Revised time scales of RNA virus evolution based on spatial information

Abstract

The time scales of pathogen evolution are of major concern in the context of public and veterinary health, epidemiology and evolutionary biology. Dating the emergence of a pathogen often relies on estimates of evolutionary rates derived from nucleotide sequence data. For many viruses, this has yielded estimates of evolutionary origins only a few hundred years in the past. Here we demonstrate through the incorporation of geographical information from virus sampling that evolutionary age estimates of two European hantaviruses are severely underestimated because of pervasive mutational saturation of nucleotide sequences. We detected very strong relationships between spatial distance and genetic divergence for both Puumala and Tula hantavirus-irrespective of whether nucleotide or derived amino acid sequences were analysed. Extrapolations from these relationships dated the emergence of these viruses most conservatively to at least 3700 and 2500 years ago, respectively. Our minimum estimates for the age of these hantaviruses are ten to a hundred times older than results from current non-spatial methods, and in much better accordance with the biogeography of these viruses and their respective hosts. Spatial information can thus provide valuable insights on the deeper time scales of pathogen evolution and improve our understanding of disease emergence.

Assessing Genome-Wide Diversity in European Hantaviruses through Sequence Capture from Natural Host Samples

“…We used samples from TULV-positive common voles and PUUV-positive bank voles from different locations in Central Europe, chosen to cover the high genetic variability in the region, together with published genome data ( Figure 1, Table 1, Table S1; see [24]). Total RNA was extracted from lung tissue used in previous studies with a modified QIAzol protocol as described in [27] and partial S-segment sequences were generated with Sanger sequencing [23,24,28]. RNA concentration was measured for each sample using the Qubit RNA BR Assay Kit (Invitrogen, Basel, Switzerland) and RNA quality was determined on a Fragment Analyzer CE12 (Advanced Analytics, Agilent, Santa Clara, CA, USA).…”

“…Shotgun sequence read data are from [26]. different locations in Central Europe, chosen to cover the high genetic variability in the region, together with published genome data ( Figure 1, Table 1, Table S1; see [24]). Total RNA was extracted from lung tissue used in previous studies with a modified QIAzol protocol as described in [27] and partial S-segment sequences were generated with Sanger sequencing [23,24,28].…”

“…Phylogenetic trees were inferred for nucleic acid sequences in MrBayes v.3.2.6 [40] on CIPRES [41] using mixed models. Individual nucleotide substitution rate priors in MrBayes were used (see [24]). Nucleotide sequence data were partitioned into 1st + 2nd and 3rd codon position with the evolutionary rate unlinked across partitions.…”

“…Due to the error-prone replication by the RNA polymerase, the mutation rates in hantaviruses are typically very high [4,22,23]. At the regional and continental geographic scale, this has been shown to lead to very strong mutational saturation in hantavirus sequence datasets resulting in too young age estimates and potentially biased phylogenetic analyses [24]. Full genome sequences are preferable to partial segments to most accurately determine evolutionary relationships and to assess the importance of recombination, reassortment and adaptive processes in the history of hantaviruses [23][24][25], but, to date, efficient protocols for genomic sequencing are not available.…”

Common vole (Microtus arvalis) and bank vole (Myodes glareolus) derived permanent cell lines differ in their susceptibility and replication kinetics of animal and zoonotic viruses

“…Large-scale sequence analyses of TULV strains indicated wellseparated genetic clades with defined geographical distribution, e.g. Moravia strain belongs to the Eastern South (EST.S) clade in the Eastern evolutionary lineage of the common vole (8,9). The zoonotic potential of TULV is controversially discussed (10).…”