The section Multicaulia is the largest clade in the genus Hedysarum L. (Fabaceae). The species included in this section are distributed mainly in Central Asia (Malyshev et al., 2012) and are actively used in medicine because they contain various biologically active compounds (Vysochina et al., 2011; Dong et al., 2013; Liu Y. et al., 2019). The high content of mangiferin (a natural xanthone) in Hedysarum gmelinii Ledeb., H. grandiflorum Pall., and H. setigerum Turcz. ex Fisch. et Meyer makes them promising raw materials for obtaining antiviral and antibacterial agents (Neretina et al., 2002; Kuprina and Lutsky, 2014; Imachuyeva et al., 2020). In addition, some Hedysarum species are able to provide high increases in biomass in summer, and they are valuable forage plants (Fedtschenko, 1948; Sa et al., 2010; Malyshev et al., 2012). To preserve the gene pool of rare and valuable species of Hedysarum, their reintroduction was performed, and also synthetic populations and in vitro collections were developed (Maslova et al., 2019; Avramova and Cherepanova, 2020; Suprun et al., 2020).

The taxonomy and phylogeny of the sect. Multicaulia, similar to the entire genus Hedysarum, are still ambiguous. According to phylogenetic analyses using nuclear (ITS) and plastid DNA sequences (trnL—trnF and matK), two main lineages in the genus Hedysarum, the Hedysarum s.s. clade (Gamotion) and the Sartoria clade (Multicaulia), are presented (Choi and Ohashi, 2003; Duan et al., 2015; Liu et al., 2017; Nafisi et al., 2019). The Hedysarum sect. Multicaulia includes three subsections Multicaulia, Subacaulia, and Crinifera (Choi and Ohashi, 2003). However, some taxonomists are assigned Subacaulia and Multicaulia to different sections (Fedtschenko, 1948; Malyshev et al., 2012). Investigation of species genome diversity within the genus Hedysarum by means of molecular AFLP and ISSR markers detected a high level of intraspecific genetic polymorphism (Marghali et al., 2005; Bushman et al., 2007; Zvyagina and Dorogina, 2013; Qiang et al., 2018). Considerable variability of morphological features observed in the species of the section Multicaula often prevents their accurate identification, especially in the areas where their ranges overlap. The population structure analysis in East European (H. grandiflorum Pall., H. biebersteinii Zertova, and H. argyrophyllum Ledeb.) and South Siberian (H. setigerum and H. gmelinii) species of the sect. Multicaulia, performed based on the molecular genetic (ISSR) markers, did not clearly distinguish species with overlapping ranges (Schanzer and Suprun, 2012; Zvyagina et al., 2016).

In species of the sect. Multicaulia, a basic chromosome number of x = 8 was mostly revealed using monochrome chromosome staining (Choi and Ohashi, 2003; Arslan et al., 2012; Zvyagina et al., 2016; Kumar et al., 2018). Besides, in the karyotype of diploid Siberian species H. sangilense Krasnob., supernumerary small chromosomes were detected (Krogulevich and Rostovtseva, 1984). However, for some species, such as the endemic H. dahuricum Turcz. ex B. Fedtsch., growing in Eastern Siberia and Mongolia (Kurbatsky, 1994), chromosome numbers have not been determined yet. The analysis of chromosome C-banding patterns, performed in the karyotype of H. coronarium L. originating from North-East Algeria, revealed three types of bands (terminal, intercalary, and pericentromeric) (Issolah et al., 2006). In four Algerian species, Hedysarum carnosum Desf., H. spinosissimum L., H. pallidum Desf., and H. naudinianum Coss., 35S rDNA loci, localized in one chromosome pair, and different numbers and position of 5S rDNA clusters (in one or two pairs) were detected by fluorescent in situ hybridization (FISH) (Benhizia et al., 2021).

The taxonomic status of H. gmelinii and H. setigerum, having overlapping ranges and high morphological similarity, is still unclear, and H. setigerum is sometimes considered as a subspecies of closely related H. gmelinii (H. gmelinii spp. setigerum) or as a separate species (Fedtschenko, 1948; Kurbatsky, 1994; Sa et al., 2010). In H. setigerum and H. gmelinii, chromosome numbers varied greatly in different reports, and chromosome numbers 2n = 14, 16, 28, 32, 48, and 56 were revealed (Gatsuk, 1967; Plennik and Rostovtseva, 1977; Malahova and Kurbatsky, 1992; Philippov et al., 2008; Cherkasova, 2009). Recently, FISH with 35S rDNA and 5S rDNA, performed in several species of the sect. Hedysarum, has demonstrated that 35S rDNA and 5S rDNA clusters could be effective molecular chromosomal markers facilitating precise identification of morphologically similar species of the sect. Hedysarum (Yurkevich et al., 2021). Also, various tandem repeats are applied as chromosomal markers to detect intra- and interspecific diversities in plant genomes, chromosomal rearrangements, and the evolutionary pathways in various taxa, including Fabaceae species (Pamponét et al., 2019; Ávila Robledillo et al., 2020; Campomayor et al., 2021; Waminal et al., 2021). The use of such molecular chromosomal markers for karyotype analyses in species from the sect. Multicaulia will make it possible to specify their taxonomy, chromosome numbers, and ploidy status.

Due to the diversity of repetitive DNA sequences, plant genomes vary greatly in composition and size (McCann et al., 2020). Comparative repeatome analyses in related species provide new insight into the organization and divergence of their genomes (Liu Q. et al., 2019; McCann et al., 2020; Zwyrtková et al., 2020). For the understanding of species relationships within the sect. Multicaulia, further studies of genomic diversity are needed. In particular, intra- and interspecific variabilities in composition and genomic organization of transposable elements and satDNA should be explored in different Hedysarum species as well as in accessions from various growing areas.

To explore intra- and interspecific genome diversities and clarify the species relationships within the section Multicaulia, we performed a comparative characterization of repeatomes of H. grandiflorum, H. zundukii Peschkova, and H. dahuricum, which included the genome-wide bioinformatic analysis by RepeatExplorer/TAREAN pipelines. We also carried out FISH mapping of the identified tandem DNAs on chromosomes of these species and also three other related species of the sect. Multicaulia from different regions of Eurasia. Additionally, we explored available data on these species distributions within Eurasia to construct the integrated schematic map of their habitats.

The widespread genus Hedysarum involves many valuable medicinal and fodder species (Fedtschenko, 1948; Sa et al., 2010; Dong et al., 2013; Liu Y. et al., 2019). Most species of the sect. Multicaulia are distributed in southern Siberia and Central Asia. The range of the studied species of the sect. Multicaulia is divided into two non-overlapping areas: (1) H. grandiflorum, H. razoumowianum, and H. gmelinii grow in the East-European region, and H. zundukii, H. dahuricum, H. setigerum, and H. gmelinii distributed in the South-Siberian region, i.e., H. gmelinii can grow in both regions and its range overlaps with all studied species. Hedysarum grandiflorum, H. razoumowianum, and H. zundukii are considered to be rare and endangered taxa (KMK Publ, 2008). As shown on the constructed map, the species of the sect. Multicalia can occupy both wide and very narrow areas. Hedysarum grandiflorum is a rare but wide-ranging species, whereas both H. razoumowianum and H. zundukii occupy narrow areas. The ranges of H. setigerum and H. gmelinii are partially overlapped, and morphological similarities between these closely related species made controversial to their taxonomy (Fedtschenko, 1948; Kurbatsky, 1994; Sa et al., 2010).

It was previously reported that x = 8 is the basic chromosome number for species of the sect. Multicaulia (Arslan et al., 2012; Choi and Ohashi, 2003; Duan et al., 2015; Liu et al., 2017; Nafisi et al., 2019). In this study, we confirmed this basic chromosome number for the studied species, including H. dahuricum, which karyotype was explored for the first time. In species from the sect. Multicaulia, variations in the number of chromosomes in karyotypes were previously described with the use of monochrome staining (Gatsuk, 1967; Plennik and Rostovtseva, 1977; Malahova and Kurbatsky, 1992; Philippov et al., 2008; Cherkasova, 2009). We also detected diploid, tetraploid, and hexaploid karyotypes in H. gmelinii specimens from different geographical regions. At the same time, in the studied H. setigerum specimens, we revealed only tetraploid karyotypes although variations in the number of chromosomes in karyotypes (14, 32, and 48) were previously reported for this species (Gatsuk, 1967; Plennik and Rostovtseva, 1977). Moreover, we revealed small supernumerary B chromosomes in H. zundukii, H. gmelinii, and H. setigerum. B chromosomes were previously revealed in the karyotype of a diploid Siberian species H. sangilense (Krogulevich and Rostovtseva, 1984). These chromosomes are characterized by a mosaic distribution within the population as well as in individual plants (Camacho et al., 2000; Houben et al., 2014; D’Ambrosio et al., 2017). In our study, the number of B chromosomes also varied in karyotypes of H. zundukii, H. gmelinii, and H. setigerum. Besides, we often detected dispersed 35S rDNA and, sometimes, revealed satDNA hybridization signals on these supernumerary chromosomes. It was earlier reported that B chromosomes could contain rDNA genes and/or tandem repeats (D’Ambrosio et al., 2017; Marques et al., 2018; Ebrahimzadegan et al., 2019). The performed comparative molecular cytogenetic analysis and also detection of B chromosomes in the karyotypes of H. zundukii, H. gmelinii, and H. setigerum allowed us to establish the number of chromosomes in the main sets of these species.

In karyotypes of all studied diploid species, we revealed one chromosome pair bearing a major 35S rDNA cluster and one pair with a 5S rDNA cluster. At the same time, the chromosomal distribution of these molecular markers differed from that revealed in the species of the sect. Hedysarum (Yurkevich et al., 2021). The analysis of chromosome morphology, as well as patterns of chromosomal distribution of 35S rDNA, allowed us to confirm the tetraploid and/or hexaploid nature of H. setigerum and H. gmelinii specimens and also demonstrate close genome relationships among studied species of the sect. Multicaulia.

The performed comparative bioinformatic analysis of repeatomes of the species from the sect. Multicaulia in this study demonstrated a high similarity in repeatome composition of H. grandiflorum, H. zundukii, and H. dahuricum and also established common features in their repeatomes. In genomes of eukaryotes, retrotransposons (Class I) are the most abundant transposable elements. Within the legume family, their composition can vary in different species due to the predominant number of Ty1 Copia (Istvánek et al., 2014; Jegadeesan et al., 2021) or Ty3-Gypsy elements (Kreplak et al., 2019; Lonardi et al., 2019). In the studied species of the sect. Multicaulia, mobile elements of Class I made up the majority of their repetitive DNA (20–24%), and Ty3-Gypsy retroelements were almost 1.5–2 times more abundant when compared to Ty1-Copia elements. Retrotransposons are known to be replicated with the copy and paste mechanisms, and they can be accumulated in nuclear genomes (Bennetzen and Wang, 2014). The plant species having small genome sizes contain fewer LTR retrotransposons compared to plants with large genomes (Vitte and Bennetzen, 2006; Bennetzen and Wang, 2014; Wang et al., 2021). Therefore, variations in genome sizes, observed within legume family, can be explained by variability in the content of LTR retrotransposons (Macas et al., 2015; Ellis and Vershinin, 2020). In diploid Hedysarum species with 2n = 2x = 16, the amount of nuclear DNA ranged within 2C = 1.26–3.4 pg (Benhizia et al., 2021), which was relatively small for plants (Vitte and Bennetzen, 2006; Bennetzen and Wang, 2014; Wang et al., 2021). This fact was quite consistent with the ratio of low-copy and repetitive DNA sequences detected in genomes of the studied species of the sect. Multicaulia.

In different satellite DNA families, a rather high rate of genomic changes was revealed, and satellite DNAs can be either species-specific or common to a certain group of related species (Garrido-Ramos, 2015). Most Fabaceae species are characterized by a large number of various satellite repeats (Neumann et al., 2012; Macas et al., 2015; Ávila Robledillo et al., 2020). Despite the fact that the number of identified tandem DNAs differed in H. grandiflorum, H. zundukii, and H. dahuricum, the main set of common tandem DNA repeats was homologous and their monomer sequences were mostly identical in length.

Thus, the repeatomes of the studied related species of the sect. Multicaulia have more common satellite repeats than species-specific ones, which could be related to their common origin. The revealed close relationship between genomes of the species from two subsections Multicaulia (H. dahuricum, H. razoumovianum, H. setigerum, and H. gmelinii) and Subacaulia (H. grandiflorum and H. zundukii) confirms the results of phylogenetic studies reported earlier (Choi and Ohashi, 2003; Duan et al., 2015; Liu Y. et al., 2019).

SatDNA is often associated with heterochromatin and localized in the certain chromosome regions, which allows it to be studied using various cytogenetic techniques, such as FISH (Mehrotra and Goyal, 2014; Biscotti et al., 2015). In some Fabaceae species, several (2–12) satellite DNA repeats were previously localized in the pericentromeric regions of chromosomes, and most of these repeats were species-specific (Ávila Robledillo et al., 2020). According to our FISH results, four common tandem DNA repeats (Hz 9, Hz 2, Hz 44, and Hz 96) presented similar pericentromeric co-localization on the chromosomes of the studied species, which allowed us to clarify the chromosome morphology and also confirm a close genomic relationship between these species.

Hz 75, Hz 59, and Hz 6 repeats were FISH mapped predominantly in the terminal regions of chromosomes in the studied species except H. grandiflorum in the genome of which Hz 59 was not detected by TAREN. At the same time, Hz 75 and Hz 6, which were FISH mapped on chromosomes of H. dahuricum and H. grandiflorum, were also not detected by TAREN in their repeatomes. This might be due to some features of the used sequencing method, subsequent bioinformatic processing, as well as to the prevalence of satDNA in the genomes of these species. These results show that cytogenetic studies are important for the investigation of plant genomes, since they refine the information obtained after bioinformatics analysis.

SatDNA is considered to be involved in the main processes of formation of the most important chromosomal structures, e.g., DNA packaging and chromatin condensation, and it was reported to represent recombination “hotspots” of genome reorganization (Plohl et al., 2012; Biscotti et al., 2015). The content of satDNA can vary in plant genomes even between generations, which results in high polymorphism in the length of satellite arrays (Plohl et al., 2012; Macas et al., 2015; Ávila Robledillo et al., 2020). Currently, the patterns of distribution of satDNAs as chromosome markers are widely used to detect rearrangements, to identify chromosomes and subgenomes in karyotypes of diploid and polyploid plants, as well as to study the paths of chromosomal evolution of related taxa (Samoluk et al., 2017; Belyayev et al., 2019; Liu Q. et al., 2019; Mata-Sucre et al., 2020). In the studied species from the section Multicaulia of the genus Hedysarum, FISH patterns of chromosomal distribution of Hz 6, both separately or together with any of the pericentromeric probes (Hz 9, Hz 2, Hz 96, and Hz 44), were chromosome-specific, which allowed us to identify all chromosome pairs in karyotypes. At the same time, these molecular chromosomal markers did not reveal chromosomal rearrangements in the karyotypes of the studied species, as it was earlier found in other species (Belyayev et al., 2019; Amosova et al., 2021; Waminal et al., 2021). On chromosomes of the studied species, Hz 6 presented a specific distribution pattern with permanent sites, which were localized in the subtelomeric regions of three pairs of chromosomes, and also, several polymorphic sites were detected on the remaining chromosome pairs. Thus, the permanent sites of Hz 6 in combination with any of the pericentromeric sites of Hz 9, Hz 2, Hz 96, or Hz 44 were the effective molecular markers for chromosome identification. Polymorphic sites of Hz 6, Hz 75, and Hz 59, detected in the terminal regions of several chromosomes, were valuable molecular chromosome markers to analyze intra- and interspecies chromosomal variabilities in the studied species. Moreover, the tandem DNAs Hz 9, Hz 2, Hz 96, Hz 44, and Hz 6 could be useful for comparative cytogenetic studies to clarify the evolutionary relationships within Hedysarum.

The species H. setigerum and H. gmelinii are rather similar in morphological characters, and taxonomists identify H. setigerum either as a separate species (Fedtschenko, 1948) or as a subspecies of H. gmelinii (Kurbatsky, 1994; Sa et al., 2010). In Southern Siberia, both species have overlapping ranges and occupy similar habitats (Fedtschenko, 1948; Malyshev et al., 2012). In both H. gmelinii and H. setigerum, the performed FISH analysis demonstrated pattern similarity in chromosomal distribution of major 35S rDNA and 5S rDNA clusters, minor 5S rDNA sites, and also tandem DNA repeats. Our findings are consistent with the results of the ISSR analysis, which indicated a close relationship between H. setigerum and H. gmelinii (Zvyagina et al., 2016), and also with the reported earlier suggestion that H. setigerum was most likely a subspecies of H. gmelinii (Kurbatsky, 1994; Sa et al., 2010; Zvyagina et al., 2016). H. dahuricum has overlapping areas with H. gmelinii and H. setigerum, and it was included in the Hedysarum gmelinii group (Sa et al., 2010). In our study, FISH-based patterns of chromosomal localization of Hz 6 in H. dahuricum were similar to those revealed in H. setigerum and H. gmelinii, which was consistent with previously reported data on a close relationship between H. dahuricum and the yellow-flowered form of H. gmelinii (Fedtschenko, 1948). Morphologically well-differentiated, H. razoumovianum and H. grandiflorum occupy the overlapped region within Eastern Europe though they are considered to belong either to different sections (Fedtschenko, 1948) or to subsections Multicaulia and Subacaulia (Choi and Ohashi, 2003). Our findings demonstrate that the karyotypes of Eastern European H. razoumovianum are rather similar to karyotypes of South Siberian species belonged to the subsect. Multicaulia. Previously, the morphological similarity between H. gmelinii and H. razoumovianum was described, which was believed to indicate their relationship and the common geographical origin from the Asian ancestor closely related to H. gmelinii (Gorchakovsky, 1969).

It was suggested that morphological similarity of the Eastern European species H. grandiflorum (subsect. Subacaulia) and the South Siberian H. zundukii (subsect. Subacaulia) could be related to their growth in similar conditions on carbonate rocks (Peshkova, 2001). The relict species H. zundukii occupies a very narrow range, and it was reported that its discrete distribution within this range was due to its adaptation to certain environmental conditions (Karnaukhova et al., 2008). Nevertheless, for the natural populations of H. zundukii, some heterogeneities of morphological characters were described (Cherkasova, 2008). On the other hand, H. zundukii exhibits low polymorphism in seed storage proteins, which is typical for species with a small range and small population size (Konichenko and Selytina, 2011).

All South Siberian species of the subsect. Subacaulia, including H. zundukii, are believed to be related to H. gmelinii (Kurbatsky, 1994). Among the studied diploid (2n = 16) species, H. zundukii exhibited the greatest variety of tandem DNA repeats and the most significant intraspecific variability in patterns of chromosomal localization of molecular markers. Moreover, the minor 35S rDNA sites as well as maximum sites of Hz 6 were detected on chromosomes of H. zundukii. At the same time, according to the patterns of chromosomal localization of Hz 6 repeat, H. zundukii (subsect. Subacaulia) was more similar to H. setigerum and H. gmelinii (subsect. Multicaulia) compared to H. grandiflorum (subsect. Subacaulia).

Considering that H. zundukii, H. setigerum, and H. gmelinii occupy the same areas of the western coast of Lake Baikal, and the plants with intermediate (between H. zundukii and H. gmelinii) traits have already been found (Kurbatsky and Malahova, 1992), it is likely that spontaneous interspecific hybridization could increase the genetic intraspecific diversity in the relict species H. zundukii.

Thus, our comprehensive comparative study of genomes of six species of the sect. Multicaulia of the genus Hedysarum detected a close relationship among their genomes (regardless of the region of their growth and the range size), indicating a common origin of these species. At the same time, based on the analysis of intra- and interspecific variabilities in patterns of chromosomal distribution of molecular markers (35S rDNA, 5S rDNA, and tandem DNA repeats), we could subdivide the studied species into four groups: (1) H. zundukii (subsect. Subacaulia), (2) H. setigerum and H. gmelinii (subsect. Multicaulia), (3) H. dahuricum, H. razoumovianum (subsect. Multicaulia), and (4) H. grandiflorum (subsect. Subacaulia). Also, a comparative analysis of genomes of the studied species allowed us to detect tetraploid and hexaploid forms in H. setigerum and H. gmelinii and also confirm the taxonomic status of H. setigerum as a subspecies of H. gmelinii. Our findings indicate the validity of combining the sections Multicaulia and Subacaulia into one section Multicaulia.

The comparison of repeatomes of H. grandiflorum, H. dahuricum, and H. zundukii revealed species-specific differences in genome composition, and also high sequence similarity in the identified satDNAs. FISH mapping of the identified tandem DNA repeats on chromosomes of six Hedysarum species allowed us to assess genome diversity within the section Multicaulia and determine new effective molecular chromosome markers especially important for comparative karyotypic studies. In all studied species, we revealed intra- and interspecific variabilities in patterns of chromosomal distribution of the detected chromosome markers and constructed species karyograms. In H. gmelinii and H. setigerum, similar subgenomes were detected confirming the polyploid status of their genomes. Our findings demonstrated a close relationship among genomes of six studied species indicating their common origin and also confirmed the taxonomic status of H. setigerum as a subspecies of H. gmelinii as well as the validity of combining the sections Multicaulia and Subacaulia into one section Multicaulia.

The datasets presented in this study can be found in online repositories. The name of the repository and accession number can be found below: National Center for Biotechnology Information (NCBI) BioProject database under accession number PRJNA811959 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA811959).

OY and OM contributed to methodology and conceptualization. TS, SZ, IS, SS, and NS contributed to formal analysis. IS and NS contributed to plant materials. OY, TS, SZ, IS, NS, SS, AA, and OM contributed to investigation and contributed to writing—original draft. OM contributed to supervision. OY, TS, SZ, IS, NS, SS, and AA contributed to visualization. OY, AA, and OM contributed to writing, reviewing, and editing the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.