Sporadic hepatitis E infection in southern China

Abstract

China is a hepatitis E endemic region. For many years, the Burmese-like hepatitis E virus (HEV) strain was considered to be responsible for hepatitis E in China. However, in 1993, a novel sporadic divergent HEV isolate (G-9 and G-20) was found in two patients with hepatitis E in Guangzhou, southern China. Subsequently, the Guangzhou divergent (GZD) isolate was also found in other patients in Guangzhou and even in Taiwan. The Taiwan patients had a history of traveling to China before onset of the disease. These findings indicate that the GZD isolate is responsible for some of these sporadic HEV infections. The discovery demonstrates the presence of significantly divergent variant and has led to the identification of a number of additional isolates from many other regions. To investigate more extensively the GZD isolate and its role in sporadic HEV infection in southern China, a relevant fecal strain (93G strain) of the G-9 isolate was isolated by A549 cell culture and identified immunologically and molecular biologically in this study. In the isolation and identification of the 93G strain in cell culture, cytopathic effects (CPE), morphology and morphogenesis of the 93G strain were observed by light microscope (LM), electron microscopy (EM) and immune electron microscopy (IEM) respectively. The results showed that 93G virus could be propagated in the A549 cells. The infected cells became round and lysed gradually until all the cells in the monolayer were destroyed. Through EM observation, progressive development of local vesicles, virions accumulation in crystalline arrays and viroplasmic focus were seen in cytoplasm of infected cells. Replication and assembly of the newly generated viruses were closely associated with CPE. The viral particles were about 29~32 nm in diameter and can be aggregated by a specific antibody to HEV Chinese Xinjiang strain (87A) as observed by IEM or neutralized by serum from the patient with hepatitis E and were similar morphologically to HEV from other sources. 93G viral RNA was amplified by reverse transcriptase and polymerase chain reaction (RT-PCR) and showed high homology (93%) with that of the G-9 isolate but was also as divergent from the Burmese-like isolates, Mexican isolate and US isolates, as was G-9. The result suggests the A549 cell line is a reliable cell culture system for HEV replication. The establishment of this reliable system is very important for further study of the virus. An indirect fluorescent antibody assay was developed to examine HEV antigen in the infected cells, in the cytoplasm and on the cell membrane. The 58 kDa and 82 kDa native structural proteins of the 93G virus were identified by Western blotting. To compare immunoreactivity to the Burmese-like strain and to the 93G strain, an Enzyme-liked immunosorbent assay (ELISA) was developed based on the 58 kDa native structural protein to detect IgM-class antibody to 93G (93G-IgM) in 62 patients with acute nonA to nonC hepatitis. The result was compared with their anti-HEV status established by commercially available ELISA test which is based on recombinant fusion proteins expressed from antigenic portions of the putative structural proteins of both HEV Burma and Mexico prototypic strains. The patients with the Burmese-like isolate infection or with Guangzhou-like isolate infection had a similar immunoreactivity to HEV. The accordant rate of 93G-IgM was 82% with HEV-IgM and 48% with HEV-IgG. The correlation of the three antibodies to HEV RNA was in decreasing order, HEV-IgM (97%), 93G-IgM (90%) and HEV-IgG (71%). These results indicate that the patients with the GZD infection were not altered immunoreactivity to the Burmese stain and proven that the commercially available immunoassay is adequate for diagnosis of HEV infection in southern China. RT-nPCR is a useful technique for amplification of small amounts of HEV RNA and was used for confirmation of HEV infection in this study. It was found the Burmese original primers could not amplify the HEV gene in two patients with GZD infection. In order to avoid false-negative result and to rapidly screen HEV gene in the patients with the Burmese-like isolate infection or with Guangzhou-like isolate infection, a HEV probe was prepared based on the nucleotide sequence of the GZD. Both the Burmese-like isolate and the GZD isolate could be hybridized with the HEV probe prepared. The dot-hybridization test could detect majority of the patients with HEV infection although it was not as sensitive as RT-nPCR. Further, the latter was used to confirm the HEV infection in this study. However, two samples were found to have inconsistent results of HEV RNA as detected with RT-nPCR, even when two different sets of HEV-specific primers or two different RT-nPCR systems were used. Analysis with the computer software DNASIS showed that a RNA secondary structure might be present, leading to misalignment of the HEV sequences and thus unexpected amplicon. The conventional reverse transcription condition was subsequently modified to eliminate RNA secondary structure and an expected amplicon was subsequently obtained. Our study may provide unprecedented evidence that spontaneous secondary structure formed in RT-PCR reaction can influence the outcome of RNA gene amplification and should be considered an important factor when designing primers and adopting protocols for RNA gene amplification. The commercially available ELISA test was used to evaluate the anti-HEV status in 153 patients with acute hepatitis, admitted to the municipal infectious disease hospital of Guangzhou from 1994 to 1998 and 481 non-hepatitis individuals from Guangzhou and Hong Kong from 1999 to 2000. Anti-HEV seroreactivity was found unexpected high (up to 42%) amongst the non-hepatitis population, particularly in elderly patients and those with chronic disease. Using 56 identified patients with hepatitis E as controls, it was found that the low-level anti-HEV and single HEV-IgG were responsible for such observation. The result indicates unspecific seroreactions to HEV existing in the test. Thus, while the commercial test is used for diagnosis of HEV infection, the result should be excluded unspecific reactivity, particularly in non-hepatitis population who are elderly or with chronic disease. Phylogenesis of 45 HEV isolates found during 1994 to 2000 was analyzed to investigate molecular epidemiology and evolution of the HEV isolates in southern China. The result of the phylogenetic analysis showed that there were two genotypes of HEV present: 43 of them belonged to one sub-genotype of the Burmese-like isolates and 2 were GZD isolates like 93G strain. The GZD isolates were only found in 1994. They along with the 93G strain formed a unique genotype. The GZD isolates more closed to the Burmese strain BUR2 and Chinese Xinjiang strain CH4, and relatively closed to CTI than other groups when compared with the reported five HEV genotypes. After 1994, only the Burmese-like isolates were found to be causative agents for sporadic hepatitis E in southern China. No major shift from one genotype to another was noted since then. Isolation, characterization and phylogenetic analysis of the HEV sporadic isolates in southern China have shown the existent of divergent variants. Adopting adequate methodologies may help to provide rapid and accurate diagnosis for sporadic HEV infection.