Characterization of Chromosomal Rearrangement in New Wheat — Thinopyrum intermedium Addition Lines Carrying Thinopyrum — Specific Grain Hardness Genes

The wild species, Thinopyrum intermedium. (Genome StStJSJSJJ), serves as a valuable germplasm resource providing novel genes for wheat improvement. In the current study, non-denaturing fluorescence in situ hybridization (ND-FISH) with multiple probes and comparative molecular markers were applied to characterize two wheat-Th. intermedium chromosome additions. Sequential ND-FISH with new labeled Th. intermedium specific oligo-probes were used to precisely determine the chromosomal constitution of Th. intermedium, wheat—Th. intermedium partial amphiploids and addition lines Hy36 and Hy37. The ND-FISH results showed that the added JS-St translocated chromosomes in Hy36 had minor Oligo-5S ribosomal DNA (rDNA) signals at the short arm, while a pair of J-St chromosomes in Hy37 had major Oligo-pTa71 and minor Oligo-5S rDNA signals. The 90K SNP array and PCR-based molecular markers that mapped on wheat linkage group 5 and 3 facilitated the identification of Thinopyrum chromosome introgressions in the addition lines, and confirmed that added chromosomes in Hy36 and Hy37 were 5JSS.3StS and 5JS.3StS, respectively. Complete coding sequences at the paralogous puroindoline-a (Pina) loci from Th. intermedium were cloned and localized on the short arm of chromosome 5JS of Hy36. Line Hy36 showed a reduction in the hardness index, which suggested that Th. intermedium-specific Pina gene sequences may be associated with the softness trait in wheat background. The molecular cytogenetic identification of novel wheat—Th. intermedium derivatives indicated that the frequent chromosome rearrangement occurred in the progenies of wheat-Thinopyrum hybridization. The new wheat-Thinopyrum derived lines may increase the genetic diversity for wheat breeding.


Introduction
Thinopyrum intermedium (Host) Barkworth & D.R. Dewey, a hexaploid perennial species, is widely distributed in Europe, western Asia, and northern Africa [1].The great geographical and climatic diversity of the native distribution has resulted in high variability and accumulated genetic rearrangement among Th.intermedium subspecies [2].The relative ease of crossability between wheat and Th.intermedium has led to studies involving the transfer of valuable genetic traits from Th. intermedium to wheat [2,3].Different types of wheat-Th.intermedium partial amphiploids have Agronomy 2019, 9, 18 2 of 15 been developed [4][5][6][7][8], and a number of wheat-Th.intermedium chromosome addition, substitution, and translocation lines were produced for the localization of novel gene(s) for disease resistances and agronomic traits in diversified wheat background [9][10][11][12].The genome in situ hybridization (GISH) revealed that Th. intermedium consists of three genomes named J, J S , and St, among which the St genome shows a high degree of similarity to that of Pseudoroegneria strigosa [2,4], while the J and J S genomes appear to be related to the modified Th. elongatum/ Th. bessarabicum/ Dasypyrum villosum genomes [13][14][15][16].So far, the complete set of Th. intermedium chromosomes involving seven linkage groups and three individual genomes still have not been transferred to a wheat background [12].Thus continuous production and precise characterization of new wheat-Th.intermedium addition lines will be important for wheat breeding programs, including grain quality enhancement by chromosome engineering.
Grain endosperm texture responsible for hardness or softness is an important characteristic for wheat quality improvement, since the hard-textured grains require more grinding energy than soft-textured grains to reduce endosperm into flour [17].The flours from hard wheat are preferred for yeast-leavened bread baking, while flours from soft wheat are preferred for manufacturing cookies and cakes [18].Grain hardness is inherited and controlled primarily by the Hardness locus (Ha), and the Puroindoline genes in Ha locus were located in the extreme distal end of chromosome 5D in common wheat [17,18].The Puroindoline homologous alleles and mutations from barley [19] and wheat related species including Secale cereale [20], Dasypyrum villosum [21] and S. africanum [22] have been introduced into common wheat for increasing the genetic diversity of the wheat hardness locus.The identification of novel Puroindoline alleles from Thinopyrum species will be useful in improving the wheat quality for end-use products and better understanding Puroindoline gene function.
The objectives of the present study were to characterize new wheat-Th.intermedium chromosome addition lines by fluorescence in situ hybridization (FISH) and molecular markers, and to target the Puroindoline gene sequences on specific Th.intermedium chromosomes.The chromosomal rearrangement of Th. intermedium derived lines and new Th.intermedium specific markers are potentially useful in wheat quality improvement.

Plant Materials
Th. intermedium accession PI440043 (StJ S J genomes, 2n = 6x = 42) was obtained from the National Small Grains Collection at Aberdeen, Idaho, USA.Wheat-Th.intermedium addition lines Z series, the octoploid wheat and Th.intermedium partial amphiploid Zhong 5 and Zhong 2 [23] were provided by Dr. Zhijian Chang of Shanxi Academy of Agricultural Sciences, China.The wheat-Th.intermedium addition lines L series originated from TAF46 were from National Plant Germplasm System (NPGS), United States Department of Agriculture (USDA).Two wheat Chinese Spring (CS)-Th.intermedium addition named Hy36 and Hy37 were selected from a BC 1 F 6 generation of the wheat CS crosses to Zhang 5 and Zhong 2, respectively.

90K SNP Arrays Analysis
Plants from lines CS, Hy36 and Hy37 were grown in a greenhouse at 23 • C under 12-h photoperiod.At the two-leaf stage, the leaves were used for the isolation of genomic DNA by CTAB methods [30].The DNA of plants was subjected to genome-wide scanning using an Illumina wheat 90K iSelect SNP array (containing 81,587 SNPs) using a combination of GenomeStudio v2011.1, as described [31].All of the genotyping experiments were performed by CapitalBio Corporation (Beijing, China).SNP genotype calling and clustering were performed as described by Yuan et al. [32].

Molecular Marker Analysis
DNA was extracted from the young leaves of CS, Hy36, Hy37, Th. intermedium, and Zhong 5, using a sodium dodecyl sulfate (SDS) protocol [14].PCR-based Landmark Unique Gene (PLUG) primers [33] and CINAU (Cytogenetics Institute, Nanjing Agricultural University, Nanjing, China) markers [34] were designed from rice genomic DNA sequences specific for the wheat A, B, and D syntenic regions.Polymerase chain reaction (PCR) was performed in an Icycler Thermal Cycler (Bio-Rad Laboratories, Emeryville, CA, USA) in a 25 µ: reaction, containing 10 mmol Tris-HCl (pH 8.3), 2.5 mmol MgCl 2 , 200 µmol of each dNTP, 100 ng template DNA, 0.2 U Taq polymerase (Takara, Japan), and 400 nmol of each primer.The cycling parameters were 94 • C for 3 min for denaturation, followed by 35 cycles at 94 • C for 1 min, 55 • C for 1 min, 72 • C for 2 min, and a final extension at 72 • C for 10 min.The amplified products were restriction enzyme-digested and electrophoresis was conducted as described by Li et al. [26].The markers for chromosomal physical location were searched using the wheat genome database The International Wheat Genome Sequencing Consortium hole genome assembly (IWGSC) WGA v0.4 from https://urgi.versailles.inra.fr/blast/,and confirmed by PCR.

Cloning and Sequencing of the Pina Gene
Primers for Pina (Pina-F and Pina-R) and PCR protocol followed the reports of Li et al. [21].The target genes identified by PCR were excised from 1.0% agarose gels, and purified using a gel extraction kit (Qiagen, Valencia, CA, USA).The purified products were ligated into the pT7 Blue R-Vector (Novagen, Madision, WI, USA) using T4 DNA ligase, and then introduced into Escherichia coli DH5α using heat shock transformation.Nucleotide sequencing was performed on a polyacrylamide gel with the ABI prism 377 sequencer (Perkin Elmer, Waltham, MA, USA) as an automated fluorescent sequencing system.The nucleotide sequences based on two identically independent sequences of each clone were assembled, and the sequence alignment among different alleles and neighbor-joining (NJ) phylogenetic tree construction was carried out using BIOEDIT software v7.0.5 (http://www.mbio.ncsu.edu/Bioedit)[35].

Agronomic Traits Survey and Grain Hardness Tests
Field trials were performed by each plot contained three 2 m rows spaced 20 cm apart with three replicates at the Xindu Experimental Station, Chengdu, China during the 2016-2018 wheat growing seasons.The protein, wet gluten content, test weight, water absorption, and Zeleny sedimentation value of whole grains were determined using the near-infrared spectroscopy DA7250 (Perten, Sweden), according to the approved methods at Northwest A & F University, China.The hardness of grains was determined by a Perten Single Kernel Characterization System (SKCS), and the classification of hardness types referred to the report of Li et al. [21].

FISH of Th. Intermedium by Multiple Oligo Probes
Sequential FISH analyses of metaphase cell of Th. intermedium accession PI440043 were done by using multiple probes including Oligo-B11, Oligo-pSc119.2,Oligo-pTa535, Oligo-pTa71 [25], Oligo-pSt122 [12], Oligo-3A1 [36], Oligo-pDb12H, and Oligo-5S rDNA.The new Oligo-pDb12H was used to distinguish the J S genome, instead of the conventional LTR probe pDb12H [14,28].As shown in Figure 1a, the Oligo-pDb12H was able to clearly classify the 14 J S chromosomes of Th. intermedium, while the full abundant signals by Oligo-B11 were St-chromosomes, and the mostly telomeric and sub-telomeric signals by Oligo-B11 were J-chromosomes.Thus, the probes Oligo-B11 and Oligo-pDb12H can be used to distinguish the Th.intermedium chromosomes into three groups, each with 14 chromosomes, by using ND-FISH (Figure 1a).The sequential ND-FISH demonstrated that a total of 32 Th.intermedium chromosomes had clear signals of the probes Oligo-pSc119.2and Oligo-pTa535 (Figure 1b), and probes Oligo-5SrDNA, Oligo-3A1 (Figure 1c), Oligo-pTa71, and Oligo-pSt122 signals (Figure 1d) were observed in 12, four, six, and 24 Th.intermedium chromosomes, respectively.The ND-FISH patterns from the above probes were used to construct the Th.intermedium karyotypes (Figure 1e).The probes Oligo-pTa71 and Oligo-5SrDNA representing 18S-5.8-45SrDNA and 5S rDNA were mostly localized in the chromosomes of groups 1 and 5. Based on the different arm ratios between the two groups of chromosomes, they were presumably designated as 1J, 5J, 5J S , and 1St (Figure 1).

FISH of Th. Intermedium by Multiple Oligo Probes
Sequential FISH analyses of metaphase cell of Th. intermedium accession PI440043 were done by using multiple probes including Oligo-B11, Oligo-pSc119.2,Oligo-pTa535, Oligo-pTa71 [25], Oligo-pSt122 [12], Oligo-3A1 [36], Oligo-pDb12H, and Oligo-5S rDNA.The new Oligo-pDb12H was used to distinguish the J S genome, instead of the conventional LTR probe pDb12H [14,28].As shown in Figure 1a, the Oligo-pDb12H was able to clearly classify the 14 J S chromosomes of Th. intermedium, while the full abundant signals by Oligo-B11 were St-chromosomes, and the mostly telomeric and sub-telomeric signals by Oligo-B11 were J-chromosomes.Thus, the probes Oligo-B11 and Oligo-pDb12H can be used to distinguish the Th.intermedium chromosomes into three groups, each with 14 chromosomes, by using ND-FISH (Figure 1a).The sequential ND-FISH demonstrated that a total of 32 Th.intermedium chromosomes had clear signals of the probes Oligo-pSc119.2and Oligo-pTa535 (Figure 1b), and probes Oligo-5SrDNA, Oligo-3A1 (Figure 1c), Oligo-pTa71, and Oligo-pSt122 signals (Figure 1d) were observed in 12, four, six, and 24 Th.intermedium chromosomes, respectively.The ND-FISH patterns from the above probes were used to construct the Th.intermedium karyotypes (Figure 1e).The probes Oligo-pTa71 and Oligo-5SrDNA representing 18S-5.8-45SrDNA and 5S rDNA were mostly localized in the chromosomes of groups 1 and 5. Based on the different arm ratios between the two groups of chromosomes, they were presumably designated as 1J, 5J, 5J S , and 1St (Figure 1).

An Updated FISH Karyotype of Zhong 5 and Zhong 2
Cytological analysis of wheat-Th.intermedium partial amphiploid Zhong 5 with 2n = 56 were identified by sequential ND-FISH using probes Oligo-B11, Oligo-pDb12H, Oligo-pSc119.2,Oligo-pTa535, Oligo-pTa71, Oligo-3A1, and Oligo-(GAA) 7 (Figure 2).The 14 chromosomes showed hybridization signals of probe Oligo-B11, and eight of 14 chromosomes had the signals of Oligo-pDb12H including two chromosomes with the signal of the entire arms, and six chromosomes with the signal only on one arm (Figure 2a).It indicated that a total of 14 Th.intermedium chromosomes, including six St, two J S and six St-J S translocations, were observed in Zhong 5. Based on the comparison of current FISH patterns of Zhong 5 to CS by the probes Oligo-pSc119.2and Oligo-pTa535 [25], we found that 39 of the 42 wheat chromosomes from Zhong 5 had normal FISH patterns, except that a strong Oligo-pSc119.2signal appeared on the end of the chromosome 1AL, and a clear Oligo-pSc119.2signal on chromosome 2AS, as well as a distinct Oligo-pTa535 signal added on chromosome 3AS (Figure 2b).The probe Oligo-pTa71 hybridized to the nucleolus regions of both wheat and Thinopyrum chromosomes.The chromosomes 1B and 6B and two Th.intermedium chromosomes of Zhong 5 displayed hybridization signals by Oligo-pTa71 (Figure 2c).The FISH were conducted in the Zhong 5 derived addition lines Z1 to Z6, by using Oligo-pSc119.2,Oligo-pTa535 and Oligo-pSt122.The chromosomes 1St in Z3, 7J S in Z4 and Z6, 2St.J S translocation in Z5 were defined, respectively (Figure 2e).The remaining three pairs of Thinopyrum chromosomes in Zhong 5 were designated as E, C, and A with unidentified linkage group (Figure 2e).Similarly, the sequential ND-FISH with above eight Oligo probes were applied to identify the chromosome constitution of Th. intermedium chromosomes in Zhong 2, with 54 chromosomes (Figure S1).The karyotype of FISH patterns of six pairs of Th. intermedium chromosomes of Zhong 2 were shown in Figure 2f.A pair of J S chromosomes with the terminal region on the short arm contained strong signals from Oligo-pSc119.2 and Oligo-pSt122, which reassembled the 4J S chromosome [12].An undetermined J S chromosome was designated as F, and the remaining four pairs of chromosomes were consistent to those from Zhong 5. Therefore, Zhong 2 contained six J S , two St, and four St-J S chromosomes of Th. intermedium in the wheat background (Figure 2f).

ND-FISH of Lines Hy36 and Hy37
The mitotic metaphase cells of wheat-Th.intermedium derived lines Hy36 and Hy37 were identified by using ND-FISH with the probes Oligo-B11 and Oligo-pDb12H.Both Hy36 and Hy37 contained 44 chromosomes including a pair of Th. intermedium chromosomes, and only the short arm of the alien chromosome in Hy36 has Oligo-pDb12H hybridization signals (Figure 3a,e).Compared to the length (8.0 µm) of wheat chromosome 3B, the averaged relative length of Thinopyrum chromosomes in Hy36 and Hy37 were 5.0 µm and 4.6 µm, respectively.The short arm of the Th.intermedium chromosomes in Hy37 was a short satellite with clear nucleolus regions.FISH using probes Oligo-pSc119.2and Oligo-pTa535 showed that the added Th. intermedium chromosomes in both Hy36 and Hy37 have weak Oligo-pSc119.2signals on the long arm (Figure 3b,f).The FISH signals from Oligo-5SrDNA were observed in the interstitial regions of the short arms of Th. intermedium chromosomes in both Hy36 and Hy37 (Figure 3c,g).The alien chromosome in Hy37 has strong Oligo-3A1 hybridization signals on the short arm closed to centromere (Figure 3g).The sequential ND-FISH with Oligo-(GAA) 7 revealed that the added Th. intermedium chromosomes of Hy36 had strong hybridization signals on the pericentromeric region on the short arm (Figure 3d).Oligo-pTa71 had the FISH signals on the alien chromosome of Hy37 (Figure 3h).The ND-FISH patterns of the added Th. intermedium chromosomes in Hy36 and Hy37 were indicated in Figure 3i,j, respectively.It is likely that the short arm of chromosome C contributed to the long arm of alien chromosome in both Hy36 and Hy37, while the short arm of alien chromosome in Hy37 were originated from the short arm of chromosome E in Zhong 5, and the short arm of alien chromosome in Hy36 originated from short arm of chromosome F in Zhong 2. Therefore, the sequential ND-FISH with multiple probes can be used to trace the chromosome transmission or variation during the processes of hybridization and selection.
with multiple probes can be used to trace the chromosome transmission or variation during the processes of hybridization and selection.

Molecular Markers Validation of Linkage Groups
The 90K iSelect SNP array was performed to detect the polymorphisms among CS, Hy36, and Hy37.A total of 7470 SNP markers showed the polymorphism among CS, Hy36, and Hy37.Most of markers (90.89%) from Hy36 and Hy37 were identical to those from CS, implying the two lines were closely related to CS.The 32.4% categorized heterozygous SNPs from both Hy36 and Hy37 were distributed on the chromosomes 3D, 5A, 5B, and 5D.These highly heterozygous SNPs may be resulted from the homologous groups 3 and 5 of Thinopyrum chromosomes added in Hy36 and Hy37.
We analyzed the amplification patterns of the DNAs from Zhong5, Zhong 2, Hy36, Hy37, and CS, respectively, by using 350 PLUG markers [33] which were located on seven wheat homoeologous groups and 890 CINAU markers [34] derived from all D. villosum chromosomes (Figure 4).The 51 markers from the short arm of group 5, and the 44 markers from the short arm of group 3 generated Th. intermediumspecific bands in Hy36 and Hy37, respectively.As shown in Figure 4a, the Th.intermediumspecific bands in Hy36 were identical to Zhong 5, and those in Hy37 were identical to Zhong 2. Among 51 markers located on short arm group 5, a total of 34 and 22 markers produced Th. intermedium-specific polymorphic bands in Hy36 and Hy37, respectively (Figure 4b).The results showed that these two group 5 Th.intermedium chromosomes were genetically different between Hy36 and Hy37.A total of 30 and 28 of 44 markers located on short arm of group 3 produced Th. intermedium-specific polymorphic bands in Hy36 and Hy37, respectively (Figure 4b).The FISH results and molecular marker surveys demonstrated that the Th.intermedium chromosomes in Hy36 and Hy37 were 5J S S.3StS and 5JS.3StS, respectively.

Molecular Markers Validation of Linkage Groups
The 90K iSelect SNP array was performed to detect the polymorphisms among CS, Hy36, and Hy37.A total of 7470 SNP markers showed the polymorphism among CS, Hy36, and Hy37.Most of markers (90.89%) from Hy36 and Hy37 were identical to those from CS, implying the two lines were closely related to CS.The 32.4% categorized heterozygous SNPs from both Hy36 and Hy37 were distributed on the chromosomes 3D, 5A, 5B, and 5D.These highly heterozygous SNPs may be resulted from the homologous groups 3 and 5 of Thinopyrum chromosomes added in Hy36 and Hy37.
We analyzed the amplification patterns of the DNAs from Zhong5, Zhong 2, Hy36, Hy37, and CS, respectively, by using 350 PLUG markers [33] which were located on seven wheat homoeologous groups and 890 CINAU markers [34] derived from all D. villosum chromosomes (Figure 4).The 51 markers from the short arm of group 5, and the 44 markers from the short arm of group 3 generated Th. intermedium-specific bands in Hy36 and Hy37, respectively.As shown in Figure 4a, the Th.intermedium-specific bands in Hy36 were identical to Zhong 5, and those in Hy37 were identical to Zhong 2. Among 51 markers located on short arm group 5, a total of 34 and 22 markers produced Th. intermedium-specific polymorphic bands in Hy36 and Hy37, respectively (Figure 4b).The results showed that these two group 5 Th.intermedium chromosomes were genetically different between Hy36 and Hy37.A total of 30 and 28 of 44 markers located on short arm of group 3 produced Th. intermedium-specific polymorphic bands in Hy36 and Hy37, respectively (Figure 4b).The FISH results and molecular marker surveys demonstrated that the Th.intermedium chromosomes in Hy36 and Hy37 were 5J S S.3StS and 5JS.3StS, respectively.

Molecular Cloning of Th. Intermedium-Specific Pina-like Genes
The primer pair Pina-F/Pina-R for targeting puroindoline alleles were used to amplify the DNA of Th. intermedium, Hy36, Hy37 and D. villosum.The amplicon of about 500 bp for Pina gene was cloned and sequenced.A total of 20 clones from Th. intermedium, 25 each from Hy36 and Hy37, and

Molecular Cloning of Th. Intermedium-Specific Pina-like Genes
The primer pair Pina-F/Pina-R for targeting puroindoline alleles were used to amplify the DNA of Th. intermedium, Hy36, Hy37 and D. villosum.The amplicon of about 500 bp for Pina gene was cloned and sequenced.A total of 20 clones from Th. intermedium, 25 each from Hy36 and Hy37, and ten from D. villosum were sequenced.Two clones from Hy36 (Genbank numbers: MH745807 and MH745808) were different from the reported wheat Pina sequences.All clones from Hy37 were identical to the registered wheat Pina gene sequences.Four redundant Th. intermedium and three D. villosum sequences were also obtained, respectively.The conserved tryptophan-rich region (WRWWKWWK) of the predicted PINA protein sequences was located between cysteine residues 3 and 4, and the substitution Agronomy 2019, 9, 18 8 of 15 from glutamine to arginine at position 69 (K65R) was identified from Th. intermedium PINA for the first time (Figure S2).The Pina sequences from wheat, Secale species, and the present Th.intermedium and D. villosum were used to construct a phylogenetic tree.As shown in Figure 5, the Pina gene sequences from Secale were first separated, and Pina genes of wheat were categorized as the second subgroup.The Pina sequences from D. villosum, Th. intermedium, and the Hy36 were clustered as the third subgroup.The results supported that the genome of D. villosum was genetically close to Th. intermedium genomes.
ten from D. villosum were sequenced.Two clones from Hy36 (Genbank numbers: MH745807 and MH745808) were different from the reported wheat Pina sequences.All clones from Hy37 were identical to the registered wheat Pina gene sequences.Four redundant Th. intermedium and three D. villosum sequences were also obtained, respectively.The conserved tryptophan-rich region (WRWWKWWK) of the predicted PINA protein sequences was located between cysteine residues 3 and 4, and the substitution from glutamine to arginine at position 69 (K65R) was identified from Th. intermedium PINA for the first time (Figure S2).The Pina sequences from wheat, Secale species, and the present Th.intermedium and D. villosum were used to construct a phylogenetic tree.As shown in Figure 5, the Pina gene sequences from Secale were first separated, and Pina genes of wheat were categorized as the second subgroup.The Pina sequences from D. villosum, Th. intermedium, and the Hy36 were clustered as the third subgroup.The results supported that the genome of D. villosum was genetically close to Th. intermedium genomes.

Agronomic Traits and Grain Quality Observation
The plants Hy36, Hy37, CS, and Zhong 5 were grown under field conditions, and their agronomic traits and grain quality were measured (Table 1, Figure S3).The Hy36 line showed a higher number of spikelet per spike, longer grain length, but a shorter spike length compared to those of Hy37.The plant height and the tiller per plant of Hy36 and Hy37 were significantly reduced compared to CS.The kernel protein contents, wet gluten content, and Zeleny value of Hy36 and Hy37 were significantly increased, in comparison with those in wheat parent CS, showing the end-product quality improved.Hy36 became soft, according to the SKCS hardness index, comparing to Hy37 and the recurrent parent CS, to an extent of 17.4%.The results imply that the addition of Th. intermedium chromosome 5J S in Hy36 may reduce the hardness.The increase of protein and gluten contents in Hy36 and Hy37 proved that the introgression of Th. intermedium group 5 and 3 chromatin may be associated with grain end-use quality characteristics.

Agronomic Traits and Grain Quality Observation
The plants Hy36, Hy37, CS, and Zhong 5 were grown under field conditions, and their agronomic traits and grain quality were measured (Table 1, Figure S3).The Hy36 line showed a higher number of spikelet per spike, longer grain length, but a shorter spike length compared to those of Hy37.The plant height and the tiller per plant of Hy36 and Hy37 were significantly reduced compared to CS.The kernel protein contents, wet gluten content, and Zeleny value of Hy36 and Hy37 were significantly increased, in comparison with those in wheat parent CS, showing the end-product quality improved.Hy36 became soft, according to the SKCS hardness index, comparing to Hy37 and the recurrent parent CS, to an extent of 17.4%.The results imply that the addition of Th. intermedium chromosome 5J S in Hy36 may reduce the hardness.The increase of protein and gluten contents in Hy36 and Hy37 proved that the introgression of Th. intermedium group 5 and 3 chromatin may be associated with grain end-use quality characteristics.

Discussion
The ND-FISH based chromosome painting technique with the labelled oligonucleotide (oligo) is widely applied, due to obvious advantages over conventional FISH procedures, such as easy probes synthesis, flexible design, relatively low cost, and high-throughput [25,30,[36][37][38][39].This technique has been successfully used for distinguishing different genomes, such as Oligo-1162 for Secale cereale [30] and Oligo-B and Oligo-D for wheat B-and D-chromosomes [39], respectively.GISH by using D. villosum genomic DNA as a probe [13,15], and FISH using Dasypyrum specific LTR pDb12H probe can distinguish the J S chromosomes of Th. intermedium genome [12,14,40].In the present study, the hybridization patterns of ND-FISH using probe Oligo-pDb12H can substitute the conventional FISH with the pDb12H probe.All 14 J S -chromosomes were clearly visualized by ND-FISH with Oligo-pDb12H in Th. intermedium (Figure 1).Another new LTR probe B11 [29] was also converted to the Oligo-probe, which was helpful for detecting wheatgrass fragments in common wheat by using ND-FISH.Our results on the constitution of alien chromosomes in wheat-Th.intermedium partial amphiploid Zhong 5 (Figure 2) are identical to the St-genome DNA-based GISH and FISH [5,36,41].Taking advantage of a combination of Oligo-pDb12H and Oligo-B11 for multicolor ND-FISH, we can effectively distinguish St, J S , and J-chromosomes of Th. intermedium, and precisely trace the specific Th.intermedium chromosome segments in large numbers of wheat-Th.intermedium hybridization progenies.
By using GISH with St genomic DNA as a probe, Tang et al. [42] observed that Zhong 5 has three pairs of St chromosomes, two pairs of J S chromosomes, and two pairs of St-J S reciprocal translocation chromosomes.Chen et al. [41] reported that Zhong 5 consists of four St and two J S chromosomes, plus two J S -W translocations, and six St and J S or J translocated chromosomes.Han et al. [43] demonstrated that Zhong 5 contains 42 wheat chromosomes plus 14 Th.intermedium chromosomes by multi-color GISH.In the present study, we observed that the composition of Th. intermedium chromosomes in Zhong 5 is six St-chromosomes, two J S -chromosomes, and six St/J S translocated chromosomes.Recently, Lang et al. [36] found that Zhong 5 contains two pairs of J S -W translocations, which were confirmed as T3DS-3AS.3AL-7JS S and T3AL-7J S S.7J S L translocations.Consequently, Zhong 5 may have different kinds of karyotypes.Among them, one has normal 7J S , and the other has translocation between 7J S and 3A.The above 7J S -wheat chromosomes in Zhong 5 were transmitted to wheat-Th.intermedium addition line Z4 [36].The relatively high variation of Th. intermedium chromosome constitution in Zhong 2 was also observed [41,43].The karyotypic variation found in the wheat-Thinopyrum partial amphiploids is due to the high frequency of paired wheat and Th.intermedium chromosomal rearrangement during the period of maintenance by successive self-selection [5][6][7][44][45][46].
The localization of rDNA in the wheat and Triticeae species was observed by cytological methods [47].The probes pTa794, a 420 bp sequence of wheat encoding the 5S rRNA gene [27], and pTa71, a 9 kb sequence that encodes 18S-5.8S-26SrRNA genes [48] are commonly used for FISH analyses.The 45S rDNA sites preferentially localize on the chromosomes of homoeologous groups 1, 5, and 6 [49], while the 5S rDNA sites appear on the short arms of groups 1 and 5 chromosomes [50].Li et al. [51] reported that six or eight major pTa71 signals of rDNA loci are detected in 42 chromosomes of three Th.intermedium accessions, with two or three pairs located in terminal parts, one pair of loci located interstitially.Mahelka et al. [15] used FISH with pTa794 probe to reveal that three or four pairs of major loci were detected in all accessions of Th. intermedium.In the present study, we used the Oligo-5SrDNA representing pTa794 probe for ND-FISH, and revealed that the signals of Oligo-5S rDNA were presented on the ends of two pairs of St-chromosomes, one pair of J S -chromosome and two pairs of J-chromosomes in Th. intermedium accession (Figure 1).We suggested that a pair of major 5SrDNA loci in J S chromosomes were probably lost during the evolution of the hexaploid species of Th. intermedium.Moreover, the FISH signals of Oligo-3A1 were located on 5AL, 5BL and 5DS [36].In the present study, we found the hybridization signals of Oligo-3A1 on 5JS and 5J S L (Figure 1).The varied distribution patterns of FISH hybridization signals of Oligo-pTa71, Oligo-5SrDNA and Oligo-3A1 sites among different wheat and Th.intermedium chromosomes' arms (Figures 1 and 2) indicate the complexity of chromosome rearrangement during the evolutionary process of both allopolyploid wheat and Th.intermedium species.
Chromosome rearrangement is commonly observed in wheat-Th.intermedium derivatives.Different St-J S or J-St genome chromosomes as either Robertsonian, intercalary, or terminal interchanges are observed in several wheat-Th.intermedium partial amphiploids and addition lines [2].The wheat-Th.intermedium addition lines Yi 4212, HG295, Z1-Z6 originated from Zhong 5 have extensively rearranged Th. interemedium chromosomes [10,36,41].We previously observed non-Roberstanian St-J S translocations among homologous group 4 chromosomes in wheat-Th.intermedium substitution lines X479 and X482, by GISH-FISH and molecular marker analysis [52].The present study also showed that the lines Hy36 and Hy37 contained the translocation between chromosomes 5J S , 5J with 3St, respectively.It is likely that St-genome is closely related to J or J S genomes than between J and J S , which is resulted in the high frequency occurrences of St-J and St-J S chromosome translocation [2,51].
Kernel texture controlled by the hardness (Ha) locus including Puroindoline genes in wheat has long been recognized as a highly heritable trait [53].The hardness loci, including Pina and Pinb genes are reported in species of genera such as Triticum, Aegilops, Hordeum, Secale, and Avena [17,54].These genes were not detected in species of other Poaceae subfamilies, such as rice, maize, or sorghum [53].The sequence variation of the Pina locus constitutes an excellent model for studying the mechanisms of evolution in the tribe Triticeae [54,55].In the present study, the phylogenetic analysis indicates that the distinct evolutionary patterns of Pina genes occurred among the different Triticeae genomes (Figure 5).The Pina gene sequences from Th. intermedium showed great variation, possibly due to hexaploid genome or the cross pollination which resulted in heterozygosity of genes in the Th.intermedium accession.The Puroindoline genes were localized on the group 5 chromosomes of all Triticeae diploids.However, the Puroindoline genes have been deleted from chromosomes 5A and 5B, and only a gene was found on the short arm of chromosome 5D of common wheat [54].In the present study, we found that the chromosome 5J S in Hy36 contained the Th.intermedium specific Pina gene sequences, however, chromosome 5J in Hy37 had no Pina genes.The evolutionary dynamics of Pina genes along with the chromosome rearrangements in polyploid Triticeae species including Th. intermedium genomes needs to be further investigated.Moreover, the hordoindolines on chromosome 5HS of Hordeum chilense plays a role in reducing grain hardness in a wheat genetic background [19].The Dasypyrum villosum derived Pina genes are associated with soft grain texture [22].Our previous studies indicated that the introduction of Pina gene from S. africanum chromosome 5R afr S to wheat also clearly reduced the grain hardness [21].Gazza et al [56] also reported that wheat-Th.intermedium introgression lines inherited PINA proteins from Th. intermedium, and produced medium-hard kernels.Therefore, it is reasonable to find that Hy36 with 5J S S.3StS chromosome displayed soft grain texture, with an approximately 17% reduction in the SKCS hardness index compared with that of its parent CS and Hy37 with 5JS.3StS chromosome.Furthermore, the Puroindoline genes have been proven to have a broad range of antibacterial and antifungal activities, the effects of Pina genes on wheat rust are also recognized [57].Further studies will be conducted to develop different types of wheat 5J S translocation lines by chromosome manipulation, and to reveal the expression between Th. intermedium derived genes and wheat native genes for better quality and elite plant defense characteristics.As shown in Table 1, Hy36 and Hy37 showed shorter plant height and tillers per plant for tight plant architectures, which may overcome the defects of the wheat line Chinese Spring, for direct yield improvement [58].Moreover, Hy36 and Hy37 displayed higher grain wet gluten content, stable time, and sedimentation value, which are in favor of dough processing quality.The deviation of morphology and grain quality between Hy36 and Hy37 were mostly due to the different chromosome arms of 5Js and 5J, and their interaction with wheat background.It is convincingly shown that the wheat-Th.intermedium hybrids are useful for future wheat breeding for yields, end-use quality, market classification, and economic benefits [3,11,12].

Figure 4 .
Figure 4. PCR profiling (a) and physical localization (b) of Th. intermedium-polymorphic markers in wheat chromosomes 5B and 3B by a BLAST search the reference genomic sequences version 1.0 at

Figure 5 .
Figure 5.A phylogenetic tree generated from the nucleotide sequence alignment of the Pina genes by neighbor joining method by 1000 bootstrap replicates.Scale bar 0.01 = 1% difference among nucleotide sequences.

Figure 5 .
Figure 5.A phylogenetic tree generated from the nucleotide sequence alignment of the Pina genes by neighbor joining method by 1000 bootstrap replicates.Scale bar 0.01 = 1% difference among nucleotide sequences.

Table 1 .
Agronomic and grain quality traits of wheat-Th.intermedium lines and the wheat parent CS.

Table 1 .
Agronomic and grain quality traits of wheat-Th.intermedium lines and the wheat parent CS.