Development and evaluation of a pseudovirus-luciferase assay for rapid and quantitative detection of neutralizing antibodies against enterovirus 71

Abstract

The level of neutralizing antibodies (NtAb) induced by vaccine inoculation is an important endpoint to evaluate the efficacy of EV71 vaccine. In order to evaluate the efficacy of EV71 vaccine, here, we reported the development of a novel pseudovirus system expression firefly luciferase (PVLA) for the quantitative measurement of NtAb. We first evaluated and validated the sensitivity and specificity of the PVLA method. A total of 326 serum samples from an epidemiological survey and 144 serum specimens from 3 clinical trials of EV71 vaccines were used, and the level of each specimen's neutralizing antibodies (NtAb) was measured in parallel using both the conventional CPE-based and PVLA-based assay. Against the standard neutralization assay based on the inhibition of the cytopathic effect (CPE), the sensitivity and specificity of the PVLA method are 98% and 96%, respectively. Then, we tested the potential interference of NtAb against hepatitis A virus, Polio-I, Polio-II, and Polio-III standard antisera (WHO) and goat anti-G10/CA16 serum, the PVLA based assay showed no cross-reactivity with NtAb against other specific sera. Importantly, unlike CPE based method, no live replication-competent EV71 is used during the measurement. Taken together, PVLA is a rapid and specific assay with higher sensitivity and accuracy. It could serve as a valuable tool in assessing the efficacy of EV71 vaccines in clinical trials and disease surveillance in epidemiology studies.

Figure 1. Optimization of assay parameters.

(A) Determination of RD cell density. RD cells (5×10⁴ cells per well) were incubated in a 96-well plate for 12 h with various dilutions of pseudovirus (from 2×10³ to 2×10⁷ copies per well) in a final volume of 200 μl per well. The data showed that the RLUs decreased with pseudotype in a dose dependent manner. The variation coefficient was lower than 5% for 5×10⁴ and 1×10⁵ cells per well at all the doses of pseudovirus. However, when the RD cell density was 2×10⁴ cells per well, the variation coefficient of RLU was higher than 15% with lower pseudovirus dosage. So the 5×10⁴ cells per well was chosen as the optimized RD cell density. (B) Evaluation of luciferase activity at various incubation times. RD cells (5×10⁴ cells per well) were incubated with 10⁶ copies pseudovirus per well in a 96-well plate for various time points, from 0.5 to 24 h, and each luciferase reading shown is the average of eight replicates. The results indicated that luciferase activity increased gradually with the extension of the incubation over time and reached a plateau at 12 h after incubation. therefore, 12 h was chosen as the optimized incubation time. (C) Infection curve of EV71(FY)-Luc. RD cells were incubated at 35°C together with a range of EV71 pseudovirus concentrations in a final volume of 200 μl per well in 96-well plates. Luciferase activity was measured 12 h post-infection. Each data point represents the average of eight replicates. Linear regression analysis was conducted using Graph Pad Prism. The results showed that when the dose of pseudovirus were within the range of 1×10⁴–10⁷ copies per well, there was a strong correlation between the viral dose and luciferase activity.

Figure 2. Comparison between CPE and PVLA measurements.

(A) Correlation analysis was performed for 180 double-positive EV71-NtAb samples tested by both the CPE and PVLA methods. The solid line represents the linear regression fitting of the data. The result demonstrated that there was a good relativity between the two assays (r = 0.91, p<0.0001). (B) A Bland-Altman difference plot of EV71-NtAb measurements was constructed using the results of the CPE and PVLA methods. The x-axis corresponds to the average (log₁₀) concentration of EV71-NTAb determined by CPE or PVLA, and the y-axis is a measure of the difference between the concentrations as determined by CPE and PVLA. The solid line represents the mean value, while dashed lines represent the 95% confidence limits. The results indicated that the new method has a high-degree of agreement with the “gold standard” assay. Only four values fell outside ±2 standard deviations (−1.003 to 0.809log₁₀).

Figure 3. Comparison of EV71-NTAb concentration between CPE and PVLA measurements.

Sera were obtained from phase I clinical trials of inactivated EV71 vaccine which were provided by three vaccine manufacturers. NtAb concentration(U/ml) on day 0(pre-vaccination) and day 28(after the first dose, post-vaccination) were tested by the CPE and PVLA methods. The paired t-test showed that the NtAb concentrations determined by the two detection systems were not significantly different (P>0.05).