Molecular Phylogenetics and Evolution 41 (2006) 288–294 www.elsevier.com/locate/ympev Molecular phylogenetic relationships of Xiphidiopicus percussus, Melanerpes, and Sphyrapicus (Aves: Picidae) based on cytochrome b sequence Lowell C. Overton a,¤, Douglas D. Rhoads b a Department of Biological Sciences, University of Windsor, Windsor, Ont., Canada N9B 3P4 b Department of Biological Sciences, University of Arkansas, Fayetteville, AR 72701, USA Received 25 November 2005; revised 10 May 2006; accepted 15 May 2006 Available online 22 May 2006 Abstract The endemic woodpecker, Xiphidiopicus percussus, from Cuba has been postulated as the sister taxon to the Hispaniolan woodpecker (Melanerpes striatus) and its relationships to the genera Sphyrapicus and Melanerpes have been speculated. We used mitochondrial cyto- chrome b sequences from a collection of New World picids to investigate the phylogenetic relationships among these species using maxi- mum parsimony and maximum likelihood approaches. Our data suggest that X. percussus is the sister taxon to the Melanerpes woodpeckers, which appear to group into a single distinct clade. Xiphidiopicus percussus is not the sister taxon to M. striatus as has been postulated [Olson, S., 1972. The generic distinction of the Hispaniolan Woodpecker, Chryserpes striatus (Aves: Picidae). Proc. Biol. Soc. Wash. 85, 499–508]. The genus Sphyrapicus appears to have diverged earlier than Xiphidiopicus. Divergence estimates from the cyto- chrome b sequences indicate that Xiphidiopicus probably diverged sometime in the late Miocene-early Pliocene, and the endemic contem- porary species X. percussus on Cuba may be a relict from a group that originated in Central America or North America. © 2006 Elsevier Inc. All rights reserved. Keywords: Cytochrome b; Xiphidiopicus percussus; Cuba; Endemic 1. Introduction ulets are the sister group to the woodpeckers and the wry- necks are the sister group to the piculet–woodpecker clade The Order Piciformes comprises the Jacamars (Galbuli- (Sweirczewski and Raikow, 1981; Short, 1982; Burton, dae), PuVbirds (Bucconidae), Barbets (Capitonidae), Tou- 1984). cans (Rhamphastidae), Honeyguides (Indicatoridae), and The monophyly of woodpeckers and their sister status to true woodpeckers (Picidae) (Short, 1982). The family Pici- piculets have never been seriously disputed. However, the dae consists of the subfamilies Picinae (woodpeckers), Jyn- placement of certain genera and species within the Picinae ginae (wrynecks), and the Picuminae (piculets). has not yet been established. One explanation given for the Woodpeckers comprise approximately 218 species within poorly resolved classiWcation of many woodpecker genera 23 genera in the subfamily Picinae and are fairly cosmopol- is rapid diversiWcation of the basal picinae lineages from an itan in distribution (except Australia and Antarctica). The early common ancestor may have obscured the develop- accepted relationships among the three subfamilies are: pic- ment of synapomorphies during incipient diVerentiation (Webb and Moore, 2005). Mitochondrial DNA analyses from DeWllipis and Moore (2000); Prychitko and Moore * Corresponding author. Present address: Institute of Environmental (2000), Weibel and Moore (2002a,b), and Webb and Moore Health Sciences, Wayne State University, 2727 Second Avenue, Detroit, (2005) have provided a substantial database of DNA MI 48201, USA. E-mail addresses: [email protected], [email protected] sequences for many woodpecker species and genera, much (L.C. Overton). of which is in disagreement with Short’s (1982) taxonomic 1055-7903/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2006.05.014 L.C. Overton, D.D. Rhoads / Molecular Phylogenetics and Evolution 41 (2006) 288–294 289 scheme for picids, and Olson’s (1983) polyphyletic argu- proven useful in a number of studies involving taxa of vary- ment for woodpeckers. The molecular evidence from these ing evolutionary depths, from population level analyses studies shows that woodpeckers (excluding barbets, jaca- (Smith and Patton, 1991; Patton and Smith, 1994) to stud- mars, toucans, and honeyguides) comprise a monophyletic ies of much older (>50 mya) divergences (Meyer and Wil- group, with Melanerpes and Sphyrapicus as sister genera, son, 1990). As woodpeckers are considered to have and Colaptes, Dryopcopus, Piculus, Geocolaptes, Campephi- diverged relatively rapidly over short periods of time (DeWl- lus, Celeus, and Chrysocolaptes forming a separate distinct lipis and Moore, 2000; Moore and DeFilippis, 1997; Webb group closest to a Melanerpes–Sphyrapicus–Picoides–Venil- and Moore, 2005), and since nucleotide sequence from the iornis–Dendropicos subclade. cytochrome b gene has been used successfully to examine Within the picid woodpeckers, however, the placement the phylogenetic relationships for most piciforms (DeWllipis of the monotypic species Xiphidiopicus percussus from and Moore, 2000; Moore and DeFilippis, 1997; Prychitko Cuba has been of question for some time. The species has and Moore, 2000; Weibel and Moore, 2002a,b; Webb and been thought to be nearest Melanerpes and Sphyrapicus, Moore, 2005) and other vertebrate species (e.g., Irwin et al., although some speculation regarding where X. percussus 1991; Graybeal, 1993), we felt that the cytochrome b gene sits relative to both of those genera exists. Short (1974) would be an appropriate genetic marker for this study. thought it showed some resemblance to Melanerpes formi- civorous and some species of Sphyrapicus, and may have 2. Materials and methods diverged earlier from a group that gave rise to Melanerpes (including M. striatus from Hispaniola). However, it has 2.1. Taxon sampling been suggested that plumage convergence among wood- peckers has caused errors in generic-level classiWcations Species chosen for inclusion in the phylogenetic analyses (e.g., Veniliornis and Piculus) (Webb and Moore, 2005). As were from the Melanerpes, Sphyrapicus, Picoides, Venilior- the Melanerpes woodpeckers are the closest relatives to the nis, and Dendropicos genera. The following species were sapsuckers (Sphyrapicus) (DeWllipis and Moore, 2000; sampled for this study: Sphyrapicus varius, M. striatus, Mel- Moore and DeFilippis, 1997; Weibel and Moore, 2002a,b; anerpes carolinus, Melanerpes supercilliaris, Melanerpes por- Prychitko and Moore, 2000; and Webb and Moore, 2005), toricensis, and X. percussus (GenBank Accession Nos.: it’s possible that X. percussus is a relict that shares common AF441649–AF441659). Sequence data from a minimum of ancestry to Melanerpes; however, no analysis to address two individuals per species was chosen for reproducibility, this hypothesis has ever been done. with the exception of X. percussus for which only one indi- Olson (1972) proposed that X. percussus and M. striatus vidual was used. Samples (blood or tissue) of Melanerpes, (previously Chryserpes striatus) from Hispaniola were Sphryapicus, and Xiphiodiopicus were obtained from Weld probably sister species based on osteological evidence. He collections by the Wrst author, collaborators, or from tissue felt that M. striatus was too distinct to be considered close collections at the Museum of Vertebrate Zoology (UC to any other Melanerpes species and as such suggested plac- Berkeley) and Louisiana State University Museum of Nat- ing M. striatus nearest Xiphidiopicus, indicating the possi- ural Science. To increase the phylogenetic robustness and bility of M. striatus not being a member of Melanerpes. statistical power of our tree topologies we included the Olson (1972) argued that X. percussus and M. striatus be cytochrome b sequences from GenBank of Veniliornis call- removed from a “melanerpine” grouping and placed near onotus (AY942892), Veniliornis nigriceps (AY942893), Pico- the African and Neotropical members of the “campether- ides albolarvatus (AY942887), Dendropicos griseocephalus ine–colaptine” assemblage (Campethera, Geocolaptes, Den- (AY942884), Dendropicus fuscescens (AY942883) from dropicos, Colaptes, Piculus, and Picoides) (Olson, 1972). Webb and Moore (2005), and Picoides pubescens Olson (1972), however, did not perform any phylogenetic (AF389324) from Weibel and Moore (2002a). All of these reconstruction of the characters he used to propose these species are the closest relatives to Melanerpes and Sphyrapi- relationships. cus, so we expected that their inclusion would provide ade- Although the evolutionary placement of X. percussus quate resolution of the phylogenetic position of X. has been the subject of speculation for some time, no percussus. We rooted our trees with Veniliornis callonotus molecular analysis of this taxon and its relationships to (Webb and Moore, 2005) to evaluate the Xiphidiopicus– other woodpecker genera has ever been reported. We used Melanerpes–Sphyrapicus relationships. nucleotide sequence data from the mitochondrial gene cytochrome b to assess the phylogenetic placement of X. 2.2. DNA sequencing percussus relative to other species of the woodpeckers (including M. striatus) within the subfamily Picinae. The Genomic DNA was extracted from tissue or blood sam- cytochrome b gene codes for the central subunit of the ubi- ples by lysis in 0.5% SDS, phenol:chloroform:isoamyl alco- quinol: cytochrome c reductase (bc1 complex) present hol (25:24:1) extractions, and ethanol precipitation. within the inner mitochondrial membrane as part of com- Extracts were then digested with 100 g/ml RNAse A at plex III of the electron transport chain (Hauska et al.,