Phylogenetic relationships among Synallaxini spinetails (Aves: Furnariidae) reveal a new biogeographic pattern across the Amazon and Paraná river basins

Molecular Phylogenetics and Evolution 78 (2014) 223–231

Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev

Phylogenetic relationships among Synallaxini spinetails (Aves: Furnariidae) reveal a new biogeographic pattern across the Amazon and Paraná river basins Santiago Claramunt Department of Ornithology, American Museum of Natural History, Central Park West at 79th Street, New York, NY 10024, USA

a r t i c l e

i n f o

Article history: Received 17 January 2014 Revised 5 May 2014 Accepted 6 May 2014 Available online 24 May 2014 Keywords: Furnariidae Phylogenetics Biogeography Amazon basin Paraná river Z-linked markers

a b s t r a c t Relationships among genera in the tribe Synallaxini have proved difficult to resolve. In this study, I investigate relationships among Synallaxis, Certhiaxis and Schoeniophylax using DNA sequences from the mitochondrion and three nuclear regions. I implemented novel primers and protocols for amplifying and sequencing autosomal and sex-linked introns in Furnariidae that resolved basal relationships in the Synallaxini with strong support. Synallaxis propinqua is sister to Schoeniophylax phryganophilus, and together they form a clade with Certhiaxis. The results are robust to analytical approaches when all genomic regions are analyzed jointly (parsimony, maximum likelihood, and species-tree analysis) and the same basal relationships are recovered by most genomic regions when analyzed separately. A sister relationship between S. propinqua, an Amazonian river island specialist, and S. phryganophilus, from the Paraná River basin region, reveals a new biogeographic pattern shared by at least other four pairs of taxa with similar distributions and ecologies. Estimates of divergence times for these five pairs span from the late Miocene to the Pleistocene. Identification of the historical events that produced this pattern is difficult and further advances will require additional studies of the taxa involved and a better understanding of the recent environmental history of South America. A new classification is proposed for the Synallaxini, including the description of a new genus for S. propinqua. Ó 2014 Elsevier Inc. All rights reserved.

1. Introduction The Furnariidae (Aves: Passeriformes) represents a spectacular case of a continental adaptive radiation (Remsen, 2003; Claramunt, 2010). Molecular phylogenetic work in recent years has been fundamental to unravel the processes of diversification that led to this radiation (Fjeldså et al., 2005; Moyle et al., 2009; Irestedt et al., 2009; Derryberry et al., 2011; Claramunt et al., 2012a, b; Ohlson et al., 2013). Nevertheless, despite comprehensive taxon sampling and multilocus datasets, some relationships turned out difficult to resolve. This is the case of basal relationships among Synallaxis spinetails and allies (Tribe Synallaxini sensu Moyle et al., 2009), one of the most diverse and ubiquitous groups in the Furnariidae. Close relationships among Synallaxis, Schoeniophylax, Certhiaxis, Poecilurus, and Gyalophylax were first identified by a cladistic analysis of nest architecture (Zyskowski and Prum, 1999). Like other spinetails, these genera build a globular nest made with sticks, with an elongated entrance tube, but in addition, their nests are thatched with extra layers of sticks and E-mail address: [email protected] http://dx.doi.org/10.1016/j.ympev.2014.05.011 1055-7903/Ó 2014 Elsevier Inc. All rights reserved.

plant material placed over the brood chamber, a derived character of the group that gives the nest a characteristics teapot shape (Zyskowski and Prum, 1999). Sequences of the recombination reactivation genes (RAG-1, RAG-2, Moyle et al., 2009) and a combination of cytochrome b and five short nuclear intros (Irestedt et al., 2009) found support for the Synallaxini and a sister relationship between Certhiaxis and Schoeniophylax. An analysis combining RAG and mitochondrial sequences and most of the species in these genera did not recovered the later group and revealed that Poecilurus, Gyalophylax, and Siptornopsis were actually nested within a well-supported ‘core Synallaxis’ clade, and that Synallaxis propinqua was not part of the core Synallaxis clade but sister to Schoeniophylax phryganophilus (Derryberry et al., 2011). However, basal relationships in the Synallaxini were either incongruent or poorly supported across these studies (see also Ohlson et al., 2013), precluding taxonomic changes and inferences about the process of diversification in the group. A sister relationship between S. propinqua and S. phryganophilus has never been suspected before given their disparate phenotypes (Fig. 1). S. propinqua shows a typical Synallaxis morphology, including uniformly gray to grayish brown plumage with contrasting

224

S. Claramunt / Molecular Phylogenetics and Evolution 78 (2014) 223–231

Fig. 1. Schematic representation of Synallaxis propinqua and Schoeniophylax phryganophila, and their distributions. S. propinqua inhabits Amazonian river islands, whereas S. phryganophila inhabits both floodplain and upland habitats in the Paraná River basin and neighboring areas, including an isolated population in the San Francisco River basin in Northeastern Brazil. The divide between the Amazon and Paraná River basins (dashed line) and other geographic features mentioned in the text are also indicated.

rufous wings and tail, a black throat patch, and a tail with 10 pointy rectrices with loosely integrated barbs (Vaurie, 1980; Remsen, 2003, Fig. 1). In contrast, S. phryganophilus has multiple traits that distinguish it from all Synallaxis species, including dark streaks on the dorsal plumage, a longer tail with well-integrated vanes, and the most conspicuous throat plumage in the family (Vaurie, 1980; Remsen, 2003, Fig. 1). From a biogeographical standpoint, the distribution of these sister species does not correspond to any described pattern: S. propinqua is a river-island specialist in major Amazonian rivers (Remsen and Parker, 1983; Rosenberg, 1990), and S. phryganophilus inhabit scrublands and woodlands mainly in the Paraná river basin (Remsen, 2003, Fig. 1). Therefore, confirmation of a sister relationship between these two species may provide new perspectives on phenotypic and geographic diversification in Neotropical birds. In this study, I investigate basal relationships among Synallaxis and allies using 6.8 kb of DNA sequences from the mitochondrion and three nuclear regions. I implemented novel primers and protocols for amplifying and sequencing autosomal and sex-linked introns in Furnariidae that can resolve relationship in rapid radiations. Using multiple genomic regions increases the chance of recovering the species tree, as opposed to the gene tree of an individual locus (Maddison, 1997; Liu et al., 2009a). In addition, loci on the Z chromosome have shorter coalescent times and lower recombination rates compared to autosomal loci, two useful characteristics for phylogenetics (Peters et al., 2005; Barker et al., 2008; Corl and Ellegren, 2013). With this expanded dataset, I provide a solid phylogenetic framework for basal relationships in the Synallaxini, and propose a new classification, including the description of a new genus. In addition, the relationships revealed suggest a new biogeographic pattern for Neotropical birds, the basic characteristics of which are described and discussed here for the first time.

2. Material and methods 2.1. Taxon sampling and molecular markers I selected species to represent all major clades among Synallaxis and allies: 12 core Synallaxis species including the type species of former genera Poecilurus, Siptornopsis and Gyalophylax (now merged into Synallaxis), S. propinqua, S. phryganophilus, and the two species of Certhiaxis. As outgroup, I included representatives of the most closely related genera Spartonoica, Pseudoseisura, and Pseudasthenes (Derryberry et al., 2011). In addition to three mitochondrial genes used in previous studies (ND2, ND3, and COII, Derryberry et al., 2011), I included eight nuclear introns. Two of them, Aconitase 1 intron 9 (ACO1) and muscle-specific kinase intron 3 (MUSK), are located on the Z chromosome, one of the sex chromosomes in birds. The other introns used are located in two autosomal genes: introns 5–7 of the beta fibrinogen gene (FGB), and introns 5–11 of the glyceraldehyde 3-phosphate dehydrogenase gene (G3PDH). Amplification of mitochondrial genes followed Claramunt et al. (2010). ACO1, MUSK, G3PDH, and FGB introns were amplified and sequenced using published as well as newly designed primers (Table 1). In the case of the G3PDH gene, whereas previous studies in suboscines used only intron 11 (Fjeldså et al., 2003, 2005), I designed primers for amplifying two longer fragments including introns 5–11, and intervening exons. ACO1, MUSK, and G3PDH where amplified using three PCR primers: a small quantity of a third primer that anneals slightly outside the target sequence was added to the master mix to increase specificity (Hillis et al., 2005). PCR reactions were done using standard methods in 15-lL total volume, including 0.75 lL of each main primer and, for ACO1, MUSK, and G3PDH, 0.15 lL of the third primer. Thermocycling conditions started with an initial denaturation at 95 °C for

225

S. Claramunt / Molecular Phylogenetics and Evolution 78 (2014) 223–231 Table 1 Primers used for amplification and sequencing, including internal sequencing primers. Gene

Name

Use

Direction

Sequence (50 –30 )

References

ACO1 intron 9

I9F2 I9R I9F

PCR main PCR main PCR third

Forward Reverse Forward

CTCCTCTCAGGATCCAGACTT CTGCAGCAAGGCACAACAGT CTGTGGGAATGCTGAGAGATTT

Kimball et al. (2009) Barker et al. (2008) Barker et al. (2008)

I9FintSyn I9RintSyn

Internal Internal

Forward Reverse

CTTCTGTGTCCAGGAAACTCAAG GCAAACCTAAACACAAGCTACAGG

This study This study

MUSK intron 3

I3F I3R I3R2

PCR main PCR main PCR third

Forward Reverse Reverse

CTTCCATGCACTACAATGGGAAA CTCTGAACATTGTGGATCCTCAA TAGGCACTGCCCAGACTGTT

Kimball et al. (2009) Kimball et al. (2009) Kimball et al. (2009)

G3PDH introns 5–8

5Fa 9R 5Fb

PCR main PCR main PCR third

Forward Reverse Forward

GGTGCTGAGTATGTTGTGGAGTC GGTCMACAACAGACACATTAGG GCGTGACCCCAGCAACATC

This study This study This study

G3PDH introns 9–11

9F 12Ra 12Rb

PCR main PCR main PCR third

Forward Reverse Reverse

CCTAATGTGTCTGTTGTKGACC CTTGGATGCCATGTGGACC TGTGTGCTCAGCTTACTCC

This study This study This study

11F 11R

Internal Internal

Forward Reverse

GCCATTCCTCCACCTTTGATGC AACCAGCTTGACAAAATGGTC

This study This study

FGB intron 5

Fib5 Fib6 Fib6R3

PCR main PCR main PCR third

Forward Reverse Reverse

CGCCATACAGAGTATACTGTGACAT GCCATCCTGGCGATTCTGAA CTTCCGAAATTAACACTGCC

Kimball et al. (2009) Kimball et al. (2009) This study

FGB intron 7

Fib.7F Fib.7R

PCR main PCR main

Forward Reverse

TGATGGAAGGAGCTTCACAG TTGGCTGATTTTGTCATTTCC

Kimball et al. (2009) Kimball et al. (2009)

2 min, followed by 35 cycles consisting of: 20 s denaturation at 95 °C, 15 s annealing at 50 °C (MUSK, FGB-7) or 55 °C (ACO1, G3PDH, FGB-5), and 60 s (MUSK, FGB) or 80 s (ACO1, G3PDH) extension at 72 °C; a final extension of 3 (MUSK, FGB) or 5 min (ACO1, G3PDH) at 72 °C. New sequences were deposited in GenBank. Specimen data and GenBank accession numbers are detailed in Supplementary Table 1.

2.2. Phylogenetic analysis Phylogenetic analyses were performed using parsimony, maximum-likelihood, and species-tree methods. A parsimony analysis of the concatenated dataset was run in PAUP (Swofford, 2002) as a non-parametric alternative to probabilistic methods. Heuristic searches consisted of 100 runs of stepwise random taxon-additions followed by tree-bisection-reconnection branch swapping rearrangements with a maximum of 1000 optimal trees kept in each replicate. Clade support was assessed using non-parametric bootstrapping with 1000 replicates and the same heuristic-search parameters except that only 10 rounds of random additions were used. Maximum-likelihood inference was performed in RAxML version 7.3.2 (Stamatakis et al., 2005) using the general time-reversible model of nucleotide substitution with rate heterogeneity among sites modeled by a gamma distribution (GTR + C). I evaluated various partitioning regimes and used model selection techniques to choose the optimal model (Sullivan and Joyce, 2005). The four genomic regions (mtDNA, FGB, Z-linked, and G3PDH) were always treated in different partitions. Within some of these regions, I explored partitioning by coding versus non-coding sequences (G3PDH), by gene (mtDNA, Z-linked) and by codon position (G3PDH and mtDNA): I obtained maximum likelihood values for the different partition regimes in RAxML and calculated Akaike’s information criterion (second-order estimator AICc) including branch lengths in addition to substitution model parameters as estimated parameters and using the number of variable sites in the alignment as sample size. The optimal models consisted on mitochondrial sequences partitioned by gene and codon position (nine partitions), the G3PDH gene partitioned by non-coding and coding regions and third codon positions separated (three partitions), and Z-linked intros as a single partition. Using these partitioning schemes, I generated trees for the concatenated dataset and for each genomic

region separately. No further attempt was made to generate trees for individual genes or introns; neighboring sequences are expected to be linked to some degree, and the joint analysis of neighboring sequences helps estimating gene-trees with more accuracy. Clade support was assessed using the fast bootstrap algorithm implemented in RAxML (Stamatakis et al., 2008). Finally, species tree methods take into account the discordance among gene genealogies due to the stochasticity of the coalescence process (Liu et al., 2009a). I used the STAR algorithm, which estimates the topology of the species tree based on the rank of clades in a sample of gene trees (Liu et al., 2009b). Like other species tree methods based on summary statistics, STAR does not require modeling ancestral population sizes or relative substitution rates (Liu et al., 2009b; Allman et al., 2013); therefore, it is useful for analyzing coalescent events that occurred deep into the past. The method is statistically consistent and proved effective in recovering the species tree in simulations and empirical datasets (Liu et al., 2009b; Allman et al., 2013). STAR analysis was performed in the STRAW web server (Shaw et al., 2013) using 1000 bootstrap replicates of each genomic region from the maximum likelihood analysis. Clade support was assessed using a two-step bootstrap procedure to take into account the hierarchical nature of multilocus datasets (genes and sites within genes; Seo, 2008). In addition to comparing concatenation with species tree methods, I evaluated whether the four genomic regions converged to a common topology using a graphical method (Hillis et al., 2005). First, I calculated pairwise distances between tree topologies generated for each genomic region using the ‘path difference’ metric (Steel and Penny, 1993), based on the Euclidean distance of the vectors of number of branches separating tips in a tree (path lengths). The path difference metric was calculated using function ‘treedist’ in R’s ‘phangorn’ library (Schliep, 2011). Then, the distance matrix was reduced to two dimensions using Kruskal’s non-metric multidimensional scaling implemented in the R function ‘isoMDS’ (library ‘MASS’, Venables and Ripley, 2002).

2.3. Biogeography and divergence times To evaluate whether the relationship between S. propinqua and S. phryganophilus corresponds to a more general biogeographic pattern, I focused on other Amazonian island-specialist as

226

S. Claramunt / Molecular Phylogenetics and Evolution 78 (2014) 223–231

determined by Rosenberg (1990). I identified the sister taxon to each island-specialist using recent phylogenetic studies and determined geographic distributions and habitats. I also evaluated whether the biogeographic pattern is shown by species in other Amazonian river habitats (Remsen and Parker, 1983). As a preliminary assessment of historical congruence among species showing the biogeographic patter, I estimated divergence times for taxon pairs using a relaxed mitochondrial 2% rule. Source of sequences for taxon pairs and outgroup species are listed in Supplementary Table 2. I used cytochrome b or ND2 sequences in a Bayesian MCMC sampler implemented in BEAST2 (Bouckaert et al., 2013). Alignments were treated as a single partition with a GTR + C model of substitution. Rate heterogeneity across lineages was modeled with a relaxed lognormal clock (Drummond et al., 2006). Priors were set to defaults (as in Beauti 2.0) except for the mutation rate prior, which was set as a log-normal (logmean = 4.6, log-standard deviation = 0.25) that mimics the distribution of rates found by Weir and Schluter (2008) for a variety of bird groups (mean: 0.01, 95% CI between 0.007 and 0.018 substitutions per site per million year). In this way, the mitochondrial 2% clock is relaxed not only across sites in the alignment (C model) and across lineages (relaxed lognormal clock) but also because its basic rate is not fixed. Although Weir and Schluter (2008) estimates are based on sequences of the cytochrome b gene, the ND2 gene evolves at similar overall rates in birds (Johnson and Sorenson, 1998; Powell et al., 2013). MCMC chains were run for 20 million generations and resulting parameter estimates were checked for stationary and sample sufficiency in Tracer 1.5. 3. Results 3.1. Basal relationships in the Synallaxini The four genomic regions showed heterogeneity in topology and support (Fig. 2). Strongly supported clades tended to be the same across markers and involved sister species relationships: C. cinnamomeus and C. mustelinus, S. stictothorax and S. hypochondriaca, and S. propinqua and S. phryganophilus. Beta fibrinogen was the only marker in which the core Synallaxis was not monophyletic. Z-linked and G3PDH showed a sister relationship between Certhiaxis and the S. propinqua-Schoeniophylax clade; these two

clades formed a paraphyletic group in the mtDNA tree, and are distantly related in the fibrinogen tree. In agreement with these qualitative observations, beta fibrinogen trees are the most different in topological tree space (Fig. 3). Among the other three markers, the overlap in the bootstrap sample suggests more similar phylogenetic information among nuclear markers (Fig. 3). When all markers were analyzed together, different analytical methods resulted in similar topologies and clade support (Fig. 4). In fact, topologies differed only in poorly supported nodes (mostly nodes bellow 50% bootstrap). Conflicting nodes with low bootstrap support (less than 67%) involved S. hellmayri, which was sister to a red-capped clade composed of S. brachyura, S. ruficapilla, S. spixi, and S. albescens in the maximum likelihood and species tree analyses, but sister to the later three only in the parsimony analysis. Across methods, S. propinqua was sister to S. phryganophilus, and together they were sister to the genus Certhiaxis. The concatenated likelihood approach resulted in a topology most similar to the mitochondrial dataset, although bootstrap replicates approached the topologies of nuclear genes (Fig. 3). The STAR species tree occupied a more central position, nearly equidistant to mtDNA, Z-linked, and G3PDH partitions in tree space (Fig. 3). 3.2. Biogeography and divergence times Of the 18 Amazonian river island specialist listed by Rosenberg (1990), five have sister species or clades distributed in the Paraná River basin. One is S. propinqua, discussed above. Another furnariid shows a similar patter: Furnarius minor is sister to a clade formed by F. cristatus from dry Chaco woodlands and F. rufus from campos in the Paraná basin and adjacent areas (Remsen, 2003; Derryberry et al., 2011). The other three species are tyrant flycatchers (family Tyrannidae). Elaenia pelzelni is sister to E. spectabilis (Rheindt et al., 2008) from riverine thickets and forests in the Paraná basin and adjacent parts of the Cerrado (Fitzpatrick et al., 2004). Serpophaga hypoleuca is sister to S. subcristata and S. munda (Rheindt et al., 2008), widespread in the Paraná River basin region (Fitzpatrick et al., 2004). Stigmatura napensis is sister to S. budytoides (Rheindt et al., 2008), from dry Chaco and Espinal scrublands (Fitzpatrick et al., 2004). Interestingly, like S. phryganophilus, both species of Stigmatura also have a subspecies in the San Francisco River basin, in northeastern Brazil.

Fig. 2. Phylogenetic relationships among 16 species in the Synallaxini from the independent analysis of four genomic regions: mitochondrial DNA (genes ND2, ND3, and COII), Z-linked aconitase 1 intron 9 (ACO1) and muscle-specific kinase intron 3 (MUSK), introns 9–11 of the glyceraldehyde 3-phosphate dehydrogenase gene (G3PDH), and introns 5–7 of the beta fibrinogen gene (FGB). Trees are from a maximum likelihood optimization in RAxML, with bootstrap support values on nodes (outgroup not shown).

227

S. Claramunt / Molecular Phylogenetics and Evolution 78 (2014) 223–231

Dimension 2

10

0

-10

-20

-10

0

10

20

Dimension 1

30

40

-10

0

10

20

Dimension 3

Fig. 3. Non-metric multidimensional scaling analysis of path difference distances among alternative trees: maximum likelihood trees (large black symbols) and bootstrap samples (small colored symbols) for four genomic regions (triangles = FGB, squares = mitochondrial, diamonds = Z-linked, downward triangles = G3PDH), a maximum likelihood concatenated analysis (black star) and its bootstrap sample (gray dots), and a species-tree analysis (white star). FGB omitted on second plot.

Fig. 4. Joint analyses of four genomic regions for 16 species in the Synallaxini using three different inference methods: maximum parsimony (MP), concatenated maximum likelihood (ML), and a species-tree method (STAR). The outgroup was omitted from the figure.

All island specialists inhabit relatively young islands with early successional vegetation. Like S. propinqua, S. hypoleuca and S. napensis inhabit Tessaria shrubs; F. minor forages on the ground and E. pelzelni inhabits low Cecropia woodlands (Rosenberg, 1990; Remsen, 2003). The Paraná basin counterparts can be found in a variety of habitats far from rivers. However, many of them are common in habitats closely associated with rivers, such as gallery forest borders (E. spectabilis, F. rufus, S. subcristata) and riverine thickets (E. spectabilis, S. subcristata, S. phryganophilus, Ridgely and Tudor, 1994; Remsen, 2003; Fitzpatrick et al., 2004). When far from rivers, they are usually in vegetation formations that develop over alluvial plains such as Espinal woodlans and arid scrub (F. cristatus, S. budytoides, S. subcristata) in the old sedimentary basins of the

Chaco and Pampas regions (Ridgely and Tudor, 1994; Remsen, 2003; Fitzpatrick et al., 2004). Therefore, they seem to have a loose association with current or ancient aquatic systems. Other birds show an overall similar patter except that the Amazonian taxon is not restricted to islands but inhabits a wider variety of river-created habitats (see Remsen and Parker, 1983; Rosenberg, 1990) and is distributed beyond the Amazon basin, either extending north into the Orinoco basin or the Guiana Shield. The taxon centered on the Paraná basin also usually extends its geographic range into the Caatinga region, or even reaching the Amazon River Delta. This pattern is shown by Hydropsalis climacocerca and H. torquata (Caprimulgidae, Han et al., 2010), Brotogeris versicolurus and B. chiriri (Psittacidae, Ribas et al., 2009), Cyanocorax violaceus and

228

S. Claramunt / Molecular Phylogenetics and Evolution 78 (2014) 223–231

the pair C. cyanomelas-C. cristatellus (Corvidae, Bonaccorso et al., 2010), Paroaria gularis and the pair P. capitata–P. cervicalis (Thraupidae, Dávalos and Porzecanski, 2009), Chrysomus icterocephalus and C. ruficapillus (Icteridae, Lanyon, 1994; Powell et al., 2013), and Agelasticus xanthophthalmus and A. cyanopus (Icteridae, Lanyon, 1994; Powell et al., 2013). The Bayesian analysis of divergence times using a relaxed mitochondrial clock indicated a crown age of about 8.8 Ma (credible interval: 5.4–13.8 Ma) for the Synallaxini, and 6.0 Ma (3.5–9.5 Ma) for the divergence between S. propinqua and S. phryganophilus. For other taxon pairs showing the Amazonia-Paraná pattern, divergence estimates ranged from the mid Miocene (Stigmatura) to the Pleistocene (Serpophaga) with no signs of clumping on particular times suggesting simultaneous divergence (Fig. 5).

robust to analytical approaches when all genomic regions were analyzed jointly and was recovered by the majority of the genomic regions when analyzed separately. A sister relationship between S. propinqua and S. phryganophilus is remarkable given their differences in morphology and plumage (Fig. 1). Perhaps more perplexing is the similarity between S. propinqua and true Synallaxis species. The only morphological characteristic that consistently distinguishes S. propinqua from Synallaxis is its tail feathers, whose tips are more strongly pointed (Vaurie, 1980, pers. obs.). S. phryganophilus, on the other hand, is the most disparate member of the Synallaxini (Vaurie, 1980; Remsen, 2003, Fig. 1). The only aspect in which S. propinqua and S. phryganophilus resemble one another, compared to other Synallaxini, is in the low guttural rattles emitted as part of some of their vocalizations (B. O’Shea and L. Naka, pers. com.). The contrasting plumage differences between S. propinqua and S. phryganophilus could be related to differences in vegetation density and cover. S. propinqua inhabits dense Tessaria thickets, where it is usually out of sight (Rosenberg, 1990; Ridgely and Tudor, 1994). S. phryganophilus inhabits a variety of habitats including thickets and open woodlands, sometimes far from rivers (Ridgely and Tudor, 1994; Remsen, 2003). Although usually secretive like other spinetails, S. phryganophilus is more conspicuous in its habitat, and this could explain its more patterned plumage. I suggest that its colorful throat may be correlated with a more prominent role of visual cues in intraspecific communication, and its striped back may be a cryptic pattern to avoid aerial predators. Field observations and experiments can be conducted to test these hypotheses.

4. Discussion 4.1. Basal relationships in the Synallaxini With the use of long sequences from four independent genomic regions, I confirmed a sister relationship between S. propinqua and S. phryganophilus, which was suggested by previous analyses of mitochondrial sequences but without statistical support (Derryberry et al., 2011). This result is not only evident across analytical methods (Fig. 4) but also accross the four genomic regions analyzed separately (Fig. 2). The addition of nuclear makers also provided strong support for basal relationships in the Synallaxini that were not resolved with confidence in previous analyses (Moyle et al., 2009; Irestedt et al., 2009; Derryberry et al., 2011; Ohlson et al., 2013). The clade formed by S. propinqua and S. phryganophilus is sister to the genus Certhiaxis, and together form a clade that is sister to the core Synallaxis clade (Fig. 4). This result was

4.2. Biogeography Confirmation of a sister relationship between S. propinqua and S. phryganophilus validates a scenario in which an Amazonian

mega-wetlands in W Amazonia

Fitzcarrald arch

Serpophaga hypoleuca S. munda S. subcristata |

|

Elaenia pelzelni E. spectabilis |

|

Furnarius minor F. cristatus F. rufus |

|

Synallaxis propinqua Schoeniophylax phryganophila |

|

Stigmatura napensis S. budytoides |

Quat.

0

Pliocene

5

Miocene

10

15

Divergence time (Ma) Fig. 5. Divergence times for five taxon pairs that show a pattern of distribution in which an Amazonian river island specialist is sister to a taxon distributed in the Paraná River basin and adjacent areas. Estimates were derived from mitochondrial sequences for a more inclusive clade and assuming a relaxed 2% mitochondrial clock in the program BEAST. The light shaded area indicates uncertainty regarding the end of Amazonian mega-wetlands (Campbell et al., 2006; Hoorn et al., 2009). The time of rise of the Fitzcarrald arch is marked with a dashed line. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

S. Claramunt / Molecular Phylogenetics and Evolution 78 (2014) 223–231

island-specialist is sister to a species in the Paraná River basin. Moreover, this pattern is also shared by four other Amazonian island-specialists. A central question is whether this shared pattern is the result of a common history, for example, a vicariance event that resulted in simultaneous speciation. Both basins supported extensive aquatic systems during the Miocene along the foreland basins east of the Andes: the Pebas system and the Paranense sea (Räsänen et al., 1995; Hernández et al., 2005; Hulka et al., 2006; Hoorn et al., 2009). During the latest Miocene, Western Amazonia was characterized by dynamic riverine systems with extensive wetlands and alluvial plains in a landscape of marshlands, savannas, and gallery forests (Latrubesse et al., 2010). Extensive alluvial fans (megafans) covered the Andean foreland basins, from the Orinoco to the Chaco (Wilkinson et al., 2006). This rich mix of dynamic aquatic systems would have provided good conditions for riverine and floodplain successional plant communities and birds adapted to such habitats. It is thus plausible that ancestors of Amazonian and Paraná taxon pairs were distributed across the two basins in riverine or floodplain habitats. The event that separated hypothetical widespread ancestors is more difficult to identify either temporally or spatially. One difficulty is the disagreements on the paleogeographic evolution of Western Amazonia. Some suggest that extensive aquatic systems persisted until the end of the Pliocene, about 2.5 Ma ago (Campbell et al., 2006). Others, in contrasts, suggest that extensive aquatic systems were gradually substituted by rivers and upland forests by the Late Miocene, about 7 Ma ago (Hoorn et al., 2009). Regarding the geographic location of the hypothetical vicariance barrier, the separation of the two basins likely occurred in the Bolivian forelands, where the Andean orogenic belts are closer to the Brazilian Shield. The paleogeography of this region is complex and river systems changed configuration extensively due to river-capture events and megafan dynamics (Lundberg et al., 1998; Wilkinson et al., 2006). For example, the Chapare buttress in Central Bolivia separated the Paraná and Amazon basins during the early and middle Miocene (Fig. 1, Lundberg et al., 1998); some time after 10 Ma ago, a river capture event led to the shift of the divide farther south, in the Michicola arch, where it is today (Fig. 1, Lundberg et al., 1998). In addition, climatic conditions in this region are seasonally dry (Hulka et al., 2006; Jeffery et al., 2012), thus facilitating the separation of aquatic systems. Another possibility is that the vicariance event was the result of the separation of formerly contiguous habitat rather than the division of the two river basins. In particular, the development of the tall tropical forest that covers Western Amazonia today would have separated open riverine habitats from open habitats away from the river. As tall tropical forest began to cover former marshlands and floodplains, bird of open and semi-open habitats were forced to either remain in young successional vegetation in river islands and banks, or move to drier upland areas. This alternative is supported by distributional patterns of the taxa involved, which coincide more with the limit between biomes (Amazonian forests versus savannas and scrublands) rather than the divide between the two river basins. Taxa centered on the Paraná River basin usually extend their geographic range into the Beni savannas, in northern Bolivia, well into the Amazon basin. However, watershed divides within the Amazon basin can also explain this pattern. For example, the rise of the Fitzcarrald arch (Fig. 1), that divided Andean foreland basins in southeastern Peru ca. 4 Ma ago (Espurt et al., 2010) may have triggered speciation in ancestors restricted to lowland floodplains. In any case, divergence time estimates do not concentrate on a narrow temporal range, suggesting that more than one vicariance event may have been involved. Only the two oldest divergences (S. propinqua–S. phryganophila, Stigmatura) are compatible with an effect of landscape transformations in the late Miocene

229

(Hoorn et al., 2009). A vicariance event around 4 Ma, coincident with the rise of the Fitzcarrald arch, seems plausible for the three pairs in the mid range (S. propinqua–S. phryganophila, Furnarius, Elaenia). Finally, the most recent divergences (Furnarius, Elaenia, and Serpophaga) are consistent with an effect of a late Plioceneearly Pleistocene substitution of major aquatic systems by upland tropical forests in Western Amazonia (Campbell et al., 2006). Drier conditions at the onset of the ice ages may have induced a definite break or wider gap between Amazonian and Paraná riverine habitats. Further insights can only be gained by inclusion of more taxa and developing more precise divergence time estimates together with a better understanding of the landscape history of Western South America. 4.3. Classification of the Synallaxini The well-supported phylogenetic framework established in the present study calls for changes in the taxonomic classification of the group, in particular, regarding the redefinition of Synallaxis and the treatment of S. propinqua. Even within a strict phylogenetic paradigm, trees of relationships can be converted into a classification in multiple ways. Here I discus alternatives and propose a classification that (1) reflects phylogenetic relationships, (2) results in diagnosable and phenotypically cohesive named taxa, and (3) minimizes taxonomic changes in relation to the traditional classification. One possibility for establishing a monophyletic Synallaxis is subsuming Schoeniophylax and Certhiaxis into an expanded Synallaxis, but this would result in an exceedingly heterogeneous group, considering the characters of Schoeniophylax described above. The only diagnostic trait associated with an expanded Synallaxis would the thatched nests (Zyskowski and Prum, 1999), a behavioral trait that cannot be evaluated in individual birds or specimens. In addition, the genus Certhiaxis has a long history in ornithology and has been invariably used in its current sense since Cory and Hellmayr (1925) classification. Lumping Certhiaxis into Synallaxis would not only disrupt taxonomic stability but also affect a specific epithet, since there would be homonymy between Certhiaxis cinnamomeus (Gmelin 1788) and Synallaxis cinnamomea Lafresnaye 1843. For these reasons, it is better to maintain these three genera separated and exclude S. propinqua from Synallaxis. The classification of S. propinqua requires further consideration. The simplest solution would be to place it with its sister species, in the genus Schoeniophylax. However, this would deprive Schoeniophylax of any diagnostic characteristic. As noted above, the differences between S. propinqua and S. phryganophilus are as extreme as they can be within the Synallaxini. The alternative is to erect a new genus for S. propinqua. Creating a monotypic genus for S. propinqua may not be desirable on purely phylogenetic grounds; a monotypic genus would be redundant with the specieslevel taxon and will fail to convey phylogenetic relationships (the sister relationship between S. propinqua and S. phryganophilus would not be represented in the classification) (de Queiroz and Gauthier, 1992). On the other hand, placing S. propinqua and S. prhyganophilus in different genera would be in consonance with traditional generic classifications in birds, in which genera are relatively uniform or at least do not show marked phenotypic gaps among species. In addition, the sister relationship between S. propinqua and S. prhyganophilus can be represented using ‘phyletic sequencing’ (Nelson, 1972; Cracraft, 1974, see below). For these reasons, I describe a new genus for S. propinqua, Mazaria, genus novum Type species: Synallaxis propinqua Pelzeln 1859. Diagnosis: Medium-sized spinetail with short, rounded wings and intermediate tail length (about 20% longer than the wings),

230

S. Claramunt / Molecular Phylogenetics and Evolution 78 (2014) 223–231

overall grayish brown plumage, black throat patch, and a rufous tail composed of ten pointy rectrices. It can be distinguished from Synallaxis by the shape of tail feathers tips, which are long cuspidate in Mazaria (i.e. the distal tapering of the vanes occupies ca. 20% of the feather length and ends in a sharp point) rather than short attenuate like in most Synallaxis species (i.e. the distal tapering of the vanes occupies <15% of the feather length and ends in a blunt tip). Included species: Mazaria propinqua (Pelzeln 1859), new combination. Etymology: I dedicate this new genus to my colleague and friend Juan Mazar Barnett, in recognition to his multiple talents, candid spirit, and comradeship. He left us too early and Neotropical ornithology will not be the same without him. Finally, phylogenetic relationships among Synallaxini genera can be represented entirely yet in a simple classification by using phyletic sequencing, in which the taxon listed first is sister to all taxa listed below (Nelson, 1972; Cracraft, 1974): Tribe Synallaxini Genus Synallaxis Genus Certhiaxis Genus Mazaria Genus Schoeniophylax

Acknowledgments I thank Jeff Groth, Jonas Lai, and William M. Mauck for support and advise in the laboratory. Brian O’Shea and Luciano Naka pointed out vocal similarities between S. propinqua and S. prhyganophilus. Joel Cracraft, Luciano Naka, and two anonymous reviewers provided useful comments on the manuscript. My work at the AMNH was supported by a Chapman Postdoctoral Fellowship. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ympev.2014. 05.011. References Allman, E.S., Degnan, J.H., Rhode, J.A., 2013. Species tree inference by the STAR method and its generalizations. J. Comp. Biol. 20, 50–61. Barker, F.K., Vandergon, A.J., Lanyon, S.M., 2008. Assessment of species limits among yellow-breasted meadowlarks (Sturnella spp.) using mitochondrial and sexlinked markers. Auk 125 (4), 869–879. Bonaccorso, E., Peterson, T., Navarro-Sigüenza, A.G., Fleischer, R.C., 2010. Molecular systematics and evolution of the Cyanocorax jays. Mol. Phylogenet. Evol. 54, 897–909. Bouckaert, R., Heled, J., Kuhnert, D., Vaughan, T.G., Wu, C.H., Xie, D., Suchard, M.A., Rambaut, A., Drummond, A.J., 2013. BEAST2: a software platform for Bayesian evolutionary analysis. . Campbell, K.E., Frailey, C.D., Romero-Pittman, L., 2006. The Pan-Amazonian Ucayali Peneplain, Late Neogene sedimentation in Amazonia, and the birth of the modern Amazon river system. Palaeogeogr. Palaeoclimatol. Palaeoecol. 239, 166–219. Claramunt, S., 2010. Discovering exceptional diversifications at continental scales: the case of the endemic families of Neotropical suboscine passerines. Evolution 64 (7), 2004–2019. http://dx.doi.org/10.1111/j.1558-5646.2010.00971.x. Claramunt, S., Derryberry, E.P., Chesser, R.T., Aleixo, A., Brumfield, R.T., 2010. Polyphyly of Campylorhamphus, and description of a new genus for C. pucherani (Dendrocolaptinae). Auk 127 (2), 430–439. http://dx.doi.org/10.1525/ auk.2009.09022. Claramunt, S., Derryberry, E.P., Brumfield, R.T., Remsen Jr., J.V., 2012a. Ecological opportunity and diversification in a continental radiation of birds: climbing adaptations and cladogenesis in the Furnariidae. Am. Naturalist 179 (5), 649– 666. http://dx.doi.org/10.1086/664998. Claramunt, S., Derryberry, E.P., Remsen Jr., J.V., Brumfield, R.T., 2012b. High dispersal ability inhibits speciation in a continental radiation of passerine birds. Proc. Roy. Soc. B 279, 1567–1574. http://dx.doi.org/10.1098/rspb. 2011.1922.

Corl, A., Ellegren, H., 2013. Sampling strategies for species trees: the effects on phylogenetic inference of the number of genes, number of individuals, and whether loci are mitochondrial, sex-linked, or autosomal. Mol. Phylogenet. Evol. 67, 358–366. Cory, C.B., Hellmayr, C.E., 1925. Catalogue of birds of the Americas and the adjacent islands, Part IV, Furnariidae-Dendrocolaptidae. Field Mus. Nat. Hist. Zool. Ser. 13 (3), 1–390. Cracraft, J., 1974. Phylogenetic models and classification. Syst. Zool. 23, 71–90. Dávalos, L.M., Porzecanski, A.L., 2009. Accounting for molecular stochasticity in systematic revisions: species limits and phylogeny of Paroaria. Mol. Phylogenet. Evol. 53, 234–248. de Queiroz, K., Gauthier, J., 1992. Phylogenetic taxonomy. Annu. Rev. Ecol. Syst. 23, 449–480. Derryberry, E.P., Claramunt, S., Chesser, R.T., Remsen Jr., J.V., Cracraft, J., Aleixo, A., Brumfield, R.T., 2011. Lineage diversification and morphological evolution in a large-scale continental radiation: the Neotropical ovenbirds and woodcreepers (Aves: Furnariidae). Evolution 65, 2973–2986. Drummond, A.J., Ho, S.Y.W., Phillips, M.J., Rambaut, A., 2006. Relaxed phylogenetics and dating with confidence. PLoS Biol. 4, e88. Espurt, N., Baby, P., Brusset, S., Roddaz, M., Hermoza, W., Barbarand, J., 2010. The Nazca Ridge and uplift of the Fitzcarrald Arch: implications for regional geology in northern South America. In: Hoorn, C., Wesselingh, F.P. (Eds.), Amazonia: Landscape and Species Evolution: A Look into the Past. Blackwell, pp. 89–100. Fitzpatrick, J.W., Bates, J.M., Bostwick, K.S., Caballero, I.C., Clock, B., Farnsworth, A., Hosner, P., Joseph, L., Langham, G., Lebbin, D., Mobley, J.A., Robbins, M.B., Scholes, E., Tello, J.G., Walther, B., Zimmer, K.J., 2004. Family Tyrannidae (Tyrant Flycatchers): species accounts. In: Hoyo, del J., Elliot, A., Sargatal, J. (Eds.), Handbook of the Birds of the World, vol. 9. Cotingas to Pipits and Wagtails, vol. 9. Lynx Edicions, Barcelona, pp. 258–462. Fjeldså, J., Zuccon, D., Irestedt, M., Johansson, U.S., Ericson, P.G., 2003. Sapayoa aenigma: a NEW WORLD representative of ‘Old World suboscines’. Proc. R. Soc. Lond. B 270, 238–241. Fjeldså, J., Irestedt, M., Ericson, P.G., 2005. Molecular data reveal some major adaptational shifts in the early evolution of the most diverse avian family, the Furnariidae. J. Ornithol. 146, 1–13. Han, K., Robbins, M.B., Braun, M.J., 2010. A multi-gene estimate of phylogeny in the nightjars and nighthawks (Caprimulgidae). Mol. Phylogenet. Evol. 55, 443–453. Hernández, R.M., Jordan, T.E., Dalenz Farjat, A., Echavarría, L., Idleman, B.D., Reynolds, J.H., 2005. Age, distribution, tectonics, and eustatic controls of the Paranense and Caribbean marine transgressions in southern Bolivia and Argentina. J. S. Am. Earth Sci. 19, 495–512. Hillis, D.M., Heath, T.A., St John, K., 2005. Analysis and visualization of tree space. Syst. Biol. 54, 471–482. Hoorn, C., Wesselingh, F.P., Hovikoski, J., Guerrero, J., 2009. The development of the Amazonian mega-wetland (Miocene; Brazil, Colombia, Peru, Bolivia). In: Hoorn, C., Wesselingh, F.P. (Eds.), Amazonia: Landscape and Species Evolution: A Look into the Past. Wiley-Blackwell. Hulka, C., Gräfe, K.-U., Sames, B., Uba, C.E., Heubeck, C., 2006. Depositional setting of the middle to late Miocene Yecua formation of the Chaco foreland basin, Southern Bolivia. J. S. Am. Earth Sci. 21, 135–150. Irestedt, M., Fjeldså, J., Dalén, L., Ericson, P.G.P., 2009. Convergent evolution, habitat shifts and variable diversification rates in the ovenbird-woodcreeper family (Furnariidae). BMC Evol. Biol. 9, 268. Jeffery, M.L., Poulsen, C.J., Ehlers, T.A., 2012. Impacts of Cenozoic global cooling, surface uplift, and an inland seaway on South American paleoclimate and precipitation d18O. Geol. Soc. Am. Bull. 124, 335–351. Johnson, K.P., Sorenson, M.D., 1998. Comparing molecular evolution in two mitochondrial protein coding genes (cytochrome b and nd2) in the dabbling ducks (Tribe: Anatini). Mol. Phylogenet. Evol. 10, 82–94. Kimball, R.T., Braun, E.L., Barker, F.K., Bowie, R.C., Braun, M.J., Chojnowski, J.L., Hackett, S.J., Han, K.L., Harshman, J., Heimer-Torres, V., Holznagel, W., Huddleston, C.J., Marks, B.D., Miglia, K.J., Moore, W.S., Reddy, S., Sheldon, F.H., Smith, J.V., Witt, C.C., Yuri, T., 2009. A well-tested set of primers to amplify regions spread across the avian genome. Mol. Phylogenet. Evol. 50, 654–660. Lanyon, S.M., 1994. Polyphyly of the blackbird genus Agelaius and the importance of assumptions of monophyly in comparative studies. Evolution 48, 679–693. Latrubesse, E.M., Cozzuol, M., da Silva-Caminha, S.A.F., Risgby, C.A., Absy, M.L., Jaramillo, C., 2010. The Late Miocene paleogeography of the Amazon Basin and the evolution of the Amazon River system. Earth Sci. Rev. 99, 99–124. Liu, L., Yu, L., Kubatko, L., Pearl, D.K., Edwards, S.V., 2009a. Coalescent methods for estimating phylogenetic trees. Mol. Phylogenet. Evol. 53, 320–328. Liu, L., Yu, L., Pearl, D.K., Edwards, S.V., 2009b. Estimating species phylogenies using coalescence times among sequences. Syst. Biol. 58, 468–477. Lundberg, J.G., Marshall, L.G., Guerrero, J., Horton, B., Malabarba, M.C.S.L., Wesselingh, F., 1998. The stage for Neotropical fish diversification: a history of tropical South American rivers. In: Malabarba, L.R., Reis, R.E., Vari, R.P., Lucena, Z.M., Lucena, C.A.S. (Eds.), Phylogeny and Classification of Neotropical Fishes. Edipucrs, Porto Alegre, pp. 13–48. Maddison, W.P., 1997. Gene trees in species trees. Syst. Biol. 46, 523–536. Moyle, R.G., Chesser, R.T., Brumfield, R.T., Tello, J.G., Marchese, D.J., Cracraft, J., 2009. Phylogeny and phylogenetic classification of the antbirds, ovenbirds, woodcreepers, and allies (Aves: Passeriformes: infraorder Furnariides). Cladistics 25, 1–20. Nelson, G.J., 1972. Phylogenetic relationship and classification. Syst. Zool. 21, 227– 231.

S. Claramunt / Molecular Phylogenetics and Evolution 78 (2014) 223–231 Ohlson, J.I., Irestedt, M., Ericson, P.G.P., Fjeldså, J., 2013. Phylogeny and classification of the New World suboscines (Aves, Passeriformes). Zootaxa 3613, 001–035. Peters, J.L., McCracken, K.G., Zhuravlev, Y.N., Lu, Y., Wilson, R.E., Johnson, K.P., Omland, K.E., 2005. Phylogenetics of wigeons and allies (Anatidae: Anas): the importance of sampling multiple loci and multiple individuals. Mol. Phylogenet. Evol. 35, 209–224. Powell, A.F.L.A., Barker, F.K., Lanyon, S.M., 2013. Empirical evaluation of partitioning schemes for phylogenetic analyses of mitogenomic data: an avian case study. Mol. Phylogenet. Evol. 66, 69–79. Räsänen, M.E., Linna, A.M., Santos, J.C.R., Negri, F.R., 1995. Late Miocenen tidal deposits in the Amazonian foreland basin. Science 269, 386–390. Remsen Jr., J.V., 2003. Family furnariidae (ovenbirds). In: del Hoyo, J., Elliott, A., Christie, D.A. (Eds.), Handbook of the Birds of the World, vol. 8, Broadbills to Tapaculos. Lynx Edicions, Barcelona, pp. 162–357. Remsen, J.V., Parker III, T.A., 1983. Contribution of river-created habitats to bird species richness in Amazonia. Biotropica 15, 223–231. Rheindt, F.E., Christidis, L., Norman, J.A., 2008. Habitat shifts in the evolutionary history of a Neotropical flycatcher lineage from forest and open landscapes. BMC Evol. Biol. 8, 1193. Ribas, C.C., Miyaki, C.Y., Cracraft, J.L., 2009. Phylogenetic relationships, diversification and biogeography in Neotropical Brotogeris parakeets. J. Biogeogr. 36, 1712–1729. Ridgely, R.S., Tudor, G., 1994. The Birds of South America, vol. 2. The Suboscines Passerines. University of Texas Press, Austin. Rosenberg, G.H., 1990. Habitat specialization and foraging behavior by birds of Amazonian river islands in northeastern Peru. Condor 92, 427–443. Schliep, K.P., 2011. Phangorn: phylogenetic analysis in R. Bioinformatics 27, 592–593.

231

Seo, T., 2008. Calculating bootstrap probabilities of phylogeny using multilocus sequence data. Mol. Biol. Evol. 25, 960–971. Shaw, T., Ruan, Z., Glenn, T., Liu, L., 2013. STRAW: a web server for species tree analysis. Nucleic Acids Res. 41 (W1), W238–W241. http://dx.doi.org/10.1093/ nar/gkt377. Stamatakis, A., Ludwig, T., Meier, H., 2005. RAxML-III: a fast program for maximum likelihood-based inference of large phylogenetic trees. Bioinformatics 21, 456–463. Stamatakis, A., Hoover, P., Rougemont, J., 2008. A rapid bootstrap algorithm for the RAxML web-servers. Syst. Biol. 75, 758–771. Steel, M.A., Penny, P., 1993. Distributions of tree comparison metrics: some new results. Syst. Biol. 42, 126–141. Sullivan, J., Joyce, P., 2005. Model selection in phylogenetics. Annu. Rev. Ecol. Evol. Syst. 36, 445–466. Swofford, D.L., 2002. PAUP. Phylogenetic Analysis Using Parsimony (and Other Methods). Version 4. Sinauer Associates, Sunderland, Massachusetts. Vaurie, C., 1980. Taxonomy and geographical distribution of the Furnariidae (Aves, Passeriformes). Bull. Am. Mus. Nat. Hist. 166, 1–357. Venables, W.N., Ripley, B.D., 2002. Modern Applied Statistics with S, fourth ed. Springer. Weir, J.T., Schluter, D., 2008. Calibrating the avian molecular clock. Mol. Ecol. 17, 2321–2328. Wilkinson, M.J., Marshall, L.G., Lundberg, J.G., 2006. River behavior on megafans and potential influences on diversification and distribution of aquatic organisms. J. S. Am. Earth Sci. 21, 151–172. Zyskowski, K., Prum, R.O., 1999. Phylogenetic analysis of the nest architecture of Neotropical ovenbirds (Furnariidae). Auk 116 (4), 891–911.