• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of annbotAboutAuthor GuidelinesEditorial BoardAnnals of Botany
Ann Bot. Aug 2007; 100(2): 249–260.
Published online Jun 11, 2007. doi:  10.1093/aob/mcm093
PMCID: PMC2735307

Recent Natural Hybridization between Two Allopolyploid Wheatgrasses (Elytrigia, Poaceae): Ecological and Evolutionary Implications

Abstract

Background and Aims

Natural hybridization was investigated between two predominantly allohexaploid wheatgrasses, weedy Elytrigia repens and steppic E. intermedia, with respect to habitats characterized by different degrees of anthropogenic disturbance.

Methods

Using flow cytometry (relative DNA content), 269 plants from three localities were analysed. Hybrids were further analysed using nuclear ribosomal (ITS1-5·8S-ITS2 region) and chloroplast (trnT-F region) DNA markers in addition to absolute DNA content and chromosome numbers.

Key Results

Weedy E. repens was rare in a steppic locality whereas E. intermedia was almost absent at two sites of agricultural land-use. Nevertheless, hybrids were common there whereas none were found at the steppic locality, underlining the importance of different ecological conditions for hybrid formation or establishment. At one highly disturbed site, > 16 % of randomly collected plants were hybrids. Hexaploid hybrids showed intermediate genome size compared with the parents and additive patterns of parental ITS copies. Some evidence of backcrosses was found. The direction of hybridization was highly asymmetric as cpDNA identified E. intermedia as the maternal parent in 61 out of 63 cases. Out of nine nonaploid cytotypes (2n = 9x = 63) which likely originated by fusion of unreduced and reduced gametes of hexaploids, eight were hybrids whereas one was a nonaploid cytotype of E. repens. The progeny of one nonaploid hybrid demonstrated gene flow between hexaploid and nonaploid cytotypes.

Conclusions

The results show that E. repens and E. intermedia frequently cross at places where they co-occur. Hybrid frequency is likely influenced by habitat type; sites disturbed by human influence sustain hybrid formation and/or establishment. Hexaploid and nonaploid hybrid fertility is not negligible, backcrossing is possible, and the progeny is variable. The frequent production of new at least partially fertile cyto- and genotypes provides ample raw material for evolution and adaptation.

Key words: Triticeae, Poaceae, Elytrigia repens, Elytrigia intermedia, hybridization, polyploidy, chloroplast DNA, internal transcribed spacer, genome size, adaptation

INTRODUCTION

Hybridization is perceived as an important phenomenon in plant speciation (e.g. Arnold, 1997; Rieseberg et al., 2003; Gross and Rieseberg, 2005). This is mainly evident in hybrids emerging from hybridization involving at least one species non-indigenous to the respective area (Abbott, 1992). Such cases are usually well documented and carefully studied, because they represent examples of speciation caught in the act (Ownbey, 1950; Rieseberg et al., 1990; Gray et al., 1991; Ashton and Abbott, 1992; Soltis et al., 1995; Krahulec et al., 2005; Mandák et al., 2005). Species co-occurring at the same locality for a longer time may also hybridize; however, their hybrids may be more easily overlooked or misidentified when the parental species are morphologically similar and the morphology of hybrids is overlapping with that of the parental species. Elytrigia repens and E. intermedia (Poaceae), on which this study focuses, are examples of such a potential underestimation of hybridization in their native area.

Both species are perennial, outcrossing allopolyploid grasses belonging to the wheat tribe Triticeae (Dewey, 1984; Löve, 1984). The tribe is especially well-known for the economic importance of its three major crops: wheat, barley and rye. The tribe's structure is highly reticulate, with distinct genomes/gene lineages occurring within many polyploid, but also within some diploid species, which is a consequence of ancient hybridization events, introgression, lineage sorting of ancestral variation, multiple origins of particular species, or a combination of these (Kellogg et al., 1996; Mason-Gamer, 2004). These processes resulted in a strong ecological, morphological and genetic resemblance of many Triticeae taxa (Stebbins, 1956; Dewey, 1984). Their ability to hybridize with each other is so common that Stebbins noted: ‘So many hybrid combinations in one group is unparalleled in the higher plants.’ (Stebbins, 1956). One consequence of a reticulate structure is that if subsequent hybridization between genetically related species occurs, fertility of the hybrids can be enhanced because their chromosomes may pair more readily, and polyploidization generally provides an effective way to escape from sterility (Stebbins, 1940). Within the wheat tribe, about three-quarters of the taxa are of polyploid origin (Löve, 1984).

The predominantly hexaploid Elytrigia repens and E. intermedia are no exceptions in this respect, and their ability to hybridize with many other species of the tribe has been observed (Dewey, 1984, and references therein; Assadi and Runemark, 1995). Moreover, E. intermedia is able to cross with wheat. This hybrid (× Trititrigia cziczinii Tsvel.) was originally described by Tsitsin (1960) and taxonomically validated by Tsvelev (1973). Since then, many experimental hybridization studies have followed (Sharma and Gill, 1983; Franke et al., 1992; Chen et al., 2001; Han et al., 2003) in order to transfer desirable traits of the wild grass into the wheat genome (Sharma et al., 1995; Friebe et al., 1996; Fedak and Han, 2005). However, natural hybrids between wheat and E. intermedia have not been observed so far. Should those be discovered, hybridization of E. repens, one of the most troublesome weeds on cultivated land worldwide (Palmer and Sagar, 1963), with the comparably rare E. intermedia, combined with abundant production and at least partial fertility of E. repens × E. intermedia hybrids, might have a considerable impact on risk assessment of genetically modified wheat. Therefore, knowledge about the frequency of hybridization between E. repens and E. intermedia in nature is not only of interest for science, but also for economy and agriculture. In this context, ecological parameters that could facilitate hybrid formation need to be investigated.

Both Elytrigia congeners differ ecologically; however, mainly due to the wide ecological amplitude of E. repens, they also co-occur in some types of habitats such as field margins and steppic grasslands in warmer regions. This study focuses on several such sites. Although hybridization between them occurs in central Europe–their hybrid was originally described from the area of the present Czech Republic–it has attracted little attention. Except for morphology or chromosome pairing in artificial hybrids (Berchtold and Opitz, 1836; Prokudin and Druleva, 1971; Melderis, 1980; Assadi and Runemark, 1995), not much proven evidence of natural hybridization nor of its frequency is currently available. As a consequence of hybridization, the species' introgressive potential could lead to the transfer of ecological adaptations between species (Stutz and Thomas, 1964; Arnold and Bennett, 1993; Kim and Rieseberg, 1999; Mahelka, 2006).

The acknowledged ease with which Elytrigia repens and E. intermedia can hybridize might suggest that they are closely related. Cytogenetic studies revealed the preliminary genome constitution of hexaploid cytotypes (2n = 6x = 42) of both species. In E. repens, it was determined as StStH, where St and H designate Pseudoroegneria (Nevski) Á. Löve and Hordeum L. genomes, respectively (Assadi and Runemark, 1995). The genome constitution of E. intermedia was determined as EeEeSt (Liu and Wang, 1993) or EeEbSt (Chen et al., 1998) with Ee and Eb designating the closely related Thinopyrum elongatum and Th. bessarabicum genomes. More recent studies revealed that both species might have still more complex genomic histories. Mason-Gamer (2004) and Mason-Gamer et al. (2005) found at least five distinct lineages in the genome of E. repens, revealing the reticulate and possibly polyphyletic origin of this species. Recently, new insights in the genome composition of E. intermedia became available (Kishii et al., 2005), but not all potential genome donors have been identified yet. These data show that the two species represent distinct genetic entities that probably share only the St genome from Pseudoroegneria. Furthermore, at least some accessions might be further influenced by hybridization and introgression so that both species might actually have multiple origins.

As only two morphological characters, which show large intraspecific variation and frequent overlapping of character values, distinguish between E. repens and E. intermedia (Melderis, 1980; Barkworth and Dewey, 1985; Kubát et al., 2002), identification of hybridogenous plants by morphology alone is difficult. As a prerequisite for evaluating the frequency of hybrids in the field, genome size measurements were recently established as a reliable means of identifying both parents and their hybrids (Mahelka et al., 2005).

In the present paper, using nuclear ribosomal (ITS1-5·8S-ITS2 region) and chloroplast (trnT-F region) DNA markers in addition to genome size and chromosome numbers, evidence of natural hybridization between E. repens and E. intermedia is reported. In particular (a) the frequency of hybridization among hexaploid cytotypes with respect to habitat types, i.e. between natural steppic grassland and the agricultural landscape, is compared; (b) the origin of nonaploid (2n = 9x = 63) cytotypes is proposed; (c) evidence of natural hybridization between nonaploid and hexaploid cytotypes is presented; (d) the maternal origin of hybrids and nonaploids is identified; and (e) the impact of hybridization within habitats of different land-use is discussed.

MATERIALS AND METHODS

Study species

Elytrigia repens (L.) Nevski [syn. Agropyron repens (L.) P. Beauv., Elymus repens (L.) Gould] is widespread throughout the territory of the Czech Republic and ranges from lowlands to the mountain belt. It occupies all man-made habitats and arable ground, and also occurs on such natural habitats as steppic grasslands and wet meadows (Chytrý and Tichý, 2003; authors' observations).

Elytrigia intermedia (Host) Nevski [syn. Agropyron intermedium (Host) P. Beauv., Thinopyrum intermedium (Host) Barkworth et D.R. Dewey] has a more limited distribution, strongly corresponding with the occurrence of steppic habitats. It colonizes dry and warm habitats like steppes and base-rich rocks and also pine forests on sandy ground, vineyards, orchards and field margins in warm regions of the Czech Republic (Chytrý and Tichý, 2003; authors' observations).

Both species are morphologically variable (Mizianty and Szczepaniak, 1997; Assadi, 1998; Mizianty et al., 2001). Constancy of the morphological characters, their taxonomic significance, correlation with ecological preferences or with genetic variation, remain unexplored. Both species occur predominantly at hexaploid level in the Czech Republic. Aside from hexaploids, several nonaploids (2n = 9x = 63) were found earlier (Mahelka et al., 2005). In places where natural or semi-natural habitats with E. intermedia come into contact with agricultural land-use, both species co-occur and hybridize. The hybrid was originally described as Agropyron × mucronatum Opiz (Berchtold and Opiz, 1836) [syn. Elytrigia mucronata (Opiz) Prokudin]. The morphology of hybrids is intermediate between the parental species but sometimes overlaps with one or the other parent.

All plants used in this study are cultivated in the experimental garden at the Institute of Botany, Průhonice, Czech Republic.

Sampling strategy

The plant material analysed was divided into two sets. (1) To compare between hybridization frequency in a natural habitat and the agricultural landscape, three localities (A–C; described below) were chosen comprising two different habitat types, from which a total of 269 plants was collected (Table 1). At localities A and B, plants were collected predominantly at transect points without bias towards flowering plants. (2) To these were added 33 hexaploid hybrids and five nonaploids from the authors' collection (Mahelka et al., 2005) for more detailed investigation.

Table 1.
Localities and distribution of species, hybrids and nonaploids

Locality A. ‘Pouzdřany’, a steppic slope in a protected area, characterized by a community of Festucion valesiaceae Klika 1931, represents the conserved, natural steppic habitat. The south-facing aspect of the site often causes plants to suffer from droughts. Two transects were sampled (one plant per 5 m): (1) on the top part of the slope, parallel to the boundary between the steppic habitat and an abandoned field (38 plants); (2) from the top to the bottom of the slope along a footpath (74 plants).

Locality B. ‘Valtice’, a vineyard in an agricultural landscape. (a) Three transects were carried out: (1) within the vineyard (12 plants; one per 24 m); (2) along a path adjoining the vineyard (36 plants; one per 5 m); (3) a shrubby vineyard margin ending in a steppic locality adjacent to a cultivated field, transect in orthogonal direction to (1) and (2) (61 plants; one per 8 m). (b) An additional 20 plants were collected at the adjacent steppic locality in order to cover as much of the morphological variation as possible.

Locality C. ‘Dolní Dunajovice’–agricultural landscape, characterized by an alternation of vineyards and cultivated and abandoned fields. Due to discontinuous occurrence of Elytrigia species, 28 plants were collected to cover the study area.

Test of hexaploid hybrids' and nonaploids' seed fertility

To assess fertility, all available spike-forming hexaploid hybrids and nonaploids were tested for seed fertility and germinability (18 hexaploids: H3, H6, H8, H12, H13, H19, H20, H22–H25, H30, H34, H39, H44, H56, H58, H63; five nonaploids: N3, N5, N6, N8, N9). In the autumn of 2003, spikelets were collected in the experimental garden and flower numbers in spikelets and developed caryopses (one-seeded fruit), if any, were counted. Fertility was calculated as the ratio between caryopses and flowers. Caryopses were tested for germination ability in pots in a greenhouse. Five randomly selected samples of each parental species were tested in the same way and used as a control. Because of their high fertility, only ten randomly selected caryopses were tested for germinability.

Progeny of a nonaploid

The progeny of one hybridogenous nonaploid plant (N7, locality C) was investigated to determine the ratio of offspring. In 2002, a total of 195 spikelets was collected from the plant in the field. They produced 20 fully developed caryopses all of which germinated in pots in a greenhouse. Eight of the seedlings died at the 2–6 leaf stage. The other 12 were transferred to the experimental garden and maintained for subsequent analyses.

Genome size analyses

Relative DNA content was measured in all plants for their identification. For determination of absolute DNA content of the whole chromosome complement (holoploid genome size sensu Greilhuber et al., 2005; hereafter for brevity the term genome size will be used) of hexaploid hybrids, nonaploids and the nonaploid's offspring, specimens with close but non-overlapping genome size compared with the material analysed were employed as internal standards: Triticum aestivum L. var. lutescens (Alef.) Mansf. ‘Bezostaja 1’ (2C = 34·4 pg; Mahelka et al., 2005) for hexaploid hybrids and Vicia faba (2C = 26·9 pg; Doležel et al., 1992) for nonaploids. Because of the considerable variation in genome size of the nonaploid's offspring, both internal standards were used (Table 2). All procedures followed Mahelka et al. (2005).

Table 2.
Holoploid genome size, chromosome numbers, morphological identification, chloroplast DNA haplotypes, ITS variants and possible gamete compositions of nonaploid plants and one nonaploid's progeny

Chromosome counting

Chromosome numbers of the four nonaploids and the nonaploid's progeny (12 plants) were counted as described previously (Mahelka et al., 2005). Additionally, three hexaploid hybrids with DNA content deviating most from the values typical of hybrids (nos H1, H2, H63) were counted to verify that the plants were not aneuploid.

DNA isolation

DNA was isolated as described in Štorchová et al. (2000), but fresh leaves were crushed in liquid nitrogen. Quality and yield of the isolated DNA were checked on agarose gels.

Analysis of chloroplast DNA

Based on the knowledge of cpDNA variation in the Triticeae (Mason-Gamer et al., 2002, and references therein), the trnL intron and the trnL-trnF intergenic spacer proved to be the most variable regions known so far. A set of Elytrigia repens (= Elymus repens) data was retrieved from GenBank (accession numbers AY362786–91), but no sequence for Elytrigia intermedia (synonyms included) was available. Therefore intraspecific variation was assessed by sequencing these parts for ten samples of each ‘pure’ parental species, selected according to the following criteria: (a) relative nuclear DNA content matching the range for a given species (genome size); (b) unambiguous determination on the basis of morphological characters; (c) representative geographic distribution, plants chosen from distant sites to assess intraspecific variability within the study area.

The trnL-trnF region was PCR-amplified as follows: reaction volumes of 50 µL contained 5 µL of Mg2+-free reaction buffer, 1·5 mm MgCl2, 200 µm of each dNTP, 0·5 µm of each primer (c and f; Taberlet et al., 1991), 5–10 ng of genomic DNA, and 1 unit of Taq DNA-polymerase (Fermentas, Ontario, Canada). The thermocycling profile was as follows: 94 °C/4 min, 40 × (94 °C/30 s, 53 °C/30 s, 72 °C/1·5 min), 72 °C/10 min. PCR products were purified using the QIAquick® PCR purification kit (Qiagen, Hilden, Germany) and sequenced (GATC Biotech, Konstanz, Germany) using the PCR primers. Electropherograms were edited, and alignments adjusted manually in BioEdit (Hall, 1999). Sequences representing all the variation found were deposited in GenBank (accession numbers DQ912406–10).

Because of low variability between the parental species, the trnT-trnL and rpl20-rps12 intergenic spacers were also analysed. PCR amplification of the trnT-L was as follows: reaction volumes of 50 µL contained 5 µL of Mg2+-free reaction buffer, 2·5 mm MgCl2, 200 µm of each dNTP, 1 µm of each primer (a and b; Taberlet et al., 1991), 5–10 ng of genomic DNA, and 1 unit of Taq DNA-polymerase. The thermocycling profile was: 94 °C/3 min, 35 × (94 °C/1 min, 46·5 °C/1 min, 72 °C/1 min), 72 °C/10 min. Purification, sequencing and alignment were done as above (GenBank accession numbers DQ914534–36). A single position differed for some E. intermedia samples. It created an AclI restriction site. Restriction digests were performed using 12 µL of PCR product, 5 units of AclI enzyme, and 1/10 reaction volume of Tango® buffer (Fermentas), and incubated overnight at 37 °C. The products were separated on 1·5 % agarose gels, stained with ethidium bromide, and visualized by UV. Initial screening of chloroplast haplotypes was done by PCR–RFLP and all samples not showing the E. intermedia-specific mutation in the trnT-L were sequenced for trnL-F.

The rpl20-rps12 region was amplified as described by Kaplan and Fehrer (2006) (one sample per species sequenced, GenBank accession numbers DQ914537–38).

Nuclear ribosomal DNA (ITS) analyses

Three samples of each parental species were chosen according to the criteria described above and assessed for intra/interspecific variability (GenBank accession numbers DQ859048–54). PCR amplification of the ITS region was as follows: reaction volumes of 50 µL contained 5 µL of Mg2+-free reaction buffer, 2·5 mm MgCl2, 100 µm of each dNTP, 0·2 µm of each primer (ITS 4 and ITS 5; White et al., 1990), 5–10 ng of genomic DNA, and 1 unit of Taq DNA-polymerase. The thermocycling profile was as follows: 94 °C/5 min, 35 × (94 °C/30 s, 51 °C/30 s, 72 °C/1 min), 72 °C/10 min. As this primer combination yielded some ITS sequences of an unspecified endophytic fungus (GenBank accession number DQ987703), ITS 5 was replaced with a newly designed Poaceae-specific primer (ITS-Poa-f, 5′-aaggatcattgtcgtgacg-3′) spanning the 3′ part of 18S rDNA and the 5′ end of ITS 1. PCR products were sequenced with the ITS 4 primer. The two species were distinguished by one SmaI restriction site in E. repens and one HaeIII restriction site in E. intermedia. For the purpose of RFLP analyses, PCRs were performed in triplicate and equimolar amounts of PCR products were mixed to reduce potential effects of PCR drift and to obtain a more accurate representation of parental copy types. Restriction digests were performed as above, using 10 units of enzyme and incubating overnight at 30 °C with SmaI and at 37 °C with HaeIII. All 63 hexaploid hybrids, nine nonaploids and 12 offspring plants of a nonaploid hybrid were analysed by SmaI RFLP. The nonaploids were additionally analysed by HaeIII RFLP to confirm the contribution of E. intermedia. Previously sequenced samples of each parent served as references in the RFLPs.

PCR–RFLPs were used to simultaneously detect the representation of both parental ITS types in hybrids to infer recent hybridization events. As initial digests suggested skewed ratios of parental ITS amplificates despite using replicates to avoid PCR drift (see above), the reproducibility and sensitivity of the method were verified by a semi-quantitative approach as follows: a series of SmaI RFLPs, in which separately amplified PCR products of both parents were mixed in different ratios (98 %, 95 %, 80 %, 65 %, 50 %, 35 %, 20 %, 10 %, 5 %, 2 %) was prepared prior to digest (Fig. 1). Additionally to exclude preferential amplification of one or the other parental type in hybrids, several independent amplifications with equal amounts of mixed parental DNAs were performed and they were examined by subsequent SmaI RFLPs. No preferential amplification of either ITS type was detected in PCR–RFLPs with mixed parental DNAs in PCRs (Fig. 1, M1–M3). Also, mixing of three PCR products for each sample prior to RFLPs did ensure representative amplification: replication of a subset of PCR–RFLPs confirmed the reproducibility of the patterns with respect to the relative amounts of detected parental ITS copies (not shown). The mixed ratio series (Fig. 1, 98 to 2) allowed an approximate estimation of the relative proportions of parental ITS types present in particular hybrids (e.g. H63, Fig. 1).

Fig. 1.
SmaI ITS-RFLP of artificial PCR mixtures and hybrid H63. From left: PCR products of both parents mixed in different ratios [numbers indicate proportion (%) of E. repens in each sample; letters ‘r’ and ‘i’ refer to reference ...

RESULTS

Flow cytometric analyses and chromosome counts

Initial flow cytometric analysis (relative DNA content) revealed 265 hexaploid plants among 269 plants collected at localities A–C: 123 E. repens, 112 E. intermedia and 30 E. repens × E. intermedia hybrids. Additionally, four nonaploids occurred at locality C (Table 1). Together with material from Mahelka et al. (2005), 63 hexaploid hybrids from 20 localities and nine nonaploids from four localities were analysed for absolute DNA content.

Absolute genome sizes of hexaploid hybrids are presented in Fig. 2. All had DNA content intermediate between the parents; plants H1, H2 and H63 deviated from values typical of hybrids, and their genome sizes were approaching either parent (E. repens in H1 and H2, E. intermedia in H63). All three plants were euploid hexaploids according to chromosome counts. Absolute genome sizes and chromosome numbers of nonaploids and of the progeny of one nonaploid hybrid (N7) are given in Table 2. Among this progeny, a variety of chromosome numbers was found. Nine plants with chromosome numbers 49–52 were very similar in genome size, two plants with 54 chromosomes had higher genome sizes but were different from each other, and one plant with 63 chromosomes had the highest genome size matching the range of other natural nonaploids. These results suggest backcrossing of the mother plant with hexaploids (heptaploid P1, aneuploids P2–P11) and fusion of two reduced gametes of nonaploids (either through self- or out-pollination) (nonaploid P12).

Fig. 2.
Absolute genome sizes of hexaploid E. repens × E. intermedia hybrids. Reference values of E. repens and E. intermedia are shown in black.

Chloroplast DNA analyses

Intraspecific cpDNA variability in ten accessions of E. repens was almost absent (one substitution in trnL-F), and the present samples fell well into the variation of the GenBank sequences based on North American samples. Intraspecific variability of E. intermedia was also very low (two substitutions in trnL-F, one in trnT-L).

Even interspecific variability of cpDNA was very low for all three markers. While the rpl20-rps12 intergenic spacer was invariant, only a single mutation occurred in the trnT-L region. RFLP screening revealed it in 18 hybrids with the E. intermedia chloroplast haplotype. The trnL-F sequences differed consistently between E. repens and E. intermedia only by a 5-bp indel at a tandemly repetitive site that was identified by sequencing.

In 61 cases out of 63 hexaploid hybrids, E. intermedia was found to be the maternal parent. In samples H6 and H9, the maternal plant was E. repens. Out of the nine nonaploid hybrids, E. intermedia was identified as the maternal parent in six cases (Table 2). The nonaploid's progeny expectedly had E. intermedia-like cpDNA, confirming maternal transmission of chloroplast DNA.

Nuclear ribosomal DNA (ITS) analyses

Despite their allopolyploid origin, ITS copies of all hexaploid parental plants analysed were sufficiently homogenized to provide well-readable electropherograms by direct sequencing. Apart from a few polymorphic sites within each sequence–some reflected interspecific variation, others occurred at otherwise invariant sites–there was no intraspecific variability within both species. The parental species consistently differed from each other by 15 substitutions (2·3 % sequence divergence). Thus, ITS provided a taxon-specific marker and was usable for inferring recent hybridization events.

RFLPs revealed a small portion of undigested PCR product in E. repens (Fig. 1, left arrow) despite manifold over-digestion. Direct sequencing of this undigested fragment (accession number DQ859049) showed one substitution and an adjacent 1-bp indel, both resulting in a loss of the original restriction site, which was, however, different from the E. intermedia-specific substitution. Likewise, E. intermedia samples contained a small portion (around 1%) of ITS copies that were digested with SmaI. Direct sequencing of the approx. 500-bp fragment (Fig. 1, right arrow) and BLAST search in GenBank matched an ITS sequence similar to the cloned sequence type AF507808 of Thinopyrum intermedium (= E. intermedia) (Li et al., 2004) which is very divergent from the major copy type. These minority copies, undetectable by direct sequencing of the original PCR products, were present in several samples of both Elytrigia species analysed and suggest a small amount of different ancestral genomes that were not completely homogenized. However, their amount was so low that they did not affect the detection of hybridization between the two species.

All plants determined as hexaploid hybrids by flow cytometry expectedly displayed an additive pattern of parental ITS copies (Fig. 3). Some samples showed overrepresentation of one or the other parental copy.

Fig. 3.
SmaI ITS-RFLP of hexaploid E. repens × E. intermedia hybrids. Samples are ordered according to their genome size. Letters ‘r’ and ‘i’ refer to reference samples of E. repens and E. intermedia. Approximate lengths ...

In nonaploids, restriction digest with SmaI showed additive patterns of both parental ITS copies in eight out of nine samples (Table 2). No obvious bias between parental ITS copies was detected in any of these samples (not shown). Sample N1 displayed the RFLP pattern typical of E. repens. Restriction digest with HaeIII excluded the presence of E. intermedia ITS in this sample and confirmed the others as true hybrids (not shown). Thus, out of nine nonaploids, eight were hybrids and one (N1) was a nonaploid cytotype of E. repens.

The progeny of one nonaploid hybrid (N7) was analysed further. SmaI RFLPs displayed an additive pattern in all samples, confirming the hybridogenous origin of the offspring plants. Equal or heavily biased copy numbers of both parents were found (Fig. 4).

Fig. 4.
SmaI ITS-RFLP of the progeny of the nonaploid plant N7. Hexaploid E. repens (r) and E. intermedia (i) were used as reference samples. Samples are ordered according to their chromosome numbers. Approximate lengths of the fragments are 650, 470 and 180 ...

Frequency of hybrids and habitat type

The frequency of hybrids differed among the three localities (Table 1). At sites of agricultural land-use, hybrids were common whereas none was found at the steppic locality. One parental species was very rare at both the steppic (E. repens) and the agricultural localities (E. intermedia). The high proportion of hybrids at locality B-b likely reflects sampling focused on morphological variation.

Test of fertility

Five out of 18 hexaploid hybrids and three out of five nonaploids investigated yielded well-developed caryopses. Fertility/germinability was: H6, 1·1 %/100 %; H58, 1·3 %/50 %; H56, 1·5 %/33·3 %; H22, 1·5 %/0 %; H63, 28·8 %/60 %; N5, 2·0 %/100 %; N8, 3·5 %/33·3 %; N9, 6·5 %/83·3 %. Fertility/germinability of five pure E. repens and E. intermedia samples ranged between 20·6 and 52·5 %/10 and 100 % and 10·3 and 51·9 %/80 and 90 %, respectively. Average values were 41·5 %/62 % in E. repens and 30·9 %/88% in E. intermedia.

DISCUSSION

Parental species and ITS copy homogenization

ITS sequences of the Elytrigia repens and E. intermedia specimens analysed were sufficiently homogenized to provide a taxon-specific marker. Preliminary phylogenetic analysis using the major ITS types in a context of related species (not shown) placed the two species in divergent parts of the ITS tree. Because rDNA arrays occur at several chromosomal loci in Triticeae species (Dubcovsky and Dvořák, 1995; Li and Zhang, 2002), interlocus concerted evolution leading to homogenization of the ITS sequences must have occurred in the Elytrigia parental species and apparently is nearly complete. The small amounts of unhomogenized ITS sequences detected in both species may still allow some of the ancient hybridization events that have lead to the present genome composition of both allohexaploid grasses to be traced, but this will require a cloning approach and an efficient screening strategy to identify these rare variants.

Hybrid and putative backcross identification

Identification of putative hybrids was based on the combination of two markers–genome size and additivity of parental ITS copies. While this combined approach enabled first or early generation hybrids to be determined with a high degree of certainty (additivity of ITS copies matched the results obtained by flow cytometry for all hexaploid hybrids and additionally revealed hybrid origin of most nonaploid plants), both markers suffer from limitations for inferring introgression, particularly for the following reasons. (a) Genome size was most effective in detecting F1 hybrids in the present study species, revealing a DNA content intermediate between both parents. In contrast, later-generation hybrids or backcrosses might be more problematic to detect because the hybrids' genome size will approach that of one or the other parental species. (b) Ribosomal DNA undergoes its own intragenomic evolution (Álvarez and Wendel, 2003), and homogenization of different ITS copies itself may cause an unpredictably biased representation of particular parental copies in hybrid genomes. This process has obviously happened after the initial formation of the allohexaploid parental species (resulting in mainly one ITS variant each) in the present study, but can also occur with remarkable speed in recent hybrids, even within only two generations of backcrossing (Fuertes Aguilar et al., 1999).

Flow cytometric data suggested only three candidate plants (H1, H2, H63; Fig. 2) to be potential backcrosses. Out of these, only H63 showed a congruence of genome size and ITS data. While genome size of this sample was intermediate between the values of other hybrids and E. intermedia, ITS displayed a strong bias towards the E. intermedia type, almost corresponding with the pure E. intermedia sample (Fig. 1). Especially the comparably high fertility of this plant (28·8 %) and its E. intermedia-like morphology suggest that this plant could be a later generation backcross. This is the only good example among the present data that backcrosses among hexaploids are possible and do exist (but see also the progeny of the nonaploid hybrid N7 below).

However, gene conversion without meiotic cycles (i.e. hybrids persisting vegetatively by rhizomes) is probably unlikely or less efficient to homogenize divergent ITS copies. In this case, the variable proportions of parental ITS variants among the hexaploid hybrids (Fig. 3) may have another explanation. Up to five different genome types (out of at least seven so far discovered in both parental species) whose individual rDNA loci and copy numbers might differ can be arbitrarily recombined during hybridization. On the other hand, it cannot be excluded that hybrids with copy numbers biased towards either parental species, but of intermediate genome size, may also be backcrosses or later-generation hybrids. Theoretically, they could have arisen through hybridization of two F1 hybrids of smaller and larger genome size, resulting in F2 with intermediate genome size. However, as hybrid fertility is usually low, more complex dynamics of ITS homogenization or locus loss in F1 hybrids could as well be responsible for the biased ITS copy numbers.

Another example of proven backcrosses is the progeny of the nonaploid mother plant N7. Heavily biased copy numbers of E. repens or E. intermedia ITS suggest loss of ITS loci of one or the other parent, as indicated by the samples P1–P9, which have similar genome sizes while displaying a difference of three chromosomes, and thus having different genome composition.

As long as there is no more information about the genomic processes concerning rDNA loci in both Elytrigia species available, these scenarios remain speculative. As a note of caution, several apparent contradictions between genome size and ITS ratio indicate that there is no clear and easy correlation between these approaches and that neither of them seems to be suitable to study introgression: (a) the genome size in hybrids H1 and H2 is approaching that of E. repens (Fig. 2), but is unaccompanied by ITS sequences skewed to the E. repens type (Fig. 3); (b) varying ITS ratios exist among hexaploid hybrids of similar genome size (Fig. 3); and (c) approximately equal amounts of both parental ITS variants occur in natural nonaploids of markedly different genome size (see below).

Origin of nonaploids

Nonaploids can arise by a combination of reduced (n) and unreduced (2n) gametes of parental hexaploid species. Their origin was assessed by a combination of ITS-RFLP, genome size, cpDNA and morphology (Table 2). Contrary to expectation (2 : 1 ratio of parental genomes), no obvious bias between parental ITS copies was detected, and the results were reproducible (data not shown). The reason is unclear; not much is known about the particular intragenomic processes, but they can often be unpredictable.

One plant (N1) represented a nonaploid cytotype of E. repens. For the origin of the eight nonaploid hybrids, two plausible scenarios are proposed (Table 2). Under the first, a stronger maternal influence on the morphology of the plants seems to be apparent: out of six hybrids with lower genome size, two with E. repens cpDNA (N3, N6) morphologically resembled E. repens, whereas four with E. intermedia cpDNA (N2, N4, N5, N7) were correctly identified as hybrids. They may all have arisen from 2n (E. repens) + n (E. intermedia) gametes. Two plants with higher genome size and E. intermedia morphology also had E. intermedia cpDNA (N8, N9). They may represent a composition of 2n (E. intermedia) + n (E. repens) gametes. Under the second scenario, fusion of unreduced gametes of hexaploid hybrids (with predominantly E. intermedia-like chloroplast hapotype) with reduced gametes of either parental species is considered because hybrids are partially fertile and might more easily produce unreduced gametes than pure species due to disturbed meiosis (Ramsey and Schemske, 1998). Genome sizes in both scenarios roughly match the theoretically expected values, estimated from absolute genome sizes of parental species (Mahelka et al., 2005).

The formation of another nonaploid hybrid cytotype that probably arose by fusion of a reduced gamete of E. repens and an unreduced gamete of E. pycnantha has been described by Refoufi et al. (2005). This observation suggests that the formation of unreduced gametes in Elytrigia with subsequent hybridization with other species may not be unusual.

Hybridization between nonaploid and hexaploid cytotypes

Data on the progeny of the nonaploid hybrid N7 and a certain fertility of the nonaploids examined in the experimental garden show that at least partial fertility of nonaploids should be expected. The viability of the nonaploid's offspring in nature is unknown. No such plants were found growing spontaneously at locality C, where the nonaploid mother originated. More detailed investigation of localities with nonaploid cytotypes would be desirable in this respect. But recently, a population of heptaploid (2n = 7x = 49) cytotypes intermixed with hexaploid E. intermedia was discovered at another locality (pers. obs.). Besides heptaploids, several aneuploids (2n = 47, 48, 50) were present there, too, similarly to the N7 offspring recovered from seeds collected in the field. This suggests that such cytotypes can be viable under natural conditions and some of them may persist and take part in further hybridizations. Hybridization between different cytotypes can apparently generate a large variability of geno- and cytotypes that can serve as raw material for evolution.

Ecological and evolutionary implications of hybridization

One of the most interesting aspects concerning hybridization in general is the fate of hybrids after they have arisen. There has been a long debate about hybrid fitness relative to their parents (Barton and Hewitt, 1985; Arnold and Hodges, 1995). While studies showing decreased hybrid fitness concern mostly animals, there is an increasing number of studies on plants demonstrating that hybrids can be as fit as their parents or even surpass them, at least in some environments (Arnold and Hodges, 1995; Krahulcová et al., 1996; Wang et al., 1997; Campbell and Waser, 2001; Rieseberg et al., 2003; Campbell et al., 2005; Kirk et al., 2005a, b). Hybrid fitness in these cases rather displays genotype-by-environment interactions than consistent breakdown. While fitness of the natural hybrids was not measured, the present observations suggest that it may be superior relative to the parents at some intermediate sites, such as transition zones between steppic grasslands and agricultural land. For example, at locality C no E. intermedia plants were found out of 28 plants collected; similarly, at locality B-1 no plants of this species were found among 109 plants collected, although hybrids were present there. Due to arbitrary sampling at locality C, it is possible that E. intermedia plants were missed, but it is plausible that the species is actually rare or even absent at localities B-a and C. As E. intermedia occurs predominantly on adjacent steppes, there is ample opportunity to occasionally form hybrids with E. repens. Indeed, influence of the steppic locality adjoining the third transect on the species composition was evident: the closer to the steppe, the higher the proportion of hybrids detected (Table 1). Such hybrids could benefit from acquisition of E. repens-specific adaptations to the weedy, disturbed habitats in which E. intermedia does not occur. As E. intermedia was a mother plant of almost all hybrids, it had to be present at the site at least at the time of hybrid formation. Its rare occurrence at the sites where most hybrids were found probably resulted in a scarcity of conspecific mates and an exposure to an excess of E. repens pollen. This can at least partly explain the biased directionality of the cross at localities B-a and C. Highly asymmetric hybrid formation is not unusual and can be caused by complex genotype–environment interactions (Rieseberg et al., 1991; Krahulcová et al., 1996; Campbell and Waser, 2001; Campbell et al., 2005; Kirk et al., 2005a, b). Complete cytoplasmic incompatibility can be excluded in the present case as the reciprocal cross was possible, albeit rare.

The role of hybrids in plant speciation has been an object of discussion (Rieseberg, 1997; Gross and Rieseberg, 2005). Frequency of hybridization and fertility of hybrids are among the most important aspects in this respect. The frequency of E. repens × E. intermedia hybrids differed considerably between habitat types, suggesting that different ecological conditions may play an important role in hybrid formation and/or establishment. The present study localities represent two extreme types of habitats: a natural, conserved habitat with nearly no anthropogenic disturbance and agricultural habitats with a high degree of anthropogenic disturbance. The latter habitat type with a relaxed competition is likely to have sustained hybrid formation and/or establishment. On the other hand, hybrid formation or establishment in natural steppic populations where E. intermedia is common seems to be restricted. As the current study is based on the results from only three localities, no generalization can be made, and other aspects such as the history of particular localities have to be taken into account. The present data on hybrid seed fertility and germinability under garden conditions have rather informative character as to whether hybrids and nonaploids can produce germinable seeds in principal. However, the data do indicate that at least some F2 hybrids or backcrosses may be expected in nature as well. While male fertility was not determined in the hybrids in the present study, production of viable pollen of hybrids can be high even in cases of complete seed sterility (Mráz et al., 2005). The rather frequent occurrence of hexaploid and nonaploid hybrids in the field and their partial seed fertility suggest that hardly any pre-mating and no strong post-mating reproductive barriers exist between the two Elytrigia congeners and that hybrids could mediate gene flow in this species complex.

Successful hybridization and potential introgression to one parental species may cause transfer of genetically encoded adaptation whereby genetic diversity of species may be increased (Stutz and Thomas, 1964; Arnold and Bennett, 1993; Kim and Rieseberg, 1999). For example by heterosis and transgressive segregation, hybrid phenotypes may, through new combination of alleles, exceed their parents, at least in some environments (Rieseberg et al., 2000, 2003; Campbell et al., 2005). The possible number of allele combinations in both Elytrigia species is magnified by their allopolyploid origin. Polyploidy per se is often perceived as a process facilitating evolution and adaptation, and the increased number of genetically divergent loci that may enhance environmental adaptability is one of the most often discussed advantages of polyploids (Wendel, 2000, and references therein). According to the preliminary data on genome composition, the two Elytrigia species share only one genome, donated by Pseudoroegneria (Assadi and Runemark, 1995; Chen et al., 1998). Theoretically, the hybrid between E. repens and E. intermedia combines genetic material from up to seven different donor species. F1 hybrids between the two Elytrigia species contain a full genomic complement of both parents and thus their genetic pool may be enriched. Namely E. intermedia is known to possess many valuable traits, such as biotic and abiotic resistances, wherefore it is often used in wheat improvement (Fedak, 1999; Fedak and Han, 2005). Although E. repens is rather unexplored in this respect, its ecological amplitude is even wider than that of the former species. Mahelka (2006) showed that the response of the E. repens × E. intermedia hybrids to flooding tended to be intermediate between that of the parents. This was likely caused by enhanced rhizome production inherited from highly rhizomatous E. repens. Such an adaptation may gain high importance after ecological conditions at a locality have changed, e.g. during local floods, which are currently becoming more frequent as a consequence of low-tillage management, especially on heavy soils. In this respect, enhanced rhizome formation in hybrids compared with E. intermedia would likely be an adaptive advantage also in habitats frequently disturbed by tillage or ploughing because rhizomes as storage organs maintain damaged plants viable if fragmented and even allow further propagation. Vegetative propagation may also be important in cases of low fertility, such as in hybrids. Survival of plants at a locality for many years through vegetative propagation increases the chance of hybridization in the future, because (a) multiplication of individuals increases the probabilities simply in a mathematical way; and (b) local ecological conditions change through time whereby the chance to meet a compatible sexual counterpart increases. Moreover, via cultivation of fields, fragmented rhizomes may be easily transported over hundreds of metres from the place where they originated, increasing the chance of finding a suitable place for establishment and sexual partner to mate.

Conclusions and perspectives

In conclusion, it can be stated that E. repens and E. intermedia frequently cross at places where they co-occur. Hybrid frequency is likely to be influenced by habitat type; sites disturbed by human influence sustain hybrid formation and/or establishment. Hexaploid and nonaploid hybrid fertility is not negligible, backcrossing is possible, and the progeny is variable. These processes generate a high diversity of cyto- and genotypes that may adapt to different environmental conditions. In the light of weak reproductive barriers and frequent natural hybridization between these species, it cannot be ruled out that introgression may also have contributed to shape the ecological amplitudes of the species.

To further elucidate this, necessary prerequisites are (a) to study the (multiple?) origin and genomic composition of both parental species across the study area in more detail, e.g. by the application of single copy nuclear gene markers and/or in situ hybridization techniques, and (b) to develop a fingerprinting system that allows the identification of vegetatively propagated plants (clones). Based on that knowledge, it will be possible to develop appropriate, sufficiently sensitive markers for studying introgression.

ACKNOWLEDGEMENTS

We cordially thank two anonymous reviewers and handling editor Prof. C. A. Buerkle for valuable suggestions concerning the manuscript. We also thank Bohumil Mandák for reading the manuscript and for helpful comments and Fred Rooks for language editing. Marie Stará, L'udmila Tereková and Petr Jurkovský are appreciated for technical assistance in the DNA laboratory. Pavel Trávníček and Jan Suda kindly assisted us in flow cytometric analyses. Eva Ibermajerová, Eva Morávková and Eva Slívová are appreciated for their help in the experimental garden. This study was supported by the Czech Science Foundation (grants 206/05/0778 and 206/03/H137) and the Grant Agency of the Academy of Sciences of the Czech Republic (AV0Z60050516).

LITERATURE CITED

  • Abbott RJ. Plant invasions, interspecific hybridization and the evolution of new plant taxa. Trends in Ecology and Evolution. 1992;7:401–405. [PubMed]
  • Álvarez I, Wendel JF. Ribosomal ITS sequences and plant phylogenetic inference. Molecular Phylogenetics and Evolution. 2003;29:417–434. [PubMed]
  • Arnold ML. Natural hybridization and evolution. New York, NY: Oxford University Press; 1997.
  • Arnold ML, Bennett BD. Natural hybridization in Louisiana irises: genetic variation and ecological determinants. In: Harrison RG, editor. Hybrid zones and the evolutionary processes. Oxford: Oxford University Press; 1993. pp. 115–139.
  • Arnold ML, Hodges SA. Are natural hybrids fit or unfit relative to their parents? Trends in Ecology and Evolution. 1995;10:67–71. [PubMed]
  • Ashton PA, Abbott RJ. Multiple origins and genetic diversity in the newly arisen allopolyploid species, Senecio cambrensis Rosser (Compositae) Heredity. 1992;68:25–32.
  • Assadi M. Biosystematic studies of the Elymus hispidus (Poaceae: Triticeae) group in Iran. Nordic Journal of Botany. 1998;18:483–491.
  • Assadi M, Runemark H. Hybridisation, genomic constitution and generic delimitation in Elymus s. l. (Poaceae: Triticeae) Plant Systematics and Evolution. 1995;194:189–205.
  • Barkworth ME, Dewey DR. Genomically based genera in the perennial Triticeae of North America: identification and membership. American Journal of Botany. 1985;72:767–776.
  • Barton NH, Hewitt GM. Analysis of hybrid zones. Annual Review of Ecology and Systematics. 1985;16:113–148.
  • Berchtold GF, Opiz PM. Oekonomisch-technische Flora Böhmens. Prag: JH Pospischil; 1836. I/2.
  • Campbell DR, Waser NM. Genotype-by-environment interaction and the fitness of plant hybrids in the wild. Evolution. 2001;55:669–676. [PubMed]
  • Campbell DR, Galen C, Wu CA. Ecophysiology of first and second generation hybrids in a natural plant hybrid zone. Oecologia. 2005;144:214–225. [PubMed]
  • Chen Q, Conner RL, Laroche A, Thomas JB. Genome analysis of Thinopyrum intermedium and Thinopyrum ponticum using genomic in situ hybridization. Genome. 1998;41:580–586. [PubMed]
  • Chen Q, Conner RL, Laroche A, Ahmad F. Molecular cytogenetic evidence for a high level of chromosome pairing among different genomes in Triticum aestivum–Thinopyrum intermedium hybrids. Theoretical and Applied Genetics. 2001;102:847–852.
  • Chytrý M, Tichý L. Diagnostic, constant and dominant species of vegetation classes and alliances of the Czech Republic: a statistical revision. Folia Facultatis Scientiarum Naturalium Universitatis Masarykianae Brunensis. 2003;108:1–231.
  • Dewey DR. The genomic system of classification as a guide to intergeneric hybridization with the perennial Triticeae. In: Gustafson JP, editor. Gene manipulation in plant improvement. New York, NY: Plenum; 1984. pp. 209–279.
  • Doležel J, Sgorbati S, Lucretti S. Comparison of three DNA fluorochromes for flow cytometric estimation of nuclear DNA content in plants. Physiologia Plantarum. 1992;85:625–631.
  • Dubcovsky J, Dvořák J. Ribosomal RNA multigene loci: nomads of the Triticeae genomes. Genetics. 1995;140:1367–1377. [PMC free article] [PubMed]
  • Fedak G. Molecular aids for integration of alien chromatin through wide crosses. Genome. 1999;42:584–591.
  • Fedak G, Han F. Characterization of derivatives from wheat–Thinopyrum wide crosses. Cytogenetic and Genome Research. 2005;109:360–367. [PubMed]
  • Franke R, Nestrowicz R, Senula A, Staat B. Intergeneric hybrids between Triticum aestivum L. and wild Triticeae. Hereditas. 1992;116:225–231.
  • Friebe B, Gill KS, Tuleen NA, Gill BS. Transfer of wheat streak mosaic virus resistance from Agropyron intermedium into wheat. Crop Science. 1996;36:857–861.
  • Fuertes Aguilar J, Rosselló JA, Nieto Feliner G. Nuclear ribosomal DNA (nrDNA) concerted evolution in natural and artificial hybrids of Armeria (Plumbaginaceae) Molecular Ecology. 1999;8:1341–1346. [PubMed]
  • Gray AJ, Marshall DF, Raybould AF. A century of evolution in Spartina anglica. Advances in Ecological Research. 1991;21:1–62.
  • Greilhuber J, Doležel J, Lysák MA, Bennett MD. The origin, evolution and proposed stabilization of the terms ‘genome size’ and ‘C-value’ to describe nuclear DNA contents. Annals of Botany. 2005;95:255–260. [PubMed]
  • Gross BL, Rieseberg LH. The ecological genetics of homoploid hybrid speciation. Journal of Heredity. 2005;96:241–252. [PMC free article] [PubMed]
  • Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series. 1999;41:95–98.
  • Han FP, Fedak G, Benabdelmouna A, Armstrong K, Oullet T. Characterization of six wheat × Thinopyrum intermedium derivatives by GISH, RFLP, and multicolor GISH. Genome. 2003;46:490–495. [PubMed]
  • Kaplan Z, Fehrer J. Comparison of natural and artificial hybridization in Potamogeton. Preslia. 2006;78:303–316.
  • Kellogg EA, Appels R, Mason-Gamer RJ. When gene trees tell different stories: the diploid genera of Triticeae. Systematic Botany. 1996;21:312–347.
  • Kim SC, Rieseberg LH. Genetic architecture of species differences in annual sunflowers: implications for adaptive trait introgression. Genetics. 1999;153:965–977. [PMC free article] [PubMed]
  • Kirk H, Vrieling K, Klinkhamer GL. Maternal effects and heterosis influence the fitness of plant hybrids. New Phytologist. 2005a;166:685–694. [PubMed]
  • Kirk H, Vrieling K, Klinkhamer GL. Reproductive fitness of hybrids between Senecio jacobaea and S. aquaticus (Asteraceae) American Journal of Botany. 2005b;92:1467–1473. [PubMed]
  • Kishii M, Wang RR-C, Tsujimoto H. GISH analysis revealed a new aspect of genomic constitution of Thinopyrum intermedium. Czech Journal of Genetics and Plant Breeding. 2005;41:91–95.
  • Krahulcová A, Krahulec F, Kirschner J. Introgressive hybridization between a native and an introduced species: Viola lutea subsp. sudetica versus V. tricolor. Folia Geobotanica and Phytotaxonomica. 1996;31:219–244.
  • Krahulec F, Kaplan Z, Novák J. Tragopogon porrifolius × T. pratensis: the present state of an old hybrid population in Central Bohemia, the Czech Republic. Preslia. 2005;77:297–306.
  • Kubát K, Hrouda L, Chrtek J, Jr, Kaplan Z, Kirschner J, Štěpánek J. Klíč ke květeně České republiky [Key to the Flora of the Czech Republic] Praha: Academia; 2002.
  • Li D-Y, Zhang X-Y. Physical localization of the 18S-5·8S-26S rDNA and sequence analysis of ITS regions in Thinopyrum ponticum (Poaceae: Triticeae): implications for concerted evolution. Annals of Botany. 2002;90:445–452. [PubMed]
  • Li D-Y, Ru Y-Y, Zhang X-Y. Chromosomal distribution of the 18S-5·8S-26S rDNA loci and heterogeneity of nuclear ITS regions in Thinopyrum intermedium (Poaceae: Triticeae) Acta Botanica Sinica. 2004;46:1234–1241.
  • Liu Z-W, Wang RR-C. Genome analysis of Elytrigia caespitosa, Lophopyrum nodosum, Pseudoroegneria geniculata ssp. scythica, and Thinopyrum intermedium (Triticeae: Gramineae) Genome. 1993;36:102–111. [PubMed]
  • Löve Á. Conspectus of the Triticeae. Feddes Repertorium. 1984;95:425–521.
  • Mahelka V. Response to flooding intensity in Elytrigia repens, E. intermedia (Poaceae: Triticeae) and their hybrid. Weed Research. 2006;46:82–90.
  • Mahelka V, Suda J, Jarolímová V, Trávníček P, Krahulec F. Genome size discriminates between closely related taxa Elytrigia repens and E. intermedia (Poaceae: Triticeae) and their hybrid. Folia Geobotanica. 2005;40:367–384.
  • Mandák B, Bímová K, Pyšek P, Štěpánek J, Plačková I. Isoenzyme diversity in Reynoutria (Polygonaceae) taxa: escape from sterility by hybridization. Plant Systematics and Evolution. 2005;253:219–230.
  • Mason-Gamer RJ. Reticulate evolution, introgression, and intertribal gene capture in an allohexaploid grass. Systematic Biology. 2004;53:25–37. [PubMed]
  • Mason-Gamer RJ, Orme NL, Anderson CM. Phylogenetic analysis of North American Elymus and the monogenomic Triticeae (Poaceae) using three chloroplast DNA data sets. Genome. 2002;45:991–1002. [PubMed]
  • Mason-Gamer RJ, Burns MM, Naum M. Polyploidy, introgression, and complex phylogenetic patterns within Elymus. Czech Journal of Genetics and Plant Breeding. 2005;41:21–26.
  • Melderis A. In: Elymus. In. Tutin TG, Heywood VH, Burges NA, et al., editors. Vol. 5. Cambridge: Cambridge University Press; 1980. pp. 192–198. Flora Europaea.
  • Mizianty M, Szczepaniak M. Remarks on the Agropyron–Elymus complex (Poaceae) with special reference to its representatives in Poland. Fragmenta Floristica et Geobotanica. 1997;42:215–225.
  • Mizianty M, Frey L, Szczepaniak M. The Agropyron–Elymus complex (Poaceae) in Poland: biosystematics. In: Frey L., editor. Studies on grasses in Poland. Krakow: PAN; 2001. pp. 25–77.
  • Mráz P, Chrtek J, Fehrer J, Plačková I. Rare recent natural hybridization in Hieracium s. str.–evidence from morphology, allozymes and chloroplast DNA. Plant Systematics and Evolution. 2005;255:177–192.
  • Ownbey M. Natural hybridization and amphiploidy in the genus Tragopogon. American Journal of Botany. 1950;37:487–499.
  • Palmer JH, Sagar GR. Agropyron repens (L.) Beauv. Biological flora of the British Isles. Journal of Ecology. 1963;51:783–794.
  • Prokudin YN, Druleva IV. Pro gibridnu prirodu piriyu zagostrenogo [Elytrigia mucronata (Opiz) Prokudin]. Ukrainskii Botanichnii Zhurnal. 1971;28:712–717.
  • Ramsey J, Schemske DW. Pathways, mechanisms, and rates of polyploid formation in flowering plants. Annual Review of Ecology and Systematics. 1998;29:467–501.
  • Refoufi A, Esnault M-A, Levasseur J-E. Characterization of a novel 9-ploid hybrid (2n = 63) with four genomes in an Elytrigia complex (Poaceae) Botanical Journal of the Linnean Society. 2005;147:501–508.
  • Rieseberg LH. Hybrid origins of plant species. Annual Review of Ecology and Systematics. 1997;28:359–389.
  • Rieseberg LH, Beckstrom-Sternberg S, Doan K. Helianthus annuus ssp. texanus has chloroplast DNA and nuclear ribosomal RNA genes of Helianthus debilis ssp. cucumerifolius. Proceedings of the National Academy of Sciences of the USA. 1990;87:593–597. [PMC free article] [PubMed]
  • Rieseberg LH, Choi HC, Ham D. Differential cytoplasmic versus nuclear introgression in Helianthus. Journal of Heredity. 1991;82:489–493.
  • Rieseberg LH, Baird SJ, Gardner KA. Hybridization, introgression, and linkage evolution. Plant Molecular Biology. 2000;42:205–224. [PubMed]
  • Rieseberg LH, Raymond O, Rosenthal DM, Lai Z, Livingstone K, Nakazato T, et al. Major ecological transitions in wild sunflowers facilitated by hybridization. Science. 2003;301:1211–1216. [PubMed]
  • Sharma HC, Gill BS. New hybrids between Agropyron and wheat. 2. Production, morphology and cytogenetic analysis of F1 hybrids and backcross derivatives. Theoretical and Applied Genetics. 1983;66:111–121. [PubMed]
  • Sharma H, Ohm H, Goulart L, Lister R, Appels R, Benlhabib O. Introgression and characterization of barley yellow dwarf virus resistance from Thinopyrum intermedium into wheat. Genome. 1995;38:406–413. [PubMed]
  • Soltis PS, Plunkett GM, Novak SJ, Soltis DE. Genetic variation in Tragopogon species: additional origins of the allotetraploids T. mirus and T. miscellus (Compositae) American Journal of Botany. 1995;82:1329–1341.
  • Stebbins LG. The significance of polyploidy in plant evolution. American Naturalist. 1940;74:54–66.
  • Stebbins L. Taxonomy and the evolution of genera, with special reference to the family Gramineae. Evolution. 1956;10:235–245.
  • Štorchová H, Hrdličková R, Chrtek J, Jr, Tetera M, Fitze D, Fehrer J. An improved method for DNA isolation from plants collected in the field and conserved in saturated NaCl/CTAB solution. Taxon. 2000;49:79–84.
  • Stutz HC, Thomas LK. Hybridization and introgression in Cowania and Purshia. Evolution. 1964;18:183–195.
  • Taberlet P, Gielly L, Pautou G, Bouvet J. Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology. 1991;17:1105–1109. [PubMed]
  • Tsitsin NN. Novyi vid i novye raznovidnosti pshenitsy [New species and varieties of wheats] Byulleten Glavnogo Botanicheskogo Sada AN SSSR. 1960;38:38–41.
  • Tsvelev N. Conspectus specierum tribus Triticeae Dum. familiae Poaceae in flora URSS. Novosti Systematiki Vysshikh Rastenii. 1973;10:19–59.
  • Wang H, McArthur ED, Sanderson SC, Graham JH, Freeman DC. Narrow hybrid zone between two subspecies of big sagebrush (Artemisia tridentata: Asteraceae). IV. Reciprocal transplant experiments. Evolution. 1997;51:95–102.
  • Wendel JF. Genome evolution in polyploids. Plant Molecular Biology. 2000;42:225–249. [PubMed]
  • White TJ, Bruns T, Lee S, Taylor J. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis M, Gelfand D, Sninsky J, White T, editors. PCR protocols: a guide to methods and applications. San Diego, CA: Academic Press; 1990. pp. 315–322.

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

Formats:

Related citations in PubMed

See reviews...See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...