GoKawiilSamples 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 prevention71,72. 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% CO 2 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 phylogenies78,79,80. 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 calibrations79,83,84,85,86,87,88,89,90,91 (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
... continue reading