Evolutionary relationships and population genetics of the Afrotropical leaf-nosed bats (Chiroptera, Hipposideridae)

Abstract

The Old World leaf-nosed bats (Hipposideridae) are aerial and gleaning insectivores that occur throughout the Paleotropics. Both their taxonomic and phylogenetic histories are confused. Until recently, the family included genera now allocated to the Rhinonycteridae and was recognized as a subfamily of Rhinolophidae. Evidence that Hipposideridae diverged from both Rhinolophidae and Rhinonycteridae in the Eocene confirmed their family rank, but their intrafamilial relationships remain poorly resolved. We examined genetic variation in the Afrotropical hipposiderids Doryrhina, Hipposideros, and Macronycteris using relatively dense taxon-sampling throughout East Africa and neighboring regions. Variation in both mitochondrial (cyt-b) and four nuclear intron sequences (ACOX2, COPS, ROGDI, STAT5) were analyzed using both maximum likelihood and Bayesian inference methods. We used intron sequences and the lineage delimitation method BPP-a multilocus, multi-species coalescent approach-on supported mitochondrial clades to identify those acting as independent evolutionary lineages. The program StarBEAST was used on the intron sequences to produce a species tree of the sampled Afrotropical hipposiderids. All genetic analyses strongly support generic monophyly, with Doryrhina and Macronycteris as Afrotropical sister genera distinct from a Paleotropical Hipposideros; mitochondrial analyses interpose the genera Aselliscus, Coelops, and Asellia between these clades. Mitochondrial analyses also suggest at least two separate A peer-reviewed open-access journal B. D. Patterson et al. / ZooKeys 929: 117-161 (2020) 118 colonizations of Africa by Asian groups of Hipposideros, but the actual number and direction of faunal interchanges will hinge on placement of the unsampled African-Arabian species H. megalotis. Mitochondrial sequences further identify a large number of geographically structured clades within species of all three genera. However, in sharp contrast to this pattern, the four nuclear introns fail to distinguish many of these groups and their geographic structuring disappears. Various distinctive mitochondrial clades are consolidated in the intron-based gene trees and delimitation analyses, calling into question their evolutionary independence or else indicating their very recent divergence. At the same time, there is now compelling genetic evidence in both mitochondrial and nuclear sequences for several additional unnamed species among the Afrotropical Hipposideros. Conflicting appraisals of differentiation among the Afrotropical hipposiderids based on mitochondrial and nuclear loci must be adjudicated by large-scale integrative analyses of echolocation calls, quantitative morphology, and geometric morphometrics. Integrative analyses will also help to resolve the challenging taxonomic issues posed by the diversification of the many lineages associated with H. caffer and H. ruber.

Concordant patterns of genetic, acoustic, and morphological divergence in the West African Old World leaf‐nosed bats of the Hipposideros caffer complex

“…The former study also showed the endemic West African H. lamottei from Mount Nimba to belong within lineage B, which represents another challenge to the taxonomy of H. caffer complex. Further complexity in the cryptic diversity in H. ruber has been shown by Patterson et al (2020). Although their phylogeny corroborates the basic division into the four main lineages by Vallo et al (2008), several new sublineages were newly introduced from East Africa and together with older redefined sublineages proposed as putative species.…”

“…Aligned cytb sequences were first shortened to the length of the shortest sequence and then narrowed down to haplotypes using DnaSP 5.10 (Librado & Rozas, 2009). For assignment of the Ghanaian bats into the major mtDNA lineages A, B, C, and D and for comparison with published data from other African regions, their haplotypes were aligned with selected published haplotypes of H. caffer complex by Vallo et al (2008Vallo et al ( , 2011 and Patterson et al (2020), and outgroup taxa H. jonesi and R. landeri (Figure 2, Table S2, Alignment S1).…”

“…Although several subspecies are recognized within each species, their definition is rather vague due to substantial morphological variability. It has been suggested that further cryptic species diversity exists within the H. caffer complex (Patterson et al, 2020;Simmons, 2005;Vallo et al, 2008;Vallo et al, 2011;Monadjem et al, 2013).…”

“…Previous analysis of mitochondrial (mt) DNA has revealed the existence of four deeply divergent lineages within the H. caffer complex (Vallo et al, 2008), further split into multiple sublineages (Patterson et al, 2020), which possibly constitute separate species, rather than the two currently acknowledged. The four major lineages were labeled A, B, C, and D by Vallo et al (2008).…”

“…Lineage A comprised two forms pertaining to the smaller H. caffer morphotype, recognized due to disjunct distribution areas as H. caffer and H. tephrus. However, a recent study by Patterson et al (2020) showed that this distinction underestimates the diversity within the proper H. caffer lineage; hence, the previous subspecies designation H. caffer tephrus is adopted here for the particular sublineage occurring in the northern savanna belt. The lineages B, C, and D correspond to the larger H. ruber morphotype (forearm length >48 mm; Vallo et al, 2008;Vallo et al, 2011).…”