Development of a karyotype associated with the genome map and its applications in identification and characterization of chromosomes in peanut


 Background

Chromosomal structural variants are important materials for crop breeding and genetic research. Oligo fluorescence in situ hybridization (oligo FISH) is a useful tool for the identification of chromosomal structural variants.
Results

We developed 114 new repetitive oligos based on genome-wide tandem repeats (TRs) using genomic reference sequences of the cultivar Tifrunner and the diploid species Arachis ipaensis by bioinformatics and FISH. These oligo probes were positioned and classified into 28 types. Signals produced by representative probes from eight types were in the secondary constriction, middle arm, and terminal regions; signals of four probe types were on the B subgenome; and one probe being able to produce signals on only one pair of chromosomes could be used to recognize a special genome or chromosomes of peanut. Based on new and previous oligo probes, we developed a cocktail Multiplex #3 including FAM modified TIF-439, TIF-185-1, TIF-134-3, TIF-165-3, and TAMRA modified Ipa-1162, Ipa-1137, DP-1, and DP-5, which combined with the total genomic DNA of A. duranensis and A. ipaensis probes, and 45S rDNA and 5S rDNA probes to establish a karyotype associated with the genome map of peanut and identify 14 chromosomal structural variants.
Conclusions

The new oligo probes are useful and convenient for distinguishing peanut chromosomes or specific segments of peanut chromosomes. Comparisons of oligo sites in the karyotype and chromosome plots of electronic location revealed the characteristics of repeated sequences, and showed that the assembly of repeated sequence in peanut genomic reference sequence was incomplete. We therefor demonstrated the great potentials of the new oligo probes in facilitating chromosome identification and characterization, and provided novel materials for further study and genetic improvement of peanut.

Cultivated peanut (Arachis hypogaea L., 2n = 4x = 40) is an allotetraploid, and a major oilseed and cash crop worldwide. Many mutants have been created and investigated in peanut [3,[13][14][15][16][17]. Candidate genes for pod width [16], seed coat color [18], and other traits have been identi ed by mutant transcriptome sequencing. However, chromosomal structural variants have not been a major concern of researchers owing to the lack of detection methods.
Cytological methods, especially chromosomal banding techniques based on repeated sequences, are frequently used to detect chromosomal structural variants. Both C-banding [19] and uorescent in situ hybridization banding [20] can effectively identify chromosomal structural variants. Although 5S and 45S rDNA [21], telomeres [22], and bacterial arti cial chromosome (BAC) probes [23] have been developed in peanut, the cytogenetic markers in chromosomal identi cation is still very insu cient. Furthermore, the preparation of these markers is time-consuming and expensive.
Repetitive single-stranded oligo probe uorescence in situ hybridization (SSON FISH), based on genomic sequences provides a new solution for cost-effective chromosome identi cation [24][25][26][27]. Development of genome-wide repetitive oligo probes to establish high-resolution chromosome-painting karyotypes can accurately identify both naturally occurred and arti cially-induced chromosomal structural variants [20,[28][29][30]. The physical distribution of repeated sequences in the plant genome could reveal genome evolution and improve the quality of the genomic sequence assembly [28,31] In peanut, Du et al. [32] developed oligo probes and established karyotypes to identify a translocation region of the chromosome. They discovered that a probe dye could be made to stain chromosomes [33].
However, the signals of these karyotypes were mainly distributed in the centromeric regions with very few and even no bands were discovered on the arms and telomeric regions of the chromosomes. In addition, these cytological markers were developed based on the simpli ed sequences of the A. duranensis genome.
Genome sequencing of cultivated peanut [34][35][36] and its assumed ancestral wild species [37] has been completed in recent years, and has provided better reference sequences to design oligo probes. In the present study, the entire genomes of the cultivated peanut variety Tifrunner and the B genome donor parent, A. ipaensis, were used to develop more oligo probes that could enrich chromosome regions. In addition, we established a karyotype corresponding to chromosomes in the sequence map of peanut, to facilitate the identi cation of chromosomal structural variants.

Results
Genome-wide discovery of new repetitive oligo probes Genome sequences of Tifrunner and A. ipaensis were analyzed using the TRF software, and 4,595 and 894 repeated sequence patterns were obtained respectively from the two genomes. The length of the repeated sequences was between 4 bp and 723 bp, and the copy number was between 50 and 29162.
After CD-HIT elimination, a total of 80 and 35 tandem repeat (TR) sequences of Tifrunner and A. ipaensis, respectively, were selected to design oligos.
A total of 249 oligos were designed from 115 TRs using the Oligo 7 software, and electronically positioned by B2DSC. Oligos with speci c or good polymorphism were preserved for FISH. For example, the oligo, Ipa-1463, showed up to 1,260 repeats, a relatively highly enriched, only within the 63-77 Mbp region of one chromosome (B9) (Fig. 1a). Oligo FISH result veri ed the presence of only one pair of chromosomes with signals (Fig. 1b).
Using the same method for the FISH analysis of 249 oligos, we found that 114 oligos from 57 TRs were able to produce signals in Tifrunner (Fig. 2, Table S1). We also found that the distribution of signals among some probes were similar, while those among other probes showed signi cant differences. Based on the similarity of their signals, dual-color and sequential FISH were used to classify the 114 probes.
Probes with similar sites were classi ed into the same group. As a result, all probes under investigation were divided into 28 types (Table S2).

Development of karyotypes corresponding to the genome map
Based on oligos discovered in this study and two oligos (DP-1 from 5S rDNA and DP-5 from telomere repeats) reported by Du et al. [32], we developed a probe cocktail Multiplex #3 including FAM modi ed TIF-439, TIF-185-1, TIF-134-3, TIF-165-3, and TAMRA modi ed Ipa-1162, Ipa-1137, DP-1, and DP-5 (Table  1); combined with the total genomic DNA of A. duranensis and A. ipaensis probes, and 45S rDNA and 5S rDNA probes. A karyotype and idiogram with a greater number of signal bands was developed by sequential FISH/GISH (Fig. 3a-e). To match chromosomes in the genome sequence map of Tifrunner with chromosomes in the karyotype, electronic location of eight oligos in Multiplex #3 was performed on the B2DSC website (Fig. S1). Chromosome plots of the eight visible oligo sites in the sequence map were established (Fig. 3f). Based on the similar distributions of eight oligos on chromosomes in the karyotype and chromosome plots, we made matches between chromosomes in karyotype with those in the sequences map. To further con rm this result, sequential FISH/GISH was also conducted using two chromosome-speci c libraries from the end of chromosome A1 and A3 in A. duranensis, Multiplex #3, and the total genomic DNA of A. duranensis and A. ipaensis probes (Fig. S2). The results showed that chromosome A1 and A3 in the sequence map did correspond to two chromosomes of the karyotypes.
Based on the idiogram of the karyotype and chromosome plots, we observed similar distribution patterns of the signals. Green signals covered the centromeric regions of chromosome A1 and the smallest chromosome A8 while strong red interstitial telomeric signals were observed on the short arm of chromosome A1. Chromosomes A6, A7, and A10 had 45S or 5S rDNA sites. Noticeable green signals were observed in the telemetric regions on the upper arms of chromosomes A2, A3, and A4; however, chromosomes A3 and A4 also showed green signals in the centromeric regions. Weak green and red signals were observed on the centromere and arm of chromosome A5, and on the centromere and ends of chromosome A9. In the B subgenome, three chromosomes (B6, B7, and B8) were detected with 45S or 5S rDNA signals. Chromosome B9 displayed distinct red centromeric interstitial telomeric repeat signals, while chromosome B3 showed distinct green centromeric repeat signals. The other chromosomes showed green signals, with notable differences in the sites and strength of those signals. These ndings show that the karyotype showed signi cantly enriched signals, which could be used to identify chromosome variants.

Localization of repetitive oligo probes
Based on the karyotype, we physically mapped 28 representative oligo probes with one from each of the 28 types of oligos in Tifrunner (Fig. 4). We observed that the Ipa-1463 probe had the fewest signals, with signals on only two chromosomes. While, TIF-439 was the most widely distributed probe, as it produced signals on almost all chromosomes of the A and B subgenomes.
Most probes produced signals on both the A and B subgenomes. However, ve probes, Ipa-1463, TIF-198-2, TIF-416-3, TIF-497, and TIF-342-2 were observed to produce signals only on the B subgenome alone. Most signals of the above-mentioned probes were detected in the centromeric regions, but signals of the following six probes TIF-165-3, TIF-439, TIF-556, TIF-198-1, TIF-384-3, and TIF-185-1 were observed on the middle arm or terminal region of the chromosomes. Furthermore, signals of the two probes, Ipa-1137 and Ipa-1162, were same to the signals of 45S rDNA sites, which were distributed along the secondary constriction region (Fig. S3). With all probes, signals were detected on 69 pairs of chromosomes in the A subgenome, and 145 pairs of chromosomes in the B subgenome respectively.
Signi cantly different patterns of distribution produced among these probes were also found between karyotype and chromosome plots. Electronic location of oligo TIF-89-3 and TIF-155-5 showed that only two pairs of chromosomes had clear repeated sequences sites, but eight pairs of chromosomes showed signals via FISH using probes of them. On the contrary, a lot of chromosomes showed obvious repeated sites by electronic mapping of oligos TIF-359-3 and TIF-76-1, while only a few chromosomes showed signals via FISH using corresponding uorescent probes (Fig. 4, Fig. S4).

Identi cation of variants using repetitive oligo probes
To validate the potential of the karyotypes under investigation and to detect chromosomal structural variants, sequential FISH/GISH was used to characterize SLH and its irradiation-induced M 1 plants.
Based on the karyotype of the M 1 plant 161-1a generated with FISH/GISH, we easily found that it lost two pairs of chromosomes, A1 and B3; however, it had two translocations, which we speculate to be T1AL-3BS·3BL and T3BS-1AL·1AS, based on their characteristics (Fig. 5a-5e). Further FISH was performed using single-copy oligo library probes, which con rmed that the two translocations did occur between chromosomes A1 and B3, and were identi ed as T1AL-3BS·3BL and T3BS-1AL·1AS ( Fig. 5f-5h).
With the examination of other M 1 plants, we found 13 chromosomal variants including 17 translocations, one deletion, and eight monosomic chromosomes (Fig. S5). Ten translocations occurred between homologous chromosomes, and seven translocations occurred between non-homologous chromosomes. Chromosomes 5, 1, and 3 showed a relatively greater number of translocations (Fig. 6). It is quite obvious that the karyotyping methods with the aid of oligo probes can be effetely used to identify chromosomal translocations in the M 1 plants.

Applications of the new oligo probes
The TRs comprise a large proportion of the plant genome [40]. In the past, DNA repeats were regarded as genomic 'junk' [41]. However, various studies in recent years found that TRs could be used to gain a better understand of the evolutionary history and genomic composition of TRs [42]. They also serve as placeholders for epigenetic signals that govern heterochromatin formation, and may have function in the repair of double-strand DNA breaks [42,43]. Many repetitive DNA elements generate speci c patterns on chromosomes of various species [44][45][46]. These elements are the most common sources of probes for chromosome identi cation and cytogenetic studies in plants [26]. Du et al. [32] developed eight oligo probes of peanut based on simpli ed genome sequencing of A. duranensis, and established a highresolution chromosome painting karyotype and revealed the relationships among eight Arachis species. However, these oligo probes were neither genome-speci c nor chromosome-speci c, and it lacked probes which distributed on the arms.
In the present study, TRF software [38] was used to search TRs from reference genome sequences of the cultivated peanut variety Tifrunner and the assumed B genome donor species of the cultivated peanut A. ipaensis. Therefore, we maximized the discovery of TRs of peanut. Furthermore, 114 new oligos were developed based on genome-wide TRs. Among them, signals of eight probes, TIF-165-3, TIF-439, TIF-556, TIF-198-1, TIF-384-3, TIF-185-1, Ipa-1137, and Ipa-1162 were distributed in the secondary constriction, middle arm, or terminal region of chromosomes, which could be used to improve the karyotype of peanut. Five probes, TIF-198-2, TIF-416-3, TIF-497, TIF-342-2, and Ipa-1463, produced signals on B subgenome alone, while one probe (Ipa-1463) just produced signals on one pair of B9 chromosomes, indicated that these probes were genome-speci c or chromosome-speci c probes.
As compared with functional genes, repetitive DNA sequences are likely to have evolved under different evolutionary pressures. Thus, comparing the distributions of repetitive sequences based on FISH karyotyping can facilitate various phylogenetic schemes [31]. Telomeric and centromeric repeat probes were used to investigate the evolutionary relationship among cultivated peanut and wild species. Multiple hybridization events reportedly occur during tetraploid peanut formation [23]. Four repetitive oligo probes capable of staining the chromosomes of 51 Arachis species con rmed the relationships among the A, B, K, F, E, and H genomes [32]. In the present study, A total of 114 oligo probes derived from TRs were positioned on chromosomes of Tifrunner and A. ipaensis. However, the distribution patterns of these repeats in other Arachis species are not yet known. If hybridization can be achieved with chromosomes of other Arachis species, further evidence of genomic evolution in the genus Arachis will be provided.
Karyotypes associated with the genome map revealed the characteristics in the distribution of repeated sequences in peanut Genome sequences of peanut cultivars have been published, and have shown that the A and B genomes are partially homologous [34][35][36]. Repetitive sequences in Tifrunner account for 74% of the assembled genome sequence [35]. However, the sites, distribution, and density of repeated sequences in the genomes remain cytologically invisible. In the present study, we developed oligo probes based on TRs and established associations between the karyotype associated with genome sequence map of Tifrunner was established by oligo FISH and electronic location, which enabled a direct observation of repeated sequences on the chromosomes. Most oligo probes showed similar distributions in the karyotype as in chromosome plots, however, some oligo probes showed signi cant difference in signal distributions and lower copy numbers in the sequence map than the karyotype, indicating that the assembly of repeated sequence in the peanut genome is not complete, and the TR contents may be much higher than those of reported.
Furthermore, repeated sequences in the B subgenome occurred signi cantly more frequently than in the A subgenome. These ndings indicate that the B subgenome acquired more repeated sequences during the evolution of the two genomes. All chromosomes had a high number of repeated sequences in the centromeric or pericentromeric region. However, some showed a high number of repeated sequences in the telomeric region, indicating extensive heterochromatization in this region of the chromosomes.

Identi cation of chromosome variants for peanut research and breeding
Chromosomal structural variants have special signi cance in the research and improvement of heteroploid plants [7]. Oligo FISH provided a cost-effective technique to detect chromosomal structural variants [20]. In peanut, cytogenetic markers have previously included only 45S and 5S rDNA probes [47][48][49], which made it di cult to identify chromosomal structural variants. Although BAC and eight oligo probes have been developed in recent years, their utilization has been hindered due to lack of markers on the arms of the chromosomes.
In the present study, we established a new karyotype with more chromosome bands and speci c characteristics, which corresponded to the genome map. Based on this karyotype, we identi ed 14 chromosomal structural variant materials including translocations, deletions, and monosomic chromosomes. Translocation and deletion lines could be used in genetic research, for translocation or deletion mapping, radiation hybrid mapping, and gene mapping, among other procedures [50,51].
Furthermore, the translocations occurring between partially homologous chromosomes or heterologous chromosomes promote the exchange of DNA beyond homologous chromosomes in conventional breeding, which could lead to new bene cial traits. In future studies, we intend to produce inbred translocation lines to create homozygous translocations, which will no doubt facilitate the genetic improvement of peanut.

Conclusion
Based on the genomic reference sequences of the cultivated peanut Tifrunner and the B subgenome donor ancestor, A. ipaensis, we developed and positioned 114 new repetitive oligo probes, which were further classi ed into 28 types according to the similarity among their signals. Combined with distributions of eight oligos on chromosomes via oligo FISH and electronic location, a karyotype associated with the genome map of peanut was constructed with much enriched chromosome bands. With the aid of this karyotype, we con rmed that the assembly of repeated sequence in the published peanut reference genome was incomplete, revealed the characteristics in the distribution of repeated sequences in peanut, and identi ed 14 chromosomal structural variants. We therefor demonstrated the great potentials of the new oligo probes in facilitating chromosome identi cation and characterization, and provided novel materials for further study and genetic improvement of peanut.

Plant materials
Peanut varieties Tifrunner and Silihong (SLH) were maintained by the Henan Academy of Crop Molecular Breeding, Henan Academy of Agricultural Sciences. SLH plants at the owering stage were irradiated with a dosage of 16 Gy using a 60 Co source (Isotope Institute Co. Ltd., Henan Academy of Sciences) to create chromosomal structural variants. Variants of M 1 were selected for the present study.

Chromosome preparation
Chromosomes were prepared, as described by Du et al. [33]with minor adjustments. When roots of the peanut plant reached about 2 cm long, root tips were cut and placed in 8-hydroxyquinoline for 3 h at 25°C . It was then treated with nitrous oxide (N 2 O) at a pressure of 0.8-1.2 MPa for 1.5 h, and xed in absolute ethanol-glacial acetic acid (3:1) solution for 12 h at 25 °C. The treated root tips were then stored in a refrigerator at -20 °C until further use. The apex was resected and disintegrated in 45% glacial acetic acid for 2-3 min, compressed, and frozen at -80 °C in an ultra-low temperature refrigerator for 24 h, and then left uncovered. The root material was dehydrated in ethanol absolute for 12 h and air-dried for FISH.

Design of oligo probes
The whole genome assembly sequences of Tifrunner and A. ipaensis were downloaded from Peanutbase (https://peanutbase.org/home). The sequences of both Tifrunner and A. ipaensis genomes were analyzed to obtain tandem repeat sequences using the Tandem Repeats Finder (TRF, version 4.09) [38] based on the methods of Tang et al. [27]. Brie y, the TRF algorithm used the following alignment parameters: Match = 2; Mismatch = 7; Indel = 7; Probability of match = 80; Probability of indel = 10; Min score = 50; and Max period = 2000. Data with a period size > 4 and DNA copy number > 50 were obtained using a Python script. Tandem repeat sequences with > 75% identity were determined using the Cluster Database at High Identity with Tolerance (CD-HIT) [39] program for clustering and maintaining consensus for each cluster. Oligos 40-45 nt in length were then designed based on the tandem repeat sequences and the Oligo 7 software.
To determine the effectiveness of each designed probe, oligos were rst physically mapped onto reference sequences of chromosomes. The sequence of every obtained oligo probe was aligned using the Nucleotide Basic Local Alignment Search Tool (blastn) in the B2DSC2 server against the Tifrunner genome. The blast results were ltered using pident = 90% and qcovhsp = 90%. The physical location information was drawn using the Plot tool in B2DSC. Oligos with speci c sites or good polymorphisms were synthesized by the General Biosystems Company (Anhui, China).
Two single-copy oligo libraries (L1A-1and L3A-1) from chromosome A1 and A3 were developed using genomic sequences of A. duranensis in PeanutBase according to the method described by Du et al. [32].
Each single-copy oligo library was derived from the distal region of each chromosome. Libraries were synthesized by MYcroarray (Ann Arbor, MI, USA). The resulting library was ampli ed and labeled with biotin-16-dUTP or digoxigenin-11-dUTP, according to the MYcroarry_MYtags labeling protocol.
The FISH and sequential FISH procedures were as described by Du et al. [32]. Brie y, the slides were denatured in 70% formamide at 75 °C for 70 s, and the hybridization solution, including 3 µL of each probe, was denatured for 13 min. The slides were immersed in the hybridization solution at 37 °C in a wet box for at least 12 h, and then washed 10 times with 2 × saline-sodium citrate at 42 °C. Slides were then stained with 4', 6-diamidino-2-phenylindole (DAPI) and mounted with VECTASHIELD Mounting Medium.
Sequential FISH was performed to map the signals of oligo probes and correlate a sequenced chromosome with a cytologically identi ed chromosome. Thereafter, FISH was conducted using repetitive or single-copy oligo probes, which were then washed to remove all signals, and subsequently dried.
Genomic in situ hybridization (GISH) was then conducted using the total genomic DNA of A. duranensis and A. ipaensis probes, or FISH was conducted using 45S rDNA and 5S rDNA.