Isolation and molecular characterization of the hemagglutinin gene of H9N2 avian influenza viruses from poultry in Java, Indonesia

Objective: The avian influenza virus (AIV) subtype H9N2 circulating in Indonesia has raised increasing concern about its impact on poultry and its public health risks. In this study, the H9N2 virus from chicken poultry farms in Java was isolated and characterized molecularly. Materials and Methods: Thirty-three pooled samples of chicken brain, cloacal swab, trachea, and oviduct were taken from multiple chickens infected with AIV in five regions of Java, Indonesia. The samples were isolated from specific pathogenic-free embryonated eggs that were 9 days old. Reverse transcription polymerase chain reaction and sequencing were used to identify H9N2 viruses. Results: This study was successful in detecting and characterizing 13 H9N2 isolates. The sequencing analysis of hemagglutinin genes revealed a 96.9%–98.8% similarity to the H9N2 AIV isolated from Vietnam in 2014 (A/muscovy duck/Vietnam/LBM719/2014). According to the phylogenetic analysis, all recent H9N2 viruses were members of the lineage Y280 and clade h9.4.2.5. Nine of the H9N2 isolates studied showed PSKSSR↓GLF motifs at the cleavage site, while four had PSKSSR↓GLF. Notably, all contemporary viruses have leucine (L) at position 216 in the receptor-binding region, indicating that the virus can interact with a human-like receptor. Conclusion: This study described the features of recent H9N2 viruses spreading in Java’s poultry industry. Additionally, H9N2 infection prevention and management must be implemented to avoid the occurrence of virus mutations in the Indonesian poultry industry.


Introduction
Since the early 1980s, Indonesia's poultry sector has risen dramatically. The poultry industry in Indonesia is projected to be worth more than 34 billion US dollars. According to CASARED, there are approximately 67.9 million birds on 99,000 farms in Indonesia that use a commercially oriented production method [1,2].
In the last two decades, avian influenza virus (AIV) infection has become a common problem for the global poultry industry [3,4], including Indonesia. These viruses are categorized into two groups. The first group is an highly pathogenic AIV, for example, the H5N1 subtype which can cause mortality up to 100%. This subtype has become a major concern for human and animal health in Indonesia, especially the entire province of Java, since 2003 [5], and vaccinations have been routinely administered against this subtype. The second group was infected with various types of domestic and wild birds with mild to moderate infection called low pathogenic avian influenza virus (LPAIV) [6]. H9N2 belongs to this group and causes significant economic losses associated with increased mortality and decreased egg production [7]. Coinfection of bacteria such as Mycoplasma gallisepticum and Escherichia coli or other infections from other viruses such as infectious bronchitis © The authors. This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License (http://creativecommons.org/ licenses/by/4.0) (IB) and newcastle disease (ND) can increase the mortality of chickens infected with H9N2 [8]. Currently, this subtype is endemic to poultry in several African, Asian, and Middle Eastern countries [9].
There is a dearth of evidence on the characteristics of circulating AIVs subtype H9. The bulk of these virus sequences are classified as G1 (clade h9. 4 [10].
Based on the duck sera's susceptibility to recombinant hemagglutinins and neuraminidases in indirect enzyme-linked immunosorbent assay and dot blot studies, the H9N2 subtype was believed to exist in Java, Indonesia, duck population [11]. This is the first report of H9N2 outbreaks in Java, and surveillance will be required to confirm the existence of H9N2 subtypes via virus isolation and molecular characterization. In this work, we detected and examined H9N2 viruses from several regions of Java.

Materials and Methods
The study was carried out in April 2017-March 2018. The materials and methods used in this study are describe below.

Ethical approval
The current study follows the requirements established by the Indonesian Law on Animal Health Research (UU/18/2009, article 80. Ethical approval was not required for this investigation because no live animals were used.

Samples
A total of 33 commercial poultry chickens were sampled in 5 provinces of Java, Indonesia (Banten, n = 4; Central Java, n = 2; East Java, n = 9; West Java, n = 17; and Yogyakarta, n = 1) with a history of widespread depression, respiratory and neurological disorders, and decreased egg production. The poultry flocks comprised birds ranging in age from 25 to 79 weeks. Three moribunds or dead hens were randomly selected from each group and brought to the laboratory. The chicken brain, cloacal swab, trachea, and oviduct were taken and pooled for virus isolation. The samples collected are listed in Table 1.

Virus isolation
Propagation of the sample was carried out according to conventional laboratory protocols. The swab or tissue was suspended in sterile phosphate buffer saline with penicillin (200 μg/ml) and streptomycin (100 μg/ml) in a 2:10-20 ratio and centrifuged at 1,000× g for 10 min at 4°C. The supernatant was filtered via a 0.22 μm membrane filter and inoculated into 9-day-old specific pathogenic-free (SPF) embryonated eggs via the allantoic route. The eggs were incubated at 37°C for 48-72 h and daily embryo death was monitored. Eggs that died before 24 h were discarded because the virus considered was non-specific. Harvested allantoic fluid was used in the hemagglutination (HA) experiment. Three serial passages were conducted prior to the selection of negative samples [12].

Detection of viral RNA using reverse transcription-polymerase chain reaction (RT-PCR) test
Allantoic fluid was extracted using a total ribonucleic acid (RNA) mini kit (Geneaid, Taiwan). MyTaqTM OneStep RT-PCR kit (Bioline ® , Taunton) was used to test the RNA for H9N2 viruses by RT-PCR utilizing H9 and N2-specific primers ( Table 2). The following thermal profile was used to amplify the genes: 48°C for 20 min, 95°C for 2 min, followed by 40 cycles of 95°C for 10 sec, 52°C for 10 sec, and 72°C for 2 min. At the conclusion of the amplification, the final stage was carried out at 72°C for 10 min. Electrophoresis was used to visualize the amplified samples. The usual size marker is 100 bp (Geneaid, Taipei, Taiwan). Observations were conducted with the aid of a UV transilluminator. Additionally, real-time polymerase chain reaction (RT-PCR) assays for H5N1, Newcastle disease virus (NDV), and infectious bronchitis virus (IBV) were performed [17].

Sequencing
Following good RT-PCR results for the H9 gene, the HA gene was amplified and sequenced using the MyTaq One- Step RT-PCR kit (Bioline) with the HA9-F and HA9-R primers (Table 2). Electrophoresis was used to separate the PCR products, and the desired band was purified for sequencing (by First BASE Laboratories Sdn Bhd, Malaysia). Determination of the nucleotide and amino acid sequences for the HA gene from a recently isolated strain using Bioedit v.7 and alignment with ClustalW. MEGA 7 was used to create the phylogenetic tree of contemporary viruses using the neighbor-joining method [18] with 1,000 alignment repeats [19]. Clades were defined using the genetic distance between members and the topology of the phylogenetic tree. We studied the molecular properties of amino acid sequence derivation from 13 nucleotide sequences. The nucleotide sequences of hemagglutinin genes were compared between the isolates from this investigation, isolated studies with prototype H9N2 strains, and isolated studies with H9N2 subtype vaccine strain China (chicken/ Guangdong/SS/94, chicken/Shanghai/F/98, and chicken/ Shandong/6/96).
This study investigated several critical areas, including receptor-binding sites (RBS) on the left edge of the binding pocket, the right edge of the binding pocket, and the cleavage site. Additionally, certain amino acid residues (54, 80

Isolation and molecular identification of recent viruses
The result revealed that 18 samples were HA-positive after three passaged in embryonated chicken eggs with mean  (Table 3), while the findings of RT-PCR showed only 13 samples were positive for H9N2 AIV (Fig. 1). Six of them were single infections of H9N2 AIV, while the seven positive samples were multiple infections by many coinfected viruses such as H5N1 (15.4%), NDV (38.5%), or IBV (7.6%). (7.6%). Regarding the distribution of positive H9N2 cases across different months, we observed that two samples (15.4%) were positive in April, two (15.4%) in May, one (7.7%) in July, one (7.7%) on August, six (48.2%) in September, and one (7.7%) in October 2017.

Sequence and phylogenetic analysis
The RT-PCR experiment detected the presence of a complete HA gene with an amplicon size of 1,683 bp in the allantoic fluid (Fig. 2). The HA gene's 13 most recent nucleotide sequences were deposited in GenBank with accession numbers MG957200, MG957201, MG957202, MG957203, MG957204, MG957205, MG957206, MG957207, MG957208, MG957209, MG957210, MG957211, and MG957212. A phylogenetic tree analysis based on the HA gene was created on the sequences of 13 recent isolates in our investigation and prior AIV H9N2 viruses [20]. Phylogenetic analysis revealed that all recent viruses belonged to clade h9.4.2.5 (Fig. 3), and a comparison of the 13 isolates' HA sequences at the nucleotide level revealed 96.2%-100% identity with the lineage Y280 reference strain (

Analysis of deduced amino acid sequences
The characteristics of LPAIV seen from the sequence of HA segments reveal a monobasic amino acid motif at the cleavage site (PSRSSRGLF) in 9 out of 13 study isolates, while 4 other isolates, A/chicken/Banten-01/17 (Accession No. MG957200), A/chicken/Banten-02/17 (Accession No. MG957201), A/chicken/Banten-04/17 (Accession No. MG957202), and A/chicken/EastJava-05/17 (Accession No. MG957210), showed a non-synonymous substitution of the amino acid (R317K) at this cleavage site to make PSKSSR↓GLF. All isolates in the present study showed H173N and Q217M mutations at receptor-binding sites while comparing the reference strains in GenBank. The E180A substitution was found in only five of our isolates, while the remaining seven isolates contain the E180T substitution ( Table 5).
The antigenic epitope of HA of H9N2 is mapped in two parts: site I and site II [21]. The antigenic site of the HA gene in all isolates obtained the same site I motif on the position residues of 125, 147, and 152, respectively, serine (S), lysine (K), and proline (P). While in site II, there are two kinds of motives at positions 135, 183, and 216. The first motif is aspartic acid (D), asparagine (N), and leucine (L), owned by 11 isolates in this study (Table 5). Unlike the others, N183D substitution occurs in two isolates from West Java, namely A/chicken/WestJava-04/17 (Accession  No. MG957203) and A/chicken/WestJava-06/17 (Accession No. MG957204). One non-synonymous amino acid substitution (K164N) occurred in A/chicken/ Banten-01/17 (Accession No. MG957200), A/chicken/  (Fig. 4).
The potential glycosylation site of the HA gene analysis from 13 study isolates revealed 6 sites with N-X-T / S motif (where X is any amino acid except proline) in the  Table 6).

Discussion
Detection of H5N1 AIV from H9N2 positive cases indicated continuous co-circulation of the two subtypes in commercial chicken flocks. These study results agreed with previous studies [22] and provided a possible explanation for increased infectivity of H9N2 AIV over time due to genetic reassortment with the H5N1 subtype [23,24]. Table 3. Results of isolation and identification using RT-PCR.       The H9N2 virus has spread widely in poultry throughout the Asian region, providing the fact that the virus is potentially rapidly evolving with virulence rising [25,26,27]. It is similar to the report in Java. Lesser than a year ago, H9N2 AIV was found in all provinces of layer farms in Java. The presence of H9N2 in commercial farms of Java may indicate some defect in applying biosecurity [28].
The motif of the cleavage site of the recent isolate study was different from those of clades representative isolates. All H9N2 AIV fit the characteristic of LPAI. The proteases (such as trypsin), which are only secreted by the respiratory organs and intestines, will recognize a single arginine (R) residue [29]. While vaccination is the optimum technique for preventing AIV, earlier research indicates that inadequate protective immunization may result in an antigenic drift of the H9N2 virus [31]. To assess if a vaccination strain should be updated in response to the virus that is circulated in the field, an antigenic link between the commercial vaccine strain and the emerging recent virus should be evaluated [32].
The receptor-binding specificity was associated with amino acids in the RBS of HA proteins. In general, AIVs specifically recognize sialylα2,3-galactose receptors, while the human virus recognizes the sialyl-α2,6-galactose receptor [33]. Hemaglutinine proteins of H9N2 viruses isolated from chickens in a recent study had histidine (H), glutamic acid (E), glutamine (Q), and glutamine (Q) at 173, 180, 216, and 217 positions at the RBS [34]. The amino acid substitution in this study had previously appeared in Chinese isolates (A/chicken/Guangdong /LGQ02/2014). The presence of asparagine (N) at 173 positions is owned by the AIV subtype H9N2 belonging to clade h9.4.2, whereas methionine (M) at position 217 can only be encountered in clade h9.4.2.5.
Leucine (L) at position 216 was found mainly in the HA protein of H2 and H3 viruses isolated from humans. In a previous study, Leucine (L) at 216 caused a van der Waals bond with the C-6 atom to form a sialyl-α2,6-galactose linkage. HA having leucine (L) at position 216 tends to bind to sialyl-α2,6-galactose, while Q in that position tends to bind to sialyl-α2,3-galactose. The substitution of Q to L in position 216 allows the H9N2 virus to replicate more efficiently in human cell culture [34]. The substitution of Q to L at position 216 is one of the genetic changes that occurs during the adaptation of avian influenza strains in pigs [35]. The existence of a backyard poultry system exacerbates the condition. Infected poultry without any clinical symptoms may increase the likelihood of exposure in humans [25].
The N183D mutation with Q216L at antigenic site II is involved in inhibiting viral interactions with epitope-specific antibodies and can result in mutant virus escape; both of these residues are identical to those in the most closely related vaccine. The 127 amino acid residue is another critical location in the HA H9N2 gene [21]. In a recent Table 6. Comparison of the PGS in HA proteins of 13 H9N2 isolates. investigation, all H9N2 viruses from Java showed residual S127, which was identical to that of the H5 subtype [36]. Recent research [37] identified strains of H9 variations having the S127N mutation, which may increase the pathogenicity of SPF hens, while 13 isolates lacked the mutation. In general, N-linked glycosylation has a substantial effect on the virus' biological features. It may be impacted by the N-X-S/T-X amino acid structure [38]. Additionally, earlier research has demonstrated that alterations in glycosylation residues near HA protein cleavage sites might influence viral virulence [39]. Previous research has established that N-linked glycosylation of the HA protein has an effect on host innate immune system evasion and survival, as seen in human influenza virus, H1 and H3 subtypes [40]. Changes in the glycosylation status of the antigenic protein of the influenza virus HA region alter virus binding to antibodies, whereas changes in the glycosylation site at numerous places have no effect on the structure and function of HA [41]. There was a potential glycosylation site (PGS) within the antigenic area, NVS, between locations 123 and 125 in this study. Previous research with H1N1 viruses revealed that the presence of K123N mutations results in immune evasion via glycosylation [42].

Conclusion
In five regions of Java, the H9N2 virus was isolated from chickens. The H9N2 virus that recently circulated in Java is classified as clade h9.4.2.5 and lineage Y280. Additionally, all recent viruses have a genetic link with viruses from China and Vietnam.