*2.5. Comparison of CficCl-61-40 and Proposed HOR Unit CacuCl-1-117 Chromosomal Distribution*

For further confirmation of the existence of HOR units' in the genomes of *Chenopodium* species an alternative method of FISH was used. Two distant clusters according to phylogenetic analysis, cluster 61 of *C. ficifolium* (CficCl-61-40) and cluster 1 of *C. acuminatum* (CacuCl-1-117) from the RE output, were selected as sources for in situ probes for comparative molecular cytogenetic analysis (Figure 5, supplementary data 1 and 5). FISH experiments were performed to verify if (i) *C. acuminatum*-specific tandem repeats that were proposed to be HOR units (CacuCl-1-117) are species-specific and do not hybridize to chromosomes of the other six species and (ii) if the chromosomal positions of the *C. acuminatum*-specific tandem repeat (CacuCl-1-117) are similar to or different from the positions of the tribe-specific repeat (CficCl-61-40) on the chromosomes of

**3. Discussion** 

lineages.

CficCl-61-40 satDNA family structure (Figure 2).

*C. acuminatum*. It should be noted, however, that accurate FISH-based karyotyping and chromosome mapping of CficCl-61-40 satDNA family tandem repeats is challenging in *Chenopodium* due to the small chromosome sizes and to the large number of clusters (Table 2, Figure 5). *Int. J. Mol. Sci.* **2019**, *20*, x 9 of 18

**Figure 4.** Agarose gel electrophoresis of PCR products obtained with primers designed from consensus monomer sequence of proposed high order repeat (HOR) units for determination of their physical counterparts. Cloned DNA fragments are shown by asterisks. The far-right line is an example of negative amplification of a computer-generated proposed HOR unit. **Figure 4.** Agarose gel electrophoresis of PCR products obtained with primers designed from consensus monomer sequence of proposed high order repeat (HOR) units for determination of their physical counterparts. Cloned DNA fragments are shown by asterisks. The far-right line is an example of negative amplification of a computer-generated proposed HOR unit.

*2.5. Comparison of CficCl-61-40 and Proposed HOR Unit CacuCl-1-117 Chromosomal Distribution*  For further confirmation of the existence of HOR units' in the genomes of *Chenopodium* species an alternative method of FISH was used. Two distant clusters according to phylogenetic analysis, cluster 61 of *C. ficifolium* (CficCl-61-40) and cluster 1 of *C. acuminatum* (CacuCl-1-117) from the RE output, were selected as sources for *in situ* probes for comparative molecular cytogenetic analysis (Figure 5, supplementary data 1 and 5). FISH experiments were performed to verify if (i) *C. acuminatum*-specific tandem repeats that were proposed to be HOR units (CacuCl-1-117) are species-specific and do not hybridize to chromosomes of the other six species and (ii) if the chromosomal positions of the *C. acuminatum*-specific tandem repeat (CacuCl-1-117) are similar to or different from the positions of the tribe-specific repeat (CficCl-61-40) on the chromosomes of *C.*  FISH experiments confirmed species specificity and sometimes separate chromosomal positions of newly formed HOR units. Probe CficCl-61-40 hybridized to the chromosomes of all analyzed species except *C. iljinii* (similarity, copy number, or both of the particular FISH probe in *C. iljinii* genome is likely much less in comparison with other species), which demonstrates the presence of a tribe-specific satellite, while CacuCl-1-117 hybridized only to chromosomes of *C. acuminatum* with no signal on the chromosomes of the other six species (Figure 5, supplementary data 5, the minor green signal in *C. pamiricum*, *C. suecicum* and *C. vulvaria* in supplementary data 5 is epifluorescence). In addition, the simultaneous hybridization of CacuCl-1-117 and CficCl-61-40 on the chromosomes of *C. acuminatum* shows that in many cases these tandem arrays form separate clusters that create a species-specific chromosomal pattern (Figure 5). *Int. J. Mol. Sci.* **2019**, *20*, x 10 of 18

**Figure 5.** Chromosomal distribution CficCl-61-40 satDNA family sequences. CficCl-61-40 is labelled red; *C. acuminatum*-specific HOR unit CacuCl-1-117 is labelled green. Bar represent 5 μm. **Figure 5.** Chromosomal distribution CficCl-61-40 satDNA family sequences. CficCl-61-40 is labelled red; *C. acuminatum*-specific HOR unit CacuCl-1-117 is labelled green. Bar represent 5 µm.

investigated CficCl-61-40 satDNA family is the most abundant and oldest component of the *Chenopodium* genome, given that related sequences were found in both *Chenopodium* and *Beta* species. Regarding these two genera, it is essential to note that the genome of *Beta* should be recognized as more static, at least because it contains many fewer species (approximately 7–8 species in total, [33]) in comparison with *Chenopodium* (approximately 150 species [24]). Alignment of the satellite monomers allowed identification of the ancestral DNA fragment of 37 bp that showed 100% identity between *B. corolliflora* from one side and *C. bryoniifolium* and *C. vulvaria* from the other (supplementary data 2). The latter two are species that split off early and possess a modified sequence that is still recognizable by BLAST as a ~40 bp variant of the ancestral monomer. The identified DNA fragment served as a benchmark for our subsequent analyses, in which we intended to characterize intra-unit evolutionary transformations in the diverse *Chenopodium*

Remarkably, the evolutionary history of the *C. album* aggregate revealed by cpDNA spacers and two low-copy genes [27] correlates fairly well with significant paleoclimatic events. Thus, the early differentiation coincides with the beginning of the Miocene Climatic Optimum in the Burdigalian Age (approximately 20 Mya) (Figure 1). Clade H (*C. vulvaria*) separated upon transition between the Serravallian and Tortonian Ages, ~11 Mya. However, the main lineages were formed in the Pliocene, when due to a cooler and dry, seasonal climate, grasslands spread on all continents, and savannahs and deserts appeared in Asia and Africa. Subsequent speciation within the lineages and the appearance of the majority of polyploids occurred in the Quaternary Period, when the glacial and interglacial epochs succeeded each other. During this time, since there were no places on Earth with identical climate history and since the species of aggregate were spread widely, the CficCl-61-40 satDNA arrays evolved divergently. Excluding clade H, which split off early and is now very different, k-mer-based distance estimation of basic monomer show the most significant differences in genomes of species from clades A and D. It is most likely that both lineages separated early from the ancestral group and evolved independently. This is consistent with the present species distribution ranges and with molecular phylogenetic data [26,27]. However, the pace of evolution of these clades was probably different and is most likely connected with the climatic history of the species distribution areas. In clades B and E, the species are much more similar in the

The concept of "molecular drive" [19] postulates that mutations can gradually spread throughout a satDNA family by several of ubiquitous mechanisms of DNA turnover (homogenization) and become fixed in a population. SatDNA families can show a rapid rate of
