Viruses, cells, and animals
The H7 AIV used in the study was a low-pathogenic strain constructed with the internal gene fragments of A/Puerto Rico/8/34 (H1N1) as the backbone through reverse genetics. H7 positive hybridoma cell line 2F8 was cultured in RPMI-1640 medium supplemented with 10% foetal calf serum (Gibco-BRL, USA) at 37 °C with 5% CO2. A totoal of nine six-week-old female BALB/c mice weighing between 22 and 25 g were purchased from the Sino-British SIPPR/BK Lab Animal Ltd (Shanghai, China). The mice were fed with standard commercial diet and raised in individually ventilated cages in a clean facility at our laboratory.
Serum samples
The experimental standard sera were obtained from our laboratory as described previously [18]. A total of thirty-six three-month-old specific-pathogen-free (SPF) chickens were purchased from Shennong Company in Zhejiang province of China. They were randomly divided into eighteen groups and reared separately in SPF isolators to prepare experimental standard sera. Seventeen groups of SPF chickens were sequentially immunized with inactivated AIV (H1N1, H2N2, H3N8, H4N6, H5N1, H6N5, H7N3, H8N4, H9N2, H11N9, H12N5, H13N6 and H14N5), Newcastle disease virus (NDV), avian infectious bronchitis virus (IBV), infectious bursal disease virus (IBDV) and avian leukosis virus-J subgroup (ALV-J) emulsified 1:1 in complete Freund’s adjuvant by subcutaneous injection. The above virus strains were preserved at -80 °C at the author’s laboratory. Another group of two SPF chickens were injected with phosphate buffered saline (PBS) as a negative control. After two weeks, a second immunization was administered with the above inactivated viruses emulsified 1:1 in incomplete Freund’s adjuvant. A booster was administered at a two-week interval. Standard monospecific antisera were prepared from the blood collected 10 days after the booster immunization (Table 1). Standard negative sera were harvested from SPF chickens prior to immunization. After the blood was collected from leg veins of SPF chickens, they were euthanized in CO2. These experimental sera were separated from whole blood by centrifugation at 2,000 × g for ten minutes after overnight incubation at 4 °C and then stored at -40 °C until use. They were previously evaluated by HI assays or commercial ELISA kits and then subjected to the cELISA assay.
Table 1
The information profile of standard positive chicken antisera
| Strain name | Strain description | Highly similar sequencesa (≥ 99%) | HI titer (2x) |
| AIV-H1N1 P2009 | H1N1 | MH061695.1 | 10 |
| AIV-H2N2 21103 | H2N2 | L11134.1 | 10 |
| AIV-H3N8 11102 | H3N8 | CY005816.1 | 8 |
| AIV-H4N6 20411 | H4N6 | GU052381.1 | 7 |
| AIV-H5N1 060315 | H5N1 | JX565019.1 | 9 |
| AIV-H6N5 20411 | H6N5 | CY014656.1 | 9 |
| AIV-H7N3 201369 | H7N3 | JQ906576.1 | 6 |
| AIV-H8N4 20413 | H8N4 | CY014659.1 | 9 |
| AIV-H9N2 201313 | H9N2 | KF059279.1 | 10 |
| AIV-H11N9 21103 | H11N9 | CY014687.1 | 8 |
| AIV-H12N5 11103 | H12N5 | GU052216.1 | 5 |
| AIV-H13N6 11103 | H13N6 | CY014694.1 | 8 |
| H14-A1131028 | H14N5 | KF986854.1 | 10 |
| NDV (Lasota) | NDV | DQ195265.1 | 9 |
| IBV-J (F8)050309 | IBV | FJ849834.1 | - |
| IBDV-NB(F7) | IBDV | AY319768.2 | - |
| ALV-J | ALV-J | KM655820.1 | - |
| aAccession number from the GenBank databases |
In this study, a total of 400 clinical serum samples from chicken, duck and peacock were obtained. The 260 chicken and duck clinical serum samples used in this study were collected from poultry farms of Jiangsu provinve in China, and categorized into positive or negative sera by HI assay. The 140 peacock clinical serum samples were gathered from Nanjing Hongshan Forest Zoo, of which the 100 positive peacock clinical sera verified by HI assay were collected sequentially one month after the imunization of reassortant AIV (H5 + H7) trivalent vaccine, and 40 negative peacock clinical sera were collected before immunization.
Identity and preparation of antigen
Each 10-day-old SPF embryonated chicken egg was inoculated via the allantoic cavity with one strain of H7 AIV diluted in 100 µL 0.01 M sterile PBS. During the incubation period at 37 °C for 2–3 days, eggs were monitored twice daily for embryo mortality. Embryos which died within 24 h were discarded. After eggs were chilled at 4 °C overnight, allantoic fluids confirmed by hemagglutination (HA) test were aseptically harvested and combined. The allantoic fluids of H7 viruses were further inactivated by administration of β-propiolactone diluted at a proportion of 1:2,000 with PBS at 4 °C for 12 h. An inactivated status was detected by antigen challenge in 10-days-old SPF eggs for 3 days to check for live viruses by HA test. The validated inactivated virus suspensions were further clarified by centrifugation at 12,000 × g at 4 °C for 10 minutes. Virus supernatants were collected and pelleted by ultracentrifugation at 110,000 × g at 4 °C for 1.5 h and subsequently the virus precipitates were suspended by 3–6 mL PBS. Thereafter, the precipitated viruses were ultracentrifuged at 160,000 × g at 4 °C for 3 h by using the sucrose gradient centrifugation to remove impurities of different densities. Finally the purified viruses underwent ultracentrifugation at 160,000 × g at 4 °C for 2.5 h to remove sucrose followed by suspension with 2 mL PBS. The successfully purified viruses were evaluated and stored at -80 °C. All viral manipulations were performed under appropriate biosafety level 2 laboratory conditions.
Preparation of anti-H7-HA1 mAbs
The mAb 2F8 directed against recombinant H7-HA1 protein was produced as described previously [17]. In brief, six-week-old female BALB/c mice were first injected at multiple sites subcutaneously with 50 µg of recombinant H7-HA1 protein mixed with an equal volume of Freund’s complete adjuvant, and then injected intraperitoneally with 50 µg of recombinant H7-HA1 protein mixed with an equal volume of Freund’s incomplete adjuvant for two times at 14 and 28 days after the first injection. Three days after the final injection with 100 µg of the protein, spleen cells from the immunized mice were fused with myeloma cells (sp2/0), and the positive hybridoma were screened by indirect ELISA results against recombinant H7-HA1 protein. The selected clones were further subcloned four times by limiting dilution to obtain stable monoclone, and confirmed by western blot assay, immunofluorescence assay, HI assay, blocking assay and netrulization assay.
Each mouse received an intraperitoneal inoculation of 0.5 mL liquid paraffin. Following 7 days, hybridoma cells were diluted in serum-free RPMI-1640 medium. The mAb 2F8 ascites were prepared by injecting intraperitoneally 5 × 106 diluted positive hybridoma cells into each paraffin-primed six-week-old BALB/c mice. Following a further 7 days, mouse ascites production were monitored daily. Ascites were considered to have generated when obvious abdominal swelling was observed and skin tension was palpable. At the end of the experiment, all the mice were first anesthetized with isofluorane and then euthanized by cervical dislocation. Subsequently, the mAb 2F8 was purified from ascites via caprylic acid-ammonium sulfate precipitation followed by HiTrap Protein G affinity chromatography (Amersham, Sweden) according to the manufacturer’s instructions.
Establishment of the cELISA immunoassay
The cELISA assay runs according to the workflow illustrated in Fig. 1. 96-well microtiter plates (JET BIOFIL, China) were coated overnight at 4 °C with 100 µL coating buffer (0.05 M carbonate/bicarbonate buffer, pH = 9.6) containing the purified H7 inactivated whole virus particles at working concentration. After five rapid washes with 300 µL PBS buffer containing 0.05% Tween-20 (PBST), plates were then blocked with 300 µL blocking solution (PBS buffer containing 5% skimmed milk) at 37 °C for 1 h. After rinsing five times with PBST, equal volumes of unknown serum samples and competitive mAb diluted to working concentration with blocking buffer were mixed and incubated simultaneously at 37 °C for 1 h. Wells added with competitive mAb mixed with standard monospecific antisera, negative sera or no serum served as positive, negative or blank control respectively. To provide results statistically valid, each serum sample was tested in triplicate. After washing steps for five times, 100 µL horseradish peroxidase-labeled goat anti-mouse IgG conjugate diluted to an optimal dilution was placed in each well at 37 °C for 30 minutes, followed by the same washing steps as described above. After 100 µL tetramethylbenzidine (TMB) substrate was added to each well followed by incubation in the dark at room temperature for 5 minutes, 50 µL stop solution (2 M sulfuric acid) per well was added to terminate the colorimetric reaction and optical density at 450 nm (OD450) were measured by an automated multimode reader. Results were interpreted as the percentage of inhibition (PI) calculated according to the following formula: % inhibition = ((OD450 of mAb-OD450 of serum sample)/OD450 of mAb) × 100%.
To standardize the assay, the optimal concentrations of inactivated antigen and mAb were determined by means of checkerboard titration. The coating antigen concentrations of 2, 4, 6 and 8 µg/mL and the two-fold serial dilutions of the purified mAb ascites fluids varied from 0.3125 µg/mL to 1.25 µg/mL were tested respectively. The optimal parameters were determined by the highest PI value and then used in the following assays. Subsequently, a receiver-operating characteristic (ROC) curve analysis was carried out to determine the cut-off points and assess the discriminating performance of the established cELISA assay for H7 antibody detection with HI assay as the gold standard test.
Specificity and sensitivity for H7 antibody detection
A panel of standard monospecific antisera produced against other non-H7 AIVs, NDV, IBV, IBDV and ALV-J were used to assess the specificity of the cELISA assay. In order to test the sensitivity of established cELISA assay to detect H7 antibodies, two-fold serial dilutions of chicken, duck and peacock hyperimmune sera against H7 subtype starting from one-quarter were subjected to the cELISA by comparison to homologous HI assay. The chicken hyperimmune sera used for sensitivity test were selected from both experimental standard sera and field sera samples confirmed by HI assay; the duck and peacock hyperimmune sera were selected from field sera samples confirmed by HI assay.
Repeatability test
The same positive and negative chicken, duck and peacock sera verified by HI assay were tested by the same batch and different batches of constructed ELISA plates to analyze the reproducibility and reliability of the cELISA immunoassay. The chicken sera were selected from prepared experimental standard sera and field sera samples verified by HI assay; the duck and peacock sera were selected from clinical sera samples verified by HI assay. The mean PI values, standard deviation (SD) and coefficient of variation (CV) were calculated to measure the uniformity within and between batches of the cELISA assay. The CV is defined as the ratio of the SD to the mean.
Clinical applicability study
For evaluation of clinical application performance, a pool of 400 field serum samples from chicken, duck and peacock was tested for the presence of H7 antibody by the cELISA assay and homologous HI test in parallel. Subsequently, The diagnostic sensitivity, specificity and accuracy of the cELISA assay compared to HI test were calculated by the following formulas: Sensitivity = True positive/(True positive + False negative) × 100%, Specificity = True negative/(False positive + True negative) × 100% and Accuracy=(True positive + True negative)/total number of serum samples tested × 100%.
HI assay
The HI assay was performed in microtiter plates as described previously with 1% chicken red blood cells and 8 hemagglutinin units (8 HAU) of virus [19]. One HA unit is represented by the highest dilution of virus giving complete hemagglutination. The HI titer is expressed as the highest dilution of serum resulting in complete inhibition of 8 HAU of virus. HI titer was regarded as being positive if there is inhibition at a serum dilution of 1/16 or more against 8 HAU of virus. In order to validate the results of the established cELISA for H7 antibody detection, all sera were tested by both HI assay and the cELISA for H7 antibody detection in parallel.
Statistical analysis
The area under ROC curves (AUC) and optimal cut-off points of the cELISA assay were determined with HI assay as the gold standard method by using Graphpad Prism Software Version 5.0. The AUC value describes an overall summary statistic of diagnostic accuracy which can distinguish between non-informative (AUC = 0.5), less accurate (0.5 < AUC ≤ 0.7), moderately accurate (0.7 < AUC ≤ 0.9), highly accurate (0.9 < AUC < 1) and perfect tests (AUC = 1). The optimal cutoff points were ascertained as the serum antibody titers corresponding to the PI values at which both specificity and sensitivity of the assay were maximized.
The correlation between the cELISA and HI test was interpreted by the Pearson correlation coefficient (r) which varies between − 1 and + 1. Zero implies there is no correlation while 1 implies a perfect correlation. The strength of the correlation rises from 0 to + 1, and from 0 to -1. The p-value manifests the probability that the strength of correlation may occur by chance.
The strength of agreement between the cELISA and homologous HI test for field serum sample testing was assessed by Cohen’s κ value through IBM SPSS Statistics 23. The κ values were interpreted in line with the criteria given by Landis and Koch [20]. (κ ≤ 0.00 was designated as poor agreement, 0.00 < κ ≤ 0.20 slight agreement, 0.21 < κ ≤ 0.40 fair agreement, 0.41 < κ ≤ 0.60 moderate agreement, 0.61 < κ ≤ 0.80 substantial agreement, and 0.81 < κ < 1.00 almost perfect agreement.)