Nuclear DNA Content, Selected Morphological and Anatomical Traits of Narcissus Cultivars and Breeding Clones

: The genus Narcissus belongs to the family Amaryllidaceae. This genus has been the subject of numerous cytological and cytometric studies and have shown enormous variation in terms of genome size, ploidy level, and even the basic chromosome number. The basic chromosome numbers are 5 or 7, but 10, 11, and 12 have been recorded as well. Most narcissus cultivars are euploid tetraploids. There are also numerous triploids. Some cultivars are aneuploid such as tetraploids or triploids, with missing chromosomes or possessing additional chromosomes. Due to their very complex parentage, cultivars have various numbers of chromosomes not found in the species. In this publication, we present a study on the genome size and assessment of the likely ploidy level of 38 cultivars and breeding clones of Narcissus in relation to their selected morphological traits and information on their parental forms. For the ﬁrst time, 12 Polish cultivars and breeding clones of narcissus were the subject of such an evaluation. Perianth diameter, leaf length, and width were evaluated and rated with notes according to the descriptor of the International Union for the Protection of New Varieties of Plants. Stomatal density and stomata length were measured using light microscopy. Analysis of genome size was carried out using ﬂow cytometry. For three selected genotypes, the chromosome number was counted. Our results lead to the general conclusion that the morphological traits studied and nuclear DNA content can be useful for determining the possible ploidy level of narcissi. The information on the origin and parental forms of narcissi can be helpful in determining the ploidy level of narcissi. However, clear conﬁrmation of ploidy level requires veriﬁcation of chromosome number and preferably karyotyping. The results obtained are a prelude to further studies.


Introduction
Narcissus L. (narcissus, daffodil) is one of the most important spring-flowering bulbous crops cultivated worldwide. Uses of Narcissus includes bulb production, the cultivation of cut flowers from the ground, cut and potted flowers from greenhouses, and the use of bulbs in gardens and urban green areas. The largest producer of Narcissus bulbs is the United Kingdom with an area of over 4000 ha [1]; second are the Netherlands with an area of 1280 ha [2]. The narcissus industry in United Kingdom is estimated to have an annual output value of around 45 million pounds [1]. Bulbs of this crop on the area larger than 50 ha are also produced in Australia, Canada, Israel, USA, and Poland [3]. A huge number of cultivars of the genus Narcissus are grown worldwide. The last International Daffodil Register and Classified List of Royal Horticultural Society, London, UK, contains 26 thousand names of genotypes ordered in 13 groups (divisions) of horticultural classification [4]. Commercial cut-flower production is dominated by cultivars from the trumpet (i.e., 'Dutch Master', 'Golden Harvest') and large-cupped (i.e., 'Carlton') daffodil divisions, and the most important cultivar in pot production is 'Tète-ā-Tète' from division 12 (other daffodil cultivars; former division 6-cyclamineus daffodil cultivars). More than 50% of newly bred and registered cultivars in recent years also belong to the trumpet and large-cupped daffodil divisions [5].
The genus Narcissus belongs to the family Amaryllidaceae and has a mainly Mediterranean distribution. The species of Narcissus originate primarily from the Iberian Peninsula, they also occur in south-western France, northern Africa and eastwards to Greece, in Asia Minor and Syria [3,6]. N. tazetta L. originated from the Mediterranean region, is doubtless subspontaneous in Persia and Cashmir [7] and was also introduced to China and Japan in ancient times and known now as N. tazetta var. chinensis (M. Roem.) Masam. & Yanagih [6].
In horticultural production, narcissi are propagated vegetatively by means of daughter bulbs (offsets), but also by bulb segmentation (chipping), twin-scaling, and micropropagation in vitro [3,6]. Production from seeds is, of course, the method used by breeders and is also important for species that form few offsets [3]. In nature, species reproduce in both ways (i.e., by producing progeny bulbs and by seeds). There are great taxonomic difficulties with the genus Narcissus since many species have been cultivated for a long time, there has been extensive selection of cultivars and ecotypes, and there has been extensive natural hybridization in their original environment, with intentional breeding by humans for several hundred years [8]. This situation has promoted intensive genetic changes, including those related to chromosome number, DNA content, and plant ploidy levels. Marques et al. even suggest that natural hybrid populations are composed of a mixture of markedly different hybrid genotypes produced either by structural chromosome changes, consistent with classic cytogenetic studies in Narcissus, or by transposon-mediated events [9].
Most narcissus cultivars are euploid tetraploids. However, there are also numerous triploids [15,16]. Some cultivars are either aneuploid tetraploids or triploids with missing chromosomes or possesing additional chromosomes. According Ramanna et al. [24], among ornamental geophytes there is a tendency to replace diploids with polyploids cultivars, which is especially visible in Narcissus and Lilium. Due to the very complex parentage, cultivars have various numbers of chromosomes not found in the species, e.g., some diploid cultivars have been reported to possess 24 chromosomes ('Dutch Master') and 28 or 29 chromosomes ('Beersheba') [19,24]. In addition, a higher number of chromosomes (28) was reported by other researchers [25,26]. On the other hand, the number of 24 chromosomes minus or plus 1-2 chromosomes have been recorded for triploid cultivars, e.g., 'Tète-ā-Tète' and 'Cheerfulness' based on karyotype analysis [18]. Thus, knowing the chromosome number and/or genome size it difficult to state clearly the ploidy level without knowledge of cultivar parentage and appearance.
It is common knowledge that, in many cases, higher ploidy level of ornamental plants manifests itself in bigger flowers, bigger leaves, and better plant vigour in relation to their diploid counterparts (which was confirmed for Hemerocallis [27,28] and Tulipa [24,29]). Many studies also confirm the relationship between the size of stomata and stomata density with the level of ploidy of ornamental plants, including geophytes. There are reports that as ploidy levels increase, the stomatal density decreases [30] but stomata size increases [27,28,30,31]. It can also be assumed that for the genus Narcissus, such relationships exist and morphological traits can be found that can become markers of ploidy level.
In this publication, we present a study on the genome size and assessment of the likely ploidy level of 38 cultivars and breeding clones of narcissus (Narcissus L.) in relation to their selected morphological and anatomical traits, and information on their parental forms. For the first time, 12 Polish cultivars and breeding clones of narcissus were the subject of such studies.
Studying the genome size and assessing the ploidy level of Narcissus cultivars collected in the genebank, representing most groups of the horticultural classification of the genus Narcissus, has a scientific but also a practical aspect for use in breeding aimed at obtaining plants with more robust and showy flowers. Finding morphological markers related to the ploidy level of plants may be helpful in developing a rapid identification method useful for cultivar recognition and/or breeding as a preliminary test for ploidy assessment.

Plant Material
Thirty-eight genotypes (cultivars and breeding clones) of narcissus (Narcissus L.), including twelve Polish ones, derived from the genebank located at the National Institute of Horticultural Research, Skierniewice, Poland, were used for the study. All genotypes were grown in the field and included in the genebank at least five years before the study. Evaluated narcissus accessions belonged to different horticultural classification's divisions according to The International Daffodil Register & Classified List of the Royal Horticultural Society (RHS), London, UK [4] (Table 1). All cultivars and clones included in the study are listed alphabetically in Table 1 together with the names of the horticultural classification division, the flower color code, and information on their parental forms and year of registration or breeder's creation. The used international flower color code according to RHS [4] provides the information about the basic colors of the flower perianth segments and the corona: W, white or whitish; G, green; Y, yellow; P, pink; O, orange; R, red. The letter(s) before the hyphen describe the perianth segments, and after the hyphen they describe the corona. The genotypes tested represented 6 out of 13 divisions of the RHS horticultural classification, and most belonged to the divisions of large-cupped daffodil cultivars and trumpet daffodil cultivars, which are still the most abundantly represented in the horticultural market and still bred for the most part [5].

Phenotype Evaluation of Plants
The phenotype evaluation of 38 genotypes of narcissus based on three selected characteristics of the International Union for the Protection of New Varieties of Plants (UPOV) descriptor for the genus Narcissus was made [32]. The perianth diameter, leaf length, and width of fifteen plants of each genotype were evaluated and rated with notes according to the UPOV descriptor [32]. Perianth diameter and notes were determined according to the following rules: <6 cm (small, note 3); 6-9 cm (medium, 5); >9 cm (large, 7), leaf length: <20 cm (short, note 3); 20-30 cm (medium, 5); >30 cm (long, 7) and leaf width: Agronomy 2022, 12, 648 4 of 12 <1 cm (narrow, note 3); 1-2 cm (medium, 5); >2 cm (broad, 7). The three morphological traits above were chosen due to their frequent correlation with the ploidy level of many ornamental plant species.

Stomatal Density and Stomata Length
Stomata were measured using light microscopy. The leaf samples (2 cm in length) were collected from the part distant 10 cm from the leaf tip, in mid-May, after the flowering period. The abaxial epidermis was isolated with a transparent adhesive tape and stained with toluidine blue and next mounted on slides for microscopic observations according to the procedure of Dyki and Habdas (1996) [33]. The stomata lengths were determined for three leaves (×10 stomata) of each genotype counted and measured using a Nikon Eclipse 80i microscope with the program NIS-Elements BR 2.30 at 200 times magnification. For each sample, the stomatal density (i.e., number of stomata per field of view) was counted on the basis of the 15 fields of view.

Nuclear DNA Content
Analysis of genome size was carried out using flow cytometry (FCM/PI) (CyFlow PA, Partec, Münster, Germany). Samples were taken in mid-Mai from six leaves collected from six plant of each analyzed cultivar. Leaf tissue (0.5-1 cm 2 ) was chopped together with a piece (1 cm 2 ) of plant internal standard in a Petri dish in 0.5 mL nuclei isolation Galbraith's buffer [34] to which propidium iodide (50 µg/mL) and RNasa (50 µg/mL) were added [35]. As an internal standard, the young leaves of Vicia faba 'Inovec' (2C = 26.9 pg DNA) were used. The exception was the cultivar with the smallest genome, 'Bridal Crown' (27. 69 pg), which was analyzed with the standard of Agave americana (15.9 pg) since the peaks of V. faba 'Inovec' overlapped with those of the narcissus genotype tested. After adding 1.5 mL of the isolation buffer, the samples were filtered through a 30 µm filter and incubated for Agronomy 2022, 12, 648 6 of 12 30 min in room temperature. The fluorescence of the nuclei was measured using CyFlow Ploidy Analyser with CyView software (CyFlow PA, Partec, Münster, Germany) with an Nd-YAG green laser at 532 nm. Data were analysed by means of CyView software. The 2C DNA content of a sample was calculated as the sample peak mean divided by the standard plant peak and multiplied by the amount of DNA of the standard plant. Samples with at least 5000 nuclei were measured for six leaves (one leaf of each plant) with two runs from each nuclei isolation extract.

Chromosome Number
For three selected genotype the chromosome number were counted. Our selection included three cultivars representing different ploidy levels ('Dutch Master' as a diploid, 'Yellow Cheerfulness' as a triploid and 'Ice Follies' as a tetraploid). Bulb roots were dipped in a 0.25% aqueous solution of colchicine for 20-24 h. The root tips were macerated with 0.1 M HCl for 30 min at 60 • C, followed by isolation and fixation of them for 24 h in a mixture of acetic acid and ethanol (1:3). Crushed, microscopic slides of the roots with dry ice were stained with 2% aceto-orcein for 24 h. The chromosome counts were made using a Nikon Eclipse 80i microscope. The evaluation of metaphase chromosomes in the cells of root meristems was carried out. At least three metaphases were observed and compared in one root.

Statistical Analyses
To compare the degree of variation in nuclear DNA content, the standard deviation (SD) was determined. The results of number and length of stomata were analyzed by the analysis of variance (ANOVA) and post hoc Duncan's Multiple Range test at the 5% significance level, using SPSS, the version PS IMAGO 4.0 (IBM Statistics 24).

Phenotype Evaluation of Plants
Among all of the assessed genotypes of narcissus, only three cultivars were characterized by a small diameter of the perianth (<6 cm): double (division 4) 'Bridal Crown' and 'Yellow Cheerfulness', and dwarf 'Tète-ā-Tète' from division 12 (other daffodil cultivars), large (6-9 cm) by 13 genotypes, and medium (>9 cm) by the remaining 22 genotypes (Table 1). Only one genotype, the dwarf cultivar 'Tète-ā-Tète' had short leaves (<20 cm in length). Five genotypes had medium length leaves (20-30 cm), and thirty-two genotypes had long leaves. Regarding leaf width, the dwarf 'Orange Prince' from division 8 (tazetta daffodil cultivars) had narrow leaves (<1 cm), 2 breeding clones (0.919-a and 7/97) had wide leaves (>2 cm), all other genotypes had medium leaf width (1-2 cm). Some of the traits described appear to be related to the level of ploidy of the genotype (Tables 1 and 2). Among others, the diploid cultivar 'Bridal Crown' and triploid cultivars 'Yellow Cheerfulness' and 'Tète-ā-Tète' were characterized by small perianth diameter. Additionally, the cultivar 'Tète-ā-Tète' had short leaves and the triploid 'Orange Prince' was characterized by narrow leaves. Tetraploids had medium to high values of the morphological traits studied. This correlation was not confirmed only for the triploid cultivar 'Caruso', which had medium perianth diameter and long and medium leaf length and width, respectively. Obtained results confirmed other reports concerning phenotype evaluation of higher ploidy level of ornamental geophytes in relation to their diploid counterparts. Leaves and flowers of obtained tetraploid plants of daylilies (Hemerocallis) were significantly larger in comparison to their diploid parent cultivars [27,28]. Also, polyploid tulips, especially triploid cultivars, proved to be superior to their diploid parents, usually having larger flower size, sturdier stems, broader and thicker leaves, or more compact plants [24,29]. Among ornamental geophytes, there is a tendency to replace diploids with polyploid cultivars, a trend which is especially visible in Narcissus, Lilium, and Hemerocallis [24,36,37]. At present, nearly 75% of Narcissus cultivars are tetraploids while the diploids and triploids amount to only about 12% each [38]. Polyploids, including tetraploids of many ornamental crops, have taken a Agronomy 2022, 12, 648 7 of 12 leading position among the cultivars due to desirable traits such as vigorous growth and larger flowers, sometimes with more intense color or other interesting characteristics. In the case of polyploidization of diploid cultivar of Tulipa [31] smaller flowers and fragile stems were obtained in tetraploids. Despite these disadvantages, the obtained tetraploids are characterized by a compact plant habit and more fringed tepals, which can be considered advantages. Our results do not always strongly indicate a relationship between the morphological traits studied and the level of ploidy, but they do indicate such a tendency.

Stomatal Density and Stomata Length
The stomatal density ranged from 5.80 (for tetraploid 'Moneymaker') to 18.40 (for diploid 'Bridal Crown') ( Table 2). This confirms other reports that as ploidy levels increase, the stomatal density decreases. The stomatal density was reduced in tetraploid plants of Lilium regale compared to diploid [30] and in polyploids of Tagetes erecta [39]. A relatively high number of stomata (12.47) was also recorded for the cultivar 'Lajkonik', which could be considered a tetraploid based on the amount of cDNA. This could indicate either a different level of ploidy in this cultivar or a lack of correlation between a higher level of ploidy and a lower number of stomata. The predominant majority of the values obtained are between 6.73 and 8.73 pcs.
The size of stomata (the average length of 30 measurements) ranged from 46.1 for diploid cultivar 'Bridal Crown' to 60.9 and concern tetraploids (Figure 1A-C; Table 2). The difference between the shortest stomata and the longest was nearly 34%. The smallest stomata was less than in 34 of the 38 genotypes tested. Our results confirm other reports. Stomata of tetraploid plants of daylilies (Hemerocallis) were longer by 30-35% than those of diploid parent cultivars [27,28]. In the case of tulips (Tulipa) stomata were longer by 30% in tetraploids as compared to diploids [31]. An increase in the length of stomata was found also in polyploid plantlets of lilies (L. regale) [30] and African marigold (T. erecta) [39]. The study of McGoey et al. [40] in the case of Crataegus shows that stomata size may be useful in differentiating between tetraploid C. douglasii and diploid and triploid C. suksdorfii. They found that stomata differed between the two species, with C. douglasii having larger average stomata sizes than C. suksdorfii. found also in polyploid plantlets of lilies (L. regale) [30] and African marigold (T. erecta) [39]. The study of McGoey et al. [40] in the case of Crataegus shows that stomata size may be useful in differentiating between tetraploid C. douglasii and diploid and triploid C. suksdorfii. They found that stomata differed between the two species, with C. douglasii having larger average stomata sizes than C. suksdorfii.

Nuclear DNA Content
For 33 tested genotypes (12 Polish cultivars and breeding clones and 21 foreign cultivars), the DNA content ranged from 46.89 to 52.2 pg (Table 2), which indicates that they can be preliminary considered tetraploids based on the results for the reference cultivar 'Ice Follies' with a known from the literature number of chromosomes 2n = 4x = 28 [16,17,26] and 48.80 pg of nuclear DNA [17]. For four cultivars, the nuclear DNA content ranged from 39.21 pg to 41.42 pg and these cultivars can be considered triploids. In reference triploid cultivars 'Tête-à-Tête' and 'Yellow Cheerfulness', having 2n = 3x = 24 chromosomes, 2C value was 39.21 and 40.80 pg, respectively, which were very close to the reported by Zonneveld [11]. In addition, more recent studies confirmed that cultivar 'Tèteā-Tète' (originated from crossing diploid hybrid Cyclataz 2n = 2x = 17, what means crossing of Narcissus cyclamineus 'Soleil d'Or' and N. tazetta, with unknown diploid parent with 2n = 14), has indeed two genomes of N. cyclamineus and one genome of N. tazetta together with a B chromosome (2n = 3x = 24 + 1B) [14,34]. Based on the 2C value,

Nuclear DNA Content
For 33 tested genotypes (12 Polish cultivars and breeding clones and 21 foreign cultivars), the DNA content ranged from 46.89 to 52.2 pg (Table 2), which indicates that they can be preliminary considered tetraploids based on the results for the reference cultivar 'Ice Follies' with a known from the literature number of chromosomes 2n = 4x = 28 [16,17,26] and 48.80 pg of nuclear DNA [17]. For four cultivars, the nuclear DNA content ranged from 39.21 pg to 41.42 pg and these cultivars can be considered triploids. In reference triploid cultivars 'Tête-à-Tête' and 'Yellow Cheerfulness', having 2n = 3x = 24 chromosomes, 2C value was 39.21 and 40.80 pg, respectively, which were very close to the reported by Zonneveld [11]. In addition, more recent studies confirmed that cultivar 'Tète-ā-Tète' (originated from crossing diploid hybrid Cyclataz 2n = 2x = 17, what means crossing of Narcissus cyclamineus 'Soleil d'Or' and N. tazetta, with unknown diploid parent with 2n = 14), has indeed two genomes of N. cyclamineus and one genome of N. tazetta together with a B chromosome (2n = 3x = 24 + 1B) [14,34]. Based on the 2C value, two other cultivars 'Caruso', 'Orange Prince' could be also triploids. On the other hand, in the cultivar 'Orange Prince' with a DNA content of 41.42 pg, having in its pedigree N. poeticus and N. tazetta, the number of chromosomes may be 2n = 26.
The lowest content of nuclear DNA (27.69 pg) was recorded for the cultivar 'Bridal Crown' from the group of double narcissi, which may confirm the literature data on the 2C DNA value of 28.2 pg and the number of chromosomes 2x = 17 [11]. This cultivar may be considered a diploid. Most of the diploid cultivars possess 14 chromosomes. However, in some diploid cultivars, the number of chromosomes ranges from 10 to 28. This is due to the fact that the origin of the majority of cultivars is very complex. The cultivar 'Bridal Crown' is a sport of the cultivar 'L'Innocence' with 2n = 17, which originated from a cross between N. poeticus (2n = 14) and N. tazetta (2n = 4x = 20) or cultivars derived from these species. None of the Polish cultivars or breeding lines turned out to be diploid.
Surprisingly, two of the cultivars-'Dutch Master' and 'Fortissimo'-showing 2C values of approximately 48 pg having 24 [19] or 28 [25,26] chromosomes ('Dutch Master') and 28 [19] chromosomes ('Fortissimo') have been considered diploids based on karyotype analysis [19]. The origin of both cultivars is unfortunately unknown. The explanation why the plants with a high 2C value and a large number of chromosomes indicating a tetraploid level appears to be diploid can be found in the phenomenon of diploidization. This phenomenon involves many changes in genome organization that ultimately restore bivalent chromosome pairing and disomic inheritance [41].
Examples of the flow cytometry histograms for selected Narcissus genotypes with different nuclear DNA contents and thus with probably different ploidy levels are shown in Figure 2. Examples of the flow cytometry histograms for selected Narcissus genotypes with different nuclear DNA contents and thus with probably different ploidy levels are shown in Figure 2.

Chromosome Number
The microscopic images make it possible to estimate the number of chromosomes in diploid cultivar 'Dutch Master' at 28 ( Figure 3A), which confirms the literature reports on chromosomes number [25,26]. However, this is also different from that (5) as reported by Sun et al. [19]. For the triploid cultivar 'Yellow Cheerfulness', the number of chromosomes could be estimated at 24 (Figure 3B), which confirms the literature reports [15,17,25] and our own research by flow cytometry analysis carried out for this cultivar. For the reference tetraploid cultivar 'Ice Follies', the number of chromosomes could be estimated at 28 ( Figure 3C), which confirms the literature reports on chromosomes 2n = 4x = 28 [16,17,26] and studies conducted independently by flow cytometry. Our studies on chromosome number should be regarded as preliminary only. Further studies will be carried out on all cultivars and breeding clones evaluated. In addition, as seen in the example of the cultivar 'Dutch Master' previously considered a tetraploid, it would be very important to carry out karyotype analyses of all cultivars.

Chromosome Number
The microscopic images make it possible to estimate the number of chromosomes in diploid cultivar 'Dutch Master' at 28 ( Figure 3A), which confirms the literature reports on chromosomes number [25,26]. However, this is also different from that (5) as reported by Sun et al. [19]. For the triploid cultivar 'Yellow Cheerfulness', the number of chromosomes could be estimated at 24 ( Figure 3B), which confirms the literature reports [15,17,25] and our own research by flow cytometry analysis carried out for this cultivar. For the reference tetraploid cultivar 'Ice Follies', the number of chromosomes could be estimated at 28 ( Figure 3C), which confirms the literature reports on chromosomes 2n = 4x = 28 [16,17,26] and studies conducted independently by flow cytometry. Our studies on chromosome number should be regarded as preliminary only. Further studies will be carried out on all cultivars and breeding clones evaluated. In addition, as seen in the example of the cultivar 'Dutch Master' previously considered a tetraploid, it would be very important to carry out karyotype analyses of all cultivars.

Conclusions
Our results lead to the general conclusion that the morphological traits studied and nuclear DNA content can be helpful for determining the possible ploidy level of narcissus, as well the information on the origin and parental forms. The results obtained are a prelude to further studies, especially in the assessment of chromosome number and karyotype, in order to ascertain in cases of doubt the true level of ploidy. In detail, we propose the following conclusions: 1. Flower diameter, leaf length and width, and density and size of stomata are in many cases correlated or indicated a tendency towards the ploidy level of Narcissus genotypes evaluated and could be used as a part of the system of morphological markers (but absolutely after additional research). 2. Our results of nuclear DNA content confirm the literature reports, which are known for 7 of the 38 genotypes studied. The remaining genotypes without literature references, where relatively high values of 2C DNA were obtained that may tentatively suggest tetraploids, should be verified by chromosomes counting. 3. Clear confirmation of ploidy level requires verification of chromosome number and preferably karyotyping.

Conclusions
Our results lead to the general conclusion that the morphological traits studied and nuclear DNA content can be helpful for determining the possible ploidy level of narcissus, as well the information on the origin and parental forms. The results obtained are a prelude to further studies, especially in the assessment of chromosome number and karyotype, in order to ascertain in cases of doubt the true level of ploidy. In detail, we propose the following conclusions:

1.
Flower diameter, leaf length and width, and density and size of stomata are in many cases correlated or indicated a tendency towards the ploidy level of Narcissus genotypes evaluated and could be used as a part of the system of morphological markers (but absolutely after additional research).

2.
Our results of nuclear DNA content confirm the literature reports, which are known for 7 of the 38 genotypes studied. The remaining genotypes without literature references, where relatively high values of 2C DNA were obtained that may tentatively suggest tetraploids, should be verified by chromosomes counting.

3.
Clear confirmation of ploidy level requires verification of chromosome number and preferably karyotyping.  Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: Data sharing not applicable.