Historic and contemporary museum specimens implicate Northern Red-backed Vole ( Clethrionomys rutilus ) as borealpox host as early as 1990s

27 28 Borealpox virus (BRPV; formerly Alaskapox) is an orthopoxvirus that has caused seven reported 29 human infections in Alaska since 2015, including a fatal case in 2023. The natural reservoir of 30 BRPV is unknown, although previous investigations have raised the possibility of wild small 31 mammals transmitting the virus to humans, either through direct contact or via domestic cats and 32 dogs. To understand which species may be involved in the maintenance and/or spillover of 33 BRPV in Alaska, we trapped and sampled wild small mammals (including voles, shrews, and 34 squirrels) in 2021 and 2024 near reported human case locations in Fairbanks and the Kenai 35 Peninsula, respectively. We found evidence of previous exposure to orthopoxviruses in five 36 species (including the House Mouse, Mus musculus) and detected BRPV DNA as well as viable 37 virus in Northern Red-backed Voles (Clethrionomys rutilus). Further, screening of tissues from 38 historical museum specimens revealed BRPV DNA in C. rutilus specimens collected in Denali 39 National Park and Preserve in 1998 and 1999, 17 years before the first reported human case of 40 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint BRPV. Phylogenomic analysis of all human and animal BRPV isolates strongly supports the 41 hypothesis of local human infections through multiple spillover events. These findings suggest 42 C. rutilus as a possible reservoir species for BRPV and indicate that BRPV has been present in 43 Alaskan wild small-mammal populations for at least 25 years. Our study highlights the potential 44 of museum collections to elucidate the temporal, spatial, and host ranges of emerging pathogens. 45 Further museum- and field-based sampling will clarify the true geographic range of BRPV, 46 which is closely related to Old World orthopoxviruses and may be circulating beyond North 47 America. 48 49

50 51 The genus Orthopoxvirus (family Poxviridae) contains ten recognized species of double-stranded 52 DNA viruses that cause disease in a wide range of mammalian hosts, including humans (Bonwitt 53 et al. 2022). Notably, orthopoxviruses (OPXVs) include variola virus (VARV), which causes 54 smallpox, and monkeypox virus (MPXV), the causative agent of monkeypox, which has recently 55 caused widespread human outbreaks with sustained human-to-human transmission (Bonwitt et 56 al. 2022; WHO 2022). OPXVs may be emerging at higher rates due to increased host mixing and 57 human-wildlife contact driven by climate and land-use change (Thomassen et al. 2013), as well 58 as heightened human susceptibility following the cessation of smallpox vaccination after its 59 global eradication in 1980 (Diaz 2021). 60 61 The most recently described OPXV is borealpox virus (BRPV), formerly known as Alaskapox 62 virus (Gigante et al. 2019). BRPV was first detected after a patient presented with a lesion near 63 Fairbanks, Alaska, USA in July 2015 (Springer et al. 2017). Over the following eight years, five 64 additional, epidemiologically unrelated cases were reported in Fairbanks, all involving self-65 limiting illness (Mooring et al. 2020; Mooring et al. 2021; Mooring et al. 2025). However, in 66 September 2023, the first fatal case was reported in an elderly, immunocompromised patient on 67 the Kenai Peninsula in Southcentral Alaska (~320 miles straight-line distance SSE of Fairbanks), 68 expanding the known geographic range of this virus and demonstrating its potential for causing 69 fatal outcomes, particularly in severely immunocompromised individuals (Rogers et al. 2025). 70 Furthermore, the viral sequence recovered from this patient was distinct from sequences isolated 71 from prior cases in the Fairbanks area (Rogers et al. 2025), suggesting genetic variation across 72 space as is the case in OPXVs with wide host ranges such as cowpox virus (CPXV) and zoonotic 73 MPXV (Reynolds et al. 2018). Unraveling the geographic origins and range, host species, and 74 transmission pathways of BRPV is important for understanding and mitigating this viral 75 pathogen. 76 77 In the initial 2015 investigation of the first BRPV case, inadvertent viral importation from 78 outside of Alaska could not be excluded (Springer et al. 2017). However, the subsequent cases in 79 the Fairbanks area, the patients’ lack of out-of-state travel, and the greater genetic diversity seen 80 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint in the Kenai isolate strongly indicate that the virus is circulating in at least one non-human 81 reservoir in Alaska. This is particularly noteworthy given the phylogenetic relationship between 82 BRPV and other OPXVs. The genus Orthopoxvirus contains two reciprocally monophyletic 83 clades: “Old World” and “New World” OPXVs, which are thought to have diverged from a 84 common ancestor approximately 42,000 years ago (Babkin et al. 2022). Despite being detected 85 only in Alaska, BRPV isolates share a most recent common ancestor with Old World OPXVs 86 and constitute a reciprocally monophyletic sister clade with an estimated divergence date of 87 approximately 19,000 years ago (Babkin et al. 2022). The geographic range of BRPV remains 88 unknown, and it may be circulating beyond Alaska and North America. 89 90 The natural reservoir (sensu Viana et al. 2014) of BRPV is also unknown, but evidence from 91 cases, wildlife screening, and the genomic diversity within BRPV suggests small mammals play 92 a role (Mooring et al. 2025). All known patients lived in low population-density, forested areas 93 and reported contact with domestic animals (i.e., dogs and cats), many of which were known to 94 hunt small mammals such as rodents and shrews (i.e., pet dogs and cats that purportedly hunted 95 small mammals) (Springer et al. 2017; Mooring et al. 2020; Mooring et al. 2021; Mooring et al. 96 2025; Rogers et al. 2025). Furthermore, all reported cases occurred in the late summer to early 97 fall, when outdoor recreational and subsistence activities (e.g., hunting and foraging) increase 98 across Alaska; three out of the seven patients reported berry picking prior to symptom onset. 99 These activities increase opportunities for direct wildlife contact or indirect exposure via fomites, 100 both established pathways for OPXV transmission (Bonwitt et al. 2022). In 2015, samples from 101 small mammals collected near the first patient’s home were screened for OPXV DNA by PCR 102 (serological testing was not conducted); all samples were negative (Springer et al. 2017). During 103 a 2020 investigation following a reported case in the Fairbanks area, 176 small mammals were 104 captured and sampled. Anti-OPXV IgG antibodies were detected in serum samples from 28 of 105 147 Northern Red-backed Voles (Clethrionomys rutilus; 19.0%), one of four Northern Flying 106 Squirrels (Glaucomys sabrinus; 25.0%), and one of three American Red Squirrels (Tamiasciurus 107 hudsonicus; 33.3%) (Mooring et al. 2025). Viral DNA was amplified from tissue samples of 12 108 of 149 Northern Red-backed Voles (8.1%) and one of 14 Masked Shrews (Sorex cinereus; 109 7.1%). Among PCR-positive samples, viable BRPV was detected in one Northern Red-backed 110 Vole and one Masked Shrew. 111 112 The emergence of BRPV in North America raises pressing questions about its origins and actual 113 geographic distribution, which additional field-based sampling can help address. A 114 complementary approach involves the screening of historical specimens in natural history 115 museums. These collections are valuable but largely underutilized resources for pathogen 116 reservoir identification, with the potential to expand the known spatial, taxonomic, and temporal 117 distributions of viruses of concern (Yates et al. 2002; Colella et al. 2021). In this study, we aimed 118 to identify the reservoir host(s) for BRPV through both field- and museum-based screening. 119 First, we sampled small mammals in and around the properties of human cases and other non-120 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint epidemiologically linked locations near Fairbanks and Kenai, Alaska. Here, we sought to expand 121 the taxonomic and geographic scope of previous sampling efforts (Springer et al. 2017; Mooring 122 et al. 2025) by screening a larger sample of small mammals, following similar field studies of 123 MPXV and Akhmeta virus, another recently described OPXV (Doty et al. 2017; Doty et al. 124 2019). To broaden the temporal scope, we also screened tissue samples from historical 125 specimens housed at the University of Alaska Museum (UAM), targeting species based on 126 preliminary results from field-based sampling (Mooring et al. 2025). 127 128

129 130 Small mammals trapped in 2021 and 2024 131 132 In September 2021, 202 wild small mammals representing at least six species were trapped and 133 sampled in the Fairbanks and North Pole regions of Interior Alaska (Table 1). The majority were 134 Northern Red-backed Voles (n = 147), followed by shrews (Sorex spp.; 21), American Red 135 Squirrels (15), Northern Flying Squirrels (12), voles (Microtus sp.; 6), and a single Woodchuck 136 (Marmota monax). Additionally, a series of domestic/commensal rats (Rattus sp.; 7) obtained by 137 the Alaska Department of Fish and Game (and subsequently deposited in the UAM’s Mammal 138 Collection) from the Fairbanks area were also sampled. In August 2024, 135 wild small 139 mammals of at least six species were trapped and sampled in the Kenai Peninsula area of Alaska 140 (Table 1). Similarly, the majority were Northern Red-backed Voles (83), followed by shrews 141 (37), American Red Squirrels (11), Northern Bog Lemmings (Synaptomys borealis; 2), one 142 House Mouse (Mus musculus), and one Ermine (Mustela erminea). 143 144 In 2021, 20 Northern Red-backed Voles (13.6%) were IgG-positive for OPXV antibodies by 145 ELISA, as were five American Red Squirrels (33.3%) and seven Northern Flying Squirrels 146 (58.3%). However, OPXV DNA was only detected in seven Northern Red-backed Voles (4.8%). 147 In 2024, seven Northern Red-backed Voles (8.4%) were IgG-positive for OPXV antibodies by 148 ELISA, as were one shrew (2.7%), one American Red Squirrel (9.1%), and one House Mouse 149 (100%). Once again, the only PCR-positive animal was a Northern Red-backed Vole (1.2% of 83 150 C. rutilus). Viable virus was isolated from one PCR-positive Northern Red-backed Vole sample 151 collected in 2024. Sequencing was conducted on this viral isolate and one PCR-positive Northern 152 Red-backed Vole sample from 2021. None of the PCR-positive animals from either sampling 153 year had visible lesions suggestive of a poxvirus infection. 154 155 156 157 158 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint Table 1. Results of serological assays, PCR assays, and viral culture attempts on samples taken 159 from wild small mammals trapped in 2021 and 2024. A breakdown of results by trap site is 160 included in Supplementary Data S1. 161 162 163 Historical voucher specimen investigations 164 165 We screened a total of 285 frozen tissue samples from vole, shrew, and squirrel specimens 166 housed at UAM for OPXV DNA. Four of the 201 samples of Northern Red-backed Voles were 167 positive for OPXV DNA (Table 2). Two of these four PCR-positive samples were taken from the 168 same vole specimen (UAM 51567), and both yielded viable virus (Table 3). All three specimens 169 (UAM 51528, UAM 51567, UAM 73062) were collected at Rock Creek Site in Denali National 170 Common name Species Sample size ELISA positives (%) PCR positives (%) Viable virus (% Fairbanks and North Pole, Alaska; September 2021 Northern Red-backed Vole Clethrionomys rutilus 147 20 (13.6) 7 (4.8) 0 Shrew Sorex spp. 21 0 0 0 American Red Squirrel Tamiasciurus hudsonicus 15 5 (33.3) 0 0 Northern Flying Squirrel Glaucomys sabrinus 12 7 (58.3) 0 0 Rat Rattus sp. 7 0 0 0 Vole Microtus sp. 6 0 0 0 Woodchuck Marmota monax 1 0 0 0 Totals 209 32 (15.3) 7 (3.3) 0 Kenai Peninsula, Alaska; August 2024 Northern Red-backed Vole Clethrionomys rutilus 83 7 (8.4) 1 (1.2) 1 (1.2) Shrew Sorex spp. 37 1 (2.7) 0 0 American Red Squirrel Tamiasciurus hudsonicus 11 1 (9.1) 0 0 Northern Bog Lemming Synaptomys borealis 2 0 0 0 House Mouse Mus musculus 1 1 (100.0) 0 0 Ermine Mustela erminea 1 0 0 0 Totals 135 10 (7.4) 1 (0.7) 1 (0.7) 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint Park and Preserve, Alaska, in July 1998, August 1998, and August 1999, respectively (Fig. 1). A 171 frozen liver sample was positive from each specimen, as well as a frozen mixed-tissue sample 172 from UAM 51567. Frozen spleen and mixed-tissue samples from UAM 73062 were also 173 screened, and neither was positive for OPXV DNA. 174 175 Table 2. Results of PCR and viral culture attempts on samples from museum voucher specimens. 176 Collection date, locality, and tissue type of each sample are included in Supplementary Data S1. 177 178 Common name Species Samples PCR positives (%) Viable virus (%) Northern Red-backed Vole Clethrionomys rutilus 201 4 (2.0) 2 (1.0) Tundra Vole Alexandromys oeconomus 22 0 0 Insular Vole Microtus abbreviatus 6 0 0 Vole Microtus sp. 1 0 0 Masked Shrew Sorex cinereus 43 0 0 Tundra Shrew Sorex tundrensis 8 0 0 Montane Shrew Sorex monticolus 3 0 0 American Red Squirrel Tamiasciurus hudsonicus 1 0 0 Totals 285 4 (1.4) 2 (0.7) 179 180 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint Table 3. Tissue type, collection date, and locality of OPXV PCR-positive Northern Red-backed 181 Vole (Clethrionomys rutilus) museum voucher specimens. 182 183 184 185 Fig. 1. Localities of museum specimens sampled in our study. Diamonds indicate PCR-positive 186 specimens and smaller black dots indicate all other sampling sites. 187 188 189 Catalog number Tissue type Viable virus isolated? Locality Coordinates Collection date Sex UAM 51528 Liver No Rock Creek Site, Denali National Park, Alaska 63.729636, - 148.982167 21 July 1998 Female UAM 51567 Liver Yes Rock Creek Site, Denali National Park, Alaska 63.727261, - 148.980936 20 August 1998 Male Mixed tissue Yes UAM 73062 Liver No Rock Creek Site, Denali National Park, Alaska 63.730683, - 148.984308 31 August 1999 Male 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint Phylogenetic analysis of borealpox virus whole-genome sequences 190 191 A whole-genome maximum-likelihood tree revealed that isolates from Fairbanks human cases 192 (2015-2023) form a well-supported clade with small mammal specimens from Fairbanks C. 193 rutilus (2020, 2021) and S. cinereus (2020) sampled in this and an earlier study (Mooring et al. 194 2025). There was also strong support for a separate clade containing isolates from the 2023 195 Kenai human case and C. rutilus sampled from Kenai in 2024. The BRPV isolate from the C. 196 rutilus specimen collected in Denali National Park and Preserve in 1998 (UAM 51567) fell 197 within the Fairbanks clade, clustering with human cases from 2022 and 2023 (Fig. 2). Notably, 198 both Bayesian and maximum-likelihood analyses did not support monophyly of human-derived 199 sequences, suggesting multiple separate spillover events. 200 201 202 203 Fig. 2. Maximum-likelihood tree based on whole-genome sequences from BRPV isolates, 204 constructed using a GTR+F model in IQ-TREE (Nguyen et al. 2015). Blue labels indicate human 205 and vole isolates from the Kenai Peninsula, while the red label indicates the isolate from UAM 206 51567. The vole sequence from Fairbanks 2021 was sequenced directly from DNA extracted 207 from tissue. Bootstrap support values are shown at each node. 208 209

210 an 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint 211 Borealpox virus (BRPV) is a recently identified OPXV that has repeatedly spilled over into 212 human populations in Alaska since first reported in 2015. To better understand the zoonotic 213 reservoir of BRPV, we trapped and sampled wild small mammals in areas of known BRPV 214 spillover in Alaska—the Interior Region’s Fairbanks region in September 2021 and the Kenai 215 Peninsula in 2024—and screened tissue and blood samples for OPXV DNA and OPXV 216 antibodies, respectively. We also screened tissue samples associated with historical museum 217 specimens of suspected host species (collected 1991-2019 and housed at UAM) for OPXV DNA. 218 219 As in earlier field-based surveys (Mooring et al. 2025), most seropositive animals in our study 220 were Northern Red-backed Voles (n = 27), with 13.6% seropositivity and 8.4% seropositivity for 221 OPXV antibodies in Fairbanks (2021) and Kenai (2024), respectively. American Red Squirrels, 222 Northern Flying Squirrels, an unspecified shrew, and a House Mouse were also seropositive 223 (Table 1). The IgG ELISA used in this study is not BRPV-specific, so these results may reflect 224 past infection with other OPXVs, which could potentially circulate undetected in these small-225 mammal populations; however, no other OPXVs have been identified to date in Alaska. Other 226 OPXVs thought to circulate in North American small mammals have not been detected in 227 Alaska, including raccoonpox, volepox, and skunkpox viruses (Alexander et al. 1972; Regnery 228 1987; Emerson et al. 2009). Consequently, the detection of OPXV antibodies in a House Mouse 229 (M. musculus) is notable; should this reflect a prior BRPV infection, it suggests the potential for 230 mouse-human transmission risk in domestic settings, which may be higher than the risk of 231 transmission from other small mammals that are less likely to be found in human dwellings. The 232 distribution and prevalence of M. musculus in Alaska is very poorly understood, but museum 233 voucher specimens have been collected from or near Fairbanks, Fort Yukon, Anchorage, the 234 Kenai Peninsula, Kodiak Island, St. Paul Islands, Unalaska Island, Kiska Island, and multiple 235 localities in Southeast Alaska (MacDonald and Cook 2009). Predictive modeling based on viral 236 genomic features has implicated the genus Mus as potentially suitable borealpox hosts (Tseng et 237 al. 2025). Expanded serological and molecular testing of mice in and around the homes of BRPV 238 patients, and throughout Alaska, will be critical for investigating this hypothesis. 239 240 Only Northern Red-backed Voles tested positive for OPXV DNA (seven in 2021 and one in 241 2024), two of which were confirmed as BRPV by DNA sequencing (Figure 2.) Viable virus was 242 isolated from the positive sample collected in 2024. This supports prior evidence that the 243 Northern Red-backed Vole, which is also commonly found in and around human dwellings, is 244 involved in the maintenance and circulation of BRPV in wildlife populations in Alaska (Mooring 245 et al. 2025). To further investigate the reservoir of BRPV, future research should include 246 longitudinal study of potential host species, confirmation of long-term viral maintenance in these 247 populations, and expanded screening of species beyond Northern Red-backed Voles. Our study 248 is limited by small sample sizes of most species and a heavy bias toward C. rutilus based on the 249 availability of samples to screen and previous evidence of BRPV circulation in this species. 250 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint Other species—including squirrels, a mouse, and a shrew—were seropositive in our study, and 251 the screening of larger sample sizes of these species may reveal evidence of PCR positivity and 252 viable virus. A recent study suggests squirrels as possible reservoirs of MPXV, a related OPXV 253 (Riutord-Fe et al. 2026), and future work on BRPV should continue to investigate the possibility 254 of other small mammal species playing a role in its maintenance and transmission. 255 256 While future field studies of BRPV dynamics in wild mammals will be critical for elucidating 257 the reservoir species, looking for evidence of past BRPV circulation in Alaska can complement 258 contemporary data and further our understanding of the ecology of this virus. To do this, we 259 screened 285 tissue samples from historical museum specimens of voles, shrews, and squirrels 260 for OPXV DNA, and found four positive samples—all from Northern Red-backed Voles (Table 261 2). DNA sequencing confirmed the presence of BRPV DNA in one of these samples (Figure 2.). 262 Positive samples were collected from Denali National Park and Preserve in 1998 and 1999, over 263 15 years before the first reported case of human BRPV infection (Table 3). Denali National Park 264 and Preserve is roughly equidistant from the known spillover locations of BRPV (Fairbanks and 265 Kenai Peninsula) (Fig. 1). The detection of BRPV DNA in samples from the 1990s and in a 266 different site from known cases confirms the wider circulation of BRPV in Alaskan small-267 mammal populations over a longer period of time than previously known. Phylogenetically, the 268 isolate from this museum specimen fell within a clade of contemporary human and small-269 mammal isolates from the Fairbanks area (Fig. 2). Another separate, reciprocally monophyletic 270 clade contains isolates from the Kenai Peninsula human case in 2023 and C. rutilus sampled 271 from case patient’s residence in 2024, suggesting that the Kenai patient was likely infected 272 locally (Fig. 2). Our findings support the hypothesis of a long-term, established wildlife reservoir 273 with occasional spillover into human populations and potentially other mammal populations. 274 Initial investigations of human cases could not exclude the possibility of recent BRPV 275

into Alaska (Springer et al. 2017). However, our study suggests that BRPV has been 276 present in Alaska for at least 25 years (and likely much longer) with multiple separate spillover 277 events since 2015, though the drivers of these events remain unknown. It is possible that 278 previous spillover events went undetected due to mild, self-limiting illness or symptoms being 279 mistaken for other rashes (Mooring et al. 2020; Mooring et al. 2021). Alternatively, spillover 280 may have become more likely in the past decade due to the effects of climate and land-use 281 change on the distributions and cross-species interactions of small mammals (Baltensperger et al. 282 2024), as is the case for MPXV (Thomassen et al. 2013), and/or waning population-level 283 immunity against OPXVs following the cessation of smallpox vaccination (Diaz 2021). 284 285 The Northern Red-backed Vole is found in Alaska, Canada, northern Russia, and Scandinavia 286 (Linzey et al. 2020). If this species is the primary reservoir of BRPV, our findings suggest that 287 BRPV may be circulating beyond North America, in C. rutilus as well as in related species that 288 hybridize with C. rutilus in contact zones (Runck et al. 2009; Wiens and Colella 2025). A 289 broader awareness of BRPV symptoms and diagnostics will be critical to identify other spillover 290 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint events in Alaska and the broader arctic region. Further molecular and serological surveillance of 291 wild mammal populations in Alaska, as well as longitudinal studies of suspected host species 292 like Northern Red-backed Voles, will be vital to identifying the drivers of the apparently sudden 293 spillover of BRPV into human populations. 294 295 When federal public health agencies partner with non-governmental organizations such as 296 natural history museums and academic institutions, investigations are strengthened by leveraging 297 the resources, capabilities, and expertise of all partners. Our study illustrates the value of 298 museum specimens and their unique potential to elucidate the origins, reservoirs, and dynamics 299 of emerging viruses. As demonstrated here, disease surveillance in historical museum collections 300 can expand our understanding of the temporal, spatial, or host taxonomic range of newly 301 discovered pathogens (Juman et al. 2025; Cronin et al. 2025). This work can clarify whether a 302 virus was recently introduced to a given region or whether it has been endemic for decades and 303 present in additional nearby areas, as is the case for BRPV in Alaska. The continued collection 304 and preservation of voucher specimens will allow future researchers to retrospectively 305 understand pathogen emergence in the context of unprecedented global change. 306 307

308 Small-mammal trapping and specimen collection 309 Small mammals were trapped at nine sites around the Fairbanks and North Pole regions of 310 Alaska during September 6-14, 2021 (Fig. 1). Sites included homes of the two human cases 311 reported in 2021 and locations within seven miles of the case’s home where permission was 312 received from landowners. Other potential exposure locations such as homes of family members 313 were also included. All sites were in mixed evergreen and deciduous forests, either on public 314 land (n = 2) or low-density residential areas (n = 7). 315 Following the identification of the BRPV case on the Kenai Peninsula in late 2023, small 316 mammals were also trapped at three sites in the Kenai/Soldotna/Sterling area during August 15-317 22, 2024 (Fig. 1). One site was on public land north of the Kenai Municipal Airport and was 318 partially boggy while the other two sites were in low-density residential areas with mixed 319 evergreen and deciduous forests, including the home of the case patient. 320 Trapping was primarily conducted using standard (7.6 x 8.9 x 22.9 cm) Sherman live traps (H.B. 321 Sherman Traps, Inc., Tallahassee, FL, USA), which were deployed for a total of 1384 trap-nights 322 (i.e., the sum of the number of traps placed each night over the course of trapping) in 2021 and 323 1825 trap-nights in 2024. Tomahawk live traps (Model 102, Tomahawk Live Trap, Hazelhurst, 324 WI, USA) were used for larger mammals such as squirrels, and pitfall traps were deployed to 325 catch shrews. The Tomahawk and Sherman traps were baited with a mixture of peanut butter and 326 oats, with carrots included in Fairbanks due to colder temperatures. Trapped animals were 327 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint anesthetized with isoflurane then humanely euthanized and examined prior to examination for 328 skin lesions and collection of blood and tissue samples (liver, skin, pooled heart and lung, and 329 pooled kidney and spleen; lesion material if observed). All trapping and animal handling 330 followed protocols approved by the CDC and UAF’s Institutional Animal Care and Use 331 Committees (CDC: 3183DOTMULX, 3400DOTMULX; UAF: 152295). Tissue specimens were 332 also obtained from roadkill animals collected by the UAM researchers and from animals 333 collected by the Alaska Department of Fish and Game. All voucher specimens and their 334 associated tissues have been deposited and catalogued in UAM’s Mammal Collection 335 (Supplementary Data S1). 336 Museum specimen sampling 337 338 We screened tissue samples from specimens of voles, shrews, and squirrels archived at UAM. 339 The following species were selected for screening based on the availability of tissue samples, as 340 well as preliminary results about seropositivity and molecular BRPV detection in wild small 341 mammals: Northern Red-backed Voles (n = 201 samples), Tundra Voles (Alexandromys 342 oeconomus; 22), Singing Voles (Microtus miurus; 6), a vole of undetermined species (Microtus 343 sp.; 1), Masked Shrews (43), Tundra Shrews (Sorex tundrensis; 8), Montane Shrew (Sorex 344 monticolus; 3), and an American Red Squirrel (1). The specimens were collected at various 345 locations around Fairbanks, Kenai, and Denali National Park and Preserve from 1991-2019 (Fig. 346 1; Supplementary Data S1). 347 348 Historical samples archived at UAM were flash-frozen without a buffer and housed in liquid 349 nitrogen-cooled cryovats that maintain vapor-phase nitrogen at -170°C. The work surface and 350 tools were cleaned with 10% bleach. The box containing the appropriate tissues was removed 351 from the cryovat freezer and placed on an insulated cold-block while outside the cryovat. Using a 352 separate set of cleaned instruments for each tube, a small subsample (roughly 0.1 gram) of tissue 353 was removed from each cryotube and transferred into appropriately labeled tubes. Between 354 sampling, instruments were cleaned in 10% bleach solution and then rinsed with water and dried. 355 After subsampling, the original samples were returned to the cryovat freezer, and the subsamples 356 were placed in a labeled box on a second insulated cold-block. Once all subsampling was 357 completed, the box of subsamples was stored in an ultra-low freezer until shipment. Samples 358 were shipped on dry ice. 359 360 Laboratory Diagnostics 361 Tissue samples were homogenized in 500μ l of PBS with 250μ l of 1mm zirconia silica beads 362 (Biospec) and a Mini BeadBeater 24 following the manufacturer’s recommendations. 100μ l of 363 tissue homogenate was used for DNA extraction with a MagMAX deep-well magnetic processor 364 (ThermoFisher Scientific, https://www.thermofisher.com) using the MagMAX DNA Multi-365 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint Sample Ultra kit. DNA samples were assessed for the presence of OPXV with the CDC OPXV 366 generic real-time PCR assay (Li et al. 2006). 367 A modified anti-OPXV IgG enzyme-linked immunosorbent assay (ELISA) was conducted on 368 sera obtained from cardiac punctures or on dried blood spots collected on Nobuto filter paper 369 (Advantec, San Diego, CA) as previously described (Hutson et al. 2009, Doty et al. 2017). 370 Samples were considered positive if the OD value passed the cut-off value in at least two 371 consecutive dilutions (1:100 and 1:200). 372 Viral isolation and DNA sequencing 373 Following PCR, samples with OPXV DNA amplification were added to cell culture for virus 374 isolation. Remaining tissue homogenate from the DNA extraction process was added to BSC-40 375 cell monolayers (African Green Monkey Chlorocebus sabaeus kidney cell line) in T-25 cell 376 culture flasks and incubated at 37°C with 5% CO2 (Hutson et al. 2013). DNA was then extracted 377 from propagated virus and sheared to target 500 bp fragments on a Covaris S220 instrument 378 (Covaris, Woburn, Massachusetts, USA). The Swift Accel-NGS 2S DNA Library Kit was used 379 for library preparation with dual indexing following the manufacturer’s instructions (Swift 380 Biosciences, product no longer available). Libraries were visualized with an Agilent 2200 Tape 381 Station (Agilent Technologies, Santa Clara, California, USA), followed by sequencing on an 382 Illumina MiSeq machine with the MiSeq Reagent v3 600 cycle kit (Illumina, San Diego, 383 California, USA). 384 Genome Assembly 385 Reads were trimmed to a minimum length of 50 bp and minimum quality 20 with FaQCs v1.34 386 (https://github.com/LANL-Bioinformatics/FaQCs), and three nucleotides were removed from 387 each end. The 2015 BRPV reference genome (GenBank accession MN240300.1) was used for 388 mapping, with bwa mem v0.7.17 (https://github.com/lh3/bwa). Mapped reads were then 389 extracted with samtools v.1.9 (https://github.com/samtools/samtools) and de novo assembled 390 with SPAdes v3.13.0 (https://github.com/ablab/spades), using the --careful flag and --cov-cutoff 391 off. Contigs with coverage >10 were manually assembled in Geneious Prime 2023.0.4 392 (Dotmatics, Boston, Massachusetts, USA) after mapping. Draft genomes were then edited in 393 repeat regions at approximate positions 150, 163, 174, and 200 kb to match repeat lengths from 394 the 2015 BRPV genome, and only the left ITR and 500 bp of the right ITR were retained. 395 Trimmed reads were mapped back to the draft genomes, and a polished genome was generated 396 with ivar v0.1, using the conditions above except requiring a minimum read depth of 10. Finally, 397 to produce a final genome, ITRs were copied from the left to the right end. Annotations from the 398

genome were transferred using Liftoff (https://github.com/agshumate/Liftoff). 399 400 Phylogenetic Analysis 401 402 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint Genome alignments were performed with MAFFT v7.490 in Geneious Prime 2023.0.4 using 403 FFT-NS-i x2 and final alignments are provided as supplementary files (Supplementary Data S2). 404 Bayesian phylogenetic trees were generated in BEAST v2.7.7 (Bouckaert et al. 2019) in two runs 405 with the following parameters based on Gigante et al. (2019): GTR+I+G nucleotide substitution 406 model (4 gamma categories, 35% invariant), relaxed lognormal clock (exponential distribution of 407 ucldStdev prior with mean = 0.333), and Yule model prior until all parameters exhibited ESS > 408 200 after 10% burn-in, visualized in Tracer. Default parameters were used unless specified. Run 409 log and tree files were combined using LogCombiner after 10% burn-in. A maximum clade 410 credibility tree was estimated in TreeAnnotator based on sampling frequency of 1000 and 10% 411 burn-in. We also estimated a maximum-likelihood tree on the whole genome using IQ-TREE 412 (Nguyen et al. 2015), with branch supports obtained using the ultrafast bootstrap method and 413 1000 replicates. The nucleotide substitution model for the maximum-likelihood tree (GTR+F) 414 was determined using ModelFinder implemented through IQ-TREE (Kalyaanamoorthy et al. 415 2017). Vaccinia virus (GenBank AY603355) was used as an outgroup. 416 417

418 419 This research was supported by generous donations to the University of Alaska Museum’s 420 Mammal Collection from the Jay Pritzker Foundation and Nancy Eliason. MMJ was supported 421 by a Gates Cambridge Scholarship enabled by grant OPP1144 from the Bill & Melinda Gates 422 Foundation. CNM, AM, and AC were supported in part by appointment to the Research 423 Participation Program at the Centers for Disease Control and Prevention, administered by the 424 Oak Ridge Institute for Science and Education through an interagency agreement between the 425 U.S. Department of Energy and CDC. MMM was supported by an NSF CAREER Award 426 (2238801). The authors thank Mallory Gulbranson and Kyndall B.P. Hildebrandt at the 427 University of Alaska Museum for access to frozen tissues in their care and assistance with 428 subsampling. The authors also thank Nicholas Fowler from Alaska Department of Fish and 429 Game for his help with accessing sampling sites on the Kenai Peninsula. 430 431 Disclaimer: The findings and conclusions in this report are those of the authors and do not 432 necessarily represent the official position of the Centers for Disease Control and Prevention. 433 434

435 436 Alexander, A. D., et al. Survey of wild mammals in a Chesapeake Bay area for selected 437 zoonoses. Journal of Wildlife Diseases 8, 119-126 (1972). 438 439 Babkin, I. V., Babkina, I. N. & Tikunova, N. V. An update of orthopoxvirus molecular evolution. 440 Viruses 14, 388 (2022). 441 442 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint Baltensperger, A. P., Lanier, H. C., & Olson, L. E. Extralimital terrestrials: A reassessment of 443 range limits in Alaska’s land mammals. PLOS ONE 19, e0294376 (2024). 444 445 Bonwitt, J., Doty, J. B., McCollum, A. M. & Nakazawa, Y. Zoonotic orthopoxviruses: 446 Innocuous rash or global public health threat? In: Sing, A. (eds) Zoonoses: Infections Affecting 447 Humans and Animals (Springer, Cham, 2022). 448 449 Bouckaert, R., et al. BEAST 2.5: An advanced software platform for Bayesian evolutionary 450 analysis. PLOS Computational Biology 15, e1006650 (2019). 451 452 Colella, J. P., et al. Leveraging natural history biorepositories as a global, decentralized, 453 pathogen surveillance network. PLOS Pathogens 17, e1009583 (2021). 454 455 Cronin, C. E., Juman, M. M., Richardson, A. J., et al. Biodiversity databases as underutilized 456 resources for pathogen discovery: a quantitative synthesis of bat and rodent tissue collections in 457 natural history museums. Preprint at https://doi.org/10.1101/2025.11.28.691153 (2025). 458 459 Diaz, J. H. The disease ecology, epidemiology, clinical manifestations, management, prevention, 460 and control of increasing human infections with animal orthopoxviruses. Wilderness & 461 Environmental Medicine 32, 528-536 (2021). 462 463 Doty, J. B., et al. Assessing Monkeypox virus prevalence in small mammals at the human-animal 464 interface in the Democratic Republic of the Congo. Viruses 9, 283 (2017). 465 466 Doty, J. B., et al. Isolation and characterization of Akhmeta virus from wild-caught rodents 467 (Apodemus spp.) in Georgia. Journal of Virology 93 (2019). 468 469 Emerson, G. L., et al. The phylogenetics and ecology of the orthopoxviruses endemic to North 470 America. PLOS ONE 4, e7666 (2009). 471 472 Gigante, C. M., et al. Genome of Alaskapox virus, a novel orthopoxvirus isolated from Alaska. 473 Viruses 11, 708 (2019). 474 475 Hutson, C. L., et al. A prairie dog animal model of systemic orthopoxvirus disease using West 476 African and Congo Basin strains of monkeypox virus. Journal of General Virology 90, 323-333 477 (2009). 478 479 Hutson, C. L., et al. Transmissibility of the monkeypox virus clades via respiratory transmission: 480 investigation using the prairie dog-monkeypox virus challenge system. PLOS ONE 8, e55488 481 (2013). 482 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint 483 Juman, M. M., et al. Museum collections and machine learning guide discovery of novel 484 coronaviruses and paramyxoviruses. Preprint at https://doi.org/10.1101/2025.09.11.675601 485 (2025). 486 487 Kalyaanamoorthy, S., Minh, B. Q., Wong, T. K. F., von Haeseler, A., & Jermiin, L. S. 488 ModelFinder: fast model selection for accurate phylogenetic estimates. Nature Methods 14, 587-489 589 (2017). 490 491 Li. Y., Olson, V. A., Laue, T., Laker, M. T., & Damon, I. K. Detection of monkeypox virus with 492 real-time PCR assays. Journal of Clinical Virology 36, 194-203 (2006). 493 494 Linzey, A. V., Henttonen, H., Sheftel, B. & Batsaikhan, N. Myodes rutilus (amended version of 495 2016 assessment). The IUCN Red List of Threatened Species 2020, e.T4975A164372228 496 (2020). Accessed on 14 February 2025. 497 498 MacDonald, S. O. & Cook, J. A. Recent Mammals of Alaska (University of Alaska Press, 499 Fairbanks, 2009). 500 501 Mooring, E. Q., Olson, L. E. & Werle, Z. Second case of novel Orthopoxvirus infection in a 502 Fairbanks-area resident. State of Alaska Epidemiology Bulletin 11 (2020). 503 504 Mooring E. Q., Newell, K. G., & Werle, Z. Additional cases of Orthopoxvirus infection in 505 Fairbanks-area residents, 2021. State of Alaska Epidemiology Bulletin 13 (2021). 506 507 Mooring, E. Q., et al. Six cases of Borealpox and evidence of a zoonotic source—Alaska, 2020–508 2023. Clinical Infectious Diseases ciaf497 (2025). 509 510 Nguyen, L. T., Schmidt, H. A., von Haeseler, A. & Minh, B. Q. IQ-TREE: a fast and effective 511 stochastic algorithm for estimating maximum-likelihood phylogenies. Molecular Biology and 512 Evolution 32, 268–274 (2015). 513 514 Regnery, D. C. Isolation and partial characterization of an orthopoxvirus from a California vole 515 (Microtus californicus). Archives of Virology 94, 159-162 (1987). 516 517 Reynolds, M. G, Guagliardo, S. A. J., Nakazawa, Y. J., Doty, J. B., & Mauldin, M. R. 518 Understanding orthopoxvirus host range and evolution: from the enigmatic to the usual suspects. 519 Current Opinion in Virology 28, 108-115 (2018). 520 521 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint Riutord-Fe, C., et al. Transmission of MPXV from fire-footed rope squirrels to sooty mangabeys. 522 Nature (2026). 523 524 Rogers, J. H., et al. Fatal borealpox in an immunosuppressed patient treated with antivirals and 525 vaccinia immunoglobulin—Alaska, 2023. Clinical Infectious Diseases 80, 1053–1059 (2025). 526 527 Runck, A. M., Matocq, M. D., & Cook, J. A. Historic hybridization and persistence of a novel 528 mito-nuclear combination in red-backed voles (genus Myodes). BMC Evolutionary Biology 9, 529 114 (2009). 530 531 Springer, Y. P., et al. Novel orthopoxvirus infection in an Alaska resident. Clinical Infectious 532 Diseases 64, 1737–1741 (2017). 533 534 Thomassen, H. A., et al. Pathogen-host associations and predicted range shifts of human 535 Monkeypox in response to climate change in Central Africa. PLOS ONE 8, e66071 (2013). 536 537 Tseng, K. K., et al. Viral genomic features predict Orthopoxvirus reservoir hosts. 538 Communications Biology 8, 309 (2025). 539 540 Viana, M., et al. Assembling evidence for identifying reservoirs of infection. Trends in Ecology 541 & Evolution 29, 270–279 (2014). 542 543 WHO. Multi-country monkeypox outbreak: situation update. Disease Outbreak News (2022). 544 Available at: https://www.who.int/emergencies/disease-outbreak-news/item/2022-DON396. 545 Accessed on 24 February 2026. 546 547 Wiens, B. J. & Colella, J. P. Unidirectional mitochondrial introgression despite limited nuclear 548 admixture in North American red-backed voles, Clethrionomys rutilus and C. gapperi. Ecology 549 and Evolution, 15:e72603 (2025). 550 551 Yates, T. L., et al. The ecology and evolutionary history of an emergent disease: Hantavirus 552 pulmonary syndrome. BioScience 52, 989 (2002). 553 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.22.713527doi: bioRxiv preprint