Antigens, antibodies, buffers and samples

The CCV antigen (nucleoprotein, N protein, expressed in E. coli) and CPV-2 antigen (expressed in VP2 protein, expressed in E. coli) and their paired antibodies were prepared by our team. Eu3+ and Sm3+ labeling reagents were obtained from PerkinElmer (Norwalk, USA). The reagents and buffer used in the study were prepared in the laboratory. Sixty-two fecal samples were obtained from suspected CCV-infected dogs, 75 fecal samples were obtained from suspected CPV-2-infected dogs, and 150 healthy fecal samples were obtained from Guangzhou Fu Mao Pet Hospital (Guangzhou, China).

Coating and labeling procedure

CCV (No. 8D6 and 7E10) and CPV-2 (No. P2A3 and P3F6) paired antibodies were used for coating and labeling, respectively. Briefly, 8D6 and P2A3 antibodies were added to the 96-well microplate at a certain concentration and proportion, and the cells were incubated for 2 h at 37 °C. The microplate was blocked with blocking buffer (phosphate buffered saline (PBS) supplemented with 5% bovine serum albumin (BSA)) for 1 h at 37 °C. After drying, the microplate was stored under vacuum at 4 °C. Briefly, Eu3+ and Sm3+ labels were used for 7E10 and P3F6 antibody labeling, respectively. Eu3+/Sm3+ labeling reagent was added to the 7E10 and P3F6 antibodies, which were gently shaken at 4 °C and subsequently purified to obtain the Eu3+-labeled 7E10 antibodies and Sm3+-labeled P3F6 antibodies.

Optimization of detection conditions

The reaction conditions were optimized according to a one-step procedure. The conditions that needed to be optimized included the concentration and proportion of 8D6 and P2A3 antibodies, the amount and proportion of labeling antibody to Eu3+/Sm3+ labels in the labeling procedure, the amount of Eu3+-labeled 7E10 antibodies and Sm3+-labeled P3F6 antibodies, the immunoreaction temperature and time, the washing time and the enhancement solution volume.

The optimized conditions were as follows: 8D6 antibody coating amount: 1 µg/mL, 100 µL/well; P2A3 antibody coating amount: 1.5 µg/mL, 100 µL/well; ratio of Eu3+ labeling reagent to 7E10 antibodies: 500 µg to 1.5 mg; ratio of Sm3+ labeling reagent to P3F6 antibodies: 500 µg to 2 mg; amount of Eu3+ labeling 7E10 antibodies: 90 µL; amount of Sm3+ labeling P3F6 antibodies: 110 µL; 6 washes; immunoreaction time: 50 min; and corresponding immunoreaction temperature: 37 °C.

Kit preparation

The TRFIA kit included the following components: a coated 96-well microplate (vacuum sealing), Eu3+-labeled 7E10 antibodies, Sm3+-labeled P3F6 antibodies, sample diluent buffer (PBS), a series of standards (0, 0.1, 1, 10, 100, 200 and 500 ng/mL), washing buffer, enhancement solution and instructions.

Test procedure

The test procedure was performed on a time-resolved analyzer, and the parameters were set in advance. The test procedure was as follows: 50 µL of sample, 90 µL/well of Eu3+-labeled 7E10 antibodies and 110 µL/well of Sm3+-labeled P3F6 antibodies were added to the coated 96-well plates and then incubated for 50 min at 37 °C. After 5 washes, 200 µL/well of enhancement solution was added to the wells, after which a time-resolved analyzer automatically measured the fluorescence values and calculated the CCV/CPV-2 concentration using built-in standard curves. A schematic diagram of the TRFIA detection principle and steps are shown in Fig. 1.

Fig. 1

Schematic diagram of the TRFIA detection principle and steps

Laboratory sensitivity assay

The CCV/CPV-2 antigen was diluted to concentrations of 0, 0.2, 2, 20, 200, 400 and 1000 ng/mL, and then, an equal volume of antigen was mixed to prepare a series of standards (0, 0.1, 1, 10, 100, 200 and 500 ng/mL). A series of standards were detected from the samples using a TRFIA kit, after which the corresponding fluorescence values were obtained. Three replicates were performed for each concentration. Standard curves were plotted using linear regression and log-log regression. The 0 ng/mL standard was detected 20 times as the sample, and the mean and standard deviation (SD) obtained from the detection were used for determining sensitivity (“sensitivity = mean + 2 × SD”) [16].

Determination of the laboratory reference interval

A total of 150 healthy fecal samples were used for reference interval determination. After dissolution, the fecal supernatant was added to the coated 96-well microplate, and the TRFIA was performed. The CCV/CPV-2 concentration values were tested for normality using SPSS 20.0, and cutoff values were obtained according to the following formula: cutoff = mean + 1.64 × SD [17]. The reference interval was determined through the cutoff value.

Laboratory specificity assay

Several common infectious disease samples from dogs were used for specificity experiments, including the following positive (P) clinical samples: canine distemper virus (CDV, 10 samples), canine parainfluenza virus (CPIV, 15 samples), canine adenovirus type 1 (CAV-1, 8 samples), and rabies virus (6 samples); 150 negative (N, healthy) samples; and the specific recombinant antigens CDV, CPIV, CAV-1 and rabies virus. The above clinical samples were confirmed by PCR testing and clinical symptoms. The samples were diluted/dissolved in PBS, after which TRFIA was performed. According to the reference interval, we determined the negativity and positivity of the samples and obtained the specificity results.

Laboratory accuracy assay

Three concentrations of CCV/CPV-2 antigens (10 ng/mL, 100 ng/mL and 500 ng/mL) were added to the healthy fecal samples and detected using a TRFIA kit. Four replicates were performed for each sample. The coefficient of variation (CV) and recovery were calculated for accuracy validation. CV (%) = SD/mean × 100%. Recovery (%) = (determined concentration-basal concentration)/spiked concentration × 100%. The basal concentrations of CCV and CPV-2 were 4.23 ng/mL and 5.12 ng/mL, respectively.

Laboratory stability assay

The kits were stored in the dark for 20 days at 4 °C or for 7 days at 37 °C. The CCV/CPV-2 antigen (100 ng/mL) was utilized for daily testing, after which the fluorescence values were recorded, and the curves were plotted to evaluate the stability of the kits.

Methodology comparison and clinical sensitivity/specificity evaluation

Sixty-two fecal samples from suspected CCV-infected dogs and 75 fecal samples from suspected CPV-2-infected dogs were detected using PCR and a TRFIA kit, respectively. The results of the PCR method served as the detection criteria. Clinical sensitivity = true positive/(true positive + false-negative) × 100%), clinical specificity = true negative/(true negative + false-positive) × 100%). Pearson’s chi-square test was used to analyze the correlations between the two methods. The null hypothesis is that the two methods exhibit differences.

Statistical methods

The data were statistically analyzed using SPSS 20.0. Pearson correlation analysis was performed, and Bland‒Altman plots and figures were constructed and plotted using GraphPad Prism 5 (GraphPad Software, USA). All the results are presented as the means ± SDs.