A20 Sample preparation for whole-genome next-generation sequencing (NGS) of hepatitis C virus (HCV) routine RNA samples

Abstract

Next-generation sequencing (NGS) is a technique that can capture the variability of viral populations in transmission studies. The conventional sample preparation for NGS, based on amplicons, is a potential source of errors, derived from the variable affinity of specific primers for different viral variants and from irregular DNA polymerase efficiency. In this context, we propose a more reliable method for viral whole genome sample preparation, starting from nucleic acids obtained and stored with conventional procedures. Our goal was to obtain complete hepatitis C virus (HCV) genome sequences to subsequently perform extensive phylogenetic analyses. Additionally, we aimed to test the effectiveness of nuclease treatment used to remove contaminating host DNA. Nucleic acids were obtained from almost cell-free blood plasma of HCV-infected patients. As a source for Illumina library preparation, double-stranded cDNA was generated using random primers. The HCV genome was not amplified before library preparation, avoiding possible biases derived from unequal copying. To get rid of possible host contaminants in the samples, a DNase treatment step was added. Libraries were paired-end sequenced on the Illumina platform using MiSeq reagent kit v3. After conservative filtering of contaminant human reads by alignment with the human reference genome using Burrows-Wheeler Aligner (BWA), the remaining reads were mapped to the HCV reference genome using BWA. Primary maximum likelihood phylogenetic analyses were performed using ClustalW and IQTREE to infer the phylogenetic relationships of the sequenced samples in the context of complete genome sequences of the same genotype. NGS sample preparation method of HCV from blood plasma was established. Complete genome sequences of HCV could be obtained with variable coverage depending on the viral load of plasma samples. No significant reduction of host DNA proportion in DNase treated samples in comparison to the controls was observed. The new sequences clustered within the Los Alamos National Laboratory database-deposited HCV subtype 4d samples. The method can be used to obtain full-length sequences of HCV from nucleic acid samples not previously planned for NGS. No improvement was observed when DNase pre-treatment of nucleic acids extracted from blood plasma was performed.