Efficient and unbiased metagenomic recovery of RNA virus genomes from human plasma samples

Carmen F Manso,Jean L Mbisa,David F Bibby

doi:10.1038/s41598-017-02239-5

Abstract

RNA viruses cause significant human pathology and are responsible for the majority of emerging zoonoses. Mainstream diagnostic assays are challenged by their intrinsic diversity, leading to false negatives and incomplete characterisation. New sequencing techniques are expanding our ability to agnostically interrogate nucleic acids within diverse sample types, but in the clinical setting are limited by overwhelming host material and ultra-low target frequency. Through selective host RNA depletion and compensatory protocol adjustments for ultra-low RNA inputs, we are able to detect three major blood-borne RNA viruses – HIV, HCV and HEV. We recovered complete genomes and up to 43% of the genome from samples with viral loads of 104 and 103 IU/ml respectively. Additionally, we demonstrated the utility of this method in detecting and characterising members of diverse RNA virus families within a human plasma background, some present at very low levels. By applying this method to a patient sample series, we have simultaneously determined the full genome of both a novel subtype of HCV genotype 6, and a co-infecting human pegivirus. This method builds upon earlier RNA metagenomic techniques and can play an important role in the surveillance and diagnostics of blood-borne viruses.

Highlights

Single-stranded RNA viruses exhibit exceptional genetic diversity due to low fidelity replication mechanisms[1, 2]
In light of the large and ever-increasing number of human RNA virus pathogens, it is perhaps unsurprising that standard serological assays and nucleic acid tests suffer from a lack of sensitivity to diverse variants of target viruses, overlook the presence of new or unexpected viruses, and provide only limited information about those targets they do successfully detect
Most existing protocols achieve a maximum of two of these aims, but difficulties in selectively isolating viral RNA species and short read sequences from those of the super-abundant host nucleic acid have limited the utility of metagenomic approaches in diagnostic virology

Summary

Introduction

Single-stranded RNA viruses exhibit exceptional genetic diversity due to low fidelity replication mechanisms[1, 2]. As a group, they constitute the major source of emerging infections in humans such as Ebola, chikungunya, Zika, West Nile virus and Middle East respiratory syndrome-related coronavirus[3–7]. With new human RNA viruses being discovered each year[8], the serological and nucleic acid amplification techniques that have dominated virus diagnostics for many years are becoming increasingly unable to respond to the ever-expanding range of pathogens. Blood and transplant products are routinely screened for these and other viruses Even within these two relatively well-characterised virus species, new genotypic divisions and recombinant variants are being discovered[11–13], and questions have been raised regarding the reliability of commercial assays in detecting new strains[14–16]. In the absence of significant host depletion, genome coverage and depths are often low even at high viral copy numbers[20–23]

Methods

Results

Conclusion