Genetic diversity and phylogeography of Seewis virus in the Eurasian common shrew in Finland and Hungary

Recent identification of a newfound hantavirus, designated Seewis virus (SWSV), in the Eurasian common shrew (Sorex araneus), captured in Switzerland, corroborates decades-old reports of hantaviral antigens in this shrew species from Russia. To ascertain the spatial or geographic variation of SWSV, archival liver tissues from 88 Eurasian common shrews, trapped in Finland in 1982 and in Hungary during 1997, 1999 and 2000, were analyzed for hantavirus RNAs by reverse transcription-polymerase chain reaction. SWSV RNAs were detected in 12 of 22 (54.5%) and 13 of 66 (19.7%) Eurasian common shrews from Finland and Hungary, respectively. Phylogenetic analyses of S- and L-segment sequences of SWSV strains, using maximum likelihood and Bayesian methods, revealed geographic-specific genetic variation, similar to the phylogeography of rodent-borne hantaviruses, suggesting long-standing hantavirus-host co-evolutionary adaptation.

Previous analysis of the full-length S and partial M and L segments of a newfound hantavirus, designated Seewis virus (SWSV), detected in the Eurasian common shrew (Sorex araneus), captured in the Swiss canton of Graubünden [13], corroborates earlier reports of hantavi-ral antigens in this shrew species from Russia, Belgium and the former Yugoslavia [14][15][16]. As its name implies, the Eurasian common shrew (Subfamily Soricinae) is among the most widely dispersed small mammal species in Eurasia. Its vast geographic range, which extends throughout Northern Europe, including Scandinavia and Great Britain (but excluding Ireland), and across Russia (Fig. 1), provided an opportunity to investigate the genetic diversity and phylogeography of SWSV.
An exception was the partial S-segment sequence of SWSV strain DGR18890 from Oulun Lääni, which was highly incongruent, showing marked divergence of nearly 20% at the nucleotide and amino acid levels ( Table 2). Analysis, using multiple recombination-detection methods, including GENECONV, Bootscan, Chimaera, 3SEQ, RDP, SiScan, MaxChi and HyPhy Single Recombinant Breakpoint [17], failed to disclose any evidence of recombination. However, analyses of full-length genomic sequences of SWSV strains would be required to demonstrate intralineage recombination events. Apart from the above-mentioned incongruity, the inability to amplify the S segment in six of the 25 L-segment RT-PCR positive tissues, despite repeated attempts using numerous primers, may be the result of low viral titers or inadequate sensitivity of the PCR primers. Intensive efforts are ongoing to resolve this important issue.
Because shrews are inherently difficult to identify by morphological features alone, host verification of SWSVinfected shrews was confirmed by analyzing voucher specimens and sequencing the entire 1,140-base pair cyto-chrome b gene of mitochondrial DNA (mtDNA), amplified by PCR, using previously described universal primers (5'-CGAAGCTTGATATGAAAAACCATCGTTG-3' and 5'-GCAGCCCCTCAGAATGATATTTGTCCAC-3'). mtDNA sequences were deposited into GenBank (GQ374412-GQ374437), and the identities of the 25 hantavirus-infected hosts were assessed using a Bayesian approach (5 million generation with burn-in of 5000 discarded) that was mid-point rooted (tree not shown). All SWSV-infected shrews were confirmed as Sorex araneus. However, the Eurasian common shrew exhibits significant chromosomal polymorphism throughout its geographic range [31]. Previous studies suggest that several chromo-   Because the original report of SWSV was based on a single Eurasian common shrew from Switzerland [13], there has been understandable reluctance in fully accepting this hantavirus-soricid association. Data from the present study, however, provide compelling evidence that this soricine shrew species harbors SWSV across its broad geographic range. As further support, in a separate study, Sorex araneus, as well as the tundra shrew (Sorex tundrensis) and Siberian large-toothed shrew (Sorex daphaenodon), have been shown to harbor genetic variants of SWSV in six widely separated administrative regions of Western and Eastern Siberia [32]. Similarly, the American water shrew (Sorex palustris), Trowbridge's shrew (Sorex trowbridgii) and vagrant shrew (Sorex vagrans) in North America harbor genetic variants of Jemez Springs virus (H.J. Kang and R. Yanagihara, unpublished), which was originally found in the dusky shrew [3]. When viewed within this context, the demonstration of SWSV in Eurasian common shrews from Finland and Hungary lends support to the hypothesis that common ancestral hantaviruses established themselves in ancestors of present-day soricine shrew species, with subsequent cross-species transmission and local host-specific adaptation.
As noted, SWSV RNAs were found in Eurasian common shrews captured in Finland more than 25 years ago. Analysis of hantavirus sequences amplified from tissues of Eurasian common shrews and other soricine shrew species more recently trapped in these same sites in Finland would be extremely valuable, in providing insights into the evolutionary rate of SWSV. Such studies are now underway.
The emerging story of previously unrecognized hantaviruses in soricomorphs has been greatly accelerated by the availability of an extensive, meticulously curated, smallmammal frozen-tissue collection, housed at the Museum of Southwestern Biology. That is, while these tissues were not collected for the purposes of our current and past studies, their ready accessibility has facilitated the rapid acquisition of new knowledge about the spatial distribution of hantaviruses in nonrodent reservoir hosts [2,3,12]. As such, these opportunistic studies provide convincing justification and strong testament for the establishment and long-term maintenance of these repositories for future scientific inquiry. Additional hantaviruses and other zoonotic agents are likely to be successfully mined from such banked tissues, by employing powerful microarray and ultra high-throughput sequencing technologies.

Competing interests
The authors declare that they have no competing interests.
Phylogenetic tree generated by the Bayesian method, under the best-fit GTR+I+Γ model of evolution, based on the L-genomic segment of SWSV and other well-characterized hantaviruses