Insights into longevity and virus-driven adaptation from <i>Myotis</i> bat genomes

Bats (order Chiroptera) represent approximately 20% of all known mammalian species and are one of the most phenotypically diverse clades of mammals⁴. Since their emergence 60 million years ago⁵, many bat lineages have independently evolved a wide variety of life history strategies and phenotypic traits, including exceptional longevity, viral tolerance and immune defences^2,3. Systems in which shared traits have evolved de novo multiple times are powerful resources for dissecting the genetic basis of phenotypes. The largest genus of bats, Myotis, emerged approximately 33 million years ago⁶ and encompasses over 139 described species spanning 6 continents and a wide range of ecological niches⁷. Myotis species demonstrate some of the most extreme variation in lifespan among mammals^1,8, including a sixfold difference in lifespan between the longest-lived species⁹ (Myotis brandtii, 42 years; Fig. 1a) and the shortest-lived species¹⁰ (Myotis nigricans, 7 years), which diverged approximately 10.6 million years ago¹¹. Moreover, Myotis species are representative of bats’ notable immune mechanisms that enable viral tolerance and pathogen resistance¹² contributing to their role as key zoonotic reservoirs^2,13.

To study how longevity and infectious disease resistance have evolved in Myotis, we used an integrated field-to-functional-genomics approach to assemble near-complete genome assemblies for eight Myotis species with primary cell culture resources for functional validation. We identify copy-number-variable genes associated with RNA viral tolerance and stress response, as well as a trans-species copy number polymorphism of a key immune factor, PKR. Consistent with the extreme lifespans of Myotis relative to its body size, we identified selective evolutionary signatures in genes associated with longevity- and cancer-related processes. In contrast to humans and other primates, in which virus adaptation is driven by interactions with RNA viruses, we find that modes of virus adaptation in bats differ between DNA and RNA viruses. Together, our results highlight pleiotropic adaptations contributing to the lifespan and immune phenotypes of Myotis bats.

We investigated the landscape of structural variation within the tightly conserved Myotis karyotype. With only 6 exceptions across over 60 studied species, all Myotis have a conserved 2n = 44 karyotype—a remarkable phenomenon for a genus spread across six continents and 33 million years of divergence^4,6. The Myotis karyotype comprises three large autosomes; one small metacentric autosome; 17 small telocentric autosomes; and metacentric X and Y chromosomes¹⁶. Consistent with this broad cytological conservation, we find large scale synteny across the Nearctic Myotis in this study (Fig. 1e). However, structural variants (SVs), including inversions, duplications and translocations, are relatively common within chromosomes, especially adjacent to putative centromeric regions (Fig. 1e,f). This includes an approximately 20 Mb block at the subtelomeric end of chromosome V15 that displays frequent and recurrent inversions and translocations across the Nearctic Myotis, and contains a 10 Mb locus that was recently identified as a potential target of recent selection by adaptive introgression¹⁸ (Fig. 1e (red bar) and Extended Data Fig. 3d).

Using SyRI¹⁹, we identified between 6,813 and 8,013 SVs per Nearctic Myotis genome relative to their outgroup, M. myotis; 97–99% of these events were under 10 kb. In the three large autosomes, which constitute around 30% of the genome, we catalogued an average of 509 SVs (Supplementary Table 3). By contrast, in the small autosomes, constituting around 65% of the genome, we observed an average of 316 events, highlighting the distinct structural evolution between these chromosome types with SVs found at twofold higher density in larger autosomes (Supplementary Table 3). However, large (at least 10 kb) duplications, inverted duplications and inverted translocations were more common on small autosomes compared with on the large autosomes (Supplementary Table 3). We also quantified the distribution of transposable elements (TEs) across chromosomes (Fig. 1f,g). Consistent with previous observations in other Chiropteran genomes²⁰, long interspersed nuclear elements appeared to be enriched around the predicted centromere (Fig. 1f), with simulations showing that this trend is primarily limited to the long, metacentric autosomes (Extended Data Fig. 3a and Supplementary Table 3). The concentration of segmental duplications was also significantly correlated with TE density in each species (linear regression, P < 0.001; Fig. 1f,h and Extended Data Fig. 3b), highlighting the importance of TEs in facilitating structural evolution. Together, these results highlight high levels of structural variation occurring in the context of a highly constrained karyotype.

Among the SVs identified in Nearctic Myotis, the gene PKR—previously shown to be duplicated in certain Myotis²¹—stood out as an interferon-stimulated gene with antiviral activity against both DNA and RNA viruses. Using our near-complete genome assemblies, we fully resolved the sequence and structure of the two previously known haplotypes: H1, containing a single copy of PKR (PKR2); and H2, containing two tandemly duplicated copies of PKR (PKR1 and PKR2; Fig. 2a). We also identified a third haplotype, H3, with three tandem duplicates of PKR (PKR1, PKR2 and a third copy) present only in Myotis californicus. Notably, while 7 out of 9 Myotis species carried duplicated haplotypes, 5 of these cases were heterozygous for the duplicated haplotype (that is, H1/H2 or H2/H3; Fig. 2b). Two Myotis individuals (M. lucifugus and Myotis evotis) carried only non-duplicated haplotypes (that is, H1/H1; Fig. 2b). To determine the evolutionary history of the duplicates, we used GeneRax²² to construct a tree from alignments of all PKR gene copies across Nearctic Myotis, using Pipistrellus pygmaeus as a non-Myotis outgroup (Fig. 2c). We found that PKR2 is the ancestral copy of PKR, and that PKR1 originated from a single duplication event at the root of Myotis. These results highlight that both the duplicated and unduplicated haplotypes have probably been segregating for tens of millions of years, representing an ancient trans-species polymorphism.

PKR is a stress response and innate immune factor that becomes active after detection of double-stranded RNA through sensing, autophosphorylation and dimerization²³. Active PKR shuts down protein translation and restricts viral replication²³. Although PKR1 and PKR2 are both expressed in basal or interferon-stimulated Myotis velifer cells, only their independent functional impacts have previously been investigated²¹. Given the co-expression and the complex haplotype diversity that we identified in Myotis, we set out to determine whether the two PKR copies act additively, synergistically or as dominant negatives in several essential functions. We investigated the functional impact of the duplicates’ co-expression on total PKR protein expression levels, cell viability, inhibition of translation, antiviral restriction and homodimer/heterodimer formation (Fig. 2d–h). To pinpoint the contribution of PKRs, we used a heterologous cell system²¹ with PKR-KO HeLa cells transfected with either an empty pSG5 plasmid (control), or pSG5 encoding PKR1, PKR2 or PKR1 + PKR2 (1:1 ratio) from Myotis myotis or M. velifer. In this system, PKR transfection leads to its autoactivation²¹.

First, we found that, although Myotis Flag–PKR1 was expressed at lower levels than Flag–PKR2, PKR1 + PKR2 co-expression did not affect the overall steady-state protein expression levels of PKRs (Extended Data Fig. 4a). Second, we performed cell viability assays at two doses of transfected PKRs and found that cell viability was only affected at a high dose (Extended Data Fig. 4d). Third, using a luciferase reporter translational assay, we found that PKR1 + PKR2 coexpression led to an intermediate translation shutdown, showing neither synergistic nor dominant negative effect of the PKR duplicates (Fig. 2d). Fourth, we tested PKR1 + PKR2 co-expression effects on infections by two different model RNA viruses: VSV-GFP (vesicular stomatitis virus encoding a GFP reporter, a representative of Rhabdoviridae family that is common in bats, including Myotis) and SINV-GFP (Sindbis virus, a representative of Togaviridae) (Fig. 2e,f and Extended Data Fig. 4c,e). We found that PKR1 + PKR2 restricted viral infections to a similar extent—or slightly lower in the case of VSV—to PKR1 or PKR2 alone (Fig. 2e,f and Extended Data Fig. 4c,e; P > 0.05). While PKR1 homodimerized, PKR1 and PKR2 did not form heterodimers (Fig. 2g,h and Extended Data Fig. 4b in the SINV condition). This suggests that, since their duplication, divergence between Myotis PKR copies has been particularly strong at the protein–protein interface hindering their ability to form heterodimers.

These results suggest that PKR1 and PKR2 do not exhibit dominant-negative or synergistic effects but, instead, act additively in their primary cellular effector functions. While their duplication may broaden resistance to certain DNA and RNA viral antagonists²¹, the increased cell toxicity observed at high PKR doses suggests a trade-off that could explain the absence of PKR duplication in other mammals. This trade-off may also account for the persistence of both duplicated and non-duplicated haplotypes in Myotis. Consistent with this, PKR1, which is generally more potent in our experiments, is expressed at significantly lower levels than PKR2 in M. velifer cells²¹. Together, these results reveal gene duplicates still segregating across several Myotis species, probably shaped by a balance between antiviral defence and cell toxicity.

To further examine how viruses may have shaped the genomes of Myotis bats, we tested for adaptive signatures in virus-interacting proteins (VIPs) in Myotis and other bats. VIPs are host proteins that physically interact with viral proteins (such as CD45; Fig. 3a), and can be proviral (contributing to viral infection; for example, viral receptors), antiviral (protective against viral infection; such as interferons) or both, depending on infection stage and virus type. VIPs commonly have major roles in basic host function and, as a result, most host cell biological pathways include VIPs. Previous studies investigating positive selection across mammals have found an enrichment for adaptation among a set of 5,527 manually curated VIPs, defined as host proteins that have at least one experimentally verified physical interaction with a viral protein, RNA or DNA²⁴.

By calculating an enrichment score from the ratio of positive selection in VIPs compared with their matched control genes using BUSTED-MH²⁵, we found that like other mammals, Myotis show an enrichment for adaptation at VIPs (Fig. 3b–d and Supplementary Table 5). Physical host–virus interactions may not always result in fitness effects in the host. We therefore repeated our analysis using a gene set restricted to VIPs with experimental evidence of specific pro- or anti-viral effects, and therefore with a stronger expectation of fitness effects. We expect that, if viruses were not the primary driver of adaptation in VIPs, then there would be no difference in the enrichment of selection in VIPs with versus without demonstrated pro- or anti-viral effects. However, we observed an even stronger significant elevation in the ratio of positive selection in these proviral and antiviral VIPs (Fig. 3b and Supplementary Table 5), but no elevation in this ratio in VIPs without experimental evidence of specific pro- or anti-viral effects (Fig. 3c and Supplementary Table 5). This is consistent with the expectation of viral interaction as the cause of enrichment of positive selection in VIPs in bats²⁶. We repeated this analysis using a dataset of 47 publicly available non-Myotis bat genomes, and confirmed these same patterns across bats more broadly, even when excluding Myotis genomes (Extended Data Fig. 5a).

Previous work has suggested that bats may have different physiological responses to DNA and RNA viruses²⁷. To determine whether this was reflected in genomic VIP adaptation, we compared the enrichment of positive selection in VIPs that interact only with DNA viruses (DNA-only VIPs) with those that interact only with RNA viruses (RNA-only VIPs). Notably, we found that VIP adaptation in Myotis and other bats is driven by selection only in DNA VIPs (Fig. 3e and Extended Data Fig. 5b) in contrast to the observed pattern in RNA VIPs, which show no evidence of genome-wide enrichment in adaptation (Fig. 3f and Extended Data Fig. 5c).

Adaptation to viral pathogens can also occur through gene copy-number changes^28,29,30. We tested whether VIPs were enriched among genes that were recently gained or lost throughout Myotis using a method described previously³¹ to test the cumulative per-gene, per-branch copy-number changes and birth–death rates of VIPs compared with other non-VIP genes. While the birth–death rates of VIP and non-VIP genes were comparable (P = 0.071; Extended Data Fig. 5d), VIP genes were significantly more likely to have undergone expansions and/or contractions on at least one branch of the Myotis family (P < 0.001; Fig. 3g). Furthermore, we find that this pattern is driven exclusively by copy-number changes in RNA-only VIPs, and not by changes in copy number of DNA-only VIPs (Fig. 3h,i). This suggests that, relative to the general variation in gene family birth–death rates across species, RNA VIPs are more dynamic across the Nearctic Myotis as a whole.

In contrast to what we observe in bats, VIP adaptation in humans is driven by positive selection in RNA—and not DNA—VIPs³². To investigate whether DNA VIP-driven adaptation in bats is exceptional among mammals, we replicated these analyses across four other large mammalian clades that are well represented among publicly available mammalian genomes: primates, Glires, Euungulata and Carnivora. We found that, while other mammalian orders show a mix of adaptation enrichments in both RNA and DNA VIPs, none show the absence of genome-wide enrichment of protein-coding adaptation in RNA VIPs observed in bats (Extended Data Fig. 5a–c). These results highlight that bats, including Myotis, exhibit unique modes of adaptation to DNA viruses compared to RNA viruses, in contrast to all other mammals.

Beyond their immune adaptations, bats are exceptional as the longest-lived clade of mammals after correcting for their body size¹, demonstrating an approximately 11× range of lifespans within an approximately 650× range of body sizes^8,33. Myotis species in particular exhibit the full dynamic range of bat lifespans within a narrow range of body sizes. While bats have been noted as an exception to the strong allometric scaling (positive correlation with body size) of lifespan otherwise seen across mammals and other metazoans, this exception has not been tested using phylogenetically corrected statistics leveraging well-resolved phylogenies.

To test the hypothesis of non-allometric scaling of lifespan in bats, we modelled the evolution of body size and lifespan independently across a supertree of over 1,000 mammals³⁴ (Fig. 4a,b, Extended Data Fig. 6a,b and Supplementary Table 6). We generally observed agreement between evolutionary patterns of body size and lifespan, such as in whales (Cetacea)³⁵, elephantids (Proboscidea)³⁶ and primates³⁷ (Fig. 4a,b). However, while only minor changes were observed in their body size, we observed some of the largest and most-rapid changes in lifespan across mammals in bats (Fig. 4a,b and Extended Data Fig. 6a,b and Supplementary Table 6). These changes are largely independent between species and genera, consistent with the theory of multiple independent increases in lifespan across bats. We used phylogenetically corrected generalized linear models and analysis of covariance to quantify the relationship between body size and lifespan across mammals. We find that bats experience a 40% greater increase in lifespan per 1% increase in body size compared with non-bat mammals (0.223% increase in lifespan versus 0.159% increase in lifespan per 1% increase in mass); however, these rates were not significantly different after phylogenetic correction (Extended Data Fig. 6c,d; phylogenetic analysis of covariance, P = 0.29). In Myotis, we saw many of the fastest increases in lifespan relative to their most recent ancestor, as measured by the change in their lifespan over the divergence time from their most recent ancestor (Δlifespan): M. brandtii (8.6 Δlifespan, 99th percentile), M. lucifugus (8.4 Δlifespan, 99th percentile), M. myotis (2.5 Δlifespan, 93nd percentile), Myotis grisescens (1.3 Δlifespan, 92nd percentile) and the Myotis common ancestor (1.7 Δlifespan, 89th percentile) (Fig. 4b, Extended Data Fig. 6b and Supplementary Table 6). Together, these results demonstrate that Myotis bats and their recent ancestors have evolved some of the most extreme increases in lifespan among mammals, despite similar allometric lifespan scaling of bats and mammals after phylogenetic correction.

Rapid changes in body size and lifespan can have major implications for the evolution of cancer risk and resistance across mammals³⁸. While lifetime cancer risk scales proportionally to both body size and lifespan within species^38,39, there is little to no correlation between body size, lifespan and cancer risk across species^39,40. This observation, known as Peto’s paradox⁴⁰, suggests that species with more cells or longer lifespans have adapted to reduce their cancer risks.

We hypothesized that the extreme changes in lifespan observed throughout the Myotis phylogeny should exert a selective pressure on genes associated with cancer processes. Using aBSREL⁴¹ to test for selection at terminal and internal branches within Myotis, we found that, per node, an average of 5.46% and 20.8% of protein-coding genes had at least one region under positive or negative selection, respectively (Supplementary Table 7). Genes under either positive or negative selection were enriched for several pathways in immunity, cancer and ageing, with many intersecting multiple of these processes, suggesting possible pleiotropic selective pressures (Supplementary Table 7).

We quantified the proportion of cancer-associated pathways⁴² over-represented among genes under positive selection throughout the phylogeny (Fig. 4c,d). Among genes under positive selection, we observed that most nodes within Nearctic Myotis were enriched for cancer hallmark pathways, especially at the recent ancestors of the longest-lived species (for example, M. lucifugus + M. occultus; Fig. 4c). Furthermore, when considering the union of all genes under positive selection in the evolutionary history of each species since their common ancestor, we observed significant enrichments in the representation of cancer-associated pathways in many of the lineages with the greatest cumulative increases in lifespan (M. lucifugus, M. occultus, M. evotis and M. yumanensis; Fig. 4d and Extended Data Fig. 6e).

The longest-lived bat in our study, M. lucifugus, had an over-representation of pathways specifically associated with DNA double-stranded break repair when looking at both lineage-wide and node-specific enrichments in positive selection using the Reactome database⁴³ (Fig. 4e and Supplementary Table 7). This includes 35 out of 65 genes in the high-fidelity ‘Homologous Recombination Repair’ Reactome pathway, and 21 out of 37 members of the ‘Homology-Directed Repair through Single-Strand Annealing’ Reactome pathway (Fig. 4e, Extended Data Fig. 6e and Supplementary Table 7). These results suggested that M. lucifugus might have an enhanced response to DNA double-stranded breaks relative to other bats. To test this hypothesis, we assessed the tolerance of M. lucifugus primary skin fibroblasts to neocarzinostatin (NCS), a potent radiomimetic agent that induces DNA double-stranded breaks⁴⁴ (Fig. 4f and Supplementary Table 8), compared with primary skin fibroblasts derived from M. evotis and three non-Myotis bats (Eidolon helvum, Pteropus rodricensis and Rousettus lanosus). At low doses of NCS, M. lucifugus was the only species with sensitivity to NCS after 24 h, with a drop in viability and concomitant increase in apoptosis. At high doses, M. lucifugus had the highest level of apoptosis and the greatest drop in viability of all the bats tested. This is consistent with other long-lived species, including elephants^45,46, naked mole rats⁴⁷ and bowhead whales⁴⁸, in which longevity is associated with an increased ability to clear out damaged cells.

To better understand the genes that may be driving this difference, we examined RNA expression in M. lucifugus cells after 6 h and 18 h of treatment with 100 nM NCS. At 6 h we observed upregulation of genes associated with cell cycle arrest (such as MDM2 and CDKN1A) and cell death (such as BAX), and downregulation of genes in pathways associated with cell division, growth and DNA damage repair and synthesis. At 18 h, we observed upregulation of genes in pathways associated with cellular stress response and continued downregulation of the cell cycle (Fig. 5a–d, Extended Data Fig. 7a and Supplementary Table 9). However, we observed that the strongest pathway enrichment in our DNA-damage RNA-sequencing data were genes associated with the innate immune response, including genes in the PI3K–AKT pathway (Fig. 5c,d and Supplementary Table 9). We hypothesized that the observed relationship between the unique DNA damage response of M. lucifugus may be due to pleiotropy with DNA VIPs. Indeed, we observed that there was a highly significant enrichment for genes that are differentially expressed after NCS treatment and DNA-only VIPs (P = 3.34 × 10⁻⁵, α = 0.0245; Fig. 5e and Extended Data Fig. 7b). The genes at the intersection of DNA-only VIPs and NCS response are over-represented in pathways associated with the cell cycle, senescence, transcription and DNA damage repair (Fig. 5f). We repeated these analyses considering only genes under positive selection in M. lucifugus. The intersection between DNA-only VIPs and NCS differentially expressed genes remained highly significant (P = 1.88 × 10⁻³; Extended Data Fig. 7c), with only the pathways cell cycle, DNA damage repair and deubiquitination enriched at false-discovery rate (FDR) ≤ 0.05 (Fig. 5g). DNA-only VIP genes, which are not NCS responsive, did not show these signatures (Extended Data Fig. 7d–g). Together these results suggest that selection on the innate immune response may drive agonistic pleiotropy in ageing-associated traits such as DNA damage response and cell cycle regulation.

In addition to evolving true flight, bats are notable for their long lifespan^8,33, stress tolerance^39,49 and viral tolerance^2,50. The genes and mechanisms underlying these highly complex and pleiotropic phenotypes can be challenging to identify, especially for rapidly evolving phenotypes, such as host–pathogen interactions. Here we identify patterns of adaptation contributing to longevity, cancer resistance and viral interactions in bats, and demonstrate a unique DNA damage response in primary cells of the long-lived M. lucifugus.

The evolution of body size and lifespan across mammals has major implications for the co-evolution of cancer risk and resistance. We find altered scaling of longevity in Myotis that is predicted to have serious consequences for their intrinsic, per-cell cancer risk. Similar to other systems in which the evolution of cancer resistance has been driven largely by rapid changes in body size^36,51,52, the rapid and repeated changes in lifespan across an order of magnitude observed in Myotis are hypothesized to result in strong selective pressures on lifetime cancer suppression to avoid malignancies^33,38,53. This is supported by recent meta-analyses on cancer risk in vertebrates^39,54 demonstrating little to no correlation between neoplasias and either species body size or lifespan, as well as by studies showing weak correlation or no correlation between mutation rates and either species body size or lifespan, respectively^55,56,57.

Bats are also important models for understanding infectious disease response and immune adaptation owing to their tolerance to viruses and role as zoonotic reservoirs^13,27,58. Here we show that, while bats have adapted to both DNA and RNA viruses, they have done so by different modes—protein evolution versus copy-number changes, respectively. This is in contrast to humans and other primates, in which protein coding adaptations to RNA viruses dominate^32,59. Moreover, we demonstrate complex patterns of structural variation in immune genes, including a segregating duplication of PKR, which encodes a major protein involved in the antiviral innate immune system. PKR has functional relevance in its activity against both DNA and RNA viruses²¹. We further demonstrate the functional implications of copy-number variation in PKR, including an additive effect on cellular effector functions.

Multiple hypotheses have been proposed to connect the particular physiology and ecology of bats with the evolution of notable adaptations such as viral infection tolerance and defence, stress tolerance and exceptional longevity³. Many theories of the driving causes of disease resistance and longevity in bats have been proposed, including the evolution of flight^1,60,61,62, the disposable soma hypothesis⁶³, metabolic state⁶⁴, torpor⁸ and other environmentally driven adaptations^65,66. Moreover, many studies have highlighted the intersection of these traits, including links between hibernation, DNA repair, longevity^8,58 and infectious disease resistance^8,27,58,67. Our results support the importance of agonistic pleiotropy in shaping bat evolution, whereby genetic adaptations for many specific traits (such as DNA virus innate immunity) may prove beneficial to other seemingly unrelated traits (such as cancer resistance, cellular homeostasis and longevity). Importantly, bats are also the only known mammal to have active DNA transposons⁶⁸, the suppression of which may also contribute to signatures of selection in DNA-interacting VIPs. Our findings on pervasive selection on DNA-only VIPs and the extreme evolution of longevity-associated in Myotis and other bats support a broader hypothesis that traits such as cancer risk, cellular homeostasis and antiviral response have evolved in tandem due to pleiotropic selection at coinciding points in the evolutionary history of bats.

Samples and materials

All sample collection efforts were performed according to approved animal use protocols by the University California, Berkeley and the University of Arizona. All samples collected in this study were collected under scientific collection permits from either the Arizona Game and Fish Department (SP405504, SP407155, SP403977 and SP407113) or the California Department of Fish and Wildlife (S-220230001-22025-001) (Supplementary Table 1). Bats were sampled using standard mist-netting procedures, including taking standard body measurements, following USGS recommendations for White-Nose Syndrome and COVID-19 prevention^71,72. For M. lucifugus, the donor individual was field-caught in California and transported to the Genetics Laboratory of the California Department of Fish and Wildlife, where they were euthanized by isofluorane. The M. velifer individual was caught in Arizona and euthanized in the field by isoflurane. For both M. lucifugus and M. velifer, tissues were collected and preserved through flash-freezing in liquid nitrogen. All other genomes were generated from primary cell lines of heterogametic individuals whenever possible to ensure assembly of all sex chromosomes. Additional information can be found in the Supplementary Information and in Supplementary Table 1. All cell lines were tested routinely for mycoplasma contamination using MycoStrip (InVivogen).

All PKR experiments were performed using HeLa PKR-KO cells (provided by A. Geballe)⁷³. The cells were maintained at 37 °C under 5% CO₂ and cultured in DMEM supplemented with 5% FBS, 1% penicillin–streptomycin mix and 1 μg ml⁻¹ puromycin (Sigma-Aldrich). All transfections were performed 24 h after seeding, using 3 µl of TransIT-LT1 Transfection Reagent (Mirus Bio) per 1 µg of DNA and Opti-MEM medium. We used previously generated pSG5-Flag×2 vectors encoding either M. myotis PKR1 (GenBank: OP006550), M. myotis PKR2 (GenBank: OP006559), M. velifer PKR1 (GenBank: OP006558) or M. velifer PKR2 (GenBank: OP006557)²¹. Plasmids encoding the interferon-stimulated gene ISG20 (ref. ⁷⁰) and a constitutively active variant of the sterile alpha-motif-domain-containing protein 9-like SAMD9L-F886Lfs*11 (referred to here as SAMD9L)⁶⁹ were used as controls in viral infections and cell translation experiments, respectively.

Near-complete genome assembly and annotation

Details on genome assembly, including DNA and RNA extraction, preparation of PacBio HiFi, Omni-C (Dovetail Genomics) and RNA-sequencing libraries, genome assembly, annotation and manual curation, are provided in the Supplementary Information.

Structural variation

To understand the genomic distribution of SVs, including segmental duplication events, we used SyRI (Synteny and Rearrangement Identifier¹⁹). After masking repetitive regions such as telomeres and centromeres, the primary 22 scaffolds corresponding to the autosomes of the Nearctic Myotis genomes were mapped to each other in the correct orientations using minimap2 (ref. ⁷⁴). We ran SyRI on the resulting files and plotted the results with plotsr⁷⁵.

Phylogenetics

A phylogeny of all 536 mammals in our alignments was generated using IQTREE⁷⁶ (v.2.3.1) using all gene alignments with the settings ‘-B 1000 -m GTR+F3x4+R6’. Gene trees were generated from gene alignments to exclude alignments with less than 50% gaps in the sequence and 4 or more species represented by using IQTREE with the settings ‘--wbtl --bnni --alrt 1000 -B 1000 --safe’. The best substitution models for each gene were saved as a NEXUS file. As the bootstrap values for this tree were unanimously 100%, we ran ASTRAL-IV⁷⁷ (v.1.24.4.7) using the settings ‘-t 54 -u 2 -C’ and confirmed that our phylogeny agreed with other previously published Eutherian phylogenies^78,79,80. The Chiroptera portion of our phylogeny was time calibrated using MCMCtree⁸¹ and PAML⁸² (v.4.10.0) with the bat-subset of our codon alignments and using fossil calibrations^{79,83,84,85,86,87,88,89,90,91} (Supplementary Table 2). We ran MCMCtree twice to generate the Hessian matrix and confirm convergence, and ran ten independent chains using the out.BV file from the first run. Finally, the output files of all ten chains were combined to compute the final divergence time estimates.

Ancestral body size and lifespan reconstruction

To examine how body size and lifespan have evolved over time in mammals, we used a super-phylogeny of mammal species³⁴ subsampled to contain only species with extant body size and lifespan data collected from AnAge¹⁵ and PanTHERIA⁹². Ancestral body sizes and lifespans were simulated separately using StableTraits⁹³ and the settings ‘--iterations 100000 --skip 1000 --chains 8’. Estimates were further validated by comparing our initial results to a second run of StableTraits using the same settings.

Selection scans and evolutionary rates

aBSREL

To conservatively test for branch-specific selection, we used aBSREL^41,94 (v.2.5.48) to test for selection at each branch within the Nearctic Myotis clade for 15,734 gene alignments spanning 536 mammals. These genes were identified as 1:1 orthologues across the full alignment, with no more than 50% sequence gaps and at least 4 species present in the alignment. For each gene alignment, the full phylogeny was trimmed to match the species present, and the HyPhy script label-tree.bf was used to highlight all nodes within Nearctic Myotis as foreground branches. aBSREL was then run using the pruned and highlighted tree and gene alignments on the command line with the options ‘--code Universal --branches Foreground’. We defined a gene as under selection if one or more regions within the gene demonstrated a selective pressure with an FDR-corrected P ≤ 0.05. A gene was specifically identified as under positive selection if at least one of these significant regions had an ω > 1.

BUSTED

To quantify the total amount of positive selection across the Myotis tree or the different species trees used in this Article, we used an improved version of the BUSTED^25,94 test called BUSTED-MH (details on BUSTED versus BUSTED-MH are provided in the Supplementary Information). We applied BUSTED-MH to 19,646 Myotis orthologous CDS alignments with at least five orthologues. These orthologues are cases in which the Orthofinder gene trees coincide with the species tree. This effectively removes issues regarding whether we should use the gene or the species tree, at the cost of removing 2,110 genes from the Myotis selection analysis. Similarly, we applied BUSTED-MH to 17,469 non-Myotis bat alignments with at least five orthologues. The species in these two sets (Myotis bats and non-Myotis bats) are mutually exclusive. This includes a subset of 14,091 alignments with orthologues present in two thirds of the non-Myotis bat species, regardless of the genes’ presence/absence in the Myotis-only complement. This is specifically used to show that patterns of virus-driven adaptation are representative of all bats. We also tested 17,890 primate alignments with at least five orthologues with BUSTED-MH, as well as 19,311 Glire, 18,000 Carnivora and 18,504 ungulate alignments. Similarly, HyPhy MEME (with --resample 100 to increase power) was run to identify individual codon sites subject to diversifying selection in the CD45 gene for visualization with host protein–virus contact sites.

Enrichment scores for VIPs

We expanded on the set of VIPs reported previously²⁶ (5,291 VIPs) by searching the literature for more recent publications reporting VIPs, with a final dataset of 5,527 VIPs. These VIPs (all VIPs) were further subcategorized into four groups on the basis of their interactions with either DNA viruses and/or RNA viruses: DNA VIPs, DNA-only VIPs, RNA VIPs and RNA-only VIPs (Supplementary Table 5). For each VIP set, we selected sets of matching control genes by controlling for 16 different. Sets of control genes were resampled in a bootstrap procedure to generate 95% confidence intervals for sets of genes across a range of P values; additional details are provided in the Supplementary Information.

Gene duplications and CAFE analysis

To quantify patterns of gene duplication and loss, we quantified the copy number of genes with human orthologues from our gene annotations for each Nearctic Myotis genome. To calculate per-gene expansion and loss rates and their statistical significance, we ran CAFE⁹⁵ (v.5) on the previously described set of copy-number counts using our time-calibrated species tree pruned to include only the nine Nearctic Myotis species. CAFE uses a birth-death model of gene family evolution to investigate changes in gene family size accounting for evolutionary history using the species phylogeny. Evolutionary branch length is incorporated into the model expectations, therefore accounting for potential bias resulting from differences in branch lengths. We ran CAFE on the subset of genes with two or more copies in at least one species using a Poisson distribution for the root frequency (-p), first generating an error model to correct for genome assembly and annotation error (-e). We compared the base model (each gene family belongs to the same evolutionary rate category) to gamma models (each gene family can belong to one of k evolutionary rate categories) with different values of k from 2 to 13. A final gamma model with k = 9 was chosen to balance model log likelihood with the number of gene families for which the optimizer failed. The model was run three separate times to ensure convergence.

To understand whether genes in these pathways have higher birth–death rates or are more likely to have significant changes in gene copy number compared with that expected relative to other genes, we evaluated the gene copy birth–death rate λ and the number of genes significantly expanded or contracted in copy number on at least one branch within our Nearctic Myotis phylogeny using a bootstrap procedure. Following ref. ³¹, we tested whether VIP genes in particular underwent significant copy-number changes or had significantly different birth-death rates than non-VIP genes. For each category of VIP genes (all VIPs, DNA VIPs, DNA-only VIPs, RNA VIPs and RNA-only VIPs), we generated 100 bootstrap sets of control non-VIP genes with the same number of genes as the corresponding VIP gene set. We ran CAFE on each set of VIP genes and the corresponding control non-VIP genes to infer per-gene birth–death rates and per-gene, per-branch expansion/loss events.

Assessment of DNA double-stranded break tolerance

We assessed each species’ tolerance to DNA double-stranded breaks by measuring viability, cytotoxicity and apoptosis across a range of doses of NCS, a radiomimetic drug. We measured dose–response curves in wing-derived primary dermal fibroblasts across five bat species (M. lucifugus, n = 8; M. evotis, n = 8; R. lanosus, n = 2; E. helvum, n = 2; P. rodricensis, n = 2) using the multiplexed ApoTox-Glo assay (Promega). Using 96-well plates, 2 individuals and 11 doses were assessed simultaneously with 4 technical replicates. The results were normalized to treatment controls for each individual in R (Code availability).

PKR1/2 characterization in PKR-KO HeLa cells

Western blot

Details on PKR western blotting are provided in the Supplementary Information.

Luciferase reporter assays

Details on PKR luciferase reporter assays are provided in the Supplementary Information.

PKR co-IP

HeLa PKR-KO cells were transfected with 1.25 µg plenti6 HA-tagged M. myotis PKR1 plasmid per million cells and 1.25 µg of either plenti6 MYC empty vector, MYC-tagged M. myotis PKR1 or MYC-tagged M. myotis PKR2 plasmid using TransIT-LTI transfection reagent (Mirus Bio). The next day, some wells were infected with Sindbis virus expressing GFP (SINV-GFP) at a multiplicity of infection (MOI) of 2 for 24 h. Cells were then scraped with cold PBS and pelleted. For the IP, cells were lysed in 500 µl IP buffer (50 mM Tris HCl pH 7.5, 140 mM NaCl, 6 mM MgCl₂, 0.1% NP40) supplemented with RNase (RiboLock, Fisher Scientific) and protease (complete EDTA-free protease inhibitor cocktail, Sigma-Aldrich) inhibitors for 10 min on ice, then centrifuged at 12,000g for 10 min at 4 °C. In total, 5% of the volume was kept for input, and the rest was incubated with 40 µl µMACS anti-MYC MicroBeads (Miltenyi Biotec) for 1 h at 4 °C with constant rotation. The samples were then loaded onto µMACS columns placed in the magnetic field of a µMACS Separator (Miltenyi Biotec), washed four times with cold IP buffer and eluted with the µMACS denaturing elution buffer. Proteins were denatured in elution buffer for 5 min at 95 °C, then loaded onto a 4–20% BioRad Criterion TGX Stain-Free precast gel and transferred onto an Amersham Protran nitrocellulose membrane (Sigma-Aldrich) for 1 h. Membranes were blocked for 1 h in 5% milk in PBS (Euromedex) supplemented with 0.2% Tween-20 (Thermo Fisher Scientific) and were incubated with mouse anti-MYC monoclonal antibodies (Abcam, 9E10, ab32), then secondary anti-mouse IgG antibodies conjugated with HRP (Sigma-Aldrich, A4416), or with rat anti-HA antibodies conjugated to HRP (Roche, Sigma-Aldrich, 12013819001). Images were taken on the Fusion FX imager (Vilber) with SuperSignal West Femto Chemiluminescent Substrate (Thermo Fisher Scientific).

Cell viability assay

HeLa PKR-KO cells were transfected 24 h after plating in 96 well plates (n = 10,000 cells per well), with 100 or 200 ng of pSG5 plasmid: empty or coding for M. myotis or M. velifer PKR1, PKR2 or PKR1 + 2 equal mix (50%–50%). At 24 h after transfection, positive control cells were treated with an apoptosis-inducing drug (etoposide) at different doses (250, 200 or 100 µM). Then, at 48 h after transfection, cells were collected and lysed to quantify the luminescence signal according to CellTiter-Glo Luminescent Cell Viability Assay (Promega) kit protocol.

VSV and SINV infections

VSV infections

In total, 200,000 cells were plated in 12-well plates and transfected 24 h after with 350 ng of pSG5 plasmid: empty; encoding M. myotis or M. velifer PKR1, PKR2 or equal input of PKR1 and PKR2 (175 ng per plasmid); or a plasmid encoding ISG20, due to its known antiviral activity against VSV as positive control⁷⁰. Cells were infected at 24 h after transfection with replicative VSV-GFP virus⁹⁶ at a MOI of 3. Cells were fixed with 4% paraformaldehyde at 16–18 h after infection. VSV infection was quantified by measuring the percentage of GFP-positive cell populations using the BD FACSCanto II Flow Cytometer (SFR BioSciences). Fold change results were normalized to the empty pSG5 condition across at least three independent experiments.

SINV infections

HeLa PKR-KO cells were transfected with 5 µg pSG5 empty vector, M. myotis PKR1, M. myotis PKR2 or 2.5 µg M. myotis PKR1 + 2.5 µg M. myotis PKR2 per million cells using TransIT-LTI transfection reagent (Mirus Bio). The next day, some wells were infected with SINV-GFP at MOI 0.2. Cells were then placed into the CellCyte X live cell imaging system (Cytena) and pictures of every well were taken every 2 h for 48 h. The fraction of GFP⁺ cells over the total cell area was measured and averaged from 6 photos of 2 individual wells per condition, and repeated for a total of 3 independent experiments.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

All sequencing data and genomes generated in this study are available under NCBI BioProject accessions PRJNA973719, PRJNA1035541 and PRJNA1295526. Annotations generated in this study are available at GitHub (https://github.com/docmanny/myotis-gene-annotations). All other data are archived on Dryad⁹⁷ (https://doi.org/10.5061/dryad.02v6wwqfh).

All custom code used herein is available at GitHub (https://github.com/sudmantlab/MyotisGenomeAssembly).

1. Austad, S. N. & Fischer, K. E. Mammalian aging, metabolism, and ecology: evidence from the bats and marsupials. J. Gerontol. 46, B47–B53 (1991).

   Article  CAS  PubMed  Google Scholar 

2. Irving, A. T., Ahn, M., Goh, G., Anderson, D. E. & Wang, L.-F. Lessons from the host defences of bats, a unique viral reservoir. Nature 589, 363–370 (2021).

   Article  ADS  CAS  PubMed  Google Scholar 

3. Cooper, L. N. et al. Bats as instructive animal models for studying longevity and aging. Ann. N. Y. Acad. Sci. 1541, 10–23 (2024).

   Article  ADS  PubMed  PubMed Central  Google Scholar 

4. Wilson, D. E. & Reeder, D. M. Mammal Species of the World: A Taxonomic and Geographic Reference (JHU Press, 2005).

5. Teeling, E. C. Bats (Chiroptera). In The Timetree of Life (eds Hedges, S.B. & Kumar, S.) 499–503 (Oxford Univ. Press, 2009).

6. Gunnell, G. F., Smith, R. & Smith, T. 33 million year old Myotis (Chiroptera, Vespertilionidae) and the rapid global radiation of modern bats. PLoS ONE 12, e0172621 (2017).

   Article  PubMed  PubMed Central  Google Scholar 

7. Morales, A. E., Ruedi, M., Field, K. & Carstens, B. C. Diversification rates have no effect on the convergent evolution of foraging strategies in the most speciose genus of bats, Myotis. Evolution 73, 2263–2280 (2019).

   Article  PubMed  Google Scholar 

8. Wilkinson, G. S. & Adams, D. M. Recurrent evolution of extreme longevity in bats. Biol. Lett. 15, 20180860 (2019).

   Article  PubMed  PubMed Central  Google Scholar 

9. Podlutsky, A. J., Khritankov, A. M., Ovodov, N. D. & Austad, S. N. A new field record for bat longevity. J. Gerontol. A 60, 1366–1368 (2005).

   Article  Google Scholar 

10. Wilson, D. E. & Tyson, E. L. Longevity records for Artibeus Jamaicensis and Myotis Nigricans. J. Mammal. 51, 203 (1970).

    Article  Google Scholar 

11. Ruedi, M. et al. Molecular phylogenetic reconstructions identify East Asia as the cradle for the evolution of the cosmopolitan genus Myotis (Mammalia, Chiroptera). Mol. Phylogenet. Evol. 69, 437–449 (2013).

    Article  PubMed  Google Scholar 

12. He, X. et al. Establishment of myotis myotis cell lines—model for investigation of host-pathogen interaction in a natural host for emerging viruses. PLoS ONE 9, e109795 (2014).

    Article  ADS  PubMed  PubMed Central  Google Scholar 

13. Guth, S. et al. Bats host the most virulent-but not the most dangerous-zoonotic viruses. Proc. Natl Acad. Sci. USA 119, e2113628119 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

14. Stadelmann, B., Lin, L.-K., Kunz, T. H. & Ruedi, M. Molecular phylogeny of new world Myotis (Chiroptera, Vespertilionidae) inferred from mitochondrial and nuclear DNA genes. Mol. Phylogenet. Evol. 43, 32–48 (2007).

    Article  CAS  PubMed  Google Scholar 

15. Tacutu, R. et al. Human ageing genomic resources: integrated databases and tools for the biology and genetics of ageing. Nucleic Acids Res. 41, D1027–D1033 (2013).

    Article  CAS  PubMed  Google Scholar 

16. Sotero-Caio, C. G., Baker, R. J. & Volleth, M. Chromosomal evolution in Chiroptera. Genes 8, 272 (2017).

    Article  PubMed  PubMed Central  Google Scholar 

17. Manni, M., Berkeley, M. R., Seppey, M. & Zdobnov, E. M. BUSCO: assessing genomic data quality and beyond. Curr. Protoc. 1, e323 (2021).

    Article  PubMed  Google Scholar 

18. Foley, N. M. et al. Karyotypic stasis and swarming influenced the evolution of viral tolerance in a species-rich bat radiation. Cell Genom. 4, 100482 (2024).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

19. Goel, M., Sun, H., Jiao, W.-B. & Schneeberger, K. SyRI: finding genomic rearrangements and local sequence differences from whole-genome assemblies. Genome Biol. 20, 277 (2019).

    Article  PubMed  PubMed Central  Google Scholar 

20. de Sotero-Caio, C. G. et al. Centromeric enrichment of LINE-1 retrotransposons and its significance for the chromosome evolution of Phyllostomid bats. Chromosome Res. 25, 313–325 (2017).

    Article  PubMed  Google Scholar 

21. Jacquet, S. et al. Adaptive duplication and genetic diversification of protein kinase R contribute to the specificity of bat-virus interactions. Sci. Adv. 8, eadd7540 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

22. Morel, B., Kozlov, A. M., Stamatakis, A. & Szöllősi, G. J. GeneRax: a Tool for species-tree-aware maximum likelihood-based gene family tree inference under gene duplication, transfer, and loss. Mol. Biol. Evol. 37, 2763–2774 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

23. García, M. A. et al. Impact of protein kinase PKR in cell biology: from antiviral to antiproliferative action. Microbiol. Mol. Biol. Rev. 70, 1032–1060 (2006).

    Article  PubMed  PubMed Central  Google Scholar 

24. Enard, D., Cai, L., Gwennap, C. & Petrov, D. A. Viruses are a dominant driver of protein adaptation in mammals. eLife 5, e12469 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

25. Murrell, B. et al. Gene-wide identification of episodic selection. Mol. Biol. Evol. 32, 1365–1371 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

26. Souilmi, Y. et al. An ancient viral epidemic involving host coronavirus interacting genes more than 20,000 years ago in East Asia. Curr. Biol. 31, 3704 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

27. Brook, C. E. & Dobson, A. P. Bats as ‘special’ reservoirs for emerging zoonotic pathogens. Trends Microbiol. 23, 172–180 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

28. Kondrashov, F. A. in Evolution after Gene Duplication (eds Dittmar, K. & Liberles, D.) 57–76 (John Wiley & Sons, 2011).

29. Rastogi, S. & Liberles, D. A. Subfunctionalization of duplicated genes as a transition state to neofunctionalization. BMC Evol. Biol. 5, 28 (2005).

    Article  PubMed  PubMed Central  Google Scholar 

30. Assis, R. & Bachtrog, D. Neofunctionalization of young duplicate genes in Drosophila. Proc. Natl Acad. Sci. USA 110, 17409–17414 (2013).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

31. Huang, Z. et al. Duplications of human longevity-associated genes across placental mammals. Genome Biol. Evol. 15, evad186 (2023).

    Article  PubMed  PubMed Central  Google Scholar 

32. Enard, D. & Petrov, D. A. Evidence that RNA viruses drove adaptive introgression between Neanderthals and modern humans. Cell 175, 360–371 (2018).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

33. Brunet-Rossinni, A. K. & Austad, S. N. Ageing studies on bats: a review. Biogerontology 5, 211–222 (2004).

    Article  CAS  PubMed  Google Scholar 

34. Upham, N. S., Esselstyn, J. A. & Jetz, W. Inferring the mammal tree: species-level sets of phylogenies for questions in ecology, evolution, and conservation. PLoS Biol. 17, e3000494 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

35. Pyenson, N. D. & Sponberg, S. N. Reconstructing body size in extinct crown Cetacea (neoceti) using allometry, phylogenetic methods and tests from the fossil record. J. Mamm. Evol. 18, 269–288 (2011).

    Article  Google Scholar 

36. Vazquez, J. M. & Lynch, V. J. Pervasive duplication of tumor suppressors in Afrotherians during the evolution of large bodies and reduced cancer risk. eLife 10, e65041 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

37. Tillquist, R. C., Shoemaker, L. G., Knight, K. B. & Clauset, A. The evolution of primate body size: left-skewness, maximum size, and Cope’s Rule. Preprint at bioRxiv https://doi.org/10.1101/092866 (2016).

38. Nunney, L. Commentary: the multistage model of carcinogenesis, Peto’s paradox and evolution. Int. J. Epidemiol. 45, 649–653 (2016).

    Article  PubMed  Google Scholar 

39. Vincze, O. et al. Cancer risk across mammals. Nature 601, 263–267 (2022).

    Article  ADS  CAS  PubMed  Google Scholar 

40. Peto, R. Quantitative implications of the approximate irrelevance of mammalian body size and lifespan to lifelong cancer risk. Philos. Trans. R. Soc. Lond. B 370, 20150198 (2015).

    Article  Google Scholar 

41. Smith, M. D. et al. Less is more: an adaptive branch-site random effects model for efficient detection of episodic diversifying selection. Mol. Biol. Evol. 32, 1342–1353 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

42. Hanahan, D. Hallmarks of cancer: new dimensions. Cancer Discov. 12, 31–46 (2022).

    Article  CAS  PubMed  Google Scholar 

43. Milacic, M. et al. The reactome pathway knowledgebase 2024. Nucleic Acids Res. 52, D672–D678 (2024).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

44. Ohtsuki, K. & Ono, Y. in Neocarzinostatin (eds Maeda, H. et al.) 129–154 (Springer, 1997).

45. Vazquez, J. M., Sulak, M., Chigurupati, S. & Lynch, V. J. A zombie LIF gene in elephants is upregulated by TP53 to induce apoptosis in response to DNA damage. Cell Rep. 24, 1765–1776 (2018).

    Article  CAS  PubMed  Google Scholar 

46. Sulak, M. et al. TP53 copy number expansion is associated with the evolution of increased body size and an enhanced DNA damage response in elephants. eLife 5, e11994 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

47. MacRae, S. L. et al. DNA repair in species with extreme lifespan differences. Aging 7, 1171–1184 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

48. Firsanov, D. et al. Evidence for improved DNA repair in the long-lived bowhead whale. Nature 648, 717–725 (2025).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

49. Chionh, Y. T. et al. High basal heat-shock protein expression in bats confers resistance to cellular heat/oxidative stress. Cell Stress Chaperones 24, 835–849 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

50. Hayman, D. T. S. et al. Ecology of zoonotic infectious diseases in bats: current knowledge and future directions. Zoonoses Publ. Health 60, 2–21 (2013).

    Article  CAS  Google Scholar 

51. Tollis, M. et al. Return to the sea, get huge, beat cancer: an analysis of cetacean genomes including an assembly for the humpback whale (Megaptera novaeangliae). Mol. Biol. Evol. 36, 1746–1763 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

52. Nair, N. U. et al. Cross-species identification of cancer resistance-associated genes that may mediate human cancer risk. Sci. Adv. 8, eabj7176 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

53. Caulin, A. F., Graham, T. A., Wang, L.-S. & Maley, C. C. Solutions to Peto’s paradox revealed by mathematical modelling and cross-species cancer gene analysis. Philos. Trans. R. Soc. Lond. B 370, 20140222 (2015).

    Article  Google Scholar 

54. Compton, Z. T. et al. Cancer prevalence across vertebrates. Cancer Discov. https://doi.org/10.1158/2159-8290.CD-24-0573 (2024).

55. Beichman, A. C., Zhu, L. & Harris, K. The evolutionary interplay of somatic and germline mutation rates. Annu. Rev. Biomed. Data Sci. 7, 83–105 (2024).

    Article  PubMed  PubMed Central  Google Scholar 

56. Cagan, A. et al. Somatic mutation rates scale with lifespan across mammals. Nature 604, 517–524 (2022).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

57. Bergeron, L. A. et al. Evolution of the germline mutation rate across vertebrates. Nature 615, 285–291 (2023).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

58. Morales, A. E. et al. Bat genomes illuminate adaptations to viral tolerance and disease resistance. Nature 638, 449–458 (2025).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

59. Enard, D. & Petrov, D. A. Ancient RNA virus epidemics through the lens of recent adaptation in human genomes. Philos. Trans. R. Soc. Lond. B 375, 20190575 (2020).

    Article  CAS  Google Scholar 

60. O’Shea, T. J. et al. Bat flight and zoonotic viruses. Emerg. Infect. Dis. 20, 741–745 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

61. Levesque, D. L., Boyles, J. G., Downs, C. J. & Breit, A. M. High body temperature is an unlikely cause of high viral tolerance in bats. J. Wildl. Dis. 57, 238–241 (2021).

    Article  PubMed  Google Scholar 

62. Matsuda, Y. & Makino, T. Comparative genomics reveals convergent signals associated with the high metabolism and longevity in birds and bats. Proc. Biol. Sci. 291, 20241068 (2024).

    CAS  PubMed  PubMed Central  Google Scholar 

63. Kirkwood, T. B. L. The Disposable Soma Theory. In The Evolution of Senescence in the Tree of Life (eds Shefferson, R. P. et al.) 23–39 (Cambridge Univ. Press, 2017).

64. Toshkova, N. et al. Temperature sensitivity of bat antibodies links metabolic state of bats with antigen-recognition diversity. Nat. Commun. 15, 5878 (2024).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

65. Gutiérrez-Guerrero, Y. T. et al. Genomic consequences of dietary diversification and parallel evolution due to nectarivory in leaf-nosed bats. Gigascience 9, giaa059 (2020).

    Article  PubMed  PubMed Central  Google Scholar 

66. Pei, G., Balkema-Buschmann, A. & Dorhoi, A. Disease tolerance as immune defense strategy in bats: one size fits all? PLoS Pathog. 20, e1012471 (2024).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

67. Wang, X. et al. Myotis bat STING attenuates aging-related inflammation in female mice. Zool. Res. 45, 961–971 (2024).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

68. Ray, D. A. et al. Multiple waves of recent DNA transposon activity in the bat, Myotis Lucifugus. Genome Res. 18, 717–728 (2008).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

69. Legrand, A. et al. SAMD9L acts as an antiviral factor against HIV-1 and primate lentiviruses by restricting viral and cellular translation. PLoS Biol. 22, e3002696 (2024).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

70. Wu, N. et al. The interferon stimulated gene 20 protein (ISG20) is an innate defense antiviral factor that discriminates self versus non-self translation. PLoS Pathog. 15, e1008093 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

71. Sleeman J. NWHC Operations During the COVID-19 Pandemic and Information About Coronaviruses in Wildlife (Canadian Wildlife Health Cooperative, 2020).

72. White-Nose Syndrome Disease Management Working Group. National white-nose syndrome decontamination protocol - March 2024. USDA US Forest Service https://www.fs.usda.gov/sites/nfs/files/legacy-media/r04/WNS%20Decon%20Protocol-revised%20March%202024.pdf (2024).

73. Child S. J. et al. Antagonism of the protein kinase R pathway in human cells by rhesus cytomegalovirus. J. Virol. 92, e01793-17 (2017).

74. Li, H. New strategies to improve minimap2 alignment accuracy. Bioinformatics 37, 4572–4574 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

75. Goel, M. & Schneeberger, K. plotsr: visualizing structural similarities and rearrangements between multiple genomes. Bioinformatics 38, 2922–2926 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

76. Minh, B. Q. et al. IQ-TREE 2: new models and efficient methods for phylogenetic inference in the genomic era. Mol. Biol. Evol. 37, 1530–1534 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

77. Zhang, C. & Mirarab, S. Weighting by gene tree uncertainty improves accuracy of quartet-based species trees. Mol. Biol. Evol. 39, msac215 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

78. Romiguier, J., Ranwez, V., Delsuc, F., Galtier, N. & Douzery, E. J. P. Less is more in mammalian phylogenomics: AT-rich genes minimize tree conflicts and unravel the root of placental mammals. Mol. Biol. Evol. 30, 2134–2144 (2013).

    Article  CAS  PubMed  Google Scholar 

79. Foley, N. M. et al. A genomic timescale for placental mammal evolution. Science 380, eabl8189 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

80. Korstian, J. M., Paulat, N. S., Platt, R. N., Stevens, R. D. & Ray, D. A. SINE-based phylogenomics reveal extensive introgression and incomplete lineage sorting in Myotis. Genes 13, 399 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

81. Dos Reis, M. & Yang, Z. Bayesian molecular clock dating using genome-scale datasets. Methods Mol. Biol. 1910, 309–330 (2019).

    Article  PubMed  Google Scholar 

82. Yang, Z. PAML 4: phylogenetic analysis by maximum likelihood. Mol. Biol. Evol. 24, 1586–1591 (2007).

    Article  CAS  PubMed  Google Scholar 

83. Jebb, D. et al. Six reference-quality genomes reveal evolution of bat adaptations. Nature 583, 578–584 (2020).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

84. Rietbergen, T. B. et al. The oldest known bat skeletons and their implications for Eocene chiropteran diversification. PLoS ONE 18, e0283505 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

85. Phillips, M. J. Geomolecular dating and the origin of placental mammals. Syst. Biol. 65, 546–557 (2016).

    Article  PubMed  Google Scholar 

86. Gunnell, G. F. & Simmons, N. B. Fossil evidence and the origin of bats. J. Mamm. Evol. 12, 209–246 (2005).

    Article  Google Scholar 

87. Eiting, T. P. & Gunnell, G. F. Global completeness of the bat fossil record. J. Mamm. Evol. 16, 151–173 (2009).

    Article  Google Scholar 

88. Storch, G., Sigé, B. & Habersetzer, J. Tachypteron franzeni n. gen., n. sp., earliest emballonurid bat from the Middle Eocene of Messel (Mammalia, Chiroptera). Palaontol. Z. 76, 189–199 (2002).

    Article  Google Scholar 

89. Ravel, A. et al. A new large philisid (Mammalia, Chiroptera, Vespertilionoidea) from the late Early Eocene of Chambi, Tunisia. Palaeontology 55, 1035–1041 (2012).

    Article  Google Scholar 

90. Lim, B. K. Review of the origins and biogeography of bats in South America. Chiropt. Neotrop. 15, 391–410 (2009).

    Google Scholar 

91. Morgan, G. S. & Czaplewski, N. J. A new bat (Chiroptera: Natalidae) from the Early Miocene of Florida, with comments on natalid phylogeny. J. Mammal. 84, 729–752 (2003).

    Article  Google Scholar 

92. Jones, K. E. et al. PanTHERIA: a species-level database of life history, ecology, and geography of extant and recently extinct mammals. Ecology 90, 2648–2648 (2009).

    Article  Google Scholar 

93. Elliot, M. G. & Mooers, A. Ø. Inferring ancestral states without assuming neutrality or gradualism using a stable model of continuous character evolution. BMC Evol. Biol. 14, 226 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

94. Kosakovsky Pond, S. L. et al. HyPhy 2.5-A customizable platform for evolutionary hypothesis testing using PHYlogenies. Mol. Biol. Evol. 37, 295–299 (2020).

    Article  PubMed  PubMed Central  Google Scholar 

95. Mendes, F. K., Vanderpool, D., Fulton, B. & Hahn, M. W. CAFE 5 models variation in evolutionary rates among gene families. Bioinformatics 36, 5516–5518 (2021).

    Article  PubMed  Google Scholar 

96. Ostertag, D., Hoblitzell-Ostertag, T. M. & Perrault, J. Overproduction of double-stranded RNA in vesicular stomatitis virus-infected cells activates a constitutive cell-type-specific antiviral response. J. Virol. 81, 503–513 (2007).

    Article  CAS  PubMed  Google Scholar 

97. Lauterbur, M.E. et al. Data for ‘Insights into longevity and virus-driven adaptation from Myotis bat genomes’. Dryad https://doi.org/10.5061/dryad.02v6wwqfh (2026).

Download references

Author notes

1. Juan M. Vazquez

   Present address: Department of Biology, Pennsylvania State University, State College, PA, USA

2. M. Elise Lauterbur

   Present address: Department of Biology, University of Vermont, Burlington, VT, USA

3. Saba Mottaghinia

   Present address: Max Delbrück Center for Molecular Medicine in the Helmholtz Association, Berlin, Germany

4. Devaughn Fraser

   Present address: Wildlife Diversity Program, Wildlife Division, Connecticut Department of Energy and Environmental Protection, Burlington, CT, USA

5. William Thomas

   Present address: Colossal Biosciences, Dallas, TX, USA

6. These authors contributed equally: Juan M. Vazquez, M. Elise Lauterbur

7. These authors jointly supervised this work: Lucie Etienne, David Enard, Peter H. Sudmant

Authors and Affiliations

1. Department of Integrative Biology, University of California, Berkeley, Berkeley, CA, USA

   Juan M. Vazquez, Dakota McMillan, Stacy Li, Jose Pablo Vazquez-Medina & Peter H. Sudmant

2. Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, AZ, USA

   M. Elise Lauterbur, Melanie Bucci, Genavieve Gray-Sandoval & David Enard

3. Centre International de Recherche en Infectiologie (CIRI), Inserm U1111, Université Claude Bernard Lyon 1, CNRS UMR5308, Ecole Normale Supérieure ENS de Lyon, Université de Lyon, Lyon, France

   Saba Mottaghinia, Sarah Maesen, Amandine Le Corf, Clara Loyer, Carine Rey & Lucie Etienne

4. Université de Strasbourg, Architecture et Réactivité de l’ARN, Institut de Biologie Moléculaire et Cellulaire du CNRS, Strasbourg, France

   Léa Gaucherand & Sébastien Pfeffer

5. Department of Molecular and Cellular Biology, University of California, Berkeley, Berkeley, CA, USA

   Michael Singer, Sarah Villa & Melissa Sui

6. Wildlife Genetics Research Unit, Wildlife Health Laboratory, California Department of Fish and Wildlife, Sacramento, CA, USA

   Devaughn Fraser, Samantha L. R. Capel & Michael Buchalski

7. Department of Biology, California State Polytechnic University, Humboldt, Arcata, CA, USA

   Zeinab R. Haidar

8. Department of Pathology and Clinical Laboratories, University of Michigan, Ann Arbor, MI, USA

   Melissa Han, William Kohler & Richard Miller

9. Department of Biological Sciences, Smith College, Northampton, MA, USA

   Tanya M. Lama

10. Department of Science and Biotechnology, Berkeley City College, Berkeley, CA, USA

    Dakota McMillan

11. Center for Computational Biology, University of California, Berkeley, Berkeley, CA, USA

    Stacy Li & Johnathan Lo

12. Bat Conservation International, Austin, TX, USA

    Kathleen Slocum

13. Department of Plant and Microbial Biology, University of California, Berkeley, Berkeley, CA, USA

    Melissa Sui & Rachel Brem

14. Department of Ecology and Evolution, Stony Brook University, Stony Brook, NY, USA

    William Thomas

15. Bat Survey Solutions, Tucson, AZ, USA

    Janet Debelak Tyburec

Contributions

D.E. and P.H.S. contributed equally. Conceptualization: J.M.V., M.E.L., D.E., P.H.S., L.E., D.F. and M. Buchalski Methodology: J.M.V., M.E.L., D.E., P.H.S., L.E. and G.G.-S. Software: J.M.V., M.E.L., D.E., P.H.S., L.E. and T.M.L. Validation: J.M.V., M.E.L., D.E., P.H.S., L.E. and J.P.V.-M. Formal analysis: J.M.V., M.E.L., D.E., P.H.S., L.E., S.M., A.L.C., D.M. and S.V. Investigation: J.M.V., M.E.L., S. Mottaghinia, M. Bucci, D.F., G.G.-S., L.G., Z.R.H., M.H., W.K., T.M.L., A.L.C., C.L. S. Maesen, D.M., S.L., J.L., C.R., S.L.R.C., M. Singer, K.S., W.T., J.D.T., S.V., M. Sui, R.B., R.M., M. Buchalski, J.P.V.-M., S.P., L.E., D.E. and P.H.S. Resources: J.M.V., M.E.L., D.E., P.H.S., L.E., M. Bucci, D.F., G.G.-S., L.G., Z.R.H., M.H., W.K., T.M.L., A.L.C., S. Mottaghinia, S. Maesen, D.M., S.L., J.L., C.R., S.L.R.C., M. Singer, K.S., W.T., J.D.T., S.V., R.M. and M. Buchalski Data curation: J.M.V., M.E.L., S. Mottaghinia, M. Bucci, D.F., G.G.-S., L.G., Z.R.H., M.H., W.K., T.M.L., A.L.C., S. Maesen, D.M., S.L., J.L., C.R., S.L.R.C., M. Singer, K.S., W.T., J.D.T., S.V., R.M., M. Buchalski, J.P.V.-M., S.P., L.E., D.E. and P.H.S. Writing  (original draft): J.M.V., M.E.L., D.E., P.H.S., L.E., S. Mottaghinia and J.L. Writing  (review and editing): J.M.V., M.E.L., S. Mottaghinia, M. Bucci, D.F., G.G.-S., L.G., Z.R.H., M.H., W.K., T.M.L., A.L.C., S. Maesen, D.M., S.L., J.L., C.R., S.L.R.C., M. Singer, K.S., W.T., J.D.T., S.V., M. Sui, R.B., R.M., M. Buchalski, J.P.V.-M., S.P., L.E., D.E. and P.H.S. Visualization: J.M.V., M.E.L., D.E., P.H.S., L.E., S. Maesen and A.L.C. Supervision: J.M.V., M.E.L., D.E., P.H.S. and L.E. Project administration: J.M.V., M.E.L., D.E., P.H.S. and L.E. Funding acquisition: J.M.V., M.E.L., D.E., P.H.S. and L.E. As co-contributors, M.E.L. and J.M.V. reserve the right to order their names interchangeably when citing this work.

Corresponding authors

Correspondence to Lucie Etienne, David Enard or Peter H. Sudmant.

Competing interests

The authors declare no competing interests.

Peer review information

Nature thanks the anonymous reviewers for their contribution to the peer review of this work. Peer reviewer reports are available.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions