Skip to main content
NCBI home page
Annals of Botany logoLink to Annals of Botany
. 2016 Dec 30;119(1):95–107. doi: 10.1093/aob/mcw222

Genome origin, historical hybridization and genetic differentiation in Anthosachne australasica (Triticeae; Poaceae), inferred from chloroplast rbcL, trnH-psbA and nuclear Acc1 gene sequences

Li-Na Sha 1,2,†, Xing Fan 1,†, Xiao-Li Wang 3, Zhen-Zhen Dong 1, Jian Zeng 4, Hai-Qin Zhang 1, Hou-Yang Kang 1, Yi Wang 1, Jin-Qiu Liao 3, Yong-Hong Zhou 1,2,*
PMCID: PMC5218373  PMID: 28040673

Abstract

Background and Aims Anthosachne Steudel is a group of allopolyploid species that was derived from hexaploidization between the Asian StY genome Roegneria entity and the Australasia W genome Australopyrum species. Polyploidization and apomixis contribute to taxonomic complexity in Anthosachne. Here, a study is presented on the phylogeny and evolutionary history of Anthosachne australasica. The aims are to demonstrate the process of polyploidization events and to explore the differentiation patterns of the St genome following geographic isolation.

Methods Chloroplast rbcL and trnH-psbA and nuclear Acc1 gene sequences of 60 Anthosachne taxa and nine Roegneria species were analysed with those of 33 diploid taxa representing 20 basic genomes in Triticeae. The phylogenetic relationships were reconstructed. A time-calibrated phylogeny was generated to estimate the evolutionary history of A. australasica. Nucleotide diversity patterns were used to assess the divergence within A. australasica and between Anthosachne and its putative progenitors.

Key Results Three homoeologous copies of the Acc1 sequences from Anthosachne were grouped with the Acc1 sequences from Roegneria, Pseudoroegneria, Australopyrum, Dasypyrum and Peridictyon. The chloroplast sequences of Anthosachne were clustered with those from Roegneria and Pseudoroegneria. Divergence time for Anthosachne was dated to 4·66 million years ago (MYA). The level of nucleotide diversity in Australasian Anthosachne was higher than that in continental Roegneria. A low level of genetic differentiation within the A. australasica complex was found.

Conclusions Anthosachne originated from historical hybridization between Australopyrum species and a Roegneria entity colonized from Asia to Australasia via South-east Asia during the late Miocene. The St lineage served as the maternal donor during the speciation of Anthosachne. A contrasting pattern of population genetic structure exists in the A. australasica complex. Greater diversity in island Anthosachne compared with continental Roegneria might be associated with mutation, polyploidization, apomixis and expansion. It is reasonable to consider that A. australasica var. scabra and A. australasica var. plurinervisa should be included in the A. australasica complex.

Keywords: Phylogeny, genetic differentiation, genome origin, Anthosachne, apomixis

INTRODUCTION

Polyploidy, resulting from either intraspecific genome duplication (autopolyploidy) or combination of two or more distinct genomes (allopolyploidy), is generally viewed as an important driver of plant evolution and the clearest route to sympatric speciation in natural populations (Adams and Wendel, 2005). Polyploidy promotes variability through change in the chromosomal number per se, increased genetic diversity and genomic restructuring, potentially resulting in beneficial new phenotypes and evolutionary innovation in physiological and ecological flexibility (Baack and Rieseberg, 2007; Ramsey and Ramsey, 2014). Moreover, unreduced female gametophytes following polyploidy can be produced through meiotic abnormality in the megasporocyte (diplospory), allowing an evolutionary transition from sexual to asexual reproduction via apomixis (Ozias-Akins and van Dijk, 2007; Cosendai et al., 2013). Apomixis is typically associated with polyploidy, since many apomictic taxa are of allopolyploid origin (Sochor et al., 2015). Recent reviews focused on the role of polyploidy in divergence and speciation within sexually reproducing plants (Rieseberg and Willis 2007; Hegarty and Hiscock 2008; Soltis, 2013), yet these processes in apomictic polyploids have been less well studied (Hörandl and Paun 2007; Robertson et al., 2010). In addition, relatively few attempts have been made to evaluate and compare genetic differentiation between apomictic and sexual relatives, in which life history may impact on the pattern and level of genetic variation.

Anthosachne Steudel, a polyploid perennial genus in the wheat tribe (Poaceae: Triticeae), includes three species native to Australia: Anthosachne australasica Steudel, Anthosachne aprica (Löve et Connor) C. Yen et J. L. Yang and Anthosachne multiflora (Banks et Solander ex Hook. f.) C. Yen et J. L. Yang. Anthosachne was distinguished as a separate genus from traditional Elymus sect. Anthosachne, with A. australasica Steudel as the type species (Yen et al., 2006). Morphologically, Anthosachne differs from most genera of the Triticeae in its tendency to have rather slender culms with drooping inflorescences and large, widely spaced, solitary spikelets (Yen et al., 2006; Barkworth and Jacobs, 2011). The habits preferred by Anthosachne species are dry alpine meadow and coastal desert. Cytologically, all the species of Anthosachne are allohexaploids (2n = 6x = 42) with StYW genomes (Torabinejad and Mueller, 1993; Yen et al., 2006). The St and W genomes are derived from Pseudoroegneria (Nevski) Á Löve and Australopyrum (Tzvelev) Á Löve, respectively (Löve, 1984; Wang et al., 1994). It is unknown where the Y genome originates, although it is a fundamental Anthosachne genome (Torabinejad and Mueller, 1993; Yen et al., 2006). Dewey (1984) considered that the Y genome has its origin in Central Asia or the Himalaya region, and may be extinct. Extensive cytogenetic and molecular studies have suggested that the St (Pseudoroegneria), W (Australopyrum), P (Agropyron Gaertn.) and Xp (Peridictyon O. Seber, S. Frederiksen & C. Baden) genomes are potential donors of the Y genome (Lu et al., 1992; Liu et al., 2006; Okito et al., 2009; Sun and Komatsuda, 2010; Dou et al., 2013; Fan et al., 2013a). Accumulated evidence has suggested that speciation of the Anthosachne polyploid was derived from hybridization between tetraploid Roegneria C. Koch (2n = 4x = 28, StY) and diploid Australopyrum (2n = 2x = 14, W) species (Torabinejad and Mueller, 1993; Lu et al., 1995; Yen et al., 2006; Fan et al., 2013a). Roegneria, a large genus in the wheat tribe, includes about 100 species distributed throughout Asia (Yen and Yang, 2011). Morphologically, Roegneria species have a single spikelet per rachis node with lanceolate glumes and short and broader palea. Cytologically, two ploidy levels, tetraploid (2n = 4x = 28) and hexaploid (2n = 6x = 42, StStY), were recognized in Roegneria species. Following the taxonomic delimitation of genomic constitutions by Löve (1984), the species with the StY genomes were included in Roegneria (Yen and Yang, 2011). Tetraploid Roegneria taxa include about 30 species that are of Asian origin and distributed in the Qinghai–Tibetan Plateau, Central Asia, East Asia, West Asia and South-east Europe (Yen and Yang, 2011). The species of Anthosachne and Australopyrum are restrictedly distributed in some regions of Australasia. Despite the suggestion that tetraploid Roegneria was considered to be the StY genome donor of Anthosachne, the processes that have driven polyploid diversification and speciation, especially with regards to which tetraploid species were involved in hexaploid evolution in island Anthosachne, remain unclear.

Anthosachne australasica, which is the only known apomictic species in the Triticeae (Dewey, 1984; Yen et al., 2006), includes three varieties, A. australasica var. scabra, A. australasica var. plurinervisa and A. australasica var. typica (Yen et al., 2006). Both A. australasica and A. australasica var. typica were previously described as Elymus rectisetus (Yen et al., 2006; Barkworth and Jacobs, 2011). Anthosachne australasica var. scabra and A. australasica var. plurinervisa were included in the Elymus scaber complex (Wang and Henwood, 1999; Yen et al., 2006). Several infraspecific phenodemes associated with obligately apomicts, predominantly apomicts and facultatively apomicts, have been recognized in A. australasica (Connor, 2005). Variation in apomictic patterns, together with polyploidization, thus contributes to taxonomic complexity in A. australasica. Several systematists (Wheeler et al., 1982; Jessop, 1986) treated A. australasica as a long-awned morphologically distinct entity, which was included in A. australasica var. scabra. Wang and Henwood (1999) and Murphy (2003) suggested that A. australasica var. plurinervisa should be elevated to species rank, and A. australasica and A. australasica var. scabra should be conspecific relatives in origin. Combining morphology, cytogenetics and distribution evidence, Yen et al. (2006) suggested that A. australasica var. scabra, A. australasica var. plurinervisa and A. australasica var. typica should be included in the rank of the species A. australasica. However, Barkworth and Jacobs (2011) advocated that A. australasica var. scabra and A. australasica var. plurinervisa should be treated as A. scabra (R.Br.) Nevski and A. plurinervis (Vickery) Barkworth & S. W. L. Jacobs, respectively. Therefore, the definitions, precise taxonomic ranks of A. australasica and its varieties, and their phylogenetic relationships are under discussion. The main problems focus on whether these taxa should be treated as independent species or clumped into the A. australasica complex. In addition, the adaptive and evolutionary potential of Anthosachne apomicts compared with sexual Roegneria taxa are less documented.

Phylogenetic analysis is routinely applied to illustrate evolutionary and taxonomic questions. Molecular phylogenetics allows the testing of biogeographic hypotheses by comparing the estimated node ages of a disjunct group against the estimated age of a historical climatic or geographic event. By combining an effect of a historically geographic event with time-calibrated phylogenies, recent studies have indicated that variation in geographic characters determined the speciation, richness and diversity of plant clades, and hence indirectly influenced plant diversification and evolution (Egan and Crandall, 2008; Särkinen et al, 2012; Sha et al., 2016). Single- or low-copy nuclear genes are less likely to be subject to concerted evolution, thus making themselves ideal tools for tracing the origin of polyploids (Kim et al., 2008), identifying the genome donors (Mason-Gamer, 2004), demonstrating the pattern of polyploidization and diversification (Brassac and Blattner, 2015), tracing a reticulate history and speciation time in polyploids (Marcussen et al., 2015), and clarifying taxonomic treatments (Fan et al., 2013a). Comparative phylogenies between single- or low-copy nuclear and chloroplast sequences have become a useful tool to investigate plant polyploid phylogeny (Mason-Gamer, 2004; Liu et al., 2006; Sha et al., 2016). Previous chloroplast DNA (cpDNA) data have indicated that despite variation in ploidy levels in Elymus L. sensu lato (s.l.; including Anthosachne), Pseudoroegneria (St genome) served as the maternal donor during the polyploid speciation of Elymus s.l. (Mason-Gamer et al., 2002; Liu et al., 2006; Dong et al., 2013), which allows phylogenetic relationships among the St-containing species to be elucidated on the basis of orthologous comparison. In this study, we sequenced and analysed the single-copy nuclear Acc1 gene and plastid rbcL gene and trnH-psbA region sequences for 60 accessions of A. australasica and their putative donors to explore the origin and relationships of the polyploid Anthosachne species. Based on the combined nuclear and chloroplast sequences of the St genome lineage, a time-calibrated phylogenetic tree was generated to estimate the evolutionary history of Anthosachne. The objectives were: (1) to reveal the phylogenetic relationships between Anthosachne and Roegneria species; (2) to identify the origin of A. australasica; (3) to investigate intraspecific relationships of A. australasica; and (4) to compare the genetic differentiation between sexual Roegneria and apomictic A. australasica.

MATERIALS AND METHODS

Plant materials

A total of 69 polyploids, comprising 60 A. australasica (StYW genomes) accessions and nine Roegneria (StY genomes) species, were analysed together with 33 diploid taxa representing 20 basic genomes in the Triticeae. Sample information (including names, genomes, accession number, origin and abbreviation) and GenBank accession data are listed in Supplementary Data Table S1. The Acc1 sequences of 27 diploid species and seven tetraploid StY species, and 29 rbcL and 28 trnH-psbA sequences of diploid species, along with the sequences from the genus Bromus L. were obtained from published data (NCBI: http://www.ncbi.nlm.nih.gov/). The seed materials with PI and TA numbers were kindly provided by the American National Plant Germplasm System (Pullman, WA, USA), while the others were collected by ourselves. The plants and voucher specimens are deposited at the Herbarium of Triticeae Research Institute, Sichuan Agricultural University, China (SAUTI).

DNA isolation, amplification and sequencing

Total DNA was extracted from fresh leaves of plants using the CTAB (cetyltrimethylammonium bromide) method (Doyle and Doyle, 1987). The chloroplast rbcL gene fragment was amplified with the primers rbcL1 and rbcL2 (Kress and Erickson, 2007). The rbcL sequence amplification was performed in a total reaction of volume of 25 μL of mixture containing 1× reaction buffer, 1·5 mm MgCl2, 1 μm of each primer, 200 μm of each dNTP [TaKaRa Biotechnology (Dalian) Co., Ltd, Liaoning, China], 20–40 ng of template DNA, 1·25U of ExTaq Polymerase (TaKaRa) and sterile water to the final volume. The PCR profile included an initial pre-melt of 4 min at 95 °C, followed by 30 cycles of 1 min denaturation (94 °C), 1 min annealing (55 °C) and 90 s extension (72 °C), followed by a 10 min final extension at 72 °C. The trnH-psbA region was amplified using the primers trnH1 and trnH2 of Shaw and Small (2005). The PCR protocol differed from that of rbcL in that it included an initial pre-melt of 2 min at 94 °C, followed by 35 cycles of 1 min denaturation (94 °C), 1 min annealing (46 °C) and 2 min elongation (72 °C), and ending with a final extension at 72 °C. The Acc1 gene was amplified by PCR using the primers AccF1 and AccF2 (Huang et al., 2002). PCR for the Acc1 gene amplification was conducted under cycling conditions reported previously (Huang et al., 2002). The PCR products for both nuclear Acc1 and chloroplast rbcL and trnH-psbA fragments were visualized on 1 % agarose gels, purified by an ENZA™ gel extraction kit (Omega, Norcross, GA, USA) and then cloned into pMD19-T vector (TaKaRa, Dalian, China) according to the manufacture’s instructions. Ten to fifteen clones per allopolyploid were picked and sequenced to obtain all the possible Acc1 sequences from the putative donor species. DNA was commercially sequenced in both directions by the Beijing Genomics Institute (Beijing, China).

Data analysis

Multiple sequences were aligned using ClustalX (Thompson et al., 1999), with manual adjustment. The sequence statistics, including nucleotide substitutions, the transition/transversion (TS/TV) ratio and variability of the sequences, were calculated by MEGA 4 (Tamura et al., 2007).

The chloroplast rbcL and trnH-psbA regions were combined into a common matrix (combined cpDNA). Two data matrices, including Acc1 and combined cpDNA data, were used separately to carry out phylogenetic analyses. Phylogenetic analyses were conducted using the maximum likelihood (ML) method with PAUP*4·0b10 (D. L. Swofford, Sinauer Associates, http://www.sinauer.com) and the Bayesian inference (BI) method with MrBayes version v3.1.2 (Huelsenbeck and Ronquist, 2001). The evolutionary model used for the phylogenetic analyses was determined using ModelTest v3.0 with the Akaike information criterion (AIC) (Posada and Crandall, 1998). The optimal models were GTR + G + I for the two data sets. In ML analysis, heuristic searches were employed [ACCTRAN; starting tree based on Neighbor–Joining reconstruction; TBR (tree bisection and reconnection) branch swapping; and STEEPEST DESCENT]. Statistical support for nodes in ML analysis was estimated by bootstrap support (BS) (Felsenstein, 1985). BI analyses were carried out under the same evolutionary model as ML analysis. Two sets of four chains were run for 20 million generations, and trees were sampled and saved every 1000th generation. A majority rule consensus tree was calculated for the two data sets after discarding the first 25 % of the trees as ‘burn-in’. The statistical confidence in nodes was evaluated by posterior probabilities (PPs). A PP-value <90 % was not included in the figures.

Phylogenetic analysis of cpDNA sequences has suggested that the Pseudoroegneria species (St genome) served as the maternal donor during the polyploid speciation of the St-containing species (StY, StH, StYH, StYP and StYW genomes) (Liu et al., 2006; Dong et al., 2013). To generate a time-calibrated tree on the basis of orthologous comparison, the St-copy sequences of the Acc1 and the sequences of chloroplast rbcL and trnH-psbA regions within Anthosachne were combined into a common matrix (combined St sequences) and analysed with its relatives in Triticeae. Clock-like evolution of combined St sequences within Anthosachne and its putative relative species was evaluated with a likelihood ratio test comparing the likelihood scores from the unconstrained and clock-constrained analyses, implemented in PAUP*4·0b10. Divergence times with 95 % confidence intervals (CIs) were estimated using the Bayesian relaxed molecular clock method, implemented in BEAST v1.4.6 (Drummond and Rambaut, 2007). The lack of fossils for the Triticeae precluded a direct calibration of tree topologies. Instead, molecular dating was based on the divergence time for the basal-most split in the Triticeae (Marcussen et al., 2014). Priors on the Triticeae crown age [15·32 ± 0·34 million years ago (MYA)] were set as inferred by Marcussen et al. (2014). Calibration points were performed using a relaxed uncorrelated lognormal molecular clock. The analysis was run using the Yule species tree prior, as well as the piecewise linear and constant root population model. Three independent analyses were computed for 10 000 000 generations each under the GTR + G + I model (with the associated parameters specified by ModelTest v3.0 as the priors), sampling the states every 1000 generations. Tracer v1.4 (Rambaut and Drummond, 2007) was used to ensure the convergence of the mixing in terms of the effective sample size (ESS) values and the coefficient rate. Appropriate burn-ins were estimated from each trace file, discarded and all analyses were combined with LogCombiner (http://beast.bio.ed.ac.uk/LogCombiner). Tree files were compiled into a maximum clade credibility tree using TreeAnnotator (http://beast.bio.ed.ac.uk/TreeAnnotator) to display mean node ages and highest posterior density (HPD) intervals at 95 % for each node. Trees were then viewed in FigTree v. 1.3.1 (http://tree.bio.ed.ac.uk/).

Nucleotide diversity was estimated as π, which is based on the average number of pairwise differences among sequences (Tajima, 1989), and as Watterson’s (1975) θw, which is based on the number of segregating sites (S) corrected for sample size. To assess the divergence and genetic relationships among Anthosachne taxa and between allohexaploids and its putative progenitors, nucleotide diversity calculated by the number of fixed differences (SF), the numbers of shared polymorphisms (SS), nucleotide substitutions per site (Dxy), net nucleotide substitutions per site (Da) and comparison in polymorphic and monomorphic sites (Spop1, mutations polymorphic in population 1 but monomorphic in population 2; Spop2, mutations polymorphic in population 2 but monomorphic in population 1) were estimated with DnaSP 4·10·9 (Rozas et al., 2005). Tests of neutrality including Tajima’s, and Fu and Li’s D statistic were performed as described by Tajima (1989) and Fu and Li (1993). Significance of D-values was estimated with the simulated distribution of random samples (1000 steps) using a coalescence algorithm assuming neutrality and population equilibrium (Hudson, 1990). The haplotypic richness was calculated using the rarefaction method, standardized for the lower sample size, implemented in Contrib 1·02 (Petit et al., 1998). Two measures of population differentiation, GST and FST, were compared by the programs DnaSP 4·10·9 (Rozas et al., 2005) and Arlequin 3·1 (Excoffier et al., 2005). GST makes use of haplotype frequencies and the FST index is a relative measure of population differentiation, with all estimates weighted by sample size.

RESULTS

Phylogenetic analyses

Acc1 data

A total of 111 Acc1 homoeologous sequences, comprising 53 St-copy, 48 W-copy and ten Y-copy types, were obtained from 60 sampled Anthosachne accessions. In theory, 180 single-copy homoeologous sequence should be isolated from 60 hexaploids. In many cases, two (St and W types) or three distinct Acc1 sequence derived from their ancestral genome donors were detected from each accession. The missing genome homoeologous types might result from a sampling artefact, or they might show the occasional loss of one copy of the gene, through either homoeologous recombination or deletion. In addition, the possibility also cannot be ruled out that missing genome types were not sufficiently picked by the screen of 10–15 positive clones. Sampling artefact is the most straightforward explanation, as it would result in a random and balanced sampling for each genome type.

The aligned Acc1 sequences yielded a total of 776 characters, of which 299 were variable characters and 178 were informative. The Acc1 gene tree constructed by the ML method (–ln likelihood = –4739·3257; base frequencies A, 0·3539; C. 0·2143; G. 0·1746; T, 0·2571; shape = 0·7934; pinvar = 0·1778) was almost identical to the Bayesian consensus tree. The tree illustrated in Fig. 1 is the phylogenetic tree with collapsed clades, and the full tree is shown in Supplementary Data Fig. S1. In the Acc1 tree, the St-copy, W-copy and Y-copy sequences from Anthosachne accessions were split into tree well supported clades (>70 % BS; >90 % PP), clade I, clade II and clade III (Fig. 1; Fig. S1). The clade I included the St-copy sequences of Anthosachne (StYW) and Roegneria (StY) and the sequences from Pseudoroegneria (St) (97 % BS and 100 % PP). Three Pseudoroegneria species (P. stipifolia, P. libanotica and P. strigosa) and Roegneria caucasica were placed outside the sub-clade including 53 Anthosachne accessions (25 A. australasica, 23 A. australasica var. scabra and five A. australasica var. plurinervisa), one Pseudoroegneria species (P. spicata) and six Roegneria species (R. grandis, R. amurensis, R. stricta, R. ciliaris, R. pendulina and R. japonensis). Clade II contained the W-copy sequences of Anthosachne and the sequence from Australopyrum retrofractum (W) (76 % BS and 97 % PP). In this clade, 24 A. australasica, 20 A. australasica var. scabra and four A. australasica var. plurinervisa formed a monophyletic sub-clade, and Australopyrum retrofractum was placed outside the sub-clade. Clade III consisted of the Y-copy sequences from ten Anthosachne accessions (two A. australasica, six A. australasica var. scabra, and two A. australasica var. plurinervisa) and six Roegneria (StY) species (98 % BS and 100 % PP).

Fig. 1.

Fig. 1.

Open in a new tab

Phylogenetic tree inferred from the Acc1 sequences of Anthosachne taxa and the sequences of its affinitive species within Triticeae, under the GTR + G + I model. The numbers above and below the branches indicate bootstrap values >50 % and Bayesian posterior probability values >90 %. The upper case letters in parentheses indicate the genome type of the species. The clade consisting of Anthosachne and its putative donor is highlighted by a collapsed clade. For the full tree, please see Fig. S1.

Combined cpDNA data

The combined rbcL and trnL-F data yielded a total of 1345 characters, of which 136 were variable characters and 42 were informative. The ML tree (–ln likelihood = –3268·9553; base frequencies A, 0·3191; C, 0·1880; G, 0·2051; T, 0·2878; shape = 0·9909; pinvar = 0·6962) inferred from the combined rbcL and trnH-psbA data was almost identical to the Bayesian consensus topology. The phylogenetic tree with collapsed clades is shown in Fig. 2 (for the full tree, see Supplementary Data Fig. S2). In the phylogenetic tree (Fig. 2; Fig. S2), all the sampled Anthosachne accessions, Roegneria species, Thinopyrum bessarabicum, Lophopyrum elongatum, Dasypyrum villosum and Pseudoroegneria spicata formed one clade with 67 % BS and 90 % PP. Dasypyrum villosum and P. spicata were at the base of the clade. In this clade, all the accessions of Anthosachne and Roegneria anthosachnoides formed one sub-clade with good statistical support (83 % BS and 98 % PP).

Fig. 2.

Fig. 2.

Open in a new tab

Phylogenetic tree of Anthosachne species and its affinitive species inferred from the combined chloroplast rbcL and trnH-psbA sequences, under the GTR + G + I model. The numbers above and below the branches indicate bootstrap values >50 % and Bayesian posterior probability values >90 %. Upper case letters in parentheses indicate the genome type of the species. The clade including Anthosachne and its putative donor is highlighted by a collapsed clade. For the full tree, please see Fig. S2.

Divergence dating

The BEAST analyses of the combined rbcL, trnH-psbA and the St-copy Acc1 sequences within Anthosachne and its putative tetraploid and diploid species generated a time-calibrated tree (Fig. 3). Under a lognormal relaxed clock, the coefficient of rate variation was estimated to be 0·804 (95 % CI 0·597–1·015), indicating that a relaxed clock was appropriate. The birth rate indicated by the Yule prior is 0·282 (95 % CI 0·215–0·360). The mean ages with 95 % CIs are indicated in the chronogram (Fig. 3). The Triticeae crown clade age (15·09 MYA, 95 % CI 12·13–18·5) fitted to our prior. The age of Hordeum was estimated to be 11·2 MYA (95 % CI 8·27–15·09), which is consistent with the previous suggestion that the split between the Hordeum and Aegilops/Triticum lineage ranged from 9 to 13 MYA (Huang et al., 2002; Marcussen et al., 2014; Middleton et al., 2014). In the time-calibrated tree, all the sampled Anthosachne taxa formed one monophyletic group (100 % PP), and all the Roegneria species and P. spicata were placed outside this monophyletic group (100 % PP). Time calibration analysis demonstrated that the time to the most recent common ancestor (tMRCA) of StYW genome Anthosachne species was dated to 4·66 MYA (95 % CI 3·34–6·68), and the divergence time of the St genome lineages was 5·55 MYA (95 % CI 3·92–7·74).

Fig. 3.

Fig. 3.

Open in a new tab

A time-calibrated tree inferred from the combined rbcL, trnH-psbA and St-copy Acc1 sequences of Anthosachne taxa and its putative donors using a Bayesian relaxed clock method in BEAST. Three Anthosachne taxa were labelled with blue, red and green, respectively. Two asterisks above the branch indicate the posterior probability values >95 %, and posterior probability values of 90–95 % are labelled with one asterisk. The monophyletic groups of Anthosachne and Roegneria re labelled with a pink and green bar, respectively.

Nucleotide polymorphism

Based on the rbcL, trnH-psbA and Acc1 data, nucleotide polymorphism including number of segregating sites (S), haplotype diversity (Hd), and measures of nucleotide diversity (π and θw) were separately calculated for Anthosachne and Roegneria (Table 1). The rbcL haplotype diversity within Anthosachne ranged from 0·417 to 0·865 and the haplotypic richness (Hr) from 1·000 to 2·958. Estimates of nucleotide diversity (π) for the rbcL data within Anthosachne ranged from 0·0005 to 0·0032, and Watterson’s θ (θw) from 0·0007 to 0·0086. The trnH-psbA haplotype diversity within Anthosachne varied from 0 to 0·738 and the haplotypic richness from 0 to 2·296. Levels of nucleotide diversity for the trnH-psbA data within Anthosachne varied from 0 to 0·0059, and θw varied from 0 to 0·0120. For the Acc1 data, all the Anthosachne population consisted of heterozygous haplotype diversity ranging from 0·9 to 0·992, and the haplotypic richness ranged from 3·000 to 3·921. Estimates of nucleotide diversity within Anthosachne ranged from 0·0063 to 0·0107 and θw from 0·0066 to 0·0162. Nucleotide polymorphism between Anthosachne and Roegneria was also compared. The number of segregating sites in Anthosachne (the rbcL data, 35; the trnH-psbA data, 36; the Acc1 data, 62) was higher than that in Roegneria (the rbcL data, 3; the trnH-psbA data, 13; the Acc1 data, 10). The Hr value in Anthosachne was 2·526 for the rbcL data, 2·012 for the trnH-psbA data and 3·895 for the Acc1 data, while the Hr value in Roegneria was 1·111 for the rbcL data, 3·778 for the trnH-psbA data and 3·524 for the Acc1 data. The estimates of nucleotide diversity in Anthosachne were π = 0·0025 and θw = 0·0102 for the rbcL data, π = 0·0041 and θw = 0·0136 for the trnH-psbA data and π = 0·0102 and θw =0·0194 for the Acc1 data, while the estimates of nucleotide diversity in Roegneria were π = 0·0009 and θw = 0·0015 for the rbcL data, π = 0·0035 and θw =0·0084 for the trnH-psbA data and π = 0·0049 and θw =0·0056 for the Acc1 data. The Tajima’s and Fu and Li’s D tests were conducted on each of the data sets (Table 1), and all tests had negative values of the neutrality statistic D. The Tajima’s and Fu and Li’s D values estimated from the rbcL data were –2·5162 (P < 0·01) and –5·5027 (P < 0·05) for Anthosachne, and –1·513 (P > 0·05) and –1·6827 (P > 0·05) for Roegneria. The same values calculated by the trnH-psbA data were –2·2993 (P < 0·01) and –5·8918 (P < 0·01) for Anthosachne, and –1·363 (P > 0·05) and –1·8633 (P < 0·05) for Roegneria. For the Acc1 data, the Tajima’s and Fu and Li’s D values were –1·6495 (P < 0·05) and –1·4813 (P < 0·05) for Anthosachne, and –0. 6707 (P > 0·05) and –1. 0532 (P > 0·05) for Roegneria.

Table 1.

Estimates of nucleotide diversity and test statistics at rbcL, trnH-psbA and Acc1 loci in Anthosachne species and its putative tetraploid genome donor

Locus Taxa n S Hd* Hr† π θw Fu and Li’s D Tajima’s D
rbcL Anthosachne
AU 743 25 0·865 2·958 0·0032 0·0086 –3·9304 –2·2547
(P < 0·05) (P < 0·01)
SA 743 9 0·684 2·087 0·0015 0·0033 –2·3995 –1·8525
(P < 0·05) (P < 0·05)
PL 743 1 0·400 1·000 0·0005 0·0007 –0·8165 –0·8165
(P > 0·05) (P > 0·05)
Total 743 35 0·773 2·526 0·0025 0·0102 –5·5027 –2·5162
(P < 0·05) (P < 0·01)
Roegneria 743 3 0·417 1·111 0·0009 0·0015 –1·6827 –1·5130
(P > 0·05) (P > 0·05)
trnH-psbA Anthosachne
AU 569 16 0·738 2·296 0·0035 0·0070 –3·0201 –1·6583
(P < 0·05) (P < 0·05)
SA 570 25 0·649 2·053 0·0059 0·0120 –3·1515 –1·9209
(P < 0·01) (P < 0·01)
PL 570 1 0 0 0 0 0 0
Total 569 36 0·656 2·012 0·0041 0·0136 –5·8918 –2·2993
(P < 0·01) (P < 0·01)
Roegneria 568 13 0·972 3·778 0·0035 0·0084 –1·8633 –1·3630
(P < 0·05) (P > 0·05)
Acc1 Anthosachne
AU 704 43 0·990 3·900 0·0107 0·0162 –1·0568 –1·2937
(P > 0·05) (P > 0·05)
SA 729 40 0·992 3·921 0·0104 0·0149 –0·7115 –1·1742
(P > 0·05) (P > 0·05)
PL 731 10 0·900 3·000 0·0063 0·0066 –0·2982 –0·2982
(P > 0·05) (P > 0·05)
Total 703 62 0·989 3·895 0·0102 0·0194 –1·4813 –1·6495
(P < 0·05) (P < 0·05)
Roegneria 730 10 0·952 3·524 0·0049 0·0056 –1·0532 –0·6707
(P > 0·05) (P > 0·05)
Open in a new tab

AU, A. australasica; SA, A. australasica var. scabra; PL, A australasica var. plurinervisa; n, the number of the sites (excluding sites with gaps/missing data); S, the number of segregating sites; π, the average pairwise diversity; θw, the diversity based on the number of segregating sites.

*

Haptotypic diversity

†

Haplotypic richness with rarefaction equal to five.

Levels of intraspecific differentiation of Anthosachne taxa

Estimates for DNA divergence among Anthosachne taxa are shown in Table 2. More polymorphic sites were observed in A. australasica (Spop1 = 77) than in A. australasica var. scabra (Spop1 = 52). Comparison between A. australasica and A. australasica var. plurinervisa and between A. australasica var. scabra and A. australasica var. plurinervisa showed more polymorphic sites in A. australasica (Spop1 = 69) and A. australasica var. scabra (Spop1 = 97) than in A. australasica var. plurinervisa (Spop2 = 2 and 3), respectively. Higher pairwise nucleotide divergence was found between A. australasica and A. australasica var. scabra or A. australasica var. plurinervisa (AU–SA, Dxy = 0·00647, Da = 0·00024; AU–PL, Dxy = 0·00447, Da = 0·00004) than between A. australasica var. scabra and A. australasica var. plurinervisa (SA–PL, Dxy = 0·00478, Da = 0·00010).

Table 2.

Divergence among Anthosachne taxa based on the combined rbcL, trnH-psbA and Acc1 data

Spop1 Spop2 Dxy Da SS SF
AU (population 1) – SA (population 2) 77 52 0·00647 (0·0007) 0·00024 (0·0002) 24 0
AU (population 1) – PL (population 2) 69 2 0·00447 (0·0009) 0·00004 (0·0001) 7 0
SA (population 1) – PL (population 2) 97 3 0·00478 (0·0011) 0·00010 (0·0012) 6 0
Open in a new tab

The numbers in parentheses represent the standard error

AU, A. australasica; SA, A. australasica var. scabra; PL, A australasica var. plurinervisa; Spop1, mutations polymorphic in population 1 but monomorphic in population 2; Spop2, mutations polymorphic in population 2 but monomorphic in population 1; Dxy, the average number of nucleotide substitution per site between the two species. Da, the number of net nucleotide substitution per site; SS.the number of polymorphic sites shared by the two populations; SF the number of fixed differences between the two populations.

Population differentiation using GST and FST among Anthosachne taxa was also estimated (Table 3). The GST and FST parameters were –0·0008 and 0·0451 for comparison between A. australasica and A. australasica var. scabra, respectively. The values were GST = 0·0552 and FST = 0·0121 for comparison between A. australasica and A. australasica var. plurinervisa. The same parameters were GST = 0·0537 and FST = –0·0206 for comparison between A. australasica var. scabra and A. australasica var. plurinervisa.

Table 3.

Pairwise population differentiation as measured by GST and FST

GST FST
AU (population 1) – SA (population 2) –0·0008 0·0451
AU (population 1) – PL (population 2) 0·0552 0·0121
SA (population 1) – PL (population 2) 0·0537 −0·0206
Open in a new tab

AU, A. australasica; SA, A. australasica var. scabra; PL, A australasica var. plurinervisa.

DISCUSSION

The ancestors during the polyploid speciation of Anthosachne

Phylogenetic analysis based on Acc1 sequences showed that the St and W genome homoeologous types were separately grouped with the sequences of Pseudoroegneria and Australopyrum with high statistical support (BS > 70 % and PP > 90 %), and combined cpDNA data also indicated that Anthosachne and Pseudoroegneria species were in one clade. This is in agreement with a previous cytogenetic suggestion that the Pseudoroegneria and Australopyrum species served as the St and W genome diploid donors during the polyploid speciation of Anthosachne species (Dewey, 1984; Torabinejad and Mueller, 1993). A single nuclear marker region, however, was unable to resolve all taxon relationships among StYW genome polyploids. The major difficulty lies in the uncertainty of intraspecific relationships due to the non-monophyly of each taxon in the Acc1 tree. In many cases, different taxa placed scatteredly into one clade and formed a paraphyletic grade with a number of zero-length branches (Clade I and II). As the Acc1 sequences derived from polyploids cluster with sequences obtained from specific diploid species in phylogenetic analyses, they enable the identification of the parental species involved in polyploid formation. The origin of the Y genome in Y-containing species is still unclear, despite decades of intensive efforts. Phylogenetic analysis of ITS (internal transcibed spacer) sequence data presented by Liu et al. (2006) suggested that the Y genome was probably a derivative of the St genome. Analysis of Y-containing Elymus species based on RPB2 (Sun et al., 2008) and EF-G (Sun and Komatsuda, 2010) showed that the Y genome has a different origin from the St genome. Data from the pepC (Mason-Gamer et al., 2010), GBSSI (Mason-Gamer et al., 2010), Acc1 (Fan et al., 2013a) and Pgk1 (Fan et al., 2013a) gene region indicated that the Y genome is related to the V genome in Dasypyrum villosum, the Q genome in Heteranthelium piliferum and the Xp genome in Peridictyon sanctum. In the present Acc1 tree, the Y-type sequences from Anthosachne and Roegneria were grouped with the V genome in D. villosum (100 % BS and 100 % PP), and the Xp genome in P. sanctum was placed outside the group with strong statistical support (80 % BS and 98 % PP), which is consistent with previous studies (Mason-Gamer et al., 2010; Fan et al., 2013a). Our results supported the hypothesis that the Y genome was not derived from the Pseudoroegneria species.

Anthosachne has its origin through a natural polyploidization process, which originated from hybridization between tetraploid Roegneria and diploid Australopyrum species (Torabinejad and Mueller, 1993; Lu et al., 1995; Yen et al., 2006; Fan et al., 2013a). The present phylogenetic analysis based on the combined cpDNA sequences showed that Anthosachne was grouped with Pseudoroegneria and Roegneria, indicating that Pseudoroegneria and Roegneria are closely related to Anthosachne, and the St lineage from Roegneria served as the maternal chloroplast parent during the speciation of Anthosachne. This result, in conjunction with the biparentally inherited Acc1 tree (Fig. 1; Fig. S1), implies that Australopyrum (W) species were the paternal parents for the Anthosachne species.

Genus Roegneria is the donor of StY genomes to Anthosachne and contains approx. 30 species that distributed in the Qinghai–Tibetan Plateau, Central Asia, East Asia, West Asia, and South-east Europe (Yen and Yang, 2011). In the calibrated phylogenetic tree inferred from the combined St sequences herein, Anthosachne formed one monophyletic group, followed by Pseudoroegneria and Roegneria. The combined cpDNA data showed that all the sampled Anthosachne taxa were grouped with Roegneria anthosachnoides from south-west of China with 98 % BS and 100 % PP. These results are in good agreement with previous studies by Liu et al. (2006) based on analysis of chloroplast trnL-F sequences of 45 accessions of Elymus L. s.l. (including 25 Roegneria speceis with StY genomes), where A. australasica and three Roegneria species (including R. anthosachnoides) from south-west of China were in one clade. Given the present data, it is suggested that the contributor to StY genomes of Anthosachne is most probably related to current Roegneria species from south-west of China.

Because hybridization requires physical proximity, it is pertinent to assess polyploidization events in Anthosachne from a geographic perspective. Anthosachne and its donor of the W genome (Australopyrum) are Australasian native Triticeae genera, and the donor of the StY genomes (Roegneria) is of Asian origin (most Roegneria species are distributed in south China), indicating that Anthosachne has its origin through a historical hybridization event between ancestral Roegneria and Australopyrum species in Australasia. The occurrence of ancestral Roegneria in Australasia might result from either a Gondwanan origin or a migration event. Geological evidence indicated that the creation of Australia and New Zealand resulted from the break-up of Gondwana, and Australia had a long history of separation from Antarctica, with full separation in the early Oligocene (35 MYA) (Upchurch, 2008). However, grass pollen and fossil plants did not appear in the fossil record in Australia until the Miocene (5–23 MYA) (Kershaw et al., 1994). Our molecular dating analysis showed that the most recent common ancestor (tMRCA) of the StYW genome Anthosachne species, StY genome Roegneria tetraploids and Pseudoroegneria diploid was dated to 5·55 MYA, and the divergence time of Anthosachne was 4·66 MYA, which is in agreement with the suggestion that the origin of Roegneria was prior to the origin of Anthosachne. Combined with geological and fossil plant evidence, it can be suggested that the timing of the continental separation of Australia from the break up of Gondwana was significantly prior to that of the entry of Roegneria into Australia. We thus prefer the explanation of migration to Australia. Previous studies based on morphological and geographic data (West et al., 1988) and DNA information (Appels and Baum, 1992; Fan et al., 2013b) implied a Eurasian origin of the Triticeae. Migration leading to the present distribution of most Triticeae plants involved transcontinental migrations and long-distance dispersals (Blattner, 2006; Fan et al., 2013b), which may explain the occurrence of ancestral Roegneria in Australasia. Evidence from geological and plate tectonic evolution has suggested that the Australian plate collided with Sundaland (South-east Asia and the Indonesian archipelago) in the Miocene (Powell et al., 1981; Pubellier et al., 2003), which led to the emergence of islands and land bridges. Land bridges repeatedly connected some areas, including New Guinea and Australia, and Indochina, Sumatra, Java and Borneo on the Sundaland shelf, allowing overland migration among the Asian mainland, Sundaland and Australia in the late Miocene (5–10 MYA) (Nauheimer et al., 2012). Therefore, it is possible that the ancestral Roegneria allotetraploid might have colonized by transcontinental migrations and long-distance dispersals from Asia to Australasia via South-east Asia during the late Miocene. This provided physical proximity for allohexaploidization between the Roegneria entity and Australopyrum species in Australasia during the Pliocene and for the range expansions of the A. australasica complex (Fig. 4). However, the reason for the extinction of Roegneria in Australasia is uncertain. Because sexual and apomictic plants often occur in close proximity, gene flow via pollen from apomicts to outcrossing plants may lead to the spread of apomixis genes into the outcrossing population, compromising the integrity of the outcrossing population and placing the outcrossing population at risk of extinction (van Dijk and van Damme, 2000; Noyes, 2006). Since Roegneria species are outcrossing sexual plants, it is likely that apomixis genes from Anthosachne taxa may be transmitted into outcrossing Roegneria ancestors through backcrossing, leading to the extinction of Roegneria in Australasia.

Fig. 4.

Fig. 4.

Open in a new tab

Scheme de monstrating the process of the colonization of Roegneria entity during the late Miocene, the speciation of Anthosachne polyploids during the Pliocene and subsequent range expansions within the A. australasica complex.

Intraspecific relationships within A. australasica complex

Anthosachne australasica, A. australasica var. scabra and A. australasica var. plurinervisa were previously described as E. rectisetus, E. scaber and E. scaber var. plurinervis, respectively (Wang and Henwood, 1999; Yen et al., 2006), and fertile hybrids were formed among them (Torabinejad et al., 1987; Torabinejad and Mueller, 1993; Yen et al., 2006). It has long been a question whether to treat A. australasica and A. australasica var. scabra as distinct species or taxa within a single species. Analysis of morphological characteristics regarded A. australasica as synonymous with long-awned forms of A. australasica var. scabra (Löve and Connor, 1982). In a series of experiments, Hair (1956) recorded the occurrence of pseudogamous diplosporous apomixis in A. australasica, which is the only embryologically documented apomict in the Triticeae (Dewey, 1984). Crane and Carman (1987) reported that long-awned forms of A. australasica var. scabra were also apomictic. Data from morphology, embryology, cytology and pollen characteristics suggested that A. australasica and A. australasica var. scabra should be recognized at the rank of the species (Murphy, 2003). Based on morphological and cytogenetic evidence, Yen et al. (2006) proposed that A. australasica var. scabra and A. australasica var. plurinervisa should be included at the rank of variety in the species A. australasica. In this study, estimates of nucleotide divergence among A. australasica, A. australasica var. scabra and A. australasica var. plurinervisa ranged from 0·00447 (between A. australasica and A. australasica var. plurinervisa) to 0·00647 (between A. australasica and A. australasica var. scabra), and no fixed difference was observed among Anthosachne taxa (Table 2), indicating a low level of genetic differentiation among Anthosachne taxa. Such a low level of genetic differentiation is further strengthened by the present estimates of pairwise population differentiations (GST and FST), where in all population pairs, no more than 5·5 % of total nucleotide diversity was attributable to taxa divergence (Table 3). A low level of genetic differentiation is indicative of a single reproductive unit, and may also explain the lack of phylogenetic resolution within Anthosachne taxa, as shown in the present phylogenetic tree with a high number of zero-length branches. In addition, phylogenetic analysis of the W-copy Acc1 gene, cpDNA sequences and time-calibrated data showed a monophyletic origin of Anthosachne, indicating that the genus might originate through a single allopoyploidization event. In this case, the initially very restricted population size of the founder, together with partial apomixis preventing recombination by perpetuating genetic constitution of female gametes, might result in low genetic differentiation among Anthosachne taxa. Combining previous cytogenetic results (Torabinejad et al., 1987; Torabinejad and Mueller, 1993; Yen et al., 2006), it is reasonable to recognize A. australasica var. scabra and A. australasica var. plurinervisa as A. australasica complex. From a standpoint of reproductive biology, we tend to support the system of classification proposed by Yen et al. (2006).

Comparisons of nucleotide diversity at the population level based on the rbcL, trnH-psbA, and Acc1 data showed that A. australasica possesses a much higher level of nucleotide variation than A. australasica var. scabra and A. australasica var. plurinervisa (Table 1). This suggests that a contrasting pattern of population genetic structure exists for the three taxa. A different level of genetic diversity and contrasting population genetic structure might be explained by factors such as reproductive system, population size and history, and habitat that are diverged markedly among population (Zhou et al., 2008). Theoretical and empirical investigation suggested that outcrossing taxa possess a higher level of genetic diversity and lower genetic differentiation among populations than the predominantly inbreeding taxa (Zhou et al., 2008). Anthosachne australasica is currently described as apomictic with long awns, A. australasica var. scabra taxa is a mix of both sexual and apomictic individuals and A. australasica var. plurinervisa is exclusively sexual (Torabinejad et al. 1987; Wang and Henwood, 1999; Murphy, 2003; Yen et al., 2006). Within the A. australasica complex, apomixis is much more prevalent among outcrossing taxa than inbreeding taxa (Löve and Connor, 1982; Yen et al., 2006). The apomictic taxa were observed to be outcrossing in nature, and the taxa with sexual reproduction are inbreeding plants (Yen et al., 2006). Genetic variation in a selfing species would be reduced by decreasing effective population size and eliminating the effective rate of recombination (Charlesworth, 2003). Phytogeographically, A. australasica and A. australasica var. scabra are distributed in Australia and New Zealand, while A. australasica var. plurinervisa is restricted to New South Wales and Queensland of Australia (Yen et al., 2006), implying a limited population size in A. australasica var. plurinervisa. The habitats preferred by A. australasica are mountainous meadow, steppe desert and dry slopes on the coast. However, A. australasica var. plurinervisa grows in the region with black alkaline clay. Taking all this into consideration, it is not unexpected that apart from the factor of a difference in natural distribution and habitats within the A. australasica complex, the outcrossing and apomictic A. australasica possess a higher level of genetic diversity than facultatively autogamous A. australasica var. scabra and inbreeding A. australasica var. plurinervisa.

Differentiation of the St genome following an allohexaploid speciation event

The St-type sequences of both nuclear Acc1 gene and cpDNA were obtained from nearly all the polyploid species, which allows analysis of genetic differentiation between Roegneria and Anthosachne to be elucidated on the basis of orthologous comparison. On the basis of three data sets (the rbcL, trnH-psbA and Acc1 gene data), the level of nucleotide diversity (Tajima’s π and Watterson’s θw) in Anthosachne in Australasia was higher than that in continental Roegneria, and Tajima’s and Fu and Li’s D statistic suggests a departure from the equilibrium neutral model at these loci, with an excess of rare sequence variants in Anthosachne species. The genetic diversity patterns observed are therefore inconsistent with the traditional assumption that oceanic island populations exhibit lower levels of genetic variation than their continental relatives (Barrett, 1996; Fernández-Mazuecos and Vargas, 2011; Désamoré et al., 2012). There are three possible explanations for this inverted pattern of genetic diversity. The first one is mutation because mutation was considered to be an important factor that can increase genetic variation in a population and is of special importance in apomictic populations (Nybom, 2007; Lo et al., 2009). It was reported that levels of heterozygosity in genetic diversity could be increased by independent mutations in different alleles at different loci, and such mutations may appear more rapidly and accumulate over time, particularly in higher apomictic polyploids (Lo et al., 2009). Anthosachne is the only apomictic taxon in Triticeae. Genetic diversity could be increased by mutation at rbcL, trnH-psbA and Acc1 loci in apomictic Anthosachne taxa compared with no apomictic Roegneria taxa. The second explanation is that greater diversity in Anthosachne could reflect the evolutionary potential of asexuality following an allopolyploidization event. Recent evidence showed that diversity can be rapidly generated in early stages of allopolyploidy, due to genome reorganization (Fledman and Levy, 2012; Soltis et al., 2015). Apart from potential hybrid vigour, apomictic polyploids can take advantage not only of uniparental reproduction, lowered cost of sex and maintaining adapted genotypes, but also of reproduction by seeds including better dispersal ability compared witho vegetative reproduction (Hörandl and Paun, 2007). This leads to great ecological and evolutionary success of apomictic polyploids. Consequently, as the new apomictic species in Anthosachne established itself in new habitats (oceanic island) during the Pliocene, the populations expanded and further diversified, harbouring high diversity generated from an allopolyploidization event within and among taxa as seen today. This is in line with the present discovery that the island Anthosachne shows high nucleotide diversity compared with the mainland Roegneria. Thirdly, it cannot be ruled out that Roegneria population expansion could result in lower diversity in Roegneria compared with Anthosachne. Population expansion might have accelerated the fixation of mildly deleterious replacement mutations that became effectively neutral in Roegneria populations, as indicated by the present analysis of Tajima’s and Fu and Li’s D statistics. In addition, much of the continental distribution of the Roegneria species is likely to be the result of a recent range expansion.

SUPPLEMENTARY DATA

Supplementary data are available online at www.aob.oxfordjournals.org and consist of the following. Table S1: Anthosachne taxa and other related genera in Triticeae used in this study. Figure S1: full phylogenetic tree inferred from the Acc1 sequences of Anthosachne taxa and the sequences of its affinitive species within Triticeae, under the GTR + G + I model. Figure S2: full phylogenetic tree of Anthosachne species and its affinitive species inferred from the combined chloroplast rbcL and trnH-psbA sequences, under the GTR + G + I model.

Supplementary Data
supp_119_1_95__index.html (1,003B, html)

ACKNOWLEDGEMENTS

This study was funded by the National Natural Science Foundation of China (Nos 31200252, 31270243 and 31670387), Special Fund for Protection and Utilization of Crop Germplasm Resources in the Public Interest of China (No. 2016NWB030-02) and the Science and Technology Bureau (No. 2060503) and Education Bureau of Sichuan Province. We are very grateful to the American National Plant Germplasm System (Pullman, Washington, USA) for providing part of the seed materials.

LITERATURE CITED

  1. Adams KL, Wendel JF. 2005. Novel patterns of gene expression in polyploid plants. Trends in Genetics 21: 539–543. [DOI] [PubMed] [Google Scholar]
  2. Appels R, Baum B. 1992. Evolution of the Nor and 5SDna loci in the Triticeae, In: PS Soltis, DE Soltis, JF Doyle, eds. Molecular systematics of plants. New York: Chapman & Hall, 92–116. [Google Scholar]
  3. Baack EJ, Rieseberg LH. 2007. A genomic view of introgression and hybrid speciation. Current Opinion in Genetics and Development 17: 513–518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Barkworth ME, Jacobs SWL. 2011. The Triticeae (Gramineae) in Australasia. Telopea 13: 37–56. [Google Scholar]
  5. Barrett SCH. 1996. The reproductive biology and genetics of island plants. Philosophical Transactions of the Royal Society B: Biological Sciences 351: 725–733. [Google Scholar]
  6. Blattner FR. 2006. Multiple intercontinental dispersals shaped the distribution area of Hordeum (Poaceae). New Phytologist 169: 603–614. [DOI] [PubMed] [Google Scholar]
  7. Brassac J, Blattner FR. 2015. Species-level phylogeny and polyploid relationships in Hordeum (Poaceae) inferred by next-generation sequencing and in silico cloning of multiple nuclear loci. Systematic Biology 64: 792–808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Charlesworth D. 2003. Effects of inbreeding on the genetic diversity of populations. Philosophical Transactions of the Royal Society B: Biological Sciences 358: 1051–1070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Connor HE. 2005. Flora of New Zealand: Gramineae Supplement II*: Pooideae (Poeae, Triticeae). New Zealand Journal of Botany 43: 493–507. [Google Scholar]
  10. Cosendai AC, Wagner J, Ladinig U, Rosche C, Hörandl E. 2013. Geographical parthenogenesis and population genetic structure in the alpine species Ranunculus kuepferi (Ranunculaceae). Heredity 110: 560–569. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Crane CF, Carman JG. 1987. Mechanisms of apomixis in Elymus rectisetus from eastern Australia and New Zealand. American Journal of Botany 74: 477–496. [Google Scholar]
  12. Désamoré A, Laenen B, González-Mancebo JM, et al. 2012. Inverted patterns of genetic diversity in continental and island populations of the heather Erica scoparia s.l. Journal of Biogeography 39: 574–584. [Google Scholar]
  13. Dewey DR. 1984. The genomic system of classification as a guide to intergeneric hybridization with the perennial Triticeae In: JP Gustafson, ed. Gene manipulation in plant improvement. New York; Columbia University Press, 209–279. [Google Scholar]
  14. van Dijk P, van Damme J, 2000. Apomixis technology and the paradox of sex. Trends in Plant Sciences 5: 81–84 [DOI] [PubMed] [Google Scholar]
  15. Dong ZZ, Fan X, Sha LN, et al. 2013. Phylogeny and molecular evolution of the rbcL gene of St genome in Elymus sensu lato (Poaceae: Triticeae). Biochemical Systematics and Ecology 50: 322–330. [Google Scholar]
  16. Dou Q, Wang RR, Lei Y, Yu F, Li Y, Wang H. 2014. Genome analysis of seven species of Kengyilia (Triticeae: Poaceae) with FISH and GISH. Genome 56: 641–649. [DOI] [PubMed] [Google Scholar]
  17. Doyle JJ, Doyle JL. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11–15. [Google Scholar]
  18. Drummond AJ, Rambaut A. 2007. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology 7: 214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Egan AN, Crandall KA. 2008. Divergence and diversification in North American Psoraleeae (Fabaceae) due to climate change. BMC Biology 6: 1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Excoffier L, Laval G, Schneider S. 2005. Arlequin ver. 3.0: an integrated software package for population genetics data analysis. Evolutionary Bioinformatics 1: 47–50. [PMC free article] [PubMed] [Google Scholar]
  21. Fan X, Sha LN, Dong ZZ, et al. 2013a. Phylogenetic relationships and Y genome origin in Elymus L. sensu lato (Triticeae; Poaceae) based on single-copy nuclear Acc1 and Pgk1 gene sequences. Molecular Phylogenetics and Evolution 69: 919–928. [DOI] [PubMed] [Google Scholar]
  22. Fan X, Sha LN, Yu SB, et al. 2013b. Phylogenetic reconstruction and diversification of the Triticeae (Poaceae) based on single-copy nuclear Acc1 and Pgk1 gene data. Biochemical Systematics and Ecology 50: 346–360. [Google Scholar]
  23. Feldman M, Levy AA. 2012. Genome evolution due to allopolyploidization in wheat. Genetics 192: 763–774. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Felsenstein J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783–791. [DOI] [PubMed] [Google Scholar]
  25. Fernández-Mazuecos M, Vargas P. 2011. Genetically depauperate in the continent but rich in oceanic islands: Cistus monspeliensis (Cistaceae) in the Canary Islands. PLoS One 6: e17172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Fu YX, Li WH. 1993. Statistical tests of neutrality of mutations. Genetics 133: 693–709. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Hair JB. 1956. Subsexual reproduction in Agropyron. Heredity 10: 129–160. [Google Scholar]
  28. Hegarty MJ, Hiscock SJ. 2008. Genomic clues to the evolutionary success of polyploid plants. Current Biology 18: R435–R444. [DOI] [PubMed] [Google Scholar]
  29. Hörandl E, Paun O. 2007. Patterns and sources of genetic diversity in apomictic plants: implications for evolutionary potentials and ecology In: E Hörandl, U Grossniklaus, PJ Van Dijk, T Sharbel, eds. Apomixis: evolution, mechanisms and perspective. Ruggell, Liechtenstein: ARG-Gantner, 169–194. [Google Scholar]
  30. Huang S, Sirikhachornkit A, Su X, et al. 2002. Genes encoding plastid acetyl-CoA carboxylase and 3-phosphoglycerate kinase of the Triticum/Aegilops complex and the evolutionary history of polyploid wheat. Proceedings of the National Academy of Sciences, USA 99: 8133–8138. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Hudson RR. 1990. Gene genealogies and the coalescent process In: D Futuyma, J Antonovics, eds. Oxford surveys in evolutionary biology. New York: Oxford University Press, 1–44. [Google Scholar]
  32. Huelsenbeck JP, Ronquist FR. 2001. MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754–755. [DOI] [PubMed] [Google Scholar]
  33. Jessop JP. 1986. Elymus L In: JP Jessop, HR Toelken, eds. Flora of South Australia, vol. 4 Adelaide: South Australian Government Printing Division, 1882–1884. [Google Scholar]
  34. Kershaw AP, Martin HA, McEwenMmason JRC. 1994. The Neogene: a period of transition In: RS Hill, ed. Australian vegetation history. Cambridge: Cambridge University Press, 299–327. [Google Scholar]
  35. Kim ST, Sultan SE, Donoghue MJ. 2008. Allopolyploid speciation in Persicaria (Polygonaceae): insights from a low-copy nuclear region. Proceedings of the National Academy of Sciences, USA 105: 12370–12375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Kress WJ, Erickson DL. 2007. A two-locus global DNA barcode for land plants: the coding rbcL gene complements the non-coding trnH-psbA spacer region. PLoS One 2: e508. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Liu QL, Ge S, Tang HB, Zhang XL, Zhu GF, Lu BR. 2006. Phylogenetic relationships in Elymus (Poaceae: Triticeae) based on the nuclear ribosomal internal transcribed spacer and chloroplast trnL-F sequences. New Phytologist 170: 411–420. [DOI] [PubMed] [Google Scholar]
  38. Lo EYY, Stefanovic S, Dickinson TA. 2009. Population genetic structure of diploid sexual and polyploid apomictic hawthorns (Crataegus; Rosaceae) in the Pacific Northwest. Molecular Ecology 18: 1145–1160. [DOI] [PubMed] [Google Scholar]
  39. Löve Á. 1984. Conspectus of the Triticeae. Feddes Repert 95: 425–521. [Google Scholar]
  40. Löve Á, Connor HE. 1982. Relationships and taxonomy of New Zealand wheatgrasses. New Zealand Journal of Botany 20: 169–186. [Google Scholar]
  41. Lu BR, Salomon B, von Bothmer R. 1992. Cytogenetic studies of progeny from the intergeneric crosses Elymus × Hordeum and Elymus × Secale. Genome 33: 425–432. [Google Scholar]
  42. Lu BR, Salomon B, von Bothmer R. 1995. Interspecific hybridizations with Elymus confusus and E. dolichatherus, and their genomic relationships (Poaceae: Triticeae). Plant Systematics and Evolution 197: 1–19. [Google Scholar]
  43. Marcussen T, Sandve SR, Heier L, Spannagl M, Pfeifer M, Jakobsen KS. 2014. Ancient hybridizations among the ancestral genomes of bread wheat. Science 345: 1250092. [DOI] [PubMed] [Google Scholar]
  44. Marcussen T, Heier L, Brysting AK, Oxelman B, Jakobsen KS. 2015. From gene trees to a dated allopolyploid network: insights from the angiosperm genus Viola (Violaceae). Systematic Biology 64: 84–101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Mason-Gamer RJ. 2004. Reticulate evolution, introgression, and intertribal gene capture in an allohexaploid grass. Systematic Biology 53: 25–37. [DOI] [PubMed] [Google Scholar]
  46. Mason-Gamer RJ, Orme NL, Anderson CM. 2002. Phylogenetic analysis of North American Elymus and the monogenomic Triticeae (Poaceae) using three chloroplast DNA data sets. Genome 45: 991–1002. [DOI] [PubMed] [Google Scholar]
  47. Mason-Gamer RJ, Burns MM, Naum M. 2010. Phylogenetic relationships and reticulation among Asian Elymus (Poaceae) allotetraploids: analyses of three nuclear gene trees. Molecular Phylogenetics and Evolution 54: 10–22. [DOI] [PubMed] [Google Scholar]
  48. Middleton CP, Senerchia N, Stein N, et al. 2014. Sequencing of chloroplast genomes from wheat, barley, rye and their relatives provides a detailed insight into the evolution of the Triticeae tribe. PLoS One 9: e85761. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Murphy MA. 2003. Relationships among taxa of Elymus (Poaceae: Triticeae) in Australia: reproductive biology. Australian Systematic Botany 16: 633–642. [Google Scholar]
  50. Nauheimer L, Boyce PC, Renner SS. 2012. Giant taro and its relatives: a phylogeny of the large genus Alocasia (Araceae) sheds light on Miocene floristic exchange in the Malesian region. Molecular Phylogenetics and Evolution 63: 43–51. [DOI] [PubMed] [Google Scholar]
  51. Noyes RD. 2006. Intraspecific nuclear ribosomal DNA divergence and reticulation in sexual diploid Erigeron strigosus (Asteraceae). American Journal of Botany 93: 470–479. [DOI] [PubMed] [Google Scholar]
  52. Nybom H. 2007. Unique reproduction in dogroses (Rosa sect. Caninae) maintains successful and highly heterozygous genotypes In: E Hörandl, U Grossniklau, PJ Van Dijk, TF Sharbel, eds. Apomixis: evolution, mechanisms, and perspectives. Ruggell, Liechtenstein: ARG-Gantner, 281–298. [Google Scholar]
  53. Okito P, Mott IW, Wu Y, Wang RR-C. 2009. A Y genome specific STS marker in Pseudoroegneria and Elymus species (Triticeae: Gramineae). Genome 52: 391–400. [DOI] [PubMed] [Google Scholar]
  54. Ozias-Akins P, van Dijk PJ. 2007. Mendelian genetics of apomixis in plants. Annual Review of Genetics 41: 509–537. [DOI] [PubMed] [Google Scholar]
  55. Petit RJ, Mousadik AE, Pons O. 1998. Identifying populations for conservation on the basis of genetics markers. Conservation Biology 12: 844–855. [Google Scholar]
  56. Posada D, Crandall KA. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817–818. [DOI] [PubMed] [Google Scholar]
  57. Powell CM, Johnson BD, Veevers JJ. 1981. The Early Cretaceous break-up of Eastern Gondwanaland, the separation of Australia and India, and their interaction with South-east Asia In: A Keast, ed. Ecological biogeography of Australia. The Hague: DW Junk, 15–29. [Google Scholar]
  58. Pubellier M, Ali J, Monnier C. 2003. Cenozoic plate interaction of the Australia and Philippine Sea Plates: ‘hit-and-run’ tectonics. Tectonophysics 363: 181–199. [Google Scholar]
  59. Rambaut A, Drummond AJ. 2007. Tracer V1.4 http://beast.bio.ed.ac.uk/Tracer.
  60. Ramsey J, Ramsey TS. 2014. Ecological studies of polyploidy in the 100 years following its discovery. Philosophical Transactions of the Royal Society B: Biological Sciences 369: 20130352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Rieseberg LH, Willis JH. 2007. Plant speciation. Science 317: 910–914. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Robertson A, Rich TC, Allen AM, et al. 2010. Hybridization and polyploidy as drivers of continuing evolution and speciation in Sorbus. Molecular Ecology 19: 1675–1690. [DOI] [PubMed] [Google Scholar]
  63. Rozas J, Sánchez-DelBarrio JC, Messeguer X, Rozas R. 2005. DNA sequence polymorphism, version 4.10.4. DNAsp4 computer software. Barcelona University, Barcelona, Spain. [Google Scholar]
  64. Särkinen T, Pennington RT, Lavin M, Simon MF, Hughes CE. 2012. Evolutionary islands in the Andes: persistence and isolation explain high endemism in Andean dry tropical forests. Journal of Biogeography 39: 884–900. [Google Scholar]
  65. Sha LN, Fan X, Zhang HQ, et al. 2016. Phylogeny and molecular evolution of the DMC1 gene in the polyploid genus Leymus (Triticeae: Poaceae) and its diploid relatives. Journal of Systematics and Evolution 54: 250–263. [Google Scholar]
  66. Shaw J, Small RL. 2005. Chloroplast DNA phylogeny and phylogeography of the North American plums (Prunus subgenus Prunus section Prunocerasus, Rosaceae). American Journal of Botany 92: 2011–2030. [DOI] [PubMed] [Google Scholar]
  67. Sochor M, Vašut RJ, Sharbel TF, Trávníček B. 2015. How just a few makes a lot: Speciation via reticulation and apomixis on example of European brambles (Rubus subgen. Rubus, Rosaceae). Molecular Phylogenetics and Evolution 89: 13–27. [DOI] [PubMed] [Google Scholar]
  68. Soltis PS. 2013. Hybridization, speciation and novelty. Journal of Evolutionary Biology 26: 291–293. [DOI] [PubMed] [Google Scholar]
  69. Soltis PS, Marchant DB, Van de Peer Y, Soltis DE. 2015. Polyploidy and genome evolution in plants. Current Opinion in Genetics and Ddevelopment 35: 119–125. [DOI] [PubMed] [Google Scholar]
  70. Sun GL, Ni Y, Daley T. 2008. Molecular phylogeny of RPB2 gene reveals multiple origin, geographic differentiation of H genome, and the relationship of the Y genome to other genomes in Elymus species. Molecular Phylogenetics and Evolution 46: 897–907. [DOI] [PubMed] [Google Scholar]
  71. Sun GL, Komatsuda T. 2010. 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. Molecular Phylogenetics and Evolution 56: 727–733. [DOI] [PubMed] [Google Scholar]
  72. Tajima F. 1989. Statistical method for testing the neutral mutation of hypothesis by DNA polymorphism. Genetics 123: 585–595. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Tamura K, Dudley J, Nei M, Kumar S. 2007. MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0. Molecular Biology and Evolution 24: 1596–1599. [DOI] [PubMed] [Google Scholar]
  74. Thompson JD, Plewniak F, Poch O. 1999. A comprehensive comparison of multiple sequence alignment programs. Nucleic Acids Research 27: 2682–2690. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Torabinejad J, Carman JG, Crane CF. 1987. Morphology and genome analyses of interspecific hybrids of Elymus scabrus. Genome 29: 150–155. [Google Scholar]
  76. Torabinejad J, Mueller RJ. 1993. Genome analysis of intergeneric hybrids of apomictic and sexual Australian Elymus species with wheat, barley and rye: implications for the transfer of apomixes to cereals. Theoretical and Applied Genetics 86: 288–294. [DOI] [PubMed] [Google Scholar]
  77. Upchurch P. 2008. Gondwanan break-up: legacies of a lost world? Trends in Ecology and Evolution 23: 229–236. [DOI] [PubMed] [Google Scholar]
  78. Wang RR-C, von Bothmer R, Dvorak J, Fedak G, Linde-Laursen I, Muramatsu M. 1994. Genome symbols in the Triticeae In: RR-C Wang, KB Jensen, C Jaussi, eds. Proceedings of the 2nd International Triticeae Symposium. Utah State University Press, 29–34. [Google Scholar]
  79. Wang S, Henwood MJ. 1999. The taxonomic utility of micromorphological characters in Australian and New Zealand Elymus species (Poaceae). Telopea 8: 351–362. [Google Scholar]
  80. Watterson GA. 1975. On the number of segregation sites in genetic models without recombination. Theoretical Population Biology 7: 256–276. [DOI] [PubMed] [Google Scholar]
  81. West JG, Mclntyre CL, Appels R. 1988. Evolution and systematic relationships in the Triticeae (Poaceae). Plant Systematics and Evolution 160: 1–28. [Google Scholar]
  82. Wheeler DJB, Jacobs SWL, Norton BE. 1982. Grasses of New South Wales. Armidale: University of New England Publishing Unit. [Google Scholar]
  83. Yen C, Yang JL, Baum BR. 2006. Biosystematics of Triticeae, Vol. 3 Beijing, Chinese Agriculture Press, Beijing. [Google Scholar]
  84. Yen C, Yang JL. 2011. Biosystematics of Triticeae, Vol. 4 Beijing: Chinese Agriculture Press. [Google Scholar]
  85. Zhou HF, Zheng XM, Wei RX, Second G, Vaughan DA, Ge S. 2008. Contrasting population genetic structure and gene flow between Oryza rufipogon and Oryza nivara. Theoretical and Applied Genetics 117: 1181–1189. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Data
supp_119_1_95__index.html (1,003B, html)
supp_mcw222_suppl_data.zip (380.9KB, zip)

Articles from Annals of Botany are provided here courtesy of Oxford University Press

RESOURCES