A supermatrix approach provides a comprehensive genus-level phylogeny for Gentianales

Gentianales consist of Apocynaceae, Gelsemiaceae, Gentianaceae, Loganiaceae, and Rubiaceae, of which the majority are woody plants in tropical and subtropical areas. Despite extensive efforts in reconstructing the phylogeny of Gentianales based on molecular data, some interfamily and intrafamily relationships remain uncertain. We reconstructed the genus-level phylogeny of Gentianales based on the supermatrix of eight plastid markers (rbcL, matK, atpB, ndhF, rpl16, rps16, thetrnL-trnF region, and atpB-rbcL spacer) and one mitochondrial gene (matR) using maximum likelihood. The major clades and their relationships retrieved in the present study concur with those of previous studies. All of the five families of Gentianales are monophyletic with strong support. We resolved Rubiaceae as sister to the remaining families in Gentianales and showed support for the sister relationship between Loganiaceae and Apocynaceae. Our results provide new insights into relationships among intrafamilial clades. For example, within Rubiaceae we found that Craterispermeae were sister to Morindeae + (Palicoureeae + Psychotrieae) and that Theligoneae were sister to Putorieae. Within Gentianaceae, our phylogeny revealed that Gentianeae were sister to Helieae and Potalieae, and subtribe Lisianthiinae were sister to Potaliinae and Faroinae. Within Loganiaceae, we found Neuburgia as sister to Spigelieae. Within Apocynaceae, our results supported Amsonieae as sister to Melodineae, and Hunterieae as sister to a clade comprising Plumerieae + (Carisseae + APSA). We also confirmed the monophyly of Perplocoideae and the relationships among Baisseeae + (Secamonoideae + Asclepiadoideae).     

CLADISTICS AND FAMILY LEVEL CLASSIFICATION OF THE GENTIANALES

Abstract— The most recent classification of the angiosperm order Gentianales (Thorne, 1992) includes four principal families: Apocynaceae, Gentianaceae, Loganiaceae, and Rubiaceae. Ever since Bentham (1857) the status of Loganiaceae has been questioned, and several segregates of that family have been proposed both before and after his treatment. In this study we present cladistic results that show Loganiaceae, sensu lato, to be a paraphyletic group definable only by plesiomorphies, with members showing closest relationships to other families of the order. As the impact of different character-state representations of polymorphic terminals remains largely untested, our morphological and phytochemical data were analysed both with restricted polymorphism coding as well as with the monomorphic "subtaxon" recoding method of Nixon and Davis (1991). Both approaches yield highly compatible results, and we here discuss a new classification of the Gentianales based on (i) monophyletic groups identified by outgroup analysis, and (ii) the maximal portrayal of evidence provided by subtaxon polymorphism recoding. Most prominently, the Loganiaceae sensu lato are divided into four segregate families, two previously named (Loganiaceae sensu stricto and Strychnaceae), and two defined as a result of this study (Gelsemiaceae, L. Struwe & V. A. Albert, stat. nov. and Geniostomaceae, L. Struwe & V. A. Albert, fam. nov.). Apocynaceae (incl. Asclepiadaceae), Gentianaceae (incl. Loganiaceae—Potalieae), and Rubiaceae remain as monophyletic families. Outgroup analysis supports both the monophyly of the Gentianales as well as the exclusion from the order of Buddleja, Desfontainia, Plocosperma, Polypremum , and Retzia (all Loganiaceae sensu Leeuwenberg and Leenhouts).     

More characters or more taxa for a robust phylogeny--case study from the coffee family (Rubiaceae)

Using different data sets mainly from the plant family Rubiaceae, but in parts also from the Apocynaceae, Asteraceae, Lardizabalaceae, Saxifragaceae, and Solanaceae, we have investigated the effect of number of characters, number of taxa, and kind of data on bootstrap values within phylogenetic trees. The percentage of supported nodes within a tree is positively correlated with the number of characters, and negatively correlated with the number of taxa. The morphological analyses are based on few characters and weakly supported trees are expected. The percentage of supported nodes is also dependent on the kind of data analyzed. In analyses of Rubiaceae based on the same number of characters, RFLP data give trees with higher percentage of supported nodes than rbcL and morphological data. We also discuss the support values for particular nodes at the familial and subfamilial levels. Two new data sets of ndhF and rbcL sequences of Rubiaceae are analyzed and together with earlier studies of the family we can conclude that the monophyly of the Rubiaceae is supported and within the family there are three well supported, but not easily characterized, large subfamilies, Rubioideae, Cinchonoideae s.s. and Ixoroideae s.l. There are also a few genera (Luculia and Coptosapelta) unclassified to subfamily.     

PHYLOGENY OF THE RUBIACEAE AND THE LOGANIACEAE: CONGRUENCE OR CONFLICT BETWEEN MORPHOLOGICAL AND MOLECULAR DATA?

Phylogenetic analyses of 33 genera of Rubiaceae were performed using morphological and a few chemical characters. Parsimony analysis based on 29 characters resulted in eight equally parsimonious trees, with a consistency index of 0.40 and a retention index of 0.69. These results were compared to a phylogenetic analysis of the same genera based on chloroplast DNA restriction site data. There are discrepancies between the two analyses, but if we consider groupings reflected in the present classification there is much congruency. With the exception of four genera, all the genera are positioned in the same group of taxa in the two analyses. Clades of taxa representing three of the four subfamilies (~the Antirheoideae, ~the Rubioideae, and the ~Ixoroideae) are monophyletic, while the fourth subfamily Cinchonoideae is shown to be paraphyletic. Both analyses support a widened tribe Chiococceae, including the former subtribe Portlandiinae (Condamineeae). Furthermore, in both analyses the tribe Hamelieae is placed outside the subfamily Rubioideae where it is now housed. In search for the most plausible sister group to the Rubiaceae, the genus Cinchona (Rubiaceae) was analyzed together with 13 genera of the Loganiaceae, Nerium (Apocynaceae), and Exacum (Gentianaceae). Cornus (Comaceae), Olea (Oleaceae), and these two genera together were used as outgroups. The analysis, including 25 characters, 16 taxa, and with Cornus and Olea together as an outgroup, resulted in four equally parsimonious trees, with a consistency index of 0.53 and a retention index of 0.62. The non-Loganiaceae taxa Cinchona (Rubiaceae), Nerium (Apocynaceae), and Exacum (Gentianaceae) were all found to have their closest relatives within the Loganiaceae indicating that the Loganiaceae are paraphyletic and ought to be reclassified. As a result of the morphological data the most plausible sister group to the Rubiaceae is the tribe Gelsemieae of the Loganiaceae.     

The phylogeny of the Asteridae sensu lato based on chloroplast ndhF gene sequences

A phylogenetic study of Asteridae sensu lato was conducted based on chloroplast ndhF gene sequences for 116 ingroup and 13 outgroup species. Prior molecular studies based on rbcL sequences identified terminal groups corresponding to families, but were unable to resolve relationships among them. These results are largely consistent with earlier rbcL studies, but provide much greater resolution and stronger bootstrap support throughout the tree. The parsimony analysis found eight equally parsimonious trees, all of which recognize four major clades with the following relationship: (Cornales (Ericales (Euasterids I, Euasterids II))). Euasterids I includes (Garryales ((Solanales, Boraginaceae) (Gentianales, Lamiales))), although with weak support for relationships among the named clades. Euasterids II includes (Aquifoliales (Asterales (Apiales, Dipsacales))) with strong support for these relationships. Relationships within Ericales are weakly supported and merit further attention.     

Multiple origin of the tropical forest tree family Icacinaceae

Analyses of DNA sequences from four genes (ndhF, rbcL, atpB, and 18S rDNA) and morphological data show that the members of the tropical forest tree family Icacinaceae do not have a common origin. All of the genera earlier placed in Icacinaceae are euasterids but placed in the following three different orders: Garryales, Aquifoliales, and Apiales. Icacina and related genera are members of Garryales and, pending more data, are still best treated as Icacinaceae (sensu stricto). The genera related to Aquifoliales are placed in Cardiopteridaceae and a new family, Stemonuraceae. The genus Pennantia is a member of Apiales and the family Pennantiaceae is recognized. Morphological characters delimiting these groups are discussed. Twenty-six new ndhF sequences were obtained for the study (25 from former Icacinaceae and 1 from Cardiopteris).     

Morphological and ultrastructural diversity of orbicules in Gentianaceae

Minute granules of sporopollenin, called orbicules, can be observed on the innermost tangential and/or radial walls of secretory tapetum cells. Orbicules were investigated in 53 species of 34 Gentianaceae genera using light microscopy, scanning electron microscopy and transmission electron microscopy. This selection covered all different tribes and subtribes recognized in Gentianaceae (87 genera, +/-1650 species). Orbicules were found in 38 species (23 genera) distributed among the six tribes recognized in Gentianaceae. The orbicule typology is based on those described previously in Rubiaceae. Of the six orbicule types described previously, Type II orbicules are lacking. Type III orbicules are most common (17 species). Hockinia Gardner is the only representative with Type I orbicules. The number of representatives with orbicules belonging to the other orbicule types are equally distributed among the species studied: seven species possess Type IV orbicules, six species Type V and six species Type VI. The systematic usefulness of this typology is discussed in comparison with the latest systematic insights within the family, and palynological trends in Gentianaceae. Orbicule data have proven to be useful for evaluating tribal delimitation within Rubiaceae and Loganiaceae s.l.; however, they seem not to be useful for tribal delimitation in Gentianaceae. In the tribes Potalieae and Gentianeae orbicule data may be useful at subtribal level.     

Discovery of paralogous nuclear gene sequences coding for the second-largest subunit of RNA polymerase II (RPB2) and their phylogenetic utility in gentianales of the asterids

Paralogous sequences of the RPB2 gene are demonstrated in the angiosperm order Gentianales. Two different copies were found by using different PCR primer pairs targeting a region that corresponds to exons 22-24 in the Arabidopsis RPB2 gene. One of the copies (RPB2-d) lacks introns in this region, whereas the other has introns at locations corresponding to those of green plants previously investigated. When analyzed with other available RPB2 sequences from this region, all 28 RPB2-d sequences obtained from the Gentianales and the four sequences from the Lamiales form a monophyletic group, together with a previously published tomato cDNA sequence. The substitution patterns, relative rates of change, and nucleotide compositions of the two paralogous RPB2 exon regions are similar, and none of them shows any signs of being a pseudogene. Although multiple copies of similar, paralogous sequences can confound phylogenetic interpretations, the lack of introns in RPB2-d make a priori homology assessment easy. The phylogenetic utility of RPB2-d within the Gentianales is evaluated in comparison with the chloroplast genes ndhF and rbcL. The hierarchical information in the RPB2-d region sequenced is more incongruent with that of the plastid genes than the plastid genes are with each other as determined by incongruence length difference tests. In contrast to the plastid genes, parsimony-informative third codon positions of RPB2 have a significantly higher rate of change than first and second positions. Topologically, the trees from the three genes are similar, and the differences are usually only weakly supported. In terms of support, RPB2 gives the highest jackknife support per sequenced nucleotide, whereas ndhF gives the highest Bremer support per sequenced nucleotide. The RPB2-d locus has the potential to be a valuable nuclear marker for determination of phylogenetic relationships within the euasterid I group of plants.     

Phylogenetic structure in the grass family (Poaceae): evidence from the nuclear gene phytochrome B

Phylogenetic analyses of partial phytochrome B (PHYB) nuclear DNA sequences provide unambiguous resolution of evolutionary relationships within Poaceae. Analysis of PHYB nucleotides from 51 taxa representing seven traditionally recognized subfamilies clearly distinguishes three early-diverging herbaceous "bambusoid" lineages. First and most basal are Anomochloa and Streptochaeta, second is Pharus, and third is Puelia. The remaining grasses occur in two principal, highly supported clades. The first comprises bambusoid, oryzoid, and pooid genera (the BOP clade); the second comprises panicoid, arundinoid, chloridoid, and centothecoid genera (the PACC clade). The PHYB phylogeny is the first nuclear gene tree to address comprehensively phylogenetic relationships among grasses. It corroborates several inferences made from chloroplast gene trees, including the PACC clade, and the basal position of the herbaceous bamboos Anomochloa, Streptochaeta, and Pharus. However, the clear resolution of the sister group relationship among bambusoids, oryzoids, and pooids in the PHYB tree is novel; the relationship is only weakly supported in ndhF trees and is nonexistent in rbcL and plastid restriction site trees. Nuclear PHYB data support Anomochlooideae, Pharoideae, Pooideae sensu lato, Oryzoideae, Panicoideae, and Chloridoideae, and concur in the polyphyly of both Arundinoideae and Bambusoideae.     

Disintegration of the scrophulariaceae

A molecular systematic study of Scrophulariaceae sensu lato using DNA sequences of three plastid genes (rbcL, ndhF, and rps2) revealed at least five distinct monophyletic groups. Thirty-nine genera representing 24 tribes of the Scrophulariaceae s.l. (sensu lato) were analyzed along with representatives of 15 other families of Lamiales. The Scrophulariaceae s.s. (sensu stricto) include part or all of tribes Aptosimeae, Hemimerideae, Leucophylleae, Manuleae, Selagineae, and Verbasceae (= Scrophularieae) and the conventional families Buddlejaceae and Myoporaceae. Veronicaceae includes all or part of tribes Angelonieae, Antirrhineae, Cheloneae, Digitaleae, and Gratioleae and the conventional families Callitrichaceae, Globulariaceae, Hippuridaceae, and Plantaginaceae. The Orobanchaceae include tribes Buchnereae, Rhinantheae, and the conventional Orobanchaceae. All sampled members of Orobanchaceae are parasitic, except Lindenbergia, which is sister to the rest of the family. Family Calceolariaceae Olmstead is newly erected herein to recognize the phylogenetic distinctiveness of tribe Calceolarieae. The Calceolariaceae are close to the base of the Lamiales. The Stilbaceae are expanded by the inclusion of Halleria. Mimulus does not belong in any of these five groups.     

Placing the woody tropical genera of Polygonaceae: A hypothesis of character evolution and phylogeny

? Premise of the study: Taxonomic groups have often been recognized on the basis of geographic distinctions rather than accurately representing evolutionary relationships. This has been particularly true for temperate and tropical members from the same family. Polygonaceae exemplifies this problem, wherein the woody tropical genera were segregated from temperate members of the family and placed in the subfamily Polygonoideae as two tribes: Triplarideae and Coccolobeae. Modern phylogenetic studies, especially when inferred from many lines of evidence, can elucidate more probable hypotheses of relationships. This study builds on previous work in the family and aims to test the traditional classification of the tropical woody taxa, which have been understudied and undersampled compared to their temperate relatives. ? Methods: A phylogenetic study was undertaken with expanded sampling of the tropical genera with data from five plastid markers (psbA-trnH, psaI-accD, matK, ndhF, and rbcL), nuclear ribosomal DNA (ITS) and morphology. ? Key results: Results support the placement of nine of 12 genera of the Triplarideae and Coccolobeae within Eriogonoideae, in which these genera form a paraphyletic assemblage giving rise to Eriogoneae. The remaining woody tropical genera excluded from Eriogonoideae occur in the paleotropics. ? Conclusions: Traditional characters used to delimit Coccolobeae and Triplarideae are not useful for defining monophyletic groups. The six-tepal condition is derived from the five-tepal condition, and unisexual flowers have arisen multiple times in different sexual systems. Ruminate endosperm has arisen multiple times in the family, suggesting this character is highly plastic.     

Phylogenetic relationships among Acanthaceae: evidence from noncoding trnL-trnF chloroplast DNA sequences

We used sequence data from the intron and spacer of the trnL-trnF chloroplast region to study phylogenetic relationships among Acanthaceae. This region is more variable than other chloroplast loci that have been sequenced for members of Acanthaceae (rbcL and ndhF), is more prone to length mutations, and is less homoplasious than these genes. Our results indicate that this region is likely to be useful in addressing phylogenetic questions among but not within genera in these and related plants. In terms of phylogenetic relationships, Elytraria (representing Nelsonioideae) is more distantly related to Acanthaceae sensu stricto (s.s.) than Thunbergia and Mendoncia. These last two genera are strongly supported as sister taxa. Molecular evidence does not support monophyly of Acanthaceae s.s., although there is strong morphological evidence for this relationship. There is strong support for monophyly of four major lineages within Acanthaceae s.s.: the Acanthus, Barleria, Ruellia, and Justicia lineages as here defined. The last three of these comprise a strongly supported monophyletic group, and there is weaker evidence linking the Ruellia and Justicia lineages as closest relatives. Within the Acanthus lineage, our results confirm the existence of monophyletic lineages representing Aphelandreae and Acantheae. Lastly, within the Justicia lineage, we develop initial hypotheses regarding the definition of sublineages; some of these correspond to earlier ideas, whereas others do not. All of these hypotheses need to be tested against more data.     

A Revised Time Tree of the Asterids: Establishing a Temporal Framework For Evolutionary Studies of the Coffee Family (Rubiaceae)

Divergence time analyses in the coffee family (Rubiaceae) have all relied on the same Gentianales crown group age estimate, reported by an earlier analysis of the asterids, for defining the upper age bound of the root node in their analyses. However, not only did the asterid analysis suffer from several analytical shortcomings, but the estimate itself has been used in highly inconsistent ways in these Rubiaceae analyses. Based on the original data, we here reanalyze the divergence times of the asterids using relaxed-clock models and 14 fossil-based minimum age constraints. We also expand the data set to include an additional 67 taxa from Rubiaceae sampled across all three subfamilies recognized in the family. Three analyses are conducted: a separate analysis of the asterids, which completely mirrors the original asterid analysis in terms of taxon sample and data; a separate analysis of the Gentianales, where the result from the first analysis is used for defining a secondary root calibration point; and a combined analysis where all taxa are analyzed simultaneously. Results are presented in the form of a time-calibrated phylogeny, and age estimates for asterid groups, Gentianales, and major groups of Rubiaceae are compared and discussed in relation to previously published estimates. Our updated age estimates for major groups of Rubiaceae provide a significant step forward towards the long term goal of establishing a robust temporal framework for the divergence of this biologically diverse and fascinating group of plants.     

Molecular phylogenetic analysis of Passifloraceae sensu lato (Malpighiales) based on plastid and nuclear DNA sequences

A phylogenetic analysis of Passifloraceae sensu lato was performed using rbcL, atpB, matK, and 18S rDNA sequences from 25 genera and 42 species. Parsimony analyses of combined data sets resulted in a single most parsimonious tree, which was very similar to the 50% majority consensus tree from the Bayesian analysis. All nodes except three were supported by more than 50% bootstrap. The monophyly of Passifloraceae s.l. as well as the former families, Malesherbiaceae, Passifloraceae sensu stricto, and Turneraceae were strongly supported. Passifloraceae s.s. and the Turneraceae are sisters, and form a strongly supported clade. Within Passifloraceae s.s., the tribes Passifloreae and Paropsieae are both monophyletic. The intergeneric relationships within Passifloraceae s.s. and Turneraceae are roughly correlated with previous classification systems. The morphological character of an androgynophore/gynophore is better used for characterizing genera grouping within Passifloraceae s.s. Other morphological characters such as the corona and aril are discussed.     

Phylogeny of the Celastraceae inferred from 26S nuclear ribosomal DNA, phytochrome B, rbcL, atpB, and morphology

Phylogenetic relationships within Celastraceae (spindle-tree family) were inferred from nucleotide sequence characters from the 5' end of 26S nuclear ribosomal DNA (including expansion segments D1-D3; 84 species sampled), phytochrome B (58 species), rbcL (31 species), atpB (23 species), and morphology (94 species). Among taxa of questionable affinity, Forsellesia is a member of Crossosomataceae, and Goupia is excluded from Celastraceae. However, Brexia, Canotia, Lepuropetalon, Parnassia, Siphonodon, and Stackhousiaceae are supported as members of Celastraceae. Gymnosporia and Tricerma are distinct from Maytenus, Cassine is supported as distinct from Elaeodendron, and Dicarpellum is distinct from Salacia. Catha, Maytenus, and Pristimera are not resolved as natural genera. Hippocrateaceae (including Plagiopteron and Lophopetalum) are a clade nested within a paraphyletic Celastraceae. These data also suggest that the Loesener's classification of Celastraceae sensu stricto and Hallé's classification of Hippocrateaceae are artificial. The diversification of the fruit and aril within Celastraceae appears to be complex, with multiple origins of most fruit and aril forms.     

Family-level relationships of Onagraceae based on chloroplast rbcL and ndhF data

Despite intensive morphological and molecular studies of Onagraceae, relationships within the family are not fully understood. One drawback of previous analyses is limited sampling within the large tribe Onagreae. In addition, the monophyly of two species-rich genera in Onagreae, Camissonia and Oenothera, has never been adequately tested. To understand relationships within Onagraceae, test the monophyly of these two genera, and ascertain the affinities of the newly discovered genus Megacorax, we conducted parsimony and maximum likelihood analyses with rbcL and ndhF sequence data for 24 taxa representing all 17 Onagraceae genera and two outgroup Lythraceae. Results strongly support a monophyletic Onagraceae, with Ludwigia as the basal lineage and a sister-taxon relationship between Megacorax and Lopezia. Gongylocarpus is supported as sister to Epilobieae plus the rest of Onagreae, although relationships within the latter clade have limited resolution. Thus, we advocate placement of Gongylocarpus in a monogeneric tribe, Gongylocarpeae. Most relationships within Onagreae are weakly resolved, suggesting a rapid diversification of this group in western North America. Neither Camissonia nor Oenothera appears to be monophyletic; however, increased taxon sampling is needed to clarify those relationships. Morphological characters generally agree with the molecular data, providing further support for relationships.