Patterns of lineage diversification in the genus Naso (Acanthuridae)

Patterns of lineage diversification in the genus Naso (Acanthuridae)

MOLECULAR PHYLOGENETICS AND EVOLUTION Molecular Phylogenetics and Evolution 32 (2004) 221–235 www.elsevier.com/locate/ympev Patterns of lineage diver...

563KB Sizes 0 Downloads 77 Views

MOLECULAR PHYLOGENETICS AND EVOLUTION Molecular Phylogenetics and Evolution 32 (2004) 221–235 www.elsevier.com/locate/ympev

Patterns of lineage diversification in the genus Naso (Acanthuridae) Selma O. Klanten,a,* Lynne van Herwerden,a J. Howard Choat,a and David Blairb a

Molecular Ecology and Evolution Laboratory, School of Marine Biology and Aquaculture, James Cook University, Townsville, Qld 4811, Australia b School of Tropical Biology, James Cook University, Townsville, Qld 4811, Australia Received 10 July 2003; revised 5 November 2003 Available online 17 January 2004

Abstract The evolutionary history of the reef fish genus Naso (F. Acanthuridae) was examined using a complete species-level molecular phylogeny of all recognized (19) species based on three loci (one nuclear ETS2 and two mitochondrial 16S, cyt b). This study demonstrates that distinct foraging modes and specialized body shapes arose independently at different times in the evolutionary history of the genus. Members of the subgenus Axinurus, characterized by a scombriform morphology, caudal fin structure and pelagic foraging mode, were consistently placed basal to the remaining Naso species, suggesting that pelagic foraging is plesiomorphic and benthic foraging derived in this genus. We used a genus-level phylogeny (nuclear marker, ETS2), which included several taxa from all other acanthurid genera, to obtain a range of age estimates for the most recent common ancestor of the genus Naso. These age estimates (range of 52–43.3 MY) were then used to estimate divergence times (by nonparametric rate smoothing method) of the node giving rise to extant Naso species using the combined sequence data (from all loci). The reconstruction of the pattern of divergence of extant species indicates two sequences of events. The basal species characterized by pelagic foraging modes arose during the Eocene and Oligocene. Most of the remaining Naso species, including those characterized by benthic foraging, arose over a period of 20 MY during the Miocene. Diversification during this period was associated with major plate tectonic and glaciation events, resulting in changes in sea level, ocean temperature and productivity regimes. Regardless of the foraging mode exhibited, all species of Naso have a caudal propulsive unit similar to that observed in pelagic scombriform fishes, a legacy of the basal position of the subgenus Axinurus in the phylogeny of the genus. Ó 2003 Elsevier Inc. All rights reserved. Keywords: Foraging mode; Molecular phylogeny; Acanthuridae; Fossil record; Lineage diversification; Morphology; Nonparametric rate smoothing

1. Introduction The suborder Acanthuroidei consists of six families of tropical fishes displaying a high level of structural and morphological diversity (Randall, 2002; Tyler et al., 1989; Winterbottom and McLennan, 1993). Most species are deep-bodied laterally compressed fishes with a benthic foraging mode that feed on a variety of sessile and motile invertebrates and marine plants and are strongly associated with reef environments. However, a significant minority of acanthuroid fishes display a contrasting life-style, foraging in open water for planktonic and small nektonic prey. The most distinctive pelagic foraging taxon is represented by the monotypic * Corresponding author. Fax: +61-07-4725 1570. E-mail address: [email protected] (S.O. Klanten).

1055-7903/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2003.11.008

genus Luvarus (Family Luvaridae) confined to open waters (Tyler et al., 1989). Pelagic foraging has also been recorded from the family Acanthuridae, especially in the genera Naso and Acanthurus (Jones, 1968; Randall, 2002). The family Acanthuridae comprises 6 extant genera containing 80 species, Naso (19 recognized species), Prionurus (7), Paracanthurus (1), Zebrasoma (7), Acanthurus (37), and Ctenochaetus (9) (Randall, 2002). Relationships amongst the six genera are well documented from morphological (Winterbottom and McLennan, 1993) and molecular (Clements et al., 2003; Tang et al., 1999) studies. Moreover, the morphological reconstruction of generic relationships within the suborder (Winterbottom and McLennan, 1993) is congruent with the reconstruction based on mitochondrial sequences (Clements et al., 2003). Two primary issues remain. First, there is a need to

222

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

clarify the patterns of evolutionary diversification within genera. This requires species-level phylogenies, especially in the genera Naso and Acanthurus where the component species exhibit considerable ecological diversification (Borden, 1998; Winterbottom and McLennan, 1993). Second, the temporal pattern of diversification, especially in these genera, needs further clarification (Clements et al., 2003). Although none of the extant genera of acanthurids occurs as fossils, the family has an informative fossil record, especially from the early Tertiary, a period when critical events in acanthurid lineage diversification must have occurred (Clements et al., 2003; Sorbini and Tyler, 1998; Tyler, 2000). A number of tropical reef fish phylogenies have been inferred from molecular data (e.g., Bernardi and Crane, 1999; Bernardi et al., 2000; Clements et al., 2003; Craig et al., 2001; Lo Galbo et al., 2002; Ruber et al., 2003; Streelman et al., 2002; Tang, 2001; Tang et al., 1999; Thacker, 2003; Wang et al., 2001). Most of these examined high-level relationships using mtDNA (sequencing 1–4 gene regions) rather than species-level relationships. Some of the studies (e.g., Bernardi et al., 2000; Ruber et al., 2003; Streelman et al., 2002) examined specific features, such as foraging modes, in relation to the phylogeny obtained, an issue also examined and addressed in this study. Two genus-level molecular phylogenies for the family Acanthuridae (surgeonfishes) have been published (Clements et al., 2003; Tang et al., 1999). However, these studies used only a few species of Naso (unicornfish), the genus that is the focus of this study. The 19 described species in the genus Naso (the only extant member of the subfamily Nasinae) exhibit a variety of morphologies including striking extensions of the frontal region of the head either as horns or bulbous protuberances (Randall, 2002; Smith, 1966). In addition, body proportions vary amongst the different species, ranging from relatively deep-bodied forms with an extended snout to relatively slender streamlined forms with body proportions similar to those of scombrid fishes (here termed scombriform morphology). Smith (1966) in a taxonomic survey of the subfamily Nasinae divided the genus Naso into three genera, Naso, Axinurus, and Callicanthus, based on a number of morphological features: two peduncular plates and frontal horn/protuberance in Naso; a single peduncular plate in Axinurus; dentition and the absence of a frontal horn in Callicanthus. In subsequent studies of the Nasinae, Randall (1994, 2001) did not retain these genera but noted that species placed by Smith (1966) in Axinurus were morphologically distinct from other members of the genus Naso. Previous studies (Borden, 1998; Winterbottom, 1993; Winterbottom and McLennan, 1993) identified distinctive morphologies associated with benthic foraging within reef habitats and with pelagic foraging in open

water. A major focus of these studies, recently summarized by Brooks and McLennan (2002) has been the evolutionary relationships among species exhibiting these different foraging modes. The internal structures associated with pelagic foraging in Nasinae have been described by Tyler (1970, 2000), Tyler et al. (1989), and Tyler and Sorbini (1998). These are primarily associated with the caudal propulsive unit. Key features are: a narrow caudal peduncle; caudal fin with a high aspectratio; fusion of the hypural bones into a single plate with bases of the caudal fin rays overlapping the hypural bone (hypurostegy) resulting in a consolidated caudal skeleton similar to that of scombrids and istiophorids (Tyler et al., 1989). The pattern of occurrence suggests that these features have evolved repeatedly in other open-water pelagic fishes, indicative of convergence to common requirements of open-water foraging. Winterbottom and McLennan (1993) argued that benthic foraging targeting macroscopic algae is plesiomorphic in the Acanthuridae (Brooks and McLennan, 2002) and pelagic foraging (plankton feeding) derived. They considered that planktivorous foraging evolved independently from herbivorous ancestors in both Naso and Acanthurus but that in the absence of species-level phylogenetic hypotheses it could not be determined how many times planktivory originated within Naso. Borden (1998) developed a species level phylogeny of Naso based on 15 species and 13 character states (3 osteological, 10 based on soft tissue). He also concluded that (i) herbivory (benthic foraging) was ancestral in Naso, a view reflected in the basal position of a suite of deepbodied reef-associated species in his phylogeny; (ii) pelagic foraging (zooplanktivory) was a derived condition appearing once in his phylogeny; (iii) horn development was ‘‘haphazard’’ occurring at a number of points within the phylogeny. However, Borden (1998) also noted that the phylogeny was only partially resolved and that further studies and increased taxon sampling were required to resolve existing polytomies and species-level relationships associated with pelagic foraging. Fossil acanthurid fishes including pelagic species were examined by Tyler and Sorbini (1998) and Tyler (2000). Tyler (2000) established a phylogeny of the Nasinae demonstrating that the extant Naso (Indo-Pacific Ocean) and its sister taxon Eonaso (Caribbean, possibly Oligocene) were a sister group to Arambourgthurus (Oligocene, Tethys), Sorbinithurus (Eocene, Tethys), and Marosichthys (Miocene, Celebes, West Pacific). Arambourgthurus is a distinctive genus of nasines displaying a scombriform morphology and caudal propulsive unit with extensive hypurostegy. Such a modified caudal fin structure is also seen in Naso and Eonaso although in these genera hypurostegy is rarely strongly developed (Tyler, 2000). This caudal fin occurs throughout the Nasinae including species with a benthic reef-associated foraging mode. A capacity for pelagic foraging appears

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

to be widespread within the subfamily. We concur with the suggestion of Borden (1998) that a more detailed species-level analysis of the genus is required to resolve this issue and we provide that here. The temporal distribution of acanthurid fossils (Blot and Tyler, 1991; Sorbini and Tyler, 1998; Tyler, 1999, 2000; Tyler and Bannikov, 2000) strongly suggests that initial diversification within the Acanthuridae occurred very early in the Tertiary (Clements et al., 2003). Proacanthurus (subfamily Acanthurinae) (Sorbini and Tyler, 1998), the morphologically primitive sister taxon of the recent genera Paracanthurus, Zebrasoma, Acanthurus, and Ctenochaetus, is of early-Eocene age as is the Nasinae genus Sorbinithurus (52 MY according to Tyler, 2000), demonstrating that the Nasinae existed at this time. In this study, molecular markers were used to generate the first comprehensive species-level phylogeny of the genus Naso. To infer both the position and the age of the most recent common ancestor (MRCA) of the genus Naso, an additional 18 acanthuroid species (from 5 genera) were sequenced for a nuclear marker, ETS2. Producing a complete phylogeny of the genus Naso allowed for the examination of the following four hypotheses, that can be developed from the work of previous authors. 1. The morphological tree of Borden (1998) is congruent with the molecular tree generated in this study. 2. Benthic foraging on macroscopic algae is plesiomorphic in the genus Naso (pelagic foraging is derived) (Borden, 1998; Winterbottom, 1993; Winterbottom and McLennan, 1993). 3. The presence of frontal horns and other cephalic structures does not reflect evolutionary relationships within the genus Naso (Borden, 1998). 4. All members of the subgenus Axinurus (N. minor, N. thynnoides and N. caeruleacauda) constitute a single morphologically defined group within the genus Naso as proposed by Smith (1966), and Randall (1994, 2001). In addition to the four hypotheses tested, we examine the timing of lineage diversification of the genus Naso to relate this to major plate tectonic and climatic events.

2. Materials and methods 2.1. Sampling Specimens for this study were either collected by spearing from a wide geographic range or acquired from museum collections (Table 1). At least two individuals per species with the exception of Naso reticulatus and Luvarus imperialis (Table 1) were analysed. In addition to samples from the 19 recognized Naso species, we had samples from one undetermined, possibly new (but undescribed) species of Naso. Tissue

223

samples (pectoral fin) of freshly speared fish were immediately placed into 80% ethanol or salt-saturated 20% DMSO (dimethyl sulfoxide). Muscle tissue of museum specimens had been preserved either in 95–100% ethanol or had been fixed in 10% formalin prior to preservation in 55% isopropanol. Total DNA was extracted from fin tissue using proteinase K digestion and standard salt–chloroform extraction procedures (Sambrook et al., 1989). DNA from formalin fixed museum tissue was extracted as described previously by Klanten et al. (in press). 2.2. Markers used and PCR amplification Two different categories of molecular markers were used to explore the acanthurid phylogeny: one nuclear marker (ETS2) and two mitochondrial (mt) markers (16S and cytochrome b). ETS2 is an intron from a nuclear oncogene. It was originally used as a conserved mammalian single-locus DNA marker (Lyons et al., 1997), but the primers, based on flanking exonic regions, also amplify fish DNA. This marker has been successfully used for serranids (van Herwerden et al., 2002). We used ETS2 by itself to confirm the relationship among all species of Naso and the remaining extant acanthurid genera and to identify and date the earliest diverging lineages of Naso using fossil dates. A nuclear marker has not featured previously in acanthurid studies, which used only mtDNA (Clements et al., 2003; Tang et al., 1999). All three markers (ETS2, 16S, and cyt b) were used to obtain a detailed resolution of species relationships within the genus Naso. The nuclear intron ETS2 (ETS2F 50 -AGC TGT GGC AGT TTC TTC TG-30 and ETS2R 50 -CGG CTC AGC TTC TCG TAG-30 (Lyons et al., 1997) and two mt markers, 16S rRNA (LR-J-12887 50 CCG GTC TGA ACT CAG ATC ACG T-30 and LRN-13398 50 -CGC CTG TTT ACC AAA AAC AT-30 (Simon et al., 1994) and cytochrome b (L14841 50 -AAA AAG CTT CCA TCC AAC ATC TCA GCA TGA TGA AA-30 and H15149 50 -AAA CTG CAG CCC CTC AGA ATG ATA TTT GTC CTC A-30 (Kocher et al., 1989) were PCR-amplified for all species. Each 20 ll PCR volume contained 2.5 mM Tris–Cl (pH 8.7), 5 mM KCl(NH4 )2 SO4 , 200 lM each dNTP, MgCl2 (1.5– 6.0 mM for fresh specimens), 10 lM each primer, 1 U of Taq Polymerase (Qiagen) and 10 ng template DNA. ETS2 and cyt b amplifications followed the same basic cycling: initial denaturing for 2 min at 94 °C was followed by 30 cycles, the first 5 cycles at 94 °C for 30 s, 30 s at the highest annealing temperature (ETS2 53 °C, cyt b 50 °C), followed by 1 min 30 s at 72 °C. An additional 25 cycles were performed, at the same denaturing and extension times and temperatures as before, but at a lower Ta °C (ETS2 51 °C, cyt b 48 °C). 16S rRNA was amplified using touchdown PCR in three phases of 5, 5, and 20 cycles per annealing temperature (51–49–47 °C).

224

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

Table 1 List of all species, source (including museum specimens indicated by *) and GenBank accession numbers for each marker used Species

Acanthurus blochii Acanthurus dussumieri Acanthurus nigricans Acanthurus nigrofuscus Acanthurus nubilus Acanthurus olivaceus Acanthurus pyroferus Acanthurus triostegus Acanthurus xanthopterus Ctenochaetus binotatus Ctenochaetus striatus Luvarus imperialis* Naso annulatus Naso brachycentron Naso brevirostris Naso caeruleacauda* Naso caesius Naso elegans Naso fageni Naso hexacanthus Naso lituratus Naso lopezi Naso maculatus Naso mcdadei* Naso minor Naso reticulatus* Naso thynnoides Naso tonganus Naso tuberosus Naso unicornis Naso vlamingii Naso undetermined Paracanthurus hepatus Prionurus microlepidotus Prionurus scalprum Zanclus cornutus Zebrasoma scopas Zebrasoma veliferum

Source

Lizard Isl., GBR Lizard Isl., GBR Lizard Isl., GBR Lizard Isl., GBR Kimbe Bay, PNG Lizard Isl., GBR Lizard Isl., GBR Lizard Isl., GBR Lizard Isl., GBR Lizard Isl., GBR Lizard Isl., GBR AMS I38647002 Lizard Isl., GBR Amirante, Seychelles Lizard Isl., GBR ROM 67197 Lizard Isl., GBR Amirante, Seychelles Ningaloo Reef, WA Lizard Isl., GBR Lizard Isl., GBR Selinog, Philippines Hawaii LFNMMST-62 Apo Isl., Philippines BPBM 23428 La Digue, Seychelles Lizard Isl., GBR Amirante, Seychelles Lizard Isl., GBR Lizard Isl., GBR One Tree Isl., GBR Amirante, Seychelles Broughton Isl., NSW Taiwan Lizard Isl., GBR Lizard Isl., GBR Lizard Isl., GBR

GenBank Accession Nos. ETS2

16S

Cyt b

AY264685 AY264686 AY264687 AY264688 AY264689 AY264690 AY264691 AY264692 AY264693 AY264694 AY264695 AY264678 AY264696 AY264697 AY264698 AY264699 AY264700 AY264701 AY264702 AY264703 AY264704 AY264705 AY264706 AY264707 AY264708 AY264709 AY264710 AY264711 AY264712 AY264713 AY264714 AY264715 AY264680 AY264681 AY264682 AY264679 AY264683 AY264684

– – – – – – – – – – – – AY264605 AY264606 AY264607 AY264608 AY264609 AY264610 AY264611 AY264612 AY264613 AY264614 AY264615 AY264616 AY264617 AY264618 AY264619 AY264620 AY264621 AY264622 AY264623 AY264624 – – – – – –

– – – – – – – – – – – – AY264643 AY264644 AY264645 AY264646 AY264647 AY264648 AY264649 AY264650 AY264651 AY264652 AY264653 AY264654 AY264655 AY264656 AY264657 AY264658 AY264659 AY264660 AY264661 AY264662 – – – – – –

AMS, Australian Museum, Sydney; ROM, Royal Ontario Museum, Canada; LFNMMST, National Museum of Science and Technology, Taiwan; BPBM, Bernice P. Bishop Museum, Hawaii; GBR, Great Barrier Reef; PNG, Papua New Guinea; NSW: New South Wales.

PCR products of 16S and cyt b were purified by isopropanol precipitation and sequenced directly in both directions using dye terminator chemistry (ABI) following manufacturerÕs instructions. PCR products of ETS2 had to be gel-purified on 2% agarose gels as two bands (500 and 300 bp sizes) appeared routinely. The 500 bp fragment was excised and purified in a column following manufacturerÕs protocol (Qiagen). The template was then sequenced directly as described above. Two examples of the 300 bp fragment were sequenced to check if this fragment was from the same gene family. The 300 bp sequences aligned with the 500 bp fragment but had several large deletions. Sequences were analysed on an automated ABI Prism 377 sequencer at Griffith University Sequencing Facility.

Museum specimens were amplified using a highfidelity Taq Polymerase (ProofStart Qiagen) following manufacturerÕs instructions (Qiagen). Magnesium sulphate (MgSO4 ) concentrations had to be re-optimized for each genus and marker, as did annealing temperatures. DNA template was increased to 100 ng and PCR products had to be gel-purified in 2% agarose gel, then column-purified following manufacturerÕs instructions (Qiagen) prior to direct sequencing as before (Klanten et al., in press). 2.3. Phylogenetic analysis All sequences (forward and reverse) were edited in GeneDoc (Nicholas and Nicholas, 1997) prior to alignment using ClustalW in DAMBE (Xia, 2000).

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

Gaps were treated as missing data. Saturation of nucleotide substitution was checked using a test by Xia et al. (2003), implemented in DAMBE version 4.2.8 (Xia, 2000). This test calculates if the observed index of substitution saturation (Iss) of nucleotide sequences is significantly lower than the critical value (Iss.c) calculated for the same sequences where Iss.c ¼ the value at which saturation occurs (Xia et al., 2003). The null hypothesis of this test is that there is saturation (Iss P Iss.c). The overall genus-level (38 taxa) relationship was inferred from ETS2 sequences using three different methods: maximum likelihood (ML), maximum parsimony (MP) implemented in PAUP* 4.0b10 (Swofford, 1998), and a Bayesian inference (BI) approach implemented in Mr. BAYES version 3.0B4 (Huelsenbeck and Ronquist, 2001). The best substitution model for all ML and BI analyses were identified using a likelihood approach implemented in Modeltest 3.06 (Posada and Crandall, 1998). Sequence data of all three markers were combined for members of the genus Naso and tested using the partition homogeneity test (see Section 3, also following Yoder et al., 2001) for congruence of tree lengths from each data partition, implemented in PAUP* 4.0b10. Trees of Naso were inferred using the same three methods. The subtree–pruning–regrafting (SPR) branch-swapping algorithm was used in the ML analyses (150 bootstrap replicates). Parsimony searches utilized a heuristic search (1000 bootstrap replicates, 50% majority rule consensus of all equally parsimonious, shortest trees) with the tree bisection reconnection (TBR) branch-swapping algorithm. The Bayesian analysis was performed using a Markov chain Monte Carlo (MCMC) search with 4 chains for 1 million generations. Trees were sampled every 100 generations, and the first 100,000 generations were discarded as burn-in. At the generic level (38 species) Zanclus cornutus and L. imperialis and at the species-level (21 species) Zebrasoma scopas were specified for outgroup rooting of phylogenetic trees. All resulting topologies were tested for significant difference from the best tree selected by PAUP* (p 6 0.05 significance level,  ln L for ML, tree length for MP) using Shimodaira–Hasegawa (ML) or Kishino– Hasegawa (MP) tests implemented in PAUP* 4.0b10 (Swofford, 1998). 2.4. Hypothesis testing With the exception of Dayton et al. (1994), Dayton (2001), and Clements et al. (2003), all hypotheses of species relationships within the genus Naso are developed from studies based on morphological data (Borden, 1998; Winterbottom and McLennan, 1993). We examined these hypotheses by testing whether topologically constrained trees were consistent with the hypotheses below. This was done, by evaluating the constrained trees against the molecular sequences. If

225

constrained trees were significantly worse than the best tree inferred from the molecular sequences, then the hypothesis was rejected. Hypothesis 1. The morphological tree of Borden (1998) is congruent with the molecular tree generated in this study. In order to compare these trees, four species (N. tonganus, Naso undetermined, N. elegans and N. reticulatus) were excluded from the combined molecular data set so that it included the same taxa as the morphological tree. BordenÕs, (1998, Fig. 7, p. 109) tree topology was evaluated against the molecular data using MacClade 4.03 (Maddison and Maddison, 2001). The resulting tree length was compared to the most parsimonious trees of the molecular analysis using the pairwise Kishino–Hasegawa test (KH) and nonparametric tests (Wilcoxon signed-ranks and sign test) implemented in PAUP* 4.0b10 (Swofford, 1998). Hypothesis 2. Benthic foraging on macroscopic algae is plesiomorphic within the genus Naso (pelagic foraging is derived) (Borden, 1998; Winterbottom and McLennan, 1993). Hypothesis 3. The presence of frontal horns and other cephalic structures does not reflect evolutionary relationships within the genus Naso (Borden, 1998). Hypothesis 4. All members of the subgenus Axinurus (N. minor, N. thynnoides, and N. caeruleacauda) constitute a single morphologically defined group within the genus (Randall, 2001; Smith, 1966) and are basal to other Naso species (Dayton, 2001). Hypotheses 2–4 were tested by entering trees with appropriate ‘‘backbone’’ constraints into PAUP* and allowing the program to find the best trees given the constraint, using a heuristic search under parsimony. Hypothesis 2 was tested by constraining all eight benthic foraging species (Table 2) to form a monophyletic clade. Similarly, to test Hypotheses 3 and 4, the four horned species and all members of the subgenus Axinurus (Table 2), respectively, were constrained into monophyletic clades. Lengths of the constrained trees were compared against that of the most parsimonious unconstrained tree using the KH and nonparametric tests implemented in PAUP* 4.0b10 (Swofford, 1998). 2.5. Timing of divergence Recently, Glazko and Nei (2003) showed that by concatenating several gene fragments a more robust estimate of divergence time can be obtained, especially if unlinked genes are included. To obtain estimates of the timing of lineage diversification in Naso, a two-step protocol was followed. The first step was to obtain a

226

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

Table 2 Adult foraging modes for all Naso species with source references Species

Adult foraging mode

Source reference

Naso annulatus

Pelagic and benthicb

Naso brachycentron Naso brevirostris

Benthica;b

Naso caeruleacauda Naso caesius Naso elegans Naso fageni Naso hexacanthus Naso lituratus

Pelagicc

Choat et al. (2002) Choat and Clements (1998) Choat et al. (2002) Randall (2002)

Naso Naso Naso Naso Naso Naso Naso Naso Naso

Pelagic Pelagic Benthica Pelagicc Pelagic Pelagicc Benthica Benthica Benthicab

lopezi maculatus mcdadei minor reticulatus thynnoides tonganus tuberosus unicornis

Benthic and pelagicb

Pelagic Benthica Benthica Pelagic Benthica

Naso vlamingii

Benthic and pelagic

Naso undetermined

Pelagic

Randall (2002) Randall (2002) This study Choat et al. (2002) Clements and Choat (1995) Randall (2002) Randall (2002) Randall (2002) Randall (2002)

estimating ages for each node, local rate estimates for each branch and a mean value of these divergence rates (substitution/site/unit time). Once a range of age estimates was obtained for the MRCA of Naso, the second step involved generating estimates of divergence times for each node in the tree of Naso species. For this, the same methodology as described above was followed using the tree inferred from combined sequence data (ETS2, 16S, and cyt b). This allowed a curve, illustrating the accumulation of extant lineages over evolutionary time, to be constructed for the genus Naso for each of the age estimates (MY). The curve was generated following a similar methodology developed by Barraclough and Nee (2001). The accumulation of lineages giving rise to all extant Naso species from the MRCA node to the tip of the tree was plotted against time (MY) (both axes used an arithmetic scale).

3. Results Randall (2002) Randall (2002) This study Choat et al. (2002) Choat et al. (2002)

a

Species constrained into benthic clade (Hypothesis 2). Horned clade (Hypothesis 3). c Axinurus clade (Hypothesis 4). * Pelagic: probably foraging on planktonic matter. b

range of dates (in MY) for the MRCA (most recent common ancestor) of the genus Naso. For this, the genuslevel phylogeny inferred using ETS2 sequences was used. Two nodes, one basal to Luvarus and the other to the family Acanthuridae were fixed (following Sanderson, 1997, 2003) with ages of 55 and 52 MY, respectively. These dates are based on ages of related fossil fish (Proluvarus necopinatus, 55 MY, Proacanthurus tenuis and Sorbinithurus sorbinii, both 52 MY) (see Bannikov and Tyler, 1995; Sorbini and Tyler, 1998; Tyler, 1999, 2000). To establish if sequences accumulate substitutions in a clock-like manner, they were analysed with the Langley– Fitch (LF) (Langley and Fitch, 1974) method, which uses maximum likelihood under the assumption of a molecular clock as implemented in r8s version 1.60 (Sanderson, 2003). This method generates a v2 value, which if nonsignificant (p > 0:05), indicates that sequences act in a clock-like manner. If however this hypothesis is rejected and sequences do not conform to a strict molecular clock, a nonparametric rate smoothing (NPRS) approach can be used (Sanderson, 1997) implemented in r8s. The latter approach relaxes the stringent assumptions of a molecular clock (Sanderson, 1997), while still

3.1. Sequences obtained One sequence per species was deposited in GenBank (Table 1), because both individuals analysed for a species had the same sequence. The alignment of the nuclear marker, ETS2 had 419 bp of which 27% were parsimony-informative. The first 81 sites of ETS2 were exonic and the remaining 338 bp (including alignment gaps) were from the intron. Gaps tend to be genusspecific. Only two gaps needed to be inserted to align sequences from Naso species, but as many as 30 gaps were required in the case of some Acanthurus species. The two mtDNA alignments, 16S and cyt b were 574 bp (including gaps) and 339 bp (no gaps), respectively. ETS2 combined with both mtDNA markers produced 1332 bp of sequence with 10% parsimony-informative sites. 3.2. Genus-level phylogeny ETS2 sequences were not saturated with an Iss-value, 0.214 < Iss.c of 0.791, significantly lower (p < 0:0001) than the critical value. The same topology was generated for the 38 taxa whether analysed by maximum parsimony (MP), maximum likelihood (ML) or Bayesian inference (BI). For MP, 259 trees were equally parsimonious with a tree length of 320, high consistency (CI: 0.772) and retention (RI: 0.927) indices. For ML (HKY + C substitution model) analysis, 9823 trees were generated, the best tree (selected by PAUP and checked by SH-test) had a log likelihood score of ln L )2356.171. The genus Naso was monophyletic and formed a sister group to the remaining five acanthurid genera (Paracanthurus, Zebrasoma, Prionurus, Acanthurus, and Ctenochaetus) (Fig. 1). This pattern of generic relationships is similar to that described by Clements et al.

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

(2003). Overall, bootstrap support for the main clades was high for all analyses (P91) (Fig. 1). The generic configuration was retrieved in association with the full phylogeny of Naso and provides a basis for determining the early history of the genus. While bootstrap support within the genus Naso was relatively weak for MP and ML analyses, Bayesian posterior probabilities generated stronger values for all major clades (Fig. 1). Members of the subgenus Axinurus (N. caeruleacauda, N. thynnoides and N. minor) were basal to the remaining Naso species. ETS2 sequences did not perform in a clock-like manner (v2 ¼ 164:98, d.f. 37, p < 0:001). Consequently,

227

estimates of ages leading to different genera were obtained by nonparametric rate smoothing (NPRS). The average divergence rate (%/MY) was 0.21% (with SD 0.55%) and a range of local rates: min 0.0 and max 3.3%. The MRCA of Naso had a max age of 52 MY and a min of 43.3 MY: thus the average was 47.6 MY. The MRCA of the genus Prionurus had an estimated age range between max 49.7 to min 45.9 MY and that of Zebrasoma ranged between max 36.9 to min 31.6 MY. The Acanthurus–Ctenochaetus clade was in-between the latter two genera, but younger than Naso with age estimates ranging between max. 47.8 to min. 31.1 MY (not shown).

Fig. 1. Phylogenetic tree of 38 acanthuroid species obtained by ML analysis of ETS2 sequences. Tree is outgroup-rooted with Luvarus and Zanclus. Bayesian posterior probabilities (in percent) are indicated above branches, bootstrap values (> 50) for MP and ML below branches (upper and lower values, respectively).

228

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

3.3. Species-level phylogeny All three markers were combined for the species-level phylogeny. The partition homogeneity test score was low (P ¼ 0:02) for all three markers (21 species) suggesting that there are incongruent tree length differences between partitions and that data partitions should not be combined. When only the two mtDNA markers were combined, the partition homogeneity test score P ¼ 0:3, indicated no incongruence between the two mtDNA markers. The only conflict between the nuclear (ETS2) and the two mtDNA markers was in the placement of N. reticulatus (compare Figs. 1 and 2). We therefore, ran the combined data set again excluding N. reticulatus and obtained a partition homogeneity test score of P ¼ 0:41. A single base change (A to G) produced this difference in placement of N. reticulatus with ETS2. Therefore, all three markers were combined despite the low partition homogeneity score (see also Yoder et al., 2001). No saturation was evident in the combined sequences, with Iss < Iss.c (0.1785 < 0.7779, p < 0:0001). The MP analysis produced two equally parsimonious trees of length 519, CI: 0.617, RI: 0.640. The ML analysis (GTR + I + C substitution model) generated 185 trees of which the best had a log likelihood score ln L )4455.934. BI retrieved a similar topology (Fig. 2) as MP and ML. The combined sequences resolved inter-specific relationships within the genus and segregated all Naso species as follows (Fig. 2). The subgenus Axinurus was basal but paraphyletic for MP and ML (not shown). BI however, grouped members of Axinurus in a monophyletic clade with little support (<50) (Fig. 2). The remaining species segregated into four distinct sub-clades, which were identified as: N. annulatus sub-clade (1), N. brevirostris sub-clade (2), N. elegans sub-clade (3), and N. maculatus sub-clade (4). Bootstrap values were high for all four sub-clades by all three methods (Fig. 2). Subclades 2–4 were collectively sister to the N. annulatus subclade (Fig. 2). For the MP and ML topologies, these three sub-clades (2–4) essentially arose from a polytomy. The N. annulatus sub-clade (1) consists of seven species, five of which are exclusively benthic browsing species (N. brachycentron, N. tonganus, N. tuberosus, N. mcdadei, and N. fageni), one is a pelagic foraging species (N. lopezi) and the large horned species, N. annulatus, forages on the benthos as a juvenile and in the pelagic zone as an adult (Fig. 2, also Table 2). The N. brevirostris sub-clade (2) consists of five species. N. brevirostris is similar in morphology (horned) and feeding ecology to N. annulatus, whilst N. vlamingii is a largely benthic-foraging species that also consumes fish faeces (Table 2). The remaining three species in this clade are pelagic foraging species (N. hexacanthus, N. caesius, and Naso undetermined). The N. elegans sub-clade (3) is an exclusively benthic foraging clade and includes both horned (N. unicornis) and non-horned (N. lituratus,

N. elegans) species. Members of the N. maculatus subclade (4) are probably pelagic foraging species although ecological details are unclear (Fig. 2, Table 2). The majority of Naso species occur throughout the Indo-Pacific Ocean (Fig. 2). For a few sister species such as N. tuberosus–N. tonganus (sub-clade 1) and N. elegans– N. lituratus (sub-clade 3), distinct distribution ranges (by ocean basins) have been recorded (Johnson, 2002; Randall, 2002). For other sister pairs (e.g., N. mcdadei– N. fageni and N. reticulatus–N. maculatus) only limited information about their distribution ranges is available (see also Kuiter and Debelius, 2001; Randall, 2002). Having established a well-resolved molecular phylogeny it is now possible to test various hypotheses about the evolution of morphology and foraging modes. 3.4. Hypothesis testing Hypothesis 1. The morphological tree of Borden (1998) is congruent with the molecular tree generated in this study. The forced tree topology was 147 steps longer, when evaluated against the best topology inferred from molecular sequences. It was thus significantly worse according to both KH and nonparametric tests (Wilcoxon signed-ranks and sign test) (Table 3). BordenÕs (1998) morphological phylogeny formed the basis for the argument that benthic foraging is plesiomorphic (Fig. 3A) and pelagic foraging is derived at one point in the phylogeny. This contrasts with the molecular phylogeny where pelagic foraging appears to be plesiomorphic (Figs. 2 and 3B). Therefore, alternative scenarios were tested statistically. These hypotheses were tested against the most parsimonious MP tree (tree length ¼ 519) obtained from the combined molecular sequences including all 20 Naso species with Z. scopas as outgroup. Hypothesis 2. Benthic foraging on macroscopic algae is plesiomorphic within the genus Naso (pelagic foraging is derived) (Borden, 1998; Winterbottom, 1993; Winterbottom and McLennan, 1993). This hypothesis was rejected for both cases, when the constrained topologies (either for a herbivorous or a pelagic clade) were compared to the MP molecular tree (Table 3). Hypothesis 3. The presence of frontal horns and other cephalic structures does not reflect evolutionary relationships within the genus Naso, i.e., this trait occurs independently in different clades (Borden, 1998). The existence of a cephalic horn clade was rejected (Table 3), thereby confirming BordenÕs hypothesis. Hypothesis 4. All members of the subgenus Axinurus constitute a single morphologically defined group within the genus (Randall, 2001; Smith, 1966) and are basal to

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

229

Fig. 2. Topology obtained using combined data (ETS2, 16S and cyt b) and Bayesian inference for 20 Naso species plus Z. scopas as the outgroup. Bayesian posterior probabilities (in percent) are indicated above branches, bootstrap values (>50) for MP and ML below branches (upper and lower values respectively). Numbers refer to the sub-clades: 1 for the N. annulatus sub-clade, 2 for the N. brevirostris sub-clade, 3 for the N. elegans subclade, 4 for the N. maculatus sub-clade. Foraging modes are indicated, pelagic as black and benthic as dotted lines. Body shapes are silhouetted and distribution ranges indicated. IPO, Indo-Pacific Ocean; PO, Pacific Ocean; IO, Indian Ocean; WIO, West Indian Ocean. *Two specimens collected from One Tree Island, Great Barrier Reef.

other Naso species (Dayton, 2001). This hypothetical tree topology was the only one not rejected by all three tests (Table 3). 3.5. Estimation of temporal pattern of lineage diversification for Naso species Age estimates for the MRCA of Naso were obtained from the nuclear marker (range 52–43.3, mean 47.6 MY) (Table 4). The combined sequences did not act in a clock-

like manner (v2 ¼ 59:18, d.f. 19, p < 0:001). A NPRS method was therefore used and divergence times were estimated for each node assuming each of three different ages in turn (Fig. 4B). The mean % divergence rates were: 0.063% (SD 0.027%) for a 52 MY MRCA; 0.068% (SD 0.03%) for a 47.6 MY MRCA and 0.075% (SD 0.033%) for the youngest, 43.3 MY, estimated age of the MRCA. Different curves were generated for each age estimate, which displayed similar divergence patterns over evolutionary time (Fig. 4B). Basal lineages in Naso (especially

230

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

Table 3 Tests of four hypotheses, including constrained tree length with most parsimonious molecular MP tree length of this study in brackets, the difference to the most parsimonious molecular tree (number of extra steps), p values for KH and nonparametric tests (Wilcoxon signed-ranks, winning-sites (sign) test) Hypotheses

Constrained tree length

No. of extra steps

KH-test (2 tailed)

Wilcoxon signed-rank test

Winning-sites (sign) test

1. Morphology 2. Foraging mode Benthic clade Pelagic clade 3. Horned clade 4. Axinurus monophyly

604 (457)

147

<0.0001

<0.0001

<0.0001

30 34 46 3

0.0001 <0.0001 <0.0001 0.57

0.0001 0.0001 <0.0001 0.56

0.0002 <0.0001 <0.0001 0.70

*

549 553 565 522

(519) (519) (519) (519)

Significance-level p < 0:001.

Fig. 3. BordenÕs (1998) morphological phylogeny enforced onto the best molecular MP tree for the combined ETS2, 16S, and cyt b data set. (A) BordenÕs (1998) constrained morphology tree. (B) Best parsimony tree for the equivalent species of Naso. At the time of BordenÕs analysis N. tonganus was identified as a single widespread species N. tuberosus. Foraging modes are indicated (refer to Fig. 2).

members of Axinurus), all exhibiting a pelagic foraging mode (plesiomorphic state), arose from the early Eocene to early Oligocene (nodes a!d, all three curves, Figs. 4A and B). The majority of the MRCAs to extant lineages however, arose throughout the Miocene (nodes g/h!r, Fig. 4B), subsequent to the Mi-1 glaciation. It is during the early Miocene that ancestors of extant species with a benthic foraging mode arose (e.g., node j, N. annulatus clade). All three curves suggest that there were two extended periods during which no extant lineages arose (a and b, Fig. 4B). The first of these was in the mid-late Oligocene and the second in mid-late Miocene (Fig. 4B). The most recent diversification of extant lineages occurred during the Plio/Pleistocene (nodes s and t) giving

rise to extant species displaying a pelagic foraging mode (Figs. 4A and B).

4. Discussion The primary issues arising from the phylogenetic reconstruction and history of the genus Naso are as follows: 1. The relationship between structure, external morphologies, foraging modes and lineage diversification. 2. The relationship between major plate tectonic, glaciations, palaeoclimatic events, and the evolutionary diversification of the genus.

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

231

Table 4 Estimated ages of the most recent common ancestor (MRCA) giving rise to Naso lineages as generated by nonparametric rate smoothing implemented in r8s (Sanderson, 2003) Nodes (Fig. 4)

Max. age (52 MY)

Mean age (47.6 MY)

Min. age (43.3 MY)

Giving rise to extant species

a b c d e f g h j k l m n o p q r s t

52 46.1 43.8 37.05 35.83 34.21 24.24 23.75 23.01 21.83 20.19 19.86 18.48 16.8 14.19 7.18 5.62 3.24 1.71

47.6 42.2 40.1 33.92 32.8 31.31 22.71 21.74 21.57 19.98 18.48 18.18 16.91 15.37 12.99 6.57 5.15 2.97 1.56

43.3 38.39 36.47 30.85 29.84 28.49 20.66 19.78 19.62 18.18 16.81 16.54 15.38 13.99 11.81 5.98 4.68 2.7 1.42

Root N. minor

4.1. Overall phylogeny, morphology and foraging modes The pattern of generic relationships within the family Acanthuridae, as retrieved by the nuclear marker ETS2, confirmed those of previous studies (Clements et al., 2003; Tang et al., 1999; Winterbottom and McLennan, 1993). The genus Naso is monophyletic and forms a sister clade to the remaining five extant acanthurid genera. The pattern retrieved by a new marker (ETS2) emphasizes the underlying agreement of generic relationships that is reflected by both morphological and molecular data (Clements et al., 2003). However, in marked contrast, the sequence data provided strong evidence against the species relationships within Naso as proposed by Borden (1998) on the basis of morphology. The difference between morphological and molecular topologies was most clearly seen in the relationship between benthic and pelagic foraging species. The primary morphology-based hypothesis of species relationships within Naso (benthic foraging is plesiomorphic) was clearly rejected by the sequence data. This does not, however, refute a hypothesis that benthic foraging may be plesiomorphic in other acanthurid genera (Winterbottom and McLennan, 1993), such as Acanthurus. Second, the sequence data supported the hypothesis that frontal horns do not reflect evolutionary relationships within the genus (Borden, 1998). The subgenus Axinurus was basal (as suggested by Dayton, 2001), in all trees examined moreover we could not reject the hypothesis of monophyly of Axinurus on the basis of available sequence data (Randall, 1994; Smith, 1966). Distinctive structural features, such as cephalic horns, tuberosities and scombriform morphology were dis-

N.caeruleacauda, N. thynnoides

N. lopezi N. brevirostris N. maculatus, N. reticulatus N. unicornis N. vlamingii N. N. N. N. N. N.

annulatus, N. brachycentron fageni, N. mcdadei tonganus, N. tuberosus elegans, N. lituratus hexacanthus caesius, Naso undetermined

tributed amongst the four sub-clades in a manner suggesting independent origins of or, in the case of scombriform morphology, retention of these features. Significantly, occurrence of species with streamlined scombriform morphologies within three of the Naso clades, including the basal subgenus Axinurus, suggests that this plesiomorphic morphology has been retained and other morphologies have arisen repeatedly and independently. The distinctive morphological features that characterize Naso appear to be prone to convergence. They do not provide a reliable basis for classification within the genus. 4.2. Rate estimates for lineage diversification The average rate estimate obtained for Naso sequences is low (0.065% divergence/MY) compared to other fish studies (e.g., between 0.05 and 1.32% divergence/MY Alvez-Gomes, 1999; Bermingham et al., 1997; Kumazawa and Nishida, 2000; Machordom and Doadrio, 2001), which used at least two mt genes and also used either fossil calibrations and/or geological events. However, those studies based their calibrations exclusively on mtDNA sequences whereas this study concatenated three gene regions (including one nuclear) (Glazko and Nei, 2003). 4.3. Timing of divergence, cladogenesis, and biogeographic history Complex perciform lineages that constitute the modern reef fish fauna proliferated in the early to mid Eocene, although basal perciform lineages must have

232

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

Fig. 4. (A) Same topology as shown in Fig. 2, but presented as an ultrametric tree. Foraging modes are indicated (refer to Fig. 2). Small letters (a–t) indicate nodes leading to lineages. (B) Curves illustrating the diversification of Naso lineages through evolutionary time. The x-axis depicts relative time since the ancestral root node, values in MY. The y-axis illustrates the number of Naso lineages. Eustatic sea levels with present day level (dotted line), Oi-1 and Mi-1 glaciation times (arrows) modified from Haq et al. (1987), Hallam (1984), Pickering (2000), and Zachos et al. (2001a) are shown. Thick bars, a and b, indicate periods of no diversification. Small letters (a–t) as above (see A) indicated on the curve are the same across the horizontal plane for all three curves.

been present earlier (Bellwood, 1996; Bellwood and Wainwright, 2002; Tyler et al., 1989). Among the morphologically specialized Eocene acanthuroid fishes, were 11 genera of acanthurids, 2 genera of luvarids, 3 genera of siganids, and a zanclid (Bannikov and Tyler, 1995; Tyler, 1997, 1999, 2000). The tropical shallow-water reef environment of this period was particularly favourable to reef fishes. Sea levels were substantially higher (150– 200 m) than at present (Hallam, 1984; Haq et al., 1987) and sea-surface temperatures in low latitude habitats

were uniformly high (also called greenhouse period Stanley, 2001) in association with the 55 MY early Eocene climatic optimum (Zachos et al., 2001a). Biological reef systems, including both algae and coral dominated reefs, achieved their greatest latitudinal spread of the Tertiary at this time (Kiessling, 2001). This period of rapid proliferation of reef fish lineages was followed by relative stasis of the fauna (Bellwood, 1996; Bellwood and Wainwright, 2002) for much of the remainder of the Tertiary. More recent episodes of diversification of reef

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

fishes occurred during the period of fluctuating sea levels throughout the Pliocene/Pleistocene (Clements et al., 2003; McMillan and Palumbi, 1995; Palumbi, 1997). The basal Naso species that arose during the Eocene were pelagic foragers. The subsequent diversification leading to the variety of morphological and foraging modes that characterizes the present fauna occurred post-Eocene/Oligocene. The critical episode of lineage diversification of Naso commenced in the very late Oligocene/early Miocene. This coincides with the brief, but intense Mi-1 glaciation (Zachos et al., 2001b), a period of profound modification to shallow-water marine habitats associated with increased glaciations, falling temperatures, fluctuating sea levels and substantial increase in ocean productivity (Zachos et al., 2001a,b). Algal reefs may have dominated during such glacial periods (Kiessling, 2001), however the mid Miocene saw a rise in sea levels and an increase in coral reef systems covering a range similar to that of present day tropical coral reefs (Kiessling, 2001). Ancestors of generalist species such as N. brevirostris (node h, Fig. 4), which display both benthic and pelagic foraging modes, arose for the first time in the early Miocene. The MRCA leading to the N. annulatus clade (node j), the majority of which are benthic feeding, diverged during the early Miocene, probably an example of ecological diversification in the new algal-reef dominated environment. Subsequent lineages arising during the remainder of the Miocene include species that forage exclusively on algae (N. elegans clade), species that are pelagic foraging (N. maculatus clade) and generalist species (N. vlamingii and N. annulatus). All fossil representatives of the subfamily Nasinae with the exception of Eonaso (sister to Naso) occur in the Tethyan, Sorbinithurus (Eocene), Arambourgthurus (Oligocene) or western Pacific deposits Marosichthys (Miocene) (Tyler, 2000). The fossil acanthurid, Eonaso, closely related to Naso (Tyler, 2000) was recorded from the tropical Atlantic Ocean (Antilles, Caribbean) (Tyler and Sorbini, 1998; Tyler, 2000). The age of this important fossil remains unresolved. As with many groups of reef fishes, early diversification of the Nasinae occurred in the Tethys (Tyler, 2000), with the most subsequent taxa colonizing eastwards into the Indo-Pacific Ocean. The most parsimonious explanation accounting for the location of Eonaso, is westward colonization into the tropical Atlantic and subsequent extinction. It seems likely that the extinction of Eonaso as the only Nasinae species in the tropical Atlantic occurred during events associated with the Mi-1 glaciation, a period when widespread extinctions of corals occurred in the Caribbean (Zachos et al., 2001a). It is of interest to note that the well-preserved fossil of Eonaso shows clear evidence of scombriform caudal structure but an external morphology consistent with benthic foraging (Tyler and Sorbini, 1998).

233

4.4. Comparison to other acanthurids The pattern of diversification of lineages within the genus Naso contrasts with the evolutionary pattern of reef fishes in general and other acanthurids in particular. Evolution in the Acanthuridae was characterized by two episodes of rapid diversification. First, in the early to mid-Eocene a complex of benthic browsing taxa arose (Blot and Tyler, 1991; Tyler et al., 1989) and second, there was a much more recent proliferation of grazing species within the genera Acanthurus and Ctenochaetus (Clements et al., 2003; Tang et al., 1999). In contrast to the pattern of rapid Eocene diversification seen in other acanthuroids, the curve for the genus Naso suggests a slow accumulation of lineages during the Eocene, none during the Oligocene, followed by an increase during the Miocene. Unlike Acanthurus and Ctenochaetus there is little evidence of rapid and recent speciation of Naso. With the exception of Sorbinithurus (Tyler, 1999, 2000), the subfamily Nasinae was not represented in the Eocene Monte Bolca beds. This may reflect the pelagic foraging mode of the basal lineages. Similar conclusions were reached concerning the early evolution of the luvarid acanthuroids (Bannikov and Tyler, 1995). The main episode of diversification of Naso occurred during the Miocene, represented by a period when shallow reef systems at all latitudes were subject to major disturbances and faunal turnover (Prothero, 1994). Although many of the present day species of Naso are consistently associated with coral reefs, with the exception of the smaller pelagic species, they are rarely abundant compared to Acanthurus and Ctenochaetus, and many taxa (e.g., N. fageni and N. mcdadei) occur in deeper non-reef environments, unlike Acanthurus and Ctenochaetus that are closely associated with reef environments.

5. Conclusion On the basis of this comprehensive molecular phylogeny, tests of morphology and foraging modes, the extensive acanthurid fossil record, and the Cenozoic record of tectonic and climatic change we conclude: The genus Naso had an extended evolutionary history from the Eocene to the present day. The most parsimonious interpretation of this history is that the basal group was represented by species displaying a pelagic foraging mode with scombriform morphology and caudal propulsive unit. The distinctive caudal structure has been retained throughout the genus and partly explains the repeated appearance of the pelagic foraging mode amongst the sub-clades. Over the first 25 MY of this evolutionary history little diversification of body form or foraging mode occurred. Subsequent diversification resulted in a number of clades in which species

234

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235

displaying benthic foraging modes (including those with distinctive cephalic horns) arose independently. The greatest diversification of Naso lineages occurred during the Miocene coincident with major changes to global climate, ocean and reef ecosystems. This pattern of diversification stands in contrast to patterns observed in other acanthurid taxa in which major episodes of diversification occurred in both the Eocene and the Pliocene/Pleistocene (Clements et al., 2003).

Acknowledgments We thank two anonymous reviewers for constructive suggestions that improved this publication. We thank several museums and institutions for loan of material: Australian Museum Sydney; Royal Ontario Museum, Canada; Smithsonian Institution National Museum of Natural History, Washington; Bishop Museum, Hawaii; J.L.B. Smith Institute of Ichthyology, Durban; CSIRO Division of Fisheries, Hobart; Queensland Museum, Brisbane; Silliman University Angelo King Center for Research and Environmental Management, Philippines and National Museum of Science and Technology, Taiwan. Additional specimens were collected by J. Ackerman, W. Robbins, G.P. Jones, J. Kritzer, D.R. Robertson, J. Johnson, J. Randall, K.D. Clements, L.-S. Chen, A. Alcala, A. Maypa, and A. Lo-Yat. The manuscript was improved by discussions with K. Clements, M. van Oppen, R. Winterbottom, L. Atkinson, and P. Munday. Assistance and support in the laboratory by L. Bay, J. Gardiner and C. Dudgeon. This project was supported and funded by The Australian Institute of Marine Science (M. Meekan); The Seychelles Fishing Authority (E. Grandcourt and J. Robinson); The Queensland Government/Smithsonian Institution Collaborative Research Program on Reef Fishes; James Cook University internal funding scheme and The Smithsonian Institution. James Cook University Ethics Approval No. is A503.

References Alvez-Gomes, J.A., 1999. Sytematic biology of gymnotiform and mormyriform electric fishes: phylogenetic relationships, molecular clocks and rates of evolution in the mitochondrial rRNA genes. J. Exp. Biol. 202, 1167–1183. Bannikov, A.F., Tyler, J.C., 1995. Phylogenetic revision of the fish families Luvaridae and Kushlukiidae (Acanthuroidei), with a new genus and two new species of Eocene Luvarids. Smithson Contr. Paleobiol. 81, 45. Barraclough, T.G., Nee, S., 2001. Phylogenetics and speciation. Trends Ecol. Evol. 16, 391–399. Bellwood, D.R., 1996. The Eocene fishes of Monte Bolca—the earliest coral reef fish assemblage. Coral Reefs 15, 11–19. Bellwood, D.R., Wainwright, P.C., 2002. The history and biogeography of fishes on coral reefs. In: Sale, P.F. (Ed.), Coral Reef Fishes,

Dynamics and Diversity in a Complex Ecosystem. Academic Press, San Diego, pp. 5–32. Bermingham, E., McCafferty, S.S., Martin, A.P., 1997. Fish biogeography and molecular clocks: perspectives from the Panamanian Isthmus. In: Kocher, T.D., Stepien, C.A. (Eds.), Molecular Systematics of Fishes. Academic Press, California, pp. 113–128. Bernardi, G., Crane, N.L., 1999. Molecular phylogeny of the humbug damselfishes inferred from mtDNA sequences. J. Fish Biol. 54, 1210–1217. Bernardi, G., Robertson, D.R., Clifton, K.E., Azzurro, E., 2000. Molecular systematics, zoogeography, and evolutionary ecology of the Atlantic Parrotfish genus Sparisoma. Mol. Phylogenet. Evol. 15, 292–300. Blot, J., Tyler, J.C., 1991. New genera and species of fossil surgeon fishes and their relatives (Acanthuroidei, Teleostei) from the Eocene of Monte Bolca, Italy, with application of the Blot formula to both fossil and recent forms. Studi e Ricerche sui Giacimenti Terziari di Bolca 6, 13–92. Borden, W.C., 1998. Phylogeny of the Unicornfishes (Naso, Acanthuridae) Based on Soft Anatomy. Copeia 1, 104–113. Brooks, D.R., McLennan, D.A., 2002. The Nature of Diversity. The University of Chicago Press, Chicago. Choat, J.H., Clements, K.D., 1998. Vertebrate herbivory in marine and terrestrial environments: a nutritional ecology perspective. Annu. Rev. Ecol. Syst. 29, 375–403. Choat, J.H., Clements, K.D., Robbins, W.D., 2002. The trophic status of herbivorous fishes on coral reefs—I: dietary analyses. Mar. Biol. 140, 613–623. Clements, K.D., Choat, J.H., 1995. Fermentation in tropical marine herbivorous fishes. Physiol. Zool. 68, 355–378. Clements, K.D., Gray, R.D., Choat, J.H., 2003. Rapid evolutionary divergence in reef fishes of the family Acanthuridae (Perciformes; Teleostei). Mol. Phylogenet. Evol. 26, 190–201. Craig, M.T., Pondella, D.J., Frank, J.P.C., Hafner, J.C., 2001. On the status of the Serranid fish genus Epinephelus: evidence for paraphyly based upon 16S rDNA sequence. Mol. Phylogenet. Evol. 19, 121–130. Dayton, C., Santayana, A.C., Lacson, J.M., 1994. Genetic evidence for reproductive isolation of the recently described unicornfish Naso caesius and its sibling N. hexacanthus. Mar. Biol. 118, 551–554. Dayton, C.L., 2001. Genetic Evolution among selected members of the genus Naso (Nasinae), unicornfishes from Guam. Mar. Biol. 139, 771–776. Glazko, G.V., Nei, M., 2003. Estimation of divergence times for major lineages of primate species. Mol. Biol. Evol. 20, 424–434. Hallam, A., 1984. Pre-quaternary sea-level changes. Annu. Rev. Earth Planet. Sci. 12, 205–243. Haq, B.U., Hardenbol, J., Vail, P.R., 1987. Chronology of fluctuating sea levels since the Triassic. Science 235, 1156–1167. Huelsenbeck, J.P., Ronquist, F., 2001. MR. BAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17, 754–755. Johnson, J.W., 2002. Naso mcdadei, a new species of unicornfish (Perciformes: Acanthuridae), with a review of the Naso tuberosus species complex. Aust. J. Zool. 50, 293–311. Jones, R.S., 1968. Ecological relationships in Hawaiian and Johnston Island Acanthuridae (Surgeonfishes). Micronesica 4, 309–361. Kiessling, W., 2001. Paleoclimatic significance of Phanerozoic reefs. Geology 29, 751–754. Klanten, S.O., van Herwerden, L., Choat, J.H., in press. Acquiring DNA sequences from formalin fixed museum specimens: application for reef fish. Bull. Mar. Sci. Kocher, T.D., Thomas, W.K., Meyer, A., Edwards, S.V., Paabo, S., Villablanca, F.X., Wilson, C., 1989. Dynamics of mitochondrial DNA evolution in animals: amplification and sequencing with conserved primers. Proc. Natl. Acad. Sci. 86, 6196–6200. Kuiter, R.H., Debelius, H., 2001. Surgeonfishes. Rabbitfishes and their Relatives. TMC Publishers, Chorleywood, UK.

S.O. Klanten et al. / Molecular Phylogenetics and Evolution 32 (2004) 221–235 Kumazawa, Y., Nishida, M., 2000. Molecular phylogeny of osteoglossids: a new model for Gondwanian origin and plate Tectonic transportation of the Asian Arowana. Mol. Biol. Evol. 17, 1869– 1878. Langley, C.H., Fitch, W.M., 1974. An examination of the constancy of the rate of molecular evolution. J. Mol. Evol. 3, 161–177. Lo Galbo, A.M., Carpenter, K.E., Reed, D.L., 2002. Evolution of tropic types in emperor fishes (Lethrinus, Lethrinidae, Percoidei) based on cytochrome b gene sequence variation. J. Mol. Evol. 54, 754–762. Lyons, L.A., Laughlin, T.F., Copeland, N.G., Jenkins, N.A., Womack, J.E., OÕBrien, S.J., 1997. Comparative anchor tagged sequences (CATS) for integrative mapping of mammalian genomes. Nat. Genet. 15, 47–56. Machordom, A., Doadrio, I., 2001. Evidence of a Cenozoic BelticKabilian connection based on freshwater fish phylogeography (Luciobarbus, Cyprinidae). Mol. Phylogenet. Evol. 18, 252–263. Maddison, D.R., Maddison, W.P., 2001. MacClade 4: Analysis of phylogeny and character evolution, version 4.03. Sinauer Associates, Sunderland, MA. McMillan, W.O., Palumbi, S.R., 1995. Concordant evolutionary patterns among Indo-West Pacific Butterflyfishes. Proc. R. Soc. Lond. B 260, 229–236. Nicholas, K.B., Nicholas, H.B.J., 1997. GeneDoc: a tool for editing and annotating multiple sequence alignments, version 2.5. Palumbi, S.R., 1997. Molecular biogeography of the Pacific. Coral Reefs 16 (Suppl.), S47–S52. Pickering, K.T., 2000. The Cenozoic world. In: Culver, S.J., Rawson, P.F. (Eds.), Biotic Response to Global Change. Cambridge University Press, pp. 20–34. Posada, D., Crandall, K.A., 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14, 817–818. Prothero, D.R., 1994. The late Eocene–Oligocene extinctions. Annu. Rev. Earth Planet. Sci. 22, 145–165. Randall, J.E., 1994. Unicornfishes of the subgenus Axinurus (Perciformes: Acanthuridae: Naso), with description of a new species. Copeia 1, 116–124. Randall, J.E., 2001. Naso reticulatus, a New Unicornfish (Perciformes: Acanthuridae) from Taiwan and Indonesia, with a key to the species of Naso Zool. Studies 40, 170–176. Randall, J.E., 2002. Surgeonfishes of HawaiÕi and the World. Mutual Publishing and Bishop Museum Press. Ruber, L., Van Tassell, J.L., Zardoya, R., 2003. Rapid speciation and ecological divergence in the American seven-spined gobies (Gobiidae, Gobiosomatini) inferred from a molecular phylogeny. Evolution 57, 1584–1598. Sambrook, J., Fritsh, E.F., Maniatis, T., 1989. Molecular Cloning, A Laboratory Manual. Cold Spring Harbour Laboratory Press, pondy. Sanderson, M.J., 1997. A nonparametric approach to estimating divergence times in the absence of rate constancy. Mol. Biol. Evol. 14, 1218–1231. Sanderson, M.J., 2003. r8s: inferring absolute rates of molecular evolution and divergence times in the absence of a molecular clock. Bioinformatics 19, 301–302. Simon, C., Frati, F., Beckenbach, A., Crespi, B., Liu, H., Flook, P., 1994. Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Ann. Entomol. Soc. Am. 87, 651–701. Smith, J.L.B., 1966. Fishes of the sub-family Nasinae with a synopsis of the Prionurinae. Ichthyolog. Bull. 32, 635–682. Sorbini, L., Tyler, J.C., 1998. A new species of the Eocene surgeon fish genus Pesciarichthys from Monte Bolca, Italy (Acanthuridae), with comments on Caudal Peduncle Armature and Supraneurals in Acanthurids. Studi e Ricerche sui Giacimenti Terziari di Bolca 7, 21–34. Stanley, G.D.J., 2001. In: Landman, N.H., Jones, D.S. (Eds.), The History and Sedimentology of Ancient Reef Systems. Kluwer Academic/Plenum Publisher, New York.

235

Streelman, J.T., Alfaro, M., Weastneat, M.W., Bellwood, D.R., Karl, S.A., 2002. Evolutionary history of the parrotfishes: biogeography, ecomorphology, and comparative diversity. Evolution 56, 961–971. Swofford, D.L., 1998. PAUP* Phylogenetic Analysis Using Parsimony (* and Other Methods). 4.0b10. Sinauer Associates, Sunderland, MA. Tang, K.L., Berendzen, P.B., Wiley, E.O., Morrissey, J.F., Winterbottom, R., Johnson, G.D., 1999. The phylogenetic relationships of the Suborder Acanthuroidei (Teleostei: Perciformes) based on molecular and morphological evidence. Mol. Phylogenet. Evol. 11, 415–425. Tang, K.L., 2001. Phylogenetic relationships among Damselfishes (Teleostei: Pomacentridae) as determined by mitochondrial DNA data. Copeia 3, 591–601. Thacker, C.E., 2003. Molecular phylogeny of the gobioid fishes (Teleostei: Perciformes: Gobioidei). Mol. Phylogenet. Evol. 26, 354–368. Tyler, J.C., 1970. Osteological aspects of interrelationships of surgeon fish genera (Acanthuridae). Proc. Acad. Natl. Sci. Philadelphia 122, 87–124. Tyler, J.C., 1997. The Miocene fish Marosichthys, a putative tetraodontiform, actually a perciform surgeon fish (Acanthuridae) related to the recent Naso. Beaufortia 47, 1–10. Tyler, J.C., 1999. A new genus and species of surgeon fish (Acanthuridae) with a unique dorsal-fin Pterygiophore arrangement from the Eocene of Monte Bolca, Italy. Studi e Ricerche sui Giacimenti Terziari di Bolca 7, 245–256. Tyler, J.C., 2000. Arambourgthurus, a new genus of hypurostegic surgeonfish (Acanthuridae) from the Oligocene of Iran, with a phylogeny of Nasinae. Geodiversitas 22, 525–537. Tyler, J.C., Bannikov, A.F., 2000. A new species of the surgeon fish genus Tauichthys from the Eocene of Monte Bolca, Italy (Perciformes, Acanthuridae). Bollettino del Museo Civico di Storia Naturale di Verona 24, 29–36. Tyler, J.C., Sorbini, L., 1998. On the relationships of Eonaso, an Antillean fossil surgeon fish (Acanthuridae). Studi e Ricerche sui Giacimenti Terziari di Bolca 7, 35–42. Tyler, J.C., Johnson, G.D., Nakamura, I., Collette, B.B., 1989. Morphology of Luvarus imperialis (Luvaridae), with a phylogenetic analysis of the Acanthuroidei (Pisces). Smithson. Contr. Zool. 485, 1–78. van Herwerden, L., Davies, C.R., Choat, J.H., 2002. Phylogenetic and evolutionary perspectives of the Indo-Pacific grouper Plectropomus species on the Great Barrier Reef, Australia. J. Fish Biol. 60, 1591– 1596. Wang, H.Y., Tsai, M.P., Dean, J., Lee, S.C., 2001. Molecular phylogeny of gobioid fishes (Perciformes: Gobioidei) based on mitochondrial 12S rRNA sequences. Mol. Phylogenet. Evol. 20, 390–408. Winterbottom, R., 1993. Myological evidence for the phylogeny of recent genera of surgeonfishes (Percomorpha, Acanthuridae), with comments on the Acanthuroidei. Copeia 1, 21–39. Winterbottom, R., McLennan, D.A., 1993. Cladogram versatility: evolution and biogeography of Acanthuroid fishes. Evolution 47, 1557–1571. Xia, X., 2000. Data Analysis in Molecular Biology and Evolution (DAMBE), version 4.2.8. Xia, X.H., Xie, Z., Salemi, M., Chen, L., Wang, Y., 2003. An index of substitution saturation and its application. Mol. Phylogenet. Evol. 26, 1–7. Yoder, A.D., Irwin, J.A., Payseur, B.A., 2001. Failure of the ILD to determine data combinability for Slow Loris phylogeny. Syst. Biol. 50, 408–424. Zachos, J., Pagani, M., Sloan, L., Thomas, E., Billups, K., 2001a. Trends, rhythms, and aberrations in global climate 65 Ma to present. Science 292, 686–692. Zachos, J.C., Shackleton, N.J., Revenaugh, J.S., Paelike, H., Flower, B.P., 2001b. Climate response to orbital forcing across the Oligocene–Miocene boundary. Science 292, 274–278.