| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Systematics |
2Department of Biological Sciences, Northern Illinois University, DeKalb, Illinois 60115-2861 USA; and 3Department of Botany, Iowa State University, Ames, Iowa 50011-1020 USA
Received for publication January 4, 2001. Accepted for publication May 1, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Kranz • molecular phylogenetics • non-Kranz • Paniceae • Poaceae • rpoC2 insert
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Hattersley and Watson, 1992
). The more prevalent C3 photosynthesis is correlated with three characters: (1) irregular chlorenchyma; (2) starch formation confined to the mesophyll; and (3) four or more chlorenchyma cells between adjacent bundle sheaths. In C4 species, the starch-forming chlorenchyma is restricted to a bundle sheath, and there are fewer cells between adjacent bundle sheaths (two to four) than in C3 grasses. The C3 state is generally considered to be plesiomorphic. However, with regard to Paniceae, Clayton and Renvoize (1986
, p. 260) urge "some caution ... in regarding the non-Kranz genera as primitive in any absolute sense...."
There are also distinguishable C4 subtypes. Decarboxylation of four-carbon compounds in C4 species is catalyzed either by NADP-malic enzyme (NADP-ME), NAD-malic enzyme (NAD-ME), or PEP-carboxykinase (PCK) (see Hattersley and Watson [1992]
and Sinha and Kellogg [1996]
for physiological differences between subtypes). The most common condition in Paniceae is NADP-ME, although PCK species are not rare.
Photosynthetic physiology is nearly invariant within entire subfamilies of Poaceae with only three exceptions. Nine of the 12 subfamilies (following Grass Phylogeny Working Group [GPWG], 2000
, in press) are each wholly non-Kranz. All Kranz grasses are found in a large and diverse lineage known as the PACCAD clade (GPWG, 2000, in press). Aristidoideae are Kranz except for Sartidia (four species). All chloridoid grasses are Kranz except for two species (Eragrostis walteri and Merxmuellera rangei). Panicoideae as a whole, and Paniceae in particular, are uniquely variable, comprising many taxa which are C3 as well as taxa of all C4 subtypes. This variation affords a unique opportunity for phylogenetic study. Hypotheses of independent origin and putative ancestral states for each subtype, which are generally unknown, can be tested in a phylogenetic framework.
The Poaceae has a generally well-studied phylogeny (e.g., GPWG, 2000, in press), however, the most intensively sampled phylogenetic work in Paniceae published to date is limited to two studies. The first, a cladistic study of 67 exomorphological and anatomical characters taken largely from herbarium specimens of 110 taxa, was conducted by Zuloaga, Morrone, and Giussani (2000)
. The authors considered this substantial study to be "preliminary," undoubtedly because of the size of Paniceae, and support statistics for the cladogram were not reported. Of note was the fact that Panicum, as expected, was polyphyletic. For example, Panicum subgenus Megathyrsus clustered with a Urochloa/Eriochloa clade, whereas Panicum section Laxa was in a somewhat more basal clade with Arthropogon and other taxa with fusoid-like cells.
The second, a molecular study (sequences of the plastid trnL-F region) of 32 species plus outgroups, was conducted by Gómez-Martinez and Culham (2000)
. Some of their main conclusions and discoveries about Paniceae were the following: (a) Setariinae and Digitariinae were both found to be paraphyletic with other genera of Paniceae; (b) the identification of a "setae-bearing clade" (Cenchrus, Pennisetum, and Setaria); (c) a monophyletic clade of species that are exclusively C4 PCK, including Panicum maximum and Brachiaria; (d) a polyphyletic Panicum that supports Zuloaga, Morrone, and Giussani's conclusions (2000)
; and (e) moderately supported monophyly of each of Paspalum and Brachiaria. We present here a complementary molecular analysis of Paniceae utilizing a different plastid locus, the grass-specific insert found in rpoC2, which encodes subunit C2 of a plastid-encoded RNA polymerase.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Amplified products were purified and concentrated using Microcon-YM100 concentrators (Millipore, Bedford, Massachusetts, USA) following the manufacturer's instructions. Automated sequencing of these templates was performed (Core Laboratory Facility, Biological Sciences, Northern Illinois University, DeKalb, Illinois, USA) with the same primers that were used for amplification. Both strands were sequenced.
Molecular phylogenetic studies conducted broadly across Poaceae have suggested that Danthoniopsis is either the immediate sister group to the bulk of the other panicoid grasses or a member of the sister clade (Clark, Zhang, and Wendel, 1995
; Spangler et al., 1999
; Mathews, Tsai, and Kellogg, 2000
). The rpoC2 insert sequence of Danthoniopsis dinterii was thus also sequenced and was specified as the outgroup for phylogenetic analyses.
Nucleotide sequences were translated with Gene Inspector (version 1.5, Textco, West Lebanon, New Hampshire, USA) prior to alignment. A rule-governed alignment of the amino acid sequences was performed following Barker, Linder, and Harley (1999)
. These rules were devised to preserve heptameric amino acid repeats commonly found in the rpoC2 insert, which are most likely the result of slipped-strand mispairing mutations (Cummings and Kellogg, 1994
). We further followed a conservative approach to alignment, recognizing new heptameric insertions only if they could not otherwise be derived by one or two single nucleotide substitutions from existing adjacent sequences. Also aligned were rpoC2 insert sequences of Brachiaria serrata, Hyparrhenia hirta, Panicum maximum, Rhynchelytrum repens, and Saccharum officinarum previously determined by Barker, Linder, and Harley (1999)
; and one of Pennisetum sp. determined by Cummings, King, and Kellogg (1994)
. Representatives of three tribes, Paniceae (57 spp.), Andropogoneae (2 spp.), and Arundinelleae (1 outgroup sp.) were thus included. After the amino acid sequences had been aligned, a corresponding alignment of nucleotide sequences was generated.
PAUP* 4.0b4a (Swofford, 2000
) analyses were performed on the aligned nucleotide sequences. Gaps were scored as "missing." Gapped positions found in half or more of the species were excluded from the phylogenetic analysis, an approach that minimized the detrimental impact of excessive amounts of missing values on phylogenetic analysis while still preserving character data found for the majority of the taxa. Gaps were encoded as binary characters, appended to the aligned nucleotide sequences, and included in the phylogenetic analysis. Parsimony analysis of these data was a computer-intensive activity that consumed 400 h of computing time (plus >600 h of preliminary analyses) on a cluster of four desktop computers. To facilitate the analysis, tasks were subdivided into smaller parts that were run separately, each with specified limits on MAXTREES. Eighty-three such random addition sequence replicates, each with a different specified starting seed and tree-bisection-reconnection (TBR) swapping, were run with MAXTREES limited to 10 000. The results of 28 of these 83 subanalyses found sets of less parsimonious trees and were discarded. One additional replicate was also run with MAXTREES set to 500 000. The separate tree files were then combined with concurrent elimination of duplicate trees. A bootstrap analysis was run with 5000 subsamples, TBR swapping, and MAXTREES set at 1000.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
.
|
Exclusion of the ten gaps found in >50% of the taxa resulted in analysis of 450 nucleotide sites together with 34 binary-coded gaps, giving 484 total characters. Of these, 143 were parsimony informative, including 114 of the nucleotide sites and 29 of the encoded gaps. Once duplicate trees were eliminated from the trees combined from the 56 separate analyses, 1 x 106 equally parsimonious trees of length 496 remained (see consensus tree, Fig. 1). Retention index (RI) was 0.7156 and consistency index (CI; excluding uninformative characters) was 0.4571.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
and those of our companion paper in this issue, the ndhF study of Giussani et al. (2001)
. Some new relationships were also suggested. Of particular interest is the distribution of C3 and C4 taxa across our rpoC2 phylogeny (Fig. 1).
A clade of four taxa with non-Kranz species (Steinchisma [which includes anatomically intermediate C3-C4 species], Plagiantha, Otachyrium, and Panicum laxum) clustered with a fifth Kranz species, Leptocoryphium lanatum, and was strongly supported (bootstrap value = 94%). A similar association was noted by Giussani et al. (2001)
. The phylogenetic association between Plagiantha and Otachyrium has been previously noted (Zuloaga, Morrone, and Giussani, 2000
). Further, a unique morphological synapomorphy has been previously recognized in the three most derived genera in this clade, Steinchisma, Plagiantha, and Otachyrium. Species in these genera are characterized by a distinct modification of the palea of the lower floret to produce enlarged keels or flanks at maturity (Clayton and Renvoize, 1986
). These compelling morphological and molecular synapomorphies strongly support the monophyly of this "keeled clade."
The keeled clade is in a polytomy with two other groups toward the base of the tree, a strongly supported Arthropogon lanceolatus and Panicum prionitis (bootstrap value = 90%), and a weakly supported clade of Paspalum (three species) that is paraphyletic with Thrasya (two species). Clayton and Renvoize (1986
, p. 289) observed that the distinction between these latter two genera is "rather arbitrary," which is consistent with the topology of the tree of Fig. 1.
Non-Kranz species were also found in seven other clades. One of these, Oplismenus hirtellus, was resolved as the sister taxon to all other ingroup species, contra Giussani et al. (2001)
. Clayton and Renvoize (1986)
previously recognized the plesiomorphic nature of Oplismenus. If Oplismenus hirtellus is, in fact, the sister group to the Paniceae, then the non-Kranz C3 condition could be interpreted as plesiomorphic for the entire tribe. Each of the seven non-Kranz clades appears to have had Kranz ancestors at some point, suggesting multiple losses of the Kranz condition in Paniceae (though of the seven, only the Kranz ancestry of the keeled clade was strongly supported). Such a finding is consistent with the observation by Clayton and Renvoize (1986)
that Paniceae have tended to radiate into mesic habitats where selection pressures favoring C4 metabolism are minimized.
Paradoxically, the non-Kranz Panicum koolauense (Panicum subgen. Dichanthelium) was associated with the two representative Andropogoneae (a wholly Kranz taxon) in our analysis, and this clade was embedded in moderately derived Paniceae. Spangler et al. (1999)
, Gómez-Martinez and Culham (2000)
, and Giussani et al. (2001)
all found a strongly supported sister group relationship between Andropogoneae and the Kranz Arundinella, the latter of which was not included in our study. Giussani et al. (2001)
found a strongly supported relationship between P. koolauense and P. sabulorum, the latter of which species was not included here; however the position of this clade in their analysis was not strongly supported. Our results do suggest that Andropogoneae and Paniceae are paraphyletic; however, the weakly supported relationship between Panicum koolauense and Andropogoneae requires further study as it may be an artifact of the sampling in this study.
The five species of Digitaria were strongly supported as monophyletic (bootstrap value = 96%) though the relatively distant and well-supported position of Leptocoryphium indicated that Digitariinae, in which these two genera are traditionally classified, was polyphyletic. Clayton and Renvoize (1986)
suggested that Digitaria was phylogenetically basal in Paniceae, whereas the genus is considerably more derived in our rpoC2 phylogeny.
Thirteen species of Panicum were included in this analysis. Six of these, P. fauriei, P. mystasipum, P. nephelophilum, P. pedersenii, P. repens, and P. tenuifolium, all subg. Panicum, associated together into a very weakly supported clade (bootstrap <50%). The remaining seven species were variously allied with diverse elements of Panicoideae, though again with weak support. The association of P. maximum with Urochloa/Eriochloa in the PCK clade and P. laxum's position in the same general clade as Arthropogon lanceolatus are noteworthy; both positions were observed by Zuloaga, Morrone, and Giussani (2000)
and both relationships were recovered by Giussani et al. (2001)
. Either rpoC2 has insufficient phylogenetic information to resolve a monophyletic Panicum or any of the subgenera sensu Zuloaga (1986)
therein, or Panicum is extraordinarily polyphyletic. The correlated morphophysiological synapomorphies of P. maximum (see below) and P. laxum with other members of their respective clades suggests that Panicum has at least some degree of polyphyly (Gómez-Martinez and Culham, 2000
; Zuloaga, Morrone, and Giussani, 2000
; Giussani et al., 2001
).
The rpoC2 data offered weak support for two other groups identified by Gómez-Martinez and Culham (2000)
and Giussani et al. (2001)
. The "setae-bearing clade" (Cenchrus, Pennisetum, and Setaria together with Paspalidium, Stenotaphrum, and Brachiaria serrata in our analysis) was monophyletic in the majority rule consensus (Fig. 1). Significantly, Cenchrus and Pennisetum, which Clayton and Renvoize (1986
, p. 304) described as "not sharply separated" were paraphyletic with each other within this clade. Also resolved was the PCK clade (Brachiaria mutica, Chaetium bromoides, Eriochloa punctata, Melinis minutiflora, Panicum maximum, Rhynchelytrum repens, and two species of Urochloa). Note that Rhynchelytrum, which was sister taxon to Melinis, is sometimes synonymized with the latter (Dallwitz, Paine, and Zurcher, 1999
). The rpoC2 insert sequence of Brachiaria serrata, which clustered neither with its congener (B. mutica) nor with other members of the PCK clade, was obtained from a previously published study (Barker, Linder, and Harley, 1999
). The unexpected association of this species with Pennisetum in the setae-bearing clade needs further study.
A distinctive linear hilum has been observed in two of the genera included here, Tatianyx and Mesosetum (Sendulsky, Filgueiras, and Burman, 1986
). Species of these two genera were clustered together in our analysis albeit with only weak support, suggesting that this character might be synapomorphous for these taxa. The moderately supported association of Altoparadisium chapadense to this clade (bootstrap value = 78%) is, to our knowledge, a new indication of relationship and was also recovered in the ndhF study of Giussani et al. (2001).
In spite of the somewhat limited phylogenetic information in the rpoC2 insert sequences, correlations were nonetheless found between clades identified from the phylogenetic analysis of this region and morphology (e.g., the "linear hilum," "keeled," and "setae-bearing" clades), physiology (the PCK clade) as well as with other plastid phylogenies (Gómez-Martínez and Culham, 2000
; Giussani et al., 2001
). Increased sampling of molecular characters for this large tribe may rearrange or increase support for basal nodes. As previously reported, we found that the gaps in the rpoC2 insert sequences are a substantial component of the phylogenetic information that the region has to offer. Our molecular phylogenetic results offer further impetus for a taxonomic reorganization of Paniceae.
|
FOOTNOTES |
|---|
4 Author for correspondence and reprint requests (mel-duvall{at}niu.edu
).
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Brown W. 1977 The Kranz syndrome and its subtypes in grass systematics. Memoirs of the Torrey Botanical Club 23: 1-97
Clark L. W. Zhang J. Wendel 1995 A phylogeny of the grass family (Poaceae) based on ndhF sequence data. Systematic Botany 20: 436-460
Clayton W. S. Renvoize 1986 Genera Graminum, grasses of the world. Kew Bulletin Additional Series 13: 1-389
Cummings M. L. King E. Kellogg 1994 Slipped-strand mispairing in a plastid gene: rpoC2 in grasses (Poaceae). Molecular Biology and Evolution 11: 1-8[Abstract]
Dallwitz M. J. T. A. Paine E. J. Zurcher 1999 User's guide to the DELTA editor. http://biodiversity.uno.edu/delta/. Accessed August 2000
Giussani L. M. J. H. Cota-Sánchez F. O. Zuloaga E. A. Kellogg 2001 A molecular phylogeny of the grass subfamily Panicoideae (Poaceae) shows multiple origins of C4 photosynthesis. American Journal of Botany 88: 1993-2012
Gómez-MartÍnez R. A. Culham 2000 Phylogeny of the subfamily Panicoideae with emphasis on the tribe Paniceae: evidence from the trnL-F cpDNA region. In S. Jacobs and J. Everett [eds.], Grasses, systematics and evolution, 136–140. CSIRO, Collingwood, Victoria, Australia
Grass Phylogeny Working Group. 2000 A phylogeny of the grass family (Poaceae) as inferred from eight character sets. In S. Jacobs and J. Everett (editors), Grasses: systematics and evolution, 3–7. CSIRO Publishing, Collingwood, Victoria, Australia
Grass Phylogeny Working Group. In press Phylogeny and subfamilial classification of the grasses (Poaceae). Annals of the Missouri Botanical Garden.
Hattersley P. L. Watson 1992 Diversification of photosynthesis. In G. Chapman [ed.], Grass evolution and domestication, 38–116. Cambridge University Press, London, UK
Mathews S. R. Tsai E. Kellogg 2000 Phylogenetic structure in the grass family (Poaceae): evidence from the nuclear gene phytochrome B. American Journal of Botany 87: 96-107
Sendulsky T. T. Filgueiras A. Burman 1986 Fruits, embryos, and seedlings. In T. Soderstrom, K. Hilu, C. Campbell, and M. Barkworth [eds.], Grass systematics and evolution, 31–36. Smithsonian Institution Press, Washington, D.C., USA
Sinha N. E. Kellogg 1996 Parallelism and diversity in multiple origins of C4 photosynthesis in the grass family. American Journal of Botany 83: 1458-1470[CrossRef][ISI]
Spangler R. B. Zaitchik E. Russo E. Kellogg 1999 Andropogoneae evolution and generic limits in Sorghum (Poaceae) using ndhF sequences. Systematic Botany 24: 267-281[CrossRef][ISI]
Swofford D. L. 2000 PAUP*. Phylogenetic analysis using parsimony (* and other methods), version 4. Sinauer Associates, Sunderland, Massachusetts, USA
Zuloaga F. 1986 Systematics of New World species of Panicum (Poaceae: Paniceae). In T. Soderstrom, K. Hilu, C. Campbell, and M. Barkworth [eds.], Grass systematics and evolution, 287–309. Smithsonian Institution Press, Washington, D.C., USA
Zuloaga F. O. Morrone L. Giussani 2000 A cladistic analysis of the Paniceae: a preliminary approach. In S. Jacobs and J. Everett [eds.], Grasses, systematics and evolution, 123–135. CSIRO, Collingwood, Victoria, Australia
This article has been cited by other articles:
|
|
 |
|
  D. G. Ibrahim, T. Burke, B. S. Ripley, and C. P. Osborne A molecular phylogeny of the genus Alloteropsis (Panicoideae, Poaceae) suggests an evolutionary reversion from C4 to C3 photosynthesis Ann. Bot., January 1, 2009; 103(1): 127 - 136. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Reinheimer, R. Pozner, and A. C. Vegetti Inflorescence, spikelet, and floral development in Panicum maximum and Urochloa plantaginea (Poaceae) Am. J. Botany, April 1, 2005; 92(4): 565 - 575. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. S. Aliscioni, L. M. Giussani, F. O. Zuloaga, and E. A. Kellogg A molecular phylogeny of Panicum (Poaceae: Paniceae): tests of monophyly and phylogenetic placement within the Panicoideae Am. J. Botany, May 1, 2003; 90(5): 796 - 821. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Goel, Z. Chen, J. A. Conner, Y. Akiyama, W. W. Hanna, and P. Ozias-Akins Delineation by Fluorescence in Situ Hybridization of a Single Hemizygous Chromosomal Region Associated With Aposporous Embryo Sac Formation in Pennisetum squamulatum and Cenchrus ciliaris Genetics, March 1, 2003; 163(3): 1069 - 1082. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) indicate nodes that were also found in the strict consensus of these trees. Asterisks mark non-Kranz species (or in the case of Steinchisma hians, a C3-C4 intermediate). Arrowhead indicates the strongly supported clade containing species with modified lower paleas and/or that are non-Kranz. Some of the major clades are labelled with the following identifiers. "PCK" = PCK clade; "SETAE" = setae-bearing clade; "Pan" = core Panicum clade; "Dig" = Digitaria clade; "KEEL" = keeled clade