Development and validation of an improved RT-PCR assay with nested universal primers for detection of hepatitis E virus strains with significant sequence divergence

Abstract

Recent studies revealed that hepatitis E virus (HEV) genomes are more variable than previously thought and well-conserved regions suitable for designing universal primers are limited. In this study, based on alignment of 70 full-length HEV sequences of genotypes 1–4, a part of the ORF2/ORF3 overlapping region was found to be the best target region for PCR amplification of various HEV strains. Using the newly designed primers, an RT-PCR method (ORF2/3-137 PCR) that amplifies a 137-nucleotide (nt) sequence within the ORF2/ORF3 overlapping region and is capable of amplifying all known HEV sequences was developed. When compared with the previous RT-PCR method (ORF2-457 PCR) that amplifies a 457 nt ORF2 sequence, ORF2/3-137 PCR was two to three times more sensitive than ORF2-457 PCR upon testing serial dilutions of three HEV RNA-positive serum samples. The ORF2/3-137 PCR assay could detect viraemia in five patients with acute or fulminant hepatitis E 3–14 days longer than ORF2-457 PCR after disease onset. All 41 ORF2-457 PCR-positive serum samples of various genotypes tested positive for HEV RNA by the ORF2/3-137 PCR assay. Since the amplicons of ORF2/3-137 PCR contain variable sequences, a phylogenetic tree of the ORF2/3-137 products could clearly distinguish the different HEV genotypes.

Introduction

Hepatitis E virus (HEV), the causative agent of hepatitis E, has a single-stranded, positive-sense RNA genome within a non-enveloped capsid (Purcell and Emerson, 2001). The genome is approximately 7.2 kb in size and consists of a short 5′-untranslated region (UTR), three partially overlapping open reading frames (ORF1–3) and a short 3′-UTR terminated by a poly(A) tract (Tam et al., 1991). HEV is classified as the sole member of the genus Hepevirus in the family Hepeviridae (Emerson et al., 2004). In developing countries, HEV is an important public health concern because it causes waterborne outbreaks as well as sporadic hepatitis (Purcell and Emerson, 2001). Recently, besides imported cases of hepatitis E, locally acquired sporadic cases of hepatitis E have been reported in industrialized countries (Harrison, 1999, Schlauder and Mushahwar, 2001, Smith, 2001, Okamoto et al., 2003, Mansuy et al., 2004, Ijaz et al., 2005), where zoonotic transmission of HEV has been suggested (Meng et al., 1997, Meng et al., 1998, Meng et al., 2002, Meng, 2003, Nishizawa et al., 2003, Okamoto et al., 2001, Tei et al., 2003, Yazaki et al., 2003).

Sequence analyses of HEV isolates revealed extensive heterogeneity of the genome and four major genotypes (1–4) have been identified (Schlauder and Mushahwar, 2001). HEV is distributed worldwide and a difference in geographic distribution of the genotypes has been noted (Schlauder and Mushahwar, 2001, Lu et al., 2006). Genotype 1 was isolated from tropical and subtropical countries in Asia and Africa, and genotype 2 was described in Mexico and Africa. Genotype 3 is most widely distributed in the world and has been isolated in North and South America, Europe, Oceania, and Asia. Genotype 4 has been found exclusively in Asia including China, Japan, Taiwan, and Vietnam. Of interest, there is also a difference in the hosts infected by each genotype: genotypes 1 and 2 have been isolated exclusively from humans thus far, whereas genotypes 3 and 4 have been recovered from not only humans but also several species of animals including pigs, wild boars, a wild deer, and a wild mongoose (Meng et al., 1997, Okamoto et al., 2001, Meng, 2003, Tei et al., 2003, Sonoda et al., 2004, Nishizawa et al., 2005, Nakamura et al., 2006). In some geographic areas where HEV of genotype 1 or 2 is prevalent in humans, HEV of genotype 3 or 4 has been identified in pigs (Arankalle et al., 2002, Cooper et al., 2005).

It was reported that the inter-genotype diversity over the entire genome of HEV is 23.6–27.7% (Zhai et al., 2006), and the intra-genotype diversities of genotypes 1, 3, and 4 are as high as 11.8, 19.3, and 17.0%, respectively (Inoue et al., 2006b). The presence of markedly heterogeneous sequences makes it difficult to design primers suitable for sensitive detection of all four genotypes of HEV by RT-PCR. At present, a total of 70 entire or nearly entire genomic sequences of HEV isolates are available from the GenBank/EMBL/DDBJ databases. When the sequences of previously reported universal primers for the detection of HEV RNA (Erker et al., 1999, Huang et al., 2002, Mizuo et al., 2002, Takahashi et al., 2003a, Jothikumar et al., 2006) were compared with the 70 HEV sequences, a set of primers and a probe designed within the ORF2/ORF3 overlapping region of the HEV genome for real-time RT-PCR (Jothikumar et al., 2006) was the best in view of being positioned in a highly conserved area. However, the amplified product of the real-time RT-PCR is too short for determining the genotype. In the present study, based on the analysis of 70 entire or nearly entire HEV sequences, nested universal primers were designed within the ORF2/ORF3 overlapping region for PCR that was capable of amplifying HEV sequences of all four genotypes and in which the amplified products could be used for HEV genotyping. The validity of this method for sensitively detecting HEV RNA was compared with that of the previous RT-PCR method that amplifies a 457 nt sequence within ORF2 (Mizuo et al., 2002).