Identification and characterization of satellite DNAs in a diploid of P. malaca
Primary screening of the chromosome number showed that the varied polyploids were existed in the tested Poa species. Specially, the diploidy form was identified in the population of P. malaca. Taking think that the diploid form contains less genome content than the tetraploid or other high polyploidy, we selected the diploid form of P. malaca for the candidate for high-throughput sequencing. The Illumina HiSeq data (genome coverage c. 30.3%) from diploid form of P. malaca was applied to the RepeatExplorer pipeline clustering tool. Four putative satellite DNAs were identified with the monomer 365bp, 189bp, 326bp, and 353bp from the output graphs. The satellite 365bp has the most abundance about 0.85% of the genome, while the 189bp, 326bp, and 353bp satellite has about 0.52%, 0.40%, and 0.24% respectively. Additionally, to test the similarity between the identified satellites with other repetitive sequences, nucleotide BLAS was conducted in the gene bank of NCBI. (https://www.ncbi.nlm.nih.gov/). The results showed that the 365bp satellite has a high similarity with the tandem repeat PpTr-1 identified in P. pratensis (KY618838.1) with an identity of 98.8% and 100% coverage, and the 189bp satellite has a high similarity with the tandem repeat PpTr-2 and PpTr-3 in P. pratensis (KY618841.1 and KY618840.1) with the identity of 98.9% and 100% coverage both. However, no significant similarity was found for both 326bp and 353bp satellite. Thus, we thought that the 326bp and 353bp satellites are two novel satellite DNAs identified in Poa, and we named them as Poa-362 and Poa-353.
Examination of chromosomal distribution of the four above satellites carried out in the diploid form of P. malaca with the chromosome number of 14. Since the satellite 365bp and 189bp showed the high similarity with the previous identified tandem repeatsPpTr-1, PpTr-2 and PpTr-3 in P. pratensis, the clone sequence probes PpTr-1 and PpTr-3  were used for representing the satellite 365bp and 189 respectively in this study. Oligonucleotide FISH probes were designed for Poa-362 and Poa-353 from the sequences of the monomers using DNAman software package (Lynnon Biosoft, Quebec, Canada) (Table 1). The FISH patterns showed that PpTr-1, PpTr-3, and Poa-362 produced high intensity signals of 12, 3, and 2 respectively, which are all located in the terminal positions of the chromosomes. Relatively, Poa-353 produced less and weak hybridization signals of 2~3 in the subtelomeric and intercalary regions (Fig.1 a1-d1). The opulence of the four satellites detected by FISH is appropriate to the abundance of those derived from the high-throughput data analysis in some extent.
Characterization of satellite DNA across different Poa species
In total, 12 different Poa species are involved in this study. Additionally, P. pratensis contains one subspecies P. pratensis Var. anceps, and four P. pratensis cultivars ‘Qinghai’, ‘Park’, ‘Geronimo’, and ‘Sapphire’. The investigated species appeared different ploidy levels with one diploidy form, 10 tetraploid species with 28 chromosomes, and 2 high polyploidy species (P. subfastigiana and P. pratensis) with a variable chromosome number (Table 2). All the probes PpTr-1, PpTr-3, and Poa-362 produced hybridization signals in the subtelomeric regions across all the investigated positive species (Fig.1 a, b and c). However, hybridizations of Poa-353 were detected in subtelomeric, intercalary, and pericentric regions in the investigated samples (Fig. 1 c). It suggests that the genomic distribution pattern of Poa-353 is totally different from those of the other three satellites in genus Poa.
The number of hybridization sites between different satellites is varied. About 14, 4, 8, and 12 hybridizations were detected by PpTr-1, PpTr-3, Poa-362, and Poa-353 respectively in an average in the total investigated samples, although varied hybridization numbers were observed in the different species. The PpTr-1 presents the intense hybridization in each species, and high sites number varied from 6~26 (Fig. 1 a, Table 2). In the tetraploid Poa species, the PpTr-1 produced about 15 hybridization sites averagely, with more than 20 hybridization sites in P. malaca, P. sinoglauca and P. orinosa. In the high polyploidy species, the PpTr-1 probed about 12 hybridization sites averagely (Table 2). The PpTr-3 produced intense hybridizations in all species except in P. orinosa and P. paucifolia. The PpTr-3 probed about 1.5 hybridization sites in tetraploid species, whereas about 7 in high polyploidy species averagely. The hybridization sites of Poa-362 were detected in all tested species except in P. pratensis cultivar ‘Geronimo’ and ‘Sapphire’ (Table 2). The average of 12 hybridization sites of Poa-362 was probed in tetraploid species with an exceptional number of 24 in P. paucifolia, while about 1 site was detected in the high polyploidy species averagely. The hybridization sites of Poa-353 were detected in all tested species except in P. elanata (Table 2). The average of 12 hybridization sites of Poa-353 of 8 was probed in tetraploid species with an exceptional number of 20 in P. paucifolia, while about 22 sites were detected in the high polyploidy species with an exceptional small number of 5 in P. subfastigiana averagely. Roughly. The hybridization sites number of PpTr-1 and Poa-362 is decreasing by the increasing of the polyploidy, but that of PpTr-3 and Poa-353 is increasing by the increasing of the polyploidy. Specially, P. pratensis nearly contains the least number or none of Poa-362, and most number of Poa-353.
rDNA distribution patterns
rDNA (45S rDNA and 5S rDNA) are expressed, and highly tandem-repeated in plant genomes. Commonly, the distribution pattern and abundance of rDNA are highly stable in a species. The variable rDNA distribution patterns across different species can provide valuable information to infer the phylogenetic relationship between species. The phylogeny of the Poa species used in this study is not clear. It was thought that description of the rDNA patterns across species might be helpful to explain the phylogenetic relationship of the investigated species as well as the satellite DNAs.
Three different FISH patterns of rDNA were uncovered in tetraploid species, while one representative pattern was in high polyploidy species. In tetraploid, most of the species showed four chromosomes containing both 5S rDNA and 45S rDNA each in discrete sites as well as other four chromosomes including only one 45S rDNA site respectively (Fig 1. d2 and d3; Table 2). Two tetroploid species P. paucifolia and P. crymophila showed four chromosomes containing both 5S rDNA and 45S rDNA each in discrete sites as well as other two chromosomes including only one 45S rDNA site respectively (Table 2). One tetroploid species P. elanata showed four chromosomes containing both 5S rDNA and 45S rDNA each in discrete sites, two chromosomes containing other more solely 5S rDNA site, and other two chromosomes including only one 45S rDNA site respectively (Fig. d4). Identical FISH pattern of rDNA was difficult to obtain due to the variable chromosome number in each high polyploidy species. However, a common feature of FISH pattern of rDNA is distinguished from the most of the tetrploid species, which showed that 2~3 chromosomes shared with 5S rDNA and 45S rDNA sites, 1~4 chromosomes had only 5S rDNA site, and 2~6 chromosome had solely 4SrDNA site (Fig.1 d5 and d6, Table 2).