The LPAIV H5N2 (A/Ck/Taiwan/1209/03, H5N2/1209, GenBank accession No. AY573918) isolated from apparently healthy chickens in Taiwan in 2003 [4] was the source of the antigen used for mouse immunization and Western blot (WB) analysis in this study. Briefly, 10-day-old embryonated specific pathogen-free (SPF) chicken eggs (Animal Health Research Institute [AHRI], Taiwan) were inoculated via the allantoic cavity with H5N2/1209 inoculum. The allantoic fluid was collected, inactivated by 0.2% formaldehyde for 24 h, clarified by centrifugation at 2,000 × g for 15 min, and further concentrated at 70,000 × g centrifugation for 2 h. The viral pellet was resuspended with 1/100 original volume of 0.08% NaN₃-phosphate-buffered saline (PBS) and tested by HA test as described in the Avian influenza chapter of the World Organisation for Animal Health (OIE) Manual 2015 [19]. The protein concentration of the viral suspension was also determined. The A/Dk/Yunlin/04 (H5N2) isolate was chosen as the source of the sequence origin of baculovirus-expressed recombinant NP (rNP). The A/wild Dk/Tainan/1634/09 (H1N1), A/Dk/Tainan/A30/02 (H5N2), H5N2/1209, A/wild Dk/830/05 (H5N2), A/Ck/Miaoli/2904/00 (H6N1), A/Ck/Changhua/7-5/99 (H6N1), and A/Dk/Tainan/A45/03 (H7N7) were isolated from either wild birds or domestic poultry birds in Taiwan. Those isolates, together with A/Dk/HongKong/820/80 (H5N3) from Dr. Hiroshi Kida, Hokkaido University, Sapporo, Japan, were propagated in embryonated SPF eggs as antigens for WB analysis and experimental inocula as mentioned above. Fifteen Influenza A virus reference subtypes (Table 1) were provided by Dr. Kida to prepare viral-infected CAMs for IHC examination. Newcastle disease virus (NDV) 060901, as a negative control, was isolated from poultry in Taiwan. The 50% egg infectious dose (EID₅₀) of A/Ck/Changhua/7-5/99 (H6N1) allantoic fluid for experimental infection was determined.

The H5N2 (A/duck/Yunlin/04) full-length NP was amplified by reverse transcription-polymerase chain reaction (PCR) with forward primer NP-F: 5′-ATGGCGTCTCAAGGCACCAAAC-3′ and reverse primer NP-R: 5′-TTAATTGTCATACTCCTCTGCATTGTC-3′. The NP full-length gene was cut from the T&A Cloning Vector (Yeastern Biotech, Taiwan) and subcloned in a frame into the corresponding pFastBac-HT-C plasmid (Invitrogen, USA). Additionally, a 6 histidine residue was designed to tag on the 3′ end of the NP by engineering the Spe I site embedded reverse primer NP-HisSpel-R: 5′-AGCACTAGTTCAGTGGTGGTGGTGGT-3′ in company with the EcoRI site embedded forward primer NP-EcoR-F: 5′-AGGAATTCGAATGGCGTCTCAAGGCAC-3′ to obtain the PCR product and used to construct the rNP vector. Subsequently, the recombinant vector was transposed into the baculovirus (Autographa californica) genome to form NP pFastBac recombinant bacmids. Positive recombinant bacmids were used to transfect an insect cell line (Spodofera frugiperda, Sf9) for viral particle formation. All procedures for viral particle production were performed according to the manufacturer's manual (Bac-to-Bac; Invitrogen). Recombinant NP from baculovirus-infected Sf9 cells was obtained, purified by Ni-NTA Agarose (Invitrogen), and used as an antigen for WB analysis.

Twelve-week-old female BALB/c mice were subcutaneously immunized three times each with 50 µg (2¹¹ HA U/4.61 mg/mL) of inactivated H5N2/1209 at two-week intervals. Immunogen of inactivated virus was mixed with complete Freund's adjuvant for priming and incomplete Freund's adjuvant for the next two boosters. The final booster of 50 µg/0.1 mL inactivated virus in PBS was given to anesthetized mice via an intrasplenic route [9] three days before fusion. The splenocytes from mice showing the highest immunofluorescence assay (IFA) titers were fused with myeloma cells Sp2/0-Ag14 (ATCC CRL-1851) to produce hybridomas according to a standard protocol [8]. Positive hybridomas were screened by indirect IFA on H5N2/1209 plates and WB (both described below). After three rounds of subcloning, selected clones were inoculated intraperitoneally into incomplete Freund's adjuvant-primed BALB/c mice to produce ascitic fluid containing MAb. The hybridoma cultural supernatant (cul. sup.) and ascitic fluid were collected for subsequent characterization. The isotype of MAb from the hybridoma cul. sup. was analyzed by IsoStrip Isotyping Kit (Roche Diagnostics, USA) according to the manufacturer's recommendations. The care and use of the mice were approved by the Institutional Animal Care and Use Committee, National Taiwan University, Taiwan, to ensure compliance with local legal and ethical requirements.

An MDCK cell suspension (100 µL) containing 3 × 10⁴ cells was added to each well of a 96-well plate. Following overnight incubation, the 80% confluent monolayers were infected by 10 TCID₅₀/100 µL/well H5N2/1209 in C-DMEM (DMEM supplemented with 100 U/mL penicillin, 100 µg/mL streptomycin, 2 µg/mL trypsin, BSA 0.2%, HEPES 20 mM) and incubated for 22 h. Subsequently, the plates were fixed with 70% (v/v) cold acetone for 20 min and then rinsed by PBS once, air dried, and stored at −20℃. Fifty microliters of hybridoma cul. sup. per well were added to the plate and the plate was incubated at room temperature for 2 h. The cells in each well were then washed with PBS and stained with 50 µL of fluorescein-conjugated goat anti-mouse IgG, IgA, and IgM (ICN Pharmaceuticals, USA) at a 1:1,000 dilution at room temperature for 1 h. After washing, IFA results were examined under a fluorescence microscope.

The AIV concentrated allantoic fluid and purified rNP were denatured in sample buffer (125 mM Tris-base, 2 mM EDTA, 2% SDS, and 5% β-mercaptoethanol, pH 6.8) by boiling for 10 min. Denatured proteins were separated in 10% SDS-PAGE gel at 100 V for 3 h. The proteins were then transferred to an NC membrane (Bio-Rad, USA), and the blots were blocked with 10% (w/v) skim milk for 2 h at room temperature. Subsequently, the blots were incubated with anti-NP MAb (WF-4 cul. sup. diluted 5-fold) for 1 h at room temperature and then washed three times with washing buffer (1× PBS and 1% Tween 20, pH 7.4). The HRP-conjugated goat anti-mouse IgG (1:2,000 × diluted; Jackson ImmunoResearch Laboratories Inc., USA) was added and incubated for 1 h at room temperature. After washing, the membrane was developed with TMB (Kirkegaard and Perry Laboratories, USA) for 15 min at room temperature. The reaction was stopped by rinsing with distilled water.

The CAMs from fifteen eggs infected with the Influenza A virus reference subtypes and having allantoic fluids with HA activity were collected [3] as the NP source for IHC assays to mimic that of the formalin-fixed tissue for histopathological examination. In brief, eleven-day-old embryonated chicken eggs were inoculated with 0.2 mL of stock Influenza A virus reference subtypes and incubated at 37℃ for 48 h before being removed from the incubator and chilled at 4℃ overnight. After the air sac end of the eggshell was cut off with scissors, the allantoic fluid was carefully collected, using an 18 G syringe, to avoid bleeding. The embryo and its attachments were discarded. The CAM was left adhered to the eggshell wall. The HA titer of the allantoic fluid was determined as described in the OIE Manual [19]. The CAMs having allantoic fluid with HA activity were removed with forceps, fixed in 10% neutral buffered formalin, and processed for IHC examination.

Histopathological sample blocks from H6N1 experimentally infected cases, field poultry cases, and CAMs for IHC were sectioned (3 µm thick). Tissue sections were placed onto microscope slides SuperFrost Plus (Thermo Fisher Scientific Inc, USA) [l], deparaffinized in xylene, and rehydrated in graded alcohol and distilled water. The sections were treated with 100 µg/mL of proteinase K (Roche Diagnostics, Germany) in 0.6 mol/L of Tris (pH 7.5)/0.1% CaCl₂ for 15 min. Endogenous peroxidase activity and non-specific antigens were blocked with 3% hydrogen peroxidase in methanol (Merck, Germany) and PowerBlock solution (BioGenex Laboratories, USA) for 10 min each. Tissue sections were then incubated at room temperature for 60 min with MAb WF-4. The detection system was sequentially conducted with Super Enhancer reagent (BioGenex Laboratories) incubation for 30 min and with Polymer-HRP reagent (BioGenex Laboratories) incubation for 40 min at room temperature. The aminoethylcarbazole (AEC; Dako, Denmark) was the substrate chromogen, and hematoxylin was used as a counterstain. The tissues from SPF and influenza virus-inoculated chickens served as negative and positive controls, respectively.

Twelve six-week-old SPF white leghorn chickens were divided into four groups for experimental inoculation by A/Ck/Changhua/7-5/99 (H6N1) as shown in Table 2. Briefly, Groups A and D had 2 chickens each, and Groups B and C had 4 chickens each. Chickens in Groups A, B, and C were inoculated with 10⁶ EID₅₀/0.1 mL H6N1 via intravenous, intranasal, and intraocular routes, respectively. Group D was the sham control. Half of the chickens in each group were sacrificed at 3 days post-inoculation (DPI), and the other half, at 5 DPI. Brain tissue from chickens experimentally infected by Taiwan NDV isolate ND060901, which causes encephalitis similar to that induced by H6N1, was the negative control for the WF-4 IHC test. The care and use of chickens were approved by the Institutional Animal Care and Use Committee, AHRI, Council of Agricultural Executive Yuan (Taiwan) to ensure compliance with local legal and ethical requirements.

Field cases of HPAIV infection from chicken farms collected in 2012, from chicken, goose, duck, and turkey farms collected in 2015, and a duck carcass index case in 2017 were included in this study. All routine pathological and virological examinations, including sequencing for all cases mentioned above, were performed at AHRI (Taiwan).

The immunized and euthanized BALB/c mice showed an IFA titer at the minimal 400× dilution of serum when it was sacrificed on fusion day. Thirty-four MAbs were obtained by sequential cloning and screening by IFA from the IFA-positive parental hybridoma. The MAb WF-4 used in this study displayed a band with a molecular weight of about 56 kDa, similar to that of AIV NP, on WB analysis and was further characterized by WB with full-length rNP expressed by baculovirus (panel A in Fig. 1). The rNP band had an expected molecular weight of about 61 kDa, unlike that of the purified H5N2/1209 on the blot (panel A in Fig. 1).

All viral isolates from wild birds and domestic poultry, including subtypes H1, H5, H6, and H7, exhibited band sizes similar to that of H5N2/1209 on WB (panel B in Fig. 1).

At low magnification, brownish-red epithelial lining cells with flattened to cuboidal shapes were observed in the single layer allantoic membranes of all CAMs infected with one of the fifteen reference serotypes of Influenza A virus. Representative sample results are shown in Fig. 2. At higher magnification, positive signals in the cytoplasm of the epithelium of allantoic endoderm with pyknosis and amorphous debris were noted. Some red-brownish deposits were observed in the nuclear area. The NP antigen was not found in the ectoderm or mesoderm.

One chicken from Group A that was found dead at 3 DPI had swollen kidneys and heavy urate deposition on serosal surfaces throughout the coelomic cavity. The other chicken from Group A, sacrificed at 3 DPI, demonstrated similar gross signs as the dead chicken. The chickens from the other groups were all clinically healthy and grossly normal, except one; chicken No. 3 in Group B, sacrificed at 5 DPI, had swollen kidneys and gross lesions but moderate urate deposition. This chicken presented strong IHC-positive results in the renal parenchyma (panel A in Fig. 3). Scattered degeneration, necrosis, and regeneration of the renal tubular epithelium were observed (panel B in Fig. 3). The varying amount of deposition of the positive signal representing the influenza viral NP corresponded to the degenerative to necrotic process in the renal tubule (panel C in Fig. 3). The other chicken (No. 4) from Group B, also sacrificed at 5 DPI, was grossly normal but showed moderate kidney lesions and a positive IHC signal like chicken No. 3 (data not shown). The other chickens from Groups C and D all showed IHC-negative results with WF-4 (data not shown). Brain tissue from chickens infected with NDV isolate 060901 did not present positive IHC signals with WF-4 (data not shown). The field cases from 2012 were from an H5N2 HPAIV infection in poultry farms affecting layer, broiler breeder, and native chickens. Those in 2015 were from H5N2, H5N3, and H5N8 HPAIV infections affecting goose, chicken, turkey, and duck farms. We also examined a duck H5N6 HPAIV index case from 2017. The results of IHC staining were similar in all naturally infected cases. IHC-positive signals were demonstrated in all organs, such as larynx (panel D in Fig. 3), heart (panel E in Fig. 3), brain, lung, kidney, pancreas, and integumen. Interestingly, IHC-positive signal strengths were highest in the heart muscle fiber of a 2015 H5N8 infection case (panel E in Fig. 3) and the endothelial cell lining of the spleen and liver of a 2015 H5N2 case (panel F in Fig. 3). In addition, there were IHC-positive signals in goose ovary (panel G in Fig. 3) of an H5N3 infection in 2015 and in duck brain (panel H in Fig. 3) of a 2017 H5N6 infection.