Genetic Diversity and Geographic Distribution of Bat-borne Hantaviruses

The recent discovery that multiple species of shrews and moles (order Eulipotyphla, families Soricidae and Talpidae) from Europe, Asia, Africa and/or North America harbour genetically distinct viruses belonging to the family Hantaviridae (order Bunyavirales) has prompted a further exploration of their host diversification. In analysing thousands of frozen, RNAlater®-preserved and ethanol-fixed tissues from bats (order Chiroptera) by reverse transcription polymerase chain reaction (RT-PCR), ten hantaviruses have been detected to date in bat species belonging to the suborder Yinpterochiroptera (families Hipposideridae, Pteropodidae and Rhinolophidae) and the suborder Yangochiroptera (families Emballonuriade, Nycteridae and Vespertilionidae). Of these, six hantaviruses are from Asia (Xuân Sơn virus and Đakrông virus in Vietnam; Láibīn virus in China and Myanmar; Huángpí virus and Lóngquán virus in China; and Quezon virus in the Philippines); three are from Africa (Mouyassué virus in Côte d’Ivoire and Ethiopia; Magboi virus in Sierra Leone; and Makokou virus in Gabon); and one from Europe (Brno virus in the Czech Republic). Molecular identification of many more bat-borne hantaviruses is expected. However, thus far, none of these newfound viruses has been isolated in cell culture and it is unclear if they cause infection or disease in humans. Future research must focus on myriad unanswered questions about the genetic diversity and geographic distribution, as well as the pathogenic potential, of bat-borne viruses of the family Hantaviridae. Introduction As recently as a decade ago, the single exception to the strict rodent association of hantaviruses was Thottapalayam virus, a long-unclassified virus originally isolated from the Asian house shrew (Suncus murinus) (Carey et al., 1971). Analysis of the genome of Thottapalayam virus strongly supported an ancient non-rodent host origin and an early evolutionary divergence from rodent-borne hantaviruses (Song et al., 2007a; Yadav et al., 2007). Equipped with this information and employing reverse transcription polymerase chain reaction (RT-PCR), genetically distinct hantaviruses were detected in multiple species of shrews (order Eulipotyphla, family Soricidae) from widely separated geographic regions in Asia, Europe, Africa and North America. In addition to challenging the conventional view that rodents are the sole or principal reservoirs of hantaviruses, the discovery of soricidborne hantaviruses suggested that moles (order Eulipotyphla, family Talpidae) might also harbour hantaviruses, and as predicted, several talpid-borne hantaviruses have been found in Europe, Asia and North America (Yanagihara et al., 2014). Acquisition of new knowledge about the spatial and temporal distribution, host range and genetic diversity of hantaviruses in shrews and moles, and caister.com/cimb 1 Curr. Issues Mol. Biol. Vol. 39


Introduction
As recently as a decade ago, the single exception to the strict rodent association of hantaviruses was Thottapalayam virus, a long-unclassified virus originally isolated from the Asian house shrew (Suncus murinus) (Carey et al., 1971). Analysis of the genome of Thottapalayam virus strongly supported an ancient non-rodent host origin and an early evolutionary divergence from rodent-borne hantaviruses (Song et al., 2007a;Yadav et al., 2007). Equipped with this information and employing reverse transcription polymerase chain reaction (RT-PCR), genetically distinct hantaviruses were detected in multiple species of shrews (order Eulipotyphla, family Soricidae) from widely separated geographic regions in Asia, Europe, Africa and North America. In addition to challenging the conventional view that rodents are the sole or principal reservoirs of hantaviruses, the discovery of soricidborne hantaviruses suggested that moles (order Eulipotyphla, family Talpidae) might also harbour hantaviruses, and as predicted, several talpid-borne hantaviruses have been found in Europe, Asia and North America .
Acquisition of new knowledge about the spatial and temporal distribution, host range and genetic diversity of hantaviruses in shrews and moles, and more recently in bats, has been made possible largely through the generosity of museum curators and field mammalogists, who willingly granted access to their archival tissue collections. The availability of such well-curated specimens provides strong justification for the continued long-term maintenance and near-term expansion of archival tissue repositories for future investigations and innovative applications (Dunnum et al., 2017).
Phylogenetic analyses of these newfound hantaviruses indicate at least four distinct clades, with the most divergent lineage comprising hantaviruses harboured by the European mole (Talpa europaea) (Kang et al., 2009c) and several bat species (Sumibcay et al., 2012;Weiss et al., 2012;Arai et al., 2013;Guo et al., 2013;Gu et al., 2014c;Xu et al., 2015;Arai et al., 2016b;Witkowski et al., 2016;Straková et al., 2017;Těšíková et al., 2017;Arai et al., 2019a;Arai et al., 2019b). On the basis of intensive phylogenetic analysis of the full-length S-and M-genomic segments, a new taxonomic classification has been proposed, which includes four newly defined subfamilies in the family Hantaviridae (Maes et al., 2019). In addition, the realization that these Eulipotyphla-and Chiroptera-borne hantaviruses are more genetically diverse than those found in rodents, as well as the discovery of hantaviruslike sequences in fish and reptile hosts, suggests that the evolutionary history of the Hantaviridae is far more complex than previously conjectured. Thus, the dawn of a new era in hantavirology is focused on exploring the inconvenient evidence that rodents may not be the original mammalian hosts of primordial hantaviruses. Also, the once-growing complacency and indifference towards rodentborne hantaviruses is being replaced by renewed zeal to fill major gaps in our understanding about the ecology, transmission dynamics and pathogenic potential of these newly discovered, still-orphan hantaviruses, before the emergence of the next new hantavirus disease outbreak occurs .
The history of research on this once-obscure group of rodent-borne viruses has been marked by rediscovery and new beginnings . In this brief review, the genetic diversity and geographic distribution of hantaviruses from bats will be summarized in the context of hantaviruses harboured by rodents, shrews and moles to provide insights into their evolutionary origins and future risk and disease emergence.

Taxonomy
Previously classified in an unassigned order and in the genus Hantavirus of the family Bunyaviridae, hantaviruses have been recently reclassified and reassigned to a new family, designated Hantaviridae, in the order Bunyavirales (Adams et al., 2017;. Viruses belonging to the four other genera formerly in the family Bunyaviridae (namely, Nairovirus, Orthobunyavirus, Phlebovirus, and Tospovirus) are now members of new viral families: Feraviridae, Fimoviridae, Jonviridae, Nairoviridae, Peribunyaviridae, Phasmaviridae, Phenuiviridae, and Tospoviridae.
The family Hantaviridae has been further classified into four subfamilies (Actantavirinae, Agantavirinae, Mammantavirinae and Repantavirinae) ( Table 6.1). The subfamily Mammantavirinae comprises four genera (Loanvirus, Mobatvirus, Orthohantavirus and Thottimvirus), based on DEmARC analysis, using concatenated complete S and M amino acid-coding regions (Maes et al., 2019). Virus members of each genus in the subfamily Mammantavirinae, with their mammalian reservoir host category (rodent, shrew, mole and bat), are summarized in Table  6.1. In addition, recently published hantavirus-like sequences from fish and reptile, which are closely related and ancestral to hantaviruses harboured by mammals (Shi et al., 2018), have been classified into the newly created subfamilies of Actantavirinae and Agantavirinae and Repantavirinae, respectively (Table 6.1).
All rodent-borne hantaviruses belong to the genus Orthohantavirus, which also comprises nearly all of the genetically distinct hantaviruses recently detected in shrews and moles (Table 6.1). The exceptions are members of the genus Thottimvirus: Thottapalayam virus in the Asian house shrew (Suncus murinus) (Carey et al., 1971;Song et al., 2007a;Kang et al., 2011c) and Imjin virus in the Ussuri white-toothed shrew (Crocidura lasiura) (Song et al., 2009), as well as probably Uluguru virus in the Geata mouse shrew (Myosorex geata)  and Kilimanjaro virus in the Kilimanjaro mouse shrew (Myosorex zinki) . By contrast, all newfound hantaviruses harboured by bats belong to the genera Loanvirus and Mobatvirus (Table 6.1).
To minimize unnecessary confusion, the viruses newly classified in the family Hantaviridae will be referred to in this review as hantaviruses, even if the genus Hantavirus no longer exists and is not part of the revised taxonomy. Members of the family Hantaviridae are enveloped viruses possessing a tripartite genome, comprising three single-stranded, negative-sense RNA segments, designated S (small), M (medium), and L (large), which encode the nucleocapsid protein, the envelope glycoproteins (Gn and Gc), and the viral RNA-dependent RNA polymerase, respectively (Plyusnin et al., 1996;Guardado-Calvo and Rey, 2017).
A fundamental epidemiological factor in HFRS and HCPS cases is exposure to environments contaminated with urine, faeces and/or saliva from infected rodents. Thus, outbreaks of HFRS and HCPS are often associated with encroachment of rodent habitats or irruptions of reservoir rodent populations with invasion of human dwellings. Although HFRS and HCPS cases have also resulted from seemingly trivial exposures, the intimate handling of rodents does not necessarily constitute sufficient exposure. That is, while mammalogists have frequent occupational contact with rodents and are presumably at increased risk, several seroepidemiological studies have indicated  (Yanagihara et al., 1984;Vapalahti et al., 1995;Lundkvist et al., 2000;Fritz et al., 2002).
This has been corroborated in a more recent study, in which only four of 757 persons who had handled neotomine or sigmodontine rodents in North America exhibited serum IgG antibodies against Sin Nombre virus (Fulhorst et al., 2007). Also, during the height of the HCPS outbreak in the Four Corners region in 1993, forest and park service personnel showed no evidence of Sin Nombre virus infection (Vitek et al., 1996). By contrast, studies in Europe show clear associations between orthohantavirus infection and exposure to rodent excreta among certain high-risk occupation groups, such as animal trappers, forestry workers and farmers (Groen et al., 1995;Vapalahti et al., 1999;Grygorczuk et al., 2008;Mertens et al., 2011;Wróblewska-Łuczka et al., 2017;Wróblewska-Łuczkawroblewska et al., 2017), and individuals, such as hunters, whose recreational activities encroach on wildlife habitats (Deutz et al., 2003).
The principal symptoms and clinical features of both HFRS and HCPS include high fever, chills, headache, generalized myalgia, abdominal pain, and nausea and vomiting. Prominent shared characteristics of HFRS and HCPS also include thrombocytopenia (Connolly-Andersen et al., 2015) and vascular leakage, or increased endothelial cell permeability (Gorbunova et al., 2010). While HFRS varies in clinical severity, ranging from mild to severe and life threatening, with mortality ranging from < 1% to ≥ 15%, depending on the specific orthohantavirus (Yanagihara and Gajdusek, 1988;Vaheri et al., 2013;Avšič-Županc et al., 2016), HCPS is generally severe, and despite intensive care treatment, has mortality rates of 25% or higher ( Jonsson et al., 2010;Martinez et al., 2010;MacNeil et al., 2011). The clinical management of HFRS and HCPS is largely supportive, with careful fluid and electrolyte management and monitoring of cardiopulmonary and/or renal function. Dialysis may be required in some patients with severe HFRS. For HCPS patients, mechanical ventilation is frequently required, and other life-saving measures, such as extracorporeal membrane oxygenation, may be necessary (Duchin et al., 1994;Mertz et al., 2006).
Previously, the conventional view posited that each genetically distinct orthohantavirus was harboured by a single rodent species, with which it co-evolved. Mounting evidence refutes this overly simplistic paradigm and instead supports the concepts of host sharing and host switching. That is, the same orthohantavirus species may be hosted by more than one reservoir rodent species. For example, Tula virus has been reported in the common vole, southern vole (Microtus rossiaemeridionalis), field vole (Microtus agrestis), Altai vole (Microtus obscurus), European pine vole (Pitymys subterraneus) and Eurasian water vole (Arvicola amphibius) (Plyusnin et al., 1994;Song et al., 2002;Song et al., 2004;Schmidt-Chanasit et al., 2010;Schlegel et al., 2012a;Klempa et al., 2013b;Polat et al., 2018).
Apart from examples of the same orthohantavirus species being harboured by multiple rodent species, the same rodent species can also host more than one orthohantavirus species. Examples include the field vole which hosts Tula virus in Europe and a newly discovered orthohantavirus, named Tatenale virus, in the United Kingdom (Pounder et al., 2013); and the striped field mouse, which serves as the reservoir of Hantaan virus in Asia, also hosts the Kurkino and Saaremaa genotypes of Dobrava-Belgrade virus in Europe (Nemirov et al., 1999;Klempa et al., 2013a;Németh et al., 2013).

Shrews and moles as reservoir hosts
Although the isolation of Thottapalayam virus from an Asian house shrew predated the isolation of prototype Hantaan virus, and despite early reports of HFRS antigens in tissues of the Eurasian common shrew (Sorex araneus), Eurasian water shrew (Neomys fodiens) and European mole in Russia (Gavrilovskaya et al., 1983;Tkachenko et al., 1983) and the former Yugoslavia , shrews and moles (order Eulipotyphla, families Soricidae and Talpidae) have generally been dismissed as being unimportant in the phylogeography and evolution of hantaviruses.
However, guided by these long-ignored historical accounts and emboldened by the isolation of a novel hantavirus, named Imjin virus from the Ussuri white-toothed shrew captured along the Imjin River, near the demilitarized zone in the Republic of Korea (Song et al., 2009;Gu et al., 2011), an aggressive search was launched for hantavirus RNA, using a brute-force RT-PCR approach based on labour-intensive, time-consuming and trial-by-error oligonucleotide design. In analysing total RNA, extracted from more than 1500 tissues from 50 shrew species collected in Europe, Asia, North America and Africa, between 1982 and 2012, multiple genetically distinct soricid-borne orthohantaviruses have been detected (Table 6.1). These include Seewis virus in the Eurasian common shrew in Switzerland (Song et al., 2007b), Hungary and Finland (Kang et al., 2009a), Russia (Yashina et al., 2010) and Poland (Gu et al., 2014b); Ash River virus in the masked shrew (Sorex cinereus) and Jemez Springs virus in the dusky shrew (Sorex monticolus) in the USA (Arai et al., 2008a); Kenkeme virus in the flat-skulled shrew (Sorex roboratus)  and Artybash virus in the Laxmann's shrew (Sorex caecutiens)  in Russia; Sarufutsu virus in the long-clawed shrew (Sorex unguiculatus) in Japan (S. Arai et al., unpublished data); Cao Bằng virus in the Chinese mole shrew (Anourosorex squamipes) in Vietnam (Song et al., 2007c) and Taiwanese mole shrew (Anourosorex yamashinai) in Taiwan ; Camp Ripley virus in the northern short-tailed shrew (Blarina brevicauda) in the USA (Arai et al., 2007); Boginia virus in the Eurasian water shrew in Poland (Gu et al., 2013a); Bowé virus in the Doucet's musk shrew (Crocidura douceti) in Guinea (Gu et al., 2013b); Azagny virus in the West African pygmy shrew (Crocidura obscurior) in Côte d'Ivoire (Kang et al., 2011b); and Jeju virus in the Asian lesser white-toothed shrew (Crocidura shantungensis) in Korea (Arai et al., 2012). By contrast, Uluguru virus in the geata mouse shrew and Kilimanjaro virus in the Kilimanjaro mouse shrew in Tanzania  appear to be more closely related to Thottapalayam virus and Imjin virus, which have been classified in the genus Thottimvirus (Table 6.1).
Host sharing or spillover has also been found for soricid-borne orthohantaviruses. Examples include Seewis virus in the Eurasian common shrew and Eurasian pygmy shrew (Sorex minutus) in the Czech Republic and Germany (Schlegel et al., 2012b) and in Poland (Gu et al., 2014b). Also, Seewis virus has been detected in the tundra shrew (Sorex tundrensis) and large-toothed Siberian shrew (Sorex daphaenodon) in Russia (Yashina et al., 2010) and in the Mediterranean water shrew (Neomys anomalus) in Austria (N. Nowotny, unpublished data) and Poland (Gu et al., 2014b). Moreover, Jemez Springs virus, which is harboured by the dusky shrew, has been found in the vagrant shrew (Sorex vagrans), Trowbridge's shrew (Sorex trowbridgii) and American water shrew (Sorex palustris) in the USA (H.J. Kang et al., unpublished data).
Testing more than 600 tissue samples from 12 of the approximately 40 extant mole species has yielded seven genetically distinct hantaviruses: Asama virus in the Japanese shrew mole (Urotrichus talpoides) from Japan (Arai et al., 2008b); Oxbow virus in the American shrew mole (Neurotrichus gibbsii) (Kang et al., 2009b) and Rockport virus in the eastern mole (Scalopus aquaticus) (Kang et al., 2011a) from the USA; Nova virus in the European mole from Hungary (Kang et al., 2009c), France (Gu et al., 2014a), Poland (Gu et al., 2014b) and Belgium (Laenen et al., 2018); Bruges virus in the European mole from Belgium and United Kingdom (Laenen et al., 2018); Asturias virus in the Iberian mole (Talpa occidentalis) from Spain (Gu et al., 2016c); and Dahonggou Creek virus in the long-tailed mole (Scaptonyx fusicaudus) from China (Kang et al., 2016). As in rodents, the same talpid host can harbour more than one hantavirus species, as exemplified by the European mole, which hosts Nova virus and Bruges virus in Belgium (Laenen et al., 2018) and the Iberian mole, which appears to harbour Asturias virus, Bruges virus and Nova virus in Spain (Gu et al., 2016c).
Because many mole species have been unavailable for testing and sample sizes were small, numbering fewer than 10 individuals for many of the species tested, the list of seven probably represents a gross underestimation of the actual number of talpid-borne hantaviruses. More targeted searches for hantavirus RNA in mole species, such as the Altai mole (Talpa altaica), blind or Mediterranean mole (Talpa caeca), Caucasian mole (Talpa caucasica), Roman mole (Talpa romana) and Balkan mole (Talpa stankovici), will likely lead to the discovery of additional hantaviruses and provide insights into host-switching events.
Nova virus, which represents among the most highly divergent lineages (Kang et al., 2009c), has been classified in the genus Mobatvirus with several hantaviruses harboured by bats (  (Gu et al., 2014a;Gu et al., 2014b). Much like Seewis virus, which is widespread in the Eurasian common shrew throughout Europe, Nova virus probably occurs throughout the vast distribution of the European mole.

Bats as reservoir hosts
Bats (order Chiroptera) represent the second largest order of mammals, with more than 1200 species or approximately 20% of all classified mammal species worldwide. Bats are found in all continents, except Antarctica. They are the only mammals capable of controlled and sustained flight. While most bats are insectivorous or frugivorous, three extant bat species, which are all native to the Americas, feed exclusively on the blood of birds or mammals. Bats serve as natural reservoirs of many microbial pathogens, and their mobility through flight, longevity and social structures contribute to the transmission and spread of zoonotic diseases.
The order Chiroptera was formerly divided in two suborders: Megachiroptera and Microchiroptera. However, due to the paraphyly of the Microchiroptera, a new taxonomic nomenclature has been proposed, comprising the suborder Yinpterochiroptera or Vespertilioniformes (megabats or fruit bats in the family Pteropodidae in Megachiroptera and a few Microchiroptera families) and the suborder Yangochiroptera or Pteropodiformes (the remaining Microchiroptera families) (Teeling, 2009). Irrespective of the classification, bat species in both suborders have been found to host viruses in the newly created genera of Loanvirus and Mobatvirus, within the subfamily Mammantavirinae, of the family Hantaviridae, suggesting that primordial hantaviruses may have emerged in an early common ancestor of bats.
Compared to the multitude of orthohantaviruses reported from more than half of the 50 Eulipotyphla species tested , the cumulative number of newly recognized bat-borne hantaviruses is exceedingly low (Sumibcay et al., 2012;Weiss et al., 2012;Arai et al., 2013;Guo et al., 2013;Gu et al., 2014c;Xu et al., 2015;Arai et al., 2016b;Witkowski et al., 2016;Straková et al., 2017;Těšíková et al., 2017;Arai et al., 2019a;Arai et al., 2019b). The reasons for the low success rates of detecting hantavirus RNA in bat tissues are not altogether clear. One possibility is the highly divergent nature of their genomes, as well as the very focal or localized nature of hantavirus infection in bats and the small sample sizes of bat species, as well as primer mismatches, suboptimal PCR cycling conditions, and variable tissue preservation with degraded RNA (Arai et al., 2013;Gu et al., 2014c). Alternatively, bats may be less susceptible to hantavirus infection or may have developed immune mechanisms to curtail viral replication and/or persistence. In any case, as shown in Table 6.3, the full genomes of bat-borne hantaviruses are largely incomplete. Suboptimal primer design, imperfect cycling conditions, low RNA yields and poor RNA integrity (particularly in poorly preserved archival tissues collected under harsh field conditions) may have thwarted amplification and genome sequencing efforts (Gu et al., 2014c). However, while fewer bat species have been identified as reservoirs, the hantaviruses they harbour are among the most genetically diverse described to date.

Hantaviruses in bats of the suborder Yangochiroptera
The geographic distribution of four bat species belonging to the suborder Yangochiroptera, which harbour newfound orthohantaviruses, is shown in Fig. 6.1. A brief description of each hantavirus and its reservoir bat host is provided below and data are summarized in Table 6.2. Full genomes of Láibīn virus and Brno virus are available (Table 6.3).
Mouyassué virus (MOYV) (genus Loanvirus). Detected in ethanol-fixed liver tissues from two of 12 banana pipistrelles captured near Mouyassué village (N 05°22'07'' , W 03°05'37'') in Aboisso District, 130 km from Abidjan, in the extreme south-eastern region of Côte d'Ivoire during June 2011, MOYV was one of the first bat-borne hantaviruses reported (Sumibcay et al., 2012). The originally proposed three-letter abbreviation was changed to MOYV to avoid confusion with Moussa  (Quan et al., 2010). MOYV has also been found in one of nine cape serotines captured in Dhati Walel National Park (N 09°13'33'' , E 34°52'37''), at an elevation of 1427 m, in Ethiopia during February 2014 (Těšíková et al., 2017). The successful amplification of MOYV from ethanol-fixed tissues augments the potential pool of archival tissues for future exploratory studies of hantaviruses in bats, as well as other insectivorous small mammals that have shared ancestry with bats. The banana pipistrelle and cape serotine, which are distributed widely in forests and savannas across sub-Saharan Africa (Monadjem et al., 2017b(Monadjem et al., , 2017c (Weiss et al., 2012). The host species (family Nycteridae) has a wide geographic range, encompassing much of sub-Saharan Africa, with the exception of the Horn of Africa and parts of southern Africa. There is an apparently disjunct population in western Mauritania close to the border with Senegal, and an isolated record from central Mali.
Huángpí virus (HUPV) (genus Loanvirus). Sequences from a 1115-nucleotide region of the S segment and a 343-nucleotide region of the L segment are available for HUPV, which was detected in one of five Japanese house bats, captured in Huángpí District (N 30°52'30'', E 114°22'30''), one of 13 districts of Wuhan, the capital of Húběi Province in China during 2012 (Guo et al., 2013). The Japanese house bat, a member of the family Vespertilionidae, is found in the southern Ussuri region (Russia and China), the western half of China including Taiwan, Japan, the Korean Peninsula, Vietnam, Myanmar, and India ( Fig. 6.1) (Bates and Tsytsulina, 2008).
Láibīn virus (LAIV) (genus Mobatvirus). The full-length genome (1935 nucleotide S, 3908 nucleotide M and 6531-nucleotide L segment) is available for LAIV strain BT20 (Table 6.3), which was originally detected in one of 32 black-bearded  (Xu et al., 2015). LAIV (strain BT33) has also been detected in one of 74 black-bearded tomb bats captured in May 2014 in Băisè, in Guăngxī Province, bordering Vietnam (Xu et al., 2019), as well as in two of 15 black-bearded tomb bats (strains MM4377M17 and MM4378M18), trapped in November 2015 in Shwe Ba Hill Cave in the Sagaing Region of Myanmar (Arai et al., 2019a). LAIV has the longest M segment, which includes a 504-nucleotide long 3′-non-coding region. The black-bearded tomb bat is widely distributed throughout Asia and Southeast Asia (Fig.  6.1), including Brunei, Cambodia, China, India, Indonesia, Laos, Malaysia, Myanmar, Philippines, Singapore, Sri Lanka, Thailand, Timor-Leste and Vietnam .
Brno virus (BRNV) (genus Loanvirus). In testing 53 bats, which had died accidentally or which were found dead, BRNV was detected in two of 12 common noctules in Brno (N 49°12', E 16°37'), the second largest city in the Czech Republic (Straková et al., 2017). The entire genome of BRNV strain 7/2012 is available (1269-nucleotide S, 3408-nucleotide M and 6432-nucleotide L) ( Table  6.3). To date, this is the only bat-borne hantavirus reported from Europe. Common noctules have a wide Palearctic distribution ( Fig. 6.1), including Europe and southern Scandinavia to the Urals and Caucasus; Turkey to Israel and Oman; western Turkmenistan, western Kazakhstan, Uzbekistan, Kyrgyzstan, and Tajikistan to southwest Siberia and perhaps the Himalayas. Its occurrence in North Africa is questionable, and a record from Mozambique is considered dubious (Csorba and Hutson, 2016).

Hantaviruses in bats of the suborder Yinpterochiroptera
The geographic distribution of seven bat species belonging to the suborder Yinpterochiroptera, which harbour newfound orthohantaviruses is shown in Fig. 6.2. A brief description of each hantavirus and its reservoir bat host is provided below and data are summarized in Table 6 (Guo et al., 2013). Intermediate horseshoe bats are widespread throughout South Asia (Fig. 6.2), southern and central China and Southeast Asia ; Chinese rufous horseshoe bats range from northern South Asia into northern Southeast Asia, and much of central, southern and southwestern China ; and Formosan lesser horseshoe bats are reported as a Taiwanese species, but it is very similar to least horseshoe bats (Rhinolophus pusillus) in body size, echolocation call frequency and mitochondrial gene sequences. Least horseshoe bats have a very wide range from South Asia eastward to Japan, occurring also in southern and southwestern China, including Taiwan, southward through mainland Southeast Asia to Indonesia and Borneo . Its ability to also host LQUV warrants study.
Xuân  (Arai et al., 2019a). Recently, XSV strains have also been reported in ashy roundleaf bats from Guăngxī Province in China (Xu et al., 2019). Pomona roundleaf bats and ashy roundleaf bats are sympatric, but the latter species usually occurs in much lower abundance. Nevertheless, they often roost in the same caves, which may account for spillover of XSV.
The genome of prototype XSV strain VN1982B4 consists of 1748-nucleotide S, 3756-nucleotide M and 6520-nucleotide L (Table  6.3). The Hipposideros genus of the family Hipposideridae is one of the most speciose of insectivorous bats, with more than 70 species distributed across Africa, Europe, Asia and Australia. The vast geographic distribution of the Pomona roundleaf bat throughout Vietnam (Fig. 6.2 (Sumibcay et al., 2012;Weiss et al., 2012), it is very likely that many more genetically divergent mobatviruses are harboured by bat species in this large family.
Makokou virus (MAKV) (genus Mobatvirus). A partial 3582-nucleotide region of the L segment is available for MAKV, which was detected in one of 123 Noack's roundleaf bats, trapped in a cave near the city of Makokou (N 0°34', E 12°52') in Gabon during 2009 . A member of the family Hipposideridae, the Noack's roundleaf bat is one of the most common bat species in Africa ( Fig. 6.2 (Monadjem, 2017a).
Đakrông virus (DKGV) (genus Mobatvirus). The entire genome of DKGV strain VN2913B72 (1746-nucleotide S, 3622-nucleotide M and 6535-nucleotide L) (Table 6.3) has been detected in one of two Stoliczka's Asian trident bats, captured in Đakrông Nature Reserve (N 16°39'3'', E 107°2'13'') in Quảng Trị Province in Vietnam in August 2013 (Arai et al., 2019b). The Stoliczka's Asian trident bat, one of three species in the genus Aselliscus, is found in northern Southeast Asia ( Fig.  6.2), from Myanmar and southern China in the North through Thailand, Laos and Vietnam to Pulau Tioman Island, Peninsular Malaysia in the South. A closely related species, the Dong Bac's trident bat (Aselliscus dongbacana) , overlaps in body size, geographic distribution, echolocation and habitat, but orthohantavirus RNA could not be detected in this species (S. Arai et al., unpublished data).
Quezon virus (QZNV) (genus Mobatvirus). As the only hantavirus to date in a megabat, or flying fox species, QZNV, which was detected in one of 15 Geoffroy's rousettes, captured in Quezon National Park (N 13°59', E 121°55'), located approximately 130 km southeast of Manila, on Luzon Island, in the Philippines during 2009 (Arai et al., 2016b), expands the host range of hantaviruses. The Geoffroy's rousette, one of 10 species in the genus Rousettus, is a megabat or Old World fruit bat typically roosting in caves and feeding on fruit, nectar and pollen throughout Southeast Asia and in the Malesia region of Oceania, in Myanmar, Thailand, Cambodia, Laos, Vietnam, Singapore, Indonesia, Borneo, East Timor, Solomon Islands, Bismarck Archipelago, Papua New Guinea and the Philippines (Fig. 6.2). Reproductive synchrony between Geoffroy's rousettes and two closely related species, long-tongued nectar bat (Macroglossus minimus) and lesser short-nosed fruit bat (Cynopterus brachyotis), has been documented in the Philippines (Heideman and Utzurrum, 2003).

Molecular phylogeny
Previously, the segregation of hantaviruses into clades that paralleled the molecular phylogeny of their rodent hosts suggested codivergence (Plyusnin et al., 1996). Recently, this concept has been challenged on the basis of the disjunction between the evolutionary rates of the host and virus species. Preferential host switching and local host-specific adaptation have been proposed to account for the largely congruent phylogenies (Ramsden et al., 2009). However, host-switching events alone do not completely explain the co-existence and distribution of genetically distinct hantaviruses among host species in three divergent taxonomic orders of small mammals spanning across four continents (Bennett et al., 2014).
Phylogenetic analyses, based on S-, M-and L-genomic sequences, using maximum-likelihood and Bayesian methods, indicate that all bat-borne hantaviruses (or loanviruses and mobatviruses) share a common ancestry (Fig. 6.3). In all analyses, Nova virus from the European mole segregates with the bat-associated mobatviruses. The basal position of bat-and mole-borne mobatviruses and selected shrew-borne hantaviruses (or thottimviruses), such as Thottapalayam virus in the Asian house shrew and Imjin virus in the Ussuri white-toothed shrew, suggests that bats, moles and/or shrews, rather than rodents, may have served as the primordial mammalian hosts of ancestral hantaviruses (Fig. 6.3  Geographic-specific clustering was evidenced by the close phylogenetic relationship between prototype XSV VN1982 from Phú Thọ Province and XSV F42640 and XSV F42682 from neighbouring Tuyên Quang Province in northern Vietnam. On the other hand, XSV F44583, XSV 44601 and XSV 44580 from Quảng Nam province in central Vietnam clustered together. Although limited differences are present in phylogenetic trees based on each segment, tree topologies are generally congruent and supported by significant bootstrap values (> 70%) and posterior node probabilities (> 0.70).
To compare the evolutionary relationships of loanviruses, mobatviruses, orthohantaviruses and thottimiviruses with their hosts, phylogenetic trees were reconstructed for co-phylogeny mapping, using consensus topologies based on amino acid sequences of the nucleocapsid protein, Gn and Gc glycoproteins and RNA-dependent RNA-polymerase. Such tanglegrams (Fig. 6.4), constructed using TreeMap 3b1243, exhibited congruent segregation of viruses within the family Hantaviridae, according to the subfamily of their reservoir hosts, with no evidence of host switching except for Asama virus, Oxbow virus, Nova virus and Rockport virus, which are all hantaviruses harboured by moles (Arai et al., 2008b;Kang et al., 2009b;Kang et al., 2011a). Asama virus and Oxbow virus were more closely aligned to soricine shrew-borne orthohantaviruses, Rockport virus shared a common ancestry with orthohantaviruses hosted by cricetid rodents and Nova virus was phylogenetically related to batborne mobatviruses (Figs. 6.3 and 6.4). Genetic recombination and reassortment events have also played a significant role in the evolution and the currently recognized diversity of the family Hantaviridae (Lee SH et al., 2017;Castel et al., 2017;Klempa, 2018).
Based on exhaustive phylogenetic analyses, using multiple methods, all bat-borne loanviruses and mobatviruses reported to date share a common ancestry, which is consistent with their host phylogeny. However, recently, in testing blood samples, obtained from 53 bats captured in south-eastern Brazil, for hantavirus infection, the partial S segment of an orthohantavirus, showing very high sequence similarity with Araraquara virus, was amplified from a Seba's short-tailed bat (Carollia perspicillata), a widespread frugivorous bat species in the family Phyllostomidae (Sabino-Santos et al., 2018). Also, partial S segment sequences of an Araraquara virus-like orthohantavirus was detected in the urine, heart, liver, lungs, spleen and kidneys of a common vampire bat (Desmodus rotundus), a sanguivorous phyllostomids species. Phylogenetic analysis showed that this Araraquara virus-like hantavirus formed a monophyletic group with Araraquara virus strain P5/Cajuru (GenBank EF571895), amplified from the blood of a HCPS patient in San Paulo (de Sousa et al., 2008) and strain IB_SP_bol66/2171 (EU170223), amplified from a hairy-tailed bolo mouse (Necromys lasiurus) captured in San Paulo, Brazil, in June 2005 (Ramsden et al., 2008). The detection of Araraquara virus-like orthohantavirus sequences in two phyllostomids species is decidedly unexpected. That a nearly identical sequence of an Araraquara virus-like orthohantavirus was amplified from two very different bat species in a laboratory known to work with Araraquara virus raises a high degree of suspicion that this surprising observation might represent laboratory contamination or PCR carryover.
That is, this observation is reminiscent of previously reported serological evidence of hantavirus infection in the common serotine (Eptesicus serotinus) and greater horseshoe bat (Rhinolophus ferrumequinum) captured in Korea (Kim et al., 1994). Subsequent genetic analysis of hantavirus isolates from these bat species proved to be indistinguishable from prototype Hantaan virus ( Jung and Kim, 1995), indicating laboratory contamination. Nevertheless, intensified investigations by independent groups are warranted to confirm if bats serve as reservoir hosts of Araraquara viruslike orthohantaviruses. If they do, this would be somewhat akin to Rockport virus, harboured by the fossorial eastern mole, which shares a most recent common ancestor with cricetid rodent-borne orthohantaviruses. The eastern mole is sympatric and syntopic with cricetid rodent species, which serve as reservoir hosts of orthohantaviruses, suggesting a host-switching event in the distant past (Kang et al., 2011a).  (Woolhouse and Gowtage-Sequeria, 2005), a better understanding about the geographic distribution of zoonotic viruses and their hosts is vital to assess risk and to predict future viral disease outbreaks, as well as to discover previously unrecognized disease associations in the case of still-orphan viruses harboured by small mammals (Han et al., 2016;Dunnum et al., 2017). While not all viruses in search of diseases, or orphan viruses, warrant investigations to ascertain their pathogenic potential and virulence at the time of their discovery, selected viruses, particularly those related to viruses known to cause severe and lifethreatening syndromes, such as HFRS and HCPS, are worthy of high research priority. No one would have predicted that rodent-borne orthohantaviruses could cause acute renal insufficiency with varying degrees of haemorrhage and shock, as well as a rapidly progressive, frequently fatal respiratory disease.

Because many emerging viral infectious diseases have their origins in mammalian reservoir hosts
The realization that rodent-borne orthohantaviruses are capable of causing HFRS and HCPS raises the possibility that one or more newfound soricid-and talpid-associated orthohantaviruses or thottimviruses, and possibly bat-borne loanviruses and mobatviruses, may similarly cause a wide spectrum of febrile illnesses. In this regard, prospective studies of neotomine and sigmodontine rodent-borne orthohantaviruses in the early 1980s might have provided important clues about their pathogenicity and virulence long before the abrupt recognition of HCPS in 1993. In much the same way, one or more of the newly identified nonrodent-borne viruses of the family Hantaviridae may cause outbreaks of human disease and/or serve as surrogate antigens for the diagnosis of previously unrecognized diseases. Robust serological assays and other sensitive rapid diagnostic technologies, now under development, will assist in establishing if these newest members of the family Hantaviridae are pathogenic for humans. Also, studies on the genetics, transmission dynamics and disease-causing potential of one or more of the newly identified viruses in shrews, moles and bats, as well as African rodents, may better prepare the next generation of health care workers to be vigilant for the next outbreak of hantaviral disease.
That viruses of the subfamily Mammantavirinae are distributed widely across three taxonomic orders of terrestrial mammals has changed previously held dogmas about hantavirus evolution and phylogeography (Bennett et al., 2014). At the same time, although the global landscape of hantavirus distribution is far richer now than it was just a decade ago (Fig. 6.5), it is unclear if selection bias in the collection and availability, and subsequent testing of tissues from rodents, shrews, moles and bats, has distorted or unduly influenced our current concepts. For example, while rodents of multiple genera and species serve as reservoirs of orthohantaviruses, orthohantaviruses have been found in only two genera (Hylomyscus and Stenocephalemys) of rodents belonging to the subfamily Murinae in Africa, despite decades-long exploratory investigations. By contrast, in Eurasia, rodents belonging to genera in both the subfamilies Murinae (Apodemus, Bandicota, Niviventer and Rattus) and Arvicolinae (Arvicola, Eothenomys, Lemmus, Microtus, Myodes and Pitymys) are known to harbour orthohantaviruses. In the Americas, on the other hand, a dizzying diversity of orthohantaviruses has been found in rodent genera not only of the subfamilies Murinae (Rattus) and Arvicolinae (Microtus and Myodes), but also in the subfamilies Neotominae (Peromyscus and Reithrodontomys) and Sigmodontinae (Akodon, Calomys, Holochilus, Necromys, Oligoryzomys, Oryzomys, Sigmondon and Zygodontomys) (de Oliveira et al., 2014;Yanagihara et al., 2015). The caveat here is that rodents of the subfamilies Neotominae and Sigmodontinae do not exist in the Old World.
Multiple hantaviruses exhibiting far greater genetic diversity have been detected recently in shrews and moles representing genera in three subfamilies (Crocidurinae, Myosoricinae and Soricinae) of the family Soricidae and in two subfamilies (Talpinae and Scalopinae) of the family Talpidae Much work is obviously still needed to better understand the genetic diversity of viruses within the family Hantaviridae and the geographic distribution of their hosts. In particular, more intensive investigations are warranted to investigate beyond the more common families within the order Rodentia, as well as of the less well-known and low abundant genera within the order Eulipotyphla, and of the less speciose members of the suborders Yinpterochiroptera and Yangochiroptera.
Unlike the high prevalence of orthohantavirus infection reported in multiple species of shrews and moles, the absence of hantavirus infection in the majority of bat species analysed to date and the low prevalence of hantavirus RNA in only a few individuals of a given bat species would tend to argue against a long-standing hantavirus-bat host relationship, and instead support spillover or host-switching events. That is, the gleaning feeding behaviour of some bats, such as Nycteris, presents the possibility of acquired infection from excreta of well-established terrestrial reservoirs of orthohantaviruses. However, this seems highly improbable because bat-borne hantaviruses are among the most genetically diverse described to date and are phylogenetically distinct from hantaviruses harboured by rodents, shrews and moles.
With the discovery of divergent hantavirus lineages in three taxonomic orders of placental mammals, there is renewed interest in investigating their genetic diversity and geographic distribution. Newly acquired knowledge that bats harbour distinctly divergent lineages of hantaviruses emphasizes the truly complex evolutionary origins and phylogeography of a group of viruses once thought to be restricted to rodents. At this point, it would not be surprising if hantaviruses were found in small mammals belonging to other taxonomic orders, such as Erinaceomorpha (hedgehogs) and even Afrosoricida (tenrecs). Such discoveries may provide additional insights into the dynamics of hantavirus transmission, potential reassortment of genomes, and molecular determinants of hantavirus pathogenicity. As importantly, a sizable expansion of the Hantaviridae sequence database would provide valuable tools for refining diagnostic tests and enhancing preparedness for future outbreaks caused by still-orphan newfound hantaviruses.

Future research
Among the urgent questions for future research about the genetic diversity and geographic distribution of viruses within the family Hantaviridae are the following: • What other taxonomic orders serve as reservoir hosts? That is, with the discovery of hantaviruslike sequences in reptile and fish, what is the host range of Hantaviridae? And do other terrestrial mammals, such as hedgehogs and tenrecs, harbour hantaviruses? • What do the reservoir hosts have in common?
Are some hosts more permissive to being recipients of host switches? Are some loanviruses, mobatviruses, orthohantaviruses and thottimiviruses more likely to switch into new hosts? • What are the possibilities of developing more sensitive and less labour-intensive, primerindependent molecular detection tools for discovering other members of the family Hantaviridae? • What is driving the evolutionary diversification of loanviruses, mobatviruses, orthohantaviruses and thottimiviruses? How do we definitively date divergence? What are the host evolutionary relationships?
Another unanswered question is whether or not one or more of the recently detected or yet-to-be discovered shrew-and mole-borne orthohantaviruses, or bat-borne loanviruses and mobatviruses, cause infection or disease in humans. In this regard, progress has been hampered by the lack of virus isolates from non-rodent hosts. That is, while multiple orthohantaviruses have been isolated from many rodent species and adapted to growth in cell culture or laboratory-bred rodents, there are, to date, only two thottimivirus isolates from shrews, namely, Thottapalayam virus from the Asian house shrew (Carey et al., 1971) and Imjin virus from the Ussuri white-tooted shrew (Song et al., 2009), and a single mole-borne mobatvirus isolate, namely, Nova virus from the European mole (Gu et al., 2016b). There are no isolates of loanviruses or mobatviruses from bats. Future virus-isolation attempts may benefit from more innovative approaches, such as the use of reservoir host-derived cell cultures from tissues of shrews and bats (Eckerle et al., 2014).
Thus far, while there is suggestive evidence that Bowé virus and Uluguru virus (a shrew-borne orthohantavirus found in Guinea and a presumptive shrew-borne thottimivirus found in Tanzania, respectively) might cause infection in humans, as evidenced by serological tests using recombinant nucleocapsid proteins as antigens (Heinemann et al., 2016), there is no definitive proof that these or any of the newfound non-rodent-associated loanviruses, mobatviruses, orthohantaviruses and thottimiviruses cause a clinically identifiable disease or syndrome in humans .
However, it would be premature to conclude that this would be true for other newfound and still-undiscovered soricid-, talpid-or Chiropteraborne viruses of the family Hantaviridae for the very reason that the majority of rodent-borne orthohantaviruses do not cause infection or disease in humans. A significant shortcoming of any investigation in search of a rare infectious disease event is the failure to study individuals who are affected by that rare event. On the one hand, the inability to detect antibodies against a given non-rodent-borne loanvirus, mobatvirus, orthohantavirus or thottimivirus in a given study population may indicate that that particular virus does not cause infection in humans. On the other hand, this same (negative) result could mean that the study simply failed to enrol subjects exposed to that virus. In other words, if infection with a given non-rodent-borne loanvirus, mobatvirus, orthohantavirus or thottimivirus is associated with a rare or uncommon disease, one would be unable to show pathogenicity in humans. In this regard, even at the height of the 1993 HCPS outbreak in the Four Corners region, no serological evidence of Sin Nombre virus infection could be found in patients with a variety of diseases or in health care workers, parks service personnel and mammalogists. Only patients with HCPS had evidence of Sin Nombre virus infection. Thus, even with the most lethal infectious agent, one would erroneously conclude that the microbe is non-pathogenic or non-infectious, unless the 'right' patients were tested.
Sensitive serological tests have facilitated the rapid screening of wild rodents for evidence of orthohantavirus infection. And in large part, this type of pre-screening has allowed more focused RT-PCR testing of antibody-positive rodents. By contrast, there are virtually no data about the prevalence of IgG antibodies against loanviruses, mobatviruses, orthohantaviruses and thottimiviruses in shrews, moles and bats, primarily because of the unavailability of blood or serum specimens in archival collections and the lack of suitable immunological reagents. However, as in rodents, one would surmise that antibody-positive shrews, moles and bats would more likely have detectable viral RNA. To what extent the presence of neutralizing antibodies against loanviruses or mobatviruses in bats would more readily pre-select which bats to test by RT-PCR is unknown. As it now stands, this is a moot point because no hantavirus isolates are available from bats to be able to perform neutralization tests.
For answers to these questions, and myriad others, reagents need to be developed. And multidisciplinary collaborative studies must be designed to optimize specimen collection to facilitate the isolation and characterization of newfound bat-borne loanviruses and mobatviruses to better understand virus-host interactions.