Didyma

American Journal of Botany 97(7): 1156–1167. 2010.

HYBRIDIZATION AND POLYPLOIDY OF AN AQUATIC PLANT, RUPPIA (RUPPIACEAE), INFERRED FROM PLASTID AND NUCLEAR DNA PHYLOGENIES1 Yu Ito2, Tetsuo Ohi-Toma2, Jin Murata2, and Norio Tanaka3,4 2Botanical

Gardens, Graduate School of Science, The University of Tokyo, Tokyo, 112-0001, Japan; and 3Tsukuba Botanical Garden, National Museum of Nature and Science, Tsukuba, 305-0005, Japan

• Premise of the study: The monogeneric family Ruppiaceae is found primarily in brackish water and is widely distributed on all continents, many islands, and from subartic to tropical zones. Ruppia taxonomy has been confusing because of its simplified morphology yet high phenotypic plasticity and the existence of polyploidy and putative hybrids. This study addresses the current classification of species in the genus, the origin of putative hybrids and polyploids, and the distribution of Ruppia species.

• Methods: Separate molecular phylogenetic analyses using plastid DNA and nuclear-encoded PHYB data sets were performed after chromosome observations.

• Key results: The resultant trees were largely congruent between genomes, but were incongruent in two respects: the first incongruence may be caused by long outgroup branches and their effect on ingroup rooting, and the second is caused by the existence of heterogeneous PHYB sequences for several accessions that may reflect several independent hybridization events. Several morphological species recognized in previous taxonomic revisions appear paraphyletic in plastid DNA and PHYB trees.

• Conclusions: Given the molecular phylogenies, and considering chromosome number and morphology, three species and one species complex comprising six lineages were discerned. A putative allotriploid, an allotetraploid, and a lineage of hybrid origin were identified within the species complex, and a hybrid was found outside the species complex, and their respective putative parental taxa were inferred. With respect to biogeography, a remarkably discontinuous distribution was identified in two cases, for which bird-mediated seed dispersal may be a reasonable explanation.

Key words: hybridization; matK; PHYB; phylogeny; polyploidy; rbcL; rpoB; rpoC1; Ruppia; Ruppiaceae.

About 1–2% of all vascular plant species (87 families: Cook, 1996) have adapted morphologically and physiologically to aquatic environments. Aquatic plants are usually categorized into emergent, floating, or submerged types (Sculthorpe, 1967). In addition, aquatic plants may be adapted to freshwater, saltwater, or brackish water habitats. Because fresh and brackish water areas are restricted to lakes, rivers, and estuaries, populations of species adapted to these environments are highly geographically isolated. Nevertheless, it has long been 1

Manuscript received 14 June 2009; revision accepted 19 May 2010.

The authors thank H. Freitag, P. García-Murillo, J. Hansen, K. Nonaka, H. Taneda, H. Ohba, H. Kato, H. R. Na, D. Perleberg, H. J. Cho, L. Rozas, R. Upson, A. R. Ramirez, D. Bryon, and J. J. Orth for providing materials; S. W. L. Jacobs, A. Skriptsova, H.-K. Choi, Y.-P. Yang, S. R. Yadav, K. Povidisa, M. Delefosse, T. Hammer, K. Torn, G. Izzo, R. Ishikawa, and S. Stern for help with field research; K. Sasamura for technical advice on chromosome observation; and S. Gale, S. Stern, and T. Hammer for improving an earlier draft of the manuscript and two anonymous reviewers for valuable comments on the manuscript. This study is based in part on the thesis of Y.I. at the University of Tokyo, Japan. The authors are thankful for grants to N.T. from the Ministry of Education, Culture, Sports, Science, and Technology, Japan (17710194 and 21710248) and from a study project of the National Museum of Nature and Science: Integration of Systematics and Molecular Phylogenetics of All Groups of Organisms and to Y.I. from the Academic Research Grant Program (International), The University of Tokyo, Japan and Global COE Program (Integrative Life Science Based on the Study of Biosignaling Mechanisms), MEXT, Japan.

4 Author for correspondence (e-mail: ntanaka@kahaku.go.jp) doi:10.3732/ajb.0900168

known that many genera and species that are specialized in this way are distributed worldwide (e.g., Najas marina L., Potamogeton pectinatus L., Ruppia maritima L., and Zannichellia palustris L.: Sculthorpe, 1967; Les et al., 2003). Sculthorpe (1967) suggested that dispersal of propagules (seeds, turions, and tubers) by water birds may contribute to such wide geographic distributions.

The genus Ruppia, the only genus of the family Ruppiaceae, is representative of submerged aquatic plants adapted to brackish waters. The genus is generally limited to estuaries and brackish lakes as well as inland saline and alkaline lakes (Verhoeven, 1979). Unlike most aquatic plants in fresh or brackish water, the genus often occurs on oceanic islands, including Hawaii (St. John and Fosberg, 1939), Vanuatu (Hashimoto et al., 2002), and Ogasawara (Ono and Okutomi, 1985). As an example of an extreme case, R. maritima occurs from subarctic to tropical zones in both hemispheres (Den Hartog, 1971). The fruit of R. maritima is recognized as an important element in the diet of several species of water birds, and the germination rate of defecated seeds is enhanced (Figuerola and Green, 2004). Hence, it is reasonable to assume that the widespread distribution of R. maritima was caused and is maintained by bird-mediated seed dispersal.

Ruppia species are characterized by a simplified morphology of a slender body, linear leaves, and a pedunculate, spicate inflorescence with a pair of bractless, perianthless flowers, of which the peduncle and leaf morphologies have been mainly used for taxonomy. These morphological characters often show high phenotypic plasticity among taxa, among populations within a taxon, and even within populations, often leading to

American Journal of Botany 97(7): 1156–1167, 2010; http://www.amjbot.org/ © 2010 Botanical Society of America

1156

July 2010]

Ito et al.—Phylogeny of RUPPIA

taxonomic confusion (Van Vierssen et al., 1981; Hara, 1983; Aedo and Fernandez Casado, 1988). Nevertheless, several regional taxa have been described based on a comparison of limited samples from areas such as Asia (Hara, 1983), the Pacific region (St. John and Fosberg, 1939), North America (Fernald and Wiegand, 1914), and Europe (Triest and Symoens, 1991).

Van Vierssen et al. (1981) suggested that an adequate comparison using plant material from around the world would be needed to derive a more robust classification of Ruppia. Recently, Zhao and Wu (2008) reviewed the classification of Ruppia and recognized five species: R. maritima, R. cirrhosa (Petagna) Grande, R. megacarpa R. Mason, R. polycarpa R. Mason, and R. tuberosa J. S. Davis & Toml. (Table 1). Of these, R. maritima and R. cirrhosa were treated as widespread species, whereas the other three were deemed endemic to Oceania.

Most of the regional taxa in this classification were treated as synonyms of the two widespread species (e.g., R. rostellata as a synonym of R. maritima and R. drepanensis and R. spiralis as synonyms of R. cirrhosa). However, the taxonomic criteria of Zhao and Wu (2008) are not always applicable in identifying herbarium collections and live specimens (Y. Ito and N. Tanaka, personal observation). In addition, morphological intermediates with sterile flowers have been reported in local populations (Kadono, 1994), suggesting that intragenus hybridization may have occurred, further complicating its taxonomic classification.

Polyploidy occurs widely in plants, and polyploidization (allopolyplodization) following hybridization plays an important role in speciation (Arnold, 1997; Soltis and Soltis, 2009). In Ruppia, diploid (2n = 20) and tetraploid (2n = 40) races have been reported from several regions of the world, whereas triploids (2n = 30) and hexaploids (2n = 60) are only occasionally found (reviewed in Talavera et al., 1993; Table 1). In past classifications, some researchers have recognized the two widespread species, R. maritima and R. cirrhosa, as diploid and polyploid, respectively (Reese, 1962). However, others have recognized the former as both diploid and tetraploid and the latter as tetraploid (Van Vierssen et al., 1981). Still others have described the former as tetraploid and the latter as diploid (Cirujano, 1986), or recognized diploid and tetraploid entities in both species (Triest and Symoens, 1991). This confusion may have been caused by the paucity of material available for study, reflecting the ambiguity of classification for the genus based on morphology alone. As such, a molecular phylogenetic analysis based on a combined data set of plastid and nuclear DNA (ptDNA and nDNA) corresponding to uniparental and biparental inheritance would be particularly useful in elucidating the existence of hybrids and the origin of polyploids (e.g., Doyle and Doyle, 1999), as well as in differentiating taxa and their phylogenetic relationships. Furthermore, low- or single-copy genes have been used as nuclear markers for many plant groups because they are less prone to mistaken orthological assessment than those in large gene families and have been used for many plant groups (e.g., Madia/Raillardiopsis group: Barrier et al.

[1999]; Glycine tomentella polyploidy complex: Doyle et al.

[2002]; Spartina: Ainouche et al. [2004]; Eragrostis tef: Ingram and Doyle [2003]; Geinae: Smedmark et al. [2003]; Paeonia: Sang et al. [2004]; Viburnum: Winkworth and Donoghue [2004]; Vandenboschia radicans complex: Ebihara et al. [2005]; Silene: Popp et al. [2005]; Rosa carolina complex: Joly et al.

[2006]; Cardamine: Lihová et al. [2006]; Capsella: Slotte et al.

[2006]; Paniceae: Doust et al. [2007]; Spartina: Fortune et al. [2007]; Brumus: Fortune et al. [2008]; Persicaria: Kim

1157

et al. [2008]; Bryonia: Volz and Renner [2008]; Arabidopsis: Shimizu-Inatsugi et al. [2009]).

To address the taxonomic inconsistencies in Ruppia, we examined the phylogenetic relationship of Ruppia species within the context of the recent classification proposed by Zhao and Wu (2008) and investigated the role of hybridization and polyploidization in the evolution of this genus. We performed molecular phylogenetic analyses of the genus covering a broad taxonomic sample and representing a wide geographic range. In addition, chromosome numbers were determined for certain representatives, and the morphology was reevaluated in light of the phylogenetic data. We used these new lines of evidence to address the present classification of species in the genus, the origin of putative hybrids and polyploids, and the distribution pattern of Ruppia species.

MATERIALS AND METHODS Taxon sampling—In total, 46 accessions from 44 localities of Ruppia species were collected from Asia, Europe, North Africa, North America, Oceania, Pacific, and South America (Appendix 1). One widespread species, R. maritima, was also reported in coastal areas of Africa (Verhoeven, 1979), but no high-quality specimens that would have allowed DNA extraction and analysis were available. Using only the morphological key of Zhao and Wu (2008), each sample was tentatively identified as one of five species. Leaf blade width, which was recognized as a valuable character for distinguishing between R. cirrhosa, R. megacarpa, and R. polycarpa by Zhao and Wu (2008), was excluded because of its continuous variation across the remainder of the accessions. In addition, chromosome numbers noted by Zhao and Wu (2008) were not considered for species assignment because they only cited previous studies, some of which were incorrect, (e.g., 2n = 10 for R. maritima; Table 1). Two accessions of R. polycarpa from Asia and North America and one accession of R. megacarpa from Asia were identified, although both species were treated as endemic to Oceania by Zhao and Wu (2008). Positive identification of accessions collected without fruit was problematic for all taxa, except R. tuberosa, and these accessions were treated as Ruppia A to C. Potamogeton maackianus A. Benn. (Potamogetonaceae) and Syringodium isoetifolium (Asch.) Dandy (Cymodoceaceae) were used as outgroups following the phylogeny of the core alismatid families presented by Les et al. (1997).

DNA extraction, amplification, and sequencing—Total genomic DNA was extracted from silica gel-dried leaf tissue using the method of Doyle and Doyle (1987) after pretreatment with HEPES buffer (pH 8.0) (Setoguchi and Ohba, 1995). Sequences determined in the current study were registered with the DNA Data Bank of Japan (DDBJ), which is linked to GenBank, and their accession numbers are given with the sample information in Appendix 1.

Fragments of the plastid genes matK, rbcL, rpoB, and rpoC1 (collectively referred to as ptDNA) were amplified by polymerase chain reaction (PCR) using the following primer pairs. The primers for rbcL, RM_F (5′-TATTTGCAAGGGAATTAGGA-3′) and RM_R (5′-AAGCTTCACGGATAATTTCA-3′), which amplify fragments of the gene (542 bp), were newly designed based on the nucleotide sequence of R. maritima (accession number U03729) from GenBank.

For other genes, published or modified primers were used: RM_749F (5′-TTGAGCGAACACATTTCTATG-3′) modified from Whitten et al. (2000) and 8R (Ooi et al., 1995) for the matK gene (550 bp), 2f and 4r for rpoB (508 bp), and 1f and 3r for rpoC1 (508 bp) (Royal Botanic Gardens, Kew: http://www.rbgkew. org.uk/barcoding/update.html). PCR amplification was conducted using TaKaRa Ex Taq polymerase (TaKaRa Bio, Shiga, Japan), and PCR cycling conditions were 94°C for 60 s; then 30 cycles of 94°C for 45 s, 52°C for 30 s, 72°C for 60 s; and finally 72°C for 5 min. The PCR products were digested using ExoSAP-IT (GE HealthCare, Piscataway, New Jersey, USA) and amplified using the ABI PRISM Big Dye Terminator v3.1 (Applied Biosystems, Foster City, California, USA) with the same primers as those used for PCR. DNA sequencing was performed using an ABI PRISM 377 DNA Sequencer (Applied Biosystems). Complementary electropherograms were compared by eye using the software Genetyx-Win Version 3 (Software Development Co., Tokyo, Japan).

The PHYB gene (a distinct member of the phytochrome gene family) was selected as a nuclear marker, based on its phylogenetic utility as a single

40 Eur./Asia/ Ocea.

Brackish/ fresh Absent Short Sharp Short/curved Present Small Four Absent Short Sharp Short/curved Present Small Four

20 Eur./Asia/ Paci./N. Am. Brackish Absent Short Sharp Intermediate n.a. n.a.

Four

Brackish

30 Asia

“Triploid”

Absent Short Sharp Short/curved Present Small Four

Brackish

unk.

S. Am.

“Filifolia”

Absent Short Sharp Short/curved Present Small Four

Saline

unk.

Inl. Am.

“Utahian”

Brackish/ Fresh alkaline Absent Long Sharp Long/coiled Present Small Eight

20 Asia/Inl. Am.

“Occidentalis”

Absent Short Sharp Long/coiled Present Small Eight

Brackish

18 Ocea.

Four

Absent Long Flat Long/coiled Present Large

Brackish

20 Asia/Ocea.

R. megacarpa

Absent Long Flat n.a. n.a. n.a.

Four

Brackish

unk.

Asia

Hybrid

R. megacarpa (20)

R. polycarpa (18, 20)

R. polycarpa

R. megacarpa (20)

R. polycarpa (18, 20)

Present Short Sharp Long/coiled Absent Small Eight

Hypersaline

20 Ocea.

R. tuberosa

R. tuberosa (n.a.)

R. tuberosa (20, 30)

Abbreviations: Eur., Europe plus North Africa; Inl. Am., Inland North America; Ocea., Oceania; Paci., Pacific; N. Am., North America; S. Am., South America; n.a., not applicable; unk., unknown

Turions Leaf length Leaf tips Peduncles Stalks Fruits Carpels

Habitats

Characters Chromosome numbers (2n) Distribution

“Diploid”

Proposed taxa

B) Present study

“Tetraploid”

R. maritima var. brevirostris (20) R. maritima var. maritima (20, 40) R. maritima var. brevirostris (40) — R. maritima (20) R. drepanensis (20) R. maritima (40) R. cirrhosa var. cirrhosa (40) R. maritima var. maritima (20, 40) R. cirrhosa var. drepanensis (20) R. maritima var. brevirostris (20, 40) R. cirrhosa (40) R. maritima (10)

R. maritima complex

R. spiralis (40, 60) R. cirrhosa (40)

Reese (1962) Van Vierrsen et al. (1981)

Jacobs & Brock (1982) Cirujano (1986) Triest & Symoens (1991) Zhao & Wu (2008)

Taxa (Chromosome numbers [2n])

A) Past classification

(A) Past classifications with emphasis on chromosome numbers and (B) taxa proposed in the present study in Ruppia and morphological comparison of taxa proposed in the present study.

Table 1.

1158 American Journal of Botany [Vol. 97

July 2010]

Ito et al.—Phylogeny of RUPPIA

or low copy nuclear locus (Mathews et al., 2000; Simmons et al., 2001).

Fragments of a part of exon 1 of PHYB were initially amplified by PCR using the published primers of B-up and B-down (Mathews et al., 2000) for 12 representative samples. The PCR cycling conditions were 94°C for 90 s; then 35 cycles of 94°C for 45 s, 60°C for 30 s, 72°C for 90 s; and finally 72°C for 10 min. The fragments obtained were digested with ExoSAP-IT and directly sequenced. After checking the homology of the obtained sequences by conducting a BLAST search (National Center for Biotechnology Information), three specific forward and reverse primers for Ruppia were newly designed, as follows: phyB_38F (5′-CTCGCTGTTCGTGCTATCTCG-3′), phyB_ ruppiaF (5′-CCATACTTCTCCCAGATGCATTCC-3′) and phyB_ruppiaR (5′-CCATACTTCTCCCAGATGCATTCC-3′). For PCR amplification of all materials, either phyB_38F or phyB_ruppiaF was used together with Bdown to obtain 562 and 1050 bp fragments, respectively, under the following condition: 94°C for 60 s; then 25 cycles of 94°C for 45 s, 60°C for 60 s, 72°C for 90 s; and finally 72°C for 5 min. The PCR products were purified using GeneClean (BIO 101, Carlsbad, California, USA). On direct sequencing of 22 accessions, overlapping double peaks were found at the same sites for complementary strands in the electropherograms. These products were cloned using a TOPO TA Cloning kit for Sequencing (Invitrogen, Carlsbad, California, USA). At least 16 clones per sample were chosen and their sequences were determined using the same procedure as that used in the first PCR followed by direct sequencing. For the cloned sequences, nucleotides that were not detected by direct sequencing were regarded as PCR errors.

1159

Data analysis—Multiple sequences of each gene were manually aligned because no length mutation was detected. Phylogenetic analyses were independently performed for ptDNA and PHYB data sets because we detected some well-supported incongruence between these genomes (see below). The incongruence length difference test (Farris et al., 1994) was conducted to test congruence between four ptDNA genes using a partition homogeneity test with 1000 replicates in the program PAUP* 4.0b10 (Swofford, 2002). There was no significant heterogeneity indicated by this test (P value > 0.05 for all six pairs), and so the four ptDNA data sets were combined. One representative sequence was used for accessions having the identical combined sequence.

Phylogenetic inference was performed using maximum parsimony (MP) and maximum likelihood (ML) in PAUP*, as well as Bayesian inference (BI) in the program MrBayes 3.1.2 (Ronquist and Huelsenbeck, 2003). In the MP analysis, a heuristic search was performed with 100 random addition replicates involving tree-bisection-reconnection (TBR) branch swapping, with the MulTrees option in effect. The MaxTrees option was set at no limits for the analysis. Bootstrap analyses (Felsenstein, 1985) were performed using 1000 replicates with TBR branch swapping and the simple addition sequences. In the ML analysis, heuristic searches were performed with 10 random addition replicates using a best-fit model (TVM+G for PtDNA and K80+G for PHYB). Parameter values were estimated by a hierarchical likelihood ratio test in the program Modeltest 3.7 (Posada and Crandall, 1998). Bootstrap analyses were performed using 100 replicates with TBR branch swapping and as-is addition sequences under the same ML models. In the BI analysis, hierarchical likelihood ratio tests implemented in the program MrModeltest 3.7 (Nylander, 2002) were used

Fig. 1. (A) One of two most parsimonious trees of combined plastid DNA (matK, rbcL, rpoB, and rpoC1) gene sequences. (B) One of four most parsimonious trees of nuclear encoded PHYB sequences. Accessions were tentatively identified by morphology following the taxonomic criteria of Zhao and Wu (2008). DELTRAN optimization was used for branch length measures of the phylograms. Numbers above the branches indicate bootstrap support (BP) calculated in maximum parsimony (bold) and maximum likelihood (italic) analyses and those below indicate Bayesian prior probabilities (PP). BP <70 and PP <0.9 are indicated by asterisks. Boldfaced accessions have heterogeneous PHYB sequences; for these, sequence pairs are connected by a dotted line and named #1 and #2, respectively. Note that some accessions in each tree represent multiple identical accessions.

1160

[Vol. 97

American Journal of Botany

for substitution model selection (GTR+G for ptDNA and K80+G for PHYB).

The Bayesian Markov chain Monte Carlo algorithm was run for 110 000 generations in the ptDNA analysis and for 200 000 generations in the PHYB analysis until the average standard deviation of split frequency was below 0.01, with four incrementally heated chains starting from random trees and sampling one out of every 100 generations. Of the generations, the first 25% of trees (27 500 and 50 000 generations for each data set, respectively) were discarded as burn-in, and the remaining trees were used to calculate a 50% majority-rule consensus tree and to determine posterior probabilities for branches. The data matrices and the MP and ML trees are available from the TreeBASE database (http://www.treebase.org, study accession S10485).

Chromosome observations—Somatic chromosome number of a subset of samples from 15 localities was obtained by light microscopic examination.

Root tips collected in the field were pretreated with 0.002 M 8-hydroxyquinoline at 4°C overnight, and fixed with freshly mixed Carnoy’s fixative (3 : 1 ethyl alcohol : acetic acid) for at least 30 min, and then preserved at 4°C. For microscopic observation, root tips were soaked in 1 N HCl for 1 h followed by 10 min at 60°C. After being immersed in tap water, the materials were stained in a drop of 1.5% orcein acetate solution on a slide glass, and then squashed.

RESULTS Phylogenetic inferences based on ptDNA— Of all recovered accessions, those with identical sequence are represented by a single name as operational taxonomic units (OTUs), e.g., R. maritima A to G (Fig. 1A). Based on the combined data set consisting of the four genes (2108 bp), two MP trees (tree length = 279 steps, consistency index = 0.9498, retention index = 0.8955) and one ML tree with -lnL = 4434.9927 were obtained. We refer to single-accession lineages, or clades comprising identical sequences as “branches” here, and label clades of interest with the prefix “Pt” (for plastid-based lineages) or “PB” (for PHYB-based lineages). Because MP, ML, and BI analyses recovered largely congruent topologies, one of the two MP trees is shown (Fig. 1A). In the ptDNA tree, the branch Pt-I (comprising R. megacarpa from northern Japan and Australia, Ruppia A from northern Japan, and Ruppia B from Russian Far East) is the sister group of everything else. Two distinct haplotypes of R. tuberosa form a well-supported clade (clade Pt-II) that is the sister group of the remaining accessions. In one of the two MP trees, branch Pt-III (R. polycarpa A from Australia) is the sister group of a large clade with a basal polytomy: the latter clade includes one branch (Pt-VI) and two well-supported clades (Pt-IV and Pt-V).

Clade Pt-IV consists of subclades Pt-IV-1 (which includes accessions of R. maritima A from North America, R. polycarpa B from northern Japan and inland North America, and Ruppia C from northern Japan) and subclade Pt-IV-2 (which includes R. maritima B to D from Asia, Europe North America, and Pacific). Clade Pt-V consists of R. cirrhosa from Europe and R. maritima E and F from Asia and Europe; branch Pt-VI is the accession of R. maritima G from the Falkland Islands (South America).

Phylogenetic inferences based on PHYB— Of 45 Ruppia accessions, 22 have two different sequences (the remainders have one). In the phylogenetic analyses, four MP trees (tree length = 516 steps, consistency index = 0.7946, retention index = 0.8188) and one ML tree with -lnL = 4001.3640 were obtained.

One of the four MP trees that has the same branching pattern as those recovered from the ML and BI analyses is shown (Fig. 1B). The names of individual accessions correspond to those in Fig. 1A. In the PHYB tree, the clade PB-I (which contains R. tuberosa A and B, each of which yielded two distinct sequence types) shows evidence of moderate rate elevation. The branch

PB-II (which includes identical sequences derived from R. megacarpa from northern Japan and Australia, Ruppia A from northern Japan and Ruppia B from Russian Far East) is the sister group of the remaining taxa, with a single accession of R. polycarpa A from Australia (branch PB-III), the next successive sister group. The remaining accessions are divided into three major lineages: (1) PB-IV, comprising two weakly to moderately supported subclades: PB-IV-1 (57%, 56%, and 0.68 for MP, ML bootstrap supports, and Bayesian posterior probability) and PB-IV-2 (83%, 84%, and 1.00), (2) clade PB-V, with two sequence types, and (3) PB-VI with a single accession. The latter two lineages (clade PB-V and branch PB-VI) are weakly supported as sister taxa (63%, 65%, and 0.86). Ruppia A from northern Japan had two sequences, one each belonging to branch PB-II and clade PB-IV, respectively; the latter sequence was identical to that of R. polycarpa B from northern Japan and inland North America. Similarly, R. cirrhosa from Europe, R. maritima E and F from Europe, North Africa, Asia, and South America, and Ruppia C from northern Japan each comprised two sequences that fell in distinct positions on the tree (i.e., clades PB-IV and PB-V, or PB-IV and PB-VI).

Chromosome numbers— Four cytotypes, 2n = 18, 20, 30, and 40, were observed across the 15 samples (Fig. 2). Based on these observations and those of previous studies (reviewed in Talavera et al., 1993), the basic chromosome number of Ruppia is predicted to be x = 10. The lowest number, 2n = 18 observed in only R. polycarpa A from Australia is likely to be of aneuploid origin. Diploids (2n = 20) were observed from Asian, Australian, and North American material. Only one examined sample of Ruppia C from northern Japan, was triploid (2n = 30). Six accessions from Asia and Europe were found to be tetraploids (2n = 40) (Appendix 1).

DISCUSSION Comparison of plastid and PHYB phylogenetic inferences— The separate molecular phylogenetic analyses of Ruppia based on ptDNA and PHYB sequence data sets provide valuable new insights. Of these, topological incongruencies between the ptDNA and PHYB trees are especially notable. One topological incongruence involves distinctly placing PHYB sequences from individual accessions (clade PB-IV, that includes several accessions whose plastid sequences belong to branches Pt-I and Pt-VI, and clades Pt-IV and Pt-V). Similar topological incongruences between ptDNA and nDNA trees have been observed elsewhere and have been inferred to represent the effects of hybridization and/or polyploidization (e.g., Winkworth and Donoghue, 2004; Obbard et al., 2006; Ohi-Toma et al., 2006; Kim et al., 2008). The details of the inferred hybridization and polyploidization events in Ruppia are discussed later.

The alternative scenario, though it seems less likely, is the possibility of gene losses following the duplication of the PHYB locus, which occurs singly in genomes of grasses (Mathews and Sharrock, 1996), except in maize (Childs et al., 1997; Sheehan et al., 2004). If this is the case, however, the resultant tree may be expected to have two clear lineages, as is the case of GBSSI-1 and -2 in family Rosaceae (Evans et al., 2000), Adh1 and Adh2 in Paeonia (Sang et al., 2004), and WaxyA and WaxyB in Spartina (Fortune et al., 2008). That is apparently not the case in the present study; the plastid and nuclear trees are otherwise mostly congruent.

July 2010]

Ito et al.—Phylogeny of RUPPIA

1161

Fig. 2. Somatic chromosomes in representatives of Ruppia. (A) 2n = 18 (Australia). (B) 2n = 20 (China). (C) 2n = 30 (Japan). (D) 2n = 40 (Japan).

Bar indicates 10 μm.

Another instance of topological incongruence was detected concerning the earliest evolutionary splits in the genus. Such incongruence between gene trees may have a number of biological or methodological causes, e.g., mistaken orthology of duplicated genes, plastid capture via hybridization, and incomplete lineage sorting (Wendel and Doyle, 1998; Sang, 2002).

Although members of the branches, clades, and subclades detected in each of the ptDNA and PHYB trees mostly correspond (R. megacarpa and Ruppia B in Pt-I and PB-II, Pt-II and PB-I, Pt-III and PB-III, R. maritima A and R. polycarpa B in Pt-IV-1 and PB-IV-1, R. maritima B to D in Pt-IV-2 and PB-IV-2, R. cirrhosa and R. maritima E and F in Pt-V and PB-V, and Pt-VI and PB-VI), the branching patterns at the basal position in both trees clearly differed. In the ptDNA tree, the branch Pt-I defines the root of the tree, and the clade Pt-II was at the next. In contrast, clade PB-I in the PHYB tree (which corresponds to clade Pt-II) defines the root of Ruppiaceae. Given that use of a distantly related outgroup can affect the root of an ingroup (e.g., Graham et al., 2002; Shavit et al., 2007; Graham and Iles, 2009), this may be a function of long outgroup branches being attracted to relatively long branches within the ingroup, a possible instance of long-branch attraction (Felsenstein, 1978).

Systematics of Ruppia— The phylogenetic relationships inferred from molecular data of ptDNA and PHYB were compared to the classification of Zhao and Wu (2008) (summarized in Fig. 3). In the classification, only R. tuberosa corresponds to the circumscription indicated by clades Pt-II and PB-I in the molecular phylogenies. This species is morphologically unique in possessing sessile fruits and swollen turions at the end of its shoots, and it is ecologically unusual because it occurs in hypersaline lakes in Australia (Davis and Tomlinson, 1974; Jacobs and Brock, 1982). Thus, R. tuberosa is phylogenetically as well as morphologically and ecologically distinct (Table 1). Although diploid (2n = 20) and triploid (2n = 30) individuals have been reported for the species (Snoeijs and Van Der Ster, 1983), only diploids (2n = 20) were confirmed in the current study.

Other species in Zhao and Wu (2008) were not supported by ptDNA and PHYB branches, clades, or subclades. Given these molecular phylogenies, chromosome numbers, and reevaluated morphological characters, we discuss species recognition within Ruppia and propose a revised classification (Fig. 3, Table 1).

The pair of branches Pt-I and PB-II includes R. megacarpa from northern Japan and Australia, Ruppia A from northern Japan, and Ruppia B from Russian Far East (Fig. 3). Of these,

1162

American Journal of Botany

[Vol. 97

Fig. 3. Taxonomic grouping proposed in this study based on the trees presented in Fig. 1. The root positions are shown as arrowheads (outgroups trimmed for clarity). Clade PB-I is modified to fit clade Pt-II. The three well-defined Ruppia species are represented by closed bars, and the six entities of the R. maritima complex are represented by shaded bars. Four cases of putative hybrids/polyploids are represented by open or open-shaded bars. The collecting area of each accession is given as Asia, Euro, Inl. Am, N. Am, Ocea, Paci, and S. Am (respectively referring to Asia, Europe plus North Africa, inland North America, North America, Oceania, Pacific, and South America). Boldfaced accessions have heterogeneous PHYB sequences. Dotted lines connect corresponding branches, clades, and subclades between plastid DNA and PHYB trees. Less-supported branches are shown as dotted lines. Chromosome counts are given in brackets beside accessions (uncounted accessions are shown as asterisks). Note that some accessions in each tree represent multiple identical accessions.

Ruppia A also harbored a second, distantly related PHYB sequence in the subclade PB-IV-1. Because the chromosome numbers of some accessions of R. megacarpa were diploids with 2n = 20, R. megacarpa and Ruppia B are considered to be diploid R. megacarpa, with larger fruits (if any) and flat leaf tips (Table 1), and Ruppia A is probably of hybrid origin (discussed later). With the exception of whether fruits are actually set, there are no differences in morphology between R. megacarpa and the hybrid (Table 1). Although R. megacarpa was formerly recognized as a species endemic to Oceania (Mason, 1967), R. megacarpa as described here exhibits a largely discontinuous distribution in northern East Asia and southern Oceania (Fig. 4, Table 1). No R. megacarpa-like herbarium specimens have been observed from any of the areas between these two distribution centers. Although there are no sufficiently old fossil records of Ruppia, except those from the Pliocene (Zhao et al., 2004), given the divergence time of the basal lineage of Ruppiaceae (ca. 65 Myr: Janssen and Bremer, 2004) and the lack of

sequence divergence in ptDNA and PHYB among accessions from northern East Asia and southern Oceania, geographical isolation in both hemispheres in R. megacarpa may have occurred recently, possibly as a result of seed dispersal by birds.

Accessions that key to R. polycarpa according to Zhao and Wu (2008) occur at two different positions in the ptDNA and PHYB trees (Fig. 3): R. polycarpa A from Australia in branches Pt-III and PB-III, and R. polycarpa B from northern Japan and inland North America in subclades Pt-IV-1 and PB-IV-1. Zhao and Wu (2008) recognized R. polycarpa based on 4–16 (usually 6–8) carpels and coiled, long peduncles, following the conclusions of Jacobs and Brock (1982), but they did not mention leaf characteristics. The accession of R. polycarpa A has short leaves (up to 150 mm) and leaf sheaths (ca. 10 mm), whereas the accession of R. polycarpa B has conspicuously longer leaves (up to 220 mm) and leaf sheaths (ca. 65 mm). The latter is similar to R. occidentalis Watson with 4–9 carpels and long leaf sheaths (14– 57 mm) and has often been recognized in inland alkaline lakes

July 2010]

Ito et al.—Phylogeny of RUPPIA

1163

Fig. 4. Geographic distribution of Ruppia based on the accessions used in the current study. Hybrid: black; R. megacarpa: blue; R. tuberosa: yellow; R. polycarpa: red; R. maritima complex: purple.

in North America (Kaul, 1992). Hence, at least R. polycarpa A from Australia with its unique phylogenetic position and morphological distinctiveness should be treated as R. polycarpa, and R. polycarpa B should be treated separately along with its related accessions (see later). With respect to chromosome numbers, the accession of R. polycarpa A from Australia is 2n = 18, which is the same as one of two cytotypes reported (2n = 18 and 2n = 20) for R. polycarpa (Mason, 1967; Brock, 1982).

Ruppia maritima identified by its noncoiled or few-coiled short peduncles (<50 mm) was divided among four lineages in the ptDNA and PHYB trees together with Ruppia C, R. cirrhosa, and R. polycarpa B (Fig. 3): (1) R. maritima A from inland North America in subclades Pt-IV-1 and PB-IV-1, including R. polycarpa B from northern Japan and inland North America and Ruppia C from northern Japan; (2) R. maritima B to D from Asia, Europe, North America, and Pacific in subclades Pt-IV-2 and PB-IV-2; (3) R. maritima E and F from Asia, Australia, and Europe in clades Pt-V and PB-V and subclade PB-IV-2 including R. cirrhosa from Europe with long peduncles (ca. 160 mm); and (4) R. maritima G from South America in branches Pt-VI and PB-VI and subclade PB-IV-1. The accessions of R. maritima in four lineages could not be distinguished by any morphological characters; thus, accessions in clade Pt-IV to branch Pt-VI and clade PB-IV to branch PB-VI, except those of Ruppia A, are treated as part of the R. maritima complex here.

In the complex, R. maritima A to D and R. polycarpa B have one PHYB sequence, whereas Ruppia C, R. cirrhosa, and R.

maritima E to G have two distantly related PHYB sequences (Fig. 3). Given their respective chromosome numbers, R. maritima B to D, Ruppia C, and R. cirrhosa plus R. maritima E and F are defined as “Diploid”, “Triploid”, and “Tetraploid” entities, respectively, and the second PHYB sequences could possibly be the result of polyploidization (see below). Consequently, the “Tetraploid” entity includes both R. cirrhosa and R. maritima, indicating a large variation in peduncle length. Ruppia maritima G, whose cytotype is unknown, is defined here as the “Filifolia” entity, considering the taxonomic history of Ruppia in South America, where R. filifolia was recognized by Skottsberg (1916) on the basis of its many-branched filiform stems. Ruppia filifolia includes plant materials from the Falkland Islands (Moore, 1973); its second PHYB sequence appears to be of hybrid origin (see below). From an ecological viewpoint, R. maritima A, whose cytotype is unknown, is characterized by its special habitat, namely an inland salt lake in Utah, USA, and so we refer to it as the “Utahian” entity. Most other accessions of R. maritima inhabit brackish water areas along coastlines. As mentioned, R. polycarpa B seems to correspond morphologically to R. occidentalis and is therefore referred to as the diploid “Occidentalis” entity with 2n = 20 (Table 1). On the basis of past studies and herbarium specimens, the “Occidentalis” entity is distributed continuously from inland North America to Alaska, the Kuril Islands (Takahashi and Kuwahara, 1998), Sakhalin, and Hokkaido, Japan (Fig. 4). Consequently, we provisionally name “Diploid”, “Triploid”, “Tetraploid”, “Filifolia”, “Utahian”, and “Occidentalis” entities within the broader R. maritima complex.

1164

[Vol. 97

American Journal of Botany

Hybridization and polyploidization— As a number of hybridization and polyploidization events have been identified for aquatic monocots (Les and Philbrick, 1993), in Ruppia, four cases of putative hybridization, including polyploidization (at the triploid and tetraploid levels), were inferred from the ptDNA and PHYB trees. We infer these events based on several accessions that harbored heterogeneous sequences, which assigned to two different positions in the PHYB phylogeny (Fig. 3): (1) Ruppia A with PHYB sequences in the branch PB-II and the subclade PB-IV-1; (2) the “Tetraploid” entity of the R. maritima complex with PHYB sequences in the subclade PB-IV-2 and the clade PB-V; (3) the “Filifolia” entity of the R. maritima complex with PHYB sequences in the subclade PB-IV-1 and the branch PB-VI; and (4) the “Triploid” entity of the R. maritima complex with PHYB sequences in the subclade PB-IV-2 and the clade PB-V (the cytotypes of (1) and (3) are currently unknown). The origin of these hybridizations is discussed in relation to the ptDNA and PHYB trees and cytotypes.

In the first case, Ruppia A from northern Japan was found to have ptDNA sequences identical to R. megacarpa as well as PHYB sequences identical to R. megacarpa and the “Occidentalis” entity of the R. maritima complex. Therefore, it appears that Ruppia A originated through a cross between R. megacarpa as the maternal parent and the “Occidentalis” entity as the paternal parent. In one of the two populations of the hybrid, the “Occidentalis” entity occurs together with Ruppia A, whereas R. megacarpa was not found around the hybrid populations but occurs nearby in northern Japan, indicating that the hybrid was established there, followed by the disappearance of R. megacarpa (Fig. 4). Because only flowering specimens (no fruiting specimens) were collected, it could be defined as a sterile hybrid.

The second case is the “Tetraploid” entity, which formed distinct clades in the ptDNA and PHYB trees, Pt-V and PB-V, but was also represented in the subclade PB-IV-2 (Fig. 3). Because the heterogeneous PHYB sequences were not closely related, the “Tetraploid” entity cannot be considered an autotetraploid.

Rather, it is an allotetraploid formed by hybridization between different entities in the complex. In the subclade PB-IV-2, one of its PHYB sequences was nested within those of the “Diploid” entity of the R. maritima complex, indicating that subclade PBIV-2 is a parental lineage, probably paternal, even though no other parental lineage was detected. Recently, it was revealed that tetraploids of the genus Paeonia were derived from crosses between genetically differentiated geographical diploid races (Sang et al., 2004). Therefore, the sampling strategy in this study may not have included the candidate progenitors because of the extremely widespread distribution of the genus. On the other hand, it is possible that the maternal parent lineage is now extinct.

Similarly, the “Filifolia” entity formed distinct branches in the ptDNA and PHYB trees, Pt-VI and PB-VI, but it was also represented in the subclade PB-IV-1, indicating that it was formed by hybridization between different entities in the complex, though no closely related taxa were detected.

The last case is the “Triploid” entity found in Lake Ogawara, Japan. Kadono (1994) mentioned that the Ruppia species in this lake might be of hybrid origin because of its intermediate peduncle morphology between R. maritima and R. cirrhosa and because of the absence of specimens with fruits. The accession of Ruppia C had slightly elongated peduncles and flowers with four carpels (Table 1), and it was restricted to one small population (10 × 30 m). In field observations, no fruiting specimens

were observed during the fruiting season of Japanese Ruppia, but many small individuals with rhizomes were observed in the spring (Y. Ito and T. Ohi-Toma, personal observation). The accession might be a triploid hybrid that reproduces vegetatively as in triploid hybrids of many other plant groups (e.g., Takano and Okada, 2002; Ayres et al., 2008; Gresta et al., 2008). The “Triploid” entity shared the same ptDNA sequence with the “Occidentalis” entity and harbored the same two PHYB sequences as the “Tetraploid” entity from Asia and Australia. Finally, the “Occidentalis” and “Tetraploid” entities in Asia and Australia are related to the origin of the “Triploid” entity, based on DNA sequences and chromosome numbers, even though the accession does not have a PHYB sequence that is closely related to the “Occidentalis” entity. One possible explanation is that the “Triploid” entity might have subsequently undergone introgression through multiple hybridizations or backcrosses with another putative paternal taxon, the “Tetraploid” entity.

Conclusions— To develop a systematic understanding of the genus Ruppia, we performed molecular phylogenetic analyses using ptDNA and PHYB sequence data sets. Several morphological species recognized in previous taxonomic revisions appear to be paraphyletic in both trees. Given the molecular phylogenies, chromosome numbers, and reevaluated morphological characters, three species—R. megacarpa, R. polycarpa, and R. tuberosa—are well defined. The R. maritima complex comprise what we provisionally refer to here as the “Diploid”, “Triploid”, “Tetraploid”, “Filifolia”, “Utahian”, and “Occidentalis” entities. In the complex, three entities of hybrid origin have been detected: the “Tetraploid” entity is an allotetraploid that is least related to the “Diploid” entity. The “Filifolia” entity, in which the cytotype is unknown, is formed by hybridization within the complex. The “Triploid” entity is derived from the “Occidentalis” and “Tetraploid” entities. In addition, one case of hybridization between distantly related taxa was detected, and its respective putative parental taxa were inferred: this hybrid derived from R. megacarpa and the “Occidentalis” entity parentage. Therefore, many hybridization events have occurred in Ruppia; however, the number of taxa is relatively small. Although R. megacarpa has been generally considered to represent one of three species endemic to Oceania (together with R. polycarpa and R. tuberosa), a disjunct population of R. megacarpa was found in northern Far East Asia. A similar Asia–Oceania discontinuous distribution was observed in the “Tetraploid” entity as well. Bird-mediated seed dispersal is a possible explanation for the two cases of the disjunct Asia– Oceania distribution pattern, probably followed by hybridization. These results clarify that a taxonomic reappraisal of the complex will be necessary following the examination of additional material collected from around the world.

LITERATURE CITED Aedo, C., and M. A. Fernández Casado. 1988. The taxonomic position of Ruppia populations along the Cantabrian coast. Aquatic Botany 32: 187–192.

Ainouche, M. L., A. Baumel, A. Salmon, and G. Yannic. 2004.

Hybridization, polyploidy and speciation in Spartina (Poaceae). New Phytologist 161: 165–172.

Arnold, M. L. 1997. Natural hybridization and evolution. Oxford University Press, Oxford, UK. Ayres, D. R., E. Grotkopp, K. Zarerine, C. M. Sloop, M. J. Blum, J. P. Bailey, C. K. Anttila, and D. R. Strong. 2008. Hybridization between invasive Spartina densiflora (Poaceae) and native S. foliosa

July 2010]

Ito et al.—Phylogeny of RUPPIA

in San Francisco Bay, California, USA. American Journal of Botany 95: 713–719.

Barrier, M., B. G. Baldwin, R. H. Robichaux, and M. D. Purugganan.

1999. Interspecific hybrid ancestry of a plant adaptive radiation: Allopolyploidy of the Hawaiian silversword alliance (Asteraceae) inferred from floral homeotic gene duplications. Molecular Biology and Evolution 16: 1105–1113.

Brock, M. A. 1982. Biology of the salinity torelant genus Ruppia L. in saline lakes in South Australia. I Morphological variation within and between species and ecophysiology. Aquatic Botany 13: 219–248.

Childs, K. L., F. R. Miller, M. M. Cornonnier-Pratt, L. H. Pratt, P. W. Morgan, and J. E. Mul. 1997. The Sorghum photoperiod sensitivity gene, Ma3, encodes a phytochrome B. Plant Physiology 113: 611–619.

Cirujano, S. 1986. El género Ruppia L. (Potamogetonaceae) en la mancha (España). Boletim da Sociedade Broteriana, Série 2: 293–303 [in Spanish with English abstract].

Cook, C. D. K. 1996. Aquatic plant book. SPB Academic Publishing, Amsterdam, Netherlands.

Davis, J. S., and P. B. Tomlinson. 1974. A new species of Ruppia in high salinity in Western Australia. Journal of the Arnold Arboretum 55: 59–66.

Den Hartog, C. 1971. De Nederlandse Ruppia soorten. Gorteria 5: 148– 153 [in Dutch].

Doust, A. N., A. M. Penly, S. W. L. Jacobs, and E. A. Kellogg. 2007.

Congruence, conflict, and polyploidization shown by nuclear and chloroplast markers in the monophyletic “bristle clade” (Paniceae, Panicoideae, Poaceae). Systematic Botany 32: 531–533.

Doyle, J. J., and J. L. Doyle. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11–15.

Doyle, J. J., and J. L. Doyle. 1999. Nuclear protein-coding genes in phylogeny reconstruction and homology assessment: Some examples from Leguminosae. In P. M. Hollingsworth, R. M. Bateman, and R. J. Gornall [eds.], Molecular systematics and plant evolution, 229–254.

Taylor and Francis, London, UK. Doyle, J. J., J. L. Doyle, A. H. D. Brown, and R. G. Palmer. 2002.

Genomes, multiple origins, and lineage recombination in the Glycine tomentella (Leguminosae) polyploid complex: Histone H3-D gene sequences. Evolution 56: 1388–1402.

Ebihara, A., H. Ishikawa, S. Matsumoto, S.-J. Lin, K. Iwatsuki, M. Takamiya, Y. Watano, and M. Ito. 2005. Nuclear DNA, chloroplast DNA, and ploidy analysis clarified biological complexity of the Vandenboschia radicans complex (Hymenophyllaceae) in Japan and adjacent area. American Journal of Botany 92: 1535–1547.

Evans, R. C., L. A. Alice, C. S. Campbell, E. A. Kellogg, and T. A. Dickinson. 2000. The granule-bound starch synthase (GBSSI) gene in the Rosaceae: Multiple loci and phylogenetic utility. Molecular Phylogenetics and Evolution 17: 388–400.

Farris, J. S., M. Källersjö, A. G. Kluge, and C. Bult. 1994. Testing significance of incongruence. Cladistics 10: 315–319.

Felsenstein, J. 1978. Cases in which parsimony or compatibility methods will be positively misleading. Systematic Zoology 27: 401–410.

Felsenstein, J. 1985. Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39: 783–791.

Fernald, M. L., and K. M. Wiegand. 1914. The genus Ruppia in eastern North America. Rhodora 16: 119–128.

Figuerola, J., and A. J. Green. 2004. Effects of seed ingestion and herbivory by waterfowl on seedling establishment: A field experiment with wigeongrass Ruppia maritima in Doñana, south-west Spain.

Plant Ecology 173: 33–38.

Fortune, P. M., N. Pourtau, N. Viron, and M. L. Ainouche. 2008.

Molecular phylogeny and reticulate origins of the polyploid Bromus species from section Genea (Poaceae). American Journal of Botany 95: 454–464.

Fortune, P. M., K. A. Schierenbeck, A. K. Ainouche, J. Jacquemin, J. F. Wendel, and M. L. Ainouche. 2007. Evolutionary dynamics of Waxy and the origin of hexaploid Spartina species (Poaceae).

Molecular Phylogenetics and Evolution 43: 1040–1055.

1165

Graham, S. W., and W. J. D. Iles. 2009. Different gymnosperm outgroups have (mostly) congruent signal regarding the root of flowering plant phylogeny. American Journal of Botany 96: 216–227.

Graham, S. W., R. G. Olmstead, and S. C. H. Barrett. 2002. Rooting phylogenetic trees with distant outgroups: A case study from the commelinoid monocots. Molecular Biology and Evolution 19: 1769–1781.

Gresta, F., G. M. Lombardo, L. Siracusa, and G. Ruberto. 2008.

Saffron, an alternative crop for sustainable agricultural systems: A review. Agronomy for Sustainable Development 28: 95–112.

Hara, H. 1983. Comments on the east asiatic plants (12). Journal of Japanese Botany 58: 321–330 [in Japanese].

Hashimoto, T., K. Sugimura, N. Tanaka, G. Kokubugata, T. Konichi, S. Chanel, and T. Iwashina. 2002. A list of herbarium specimens from Vanuatu, collected in 2000 and 2001. Annals of the Tsukuba Botanical Garden 21: 1–23.

Ingram, A. L., and J. J. Doyle. 2003. The origin and evolution of Eragrostis tef (Poaceae) and related polyploids: Evidence from nuclear waxy and plastid rps161. American Journal of Botany 90: 116–122.

Jacobs, S. W. L., and M. A. Brock. 1982. A revision of the genus Ruppia (Potamogetonaceae) in Australia. Aquatic Botany 14: 325–337.

Janssen, T., and K. Bremer. 2004. The age of major monocot groups inferred from 800+ rbcL sequences. Botanical Journal of the Linnean Society 146: 385–398.

Joly, S., J. R. Starr, W. H. Lewis, and A. Bruneau. 2006. Polyploid and hybrid evolution in roses east of the Rocky Mountains. American Journal of Botany 93: 412–425.

Kadono, Y. 1994. Aquatic plants of Japan. Bun’ichisogo, Shuppan, Tokyo [in Japanese].

Kaul, R. B. 1992. Distribution, habitats, and taxonomy of Ruppia maritima L. and R. occidentalis S. Watson in Nebraska. Transactions of the Nebraska Academy of Sciences 16: 67–74.

Kim, S.-T., S. E. Sultan, and M. J. Donoghue. 2008. Allopolyploid speciation in Persicaria (Polygonaceae): Insights from a low-copy nuclear region. Proceedings of the National Academy of Sciences, USA 105: 12370–12375.

Les, D., M. A. Cleland, and M. Waycott. 1997. Phylogenetic studies in Alismatidae, II: Evolution of marine angiosperms (seagrasses) and hydrophily. Systematic Botany 22: 443–463.

Les, D. H., D. J. Crawford, R. T. Kimball, M. L. Moody, and E. Landolt.

2003. Biogeography of discontinuously distributed hydrophytes: A molecular appraisal of intercontinental disjunctions. International Journal of Plant Sciences 164: 917–932.

Les, D., and C. T. Philbrick. 1993. Studies of hybridization and chromosome number variation in aquatic angiosperms: Evolutionary implications. Aquatic Botany 44: 181–228.

Lihová, J., K. K. Shimizu, and K. Marhold. 2006. Allopolyploid origin of Cardamine asarifolia (Brassicaceae): Incongruence between plastid and nuclear ribosomal DNA sequences solved by a single-copy nuclear gene. Molecular Phylogenetics and Evolution 39: 759–786.

Mason, R. 1967. The species of Ruppia in New Zealand. New Zealand Journal of Botany 5: 519–531.

Mathews, S., and R. A. Sharrock. 1996. The phytochrome gene family in grasses (Poaceae): A phylogeny and evidence that grasses have a subset of the loci found in dicot angiosperms. Molecular Biology and Evolution 13: 1141–1150.

Mathews, S., R. C. Tsai, and E. A. Kellogg. 2000. Phylogenetic structure in the grass family (Poaceae): Evidence from the nuclear gene phytochrome B. American Journal of Botany 87: 96–107.

Moore, D. M. 1973. Additions and amendments to the vascular flora of the Falkland Islands. British Antarctic Survey Bulettin 32: 85–88.

Nylander, J. A. A. 2002. MrModeltest v.1.0. Program distributed by the author. Department of Systematic Zoology, Uppsala University, Uppsala, Sweden. Website http://www.ebc.uu.se/systzoo/staff/ nylander.html.

Obbard, D. J., S. A. Harris, R. J. A. Buggs, and J. R. Pannell. 2006.

Hybridization, polyploid, and the evolution of sexual systems in Mercurialis (Euphorbiaceae). Evolution 60: 1801–1815.

Ohi-Toma, T., T. Sugawara, H. Murata, S. Wanke, C. Neinhuis, and J. Murata. 2006. Molecular phylogeny of Aristolochia sensu

1166

American Journal of Botany

lato (Aristolochiaceae) based on sequences of rbcL, matK, and phyA genes, with special reference to differentiation of chromosome numbers. Systematic Botany 31: 481–492.

Ono, M., and K. Okutomi. 1985. Endemic species and vegetation of the Bonin Islands, Aboc-Sha Co., Kamakura, Japan.

Ooi, K., Y. Endo, J. Yokoyama, and N. Murakami. 1995. Useful primer designs to amplify DNA fragments of the plastid gene matK from angiosperm plants. Journal of Japanese Botany 70: 328–331.

Popp, M., P. Erixon, F. Eggens, and B. Oxelman. 2005. Origin and evolution of a circumpolar polyploid species complex in Silene (Caryophyllaceae) inferred from low copy nuclear RNA polymerase introns, rDNA, and chloroplast DNA. Systematic Botany 30: 302–313.

Posada, D., and K. A. Crandall. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817–818.

Reese, G. 1962. Zur intragenerischen taxonomie der Gattung Ruppia L. Ein cytosystematischer Beitrag. Zeimchrift fur Botanik 50: 237–264.

Ronquist, F., and J. P. Huelsenbeck. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572–1574.

Sang, T. 2002. Utility of low-copy nuclear gene sequences in plant phylogenetics. Critical Reviews in Biochemistry and Molecular Biology 37: 121–147.

Sang, T., J. Pan, D. Zhang, D. Ferguson, C. Wang, K.-Y. Pan, and D.-Y. Hong. 2004. Origins of polyploids: An example from peonies (Paeonia) and a model for angiosperms. Biological Journal of the Linnean Society 82: 561–571.

Sculthorpe, C. D. 1967. The biology of aquatic vascular plants. Edward Arnold, London, UK. Setoguchi, H., and H. Ohba. 1995. Phylogenetic relationships in Crossostylis (Rhizophoraceae) inferred from restriction site variation of chloroplast DNA. Journal of Plant Research 108: 87–92.

Shavit, L., D. Penny, M. D. Hendy, and B. R. Holland. 2007. The problem of rooting rapid radiations. Molecular Biology and Evolution 24: 2400–2411.

Sheehan, M. J., P. R. Framer, and T. P. Brutnell. 2004. Structure and expression of maize phytochrome family homeologs. Genetics 167: 1395–1405.

Shimizu-Inatsugi, R., J. Lihová, H. Iwanaga, H. Kudoh, K. Marhold, O. Savolainen, K. Watanabe, et al. 2009. The allopolyploid Arabidopsis kamchatica originated from multiple individuals of Arabidopsis lyrata and Arabidopsis halleri. Molecular Ecology 18: 4024–4048 .

Simmons,M.P.,C.C.Clevinger,V.Savolainen,P.H.Archer,S.Mathews, and J. J. Doyle. 2001. Phylogeny of the Celastraceae inferred from phytochrome B gene sequences and morphology. American Journal of Botany 88: 313–325.

Skottsberg, C. J. F. 1916. Die vegetationsverhaltnisse langs der Cordillera de los Andes S. on 41° S. Kungl. Svenska Vetenskapsakademiens Handlingar 56: 171.

Slotte, T., A. Ceplitis, B. Neuffer, H. Hurka, and M. Lascoux. 2006.

Intrageneric phylogeny of Capsella (Brassicaceae) and the origin of the tetraploid C. bursa-pastoris based on chloroplast and nuclear DNA sequences. American Journal of Botany 93: 1714–1724.

Smedmark, J. E. E., T. Eriksson, R. C. Evans, and C. S. Campbell.

2003. Ancient allopolyploid speciation in Geinae (Rosaceae):

[Vol. 97

Evidence from nuclear granule-bound starch synthase (GBSSI) gene sequences. Systematic Biology 52: 374–385.

Snoeijs, P. J. M., and H. E. Van Der Ster. 1983. Some notes on the cytotaxonomy of two Ruppia species in South Australia. Aquatic Botany 16: 297–302.

Soltis, P. S., and D. E. Soltis. 2009. The role of hybridization in plant speciation. Annual Review of Plant Biology 60: 561–588.

St.. John, H., and F. R. Fosberg. 1939. A new variety of Ruppia maritima (Ruppiaceae) from the tropical pacific. Occasional Papers of the Bernice P. Bishop Museum 15: 175–178.

Swofford, D. L. 2002. PAUP*: Phylogenetic analysis using parsimony (*and other methods), version 4.0b. Sinauer, Sunderland, Massachusetts, USA. Takahashi, H., and Y. Kuwahara. 1998. Notes on Ruppia occidentalis S. Watson from Atlasova (Alaid), the northern Kurils. Acta Phytotaxonomica et Geobotanica 49: 193–204.

Takano, A., and H. Okada. 2002. Multiple occurences of triploid formation in Globba (Zingiberaceae) from molecular evidence. Plant Systematics and Evolution 230: 143–159.

Talavera, S., P. García-Murillo, and J. Herrera. 1993. Chromosome numbers and a new model for karyotype evolution in Ruppia L. (Ruppiaceae). Aquatic Botany 45: 1–13.

Triest, L., and J.-J. Symoens. 1991. Isozyme variation in populations of the submerged halophyte Ruppia (Ruppiaceae). In L. Triest [ed.], Isozymes in water plants. Opera Botanica Belgica vol. 4, 115–132.

National Botanic Garden of Belgium, Meise, Belgium.

Van Vierssen, W., R. J. Van Wijk, and J. R. Van Der Zee. 1981.

Some additional notes on the cytotaxonomy of Ruppia taxa in western Europe. Aquatic Botany 11: 297–301.

Verhoeven, J. T. A. 1979. The ecology of Ruppia dominated communities in western Europe. I. Distribution of Ruppia representatives in relation to their autecology. Aquatic Botany 6: 197–268.

Volz, S. M., and S. S. Renner. 2008. Hybridization, polyploidy, and evolutionary transitions between monoecy and dioecy in Bryonia (Cucurbitaceae). American Journal of Botany 95: 1297–1306.

Wendel, J. F., and J. J. Doyle. 1998. Phylogenetic incongruence: window into genome history and molecular evolution. In D. E. Soltis, P. S. Soltis, and J. J. Doyle [eds.]. Molecular systematic of plants, vol. 2, DNA sequencing, 265–296. Kluwer, Boston, Massachusetts, USA. Whitten, W. M., N. H. Williams, and M. W. Chase. 2000. Subtribal and generic relationships of Maxillarieae (Orchidaceae) with emphasis on Stanhopeinae: Combined molecular evidence. American Journal of Botany 87: 1842–1856.

Winkworth, R. C., and M. J. Donoghue. 2004. Viburnum phylogeny: Evidence from the duplicated nuclear gene GBSSI. Molecular Phylogenetics and Evolution 33: 109–126.

Zhao, L.-C., M. E. Collinson, and C.-S. Li. 2004. Fruits and seeds of Ruppia (Potamogetonaceae) from the Pliocene of Yushe Basin, Shanxi, northern China and their ecological implications. Botanical Journal of the Linnean Society 145: 317–329.

Zhao, L.-C., and Z.-Y. Wu. 2008. A review on the taxonomy and evolution of Ruppia. Journal of Systematics and Evolution 46: 467–478.

July 2010]

Ito et al.—Phylogeny of RUPPIA

1167

Appendix 1. Population code, GenBank accessions (matK, rbcL, rpoB, rpoC1, and PHYB), Locations, voucher information, entities in the R. maritima complex or species proposed in the current study, and chromosome numbers. Herbaria abbreviations: The University of Tokyo = TI, Kochi Prefectural Makino Botanical Garden = MBK, National Museum of Nature and Science = TNS, Hokkaido University = SAP, National Herbarium of New South Wales = NSW. Taxon—Population code, GenBank accessions: matK, rbcL, rpoB, rpoC1, and PHYB (a = partially obtained sequences); Locations; Voucher specimens; Herbaria; Entities in the R. maritima complex (e.g., “Diploid”) or Lineages; Chromosome number.

Ruppia cirrhosa (Petagna) Grande—SKY; AB507925, AB507885, AB507965, AB508005, AB508058, AB508059; Skye Island, UK; YI01299; TNS; “Tetraploid”; 2n = 40.

R. maritima L.—ALB; AB534775, AB534784, AB534795, AB534802, AB534813; Point Aux Pins, Alabama, USA; YI01247; TNS; “Diploid”; 2n =?. ALV; AB534777, AB534785, AB534792, AB534801, AB534812; Laguna de Alvarado, Veracruz, Mexico; YI01387; TNS; “Diploid”; 2n =?. ANA; AB507905, AB507865, AB507945, AB507985, AB508028; Anatom Island, Vanuatu; TNS9516725; TNS; “Diploid”; 2n = ?.

ART; AB507906, AB507866, AB507946, AB507986, aAB508029; Artern, Germany; YI00876; TNS; “Diploid”; 2n =?. BAR; AB507909, AB507869, AB507949, AB507989, aAB508032; Olong Williams Canal, Barataria Bay, Louisiana, USA; YI01226; TNS; “Diploid”; 2n =?. CAP; AB507907, AB507867, AB507947, AB507987, aAB508030; Cape Breton Island, Nova Scotia, Canada; CAN521697; SAP; “Diploid”; 2n =?. COR; AB534778, AB534786, AB534794, AB534800, aAB534810, aAB534811; Porto Vecchio, Corsica Isl., France; YI01248; TNS; “Tetraploid”; 2n = 40. DUB; AB507915, AB507875, AB507955, AB507995, AB508038, AB508039; Dubrovnik, Croatia; YI00878; TNS; “Tetraploid”; 2n =?. FAL; AB534779, AB534787, AB534793, AB534799, AB534808, AB534809; Port Stephens, West Falkland, Falklands, U.K,; YI01251; TNS; “Filifolia”; 2n =?. GRA; AB507908, AB507868, AB507948, AB507988, AB508031; Grand Bay Area, Gulf of Mexico, Jackson, Mississippi, USA; YI01233; TNS; “Diploid”; 2n =?. HAS; AB507916, AB507876, AB507956, AB507996, aAB508040, aAB508041; Hasunuma Pond, Chiba, Japan; YI00822; TNS; “Tetraploid”; 2n = 40. JEJ; AB507917, AB507877, AB507957, AB507997, aAB508042, aAB508043; Seogwi, Jeju, South Korea; YI00879; TNS; “Tetraploid”; 2n =?. LIE; AB507918, AB507878, AB507958, AB507998,aAB508044, aAB508045; Huelva, Spain; YI00877; TNS; “Tetraploid”; 2n =?. MAA; AB507919, AB507879, AB507959, AB507999, aAB508046, aAB508047; Montecollina Bore, c. 218 km NE Lyndhurst, Strzelecki Track, South Australia, Australia; SJ9694; NSW; “Tetraploid”; 2n =?. MAH; AB507910, AB507870, AB507950, AB507990, AB508033; Maharashtra, India; YI01209; TNS; “Diploid”; 2n = 20. MAR; AB507911, AB507871, AB507951, AB507991, AB508034; Chesapeake Bay, Maryland, USA; YI00958; TNS; “Diploid”; 2n =?. MAS; AB507920, AB507880, AB507960, AB508000, AB508048, AB508049; Massa, Agadir, Morocco; YI00959; TNS; “Tetraploid”; 2n =?. MIN; AB507921, AB507881, AB507961, AB508001, aAB508050, aAB508051; Minamijima, Ogasawara Isl., Tokyo, Japan; YI00223; TNS; “Tetraploid”; 2n = 40. ODC; AB507922, AB507882, AB507962, AB508002, AB508052, AB508053; Odense, Denmark; YI01304; TNS; “Tetraploid”; 2n =?. OST; AB507923, AB507883, AB507963, AB508003, AB508054, AB508055; Sweden; YI01334; TNS; “Tetraploid”; 2n =?. ROM; AB534780, AB534783, AB534790, AB534798, AB534806, AB534807; Rome, Italy; YI01382; TNS; “Tetraploid”; 2n =?. SKA; AB507924, AB507884, AB507964, AB508004, AB508056, AB508057; Shiokawa River, Okinawa, Japan; YI00754; TNS; “Tetraploid”; 2n = 40. TAK; AB507926, AB507886, AB507966, AB508006, aAB508060, aAB508061; Takahoko Lake, Aomori, Japan; YI00579; TNS; “Tetraploid”; 2n =?. TAL; AB534781, AB534788, AB534791, AB534797, AB534804, AB534805; Tallinn, Estonia; YI01379; TNS; “Tetraploid”; 2n =?. TCH; AB507912, AB507872, AB507952, AB507992, AB508035; Chigu, Tainan, Taiwan;

YI00643; TNS; “Diploid”; 2n =?. TOD; AB507913, AB507873, AB507953, AB507993, aAB508036; Todoroki River, Okinawa, Japan; YI00755; TNS; “Diploid”; 2n = 20. UTA; AB507928, AB507888, AB507968, AB508008, AB508064; Salt Lake City, Utah, USA; YI01274; TNS; “Utahian”; 2n =?. XAO; AB507927, AB507887, AB507967, AB508007, aAB508062, aAB508063; Hasan, Vladivostok, Russia; YI00933; TNS; “Tetraploid”; 2n = 40. YHO; AB507914, AB507874, AB507954, AB507994, AB508037; Yuhong, Sanya city, Hainan Province, China; YI00743; TNS; “Diploid”; 2n = 20.

R. megacarpa S. Mason—KAM; AB507929, AB507889, AB507969, AB508009, aAB508065; Kamoko Lake, Niigata, Japan; YI00173; TNS; 2n = 20. MEA; AB507930, AB507890, AB507970, AB508010, aAB508066; Small lake, Narracorte, South Australia, Australia; SJ9681; NSW; 2n = 20.

MED; AB507931, AB507891, AB507971, AB508011, AB508067; Lakes Entrance, Victoria, Australia; SJ9712; NSW, 2n =?. MEF; AB507932, AB507892, AB507972, AB508012, aAB508068; Corunna Lake, S Narooma, New South Wales, Australia; SJ9717; NSW; 2n =?.

R. polycarpa S. Mason—DEV; AB507934, AB507894, AB507974, AB508014, AB508070; Devil’s Lake, Minnesota, USA; YI01069; TNS; “Occidentalis”; 2n =?. NDA; AB507935, AB507895, AB507975, AB508015, aAB508071; Devil’s Lake, North Dakota, USA; YI00960; TNS; “Occidentalis”; 2n =?.

POB; AB507938, AB507898, AB507978, AB508018, AB508074; Coila Creek, S Moruya, New South Wales, Australia; SJ9719; NSW; 2n = 18.

POI; AB507936, AB507896, AB507976, AB508016, AB508072; Point Lake, Minnesota, USA; YI01070; TNS; “Occidentalis”; 2n =? . RED; AB507937, AB507897, AB507977, AB508017, AB508073; Redberry Lake, Saskatchewan, Canada; YI01264; TNS; “Occidentalis”; 2n = 20. TOH_01; AB534775, AB534782, AB534789, AB534796, aAB534803; Tofutsuko Lake, Hokkaido, Japan; YI01037 ; TNS; “Occidentalis”; 2n =?.

R. tuberosa Davis & Tomlinson—TUA; AB507939, AB507899, AB507979, AB508019, AB508075, AB508076; c. 3 km N Beachport, South Australia, Australia; SJ9687; NSW; 2n =?. TUE; AB507940, AB507900, AB507980, AB508020, AB508077, AB508078; Golden Beach, Lake Reeve, Victoria, Australia; SJ9706; NSW; 2n = 20.

Ruppia spp.—KUC; AB507902, AB507862, AB507942, AB507982, AB508023, AB508024; Kuccharoko Lake, Hokkaido, Japan; YI00143; TNS; Hybrid; 2n =?. OGA; AB507904, AB507864, AB507944, AB507984, AB508027, AB508028; Ogawarako Lake, Aomori, Japan; YI00824; TNS; “Triploid”; 2n = 30. TOH_02; AB507903, AB507863, AB507943, AB507983, AB508025, AB508026; Tofutsuko Lake, Hokkaido, Japan; YI00144; TNS; Hybrid; 2n =?. XAM; AB507933, AB507893, AB507973, AB508013, AB508069; Hasan, Vladivostok, Russia; YI00939; TNS; R. megacarpa; 2n =?.

Syringodium isoetifolium (Asch.) Dandy—AB507941, AB507901, AB507981, AB508021, AB508079; Tsukenshima Island, Okinawa, Japan; YI00957; TNS. Potamogeton maackianus A. Benn.—AB559938, AB506769, AB559936, AB559939, AB559939; Yae Aye Kan, Kalaw Township, Shan State, Myanmar; N. Tanaka & al. 080052; MBK, TI.