New nuclear evidence for the oldest divergence among neognath birds: the phylogenetic utility of ZENK (i)

Abstract

To date, there is little consensus concerning the phylogenetic relationships among neognath orders, which include all extant birds except ratites and tinamous. Different data sets, both molecular and morphologic, have yielded radically different and often unresolved ordinal topologies, especially within the neoaves clade. This lack of resolution and ongoing conflict indicates a need for additional phylogenetic characters to be applied to the question of higher-level avian phylogeny. In this study, sequences of a single-copy nuclear gene, ZENK, were used to reconstruct an ordinal-level phylogeny of neognath birds. Strong support was indicated for the oldest divergence within Neognathae; the chicken- and duck-like birds formed a clade that was sister to all other modern birds. In addition, many families of traditional taxonomic orders clustered together in the ZENK tree, indicating the gene’s general phylogenetic reliability. However, within the neoaves clade, there was little support for relationships among orders, which is a result similar to all other recent molecular studies of higher-level avian phylogeny. This similarity among studies suggests the possibility of a rapid radiation of the major neoaves lineages. Despite the ongoing lack of neoaves resolution, ZENK’s sequence divergence and base composition patterns indicate its general utility as a new phylogenetic marker for higher-level avian systematics.

Introduction

Higher-level phylogenies of modern birds (subclass Neornithes) have been plagued by lack of resolution and disagreement among different data sets, both morphologic and molecular (reviewed in Cracraft and Clarke, 2001). This is especially evident regarding the relationships among the ordinal lineages of the neognath clade, which contains all extant birds except the ratites and tinamous. Recent molecular phylogenies based on DNA–DNA hybridization (Sibley and Ahlquist, 1990; van Tuinen et al., 2001), mtDNA (Johnson, 2001; Mindell et al., 1999; Mindell et al., 1997; van Tuinen et al., 2000), and nuclear DNA (Groth and Barrowclough, 1999; Johansson et al., 2001; van Tuinen et al., 2001; van Tuinen et al., 2000) have yielded different and often unresolved topologies within the neognath clade. These problematic results have been attributed to various causes, including methodological problems with DNA–DNA hybridization (reviewed in Sheldon and Bledsoe, 1993), inappropriate rate of mtDNA evolution (too fast) for resolving deep divergences (Groth and Barrowclough, 1999), insufficient number of characters (Groth and Barrowclough, 1999), and insufficient taxon sampling (Cracraft and Clarke, 2001; Johnson, 2001; van Tuinen et al., 2000). In order to address these ongoing challenges, there is a growing need for additional phylogenetic characters to apply to the question of higher-level avian phylogeny, in both separate and combined analyses. Nuclear characters have shown promise for illuminating relationships deep in the avian evolutionary tree (Groth and Barrowclough, 1999; Johansson et al., 2001; van Tuinen et al., 2001; van Tuinen et al., 2000), but as yet, few nuclear genes have been exploited for this purpose. The goal of this study was to apply a new nuclear marker to the question of deep avian phylogeny.

ZENK (Mello et al., 1992) is a single-copy nuclear transcription factor that is well-studied in the field of neurobiology (reviewed in Clayton, 1997; Ribeiro and Mello, 2000), but is relatively untested as a phylogenetic marker (Long and Salbaum, 1998). The name ZENK is an acronym for the four other names by which the gene is known: zif-268 (Christy et al., 1988), erg-1 (Sukhatme et al., 1988), NGFI-A (Milbrandt, 1987), and Krox-24 (Lemaire et al., 1988). ZENK expression has been implicitly linked to the neuronal growth and plasticity associated with learning and memory formation at the molecular level (reviewed in Ribeiro and Mello, 2000; Stork and Welzl, 1999; Tischmeyer and Grimm, 1999) and has become an indispensable marker of neuro-activity during song learning and production in birds (reviewed in Ball and Gentner, 1998; Clayton, 1997). Despite its functional role in avian physiology, there is no evidence of selection pressure acting differentially on ZENK across diverse avian lineages (Chubb, 2002).

The main goal of this study was to use ZENK sequences to reconstruct an ordinal neognath phylogeny in two steps. The first was to test the hypothesis that the chicken-like birds and duck-like birds form a clade, referred to as cohort Galloanserae, that is sister to all other modern birds, the cohort Neoaves. Despite the growing consensus about this deepest divergence within neognath birds (Caspars et al., 1997; Groth and Barrowclough, 1999; Sibley and Ahlquist, 1990; Sibley et al., 1988; van Tuinen et al., 2000; Zusi and Livezey, 2000), more molecular evidence is needed to solidify it. The second step was then to explore whether ZENK sequences could provide some resolution and support for ordinal relationships within the neoaves clade that have been lacking in other recent gene trees. (Groth and Barrowclough, 1999; Johansson et al., 2001; Johnson, 2001; Mindell et al., 1999; Mindell et al., 1997; van Tuinen et al., 2000). The ZENK gene tree combined with a complete description of sequence composition and divergence provides an in-depth assessment of ZENK’s overall phylogenetic utility.