Detecting reticulate relationships among diploid Leucanthemum Mill. (Compositae, Anthemideae) taxa using multilocus species tree reconstruction methods and AFLP fingerprinting

https://doi.org/10.1016/j.ympev.2015.06.003Get rights and content

Highlights

  • Species tree and species network reconstruction in diploid Leucanthemum species.

  • High incongruence among ten nuclear and plastid gene trees based on 454 pyrosequencing data.

  • Novel calculation of taxon-wise ‘hybrid scores’ pinpoints to taxa of possible hybrid origin.

  • Simulation of ‘hybrid scores’ assuming exclusively incomplete lineage sorting as cause of gene tree incongruence.

  • Early-diverging species with low hybrid signal; recent radiations of Leucanthemum with pronounced signal of hybridisation.

Abstract

We examined the evolutionary history of the diploid representatives of the genus Leucanthemum Mill. (Compositae, Anthemideae), which constitutes an extensive polyploid complex comprising around 41 species with ploidy levels ranging from 2x to 22x. The inference of phylogenetic relationships even on the diploid level is complicated in this genus due to the overlay of hybridisation and incomplete lineage sorting processes leading to incongruence among gene trees based on nuclear and plastid sequence information. Species tree and network reconstructions were based on gene trees from nine low-copy nuclear markers and the concatenated sequence information for five intergenic spacer regions of the chloroplast genome, either sequenced by Roche 454 pyrosequencing techniques or traditional Sanger sequencing techniques. Additional phylogenetic information came from multi-locus AFLP-fingerprinting of representative individuals of all diploid taxa under study and the subsequent analysis of AFLP patterns with Bayesian clustering and network reconstruction methods. To distinguish between hybridisation and incomplete lineage sorting, we developed and utilized a new ‘hybrid index’ calculation for individual taxa of the data set, which was compared to a simulated null-distribution assuming the occurrence of incomplete lineage sorting alone for pinpointing taxa with a significant hybrid signal. As a result, two species groups with contrasting patterns of gene flow and/or hybrid speciation signals could be identified in the diploids of Leucanthemum: (a) an early-diverging stock of allopatrically distributed diploid species with a lack of evidence for recent hybridisation events among its members and (b) a more recently radiated taxon assemblage with morphologically less clearly circumscribed taxa and a pronounced signal of gene flow among lineages and several candidate taxa, for which a homoploid hybrid origin may be considered.

Introduction

It is hardly possible to overestimate the importance of hybridisation in plant evolution and to elude the argumentation of Oberprieler (2014) that Biology’s First Law (McShea and Brandon, 2010) saying that “in the absence of selection and constraint, complexity – in the sense of differentiation among parts – will tend to increase” should be augmented by a second principle (and maybe Biology’s Second Law) that complexity does not only increase through differentiation and divergence alone but also through genetic exchange, (re)combination, and phylogenetic reticulation. We are presently experiencing a shift in perspective from the view on hybridisation as a merely destructive process that could lead to a reversal of differentiation and a loss of biodiversity toward an enforced appreciation of hybridisation as a constructive and even creative process in evolutionary biology (Abbott et al., 2013, Yakimowski and Rieseberg, 2014). With an estimated frequency of at least 25% of species that hybridise with each other (Mallet, 2005), the plant kingdom represents a well-suited domain of life for studying this paramount evolutionary process.

While the study of hybridisation processes as “collision of species” finds its analogue in the particle colliders in physics, which allow the study of the internal structure of the components of matter (Buerkle and Lexer, 2008), methods of phylogenetic reconstruction based on molecular evidence could be seen as our instruments equivalent to telescopes in astrophysics that allow us to view back into time and the evolutionary history of an organism group. However, as in physics, the two approaches are connected and microevolutionary processes like speciation and hybridisation events leave their footprint in the evolution of genomic markers that are used in turn to infer the macroevolutionary patterns of phylogenetic relationships. Two natural processes are especially noteworthy that could lead to a blurring of phylogenetic patterns among lineages (species trees) as reconstructed from underlying (and often discordant) evolutionary histories of individual molecular markers (gene trees): incomplete lineage sorting (ILS; Hudson, 1983, Tajima, 1983, Takahata, 1995, Rannala and Yang, 2008) and gene flow among lineages (Slatkin and Maddison, 1989).

Since ‘total evidence’ approaches with concatenated sequence data from markers with high levels of discordance may produce robust and well-supported, but inaccurate phylogenetic reconstructions (Kubatko and Degnan, 2007, Weisrock et al., 2012), an increasing number of methods have been proposed to estimate the correct species tree without concatenation of sequence data, especially for those cases in which ILS is the reason for incongruence among gene trees (Maddison and Knowles, 2006, Mossel and Roch, 2010, Liu, 2008, Than and Nakhleh, 2009, Liu et al., 2009, Heled and Drummond, 2010, Knowles and Kubatko, 2010, Fan and Kubatko, 2011, Leaché and Rannala, 2011, Camargo et al., 2012). Despite continuous efforts to find methods that distinguish between the effects of ILS and hybridisation (Sang and Zhong, 2000, Holland et al., 2008, Maureira-Butler et al., 2008, Joly et al., 2009, Kubatko, 2009, Kelly et al., 2010, Gerard et al., 2011, Blanco-Pastor et al., 2012, De Villiers et al., 2013, Ramadugu et al., 2013) species tree inference jointly considering the two processes remains a great challenge (Leaché et al., 2014) and constitutes a very active field in present phylogenetic systematics; especially in the light of increasing simplifications in the process of gaining huge amounts of sequence data for gene tree reconstructions through next-generation sequencing techniques (Glenn, 2011).

The genus Leucanthemum Mill. (Compositae, Anthemideae) is a large polyploid complex comprising 42 species (The Euro+Med Plantbase, 2014) with ploidy levels ranging from diploid (2x) to dodecaploid (12x), and one species [L. lacustre (Brot.) Samp.) from Portugal] even showing a chromosome number of 2n = 22x = 198 (docosaploid level). The genus is distributed all over the European continent, with one species (L. ircutianum DC.) reaching Siberia and some species introduced to many temperate regions of the northern and southern hemisphere (Meusel and Jäger, 1992). While the reticulate evolutionary history of the genus caused by allopolyploidy was demonstrated in a number of studies based on molecular data (Oberprieler et al., 2011b, Oberprieler et al., 2014, Greiner et al., 2012, Greiner et al., 2013), results obtained in the course of a recent study of sequence variation at the external transcribed spacer region of the nuclear ribosomal repeat (nrDNA ETS) in diploid representatives of the genus suggested either gene flow among or a homoploid hybrid origin of some of these species (Oberprieler et al., 2014): ETS ribotypes realized in Leucanthemum diploids were found to fall into two clusters, a plesiomorphic ETS ribotype cluster closely related to ETS ribotypes of outgroup genera (called the ‘green’ ETS ribotype cluster in Oberprieler et al., 2014) and an apomorphic cluster (‘red’ ETS ribotype cluster). While some diploid species were found being fixed for either of the two types, others exhibited an additive pattern of the two types, which was interpreted as being due to hybridisation and gene flow among diploids in the former study (Oberprieler et al., 2014).

In the present contribution, we use AFLP fingerprinting and species tree inference based on nine nuclear and one plastid gene trees to test the hypothesis that the evolutionary history of Leucanthemum diploids was influenced by gene flow caused by hybridisation or even homoploid hybrid speciation. We were especially interested in finding ways to disentangle effects of incomplete lineage sorting and hybridisation as causes for incongruence among multilocus gene trees in order to quantify the amount of incongruence caused by hybridisation alone and to pinpoint potential hybrid lineages/taxa that exceedingly contribute to a hybrid signal in the data set. In contrast to a method used by Maureira-Butler et al., 2008, Blanco-Pastor et al., 2012, Ramadugu et al., 2013 to detect potential hybrids by examining the effect of sequential taxon deletion on gene tree incongruence, we infer taxon-specific hybrid index scores by a new method based on triplet-permutations and the computation of likelihood scores for the resulting three-taxon species trees or networks following calculations described in Yu et al. (2012). Significance of the taxon-specific hybrid index scores (resulting from the joint effects of incomplete lineage and hybridisation) is then tested by comparing them with the corresponding values from coalescence simulations done under the assumption of the presence of incomplete lineage sorting alone.

Section snippets

Taxon sampling and DNA extraction

Either silica-gel dried plant material collected in the field or herbarium specimens were used in the present study (Table 1, Fig. 1). The 19 diploid Leucanthemum taxa included were represented by 39 accessions; mostly we sampled two accessions per taxon, except in the cases of L. rotundifolium (Willd.) DC. (three accessions), L. vulgare L. subsp. vulgare (three accessions), and L. ligusticum Marchetti et al. (one accession). For the sequence-based analyses, ten representatives of genera

AFLP fingerprinting

AFLP fingerprinting of 39 individuals and nine replicates with three selective primer pairs (E-ACC/M-CTAG, E-AGG/M-CTAG, E-ACA/M-CTAG) and an optimised, automated band-scoring (Holland et al., 2008) yielded 610 polymorphic loci (210, 183, and 217, respectively) in the range of 50–420 bp. An average (Euclidian) error rate of 10% was estimated among all replicates, being a reasonable rate for an automated band scoring procedure (i.e., 6–13% or 9–18% given in Holland et al., 2008). The resolution

Disentangling hybridisation and incomplete lineage sorting (ILS)

A simulation study by Leaché et al. (2014) impressively demonstrated that gene flow could have a considerable influence on species tree estimation and could bias the estimation of the species tree topology and of parameters estimated such as population sizes and divergence times. However, while species tree reconstruction methods usually account for incomplete lineage sorting (ILS) due to its nature as a process intrinsically linked with speciation events (Edwards, 2009), gene flow as the other

Acknowledgments

We would like to thank Dr. Sarah Diermeier (Regensburg; Cold Spring Harbor, NY, USA) for her help with the analysis of NGS results with the Galaxy webportal and Dr. Santiago Ortíz and Dr. Juan Rodríguez Oubiña (both Santiago de Compostela, Spain) for their support on excursions to Galicia (NW Spain). For nomenclatural advice we owe gratitude to Nick Turland (Berlin). The technical help of Mr. Peter Hummel in the molecular laboratory of the working group at the University of Regensburg is

References (141)

  • C. Oberprieler

    Book review on: Polyploidy and hybrid genomics, by Chen ZJ, Birchler JA (Eds.), Wiley-Blackwell, Oxford, UK (2013)

    J. Plant Physiol.

    (2014)
  • C. Oberprieler et al.

    Filling of eco-climatological niches in a polyploid complex – a case study in the plant genus Leucanthemum Mill. (Compositae, Anthemideae) from the Iberian Peninsula

    Flora

    (2012)
  • C. Oberprieler et al.

    The reticulate evolutionary history of the polyploid NW Iberian Leucanthemum pluriflorum clan (Compositae, Anthemideae) as inferred from nrDNA ETS sequence diversity and eco-climatological niche-modelling

    Mol. Phylogenet. Evol.

    (2014)
  • R. Abbott et al.

    Hybridization and speciation

    J. Evol. Biol.

    (2013)
  • Barrelier, J., 1714. Plantae per Galliam, Hispaniam et Italiam observatae, iconibus aeneis exhibitae […]....
  • B.R. Baum

    Combining trees as a way of combining data sets for phylogenetic inference, and the desirability of combining gene trees.

    Taxon

    (1992)
  • O.R. Bininda-Edmonds

    Trees versus characters and the supertree/supermatrix ‘paradox’

    Syst. Biol.

    (2004)
  • J.L. Blanco-Pastor et al.

    Coalescent simulations reveal hybridization and incomplete lineage sorting in Mediterranean Linaria

    PLoS ONE

    (2012)
  • D. Blankenberg et al.

    Manipulation of FASTQ data with Galaxy

    Bioinformatics

    (2010)
  • W. Bleeker et al.

    Hybrid zones between invasive Rorippa austriaca and native R. sylvestris (Brassicaceae) in Germany: Ploidy levels and patterns of fitness in the field

    Heredity

    (2005)
  • K. Bremer et al.

    Generic monograph of the Asteraceae-Anthemideae

    Bull. Natur. History Museum Lond.

    (1993)
  • C. Brochmann

    Reproductive strategies of diploid and polyploid populations of arctic Draba (Brassicaceae)

    Plant Syst. Evol.

    (1993)
  • C. Brochmann et al.

    Gene flow across ploidal levels in Draba (Brassicaeae)

    Evolut. Trends Plants

    (1992)
  • D. Bryant et al.

    Neighbor-net: an agglomerative method for the construction of phylogenetic networks

    Mol. Biol. Evol.

    (2004)
  • T.R. Buckley et al.

    Differentiating between hypotheses of lineage sorting and introgression in New Zealand alpine cicadas (Maoricicada Dugdale)

    Syst. Biol.

    (2006)
  • Burnat, E., 1916. Flore des Alpes Maritimes 6(1). Georg & CIE, Genève, Bâle,...
  • A. Camargo et al.

    Accuracy and precision of species trees: effects of locus, individual, and base-pair sampling on inference of species trees in lizards of the Liolaemus darwinii group (Squamata, Liolaemidae).

    Syst. Biol.

    (2012)
  • Candolle, AP de., 1838. Prodromus Systematis Naturalis Regni Vegetabilis, vol. 6....
  • M. Chapman et al.

    Universal markers for comparative mapping and phylogenetic analysis in the Asteraceae (Compositae)

    Theor. Appl. Genet.

    (2007)
  • L. Cheng et al.

    Bayesian semi-supervised classification of bacterial samples using MLST databases

    BMC Bioinformatics

    (2011)
  • J. Corander et al.

    Bayesian identification of admixture events using multi-locus molecular markers

    Mol. Ecol.

    (2006)
  • J. Corander et al.

    Bayesian identification of stock mixtures from molecular marker data

    Fish. Bull.

    (2006)
  • J. Corander et al.

    Enhanced Bayesian modelling in BAPS software for learning genetic structures of populations

    BMC Bioinformatics

    (2008)
  • M.P. Cummings et al.

    A genealogical approach to quantifying lineage divergence

    Evolution

    (2008)
  • M.J. De Villiers et al.

    An approach to identify putative hybrids in the ’coalescent stochasticity zone’, as exemplified in the African plant genus Streptocarpus (Gesneriaceae)

    New Phytol.

    (2013)
  • J.J. Doyle et al.

    Preservation of plant samples for DNA restriction endonuclease analysis

    Taxon

    (1987)
  • J.J. Doyle et al.

    A rapid DNA isolation procedure for small quantities of fresh leaf tissue

    Phytochem. Bull.

    (1987)
  • D.A. Earl et al.

    STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method

    Conser. Genet. Resour.

    (2012)
  • S.V. Edwards

    Is there a new and general theory of molecular systematics emerging?

    Evolution

    (2009)
  • G. Evanno et al.

    Detecting the number of clusters of individuals using the software structure: a simulation study

    Mol. Ecol.

    (2005)
  • M. Ferriol et al.

    Microsatellite evidence for low genetic diversity and reproductive isolation in tetraploid Centaurea seridis (Asteraceae) coexisting with diploid Centaurea aspera and triploid hybrids in contact zones

    Bot. J. Linn. Soc.

    (2014)
  • D. Gerard et al.

    Estimating hybridization in the presence of coalescence using phylogenetic intraspecific sampling

    BMC Evol. Biol.

    (2011)
  • B. Giardine et al.

    Galaxy: a platform for interactive large-scale genome analysis

    Genome Res.

    (2005)
  • T.C. Glenn

    Field guide to next-generation DNA sequencers

    Mol. Ecol. Resour.

    (2011)
  • J. Goecks et al.

    Galaxy: a comprehensive approach for supporting accessible, reproducible, and transparent computational research in the life sciences

    Genome Biol.

    (2010)
  • A. Gómez et al.

    Refugia within refugia: patterns of phylogeographic concordance in the Iberian Peninsula

  • R. Greiner et al.

    Phylogenetic studies in the polyploid complex of the genus Leucanthemum Mill. (Compositae, Anthemideae) based on cpDNA sequence variation

    Plant Syst. Evol.

    (2012)
  • R. Greiner et al.

    Evolution of the polyploid north-west Iberian Leucanthemum pluriflorum clan (Compositae, Anthemideae) based on plastid DNA sequence variation and AFLP fingerprinting

    Ann. Bot.

    (2013)
  • P. Griffin et al.

    A next-generation sequencing method for overcoming the multiple gene copy problem in polyploid phylogenetics, applied to Poa grasses

    BMC Biol.

    (2011)
  • Guinea E. 1953. Geografíca Botánica de Santander. –...
  • Cited by (30)

    • Hybridization and cryptic speciation in the Iberian endemic plant genus Phalacrocarpum (Asteraceae-Anthemideae)

      2021, Molecular Phylogenetics and Evolution
      Citation Excerpt :

      Due to the frequency of hybridization in plants (Ellstrand et al., 1996; Mallet, 2005; Whitney et al., 2010), the latter approach—singling out hybridization to better understand the systematic relationships within a whole group—has gained importance. This has been particularly fuelled by the wealth of molecular data, including that from high throughput sequencing technologies (Twyford and Ennos, 2012; Eaton and Ree, 2013; Escudero et al., 2014; Konowalik et al., 2015), and the availability of new analytical tools (Maureira-Butler et al., 2008; Joly et al., 2009; Kubatko, 2009; Hibbins and Hahn, 2019). However, unravelling old reticulation events is challenging even with abundant genomic data due to the evolutionary change that occurs after hybridization events (Pfeil et al., 2017).

    • Taming the Red Bastards: Hybridisation and species delimitation in the Rhodanthemum arundanum-group (Compositae, Anthemideae)

      2020, Molecular Phylogenetics and Evolution
      Citation Excerpt :

      This normalisation step could have minimised the number of fragments from paralogous and repetitive genomic regions during the lab part of our RADseq procedure, reducing the necessity for finding an optimal condition for splitting reads of such regions during de-novo assembly of reads (mainly controlled by ct). Numerous studies have shown that interspecific hybridisation is a common phenomenon in the tribe Anthemideae of the Compositae family (e.g. Lo Presti et al., 2010; Himmelreich et al., 2014; Konowalik et al., 2015; Oberprieler et al., 2019). In Rhodanthemum, some evidence exists for the occurrence of interspecific hybridisation, albeit a recent study of Wagner et al. (2019) showed that reticulate evolution played a much smaller role in the history of the genus compared to the closely related genus Leucanthemum.

    • Nuclear loci developed from multiple transcriptomes yield high resolution in phylogeny of scaly tree ferns (Cyatheaceae) from China and Vietnam

      2019, Molecular Phylogenetics and Evolution
      Citation Excerpt :

      To visualize the conflicts among the gene trees, we generated a super-network in SPLITSTREE 4.13.1 (Huson and Bryant, 2006). The super-network uses standard MRP matrix (Baum, 1992; Ragan, 1992), which is used in super tree reconstruction, as the input for the network analysis (Konowalik et al., 2015). We phased alleles statistically in PHASE 2.1.1 (Stephens et al., 2001; Stephens and Scheet, 2005) using input files assembled on the SeqPHASE web server (Flot, 2010).

    • Molecular diversity and phylogeny of Tunisian Prunus armeniaca L. by evaluating three candidate barcodes of the chloroplast genome

      2019, Scientia Horticulturae
      Citation Excerpt :

      The construction of haplotype networks for each studied chloroplastic region confirmed the topology of the obtained phylogenetic tree and proved that the rps16-trnQ region is the most informative region to differentiate among apricot and others Prunus species. In agreement with our results, Konowalik et al. (2015) were able to differentiate among 19 accessions of genus Leucanthemum with the rsp16-trnQ region which has been reported as the most informative comparing to different chloroplastic regions. In addition, Tsai et al. (2012) revealed that the rsp16-trnQ region is also the best barcode in order to differentiate among 19 orchids.

    • The role of in situ species diversification for the evolution of high vascular plant species diversity in the European Alps—A review and interpretation of phylogenetic studies of the endemic flora of the Alps

      2017, Perspectives in Plant Ecology, Evolution and Systematics
      Citation Excerpt :

      DC. may also contain an Alpine subspecific diversification, but treatment of taxonomic entities in this group differs widely between Aeschimann et al. (2004) and Greiner et al. (2012) and Konowalik et al. (2015). Table 2 summarizes the EAS diversifications identified including the number of species they comprise, and information about their ancestral area and age when formally reconstructed.

    View all citing articles on Scopus

    This paper was edited by the Associate Editor Xiao-Quan Wang.

    View full text