Infectious bursal disease virus suppresses H9N2 avian influenza viral shedding in broiler chickens

1. Infectious bursal disease virus (IBDV) causes immunosuppression in chickens, increasing their susceptibility to other infectious diseases and resulting in vaccination failure. Here, we investigated the immune-depressing effect of IBDV on H9N2 avian influenza viral infection in the broiler chickens. 2. For this purpose, chickens were divided into four groups. In group A, chickens were inoculated with IBDV at 21 days of age and H9N2 avian influenza virus (AIV) 5 days later. Groups B and C only received AIV at 26 days of age and IBDV at 21 days, respectively. The control group (D) were inoculated with normal saline at the same times. Tissue samples from different organs were collected on the days 1, 3, 6, 9, and 12 after H9N2 infection. 3. Macroscopic observation showed IBD lesions in groups A and C, including swollen bursa with the presence of gelatinous exudates, haemorrhages in the thigh muscle, edema, and nephritis. 4. Reverse Transcription-PCR was used to study H9N2 AIV dissemination, and qRT-PCR to determine viral genome copy number in different organs. A considerable titre of AIV was found in the trachea, lungs, cecal tonsils, spleens, and feces of infected chickens. The genome copy number of the virus in the trachea and lungs of group A was significantly higher than that in group B on the first day after inoculation. But in the other days post inoculation, RT-PCR did not detect the AIV genome in group A. Although there might have been some immunosuppression in group A, IBDV could interfere with AIV replication in the chickens of this group. 5. In conclusion, we propose that pre-exposure to IBDV at 3 weeks of age reduces the replication and shedding of H9N2 in broiler chicken. ARTICLE HISTORY Received 19 January 2019 Accepted 29 March 2019


Introduction
Infectious bursal disease virus (IBDV) is a member of the Birnaviridae family, which has double-stranded RNA genomes. The virus results in acute, highly contagious infections, immunosuppression and mortality in young chickens (Sharma et al. 2000;Berg 2000). Related economic losses to chicken farms are substantial. Bursal damage is the main feature of the disease (IBD or Gumboro) and is observed in affected chickens worldwide (Sharma et al. 2000). In IBD, immunosuppression increases susceptibility to other viral diseases, such as viral arthritis (Montgomery and Maslin 1991), Marek's disease (Sharma et al. 2000), infectious bronchitis (Haghighat-Jahromi et al. 2008), infectious laryngotracheitis (Rosenberger and Gelb 1978), inclusion body hepatitis (Fadly et al. 1976), and chicken infectious anaemia (Yuasa et al. 1980), as well as infections by protozoa (Giambrone et al. 1977).
The H9N2 avian influenza virus (AIV) is a low pathogenic avian influenza (LPAI) virus in broiler chickens (Bano et al., 2003) and has been reported in Asia and especially in the Middle East (Alexander 2007;Mosleh et al. 2009a). Subtypes of this virus are known to be enzootic in rural areas of Iran and are present in commercial chicken farms. Although H9N2 AIV is an LPAI virus, it has severe pathogenicity and causes a high mortality rate among farm chickens co-infected with other bacterial and viral diseases (Bano et al., 2003;Pan et al. 2012). As IBDV is involved in many such co-infections, a greater degree of influenza viral replication can be expected in IBDV-infected chickens. To determine to what extent IBDV affects sequela of H9N2 infection, this experiment studied the effects of IBDV inoculation on subsequent H9N2 AIV infection in 215 broiler chickens, in term of clinical signs, gross pathology, serology, virus dissemination in different organs, and shedding of the virus from trachea as well as in the faeces.

Birds
A total of 215, one-day-old Cobb broiler chickens were purchased from a local hatchery and reared in the experimental research unit of the Shiraz University College of Veterinary Medicine (Shiraz, Iran). All birds were kept under controlled conditions based on the Cobb breeder management catalogue and had free access to food and water throughout the experiments. The chickens received no vaccinations. The study protocol, including animal sacrifice, was approved by the research ethics committee of Shiraz University, Iran. In addition, the protocol followed the recommendations of the European Council Directive (86/609/EC) of November 24 1986, regarding standards for the protection of animals used for experimental purposes. To determine maternal immunity, ten three-day-old chickens were randomly selected and euthanised (Table 1).
Blood samples were then collected and tested for serum antibodies against IBDV and H9N2 AIV using ELISA and haemagglutination inhibition (HI), respectively.

Experimental design
At 21 days of age, the chickens were inoculated with IBDV (Table 1). One day before inoculation, five chickens were randomly selected and blood samples taken from their wing veins. The birds were then euthanised by cervical dislocation. On necropsy, tissue samples of the trachea and the bursa of Fabricius were immediately collected and preserved at −80°C until analysed. Samples were subsequently tested for possible contamination with Newcastle disease virus, AIV (Pantin-Jackwood et al. 2015), IBDV (Wu et al. 1992), and infectious bronchitis virus using RT-PCR analysis (Falcone et al. 1997), as negative disease status was required before starting inoculations. The other birds, including two hundred broiler chickens, were randomly selected and divided into two equal groups (I and II) of 100 chickens each and were reared in a separate, isolated room under controlled conditions. For induction of IBD, isolates of Shiraz IBDV (Accession number: JX983160), a highly virulent infectious bursal disease virus enzootic in the region, were used. After a 3 h fast, chickens in the treatment group (group I) were infected with 10 4.5 ELD 50 /0.5 ml of vvIBDV isolate via intra-crop inoculation. The control group (group II) chickens only received 0.5 ml of normal saline. On day three after inoculation, five chickens from each group were euthanised, and blood samples were collected. Tissue samples from the trachea and the bursa of Fabricius were then isolated and frozen at −80°C. The tissue samples were prepared and tested by RT-PCR analysis to determine the presence or absence of IBDV infection in the chickens. Trachea tissue and faeces were analysed by RT-PCR to look for AIV contamination. In addition, serum samples were collected from each group and tested by HI to confirm the RT-PCR results of AIV contamination.
On day five after inoculation, chickens in the treatment group were assigned to subgroup A (70 chickens) or C (25 chickens), and chickens in the control group were divided into subgroups B (70 chicken) and D (25 chickens). In the AIV inoculation step, each group was transferred into a separate, isolated room. Chickens in subgroups A and B were then intranasally inoculated with 10 6.5 EID 50 /0.1 ml of A/Chicken/ Iran/772/1998(H9N2) AIV, and subgroups C and D received sterile saline only. Since quantitation of AIV and determining viremia (Table 1) were the major goals in this study, the number of chickens in subgroups A and B was higher than in the other two groups.
Chickens were observed for any clinical signs of avian influenza (AI) twice daily. In each subgroup, five chickens were randomly selected and euthanised 1, 3, 6, 9, and 12 days post-inoculation (dpi). Tissue samples from the trachea, lungs, brains, spleens, caecal tonsils, and faeces were isolated and preserved at −80°C until analysed. In addition, blood samples were collected from each subgroup and tested for antibodies against influenza virus by HI assay. To trace the possible H9N2 viremia in chickens from subgroups A and B, samples were collected from five euthanised chickens every 8 h for 3 dpi. Buffy coats were isolated from the blood samples and tested for the presence and quantity of virus using RT-PCR.

RNA extraction
Total RNA in the buffy coat samples was isolated using a Cinna-pure RNA kit (Cinna-Gen Molecular Biology and Diagnostic, Tehran, Iran) according to the manufacturer's instructions. Briefly, 400 µl of lysis buffer was added to 100 µl of the sample, and the solution was mixed by vortex. Then, 300 µl of precipitant solution was added, and the solution was mixed again. A total volume of 800 µl of solution was transferred into the spin column and centrifuged at 12 000 rpm for 15 min. The columns were washed twice with the washing buffer. Total RNA was dissolved in 30 µl of RNase-free water via centrifugation and stored at −70°C. Tissue and faecal samples were prepared for RNA extraction using RNX-plus solution (Cinna-Gen, Molecular Biology and Diagnostic, Tehran, Iran). Briefly, a 10% (W/V) suspension of faeces was prepared in diethyl pyrocarbonate (DEPC) treated water and centrifuged at 250 g for 10 min at 4°C. The supernatant was removed, and tissues weighing 50-100 mg were then homogenised in RNX-plus. RNA was isolated according to the manufacturer's protocol. All RNA samples were normalised to the lowest concentration in DNase-/RNase-free distilled water before quantitation.

Purification of recombinant plasmid
Escherichia coli XL1-blue containing a plasmid ligated with a partial sequence of the avian influenza M gene was cultured in 5 ml of LB medium for 24 h. The medium was centrifuged at 250 g for 10 min, and precipitated bacteria were subjected to plasmid isolation using an AccuPrep Plasmid Extraction kit (Bioneer, Daejeon, Korea) according to the manufacturer's instructions. The number of isolated plasmids was measured in solution using a spectrophotometer (BioPhotometer; Eppendorf, Hamburg, Germany). The recombinant plasmids were used as standards in the quantitation assay.

Reverse transcription PCR
For the synthesis of cDNA, 5 µl of isolated RNA, 1 µl containing 20 pmol of H 9 forward primer (AccuPower RT PreMix kit; Bioneer) and 14 µl of DEPC distilled water were added and mixed in a lyophilised tube. The cDNA synthesis was carried out at 45°C for 1 h and 95°C for 5 min. Isolated cDNA was then collected in tubes and stored at −20°C until analysed.

Quantitative real-time PCR
The total volume of the reaction was 20 µl, and comprised 2 µl of 10x PCR buffer, 0.4 µl of dNTPs (0.2 mM), 2.4 µl of MgCl2 (6 mM), 0.2 µl of Taq DNA polymerase (1 U), 1 µl of each primer (10 pmol), 0.6 µl of TaqMan probe (6 pmol), 7.4 µl of distilled water, and 5 µl of cDNA. The PCR program used the Bio-rad Miniopticon system and was run at 95°C for 5 min, and included 42 cycles at 95°C for 15 sec and 60°C for 1 min (Mosleh et al. 2009a). Specific sequences of the isolated oligonucleotides are shown in Table 2.

HI test
To evaluate the serological status of chickens in different groups against the H 9 antigen, sera samples were collected from chickens aged 3, 20 (the day before IBDV inoculation), and 25 d. In addition, H 9 antibody titres against influenza virus were evaluated at 1, 3, 6, 9, and 12 dpi. H9N2 The AIV antigen equivalent to four haemagglutinating units (HA) was used to determine the HI activity of two-fold serially diluted test sera in a 96-well microtitre plate (Karimi et al. 2014).

ELISA
The presence of IBDV maternal antibodies was determined in chickens aged 3 d using an IDEXX ELISA kit (Westbrook, ME, USA) according to the manufacturer's instruction.

Statistical analysis
Serological analysis results of different groups were compared by two-way ANOVA followed by Tukey's range test using SPSS version 13 (SPSS, Inc., Chicago, IL, USA; Table 2). When P < 0.05, treatment effects were considered significant.

Clinical signs
Before IBDV inoculation, chickens in all groups were apparently normal. Molecular tests were negative for AI, infectious bronchitis, IBD, and Newcastle disease infection in the birds. Three days after IBDV inoculation, chickens in subgroups A and C showed clinical signs of depression, ruffled feathers and whitish diarrhoea. In addition, chickens in subgroup A showed ruffled feathers, depression, cloudy eyes, conjunctivitis, respiratory distress, coughing, sneezing, and gasping at 2 dpi, whereas those in subgroup B showed less severe signs of infection at 3 dpi. The most definite signs of infection in the respiratory system were reported at 6 dpi with H9N2 AIV and observed in chickens from subgroups A and B. However, the clinical signs of infection in chickens from subgroup B disappeared at 9 dpi, while those in subgroup A continued for up to 12 dpi. Chickens in the control subgroup (D) showed no clinical signs of infection during the study.

Macroscopic Gumboro disease lesions
At 3 dpi, five chickens were randomly selected from subgroups A and C, sacrificed, and macroscopically observed. Gross lesions in these groups included general oedematous and swollen bursa, which are associated with the presence of gelatinous exudates, haemorrhages in thigh muscle, and nephritis. In these subgroups, the bursa of Fabricius' size at 12 dpi was significantly decreased to approximately half the size of an intact bursa with few gelatinous exudates inside. Additionally, mild nephritis was observed in chickens affected by Gumboro disease.

Macroscopic avian influenza lesions
At 1 dpi with influenza, macroscopic observations of chickens in subgroups A and B showed no lesions, but did reveal some congestion in their spleens. At 3 dpi with influenza, the chickens in these groups showed clinical signs of respiratory infection, including mild airsacculitis, with the presence of congestion, and transudates in the trachea. The severity of the lesions in these chicks was increased at 6 dpi. Moreover, the bursa was extremely degenerated at this time. Nephritis and splenitis in the chicks of subgroup A were more severe than those of subgroup B. At 9 dpi, congestion, haemorrhaging and fibrinous casts were observed in lungs isolated from two chickens in subgroup A, whereas pericarditis and perihepatitis were seen in one chicken from subgroup B. Respiratory lesions were decreased in the other birds from these subgroups. There were no necropsy injuries at 12 dpi, except for mild tracheal congestion in the chicks from subgroup A. Necropsy revealed no lesions related to AIV infection in chickens from subgroups D (control) or C.

H9N2 HI antibody titre
The mean titres of antibodies after H9N2 inoculation are presented in Table 3. The log-2 transformed mean titres before inoculation were 7.2, 4.12, and 3.5 in the blood samples of chickens aged 3, 20, and 24 d, respectively. At 9 dpi, antibody titres significantly increased in subgroups A and B. In addition, mean titres in subgroup B were significantly higher than those in subgroup A at 12 dpi.

RT-PCR
Molecular tests were negative for AI, infectious bronchitis, IBD, and Newcastle disease infection one day before IBDV inoculation. To confirm IBDV infection in group I, the bursae of Fabricius were isolated from five chicks in each group and subjected to RT-PCR analyses using VP 2 gene primers 3 d after IBDV inoculation. The results showed a 743 bp amplicon, whereas group II, was negative for IBDV infection. All tested samples were negative for AIV infection at this point. On 1, 3, 6, 9, and 12 dpi with AIV, five chickens from each subgroup were euthanised, and their spleens, trachea, caecal tonsils, lungs, brains and faeces were isolated and tested for H9N2 using RT-PCR. Blood samples were isolated at 8 h intervals and tested up to 3 d after influenza infection. The results were shown in Figure 1. In subgroups A and B, H9N2 was not detected in buffy coat samples, and results were negative in subgroups C & D.

Real-time PCR
Samples with positive RT-PCR results were subjected to real-time PCR to quantify their AIV M gene copy numbers. The efficiency of the test in different tissues was approximately 98% using five 10-fold dilutions of the recombinant plasmid. Mean copy numbers of the M segment at different days after inoculation are shown in Figure 1.

Discussion
The strain of virus used for induction of Gumboro disease in the chickens was highly virulent and causes severe lymphoid  necrosis and degeneration in the bursa of Fabricius. Sharma et al. (2000) previously reported that immunosuppression by IBDV was achieved when chickens were inoculated at an early age. Since commercial broiler chickens are protected for up to three weeks by maternal antibodies, immunosuppression by IBDV is a rare occurrence. In the present study, the chickens were inoculated at 21 days. Therefore, it was not expected for the immunosuppressive effect of IBDV to be strong. However, the results showed that HI antibody titre was lower in subgroup A (treated with IBDV) than in subgroup B on day 12 after AIV infection, which may be due to IBDV immunosuppression or lower AIV replication.
In this study, no AIV shedding was observed in subgroup A, except at 1 dpi. In subgroup B, in which chickens were inoculated with only AIV, significant increases in viral shedding were observed over the course of the study. These results suggested interference between IBDV and AIV infection. Meng et al. (2011) reported that AI loads were significantly decreased in trachea isolates from specific-pathogen-free chickens fed with IFN-α as compared with those in control chickens. Additionally, in chickens infected by IBDV, cellular immunity can be less affected than is humoral immunity (Meng et al. 2011). Therefore, it can be suggested that the study of alterations in serum interleukin levels might aid understanding of the mechanism modulating interference between the two viral infections. Raj et al. (2011) reported that serum levels of IL1-β, IL2, IL8, IL12, IL17, and interferon (IFN)-α and -β were significantly increased in commercial chickens infected with IBDV (Raj et al. 2011). Therefore, increasing serum IFNs level were probably induced by IBDV infection and may inhibit AIV shedding.
Knowledge of the quantity and duration of viral shedding from animals can help clarify ambiguities in pathogenesis as well as the severity and period of the virus spreading in flocks, which may be useful in disease control. Infected chickens shed H9N2 virus via faeces more than via respiratory secretions, particularly 5-7 d after infection. Although the virus replicates in the kidneys up to 11 d after infection, the quantity of virus in this tissue was low and did not influence the duration of faecal shedding (Mosleh et al. 2009b;Kwon et al. 2008). Several factors are involved in the degree and period of AIV shedding. Some studies revealed that avian infectious bronchitis vaccines increased the duration and degree of AIV shedding and decreased the efficacy of H9N2 AIV vaccines in reducing AIV faecal shedding (Tavakkoli et al. 2011). Some gross lesions, such as subcutaneous hyperaemia and many petechial to ecchymotic haemorrhages and stunting have been reported in chicken embryos inoculated with AIV + infectious bronchitis live vaccine virus (Haghighat-Jahromi et al. 2008).
Despite the decrease in the replication and shedding of the influenza virus, the clinical signs of the disease were not considerably different between the chickens of subgroups A and B. Costa-Hurtado et al. (2014) co-infected chickens and turkeys with lentogenic Newcastle disease virus and LPAIV and reported that the replication pattern of these viruses changed, but clinical signs were unaffected. The effects on viral replication varied according to the time of infection and species. The results suggested that infection with a heterologous virus may result in temporary competition for cell receptors or competent cells for replication, and that this competition may be interferon-mediated and decrease with time (Hurtado et al., 2014). Pantin-Jackwood et al. (2015) reported that co-infection with velogenic Newcastle disease virus and LPAIV in domestic ducks did not affect clinical signs but changed the pattern of virus shedding and transmission (Pantin-Jackwood et al. 2015). It can therefore be suggested that some other factor other than viral replication is important for influencing the severity of symptoms.
However, suppression of AIV faecal shedding was the most interesting issue in the current study. Chicken age, IBDV strain, and the time interval between inoculations of the two viruses may greatly affect AIV shedding. Vaccination against IBD at three weeks of age could be a practical approach to inhibit probable influenza infection.

Conclusions
IBDV infection suppressed AIV shedding in older chickens. According to the results, infection with Gumboro virus at three weeks of age cannot be considered to exacerbate viral shedding or the symptoms of avian influenza.