Skip to main content
NCBI home page
UKPMC Funders Author Manuscripts logoLink to UKPMC Funders Author Manuscripts
. Author manuscript; available in PMC: 2014 Dec 17.
Published in final edited form as: Mol Phylogenet Evol. 2009 Mar 9;53(1):220–233. doi: 10.1016/j.ympev.2009.02.021

Molecular phylogenetic analyses of nuclear and plastid DNA sequences support dysploid and polyploid chromosome number changes and reticulate evolution in the diversification of Melampodium (Millerieae, Asteraceae)

Cordula Blöch a, Hanna Weiss-Schneeweiss a,*, Gerald M Schneeweiss b, Michael HJ Barfuss a, Carolin A Rebernig a, José Luis Villaseñor c, Tod F Stuessy a
PMCID: PMC4268500  EMSID: EMS60094  PMID: 19272456

Abstract

Chromosome evolution (including polyploidy, dysploidy, and structural changes) as well as hybridization and introgression are recognized as important aspects in plant speciation. A suitable group for investigating the evolutionary role of chromosome number changes and reticulation is the medium-sized genus Melampodium (Millerieae, Asteraceae), which contains several chromosome base numbers (x = 9, 10, 11, 12, 14) and a number of polyploid species, including putative allopolyploids. A molecular phylogenetic analysis employing both nuclear (ITS) and plastid (matK) DNA sequences, and including all species of the genus, suggests that chromosome base numbers are predictive of evolutionary lineages within Melampodium. Dysploidy, therefore, has clearly been important during evolution of the group. Reticulate evolution is evident with allopolyploids, which prevail over autopolyploids and several of which are confirmed here for the first time, and also (but less often) on the diploid level. Within sect. Melampodium, the complex pattern of bifurcating phylogenetic structure among diploid taxa overlain by reticulate relationships from allopolyploids has non-trivial implications for intrasectional classification.

Keywords: Asteraceae, Dysploidy, ITS, matK, Melampodium, Phylogeny, Polyploidy, Reticulate evolution

1. Introduction

Chromosome evolution, involving both number (polyploidy and dysploidy) and structural changes (e.g., inversions, translocations), as well as hybridization and introgression, are recognized as important aspects of plant speciation (Rieseberg, 2001; Schubert, 2007; Leitch and Leitch, 2008). A requisite for assessing the role of chromosomal change in a given group is to have a sound hypothesis of the group’s phylogeny (Rieseberg, 2001). It is important to know whether chromosome base numbers are correlated with phylogenetic lineages, as is sometimes the case (e.g., Schneeweiss et al., 2004a,b; Hansen et al., 2006; Hidalgo et al., 2007), or whether they are independent (e.g., Baldwin and Wessa, 2000; Mast et al., 2001; Yuan et al., 2004; Ellison et al., 2006). This allows their causative role in diversification to be interpreted properly. Molecular data can provide precise estimates of phylogenetic relationships as well as evidence concerning taxa involved in hybridization at both the diploid and the polyploid level. Examples of such studies include Achillea (Guo et al., 2004, 2006), Glycine (Doyle et al., 2003), Helianthus (Rieseberg, 1991; Rieseberg et al., 2007), Nicotiana (Lim et al., 2004), and Paeonia (Ferguson and Sang, 2001).

A suitable group for investigating the evolutionary role of chromosome number changes and reticulation is the genus Melampodium (Asteracaeae). It is medium-sized and comprises 40 annual and perennial species (Stuessy, 1972; Turner, 1988, 1993, 2007) centered in tropical and subtropical Mexico and Central America with five species distributed in the adjacent southwestern United States and three species scattered in Colombia and Brazil. With the exception of the only recently described M. moctezumum (Turner, 2007), all species have now been counted chromosomally and the following haploid chromosome numbers have been reported (Stuessy, 1968, 1970b, 1971, 1972; Keil and Stuessy, 1975, 1977; H. Weiss-Schneeweiss et al., unpubl.): n = 9, 10, 11, 12, 14, 18, 20, 23, 24, 27, 28, 30, 33. Melampodium is closely related to Acanthospermum (six species in the Americas and on the Galapagos Islands; Stuessy, 1970a) and Lecocarpus (three to four species endemic to the Galapagos Islands; Eliasson, 1971; Adsersen, 1980; Sønderberg Brok and Adsersen, 2007), with which it shares functionally staminate disk florets and pistillate ray florets as well as inner phyllaries (involucral bracts) each tightly enclosing and fused with a single ray achene (Stuessy, 1970a). The generic distinctness of these groups, which together constitute a generic complex classified as a separate subtribe Melampodiinae (Hoffmann, 1890; Panero, 2007), only recently moved from tribe Heliantheae s.s. to tribe Millerieae (Panero, 2007; Baldwin, in press), has never been seriously doubted. It has been suggested, however, that Acanthospermum and Lecocarpus might have been derived from within Melampodium (Stuessy, 1971).

A previous intuitive phylogenetic hypothesis (Stuessy, 1972), which was tested by cladistic (Stuessy, 1979) and phenetic (Stuessy and Crisci, 1984) analyses of morphological characters, suggested that basic chromosome numbers correspond well with delimitation of sections. Four sections have unique chromosome base numbers (sections Zarabellia, Melampodium, Serratura, and Bibractiaria with x = 9, 10, 12, and 14, respectively), whereas two (sections Alcina and Rhizomaria) share x = 11 (Stuessy, 1971; H. Weiss-Schneeweiss et al., unpubl.). Dysploidy is not restricted to the diploid level but also occurs at the polyploid level as evidenced by n = 23 derived from n = 12 in M. dicoelocarpum (Stuessy, 1971).

Stuessy (1971) proposed x = 10 as the ancestral chromosome base number in the genus because it is found in the morphologically highly variable and most species-rich sect. Melampodium (hence divided into the five series Cupulata, Leucantha, Longipila, Melampodium, and Sericea; Stuessy, 1972), and correlates with occurrence of the presumably primitive type of conspicuous and clearly differentiated sterile ovary of the functionally male disk florets (Stuessy, 1972), otherwise found in Acanthospermum and Lecocarpus. The other chromosomal lines, which share the presumably derived character of disk florets with diminutive and undifferentiated sterile ovaries, were suggested to be derived from x = 10 by either loss (x = 9) or gain (x = 11 and x = 12) of chromosomes (Stuessy, 1971). In conflict with the above hypothesis, however, is the presence of x = 11 in the related genera Acanthospermum and Lecocarpus (Stuessy, 1971; Keil et al., 1988; H. Weiss-Schneeweiss, unpubl.).

Polyploidy (both on tetraploid and hexaploid levels) has played an important role in diversification of Melampodium with polyploidy being known in 16 species (40% of the genus). Of those, seven are uniformly tetraploid and five uniformly hexaploid, whereas intraspecific cytotype mixtures of diploid and tetraploid cytotypes and of tetraploid and hexaploid cytotypes are known from three and one species, respectively (Stuessy, 1971; Stuessy et al., 2004; H. Weiss-Schneeweiss et al., unpubl.). Among polyploids, both autopolyploid (M. aureum, and tetraploid cytotypes of M. cinereum and M. leucanthum, Stuessy, 1971; Stuessy et al., 2004) and allopolyploid (M. sericeum, Stuessy, 1971; M. paniculatum, Stuessy and Brunken, 1979) origins have been suggested.

To establish a sound phylogenetic framework as basis for a better understanding of roles of chromosome number change and reticulate evolution in diversification of Melampodium, we generated and analyzed sequence data from the nuclear ITS region as well as the plastid matK gene. Internal transcribed spacers (ITS1 and ITS2) of nuclear ribosomal DNA have been frequently and successfully used for phylogenetic studies in Asteraceae (e.g., Kimball and Crawford, 2004; Samuel et al., 2006) and in tribe Heliantheae s.l., in particular (e.g., Balsamorhiza and Wyethia, Moore and Bohs, 2003; Dahlia, Gatt et al., 2000; Saar et al., 2003; Madiinae, Baldwin and Wessa, 2000; Montanoa, Plovanich and Panero, 2004). Despite legitimate criticisms concerning, among others, concerted evolution, gene silencing and conversion, or their labile nature in the genome (Álvarez and Wendel, 2003), ITS is still one of the most useful phylogenetic markers in various plant groups (Nieto Feliner and Roselló, 2007). Plastid matK region has also been used successfully for species-level relationships in Asteraceae (Samuel et al., 2003, 2006), although in this family this sequence has mostly been used for phylogenetic studies at the intergeneric level and above (e.g., Bayer et al., 2000, 2002).

The current study analyzes the phylogenetic relationships among all known species of Melampodium. Specifically, we address the following questions: (1) What are the phylogenetic relationships among Melampodium, Acanthospermum and Lecocarpus, and is Melampodium monophyletic? (2) How well does the current taxonomic classification (Stuessy, 1972) reflect phylogenetic relationships among the species? (3) Are the chromosome base numbers predictive of evolutionary lineages? (4) Which modes of polyploidization (auto- vs. allopolyploidy) occurred, and which parental species were involved?

2. Materials and methods

2.1. Field and laboratory methods

One to several populations of all currently recognized species and varieties of Melampodium were collected in the United States, Mexico and Costa Rica (Table 1). Lecocarpus accessions used for molecular analyses were grown in the Botanical Garden of the University of Vienna, whereas Acanthospermum and Melampodium moctezumum samples were obtained from herbarium specimens (Table 1). Closely related genera (Stuessy, 1970a; Baldwin et al., 2002; Rauscher, 2002) collected in Mexico were selected as outgroups (Table 1). Unless otherwise noted, voucher specimens are deposited in MEXU and WU (Table 1). Chromosome numbers and karyotypes of nearly all Melampodium accessions used in this study have been checked in root tip meristematic cells, and occasionally also in meiotic pollen mother cells in young flower buds using standard Feulgen staining (Weiss-Schneeweiss et al., 2007); chromosomal data will be published elsewhere.

Table 1.

Species names, localities, voucher numbers, ploidy levels (H. Weiss-Schneeweiss et al., unpubl.), and GenBank accession numbers of the analyzed taxa. All vouchers deposited in WU and MEXU unless otherwise indicated; Countries: A, Argentina; CR, Costa Rica; E, Ecuador; M, México; USA, United States of America. Collectors: AR, A.L. Reina; CB, C. Blöch; CR, C.A. Rebernig; CSB, Camilla Sønderberg Brok, EO, E. Ortiz B.; GF, G. Flores; HA, H. Adsersen; IC, I. Calzada; IS, I. Sánchez; JC, J. Calonico; JV, J.L. Villasenor; JM, J.M. Morales; LA, Loran Anderson; MB, M.H.J. Barfuss; ML, M. Lenko; TD, T.R. Van Devender; TS, T.F. Stuessy.

Taxon (chromosome base number or ploidy level) Accession Collection details, voucher numbers GenBank accession numbers
ITS matK
Melampodiinae
Melampodium
 Sect. Melampodium (x = 10)
  Ser. Melampodium
   M. americanum L (2x) 1 M, Michoacán, 2005; TS, JV, CR & IC, 18592. FJ696977 FJ697080
2 M, Colima, 2005; TS, JV, CR & IC, 18609. FJ696978, FJ696979 FJ697081
   M. diffusum Cass. (2x) 1 M, Oaxaca, 2005; TS, JV, CR & IC, 18666. FJ696975 FJ697082
2 M, Guerrero, 2005; TS, JV, CR & IC, 18669. FJ696976 FJ697083
   M. linearilobum DC. (2x) 1 M, Michoacán, 2005; TS, JV, CR & IC, 18593. FJ696983 FJ697088
2 M, Oaxaca, 2005; TS, JV, CR & IC, 18661. FJ696982 FJ697089
   M. longipes (A. Gray) B.L Rob. (2x) 1 M, Nayarit, 2005; TS, JV, CR & IC, 18619. FJ696984 FJ697087
2 M, Nayarit, 2005; TS, JV, CR & IC, 18621. FJ696985 FJ697086
   M. mayfieldii B.L. Turner (4x) 1 M, Colima, 2005; TS, JV, CR & IC, 18613. FJ697018 FJ697112
2 M, Jalisco, 2006; TS, JV, CB & EO, 19019. FJ697019–FJ697021 FJ697113
   M. pilosum Stuessy (2x) 1 M, Michoacán, 2005; TS, JV, CR & IC, 18587. FJ696981 FJ697084
2 M, Michoacán, 2005; TS, JV, CR & IC, 18590. FJ696980 FJ697085
  Ser. Leucantha
   M. argophyllum (A. Gray ex B.L. Rob.) 1 M, Nuevo León, 2006; TS, JV, CR & CB, 19059. FJ697009 FJ697110
    S.F. Blake (6x) 2 M, Nuevo León, 2006; TS, JV, CR & CB, 19060. FJ697010–FJ697013 FJ697111
   M. cinereum DC. var. cinereum (2x, 4x) 1 USA, Texas, Frio Co, 2005; TS & CR, 18688A. FJ697006 FJ697101
2 USA, Texas, Zapata Co, 2005; TS & CR, 18694A. FJ697008 FJ697102
3 USA, Texas, Jim Hogg Co, 2005; TS & CR, 18698S. FJ697007 FJ697103
   M. cinereum DC. var. hirtellum Stuessy (2x) 1 M, Coahuila, 2006; TS, JV, CR & CB, 19057. FJ697015 FJ697104
2 M, Nuevo León, 2006; TS, JV, CR & CB, 19061. FJ697014 FJ697105
   M. cinereum DC. var. ramosissimum DC. 1 M, Tamaulipas, 2006; TS, JV & CB, 19063. FJ697016 FJ697106
    (A. Gray) (2x) 2 M, Tamaulipas, 2006; TS, JV & CB, 19064. FJ697017 FJ697107
   M. leucanthum Torr. & A. Gray (2x, 4x) 1 USA, Texas, Medina Co, 2005; TS & CR, 18687. FJ697005 FJ697108
2 USA, Arizona, Graham Co, 2006; CR & ML, 18800. FJ697004 –
3 USA, Arizona, Yavapai Co, 2006; CR & ML, 18808. FJ697003 FJ697109
  Ser. Sericea
   M. longicorne A. Gray (6x) 1 USA, Arizona, Pima Co, 2006; CR & MB, 18823. FJ697000 FJ697098
2 USA, Arizona, Pima Co, 2006; CR & MB, 18826. FJ697001, FJ697002 FJ697099
   M. nayaritense Stuessy (4x) 1 M, Nayarit, 2008; JV, GF & EO, 1575. FJ696992 FJ697091
2 M, Nayarit, 2008; JV, GF & EO, 1577. FJ696994–FJ696996 FJ697090
3 M, Nayarit, 2008; JV, GF & EO, 1579. FJ696993 FJ697092
   M. pringlei B.L. Rob. (6x) 1 M, Oaxaca, 2005; TS, JV, CR & IC, 18637. FJ696990, FJ696991 FJ697097
2 M, Oaxaca, 2005; TS, JV, CR & IC, 18650. FJ696988 FJ697094
   M. sericeum Lag. (6x) M, Michoacán, 2005; TS, JV, CR & IC, 18572. FJ696986, FJ696987 FJ697093
   M. strigosum Stuessy (4x) 1 USA, Texas, Jeff Davis Co, 2005; CR & ML, 18728. FJ696997, FJ696998 FJ697095
2 M, Queretaro, 2006; TS, JV & CB, 19073. FJ696999 FJ697096
  Ser. Cupulata
   M. appendiculatum B.L. Rob. (2x) M, Sonora, 2006; TS, JV & CB, 19046. FJ697030 FJ697116
   M. cupulatum A. Gray (2x) 1 M, Sinaloa, 2006; TS, JV & CB, 19044. FJ697031 FJ697114
2 M, Sonora, 2006; TS, JV & CB, 19048. FJ697032 FJ697115
   M. glabribracteatum Stuessy (2x) M, Oaxaca, 2005; TS, JV, CR & IC, 18654. FJ696989 FJ697100
   M. moctezumum B.L. Turner M, Sonora, 2006; TD & AR, 2007-706 (TEX). FJ789805, FJ789806 FJ789803
   M. rosei B.L. Rob. (2x) 1 M, Sinaloa, 2006; TS, JV, CB & EO, 19036. FJ697025 FJ697121
2 M, Sinaloa, 2006; TS, JV & CB, 19043. FJ697023, FJ697024 FJ697122
3 M, Sinaloa, 2006; TS, JV & CB, 19049. FJ697022 –
4 M, Sinaloa, 2006; TS, JV, CB & EO, 19025. FJ697026 –
   M. sinuatum Brandegee (2x) M, Baja California, 2006; TS & JV, 19037. FJ697029 FJ697136
   M. tenellum Hook.f. & Arn. (2x) 1 M, Nayarit, 2006; TS, JV, CB & EO, 19020. FJ697028 FJ697117
2 M, Nayarit, 2006; TS, JV, CB & EO, 19023. FJ697027 FJ697118
  Ser. Longipila
   M. longipilum B.L. Rob. (2x) 1 M, Oaxaca, 2005; TS, JV, CR & IC, 18630. FJ696972, FJ696973 FJ697119
2 M, Oaxaca, 2005; TS, JV, CR & IC, 18653. FJ696974 FJ697120
 Sect. Bibractiaria (x = 14)
   M. bibracteatum S. Watson (4x) 1 M, México, 2005; TS, JV, CR & IC, 18565. FJ697056 FJ697145
2 M, Durango, 2006; TS, JV, CR & CB, 19052. FJ697057 FJ697146
   M. repens Sessé & Moc. (2x, 4x) 1 M, Morelos, 2005; TS, JV, CR & IC, 18563. FJ697059 FJ697147
2 M, Oaxaca, 2005; TS, JV, CR & IC, 18639. FJ697058 FJ697148
 Sect. Zarabellia (x = 9)
   M. gracile Less. (2x) 1 M, Michoacán, 2005; TS, JV, CR & IC, 18586. FJ697072 FJ697162
2 M, Guerrero, 2005; TS, JV, CR & IC, 18674. FJ697073 FJ697163
   M. longifolium Cerv. ex Cav. (2x) 1 M, Oaxaca, 2005; TS, JV, CR & IC, 18629. FJ697068 FJ697142
2 M, Oaxaca, 2005; TS, JV, CR & IC, 18633. FJ697067 FJ697143
3 M, México D.F., 2006; TS, JV & CB, 19074. – FJ697141
   M. microcephalum Less. (2x) 1 M, Michoacán, 2005; TS, JV, CR & IC, 18569. – FJ697156
2 M, Oaxaca, 2005; TS, JV, CR & IC, 18641. – FJ697157
3 M, Oaxaca, 2005; TS, JV, CR & IC, 18644. – FJ697158
4 M, Oaxaca, 2005; TS, JV, CR & IC, 18651. FJ697070 FJ697161
5 M, Oaxaca, 2005; TS, JV, CR & IC, 18658. – FJ697159
6 M, Sinaloa, 2006; TS, JV, CB & EO, 19030. FJ697071 FJ697160
   M. mimulifolium B.L. Rob. (2x) M, Oaxaca, 2005; TS, JV, CR & IC, 18656. FJ697069 FJ697144
   M. paniculatum Gardner (4x, 6x) 1 M, Chiapas, 2008; JV, EO & JC 1589. FJ697065, FJ697066 –
2 M, Chiapas, 2008; JV, EO & JC, 1591. FJ697063 FJ697164
3 M, Chiapas, 2008; JV, EO & JC, 1593. FJ697064 FJ697165
4 FL2935 (OS). FJ697060–FJ697062 –
 Sect. Rhizomaria (x = 11)
   M. aureum Brandegee (6x) 1 M, Michoacán, 2005; TS, JV, CR & IC, 18576. FJ696970 FJ697151
2 M, Oaxaca, 2005; TS, JV, CR & IC, 18635. FJ696971 FJ697152
   M. montanum Benth. var. montanum (2x) 1 M, Oaxaca, 2005; TS, JV, CR & IC, 18640. FJ696967 FJ697153
   M. montanum Benth. var. viridulum Stuessy (2x) 1 M, Oaxaca, 2005; TS, JV, CR & IC, 18646. FJ696968 FJ697154
2 M, Oaxaca, 2005; TS, JV, CR & IC, 18655. FJ696969 FJ697155
 Sect. Alcina (x = 11)
   M. glabrum S. Watson (2x) 1 M, Michoacán, 2005; TS, JV, CR & IC, 18598. FJ697036 FJ697126
2 M, Michoacán, 2005; TS, JV, CR & IC, 18624. FJ697035 FJ697125
   M. nutans Stuessy (2x) 1 M, Michoacán, 2005; TS, JV, CR & IC, 18591. FJ697034 FJ697124
2 M, Oaxaca, 2005; TS, JV, CR & IC, 18664. FJ697033 FJ697123
   M. perfoliatum Stuessy (Cav.) H.B.K. (2x) 1 M, Jalisco, 2005; TS, JV, CR & IC, 18604. FJ697038 FJ697149
2 M, Oaxaca, 2005; TS, JV, CR & IC, 18652. FJ697037 FJ697150
 Sect. Serratura (x = 12)
   M. costaricense Stuessy (4x) 1 CR, Prov. San José, 2006; TS, JV & IS, 19076. FJ697051 FJ697129
2 CR, Prov. San José, 2006; TS, JV & IS, 19084. FJ697052 FJ697130
   M. dicoelocarpum B.L. Rob. (2x, 4x) 1 M, Michoacán, 2005; TS, JV, CR & IC, 18588. FJ697039 FJ697134
2 M, Michoacán, 2005; TS, JV, CR & IC, 18595. FJ697041–FJ697043 –
3 M, Jalisco, 2005; TS, JV, CR & IC, 18603. FJ697040 FJ697135
   M. divaricatum (Rich. in Pers.) DC. (2x) 1 M, Michoacán, 2005; TS, JV, CR & IC, 18594. FJ697044 FJ697131
2 M, Michoacán, 2005; TS, JV, CR & IC, 18601. FJ697045 –
3 M, Oaxaca, 2005; TS, JV, CR & IC, 18668. – FJ697132
4 CR, Prov. San José, 2006; TS, JV & IS, 19086. – FJ697133
   M. northingtonii B.L. Turner (4x) 1 M, Oaxaca, 2005; TS, JV, CR & IC, 18659. FJ697054, FJ697055 FJ697139
2 M, Oaxaca, 2005; TS, JV, CR & IC, 18660. FJ697053 FJ697140
   M. sinaloense Stuessy (4x) 1 M, Sinaloa, 2006; TS, JV, CB & EO, 19026. FJ697048 FJ697127
2 M, Sinaloa, 2006; TS, JV, CB & EO, 19027. FJ697049, FJ697050 FJ697128
   M. tepicense B.L. Rob. (2x) 1 M, Nayarit, 2005; TS, JV, CR & IC, 18615. FJ697047 FJ697137
2 M, Nayarit, 2005; TS, JV, CR & IC, 18617. FJ697046 FJ697138
Acanthospermum (x = 11)
   A. australe Kuntze Rauscher, 2002. AF465844 –
   A. hispidum DC. 1 A, Jujuy, 1993; TS & JM, 12956 WU. FJ696965 FJ789804
2 USA, Florida; LA, 3481, KSC. FJ696964 –
   A. microcarpum B.L. Rob. Rauscher, 2002 AF465845 –
Lecocarpus (x = 11)
   L. lecocarpoides (B.L. Rob. & Greenm.) Cronquist & Stuessy E, Galápagos, Osborn, 2001; CSB & HA, Lam1, DK. – FJ697078
   L. pinnatifidus Decne. E, Galápagos, Floreana, 2001; CSB & HA, Lam6, DK. – FJ697075
   L. sp. E, Galápagos; HA, s.n., DK & WU. FJ696966 –
Outgroups
   Acmella oppositifolia (Lam.) R.K. Jansen M, México, 2006; TS, JV, CB & EO, 19005. – FJ697074
   Galinsoga parviflora Cav. M, México, 2006; TS, JV, CB & EO, 19004. FJ696962 FJ697076
   Milleria quinqueflora L. M, Jalisco, 2006; TS, JV, CB & EO, 19016. FJ696961 FJ697077
   Siegesbeckia flosculosa L’Hér. Rauscher, 2002. AF465888 –
   Smallanthus maculatus (Cav.) H.Rob. M, Querétaro, 2006; TS, JV & CB, 19072. FJ696963 FJ697079
   Trigonospermum melampodioides DC. Rauscher, 2002. AF465906 –
Open in a new tab

Total genomic DNA was extracted from silica-dried leaf material or from herbarium specimens according to the CTAB-procedure (Doyle and Doyle, 1987) with some modifications (Tel-Zur et al., 1999). Ground plant material was washed 2–5 times with the sorbitol solution to remove polysaccharides (Tel-Zur et al., 1999). Some extracts were additionally purified with appropriate buffers of the nexttec™ Genomic DNA Isolation Kit for Plants Maxi (β-version; nexttec, Leverkusen, Germany) according to the manufacturer’s protocol.

The nuclear ITS region (partial 18S rRNA gene, ITS1, 5.8S rRNA gene, ITS2, and partial 26S rRNA gene) was amplified using primers given in Table 2. The trnK intron including the complete matK gene was amplified in one, two, three or six overlapping partitions, depending on material quality, using specific internal primers (Table 2). Polymerase chain reactions were carried out using 0.4 mM of each primer, ReddyMix PCR Master Mix (Abgene, Vienna, Austria) including 2.5 mM MgCl2 with the addition of 4% dimethyl sulfoxide (DMSO) for ITS or 0.02% bovine serum albumin (BSA) for matK. All PCR reactions were performed on an ABI thermal cycler 9700 (Applied Biosystems, Foster City, CA, USA) with initial 5 min at 80 °C followed by 36 cycles each of 30 s at 94 °C, 30 s at 52 °C (matK) or at 60 °C (ITS), and 1–2.5 min at 72 °C (depending on the size of the amplified fragment) followed by a final elongation at 72 °C for 10 min. Amplified fragments were checked on 1% agarose gel and purified using exonuclease I (ExoI) and calf intestine alkaline phosphatase (CIAP) according to the manufacturer’s protocol (Fermentas, St. Leon-Rot, Germany). The purified fragments were directly sequenced using dye terminator chemistry following the manufacturer’s protocol (Applied Biosystems). The cycle sequencing reactions were performed using the same primers as for the PCR amplifications and internal primers where appropriate (Table 2). Sequencing reactions were run on a 3130xl Genetic Analyzer automated capillary sequencer (Applied Biosystems). Sequences were assembled in AutoAssembler ver. 1.4.0 (Applied Biosystems).

Table 2.

Primers used for amplification and sequencing of ITS and matK regions.

Primer Primer sequence Reference
trnK570 fwd. 5′-TCC AAA ATC AAA AGA GCG ATT GG-3′ Samuel et al. (2005)
matK850 rev. 5′-TTT CCT TGA TAC CTA ACA TAA TGC ATG-3′ Gruenstaeudl et al., in press.
matK700 fwd. 5′-CAA TCT TCT CAC TTA CGA TCA ACA TC-3′ Gruenstaeudl et al., in press.
matK1710 rev. 5′-GCT TGC ATT TTT CAT TGC ACA CG-3′ Samuel et al. (2005)
matK550 rev. 5′-GAC TAT CCC AAT TAT GAC ACT C-3′ Gruenstaeudl et al., in press.
matK350 fwd. 5′-ATC TTC CCT AGA AAG GAA AGG GG-3′ Gruenstaeudl et al., in press.
matK1200 rev. 5′-TAT CAG AAT CTG ATA AAT CGG CCC-3′ Gruenstaeudl et al., in press.
matK1000 fwd. 5′-CCC TTG ACT TTC TGG GTT ATC G-3′ Gruenstaeudl et al., in press.
matK1450 rev. 5′-GAA GAA ACT CTT GGA AAG GTC AAG G-3′ Gruenstaeudl et al., in press.
matK1300 fwd. 5′-CTT GTG CTA GAA CTT TAG CTC GTA AG-3′ Gruenstaeudl et al., in press.
AB101 fwd. (17SE) 5′-ACG AAT TCA TGG TCC GGT GAA GTG TTC G-3′ Sun et al. (1994)
AB102 rev. (26SE) 5′-TAG AAT TCC CCG GTT CGC TCG CCG TTA C-3′ Sun et al. (1994)
ITS3 fwd. 5′-GCA TCG ATG AAG AAC GCA GC-3′ White et al. (1990)
ITS6 rev. 5′-ATG GTT CGC GGG ATT CTG CAA TTC ACA CC-3′ This study
ITS5 fwd. 5′-GGA AGT AAA AGT CGT AAC AAG G-3′ White et al. (1990)
Open in a new tab

ITS sequences of diploid accessions that showed double/multiple peaks, as well as of all polyploid accessions, were cloned using the pGEM-T-easy vector systems and JM109 competent cells (Promega, Madison, WI, USA) following manufacturer’s instructions. Inserts of 6–18 positive clones (depending on the ploidy level: 6 clones per diploid genome) were amplified using colony-PCR with universal M13 primers whereby recombinant colonies were added directly into the PCR mastermix and inserts amplified using reagents and conditions described in Park et al. (2007). All sequences are deposited in GenBank (Accession Nos. FJ696961–FJ697073 and FJ789805–FJ789806 for ITS; FJ697074–FJ697165 and FJ789803–FJ789804 for matK; Table 1).

2.2. Alignment and phylogenetic analyses

Alignments were generated with Muscle 3.6 (Edgar, 2004) using default settings and improved by visual refinement using the program BioEdit 7.0.9.0 (Hall, 1999). The potential occurrence of pseudogenes among ITS copies was assessed via checking for the conserved angiosperm motif GGCRY–(4 to 7 N)–GYGYCAAGGAA (Liu and Schardl, 1994) in ITS1, GAATTGCAGAATCC within the 5.8S rDNA (Jobes and Thien, 1997), and the presence of the conserved (C1–C6) and variable (V1–V6) domains determined for plant ITS2 sequences (Hershkovitz and Zimmer, 1996). Sequences lacking any of these motifs were considered pseudogenes, and ITS sequencing was repeated using cloning as described above.

Nuclear and plastid sequence data were analyzed separately with indels treated as missing data or with indels coded using the modified complex indel coding (MCIC; Müller, 2006) as implemented in the program Seqstate 1.36 (Müller, 2005). As the method of indel coding used here involves a step matrix, the respective data set is not amenable to likelihood methods. Maximum parsimony analyses were performed using PAUP* 4.0b10 (Swofford, 2001) treating all characters as equally weighted. Heuristic searches included 1000 replicates of random sequence addition, tree bisection reconnection (TBR) branch swapping, and MulTrees on, but permitting no more than 10 trees to be held in each step. Trees were rooted using taxa outside Melampodiinae (Baldwin et al., 2002). Nodal support was assessed via bootstrap values (BS; Felsenstein, 1985), which were calculated using PAUP* 4.0b10 with 10,000 bootstrap replicates each with 20 random sequence addition replicates holding maximally 10 trees per replicate, SPR branch swapping, and MulTrees on.

The Bayesian analyses were conducted using MrBayes 3.1.2 (Ronquist and Huelsenbeck, 2003). The best-fit substitution models were determined using MrModeltest 2.2 (Nylander, 2004, program distributed by the author, Uppsala University, Uppsala). Initially, different partitioning schemes of the data set were tested, and since they all resulted in very similar topologies with comparable posterior probabilities, differences being restricted to poorly resolved and insufficiently supported regions (data not shown), the following partition scheme and substitution models were used for the final analyses: two partitions (the genic and the spacer regions of the ribosomal cistron) with K80 + Γ and GTR + Γ substitution models, and three partitions (trnK intron, the combined first and second codon position of the matK gene, the third codon position of the matK gene) with a F81 + Γ model for the first two and a GTR + Γ model for the third partition. The MCMC settings for all Bayesian analyses consisted of four runs with four chains each (three heated ones using the default heating scheme) for 5 × 106 generations sampling every 1000th generation, using default priors and estimating all parameters during the analysis. The first 10%, which was well after the chains had reached stationarity as judged from plots of the likelihood and of all parameters and from split variances being <0.01, were discarded as burn-in. A majority rule consensus tree was constructed from the posterior set of 18,000 trees. Again, trees were rooted using non-Melampodiinae members of tribes Millerieae and Heliantheae.

The combinability of ITS and matK was tested using the Incongruence Length Difference (ILD) test (Farris et al., 1994) implemented as partition-homogeneity test in PAUP treating gaps as missing data and using 1000 partition replicates each comprising 100 random sequence addition replicates, and TBR branch swapping and keeping one tree each step. After exclusion of invariable characters, combinability was tested for (1) the whole data sets, (2) for data sets without M. nutans, M. glabrum and M. longipilum, which were resolved at conflicting positions in the different markers (see Section 3), and (3) data sets where additionally all polyploid taxa were excluded, as these might be of allopolyploid origin with potentially conflicting positions.

Conflicts and incongruences between topologies of both marker sets were visualized via consensus networks (Holland et al., 2004) as implemented in SplitsTree 4 (Huson and Bryant, 2006) using the default settings. In order to aid legibility, each species was reduced to one randomly chosen accession (except in cases of lack of species monophyly, where accordingly more accessions were retained), and the posterior set of each marker was thinned 360-fold resulting in 50 trees per marker and 100 trees in total.

Alternative phylogenetic hypotheses, specifically concerning the monophyly of currently recognized genera and sections, were tested in a Bayesian framework using Bayes factors (BF; Suchard et al., 2001). Marginal likelihoods (including their Monte Carlo error: Suchard et al., 2003; Redelings and Suchard, 2005) and BFs were calculated with Tracer 1.4 (available from http://evolve.zoo.ox.ac.uk/). As test statistic we used the widely applied 2 × lnBF, considering 2 × lnBFmodel 1 vs. model 2 > 10 as strong support for model 1 (Kass and Raftery, 1995).

3. Results

3.1. ITS

All sequences were checked for the presence of conserved angiosperm motifs (Liu and Schardl, 1994; Hershkovitz and Zimmer, 1996; Jobes and Thien, 1997). In cases where clones possessing those motifs were found, clones lacking any of these motifs were considered pseudogenes and excluded from further analyses. Since all cloned sequences of M. longifolium and M. mimulifolium showed an aberration (deletion) from the conserved angiosperm motif, they all were retained for the analyses. The conserved and variable domains described previously for ITS2 (Hershkovitz and Zimmer, 1996) could be identified in all obtained sequences, although slight changes to the published motifs were frequent. Eventually, the ITS data matrix included 115 samples (accessions and clones) from Melampodiinae, representing all Melampodium species, three species (four accessions) of Acanthospermum, two accessions of taxa of Lecocarpus, and one species each of Galinsoga, Milleria, Siegesbeckia, Smallanthus and Trigonospermum as outgroup. Sequences consisted of 91 bp from the 3′-end of the 18S rRNA gene, 254–261 bp ITS1, 158–159 bp 5.8S rRNA gene, 209–229 bp ITS2 and 628 bp from the 5′-end of the 26S rRNA gene. The final aligned matrix included 828 nucleotide characters (407 and 329 being variable and parsimony informative, respectively) and 24 coded indels of which 21 were parsimony informative (Table 3).

Table 3.

Sequence statistics for ITS and matK.

Length Var. char% GC% MSD IG (%) MSD IG vs. Ac.–Le. (%) MSD IG + Ac.–Le. vs. OG (%)
ITS1 254–261 70.40 49.00 31.30 23.20 30.40
ITS2 209–229 68.00 52.50 32.70 28.80 32.80
Coding rRNA regions 163–312 15.40 52.70 – – –
matK–trnH 1831–1904 19.74 33.94 7.30 6.50 5.80
matK gene 1479–1530 20.63 32.34 10.80 7.00 8.80
Open in a new tab

Abbreviations: GC% = GC-content in percent; MSD, maximum sequence divergence; IG, ingroup taxa; Ac.–Le., Acanthospermum and Lecocarpus; OG, outgroup taxa.

Maximum parsimony analyses with gaps treated as missing data and with gaps coded as separate characters gave nearly identical tree topologies with highly similar nodal support (data not shown); therefore, only results from the second approach are presented. The heuristic search resulted in 3470 equally parsimonious trees with a length of 1348 steps (consistency index excluding uninformative characters 0.47; retention index 0.89). The strict consensus tree is topologically very similar to the majority rule consensus tree from the Bayesian analysis (harmonic mean −ln = −8636.03), differences being only a few insufficiently supported nodes (Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Phylogenetic relationships of species of Melampodium and related genera inferred from Bayesian (solid lines) and maximum parsimony analysis (dotted lines) of the nuclear ITS region. Branches collapsing in the strict consensus tree are indicated by arrowheads. Numbers at nodes are bootstrap values/posterior probabilities. Numbers after species names refer to different accessions (Table 1) and to clone numbers (after dash). Polyploid taxa are indicated in bold (chromosome number of M. moctezumum not known). Clades discussed in text are indicated by Roman numerals. The basic chromosome numbers (gray bars), current sectional classification of the genus (normal font), and the series classification of sect. Melampodium (italics) are indicated. A, sect. Alcina; BIB, sect. Bibractiaria; MEL, sect. Melampodium; RHI, sect. Rhizomaria; ZAR, sect. Zarabellia; CUP, C, ser. Cupulata; LEU, ser. Leucantha; L, LON, ser. Longipila; M, MEL, ser. Melampodium; SER, ser. Sericea.

The clade of a paraphyletic Acanthospermum and a monophyletic Lecocarpus (clade VI, bootstrap [BS]/posterior probability [PP] 99/1.00) was nested within Melampodium (BS/PP 82/1.00), rendering the latter genus paraphyletic (Fig. 1). The alternative hypothesis of a monophyletic Melampodium is strongly rejected by 2 × ln BF of −28.64 (Table 4). Within Melampodium, several well-supported clades (BS/PP 96–100/1.00) can be distinguished (labeled from I to VII in Fig. 1; clade I′ is not inferred from the plastid data [see below]). Their relationships to one another and to some single species clades are, however, poorly resolved and insufficiently supported. These clades only partly agree with current sectional classifications and thus with chromosome base number distribution. Clade I′, which is a weakly supported sister group to the remaining ingroup taxa (BS/PP 52/0.81), consists of the sister groups M. longipilum of sect. Melampodium and the two species of sect. Rhizomaria (clade I). Clade II comprises M. mimulifolium and M. longifolium of sect. Zarabellia. The remaining species of this section (clade III) are found in a moderately supported group (BS/PP 56/1.00), which additionally includes M. perfoliatum of sect. Alcina, sect. Bibractiaria (clade IV), and sect. Serratura (clade V), the latter two forming a poorly supported clade (BS/PP <50/0.85). The alternative hypothesis of a monophyletic sect. Zarabellia is strongly rejected by 2 × ln BF of −24.24 (Table 4). Clade VII is congruent with sect. Melampodium with the exception of M. longipilum, which instead belongs to clade I (the hypothesis of a monophyletic sect. Melampodium is strongly rejected as evidenced by 2 × ln BF of −53.98; Table 4). The three species, which form single species clades with uncertain affinities to the other clades, are M. glabrum, M. nutans and M. perfoliatum (the latter with some ties to clades III-V, see above) and together constitute sect. Alcina, for which monophyly is strongly rejected (2 × lnBF of nuclear −57.72; Table 4). Concluding so far, ITS data supported only three of the currently recognized six sections (Stuessy, 1972) as monophyletic (sects. Bibractiaria, Serratura, Rhizomaria), whereas sects. Melampodium and Zarabellia are biphyletic and sect. Alcina is polyphyletic.

Table 4.

Marginal likelihoods and their Monte Carlo error as well as the test statistic 2 × ln BF for several taxonomic hypotheses, tested separately for each marker. The compared hypotheses (unconstrained vs. alternative) are arranged in rows. 2 × ln BFunconstrained vs. aiternative < −10 is regarded as strong support against the alternative hypothesis.

Unconstrained Monophyletic genus
Melampodium 		Monophyletic sect.
Melampodium 		Monophyletic sect.
Zarabellia 		Monophyletic sect.
Alcina
ITS
 Marginal likelihood −8621.74 (±0.40) −8636.06 (±0.38) −8648.73 (±0.37) −8633.86 (±0.39) −8650.60 (±0.37)
 2× Ln BF – −28.64 −53.98 −24.24 −57.72
matK
 Marginal likelihood −7834.09 (±0.27) −7834.67 (±0.30) – – −7892.39 (±0.32)
 2× Ln BF – −1.16 – – −116.60
Open in a new tab

Several subclades can be distinguished within clade VII (Fig. 1). With the exception of ser. Leucantha (the clade comprising M. argophyllum, M. cinereum and M. leucanthum; BS/PP 100/1.00), none of the other series is inferred as monophyletic (the fifth series, the holotypic ser. Longipila, does not belong to clade VII; see above). Instead, species of series Cupulata, Melampodium and Sericea intermix with each other. The clade weakly suggested as sister to ser. Leucantha (BS/PP 63/0.83) comprises M. mayfieldii of ser. Melampodium and M. longicorne of ser. Sericea nested within ser. Cupulata (BS/PP 100/1.00). Melampodium glabribracteatum of ser. Cupulata is sister to a clade (BS/PP 100/1.00) of species of series Melampodium and Sericea (BS/PP 88/1.00), which themselves are grouped into two clades including members of both series (Fig. 1).

3.2. matK

The data matrix comprised 90 accessions of Melampodiinae and one accession each of the outgroup taxa Acmella, Galinsoga, Milleria and Smallanthus (Table 1). Sequence length was 222–394 bp for the trnK intron and 1479–1530 bp for the matK gene, resulting in 524 and 1545 aligned characters, respectively. Of those, 427 were variable and 299 were parsimony informative. Gap coding added another seven characters for each region, adding 12 parsimonious informative characters. Again, the two different maximum parsimony analyses gave nearly identical results, with some clades being better supported in the second analysis (data not shown); again, only results from the second approach are presented. The heuristic search resulted in 9940 equally parsimonious trees with a length of 660 steps (consistency index excluding uninformative characters 0.72; retention index 0.95; Table 3). The strict consensus tree is similar to the majority rule consensus tree from the Bayesian analysis (harmonic mean −ln = −7853.81), differences being insufficiently supported nodes (Fig. 2).

Fig. 2.

Fig. 2

Open in a new tab

Phylogenetic relationships of Melampodium and related genera inferred from Bayesian (solid lines) and maximum parsimony analysis (dotted lines) of the plastid matK gene. Branches collapsing in the strict consensus tree are indicated by arrowheads. Numbers at nodes are bootstrap values/posterior probabilities. Numbers after species names refer to different accessions (Table 1). Polyploid taxa are indicated in bold (chromosome number of M. moctezumum not known). Clades discussed in text are indicated by Roman numerals. The basic chromosome numbers (gray bars), current sectional classification of the genus (normal font), and the series classification of sect. Melampodium (italics) are indicated (abbreviations as in Fig. 1).

As in analyses of the nuclear data, several major clades are found (BS/PP 80–100/1.00), whose relationships among each other are unresolved or insufficiently supported (Fig. 2). To allow easier comparison with results from nuclear data, clade numbers are the same (clade VII′ was not inferred from the nuclear data [see above]). Although clade VI (Acanthospermum and Lecocarpus) is nested within Melampodium rendering the latter genus paraphyletic, the alternative hypothesis of a monophyletic Melampodium cannot be rejected (2 × lnBF of −1.16; Table 4). Within Melampodium, clades I (sect. Rhizomaria), II, III (both sect. Zarabellia), IV (sect. Bribractiaria), V (sect. Serratura), and VII (sect. Melampodium except M. longipilum) are supported (BS/PP 98–100/1.00). Clades II and III form a weakly supported clade (BS/PP 53/0.91) as do clades IV and V (BS/PP 55/0.93), which themselves are sister to M. perfoliatum of sect. Alcina (BS/PP 63/0.99). Clades II–V plus M. perfoliatum together with clade I constitute a well-supported group (BS/PP 100/1.00). The phylogenetic affinities of the three species of sect. Alcina, for which monophyly is strongly rejected (2 × lnBF of −116.60; Table 4), are, possibly with the exception of M. perfoliatum, unclear. Specifically, the sister-group relationship of M. glabrum to clade VI (BS/PP <50/0.63) and of M. nutans to clade VII′ (clade VII plus M. longipilum, thus being congruent with sect. Melampodium; BS/PP 80/1.00) are insufficiently supported (BS/PP 53/0.75). From results of matK sequence data, therefore, all currently recognized sections (Stuessy, 1972) with the exception of sect. Alcina are monophyletic.

Within sect. Melampodium, ser. Melampodium (except M. linearilobum; BS/PP 63/0.97), and ser. Sericea (excluding M. nayaritense; BS/PP 98/1.00) were found as sister groups (BS/PP 99/1.00). Subsequent sister groups are M. glabribracteatum of ser. Cupulata (BS/PP 88/0.93), a clade (BS/PP 100/1.00) of M. linearilobum (ser. Melampodium) and M. nayaritense (ser. Sericea; BS/PP 100/1.00), a well-supported (BS/PP 100.1.00) clade of ser. Leucantha (BS/PP 95/0.98), and the clade (BS/PP 98/1.00) of the remaining species of ser. Cupulata (BS/PP 100/1.00).

3.3. Incongruences between nuclear and plastid sequences

Visual inspection of phylogenetic trees derived from plastid and nuclear sequence data suggest considerable topological incongruence (Figs. 1 and 2). This coincides with results from ILD tests, which reject combinability of data sets after exclusion of renegade taxa (M. glabrum, M. nutans, M. longipilum), and even after additional exclusion of all polyploid taxa (Figs. 1 and 2; all P = 0.001). Instead of combining data sets, therefore, we visualize the conflicting signals in a consensus network (Fig. 3). Some of the major incongruences concern diploid taxa and clades (M. longipilum, clade II), whereas others involve polyploids. This is particularly pronounced in sect. Melampodium. The tetraploid M. mayfieldii (ser. Melampodium) and the hexaploid M. longicorne (ser. Sericea) both group with ser. Cupulata in the nuclear data, but instead with diploids of ser. Melampodium and with tetraploid M. strigosum of ser. Sericea, respectively, in the plastid data (Fig. 3). The morphologically very similar hexaploids M. sericeum and M. pringlei (both ser. Sericea) group with tetraploid M. strigosum (ser. Sericea) in the plastid data, but in ITS analyses only M. sericeum groups with M. strigosum whereas M. pringlei groups with M. linearilobum (ser. Melampodium). Conflicting positions between data sets are also seen in sect. Serratura (clade V), where the polyploids M. costaricense, M. northingtonii and M. sinaloense group with different diploids (Fig. 3).

Fig. 3.

Fig. 3

Open in a new tab

Consensus network from 50 trees each of the set of posterior trees from the ITS and the matK data set, respectively. Polyploid taxa are indicated in bold (chromosome number of M. moctezumum not known). Sectional circumscriptions (ellipses) and series memberships within sect. Melampodium (three-letter prefixes as in Figs. 1 and 2) are also shown. Scale bar represents mean edge weights.

4. Discussion

The genus Melampodium is a suitable system to investigate the role of chromosome number evolution (polyploidy and dysploidy) and reticulate evolution, appreciated as major forces in plant evolution and speciation (Sang et al., 1997; Rieseberg, 2001; Doyle et al., 2004; Leitch and Leitch, 2008). Assessing the role of chromosome number and reticulate evolution requires having a sound hypothesis of the phylogenetic relationships of the group (Rieseberg, 2001). Our aim here, therefore, is to establish the phylogenetic framework of Melampodium for further studies by testing and refining previous phylogenetic hypotheses, which were based on morphological, karyological (Stuessy, 1971, 1972, 1979; Stuessy and Brunken, 1979; Stuessy and Crisci, 1984; Stuessy et al., 2004), and phytochemical data (Seaman et al., 1980; Bohm and Stuessy, 1991).

The presence of different chromosome base numbers in Melampodium has been used previously to characterize infrageneric groups. Turner and King (1961) used chromosome numbers obtained for 26 species to distinguish sect. Melampodium with x = 10 from sect. Zarabellia with x = 9, 11, 12, and 23 (Melampodium camphoratum, which has x = 16, was later excluded from the genus to Unxia by Stuessy, 1969). In the most recent taxonomic classification (Stuessy, 1972), four out of six sections have unique chromosome base numbers (sections Zarabellia, Melampodium, Serratura, and Bibractiaria with x = 9, 10, 12, and 14, respectively; Stuessy, 1971; H. Weiss-Schneeweiss et al., unpubl.), and only sects. Alcina and Rhizomaria share the same chromosome base number x = 11 (Stuessy, 1971; H. Weiss-Schneeweiss et al., unpubl.). It is obvious, therefore, that dysploidy has played an important role in the diversification of Melampodium.

Of the currently recognized 40 Melampodium species, 39 have been counted chromosomally, and 16 species contain polyploids, 13 species exclusively so (Stuessy, 1971; H. Weiss-Schneeweiss et al., unpubl.), which underlines the importance of polyploid evolution in the genus. Based on morphological and karyological evidence as well as crossing experiments, some of these polyploids have been suggested to be of allopolyploid origin (Stuessy, 1971; Stuessy and Brunken, 1979; H. Weiss-Schneeweiss et al., unpubl.), emphasizing the importance of reticulate evolution for speciation within Melampodium.

4.1. Monophyly of Melampodium

Based on the presence of functionally staminate disk florets, pistillate ray florets and inner phyllaries each tightly enclosing and fused with single ray achenes, Melampodium, Acanthospermum and Lecocarpus have been grouped together in subtribe Melampodiinae (Stuessy, 1973). Lecocarpus differs from Melampodium and Acanthospermum by having broadly winged inner phyllaries and a shrubby habit. The latter character is often found in island groups with otherwise herbaceous relatives (e.g., Böhle et al., 1996; Kim et al., 1996). Acanthospermum differs from Melampodium by the presence of horn-like protuberances on the achenes. There are, however, some ambiguities concerning the morphological distinctness of Acanthospermum, as a similar type of achene is also found in M. longifolium (Stuessy, 1970a). Baillon (1882) submerged species of Acanthospermum and Lecocarpus into Melampodium as distinct sections, but no one has followed this suggestion. An explicit evolutionary hypothesis was put forward by Stuessy (1971), who suggested that Acanthospermum and Lecocarpus might have been derived from Melampodium, rendering the latter paraphyletic. Stuessy’s hypothesis is supported by the nuclear sequence data (Fig. 1), which clearly reject the monophyly of Melampodium in favor of paraphyly (2 × lnBF −28.64), and is at least not contradicted by the plastid data (Fig. 2 and Table 4). Acanthospermum and Lecocarpus share a chromosome base number of x = 11, and this base number also occurs in several phylogenetically disparate lineages of Melampodium (Figs. 1 and 2), which suggests that it could be a plesiomorphic character for the entire group. Taxonomically, the phylogenetic position of Acanthospermum and Lecocarpus (clade VI) might be accommodated by combining both genera (corresponding to clade VI: Figs. 1 and 2) or, pending the establishment of monophyly of Acanthospermum (but see Fig. 1), submerging them as additional two sections within Melampodium as suggested previously (Baillon, 1882). Alternatively, all three genera might be kept intact, following acceptance of paraphyly in classification as advocated by Stuessy (1997) and Hörandl (2007).

4.2. Phylogenetic significance of chromosome base numbers: Infrageneric relationships

Based on features of the inner phyllaries, early authors distinguished three (DeCandolle, 1836) or later two sections within genus Melampodium (Robinson, 1901): Eumelampodium, Zarabellia, and Alcina (with admittedly fewer species included). This classification was further refined using chromosome numbers (Turner and King, 1961; Stuessy, 1971), and the current infrageneric classification (Stuessy, 1972) is fully congruent with the distribution of chromosome base numbers, suggesting a high predictive value of this character. This is, however, only partly corroborated by molecular phylogenetic data. Sect. Bibractiaria (clade IV) and sect. Serratura (clade V) both have unique base chromosome numbers (x = 12 and x = 14, respectively; Stuessy, 1971; H. Weiss-Schneeweiss et al., unpubl.). Sect. Rhizomaria (clade I) with x = 11, a base number also found elsewhere in the genus, is also monophyletic (Figs. 1-3). Each section is well circumscribed morphologically. Sect. Bibractiaria is characterized by two outer phyllaries, sect. Serratura includes only annual species with five outer phyllaries with herbaceous margins, and section Rhizomaria includes two perennial, rhizomatous species possessing five outer phyllaries with scarious margins. A close relationship between sect. Serratura and sect. Bibractiaria has never been suggested. In both morphological phenetic and cladistic analyses sect. Rhizomaria has been found to tie strongly to sect. Melampodium (Stuessy, 1979; Stuessy and Crisci, 1984).

Contradictory evidence is found for sects. Melampodium and Zarabellia, where plastid data agree with the current taxonomy and thus distribution of chromosome base numbers (Fig. 2), but the ITS data significantly disagree (Figs. 1 and 3). Of sect. Melampodium, plastid sequence data place M. longipilum (ser. Longipila) as sister to the remainder of the section in agreement with its chromosome base number (x = 10), whereas nuclear data place it instead as sister to sect. Rhizomaria (x = 11). The close relationship of M. longipilum and sect. Rhizomaria is also strongly supported by nuclear 5S rDNA intergenic spacer and low copy nuclear gene pgiC sequences (C. Blöch et al., unpubl.). ITS sequences of M. longipilum possessed all conservative motifs, rendering the possibility of the sampled copies being pseudogenes highly unlikely. Long branch attraction artifacts are unlikely as well, because Bayesian analysis, less prone to such difficulties, indicates relationships identical to those inferred from parsimony (Fig. 1). A unique position of M. longipilum within sect. Melampodium has already been suggested by cladistic analysis of morphological data (Stuessy, 1979). Although all species of sect. Melampodium share a sterile disk ovary with marked annular constriction at the point of corolla attachment, M. longipilum differs from the others by having an unusual flattened and apically coiled adaxial appendage on the achene, ovate subentire leaves, and markedly cupulate involucres (Stuessy, 1972). Taking the mere chromosome number as evidence for M. longipilum being a member of sect. Melampodium, its conflicting position might be the result of introgression from members of sect. Rhizomaria with subsequent convergence of the 35S rDNA cistron towards the introgressing genome. Alternatively, the unique karyotype of M. longipilum, which differs from those found in the other species of sect. Melampodium by a putative fusion-type chromosome pair 1 carrying an interstitial 35S rDNA locus in the pericentromeric region of the long arm (H. Weiss-Schneeweiss, unpubl.), suggests an independent origin of x = 10 possibly derived from x = 11 as found in sect. Rhizomaria. While further data are needed to distinguish between these hypotheses, the taxonomic consequence may be to exclude M. longipilum from sect. Melampodium. Given the likely reticulate origin of this species involving members of different sections, it might eventually be segregated into its own section.

The second case of conflicting evidence for monophyly is sect. Zarabellia, which is morphologically characterized by herbs with flowering heads with 3–5 outer phyllaries, often glandular. While plastid sequence data infer this section as monophyletic, albeit with weak support (Fig. 2), ITS data significantly reject this concept and point instead to two subgroups (Fig. 1 and Table 4). Although both units share the same chromosome base number of x = 9 as a potential synapomorphy, their karyotypes differ concerning number and localization of 5S and 35S rDNA loci (H. Weiss-Schneeweiss, unpubl.) suggesting that x = 9 evolved twice independently. While phenetic analyses of morphological characters suggested a clear differentiation of the two groups (Stuessy and Crisci, 1984), only M. gracile, M. microcephalum, and M. paniculatum (clade III) form a tightly-knit evolutionary unit with all three species having only three outer phyllaries (glandular) and in which many reciprocal artificial hybridizations have been successfully performed (Stuessy and Brunken, 1979), whereas morphological synapomorphies for the morphologically disparate M. longifolium and M. mimulifolium (clade II) still remain to be found. The latter species, however, is morphologically very similar to M. gracile of the other subgroup (Stuessy, 1972). Further data are necessary to ascertain whether sect. Zarabellia is monophyletic or not and, in consequence, whether the two subclades need to be recognized as separate sections or perhaps series.

Species of sect. Alcina share a chromosome base number of x = 11, which is, however, also found in sect. Rhizomaria and the genera Acanthospermum and Lecocarpus, suggesting the plesiomorphic nature of this feature. The potential heterogeneity of sect. Alcina was already acknowledged by phenetic and cladistic analyses of morphological data (Stuessy, 1979; Stuessy and Crisci, 1984), which found M. nutans to be very distinct from the remainder of this section, M. glabrum and M. perfoliatum. This mostly concerns the presence of an achenial hood that is somewhat similar to those of sect. Melampodium plus thin stems and long petioles reminiscent of sect. Serratura. Plastid and nuclear ITS data now congruently suggest that sect. Alcina is polyphyletic (Figs. 1-3). Melampodium perfoliatum congruently ties in the vicinity of sects. Bibractiaria, Serratura and Zarabellia p.p. (clades III–V), albeit with insufficiently supported and contradictory positions (Figs. 1 and 2), but the phylogenetic positions of M. glabrum and M. nutans are essentially unresolved. In order to retain monophyletic groups, sect. Alcina in its current circumscription cannot be maintained and breaking it into three monotypic sections is one clear option.

In summary, chromosome base numbers in Melampodium are to a considerable extent indicative of phylogenetic relationships, as has also been found in other genera (Hypochaeris/Asteraceae: Cerbah et al., 1998; Samuel et al., 2003; Weiss-Schneeweiss et al., 2008; Passiflora/Passifloraceae: Hansen et al., 2006; Pennisetum/Poaceae: Martel et al., 2004; Rhaponticum/Asteraceae and related genera: Hidalgo et al., 2007). Despite some uncertainties and incongruences concerning the phylogenetic position of several lineages, it is obvious that chromosome base number changes (dysploidy) have played an important role in the evolution of Melampodium. The presence of x = 11 in many of the basal lineages, even if their positions are not identical in plastid and nuclear marker phylogenies, suggests x = 11 as the ancestral chromosome base number (maximum parsimony reconstruction, data not shown) rather than the previously hypothesized x = 10 (Stuessy, 1971).

4.3. Polyploidy

Polyploids are found in many groups of Melampodium, and one third of all Melampodium species are exclusively polyploid, three more also including polyploid cytotypes (Stuessy, 1970b, 1971; Stuessy et al., 2004; H. Weiss-Schneeweiss et al., unpubl.). Apart from the single species-clades of M. glabrum, M. longipilum, M. nutans, and M. perfoliatum, only clade II (M. longifolium and M. mimulifolium, sect. Zarabellia p.p.) and clade VI (Acanthospermum and Lecocarpus) are devoid of polyploids. Molecular phylogenetic data, in some cases strongly supported by karyological data (H. Weiss-Schneeweiss et al., unpubl.), indicate that both auto- and allopolyploidy have played significant roles in the evolution of Melampodium.

With increasing evidence for the frequent presence of intraspecific ploidy level variation (e.g., Weiss et al., 2003; Baack, 2004; Stuessy et al., 2004; Suda et al., 2007), recent years have witnessed appreciation of the role of autopolyploidy in speciation (Soltis et al., 2007). Autopolyploid speciation is well supported morphologically and karyologically for sect. Rhizomaria (Stuessy, 1971; H. Weiss-Schneeweiss et al., unpubl.), where the hexaploid M. aureum is morphologically and ecologically so similar to the diploid M. montanum that they have been treated as a single species by McVaugh (1984). In contrast, in sect. Bibractiaria autopolyploidization occurred independently in its two constituent species (Figs. 1-3). In M. repens, a prostrate herb confined to pine-oak forests, both diploids and tetraploids are known (Keil and Stuessy, 1977; the latter reported as 2n = 54, which probably is a miscount for 2n = 4x = 56), while in M. bibracteatum, an erect, subaquatic species of open wetlands, so far only tetraploids are known (H. Weiss-Schneeweiss et al, unpubl.). In species with both diploid and polyploid cytotypes (M. dicoelocarpum of sect. Serratura, and M. cinereum and M. leucanthum of sect. Melampodium), the evolutionary significance of autopolyploidisation is unclear. At least some of these polyploid lineages appear, however, to be genetically cohesive and separated, yet morphologically indistinguishable groups (C. Rebernig et al., unpubl.), as has been suggested for other diploid-autopolyploid complexes (Soltis et al., 2007).

Allopolyploidy is a common phenomenon in Melampodium. Since in allopolyploids nuclear ITS sequences may also converge towards the maternal parent (Álvarez and Wendel, 2003), the lack of incongruence between nuclear and plastid markers per se is no proof of an autopolyploid origin, and consequently from sequence data alone the number of allopolyploid origins might be underestimated. An excellent example is provided by the tetraploid M. nayaritense of sect. Sericea. In both nuclear and plastid sequence data it groups with the diploid M. linearilobum of sect. Melampodium (Figs. 1-3), which turns out to be the likely donor of the set of 20 small chromosomes, whereas the second parent, from which the other set of 20 larger chromosomes was obtained (H. Weiss-Schneeweiss et al., unpubl.), remains elusive. Although M. linearilobum and M. nayaritense have been placed in different series, Melampodium and Sericea (Stuessy, 1972), respectively, a closer relationship between both species was already suggested by phenetic and cladistic analysis of morphological data (Stuessy, 1972; Stuessy and Crisci, 1984).

In clade III (sect. Zarabellia p.p.), the diploids M. gracile and M. microcephalum have been unambiguously shown to be involved in the origin of the tetraploid M. paniculatum (Stuessy and Brunken, 1979). The molecular data show that M. microcephalum comprises different genetic lineages (Figs. 1-3), which independently hybridized with M. gracile and gave rise to a thus polytopic M. paniculatum. Against early assertions, a polytopic origin of an allopolyploid taxon is considered the rule rather than the exception (e.g., Soltis et al., 2004; Leitch and Leitch, 2008). Similarly, in sect. Serratura different lineages within the diploids M. divaricatum, M. tepicense, and M. dicoelocarpum appear to have been involved in the origin of the polyploids M. costaricense, M. sinaloense and, very likely of polytopic origin, M. northingtonii (Figs. 1 and 2).

Numerous cases of allopolyploid speciation are also evident in sect. Melampodium. Of those, only the hexaploid M. argophyllum is found in the same series as its putative parents M. cinereum and M. leucanthum (ser. Leucantha; C. A. Rebernig et al., unpubl.;Figs. 1 and 2). Others appear to be the result of hybridization between species (or their ancestors) of different series. For instance, M. mayfieldii of ser. Melampodium nests within ser. Cupulata in the nuclear ITS data (Fig. 1), but groups with M. diffusum of ser. Melampodium in the matK data (Fig. 2). A hotspot of allopolyploid speciation is the exclusively polyploid ser. Sericea. The tetraploid M. strigosum, itself likely of allopolyploid origin involving possibly ancestors of M. americanum of ser. Melampodium and M. glabribracteatum of ser. Cupulata (Figs. 1-3; C. Blöch et al., unpubl.), is clearly the parental taxon of the three hexaploids M. longicorne, M. sericeum and M. pringlei (Figs. 1-3), with the second parental species either belonging to ser. Cupulata (in case of M. longicorne, Figs. 1-3) or being M. linearilobum of ser. Melampodium (at least in case of M. pringlei, Figs. 1-3), the same species also involved with the origin of the allotetraploid M. nayaritense (see above).

Within sect. Melampodium, several taxonomic series have been distinguished (Stuessy, 1972). The monotypic ser. Longipila might best be treated in its own section (see above). When only diploids (and their autotetraploid derivatives; Stuessy et al., 2004) are considered, ser. Leucantha and ser. Melampodium are monophyletic (except for the position of M. linearilobum in the ITS dataset). Once M. glabribracteatum has been removed from ser. Cupulata and transferred to its own monotypic series, ser. Cupulata also becomes monophyletic. When allopolyploid species are considered as well, this is, however, no longer the case. For one, several species of ser. Sericea nest within ser. Melampodium (Figs. 1-3). Species of ser. Sericea are very small-headed, few flowered, inconspicuous plants adapted to higher elevations, with short ray corollas, whereas those of ser. Melampodium are much more robust in all respects and occur in lower tropical or subtropical environments. The morphological convergence of members of ser. Sericea, despite their different phylogenetic origin, suggests that these characters are directly or indirectly connected with allopolyploidization. Even if ser. Sericea were to be merged with ser. Melampodium, monophyly of a thus enlarged ser. Melampodium is still rejected because M. longicorne and M. mayfieldii clearly connect ser. Cupulata with ser. Melampodium and ser. Sericea (Fig. 3). The complex pattern of a bifurcating phylogenetic structure in diploids overlain with reticulate relationships stemming from the allopolyploids has non-trivial implications for taxonomic classification. Alternatives include eliminating recognition of different series altogether or putting allopolyploids, which have parents belonging to different series, into their own series, although this might not be morphologically diagnosable. A formal re-evaluation of current classification in the light of these new molecular data will be published elsewhere.

Acknowledgments

The authors acknowledge: the excellent laboratory technical assistance of Verena Klejna and Gudrun Kohl; the participation of Jorge Calónico, Ismael Calzada, Gabriel Flores, Michael Lenko, Enrique Ortiz, and Joaquín Sánchez in the field trips; Prof. Dr. Hennig Adsersen for material of Lecocarpus (collection permission: 3474-2005-PNG-Dir); the Botanical Garden of the University of Vienna (HBV) for maintaining the living collections; and CONAGEBIO (Comision Nacional de Gestión en Biodiversidad) for permission to collect Melampodium costaricense and M. divaricatum in Costa Rica. Grant support was provided by the Austrian Science Foundation (FWF; project 18201 to T.F.S., and Hertha-Firnberg postdoctoral fellowship T-218 to H.W.-S.), and the Commission for Interdisciplinary Ecological Studies (KIÖS) of the Austrian Academy of Sciences (2007-12 to T.F.S.).

References

  1. Adsersen H. Revision of the Galapagos endemic genus Lecocarpus (Asteraceae) Bot. Tidsskr. 1980;75:63–76. [Google Scholar]
  2. Álvarez I, Wendel JF. Ribosomal ITS sequences and plant phylogenetic inference. Mol. Phylogenet. Evol. 2003;29:417–434. doi: 10.1016/s1055-7903(03)00208-2. [DOI] [PubMed] [Google Scholar]
  3. Baack EJ. Cytotype segregation at regional and microgeographic scales. Am. J. Bot. 2004;91:1783–1788. doi: 10.3732/ajb.91.11.1783. [DOI] [PubMed] [Google Scholar]
  4. Baillon HE. Histoire des Plantes. Vol. 8. Librairie Hachette; Paris: 1882. [Google Scholar]
  5. Baldwin BG. Heliantheae alliance. In: Funk VA, Susanna A, Stuessy TF, Bayer RJ, editors. Systematics, Evolution, and Biogeography of Compositae. IAPT; Vienna: 2009. [Google Scholar]
  6. Baldwin BG, Wessa BL. Origin and relationships of the tarweed-silversword lineage (Compositae-Madiinae) Am. J. Bot. 2000;87:1890–1908. [PubMed] [Google Scholar]
  7. Baldwin BG, Wessa BL, Panero JL. Nuclear rDNA evidence for major lineages of helenioid Heliantheae (Compositae) Syst. Bot. 2002;27:161–198. [Google Scholar]
  8. Bayer RJ, Puttock CF, Kelchner SA. Phylogeny of South African Gnaphalieae (Asteraceae) based on two noncoding chloroplast sequences. Am. J. Bot. 2000;87:259–272. [PubMed] [Google Scholar]
  9. Bayer RJ, Greber DG, Bagnall NH. Phylogeny of Australian Gnaphalieae (Asteraceae) based on chloroplast and nuclear sequences, the trnL intron, trnL/trnF Intergenic spacer, matK, and ETS. Syst. Bot. 2002;27:801–814. [Google Scholar]
  10. Bohm BA, Stuessy TF. Flavonoid variation in Melampodium. Biochem. Syst. Ecol. 1991;19:677–679. [Google Scholar]
  11. Böhle U-R, Hilger HH, Martin WF. Island colonization and evolution of the insular woody habit in Echium L. (Boraginaceae) Proc. Natl. Acad. Sci. USA. 1996;93:11740–11745. doi: 10.1073/pnas.93.21.11740. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cerbah M, Coulaud J, Siljak-Yakovlev S. rDNA organization and evolutionary relationships in the genus Hypochaeris (Asteraceae) J. Hered. 1998;89:312–318. [Google Scholar]
  13. DeCandolle AP. Prodromus systematis naturalis. Vol. 5. Treuttel et würtz; Paris: 1836. [Google Scholar]
  14. Doyle JJ, Doyle JL. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. Bot. Soc. Am. 1987;19:11–15. [Google Scholar]
  15. Doyle JJ, Doyle JL, Rauscher JT, Brown AHD. Diploid and polyploid reticulate evolution throughout the history of the perennial soybeans (Glycine subgenus Glycine) New Phytol. 2003;161:121–132. [Google Scholar]
  16. Doyle JJ, Doyle JL, Rauscher JT, Brown AHD. Evolution of the perennial soybean polyploid complex (Glycine subgenus Glycine): a study of contrasts. Biol. J. Linn. Soc. 2004;82:583–597. [Google Scholar]
  17. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;32:1792–1797. doi: 10.1093/nar/gkh340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Eliasson U. Studies in Galápagos plants X. The genus Lecocarpus Decaisne. Svensk Bot. Tidsk. 1971;65:245–277. [Google Scholar]
  19. Ellison NW, Liston A, Steiner JJ, Williams WM, Taylor NL. Molecular phylogenetics of the clover genus (Trifolium-Leguminosae) Mol. Phylogenet. Evol. 2006;39:688–705. doi: 10.1016/j.ympev.2006.01.004. [DOI] [PubMed] [Google Scholar]
  20. Farris JS, Kallersjö M, Kluge AG, Bult C. Testing significance of incongruence. Cladistics. 1994;10:315–319. [Google Scholar]
  21. Felsenstein J. Confidence-limits on phylogenies—an approach using the bootstrap. Evolution. 1985;39:783–791. doi: 10.1111/j.1558-5646.1985.tb00420.x. [DOI] [PubMed] [Google Scholar]
  22. Ferguson D, Sang T. Speciation through homoploid hybridization between allotetraploids in peonies (Paeonia) Proc. Natl. Acad. Sci. USA. 2001;98:3915–3919. doi: 10.1073/pnas.061288698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Gatt MK, Hammett KRW, Murray BG. Molecular phylogeny of the genus Dahlia Cav. (Asteraceae, Heliantheae–Coreopsidinae) using sequences derived from the internal transcribed spacers of nuclear ribosomal DNA. Bot. J. Linn. Soc. 2000;133:229–239. [Google Scholar]
  24. Gruenstaeudl M, Urtubey E, Jansen RK, Samuel R, Barfuss MHJ, Stuessy TF. Phylogeny of Barnadesioideae (Asteraceae) inferred from DNA sequence data and morphology. Mol. Phylogenet. Evol. 51:572–587. doi: 10.1016/j.ympev.2009.01.023. doi:10.1016/j.ympev.2009.01.023. [DOI] [PubMed] [Google Scholar]
  25. Guo YP, Ehrendorfer F, Samuel R. Phylogeny and systematics of Achillea (Asteraceae–Anthemideae) inferred from nrITS and plastid trnL-F DNA sequences. Taxon. 2004;53:657–672. [Google Scholar]
  26. Guo YP, Vogl C, Van Loo M, Ehrendorfer F. Hybrid origin and differentiation of two tetraploid Achillea species in East Asia: molecular, morphological and ecogeographical evidence. Mol. Ecol. 2006;15:133–144. doi: 10.1111/j.1365-294X.2005.02772.x. [DOI] [PubMed] [Google Scholar]
  27. Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 1999;41:95–98. [Google Scholar]
  28. Hansen AK, Gilbert LE, Simpson BB, Downie SR, Cervi AC, Jansen RK. Phylogenetic relationships and chromosome number evolution in Passiflora. Syst. Bot. 2006;31:138–150. [Google Scholar]
  29. Hershkovitz MA, Zimmer EA. Conservation patterns in angiosperm rDNA ITS2 sequences. Nucleic Acids Res. 1996;24:2857–2867. doi: 10.1093/nar/24.15.2857. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Hidalgo O, Garcia-Jacas N, Garnatje T, Susanna A, Siljak-Yakovlev S. Karyological evolution in Rhaponticum Vaill. (Asteraceae, Cardueae) and related genera. Bot. J. Linn. Soc. 2007;153:193–201. [Google Scholar]
  31. Hoffmann O. Compositae. In: Engler A, Prantl K, editors. Die natürlichen Pflanzenfamilien. Vol. 4. Engelmann; Leipzig: 1890. pp. 210–267. [Google Scholar]
  32. Holland RB, Huber KT, Moulton V, Lockhart PJ. Using consensus networks to visualize contradictory evidence for species phylogeny. Mol. Biol. Evol. 2004;21:1459–1461. doi: 10.1093/molbev/msh145. [DOI] [PubMed] [Google Scholar]
  33. Hörandl E. Neglecting evolution is bad taxonomy. Taxon. 2007;56:1–5. [Google Scholar]
  34. Huson DH, Bryant D. Application of phylogenetic networks in evolutionary studies. Mol. Biol. Evol. 2006;23:254–267. doi: 10.1093/molbev/msj030. [DOI] [PubMed] [Google Scholar]
  35. Jobes DV, Thien LB. A conserved motif in the 5.8S ribosomal RNA (rRNA) gene is a useful diagnostic marker for plant internal transcribed spacer (ITS) sequences. Plant Mol. Biol. Rep. 1997;15:326–334. [Google Scholar]
  36. Kass RE, Raftery AE. Bayes factors. J. Am. Stat. Assoc. 1995;90:773–795. [Google Scholar]
  37. Keil DJ, Luckow MA, Pinkava DJ. Chromosome studies in Asteraceae from the United States, Mexico, the West Indies, and South America. Am. J. Bot. 1988;75:652–688. doi: 10.1002/j.1537-2197.1988.tb13488.x. [DOI] [PubMed] [Google Scholar]
  38. Keil DJ, Stuessy TF. Chromosome counts of Compositae from the United States, Mexico and Guatemala. Rhodora. 1975;77:171–195. [Google Scholar]
  39. Keil DJ, Stuessy TF. Chromosome counts of Compositae from Mexico and the United States. Am. J. Bot. 1977;64:791–798. [Google Scholar]
  40. Kim SC, Crawford DJ, Francisco-Ortega J, Santos-Guerra A. A common origin for woody Sonchus and five related genera in the Macaronesian islands: molecular evidence for extensive radiation. Proc. Natl. Acad. Sci. USA. 1996;93:7743–7748. doi: 10.1073/pnas.93.15.7743. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Kimball RT, Crawford DJ. Phylogeny of Coreopsideae (Asteraceae) using ITS sequences suggests lability in reproductive characters. Mol. Phylogenet. Evol. 2004;33:127–139. doi: 10.1016/j.ympev.2004.04.022. [DOI] [PubMed] [Google Scholar]
  42. Leitch AR, Leitch IJ. Genome plasticity and diversity of polyploid plants. Science. 2008;320:481–483. doi: 10.1126/science.1153585. [DOI] [PubMed] [Google Scholar]
  43. Lim KY, Matyasek R, Kovarik A, Leitch AR. Genome evolution in allotetraploid Nicotiana. Biol. J. Linn. Soc. 2004;82:599–606. [Google Scholar]
  44. Liu JS, Schardl CL. A conserved sequence in Internal Transcribed Spacer-1 of plant nuclear ribosomal-RNA genes. Plant Mol. Biol. 1994;26:775–778. doi: 10.1007/BF00013763. [DOI] [PubMed] [Google Scholar]
  45. Martel E, Poncet V, Lamy F, Siljak-Yakovlev S, Lejeune B, Sarr A. Chromosome evolution of Pennisetum species (Poaceae): implications of ITS phylogeny. Plant Syst. Evol. 2004;249:139–149. [Google Scholar]
  46. Mast AR, Kelso S, Richards AJ, Lang DJ, Feller DMS, Conti E. Phylogenetic relationships in Primula L. and related genera (Primulaceae) based on noncoding chloroplast DNA. Int. J. Plant Sci. 2001;162:1381–1400. [Google Scholar]
  47. McVaugh R. Compositae. In: Anderson WR, editor. Flora Novo-Galiciana: A Descriptive Account of the Vascular Plants of Western Mexico. Vol. 12. Univ. Michigan Press; Ann Arbor: 1984. [Google Scholar]
  48. Moore AJ, Bohs L. An ITS phylogeny of Balsamorhiza and Wyethia (Asteraceae: Heliantheae) Am. J. Bot. 2003;90:1653–1660. doi: 10.3732/ajb.90.11.1653. [DOI] [PubMed] [Google Scholar]
  49. Müller K. SeqState—primer design and sequence statistics for phylogenetic DNA data sets. Appl. Bioinformatics. 2005;4:65–69. doi: 10.2165/00822942-200504010-00008. [DOI] [PubMed] [Google Scholar]
  50. Müller K. Incorporating information from length-mutational events into phylogenetic analysis. Mol. Phylogenet. Evol. 2006;38:667–676. doi: 10.1016/j.ympev.2005.07.011. [DOI] [PubMed] [Google Scholar]
  51. Nieto Feliner G, Roselló JA. Better the devil you know? Guidelines for insightful utilization of nrDNA ITS in species-level evolutionary studies in plants. Mol. Phylogenet. Evol. 2007;44:911–919. doi: 10.1016/j.ympev.2007.01.013. [DOI] [PubMed] [Google Scholar]
  52. Nylander JAA. MrModeltest v2. Evolutionary Biology Centre, Uppsala University; 2004. Program distributed by the author. [Google Scholar]
  53. Panero JL. Compositae: tribe Millerieae. In: Kadereit JW, Jeffrey C, editors. Families and Genera of Vascular Plants. Flowering Plants, Eudicots, Asterales. VIII. Springer-Verlag; Berlin: 2007. pp. 477–492. [Google Scholar]
  54. Park J-M, Schneeweiss GM, Weiss-Schneeweiss H. Diversity and evolution of Ty1-copia and Ty3-gypsy retroelements in the non-photosynthetic flowering plants Orobanche and Phelipanche (Orobanchaceae) Gene. 2007;387:75–86. doi: 10.1016/j.gene.2006.08.012. [DOI] [PubMed] [Google Scholar]
  55. Plovanich AE, Panero JL. A phylogeny of the ITS and ETS for Montanoa (Asteraceae: Heliantheae) Mol. Phylogenet. Evol. 2004;31:815–821. doi: 10.1016/j.ympev.2003.10.021. [DOI] [PubMed] [Google Scholar]
  56. Rauscher JT. Molecular phylogenetics of the Espeletia complex (Asteraceae): evidence from nrDNA ITS sequences on the closest relatives of an Andean adaptive radiation. Am. J. Bot. 2002;89:1074–1084. doi: 10.3732/ajb.89.7.1074. [DOI] [PubMed] [Google Scholar]
  57. Redelings B, Suchard M. Joint Bayesian estimation of alignment and phylogeny. Syst. Biol. 2005;54:401–418. doi: 10.1080/10635150590947041. [DOI] [PubMed] [Google Scholar]
  58. Rieseberg LH. Homoploid reticulate evolution in Helianthus (Asteraceae): evidence from ribosomal genes. Am. J. Bot. 1991;78:1218–1237. [Google Scholar]
  59. Rieseberg LH. Chromosomal rearrangements and speciation. Trends Ecol. Evol. 2001;16:351–358. doi: 10.1016/s0169-5347(01)02187-5. [DOI] [PubMed] [Google Scholar]
  60. Rieseberg LH, Kim S-C, Randell RA, Whitney KD, Gross BL, Lexer C, Clay K. Hybridization and the colonization of novel habitats by annual sunflowers. Genetica. 2007;129:149–165. doi: 10.1007/s10709-006-9011-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Robinson BL. Synopsis of the genus Melampodium. Proc. Am. Acad. Arts Sci. 1901;36:455–466. [Google Scholar]
  62. Ronquist F, Huelsenbeck JP. MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformatics. 2003;19:1572–1574. doi: 10.1093/bioinformatics/btg180. [DOI] [PubMed] [Google Scholar]
  63. Saar DE, Polans NO, Sorensen PD. A phylogenetic analysis of the genus Dahlia (Asteraceae) based on internal and external transcribed spacer regions of nuclear ribosomal DNA. Syst. Bot. 2003;28:627–639. [Google Scholar]
  64. Samuel R, Gutermann W, Stuessy TF, Ruas CF, Lack HW, Tremetsberger K, Talavera S, Hermanowski B, Ehrendorfer F. Molecular phylogenetics reveals Leontodon (Asteraceae, Lactuceae) to be diphyletic. Am. J. Bot. 2006;93:1193–1205. doi: 10.3732/ajb.93.8.1193. [DOI] [PubMed] [Google Scholar]
  65. Samuel R, Kathriarachchi H, Hoffmann P, Barfuss MHJ, Wurdack KJ, Davis CC, Chase MW. Molecular phylogenetics of Phyllanthaceae: evidence from plastid matK and nuclear phyC sequences. Am. J. Bot. 2005;92:132–141. doi: 10.3732/ajb.92.1.132. [DOI] [PubMed] [Google Scholar]
  66. Samuel R, Stuessy TF, Tremetsberger K, Baeza CM, Siljak-Yakovlev S. Phylogenetic relationships among species of Hypochaeris (Asteraceae, Cichorieae) based on ITS, plastid trnL intron, trnL-F spacer, and matK sequences. Am. J. Bot. 2003;90:496–507. doi: 10.3732/ajb.90.3.496. [DOI] [PubMed] [Google Scholar]
  67. Sang T, Crawford DJ, Stuessy TF. Chloroplast DNA phylogeny, reticulate evolution, and biogeography of Paeonia (Paeoniaceae) Am. J. Bot. 1997;84:1120–1136. [PubMed] [Google Scholar]
  68. Schneeweiss GM, Palomeque T, Colwell AE, Weiss-Schneeweiss H. Chromosome numbers and karyotype evolution in holoparasitic Orobanche (Orobanchaceae) and related genera. Am. J. Bot. 2004a;91:439–448. doi: 10.3732/ajb.91.3.439. [DOI] [PubMed] [Google Scholar]
  69. Schneeweiss GM, Schönswetter P, Kelso S, Niklfeld H. Complex biogeographic patterns in Androsace (Primulaceae) and related genera: evidence from phylogenetic analyses of nuclear ITS and plastid trnL-F sequences. Syst. Biol. 2004b;53:856–876. doi: 10.1080/10635150490522566. [DOI] [PubMed] [Google Scholar]
  70. Schubert I. Chromosome evolution. Curr. Opin. Pl. Evol. 2007;10:109–115. doi: 10.1016/j.pbi.2007.01.001. [DOI] [PubMed] [Google Scholar]
  71. Seaman FC, Fischer NH, Stuessy TF. Systematic implications of sesquiterpene lactones in the subtribe Melampodiinae. Biochem. Syst. Ecol. 1980;8:263–271. [Google Scholar]
  72. Soltis DE, Soltis PS, Pires JC, Kovarik A, Tate JA, Mavrodiev E. Recent and recurrent polyploidy in Tragopogon (Asteraceae): genetic, genomic, and cytogenetic comparisons. Biol. J. Linn. Soc. 2004;82:485–501. [Google Scholar]
  73. Soltis DE, Soltis PS, Schemske DW, Hancock JF, Thomspon JN, Husband BC, Judd WS. Autopolyploidy in angiosperms: have we grossly underestimated the number of species? Taxon. 2007;56:13–30. [Google Scholar]
  74. Sønderberg Brok C, Adsersen H. Morphological variation among populations of Lecocarpus (Asteraceae) on the Galápagos Islands. Bot. J. Linn. Soc. 2007;154:523–544. [Google Scholar]
  75. Stuessy TF. Dissertation. University of Texas; Austin: 1968. A systematic study of the genus Melampodium (Compositae–Heliantheae) [Google Scholar]
  76. Stuessy TF. Re-establishment of the genus Unxia (Compositae–Heliantheae) Brittonia. 1969;21:314–321. [Google Scholar]
  77. Stuessy TF. The genus Acanthospermum (Compositae–Heliantheae–Melampodinae): taxonomic changes and generic affinities. Rhodora. 1970a;72:106–109. [Google Scholar]
  78. Stuessy TF. Chromosome studies in Melampodium (Compositae, Heliantheae) Madroño. 1970b;20:365–372. [Google Scholar]
  79. Stuessy TF. Chromosome numbers and phylogeny in Melampodium (Compositae) Am. J. Bot. 1971;58:732–736. [Google Scholar]
  80. Stuessy TF. Revision of the genus Melampodium (Compositae: Heliantheae) Rhodora. 1972;74(1-71):161–217. [Google Scholar]
  81. Stuessy TF. A systematic review of the subtribe Melampodiinae (Compositae, Heliantheae) Contrib. Gray Herb. Harvard Univ. 1973;203:65–85. [Google Scholar]
  82. Stuessy TF. Cladistics of Melampodium (Compositae) Taxon. 1979;28:179–195. [Google Scholar]
  83. Stuessy TF. Classification: more than just branching patterns of evolution. Aliso. 1997;15:113–124. [Google Scholar]
  84. Stuessy TF, Brunken JN. Artificial interspecific hybridization in Melampodium section Zarabellia (Compositae) Madroño. 1979;26:53–63. [Google Scholar]
  85. Stuessy TF, Crisci JV. Phenetics of Melampodium (Compositae, Heliantheae) Madroño. 1984;31:8–19. [Google Scholar]
  86. Stuessy TF, Weiss-Schneeweiss H, Keil DJ. Diploid and polyploid cytotype distribution in Melampodium cinereum and M. leucanthum (Asteraceae, Heliantheae) Am. J. Bot. 2004;91:889–898. doi: 10.3732/ajb.91.6.889. [DOI] [PubMed] [Google Scholar]
  87. Suchard MA, Weiss RE, Dorman KS, Sinsheimer JS. Inferring spatial phylogenetic variation along nucleotide sequences: a multiple change-point model. J. Am. Stat. Ass. 2003;98:427–437. [Google Scholar]
  88. Suchard M, Weiss R, Sinsheimer J. Bayesian selection of continuous-time Markov chain evolutionary models. Mol. Biol. Evol. 2001;18:1001–1013. doi: 10.1093/oxfordjournals.molbev.a003872. [DOI] [PubMed] [Google Scholar]
  89. Suda J, Weiss-Schneeweiss H, Tribsch A, Schneeweiss GM, Trávníček P, Schönswetter P. Complex distribution patterns of di-, tetra- and hexaploid cytotypes in the European high mountain plant Senecio carniolicus (Asteraceae) Am. J. Bot. 2007;94:1391–1401. doi: 10.3732/ajb.94.8.1391. [DOI] [PubMed] [Google Scholar]
  90. Sun Y, Skinner DZ, Liang GH, Hulbert SH. Phylogenetic analysis of Sorghum and related taxa using Internal Transcribed Spacers of nuclear ribosomal DNA. Theor. Appl. Genet. 1994;89:26–32. doi: 10.1007/BF00226978. [DOI] [PubMed] [Google Scholar]
  91. Swofford DL. PAUP*: Phylogenetic analysis using parsimony (* and other methods), Version 4.0b.10 for 32-Bit Microsoft Windows. Sinauer Associates; Sunderland, MA, USA: 2001. [Google Scholar]
  92. Tel-Zur N, Abbo S, Myslabodski D, Mizrahi Y. Modified CTAB procedure for DNA isolation from epiphytic cacti of the genera Hylocereus and Selenicereus (Cactaceae) Plant Mol. Biol. Rep. 1999;17:249–254. [Google Scholar]
  93. Turner BL. A new species of Melampodium (Asteraceae–Heliantheae) from Oaxaca, Mexico. Phytologia. 1988;64:445–447. [Google Scholar]
  94. Turner BL. A new species of Melampodium (Asteraceae, Heliantheae) from Jalisco, Mexico. Phytologia. 1993;75:136–139. [Google Scholar]
  95. Turner BL. Melampodium moctezumum (Asteraceae: Heliantheae), a new species from Sonora, Mexico. Phytologia. 2007;89:258–262. [Google Scholar]
  96. Turner BL, King RM. A cytotaxonomic survey of Melampodium (Compositae–Heliantheae) Am. J. Bot. 1961;49:263–269. [Google Scholar]
  97. Weiss H, Stuessy TF, Grau J, Baeza CM. Chromosome reports from South American Hypochaeris (Asteraceae) Ann. MO Bot. Gard. 2003;90:56–63. [Google Scholar]
  98. Weiss-Schneeweiss H, Schneeweiss GM, Stuessy TF, Mabuchi T, Park J-M, Jang C-G, Sun B-Y. Chromosomal stasis in diploids contrasts with genome restructuring in auto- and allopolyploid taxa of Hepatica (Ranunculaceae) New Phytol. 2007;174:669–682. doi: 10.1111/j.1469-8137.2007.02019.x. [DOI] [PubMed] [Google Scholar]
  99. Weiss-Schneeweiss H, Tremetsberger K, Schneeweiss GM, Parker JS, Stuessy TF. Karyotype diversification and evolution in diploid and polyploid South American Hypochaeris (Asteraceae) inferred from rDNA localization and genetic fingerprint data. Ann. Bot. 2008;101:909–918. doi: 10.1093/aob/mcn023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  100. White TJ, Bruns T, Lee S, Taylor J. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ, editors. PCR Protocols: A Guide to Methods and Applications. Academic Press Inc.; San Diego: 1990. pp. 315–322. [Google Scholar]
  101. Yuan YM, Song Y, Geuten K, Rahelivololona E, Wohlhauser S, Fischer E, Smets E, Kupfer P. Phylogeny and biogeography of Balsaminaceae inferred from ITS sequences. Taxon. 2004;53:391–403. [Google Scholar]

RESOURCES