Chloroplast DNA restriction site variation and phylogenetic interrelationships of some genera of the Heliantheae sensu lato (Asteraceae)

Phylogenetic interrelationships of 12 genera of the Heliantheae s. lat., as well as two genera of the Anthemideae and Eupatorieae, were assessed by chloroplast DNA restriction site analysis. The 14 investigated species represented Guizotia, Helianthus, Melampodium, Rudbeckia, Sigesbeckia, Tridax , and Verbesina (Heliantheae s. str.), Helenium, Hymenoxys , and Palafoxia (Helenieae), Porophyllum (Tageteae), Coreopsis (Coreopsideae), Pleurocoronis (Eupatorieae), and Anthemis , (Anthemideae). Using 9 enzymes, 39 phylogenetically informative sites were found. Wagner parsimony analysis of the data yielded 10 equally parsimonious cladograms. Well supported relationships displayed by all cladograms included (1) a clade comprising all the genera (Heliantheae s. lat. and Eupatorieae) except Anthemis (Anthemideae), (2) a clade consisting of the genera of the Heliantheae s. str. with Pleurocoronis (Eupatorieae) nested among them, (3) a subclade of the former comprising Pleurocoronis, Guizotia , and Sigesbeckia , and (4) the generic pair of Helenium and Hymenoxys.     

CHLOROPLAST DNA RESTRICTION SITE VARIATION IN THE FERN GENUS PELLAEA: PHYLOGENETIC RELATIONSHIPS OF THE PELLAEA GLABELLA COMPLEX

The cheilanthoid ferns have long resisted efforts to circumscribe well-defined, phylogenetically natural generic and infrageneric groups, presumably because of homoplastic morphologies associated with their xeric habitats. This cladistic analysis of phylogenetically informative chloroplast DNA restriction site data from 14 enzymes and seven taxa in the cheilanthoid genus Pellaea provides new insights into the phylogenetic relationships of the P. glabella complex. It also assesses the congruence of results based on restriction site data at inter- and intraspecific levels in these sexually and apogamously reproducing ferns with those of earlier morphological and isozyme analyses of the same group. Wagner parsimony yielded a single most parsimonious tree of 187 steps and 11% homoplasy, based on a data matrix of 166 restriction sites of which 66 were phylogenetically informative. Phylogenetic analysis based on user-defined stepmatrix character-state weighting of site gains over losses produced an identical single most parsimonious tree. Dollo parsimony yielded two most parsimonious trees, one of which was topologically identical to the Wagner tree. Specific and infraspecific relationships in the P. glabella complex determined by the completely independent restriction site and isozyme data sets are identical. This lends confidence to the ongoing use of restriction site data in a broader study of Pellaea and other cheilanthoid taxa and to the present conclusions that P. atropurpurea is sister to the P. glabella complex, whereas P. breweri, previously considered the closest relative of this complex, is actually more distantly related to it than are the other taxa in this study.     

Molecular phytogeny of conifers using RFLP analysis of PCR-amplified specific chloroplast genes

We investigated the molecular phylogeny of conifers using restriction endonuclease fragment length polymorphism of six polymerase chain reaction-amplified chloroplast genes — frxC, rbcL, psbA, psbD, trnK, and 16S. We detected 227 total site changes among species, representing 23, 26, 38, 48, 67, and 25 site changes in frxC, psbA, psbD, rbcL, trnK and 16S, respectively. The mean nucleotide substitution was 10.75% (SD 0.573) among species in five families. Forty maximally parsimonious trees were obtained using the Wagner parsimony method, and a 50% majority-rule consensus tree was obtained from them. Data analysis produced similar basic patterns when both the Wagner parsimony and the neighbor-joining methods were applied, and the main lineages were clearly separated. Taxaceae and Cephalotaxaceae species were used as the out-groups when applying Wagner parsimony methods. With the Wagner method, the consistency index was 0.510, the retention index was 0.879, and tree length was 435 steps. Our results indicated that Cupressaceae and Taxodiaceae are closely related families and that Sciadopitys verticillata is the basal lineage of Cupressaceae and Taxodiaceae. The neighbor-joining tree is similar to the 50% majority-rule consensus of the 40 Wagner parsimony trees except for the position of Keteleeria daversifolia, the Picea and Cedrus group, and the divergence within Cupressaceae.     

Chloroplast DNA variation and phylogeny of theRanunculaceae

Restriction site mapping of chloroplast DNA from 31 species representing 26 genera of theRanunculaceae was performed using eleven restriction endonucleases. The chloroplast genome varies in length from approximately 152 to 160 kb. Length variants are frequent in theRanunculaceae and range from usually less than 300 bp to rarely 1.5 kb. The inverted repeat is extended into the large single copy (LSC) region by 4–4.5 kb inAnemone, Clematis, Clematopsis, Hepatica, Knowltonia, andPulsatilla. Several inversions are present in the LSC-region of the cpDNA in all these genera and inAdonis. The frequency of restriction site mutations varies within the chloroplast genome in theRanunculaceae between 4 and 32 mutations per kilobase, and is lowest in the inverted repeat and the regions containing the ATPase-genes and the genespsaA, psaB, psbA, rpoB, andrbcL. A total of 547 phylogenetically informative restriction sites was utilized in cladistic analyses of the family using Wagner, Dollo, and weighted parsimony. These three parsimony analyses result in different tree topologies. Four, six, and one equally most parsimonious trees were obtained with Wagner, Dollo, and weighted parsimony, respectively. The amount of support for the monophyletic groups was evaluated using bootstrapping and decay analysis. All three parsimony methods suggest thatHydrastis is the sister group to the remainder of theRanunculaceae, and that theAnemone-Clematis group, which shares several derived cpDNA rearrangements, is monophyletic. Only a few of the traditional groups in theRanunculaceae are supported by cpDNA restriction side data. Only Dollo parsimony provides support for the hypothesis thatThalictroideae andRanunculoideae are monophyletic.     

Restriction enzyme analysis of the nuclear 45s ribosomal DNA of six cultivated Alliums (Alliaceae)

Estimates of the phylogenetic relationships among cultivated and wildAllium species would benefit from identification of objective molecular characters. Restriction fragment length polymorphisms in the nuclear 45s ribosomal DNA (rDNA) were identified among two of five accessions of each of six cultivated Alliums. Restriction enzyme sites forBamHI,DraI,EcoRI,EcoRV,SacI, andXbaI were mapped. Different lengths of the rDNA repeat unit among the cultivated Alliums were due to sizes of the intergenic spacer. Nineteen polymorphic restriction enzyme sites were discovered and used to estimate phylogenetic relationships. Cladistic analyses based on Wagner parsimony were completed without an outgroup and resulted in two equally most parsimonious trees of 22 steps. A combined analysis of differences at RE sites in the ribosomal (19 characters) and chloroplast (15 characters) DNA generated a single most parsimonious tree of 39 steps. Single trichotomies were observed at 40 and 41 steps. Strict consensus of the three trees of 41 or fewer steps consisted of a lineage forA. tuberosum, a second forA. ampeloprasum andA. sativum, and a third forA. cepa, A. fistulosum, andA. schoenoprasum. Estimates of phylogenetic relationships based on variability at restriction enzyme sites in the rDNA and chloroplast DNA agree with the classification scheme ofTraub. Because of the predominance of autapomorphies, restriction enzyme analysis of the nuclear 45s rDNA is of limited use in estimating phylogenies amongAllium sections. However it is useful in the establishment of interspecific hybridity.     

HOMOPLOID RETICULATE EVOLUTION IN HELIANTHUS (ASTERACEAE): EVIDENCE FROM RIBOSOMAL GENES

Phylogenetic relationships among the 21 taxa comprising Helianthus sect. Helianthus and three outgroup species were assessed by restriction site mapping of the 18S-25S nuclear ribosomal RNA gene family. Wagner parsimony analysis of the 41 restriction site or length mutations observed produced a single 59-step most parsimonious tree. This tree was then compared to a cytoplasmic-based plastid phylogeny for this group. Several major discrepancies were observed between the two phylogenies suggesting both recent and ancient introgression. Furthermore, three cases of diploid hybrid speciation are unambiguously documented and a fourth case is suggested. These data are interpreted to suggest that evolution in Heliathus is reticulate rather than exclusively dichotomous and branching.     

Chloroplast DNA restriction site study of Verbesina (Asteraceae: Heliantheae)

Chloroplast DNA variation was examined for 79 species of Verbesina and 24 outgroups. Two independent analyses of the data were performed. An intergeneric study to elucidate the phylogenetic relationships of Verbesina used 22 genera, most of which are regarded in the literature as closely related to Verbesina. Coreopsis and Hymenoxys served as outgroups for this analysis. The 16 6-bp (base pair) restriction endonucleases used in the intergeneric study revealed 263 phylogenetically informative sites. Wagner analyses of these characters resulted in four equally parsimonious trees with a length of 857 steps and a Consistency Index of 0.492. Results from this study indicate that Verbesina is monophyletic, a member of the tribe Heliantheae, and that its sister taxa are the mostly Mexican genera Podachaenium, Squamopappus, and Tetrachyron. The infrageneric study of Verbesina included species belonging to all of its infrageneric taxa, except the monotypic sect. Stenocarpha. The 17 6-bp restriction endonucleases used in the infrageneric study revealed 137 sites 77 of which were phylogenetically informative. Wagner analyses of these characters generated 180 equally parsimonious trees with a length of 158 steps and a Consistency Index of 0.786. The genera Podachaenium, Squamopappus, and Tetrachyron served as outgroups. Two major clades, which correspond to traditional divisions of the genus based on leaf arrangement, are supported by the study. No support was found for the monophyly of seven of the 11 sections examined; further sampling of sections Lipactinia, Ochractinia, Verbesina, and Ximenesia is needed. Results support a major reassessment of the traditional infrageneric classification of the genus and provide the basis for the reevaluation of the sectional taxonomy of Verbesina. These studies support a North American origin for the genus with several independent introductions into South America producing significant diversity, especially in the Andean region.     

Phylogenetic relationships ofCrepidiastrum (Asteraceae) of the Bonin (Ogasawara) Islands based on chloroplast DNA restriction site variation

RFLPs of cpDNA were investigated for seven species ofCrepidiastrum, of which three are endemic to the Bonin Islands. As an outgroup for rooting the tree,Paraixeris denticulata was also examined. Approximately 350 restriction sites were surveyed using 16 restriction enzymes. A total of 26 restriction site mutations were detected, and seven of them were phylogenetically informative. Wagner parsimony analysis resulted in four most parsimonious trees. In the tree obtained, the Bonin endemics are monophyletic with four mutations and a bootstrap value of 0.98 for the branch. This result agrees with that obtained from a previous electrophoretic analysis (Ito and Ono 1990), and supports the hypothesis that the three Bonin endemics have been evolved from a common ancestor. The Bonin endemics cluster together withC. lanceolatum, suggesting that the endemics andC. lanceolatum share a common ancestor, although the bootstrap probability is not very high (0.78) and thus other possibilities cannot be rejected.     

Phylogenetic analysis of pines using ribosomal DNA restriction fragment length polymorphisms

Phylogenetic relationships among 30 species of the genusPinus were studied using restriction site polymorphism in the large subunit of nuclear rDNA. Of the 58 restriction sites scored, 48 were phylogenetically informative, and the 30 species reduced to ten taxa when species with identical restriction site patterns were combined. These ten taxa corresponded to the currently recognized subsections of the genus, with the sole exception ofP. leiophylla, which was identical in its pattern of restriction sites to all three species included from subsect.Oocarpae despite its being in a different section of subg.Pinus (Pinea instead ofPinus). A measure of the proportion of phylogenetic information contained within the data set (Homoplasy Excess Ratio, or HER) revealed that the character states were significantly non-randomly distributed among the ten taxa (HER = 0.71, p < 0.01). Branchand-bound searches using either Wagner or Dollo parsimony as the optimization criterion were carried out using PAUP in order to estimate phylogenetic relationships among the ten taxa. Three taxa (Picea pungens, Tsuga canadensis, andLarix decidua) were used independently as outgroups for purposes of rooting the trees. Despite the extreme differences in the assumptions underlying the Wagner and Dollo parsimony, the two gave surprisingly similar estimates of phylogeny, with both analyses supporting the monophyly of the two major subgeneraPinus andStrobus and differing in topology only in the placement of subsect.Ponderosae within subg.Pinus. The likelihood for the Wagner tree was only slightly higher than that computed for the Dollo tree.     

Chloroplast DNA restriction site variation and phylogeny of the Berberidaceae

Comparative restriction site mapping of the chloroplast genome was performed to examine phylogenetic relationships among 27 species representing 16 genera of the Berberidaceae and two outgroups. Chloroplast genomes of the species included in this study showed no major structural rearrangements (i.e., they are collinear to tobacco cpDNA) except for the extension of the inverted repeat in species of Berberis and Mahonia. Excluding several regions that exhibited severe length variation, a total of 501 phylogenetically informative sites was mapped for ten restriction enzymes. The strict consensus tree of 14 equally parsimonious trees indicated that some berberidaceous genera (Berberis, Mahonia, Diphylleia) are not monophyletic. To explore phylogenetic utility of different parsimony methods phylogenetic trees were generated using Wagner, Dollo, and weighted parsimony for a reduced data set that included 18 species. One of the most significant results was the recognition of the four chromosomal groups, which were strongly supported regardless of the parsimony method used. The most notable difference among the trees produced by the three parsimony methods was the relationships among the four chromosomal groups. The cpDNA trees also strongly supported a close relationship of several generic pairs (e.g., Berberis-Mahonia, Epimedium-Vancouveria, etc.). Maximum likelihood values were computed for the four different tree topologies of the chromosomal groups, two Wagner, one Dollo, and one weighted topology. The results indicate that the weighted tree has the highest likelihood value. The lowest likelihood value was obtained for the Dollo tree, which had the highest bootstrap and decay values. Separate analyses using only the Inverted Repeat (IR) region resulted in a tree that is identical to the weighted tree. Poor resolution and/or support for the relationships among the four chromosomal lineages of the Berberidaceae indicate that they may have radiated from an ancestral stock in a relatively short evolutionary time.     

Phylogenetic relationships of JapaneseAster (Asteraceae, Astereae) sensu lato based on chloroplast-DNA restriction site mutations

RFLPs of cpDNA were examined for 18 species ofAster, six species ofKalimeris, two species ofMiyamayomena and one species and one variety ofHeteropappus from Japan, using 16 restriction endonucleases. Approximately 275 restriction sites were surveyed, and a total of 74 restriction site mutations was detected, and 31 of these were phylogenetically informative. Sixteen most parsimonious trees constructed from Wagner parsimony analysis indicated the polyphyly ofKalimeris andMiyamayomena sensu Kitamura;K. miqueliana belongs to a different clade from the remaining species ofKalimeris, and two species ofMiyamayomena did not make a single clade. This result suggests that the shortening or loss of pappus have happened parallelly in different evolutionary lineages. We must be careful to assess the pappus character in taxonomy and phylogeny, and it is desirable to examine their phylogenetic relationships using a molecular data.     

A PHYLOGENETIC ANALYSIS OF DIOON (ZAMIACEAE)

A phylogenetic analysis of all the intrageneric taxa of the genus Dioon Lindley (Zamiaceae) was undertaken by using chloroplast DNA restriction fragment length polymorphism analysis. Wagner parsimony analysis on a 187 character matrix yielded two equally parsimonious trees, differing only for the position of D. caputoi. The consensus tree has two well-defined major clades. The first is composed of D. mejiae, D. rzedowskii, and D. spinulosum; the second is composed of D. califanoi, D. caputoi, D. edule var. angustifolium, D. edule var. edule, D. holmgrenii, D. merolae, D. purpusii, D. tomasellii var. sonorense, and D. tomasellii var. tomasellii. A phenetic analysis of the same data showed results broadly congruent with the cladistic analysis. This resulting phylogeny is partially congruent with morphological data and is also compatible with the biogeography of the genus. Modem species of Dioon may have evolved as a consequence of a very fast succession of vicariance events that mainly occurred during the early Cenozoic. The short time between each of these events may not have allowed the accumulation of a large number of morphological synapomorphies for the groups of species.     

A chloroplast DNA Phylogenetic study of the eastern Asia-eastern North America disjunct section Rytidospermum of Magnolia (Magnoliaceae)

Chloroplast DNA restriction site variation of ten endonucleases was examined among all ten species or varieties of the eastern Asia–eastern North America disjunct section Rytidospermum of Magnolia. Representatives from seven of the other ten sections of Magnolia and four related genera (Liriodendron, Manglietia, Michelia, and Talauma) were also included in the survey. A cladistic analysis of 200 variable sites using Wagner parsimony yielded 11 equally most parsimonious trees with a consistency index of 0.793 and a retention index of 0.870. The section Rytidospermum is polyphyletic in these trees. Magnolia tripetala from the southeastern U.S. is the only American species that has a sister relationship to the Asian taxa, M. hypoleuca, M. officinalis var. officinalis, M. officinalis var. biloba, and M. rostrata. Other American taxa in the section fall into two lineages, with M. macrophylla var. macrophylla, M. macrophylla var. ashei, and M. macrophylla var. dealbata in one, and M. fraseri var. fraseri and M. fraseri var. pyramidata in the other. They are not related to the Asian species as previously believed. The relationships revealed here are in agreement with morphological, allozymic, and cross compatibility data. These results demonstrate that a robust phylogenetic hypothesis is an important prerequisite for understanding biogeographic patterns.     

Chloroplast DNA study of the generaAmbrosia s.l. andHymenoclea (Asteraceae): Systematic implications

The relationship ofAmbrosia (ragweed) andFranseria has long been debated. Their treatment as separate genera has been repeatedly challenged. In this study, chloroplast DNA restriction site variation was examined for species from bothAmbrosia andFranseria as well as taxa from the closely related genusHymenoclea. The chloroplast genomes of members of these three genera were examined using 21 restriction endonucleases and the restriction mutations were used to construct phylogenetic trees. Wagner and Dollo parsimony as well as weighted parsimony were employed to compare the different phylogenies. The results support a close relationship betweenAmbrosia andFranseria, but indicate that the two groups are well separated. Compared toFranseria, Ambrosia is a much more strongly supported group, and the results indicate thatHymenoclea is closer toFranseria than toAmbrosia. The cpDNA phylogeny was used as a framework to examine evolutionary trends in morphology and secondary chemistry.     

Using Allele Frequencies and Geographic Subdivision to Reconstruct Gene Trees within a Species: Molecular Variance Parsimony

We formalize the use of allele frequency and geographic information for the construction of gene trees at the intraspecific level and extend the concept of evolutionary parsimony to molecular variance parsimony. The central principle is to consider a particular gene tree as a variable to be optimized in the estimation of a given population statistic. We propose three population statistics that are related to variance components and that are explicit functions of phylogenetic information. The methodology is applied in the context of minimum spanning trees (MSTs) and human mitochondrial DNA restriction data, but could be extended to accommodate other tree-making procedures, as well as other data types. We pursue optimal trees by heuristic optimization over a search space of more than 1.29 billion MSTs. This very large number of equally parsimonious trees underlines the lack of resolution of conventional parsimony procedures. This lack of resolution is highlighted by the observation that equally parsimonious trees yield very different estimates of population genetic diversity and genetic structure, as shown by null distributions of the population statistics, obtained by evaluation of 10,000 random MSTs. We propose a non-parametric test for the similarity between any two trees, based on the distribution of a weighted coevolutionary correlation. The ability to test for tree relatedness leads to the definition of a class of solutions instead of a single solution. Members of the class share virtually all of the critical internal structure of the tree but differ in the placement of singleton branch tips.     

Phylogenetic systematics of New WorldIpomoea (Convolvulaceae) based on chloroplast DNA restriction site variation

Chloroplast DNA restriction site variation was studied in 31 New World species ofIpomoea, representing a majority of the New World sections and series within the genus. Using 14 endonucleases, a total of 124 phylogenetically informative restriction sites was detected. Dollo parsimony, Wagner parsimony, and bootstrap methods were employed to construct phylogenetic trees and evaluate confidence intervals of monophyletic groups. With a few exceptions, groups circumscribed on the basis of morphological variation are in agreement with groupings based on restriction site variation. Relationships between subgeneric groupings, however, disagree substantially with those proposed in the past. Although conflicting hypotheses for some intersectional relationships are not presently resolvable, cpDNA restriction site analyses propose the following refinements of existing classification schemes.Ipomoea ser.Setosae is divided into distantly related groups, as is sect.Pharbitis. SeriesTyrianthinae, a proposed segregate of sect.Pharbitis, is associated with sect.Calonyction and the Tricolor complex (subg.Quamoclit).Ipomoea sect.Batatas is segregated from other herbaceous groups of the heterogeneous subg.Quamoclit sensu lato and aligned as a derivative ofI. setosa, subg.Eriospermum. To test for homology of key characters weighted in traditional schemes, morphological features were studied with respect to their distribution on lineages defined by restriction site data. Characters such as setose sepals, foliose-pubescent sepals, and erect growth habit, among others, are interpreted as having multiple origins, while 3-locular ovaries, 4-locular ovaries, and long-haired seeds have evolved only once.     

An investigation into the application of the Wagner parsimony method in synecology

Synecological analyses are usually based on typological, phenetic and cladistic methods. The disadvantages of these techniques are shown. The application of the Wagner parsimony method to synecology is considered. All the methods need some prerequisites, viz. definitions of localities and characters (the most simple one being the presence/absence of taxa); the choice of taxonomic level of taxa; their autochthony. The application of Wagner parsimony needs a new terminology. The congruence of any environmental condition, including freshwater monitoring indices, can be tested on parsimonious trees. The Wagner parsimony method not only provides various indices (tree length, CI, HI, RC, RI) which allow the comparison of trees but also minimal trees which are direct tools in synecology.     

THE PHYLOGENY OFLECANACTIS(OPEGRAPHACEAE)

The genusLecanactis, with 24 species, has been phylogenetically analysed using cladistic parsimony methods and support tests. Morphological, anatomical and chemical data were used, comprising 38 characters. Twelve equally most parsimonious trees were obtained. The successive approximations character weighting method gave one most parsimonious tree. The ingroup,Lecanactis, is supported as monophyletic. Although parsimony jackknifing and Bremer support indicate that the trees are poorly supported, some groups are wholly or partly distinguished in both the strict consensus tree, the successive weighting tree and the Jac tree.