Development of a New Multispecies Protein A-Based ELISA for Serodiagnosis of Hepatitis E Infection with Validation in Domestic Swine

Introduction

Approximately 20 million new infections of the hepatitis E virus (HEV) occur annually worldwide, with an average of 3.3 million cases of symptomatic acute hepatitis, which may evolve into chronic hepatitis (WHO, 2023). HEV is transmitted via the oro-fecal route and can be found in several animal species, such as domestic swine, wild boar, deer, and rabbits (Wang and Meng, 2021). In Brazil, only HEV-3 has been reported in humans and animals (Dos Santos et al., 2023). Typically, the diagnosis of HEV infection depends on the detection of HEV RNA in blood and stool or by the detection of IgG or IgM antibodies by ELISA (Zhao and Wang, 2023). Commercially available HEV ELISAs are costly and limited at a global level, especially in Brazil, resulting in weakened surveillance in domestic and wild reservoirs. Here, we developed a new multispecies protein A-based ELISA HEV and validated its use in swine.

Methods

The recombinant gene containing a truncated ORF2 gene of the HEV-3 “PRsw1,” recently found in Brazil (accession number OQ433914), was expressed in a BL21 Star Escherichia coli strain and purified using an ion-exchange chromatography, as previously described (Tavares et al., 2020). Details of the HEV-3 capsid-based recombinant antigen constructed in this work, named swHEV CP2, can be found in supplementary material (Supplementary Figs. S1 and S2).

A total of 212 swine sera were obtained from farms and slaughterhouses located in the Metropolitan Region, Zona da Mata, and Agreste regions of the state of Pernambuco (Brazil) between 2017 and 2018. The study was approved by the Ethics Committee on Animal Use of the Aggeu Magalhães Institute, protocol number 90/2015.

For the in-house ELISA validation, 95 swine sera were simultaneously tested with two commercial kits: (1) PrioCHECK™ Porcine HEV Ab Strip (Thermo Scientific, Waltham, MA) and (2) ID Screen^® Hepatitis E Indirect Multi-species (Innovative Diagnostics-ID, Grabels, France) referred to simplify as the PriCHECK and ID kits in other parts of this work. Dilution curves with multiple antigen and sample combinations were accessed to determine the optimal concentrations of sample and antigen in the tests, and the best parameters were set for the validation step. The background ELISA protocol followed the previously described method (Bezerra et al., 2022).

Results

The optimal concentration for the ELISA was 250 ng of swHEV CP2 antigen per well and a sera dilution of 1:500 (Fig. 1A and 1B). A total of 79 out of 95 sera tested presented concordant results (n = 53 positives/n = 26 negatives), demonstrating substantial agreement with both commercial assays (Kappa = 0.778) (Supplementary Fig. S3). This group of samples was termed Training Group (TG). Next, we used the TG (n = 79) to calculate the optimal cut-off value (OD = 0.1796) (Fig. 1C). Out of the 53 positive samples from the TG, 50 samples tested positive (true-positive) and 3 negative (false-negative) when using the swHEV CP2 ELISA. All negative samples within the TG were also negative in our test (Fig. 1C). The detailed results for the swHEV CP2 ELISA in comparison to the commercial kits separately and combined are in the supplementary material (Supplementary Fig. S4, Supplementary Tables S1 and S2). The swHEV CP2 ELISA exhibited an area under the curve (AUC) of 0.99 (Fig. 1D), with a sensitivity of 94% (CI 95%: 84.5–98.4) and a specificity of 100% (CI 95%: 87.1–100) and these results were also confirmed by latent class model analysis (Supplementary Table S3). Additionally, there was high agreement between swHEV CP2 and both commercial kits combined (TG) (Kappa index = 91.6) (Fig. 1F and Supplementary Table S2).

OPEN IN VIEWER

FIG. 1.

swHEV CP2 ELISA standardization and validation: A) 15% polyacrylamide electrophoresis for purified swHEV CP2 (the swHEV CP2 has a six-histidines tag on its C-terminal portion and ∼60 kDa of molecular mass). B) Standardization of multiple swHEV CP2 antigen and serum sample dilutions. These tests were performed using positive and negative swine sera pools. C) Blank-adjusted optical densities of the sera from the Training Group (n = 79), assigned as positive negative according to the PrioCHECK + Innovative Diagnostics combined results (left side of the graph) or according to each gold-standard kit individually (mid and right side of the graph). The cut-off for determination of the swHEV CP2 ELISA results was 0,1796. D) Receiver operating characteristic curve using the training group with two combined methods as reference. The results show a high area under the curve (AUC), indicating the good performance of the swHEV CP2-based ELISA. E) High correlation rates were observed between the optical density of the swHEV CP2 ELISA and both reference kits, indicating proportionality between the measures with these technologies. F) Forest plot chart showing the diagnostic performance rates of the test, when compared to the training group and each reference kit individually.

The concordance with the ID kit was 84.5, whereas for the PrioCHECK, it was 57.9 (Fig. 1F and Supplementary Table S1). The optical density signal of the swHEV CP2 ELISA showed a good correlation rate with both reference kits (PrioCHECK rho = 0.89; ID rho = 0.84), indicating the robustness of our test (Fig. 1E). Using the ID kit as the reference, the positive predictive value was 93.2%, and the negative predictive value was 92.1%. Conversely, with the PrioCHECK kit, they were 93.2% and 61.8%, respectively (Fig. 1F). Lastly, seroprevalence within the swine population in Pernambuco state, determined by swHEV CP2 ELISA, was 57.5% (122/212) (CI 95%: 50.6–64.3), with the majority of positive cases observed in animals aged 6 months or older (Supplementary Tables S4 and S5).

Discussion

Our newly developed multispecies ELISA was able to detect specific antibodies against HEV in domestic pig sera accurately. In-house kits using different approaches have been developed in Brazil and neighboring countries (Gutierrez et al., 2023; Pandolfi et al., 2017). However, their reliance on specific secondary antibodies limits their applicability for testing multiple species, given the extensive range of HEV hosts. This constraint hampers the feasibility of multispecies surveillance. In addition, because different performance rates between species have already been shown in ELISAs based on protein A (Bezerra et al., 2022), additional validation of this in-house ELISA is required prior to its use in other species. Nevertheless, our new in-house ELISA demonstrated similar diagnostic performance compared to commercial tests that are costly and difficult to obtain in Brazil.

Moreover, our findings indicate a substantial circulation of the HEV among swine farms in northeastern Brazil. Despite the seroprevalence rates being slightly lower than the 82.1% found in the same region previously (de Oliveira-Filho et al., 2017), the values are within the range of 8.6– 88.4% reported for Brazil (Moraes et al., 2021).

Conclusion

We anticipate that the new ELISA test for HEV serologic diagnosis in swine will contribute to mitigate the underreporting of hepatitis E cases. Importantly, it should be noted that hepatitis E stands out due to its significant zoonotic component, with domestic pigs serving as reservoirs. The novel assay will potentially increase information on the epidemiology of HEV, thereby assisting health authorities in devising effective strategies to mitigate HEV infections. Further collaborative one-health efforts should prioritize the investigation of the possible consequences of HEV circulation, including the occurrence of zoonotic transmission and the extent of hepatic disease in humans.