analyzed to see whether it conforms to that of highly pathogenic strains (the -6 to -1 loci being most critical). Should only 2 or less basic amino acid be present there¡Aor although there are 3 basic amino acid but the amino acid at the -2 locus is not a basic amino acid¡Athen the strain is one of low pathogenicity. Should all at the - 4 to - 1 loci are basic amino acids or even though the one at -2 locus is not, but the other five are, then the strain is considered as highly pathogenic. Below are representative amino acid arrangements:

-6 -5 -4 -3 - 2 -1 G
High pathogenicity R K R K T R G
High pathogenicity K K K R G
Low pathogenicity R E T R G
Low pathogenicity R K T R G
N.B.: (1) The basic amino acid are: lysine K¡Aarginine R¡Ahistidine H (2) T = threonine, a polar amino acid (3) G= glycine¡Aa nonpolar amino acid
If the virus is a H5 or H7 subtype and the amino acid sequence at the hemagglutinin cleavage site is KRRR/G, then the virus belongs to a highly pathogenic strain (19).
The technical shortcoming of the current procedure is the difficulty in determining whether high mortality in a chicken flock is caused by a mildly pathogenic avian influenza virus per se or due to synergistic bacterial complication. So in view of the fact that most of the influenza virus isolated from high mortality flocks have been found to be mildly pathogenic, a definite diagnosis would be a lengthy process.

In reality it has been observed that the influenza virus occurring in chicken flocks mainly belong to mildly pathogenic strains, yet occasionally outbreaks with high mortality may take place.
Consequently, despite the laboratory proving the virus to be a mild strain, measures to control the disease would be as stringency as if one were dealing with a highly pathogenic influenza outbreak (such as the chicken massacre incident in Hong Kong)¡Aand this not only would give rise to huge losses in economy and foreign trade, but also burden the region with a long term negative image.

From the discovery in Hong Kong of the first case of human infected with H5N1 avian influenza virus in 1997 to February of 2003, in all over 20 cases have occurred. Since the human population lacked antibodies to the new influenza virus subtype, one would anticipate a pandemic within a short space of time, but that did not materialize, why? This can be attributed to the inclination for clinicians in general to use broad-spectrum antibiotics. Under the effect of broad-spectrum antibiotics the complicating bacteria secreting trypsin-like or substilin-like proteases are suppressed, and so those progeny viruses lacking the enzymes for infectivity cannot replicate, with the result that the range of infected cells become greatly restricted. Table 1 and Table 2 show 4 unclassified antibiotic-resistant bacteria, which can, however, produce proteolytic enzyme able to act on the hemagglutinin cleavage site. This may lead to speculation whether or not the HA cleavage site is identical to that for trypsin or substilin, but more of concern would be the fact that the benefit from supportive medication would be greatly reduced. This finding indicate an interesting subject for future research. In this study¡Acloned bacteria inside a bacterial culture chamber was placed into each of the group A and group C cell culture flasks, incubated at 37 „aC for 12 hours, after which the chambers were removed, this procedure being adopted to ensure that enough bacterial enzyme would be secreted while not depriving the cells of nutrients.

References

(1) Gan Menghou¡A1995. Avian Influenza (Chi). Beijing¡GBeijing Agriculture
University Press

(2) Yin Zhen¡ALiu Jinghua, 1997. Animal Virology (Chi)¡GScience and Technology
Press. p 704-735_

(3) Guo Yuanji. 1981. Preliminary Survey on Distribution of Influenza A Virus
Within Animals in China (Chi). Acta Microbiologia¡A21(3): 379 - 384

(4) Yu Kangzhen¡AChen Hualan, Tang Xiuying. 1998. The 1997 Hong Kong Avian
Influenza Outbreak (Chi). Chinese Journal of Animal and Poultry
Infectious Diseases,

(5) Li Haiyan, Yu Kangzhen, Xin Xiaoguang. 2001. Development and Application
of an Avian Influenza Indirect ELISA diagnostic Kit (Chi). Chinese
Journal of Preventive Veterinary Medicine. 23¡]5¡^¡G372-376

(6) Zhu Xinchan, Zhang Yong, Liao Xiangru. 1998. PCR Technological
strategy¡]Chi¡^. Biotechnology Information¡]3¡^¡G69-33

(7) Wu Hongzhuan¡AHe Dongsheng, Yang Dewei, Chen Feng, Qin Zhifeng, Wang Xing,
Zhu Daozhong, Liu Gongping, Li Huiyan, Liu Fuan. Detection of
infectious laryngotracheitis antibody in immunized chicken flocks
with indirect ELISA (Chi). Proc. Ninth Avian Medicine Subsociety
Academic Workshop of the Chinese Animal Husbandry & Veterinary
Science Society, 1998,10,21. P 201

(8) Wu Hongzhuan¡AHe Dongsheng¡AYang Dewei¡AChen Feng¡AQin Zhifeng, Wang Xing,
Zhu Daozhong, Liu Gongping, Li Huiyan, Liu Fuan. Detection of
infectious laryngotracheitis antibody in immunized chicken flocks
with indirect ELISA (Chi). Chinese Journal of Veterinary Science
and Technology. 1999, 29 ( 1) :25 - 26

(9) Waisees Yeung¡]Yang Dewei)¡ASong Yanhua¡ALiu Fuan¡C 2000. Application of
nested PCR to reveal canine parvovirus in MDCK cell lines¡]Chi).
Journal of South China Agricultural University. 3¡]21¡^¡G 81-83

(10) Waisees Yeung (Yang Dewei)¡AYang Lin¡ALiu Fuan et al. 2000. First report
of a pulmonary canine parvovirus infection and its control¡C
Journal of Zhongshan University¡C3¡G 125-128

(11) Waisees Yeung¡]Yang Dewei), Liu Fuan. 2000. Comments on ELISA testing of
canine viruses¡]Chi). Guangdong Journal of Animal and Veterinary
Science 25¡]11¡^¡G 1111-11410.

(12) Waisees Yeung¡]Yang Dewei)¡ALiu Fuan. 2000. Efficacy testing of various
canine parvovirus vaccines in clinical usage¡]Chi). Journal of
Animal and Veterinary Science. 25¡]2¡^¡G18-21

(13) Waisees Yeung¡]Yang Dewei)¡A Liu Fuan¡C 2000¡C Dog farm mosquitoes and
parvovirus (Chi). Guangdong Journal of Animal and Veterinary
Science 25¡]4): 16 -18

(14) Yang Dewei¡ALiu Fuan. 2001. Gene Vaccine and Antibiotic Resistance
Disease (Chi). Guangdong Journal of Animal and Veterinary Science
26¡]1¡^¡G31-32

(15) Yang Dewei¡A2001. Study on Parvovirus. Doctorate Thesis

(16) Song Changxu, Liu Fuan. 1995. Preliminary study on using PCR to differentiate virulent from mild strains of Newcastle disease virus (Chi).
Chinese Journal of Veterinary Science and Technology. (10): 20

(17) Huang Gengming.1999. Establishment of molecular biology diagnostic
technique to differentiate AIV, NDV and ILT (Chi). Doctorate Thesis.

(18) Waisees Yeung(Yang Dewei), Zhu Qishun, He Jianguo, Liu Fuan 2002 Report
on the detection of antibiotic resistant Proteus mirabilis contaminant
in Intervet canine vaccine product. Thedogplace\doglevel\dogcare

(19) Senne D A, Panigraphy B, Kawaoka Y, Pearson J E, Suss J et al. 1996.
Survey of the HA cleavage site sequence of H5 and H7 Avian
Influenza viruses: Amino acid sequence at the HA cleavage site as a
marker of pathogenicity potential. Avian Diseases, 40(2): 425-437

(20) Akbari O, Panjwani N, Garcia S. 1999. DNA vaccination transfection and
activation of dendritic cells as key events for immunity.
J Exp Med. 189:169-178

(21) Barbro S, Wilfried K, Uwe T. 1995. A simple touch-down polymerase Chain
reaction for the detection of canine parvorvirus and Feline
panleukopenia virus in feces. Journal of Virological Methods.
55: 427-433

(22) Lobell A, Weissert R, Storch MK. 1998. Vaccination with DNA encoding an
immunodominant myelin basic protein peptide targeted to Fc of immunoglobin G suppress experimental autoimmune encephalomyelitis.
J Exp Med: (9)1543-1548

Appendix: bacteria and cell culture co-cultivation device
This is an in vitro co-cultivation device, which comprises a cell culture flask and a bacterial culture chamber, the latter being an enclosed structure that can permit exchange of culture medium through a 0.22-micron pore size millipore membrane (34).

According to patent required description of the device (4) the bacterial culture chamber (3) features an ¡§a¡¨wall (31) and a ¡§b¡¨wall (32) compressing a ring-shaped seal (33) kept in position by screws (35) to make an enclosed structure. On the ¡§b¡¨wall is small opening (36) closed with a 0.22-micron pore size millipore membrane (34), which prevents bacteria from getting out the chamber, while allowing free passage of bacterial secretion and culture medium.

According to patent required description of the device (5), the ¡§a¡¨ wall and ¡§b¡¨wall as well as positioning screws can be made of stainless steel or heat-stabile plastics, and the ring-shaped seal can be of heat-stabile non-toxic rubber or plastics.

Procedures for nested PCR of purified virus (see references 2, 5, 6, 7, 8, 9, and 10).

Email : waisees@hotmail.com