In total, 407 individuals were analyzed: 272 adults and 135 newly metamorphosed juveniles. Juveniles were analyzed separately, as we expected a higher frequency of polyploids than in adults because tetraplois and pentaploids do not survive until adulthood (Hermaniuk et al. 2013, Christiansen 2009). Adult individuals (79 P. lessonae, 82 P. ridibundus and 111 P. esculentus) were collected from 11 natural populations in Central Europe (Table 1). Juveniles (15 P. lessonae, 52 P. ridibundus and 45 diploid, 18 triploid and 5 tetraploid P. esculentus) were obtained from in vitro artificial crosses. Their parents (P. lessonae LL, P. ridibundus RR, and P. esculentus RL, RRL, RLL) were collected from natural populations and served for 28 crosses resulting from various parental combinations (Table 2). Artificial crosses and rearing of progeny were done according to standard protocol for water frogs (Berger et al. 1994). Taxonomic identity of individuals was determined by morphology (Berger 1983, Plötner 2005, Kierzkowski et al. 2011) and at least one of the following methods: actinomycin D (AMD)-DAPI staining (Schweitzer 1976, Heppich et al. 1982, Ogielska et al. 2004), LDH electrophoresis and 17 microsatellite loci (both methods described in Hauswaldt et al. 2012).

Frogs were collected in years 2002–2009. All specimens used in this study were collected according to legal regulations concerning wild species protection under the following permits: Agency for Nature Conservation and Landscape Protection of the Czech Republic, Poodrí Protected Landscape Area, c.j. 0926/PO/2008/AOPK, permit S/0673/PO/2008/AOPK, and Polish Ministry of Environment Protection and Forestry for performing studies on protected species OP 4072/218 /96, OP 4072/218/98/4501, OP 4201/144/99, and II Local Commission for Ethics in Experiments on Animals 13/02.

Twenty-four hours before dissection, adults were injected intraperitoneally with 1 ml, and juveniles with 0.5 ml of 0.3% colchicine (Sigma- Aldrich, St. Louis, Mo., USA). Shortly before tissue preparation, the frogs were anesthetized with 0.25% water solution of 3-aminobenzoic acid ethyl ester (MS 222, Sigma-Aldrich). The intestine was dissected, hypotonized in distilled water (20 min for adults and 10 min for juveniles), and fixed in fresh-made fixative ethanol:glacial acetic acid (3:1) according to Heppich et al. (1982). Samples were stored in the fixative at –20°C until use. To obtain chromosome preparations, small pieces of the intestine epithelium were gently pressed against a slide to make tissue ‘prints’ that were immediately squashed under a coverslip in a drop of 70% acetic acid. The slides were placed on dry ice until frozen and then the coverslips were mechanically removed. The squashes were air-dried overnight and stored at –20°C. A minimum of 10 metaphases were analyzed for each individual.

Chromosome preparations of three P. esculentus individuals were made in a different way. The animals were injected intraperitoneally with 0.1% colchicine (10µl/g body weight) 2.5 h before dissection, femur bones were removed from euthanized frogs, epiphyses were clipped off, and bone marrow cavities were immediately flushed with 0.075 M KCl solution applied with a syringe at 37°C. Bone marrow tissue was pressed through a small-mesh gauze, flushed out with 0.075 M KCl solution, then placed in a centrifuge tube filled up to 7 ml with 0.075 M KCl solution, and incubated at 37°C for 20 min. Hypotonic treatment was stopped by fixation in absolute methanol:glacial acetic acid (3:1), and cell suspension was centrifuged at 1,500 rpm for 10 min. The supernatant containing fat droplets was discarded with a Pasteur pipette and fresh fixative was added up to 5 ml. The pellet was re-suspended by agitation and kept in a freezer for 20 min, then centrifuged again at 1,500 rpm for 10 min. The procedure (centrifugation, fixative exchange, and cooling) was repeated 3 times. The suspension was then transferred to a 1-ml syringe, dropped onto slides (previously cleaned in HCl:ethanol, 3:1), and finally air-dried. Chromosome preparations obtained with either protocol were suitable for all staining methods used in this study. However, chromosomes prepared from bone marrow cells were more uniformly condensed and thus more suitable for FISH than the chromosomes obtained from gut epithelium prints that varied in the degree of condensation (Zaleśna et al. 2011).

The nucleolus organizer regions (NORs) were stained in all individuals by the silver nitrate technique (Ag-NOR). Chromosome slides of 15 randomly selected individuals were sequentially stained with chromomycin A₃ (CMA₃) and AgNO₃. All individuals that displayed another number of NORs than one per a haploid set after Ag-NOR staining were examined by PI or CMA₃. For 22 individual (9 RR, 3 LL, 5RL, 1 RRL, 3 LLR, 1 RRLL) we applied the FISH method with 28S rDNA as a probe.

Silver staining. We followed the protocol of Howell and Black (1980). A few drops of freshly prepared silver nitrate buffer (0.5g AgNO₃/1ml H₂O/0.5ml gelatin solution, i.e.1g gelatin/0.5ml formic acid/50ml bi-distilled water) were applied to each preparation. Slides were covered with a nylon mesh and incubated in a humid chamber at 60°C for 1min, washed in distilled water and air-dried. If the chromosomes were poorly visible, they were counterstained with DAPI. The number of active rDNA loci was documented by simultaneous use of fluorescence and incandescent light.

Chromomycin A₃ (CMA₃). The method was used according to Schweitzer (1976). Slides were incubated in McIlvain buffer pH 7.0 with 2.5mM MgCl ₂ for 10 min and stained with CMA₃ solution (0.5 mg/mL buffer and 2.5mM MgCl₂) for 15 min in the dark, briefly rinsed in buffer and counterstained with methyl green for 15 min (0.175g methyl green/50ml buffer). After washing in buffer, slides were stained with DAPI solution (0.5µg/ml) for 10 min in the dark and briefly rinsed with buffer. Tissue prints were mounted in 50% glycerol and analyzed using fluorescence illumination. After CMA₃ staining, preparations were faded and cleaned in xylene and benzene, each bath for 2 min, and then stained with AgN0₃.

Propidium Iodode staining after denaturation of chromosomes (PI). Tissue prints were dehydrated in 70%, 85% and 95% ethanol (30 sec each wash), air-dried at room temperature and denatured in 70% deionized formamide in 2xSSC at 70°C for 3.5 min. Slides were immediately dehydrated in chilled 70% ethanol from a freezer for 2 min, then in 85% and 95% at room temperature for 30 sec. Chromosomes were stained with PI (200ng/ml).

We applied the FISH method with 28S rDNA as a probe, according to the protocol of Traut et al. (1999). Isolation of 28S rDNA was carried out with use of commercial GeneMATRIX Bio-Trace Purification Kit (Eurix). DNA amplification was performed using PCR reaction with suitable mixture: 2.5µl PCR buffer in 15 mM MgCl2 (Eurix), 1µl 5mM dNTP (Eurix), 1µl 10mM primer A (5’-TCC GTG TTT CAA GAC GGG - 3’) and 1 µl 10mM primer B (3’-ACC CGC TGA ATT TAA GCA T -5’), 1µl 1U/µl Polymerase (OptiTaq-Eurix), 1µl matrix DNA and 17.5µl water. PCR conditions included initial denaturation in 94°C for 5 min, followed by 35 cycles: 30 s denaturation (94°C), 30 s annealing (58°C), 1 min elongation (72°C), and 5 min of final elongation (72°C). The 28S rDNA probe was labelled with tetramethyl-rhodamine-6-dUTP (Roche, 11093088910) using a Nick Translation Mix according to the manufacturer’s protocol (Roche, 10976776001). Then, DNA was precipitated by ethanol for purification and concentrated by adding one-tenth volume of 3M sodium acetate and 2.5 volume of chilled 96% ethanol from a freezer. The mixture was incubated for at least 15 min at –80°C. The precipitated DNA was spun at 15,000 rpm for 15 min at 4°C and the supernatant was discarded. The pellet was washed with 50–100 ml of chilled 70% ethanol and spun at 15,000 rpm for 10 min at 4°C. DNA was dried at 37°C for 15–20 min.

Slides were incubated in 100 µg/ml DNase-free RNase A in 2x SSC for 1 h at 37° C in a humid chamber and then washed twice in 2xSSC at room temperature for 5 min, dehydrated in ethanol series (70%, 85%, and 95%, 30 s each) and air-dried. Chromosome preparations were denatured in 70% deionized formamide, 2x SSC at 70° C for 3.5 min, dehydrated immediately for 2 min in ice-cold 70% ethanol, then in 85% and 95% ethanol for 30s at room temperature, and finally air-dried. The 28S rDNA probe was dissolved in hybridization mixture consisted of 100% deionized formamide and 20% dextran sulfate in proportion 1:1. The mixture was denatured at 90° C for 5 min and then immediately placed on ice for 3 min. 20 µl of the probe was applied to each slide and covered with a coverslip. Hybridization lasted overnight in a dark humid chamber at 37° C. After hybridization, the coverslip was removed by rinsing with 50% formamide in 2xSSC at 42°C twice for 7 min and the slide was washed 3 times (7 min each) in 1xSSC at 42°C and then in 2xSSC at room temperature for 30 s. Chromosomes were counterstained with DAPI in Vectashield antifade buffer (Cambio, Cambridge, UK).

Chromosomes were analyzed in Olympus Provis AX 70 or Carl Zeiss Axioskop 20 microscopes equipped with fluorescence lamp HBO50 and appropriate filters. Images were recorded with Olympus DP30BW CCD and cooled Carl Zeiss AxioCam HRc CCD cameras and processed using AxioVision and Lucia ver. 2.0 (Laboratory Imaging) softwares.

In 66 individuals (16.2%, 13 adult and 53 juveniles), the number of NORs differed from the expected values. Lack of NORs was relatively more frequent (13.8%, 56 out of 407 individuals) than extra NORs (2.46%, 10 out of 407 individuals), regardless of genomic compositions (RR, LL, RRL and LLR) and origin of individuals. Eight diploid adult individuals (3 diploid P. esculentus, 3 P. lessonae and 2 P. ridibundus) and 43 diploid juveniles (11 P. esculentus, 6 P. lessonae, and 26 P. ridibundus) had NORs only on one chromosome of pair 10 (Fig. 2C). Two triploid adults (one LLR and one RRL) and one juvenile displayed two, instead of three NORs (Fig. 2F). Two other triploid RRL individuals (1 adult and 1 juvenile) displayed one, instead of expected three NORs in chromosomes 10 (Fig. 2G).

Additional NORs were detected in ten individuals. The extra NORs were located within additional secondary constrictions on one of the homologs of chromosomes other than 10. Extra NORs were found in one adult P. ridibundus in the distal position of the long arm of chromosome 9 (Fig. 2 B). In interphase nuclei of this individual we found the corresponding number of three Ag-nucleoli (Fig. 1D). Extra NORs were also observed on short arms of chromosomes 2 in nine triploid P. esculentus RRL: one adult and eight juveniles (Fig. 1E).

The lack of one NOR was also detected in diploid progeny of two triploid RRL females and one LLR female. After AMD/DAPI staining we discovered that in the case of the RRL females the lack of NORs was inherited together with the haploid set of the ridibundus chromosomes, whereas in the case of the LLR female it was inherited with the haploid set of the lessonae chromosomes. Among diploid progeny of these females (altogether 37 juveniles) we observed 16 individuals with two NORs and 21 individuals with only one NOR. Each individual was stained with AgNOR and seven of them were confirmed by FISH. Thus, despite that the genomes of ova were the same (R in the case of RRL and L in the case of LLR females), the females produced in fact two types of gametes – with and without NORs (Fig. 3).

Figure 1.

A The variability of size of AgNORs (black bands) on chromosomes 10 a Pelophylax ridibundus b P. lessonae c P. esculentus RL d P. esculentus RRL e P. esculentus LLR B–E Interphase nuclei of P. esculentus with Ag-nucleoli visualized as black dots by AgNO₃B diploid RL with 1 AgNOR C diploid RL with 2 AgNORs D triploid RRL with 3 AgNORs E triploid RRL with 4 AgNORs F metaphase chromosomes and interphase nuclei of the same tetraploid P. esculentus RRLL with 4 sites of hybridization with 28S rDNA probe (FISH) (arrows).

Figure 2.

Localization and number of NORs (arrows) in diploid and triploid water frogs. Left column represent diploids (A–C) and right column represent triploids (D–G). Each picture is composed of a diagram of karyotype with black dots representing NORs and metaphase plates stained with silver (AgNOR), chromamycin A₃ (CMA₃), propidium iodide after denaturation (DPI) or after fluorescent in situ hybridization with 28S rDNA probe (FISH). A P. lessonae LL with 2 NORs, AgNORs (left) and FISH (right) B P. esculentus RL with 4 NORs on chromosome 9, CMA₃ (left) and FISH (right) C P. esculentus RL with 1 NOR, AgNORs (left) and FISH (right) D P. esculentus LLR with 3 NORs, AgNORs (left) and FISH (right) E P. esculentus RRL with 4 NORs, AgNOR (left) and DPI (right) F P. esculentus RRL with 2 NORs, AgNORs (left) and FISH (right) G P. esculentus RRL with 1 NOR, CMA₃.

Figure 3.

Inheritance of NORs by progeny of triploid females RRL (right) and RLL (left). The chromosome set represented by one copy is eliminated before oogenesis (marked by X) and the double one is segregated into eggs (represented by ovals). After fertilization (in this case by ridibundus sperm), two types of progeny arises: with two (A and C) and with one (B and D) NOR. The lack of NORs are transmitted either by lessonae (white) or by ridibundus (dark grey) chromosome sets. NORs are represented by black dots.

The variability of AgNOR sizes in individuals from the same populations may reflect different amount of rDNA (Miller and Brown 1969 for Bufo marinus and Macgreor et al. 1977 for Plethodon cinereus). Various sizes of AgNORs in frogs were described by Schempp and Schmid (1981). Schmid (1982) found a correlation between sizes of bands in secondary constrictions of AgNO₃ stained chromosomes and the intensity of CMA₃ signals which were proportional to the amount of rDNA in NORs. We also found variability of NOR sizes in chromosomes 10, however Biriuk et al. (2015) reported that the differences measured as a relative length were not significant in 65 investigated individuals of diploid and triploid green frogs. The polymorphism of NORs can also reflect geographical karyotypic variations (Amaro et al. 2012, Silva et al. 2000, Quinderé et al. 2009, Bi et al. 2009). However, we found no variations of NOR number and size in water frogs deriving from different populations.

Extra NORs were also reported in Rana catesbeiana (Lithobates catesbeianus) that displayed from two to seven Ag-positive small NORs per haploid karyotype, apart from one “standard” NOR on the chromosome 10 (Schmid 1978a,b). In the same species, Vitelli et al. (1982) revealed the standard NORs on the chromosomes 10 by FISH with 18S+28S rDNA probe, whereas 5S rDNA sequences were clustered near the centromere on the short arm of chromosome 12 and corresponded to small Ag-positive bands. Other small NORs were not confirmed neither by FISH with the 18S + 28S rDNA nor by 5S rDNA probes. FISH with 18S+28S rDNA probe in Pelophylax esculentus (Vitelli et al. 1982) displayed signals in chromosome 10, and with 5S rDNA probe at telomere of short arm of chromosome 5. Moreover, FISH with 28S rDNA probe, as well as CMA₃ and PI stainings revealed that extra NORs corresponded with the extra Ag-positive signals on the chromosomes 9. Neither of these locations corresponded to extra NORs that were observed in our study.

More common was the lack of NORs (13.8% of all animal studied herein). We observed a lack of one NOR (2n and 3n) or two NORs (only in 3n), both in metaphase plates and interphase nuclei. Lack of NORs in amphibians was also reported by Schimd (1982) in Bombina variegata, Xenopus laevis, and Bufo fowleri. According to Schmid (1982) and Motovali-Bashi et al. (2004), the causes of deletion or amplification of NORs were usually unequal meiotic divisions, sister chromatid exchanges or disruption of DNA replication.

Intraspecific polymorphism of the location of extra NORs was also reported in Hyla chrysoscelis and H. versicolor (Wiley et al. 1989, Wiley and Little 2000), H. nana(Medeiros et al. 2003), Bufo terrestris (Foote et al. 1991), Agalychnis callidryas (Schmid et al. 1995), Physalaemus petersi (Lourenço et al. 1998), P. cuvieri (Quinderé et al. 2009), Phyllomedus arhodei and P. nordestina (Barth et al. 2013). Possible mechanisms of the dispersion of NOR sites were translocations involving chromosomal segments containing NORs, however such chromosome rearrangements were connected with change in morphology of the NOR chromosomes (Lourenço et al. 1998). Other possible mechanisms involved in dispersion of NORs in anuran genomes could be inversions, transpositions by mobile genetic elements containing NORs (Wiley et al. 1989, Foote et al. 1991, Lourenço et al. 1998, Schmid et al. 1995, King et al. 1990, Bruschi et al. 2012, Carvalho et al. 2014) or rDNA amplification (Barth et al. 2013). However, in the case of water frogs presented here, translocations involving chromosomal segments containing NORs on chromosomes 2 and 9 seems unlikely because chromosomes with extra NORs did not vary in length and shape from relative chromosomes in other individuals.

The lack of one NOR was detected in both RRL and LLR triploid females and was associated with the ridibundus and lessonae chromosome sets, respectively. Triploid water frogs produce haploid gametes and transmit the chromosome set that is represented by two copies (R in RRL and L in LLR) (Christiansen 2009). Because one of the doubled sets lacked NOR, we were able to trace this character and we found it in about a half of progeny whereas the other half inherited NORs (Fig. 3). This intriguing result deserves more attention and further studies because it may serve as an indicator of clonal versus recombined inheritance of the doubled chromosome sets, as was possible in triploid green toads Bufo baturae (Stöck et al. 2002, 2012). This all-triploid population is represented by males and females that carry NORs in two chromosome sets whereas the third one lacks NORs. This character, known also from other amphibian species (Schmid 1982, Motovali-Bashii et al. 2004, this study), was probably conserved in an isolated population in the Karakoram Mountains in Pakistan. Triploid Bufo baturae females produce diploid ova containing one (recombined) set with NORs whereas the set without NOR is transmitted without recombination (clonally).

Finally, we conclude that polymorphism concerning the number and localization of NORs in water frogs was characteristic of both lessonae and ridibundus genomes, and – as expected – was observed in individuals regardless of their taxonomic position (P. lessonae, P. ridibundus and P. esculentus), ploidy level (2n, 3n, 4n), genomic constitution (RR, LL, RRL, RLL and LLRR), and collection site. The number of active silver-stained NORs reflected ploidy levels: two in diploids, three in triploids, and four in tetraploids, and therefore we believe that there is no diploidization of polyploids in water frogs, as has been observed in some of polyploid amphibians (for review see Schmid et al. 2015). The variability of NORs may be hereditary, what is best represented by inheritance of lack of NORs by the progeny of triploid females.