See discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/42975195

Origin of the Y genome in and its relationship to other genomes in based on evidence from elongation factor G (EF-G) gene sequences

ARTICLE in MOLECULAR PHYLOGENETICS AND EVOLUTION · APRIL 2010 Impact Factor: 3.92 · DOI: 10.1016/j.ympev.2010.03.037 · Source: PubMed

CITATIONS READS 21 36

2 AUTHORS, INCLUDING:

Genlou Sun Saint Mary's University

106 PUBLICATIONS 1,597 CITATIONS

SEE PROFILE

Available from: Genlou Sun Retrieved on: 14 December 2015 Molecular Phylogenetics and Evolution 56 (2010) 727–733

Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier.com/locate/ympev

Origin of the Y genome in Elymus and its relationship to other genomes in Triticeae based on evidence from elongation factor G (EF-G) gene sequences

Genlou Sun a,*, Takao Komatsuda b a Biology Department, Saint Mary’s University, Halifax, NS, Canada B3H 3C3 b Genome Research Unit, National Institute of Agrobiological Sciences, Kannondai, Tsukuba 305-8602, Japan article info abstract

Article history: It is well known that Elymus arose through hybridization between representatives of different genera. Received 7 December 2009 Cytogenetic analyses show that all its members include the St genome in combination with one or more Revised 18 March 2010 of four other genomes, the H, Y, P, and W genomes. The origins of the H, P, and W genomes are known, Accepted 30 March 2010 but not for the Y genome. We analyzed the single copy nuclear gene coding for elongation factor G (EF-G) Available online 2 April 2010 from 28 accessions of polyploid Elymus species and 45 accessions of diploid Triticeae species in order to investigate origin of the Y genome and its relationship to other genomes in the tribe Triticeae. Sequence Keywords: comparisons among the St, H, Y, P, W, and E genomes detected genome-specific polymorphisms at 66 Elymus nucleotide positions. The St and Y genomes are relatively dissimilar. The phylogeny of the Y genome Elongation factor G (EF-G) Genome differentiation sequences was investigated for the first time. They were most similar to the W genome sequences. Phylogeny The Y genome sequences were placed in two different groups. These two groups were included in an Triticeae unresolved clade that included the W and E sequences as well as sequences from many annual species. The H genomes sequences were in a clade with the F, P, and Ns genome sequences as sister groups. These two clades were more closely related to each other and to the L and Xp genomes than they were to the St genome sequences. These data support the hypothesis that the Y genome evolved in a diploid species and has a different origin from the St genome. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction with one or more of four other genomes, the H, Y, P, and W gen- omes (genome designations follow recommendations of the Polyploidy is increasingly recognized as an important mode of International Triticeae Consortium, see http://herbarium.usu.edu/ speciation in (Stebbins, 1950; Grant, 1981; Rieseberg, Triticeae/genmsymb.htm). Diploid species with the H, P, and W 1997; Soltis and Soltis, 2000). Polyploids can be classified into genomes are known, but there is no known diploid with the Y allopolyploids or autopolyploids based on the origin of the dupli- genome (Dewey, 1971; Jensen, 1990; Torabinejad and Mueller, cated genomes (Stebbins, 1947). The evolutionary fate of dupli- 1993). Lu and Liu (2005) and Liu et al. (2006) interpreted the cated genes in diploid organisms has been a topic of interest for ITS sequence data as suggesting that it was derived by gradual a while (Ohno, 1970; Lynch and Conery, 2000). In polyploids, every differentiation from the St genome. This hypothesis is also sup- gene is duplicated: the fate and function of these genes are even ported by Okito et al. (2009) who used RAPD based STS-PCR more intriguing. A better understanding of the processes of poly- markers in which one accession of Pseudoroegneria spicata (Pursh) ploidization and rapid diversification of the descendants of a single A. Löve was found to have markers closely related to those of the polyploidization event is therefore of widespread evolutionary StY Elymus longearistatus (Boiss.) Tzvelev. Other studies (Mason- interest (Wendel, 2000; Soltis et al., 2003). Gamer et al., 2005; Sun et al., 2008), using sequence data from Elymus L., as delimited by Löve (1984), is a large and exclu- single copy nuclear b-amylase gene and DNA polymerase subunit sively polyploid in the tribe Triticeae. Cytogenetic analyses II (RPB2) sequence support the Dewey’s hypothesis that the Y show that all its members include the St genome in combination genome had an independent origin from a Y diploid species, all of which are now extinct or undiscovered. The purpose of our study was to try and clarify the origin of the Y genome by exam- ining differentiation of nucleotide sequences (elongation factor G) closely linked to the vrs1 locus (Komatsuda et al., 1996, 1999)in * Corresponding author. Fax: +1 902 496 8104. wider range of species, both diploid and polyploid, than has been E-mail address: [email protected] (G. Sun). included in previous studies.

1055-7903/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2010.03.037 728 G. Sun, T. Komatsuda / Molecular Phylogenetics and Evolution 56 (2010) 727–733

2. Materials and methods that were confirmed as containing the insert, the remaining 8 lL of solution was transferred to 5 mL LB broth (with antibiotics) 2.1. Plant materials and incubated at 37 °C overnight. Plasmid DNA was isolated using the GenElute™ Plasmid Mini- We sequenced 28 polyploid Elymus accessions (26 species) and prep Kit (Sigma) according to manufacture’s instruction. DNA 12 accessions (9 species) of diploid Triticeae species from five gen- was sequenced commercially at MACROGEN (Seoul, Korea). To in- era, Pseudoroegneria (St genome), (H), (P), Aus- crease quality of the data, both forward and reverse strands were tralopyrum (W), and Thinopyrum (Eb)(Table 1). In addition, we sequenced independently. retrieved sequences for 33 accessions from GenBank. Bromus ster- For tetraploid species, the amplified products appeared as one ilis and B. arvensis were used as the outgroup species. band on 1.5% agarose gel. Ten clones were sequenced in order to recover at least one copy of each of the two ancestral allelic types (St-genome like, H-genome like, or Y-genome like type). Assuming 2.2. Molecular methods no bias in cloning or PCR amplification, this gives a 99.9% chance of amplifying at least one copy of each of the two ancestral allelic DNA was extracted from young leaf tissue collected from 5 to 10 types for the allotetraploid (Jakobsson et al., 2006). plants of each accession using the method of Junghans and Metz- laff (1990). The EF-G sequences were amplified using polymerase chain reaction (PCR) with the primers of cMWG699T3-2 and 2.3. Data analysis cMWG699T7-2 (Komatsuda et al., 1999). The PCR products were cloned into the TOPO-TA kit (Invitrogen, Carlsbad, CA) according Automated sequence outputs were visually inspected using to the manufacturer’s protocol. chromatographs. Since EF-G sequences for most diploid Triticeae Ten colonies were randomly selected for screening. Each was species are available in the GenBank, we downloaded these se- transferred to 10 lL of LB broth with 0.1 mg/ml antibiotics. These quences and further constructed the phylogeny for all the diploid solutions were incubated at room temperature for 20 min before species and the Y genomic sequences. using 2 lL for PCR to check for the presence of an insert using Total of 93 sequences were aligned using CLUSTAL X (Thomp- the cMWG699T3-2/cMWG699T7-2 primers. For those solutions son et al., 1997). Sequences in allopolyploids are usually not widely

Table 1 Taxa from Bromus (B), Elymus (E), Hordeum (H), Pseudoroegneria (P), Thinopyrum (T), Agropyron (Ag), and Australopyrum (Aust) used in this study.

Species Accession No. Genomea Origin GenBank Accession No. Bromus sterilis 55777 AY836187 B. arvensis 55764 AY836186 Ag. cristatum (L.) Gaertn. PI 383534 P Kars, GU982325 Aust. retrofractum (Vickery) Á. Löve PI 533014 W New South Wales, Australia GU982345 PI 547363 W New South Wales, Australia GU982347 H. bogdanii Wilensky PI 499498 H Inner Mongolia, China GU982334 PI 499645 H Xinjiang, China GU982335 P. ferganensis (Nevski) Á. Löve H10248 St Gissar mtns, Tadzhikistan GU982369 P. gracillima (Nevski) Á. Löve PI 420842 St Former Soviet Union GU982329 P. stipifolia (Czern. ex Nevski) Á. Löve PI 325181 St Russian Federation GU982324 P. spicata (Pursh) Á. Löve PI 610986 St Utah, United States GU982354 PI 506274 St Whitman, United States GU982338 P. tauri (Boiss. and Balansa) Á. Löve PI 401330 St Toward Ahar, GU982328 T. bessarabicum (Savul. and Rayss) Á. Löve PI 531712 Eb Estonia GU982344 E. canadensis L. PI 531569 StH GU982341 PI 564908 StH Colorado, United States GU982348 E. caninus (L.) L. H3169 StH Västmanland, Sweden GU982361, GU982362 E. confuses (Roshev.) Tzvelev PI 598463 StH Russian Federation GU982350, GU982351 E. dahuricus Turcz. ex Griseb. PI 628674 StHY Xinjiang, China GU982357 E. dentatus (Hook. f.) Tzvelev PI 628702 StH Altay, Russian Federation GU982358 E. fibrosis (Schrenk) Tzvelev H10339 StH Pelkosniemi, Finland GU982370, GU982371 E. gayanus E. Desv. PI 636675 StH Argentina GU982359, GU982360 E. glaucus Buckley PI 232258 StH United States GU982320, GU982321 E. hystrix L. H5495 StH Canada GU982365, GU982366 E. lanceolatus (Scribn. and J.G. Sm.) Gould PI 236663 StH Maryland, United States GU982322, GU982323 E. multisetus (J. G. Sm.) Burtt Davy W6-20963 StH California, United States GU982316, GU982317 E. mutabilis (Drobow) Tzvelev H10340 StH Pelkosniemi, Finland GU982372, GU982373 E. sibiricus L. PI 499459 StH China GU982332, GU982333 E. trachycaulus (Link) Gould ex Shinners H3526 StH Nerungri, Russia GU982363, GU982364 PI 537323 StH Utah, United States GU982346 E. virescens Piper H10584 StH Julianehåb, Greenland GU982374, GU982375 E. wawawaiensis ined. PI 506262 StH Washington, United States GU982336, GU982337 E. abolinii (Drobow) Tzvelev PI 531554 StY Xinjiang, China GU982339, GU982340 E. antiquus (Nevski) Tzvelev PI 619528 StY Sichuan, China GU982355, GU982356 E. caucasicus (Koch) Tzvelev PI 531573 StY Estonia GU982342, GU982343 E. fedtschenkoi Tzvelev PI 564927 StY Alma Ata, Siberia GU982349 E. gmelinii (Ledeb.) Tzvelev PI 610898 StY Xinjiang, China GU982352, GU982353 E. longearistatus (Boiss.) Tzvelev PI 401280 StY North of Tehran, Iran GU982326, GU982327 E. pendulinus (Nevski) Tzvelev H8986 StY Tibet, China GU982367 E. praeruptus Tzvelev H10218 StY Gissar Mt., Tadzhikistan GU982368 E. semicostatus (Nees ex Steud.) Melderis PI 207452 StY GU982318, GU982319 E. strictus (Keng) Á. Löve PI 499476 StY Lanzhou, China GU982330, GU982331

a The genome designations are according to Wang et al. (1994). G. Sun, T. Komatsuda / Molecular Phylogenetics and Evolution 56 (2010) 727–733 729 different because the parental species often will be closely related majority rule consensus tree generated from the remaining sam- (Petersen and Seberg, 2005). Due to a high degree of similarity of ples trees. sequences in polyploids, during PCR amplification, PCR mediated recombination could occur and yield chimeric products (Saiki 3. Results et al., 1988; Cronn et al., 2002). Recombinants can easily be identi- fied as chimeras in the alignments. Alignments were inspected for 3.1. Sequence variation chimeric sequences, and no recombinants were detected. The se- quence alignment was generally straightforward in spite of numer- The amplified patterns from each species display a single band ous small insertions and deletions (1–5 bp) in the sequences which for the EF-G sequence with size of approximately 900 bp which cor- were excluded in the phylogenetic analysis, but included in the cal- responds well to the size amplified in Hordeum (Komatsuda et al., culation of genome-specific polymorphism. 1999). Theoretically, allotetraploid species should possess two cop- Dissimilarity index [DI] was calculated for each pair of genome. ies of the sequences, representing one for each genome. After multi- DI = (number of genome specific polymorphic nucleotide between ple sequence alignment and phylogenetic analysis, we successfully any pair of genome)/(total number of genome specific polymorphic recovered two distinct copies of the EF-G sequences from 19 of the nucleotide found among the compared genomes). 28 accessions of tetraploid species. A total of 61 distinct copies of se- All characters were specified as unweighted and unordered. quences from 41 accessions of Triticeae species were used for fur- Phylogenetic analysis using the maximum parsimony (MP) method ther analysis. The sequences of PI 610986 (P. spicata), one copy of was performed with the computer program PAUP* version 4 beta the sequences in E. hystrix (H5495) and the sequence AY836214 10 Win (Swofford, 2003). The most parsimonious trees were ob- (P. spicata, retrieved from GenBank) were similar to other sequences tained by a heuristic search using the Tree Bisection-Reconnection in 250 bp, but not in the remaining 650 bp. (TBR) option with MulTrees on, and ten replications of random Based on the phylogenetic analysis (see below) and comparison addition sequence with the stepwise addition option. Multiple par- with sequences from diploid species, the two sequence types were simonious trees were used to form a strict consensus tree. Overall recovered from each tetraploid species, identified as the St copy character congruence was estimated by the consistency index (CI) and the H copy for StH species, and as the St copy and the Y copy and rescaled consistency index (RC). In order to infer the robust- for StY species. Examination of sequence data for all genomes repre- ness of clades, bootstrap values with 1000 replications (Felsen- sented in the study revealed genome specific nucleotide at 66 posi- stein, 1985) were calculated by performing a heuristic search tions (Table 2), 48 of which involved a single base substitution, 3 using the TBR option with MulTrees on. Parsimony methods try with single base substitution/indel, 3 with two base pairs substitu- to minimize the number of substitutions, irrespective of the branch tion (T/A/C, A/C/G, and T/A/G), and 12 with indel. The dissimilarity lengths on the tree. matrix (Table 3) for the genomes based on these 66 positions showed Bayesian analyses were performed using MrBayes 3.1 (Ronquist that St and H were most dissimilar (dissimilarity index [DI] = 87.9%). and Huelsenbeck, 2005). Default uniform priors were used for all The Y genome was most unlike the St genome (DI = 81.8%). No model parameters (six substitution rates, four base frequencies, genome specific nucleotide polymorphism was found to distinguish proportion of invariable sites, and alpha value of gamma distribu- the Y genome from the W genome, but nucleotide variation exists tion). One cold and three incrementally heated Markov Chains between the Y genome and the W genome in other nucleotide Monte Carlo (mcmc) chains were run simultaneously each for sites. The Y is also very similar to the E genome (DI = 1.5%). 2300,000 generations with default heating value (0.2). Samples were taken every 1000 generations under the GTR model with gamma-distributed rate variation across sites and a proportion of 3.2. Phylogenetic analysis of the EF-G sequences invariable sites. The first 25% of the samples from each run were discarded as burn into ensure the stationarity of the chains. Bayes- We conducted phylogenetic analyses of all the sequence data to ian posterior probability (PP) values were obtained from a 95% determine the phylogenetic relationships of the different genomes.

Table 2 Genome specific nucleotide sequence polymorphisms detected among the different genomes.

Genome 1 2 345678910111213141516171819202122232425262728 St GTGTAACCGGTTGCGTCTAATATCTGAA H ACTCCTCTACACAGC–––CGACCTCATG Y ACT–CATCACATGGG–––AAACCTCATG P ACTCCTCCACCTGGG–––GAACCTCATG W ACT–CATCACATGGG–––AAACCTCATG E ACT–CATCACATGGG–––AAACCTCATG 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 St TTGAGTACATCAGTCATC–––C–ATTTT H CGAGCCGT CGGT CGGGCT ACAT T AT T CT Y TGAGCCGTCGGTCGCACC–––T–––––– P TGAGCCGTCGGTCGCATCACAT–ATCCT W TGAGCCGTCGGTCGCACC–––T–––––– E TGAGCTGTCGGTCGCACC–––T––––––

57 58 59 60 61 62 63 63 64 65 St CTCTTGTTGA H CTCGGTCAAG Y –––GGTCGAG P CTCGGTCGAG W –––GGTCGAG E –––GGTCGAG 730 G. Sun, T. Komatsuda / Molecular Phylogenetics and Evolution 56 (2010) 727–733

Table 3 Pairwise comparison of dissimilarity index (DI) based on the genome specific polymorphic nucleotide sequences among the genomes.

Genome St H Y P W St – H 58/66 (87.9%) – Y 54/66 (81.8%) 26/66 (39.4%) – P 47/66 (71.2%) 15/66 (22.7%) 17/66 (25.8%) – W 54/66 (81.8%) 26/66 (39.4%) 0/66 (0.0%) 17/66 (25.8%) – E 54/66 (81.8%) 26/66 (39.4%) 1/66 (1.5%) 18/66 (27.3%) 1/66 (1.5%)

Maximum parsimony analysis was conducted using Bromus sterilis ing a high degree of similarity among the St genome in and B. arvensis as outgroups. The data matrix contained 867 char- Pseudoroegneria species and the St genome in Elymus species. The acters, of which 434 were constant, 113 variable characters were tree suggests at least two phylogenetically distinct St genome do- parsimony uninformative, and 320 characters were parsimony nors to the StY species including P. tauri and P. spicata. The St gen- informative. The parsimony analysis resulted in 638 equally parsi- ome in E. longearistatus (PI 401280) and E. caucasicus (PI 531573) monious trees (CI excluding uninformative characters = 0.812; originated from P. tauri, while the St genome in E. gmelinii (PI RI = 0.952). The separate Bayesian analyses resulted in identical 610898), E. strictus (PI 499476), E. antiquus (PI 619528), E. praerup- trees (mean log-likelihood values À5899.61 and À5903.40) (data tus (H10218) and E. abolinii (PI 531554) more likely originated not shown). Except for minor differences, relationships among from P. spicata. Separation of StY genome species was not unex- the main clades were topologically identical in MP and Bayesian pected, and agrees with the phylogenetic analysis of StY species consensus trees (data not shown). Bayesian PP and maximum par- using ITS data that also divided StY genomes species into two simony bootstrap (1000 replicates) value are included on the MP clades (Liu et al., 2006). The St genome in StY species are higher strict consensus tree (Fig. 1). differentiated from each other than the St genome in StH species, The phylogenetic tree contains three large clades. One includes suggesting that StY species originated from different St gene pools. all the St containing taxa, the other all the H containing taxa plus The St sequence in E. hystrix (H5495) is phylogenetically indistin- one Y containing taxon (E. fedtschenkoi, PI 564927), and the third guishable from the two accessions of P. spicata (PI 610986, all but one of the Y containing taxa plus diploids in which other AY836214), indicating that the St genome in the E. hystrix might genomes are present, notably the E and W genomes plus most of originate from P. spicata. The phylogenetic tree clearly shows St the annual species in the study. The F, P, and Ns taxa were sister genome differentiation in Elymus species, which is in agreement to the H clade. Relationships within the three large clades were lar- with previous studies (Mason-Gamer and Kellogg, 2000; Mason- gely unresolved. Gamer et al., 2005; Sun et al., 2008). The Y copy sequence from E. fedtschenkoi (PI 564927) is sister to the H genome sequences. The Y genome taxa fall into two distinct 4.2. Relationship of the Y genome in Elymus with other species in clades. The first contained E. semicostatus, E. strictus, E. abolinii, E. Triticeae antiquus, E. pendulinus, and E. gmelinii with 100% bootstrap support (PP = 1.0). The second group comprised E. longearistatus and E. cau- Our single copy nuclear gene EF-G sequence data provided casicus with 100% bootstrap support (PP = 1.0). The St genome se- additional support for the independent origin of the Y genome quences of all Elymus species with Pseudoroegneria species in polyploid StY species. The high level of differentiation between formed a clade with 95% bootstrap support (PP = 0.99) which was the Y and St (81.8%) was detected. No genome specific nucleotide distinct from the Y and H genomic groups. Within this group, the polymorphism was observed between the Y genome and W gen- St sequences from all StH species examined here and some StY ome sequences. Our results indicated a close relationship be- species formed a weakly support subclade (54%), within which tween the Y and W genomes, and very low homology between the St genome sequences E. hystrix (H5495) formed a subclade the St and Y genomes. Phylogenetic analysis well separated the with two P. spicata accessions with 100% bootstrap value Y genome from the St genome. These results are in agreement (PP = 1.0). While the E. longearistatus (PI 401280), E. caucasicus (PI with our published RPB2 data (Sun et al., 2008), and with results 531573), and P. tauri (PI 401330) formed a separate subclade with based on b-amylase gene and starch synthase sequences (Mason- 100% bootstrap (PP = 0.99) support. Gamer et al., 2005). The Y genomic sequences formed a clade with sequences from 4. Discussion the W genome (Australopyrum), the E genome (Thinopyrum) and several annual Triticeae species including Aegilops, Triticum mono- 4.1. Differentiation of St genome in polyploid Elymus species coccum, Crithopsis, and Taeniatherum, suggesting that the Y genome may be related to those genomes. Chromosome pairing suggested The St genome present in all species of Elymus sensu lato, and is that the W genome has very low homoeology with the St and Y an important genome for this genus. The St genome donor genus, genome in Elymus (Torabinejad and Mueller, 1993). However, the Pseudoroegneria, contains approximately 15 diploid (StSt) or tetra- RPB2 data suggested that the W genome was sister to the Y gen- ploid (StStStSt,orSt1St1St2St2 or StStPP) species distributed in the ome (Sun et al., 2008). The EF-G data also show a close relationship Middle East, central , northern China and western North Amer- between the Y genome and W genome. More sequence data is ica (Dewey, 1974, 1982, 1984; Löve, 1984; Wang et al., 1986; Jen- needed as our study is the first report to reveal the relationship sen et al., 1992). Cytological data suggested that genome of the Y genome with other genomes in Triticeae. differentiation exists among the Pseudoroegneria species (Wang Genome analysis has demonstrated that E. fedtschenkoi is a StY et al., 1986). Mason-Gamer and her colleagues (Mason-Gamer genomic species. RPB2 sequence data also support StY genomic and Kellogg, 2000; Mason-Gamer et al., 2002) have demonstrated constitution for this species (Sun et al., 2008). Surprisingly, the se- that Pseudoroegneria may be paraphyletic. Phylogenetic analysis quence of the E. fedtschenkoi did not belong to the Y genome clade of EF-G sequences grouped the St copy of Elymus with Pseudoroe- (Fig. 1) and instead appears as a sister to the clade containing the H gneria species together with high bootstrap support (95%), suggest- genome from StH species. However, the RPB2 data from the same G. Sun, T. Komatsuda / Molecular Phylogenetics and Evolution 56 (2010) 727–733 731

W6-20963 Elymus multisetus (StH) PI 506262 Elymus wawawaiensis (StH) PI 236663 Elymus lanceolatus (StH) H10584 Elymus virescens (StH) H3526 Elymus trachycaulus (StH) H10339 Elymus fibrosis (StH) 61 H10340 Elymus mutabilis (StH) 0.96 PI 598463 Elymus confuses (StH) PI 636675 Elymus gayanus (StH) H3169 Elymus caninus (StH) PI 499459 Elymus sibiricus (StH) 54 PI 232258 Elymus glaucus (StH) 72 PI 420842 Pseudoroegneria gracillima (St) 1.00 PI 325181 Pseudoroegneria stipifolia (St) PI 506274 Pseudoroegneria spicata (St) PI 610898 Elymus gmelinii (StY) St 61 PI 499476 Elymus strictus (StY) 0.99 PI 619528 Elymus antiquus (StY) 95 H10218 Elymus praeruptus (StY) PI 531554 Elymus abolinii (StY) PI 207452 Elymus semicostatus (StY) 0.99 H10248 Pseudoroegneria ferganensis (St) 99 PI 610986 Pseudoroegneria spicata (St) 100 0.98 AY836214 Pseudoroegneria spicata (St) 1.0 H5495 Elymus hystrix (StH) 100 PI401330 Pseudoroegneria tauri (St) 72 99 PI 401280 Elymus longearistatus (StY) 0.99 0.99 PI 531573 Elymus caucasicus (StY) AY836199 Henrardia persica (O) W6-20963 Elymus multisetus (StH) AY458787 Hordeum brachyantherum subsp. brachyantherum (H) H3526 Elymus trachycaulus (StH) PI 499459 Elymus sibiricus (StH) H3169 Elymus caninus (StH) PI 236663 Elymus lanceolatus (StH) 76 PI 598463 Elymus confuses (StH) H10339 Elymus fibrosis (StH) 0.99 H10340 Elymus mutabilis (StH) H H10584 Elymus virescens (StH) 85 PI 506262 Elymus wawawaiensis (StH) 62 PI 499498 Hordeum bogdanii (H) 0.95 0.99 PI 499645 Hordeum bogdanii (H) PI 232258 Elymus glaucus (StH) 61 PI 628702 Elymus dentatus (StH) 0.92 H5495 Elymus hystrix (StH) 62 PI 531569 Elymus canadensis (StH) 100 0.93 PI 564908 Elymus canadensis (StH) PI 636675 Elymus gayanus (StH) 1.0 78 PI 537323 Elymus trachycaulus (StH) 0.98 PI 628674 Elymus dahuricus (StHY) AY836202 Hordeum erectifolium (H) 100 AY458786 Hordeum brachyantherum subsp. californicum (H) 76 AY028800 Hordeum intercedens (H) 1.0 PI 564927 Elymus fedtschenkoi (StY) 0.86 70 AY836196 distans (F) 100 0.99 AY836197 Eremopyrum triticeum (F) 0.99 99 AY836188 Agropyron cristatum (P) 1.0 PI 383534 Agropyron cristatum (P) 98 AY836211 Psathyrostachys fragilis subsp. villosus (Ns) 62 0.99 AY836210 Psathyrostachys fragilis subsp. fragilis (Ns) 0.93 AY836212 Psathyrostachys stoloniformis (Ns) Y PI 207452 Elymus semicostatus (StY) PI 499476 Elymus strictus (StY) 100 PI 531554 Elymus abolinii (StY) 87 0.99 PI 619528 Elymus antiquus (StY) 0.86 PI 610898 Elymus gmelinii (StY) H8986 Elymus pendulinus (StY) 100 PI 401280 Elymus longearistatus (StY) PI 531573 Elymus caucasicus (StY) 0.99 98 PI 533014 Australopyrum retrofractum (W) PI 547363 Australopyrum retrofractum (W) 95 1.0 AY836190 Australopyrum pectinatum (W) 1.0 AY836192 Australopyrum velutinum (W) 97 DQ247850 Aegilops comosa (M) 1.0 DQ247854 Aegilops uniaristata (N) 82 57 AY836189 Aegilops mutica (T) 98 59 AY836207 Aegilops tauschii (D) 0.86 63 0.99 AY836215 Aegilops speltoides (S) 1.0 0.87 0.97 DQ247852 Triticum urartu (Au) 88 DQ247856 Aegilops longissimum (SI) 52 I 92 DQ247857 Aegilops sharonensis (S ) DQ247848 Aegilops bicorn (Sb) DQ247858 Aegilops searsii (Ss) AY836194 Triticum monococcum (A) 98 AY836200 Heteranthelium piliferum (Q) 1.0 AY836209 Crithopsis delileana (K) AY836216 Taeniatherum caput-medusae (Ta) DQ247855 Aegilops markgrafii (C) 100 PI531712 Thinopyrum bessarabicum (Eb) 1.0 AY836217 Thinopyrum bessarabicum (Eb) AY836213 Secale strictum (R) 100 AY836198 serpentini (L) AY836208 Peridictyon sanctum (Xp) 1.0 AY836186 Bromus arvensis AY836187 Bromus sterilis

Fig. 1. Strict consensus tree of 638 most parsimonious trees derived from the EF-G sequences using heuristic search with TBR branch swapping. The topologies obtained by Bayesian analysis (using GTR model) are identical except for some minor difference. Numbers on branches are bootstrap values; Bayesian posterior probabilities (PP) are shown below each branch. Bromus sterilis and B. arvensis were used as an outgroup. Consistency index = 0.812, retention index = 0.952. 732 G. Sun, T. Komatsuda / Molecular Phylogenetics and Evolution 56 (2010) 727–733

DNA sample recovered a Y sequence type for E. fedtschenkoi (Sun editor Dr. Lena Hileman and two anonymous reviewers for taking et al., 2008). Since the same DNA from this species was used in extensive time and care for providing thoughtful comments on the RPB2 analysis, special attention was paid in the laboratory proce- manuscript. Dr. D. Richardson kindly edited English of the dure to rule out the contamination. This result is particularly inter- manuscript. esting because although StH and StY species are intersterile, there are many StHY allopolyploid species in the genus Elymus. The placement of the Y copy of E. fedtschenkoi sister to the H genome References suggests that gene exchange may have occured between the H and Y genome in StHY species. An alternative explanation might Claw, B., Signe, F., Ole, S., 1997. A taxonomic revision of the genus Hystrix (Triticeae, be that E. fedtschenkoi acquired the EF-G sequence from H genome ). Nord. J. Bot. 17, 449–467. Cronn, R., Cedroni, M., Haselkorn, T., Grover, C., Wendel, J.F., 2002. PCR-mediated through introgression. This phenomenon has been reported for recombination in amplification products derived from polyploidy cotton. Theor. certain loci in E. repens (Mason-Gamer, 2008). Appl. Genet. 104, 482–489. Clearly, the St genome is phylogenetically distinct from the Y Dewey, D.R., 1971. Synthetic hybrids of Hordeum bogdanii with Elymus canadensis and Sitanion hystrix. Am. J. Bot. 58, 902–908. genome. Our data did not support the suggestion of a common ori- Dewey, D.R., 1974. Cytogenetics of Elymus sibiricus and its hybrids with Agropyron gin of the St and Y genomes, instead suggesting that the Y genome tauri, Elymus canadensis, and Agropyron caninum. Bot. Gaz. 135, 80–87. and the St genome originated from phylogenetically distinct dip- Dewey, D.R., 1982. Genomic and phylogenetic relationships among North American perennial Triticeae. In: Estes, J.R. (Ed.), Grasses and Grasslands. Oklahoma loid species (Sun et al., 2008). University Press, Norman, Oklahoma, pp. 51–88. Dewey, D.R., 1984. The genomic system of classification. A guide to intergeneric 4.3. Phylogenetic relationships of Elymus species hybridization with the perennial Triticeae. In: Gustafson, J.P. (Ed.), Gene Manipulation in Plant Improvement. 16th Stadler Genetics Symposium. Plenum Publishing Corp., New York, pp. 209–280. Genome analysis has revealed a close relationship between E. Ellenskog-Staam, P., von Bothmer, R., Anamthawat-Jónsson, K., Salomom, B., 2007. longearistatus and E. caucasicus (Lu and Salomon, 1993), and sug- Genome analysis of species in the genus Hystrix (Triticeae; Poaceae). Plant Syst. Evol. 265, 241–249. gesting that both the Y and St genomes in E. longearistatus and E. Fan, X., Zhang, H.Q., Sha, L.N., Zhang, L., Yang, R.W., Ding, C.B., Zhou, Y.H., 2007. caucasicus diverged from the other Elymus StY tetraploids, espe- Phylogenetic analysis among Hystrix, Leymus and its affinitive genera (Poaceae, cially those from eastern Asia (Lu and von Bothmer, 1993). A Triticeae) based on the sequences of a gene encoding plastid acetyl-CoA similar conclusion was drawn in a study of 28 Elymus species carboxylase. Plant Sci. 172, 701–707. Felsenstein, J., 1985. Confidence limits on phylogenies, an approach using the using repetitive DNA sequences (Svitashev et al., 1996). This con- bootstrap. Evolution 39, 783–791. clusion is strongly supported by our phylogenetic analysis of EF-G Grant, V., 1981. Plant Speciation. Columbia University Press, NY, USA. sequences. A close relationship among E. semicostatus, E. strictus, Jakobsson, M., Hagenblad, J., Tavaré, S., Säll, T., Halldén, C., Lind-Halldén, C., Nordborg, M., 2006. A unique recent origin of the allotetraploid species E. abolinii, E. pendulinus and E. gmelinii revealed by EF-G data was Arabidopsis suecica, evidence from nuclear DNA markers. Mol. Biol. Evol. 23, expected since they were classified into the section Goulardia by 1217–1231. Löve (1984). Elymus antiquus was placed by him in the section Jensen, K.B., 1990. Cytology, fertility, and morphology of Elymus kengii (Keng) Tzvelev and E. grandiglumis (Keng) A. Löve (Triticeae, Poaceae). Genome 33, Anthosachne. The EF-G tree grouped this species with E. semico- 563–570. status, E. strictus, E. abolinii, E. pendulinus, and E. gmelinii together Jensen, K.B., Hatch, S.L., Wipff, K., 1992. Cytology and morphology of with 100% bootstrap (Fig. 1). This grouping is in good agreement Pseudoroegneria deweyi (Poaceae, Triticeae), a new species from the foot hills of the Caucasus Mountains (Russia). Can. J. Bot. 70, 900–909. with the genome analysis that revealed a close genomic affinity Junghans, H., Metzlaff, M., 1990. A simple and rapid method for the preparation of of E. antiquus with E. semicostatus and E. abolinii (Lu and von total plant DNA. Biotechniques 8, 176. Bothmer, 1990; Lu and Salomon, 1993). Komatsuda, T., Li, W., Tanno, K., Nakamura, I., Takaiwa, F., Oka, S., 1996. Evidence of a recombination between m locus and PCR-based cMWG699 marker on Elymus hystrix was suggested to have a close affinity with the chromosome 2 of barley. In: Slinkard, A., Scoles, G., Rossnagel, B. (Eds.), Elymus canadensis complex that contains the basic genome of StH Proceeding V international Oat Conference and VII International Barley Genetics (Dewey, 1982; Claw et al., 1997). GISH and Southern genomic Symposium. University of Saskatchewan, Saskatoon, Canada, pp. 341–342. hybridization supported this conclusion (Ellenskog-Staam et al., Komatsuda, T., Tanno, K., Salomon, B., Bryngelsson, T., von Bothmer, R., 1999. Phylogeny in the genus Hordeum based on nucleotide sequences closely linked 2007), which was also supported by nuclear rDNA ITS and single to the vrs1 locus (row number of spikelets). Genome 42, 973–981. copy plastid ACCase data (Fan et al., 2007; Zhang et al., 2008). How- Liu, Q., Ge, S., Tang, H., Zhang, X., Zhu, G., Lu, B.R., 2006. Phylogenetic relationships ever, analysis of genome-specific repetitive DNA and RAPD mark- in Elymus (Poaceae, Triticeae) based on the nuclear ribosomal transcribed spacer and chloroplast trnL-F sequences. New Phytol. 170, 411–420. ers suggested that the St genome is either not present in E. Löve, Á., 1984. Conspectus of the Triticeae. Feddes Repert 95, 425–521. hystrix or that the St genome has been considerably modified dur- Lu, B.R., Bothmer, von R., 1990. Genomic constitution of Elymus parviglumis and E. ing the evolution of the species (Svitashev et al., 1996). Our EF-G pseudonutans, Triticeae (Poaceae). Hereditas 113, 109–119. Lu, B.R., Bothmer, von R., 1993. Meiotic analysis of Elymus caucasicus, E. data support the StH constitution in E. hystrix, but suggested that longearistatus, and their interspecific hybrids with twenty-three Elymus the St genome originated from P. spicata, while other species of Ely- species (Triticeae, Poaceae). Plant Syst. Evol. 185, 35–53. mus may have originated from different populations or possibly a Lu, B.R., Salomon, B., 1993. Genomic relationships among species of the Elymus parviglumis, E. semicostatus, and E. tibeticus groups (Triticeae, Poaceae). Genetica variant of P. spicata. 90, 47–60. Discordance among gene trees for diploid species in the Triti- Lu, B.R., Liu, Q., 2005. The possible origin of the ‘‘StY” – genome Elymus, a new ceae, and a widespread pattern of reticulation among these species mechanism of allopolyploidy in plants. Czech. J. Genet. Plant Breed. 41, 58. Lynch, M., Conery, J.S., 2000. The evolutionary fate of duplicated genes. Science 290, have been reported (e.g., Petersen and Seberg, 1997; Mason-Gamer 1151–1154. and Kellogg, 2000). The focus of this study is not on the evolution Mason-Gamer, R.J., Kellogg, E.A., 2000. Phylogenetic analysis of the Triticeae using of diploid species in Triticeae, thus the relationships among diploid the starch synthase gene, and a preliminary analysis of some North American Triticeae were not addressed here. Elymus species. In: Jacobs, S.W.L., Everett, J. (Eds.), Grasses, Systematics and Evolution. CSIRO, Melbourne, pp. 102–109. Mason-Gamer, R.J., Orme, N.L., Anderson, C.M., 2002. Phylogenetic analysis of North Acknowledgments American Elymus and the monogenomic Triticeae (Poaceae) using three chloroplast DNA data sets. Genome 45, 991–1002. Mason-Gamer, R.J., Burns, M.M., Naum, M., 2005. Polyploidy, introgression, and This research was supported with grants from NSERC (238425), complex phylogenetic patterns within Elymus. Czech. J. Genet. Plant Breed. 41, the Canadian Foundation for Innovation, a Senate Research Grant 21–26. at Saint Mary’s University, and the Japan Society for the Promotion Mason-Gamer, R.J., 2008. Allohexaploidy, introgression, and the complex phylogenetic history of Elymus repens (Poaceae). Mol. Phylogenet. Evol. 47, of Science (JSPS) to G.S. Thanks go to the Regional Plant Introduc- 598–611. tion Station, USDA for kindly supplying the seeds, and to Associate Ohno, S., 1970. Evolution by Gene Duplication. Springer-Verlag, NY, USA. G. Sun, T. Komatsuda / Molecular Phylogenetics and Evolution 56 (2010) 727–733 733

Okito, P., Mott, I.W., Wu, Y., Wang, R.R.C., 2009. A Y genome specific STS marker in Sun, G.L., Ni, Y., Daley, T., 2008. Molecular phylogeny of RPB2 gene reveals multiple Pseudoroegneria and Elymus species (Triticeae, Gramineae). Genome 52, 391– origin, geographic differentiation of H genome, and the relationship of the Y 400. genome to other genomes in Elymus species. Mol. Phylogenet. Evol. 46, 897– Petersen, G., Seberg, O., 1997. Phylogenetic analyses of the Triticeae (Poaceae) based 907. on rpoA sequence data. Mol. Phylogenet. Evol. 7, 217–230. Svitashev, S., Salomon, B., Bryngelsson, T., von Bothmer, R., 1996. A study of 28 Petersen, G., Seberg, O., 2005. Phylogenetic analysis of allopolyploid species. Czech. Elymus species using repetitive DNA sequences. Genome 39, 1093–1101. J. Genet. Plant Breed. 41, 28–37. Swofford, D.L., 2003. PAUP Beta Version 4.0b2. Phylogenetic Analysis using Rieseberg, L.H., 1997. Hybrid origins of plant species. Ann. Rev. Ecol. Syst. 28, 359– Parsimony – Macintosh Version. Sinaeur Associates, MA, Sunderland. 389. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmouguin, F., Higgins, D.G., 1997. The Ronquist, F., Huelsenbeck, J.P., 2005. Bayesian analysis of molecular evolution using CLUSTAL X windows interface, flexible strategies for multiple sequence MrBayes. In: Nielsen, R. (Ed.), Statistical Methods in Molecular Evolution. alignment aided by quality analysis tools. Nucleic Acids Res. 25, 4876–4882. Springer-Verlag, pp. 183–232. Torabinejad, J., Mueller, R.J., 1993. Genome Constitution of the Australian hexaploid Saiki, R.K., Gelfand, D.H., Stoffel, S., Scharf, S.J., Higushi, R., Horn, G.T., Mullis, K.B., grass Elymus scabrus (Poaceae, Triticeae). Genome 36, 147–151. Erlich, H.A., 1988. Primer-directed enzymatic amplification of DNA with a Wang, R.R.C., Dewey, D.R., Hsiao, C., 1986. Genomic analysis of the tetraploid termostable DNA polymerase. Science 239, 487–491. Pseudoroegneria tauri. Crop Sci. 26, 723–727. Soltis, P.S., Soltis, D.E., 2000. The role of genetics and genomic attributes in the Wang, R.R.C., von Bothmer, R., Dvorak, J., Fedak, G., Linde-Laursen, I., Muramatsu, success of polyploids. Proc. Natl. Acad. Sci. USA 97, 7051–7057. M., 1994. Genome symbols in the Triticeae. In: Wang, R.R.C., Jensen, K.B., Jaussi, Soltis, D.E., Soltis, P.S., Tate, J.A., 2003. Advances in the study of polyploidy since C. (Eds.), Proceeding of the 2nd International Triticeae Symposium. Logan Utah, plant speciation. New Phytol. 161, 173–191. USA, pp. 29–34. Stebbins, G.L., 1947. Types of polyploidy, their classification and significance. Adv. Wendel, J.F., 2000. Genome evolution in polyploids. Plant Mol. Biol. 42, 225–249. Genet. 1, 403–429. Zhang, H.Q., Fan, X., Sha, L.N., Zhang, C., Yang, R.W., Zhou, Y.H., 2008. Phylogeny of Stebbins, G.L., 1950. Variation and Evolution in Plants. Columbia University Press, Hystrix and related genera (Poaceae, Triticeae) based on nuclear rDNA ITS NY, USA. sequences. Plant Biol. 10, 635–642.