Hybridization in the Subtribe Alopecurinae Dumort. (Poaceae) According to Molecular Phylogenetic Analysis: Different Ploidy Level Tells Different Origin of the Groups

Abstract

We performed next-generation sequencing of the 18S rDNA–ITS1–5.8S rDNA region along with traditional Sanger sequencing of rbcL, matK, ndhF, and ITS1–5.8S rDNA–ITS2 to clarify the hybridization pattern in the subtribe Alopecurinae and in the genus Alopecurus in particular. Our data support the hybrid origin of Alopecurus × brachystylus from hybridization between A. geniculatus (sect. Alopecurium) and A. pratensis (sect. Alopecurus). Moreover, in the rDNA of hybrid A. × brachystylus, only A. aequalis-like ribotypes from tetraploid A. geniculatus participated. Surprisingly, we found the traces of introgression of A. arundinaceus-like ribotypes not only in hybrid A. × marssonii (A. geniculatus × A. arundinaceus) but in A. aequalis s. str. as well. A high-polyploid group from the section Alopecurus, A. aggr. alpinus has undoubted hybrid origin: e. g., A. brachystachyus has rDNA from the sect. Alopecurium. Alopecurus alpinus, with its allies, is clearly distinct from other members of the sect. Alopecurus (especially by maternal line) and thus we can re-establish a previous opinion about the separate group to which A. alpinus belongs. Species from the section Colobachne (presumably Alpine grasses from Ancient Mediterranean region) probably hybridized with the A. alpinus group. Even A. myosuroides (sect. Pseudophalaris) that could be referred to the separate genus has ribotypes common with the species of the section Alopecurium (A. aequalis, A. geniculatus) in one of the accessions. Additionally, we found that the possible polyphyletic origin of the genus Limnas. Limnas stelleri is very close to Alopecurus magellanicus according to NGS data, while L. malyschevii is more or less distinct from other studied species of the genus Alopecurus.

1. Introduction

The subtribe Alopecurinae belongs to the largest tribe of the grass family, Poeae, and is widespread in the temperate zone of both Hemispheres. It comprises the genera Alopecurus L. and Limnas Trin. according to the most recent data and is closely related to the genera Beckmannia Host and Rhizocephalus Boiss. [1]. The subtribe Alopecurinae is characterized by very dense, spike-shaped panicles of regular cylindrical form [2]. Alopecurus, the main genus of the subtribe, comprises up to 70 species [2]. It is known for drastic differences in the chromosome numbers of the species: from 2n = 14 in Alopecurus aequalis Sobol. to 2n = ca. 100 in A. alpinus Vill. [3]. This polyploid range most probably originated via interspecific hybridization, which is rather common in the Poeae tribe [1,4,5]. Hybrids in the genus Alopecurus have been noted for a long time, and nothospecies have been described beginning from A. × brachystylus Peterm. (A. geniculatus L. × A. pratensis L.).

Recent phylogenetic research of the subtribe Alopecurinae confirmed its separation from the subtribe Phleinae Dumort. that has similar inflorescences and the affinity of Alopecurinae with some genera from other subtribes, in particular Arctophila (Rupr.) Andersson, Dupontia R.Br., Arctagrostis Griseb., and, only according to the nuclear gene sequence data, Arctopoa (Griseb.) Prob. [1,5,6]. Thus, inflorescence traits of the Alopecurinae members can be explained by parallel evolution when similar structures form independently in two lines of taxa originated from the common ancestor. Additionally, the similar inflorescence has the genus Lagurus L. [2] which belongs to wholly different group in the Poeae tribe—Aveninae J.Presl (Aveneae chloroplast group) [1].

Recently, the genus Alopecurus is divided into four sections: Alopecurus, Colobachne (P. Beauv.) Griseb., Alopecurium Dumort., and Pseudophalaris Tzvelev [2,7]. The section Alopecurus differs from other species by their cylindrical to ellipsoidal panicles and absence of palea [2]. This section also can be divided into two groups: tetra- and hexaploid species of A. aggr. pratensis and A. arundinaceus Poir., and high polyploid Siberian and Arcto-Alpine species A. brachystachyus M.Bieb., A. alpinus, A. borealis Trin., A. glaucus Less. with related taxa. The section Colobachne contains mostly Alpine grasses with short, wide-ellipsoidal panicles and mostly present palea. Members of the sect. Alopecurium have an annual or shortly lived perennial habit with geniculate stems and the sect. Pseudophalaris is distinguished by annual habit and glumes with winged keel [2].

Molecular phylogenetic analysis revealed some lines within Alopecurus that did not correspond with taxonomic division of the genus [1,6] and pointed at the intersectional hybridization events. Previous research of Alopecurus mainly covered American and Middle Eastern species [1,6,8,9]. Meanwhile, Asian mountains, for example, Altai Mountain Country as well as Caucasus could provide us many hybrid taxa that have not been analyzed until now (see [10] about the plant speciation centers). New methods of next-generation sequencing (NGS) allow the detection of multiple hybridization events in plants even when their morphology is uniform. Our goal of this work is to trace the hybridization events in the subtribe Alopecurinae (especially in the genus Alopecurus, taking into account hybrid mountain species from Altai-Sayan region) that could shed light on phylogenetic relationships in the subtribe, and possibly its relationships with other taxa of the tribe Poeae. For our analysis, we used the sequences of the region 18S rDNA–ITS1–5.8S rDNA obtained by NGS and the sequences of ITS1–5.8S rDNA–ITS2, matK, rbcL, and ndhF obtained via standard Sanger sequencing.

2. Results

Since hybridization is known to occur frequently in the tribe Poeae, we treated chloroplast and ITS datasets separately. In order to trace possible hybridization events and assess geographical variability, we analyzed multiple samples of the same species for some chloroplast datasets.

Phylogenetic tree built on concatenated rbcL + matK dataset shows two well-supported clades where the studied samples of the genus Alopecurus fall (Figure 1). The first clade (PP = 1, BS = 100) includes section Alopecurium (A. aequalis) and high polyploid members of the section Alopecurus (A. alpinus, A. brachystachyus). This clade is, in turn, sister to Beckmannia syzigachne with low resolution (PP = 0.56, BS = 70). The second clade (PP = 1, BS = 93) contains tetra- and exaploidy species from the sections Alopecurus (A. pratensis + A. arundinaceus), Colobachne (A. glacialis K.Koch, A. ponticus K.Koch), and one sample of A. brachystachyus from Irkutsk Oblast, Eastern Siberia, Russia. A. laguroides Balansa has an uncertain position in this large clade. Alopecurus myosuroides (sect. Pseudophalaris) falls into a separate clade (PP = 1, BS = 100) that forms a polytomy with the previously discussed clade of Alopecurus, and Apera Adans. (PP = 0.78, BS = 70). The second clade, in addition, has weakly supported grouping with the subtribe Phleinae (PP = 0.72, BS = 53). The overall placement of the Alopecurinae species on the tree built by matK sequence data separately is almost the same (Supplementary Figure S2). But in this case intersectional hybrid A. × brachystylus (A. geniculatus × A. pratensis) falls within the clade including section Alopecurium (we have no rbcL data for A. × brachystylus and thus did not include it in the analysis of concatenated sequences). Limnas stelleri Trin. has an uncertain position on the separate rbcL tree (Supplementary Figure S1) but we do not have matK sequences of L. stelleri for a combined data tree. In phylogenetic analysis based on ndhF sequences, we used fewer samples (Figure 2). The Alopecurinae clade groups with Bellardiochloa variegata (Lam.) Kerguélen (PP = 0.99, BS = 87), whereas Beckmannia syzigachne branches separately (PP = 0.99, BS = 87) within the large clade containing the genus Alopecurus, Hookerochloa hookeriana (F.Muell. ex Hook.f.) E.B.Alexeev, Festucella eriopoda (Vickery) E.B.Alexeev, and Cinna latifolia (Trevir. ex Göpp.) Griseb. (PP = 0.99, BS = 87) (Figure 2). The Alopecurus clade (PP = 0.94, BS = 82) divides into sect. Alopecurus (A. alpinus + A. brachystachyus + A. magellanicus Lam.) + sect. Alopecurium subclade (PP = 0.95, BS = 91) and a subclade that contains sect. Alopecurus (A. pratensis + A. arundinaceus) + sect. Colobachne (A. textilis Boiss.) (PP = 0.78, BS = 93) while A. myosuroides has a sister position to all other species.

ITS-inferred phylogeny presents a weakly supported clade of Beckmannia + Apera (PP = 0.67, BS = 71), subtribe Alopecurinae, and Cinna L. + Arctopoa (Griseb.) Prob. (PP = 0.63, BS = 55) (Figure 3). This clade forms polytomy with monophyletic Phleinae and Hookerochloa E.B.Alexeev + Saxipoa Soreng, L.J.Gillespie & S.W.L.Jacobs + Sylvipoa Soreng, and L.J.Gillespie & S.W.L.Jacobs + Arctagrostis Griseb. (weakly supported, PP = 0.64). The subclade of Alopecurinae s. str. (without Beckmannia) is supported only according to Bayesian analysis (PP = 0.54, BS unsupported). Rhizocephalus orientalis Boiss. is a sister to all other taxa in the Alopecurinae subclade, though only with weak support in Bayesian inference (PP = 0.54, BS unsupported). Part of the sect. Colobachne (A. brevifolius (G.Westb.) Grossh., A. glacialis, A. textilis) falls into one clade (PP = 1, BS = 99) and occupies a sister position to the clade containing other species of Alopecurus and Limnas + Arctophila (Rupr.) Andersson + Dupontia R.Br. according to Bayesian analysis (PP = 0.70). The clade Limnas + Arctophila + Dupontia is very weakly supported (PP = 0.51, BS = 57). It was discovered as the clade that is the sister to the clade-comprising sect. Alopecurus, A. vaginatus (Willd.) Trin. (sect. Colobachne), the sample of A. myosuroides (sect. Pseudophalaris) from Dagestan Republic, Northern Caucasus, Russia, and sect. Alopecurium (PP = 0.86). In addition, A. brachystachyus (high- polyploid species from the section Alopecurus) is a sister to the clade comprising section Alopecurium and does not group with its possible relative, A. alpinus. One sample of A. myosuroides from Latvia is found in the clade corresponding to the section Alopecurium, according to ITS analysis.

We analyzed the marker sequences of 18S rDNA–ITS1–5.8S rDNA obtained via NGS to investigate the hidden multiple hybridization. When the sequence was present in more than 10,000 reads per rDNA pool, we calculated the percentage for them and took these sequences as major. In other cases, when the total quantity of reads was below 10,000, we calculated their percentage for the sequences that occurred in 100 and more reads.

The ribotype networks built in TCS 1.21 and visualized in TCSBU are presented in Figure 4, Figure 5, Figure 6, Figure 7 and Figure 8. The first of the networks depicts the origin of hybrid A. × brachystylus (A. geniculatus × A. pratensis) and A. × marssoni (A. geniculatus × A. arundinaceus) with regard to the parental taxa, e. g., diploid A. aequalis (2n = 14) (Figure 4). In addition, we took into analysis A. vlassovii Trin.—a high polyploid (2n = 120) of the sect. Alopecurus, a relative of A. alpinus. We took the main ribotype of A. aequalis as the consensus sequence. Looking at the network, we can distinguish three ribotype subnetworks corresponding to the sections Alopecurium, Alopecurus, and the high-polyploid group (A. vlassovii). Alopecurus aequalis, a diploid species, has four major ribotypes (in this case, upper 100 reads per rDNA pool) (Figure 4). Its main ribotype, Ae1 (473 reads, 22%), is identical to the main ribotype of A. geniculatus (1809 reads, 34%), the main ribotypes of hybrids A. × brachystylus and A. × marssonii Hausskn. (5828 reads, 22%, and 465 reads, 22%, respectively). Ribotype Ae1 also has the minor ribotype fraction of A. vlassovii and A. brachystachyus (below 1%). The second major ribotype of A. aequalis, Ae2 (169 reads, 8%), is common with the second major ribotype of A. × marssonii (164 reads, 8%). Ribotype Ae3 (161 read, 7%) of A. aequalis also corresponds to the second major ribotype of A. × brachystylus (2331 read, 9%) as well as to the third major ribotype of A. × marssonii (158 reads, 7%), and minor ribotype of A. geniculatus (862 reads, 16%). The fourth major ribotype of diploid A. aequalis—Ae4 (140 reads, 6%) is placed between “Alopecurium” and “Alopecurus” subnetworks. This ribotype is shared with the fourth major ribotype of hybrid A. × marssonii (133 reads, 6%) and minor ribotype of A. × brachystylus (150 reads). The second major ribotype of tetraploid A. geniculatus is species-specific (G, 1072 reads, 20%). Surprisingly, the main ribotype of A. brachystachyus (Br, 3485 reads, 47%) also belongs to the “Alopecurium” subnetwork. It is also the second major ribotype of A. vlassovii (2279 reads, 11%). The main ribotype of A. vlassovii is species-specific (Vl1, 5248 reads, 24%) and forms the separate subnetwork along with its derivative, Vl2 (1006 reads, 5%). The second major ribotype of A. pratensis unites with this subnetwork (Pr2, 2763 reads, 11%). The main ribotype of A. pratensis (Pr1, 5561 read, 21%) groups with “Alopecurus” subnetwork and is homologous to the fifth major ribotype of hybrid A. × brachystylus (1087 reads, 4%). The third major ribotype of A. pratensis, Pr3 (1928 reads, 7%), is specific as well as the fourth (Pr4, 1863 reads, 7%) and the sixth (Pr6, 1049 reads, 4%) ones. Unlike this, the fifth major ribotype of A. pratensis (Pr5, 1416 reads, 5%) is identical to the third major ribotype of A. × brachystylus (2095 reads, 8%) and minor ribotype of A. arundinaceus (41 read). Two major ribotypes of A. arundinaceus, Ar1 and Ar2 (2864 reads, 18%, and 2504 reads, 15%, respectively), form a separate cluster within the “Alopecurus” subnetwork. The main ribotype Ar1 is specific, and the second major ribotype is shared with the minor ribotypes of A. aequalis (75 reads) and A. × marssonii (73 reads). The fourth major ribotype of A. × brachystylus (B, 1722 reads, 7%) is species-specific.

The second ribotype network concerns the relationships between section Alopecurium, a high-polyploid group of the type section (A. alpinus, A. magellanicus), section Pseudophalaris (A. myosuroides) and the genus Limnas (Figure 5). The studied species of the section Alopecurium do not have the ribotypes of A. alpinus and A. magellanicus in the rDNA pool. The major ribotypes of A. alpinus and A. magellanicus (Al and Am, respectively) are mostly species-specific except for the main ribotype of A. magellanicus (Am1, 2051 read, 40%) that is shared with the minor ribotype of A. alpinus (274 reads). The sample of A. pratensis Alt 16-337 has four major ribotypes forming own subnetwork (Pr1, Pr2, Pr5, Pr4) that are identical to the four ribotypes of the A. pratensis sample from the previous network. Limnas stelleri, a species from the subtribe Alopecurinae, has ribotypes that are close to the Alopecurus magellanicus/A. alpinus subnetwork. Major ribotypes of this species can be called L. The ribotypes of the samples of A. myosuroides have no connection with other samples of Alopecurus in this network, according to NGS data. The main ribotype of A. myosuroides (My1, 10,044 reads, 56%) is identical with that of A. myosuroides var. breviaristatus Asch. & Graebn. (3292 reads, 27%). Other major ribotypes (more than 1000 reads in this case) are present only in A. myosuroides var. breviaristatus (My2-2, 1971 read, 12%, My2-3, 1478 reads, 9%). They are common with minor ribotypes of A. myosuroides.

The third ribotype network shows the phylogenetic structure of A. aggr. pratensis (Figure 6). The main ribotype of A. pratensis s. str. from Altai Krai, Russia (Pr1, 5561 reads, 21%) is shared with the second major ribotype of A. pratensis subsp. alpestris (Wahlenb.) Selander (2889 reads, 11%), the second major ribotype of A. pratensis from Teberda (4775 reads, 22%), and the second major ribotype of A. pratensis from the White Sea (1732 reads, 11%). The second major ribotype of A. pratensis from Altai Krai (Pr2, 2763 reads, 11%) is specific. The third major ribotype of A. pratensis s. str. (Pr3, 1928 reads, 7%) is common with the minor ribotype of Caucasian A. pratensis from Teberda (18 reads). The fourth major ribotype of A. pratensis, Pr4 (1863 reads, 7%) is common with fourth major ribotype of A. pratensis subsp. alpestris (1611, 5%). The fifth major ribotype (Pr5, 1416 reads, 5%) is identical to the main ribotype of A. pratensis from Teberda (5359 reads, 24%), the main ribotype of A. pratensis from the White Sea (3153 reads, 21%), and the third major ribotype of A. pratensis subsp. alpestris (1903 reads, 7%) while the sixth major ribotype, Pr6 (1049 reads, 4%) is shared with minor ribotype of A. pratensis from the White Sea (49 reads). The third major ribotype of A. pratensis from Teberda (Pr7, 3539 reads, 16%) is specific as well as the third major ribotype of A. pratensis from the White Sea (Pr8, 1041 read, 7%), and the main ribotype of A. pratensis subsp. alpestris (Pa1, 9301, 37%) is common with minor fraction of A. pratensis (547 reads, below 1%).

The next picture describes the relationships between the studied samples of the sections Alopecurus (A. pratensis s. l.) and Colobachne (Figure 7). ITS1 sequences of A. pratensis s. str. samples from the White Sea and A. pratensis subsp. alpestris obtained by NGS group separately from those of A. ponticus. The main ribotype of A. pratensis from the White Sea is the same as the main ribotype of A. pratensis subsp. alpestris (Pr1). The main ribotype of A. pratensis subsp. alpestris (Pa1, 9301, 37%) is related to the ribotypes of A. pratensis but is specific on this scheme. The fourth major ribotype of A. pratensis subsp. alpestris (1611, 5%) belongs to the group Pr and is identical to the ribotype Pr4 (see above). The ribotype structure of the species of the sect. Colobachne is diverse. A. brevifolius sample from Teberda (T89) forms the peculiar subnetwork that is only distantly related to the sect. Alopecurus. Its ribotypes, Bf1–Bf4 are specific. Ribotypes of two samples of A. ponticus from Teberda (T57, T59) are almost identical except for the fourth major ribotype of A. ponticus T59 (Po4) that is shared with the minor ribotype of A. ponticus T57.

3. Discussion

When a species has substantially discordant positions between the cpDNA and nrDNA phylogenies, there is a possibility that the species may be a hybrid [11,12,13,14]. Moreover, intraspecific genome polymorphism is widely used in research for interspecific and, in some cases, intraspecific phylogenetic reconstructions, especially in cases when introgression is involved [15,16,17,18]. For example, some genera of the tribe Triticeae Dumort. (Poaceae) appeared to be of intergeneric origin themselves [19,20,21]. Molecular phylogenetic analysis of different gene sets that allowed to trace possible hybridization events proposed generic names based on the genome combinations that the species has [22]. Other works presented new data of multiple polyploid origin of many Hordeum L. species (sometimes auto- and allopolyploid species were in the same aggregate) [21].

The genus Alopecurus belongs to the subtribe Alopecurinae (tribe Poeae s. l.), which is known for widespread hybridization. Molecular phylogenetic studies of the genus showed polyphyletic placement of the previously described sections Alopecurus and Colobachne [7,23]. Some replacements were also made on the species level that can have, in fact, different explanations [7,23]. Researching the ribotype composition of Alopecurus and some taxa from allied genera, we found many ribotypes due to the active hybridization and high polyploidy in the subtribe Alopecurinae.

In the genus Alopecurus, four nothospecies were described that are usually sterile and rather rare [2]. We analyzed two of them, Alopecurus × brachystylus and A. × marssonii. A. × brachystylus is an intersectional hybrid according to taxonomists [24,25]. One of the putative parent species, the Euro–Mediterranean–South Asian A. geniculatus, belongs to the section Alopecurium, the members of which are distinguished by their characteristic narrow cylindrical panicles and geniculate stems. The second parent taxon, the Euro–Siberian–Central Asian A. pratensis, belongs to the type section, which includes species with predominantly wide cylindrical or ellipsoidal panicles and straight stems. Parental species, growing in the same habitats, can hybridize, usually forming sterile offspring [26,27]. A. × brachystylus was not previously studied by molecular phylogenetic methods. Our data confirm this statement but add some details. On the maternal side, A. × brachystylus probably originated from A. geniculatus (Figure 5). Our NGS analysis shows the presence of A. pratensis ribotypes and ribotypes that are common with diploid A. aequalis (also belonging to the sect. Alopecurium). Tetraploid A. geniculatus (2n = 28), in its turn, could originate from diploid A. aequalis and an unknown diploid progenitor (Figure 4). The second possible interpretation is that A. geniculatus has the second ribotype that passed the stages of post-hybridization transformation. Thus, the hybrid genome of A. × brachystylus has only A. aequalis ribotypes from an A. geniculatus parent. A. × marssonii is a hybrid between A. geniculatus and Predominantly Euro–Siberian–Caucasian–Central Asian meadow–coastal weakly halophilic species A. arundinaceus. Ribotype structure of A. × marssonii indicates its origin from A. geniculatus (hybrid with A. aequalis as parental species). A. arundinaceus-related ribotypes occurred only in a minor fraction of A. × marssonii. But, surprisingly, a minor ribotype fraction that is common with A. arundinaceus was also found in A. aequalis ribotypes (Figure 4). This unexpected fact can point at the cases of introgression that took place in in the fairly distant past. All trees built on the sequence data, obtained by the Sanger method, clearly show distinction of the sect. Alopecurium.

South Siberian mountain species A. vlassovii, which belongs to the high polyploid group of the sect. Alopecurus (A. aggr. borealis), has poly- and aneuploid chromosome numbers, 2n = ca. 98–130, ca. 120, and ca. 150, without predominance of any one of them [28,29,30]. According to the NGS data, A. vlassovii has two subgenomes that correspond to the A. borealis group and A. brachystachyus (being close to the sect. Alopecurium) (Figure 4). High polyploidy of A. vlassovii is clearly a result of multiple hybridization and introgression within the group and probably between the different sections. A. vlassovii can be a member of a peculiar introgressive–hybridization complex of species [31] belonging to the affinity group of A. alpinus. A. vlassovii is possible hybrid of A. alpinus s. l. and A. brachystachyus. The latter species, according to rbcL and matK sequence analysis, has different maternal taxa. The majority of A. brachystachyus are relatives of A. alpinus s. l. and other high-polyploid mountain species but some samples occupy an uncertain position (Figure 1). ITS data obtained by the Sanger method, along with the NGS data, show the relationship between A. brachystachyus and A. aequalis + A. geniculatus.

R. Soreng et al. [32] synonymized the Antarctic species Alopecurus magellanicus, Holarctic mountain-tundra species A. alpinus, and A. borealis, North Pacific mountain-meadow A. stejnegeri Vasey and endemic to the Ural Mountains, weakly halophilic A. glaucus. In our opinion, we cannot accept such a broad interpretation of A. magellanicus, since these undoubtedly closely related species nevertheless differ in both morphological and molecular genetic characteristics. A comparison of herbarium material of these species with specimens of A. magellanicus (South Georgia and the South Sandwich Islands) did not reveal absolute similarity between them. In contrast, A. magellanicus has its own species-specific characters not found in other closely related species. For example, A. magellanicus has larger spikelets (4–5 mm long) than A. alpinus, less abundant pubescence of the glumes, more developed awns, and the presence of a membranous border at the tips of the lemmas. Subantarctic A. magellanicus and Arctic A. alpinus have their own specific ribotypes (Figure 5). Only a minor ribotype fraction of A. alpinus is shared with A. magellanicus. A. magellanicus is a close relative of A. alpinus and A. brachystachyus according to ndhF sequence analysis. A. alpinus, as an ancestor of A. magellanicus, could spread across Cordilleras, this being the classic example of interpolar disjunction (see also [33]).

Alopecurus myosuroides, belonging to the sect. Pseudophalaris, stands apart from other species of the genus. The main distinction of A. myosuroides is its annual life form and the winged keel on its glumes. We see that A. myosuroides also stands separately according to the molecular phylogenetic data (Figure 1, Figure 2, Figure 3, Figure 4 and Figure 6). Morphological separateness along with distinction of the sequences can point at possible generic specificity of A. myosuroides. But one sample of A. myosuroides from Latvia falls into a clade with A. aequalis and A. geniculatus (sect. Alopecurium). This unusual position can reflect the ancient homoploid hybridization.

Another interesting fact is the position of Limnas. The genus Limnas is a member of subtribe Alopecurinae and is distinguished by more or less loose panicles and leathery-membranous glumes [2]. There are two endemic species of Limnas in Russia: Eastern Siberian Limnas stelleri and Eastern Siberian/Far Eastern Limnas malyschevii. It is interesting that Limnas stelleri turned out to be more or less closely related with North American Alopecurus alpinus and Subantarctic A. magellanicus (Figure 5). The close relationship of Limnas and Alopecurus species was previously identified by Soreng [23], using ITS data obtained by the Sanger method (Figure 3). The second species, L. malyschevii, is weakly related to the clade Dupontia + Arctophila by ITS data and is sister to Alopecurus clade by ndhF sequences (Figure 2). Such unusual placement of two Limnas species could be due to the intergeneric reticulation or even polyphyletic origin of the genus.

Molecular phylogenetic data often help in distinguishing different species in one large geographically heterogeneous complex. For example, analysis of chloroplast and nuclear (ITS) genes revealed a different position of Malaysian samples of Spiranthes sinensis (Pers.) Ames (Orchidaceae Juss.) compared with other East Asian samples [34]. They were considered as crypto-hybrids [34]. In grasses (Poaceae), even samples of Hyalopoa pontica (Balansa) Tzvelev from different but closely located gorges of Greater Caucasus (Teberda, Carachay-Cherkessia Republic) can fall into different clades [35]. Alopecurus pratensis has almost a worldwide area, occurring in many non-tropical areas. It is tetraploid with 2n = 28 [2]. It forms a group of related taxa: A. pratensis s. str., A. pratensis subsp. alpestris, A. arundinaceus, and A. brachystachyus. A. pratensis subsp. alpestris is an Arctic or hypoarctic plant with glaucous tinge of ligules and glaucous leaves. Our NGS data confirm that A. pratensis subsp. alpestris can be can be distinguished as a separate species because it has specific main ribotype (Figure 6). In addition, A. pratensis from Northern Caucasus (Teberda) and A. pratensis from the White Sea coast have peculiar major ribotypes but not main one. It can be related with the geographical variability of A. pratensis s. l. (Figure 8). In addition, geographical variability was detected by chloroplast sequence data (rbcL and matK). Samples from Stavropol Krai and Altai Republic differ from those from Irkutsk Oblast and other samples from Altai Republic (Figure 1). It is possible that Alopecurus pratensis is heterogeneous by the maternal line. A. brachystachyus, on the contrary, has different affinity according to the chloroplast sequence data; it is closer to the Arctic and Sub-Arctic high polyploid species but also can have an A. pratensis-like maternal genome (Figure 1, see above).

Species from the section Colobachne differ from the other Alopecurus members by wide ellipsoid or shortly cylindrical panicles and by predominant presence of palea [2,24]. The center of their diversity is Ancient Mediterranean and Caucasus region ([36], Figure 9).

They form polyploid range from 2n = 14 [30] to 2n = 56 [37]. We need to note that A. brachystachyus (sect. Alopecurus) was previously treated as the species of the Vaginatae group (=sect. Colobachne) as well [36]. Molecular phylogenetic analysis clearly confirms distinction of this section from high-polyploid relatives of A. alpinus and from A. brachystachyus and comparative unity of the section Colobachne on the maternal side (rbcL and matK, Figure 1 and Figure 2). At the same time, the nrDNA of the Colobachne species indicates probable hybridizations: A. vaginatus falls into a clade with A. alpinus and allied taxa (Figure 3) and is distant from other members of the sect. Colobachne. Additionally, the section Colobachne (excl. A. vaginatus) is rather distant form all other Alopecurus members based on ITS sequences (Figure 3). According to NGS data, studied species of the section Colobachne are fairly distinct from each other and do not form common ribotypes (Figure 7). We can assume that species of the sect. Colobachne probably arose from intersectional hybridization but are rather stabile themselves.

The position of two genera, Beckmannia and Rhizocephalus, remains controversial. Earlier, these genera were placed near the genus Alopecurus in the tribe Phleeae [24,38]. These genera were placed in the special subtribe Beckmanniinae. Our data present Beckmannia as the genus allied to the subtribe Alopecurinae that is closer to the latter on the chloroplast trees (Figure 1 and Figure 2). Rhizocephalus orientalis occupies an uncertain position being sister to Alopecurinae only by Bayesian method according to the ITS data (PP = 0.54) (Figure 3).

Thus, our data prove developed hybridization in the subtribe Alopecurinae, not only in the sterile nothospecies. The degree of introgressive hybridization processes varies depending on the sections and groups; e.g., some Alopecurus species do not hybridize with closely related taxa according to NGS data but have plastid DNA from different sections. We tend to re-establish the high polyploid group of Alopecurus previously named as the group Alpinae [36] that now is placed within the sect. Alopecurus. The genus Limnas is probably polyphyletic.

4. Materials and Methods

For our molecular phylogenetic analysis, we took species from all sections of the genus Alopecurus, paying special attention to the potential hybrid taxa. We included also species from other genera of the subtribe Alopecurinae s. l.: Beckmannia, Limnas, Rhizocephalus as well. We analyzed chloroplast sequences of the region matK of 53 samples of 13 species, rbcL in 61 samples of 12 species, ndhF of 19 samples of 11 species, and 27 ITS sequences of 16 species (all obtained by Sanger method). Then, we concatenated rbcL and matK sequences of the studied species when there were both sequences of the same species. The sequences of rbcL + matK regions were presented for 48 samples of 10 species. We paid more attention to the hybrid species and polyploid species that can have some differences along their habitat. Information about the species studied via Sanger method is given in

Table 1. Sequences of the species of subtribe Alopecurinae and related subtribes obtained in the present study by the Sanger method and their numbers in GenBank.

Species	Country of Origin	Genbank Number, rbcL	Genbank Number, matK	Genbank Number, ndhF	Genbank Number, ITS
Alopecurus aequalis	Russia: Yakutia Republic				PP048962
Alopecurus aequalis	Russia: Kurgan Oblast				PP048965
Alopecurus aequalis	Russia: Irkutsk Oblast				PP048966
Alopecurus aequalis	Russia: Altai Republic	PP060934	OR997845
Alopecurus aequalis	Russia: Altai Republic	PP060935	OR997846
Alopecurus aequalis	Russia: Irkutsk Oblast	PP060936	OR997847
Alopecurus aequalis	Russia: Irkutsk Oblast	PP060937
Alopecurus aequalis	Russia: Altai Republic	PP060938	OR997848
Alopecurus aequalis	Russia: Irkutsk Oblast		OR997849
Alopecurus aequalis	Russia: Altai Republic	PP060939	OR997850
Alopecurus aequalis	Russia: Irkutsk Oblast	PP060940
Alopecurus aequalis	Russia: Irkutsk Oblast	PP060941	OR997851
Alopecurus aequalis	Russia: Leningrad Oblast	PP060942	OR997852		PP048963
Alopecurus aequalis	Russia: Irkutsk Oblast	PP060943
Alopecurus aequalis	Russia: Krasnodar Krai	PP060944	OR997853		PP048964
Alopecurus aequalis	Russia: Irkutsk Oblast		OR997854
Alopecurus aequalis	Russia: Altai Republic	PP060945	OR997855
Alopecurus alpinus	Russia: Altai Republic			PP053715	PP048961
Alopecurus alpinus	Russia: Tuva Republic			PP053716
Alopecurus alpinus	Russia: Altai Republic	PP060946	OR997856
Alopecurus alpinus	Russia: Altai Republic	PP060947	OR997857
Alopecurus alpinus	Russia: Altai Republic	PP060948	OR997858
Alopecurus alpinus	Russia: Altai Republic	PP060949	OR997859
Alopecurus alpinus	Russia		OR997860
Alopecurus alpinus	Russia: Altai Republic	PP060950	OR997861
Alopecurus alpinus	Russia: Altai Republic	PP060952	OR997862
Alopecurus alpinus	Russia: Altai Republic	PP060953	OR997863
Alopecurus alpinus	Russia: Altai Republic	PP060954	OR997864
Alopecurus alpinus	Russia: Altai Republic	PP060955	OR997865
Alopecurus arundinaceus	Russia: Yakutia Republic			PP053720
Alopecurus arundinaceus	Russia: Irkutsk Oblast	PP060956	OR997866
Alopecurus arundinaceus	Russia	PP060957	OR997867
Alopecurus arundinaceus	Russia: Buryatia Republic	PP060958	OR997868
Alopecurus arundinaceus	Russia	PP060959	OR997869
Alopecurus arundinaceus	Russia	PP060960	OR997870
Alopecurus arundinaceus	Russia: Irkutsk Oblast	PP060961	OR997871
Alopecurus arundinaceus	Russia: Irkutsk Oblast	PP060962	OR997872
Alopecurus arundinaceus	Russia: Irkutsk Oblast	PP060963	OR997873
Alopecurus brachystachyus	Russia: Irkutsk Oblast	PP060964	OR997874	PP053718
Alopecurus brachystachyus	Russia: Irkutsk Oblast	PP060965
Alopecurus brachystachyus	Russia: Irkutsk Oblast	PP060966	OR997875
Alopecurus brachystachyus	Russia	PP060967
Alopecurus brachystachyus	Russia	PP060968	OR997876
Alopecurus brachystylus	Russia: Irkutsk Oblast		OR997877
Alopecurus brachystylus	Russia: Irkutsk Oblast		OR997878
Alopecurus glacialis	Russia: Kabardino-Balkaria	PP060969
Alopecurus glacialis	Russia: Kabardino-Balkaria	PP060970	OR997879		PP048960
Alopecurus laguroides	Russia: Dagestan Republic	PP060971	OR997880
Alopecurus myosuroides	Russia: Stavropol Krai	PP060972
Alopecurus myosuroides	Russia: Stavropol Krai	PP060973	OR997881
Alopecurus myosuroides	Russia: Stavropol Krai	PP060974	OR997882
Alopecurus myosuroides	Russia: Stavropol Krai	PP060975
Alopecurus myosuroides	Russia: Stavropol Krai	PP060976
Alopecurus pratensis	Russia: Altai Republic			PP053717
Alopecurus pratensis	Russia: Altai Republic	PP060977	OR997883
Alopecurus pratensis	Russia: Altai Republic	PP060978	OR997884
Alopecurus pratensis	Russia	PP060979	OR997885
Alopecurus pratensis	Russia: Irkutsk Oblast	PP060980	OR997886
Alopecurus pratensis	Russia	PP060981	OR997887
Alopecurus pratensis	Russia: Irkutsk Oblast	PP060982	OR997888
Alopecurus pratensis	Russia: Irkutsk Oblast	PP060983	OR997889
Alopecurus pratensis	Russia	PP060984
Alopecurus pratensis	Russia: Irkutsk Oblast	PP060985	OR997890
Alopecurus pratensis	Russia: Altai Republic	PP060986	OR997891	PP053719
Alopecurus pratensis	Russia: Altai Republic	PP060987	OR997892
Alopecurus pratensis	Russia: Lipetsk Oblast	PP060988	OR997893
Alopecurus pratensis	Russia: Stavropol Krai	PP060989
Alopecurus pratensis	Russia	PP060990	OR997894
Alopecurus ponticus	Russia: Stavropol Krai	PP060991	OR997895
Alopecurus ponticus	Russia: Karachay-Cherkessia	PP060951
Alopecurus ponticus	Russia: Karachay-Cherkessia	PP060992	OR997896
Alopecurus textilis	Armenia				PP048958
Alopecurus textilis	Armenia				PP048959
Beckmannia syzigachne	Russia: Altai Republic			PP053722
Beckmannia syzigachne	Russia: Altai Republic	PP060993	OR997897
Limnas malyschevii	Russia: Yakutia Republic			PP053721
Limnas stelleri	Russia: Khakassia Republic	PP060994
Milium effusum	China			PP053726
Milium effusum	China			PP053727
Phleum alpinum	Tajikistan				PP048954
Phleum alpinum	Kazakhstan			PP053724
Phleum phleoides	Russia: Altai Republic				PP048955
Phleum phleoides	Russia: Altai Krai				PP048956
Phleum phleoides	Russia: Altai Republic				PP048957
Phleum pratense	Russia: Ryazan Oblast			PP053723
Phleum pratense	Latvia			PP053725

. For NGS analysis, we used 23 species of the subtribe Alopecurinae. The list of studied species is presented in

Table 2. Summary of the Alopecurinae species used in the present NGS study and their major ribotypes.

Species	Sample ID	Country of Origin	Accession Number in Genbank Database	Total Number of Reads	Ribotype Symbol	Number of Reads	% from the Total Number of the Reads
Alopecurus aequalis	M36	Russia: Altai Krai, Mamontovsky District	PP056744–PP056764	2173	Ae1	473	22
					Ae2	169	8
					Ae3	161	7
					Ae4	140	6
Alopecurus alpinus	L22	Canada: Franklin district, Banks Island	PP056764–PP054539	10,709	Al1	3709	35
					Al2	2396	22
Alopecurus arundinaceus	22	Russia: Khakassia Republic	PP056966–PP057011	16,316	Ar1	2864	18
					Ar2	2504	15
Alopecurus brachystachyus	M13	Russia: Zabaykalsky Krai	PP057607–PP057638	7373	Br	3485	47
Alopecurus × brachystylus	21	Russia: Novgorod Oblast	PP056765–PP056849	26,133	Ae1	5828	22
					Ae3	2331	9
					Pr5	2095	8
					B	1722	7
					PR1	1087	4
Alopecurus brevifolius	L21	Russia: Karachay-Cherkessia, Teberda	PP057450–PP057540	20,957	Bf1	7380	35
					Bf2	1879	9
					Bf3	1643	8
					Bf4	1124	5
Alopecurus geniculatus	L18	Ukraine: Lviv Oblast	PP056850–PP056879	5263	Ae1	1809	34
					G	1072	20
Alopecurus magellanicus	L14	South Georgia	PP056879–PP057291	5142	Am1	2051	40
					Am2	447	9
					Am3	235	5
Alopecurus × marssonii	M11	Finland	PP057250–PP057269	1880	Ae1	465	22
					Ae2	164	8
					Ae3	158	7
					Ae4	133	6
Alopecurus myosuroides	20	Russia: Dagestan Republic	PP054402–PP054507	18,721	My1	10,044	56
Alopecurus myosuroides var. breviaristatus	L16	Russia: Krasnodar Krai	PP057377–PP057405	12,275	My1	3292	27
					My2-2	1971	12
					My2-3	1478	9
Alopecurus ponticus	L20	Russia: Karachay-Cherkessia, Teberda	PP057430–PP057449	9229	Po1	2234	24
					Po2	1691	18
					Po3	1319	14
Alopecurus ponticus	M8	Russia: Karachay-Cherkessia, Teberda	PP057577–PP057606	13,553	Po1	3562	26
					Po2	3143	23
					Po3	1406	10
					Po4	1142	8
Alopecurus pratensis	23	Russia: Altai Krai	PP057012–PP057116	26,172	Pr1	5561	21
					Pr2	2763	11
					Pr3	1928	7
					Pr4	1863	7
					Pr5	1416	5
					Pr6	1049	4
Alopecurus pratensis	M5	Russia: Karachay-Cherkessia, Teberda	PP057117–PP057186	22,148	Pr5	5359	24
					Pr1	4775	21
					Pr7	3539	16
Alopecurus pratensis	M6	Russia: Arkhangelsk Oblast, Solovetsky Islands	PP057187–PP057249	15,362	Pr5	3153	21
					Pr1	1732	11
					Pr8	1041	7
Alopecurus pratensis	L30	Russia: Altai Republic	PP057541–PP057576	14,946	Pr1	2965	20
					Pr2	2863	19
					Pr5	1485	10
					Pr4	1228	8
Alopecurus pratensis subsp. alpestris	L15	Russia: Komi Republic	PP057292–PP057376	25,290	Pa1	9301	37
					Pr1	2889	11
					Pr5	1903	7
					Pr4	1611	5
Alopecurus vlassowii	25	Russia: Altai Republic	PP056880–PP056965	21,628	Vl1	5248	24
					Br	2279	11
					Vl2	1006	5
Limnas stelleri	M17	Russia: Irkutsk Oblast	PP057639–PP057654	1704	L1	364	21
					L2	303	18
					L3	195	11
					L4	145	9
					L5	80	5

.

In the case of 10,000 reads per rDNA pool, we assumed the ribotypes more than 1000 reads per rDNA pool to be major. When the total quantity of reads was below 10,000, we took the sequences of ribotypes that occurred in 100 or more reads as menomic DNA, which was extracted from leaf material with the aid of a Qiagen Plant Mini Kit (Qiagen Inc., Hilden, Germany), according to the instruction manual. The fragments were amplified and sequenced at the Center for Shared Use “Genomic Technologies, Proteomics, and Cell Biology” of the All-Russian Research Institute of Agricultural Microbiology on an Illumina Platform MiSeq. PCR was performed in 15 μL of the reaction mixture containing 0.5–1 unit of activity of Q5^® High-Fidelity DNA Polymerase (NEB, Ipswich, MA, USA), 5 pM of forward and reverse primers, 10 ng of DNA template, and 2 nM of each dNTP (Life Technologies, ThermoScientific, Waltham, MA, USA). The fragments were amplified under the following conditions: initial denaturation at 94 °C for 1 min, followed by 25 cycles of 94 °C for 30 s, 55 °C for 30 s, 72 °C for 30 s, and a final elongation 72 °C for 5 min using ITS 1P [39] and ITS 2 [40] primers. PCR products were then purified according to the Illumina recommended method using AMPureXP (Beckman Coulter, Indianapolis, IN, USA). The libraries for sequencing were prepared according to the manufacturer’s MiSeq Reagent Kit Preparation Guide (Illumina) (http://web.uri.edu/gsc/files/16s-metagenomic-library-prep-guide-15044223-b.pdf (accessed on 6 September 2022)). They were sequenced on an Illumina MiSeq instrument (Illumina, San Diego, CA, USA) using a MiSeq^® ReagentKit v3 (600 cycles) with double-sided reading (2 × 300 n) following the manufacturer’s instructions. The sequences were trimmed with Trimmomatic [41], included in Unipro Ugene [42] as follows: PE reads, sliding window trimming with size 4, quality threshold 12, and minimal read length 130. Further, paired marker sequences were combined, dereplicated, and sorted into the ribotypes by vsearch 2.7.1 [43]. The resulting sequences represent ribotypes with certain frequency in the whole genome pool. These sequences were analyzed by TCS 1.21 [44]. The network built by algorithms of statistic parsimony was visualized and processed in TCSBU [45].

Forward and reverse sequences of ITS and chloroplast regions obtained by Sanger method were observed in Chromas Lite 2.1 (https://technelysium.com.au/wp/chromas/, accessed on 1 October 2021) and concatenated in MEGA XI [46]. The sequencing was performed according to the standard protocol provided with a BigDyeTM Terminator Kit ver. 3.1 set of reagents on the sequencer ABI PRIZM 3100 sequencer at the Center for the collective use of scientific equipment “Cellular and molecular technologies for the study of plants and fungi” of the Komarov Botanical Institute, St. Petersburg. They were aligned by Muscle [47] implemented in MEGA XI [46]. Evolutionary models were computed using jModelTest 2.1.10 [48]. For the ITS dataset, we obtained GTR + I + G, for matK TPM3uf + G, for ndhF, the evolutionary model was TVM + I + G, and for rbcL we computed TVM + G. Bayesian inference was performed using Mr. Bayes 3.2.2 [49] as follows: 3–5 million of a generation, sampling trees every 100 generations, and the first 25% trees were discarded as burn-in. ML analysis was conducted by iqtree 1.6.12 (http://www.iqtree.org/, accessed 5 March 2023) under the fast bootstrap option, 1000 generations. The resulting trees combined Bayesian and ML data. The first index was posterior probability, and the second was a bootstrap index.