| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Systematics |
Department of Biology, Duke University, Durham, North Carolina 27708-0338 USA
Received for publication April 17, 2003. Accepted for publication July 11, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Bayesian inference • bryophytes • peatmoss • phylogenetic reconstruction • Sphagnum
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Crum, 1992a
; Turner, 1993
). Many of their uses derive from the extraordinary absorptive capacity of peatmoss (as, for example, in bandages, diapers, filtration materials, feminine hygiene products), which in turn derives from the unique morphology of Sphagnum plants.
Sphagnum-dominated peatlands cover extensive areas in the boreal zone of North America, Europe, and Asia (Vitt, 2000
). These peatlands constitute an important reservoir for global carbon and currently function as a carbon sink (Gorham, 1991
). Sphagnum-dominated peatlands also have profound effects on gas fluxes that are determinants of global climate (e.g., CO, N2O, NH3, H2S, COS, and DMS), nutrient cycling, regional patterns of hydrology, and biodiversity (Gorham, 1991
; Moore, 1994
; Wieder and Yavitt, 1994
; Camill and Clark, 1998
). As is true of their anthropocentric utility, the ecological significance of boreal peatlands is directly related to the physical and chemical characteristics of peatmosses.
Gorham (1994)
outlined an agenda for future research on peatland ecology with particular reference to global change and stressed the importance of understanding how Sphagnum species differ in biological characteristics that affect community and ecosystem processes. Morphological, chemical, and life history features of different sphagna have demonstrable effects on their species-specific ecologies, including rates of photosynthesis, decomposition, vegetative growth, competitive ability, and reproductive biology (Clymo, 1963
; Clymo and Heywood, 1982
; McQueen, 1987
; Cronberg, 1993
; Johnson and Damman, 1993
; Rydin, 1993
; Rice, 1995
, 2000
; Rice and Schuepp, 1995
; Rice and Giles, 1996
). Correlations between habitat and species traits are informative, but a phylogenetic context is essential for rigorous comparative ecological studies so that the contributions of shared ancestry and similar (or differing) ecology can be differentiated (Harvey and Pagel, 1991
).
Several phylogenetic hypotheses have been put forward for the genus Sphagnum (Eddy, 1977
, 1979
; He and Aur, 1991
; Shaw, 2000a
). Rooting a phylogeny for the genus, however, presents exceptional difficulties. The isolated phylogenetic position of the peatmosses has made it difficult to interpret the polarity of evolutionary change within Sphagnum (Shaw, 2000a
). The problem is a familiar one in phylogenetic analyses dealing with isolated groups: morphological characters that distinguish species of Sphagnum have no clear homologs outside Sphagnum, so outgroup rooting of a generic phylogeny based on morphology is difficult or impossible. The same problem exists with regard to molecular data; nucleotide sequences for genomic regions that are variable enough to resolve phylogenetic relationships within Sphagnum cannot be aligned with homologous sequences from potential outgroups, including Andreaea, Takakia, or species of Bryopsida (Shaw, 2000a
). Genes that are conserved enough to align between Sphagnum and outgroups, on the other hand, are nearly invariant within Sphagnum and offer little with regard to phylogenetic topology among peatmoss species.
We sequenced eight conserved genes representing all three genomes for a sample of 24 Sphagnum species and two outgroups (Takakia and Andreaea), in an effort to root the phylogeny of Sphagnum. Although even eight genes (ca. 9 kilobases [kb]) provided too few informative characters to resolve relationships among most sphagna, two species, S. sericeum C. Muell. and S. lapazense Crum, were strongly supported as sister to the rest of the genus. This permitted a second tier of analyses in which Takakia and Andreaea were deleted from the analyses, S. sericeum and S. lapazense were coded as outgroups, and more variable genes were added to the data set. This approach yielded a resolved phylogeny rooted to S. lapazense and S. sericeum.
Sphagnum is distinct from all other mosses in numerous aspects of both sporophyte and gametophyte morphology. Indeed, Crum (2001a)
recently elevated peatmosses to the level of division, as the Sphagnophyta. For purposes of this paper, the peatmosses are classified as one of four classes within the division Bryophyta: Takakiopsida, Andreaeaopsida, Sphagnopsida, and Bryopsida. The Sphagnopsida include two orders, Ambuchananiales and Sphagnales. The Ambuchananiales contain only one species, A. leucobryoides, which was described from Tasmania in its own section of Sphagnum (Yamaguchi et al., 1990
). Crum and Seppelt (1999)
described the new genus Ambuchanania for Sphagnum leucobryoides and placed it in its own family and order. Shaw (2000a)
reported that 26S rDNA sequence data support the separation of Ambuchanania from Sphagnum. Sphagnum, in contrast, includes approximately 250–400 species and has a worldwide distribution. The informal term "peatmoss" is used in this paper with reference to members of the genus Sphagnum, which, unlike Ambuchanania, form extensive peat deposits in many parts of the world (Vitt, 2000
).
Morphological and developmental features that set Sphagnum apart from other mosses pertain to practically every stage of the life cycle. Sphagnum spores germinate to produce a protonemal stage that is filamentous only for the first few cell divisions, after which a two-sided apical cell is differentiated and the protonema becomes thalloid (illustrated in Ruhland, 1924
; von Goebel, 1930
; Anderson and Crosby, 1965
). Thalloid protonemal morphology is shared by the Sphagnopsida and Andreaeopsida and is probably a plesiomorphic character within the Bryophyta (Mishler and Churchill, 1985
). Sphagnum protonema are remarkably similar to the gametophytes of Coleochaete (Chlorophyta).
The leafy gametophore has an apical cell with three cutting faces, as in most other mosses, but the stems also have a subapical group of meristematic cells that contributes to growth in length (Ligrone and Duckett, 1998
). Sphagnum is generally easy to recognize in nature because mature gametophores have fasciculate branching, typically with 3–5 branches per fascicle, although sometimes with more (e.g., S. wulfianum Girg.) and occasionally with fewer or even none (in simplex forms such as S. cyclophyllum Sull. & Lesq. in Sull.). Branches are more or less dimorphic, with 1–2 so-called pendent branches that extend down the stem, effective for external capillary movement of water, and 1–2 "spreading" branches that diverge more widely from the stem. Branches near the stem apices are clustered and form a more or less distinct capitulum. The shape and prominence of capitula are often useful in determining species of Sphagnum (e.g., Eddy, 1977
; Andrus, 1980
; Flatberg, 2002
). The main stems of peatmosses consist of a central region of more or less homogeneous, thin-walled cells surrounded by a "wood cylinder" of thicker-walled, generally pigmented cells, and (0–)1–4 layers of enlarged, thin-walled, so-called hyalodermal cells. The number of hyalodermal layers, and also the degree of size differentiation between hyalodermal and internal cells, vary across the genus. In species of the section Sphagnum, the hyalodermal cells have fibrils deposited on the inside surfaces of the cell walls. These layers, and especially the strongly differentiated superficial layer(s), make the cross sectional anatomy of Sphagnum stems unlike that of any other moss.
The stem and branch leaves are usually differentiated in size and shape, but both are characterized by dimorphic leaf cells in which large, empty hyaline cells are enclosed in a network of narrower, chlorophyllose cells. The degree of differentiation in shape between the two cell types is variable, but the existence of dimorphic cells is one of the "hallmarks" of peatmosses. Holcombe (1984)
described the complex pattern of cellular morphogenesis, involving a unique series of asymmetric cell divisions, that gives rise to the dimorphic leaf cells in Sphagnum. The hyaline cells are involved in water absorption and storage and are variously perforated by pores and ornamented with cell wall fibrils. Fibrils are absent on the stem leaf hyaline cells of many species, but they are absent from the hyaline cells of branch leaves in only a few taxa (Warnstorf, 1911
). Cell wall ornamentation (i.e., fibrils) is another of the unique and characteristic features of peatmosses, so the absence of these fibrils in a few taxa has led to the obvious question of which, if any, cases represent a primitive absence. That question will be addressed in this paper.
Unlike those of "true" mosses (Bryopsida), the sporophytes of Sphagnum consist of a capsule (sporangium) and foot, with little or no development of a seta. The sporophyte is raised on a pseudopodium of gametophyte origin. It has an operculum like members of the Bryopsida, but there is no peristome, and dehiscence of the capsule occurs through a unique "pop-gun" mechanism that involves increasing internal air pressure as the columnella breaks down and the capsule dries (Ingold, 1965
). Development of the Sphagnum sporophyte is fundamentally different from that in other mosses in that sporogenous tissue originates from the amphithecium and the endothecium gives rise to the massive columnella. In the Bryopsida, both columnella and spores originate from endothecium. The capsule wall is solid in Sphagnum (i.e., without air spaces as in most Bryopsida), and there are numerous pseudostomata which may or may not be homologous with the stomata of true mosses (Boudier, 1988
).
These numerous unique morphological features suggest that the Sphagnopsida are isolated from all other mosses. Molecular phylogenetic analyses have consistently indicated that the Sphagnopsida constitute one of the earliest diverging lineages of bryophytes (Hedderson et al., 1998
; Newton et al., 2000
; Yatsentyuk, 2001
; Cox et al., in press
). These studies suggest that the Sphagnopsida and Takakiopsida (one genus: Takakia) may form a monophyletic group that is sister to the Andreaopsida plus Bryopsida, although morphologically, Takakia and Sphagnum have little in common. Indeed, Newton et al. (2000)
were not able to identify a single morphological synapomorphy that unites Sphagnum and Takakia. The placement of these two morphologically divergent groups in one clade could be an analytical artifact such as so-called "long branch attraction." Cox et al. (in press)
found that statistical support for monophylly of Sphagnum plus Takakia disappeared in Bayesian analyses utilizing increasingly complex models of substitution. This observation may support the view that their monophylly in other studies is artifactual.
Both Andreaea and Takakia were included in the present analyses as outgroups. The aims of this study were to resolve phylogenetic relationships among the major lineages of peatmosses and to clarify evolutionary polarity within the genus by outgroup rooting. The results of these analyses provide the framework for future investigations of ecological, morphological, chemical, and molecular evolution in peatmosses.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
. Sampling included four species of section Sphagnum, two Rigida, four Cuspidata, three Subsecunda, four Acutifolia, and two Squarrosa (for Supplemental Data, see the online version of this article). Sphagnum wulfianum was included with the aim of clarifying its relationship to the sections Acutifolia and Squarrosa. This species has generally been segregated in the monospecific section Polyclada (e.g., Isoviita, 1966
; Andrus, 1980
; Crum, 1984
; Flatberg, 2002
), but Shaw (2000a)
suggested that it might be nested within the Acutifolia. Analyses of molecular diversity across the genus Sphagnum indicated that two species, S. sericeum and S. lapazense, are extremely divergent from all other peatmoss species (Shaw et al., 2003
). Warnstorf (1911)
classified S. sericeum in the subsection Sericea with S. macrophyllum Brid., because these species lack fibrils on the branch leaf hyaline cell walls. While S. macrophyllum is clearly nested within the section Subsecunda (Shaw, 2000a
), S. sericeum may be a phylogenetically critical species (Eddy, 1977
) and was therefore included in the current analyses. Crum (2001b)
included S. lapazense in the section Sphagnum when he recently described the species from Bolivia, but like S. sericeum, S. lapazense was surprisingly divergent from all other sphagna based on two genomic regions sequenced for a study of global peatmoss biodiversity patterns (Shaw et al., 2003
).
Genomic sampling
Nucleotide sequences were obtained from 16 genomic regions. Mitochondrial intron sequences were obtained from NADH protein-coding subunits 1, 5, and 7, yielding 3220 nucleotides of mtDNA (hereafter, nad1, nad5, and nad7, respectively). From the chloroplast genome, we sequenced two photosystem II proteins (psbA, psbT), ribulose bis-phosphate carboxylase (rbcL), chloroplast ribosomal small protein 4 (rps4), transfer RNAGly (UCC) (trnG), and the trnL (UAA) 5' exon-trnF(GAA) intergenic spacer (hereafter, trnL). Nuclear sequences were obtained from the ITS1–5.8S-ITS2 region (hereafter, ITS), and 5' segment of the large subunit (26S) ribosomal RNA gene. Additional nuclear sequences were obtained for two introns in the LEAFY/FLO gene (hereafter, LEAFY1 and LEAFY2).
Three anonymous regions, assumed to be nuclear, were sequenced using primers designed for regions identified from random amplified polymorphic DNA (RAPDs). The RAPDs were generated for a sample of nine divergent Sphagnum species, including two representatives from each of the four major sections and a single representative from section Squarrosa. Two primers from the Operon 10-mer Kit A (OPA-1 and OPA-3) (QIAGEN Operon, Alameda, California, USA) were used in a standard RAPD amplification. Using a Qiagen Taq DNA Polymerase Kit (QIAGEN Operon), 50 µL reactions contained 1x buffer (with 1.5 mmol/L MgCl2), 200 µmol/L dNTPs, 4 µmol/L primer, 1 unit Taq, and 5–10 ng template DNA. Cycling conditions were 45 cycles of 1 min at 94°C, 1 min at 35°C, and 2 min at 72°C. Amplified products were concentrated to 18–20 µL and run entirely on a 1.5% agarose gel at 6 V/cm for 4.5 h. Bands that appeared to be monomorphic in size for three or more of the nine representatives were excised from the agarose gel and cleaned with a Qiagen Gel Extraction Kit (QIAGEN Operon). Fragments were then cloned using a TA Cloning Kit (Invitrogen, Carlsbad, California, USA) and sequenced with M13 primers and the ABI Prism BigDye Terminator Cycle Sequencing Ready Reaction Kit (PE Biosystems, Foster City, California, USA). Sequencing was accomplished using an ABI 3700 automated sequencer. Sequences found to be homologous were aligned, and primers were designed within conserved regions at the 5' and 3' ends of the fragments. After an accumulation of sequences generated with these primer sets, additional primers were designed so that a nested polymerase chain reaction (PCR) protocol might be pursued. None of the three "RAPD" regions identified in this manner was found to BLAST to a homologous sequence in GenBank, and they are here utilized as anonymous markers (designated RAPDa, RAPDb, and RAPDf).
Primer sequences for amplifying and sequencing all genomic regions utilized in this study are provided in Table 1. The PCR amplification was accomplished in several ways. An initial attempt was made to amplify fragments in 25-µL reaction volumes containing 1x buffer, 2.5 mmol/L MgCl2, 200 µmol/L dNTPs, 0.5 µmol/L each primer, 5% "Q" solution," 1 unit Qiagen Taq polymerase, and 0.3 µL template DNA (stock DNA or first round product; see below). Where possible (see Table 1), a nested protocol was utilized, in which the reaction just described was preceded with a 5 µL amplification with flanking primers, performed with the same reaction conditions and designated "first round product." When this regime failed to yield adequate amplification for sequencing, a similar procedure was followed using the Advantage-GC cDNA Polymerase Kit (Clontech, Palo Alto, California, USA). Here reaction conditions were 1x buffer, 1x GC Melt, 800 µmol/L dNTPs, 0.2 µmol/L each primer, 0.4x Advantage-GC cDNA polymerase mix, and 0.3 µL template DNA (stock DNA or first round product). All amplifications were accomplished using a similar thermocycling regime: 3 min at 95°C followed by 30 cycles of 1 min at 95°C, 1 min at 50°C, and 45 (+5 s/cycle) at 72°C. A final extension of 7 min at 72°C preceded a 4°C hold. Reactions were screened on 1% agarose gels and successful amplifications cleaned using a QIAquick PCR Purification Kit (QIAGEN Operon). The pair of bands resulting from amplication with the LEAFY/FLO primers were excised individually from agarose gels and cleaned with the Qiagen Gel Extraction Kit. Sequencing was accomplished as with the RAPD clones above, using amplification primers and where available, additional sequencing primers (see Table 1).
|
). Bayesian analyses were performed using MrBayes3 (Huelsenbeck and Ronquist, 2002
) and P4 (Foster, 2002
). The phylogenetic analyses were performed in a two-tiered fashion. The most conserved genes (mtDNA: nad1, nad5, nad7; cpDNA: psbA, psbT, rbcL, rps4; nrDNA: 26S) were used for an initial analysis of the 26-taxon data set containing 24 sphagna plus Takakia and Andreaea. Sequences from the more variable genes could not be aligned with those from Takakia and Andreaea. Although the conserved genes are not variable enough within Sphagnum to resolve sectional relationships, the eight-gene analysis indicated that two Sphagnum species, S. sericeum and S. lapazense, are early diverging taxa outside (and sister to) the main Sphagnum clade. Because that inference was strongly supported by MP, ML, and Bayesian analyses, a second data set was constructed in which Takakia and Andreaea were removed and additional sequence data from less conserved regions were added to the eight conserved genes for the 24 Sphagnum species. In the second analysis, S. lapazense and S. sericeum were coded as outgroups in order to clarify relationships among the remaining Sphagnum species. The most variable genes (RAPDa, RAPDb, RAPDf) could not be unambiguously aligned between "mainstream" Sphagnum species and the two Sphagnum outgroups, so these were coded as missing data for S. lapazense and S. sericeum. All of the phylogenetic analyses described below were conducted on both data sets.
Equally weighted parsimony analyses were conducted with 300 random taxon-addition replicates with tree bisection and reconnection (TBR) branch swapping. The "steepest descent" option was turned off but the "collapse branches when maximum length is equal to zero" option was invoked. Support for nodes was assessed by 300 nonparametric bootstrap replicates each with 10 random taxon-addition replicates. For maximum likelihood analyses, the best substitution model, namely the general time-reversible model with a proportion of invariant characters and other site rates modeled by a discrete gamma distribution (GTR + I + G), was determined by hierarchical likelihood ratio tests with the aid of MrModeltest 1.1b (Nylander, 2002
). Heuristic searches under maximum likelihood were conducted with 100 replicates of random taxon addition to the starting tree.
Maximum parsimony and maximum likelihood analyses implicitly or explicitly employ a single substitution model for the combined multigenic, multigenomic data set, although it is to be expected that different genes and regions are subject to different evolutionary processes and therefore evolve according to heterogeneous patterns of nucleotide substitution. It is possible to apply relatively crude but heterogeneous substitution models in MP analyses (e.g., first and second vs. third codon positions for protein-coding genes), but ML analyses under heterogeneous substitution models, while theoretically possible, are too computationally intensive to be practical. Recent advances in Bayesian approaches to phylogenetic inference permit separate models to be applied to different regions and subregions of DNA sequences (Foster, 2002
; Huelsenbeck and Ronquist, 2002
). Heterogeneous Bayesian approaches were employed here.
Best-fit models of nucleotide substitution were determined for each of the eight genomic regions for the first analysis (including Andreaea and Takakia) and for the 15 regions used in the second analysis (excluding the non-Sphagnum outgroup taxa) using MrModeltest, as above. The optimal substitution models for each region are listed in Tables 2 and 3. In the second analysis (excluding Andreaea and Takakia), the nad1 intron included only three variable sites, insufficient to estimate an optimal likelihood model. This region was excluded from the heterogeneous Bayesian analyses of the second data set.
|
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Parsimony, likelihood, and Bayesian analyses all converge on three inferences (Fig. 1): (1) Sphagnum is monophyletic, (2) S. lapazense, from Bolivia, and S. sericeum, from the Old World tropics, comprise early diverging lineages that are outside the main Sphagnum clade, and (3) these two species are highly divergent from all other Sphagnum species. Note branch lengths in Fig. 1 leading to these two species, relative to the branches within Sphagnum sensu stricto (s.s.). Figure 1 also shows how remarkably distinct Sphagnum is at the molecular level from the outgroups Takakia and Andreaea. Sphagnum lapazense and S. sericeum to some extent break up the extremely long branch leading to Sphagnum, but the genus is still highly divergent from its closest relatives. It appears from this analysis that S. lapazense is sister to Sphagnum (s.s.) plus S. sericeum, with the latter being sister to the remaining taxa of Sphagnum.
|
With Takakia and Andreaea deleted from the data set and more variable regions included in the analyses, more resolution was obtained among species within the main lineage of Sphagnum (Fig. 2). All four of the large sections of Sphagnum (Sphagnum, Cuspidata, Subsecunda, Acutifolia), as well as Rigida and Squarrosa, are strongly supported as monophyletic. Two major clades are resolved within Sphagnum s. s. One includes the sections Sphagnum, Cuspidata, and Rigida. The position of section Rigida relative to Sphagnum and Cuspidata was ambiguous in previous analyses (Shaw, 2000a
). Parsimony bootstrap support for the sister group relationship between sections Rigida and Cuspidata is moderate (69%), but the posterior probabilities from the homogeneous and heterogeneous Bayesian analyses are 100% for this relationship.
|
Although the Acutifolia were strongly supported as monophyletic, the sister group relationship between Acutifolia and Squarrosa, strongly supported by previous analyses (Shaw, 2000a
), was not overwhelming. Furthermore, the precise relationships of S. wulfianum, a species often segregated as its own section (Polyclada), is still ambiguous. Shaw (2000a)
suggested that S. wulfianum may be nested within the Acutifolia, whereas the current analyses suggest a closer relationship to the Squarrosa. Sphagnum wulfianum may be sister to the Squarrosa, as shown in Fig. 2; it may be sister to Squarrosa plus Acutifolia, or it may be sister to the Acutifolia.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
called attention to S. sericeum as possibly primitive within Sphagnum, but Crum (2001b)
classified S. lapazense in section Sphagnum when he described it as new from Bolivia. Sphagnum lapazense and S. sericeum are highly differentiated from other sphagna for every one of the 16 genes used in this study, and their isolated position in the combined analysis is therefore not the result of conflicts among different genes or genomic compartments. There is no molecular (or morphological) evidence that either species is of hybrid origin. In an analysis of global biodiversity patterns in Sphagnum based on chloroplast and nuclear DNA sequences from 436 accessions, S. lapazense and S. sericeum accounted for almost 20% of peatmoss molecular diversity even though they were represented by only three accessions (0.7% of the sample). Notwithstanding the rather unusual morphology of S. sericeum (Andrews, 1911
; Warnstorf, 1911
; Eddy, 1977
), this level of molecular differentiation could not have been predicted from morphological observations. The significance of these species to peatmoss phylogeny is comparable to the resolution of Amborella at or near the base of the angiosperms (Mathiews and Donoghue, 1999
; Qui et al., 1999
) and to the identification of Oedipodium griffithianum (Dicks.) Schwaegr. as sister to all other "true" mosses (Bryopsida) (Newton et al., 2000
; Cox et al., in press
).
Warnstorf (1911)
classified S. sericeum in its own subsection, Sericea, a taxonomic level equivalent to the groups referred to here as sections (Sphagnum, Cuspidata, Acutifolia, etc.). The primary morphological feature that forms the basis for segregating Sericea is the absence of fibrils on the hyaline cells of branch leaves (Fig. 3B, F). Fibrillose hyaline cells are almost universal in Sphagnum and are lacking in only a few species, including S. macrophyllum, S. cribosum Lindb., and S. splendens Maass from eastern North America, and S. efibrillosum A.L. Andrews and S. novo-caledoniae Paris & Warnstorf from Oceania. DNA sequence data clearly indicate that S. macrophyllum and S. cribosum are nested within the section Subsecunda (Shaw, 2000a
; A. J. Shaw, unpublished data). Warnstorf (1911)
, Eddy (1977)
, and Crum (1992b)
argued for a similar placement for S. efibrillosum and S. novo-caledoniae in section Subsecunda, based on morphology. Sphagnum splendens appears to be an aberrant phenotype in the section Cuspidata, doubtfully worth recognizing at the specific level. It is known from only two or three localities in an area that has been relatively thoroughly collected for Sphagnum. These phylogenetic and taxonomic hypotheses imply that the absence of hyaline cell wall fibrils in these species represents a secondary loss. The absence of fibrils on the hyaline cells of S. sericeum, in contrast, appears to be primitive, as argued by Eddy (1977)
. The single apical pore on each branch leaf hyaline cell of S. sericeum distinguishes this species from all other sphagna, including those few that lack fibrils on the hyaline cell walls (Fig. 3B).
|
further argued that the lack of differentiation between spreading and pendent branches, the relative similarity in size and structure of stem and branch leaves, and the weak differentiation in shape between hyaline and chlorophyllose cells of both stem and branch leaf cells, characteristic of S. sericeum (Fig. 3), also appear to be plesiomorphic. The arrangement of hyaline and chlorophyllose cells is rather irregular in the stem leaves of S. sericeum (Fig. 3F), an unusual feature that warrants developmental investigation given the highly regular pattern characteristic of most sphagna. Adjacent clusters of hyaline and chlorophyllose cells cannot reflect the kind of regular development described by Holcombe (1984)
and S. sericeum provides an opportunity to study the evolutionary origin of the complex developmental program that results in the regular differentiation of hyaline and chlorophyllose leaf cells in other sphagna. The apiculate leaf apices of stem and branch leaves in S. sericeum (Fig. 3A, B, E) constitute an additional feature that sets this species apart from almost all other species of Sphagnum. Most species of Sphagnum have extensive cell wall resorption near the leaf apices, and Eddy (1977)
considered the cuspidate apices of S. sericeum leaves, lacking wall resorption, as plesiomorphic. The apiculae of stem leaves consist of uniformly thick-walled cells in S. sericeum (Fig. 3B). It is noteworthy, on the other hand, that the cortical cells of S. sericeum branches include well-differentiated retort cells, a uniquely Sphagnum characteristic (Fig. 3C).
Sphagnum lapazense is less morphologically distinctive than is S. sericeum. When Crum (2001b)
recently described S. lapazense as a new species from Bolivia, he classified the species in section Sphagnum. Crum did note the absence of fibrils in the cortical cells of branches and stems of S. lapazense, but these fibrils are poorly developed or nearly absent in some other species of the section, especially in those from South America. The branch leaves are ovate (Fig. 4A), typical of section Sphagnum, and the stem leaves are smaller and triangular, also not noteworthy. The hyaline cells of branch leaves are relatively broad, as is typical of section Sphagnum (Fig. 4C), and the branch leaves have a marginal resorption furrow (not shown), a feature characteristic of sections Sphagnum and Rigida. The chlorophyllose cells are elliptical in cross section, similar to such section Sphagnum species as S. magellanicum Brid. One morphological character that does appear to be unique for the section Sphagnum (and not mentioned by Crum, 2001b
) is the absence of extensive cell wall resorption near the apices of branch leaves. Species of section Sphagnum regularly have the cell walls on adaxial branch leaf surfaces so extensively resorbed that the leaf surface has a roughened texture that derives from projecting wall remnants. This feature, so characteristic of section Sphagnum, is conspicuously absent from the branch leaves of S. lapazense (Fig. 4B).
|
Bayesian, ML, and MP analyses resolve two major clades within the main Sphagnum lineage, one including sections Sphagnum, Rigida, and Cuspidata and the other including the Acutifolia, Subsecunda, and Squarrosa. This topology is unlike any previous hypothesis for Sphagnum phylogeny. Eddy (1977)
provided a diagram (not a tree) illustrating his phylogenetic concepts for Sphagnum and indicated that S. sericeum is most similar to a hypothetical ancestral peatmoss. He further indicated that section Subsecunda includes primitive Sphagnum taxa derived from something like S. sericeum and that sections Cuspidata, Sphagnum, and Acutifolia are independently derived from the section Subsecunda, implying paraphylly of the Subsecunda. More derived species of section Subsecunda are hypothesized to have given rise to section Rigida. Eddy (1977)
specifically pinpointed S. robinsonii Warnst., a species of section Subsecunda, as a possible sister taxon to section Rigida. The results of the current analyses suggest that the Rigida are sister to the section Cuspidata.
The phylogenetic hypothesis of Eddy (1977)
implies that branch leaf marginal resorption furrows, shared by species of sections Sphagnum and Rigida, evolved independently. The current topology suggests that the resorption furrow might have evolved once in the ancestor of sections Sphagnum and Rigida plus Cuspidata, but this implies a loss of resportion furrows in the Cuspidata. Although resorption furrows characterize all species of sections Sphagnum and Rigida, morphologically indistinguishable furrows occur in the section Acutifolia (in S. molle Sull. from the Northern Hemisphere and in S. costae Crum from Brazil [Crum and Da Costa, 1994
]). Resorption furrows do not occur in other species of section Acutifolia, so any phylogenetic hypothesis requires homoplasious origins and/or losses of these features.
Eddy (1977
, p. 367) argued that the section Subsecunda "is usually considered to be the most primitive group among the Sphagna." He bases this contention on the relatively low degree of (spreading and pendent) branch dimorphism within branch fascicles, the commonness of stem and branch leaf isophylly in the Subsecunda, and the cosmopolitan distribution of the section. Most authors (e.g., Andrews, 1911
; Eddy, 1977
) argue that the Subsecunda and Cuspidata are most closely related among the large sections, although one of the examples used by Andrews (1911)
to argue for actually combining the two sections (S. mendocinum Sull. & Lesq. in Sull. from the American Pacific Northwest) turns out to likely be an intersectional hybrid (Shaw and Goffinet, 2000
). The phylogenetic topology presented here is ambiguous on this point. The relationship between the Subsecunda and Acutifolia plus Squarrosa is weakly supported, and it is possible that the Subsecunda are sister to the other clade (Sphagnum + Rigida + Cuspidata). The Subsecunda do not appear to be the sister group to the Cuspidata alone, however, although the topology presented here does not exclude the possibility that morphological similarities shared by Cuspidata and Subsecunda are plesiomorphic rather than convergent.
The lack of support for a sister group relationship between the sections Acutifolia and Squarrosa is surprising as this relationship was supported by previous analyses (Shaw, 2000a
) and is generally argued from morphological considerations (Eddy, 1977
; Crum, 1984
; Daniels and Eddy, 1990
). Sphagnum wulfianum is often segregated as the section Polyclada, but the section is usually thought to be closely related to the Squarrosa, Acutifolia, or both. Shaw (2000a)
argued that S. wulfianum is nested within the Acutifolia, along with S. girgensohnii Russ. and S. fimbriatum Wils. However, several additional populations of S. wulfianum now included in a more extensive data set (A. J. Shaw, unpublished data) appear more closely related to the Squarrosa than to the Acutifolia. Nevertheless, the precise position of S. wulfianum is still without strong support; S. wulfianum is either sister to the Squarrosa (which includes just S. squarrosum and S. teres), sister to the Acutifolia, or possibly sister to the Squarrosa plus Acutifolia. None of these hypotheses agree with the phylogenetic topology for Chinese species of Sphagnum presented by He and Aur (1991)
. According to their tree, the Squarrosa are sister to section Sphagnum, S. wulfianum is sister to S. subsecundum Nees, and the Acutifolia are sister to the Cuspidata. Clearly, their hypothesis is inconsistent with the topology presented here in just about every way.
This study has provided the most fully resolved and reliably polarized phylogeny for Sphagum available to date. Nevertheless, it is remarkable that even with almost 15 000 nucleotides from all three genomes and more than 850 parsimony informative characters, several nodes remain weakly supported. The early divergence of S. sericeum and S. lapazense is likely, although these two species do not appear to be at all closely related. The absence of hyaline cell fibrils is probably a plesiomorphic feature of S. sericeum, as hypothesized by Eddy (1977)
, and the lack of cell wall resorption on the adaxial leaf surfaces of S. lapazense may also be primitive. Sphagnum lapazense otherwise shares morphological features with members of the section Sphagnum, and it is therefore surprising that it does not share more with that group at the DNA level. Sphagnum lapazense, but not S. sericeum, has a marginal resorption furrow on the branch leaves, so the origin of this feature during the course of Sphagnum evolution is unclear. Well-differentiated retort cells occur in S. sericeum but not S. lapazense, so the origin(s) of these uniquely peatmoss features is likewise ambiguous. The phylogenetic topology presented here suggests that resorption furrows may have evolved early in Sphagnum, were retained in the sections Sphagnum and Rigida, lost independently in the Cuspidata and in the clade(s) including Subsecunda, Acutifolia, and Squarrosa, and regained in the Acutifolia.
It is clear that the sections Sphagnum, Rigida, and Cuspidata form a well-supported clade, but an outstanding question that remains is the relationship of the Subsecunda to this group. The results of this study strongly support monophylly of all six sections of Sphagnum and therefore do not support paraphylly of the Subsecunda as proposed by Eddy (1997). Analyses of a much more taxon-extensive data set also support monophylly of all the sections, including Subsecunda. The present results do not provide compelling evidence to negate Eddy's view that the Subsecunda comprise an early diverging lineage within the peatmosses. The status of section Polyclada (S. wulfianum) is still in question, although the Polyclada are probably worth segregating taxonomically only if S. wulfianum is ultimately resolved as sister to the Acutifolia plus Squarrosa. The current results strongly support the conclusion of Shaw (2000a)
, that the monotypic section Insulosa (S. aongstroemii Hartm.) is part of the larger section Acutifolia, although S. aongstroemii does appear to be an early diverging lineage within that section.
|
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Andrews A. L. 1911 Notes on North American Sphagnum. I. The groups. Bryologist 14: 72-75
Andrus R. 1980 Sphagnaceae (peat moss family) of New York State. Contributions to the Flora of New York State 3: 1-89
Boudier P. 1988 Différenciation structurale de l'épiderme du sporogone chez Sphagnum fimbriatum Wilson. Annales des Sciences Naturelles, Botanique 8: 143-156
Camill P. J. S. Clark 1998 Climate change disequilibrium of boreal permafrost peatlands caused by local processes. American Naturalist 151: 207-222[CrossRef][ISI]
Clymo R. S. 1963 Ion exchange in Sphagnum and its relation to bog ecology. Annals of Botany, New Series 27: 309-327
Clymo R. S. P. M. Heywood 1982 The ecology of Sphagnum. In A. J. E. Smith [ed.], Bryophyte ecology, 229–289. Chapman & Hall, London, UK
Cox C. J. B. Goffinet A. J. Shaw S. B. Boles In press Phylogenetic relationships among the mosses based on heterogeneous Bayesian analysis of multiple genes from multiple genomic compartments. Systematic Botany.
Cronberg N. 1993 Reproductive biology of Sphagnum. Lindbergia 17: 69-82
Crum H. A. 1984 Sphagnopsida, Sphagnaceae. North American Flora, ser. 2, part 11: 1-180
Crum H. A. 1992a A focus on peatlands and peat mosses. University of Michigan Press, Ann Arbor, Michigan, USA
Crum H. A. 1992b Miscellaneous notes on the genus Sphagnum. 1–2. Bryologist 95: 274-279[CrossRef][ISI]
Crum H. A. 2001a Structural diversity of bryophytes. University Michigan Herbarium, Ann Arbor, Michigan, USA
Crum H. A. 2001b Miscellaneous notes on Sphagnum. 11. Contributions from the University of Michigan Herbarium 23: 107-114
Crum H. A. P. Da Costa 1994 Sphagnum costae, a new Brazilian species related to S. molle Sull. Crytpogamie, Bryologie Lichénologie 15: 111-115
Crum H. A. R. D. Seppelt 1999 Sphagnum leucobryoides reconsidered. Contributions from the University of Michigan Herbarium 22: 29-31
Daniels R. E. A. Eddy 1990 Handbook of European sphagna, 2nd ed. Her Majesty's Stationery Office (HMSO), London, UK
Demesure B. N. Sodzi R. J. Petit 1995 A set of universal primers for amplification of polymorphic non-coding regions of mitochondrial and chloroplast DNA in plants. Molecular Ecology 4: 129-131[Medline]
Eddy A. 1977 Sphagnales of tropical Asia. Bulletin of the British Museum (Natural History) 5: 359-445
Eddy A. 1979 Taxonomy and evolution of Sphagnum. In G. C. S. Clarke and J. G. Duckett [eds.], Bryophyte systematics, 109–121. Academic Press, London, UK
Flatberg K. I. 2002 The Norwegian Sphagna: a field colour guide. Norges Tehnisk-naturvitenskapelige Universitet Vitenskapmuset. Rapport Botanisk Serie 2002: –1 1-44 + 54 plates
Foster P. 2002 P4—software and manual. Natural History Museum, London. website: http://www.nhm.ac.uk/zoology/external/p4.htm
Gorham E. 1991 Northern peatlands: role in the carbon cycle and probably responses to climatic warming. Ecological Applications 1: 182-195[CrossRef][ISI]
Gorham E. 1994 The future of research in Canadian peatlands: a brief survey with particular reference to global change. Wetlands 14: 206-215[ISI]
Harvey P. H. M. D. Pagel 1991 The comparative method in evolutionary biology. Oxford University Press, Oxford, UK
He X.-L. C.-W. Aur 1991 Cladistic analysis of Sphagnum in Northeast China. Acta Phtyotaxonomica Sinica 29: 131-141
Hedderson T. A. R. Chapman C. J. Cox 1998 Bryophytes and the origins and diversification of land plants: new evidence from molecules. In J. W. Bates, N. W. Ashton, and J. G. Duckett [eds.], Bryology for the twenty-first century, 65–77. Maney, Leeds, UK
Holcombe J. 1984 Morphogenesis of branch leaves of Sphagnum magellanicum Brid. Journal of the Hattori Botanical Laboratory 57: 179-240
Huelsenbeck J. P. F. Ronquist 2002 MrBayes version 3.0B. Available from the authors, website: http://morphbank.ebc.uu.se/mrbayes3/info.php
Ingold C. T. 1965 Spore liberation. Clarendon Press, Oxford, UK
Isoviita P. 1966 Studies on Sphagnum L. I. Nomenclatural revisions of the European taxa. Annales Botanici Fennica 7: 157-162
Johnson L. C. A. W. H. Damman 1993 Decay and its regulation in Sphagnum peatlands. Advances in Bryology 5: 249-296
Krellwitz E. C. K. V. Kowallik P. S. Manos 2001 Molecular and morphological analyses of Bryopsis (Bryopsidales, Chlorophyta) from the western North Atlantic and Caribbean. Phycologia 40: 330-339[ISI]
Ligrone R. J. G. Duckett 1998 Development of the leafy shoot in Sphagnum (Bryophyta) involves the activity of both apical and subapical meristems. New Phytologist 140: 581-595[CrossRef][ISI]
Mathiews S. M. J. Donoghue 1999 The root of angiosperm phylogeny inferred from duplicate phtyochrome genes. Science 286: 947-950
McQueen C. B. 1987 The effects of major ions on the growth of Sphagnum protonemata. Symposia Biologica Hungarica 35: 305-313
Mishler B. D. S. P. Churchill 1985 Transition to a land flora: phylogenetic relationships of the green algae and bryophytes. Cladistics 1: 305-328
Moore T. R. 1994 Trace gas emissions from Canadian peatlands and the effect of climate change. Wetlands 14: 223-228
Nadot S. R. Bajon B. Lejeune 1994 The choloroplast gene rps4 as a tool for the study of Poaceae phylogeny. Plant Systematics and Evolution 191: 27-38[CrossRef][ISI]
Newton A. E. C. J. Cox J. G. Duckett J. Wheeler B. Goffinet B. D. Mishler T. A. J. Hedderson 2000 Evolution of the major moss lineages. Bryologist 103: 187-211[CrossRef][ISI]
Nylander J. A. A. 2002 MrModeltest, version 1.1b. Available from the author, website: http://www.ebc.uu.se/systzoo/staff/nylander.html
Pacak A. Z. Szweykowska-Kulinska 2000 Molecular data concerning alloploid character and the origin of chloroplast and mitochondrial genomes in the liverwort species Pellia borealis. Journal of Plant Biotechnology 2: 101-108
Qui Y.-P. J. Lee F. Berasconi-Quadroni D. Soltis P. Soltis M. Zanis E. A. Zimmer Z. Chen V. Savolainen M. W. Chase 1999 The earliest angiosperms: evidence from mitochondrial, plastid, and nuclear genomes. Nature 402: 404-407
Rice S. K. 1995 Patterns of allocation and growth in aquatic Sphagnum species. Canadian Journal of Botany 73: 349-359[ISI]
Rice S. K. 2000 Variation in carbon dioxide discrimination within and among Sphagnum species in a temperate wetland. Oecologia 123: 1-8
Rice S. K. L. Giles 1996 The influence of water content and leaf anatomy on carbon isotope discrimination and photosynthesis in Sphagnum. Plant, Cell, and Environment 19: 118-124[CrossRef]
Rice S. K. P. H. Schuepp 1995 On the ecological and evolutionary significance of branch and leaf morphology in aquatic Sphagnum. American Journal of Botany 82: 833-846[CrossRef][ISI]
Ruhland W. 1924 Unterklasse Sphagnales. Allgemeine Verhältnisse. In A. Engler and K. Prantl [eds.], Die natürlichen Pflanzenfamilien, 70–114. W. Englemann, Leipzig, Germany
Rydin H. 1993 Mechanisms of interactions among Sphagnum species along water-level gradients. Advances in Bryology 5: 153-185
Shaw A. J. 2000a Phylogeny of the Sphagnopsida based on nuclear and chloroplast DNA sequences. Bryologist 103: 277-306[CrossRef][ISI]
Shaw A. J. 2000b Molecular phylogeography and cryptic speciation in the mosses, Mielichhoferia elongata and M. mielichhoferiana (Bryaceae). Molecular Ecology 9: 595-608[CrossRef][Medline]
Shaw A. J. C. J. Cox S. B. Boles 2003 Global patterns of peatmoss (Sphagnum) biodiversity. Molecular Ecology 12: 2553-2570[CrossRef][Medline]
Shaw A. J. B. Goffinet 2000 Molecular evidence of reticulate evolution in the peatmosses (Sphagnum), including S. ehyalinum sp. nov. Bryologist 103: 357-374[CrossRef][ISI]
Souza-Chies T. T. G. Bittar S. Nadot L. Carter E. Besin B. Lejeune 1997 Phylogenetic analysis of the Iridaceae with parsimony and distance methods using the plastid gene rps4. Plant Systematics and Evolution 204: 109-123[CrossRef][ISI]
Swofford D. L. 2001 Paup*: phylogenetic analysis using parsimony (*and other methods). version 4.0b8. Sinauer Associates, 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][ISI][Medline]
Turner R. G. 1993 Peat and people: a review. Advances in Bryology 5: 315-328
Vitt D. H. 2000 Peatlands: ecosystems dominated by bryophytes. In A. J. Shaw and B. Goffinet [eds.], Bryophyte biology, 312–343. Cambridge University Press, Cambridge, UK
von Goebel K. 1930 Organographie der Pflanzen inbesondere der Archegoniaten und Samenpflanzen. 2. Bryopyten-Pteridophyten. Gustav Fischer, Jena, Germany
Warnstorf C. 1911 Sphagnales-Sphagnaceae (Sphagnologia Universalis). In H. G. A. Engler [ed.], Das Pflanzenreich, 1–546. Wilhelm Englemann, Leipzig, Germany
Wieder R. K. J. B. Yavitt 1994 Peatlands and global climate change: insights from comparative studies of sites situated along a latitudinal gradient. Wetlands 14: 229-238[ISI]
Williams B. 1982 The healing powers of Sphagnum moss. New Scientist 95: 713-714
Yamaguchi T. R. D. Seppelt Z. Iwatsuki A. M. Buchanan 1990 Sphagnum (section Buchanania) leucobryoides sect. et sp. nov. from Tasmania. Journal of Bryology 16: 45-54
Yatsentyuk S. P. 2001 Molecular phylogeny of the Bryophyta and Lycopodophyta, according to results from some sequences of chloroplast DNA. Ph.D. dissertation, Moscow State University, Moscow, Russia
This article has been cited by other articles:
|
|
 |
|
  A. J. Shaw, S. Boles, and B. Shaw A phylogenetic delimitation of the "Sphagnum subsecundum complex" (Sphagnaceae, Bryophyta) Am. J. Botany, June 1, 2008; 95(6): 731 - 744. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
|
