Complete Genome Sequences of Human Astrovirus Prototype Strains (Types 1 to 8)

We report the complete genome sequences of the eight human astrovirus Oxford prototype strains. These sequences share 94.9% to 99.9% nucleotide identity with open reading frame 2 (ORF2) genes of astrovirus genomes previously deposited in GenBank and include the first complete genome of human astrovirus type 7.

ABSTRACT

We report the complete genome sequences of the eight human astrovirus Oxford prototype strains. These sequences share 94.9% to 99.9% nucleotide identity with open reading frame 2 (ORF2) genes of astrovirus genomes previously deposited in GenBank and include the first complete genome of human astrovirus type 7.

ANNOUNCEMENT

Astroviruses belong to the family Astroviridae, which is composed of two genera, Mamastrovirus and Avastrovirus, which include viruses that infect mammals and birds, respectively. They are small (≈28 to 35 nm) star-shaped nonenveloped icosahedral viruses with a nonsegmented single-stranded positive-sense RNA genome (1). The genome is 6,700 to 7,000 nucleotides long and contains three open reading frames (ORFs). ORF1a and ORF1b encode the nonstructural protease and RNA-dependent RNA polymerase proteins, respectively, while ORF2 encodes the viral structural protein (1), which is commonly used for genotyping of astroviruses.

Clinical symptoms of human astrovirus (HAstV) infections include headache, diarrhea, vomiting, and abdominal pain. While infections usually resolve without specific treatment, severe dehydration leading to hospitalization has been described (2–5). HAstVs were first reported in 1975 (2) after electron microscopy (EM) analysis of stool samples from hospitalized infants with acute gastroenteritis. HAstV serotypes were initially differentiated by immune EM (6, 7), and in the early 1990s, based on sequences of the capsid gene, they were classified in the family Astroviridae (8). To date, eight serotypes have been described (6, 7, 9). HAstV accounts for 2% to 9% of all sporadic cases of acute gastroenteritis in children, and HAstV-1 has been reported as the most common strain globally (1, 10–12). Coinfection with other viruses, such as rotavirus, norovirus, or sapovirus, has been reported frequently (13).

Fewer than 30 complete HAstV genome sequences are available in GenBank, including types 1 to 6 and 8. We report the complete genomes of the original HAstV Oxford reference strains, types 1 to 8.

The first HAstVs that were successfully cultured in the laboratory of John Kurtz (Oxford, United Kingdom) have since been referred to as the Oxford reference strains. The eight isolates were originally obtained from John Kurtz and John Herrmann (University of Massachusetts Medical School, Worchester, MA) and were cultured in LLCMK2D cells as described previously (14). Astrovirus RNA was extracted using the QIAamp viral RNA minikit followed by on-column DNase digestion (Qiagen). Sequence-independent, single-primer amplification (15, 16) was used to produce amplicons that were processed using the Nextera XT library preparation kit. The resulting libraries were sequenced on an Illumina MiSeq 500-cycle paired-end run.

A custom in-house bioinformatics pipeline (17) was used to process raw FASTQ data and to de novo assemble each isolate. Within the pipeline, the preprocessing steps for the FASTQ raw reads were as follows: host removal using Bowtie 2 v2.3.3.1 (18–20) followed by primer and adapter trimming and Phred quality score filtering (removing those with a score of <20) using Cutadapt v1.8.3 (21) and, finally, removal of duplicate reads using the Python script Dedup.py (22). The remaining FASTQ reads were assembled into contigs using the de novo assembler SPAdes v3.7.0 (23) and contigs were then classified into taxonomic groups using NCBI BLAST+ v2.6.0 (24). Consensus genome sequences were verified through read mapping and annotated using Geneious vR11. Complete genome sequences with an average read coverage per genome ranging from 600 × to 2,300 × were generated. The total postprocessing FASTQ reads per sample ranged from 33,000 to 112,000, with the percentage of target viral reads between 88.7 and 96.9%.

Compared to the closest corresponding complete HAstV genomes available from GenBank, pairwise nucleotide identities from the Oxford strains ranged from 99.8% for HAstV-1 (GenBank accession number L23513), 99.6% for HAstV-2 (GenBank accession number L13745), 97.6% for HAstV-3 (GenBank accession number AF141381), 94.6% for HAstV-4 (GenBank accession number AY720891), 93.7% for HAstV-5 (GenBank accession number JQ403108), 95.2% for HAstV-6 (GenBank accession number HM237363), and 97.3% for HAstV-8 (GenBank accession number AF260508). The ORF2 of HAstV-7 had a 99.9% pairwise nucleotide identity with a partial ORF2 HAstV-7 sequence available from GenBank (accession number Y08632).

Data availability.

The HAstV Oxford reference genome sequences (types 1 to 8) have been deposited in GenBank with the accession numbers MK059949 to MK059956. The postprocessed FASTQ reads have been deposited in the Sequence Read Archive with the run accession numbers SRR8444451 to SRR8444458.