Preliminary investigation of hepatitis E virus detection by a recombinase polymerase amplification assay combined with a lateral flow strip

Abstract

Hepatitis E virus (HEV) is a zoonotic pathogen that is a significant public health problem. Detecting HEV relies mainly on conventional PCR, which is time-consuming and requires sophisticated instruments and trained staff. We aimed to establish a reverse-transcription (RT)-recombinase polymerase amplification (RPA) assay (RT-RPA) combined with a lateral flow strip (LFS; RT-RPA-LFS) to rapidly detect HEV RNA in human and rabbit samples. With the optimal reaction conditions (37°C for 30 min), our assay detected as few as 1.0 × 10² copies/mL of HEV and showed no cross-reactivity with other hepatitis viruses. We tested 28 human samples (4 fecal and 24 serum samples) and 360 rabbit samples (180 fecal and 180 serum samples) with our RT-RPA-LFS assay and compared our assay to an RT-qPCR method. There was no significant difference (p > 0.05) in the test results between the 2 assays. Our RT-RPA-LFS assay detected both HEV3 and HEV4 genotypes. Our rapid, sensitive, and specific RT-RPA-LFS assay for the detection of HEV may provide a useful detection tool for limited-resource areas.

Hepatitis E virus (HEV; Hepeviridae, Paslahepevirus balayani) is the most common cause of acute viral hepatitis worldwide and is a global public health problem. ⁷ According to the World Health Organization, HEV causes ~20 million human infections, 3.3 million symptomatic cases, and 44,000 deaths annually. ⁶ HEV is classified into 8 genotypes (HEV1–8). ¹⁵ HEV infection commonly results in asymptomatic and self-limiting disease, ⁹ but can cause chronic infection and even cirrhosis in immunocompromised individuals. ¹³ During pregnancy, HEV infection is associated with a mortality rate as high as 20%. ¹⁴

HEV is recognized as an emerging zoonotic pathogen and has been identified in a large variety of animals, including swine, rabbits, wild boar, and deer. ⁴ The main animal hosts of HEV are swine and rabbits. ⁸ Rabbit HEV, which was first isolated in China and is prevalent worldwide, belongs to a specific subgenotype of HEV (HEV3ra). ³ Although most zoonotic HEV infection cases are related to HEV3 and HEV4 from swine, recent cases of HEV3ra infections in humans have been reported, ¹ raising public health concerns about HEV zoonotic transmission.

HEV infection is detected mainly through serologic assays, viral nucleic acid detection, and antigen detection. ¹⁷ Among these assays, the detection of HEV RNA is the gold standard. ¹⁷ During the acute and chronic phases of infection, HEV RNA can be detected in the serum, feces, and urine of HEV-infected patients. ⁹ Additionally, detecting HEV RNA is important for identifying HEV infection in immunocompromised individuals, in which anti-HEV antibody results can be negative. ⁵ HEV RNA is commonly detected by PCR-based methods, including reverse-transcription nested PCR (RT-nPCR) and real-time RT-PCR (RT-rtPCR) assays. ¹⁷ However, these methods usually require sophisticated instruments and trained staff, limiting their application in low-income countries. Thus, there is an urgent need to develop an HEV RNA detection assay that is rapid, sensitive, and cost-effective.

Recombinase polymerase amplification (RPA), a promising new isothermal amplification technology, has been developed to detect many pathogens, including hepatitis B virus (HBV), ¹² SARS-CoV-2 (Severe acute respiratory syndrome-related coronavirus), ¹⁶ and human immunodeficiency virus (HIV). ² In RPA, thermostable DNA polymerase is used for the amplification step at a constant temperature (37–42°C), within 30 min. ¹⁰ Then, the amplification products of RPA can be visualized by a lateral flow strip (LFS) in 10 min. We aimed to establish a reverse-transcription (RT)-RPA-LFS assay specific for HEV RNA. We also evaluated the detection sensitivity and specificity of our RPA-LFS assay.

We obtained 31 human serum and fecal samples from Zhengzhou University Fifth Affiliated University (Zhengzhou, China), including 28 samples (collected from anti-HEV IgM–positive patients) that had been submitted for HEV RNA testing, and 3 HEV-negative samples that were positive for hepatitis A virus (HAV; Hepatovirus A), HBV, or hepatitis C virus (HCV; Hepacivirus C) to be used for cross-reactivity determination. These samples were screened by PCR assays that we have described previously, ¹¹ and the products were sequenced. The ORF2 regions of the sequences were compared using BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Eight of these 28 samples were confirmed to be HEV RNA–positive, including HEV3 (n = 2, GenBank KT633715) and HEV4 (n = 6, GenBank KC163335). The viral loads of these 8 samples were also calculated as described previously. ¹¹ For the limit of detection (LOD) analysis, 1 of these 8 samples with a viral load of 1.0 × 10⁶ copies/mL was used as a standard. HAV-positive (n = 1), HBV-positive (n = 1), and HCV-positive (n = 1) samples were used for the cross-reactivity test. Serum samples that were PCR-negative for HEV (n = 20) were used as negative controls.

We also collected a fecal sample and a serum sample from each of 180 three-mo-old Japanese White laboratory rabbits (weighing 2–3 kg each) to screen for HEV RNA. These laboratory-housed rabbits were obtained from the Department of Laboratory Animal Science of the National Institutes for Food and Drug Control (Beijing, China). All samples were stored at −80°C pending use. Our study was approved by the Ethics Committee of Zhengzhou University Fifth Affiliated University (KY2021031) and the Committee of Laboratory Animal Welfare and Ethics of the National Institutes for Food and Drug Control (2022B012).

The 10% fecal suspensions were prepared by mixing fecal samples in PBS and then centrifuged at 2,400 × g for 10 min. Nucleic acids were extracted from both fecal suspensions and serum samples (EasyPure viral DNA/RNA kit; TransGen) according to the manufacturer’s instructions. RNA samples were eluted with 20 µL of distilled water and then stored at −80°C until analyzed.

The RPA primers used were designed with Primer Premier 5.0 software (Premier Biosoft) and synthesized by Sangon Biotech. The RPA primer design was in accordance with the TwistAmp RT basic kit manual (Twist Dx), with a product size of 100–200 bp and a primer size of 30–35 nt. A primer pair targeting the HEV ORF2 region (forward primer: 5′-CATTCAGCAGTATTCTAAGACATTTTATGTTC-3′; reverse primer: 5′-TTGATCACTAGCAGTGGTGTTATAGTTATATG-3′) was designed based on the nucleotide sequence of HEV strain Kernow-C1 (GenBank HQ709170). The 5′-end of the forward primer was labeled with fluorescein isothiocyanate (FITC), and the 3′-end of the reverse primer was labeled with biotin.

The RT-RPA assay was performed with a TwistAmp RT basic kit (Twist Dx). Briefly, a 50-µL reaction was prepared, including 29.5 µL of rehydration buffer, 2 µL of RNA template, 2.1 µL of forward primer, 2.1 µL of reverse primer, nuclease-free water (Solarbio) to adjust the volume to 47.5 μL, and 2.5 µL of magnesium acetate (280 mM). After being fully mixed and centrifuged briefly, the mixture was incubated at 30°C, 35°C, 37°C, or 42°C for 15, 20, 25, or 30 min. Then, 10 µL of the RT-RPA products were diluted 1:10 with buffer and added to a LFS (Genenode). The results were visualized in 5 min. Each experiment was performed in triplicate.

To assess the LOD of the RT-RPA-LFS assay, a sample with a viral load of 1.0 × 10⁶ copies/mL was 10-fold serially diluted with ddH₂O to 1.0 × 10 copies/mL. To evaluate the cross-reactivity of the RT-RPA-LFS assay, we extracted and tested the nucleic acids of HAV-, HBV-, and HCV-positive samples with our RT-RPA-LFS assay. Each experiment was performed in triplicate.

The agreement between our RT-RPA-LFS assay and our RT-qPCR assay was assessed by Cohen kappa statistics. The differences between the 2 assays were compared by the chi-square (χ²) method. All statistics were performed using SPSS v.18.0 software (IBM); p < 0.05 was considered significant.

The positive signals were most obvious by incubating at 37°C for 30 min, which was then considered the optimal reaction condition for our RT-RPA-LFS assay (Fig. 1). Test bands were visible from 1.0 × 10⁶ to 1.0 × 10² copies/mL; hence, the LOD of our assay for HEV detection was 1.0 × 10² copies/mL (Fig. 2). HEV-positive samples were positive and HAV-, HBV-, and HCV-positive samples were negative in the RT-RPA-LFS assay (Fig. 3).

Figure 1.

Optimization of conditions for a recombinase polymerase amplification assay combined with a lateral flow strip assay for hepatitis E virus detection. A. Different incubation temperatures (30°C, 35°C, 37°C, 42°C). B. Different incubation times (15, 20, 25, 30 min). C = control line; NTC = negative test control; T = test line.

Figure 2.

Sensitivity of the RT-RPA-LFS assay for hepatitis E virus (HEV) detection. To determine the sensitivity, 10-fold dilutions of HEV RNA ranging from 1.0 × 10⁶ to 1.0 × 10¹ copies/mL were detected. C = control line; T = test line.

Figure 3.

Cross-reactivity of the RT-RPA-LFS assay for hepatitis E virus (HEV) detection. The cross-reactivity was determined by testing nucleic acids extracted from HEV-, HAV-, HBV-, and HCV-positive samples. C = control line; NTC = negative test control; T = test line.

The human and rabbit samples were considered HEV RNA–positive when their Ct values were < 40 in at least 2 replicates in the RT-qPCR assay; 8 of 28 human samples were HEV RNA–positive by both our RT-RPA-LFS assay and the RT-qPCR assay (Ct values of 32.4 ± 0.1 to 37.2 ± 0.6), and 20 of 28 human clinical samples were negative by both assays (Table 1). For rabbit samples, 36 of 360 samples were HEV RNA–positive (Ct values of 28.6 ± 0.4 to 39.2 ± 0.2); 34 of 360 samples were positive in our RT-RPA-LFS assay. Results of the 2 assays were not significantly different (p > 0.05). The agreement between the 2 assays was good for both human samples (κ = 1.00) and rabbit samples (κ = 0.97; Table 2).

Table 1.

Performance comparison of a recombinase polymerase amplification assay combined with a lateral flow strip (RT-RPA-LFS) and RT-qPCR for the detection of hepatitis E virus in human samples.

Sample	Type	RT-RPA-LFS	RT-qPCR Ct value (n = 3)
1	Feces	+	32.4 ± 0.1
2	Feces	+	34.9 ± 0.5
3	Feces	+	36.7 ± 0.6
4	Feces	+	37.2 ± 0.6
5	Serum	+	32.6 ± 0.4
6	Serum	+	35.1 ± 0.9
7	Serum	+	35.5 ± 0.6
8	Serum	+	35.4 ± 0.3
9–28	Serum	−	Undetermined

Table 2.

Agreement analysis of recombinase polymerase amplification assay combined with a lateral flow strip (RT-RPA-LFS) and RT-qPCR for the detection of hepatitis E virus in human and rabbit samples.

RT-qPCR	RT-RPA-LFS		Total
	Positive	Negative
Positive	8, 34	0, 2	8, 36
Negative	0, 0	20, 324	20, 324
Total	8, 34	20, 326	28, 360

The numbers of samples are listed as “human samples, rabbit samples”.

There are several limitations in our study. First, we did not use known positive and negative samples (generated under controlled conditions). Also, a greater number of both experimentally generated and field samples will need to be used to adequately validate this test before it can be deployed in control programs.