Genome size determination and chromosome characterization of Limosella aquatica L. (Scrophulariaceae) in Japan: Insights into Japanese population

Abstract

The genome size and karyomorphology of six accessions of Limosella aquatica L. in Japan were investigated. All six accessions had a chromosome number of 2n=40. The genome size of L. aquatica was estimated using flow cytometry (FCM) and 2C of the tetraploid genome size ranged from 1.38 pg in accession Shiranuka to 1.42 pg in accession Taiki (1,178–1,216 Mbp), with 1Cx estimated to be ca. 0.35 pg (300 Mbp). The GC contents of all accessions were less than 40%. The centromeres of all chromosomes were located at median positions, while secondary constrictions were found at the distal regions of two chromosomes, resulting in small chromatin segments as satellites. The chromosome complement showed a monomodal karyotype with a gradual decrease in the chromosome length, ranging from 1.3 to 0.6 µm. The average chromosome lengths of accessions in Hokkaido and Honshu were 0.9 and 1.0 µm, respectively. The accessions in Kumamoto were exceptions, showing chromosome lengths ranging from 1.6 to 0.8 µm. The DNA base-specific banding revealed that the pericentromeric regions of all chromosomes and the satellites were chromomycin A₃ positive and 4′,6-diamidino-2-phenylindole negative (CMA⁺DAPI⁻). The 18S rDNA fluorescence in situ hybridization (FISH) result also showed that the two signals were located at exactly the same positions as the CMA⁺DAPI⁻ satellites.

Limosella L. (Scrophulariaceae) is a small aquatic genus that contains 12 species (Ito et al. 2017). Among them, Limosella aquatica L. is a cosmopolitan species distributed in cool regions of the Northern Hemisphere, and is the only species known from Japan (Yamazaki 1993). L. aquatica is generally regarded as an annual herb with prolific seed production (Salisbury 1967), while several new ramets arise asexually at the apices of stolons every year (Yamazaki 1993, Sawa et al. 2022). This species is a typical member of the flora found in mud habitats, such as lakes, ponds, shallow waters, and river margins (Matevski and Kostadinovski 2009, Son et al. 2011), which undergoes extreme fluctuation in population numbers annually (Curtis et al. 1985).

L. aquatica is widespread in Japan, ranging from Hokkaido to Kyushu, i.e., from the northernmost to the southernmost tip of the four main islands (Yamazaki 1993). However, its distribution is highly localized, and it is classified as Vulnerable (VU) in the latest Japanese Red List (Ministry of the Environment; http://www.env.go.jp/press/107905.html). The species’ first discovery site in Kitami (the district of northeast Hokkaido), from which the Japanese name ‘Kitami-sō’ is derived, appears to have unfortunately lost its habitat (Sawa et al. 2022). Due to dramatic changes in the number of individuals in the habitats, the entire localities of the species in Japan have not been precisely determined. However, local populations are recognized in several Prefectures: Hokkaido, Tochigi, Gunma, Ibaraki, Saitama, Nara, and Kumamoto (Sawa et al. 2022). Nevertheless, in Nara Prefecture, live L. aquatica specimens have not been found in the previously recognized habitat for the past seven years (Sawa et al. 2022). Despite the recognition of domestic habitats, there have been few papers dedicated to the cytogenetic study of the Limosella genus, except for chromosome numbers (Blackburn 1939, Siljak-Yakovlev et al. 2020), let alone cytogeographic research of Japanese populations. We sampled L. aquatica from most of the Japanese populations with the aim of delineating the extent of chromosome differentiation before their original habitats are lost or significantly altered by climate change and other factors.

In this study, we investigated karyomorphological differences among six populations spanning a 1,200-km latitudinal transect, covering the entire distribution range in Japan. To achieve this, we employed cytogenetic techniques, including orcein staining, DNA base-specific banding, and fluorescence in situ hybridization (FISH). Additionally, we determined the genome size and GC content of each accession using flow cytometry (FCM). Finally, we discuss chromosome evolution in L. aquatica and its related species, along with the previous data in the genus Limosella.

Materials and methods

Plant material

The plant accessions of L. aquatica are listed in Table 1. The seedlings were obtained from aseptically germinated seeds on half-strength Murashige-Skoog’s basal medium (Murashige and Skoog 1962), supplemented with 0.6% sucrose (pH 5.7 before autoclaving). They were then subcultured in the same medium. These accessions were maintained either in vitro or ex vitro at the International Research Center for Agricultural and Environmental Biology, Kumamoto University.

Table 1. The source of the investigated materials of L. aquatica collected in Japan.

Population (Accession name)	Locality
Shiranuka	Syoro river, Shiranuka, Hokkaido Prefecture (Hokkaido Island)
Toyokoro	Yudonuma, Toyokoro, Hokkaido Prefecture (Hokkaido Island)
Taiki	Oikamanaito, Taiki, Hokkaido Prefecture (Hokkaido Island)
Gyoda	Hoshi river, Gyoda, Saitama Prefecture (Honshu Island)
Tsukubamirai	Kokai river, Tsukubamirai, Ibaraki Prefecture (Honshu Island)
Kumamoto	Lake Ezu, Kumamoto, Kumamoto Prefecture (Kyushu Island)

Chromosome preparation

For DNA base-specific fluorescence staining and orcein staining, the air-dry technique of chromosome preparation was employed following the protocol of Hoshi et al. (2021). In vitro root tips were collected and pretreated with 0.05% colchicine for 2 h at 18°C before fixing in 45% acetic acid for 30 min on ice. They were hydrolyzed in a mixture of 1 M HCl and 45% acetic acid (2 : 1) at 60°C for 7 s. The hydrolyzed root-tip meristems were isolated on glass slides, and well-spread meristem cells in 45% acetic acid were squashed under coverslips. After removing the frozen coverslips, the chromosomes were air-dried at room temperature (RT).

For FISH chromosome preparation, we employed an enzymatic digestion technique in accordance with the protocol of Hoshi et al. (2019). The fixed root tips were macerated in an enzymatic mixture consisting of 0.4% Cellulase Onozuka R-10 (Yakult Pharmaceutical Industry Co., Ltd., Tokyo, Japan) and 0.2% Pectolyase Y-23 (Seishin Pharmaceutical Co., Tokyo, Japan) for 1.5 h at 37°C. After washing the root tips with distilled water, they were placed onto glass slides and squashed in 45% acetic acid. The coverslip was then removed, and the preparations were air-dried for 12 h at RT. Some slides were also used for DNA base-specific fluorescence staining and orcein staining.

DNA base specific staining and orcein staining

Two counter fluorochromes, GC-specific chromomycin A₃ (CMA) and AT-specific 4′,6-diamidino-2-phenylindole (DAPI), were used for DNA-based specific fluorescent staining. The method employed for fluorescent staining involved the sequential application of CMA and DAPI, as described by Hoshi et al. (2011) with slight modifications. The air-dried slides were mounted using a 30% glycerol solution containing 30 µg mL⁻¹ CMA in McIlvaine’s buffer (pH 7.0) supplemented with 5 mM MgSO₄, and incubated for 30 min at RT. After observation of the CMA banding pattern, the preparations were destained in 45% acetic acid for 30 min, lightly rinsed with distilled water, and then remounted using a 30% glycerol solution containing 3 µg mL⁻¹ DAPI in McIlvaine’s buffer for 10 min. Additionally, a simplified method of double staining with CMA and DAPI, based on Hoshi et al. (2019) was employed. The air-dried slides were mounted directly with a 30% glycerol mixture containing 30 µg mL⁻¹ CMA and 3 µg mL⁻¹ DAPI in McIlvaine’s buffer, supplemented with 5 mM MgSO₄. Following observation of fluorescent banding, the preparations were destained in 45% acetic acid for 30 min. For orcein staining, the air-dried slides and destained preparations were used and stained with 2% acetic orcein.

FISH procedure

Total genomic DNA was extracted from germinated seedlings using the following procedure of Hoshi et al. (2019). To get the 45S rDNA probe, we performed polymerase chain reaction (PCR) based on the method described by Hoshi et al. (2019). In this PCR, a specific region of 45S rDNA, the 18S rDNA sequences, was amplified using a primer set of 18S F (5′-AACCTGGTTGATCCTGCCAGT-3′) and 18S R (5′-TGATCCTTCTGCAGGTTCACCTAC-3′). The thermal cycle profile was an initial denaturation at 94°C for 4 min. This was followed by 35 cycles 94°C for 30 s, 53°C for 30 s, and 72°C for 60 s. The final extension step was performed at 72°C for 5 min. The PCR fragment was labeled with biotin-16-dUTP (Sigma-Aldrich Co. LLC.) using a random primed DNA labeling technique. One microgram template DNA was added to a final volume of 16 µL water, denatured by heating in a boiling water bath for 10 min, and immediately chilled in ice water. Then, the denatured DNA was incubated at 37°C for 10 h, after adding 4 µL Biotin-High Prime mixture containing 1 U µL⁻¹ Klenow polymerase, 1 mM dATP, 1 mM dCTP, 1 mM dGTP, 0.65 mM dTTP, and 0.35 mM biotin-16-dUTP. Subsequently, the DNA with the mixture was heated at 96°C for 10 min and immediately chilled in ice water. Finally, the labeled DNA was purified and collected by ethanol precipitation.

The air-dried slides for FISH were pretreated with 100 µg mL⁻¹ RNase in 2×SSC at 37°C for 1 to 3 h, briefly washed, and drained of water immediately. The slides were fixed in 4% paraformaldehyde on ice for 15 min and treated with an ethanol series (70%, 99.5%) for 5 min each at RT. They were then air-dried for 3 to 5 min. The hybridization mixtures contained 50% formamide, 10% dextran sulfate, and 4 ng µL⁻¹ labeled DNA in 2×SSC. Hybridization mixture (25 µL) was applied to a chromosome preparation, covered with a coverslip, and sealed with rubber gum. The slides were denatured at 73°C for 3 min on a hot plate and then incubated overnight at 37°C in a humid chamber. After the incubation, the coverslips were floated off, and the slides were washed in 2×SSC at 37°C for 10 min, 0.2×SSC at 37°C for 10 min, and 2×SSC with 0.2% Tween 20 at RT for 10 min, twice.

The slides were treated with a 5% (w/v) bovine serum albumin in 2×SSC at 37°C for 30 min to prevent nonspecific FISH signals. Signals were detected using 100 µL of a 2 µg mL⁻¹ avidin-Alexafluor 488 (Thermo Fisher Scientific Inc.) in 2×SSC with 0.2% Tween 20 on each slide for 3 h at 37°C in a humid chamber. Subsequently, the slides were washed twice in 2×SSC with 0.2% Tween 20 at RT for 10 min. They were also washed twice in 2×SSC at RT for 10 min. Finally, the slides were mounted using a Vectashield mounting medium (Vector Lab.) containing 10 µg mL⁻¹ DAPI.

Chromosome observation

The chromosomes stained with fluorochromes and acetic orcein were observed by an epifluorescence and optical microscope (BX51, Olympus), respectively. In the fluorescent staining, the chromosomes stained with CMA and DAPI were observed under the epifluorescence microscope with a U-MWBV filter and a U-MWU2 filter, respectively. In FISH, the signals were observed under the epifluorescence microscope with a U-MNIBA3. Digital images were taken with a DP73 digital camera (Olympus) on the microscope.

Chromosome analysis

To determine the chromosome numbers, we observed more than 30 prometaphase or metaphase cells from at least five individuals in each accession. For karyomorphological analysis, we measured the lengths of 10 chromatids from five chromosome spreads in each stage using ImageJ version 1.45s (National Institute of Health, Bethesda, MD, USA). The classification of mitotic metaphase chromosomes in somatic cells followed the method described by Levan et al. (1964).

Estimations of genome size and GC content by FCM

The nuclear DNA contents and GC contents were determined by the methods modified by Katogi and Hoshi (2022) and Šmarda et al. (2008, 2014), respectively. To determine the nuclear DNA contents, ten fully expanded leaves cultured in vitro were collected from the plants and then chopped in 1.0 mL of a nuclei extraction buffer (NE buffer) containing 50 mM Tris–HCl (pH 7.5), 50 mM Na₂SO₃, 140 mM 2-mercaptoethanol, 2 mM MgCl₂, 2% (w/v) PVP K-30, 1% (v/v) Triton X-100, and 25 µg mL⁻¹ propidium iodide (PI).

For GC content estimation, NE buffer containing 2 µg mL⁻¹ DAPI instead of PI was used. The samples were then filtered through a 48-µm nylon mesh and centrifuged at 12,000 rpm for 2 min at RT. The pellets including isolated nuclei after centrifugation were dissolved in 0.2 mL NE buffer containing 25 µg mL⁻¹ RNase. This was followed by 30 min incubation at 37°C. The DNA contents of nuclei were measured using a Guava EasyCyte 12HT microcapillary flow cytometer (Millipore). Five thousand nuclei were acquired at a flow rate of 0.12 µL s⁻¹ for each FCM measurement. At least three replicates were measured for each accession. Young leaves of Oryza sativa L. ‘Nipponbare’ (2C=0.91 pg or 777.64 Mbp, Uozu et al. 1997, International Rice Genome Sequencing Project and Sasaki 2005) were used as a reference standard to estimate genome size in absolute units. When compared with the reference standard O. sativa ‘Nipponbare’ (43.6%, International Rice Genome Sequencing Project and Sasaki 2005), the GC contents were calculated. The result of measurement of the base-unspecific PI, which was the dye used to determine absolute genome size, was used together with that of the AT-selective DAPI. Monoploid genome size (Cx-value) was calculated according to Greilhuber et al. (2005) as the absolute 2C DNA content (2C-value) of the sample divided by the ploidy level.

Data comparison and taxonomic validation

Information on the chromosome numbers and C-values was obtained from the Index to Plant Chromosome Number database (http://www.tropicos.org/Project/IPCN) and the Plant DNA C-values Database (https://cvalues.science.kew.org), respectively. Species name validation and synonymy were performed using the Plant List (http://www.theplantlist.org).

Results and discussion

Table 2 presents the cytogenetic results of the six L. aquatica accessions examined in this study. The species exhibited a chromosome number of 2n=40 (Figs. 1, S1), with no variation among the accessions. This chromosome number of 2n=40 is consistent with previous results (Blackburn 1939, Probatova and Sokolovskaya 1981, Löve and Löve 1982, Kiehn et al. 1991, Měsíček and Javůrková-Jarolímová 1992, Chepinoga et al. 2008, Delgado et al. 2015). Additionally, Blackburn (1939) reported a chromosome number of 2n=30 in a natural hybrid between L. aquatica and L. subulata Ives (2n=20) found on the shores of Kenfig Pool, South Wales, United Kingdom. It is noteworthy that the latter species falls within the same clade as L. australis R. Br. based on molecular phylogenetic analysis (Ito et al. 2017). In contrast to L. aquatica, L. australis exhibits a meiotic chromosome number of n=30 (Hair and Beuzenberg 1960, Dawson 2000), and three different mitotic chromosome numbers of 2n=20, 48, and 60 (Löve and Löve 1958, Moore 1981, Gervais and Cayouette 1985, Siljak-Yakovlev et al. 2020). Except for one report (Moore 1981), all recorded chromosome numbers for L. aquatica and its related species including the hybrid, are multiples of ten. This suggests that polyploidy may exist with a basic chromosome number of x=10, although only a few reports have mentioned polyploidy in this genus. As Salisbury (1967) indicated, L. subulata and L. aquatica were diploids and tetraploids, respectively, studied cytogenetically by Blackburn (1939), although she made no mention of ploidy in her study of chromosomes. Based on these past chromosome studies, we agree with Salisbury’s (1967) conclusion that L. aquatica is a tetraploid. Recently, Siljak-Yakovlev et al. (2020) reported that L. australis is hexaploid (2n=6x=60) with a value of 2C=1.91 pg. Our FCM results (Table 2) showed that the 2C genome sizes of L. aquatica, a tetraploid species, ranged from 1.38 pg in Shiranuka to 1.42 pg in Taiki (1,178–1,216 Mbp). This estimation corresponds to a Cx value of ca. 0.35 pg (300 Mbp). The Cx-value in L. aquatica is thus almost the same as that of the hexaploid L. australis (Cx=0.32 pg, Siljak-Yakovlev et al. 2020). In light of research evidence regarding chromosome numbers and genome sizes in the two species, it can be argued that speciation between them may have involved polyploidization without significant changes in the genome sizes of the basic chromosome complement.

Table 2. Chromosome numbers, 2C DNA and GC contents, and genome sizes of six populations of L. aquatica.

Population (Accession name)	Somatic chromosome number (2n)	2C DNA content (pg)	GC content (%)†	Genome size (Mbp)		DNA content/Chromosome (Mbp)
				2C‡	1C
Shiranuka	40	1.38	38.6±1.96	1,178a	589	29.4
Toyokoro	40	1.42	38.6±2.05	1,213cd	607	30.3
Taiki	40	1.42	38.1±1.90	1,216d	608	30.0
Gyoda	40	1.40	39.9±0.33	1,196b	598	29.9
Tsukubamirai	40	1.38	39.6±0.24	1,180a	590	29.5
Kumamoto	40	1.40	39.2±0.21	1,199bc	600	30.0

^†Values are means±SD. ^‡1 pg=854.5 Mbp (Uozu et al. 1997). Means within the same column with different letters are significantly different by the Tukey’s test (p＜0.05).

Fig. 1. Mitotic-metaphase chromosomes in the somatic cell of L. aquatica (accession Kumamoto).

The chromosomes in the same cell were stained with orcein (A), DAPI (B), and CMA (C). GC-rich sites (yellow colored) were enhanced by superimposing image (D) of CMA and DAPI. Arrows indicate terminal CMA bands, as a landmark to identify the sat-chromosome pair. Scale bar=5 µm.

The results of karyological analyses of the mitotic chromosomes are shown in Table 3. The common characteristics of the six accessions were as follows: centromeres were located at the median position of all metaphase chromosomes (Figs. 1A, S2A, Table 3), and secondary constrictions were observed at the distal regions of two chromosomes, resulting in small chromatin segments known as satellites at the ends of constricted regions (Fig. 1). When arranging each metaphase chromosome of the mitotic complement from the largest to the smallest, the third and fourth largest chromosomes had satellites (sat-chromosomes) (Fig. S2A). The metaphase complement showed a monomodal karyotype with a gradual decrease in chromosome length, ranging from 1.3 to 0.6 µm. The average chromosome lengths of the Hokkaido and Honshu accessions were 0.9 and 1.0 µm, respectively. However, the Kyushu (Kumamoto) accession exhibited longer chromosome lengths ranging from 1.6 to 0.8 µm (Table 3). It was observed that the chromosome lengths of the Kumamoto accession were significantly longer than those of the other accessions, especially for the longest chromosome length (p＜0.05) (Table 3). Nevertheless, these differences may be attributed to variations in chromatin condensation rather than DNA content. This is supported by the 2C values of all accessions, which showed no significant difference between Kumamoto (Kyushu Island, refer to Table 1) and the other sites (Table 2). Therefore, the present results of chromosome lengths and genome sizes suggest that the Kumamoto accession has a lower degree of condensation compared to the other accessions. This difference in condensation degree is presumably because of physiological responses during pretreatment of chromosome preparation. Further genome sequencing and gene analysis are expected to reveal unique genetic characteristics of the Kumamoto accession.

Table 3. Karyomorphological characteristics of six populations of L. aquatica.

Population (Accession name)	Total chromosome length (µm)†	Average length of chromosome (µm)	Chromosome length from the longest to the shortest (µm)	Longest chromosome/shortest chromosome	Karyotype formula	Number of CMA+ or DAPI− sat-chromosome in mitotic complements
Shiranuka	37.8a±1.9	0.9a	1.3a–0.7ab	1.9	40m	2
Toyokoro	36.6a±1.5	0.9a	1.3a–0.6a	1.9	40m	2
Taiki	35.3a±2.1	0.9a	1.2a–0.6a	1.9	40m	2
Gyoda	38.8ab±2.7	1.0ab	1.3a–0.7ab	1.9	40m	2
Tsukubamirai	38.8ab±1.0	1.0ab	1.3a–0.7ab	1.9	40m	2
Kumamoto	44.5b±4.9	1.1b	1.6b–0.8b	2.0	40m	2

^†Values are means±SD. Means within the same column with different letters are significantly different by the Tukey’s test (p＜0.05).

In contrast, prometaphase revealed a gradual decrease in chromosome length, except for the two shortest chromosomes (Figs. 2, S2B). According to Blackburn’s (1939) drawing of the mitotic chromosome complement, L. aquatica possessed two chromosomes that were clearly shorter than the others. Most of the illustrated chromosomes were approximately 2 µm in length, longer than the mid-metaphase and shorter than the prometaphase in our results. Thus, the early stages with fewer condensed chromosomes may allow the two small chromosomes to be distinguished from the other chromosomes. The chromosomes in the prometaphase complement ranged from 5.0 to 1.0 µm (with an average length of 2.6 µm). The longest chromosome in prometaphase was almost four times as long as that of mid-metaphase, while the shortest one in prometaphase was approximately 1.5 times larger. Additionally, the sat-chromosomes were longer than other chromosomes in prometaphase, but not in metaphase (Fig. S2). Although we have not confirmed chromosome pairing by observing meiosis, the two shortest chromosomes might form a homologous pair, as well as the two sat-chromosomes. This indicates that each homologous chromosome has a different degree of condensation.

Fig. 2. Mitotic-prometaphase chromosomes in the somatic cell of L. aquatica (accession Tsukubamirai).

The chromosomes in the same cell were stained with orcein (A), DAPI (B), and CMA (C). GC-rich sites (yellow colored) of pericentromeric sites were clearly visible in the enhanced superimposed image (D) of CMA and DAPI. Two elongated GC-rich satellites (arrows) were attached to a nucleolus (dotted circle), which was observed at this stage by orcein staining (solid arrowhead). Moreover, two chromosomes are smaller than the others (open arrowhead). Scale bar=5 µm.

The DNA base-specific fluorescence banding reported here for the first time in the genus Limosella revealed that the pericentromeric regions of all chromosomes and the satellites were CMA positive and DAPI negative (CMA⁺DAPI⁻) in L. aquatica (Figs. 1B–D, 2B–D, S1). The proportion of each metaphase chromosome with CMA⁺DAPI⁻ pericentromeric regions ranged from 10 to 20%. Fluorescence banding by CMA and DAPI is commonly employed to detect GC- and AT-rich segments on chromosomes, respectively (Schweizer 1976). Therefore, chromosome banding indicates that the L. aquatica genome is composed of a chromosome set with a relatively large amount of GC-rich segments. Additionally, the nucleolar organizer regions (NORs), which contain the GC-rich tandem repeat 45S rDNA, can generally be stained with CMA⁺DAPI⁻ (Schweizer 1976, Galasso et al. 1995). Our 18S rDNA FISH results also showed that the two signals were located at exactly the same positions as the CMA⁺DAPI⁻ satellites. NORs are usually located at secondary constrictions or satellites (e.g., Albini and Schwarzacher 1992, Weisenberger and Scheer 1995).

In contrast to the relatively large proportion of GC-rich segments observed in the chromosomes, the genomic GC contents obtained by FCM (Fig. 3, Table 2) were unexpectedly lower than that of O. sativa (43.6%, International Rice Genome Sequencing Project and Sasaki 2005), measuring less than 40%. The GC-rich regions found that the chromosome bases of L. aquatica may be attributed to the relative AT-richness of other regions.

Fig. 3. FISH of the metaphase chromosomes (accession Kumamoto).

The chromosomes were stained with DAPI (A), CMA (B), and superimposed image (C) of CMA and DAPI, and 18S rDNA FISH counter-stained with DAPI (D). Two 18S rDNA signals (arrowheads) were located on the GC-rich satellites with CMA⁺DAPI⁻ (arrows). Scale bar=5 µm.

Our study supports previous reports that L. aquatica is tetraploid (Blackburn 1939, Salisbury 1967). Polyploidy is widely acknowledged as a major mechanism of adaptation and speciation in plants (Ramsey and Schemske 1998). Overall, our cytogenetic results for L. aquatica revealed that the maximum number of chromosomes that could be paired by karyomorphological similarity, such as satellite chromosomes, was two, rather than four. As a result, we considered allopolyploids, which form bivalent chromosomes with a strict pairing in meiosis for normal gametogenesis (Soltis and Soltis 2000, Pairon and Jacquemart 2005). Indeed, our in vitro plants not only actively generated ramets, but also produced many genets from the seeds with a short generation cycle almost similar to that of Arabidopsis thaliana. Reproductive isolation between the different genomes coexisting in the polyploid is thought to be due to the differentiation of homologous chromosomes derived from the same genome during the speciation of the parents. Therefore, it is inferred that L. aquatica is an amphidiploid of hybrid origin. It shows regular segregation in meiosis, which does not allow the pairing of the derived homologous chromosomes. This is accompanied by rearrangements following polyploidy formation. Considering the importance of rDNA integrity for vital activities, L. aquatica may have lost the 45S rDNA locus from one parental genome during speciation.

The novel findings on the chromosomes of this species obtained here will provide a basis for further cytogenetic investigation of species and populations of the genus Limosella.

References

• Albini, S. M. and Schwarzacher, T. 1992. In situ localization of two repetitive DNA sequences to surface-spread pachytene chromosomes of rye. Genome 35: 551–559.
• Blackburn, K. B. 1939. The Limosella plants of Glamorgan. Part II. Chromosome and species. J. Bot. 77: 67–71.
• Chepinoga, V. V., Gnutikov, A. A., Enushchenko, I. V., and Chepinoga, A. V. 2008. In: Marhold, K. (ed.). IAPT/IOPB chromosome data 6. Taxon 57: 1267–1273.
• Curtis, T. G. F., Ryan, J. B., and McGough, H. N. 1985. The status and ecology of Limosella aquatica L. in Clare (H9) and south-east Galway (H15). Ir. Nat. J. 21: 406–407.
• Dawson, M. I. 2000. Index of chromosome numbers of indigenous New Zealand spermatophytes. N.Z. J. Bot. 38: 47–150.
• Delgado, L., Martín, F. G., and Rico, E. 2015. In: Marhold, K. (ed.). IAPT/IOPB chromosome data 20. Taxon 64: 1344–1350.
• Galasso, I., Schmidt, T., Pignone, D., and Heslop-Harrison, J. S. 1995. The molecular cytogenetics of Vigna unguiculata (L.) Walp: The physical organization and characterization of 18s-5.8s-25s rRNA genes, 5s rRNA genes, telomere-like sequences, and a family of centromeric repetitive DNA sequences. Theor. Appl. Genet. 91: 928–935.
• Gervais, C. and Cayouette, J. 1985. Liste annotée de nombres chromosomiques de la flore vasculaire du nord-est de l’Amerique IV. Nat. Can. 112: 319–331.
• Greilhuber, J., Doležel, J., Lysák, M. A., and Bennett, M. D. 2005. The origin, evolution and proposed stabilization of the terms ‘genome size’ and ‘C-value’ to describe nuclear DNA contents. Ann. Bot. 95: 255–260.
• Hair, J. B. and Beuzenberg, E. J. 1960. Contributions to a chromosome atlas of the New Zealand flora—4. Miscellaneous families. N. Z. J. Sci. 3: 432–440.
• Hoshi, Y., Homan, Y., and Katogi, T. 2021. Colchicine induction of tetraploid and octaploid Drosera strains from D. rotundifolia and D. anglica. Cytologia 86: 21–28.
• Hoshi, Y., Praphruet, R., and Hironaka, K. 2019. Genome size determination and chromosome characterization of two marigold species (Asteraceae). Cytologia 84: 37–42.
• Hoshi, Y., Yagi, K., Matsuda, M., Matoba, H., Tagashira, N., Pląder, W., Stefan, M., Nagano, K., and Morikawa, A. 2011. A comparative study of the three cucumber cultivars using fluorescent staining and fluorescence in situ hybridization. Cytologia 76: 3–10.
• International Rice Genome Sequencing Project and Sasaki, T. 2005. The map-based sequence of the rice genome. Nature 436: 793–800.
• Ito, Y., Tanaka, N., Albach, D. C., Barfod, A. S., Oxelman, B., and Muasya, A. M. 2017. Molecular phylogeny of the cosmopolitan aquatic plant genus Limosella (Scrophulariaceae) with a particular focus on the origin of the Australasian L. curdieana. J. Plant Res. 130: 107–116.
• Katogi, T. and Hoshi, Y. 2022. Determination of chromosome numbers and genome sizes in six species of Byblis (Byblidaceae). Cytologia 87: 277–280.
• Kiehn, M., Vitek, E., Hellmayr, E., Walter, J., Tschenett, J., Justin, C., and Mann, M. 1991. Beiträge zur Flora von Österreich: Chromosomenzählungen. Verh. Zool. Bot. Ges. Wien 128: 19–39.
• Levan, A., Fredga, K., and Sandberg, A. A. 1964. Nomenclature for centromeric position on chromosomes. Hereditas 52: 201–220.
• Löve, A. and Löve, D. 1958. The American element in the flora of the British Isles. Bot. Notiser 111: 373–388.
• Löve, A. and Löve, D. 1982. In: Löve, A. (ed.). IOPB chromosome number reports LXXV. Taxon 31: 344–360.
• Matevski, V. and Kostadinovski, M. 2009. New locality of species Limosella aquatica L. in the flora of the Republic of Macedonia. Hladnikia 23: 77–80.
• Měsíček, J. and Javůrková-Jarolímová, V. 1992. List of Chromosome Numbers of the Czech Vascular Plants. Academia, Praha.
• Moore, D. M. 1981. Chromosome numbers of Fuegian angiosperms. Bol. Soc. Brot. 53: 995–1012.
• Murashige, T. and Skoog, F. 1962. A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiol. Plant. 15: 473–497.
• Pairon, M. C. and Jacquemart, A.-L. 2005. Disomic segregation of microsatellites in the tetraploid Prunus serotina Ehrh. (Rosaceae). J. Am. Soc. Hortic. Sci. 130: 729–734.
• Probatova, N. S. and Sokolovskaya, A. P. 1981. Chromosome numbers of some aquatic and bank plant species of the flora in the Amur River basin in connection with the peculiarities of its formation. Bot. Z. 66: 1584–1594. (in Russian)
• Ramsey, J. and Schemske, D. W. 1998. Pathways, mechanisms, and rates of polyploid formation in flowering plant. Annu. Rev. Ecol. Syst. 29: 467–501.
• Salisbury, E. J. 1967. The reproduction and germination of Limosella aquatica. Ann. Bot. 31: 147–162.
• Sawa, S., Kanzaki, C., Kamino, N., and Yoshida, Y. 2022. A new record of Limosella aquatica (Scrophulariaceae) from Shiranuka-cho, Shironuka-gun, Hokkaido, Japan. J. Phytogeogr. Taxon. 70: 181–182. (in Japanese)
• Schweizer, D. 1976. Reverse fluorescent chromosome banding with chromomycin and DAPI. Chromosoma 58: 307–324.
• Siljak-Yakovlev, S., Lamy, F., Takvorian, N., Valentin, N., Gouesbet, V., Hennion, F., and Robert, T. 2020. Genome size and chromosome number of ten plant species from Kerguelen Islands. Polar Biol. 43: 1985–1999.
• Šmarda, P., Bureš, P., Horová, L., Foggi, B., and Rossi, G. 2008. Genome size and GC content evolution of Festuca: Ancestral expansion and subsequent reduction. Ann. Bot. 101: 421–433.
• Šmarda, P., Bureš, P., Horová, L., Leitch, I. J., Mucina, L., Pacini, E., Tichý, L., Grulich, V., and Rotreklová, O. 2014. Ecological and evolutionary significance of genomic GC content diversity in monocots. Proc. Natl. Acad. Sci. U.S.A. 111: 4096–4102.
• Soltis, P. S. and Soltis, D. E. 2000. The role of genetic and genomic attributes in the success of polyploids. Proc. Natl. Acad. Sci. U.S.A. 97: 7051–7057.
• Son, S.-W., Lee, B.-C., Yang, H.-H., and Seol, Y.-J. 2011. Distribution of five rare plants in Korea. Korean J. Plant Taxon. 41: 280–286.
• Uozu, S., Ikehashi, H., Ohmido, N., Ohtsubo, H., Ohtsubo, E., and Fukui, K. 1997. Repetitive sequences: Cause for variation in genome size and chromosome morphology in the genus Oryza. Plant Mol. Biol. 35: 791–799.
• Weisenberger, D. and Scheer, U. 1995. A possible mechanism for the inhibition of ribosomal RNA gene transcription during mitosis. J. Cell Biol. 129: 561–575.
• Yamazaki, T. 1993. Scrophulariaceae. In: Iwatsuki, K., Yamazaki, T., Boufford, D. E., and Ohba, H. (eds.). Flora of Japan IIIa. Kodansha Ltd., Tokyo. pp. 326–374.