The enigmatic monotypic crab plover Dromas ardeola is closely related to pratincoles and coursers (Aves, Charadriiformes, Glareolidae)

EVOLUTIONARY GENETICS

SHORT COMMUNICATION

The enigmatic monotypic crab plover Dromas ardeola is closely related to pratincoles and coursers (Aves, Charadriiformes, Glareolidae)

Sergio L. Pereira^I; Allan J. Baker^{I, II}

^IDepartment of Natural History, Royal Ontario Museum, Toronto, ON, Canada

^IIDepartment of Ecology and Evolution, University of Toronto, Toronto, Canada

^{Send correspondence to}

ABSTRACT

The phylogenetic placement of the monotypic crab plover Dromas ardeola (Aves, Charadriiformes) remains controversial. Phylogenetic analysis of anatomical and behavioral traits using phenetic and cladistic methods of tree inference have resulted in conflicting tree topologies, suggesting a close association of Dromas to members of different suborders and lineages within Charadriiformes. Here, we revisited the issue by applying Bayesian and parsimony methods of tree inference to 2,012 anatomical and 5,183 molecular characters to a set of 22 shorebird genera (including Turnix). Our results suggest that Bayesian analysis of anatomical characters does not resolve the phylogenetic relationship of shorebirds with strong statistical support. In contrast, Bayesian and parsimony tree inference from molecular data provided much stronger support for the phylogenetic relationships within shorebirds, and support a sister relationship of Dromas to Glareolidae (pratincoles and coursers), in agreement with previously published DNA-DNA hybridization studies.

Key words:Dromas, Charadriiformes, Bayesian tree inference, phylogenetics, systematics.

The monotypic crab plover Dromas ardeola (Aves, Charadriiformes, Dromadidae) is very unusual among shorebirds regarding many anatomical and behavioral traits (Rands, 1996). Hence, it is not surprising that its phylogenetic affinities are not well established with these characters. For example, three studies using the same set of osteological characters, but differing in the method of analysis and character coding, have recovered conflicting phylogenies that placed Dromas plus several members of the suborders Charadrii and Lari within an unresolved clade (Strauch, 1978), or as a sister lineage to a clade containing Glareolidae plus Burhinidae embedded within the former family (Mickevich and Parenti, 1980), or yet as a sister lineage to all Lari (Chu, 1995). Based on non-cladistic analyses, skeletal and morphological similarities suggested that Dromas may be closely related to thick-knees (Charadrii, Burhinidae), while plumage characters placed it closely related to avocets (Charadrii, Recurvirostridae), and burrow-nesting behavior linked it to auks (Lari, Alcidae) (reviewed in Rands, 1996; Sibley and Ahlquist, 1990). A recent cladistic analysis of an extensive anatomical data set of birds did not recover the monophyly of any of the three suborders within Charadriiformes, and placed Dromas as a sister lineage to some members of Scolopaci plus a clade containing Lari and Charadrii, but excluding jacanas (Scolopaci, Jacanidae) (Livezey and Zusi, 2007). From a molecular perspective, the phylogenetic affinities of the crab plover has only been studied under a phenetic approach using DNA-DNA hybridization experiments (Sibley and Ahlquist, 1990), which suggested a closer relationship with coursers and pratincoles (Lari, Glareolidae).

To evaluate the phylogenetic affinities of the crab plover Dromas ardeola, we performed a Bayesian phylogenetic analysis in a taxonomic subset of 2,021 anatomical characters previously published for birds (Livezey and Zusi, 2006). Taxa included in the subset (Table 1) were those for which there are DNA sequences for the same species or a congeneric species (Baker et al., 2007). The analysis was performed in MrBayes 3.1 (Ronquist and Huelsenbeck, 2003) using the Mk model of evolution. We set the command lset coding = all rates = invgamma to account for the inclusion of 1,210 invariable anatomical characters and avoid overestimation of branch lengths (Lewis, 2001). Two independent runs were performed in parallel for 2 million generations. Trees were samples in every thousand generations, and the first 201 trees were discarded after checking for convergence of algorithm.

We amplified and sequenced the nuclear RAG-1, and mitochondrial small ribosomal subunit (12S rDNA), cytochrome b (cyt b) and NADH dehydrogenase subunit 2 (ND2) genes for two crab plover specimens, following published primers and protocols (Pereira and Baker, 2004). Both L- and H-strands sequences were checked for ambiguities and a consensus sequence was created for each gene in Sequencher 4.1.2 (GeneCodes, Ann Arbor, Michigan). Consensus sequences were aligned visually in MacClade 4.0 (Maddison and Maddison, 2000). No variation was found between the two specimens, except for a third position transition in cyt b. All sequences obtained in this study were deposited in GenBank (accession numbers HM369458 to HM369458). Ambiguously aligned regions for the 12S rDNA were excluded from the analysis. The aligned molecular data set of 5,183 nucleotides contains the same genera as in the anatomical data set. We inferred the molecular phylogenetic relationships in MrBayes 3 (Ronquist and Huelsenbeck, 2003), assuming that each gene evolves following a general time-reversible model of evolution (GTR), and accounting for gamma-distributed rate variation (G) and a proportion of invariable sites (I), as suggested by the Akaike Information Criterion implemented in Modeltest 3.7 (Posada and Crandall, 1998). A codon-based partitioned model was also applied, where each codon position of protein-coding genes and non-coding positions of 12S rDNA were allowed to evolve following the GTR+G+I model. Bayesian trees were sampled as described above for the anatomical data set. We also inferred tree topology using maximum parsimony through heuristic search (branch swap = TBR, nreps = 100), and estimate branch support with 1,000 heuristic bootstrap replicates in PAUP 4.0b10 (Swofford, 2001).

Anatomical and molecular data evolve at different rates over time and across lineages. The combined phylogenetic analysis of these characters (total-evidence approach) may provide support for different parts of the phylogenetic tree, and/or reveal hidden conflict that is highly supported by one but not both data sets (Pereira and Baker, 2005). We combined the anatomical and molecular data sets and performed a Bayesian tree inference using the models of evolution described above for each individual data set.

The Bayesian analysis of the anatomical data set performed here suggested that Dromas is a sister lineage to Haematopodidae, with Posterior Probability (PP) = 0.93 (Figure 1). Many nodes have PP < 0.95, which are considered weakly supported, and the PP of the consensus tree among 88 trees present in the 95% credible interval is 0.23. The consensus Bayesian tree obtained here is considerably different from the maximum parsimony topology derived from more inclusive taxon data set (Livezey and Zusi, 2007). The parsimony tree in Livezey and Zusi (2007) did not have strongly supported nodes among most shorebirds, did not recover the three Charadriiformes suborders as monophyletic, and placed Jacanidae followed by Dromas as sister groups to the remaining shorebirds (Livezey and Zusi, 2007).

The consensus Bayesian tree inferred from the molecular data set including the same genera as in the anatomical data set (Figure 2) placed Dromas as a sister lineage to Glareolidae with posterior probability (PP) = 0.95, in agreement with DNA-DNA hybridization studies (Sibley and Ahlquist, 1990). PP of the consensus molecular tree among 42 topologies in the 95% credible interval of trees is 0.29. The relationships among the remaining taxa were identical to those of our previous study, in which Dromas was not sampled (Baker et al., 2007), except that Rissa and Rynchopus were placed as sister genera. The codon partitioned model and the parsimony tree topology was similar to that of Figure 2, except that Chlidonias and Rissa are sister genera, in exclusion of Rynchopus (PP = 0.68; bootstrap support = 74%), in agreement with our previous phylogeny including 90 Charadriiformes genera (Baker et al., 2007).

The inferred Bayesian topology derived from the total-evidence approach (Figure S1) was identical to the topology obtained from the molecular data set (Figure 2) with two exceptions: (1) the position of Turnix was similar to the topology derived from the anatomical data alone (Figure 1), with PP = 0.98; and (2) Dromas was inferred to be a sister lineage to a clade including Uria, Stercorarius, Rhyncops, Chlidonias and Rissa (PP = 0.90), as opposed to a sister lineage to Glareolidae as inferred by the molecular data set (Figure 1). Hence, the conflicting and poorly supported topologies recovered in the analyses of the anatomical data set using two distinct methods of tree inference support our previous suggestion that anatomical characters cannot confidently resolve the phylogenetic relationships among shorebirds (Pereira and Baker, 2005). In fact, retention of ancestral polymorphism or parallel evolution in phylogenetically independent lineages caused by ecological, behavioral and/or physiological constraints seems to obscure the evolutionary history of many organisms (Pereira and Baker, 2005).

In conclusion, based on molecular sequence (this study) and DNA-DNA hybridization data (Sibley and Ahlquist, 1990), the crab plover Dromas ardeola is sister group to pratincoles and coursers, as supported by Bayesian and parsimony analyses of DNA sequences of RAG-1, 12S rDNA, cyt b and ND2.

Acknowledgments

We are grateful to Giuseppe De Marchi for donating tissue samples of two specimens of Dromas ardeola to the Royal Ontario Museum, and the Field Museum for providing tissue samples for Ciconia abdimii. AJB was supported by an operating grant from the Natural Sciences and Engineering Research Council of Canada, the Royal Ontario Museum Foundation, and the National Science Foundation (Assembling the Tree of Life (AToL) Program - EF-0228693). Bayesian analyses were carried out by using the resources of the Computational Biology Service Unit from Cornell University (http://cbsuapps.tc.cornell.edu/).

Supplementary Material

The following online material is available for this article:

Figure S1 - Consensus Bayesian tree derived from the total evidence approach.

This material is available as part of the online article from http://www.scielo.br/gmb

Received: December 12, 2009; Accepted: May 14, 2010.

Associate Editor: Louis Bernard Klaczko

License information: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

• Baker AJ, Pereira SL and Paton TA (2007) Phylogenetic relationships and divergence times of Charadriiformes genera: Multigene evidence for the Cretaceous origin of at least 14 clades of shorebirds. Biol Lett 3:205-209.
• Chu PC (1995) Phylogenetic reanalysis of Strauch's osteological data set for the Charadriiformes. Condor 97:174-196.
• Lewis PO (2001) A likelihood approach to estimating phylogeny from discrete morphological character data. Syst Biol 50:913-925.
• Livezey BC and Zusi RL (2006) Higher-order phylogeny of modern birds (Theropoda, Aves, Neornithes) based on comparative anatomy: I. - Methods and characters. Bull Carnegie Mus Nat Hist 37:1-556.
• Livezey BC and Zusi RL (2007) Higher-order phylogeny of modern birds (Theropoda, Aves, Neornithes) based on comparative anatomy. II. Analysis and discussion. Zool J Linn Soc 149:1-95.
• Maddison DR and Maddison WP (2000) MacClade 4.0. Sinauer Associates Inc, Sunderland.
• Mickevich MF and Parenti LR (1980) The phylogeny of the Charadriiformes (Aves): A new estimate using the method of character compatibility analysis by J.G. Strauch. Syst Zool 29:108-113.
• Pereira SL and Baker AJ (2004) Vicariant speciation of curassows (Aves, Cracidae): A hypothesis based on mitochondrial DNA phylogeny. Auk 121:682-694.
• Pereira SL and Baker AJ (2005) Multiple gene evidence for parallel evolution and retention of ancestral morphological states in the shanks (Charadriiformes, Scolopacidae). Condor 107:514-526.
• Posada D and Crandall KA (1998) MODELTEST: Testing the model of DNA substitution. Bioinformatics 14:817-818.
• Rands MRW (1996) Family Dromadidae. In: del Hoyo J, Elliott A and Sargatal J (eds) Handbook of the Birds of the World. v. 3. Lynx Ediciones, Barcelona, pp 302-307.
• Ronquist F and Huelsenbeck JP (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19:1572-1574.
• Sibley CG and Ahlquist JE (1990) Phylogeny and Classification of Birds: A Study in Molecular Evolution. Yale University Press, New Haven, 1080 pp.
• Strauch Jr JG (1978) The phylogeny of the Charadriiformes (Aves): A new estimate using the method of character compatibility analysis. Trans Zool Soc Lond 34:263-345.
• Swofford DL (2001) PAUP*: Phylogenetic Analysis Using Parsimony (*and Other Methods) 4.0. Sinauer Associates Inc., Sunderland.

Publication Dates

• Publication in this collection
  18 Aug 2010
• Date of issue
  2010

History

• Received
  12 Dec 2009
• Accepted
  14 May 2010