Virus Genes

, Volume 45, Issue 2, pp 283–294

Complete genome characterization of avian influenza virus subtype H9N2 from a commercial quail flock in Egypt

Authors

    • National Laboratory for Veterinary Quality Control on Poultry ProductionAnimal Health Research Institute
  • Naglaa Hagag
    • National Laboratory for Veterinary Quality Control on Poultry ProductionAnimal Health Research Institute
  • Ahmed Erfan
    • National Laboratory for Veterinary Quality Control on Poultry ProductionAnimal Health Research Institute
  • Wessam Mady
    • National Laboratory for Veterinary Quality Control on Poultry ProductionAnimal Health Research Institute
  • Mohamed El-Husseiny
    • National Laboratory for Veterinary Quality Control on Poultry ProductionAnimal Health Research Institute
  • Amany Adel
    • National Laboratory for Veterinary Quality Control on Poultry ProductionAnimal Health Research Institute
  • Soad Nasef
    • National Laboratory for Veterinary Quality Control on Poultry ProductionAnimal Health Research Institute
Article

DOI: 10.1007/s11262-012-0775-0

Cite this article as:
Arafa, A., Hagag, N., Erfan, A. et al. Virus Genes (2012) 45: 283. doi:10.1007/s11262-012-0775-0

Abstract

The suspected presence of avian influenza virus subtype H9N2 in poultry in Egypt is a major concern since this subtype is widely distributed in different countries in the Middle East, here we describe the full genetic characterization of an avian influenza A virus (Qa/Egypt/11; H9N2) of subtype H9N2 that was previously isolated from a clinically normal quail flock in Giza, Egypt in May 2011. The nucleotide sequence analysis of the hemagglutinin gene of the isolated Egyptian virus showed the highest similarity with one group of recent Israeli strains (97 %) circulating from 2006–2010. Sequence homology and phylogenetic analysis indicated that the Qa/Egypt/11 isolate belonged to the A/quail/Hong Kong/G1/1997-like lineage with new mutations identified in all viral proteins. The phylogenetic analysis for the eight genes indicated placement of the Egyptian virus within the same lineage of H9N2 viruses that circulated in the region from 2006, especially with one group of recent Israeli strains. However, phylogenetic analysis of the internal genes like PB2, NP, and PA genes identified possible reassortment events for these genes with singular Israeli strains. This study indicates progressive evolution of this subtype in the Middle East region and possible mechanism of virus adaptation in land-based poultry like in quails.

Keywords

Low pathogenic avian influenzaH9N2 subtypeGenetic characterizationFull gene sequencingPhylogenetic analysis of Egyptian H9N2 virusCommercial quails

Introduction

The H9N2 subtype of avian influenza virus A (AIV), member of the Family Orthomyxoviridae, has been detected in many countries since its first isolation of the prototype strain, A/turkey/Wisconsin/1966 (Tu/WS/66), in the USA in 1966 [1]. Since the 1990s, outbreaks of H9N2 AIV in poultry have caused great economic losses in many countries in Asia and have transferred to the Middle East. Since then, the widespread and continuous presence of H9-subtype AIV throughout Europe and Asia has caused persistent problems for the poultry industry [2, 3]. Numerous outbreaks have been recorded since 1994 in Europe (Germany, Italy, and Ireland) [2], the Middle East (Iran, Pakistan, Israel, Jordan, Lebanon, United Arab Emirates and Saudi Arabia) [37] and Africa (South Africa and Tunisia) [8, 9].

H9N2 AIV is prevalent in chickens, ducks and other poultry species, but has demonstrated the ability to infect human beings [10]. Human infections with H9N2 viruses were reported in Hong Kong in 1999 as the first avian-to-human transmission; a subsequent case was reported in 2003 [11] and recent cases from Hong Kong and Bangladesh were also reported from 2008–2011 [12]. The isolated Influenza virus, A/chicken/Pakistan/2/99 (Ck/Pak/99), is closely related in all genes to the H9N2 viruses that caused human infections in Hong Kong [4]. H9N2 AIV infections in pig farms have also been confirmed in several provinces in China [13], suggesting that H9N2 AIV can cross the species barrier to infect other animals in addition to humans. Thus, transmission to other poultry species may cause significant genetic and antigenic changes [14].

Infection with H9N2 viruses often result in decreasing of the laying rate of hens, and co-infection with other viruses or bacteria can cause severe morbidity and high mortality in chickens [15, 16].

The H9N2 viruses isolated from quail in 1988 were the first evidence of H9 virus infection in land-based poultry in Asia [17]. Quails in Hong Kong have shown a high incidence of infection with influenza A viruses, particularly with H9N2 viruses [18]. Moreover, the wider host range of the avian H9N2 strains isolated from chickens, quail, pigeons, and ducks has been reported in the Hong Kong poultry markets in 1997 [19].

In China, H9N2 influenza viruses are prevalent mainly in terrestrial poultry such as chickens and ducks [20]. H9N2 influenza viruses previously isolated from poultry in China have confirmed that these viruses formed three sublineages represented by their prototype strains: A/quail/Hong Kong/G1/1997 (Qa/Hk/G1/97) or G1-like, A/duck/Hong Kong/Y280/1997 (Dk/Hk/Y280/97) or Y280-like and A/chicken/Beijing/1/1994 (Ck/BJ/1/94) or BJ94-like [21]. Since 1998, another common lineage of H9N2 viruses in eastern China was identified as the A/chicken/Shanghai/F/1998 (Ck/Sh/F/98) lineage (F 98-like lineage) [20].

Comparisons of hemagglutinin (HA) gene sequences showed that the H9N2 viruses isolated during 1998 and 1999 in Germany, Iran, Pakistan, and Saudi Arabia were closely related to those of the human and quail Qa/Hk/G1/97-like lineage isolates from Hong Kong and distinct from other phylogenetic lineages of H9. Further analysis of viruses isolated in Germany, Iran, and Saudi Arabia showed that they represent reassortants, deriving genes from other previously characterized lineages of H9N2 viruses [20]. The first waves of H9 virus infections in Israel were recorded in 2000–2003 where the outbreaks were detected in chickens, geese, ostriches, and turkeys. Phylogenetic analysis also revealed that all Israeli isolates fall into a single lineage, indicating a common origin and further subdivided into three closely related sub-groups [7].

In Egypt, a previous report was describing the isolation and identification of H9N2 virus from a commercial quail flock in May, 2011 from Giza province as well as the preliminary genetic characterization of partial HA and NA sequences has also been made [22]. Since then, extensive surveillance has been made to discover the limit of spread of this subtype in the commercial poultry sectors.

Previous unpublished, unofficial reports indicated the presence of H9N2 subtype in Egypt many years ago. Furthermore, the H9N2 has already been detected and sequenced from clinical samples collected from live bird markets in Egypt in 2003 and submitted through NAMRU3 to SEPRL, USA to complete virus characterization (Suarez, personal communication).

Therefore, the current study aimed to determine the complete genome of the strain A/quail/Egypt/113413v/2011(H9N2) or (Qa/Egypt/11) that was characterized by El-Zoghby et al. [22] from a clinically healthy quail flock in 2011 in Egypt and to identify a possible potential source of endemic H9N2 viruses and also to assess its genetic relatedness to worldwide strains.

Materials and methods

Virus propagation and molecular subtyping

The Egyptian AIV, A/quail/Egypt/113413v/2011(H9N2) or (Qa/Egypt/11), was detected in May 2011 from a white commercial clinically healthy quail farm during routine pre-slaughter surveillance for AIV [22]. Pooled swab samples were inoculated through the allantoic route of fertile chicken eggs and further antigenic subtyping was performed using specific antisera against different AIV subtypes using standard methods [23]. In this study, virus propagation was carried out in both fertile chicken eggs of 9 days old through allantoic inoculation and in Madin-Darby canine-kidney (MDCK) cells in the presence of 2.5 μg/mL trypsin treated with l-1-(topsylamido-2-phenyl) ethyl chloromethyl ketone (TPCK) according to a standard protocols [23]. Confirmation of the presence of virus after isolation has been conducted using real-time reverse transcription-polymerase chain reaction (RT-PCR) for Matrix (M) and HA genes. RNA for RT-PCR was extracted from the chorioallantoic fluid of inoculated eggs and from supernatants of infected MDCK cells 72 h after infection. The extraction of viral RNA was performed using a QiaAmp viral RNA mini kit (Qiagen, Germany) according to the manufacturer’s instructions. Extracted RNA was used for cDNA synthesis using the universal 12-mer Uni12 primer, 5′-AGCAAAAGCAGG-3′ [24]. Real-time RT-PCR was performed using a one-step method for avian influenza A virus M and H9 genes, using specific primers for H9 subtype; ATGGGGTTTGCTGCC, TTATATACAAATGTTGCAYCTG and probe FAM-TTCTGGGCCATGTCCAATGG-TAMRA using 6-carboxyfluorescein (FAM) as a fluorescent reporter dye at the 5-end and 6-carboxytetramethylrhodamine (TAMRA) as a quencher dye at the 3-end as previously described [25, 26]. The real-time RT-PCR reactions were performed on a Stratagene thermocycler machine. Confirmation of the absence of other influenza virus subtypes (H5 and H7) were also performed using the real-time RT-PCR standard protocols [27, 28].

Nucleotide sequence determination

Determination of nucleotide sequences for the eight gene segments of the Egyptian strain Qa/Egypt/11 was carried out for each amplified gene using a BigDye Terminator Kit (version 3.1; Applied Biosystems, Foster City, CA) on a 3130 Genetic Analyzer (Applied Biosystems, Foster City, CA), using specific internal primers for full-length coding gene sequencing. Primer sequences for the eight segments were described in Supplementary Table 1. The nucleotide and amino acid sequence analysis was performed for all eight genes of the isolated Egyptian strain Qa/Egypt/11 in comparison with other related strains, including the coding regions as follows: PB2, 29-2259; PB1, 37-2165; PA, 30-2119; HA, 1-1683; NP, 1-1456; NA, 27-1367; M, 1-976; and NS, 34-838. Sequence data were submitted to the GenBank database under accession numbers: JN828567-JN828574 for the PB2, PB1, PA, HA, NP, NA, MP, and NS genes, respectively.
Table 1

Nucleotide sequence identity of eight genes from Egyptian virus Qa/Egypt/11(H9N2) compared to those from other influenza virus isolates

Gene segment

Qa/Hk/G1/97

Dk/Hk/Y280/97

Hk/1073/99

Qa/Db/338/01

Tu/Is/1608/06

Ck/IS/182/08

A/ck/Is/375/07

PB2

86

82.2

86.5

85.2

94.3

94

96.5

PB1

88.6

87.3

88.3

89.2

97.5

97

95.9

PA

85.8

85.6

85.5

90.3

94.8

94.7

97.3

HA

90.9

85.9

90.8

90.2

97.5

97

89.5

NP

89.8

88

89.7

88.9

96

95.3

98

NA

92

89.4

91.9

92.4

97.8

97.3

89.7

M

94.2

92

94.5

93.3

94.2

98.3

93.2

NS

86.5

87.6

86.4

91.1

97

97.5

88.4

Percent identity was calculated based on comparison of the Egyptian virus with other strains for all eight gene segments, the nucleotides from the start codon were (PB2, 29–2259; PB1, 37–2165; PA, 30–2119; HA, 1–1683; NP, 1–1456; NA, 27–1367; M, 1–982 and NS, 34–838). The highest homology indicated in bold

Genotyping and phylogenetic analysis of Qa/Egypt/11

Sequences used for comparison in this study were obtained from GenBank and were available from the National Center for Biotechnology Information (NCBI) Influenza Viruses Resource (http://www.ncbi.nlm.nib.gov/genomes/FLU/Database). Sequences for the eight genes of strains: A/quail/Hong Kong/G1/1997 (Qa/Hk/G1/97), A/duck/Hong Kong/Y280/1997 (Dk/Hk/Y280/97), A/Hong Kong/1073/1999 (Hk/1073/99), A/chicken/Dubai/338/2001 (Ck/Db/338/01), A/chicken/Israel/182/2008 (Ck/IS/182/08), A/turkey/Israel/1608/2006 (Tu/IS/1608/06) and A/chicken/Israel/375/2007 (Ck/IS/375/07) [5, 17, 20, 29, 30] were used in this analysis for homology comparison.

The strains used for genetic analysis were of Asian origin (from Hong Kong, China, Pakistan, United Arab Emirates, Saudi Arabia, Lebanon and Israel) and African origin (from South Africa and Tunisia). The Genbank nucleotide accession numbers for the eight genes for all strains used in the analyses were described in Supplementary Table 2. Nucleotide and amino acid sequence analyses were carried out for the eight gene segments of the isolated Egyptian virus Qa/Egypt/11. The pair-wise nucleotide percent identity matrix for H9N2 isolates from Egypt and other countries was calculated using the CLUSTAL-V algorithm in the MegAlign program of the Lasergene software suite (DNASTAR, Madison, Wisconsin). Multiple sequence alignments of nucleotide and deduced amino acid sequences were generated using the CLUSTAL-W program [31] with a gap penalty of 15. Phylogenetic trees were constructed un-rooted for the eight full-length genes using the neighbor-joining method [32], the robustness of the phylogenetic analysis and significance of the branch order were determined by bootstrap analysis with 1,000 replicates using the MEGA4 analysis software [33].
Table 2

Amino acid substitution signatures in HA and NA proteins of avian influenza viruses

Strain

HA connecting peptide

HA1 GS

HA2 GS

NA Stalk deletion

NA GS

Qa/Egypt/11

A-R-S-S-R

11, 64, 87, 123, 280, 287

474

44, 61, 69, 86, 146, 200, 234

Qa/Hk/G1/97

A-R-S-S-R

11, 64, 87, 123, 200, 280, 287

474

38–39

61, 69, 86, 146, 200, 234, 402

Dk/Hk/Y280/97

A-R-S-S-R

11, 64, 123, 200, 280, 287

474

63–65

69, 86, 146, 200, 234, 402

Dk/Hk/Y439/97

A–A-S-N-R

11, 64, 123, 200, 280, 287

474

61, 69, 86, 146, 200, 234, 308, 402

Hk/1073/99

A-R-S-S-R

11, 64, 87, 123, 200, 280, 287

474

38–39

61, 69, 86, 146, 200, 234, 402

A/turkey/Israel/90710/2000

A-R-S-S-R

11, 64, 87, 123, 280, 287

474

61, 69, 146, 200, 234, 331, 402

Ck/Db/338/01

A-R-S-S-R

11, 64, 87, 123, 200, 280, 287

474

61, 86, 146, 200, 234, 402

A/chicken/Israel/909/2005

A-R-S-S-R

11, 64, 87, 123, 280, 287

474

61, 69, 146, 200, 234, 331, 402

A/avian/Saudi Arabia/910134/2006

A-R-S-S-R

11, 64, 87, 123, 280, 287

474

61, 69, 146, 200, 234, 402

A/turkey/Israel/1608/2006

A-R-S-S-R

11, 64, 87, 123, 280, 287

474

44, 61, 69, 146, 200, 234, 402

A/chicken/Israel/375/2007

A-R-S-S-R

11, 64, 87, 123, 280, 287

474

61, 69, 146, 200, 234, 331, 402

Ck/IS/182/08

A-R-S-S-R

11, 64, 87, 123, 280, 287

474

44, 61, 69, 86, 146, 200, 234, 402

A/chicken/Israel/184/2009

A-R-S-S-R

11, 64, 87, 123, 280, 287

474

44, 61, 69, 146, 200, 234, 402

A/quail/Lebanon/272/2010

A-R-S-S-R

11, 64, 87, 123, 280, 287

474

44, 61, 69, 146, 200, 234, 402

Numbering was according to H9N2 of strain Tu/WS/66

GS glycosylation site

Results

Virus characterization

To study the molecular genetics of the Egyptian virus Qa/Egypt/11, virus propagation was performed in fertile chicken eggs and on MDCK cells treated with trypsin TPCK. The results of the real-time RT-PCR for confirmation of AIV in the harvested allantoic and cell culture fluids were positive for the M gene. Further subtyping by real-time RT-PCR was negative for both H5 and H7 subtypes, but was positive for the H9 subtype for this isolate.

Genotyping and phylogenetic analysis

Nucleotide sequence analysis of the HA gene revealed the closest genetic similarity to the Israeli strains (Ck/IS/182/08 and Tu/IS/1608/06) among other Afro-Asian viruses that circulated in the Middle East region from 2006 to 2011. The percent identity based on sequences (1,683 nucleotides) encoding amino acids 1-560 of HA was 97 and 97.5 % for the two Israeli strains (Table 1). The HA homology was reached to nearly 90 % for strains Qa/Hk/G1/97, Hk/1073/99, and Ck/Db/338/01, while the lowest homology (85.9 %) was observed with the reference strain Dk/Hk/Y280/97 (Table 1).

The percent identity for other genes of the Egyptian strain was as high as 98.3 and 97.5 % for the M and NS genes, respectively with strain Ck/IS/182/08, while with strain Tu/IS/1608/06 the highest gene homology reached 97.8 % for the NA gene and 97.5 % for the PB1 gene. Whereas the homology for NP, PA, and PB2 genes was 98, 97.3, and 96.5 %, respectively with the Israeli strain A/chicken/Israel/375/2007. This indicates possible reassortment events between pools of Israeli viruses and may correlate with the origin of Egyptian strain.

The homology between Qa/Egypt/11 and A/chicken/Dubai/338/2001 was ranged from 85.2 (for PB2 gene) to 93.3 % (for M gene). The lowest identity among genes was found for the PB2 gene while the highest identity was for the M gene among all strains used in this comparison (Table 1).

The nucleotide identities of Qa/Egypt/11 compared with the reference strains from Hong Kong (Qa/Hk/G1/97, Dk/Hk/Y280/97, and Hk/1073/99) ranged from 82.2 (for the PB2 gene) to 94.5 % (for the M gene). The lowest identities were observed with the strain Dk/Hk/Y280/97 for the two genes PB2 and M (82.2–92 %) while the identities with the other two strains (Qa/Hk/G1/97 and Hk/1073/99) were similar and ranged from 85.5 % for the PA up to 94.5 % for the M genes.

Phylogenetic analysis showed that the Egyptian virus, Qa/Egypt/11, grouped in the Qa/Hk/G1/97-like lineage and shared a common ancestor with the Qa/Hk/G1/97 isolate (Fig. 1). Figure 1a shows the phylogenetic relationships between the HA gene of the recently isolated H9N2 virus from Egypt and other regional and international strains. The HA phylogenetic tree placed Qa/Egypt/11 in one group with Israeli strains and other related strains that circulating in the Middle East region (Lebanon, Saudi Arabia, Dubai, Pakistan, and China) within the Qa/Hk/G1/97-like lineage. However, the phylogenetic trees for other internal genes placed the Egyptian strain Qa/Egypt/11 with one Israeli group that contains Ck/IS/182/08 and Tu/IS/1608/06, except for the PB2, NP, and PA trees that placed Qa/Egypt/11 closer to other Israeli group that contains strain A/chicken/Israel/375/2007 (Fig. 1b–h).
https://static-content.springer.com/image/art%3A10.1007%2Fs11262-012-0775-0/MediaObjects/11262_2012_775_Fig1a_HTML.gif
https://static-content.springer.com/image/art%3A10.1007%2Fs11262-012-0775-0/MediaObjects/11262_2012_775_Fig1b_HTML.gif
https://static-content.springer.com/image/art%3A10.1007%2Fs11262-012-0775-0/MediaObjects/11262_2012_775_Fig1c_HTML.gif
Fig. 1

Phylogenetic analysis for the eight gene segments for the Egyptian strain, Qa/Egypt/11 in comparison to other strains. Genes analyzed were a HA, b NA, c PB2, d PB1, e PA, f NP, g M, and h NS

Molecular characterization of amino acids for all gene segments

HA

The length of the putative HA protein of Qa/Egypt/2011 was predicted to be 560 amino acids with a putative cleavage site motif consisting of HA1−317RSSR/GLF323−HA2, which is similar to most H9N2 AIV, including those from Pakistan, Saudi Arabia, Israel, UAE, and Hong Kong.

The current article indicated that the putative receptor-binding domain (RBD) of Qa/Egypt/11 contained, Y91 (H3:101), W143 (H3:153), T145 (H3:155), H173 (H3:183), A180 (H3:190), L184 (H3:194) and Y185 (H3:195) Supplementary Fig. 1. In addition, the receptor-binding site (RBS) at 215–218 (H3:225–228) of Qa/Egypt/11 contained “GLIG” that is similar to some strains from the Middle East (Saudi Arabia, Lebanon, and Israel) but is dissimilar to strain Ck/Db/338/01 that consisted of “GLMG”, and for strain Qa/Hk/G1/97 that was “DLQG”.

The Egyptian strain Qa/Egypt/2011 was predicted to contain seven glycosylation sites with the N-X-S/T motif, where “X” may be any amino acid except proline. Six of these sites were located in the HA1 region (11, 64, 87, 123, 280, and 287) and one site at 474 in the HA2 region. The Egyptian virus lacks one potential glycosylation site at position 200, similar to Israeli strain Ck/IS/182/08, but unlike other Asian strains Qa/Hk/G1/97, Dk/Hk/Y439/97, Dk/Hk/Y280/97, Qa/Db/338/01, and Hk/1073/99 (Table 2). Interestingly, the two additional potential glycosylation sites predicted in the Indian isolates at amino acid positions 131 and 140 [34], were not found in the Egyptian H9N2 isolate and that may add an importance to the mechanism of independent virus evolution of this subtype in Egypt.

NA

The neuraminidase stalk region of Qa/Egypt/11 did not acquire a two-residue deletion (38–39) in the stalk of NA. Also, it did not have a deletion at 63–65. The NA genes of the Egyptian H9N2 virus showed seven glycosylation sites at positions 44, 61, 69, 86, 146, 200, and 234 (Table 2). The sialic-acid-binding pocket on the hemadsorption (HB) sites (366–373, 399–404, and 431–433) of the NA protein of Qa/Egypt/11 predicted to be IKKDSRAG, DSDSWS and PQE.

PB2

The PB2 protein of Qa/Egypt/11 contains E627V substitution (Table 3). In addition, 15 distinct mutations were identified in comparison to ancestor strain Qa/Hk/G1/97, including I129T, V292T, R293W, V338I, E472D, V495A, R508Q, E543D, M570I, M575L, E627V, S643T, T683A, S714G, and V731I that characterizing the Egyptian isolate and some of recent Israeli strains from other H9N2 strains in this comparison.
Table 3

Putative amino acid differences of NP, PB2, PB1 and PB1-F2 proteins of Egyptian strain Qa/Egypt/11

Strain

NP

PB2

627

PB1

317

PB1-F2

214

371

373

430

66

82a

A/quail/Egypt/113413v/2011

R

M

T

A

V

I

S

L

A/quail/Hong Kong/G1/97

K

V

A

K

E

I

N

L

A/duck/Hong Kong/Y280/97

R

M

T

T

E

V

N

A/duck/Hong Kong/Y439/97

R

M

T

T

E

M

N

L

A/chicken/Shanghai/F/98

R

M

A

T

E

M

N

L

A/Hong Kong/1073/99

R

V

A

K

E

I

N

L

A/chicken/Pakistan/2/99

R

V

A

K

E

I

N

L

A/turkey/Israel/90710/2000

R

T

T

T

V

M

S

S

A/chicken/Dubai/338/2001

R

V

A

K

E

M

N

L

A/chicken/Israel/909/2005

R

M

T

A

V

M

S

S

A/turkey/Israel/1608/2006

R

M

I

A

V

M

S

S

A/avian/Egypt/920431/2006

R

M

T

T

E

M

N

S

A/chicken/Israel/375/2007

R

M

T

A

E

M

S

S

A/chicken/Israel/182/2008

R

M

I

A

V

M

S

S

A/chicken/Israel/184/2009

R

M

T

A

V

M

S

S

A/quail/Lebanon/272/2010

R

M

T

T

E

M

N

S

aThe site at 82 in PB1-F2 was absent in the strain A/duck/Hong Kong/Y280/97 due to premature stop codon

PB1

The Egyptian strain, Qa/Egypt/11, has multiple mutations detected and distributed throughout the entire PB1 protein, such as V114I, S152L, E178K, T182I, V200I, R211G, R386K, E390M, L395I, G399S, and S633T. These mutations characterizing the Egyptian isolate as well as they were also shown in strains from Israel [30], all these mutations accumulated in the PB1 gene supposed to play a role in the function of the polymerase during virus replication. The Egyptian virus had isoleucine at site 317 like the precursor strain Qa/Hk/G1/97 and the human virus A/Hk/1073/99 while the Israeli viruses had methionine at this position (Table 3).

PB1-F2

In this study, the deduced amino acids sequence of the Egyptian H9N2 PB1-F2 gene was similar in length to Qa/Hk/G1/97, it was 90 residues. Some other mutations were characterizing the Egyptian virus and Israeli strains, P8L, T13I, Q28L, L62P, N66S, and E70V. The position at 82 had amino acid leucine like the original Qa/Hk/G1/97 strain, but unlike many strains from Israel which had Serine at this position (Table 3).

PA

The Egyptian strain, Qa/Egypt/11, has nine specific substitutions in the PA protein; V44I, S140T, G186S, L226F, N321K, K328R, M360L, S405A and K603R. These mutations were characterizing the Egyptian and Israeli strains.

NP

The Egyptian H9N2 virus carries isoleucine at position 183 like Qa/Hk/G1/97 and other strains from Israel, also the two mutations, K430A and R446 K, have shown to be very specific for the Egyptian and Israeli viruses in comparison to other H9N2 strains. The amino acid Arginine at position 214 (K214R) shown to be very specific for the Egyptian virus while it was lysine in other H9N2 viruses in this comparison (Table 3). Although there was no difference among strains at site 372, which contains glutamic acid (E), there were two mutations adjacent to it (V371M and A373T) present in the Egyptian virus. So, any alteration in this site can affect the pathogenicity to humans, as it may differentiate the avian from human viruses [35].

MP

There were seven substitutions present in the putative M1 protein in Qa/Egypt/11; P54S, I59V, R95K, T139N, T167A, M192V, V246L, while in the M2 protein there were three substitutions; I32V, I39V and V68M. The amino acid Isoleucine at position 15 in M1 protein was found in the Egyptian as well as other strains in this comparison except for the former Egyptian virus A/avian/Egypt/920431/2006, A/quail/Lebanon/272/2010 and A/Dk/Hk/Y439/97. The substitutions at positions 10, 14, 16, 20, 28 and 57 in the M2 protein that separated the avian from human viruses were indicated in Table 4. The S31N mutation in the M2 that confers resistance to the anti-influenza drug amantadine [36] was not found in the Egyptian H9N2 virus.
Table 4

Signature amino acid substitutions of M1, M2, NS1, and NS2 proteins of Egyptian strain Qa/Egypt/11

Strain

M1

M2

NS1

NS2

15

10

14

16

20

28

57

80

84

92

103

106

127

215

227–230a

31

70

A/quail/Egypt/113413v/2011

I

L

E

G

S

V

R

A

M

D

F

M

T

L

KSEV

M

S

A/quail/Hong Kong/G1/97

I

L

G

G

S

V

R

T

V

E

L

I

N

P

EPEV

M

S

A/duck/Hong Kong/Y280/97

I

H

G

G

S

V

R

T

V

D

L

I

N

S

EPEV

M

S

A/duck/Hong Kong/Y439/97

V

P

G

E

S

I

R

T

V

D

F

M

S

P

ESEV

M

S

A/chicken/Shanghai/F/98

I

P

G

G

S

V

R

T

V

D

L

I

T

S

M

S

A/Hong Kong/1073/99

I

L

G

E

R

V

R

T

V

E

L

I

T

P

EPEV

M

S

A/chicken/Pakistan/2/99

I

L

G

E

S

V

R

T

V

E

L

I

N

P

EPEV

M

S

A/turkey/Israel/90710/2000

I

L

E

G

S

V

R

T

V

D

F

I

N

P

EPEV

M

S

A/chicken/Dubai/338/2001

I

L

G

G

S

V

R

T

V

D

F

M

T

P

M

G

A/chicken/Israel/909/2005

I

L

E

G

N

V

R

T

V

D

F

I

N

P

EPEV

M

S

A/turkey/Israel/1608/2006

I

L

G

G

S

V

R

T

V

D

F

M

T

L

KSEV

M

S

A/avian/Egypt/920431/2006

V

P

G

E

S

I

R

T

V

D

F

M

N

P

ESEV

M

S

A/chicken/Israel/375/2007

I

L

E

G

N

V

R

T

V

D

F

I

N

P

EPEV

M

S

A/chicken/Israel/182/2008

I

L

G

G

S

V

R

T

V

D

F

M

T

L

KSEV

M

S

A/chicken/Israel/184/2009

I

L

G

G

S

V

R

T

V

D

F

M

T

L

KSEV

I

S

A/quail/Lebanon/272/2010

V

P

G

E

S

I

R

T

V

D

S

I

N

P

RSEV

M

G

aThe site 227–230 in NS1 was absent in the two strains A/chicken/Shanghai/F/98 and A/chicken/Dubai/338/2001 due to premature stop codon

NS

The putative non structural (NS1) protein of Qa/Egypt/11 was predicted to comprise the full-length NS1 protein (230 amino acids), however the partial sequence (amino acids from 12 to 230) was illustrated in this study. The NS1 protein of the Egyptian virus has the C-terminal amino acids (KSEV) that comprises the PDZ domain legend (PDZ-L) in position from 227 to 230. The NS1 protein of the Egyptian strain has additional four mutations, K62R, E142D, V194I and P215L while for the NS2 protein there was only one prominent mutation, L58F that characterizing the Egyptian and Israeli H9N2 strains. Based on the analysis of the NS1 protein, the recent Egyptian isolate Qa/Egypt/11 has mutations at different sites; T80A, V84M, E92D, L103F, I106M, N127T and P215L, while there were no characteristic mutations at sites 31 and 70 of the nuclear export protein (NS2) (Table 4).

Discussion

Previous studies of H9N2 isolates from south Asia and the Middle East have shown that these viruses produce significant disease in poultry, especially when poultry are co-infected with other respiratory pathogens, such as infectious bronchitis virus [3]. Genetic and phylogenetic analyses suggest that a stable lineage of AIV has become established in poultry in southern China [37].

The current article describes the follow-up study on Qa/Egypt/11 virus based on full genetic characterization and gene sequencing for the eight segments and also represents more extensive phylogenetic analysis than the previous study [22].

Genetic and phylogenetic analysis of the HA gene of Qa/Egypt/11 showed the closest overall relationship to the Israeli strains (Tu/IS/1608/06 and Ck/IS/182/08) in comparison with other H9N2 viruses that were previously circulating in the Middle East region from 2006 to 2010. It was also found about 96 % identity in amino acid sequences between the HA genes of the 2004 Israeli viruses and the 2000 and 2002 viruses, whereas the homology between the 2000 and 2002 Israeli viruses was closer to 99 % [38]. Previous studies of the HA genes of the viruses isolated in Germany, Iran, Pakistan, and Saudi Arabia during 1998 and 1999 also showed high genetic relatedness to the HA genes of the human and quail G1-like H9N2 viruses isolated in Hong Kong (96–98 % homology) [4]. These viruses were also more distantly related to the HA genes of A/chicken/Hong Kong/G9/1997 (92 % homology) and DK/Hk/Y439/97 (85 % homology) representing two other lineages of H9N2 viruses that circulating in China and Korea since 1994 [4].

The absence of multiple basic amino acids of HA gene suggests the low pathogenic nature of Qa/Egypt/11 virus. Indeed, this virus was isolated from clinically normal birds. While H9N2 viruses cause no signs of disease in healthy quails, these viruses may cause sub-clinical infection and can be transmitted to other poultry; therefore, signs of disease may be observed in the field, especially in cases of concurrent stress or co-infection with other pathogens [15].

The amino acid residues of HA1 involved in the RBD comprises seven residues [39]. All of them are highly conserved except one mutation at position 180 displayed variations among H9N2 viruses that could be valine (V), alanine (A) or thrionine (T) [40].

The strain Qa/Egypt/11 carried a substitution Q216L, while the RBD involved in the RBS of the avian-like motif “GQQG” was found similar to that of the prototype chicken H9N2 virus reported in northern China strain “BJ/94″ and also the prototype strain Tu/WS/66 [40]. This mutation correlates with a shift in affinity of the HA from the “avian” type sialic receptors to the human type and another substitution at position 218 (228 of H3) that parallel to the receptor-binding sites of H3 human isolates [5, 41]. Due to the locations of these mutations in the RBS, substitutions at these positions will affect virus interactions with cell-surface receptors.

Moreover, the substitution Q216G (226 by H3 Numbering) that was found in the receptor-binding site of the HA1 of Tu/WS/66 and confers a high binding affinity to 2,3-linked sialic acid (SA) found in birds, but low binding affinity to the 2,6-linked SA found in mammals [42].

Variations in both HA and NA, along with receptor specificity of different phylogenetic lineages of H9N2 viruses from Asia reflect the diverse ways in which viruses adapt to various species [43], as also has been reported previously that the adaptation of H9 viruses to land-based birds can lead to strains with expanded host range [44].

The presence of the potential glycosylation sites at the HA globular head is typical of poultry viruses, but not of aquatic bird viruses [41]. The Egyptian virus has six glycosylation sites at HA1 similar to strains from the Middle East, but unlike the original strain Qa/Hk/G1/97 which had additional site at position 200. Mutations at these positions may influence the antigenic specificity of these viruses and could play a role in antigenic variation [45].

The human receptor specificity and similar substitutions in the HB sites of the NA in recent H9N2 isolates suggest the possible role of poultry species such as chickens in zoonotic transmission of influenza viruses from aquatic birds to humans. Moreover, the HB sites in the NA of Asian H9N2 viruses are considered under positive selection pressure, and may result in compatible combinations of HA and NA [42]. Mutations in the HB site of NA in avian H9N2 viruses suggest that some species of poultry could serve as an intermediate hosts in the zoonotic transmission of influenza viruses from their natural reservoir in aquatic birds to humans [42].

The neuraminidase stalk region of Qa/Egypt/11 was shown to lack a two-residue deletion (38–39) in the stalk region, which is dissimilar to the NA of Qa/Hk/G1/97 and the human virus, Hk/1073/99. This deletion is also absent in other viruses from the Middle East region [39, 42]. In addition, Qa/Egypt/11 did not have a deletion at 63–65, unlike the strains Dk/Hk/Y280/97, Ck/Sh/F/98 and A/chicken/Beijing/1/2006 (Ck/BJ/1/06). The absence of these deletions may reflect establishment of the Qa/Hk/G1/97-like lineage in the Middle East independent of the original Qa/Hk/G1/97 strain, either before or at approximately the same time the A/Qa/Hk/G1/97-like lineage became prevalent in Hong Kong [4].

The NA genes of the Egyptian H9N2 virus showed seven glycosylation sites (44, 61, 69, 86, 146, 200, and 234) similar to strains Ck/IS/182/08, Ck/Db/338/01 and the human virus Hk/1073/99, (Table 2), but unlike strains Dk/Hk/Y280/97, Ck/Sh/F/98, and Ck/BJ/1/06. The N-linked glycosidic moieties have been found to play a vital role in mediating viral infectivity, receptor-binding capacity, and cell associated host immune responses, as well as protecting critical viral protein epitopes from immune attack [46]. Moreover, the sialic-acid-binding pocket on the HB sites of the NA protein of Qa/Egypt/11 from sites 336–373, 399–404, and 431–433 (IKKDSRAG, DSDSWS and PQE) was unlike the precursor strain Qa/Hk/G1/97 (IKKDSRSG, DSDIRS, and PQE) and for Dk/Hk/Y280/97 (IKEDSRSG, DSDNWS, and PQE), but it was very similar to the recent Israeli and Afro-Asian strains. The HB site is predicted to have a role in host–receptor interactions with the virions, so the changes in the HB site are believed to be important in host adaptation of a virus from avian to mammalian subtype [47].

The rapid evolution of H9N2 viruses circulating in the Middle East region can also be detected by studying the divergence among strains, particularly in the internal genes. It is well recognized that a single mutation, E627K, in the PB2 of H5N1 virus contributes to systemic infection and impaired T cell activation in mice [48]. The putative PB2 protein of Qa/Egypt/11 does not contain E627K; however, another substitution (E627V) in this site was found. In addition, 14 distinct mutations for Qa/Egypt/11 were identified and most of these substitutions were very specific for recent Egyptian strain as well as the Israeli viruses [30]. Further analysis is needed to study the effect of these mutations in the function of the PB2, which may reflect on the replication performance of the virus and multiplication inside the host as well as virus spread.

The putative PB1-F2 protein of the Egyptian strain “Qa/Egypt/11” was expressed in full-length. The expression of the full-length PB1-F2 gene increases the pathogenicity of influenza A virus. As well as the mutation N66S detected in the Egyptian H9N2 isolate has also been found in strains from Israel that may be associated with increased viral pathogenicity [49].

Furthermore, it has been reported that amino acid residues at positions 15, 115, 121, 137, 240 in virus matrix protein are linked with increased replication in mammals or increased pathogenicity in small animal models [34].

The internal genes of the Egyptian isolate, PB1, PA, M, and NP are similar to those of other Middle Eastern strains and most likely to the Israeli viruses, but showed somewhat divergence for the precursor strain Qa/Hk/G1/97 suggesting rapid evolution and adaptation in the Middle East region. The Egyptian virus Qa/Egypt/11 possessed some of the amino acid residues that were characteristic for the human and quail G1-H9N2 viruses [4, 5]. The substitution I317, found in the PB1 gene, showed similar to the precursor strain Qa/Hk/G1/97 and to the human virus A/Hk/1073/99, furthermore it showed to be characteristic for H9N2 and H5N1 of human origin [4].

The NP is mainly initiate encapsidation of the viral genome to form a ribonucleoprotein that necessary for transcription and packaging; it interacts with other viral PB1, PB2, M1, and cellular proteins. The arginine at position 214 in NP protein showed much conserved in the Middle Eastern strains including the Egyptian virus, also other sites like I109, Q157 and E372 are linked to avian type viruses [35]. However, the Egyptian virus carries two substitutions adjacent to the E372 (M371 and T373), that indicates possible pressure at that site to mutate. The site at 430 also was linked to human viruses as it contains lysine in strains A/Hk/1073/99 and Ck/Pakistan/2/99. Interestingly, this site had alanine in Egyptian virus.

The M2 ion channel protein was relatively conserved among H9N2 strains in the Middle East. The Egyptian virus contains mutation L10 that linked to human viruses, while the two substitutions S20 and R57 are linked to avian viruses [4, 5].

The NS1 protein of Qa/Egypt/11 was expressed as full-length for the PDZ domain legend. The PDZ-L of full-length avian NS1 (227–230) can increase the virulence of AIV in humans [50, 51]. The Egyptian H9N2 isolate has PDZ domain of “KSEV” similar to recent strains from Israel. The Egyptian H9N2 isolate lacks the D92E mutation in the NS1 protein associated with increased pathogenicity of H5N1 AIV in pigs [52] and present in human strain Hk/1073/99, Qa/Hk/G1/97, and Ck/Pak/2/99. These data indicate possible transmission and variable severity of the Egyptian H9N2 virus for humans and pigs.

Recombination and reassortments are common features among new highly pathogenic influenza viruses that are the mechanism behind antigenic shift. There was no evidence of recombination or reassortment between the co-circulated H9 and H5 subtypes in Egypt found in this study (data not shown). However, the reassortment of individual genes between both subtypes remains possible. In this study, a distinct type of reassortment events within the H9N2 subtype circulating in the region were predicted for the Egyptian virus, especially in internal genes PB2, PA and NP.

The topologies of all of the eight phylogenetic trees revealed substantial genetic diversity and frequent reassortment events for the Egyptian H9N2 virus, which could be schematically identified as intrasubtype exchanges (between different H9 genetic clusters) as described by Fusaro et al. [53]. Also, the H9N2 strains from UAE showed evidence of genetic diversity where the NS and PB2 genes showed reassortment with other viral subtypes while they correlated with G1-like for the other six segments [5].

The continuous reassortment among the currently circulating avian H9N2 viruses may give an explanation that these viruses require little adaptation in mammals following acquisition of all human virus internal genes through reassortment [54]. It was indicated that the viruses from UAE represent reassortants of previously characterized H9N2 viruses, and that increase the tendency for genetic exchange among H9N2 viruses and the extent to which genetic reassortment has contributed to their genetic diversity and causing disease in domestic poultry [5].

It was clear that H9N2 avian influenza viruses occasionally infected pigs, have the potential to reassort and generate novel viruses with respiratory transmission potential in mammals [55] as well as resistance to antiviral drugs [52]. Furthermore, the phylogenetic diversity and genotypes complexity of H9N2 influenza viruses have been recognized worldwide suggesting that they have undergone extensive reassortments to generate multiple reassortants and genotypes [53, 56].

In conclusion, this article reported the detailed genetic characterization of the genes of an H9N2 AIV isolated from quails in Egypt. Contentious surveillance activities are highly recommended to detect the prevalence of this subtype in commercial poultry in Egypt. Further antigenicity and pathogenicity studies are needed to determine the pathogenic potential of this isolate in avian and animal models since HPAI H5N1 is endemic in poultry in Egypt, a situation that is considered to have potential for the evolution of AIV. The presence of a new subtype of LPAI H9N2 may add another risk factor to the poultry industry in Egypt, especially with the endemic situation of HPAI-H5N1 and the presence of other pathogens with low biosecurity level in some commercial sectors that permit easy virus transmission and adds more stress to the condensed poultry populations.

Supplementary material

11262_2012_775_MOESM1_ESM.docx (61 kb)
Supplementary material 1 (DOCX 60 kb)
11262_2012_775_MOESM2_ESM.docx (20 kb)
Supplementary material 2 (DOCX 20 kb)
11262_2012_775_MOESM3_ESM.docx (17 kb)
Supplementary material 3 (DOCX 16 kb)

Copyright information

© Springer Science+Business Media, LLC 2012