Preparation of recombinant canine NGAL
The coding sequence for canine NGAL (GenBank accession no. 491320) was adapted to E. coli codon usage using Java Codon Adaption Tool [18]. The synthetic sequence was cloned into pET28a (+) vector (Addgene, Shanghai, China) as an NdeI/BamHI fragment. The coding sequence was expressed as a C-terminally His-tagged protein under the control of T7 promoter. The recombinant vector was transformed into E. coli strain BL21 (DE3) and the expression of recombinant protein was induced overnight with 0.2 mM IPTG at 15 ℃. The bacterial cells were collected by centrifugation for 10 min at 5,000 g, suspended in PBS (1/10 culture volume), and disrupted two times at 1300 bar using a High Pressure Cell Disruptor (JNBIO, Guangzhou, China). After centrifugation for 10 min at 12,000 g, the recombinant protein was purified from the clarified cell lysate using Ni-NTA Sefinose Column (Sangon Biotech, China) and Hitrap QFF Anion Exchange Column (GE Healthcare, USA). The 6 × His tag was removed by digestion with thrombin protease (Sangon Biotech, China) by following the manufacturer’s instruction. The identity of the purified protein was determined using commercial canine NGAL ELISA Kit (Sinobest Biotech Ltd., Shanghai, China).
Generation of hybridoma cell lines
Hybridoma cell lines were generated as previously described [19, 20]. Three female BALB/c mice (6–8 weeks old) were purchased from and housed in a single cage in the SPF-free facilite of Experimental Animal Center, Nantong University. Each mouse was immunized intramuscularly with recombinant canine NGAL (50 µg/50 µL) emulsified in 50 µL of Freund’s complete adjuvant (Sigma, USA). The mice were immunized two more times with the same amount of the antigen emulsified in Freund’s incomplete adjuvant at 2-week intervals. After three immunizations, mice were boosted intravenously with 100 µg of recombinant canine NGAL without adjuvant. The immunized mice were anaesthetized by carbon dioxide and sacrificed by cervical dislocation at day 4 post boosting immunization. The splenocytes were collected and fused with SP2/0 myeloma cells at a ratio of 10:1 in the presence of polyethylene glycol. The fused cells were suspended in HAT (hypoxanthine, aminopterin and thymidine) medium and plated onto 96-well plates. Four days later, HAT medium was replaced with HT medium. The supernatants of hybridoma cells were screened by indirect ELISA using recombinant canine NGAL as the antigen. The identity of recombinant canine NGAL was confirmed using commercial canine NGAL ELISA Kit by following the manufacturer’s instruction. The positive hybridoma cells were cloned two times by dilution cloning for further study.
Characterization of canine NGAL-specific mAbs
To prepare the mAbs in large quantities, 10 female BALB/c mice (8–9 weeks old) were pre-injected with 0.5 mL of paraffin oil (Sigma-Aldrich, USA) followed by intraperitoneal injection with hybridoma cells (5 ⋅ 106 cells/0.2 mL). The ascites were collected one week later and the mAbs used as the capture and detection antibodies of simplified sandwich ELISA were purified by ammonium sulfate precipitation as previously described [21].
The isotypes of mAbs were determined using Mouse mAb Isotyping Kit (Sigma-Aldrich, USA) by following the manufacturer’s instruction. The mAbs were identified by Western blotting using canine NGAL reference. Briefly, canine NGAL was purified from dog urine samples with AKI as previously described [22]. The purified protein (50 µg) was separated by SDS-PAGE and transferred onto PVDF membrane. After blocking with 5% defatted milk powder in PBST (PBS containing 0.05% Tween 20) for 2 h at room temperature, the ascetic mAbs (1:5000 in blocking buffer) were added and incubated overnight at 4 ℃. After three-time washing in PBST, HRP-conjugated goat anti-mouse IgG (1:1000 in blocking buffer;Beyotime Biotechnology, China) was added and incubated as described. After washing again, the hybridization signals were developed using BeyoECL Plus Kit (Beyotime Technology, China) by following the manufacturer’s instruction.
The ascitic mAbs were titrated by indirect ELISA using recombinant canine NGAL as the antigen. Briefly, 96-well plates were coated overnight with the recombinant antigen (2 ∝g/mL) and blocked with PBS containing 5% BSA for 2 h at room temperature. Serially diluted mAbs were added (100 µL/well) and incubated for 1 h at room temperature. After wash three times with PBST (PBS and 0.5% Tween 20), HRP-conjugated goat anti-mouse IgG (1:2000; Beyotime, China) was added (100 µL/well) and incubated again as described. After washing three times again, freshly prepared tetramethylbenzidine (TMB) solution was added and incubated in dark for 10 min at 37 ℃. OD450 values were determined for P/N value calculation after stopping with 2 M H2SO4. The mAb affinities for canine NGAL was tested by ELISA as previously described. The Ka value was calculated according to the formula: Ka = (A0-Ai) / li ⋅ Ai [23].
Selection of antibody pair for simplified sandwich ELISA
The additivity of each mAb pair for canine NGAL was tested by competing ELISA as previously described [24]. The additive index (AI) was defined for a pair of mAbs according to the formula: AI = {[2 ⋅ A1 + 2/(A1 + A2) ]-1} ⋅ 100, where A1, A2 and A1 + 2 indicate the OD values of the first mAb, second mAb and two mAbs together in the ELISA. Provided the concentrations of mAbs are saturated for the antigen, the AI will tend to be zero if both mAbs recognize the same epitope, but close to 100 if both mAbs bind to distinct sites. The mAb pairs with high AI values were further tested by checkerboard ELISA and the best antibody combination was selected according to the highest P/N ratio.
Establishment of simplified sandwich ELISA
The simplified sandwich ELISA was established according to previously described method [8].The detection antibody was labeled with a 1:2 mAb to HRP (Thermo, USA) ratio according to the manufacturer’s instruction. To optimize the concentrations of the antibody pair, ELISA was performed with different concentrations of NC-5 mAb as the capture antibody and different dilutions of HRP-conjugated NC-3 mAb as the detection antibody. To optimize the capture antibody coating buffer, ELISA plates were coated with NC-5 mAb in carbonate buffer (pH 9.6), Tris-HCl buffer (pH 8.0) or PBS (pH 7.4). To optimize the blocking buffer, the antibody-coated plates were blocked with PBST (0.01% Tween 20 in PBS) containing 1% casein or 5% BSA for 3 h at room temperature. To optimize the antigen reaction time, the antigen-added plates were incubated for different times at room temperature. To optimize the incubation time for detection antibody, HRP-conjugated NC-3 mAb was diluted in PBST containing 1% casein or 5% BSA and incubated for different times at room temperature. Finally, ELISA plates were coated overnight with NC-5 mAb (2 µg/mL) at 4 ℃ and blocked with PBST containing 5% BSA for 2 h at room temperature. The recombinant canine NGAL or clinical samples was added and incubated for 45 min at room temperature. After washing three times with PBST, HRP-conjugated NC-3 mAb (1:8000) was added and incubated as described. After washing again, freshly prepared TMB solution was added and incubated in dark for 15 min. Subsequent to termination with 2 M H2SO4, OD450 values were measured on ELISA reader (Enspire 2300, PerkinElmer, USA).
Generation of standard curve for simplified sandwich ELISA
The standard curve for simplified sandwich ELISA was generated with different concentrations (0, 6.25, 12.5, 25, 50, 100, 200, 400, 800 and 1600 ng/mL) of recombinant canine NGAL using. Intra-assay coefficient (20 repeats) inter-assay coefficients (10 repeats) of variation values were determined using different concentrations of recombination canine NGAL (15 and 150 ng/mL)[25].
Validation of simplified sandwich ELISA
Both serum and urine samples were collected from healthy (n = 21) and AKI canine patients (n = 21) at Ai-Bi Pet Hospital and KangMei Pet Hospital (Nantong, China). The samples were detected in triplicates using the simplified sandwich ELISA and commercial canine NGAL ELISA kit (Sinobest Biotech, Shanghai, China) by following the manufacturer’s instruction. The agreements between the concentrations detected by two different ELISA were submitted to linear regression analysis using Origin 8 software (OriginLab, Northampton, USA).