High-Resolution Evolutionary Analysis of Within-Host Hepatitis C Virus Infection.

Jayna Raghwani,Richard Molenkamp,Janke Schinkel,Chieh-Hsi Wu,Cynthia K Y Ho,Oliver G Pybus,Katrina A Lythgoe,Menno De Jong

doi:10.1093/infdis/jiy747

Jayna Raghwani, Richard Molenkamp + Show 6 more

Open Access

https://doi.org/10.1093/infdis/jiy747

Copy DOI

Abstract

BackgroundDespite recent breakthroughs in treatment of hepatitis C virus (HCV) infection, we have limited understanding of how virus diversity generated within individuals impacts the evolution and spread of HCV variants at the population scale. Addressing this gap is important for identifying the main sources of disease transmission and evaluating the risk of drug-resistance mutations emerging and disseminating in a population.MethodsWe have undertaken a high-resolution analysis of HCV within-host evolution from 4 individuals coinfected with human immunodeficiency virus 1 (HIV-1). We used long-read, deep-sequenced data of full-length HCV envelope glycoprotein, longitudinally sampled from acute to chronic HCV infection to investigate the underlying viral population and evolutionary dynamics.ResultsWe found statistical support for population structure maintaining the within-host HCV genetic diversity in 3 out of 4 individuals. We also report the first population genetic estimate of the within-host recombination rate for HCV (0.28 × 10−7 recombination/site/year), which is considerably lower than that estimated for HIV-1 and the overall nucleotide substitution rate estimated during HCV infection.ConclusionsOur findings indicate that population structure and strong genetic linkage shapes within-host HCV evolutionary dynamics. These results will guide the future investigation of potential HCV drug resistance adaptation during infection, and at the population scale.

Highlights

Despite recent breakthroughs in treatment of hepatitis C virus (HCV) infection, we have limited understanding of how virus diversity generated within individuals impacts the evolution and spread of HCV variants at the population scale
A, Dashed lines correspond to locally estimated scatterplot smoothing (LOESS) regression fit through the lower and upper 95% highest posterior density estimates of Neτ, while the solid lines correspond to the smooth curve fitting to the mean Neτ over time
This pattern was weaker in individual p37, and absent in individual p53, suggesting either not all viral lineages were sampled in the blood plasma or the degree of HCV diversification varied among individuals

Summary

Methods

We have undertaken a high-resolution analysis of HCV within-host evolution from 4 individuals coinfected with human immunodeficiency virus 1 (HIV-1). We used long-read, deep-sequenced data of full-length HCV envelope glycoprotein, longitudinally sampled from acute to chronic HCV infection to investigate the underlying viral population and evolutionary dynamics. We analyzed full-length E1E2 (1680 bp) sequences from 4 individuals that were coinfected with HIV-1 and HCV. Time-scaled phylogenies and demographic histories were estimated in BEAST v.1.8 [18] using a relaxed uncorrelated log-normal distributed molecular clock [19], a codon-structured nucleotide substitution model [20], and a Bayesian skygrid coalescent prior [21]. Bayesian phylogenetic inference methods are computationally prohibitive for large sequence datasets (>1000 sequences). Phylogenies were plotted with ggtree [22], and trends in within-host HCV population dynamics were inferred with ggplot2 [23] by fitting a loess regression through the Neτ values over time (estimated from the Bayesian skygrid model, where Ne is effective population size and τ is viral generation time)

Results

Discussion

Conclusion