Karyotypic phylogeny and polyploidy variations of Paronychia (Caryophyllaceae) taxa in Turkey

Chromosomal data can provide very valuable information about karyotypic phylogeny and speciation. This is the first study on karyotype phylogeny and polyploidy variations of the genus Paronychia. In this context, the results are these: (1) in 14 taxa, the first report on chromosomes numbers; (2) in 2 taxa, equal chromosome numbers as in the previous report; (3) in all taxa, the first report of detailed karyotype analyses; (4) karyotype asymmetry data and generally symmetrical karyotypes; (5) karyotypic variations by mechanisms of dysploidy and polyploidy; and (6) phylogenetic relationships in Paronychia. The data indicate that Anatolia is an important area for the distribution of Paronychia. In light of all data, karyotype evolution is briefly summarized. The ancestral karyotype was x = 9 (millions of years ago). The karyotypes (x = 8 and x = 7) were then shaped by dysploidy. The rate of polyploidization then significantly increased in the genus. However, data should be supported by molecular analysis. In addition, the chromosome numbers of 8 species of Turkish Paronychia is still unknown. The determination of the karyological data of all species is very important to understand karyotype evolution and chromosomal phylogeny in Paronychia.

regions, is an important place for the study of interspecific phylogenetic relations .
In Paronychia, the lack of chromosomal data and the absence of karyotypic phylogeny studies had an impact on the purpose and importance of this study. In this work, the aim is to contribute to the chromosomal data and karyotypic phylogeny of Paronychia taxa using the following marks: (1) chromosome number and karyotype formula, (2) detailed chromosome measurements, (3) karyotype symmetry/asymmetry, (4) polyploidy variations, and (5) karyotypic phylogeny. Figure 1 shows a distribution map of Paroncyhia taxa. The collection information is listed below. All taxa except P. carica and P. chionaea subsp. chionaea are endemic to Turkey.

Karyotype analysis
At least 10 well-spread metaphase plates were used to determine chromosome numbers. The chromosomal measurements were obtained by KaryoType software (Altınordu et al., 2016). The following parameters were used to characterize the chromosomes: short-arm length (SA), long-arm length (LA), total chromosome length = (SA + LA), total haploid length (THL), mean chromosome length (MHL), relative length (RL) = [(LA + SA)/THL] × 100, and centromeric index (CI) = [(SA)/(LA + SA)] × 100. Karyotype formulae were determined as described by Levan et al. (1964). The ideograms were drawn based on chromosome arm length. The following 3 parameters calculated karyotype asymmetry, and these are M CA (intrachromosomal asymmetry), CV CL (interchromosomal asymmetry), and S/A I . The formulae are given below. M CA = [mean (L T -S T ) / (L T + S T )] × 100; L T : total length of long arms and S T : total length of short arms (Peruzzi and Eroğlu, 2013).

Karyological relationships and karyotypic phylogeny
CV CL = (S CL /X CL ) × 100; S CL : standard deviation and X CL : mean chromosome length (Paszko, 2006). A scatter diagram was then drawn between the intrachromosomal asymmetry (M CA ) and interchromosomal asymmetry (CV CL ).
The phylogenetic tree showing karyological relationships was drawn by bootstrap values with UPGMA, chord coefficient. The phylogenetic tree contains the comparative phylogeny of the listed species. (1) 16 taxa from the present study, (2) 4 species from previous reports, namely P. adalia (Chaudhri), P. argentea, P. echinulata

Chromosomal data
The chromosome records of 16 taxa are herein provided (Figure 2), 14 of which are reported for the first time. In addition, two of them (P. kurdica subsp. hausknechtii and P. kurdica subsp. montis-munzur) have similar numbers with previous data reported by Küpfer (1980). The chromosome numbers of Turkish Paronychia are given in Table 1, which also includes the results of present and previous studies. Six different chromosome numbers (2n = 18, 36, 52, 54, 72, and 104) are detected, 3 of which (2n = 52, 72, and 104) are new chromosome numbers for the genus. The smallest chromosome size among the taxa is 0.68 μm in P. kurdica subsp. hausknechtii. The largest chromosome size was detected in P. condensata, with 5.14 μm. The smallest total haploid length is 12.47 μm in P. kurdica subsp. hausknechtii, and the highest value is 87.48 μm in P. chionaea subsp. chionaea (Table 2).

Basic numbers, ploidy levels, and polyploidy
In genus Paronychia, there are generally 2 common basic numbers, which are x = 7 and most commonly x = 9. In this study, the basic chromosome numbers are x = 13 in P. chionaea subsp. kemaliya with ploidy levels of 4x/8x and x = 9 in the other taxa with ploidy levels of 2x, 4x, 4x/8x, and 6x (Table 1). The monoploid ideograms were generated by x = 9, and 13 are given in Figure 3.  Polyploid nature is demonstrated by the prevalence of cells with 2n = 4x = 36 (tetraploid) in 12 taxa, 2n = 8x = 72 (octoploid) in P. chionaea subsp. chionaea, and 2n = 4x/8x = 52/104 in P. chionaea subsp. kemaliya (Table 1). The chromosome arm lengths could not be measured in the octoploid P. chionaea subsp. kemaliya because the chromosomes are too small and the centromere is indeterminate.

Karyotype formulae and karyotype asymmetry
All taxa have median (m) and submedian (sm) chromosomes but not subtelocentric (st) and telocentric (t) chromosomes. Three different karyotype formulae are observed, which are M-m, m, and m-sm. In intrachromosomal asymmetry, the M CA value ranges from 2.78 (P. kurdica subsp. montis-munzur) to 19.82 (P. argyroloba), which refers to symmetric karyotypes. In interchromosomal asymmetry, the CV CL value ranges from 13.54 (P. cataonica) to 41.99 (P. kurdica subsp. hausknechtii), which refers to karyotype heterogeneity. In chromosomal type and centromeric position, the S/ A I value ranges from 1.000 to 1.222, which refers to full symmetric and symmetric karyotypes, respectively (Table  2). Figure 4 shows a phylogenetic tree including the chromosomal data of present and previous studies in Turkish Paronychia. Sixteen taxa have variable ploidy levels (4x, 6x, and 4x/8x) and shape the clade I. The tetraploid taxa are quite dominant in subclade 1. In addition, subclade 2 contains high polyploid ratios and high chromosome numbers. In clade I, P. saxatilis, P. davrazensis, P. carica, P. argyroloba, and P. aksoyii are karyologically very close to each other. In the same way, P. turcica, P. angorensis, and P. anatolica subsp. balansae are also very close to each other karyologically.

Phylogenetic analyses
Four taxa are diploid (2x) and shape the clade II. Subclade 3 contains the most symmetrical karyotypes in terms of intrachromosomal asymmetry. In contrast, P. echinulata has the most asymmetric karyotype and is located in subclade 4. Table 1 demonstrates the chromosome numbers of the taxa investigated in the present study and in previous studies. The chromosome numbers are the first to be reported for 14 taxa. The chromosome numbers of P. kurdica subsp. hausknechtii and P. kurdica subsp. montis-munzur are the same as in the previous report, which is 2n = 18 (Küpfer, 1980). However, this taxon is P. kurdica, and it may be different from our subspecies.

Basic number and karyotype formula
A basic chromosome number of x = 9 dominates in Turkish Paronychia taxa, but basic numbers of x = 5, 7, and 13 characterize several taxa. However, a basic chromosome number of x = 8 dominates in some regions such as Granada, Almeria in Spain, and Macaronesia, which is a phytogeographical region comprising the Azores, Madeira, the Canary Islands, and the Cape Verde Islands in the eastern North Atlantic. For example, the basic number is x = 8 in P. andina A. Gray, P. canariensis (L.) Link, P. depressa (Torr. & A. Gray) Nutt. ex A. Nelson, P. pulvinata A. Gray, P. sessiliflora Nutt., and P. suffruticosa (L.) Lam. (Hartman, 1972(Hartman, , 1974Fedorov, 1974;Diosdado and Pastor, 1994;Suda et al., 2003). All taxa have karyotypes containing M-m, m, and m-sm. In addition, the genus has the karyotypes with st chromosomes, which are 12m + 8sm + 2sm/st + 6st in P. argentea and m + 4sm + 6st in P. echinulata (Diosdado and Pastor, 1994).

Karyotype evolution; dysploidy and polyploidy
We believe that the ancestral basic number is probably x = 9 (Genome I), which is the dominant number. Then, the formations of the basic number such as x = 8 (Genome II) and x = 7 (Genome III) occurred with chromosomal changes such as fusion. Dysploidy is probably caused by fusion or reciprocal translocations of median chromosomes in ancestral karyotypes. Unlike the x = 9 karyotype, the karyotypes of x = 7 and 8 show lower diversity.
P. chionaea subsp. kemaliya. We are of the opinion that Anatolia has a role in the distribution to other regions of Paronychia species for 2 reasons. The majority of taxa have a basic number of x = 9 and have a high polyploidy ratio. Polyploidy originates from autopolyploidy by genome duplication in a species and allopolyploidy by genome duplication between species and has played a major role in the speciation and evolution of higher plants (Demirci Kayıran and Özhatay, 2017). Polyploidy may affect the speciation of subspecies. Metzgar et al. (2016) reported that glaciation and associated climate shifts increased polyploidy rates. Demirci Kayıran and Özhatay (2017) reported that altitudes and high latitudes might have increased the polyploidy rates, although not always. All taxa have a distribution between 950 and 3260 m. High altitudes may have affected polyploidy levels, but this is not the only reason. For example, the adjacent cells of P. chionaea subsp. kemaliya have different patterns in terms of chromosome sets, which is the mixoploidy.

Karyotype asymmetry
In intrachromosomal asymmetry, all karyotypes are symmetrical except P. echinulata. P. kurdica subsp. montismunzur, P. kurdica subsp. hausknechtii, and P. beauverdii, which have the most symmetrical karyotypes with 0 ˂ M CA ≤ 10.00, respectively. P. kurdica subsp. montismunzur and P. kurdica subsp. hausknechtii are diploid taxa (Table 2 and Figure 4). Table 3 presents a weak positive correlation between M CA values and ploidy levels (r = 0.233). In addition, the M CA value shows the correlations as weak or average in all parameters except with the karyotype formula and S/A I value, for which it correlates well. P. argentea and P. echinulata are rare species with subtelocentric chromosomes (Diosdado and Pastor, 1994), possibly due to the reciprocal translocations of the median/submedian chromosomes.
In interchromosomal asymmetry, all karyotypes are symmetrical except P. kurdica subsp. hausknechtii. P. polygonifolia, P. echinulata, P. argentea, and P. cataonica, which have the most symmetrical karyotypes with 0 ˂ CV CL ≤ 15.00, respectively. These taxa show diploidy in P. polygonifolia and P. echinulata, polyploidy in P. argentea, and high polyploidy in P. cataonica (Table 2 and Figure 4). Table 3 presents a negative average correlation between CV CL values and ploidy levels (r = -0.452). In addition, the CV CL value shows the correlations as weak or average in all parameters.
The most symmetric and asymmetric karyotypes are completely different between M CA and CV CL with a very weak correlation (r = 0.014) ( Figure 5). All taxa have symmetrical karyotypes with only median/ submedian chromosomes. As karyotype evolution progresses, chromosomal asymmetry continues to increase (Baltisberger and Hörandl, 2016). The fact that Anatolian Paronychia taxa have symmetrical karyotypes may indicate that these taxa are in the early stages of karyotype evolution. These data support our opinion that Anatolia plays an important role in the distribution of the Paronychia species. The asymmetric karyotypes are probably higher in other regions where the genus Paronychia spreads.

Conclusion
In this study, the chromosomal data, polyploidy variations, and karyotypic phylogeny of 16 Turkish Paronychia are shown. The data make up the first report for 14 taxa. The listed data provide important contributions to karyotype phylogeny and cytotaxonomy of Paronychia: (1) the majority of taxa have a basic number of x = 9, (2) high polyploidy rates, and (3) symmetrical karyotypes. The data support the fact that Anatolia is an important distribution center of Paroncyhia. However, it should be supported by molecular analysis. In addition, the chromosome numbers of 8 species from Turkey are still unknown. Determining the karyological data of all species is very important to understand karyotype evolution and chromosomal phylogeny of Paronychia.