Introduction

Comparative genomics provides a framework for identifying the molecular mechanisms underlying unique organismal adaptations, in their endless forms. To date, comparative genomic approaches have revealed the mechanisms underlying terrestrial adaptations in mudskipper fish (You et al., 2014), heat tolerance in coral (Bay, Rose, Logan, & Palumbi, 2017), cold stress tolerance in Draba (Nowak et al., 2020), and extreme longevity in naked mole rats (X. Zhou et al., 2020). In most cases the search for molecular adaptations has focused on orthologous single-copy genes, but gene loss and duplication can also be adaptive and are critical to understanding of how phenotypic adaptations evolve. Analyses based on highly contiguous genome assemblies have uncovered gene expansions likely associated with production of urushiol and anthocyanins in mango (P. Wang et al., 2020), the earliest events of gene duplication in cytoskeletal and membrane-trafficking families in eukaryotic cellular evolution (Vosseberg et al., 2020), pseudogenization in genes associated with testicular descent in afrotherian mammals (Sharma, Lehmann, Stuckas, Funke, & Hiller, 2018), gene losses associated with diving–related adaptations in cetaceans (Huelsmann et al., 2019), and losses associated with physiological and metabolic adaptations in fruit bats (Sharma, Hecker, Roscito, Foerster, Langer & Hiller, 2018). Given the importance of gene family evolution, multiple large-scale genome sequencing consortia such as the Earth BioGenome Project (Lewin et al., 2018), the Vertebrate Genomes Project (Rhie et al., 2020), and Bat1K (Teeling et al., 2018) aim to generate high-quality genome assemblies for species spanning entire clades and even the entire phylogenetic ‘Tree of Life’, thereby enabling greater confidence in analyses of gene loss and gene family evolution.
Gene family expansions and contractions are influenced by selection, including from biological factors such as pathogens. Host-pathogen interactions are shaped by reciprocal selection, an evolutionary arms race which has forced hosts to evolve complex immune defense mechanisms (Papkou et al., 2019; Sironi, Cagliani, Forni, & Clerici, 2015). Vertebrates have two types of immune response: innate immunity, which is non-specific and acts as a first line of defense; and adaptive immunity, which is highly specific and generates immune memory (Delves, Martin, Burton, & Roitt, 2017; Janeway & Travers 2001.). Several immune-related gene families that have experienced substantial evolutionary changes during mammal evolution. While many important facets of the immune system are conserved, immune gene families have high rates of evolution whether measured via substitution rate ratios or birth–death turnover (Bernatchez & Landry, 2003; Goebel et al., 2017; Minias, Pikus, Whittingham, & Dunn, 2019; Santos et al., 2016; Shultz & Sackton, 2019; Van Oosterhout, 2009). This is especially true of the Major Histocompatibility Complex (MHC), which is responsible for generating cell surface proteins that play essential functions in the adaptive immune system (Janeway & Travers 2001).
This combination of highly conserved, and highly variable components of the immune system, is particularly intriguing among bats. Among mammals, bat diversity is second only to that of rodents, and encompasses over 1,400 species that occupy a broad diversity of ecological niches on six continents (Fenton & Simmons, 2015; Nogueira et al., 2018). The success of bats is likely related to a suite of adaptations unique both to the clade as a whole and to various subclades within the Order Chiroptera. The most obvious of these is powered flight, allowing bats to occupy a unique aerial niche not utilized by any other mammal. While this unique niche limits body size, within that constraint bats have been exceptionally successful and have diversified in ways unparalleled among other mammals. For example, bats evolved virtually every mammalian dietary strategy (e.g., frugivory, carnivory, nectarivory, piscivory) and have done so in a relatively short evolutionary time frame (Dumont et al., 2012). Another less obvious but likely more interesting adaptation is the exceptional longevity and increased health span (the period of life during which an organism is in generally good health) exhibited by many bat species given their body size. Many species such as the Bechstein’s bat (Myotis bechstein) the little brown bat, Brandt’s bat (Myotis brandtii ), greater mouse-eared bat (Myotis myotis ) and greater horseshoe bat (Rhinolophus ferrumequinum ) have unexpectedly long health spans, living 30 - 40 years (Fleischer, Gampe, Scheuerlein & Kerth, 2017; Foley et al., 2018; Podlustsky, Khritankov, Ovodov & Austad, 2005; Seim et al., 2013; Wilkinson & Adams, 2019). Such longevity defies the expectation that large species are longer-lived than small species; despite constrained body size, bats live longer than other mammals of similar size (Austad & Fischer, 1991; Healy et al., 2014). Bat longevity and health span may be influenced by their exposure to extrinsic mortality factors. Powered, mostly nocturnal flight may lower bats’ exposure to some sources of extrinsic mortality, including predation (Healy et al., 2014). However, bats’ propensity to roost in large colonies also increases their exposure to another source of mortality – infections (Brook & Dobson, 2015; H. Han et al., 2015).Thus, to achieve such longevity and decreased senescence, long–lived bat populations must overcome the burden of infectious diseases.
The uniqueness of bats extends to the immune repertoire. Early in the age of whole-genome analyses, it was clear that inflammation-related gene families had expanded or contracted, and certain single–copy genes associated with immunity and cell repair had experienced selection in bats (G. Zhang et al., 2013). There is still debate as to whether bats harbor a disproportionately large number of viruses (Moratelli & Calisher, 2015; Olival et al., 2017), but there is no doubt that several recent viral intrusions into our own species ultimately originated from bat hosts (Drexler et al., 2012; Goldstein et al., 2018; Hu et al., 2017; Memish, Perlman, Van Kerkhove, & Zumla, 2020; Towner et al., 2007). This likely includes the current SARS-CoV-2 pandemic (Boni et al., 2020; Lau et al., 2020). Bats appear to have the ability to tolerate these viruses with few health impacts, hence recent studies have focused on bat comparative genomics (Jebb et al., 2020) and its emphasis on viral response (reviewed in: Gorbunova, Seluanov, & Kennedy, 2020; Hayman, 2019). Although little is known from this perspective, there is a growing body of functional analyses showing that bats are unusual among mammals in how they deal with viruses (Ahn et al., 2019; A. Banerjee et al., 2020; Miller et al., 2016; Schountz, Baker, Butler, & Munster, 2017; Xie et al., 2018).
The ‘inflammosome’ is typically highly conserved across mammals, but bats exhibit a reduced inflammatory response that may be tied to their ability to cope with viral infection while experiencing minimal impact (Pavlovich et al., 2018). For example, the PYHIN gene family, namely, appears to have been almost completely lost in bats (Ahn, Cui, Irving, & Wang, 2016; G. Zhang et al., 2013) while at least one PYHIN gene can be found in all other eutherians examined. Similarly, in bats, the inflammatory function of interferons (G. Zhang et al., 2013) appears distinct and the APOBEC3 repertoire, which is associated with anti-viral response, is expanded (Jebb et al., 2020; Hayward et al., 2018). All of these functional patterns suggest an overall dampened inflammatory reaction despite a robust immune response to viruses whose origins may lie in the gene repertoires available to bats (A. Banerjee, Rapin, Bollinger, & Misra, 2017; A. Banerjee et al., 2020).
Gene family evolution also likely plays a role in the unique dietary ecology of bats. Several studies have found a variety of mechanisms influencing dietary adaptation. For example, convergent amino acid substitutions in several lineages of frugivorous bats have occurred independently (Gutiérrez-Guerrero et al., 2020; Shen, Han, Zhang, Rossiter, & Zhang, 2012; Teeling et al., 2018; K. Wang et al., 2020), and are associated with the shift to a high-sugar diet. Another strategy has been to repurpose a given gene to accommodate such dietary shifts. (Shen, Han, Jones, Rossiter, & Zhang, 2013). With the exception of olfactory receptors (Hayden et al., 2014; Hughes et al., 2018; Tsagkogeorga, Müller, Dessimoz, & Rossiter, 2017), the roles of gene loss and gain in shaping dietary evolution of bats have not been comprehensively explored.
Here we investigate bat gene family evolution related to immunity, metabolism, and dietary adaptations, using the most extensive genomic sampling within bats to date. Despite variability in quality of assemblies, the ecological diversity of lineages for which assemblies are available allows, for the first time, an investigation of gene family evolution across 10 families, two suborders, and a complete coverage of the entire range of diets. We find two major patterns. First, system-wide gene losses related to inflammatory response and selection on genes associated with antiviral immunity appear to have influenced bat lineages. This suggests that bats— compared to other mammals such as cow, dog, horse, pig, mouse and human— have evolved complex, complementary adaptations across multiple functional pathways to simultaneously reduce inflammatory response while maintaining strong antiviral defenses, potentially underlying their suspected tolerance of viruses. Second, the move from the ancestral arthropod diet to high-sugar nectar and fruit-based diets is associated with t lineage-specific gene family expansions in metabolic gene families.