Analysis of Powassan Virus Genome Sequences from Human Cases Reveals Substantial Genetic Diversity with Implications for Molecular Assay Development

Abstract

Powassan virus (POWV) is an emerging tick-borne virus that causes severe meningoencephalitis in the United States, Canada, and Russia. Serology is generally the preferred diagnostic modality, but PCR on cerebrospinal fluid, blood, or urine has an important role, particularly in immunocompromised patients who are unable to mount a serologic response. Although the perceived poor sensitivity of PCR in the general population may be due to the biology of infection and health-seeking behavior (with short viremic periods that end before hospital presentation), limitations in assay design may also contribute. Genome sequences from clinical POWV cases are extremely scarce; PCR assay design has been informed by those available, but the numbers are limited. Larger numbers of genome sequences from tick-derived POWV are available, but it is not known if POWV genomes from human infections broadly mirror genomes from tick hosts, or if human infections are caused by a subset of more virulent strains. We obtained viral genomic data from 10 previously unpublished POWV human infections and showed that they broadly mirror the diversity of genome sequences seen in ticks, including all three major clades (lineage I, lineage II Northeast, and lineage II Midwest). These newly published clinical POWV genome sequences include the first confirmed lineage I infection in the United States, highlighting the relevance of all clades in human disease. An in silico analysis of published POWV PCR assays shows that many assays were optimized against a single clade and have mismatches that may affect their sensitivity when applied across clades. This analysis serves as a launching point for improved PCR design for clinical diagnostics and environmental surveillance.

Keywords: PCR; Powassan; deer tick virus; lineage I.

Figure 1

Phylogenetic reconstruction of Powassan virus genome sequences from ticks and humans, colored by location and source. Where node support is >50% by bootstrap approximation, nodes are annotated with blue colored circles shaded by degree of support. Outer rings provide information on the source (tick versus human) and specimen location.

Figure 2

Comparison of POWV genome sequences from humans and ticks by nucleotide and amino acid position by Fisher’s exact test. (a) Comparison of lineage I genomes by nucleotide position. (b) Comparison of lineage II genomes by nucleotide position. (c) Comparison of lineage I genomes by amino acid position. (d) Comparison of lineage II genomes by amino acid position. Horizontal dashed line represents threshold for nominal significance (p < 0.05); horizontal dotted line represents threshold for significance after correction for multiple comparisons.