Novel Synthesis and Phenotypic Analysis of Mutant Clouds for Hepatitis E Virus Genotype 1

Abstract

Many RNA viruses exist as an ensemble of genetically diverse, replicating populations known as a mutant cloud. The genetic diversity (cloud size) and composition of this mutant cloud may influence several important phenotypic features of the virus, including its replication capacity. We applied a straightforward, bacterium-free approach using error-prone PCR coupled with reverse genetics to generate infectious mutant RNA clouds with various levels of genetic diversity from a genotype 1 strain of hepatitis E virus (HEV). Cloning and sequencing of a genomic fragment encompassing 70% of open reading frame 1 (ORF1) or of the full genome from variants in the resultant clouds showed the occurrence of nucleotide mutations at a frequency on the order of 10^-3 per nucleotide copied and the existence of marked genetic diversity, with a high normalized Shannon entropy value. The mutant clouds showed transient replication in cell culture, while wild-type HEV did not. Cross-sectional data from these cell cultures supported the existence of differential effects of clouds of various sizes and compositions on phenotypic characteristics, such as the replication level of (+)-RNA progeny, the amounts of double-stranded RNA (a surrogate for the rate of viral replication) and ORF1 protein, and the expression of interferon-stimulated genes. Since mutant cloud size and composition influenced the viral phenotypic properties, a better understanding of this relationship may help to provide further insights into virus evolution and prediction of emerging viral diseases.IMPORTANCE Several biological or practical limitations currently prevent the study of phenotypic behavior of a mutant cloud in vitro We developed a simple and rapid method for synthesizing mutant clouds of hepatitis E virus (HEV), a single-stranded (+)-RNA [ss(+) RNA] virus, with various and controllable levels of genetic diversity, which could then be used in a cell culture system to study the effects of cloud size and composition on viral phenotype. In a cross-sectional analysis, we demonstrated that a particular mutant cloud which had an extremely high genetic diversity had a replication rate exceeding that of wild-type HEV. This method should thus provide a useful model for understanding the phenotypic behavior of ss(+) RNA viruses.

Keywords: error-prone PCR; genetic diversity; hepatitis E virus; mutant cloud; phenotype; quasispecies; random mutagenesis; reverse genetics.

FIG 1

Agarose gel electrophoresis profiles of Ep-PCR products and genetic diversity of HEV amplicons. (A) Agarose gel analysis of Ep-PCR products. The HEV genome was amplified from the pSK-E2 plasmid by use of either balanced dNTP concentrations (lane 1) or combinations of dinucleotides at unbalanced concentrations (lanes 2 to 17), as indicated. Detailed Ep-PCR conditions are given in Materials and Methods. The expected size of amplicons was 7.2 kb. The products were separated in a 0.8% agarose gel and visualized by staining with ethidium bromide. Lane M, 1,000-bp DNA size ladder (sizes given in kilobase pairs). (B) Proportions of mutations per 1,000 bp in full-length amplicons arising from Ep-PCRs as determined by DNA sequencing. Statistically significant differences are indicated by asterisks (**, P < 0.001; *, P < 0.05 [Fisher's exact test]). (C) Phylogenetic analysis of full-length HEV genomes. The neighbor-joining tree shows reference full-length HEV genome sequences for genotypes 1 to 4 and full-length genome sequences from this study. Support values for branches supported by at least 60% of bootstrap replicates are shown next to the nodes. Full-length genome sequences (gt1 to gt4) and the wild-type Sar55 sequence (accession no. AF444002) are indicated in a red box. The scale bar represents the genetic distance.

FIG 2

Cloud^v-transfected cells have enhanced generation of extracellular and intracellular viral RNAs. Accumulated viral RNA was measured by using quantitative RT-PCR at 3-day intervals for HEV-transfected or mock-transfected cells. HEV RNA levels (log₁₀ RNA copies per milliliter of culture supernatant [A] or per microgram of intracellular RNA [B]) are shown. Results are representative of three independent experiments, each performed in triplicate, and are means and standard deviations (SD). Statistically significant differences are indicated by asterisks (**, P < 0.001; Student t test). n.s., not significant; u.d., undetectable; dpt, days posttransfection.

FIG 3

Cloud^v-transfected cells have enhanced amounts of ORF1 protein and dsRNA. S10-3 cells at day 6 posttransfection were fixed and stained with mouse polyclonal antibodies to ORF1 (A) or with a rabbit MAb to dsRNA (B), followed by secondary staining with a goat anti-mouse or goat anti-rabbit antibody conjugated to Alexa Fluor 488 or 568 and with DAPI. Bar, 50 μm (applies to all panels). Results are representative of three independent experiments. Densitometry quantitation results for ORF1 (C) and dsRNA (D) are expressed as mean pixel densities. Cloud^v significantly increased ORF1 protein and dsRNA levels. Data are presented as means and SD for multiple areas. Statistically significant differences are indicated by asterisks (**, P < 0.001; Student t test [95% confidence interval]). n.s., not significant; u.d., undetectable.

FIG 4

Viruses produced by Cloud^v-transfected cells were infectious. Naive S10-3 cells were infected with the viruses produced by Cloud^v-, Cloud^l-, and mock-transfected cells at day 6 p.t. Cells at day 5 postinfection were fixed and stained with mouse polyclonal antibodies to ORF1, followed by secondary staining with a mouse anti-mouse or goat anti-rabbit antibody conjugated to Alexa Fluor 568 and with DAPI. Bars, 50 μm (applies to all panels). Results are representative of three independent experiments.

FIG 5

Cloud^v-transfected cells were associated with higher ISG induction levels. Induced ISG mRNAs were measured by using quantitative RT-PCR on day 5 p.t. Fold changes for OAS1 (A), viperin (B), RIG-I (C), and IFITM3 (D) mRNAs are shown. Results are representative of three independent experiments, each performed in triplicate, and are means and SD. Statistically significant differences are indicated by asterisks (**, P < 0.001; *, P < 0.05; Student t test [95% confidence interval]). n.s., not significant.