Introduction

The genus Veronica L., commonly referred to as speedwell, bird’s eye, and gypsy weed, is a large genus of Plantaginaceae (Albach et al. 2004a;Salehi et al. 2019). Veronica comprises 11 subgenus species, including Beccabunga, Chamaedrys, Cochlidiosperma, Pellidosperma, Pentasepalae, Pocilla, Pseudolysimachium, Stenocarpon, Synthyris, Triangulicapsula, and Veronica, with 450 species worldwide (Albach and Meudt 2010;Albach et al. 2004b;Garnock-Jones et al. 2007). Among these subgenera, Pseudolysimachium (W.D.J Koch) Buchenau is considered to have the morphologically diverse and systematically complex plants in Veronica (Choi 2016). Valued for their prolonged flowering and need for minimal care, Veronica species have increasing horticultural interest for indoor potting, gardening, and landscaping purposes, and some species have long been recorded as traditional medicinal plants with wound-healing and anti-inflammatory properties (Albach et al. 2008;Hawke 2010;Jeon 2012;Kang et al. 1990;Küpeli et al. 2005). Given these characteristics, the demand for Veronica in flower markets increased eightfold in the average trading volumes and by 45% with respect to unit price (Oh et al. 2019;YFMC 2018). In recent years, certain Veronica taxa native to Korea have been characterized and used as parents for breeding new cultivars and hybrids (KPNI 2020;Song et al. 2020b). In enhancing their economic value, further interest has aroused in studying the ecophysiology, taxonomy, and propagation of Veronica taxa (Jang and Noh 2020;Song et al. 2020a, 2020b). However, the breeding of native Korean taxa has been hampered to a large extent by the limited cytogenetic analyses conducted to date.

The heightened interest in the classification, phylogeny, and breeding of Veronica has been reflected in the increasing number of studies conducted on the morphological, anatomical, and molecular characteristics of species distributed in different regions (Albach and Briggs 2012;Kaplan et al. 2007;Taskova et al. 2004). Such studies are deemed necessary given that differences in chromosomal composition and number make important contributions to our understanding of plant diversification and speciation, and thus ultimately, plant evolution (Lysak et al. 2006;Schurbert 2007;Weiss-Schneeweiss and Schneeweiss 2013). Likewise, studies that yield genome size estimates, along with cytogenetic and ploidy analyses, contribute to providing insights into related taxa in wild plant groups, such as those in the genus Veronica.

Early Veronica chromosomal studies indicated a basic chromosome number of x = 9 (Albach and Chase 2004), although a considerable diversity in chromosome number has been reported in different species (x = 6, 7, 8, 9, 12, 16, 17, 20, and 21) (Agudo et al. 2009;Heitz 1926;Weiss et al. 2002;Yamazaki and Tateoka 1959). Despite these findings, there are unresolved problems regarding the chromosomal characterization of this genus, notably the accuracy of chromosome counts. High-quality chromosome preparations with sufficient spread are required for reliable karyotype analyses, particularly for annual plants, such as Veronica, that are often characterized by aneuploidy/dysploidy and unbalanced chromosome sets (Albach and Chase 2004). The use of both basic and molecular cytogenetic tools can contribute to the resolution of these issues and facilitate further chromosome characterization, such as karyotyping coupled with fluorescence in situ hybridization (FISH) (Lakshmanan et al. 2015).

FISH enables the localization and visualization of a given target DNA sequence, thereby providing valuable insights regarding physical chromosome behavior and evolution, and is typically used in marker-assisted breeding for crop improvement (Devi et al. 2005;Heslop-Harrison and Schwarzacher 1993). The technique uses repetitive sequences labeled with fluorescent dyes that produce signals when hybridized with the respective target DNA sequences (Hizume et al. 2002;Younis et al. 2015). Moreover, physical mapping based on FISH in conjunction with tandem repeats (ribosomal RNA, protein-coding genes, satellites, telomeric DNA, centromeric DNA) has been successfully applied to reveal specific distribution patterns (Biscotti et al. 2015;He et al. 2015;Pereira and Ryan 2019).

The determination of genome size has important applications in a range of biological fields, including ecology, molecular biology, systematics, and evolution (Bennet and Leitch 2005). Genome size is an important biological parameter that clarifies plant evolutionary patterns and adaptive mechanisms (Jordan et al. 2015). Gaining insights into plant genome sizes can enhance our understanding of whole-genome duplication and polyploidy events (Némorin et al. 2013), genome reduction in response to changes in selective pressure (Johnston et al. 2004), the proliferation of non-coding DNA sequences (Gregory 2005), and correlations with cell developmental factors (Gregory 2001). Moreover, Leong-Škorničková et al. (2007) have reported that sufficient differences in genome sizes, along with other evidence, are more apparent and reliable in isolating populations or separating species into discrete groups.

Worldwide, there have been in excess of 400 publications presenting data relating to chromosome numbers in taxa of the tribe Veroniceae, covering some 2,600 populations, including diploid, polyploid, aneuploid, and hybrid species and cultivars (Albach et al. 2008). Nevertheless, despite the considerable effort in determining chromosome numbers in Veronica, the accumulated genetic information is far from comprehensive due to the species' extensive distribution. In Korea, there are more than 11 wild and native taxa that are mainly distributed in remote mountainous and high-elevation regions (Song et al. 2020b), for which we have comparatively little information. Hence, in addressing this deficiency, this study conducted cytogenetic and genome size analyses of the Veronica native taxa to Korea, with the aims of determining patterns of chromosome number and karyotypes, compiling chromosome rDNA maps, and undertaking genome sizes comparisons. Data obtained regarding the cytology and genome sizes of these taxa will make a significant contribution to characterizing fundamental aspects of the phylogeny, taxonomy, and breeding of the Veronica native to Korea.

Materials and Methods

Plant materials and chromosome preparation

Young roots were harvested from the plants of seven Korean native Veronica taxa obtained from the Useful Plant Resources Center of the Korea National Arboretum, Republic of Korea, namely, Veronica kiusiana var. diamantiaca (Nakai) T. Yamaz.; Veronica pusanensis Y. N. Lee; Veronica pyrethrina Nakai; Veronica longifolia L.; Veronica nakaiana Ohwi; Veronica rotunda var. subintegra (Nakai) T. Yamaz.; and Veronica dahurica Steven. Chromosome preparations were obtained from clean harvested roots, which were initially treated with 2 mM 8-hydroxyquinoline for 5 h at 25°C, fixed overnight in Carnoy’s solution (3:1, acetic acid-ethanol, v/v) at room temperature, and preserved in 70% ethanol at 4°C. The fixed root tips were washed with distilled water prior to treating with an enzyme mixture (1% cellulose, cytohelicase, and pectolyase) at 37°C for 90 min. Thereafter, the enzyme-treated roots were transferred to 1.5 mL tubes containing Carnoy’s solution and vortexed for 20 s. The homogenized root meristems were placed on ice for 5 min and then centrifuged at 13,000 rpm. The resulting supernatant was discarded, and the pelleted material was immediately resuspended in an acetic acid–ethanol (9:1) solution. Samples of the suspensions were spread on pre-warmed (80°C) glass slides in a humid chamber and left to air-dry at room temperature.

Fluorescence in situ hybridization

FISH was performed using the procedures described by Lim et al. (2005) with some modifications. Pre-labeled oligoprobes (PLOPs) for 5S rDNA and 45S rDNA sequences were labeled following the procedures described by Waminal et al. (2018). The hybridization solution consisted of 50% formamide, 10% dextran sulfate, 20× saline sodium citrate (SSC) buffer, 50 ng/μL of each PLOP, and nuclease-free water to a total volume of 40 μL. The mixture was pipetted onto prepared chromosome slides and denatured on a slide heater at 80°C for 5 min. These slides were subsequently incubated at room temperature in a humid chamber for 30 min. Following hybridization, the slides were washed successively with 2× SSC at room temperature for 10 min, 0.1× SSC at 42°C for 25 min, and 2× SSC at room temperature for 5 min, and then dehydrated in a series of ethanol concentrations (70%, 90%, and 100%) at room temperature, each for 3 min. Thereafter, the slides were counterstained with Vectashield (H-1000; Vector Laboratories, USA) and 1 μg･mL^-1 4ʹ,6-diamidino-2-phenylindole (Roche, USA), and examined under an Olympus BX53 fluorescence microscope (Olympus, Japan) with a built-in CCD camera (CoolSNAP™ cf; Sony Corp., USA) using an oil lens (× 100 magnification).

Karyotyping

The metaphase chromosomes of the seven assessed Veronica taxa were analyzed using the KaryoType software developed by Altinordu et al. (2016). Homologous chromosomes were paired based on their rDNA signals, chromosomal size, and centromeric position, and arranged in descending order based on chromosomal length (Hwang et al. 2013). Karyotype characterization also included determinations of the centromeric index, arm ratio, and karyotype formula (Levan et al. 1964). Karyograms and idiograms were prepared using Adobe Photoshop CS6 and Adobe Illustrator (Adobe Corp., USA).

Genome size estimation using flow cytometry analysis

As the reference standard for genome size determinations, we used young plants of Veronica officinalis, with an established genome size of 880.20 Mbp (1C) (Leitch et al. 2019). Flow cytometry analysis was performed based on the methods described by Doležel et al. (2007). A healthy young leaf of approximately 20 mg obtained from seedlings was placed into a petri dish along with 700 μL of ice-cold LB01 lysis buffer for the isolation of nuclei. Having chopped the leaves using a sharp razor blade, the homogenate was mixed by repeatedly pipetting the solution up and down, avoiding air bubbles. The resulting mixture was then initially filtered through a 50-mm nylon mesh, after which the filtrate was passed through a 30-mm nylon mesh. The final homogenate sample was stained by adding 50 μL of propidium iodide, along with the simultaneous addition of 2 μL of RNase (Sigma-Aldrich, USA). Samples placed in tubes were subsequently incubated on ice for approximately 5 min prior to analysis. Tubes were inserted into a CytoFLEX flow cytometer (Beckman Coulter Inc., USA) equipped with a 50-mW 488-nm solid-state diode laser, and the peaks of standard and sample preparations were compared using CytExpert v2.3 software (Beckman Coulter Inc., USA). Calculations were performed based on external standardization, involving the consecutive isolation, staining, and analysis of nuclei from the sample and standard preparations.

Results and Discussion

Chromosome FISH karyotype

Chromosome length, karyotype formula, and FISH signal data obtained from karyotypic analyses of the seven Korean native wild Veronica taxa are shown in Table 1. All examined taxa within Veronica subg. Pseudolysimachium have diploid chromosome compositions of 2n = 2x = 34 with a basic chromosome number of x = 17 and have conspicuously small chromosomes (Fig. 1). These chromosome numbers are consistent with those of V. dahurica, V. nakaiana, V. rotunda, and V. longifolia reported by Albach et al. (2008). The ranges of the total lengths of mitotic metaphase chromosomes of the seven taxa were as follows: V. kiusiana var. diamantiaca, 2.43 to 5.09 μm; V. pusanensis, 1.88 to 3.79 μm; V. pyrethrina, 2.17 to 4.39 μm; V. longifolia, 2.34 to 4.55 μm; V. nakaiana, 1.83 to 2.99 μm; V. rotunda var. subintegra, 2.48 to 4.85 μm; and V. dahurica, 2.02 to 4.56 μm (Table 1). With respect to the haploid total length of chromosomes, those of V. kiusiana var. diamantiaca, V. rotunda var. subintegra, and V. longifolia were the longest with lengths of 54.71, 54.49, and 53.86 μm, respectively, whereas that of V. nakaiana was shortest at 39.84 μm (Table 1). In terms of chromosomal type, four of the seven taxa are characterized by 17 metacentric chromosomes, whereas among the remaining three taxa, V. nakaiana has 11 metacentric and 6 submetacentric, V. rotunda var. subintegra. Has 16 metacentric and 1 submetacentric, and V. dahurica has 14 metacentric and 2 submetacentric chromosomes (Table 1). Total chromosome length is often used in comparisons of very similar taxa (Lopez et al. 2009;Mehes-Smith et al. 2011;Mekki et al. 2007), and thus considering the variation detected in the present study, could serve as a useful character for the identification of Veronica species and cultivars.

However, we did observe slight differences in the chromosome lengths in metaphase spreads on the same slide, which we suspect could be attributable to differences in the condensation stages of the examined chromosomes. Although the use of well-condensed chromosome is considered crucial for the exact measurement of chromosome length and karyotypic formulae (Hwang et al. 2017), as Bennett (1970) has noted, different aspects of chromosome preparation, including the type of pre-treatment agents, pre-treatment duration, and time of sample collection, can give rise to considerable differences in total chromosome size.

Providing a karyotype, the phenotypic appearance of a somatic chromosome complement is particularly useful in evaluations of genetic relationships and clarification of species origin and divergence (Bhat and Wani 2017). However, for plants that are characterized by small-sized chromosomes, such as Veronica, and only taking into consideration their chromosome size and morphology, accurate karyotype analysis poses several challenges (Rho et al. 2012). With the advancement and development of the FISH technique using repetitive DNA sequences as probes, this has not only contributed to meeting these challenges, but also to clarifying genomic structures at the chromosomal level (Heslop-Harrison 2000). Among the FISH cytogenetic markers that are routinely used in plant analyses, 5S and 45S (18S-5.8S-25S) rDNA genes are utilized as they are highly conserved and have high copy numbers (Hasterok et al. 2001;Kato et al. 2005). These genes comprise tandemly repeated sequences and are typically localized at specific loci in the chromosomes of higher eukaryotes (Kubis et al. 1998). Our FISH karyotype analysis in the present study revealed interspecific differences, not only in terms of karyomorphology, but also in rDNA signal distribution in the chromosomes of the seven Veronica taxa, which accordingly enabled efficient identification of the chromosomes of each taxon.

rDNA distribution patterns

To determine patterns of rDNA distribution in the seven assessed Veronica taxa, we performed FISH analysis using pre-labeled 5S rDNA and 45S rDNA oligoprobes, which accordingly revealed distinct distribution patterns in each taxon, thereby validating the taxa specificity and efficacy of the repeat probes in taxon differentiation (Figs. 2 and 3). In all seven taxa, we detected only a single pair of signals for 5S rDNA in the metaphase chromosomes, which are in the majority of the taxa. They were observed to be localized in intercalary regions of the long arms of chromosome 3 (Fig. 3A-C, E, and F). The exceptions in this regard were V. longifolia and V. dahurica, for which the 5S rDNA signals were detected in the intercalary region of the long arm of chromosome 5 (Fig. 3D) and chromosome 4 (Fig. 3G), respectively. This implies that the number of 5S rDNA loci is conserved among the seven Veronica taxa. In contrast, we detected a notable variation among taxa with respect to both the number and chromosomal locations of 45S rDNA signals, although these were localized at terminal sites in the short arms of the chromosomes. Specifically, we observed four pairs of 45S rDNA signals in V. pusanensis and V. nakaiana, whereas six pairs were detected in V. kiusiana var. diamantiaca, V. pyrethrina, and V. longifolia, and five and eight pairs were found in V. dahurica and V. rotunda var. subintegra, respectively. Furthermore, in V. pusanensis, V. pyrethrina, V. longifolia, and V. dahurica, 5S and 45S rDNAs were localized in the same chromosomes, although in different chromosomal regions.

The plant rDNA database contains FISH rDNA data from more than 2,000 plant species (Vitales et al. 2017). Among the available accessions, Garcia et al. (2009) analyzed 2,949 karyotypes to assess relationships between rDNA locus number and distribution. They accordingly found that approximately 48% of the assessed species have a larger number of 45S rDNA than 5S rDNA loci, whereas only 19% are characterized by a larger number of 5S rDNA loci, and the remaining 33% have co-localized rDNAs or have the same number of loci. In common with a large proportion of the plants examined to date, we detected a significantly greater number of 45S rDNA loci than 5S rDNA loci in each of the Veronica taxa examined in the present study. Additionally, similar to most angiosperms (Roa and Guerra 2012), the 45S rDNA loci in these seven taxa were found to be distributed in the terminal regions of chromosomal short arms.

Given the often high variability among closely related species or genera, the numbers and positions of rDNA loci could serve as defining characteristics of a specific species or genus (Garcia et al. 2009;Singh et al. 2009). Three mechanisms could conceivably account for the observed interspecific variation in the number and distribution of rDNA loci: (i) unequal crossing over during recombination and/or transposition events, (ii) rapid chromosome structural rearrangements affecting the rDNA regions, or (iii) polyploidization events followed by the loss of DNA sequences (Badaeva et al. 2007;Mondin and Aguiar-Perecin 2011). In the present study, we found that whereas the number of 45S rDNA sites varied from four to eight pairs in the taxa examined, the number of 5S rDNA sites was comparatively stable. We speculate that this 45S rDNA locus variability could be associated with the fragility of the chromosomal regions in which these loci are located, which are prone to breakage and the emergence of gaps, thereby possibly leading to a wider variation of loci number in plants (Bustamante et al. 2014;Mancia et al. 2017).

Establishing rDNA distribution patterns could also make a valuable contribution to determining species history and elucidating phylogenetic relationships (Clarkson et al. 2005). For example, V. pusanensis is notably well differentiated from the related taxa V. dahurica and V. pyrethrina with respect to its seashore habitats and plant characteristics (Jang and Noh 2020), and correspondingly, we found that V. pusanensis differed from these latter two species in terms of the numbers and positions of rDNA loci. Furthermore, our cytogenetic data imply that V. pusanensis is more closely related to V. nakaiana, whereas V. kiusiana var. diamantiaca, V. pyrethrina, and V. longifolia L. have a closer inter-relationship compared with other taxa. Nevertheless, despite these interesting findings, they should be treated with a degree of caution, given that the data derived from cytotaxonomic studies may not invariably indicate phylogenetic relationships and should ideally be supported by credible and extensive molecular data (Guerra 2008).

Flow cytometry analysis

There exists a considerable diversity in the size of angiosperm genomes, ranging by approximately 2000-fold (Pellicer et al. 2010), and the genus Veronica is among those that are often characterized by variations in genome size, depending on geographical distribution and/or perennial or annual life forms (Albach and Greilhuber 2004). In this context, flow cytometry is currently the method of choice used to determine the nuclear DNA content of plants and is also often used for hybrid identification and ploidy level determination (Dolezel et al. 2007). In the present study, flow cytometry was used to obtain estimates of the DNA content and genome sizes of the seven Korean native Veronica taxa, with V. officinalis being used as an external standard. A comparison of peak data revealed that the 1C DNA contents of these plants ranged from 0.53 to 0.88 pg and that the haploid genome size ranged from 517.1 to 862.0 Mbp (Table 2). Among the taxa, V. nakaiana was found to have the smallest genome size, whereas the largest genomes were detected in V. longifolia and V. rotunda var. subintegra, followed by V. kiusiana var. diamantiaca, all three of which have respective genome contents and sizes greater than 0.80 pg and 800 Mbp.

All the Veronica taxa examined in the present study are Korean native perennials that are widely distributed from mountainous to coastal regions, such as V. nakaiana on Ulleung Island, V. pusanensis in Gijang, Busan, and V. kiusiana var. diamantiaca on Mts. Seorak and Geumgang (Chung et al. 2017). Furthermore, these three Korean endemics are also included in the red list of threatened species as critically endangered or endangered (Korea National Arboretum 2021). These native Veronica taxa have been reported to show distinguishable leaf formation, plant height, and other phenotypic characteristics, including leaf color, stem growth, presence of trichomes, and flowering characteristics (Song et al. 2020b). To a certain extent, variation in phenotypic characteristics is associated with the amount of genetic material, given that genome size influences the rate of cell division and ecological behavior in plants (Bennett 1987;Bennett and Smith 1991). We can accordingly speculate that the variable genome sizes observed in these Veronica taxa could be associated with their habitat diversity, life forms, and breeding systems (Albach and Greilhuber 2004;Kellogg and Bennetzen 2004).

Genome size has also been proposed to be associated with cell cycle duration and is generally shown to be positively correlated with chromosome size (Badr et al. 1987;El-Shazly et al. 2002;Hamoud et al. 1994;Narayan 1998). Large chromosomes in plants tend to reflect a proportional relationship with nuclear DNA content (Jones and Rees 1968), and in this regard, positive correlations between genome size and total chromosome length have been observed in Liliaceae (Peruzzi et al. 2009), Lathyrus sativus (Ghasem et al. 2011), and Cicer species (Hejazi 2011). The findings of previous studies have also indicated that variations in chromosome size have led to differences in genome content (Animasaun et al. 2019;López et al. 2009;Mekki et al. 2007;Mehes-Smith et al. 2011). Consistently, we found that among the taxa examined in the present study, V. nakaiana is characterized by the lowest DNA content and genome size of 0.53 pg and 517.1 Mbp, respectively, and also has the shortest total chromosome length at 39.84 μm. Similarly, among the seven Veronica taxa, V. longifolia, V. rotunda var. subintegra, and V. kiusiana var. diamantiaca were found to have the highest DNA contents and genome sizes, and also the longest total chromosome lengths of 53.86, 54.49, and 54.71 μm, respectively (Tables 1 and 2). In this regard, Mártonfiová (2013) mentioned that whereas DNA amounts remain constant, average total chromosome lengths can vary to differing extents depending on the type of pre-treatment agent and different pre-treatment times, as well as between simultaneously treated siblings. Moreover, different degrees of chromosome contraction should be taken into consideration when interpreting the relationships between total chromosome length and DNA content.

Conclusion

This is the first report on the chromosome number and karyomorphology of the seven Veronica taxa native to Korea, as well as the use of rDNAs as markers for karyotyping and identifying individual chromosomes. The results herein provide essential information for plant breeding and ongoing genomic sequencing research. In further studies, we anticipate the identification of highly abundant repeats based on sequence analysis, which may contribute to developing either genus- or species-specific cytogenetic markers for more refined karyotyping, and in turn, could advance our current understanding of the genetic constitution and genomic evolution of Veronica taxa.