A molecular timescale for galliform birds accounting for uncertainty in time estimates and heterogeneity of rates of DNA substitutions across lineages and sites
Introduction
Current knowledge about phylogenetic relationships of galliform birds (Aves: Galliformes) has been built on several types of morphological, behavioral, and molecular data. Although too numerous to be reviewed here, these studies have established the monophyly of Galliformes and its sister relationship to Anseriformes (ducks, geese, swans, and allies). This clade, called Galloanserae, is sister to all other Neognathae birds (e.g., Cracraft and Mindell, 1989, Sibley and Ahlquist, 1990). A well-accepted classification of the Order Galliformes recognize seven families: Megapodiidae (mound builders, brush turkeys, and allies), Cracidae (curassows, guans, and chachalacas), and the ‘phasianoid’ families Odontophoridae (New World quails), Numididae (guineafowl), Phasianidae (pheasants, partridges, Old World quails, and allies), Meleagrididae (turkeys), and Tetraonidae (grouse and allies) (del Hoyo et al., 1994). Phylogenetic studies have shown that Megapodiidae, followed by Cracidae, split off from the remaining galliforms near the base of the tree (e.g., Cracraft, 1981, Groth and Barrowclough, 1999). Within phasianoids, the relationship of Odontophoridae and Numididae to each other and in relation to an unresolved clade containing Phasianidae, Meleagrididae, and Tetraonidae are still unclear (e.g., Dimcheff et al., 2002, Dyke et al., 2003). Below family level, well-supported phylogenetic hypotheses have been proposed for Megapodiidae (Birks and Edwards, 2002), Cracidae (Grau et al., 2005, Pereira and Baker, 2004, Pereira et al., 2002), Tetraonidae (Drovetski, 2002, Lucchini et al., 2001), and some Phasianidae groups (Lucchini et al., 2001, Randi et al., 2000) based on molecular markers and comprehensive taxon sampling.
Besides providing a phylogenetic framework that indicates relative timing of diversification, DNA sequences can also be used to estimate absolute times of divergence among taxa if one or more calibration points are available for at least one node in the phylogenetic tree. Suitable calibration points may be derived from the fossil record, geological barriers that isolated ancestral populations, previous molecular time estimates, or a combination of these time constraints. For example, molecular time estimates indicated that Galliformes and other avian lineages originated before the Cretaceous/Tertiary boundary at 65 million years ago (mya) (Cooper and Penny, 1997, Haddrath and Baker, 2001, Kumar and Hedges, 1998, van Tuinen and Hedges, 2001). These estimates are now supported by the discovery of Vegavis iaai, a fossil anseriform from the Maastrichtian (65–71 myr) of Antarctica (Clarke et al., 2005). Cladistic analysis indicated that this fossil is more closely related to Anatidae (true ducks), which corroborates the hypothesis that at least the ancestors of palaeognaths, galliforms, and anseriforms were already separate lineages at the end of the Cretaceous. Within galliforms, molecular time estimates have also been obtained for closely related genera and species using mitochondrial and nuclear DNA sequences (Drovetski, 2003, Grau et al., 2005, Lucchini et al., 2001, Pereira and Baker, 2004, Pereira et al., 2002, Randi et al., 2000).
Among all those studies, only Haddrath and Baker (2001), Pereira and Baker (2004), and Grau et al. (2005) relaxed the assumption of a molecular clock by using methods that allow rates of DNA substitution to vary in different lineages (Sanderson, 1997, Sanderson, 2002). The development of these and other variable rate methods (e.g., Takezaki et al., 1995, Thorne et al., 1998) was prompted by the accumulation of data showing strong departures from a strict molecular clock.
In an attempt to provide a molecular timescale for major lineages of Galliformes, van Tuinen and Dyke (2004) showed that cracids and megapodes have a slower overall rate of DNA substitution compared to other taxa in the Order. Their molecular time estimates were obtained by performing multigene and/or taxonomic partition analyses of nucleotide sequences of mitochondrial ND2 and cyt b genes, using several internal fossil anchor points, and assuming an external molecular time estimate of 90 mya (van Tuinen and Hedges, 2001) for the separation of Anseriformes and Galliformes. By restricting their analysis to genetic and taxonomic partitions that did not depart from a clock-like rate, they concluded that five out of the nine selected fossils provided consistent estimates, and therefore should be useful as calibration point for galliformes. Their results indicated that family diversification started in the Cretaceous, with the separation of the lineage leading to megapodes, and probably cracids, and extended into the beginning of the Tertiary.
Although a timescale for diversification of the Galliformes is invaluable in advancing our understanding of several aspects of their evolutionary history (e.g., time of domestication of chickens and turkeys, identification of palaeobiogeographic events that may have been responsible for their diversification, evolution of genomic structures like transposable elements and pseudogenes, estimation of rates of molecular evolution), van Tuinen and Dyke’s approach is not ideal for several reasons: (1) it excluded information contained in DNA sequences by using only genetic and/or taxonomic partitions evolving in a clock-like manner; (2) it did not account for rate variation in DNA substitutions detected among galliforms; (3) they summarized their divergence times based on estimates from several smaller partitions, therefore incurring large stochastic errors (Rodríguez-Trelles et al., 2002, Thorne and Kishino, 2002); and (4) they assumed a fixed age for the separation of Anseriformes and Galliformes estimated from clock-like sequences (van Tuinen and Hedges, 2001), which has never been tested using methods that account for uncertainty in the estimation of branch lengths and in time estimates.
Here we present a follow-up analysis to refine the molecular timescale for galliforms by using Bayesian inference of divergence times. This approach has the advantage that it accounts for rate variation of DNA substitution among sites and in different lineages, uncertainty in time estimates and branch lengths, and uses fossil data as a priori information instead of as fixed point estimates. This more realistic Bayesian approach suggests that the previously favored 90 myr molecular calibration for the common ancestor of the Galloanserae may be an underestimate compared to the multifossil approach, and that better estimates of divergence times among taxa can be obtained with a multigene approach rather than by averaging estimates across individual gene partitions.
Section snippets
Sequence data
Mitochondrial DNA sequences for 12S rDNA, ND2, and cyt b for representatives of several galliforms, three anseriforms, and one ratite bird were obtained from GenBank (see Appendix A). These were the only genes for which sequences were available for the same set of species representing major galliform lineages. Alignment was performed by eye in MacClade 4.0 (Maddison and Maddison, 2000). The final alignment had 406 bp for 12S rDNA, excluding alignment gaps and ambiguously aligned sites, 996 for
Results
The Bayesian inference of phylogenetic relationships we obtained showed that most nodes were highly supported by posterior probabilities, with only three short internodes with posterior probabilities in the range of 0.54 and 0.85 (Fig. 1). Maximum likelihood bootstrap values were also high for most nodes of the tree, except for the sister relationship of Numididae (Acryllium and Numida) to Phasianidae (pheasants and partridges), including Meleagrididae (Meleagris) and Tetraonidae (Bonasa and
Phylogenetic inference
We stress that due to our limited taxonomic sampling our goal was not intended to fully estimate the phylogenetic relationships among lineages of Galliformes. However, our phylogenetic tree is compatible with previous published phylogenetic hypothesis based on molecular and morphological data (e.g., Dimcheff et al., 2002, Dyke et al., 2003) in which Megapodiidae, followed by Cracidae are the first lineages to diverge from other Galliformes. However, we also found that Odontophoridae and
Acknowledgments
This work was supported by grants to A.J.B. from the National Science and Engineering Research Council of Canada and the Royal Ontario Museum Foundation (Canada). Two anonymous reviewers provided valuable comments.
References (59)
- et al.
A phylogeny of the megapodes (Aves: Megapodiidae) based on nuclear and mitochondrial DNA sequences
Mol. Phylogenet. Evol.
(2002) - et al.
Phylogeny of Tetraoninae and other galliform birds using mitochondrial12S and ND2 genes
Mol. Phylogenet. Evol.
(2002) - et al.
Molecular phylogenetics and biogeography of Neotropical piping guans (Aves: Galliformes): Pipile Bonaparte,1856 is synonym of Aburria Reinchenbach, 1853
Mol. Phylogenet. Evol.
(2005) - et al.
Reading the entrails of chickens: molecular timescales of evolution and the illusion of precision
Trends Genet.
(2004) - et al.
Basal divergences in birds and the phylogenetic utility of the nuclear RAG-1 gene
Mol. Phylogenet. Evol.
(1999) - et al.
A molecular phylogeny of the pheasants and partridges suggests that these lineages are not monophyletic
Mol. Phylogenet. Evol.
(1999) - et al.
Historical biogeography and a mitochondrial DNA phylogeny of grouse and ptarmigan
Mol. Phylogenet. Evol.
(2001) - et al.
Calibration of galliform molecular clocks using multiple fossil and genetic partitions
Mol. Phylogenet. Evol.
(2004) Catalogue of fossil birds, part 2 (Anseriformes through Galliformes)
Bull. Florida State Mus. Biol. Sci.
(1964)- et al.
Definitive fossil evidence for the extant avian radiation in the Cretaceous
Nature
(2005)