In this study, we investigated the coevolution relationship between bats and their viral parasites: coronaviruses and paramyxoviruses. These two groups of viruses were chosen because they have been shown to be zoonotic (Rizzo et al 2017). The sequences of the RNA-dependent RNA polymerase (RdRP) gene of 60 bat coronavirus (BatCoV) isolates, the RNA polymerase large (L) gene of 36 paramyxovirus isolates, and the cytochrome B (cytB) gene of 61 bat species were used to build phylogenetic trees. ParaFit analyses were then performed to determine the relationship between coronavirus and bat genetic trees and between paramyxovirus and bat genetic trees. Both ParaFit Global and ParaFit Individual tests were performed. In the Global test, both groups of the viruses were found to have a significant coevolution relationship with their bat hosts. In the Individual test, 51 (85%) of the 60 BatCoV isolates and 7 (19%) of the 36 paramyxovirus isolates had a significant coevolution relationship with their bat hosts.
ParaFit analyses also revealed that closely related BatCoV isolates are found in closely related bat species (Figure 2). One example of such observation is that closely related BatCoV isolates 1B/CHN/EU420137, AH2011/CHN/KJ473795, 1A/CHN/EU420138, and HKU8/CHN/EU420139 are found in Miniopterus pusillus, Miniopterus fuliginosus, and Miniopterus magnater that are very close to each other. Another example is the coevolution relationship between closely related bats including Tylonycteris pachypus, Hypsugo savii, Vespertilio sinensis, Neoromicia capensis, and Ia io and the following BatCoV isolates: GX2012/CHN/KJ473822, 206645-40/ITA/MG596802, SC2013/CHN/KJ473821, NL13845/CHN/MG021451, 5038/RSA/MF593268, and PML-PHE1/RSA/KC869678. Similar coevolution relationships are found in closely related bat species Rhinolophus ferrumequinum, Rhinolophus blasii, Rhinolophus pusillus, Rhinolophus macrotis, Rhinolophus sinicus, Rhinolophus pearsonii, and Rhinolophus affinis and the following BatCoV isolates: RF1/CHN/DQ412042, JTMC15/CHN/KU182964, JL2012/CHN/KJ473811, 16BO133/ROK/KY938558, HeB2013/CHN/KJ473812, SX2013/CHN/KJ473813, Shaanxi2011/CHN/JX993987, HuB2013/CHK/KJ473814, HKU3-1/CHN/DQ022305, YN2013/CHK/KJ473816, Rs3367/CHN/KC881006, GX2013/CHN/KJ473815, Anlong-103/CHN/KY770858, Rp3/CHN/DQ071615, LYRa11/CHN/KF569996, and RaTG13/CHN/MN996532 (Figure 2). For paramyxoviruses, closely related bat species Neoromicia nanus, Myotis alcathoe, Myotis myotis, Myotis capaccinii, Myotis daubentoniid, and Myotis bechsteinii are found to carry the following closely related isolates: GH36/GHA/FJ609192, 3-320/BGR/HQ660163, N78-14/GER/HQ660166, 6-43/BGR/HQ660164, NMS09-48/GER/HQ660165, LN2012/CHN/KJ641656, and NM98-46/GER/HQ660170 (Figure 3).
As mentioned above, 85% (51/60) of BatCoV isolates but only 19% (7/36) of paramyxovirus isolates were found to have a significant coevolution relationship with their bat hosts by the ParaFit Individual test. Since significant coevolution was found in both groups of the viruses by the ParaFit Global test, this low positive individual link rate in paramyxoviruses may be due to the small sample size. However, this observation may suggest that BatCoVs adapt to their bat hosts more readily than paramyxoviruses. This possibility is supported the fact that coronaviruses are more commonly found in bats than paramyxoviruses (Liang et al. 2017). Further studies are warranted to test this hypothesis.
We postulate that that evolutionary relationship and close habitat of bat species contribute to inter-species transmission of viruses. One observation supporting this hypothesis is that bat coronavirus isolates 206645−40/IT/MG596802 (bat host: Hypsugo savii) and 206645−63/IT/MG596803 (bat host: Pipistrellus kuhlii) share 99.46% nucleotide identity in their RdRp gene and are found in bats living in the same geographical area, Italy. In addition, bat coronavirus isolates 16BO133/ROK/KY938558, JL2012/KJ473811, and JTMC15/KU182964 are highly related with >99.6% nucleotide identity in the RdRp gene and are found in bats distributing in areas near each other, including Jilin Province, China (for isolates JL2012/KJ473811 and JTMC15/KU182964) and South Korea (for isolate 16BO133/ROK/KY938558) that is close to north China.
Most bat SARSr-CoV isolates are from Rhinolophus bats, but isolates As6526/CHN/KY417142 and Yunnan2011/CHN/JX993988 are found in Aselliscus stoliczkanu and Chaerephon plicatus, respectively. Although these two bat species are phylogenetically far apart, they live in the same geographical area, Yunnan Province, China. This observation suggests that distantly related bats in the same geographical location may carry closely related BatCoVs, leading to inter-species transmission of the viruses.
We also hypothesized that coevolution of distantly related viruses with hosts living in different geographical areas is unlikely to occur. This hypothesis is supported by the observation that the BatCoV isolate BM48−31/BGR/GU190215 is distantly related to other bat SARSr-CoV isolates (Figure 2). Its host, Rhinolophus blasii, lives in west Asia, north Africa, and south Europe (Simmons 2005) that are geographically distant from Yunnan Province, China, where many of the other BatCoV isolates examined in this study originated (Hu et al. 2017). Results of previous studies suggest that Rhinolophus sinicus is the natural hosts of human SARS-CoVs (Lau et al. 2005, Ge et al. 2013, Yang et al. 2015). Many bat SARSr-CoVs are detected in Rhinolophus sinicus and Rhinolophus ferrumequinum that distributed in Asia, Afria, and Europe. However, SARSr-CoVs that are highly related to human SARS-CoVs have not been found in Rhinolophus ferrumequinum that live in Africa and Europe. The BatCoV isolate LUX16_A_24/LUX/KY502395 found in Rhinolophus ferrumequinum in Luxembourg, Europe is distantly related to the SARS-CoV isolates found in humans and civets (Pauly et al. 2017).
For paramyxoviruses, Pteropid bats have been shown to be the natural reservoir of Henipavirus (Young et al. 1996, Halpin et al. 2000, Johara et al. 2001) and are speculated to be responsible for its outbreak in Malaysia, Australia, Singapore, Philippine, India, and Bangladesh during the period of 1995 – 2015 (Murray et al. 1995, Chua et al. 1999, Chua et al. 2000, Hsu et al. 2004, Chadha et al. 2006, Arankalle et al. 2011, Ching et al. 2015). Several species of Pteropid bats, including Pteropus alecto, Pteropus conspicillatus, Pteropus giganteus, and Pteropus vampyrus, have been found to carry Henipaviurs. These bats live in southeast Asia and Oceania (Simmons 2005), where Henipavirus pandemic occurred.
Most human SARS-CoV isolates examined in this study are derived from Rhinolophus sinicus in Yunnan Province, China. As Rhinolophus sinicus is found only in China, Nepal, Vietnam, and north India (Simmons 2005), SARS-CoV outbreak has not occurred in other places such as Europe, Africa, Oceania, or America. SARS-CoV-2 is responsible for the COVID-19 pandemic (International Committee on Taxonomy of Viruses). In this study, the BatCoV isolate RaTG13/CHN/MN996532 is found to be most close to SARS-CoV-2 with >97% nucleotide identity in the RdRp gene and 96% identity at the whole genome level. There are approximately 1100 bases that are different between the genomes of BatCoV RaTG13/MN996532 and SARS-CoV-2s, suggesting that BatCoV RaTG13/MN996532 requires at least one intermediate host to transmit to humans (Zhou and Yang 2020). As the host of isolate RaTG13/MN996532 is Rhinolophus affinis residing in Yunnan Province, China, it has been speculated that SARS-CoV-2 is derived from Rhinolophus bats roosting in areas near Yunnan Province, China, such as southwest China, Myanmar, Laos, Vietnam, or other southeast Asian countries (Latinne et al 2020).
Divergence of bats can be traced back to tens of million years ago (Teeling et al. 2005, Agnarsson et al. 2011). It has been estimated that coronaviruses diverged tens of thousand years ago (Woo et al. 2012). This difference may be due to the fact that the genome of coronaviruses is RNA that is more prone to mutations than DNA. It has also been estimated that coronaviruses have been infecting birds or bats for tens of million years; this would confer the opportunity for coevolution of coronaviruses with their hosts (Wertheim et al. 2013).