| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Systematics |
2Department of Systematic Botany, Evolutionary Biology Centre, Uppsala Universitet, Norbyvägen 18D, SE-752 36 Uppsala, Sweden; 3Leslie Hill Molecular Systematics Laboratory, National Botanical Institute, Kirstenbosch, Private Bag X7, Cape Town, 7735, South Africa
Received for publication January 30, 2003. Accepted for publication April 24, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: atpB-rbcL spacer • Colchicaceae • phylogeny • rps16 intron • trnL-F
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
) showed that the family is much younger. To address biogeographical questions further within the family, a firmly supported phylogeny is required. The aim of this study is to reconstruct such a phylogeny for the family at the generic level using chloroplast DNA (cpDNA) sequence data.
The taxonomic history of the family began in 1805 when de Candolle was the first to use the family name Colchicaceae, in the Flore Française. He included six genera: Bulbocodium, Colchicum, and Merendera, which are still members of the family, and Erythronium, Tofieldia, and Veratrum, today assigned to Liliaceae, Tofieldiaceae, and Melanthiaceae, respectively. In the following 100 to 150 years, all these genera were mostly treated as a part of Liliaceae sensu lato (s.l.). Buxbaum (1925
, 1936
, 1937
) undertook a thorough morphological study of Liliaceae s.l. and erected a new subfamily, the Wurmbeoideae, in which he placed six tribes—Neodregeae: Neodregea and Onixotis (syn. Dipidax); Wurmbaeae: Wurmbaea and Anguillaria; Colchiceae: Androcymbium and Colchicum (including Bulbocodium and Merendera); Baeometrae: Baeometra; Iphigenieae: Iphigenia, Camptorrhiza (Iphigeniopsis), and Ornithoglossum; and Glorioseae: Gloriosa, Littonia, and Sandersonia. The rhizomatous genera remained in the subfamily Melanthioideae; Buxbaum arranged these into four tribes, in one of which, the Uvularieae, he placed Burchardia.
Nordenstam (1982)
summarized the taxonomic history of Colchicaceae and suggested some new tribal delimitations. The outline of his classification to a large extent followed Buxbaum's (1936)
classification. He merged Neodregeae and Wurmbaeae into one tribe, the Anguillareae, recognized by sessile flowers and ebracteate spicate inflorescences. In Buxbaum's classification, they were separated according to free or connate tepals, but Nordenstam noted the existence of intermediate species. He also recognized the genus Hexacyrtis to be intermediate between Buxbaum's two tribes Iphigenieae and Glorioseae and therefore merged them all into a single tribe, Iphigeneae. The amended tribe is characterized by racemose, axillary flowers, and monosulcate pollen grains. Buxbaum's tribes, Baeometrae and Colchiceae, were left unchanged.
In the 1990s, molecular studies (Chase et al., 1993
, 1995
; Rudall et al., 1997
) indicated some aberrant genera, such as Burchardia, Tripladenia, Uvularia, and Disporum, to be members of the Colchicaceae. Nordenstam (1998)
expanded the family to include 19 genera and roughly 225 species. He also pointed to the necessity of a new taxonomy for the family, both at the tribal and the subfamilial level. He presented the possible solution of dividing the family into two subfamilies, one including the wurmbaeoid genera, distinguished by having a corm, parallel sheathing leaves, dry capsules, and the presence of colchicine alkaloids. The other subfamily was confined to the uvularioid genera, circumscribed by their nonsheathing leaves, sometimes with reticulate nerves, rhizomes, dry or fleshy capsules, and alkaloids without a troplone ring. An outline of today's classification of Colchicaceae, including remarks on previous position of the uvularioid genera, is shown in Table 1.
|
). Subsequent molecular studies have placed the genus as a sister taxon to Tripladenia (APG, 1998
; Rudall et al., 2000
; Vinnersten and Bremer, 2001
). However, the close relationship of the two taxa has not been concordant with morphology, leaving Petermannia in an ambiguous position (J. Conran, University of Adelaide, personal communication). The problem with the position of Petermannia is that a misidentified specimen of Tripladenia was used as the source of DNA for the original "Petermannia" rbcL sequence published in 1997 (M. Chase, Royal Botanic Gardens, Kew, personal communication; J. I. Davis, Cornell University, personal communication). J. I. Davis (unpublished manuscript) has shown, in a two-gene analysis, that a positively identified specimen of Petermannia does not belong in Colchicaceae but rather forms a sister taxon to a clade of Colchicaceae, Alstroemeriaceae, and Luzuriagaceae. The genus is therefore not included in this study.
The taxonomic treatment of a few genera within Colchicaceae is still ambiguous. For example, Colchicum is frequently taken to include the genera Merendera and Bulbocodium, but some authors distinguish three genera based on style and tepal characters. We agree with Dahlgren et al. (1985)
, Persson (1993)
, and Nordenstam (1998)
in the use of a widely circumscribed genus. However, in this study we have kept the old generic names to facilitate the recognition of their placement in the phylogenetic tree. Another problematic genus is Gloriosa, as some authors regard it as a variable monotypic genus, while others recognize several different species. Once again, though we agree with the wider circumscription recognizing one species of Gloriosa, we have kept the old species names to facilitate the identification of their location within the tree.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
were included in this analysis. The only representative of the genus Iphigenia in New Zealand, I. novae-zelandiae, has some deviating features and the position in the genus has been questioned (Nordenstam, 1998
). A recent study demonstrated I. novae-zelandiae to be nested within Wurmbea (Case, 2000
), and it was therefore not included here. In the case of Colchicum, one species turned out to be a garden hybrid (C. agrippinum), but because material of Colchicum was rather scarce it was included in this analysis.
Based on previous studies (Rudall et al., 2000
; Vinnersten and Bremer, 2001
) and a preliminary analysis of 53 taxa and three outgroup taxa (Alstroemeria, Tulipa, and Trillium), Colchicaceae is considered to be a well-supported monophyletic group (100% jackknife support). Only one taxon from one of the sister families (Alstroemeriaceae) was chosen for the analysis to facilitate ease of alignment.
Amplification
The majority of the DNA extractions and sequences were performed at the Leslie Hill Molecular Systematics Laboratory, Kirstenbosch, in Cape Town, South Africa, and the rest at the Department of Systematic Botany, Uppsala, Sweden. The extractions were made from mainly fresh and silica-dried material according to the 2x cetyltrimethyl ammonium bromide (CTAB) method of Doyle and Doyle (1987)
. The DNA was purified with QIAquick polymerase chain reaction (PCR) columns (Qiagen, Santa Clarita, California, USA) according to the instructions provided by the manufacturer for cleaning PCR products. The PCR amplifications were performed with Taq polymerase (0.0025 units per reaction) (Promega, Madison, Wisconsin, USA) according to the manufacturer's protocol. Three plastid regions were amplified and sequenced. The rps16 intron was amplified using the primers of Oxelman et al. (1996)
, while the atpB-rbcL intergenic spacer region was amplified with the primers of Mannen et al. (1994)
and Savolainen et al. (1994)
. The PCR reactions consisted of 30 cycles, 1 min denaturation at 95°C, 30 s annealing at 48°C, 1 min extension at 72°C, and 7 min final extension. A few samples were amplified at Uppsala University using 35 cycles and an annealing temperature of 50° or 55°C. The intron and adjacent intergenic spacer region of trnL-trnF was amplified mainly with the same PCR parameters, but some taxa required 37 cycles and a longer annealing period. The region was either amplified in one or two pieces using the c, d, e, and f primers (Taberlet et al., 1991
) or by two new primers designed for the project (Table 2). The PCR products were purified with QIAquick PCR kit (Qiagen) according to the instructions provided by the manufacturer (using double distilled water as the eluting agent). Automated sequencing was performed on an ABI PRISM 377 automated sequencer (Applied Biosystems, Foster City, California, USA) with ABI PRISM BigDye Terminator Cycle Sequencing Ready reaction kit (Applied Biosystems) in the Leslie Hill Molecular Systematics Laboratory and a MegaBACE 1000 capillary machine (Amersham Pharmacia Biotech, Uppsala, Sweden) in the laboratory of the Systematic Botany Department in Uppsala, both according to the manufacturer's instructions. All sequences were generated using the primers used for initial amplification.
|
), with all characters weighted equally (Fitch parsimony; Fitch, 1971
). A heuristic search with 1000 random addition sequence replicates and tree bisection reconnection (TBR) branch swapping was used to find the most parsimonious reconstructions. Support values were calculated using both the jackknife (Farris et al., 1996
) and bootstrap (Felsenstein, 1985
) options implemented in PAUP* with a heuristic search, 1000 replicates, TBR swapping, five addition sequence replicates, and the MULTREES option not in effect. The three gene regions were also analyzed separately with the same settings. |
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The only genus that is not represented in this study is Kuntheria. It is a monotypic genus restricted to the rain forests of northern Queensland. The morphological attributes of the genus place it together with Schelhammera and Tripladenia. The distinguishing features separating Kuntheria from the two previous genera are a perennial stem, introrse anther dehiscence, a pedunculate umbel, typical venation, and lack of basal perianth appendages (Conran and Clifford, 1987
).
The combined analysis resulted in 324 equally most parsimonious trees. The number of trees is mainly from the low levels of variation within the South African genus Androcymbium and its complicated relationships to Colchicum. Excluding all the Androcymbium and Colchicum species except A. ciliolatum and C. autumnale, which were retained to represent the two genera, resulted in 18 equally most parsimonious trees (A. ciliolatum and C. autumnale were chosen randomly from the two stable major clades). The topologies of these 18 trees only differ with respect to relationships within Gloriosa, Ornithoglossum, and Uvularia. Either the two sequences of G. superba (one African, the other Australian) group together or the Australian G. superba forms a group with G. simplex or, in the last case, there is a trichotomy comprising all three species. Within the genus Ornithoglossum, either O. vulgare or O. undulatum is the sister taxon to the rest of the genus or they form a monophyletic group together. In Uvularia, U. sessilifolia groups either with U. grandiflora or U. perfoliata.
Of the four tribes of Wurmbeoideae (Table 1), the Baeometreae is of course monophyletic because it is monotypic. The position of the tribe as a sister to the Anguillarieae (100% jackknife, Fig. 1) confirms the close relationship between these two tribes (Buxbaum, 1936
; Nordenstam, 1982
; Dahlgren et al., 1985
). Within the Anguillarieae, it should be noted that the two species of Onixotis do not form a monophyletic group but rather a basal grade to Wurmbea. The atpB-rbcL spacer matrix even supports a clade in which Neodregea glassii and O. punctata group together with jackknife support of 75% (Fig. 3). There is another node within Wurmbea in which one of the matrices supports a slightly different resolution from that of the combined analysis. The rps16 intron matrix places Wurmbea variabilis as a sister taxon to the rest of the African wurmbeas (67% jackknife, Fig. 2), while the other matrices support the grouping of W. variabilis and W. marginata (99% jackknife in the combined analysis).
The tribe Colchiceae, comprising the two genera Androcymbium and Colchicum, forms a well-supported clade in our analysis (100% jackknife, Fig. 1). However, the relationships within the tribe are somewhat confusing. Both Colchicum and Androcymbium are well-known genera, comprising around 60 and 90 species, respectively. Morphologically, they are distinguished by the size of flowers, leaf insertion, and by the large bracts subtending the inflorescences of Androcymbium. An earlier study by Caujapé-Castells et al. (1999)
concluded Androcymbium to be monophyletic. Their sampling included only Colchicum and Androcymbium, therefore no conclusions regarding the monophyly of Androcmbium could be drawn. Our wider sampling shows Colchicum to be nested within Androcymbium. The topology of the combined analysis consists of two well-supported clades within the tribe, one comprising Androcymbium species (100% jackknife) and the other including both Colchicum (including Merendera and Bulbocodium) and Androcymbium species (95% jackknife). The separate matrices show less resolution but no contradictions. There is no easily recognized character separating the two clades either morphologically or ecologically.
Another anomaly is that Merendera schimperiana forms a clade with A. europaeum, the only European species of Androcymbium included in the analysis (71% jackknife, Fig. 1). The source material used in this study was collected in Yemen and is now grown in the botanical garden in Göteborg. To rule out contamination, a second rps16 intron sequence was generated from herbarium material, from one of the collections cited as M. schimperiana (previously M. longifolia) in Flora of Somalia (Thulin, 1995
). The two sequences of M. schimperiana form a monophyletic group (95% jackknife, Fig. 2) in the same position as in the combined analysis. M. schimperiana (= M. abyssinica) was already by Stefanoff (1926)
regarded as a Androcymbium (A. abyssinica) but few authors have followed suit.
Nordenstam included Camptorrhiza, Gloriosa, Hexacyrtis, Iphigenia, Ornithoglossum, and Sandersonia in the tribe Iphigenieae (Table 1). The molecular analysis shows this to be a paraphyletic assemblage, because the genera form four different clades in the combined tree. Camptorrhiza and Iphigenia group together with jackknife support of 100% (Fig. 1). Gloriosa and Littonia also form a well-supported clade (100% jackknife), although not as two separate genera. Gloriosa minor and L. revoilii form a clade (99% jackknife) with the rest of Gloriosa as a sister group (99% jackknife). The sister taxa to this clade of Gloriosa and L. minor is L. modesta with a support of 100% jackknife (Fig. 1). The separate matrices show no incongruence. Many authors have regarded Gloriosa as a monotypic genus with large morphological variation under the name G. superba (Field, 1972
; Thulin, 1995
; Nordenstam, 1998
). The genera Gloriosa and Littonia were separated because of the reflexed tepals and a bent style (with an angle of approximately 90°). Nordal and Bingham (1998)
described an intermediate species, G. sessiliflora, which bridged the gap between the two genera by resembling Gloriosa in general habit and Littonia in connivent tepals. The new species was also intermediate in the style character, having a slightly bent style. Unfortunately, there was no suitable material of the intermediate species available for DNA extractions.
The positions of the three remaining genera are ambiguous. In the combined analysis, taxa of Ornithoglossum group together with Sandersonia with very low support (59% jackknife), and Hexacyrtis occupies a position as sister taxon to the Androcymbium-Colchicum clade, also with low support (63% jackknife) (Fig. 1). The low support is due to the fact that the three different gene regions support different placements. In the rps16 intron tree (Fig. 2), Sandersonia and Ornithoglossum constitute a clade (64% jackknife support), while Hexacyrtis is the sister taxon to the Androcymbium-Colchicum clade (78% jackknife). In the trnL-F tree (Fig. 4), Sandersonia forms a clade with Hexacyrtis (55% jackknife support), leaving the interrelationship between this clade, Ornithoglossum, and the Androcymbium-Colchicum clade unresolved. The atpB-rbcL spacer tree (Fig. 3) shows even less resolution, leaving the position of Hexacyrtis, Ornithoglossum, Sandersonia, the Androcymbium-Colchicum, and the Gloriosa-Littonia clade unresolved. More molecular characters are needed to resolve the relationships among these taxa.
The sister group to Colchicaceae sensu stricto (Nordenstam, 1982
) (13 genera of mainly African distribution) consists of a clade of the Australian taxa Schelhammera and Tripladenia (100% jackknife). Sister to this, with a jackknife support of 97%, is a well-supported clade of Uvularia and Disporum. Uvularia is restricted to North America, while Disporum is native to Asia. Closest to the outgroup, we find the Australian genus, Burchardia. This genus is here shown to be paraphyletic with strong support (100% jackknife). Unfortunately, we were not able to obtain material from more than two of the six species; adding the other species might show the genus to be monophyletic.
The strong patterns of relationship recovered in this study demonstrate the need for a taxonomic revision of the Colchicaceae. The paraphyletic nature of the tribe Iphigenieae has to be sorted out, either by describing three new tribes—for the lineages showed not to belong in the Iphigenieae—or perhaps best by revising the entire tribal system, leaving the wurmbeaoid part of the family with two larger tribes. The generic status of Androcymbium, Colchicum and Gloriosa, Littonia also needs to be addressed. Androcymbium will have to be merged into Colchicum generating a large, morphologically well-circumscribed genus. The monotypic genus Gloriosa will be expanded to include Littonia, resulting in a more coherent genus (J. Manning and A. Vinnersten, unpublished manuscript; A. Vinnersten and J. Manning, unpublished manuscript).
|
FOOTNOTES |
|---|
4 annika.vinnersten{at}ebc.uu.se
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Brummitt R. K. 1992 Vascular plant families and genera. Royal Botanic Gardens, Kew, UK
Buxbaum F. 1925 Vergleichende Anatomie der Melanthioideae. Feddes Repertorium specierum novarum regni vegetabilis, Beiheft 29. Verlad des Repertoriums, Berlin, Germany
Buxbaum F. 1936 Die Entwicklungslinien der Lilioideae. I. Die systematische Stellung der Gattung Gagea. Botanisches Archiv 38: 213-293
Buxbaum F. 1937 Die Entwicklungslinien der Lilioideae. II. Die Wurmbaeoideae. Botanisches Archiv 38: 305-398
Caujapé-Castells J. R. K. Jansen J. Pedrola-Monfort N. Membrives 1999 Chloroplast DNA restriction site phylogeny of the genus Androcymbium (Colchicaceae). Systematic Botany 24: 581-597[CrossRef][Web of Science]
Case A. L. 2000 The evolution of combined versus separate sexes in Wurmbea (Colchicaceae). Ph.D. thesis, University of Toronto, Toronto, Ontario, Canada
Chase M. W. D. E. Soltis R. G. Olmstead 1993 Phylogenetics of seed plants: an analysis of nucleotide sequences from the plastid gene rbcL. Annals of the Missouri Botanical Garden 80: 528-580[CrossRef][Web of Science]
Chase M. W. M. R. Duvall H. G. Hills J. G. Conran A. V. Cox L. E. Eguiarte J. Hartwell M. F. Fay L. R. Caddick K. M. Cameron S. Hoot 1995 Molecular phylogenetics of Lilianae. In P. J. Rudall, P. J. Cribb, D. F. Cutler, and C. J. Humphries [eds.], Monocotyledons: systematics and evolution, 109–137. Royal Botanic Gardens, Kew, UK
Conran J. G. H. T. Clifford 1987 In G. S. Alexander [ed.], Flora of Australia, vol. 45, 387–418. Griffin Press, Netley, Australia
Dahlgren R. M. T. H. T. Clifford P. F. Yeo 1985 The families of monocotyledons. Springer-Verlag, Berlin, Germany
de Candolle A. P. 1805 Colchicaceae. In M. M. Lamarck and A. P. de Candolle [eds.], Flore française, vol. 3, 192–193. Stoupe,Paris, France
Doyle J. J. J. L. Doyle 1987 A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11-15
Farris J. S. V. A. Albert M. Källersjö D. Lipscomb A. G. Kluge 1996 Parsimony jackknifing outperforms neighbour-joining. Cladistics 12: 99-124[CrossRef][Web of Science]
Felsenstein J. 1985 Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783-791[CrossRef][Web of Science]
Field D. V. 1972 The identity of Gloriosa simplex. Kew Bulletin 25: 243-244[CrossRef]
Fitch W. M. 1971 Toward defining the course of evolution: minimum change for a specific tree topology. Systematic Zoology 20: 406-416[CrossRef][Web of Science]
Mannen J. F. A. Natalie F. Ehrendorfer 1994 Phylogeny of Rubiaceae-Rubieae inferred from the sequence of a cpDNA intergenic region. Plant Systematics and Evolution 190: 195-211[CrossRef][Web of Science]
Nordal I. G. Bingham 1998 Description of a new species, Gloriosa sessiliflora (Colchicaceae), with notes on the relationship between Gloriosa and Littonia. Kew Bulletin 53: 479-482
Nordenstam B. 1982 A monograph of the genus Ornithoglossum (Liliaceae). Opera Botanica 64
Nordenstam B. 1998 Colchicaceae. In K. Kubitzki [ed.], The families and genera of vascular plants, vol. 3, 175–185. Springer-Verlag,Berlin, Germany
Oxelman B. M. Lidén D. Berglund 1996 Chloroplast rps16 intron phylogeny of the tribe Sileneae (Chareophyllaceae). Plant Systematics and Evolution 206: 393-410[CrossRef][Web of Science]
Persson K. 1993 Reproductive strategies and evolution in Colchicum. Proceedings of the 5th Optima meeting, 394–414. Istanbul University, Istanbul, Turkey
Rudall P. J. C. A. Furness M. W. Chase M. F. Fay 1997 Microsporogenesis and pollen sulcus type in Asparagales (Lilianae). Canadian Journal of Botany 75: 408-430
Rudall P. J. K. L. Stobart W.-P. Hong J. G. Conran C. A. Furness G. Kite M. W. Chase 2000 Consider the lilies: systematics of Liliales. In K. L. Wilson and D. A. Morrison [eds.],Monocots: systematics and evolution. Proceedings of the 2nd International Conference on the Comparative Biology of Monocotyledons, 347–359. CSIRO Publishing, Melbourne, Australia
Savolainen V. J. F. Manen E. Douzery R. Spichiger 1994 Molecular phylogeny of families related to Celestrales based on rbcL 5' flanking sequences. Molecular Phylogenetics and Evolution 3: 27-37[CrossRef][Medline]
Stefanoff B. 1926 Monografiya na roda Colchicum L. Sbornik Balgarsk Akademie Nauk Sofiya 22: 1-100
Swofford D. L. 2003 PAUP*: phylogenetic analysis using parsimony, 4th version. Sinauer, Sunderland, Massachusetts, USA
Taberlet P. L. Gielly G. Pautou J. Bouvet 1991 Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105-1109[CrossRef][Web of Science][Medline]
Thulin M. 1995 Colchicaceae. In M. Thulin [ed.], Flora of Somalia, vol. 4.Royal Botanic Gardens, Kew, UK
Vinnersten A. K. Bremer 2001 Age and biogeography of major clades in Liliales. American Journal of Botany 88: 1695-1703
This article has been cited by other articles:
|
|
 |
|
  J. Shaw, E. B. Lickey, J. T. Beck, S. B. Farmer, W. Liu, J. Miller, K. C. Siripun, C. T. Winder, E. E. Schilling, and R. L. Small The tortoise and the hare II: relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis Am. J. Botany, January 1, 2005; 92(1): 142 - 166. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
