Characterization of a new hexaploid triticale 6D(6A)substitution line with increased grain weight and decreased spikelet number

Zhiyu Feng, Zhongqi Qi, Dejie Du, Mingyi Zhng, Aiju Zho, Zhorong Hu,Mingming Xin, Yingyin Yo, Huiru Peng, Qixin Sun,Zhongfu Ni,*

aKey Laboratory of Crop Heterosis and Utilization, State Key Laboratory for Agrobiotechnology, Beijing Key Laboratory of Crop Genetic Improvement,College of Agronomy and Biotechnology, China Agricultural University,Beijing 100193,China

bInstitute of Wheat Research,Shanxi Academy of Agricultural Sciences,Linfen 041000,Shanxi,China

cHebei Crop Genetic Breeding Laboratory,Institute of Cereal and Oil Crops,Hebei Academy of Agriculture and Forestry Sciences,Shijiazhuang 050035,Hebei,China

Keywords:In situ hybridization Spikelet number Substitution line Thousand-grain weight Triticale

ABSTRACT Hexaploid triticale (×Triticosecale, AABBRR) is an important forage crop and a promising energy plant.Transferring D-genome chromosomes or segments from common wheat(Triticum aestivum)into hexaploid triticale is attractive in improving its economically important traits. Here, a hexaploid triticale 6D(6A) substitution line Lin 456 derived from the cross between the octoploid triticale line H400 and the hexaploid wheat Lin 56 was identified and analyzed by genomic in situ hybridization (GISH), fluorescence in situ hybridization (FISH), and molecular markers. The GISH analysis showed that Lin 456 is a hexaploid triticale with 14 rye(Secale cereale)chromosomes and 28 wheat chromosomes, whereas non-denaturing fluorescence in situ hybridization (ND-FISH) and molecular marker analysis revealed that it is a 6D(6A) substitution line. In contrast to previous studies,the signal of Oligo-pSc119.2 was observed at the distal end of 6DL in Lin 456.The wheat chromosome 6D was associated with increased grain weight and decreased spikelet number using the genotypic data combined with the phenotypes of the F2 population in the three environments.The thousand-grain weight and grain width in the substitution individuals were significantly higher than those in the non-substitution individuals in the F2 population across the three environments. We propose that the hexaploid triticale 6D(6A) substitution line Lin 456 can be a valuable and promising donor stock for genetic improvement during triticale breeding.

1. Introduction

Triticale (×Triticosecale Wittmack) is an artificial grass crop produced by hybridization between wheat (Triticum spp.) and rye (Secale cereale L.) [1]. Numerous triticale lines with various ploidy levels and genome combinations have been created and released, such as octoploid (AABBDDRR), hexaploid(AABBRR), and tetraploid (DDRR) triticales [2-8]. Compared to octoploid and tetraploid triticale, hexaploid triticale has been the most successful type because of its reproductive stability,superior vigour, abiotic stress tolerance, and high lysine content [1,9-13]. The large-scale triticale development programs provide many triticale germplasms with wide adaptation to a range of environments, especially at CIMMYT,Mexico, whereas the genetic diversity in those programs is extremely narrow [1]. Thus, it is necessary to continually produce new triticale varieties with superior agronomic traits to enrich the available genetic pool.

The D-genome chromosomes are considered to contribute significantly to the success of cultivated wheat [14-18].Correspondingly, hexaploid triticale can be improved by introducing the D-genome chromosomes from common wheat. For example, the introgression of wheat chromosome 1D improved baking quality of hexaploid triticale [19,20]. The smallest increase in the number of sprouted kernels was observed in cv.Presto substitution lines 3D(3R),2D(2R),and 6D(6R)[21]. A hexaploid triticale 4D(4B)substitution line confers superior stripe rust resistance[20]. Many desirable characteristics of hexaploid triticale have been associated with the presence of 2D(2R) [22-25]. The selective advantage of the 6D(6A) substitution can be explained by its better tolerance to aluminium relative to complete triticale (AABBRR) [26].Collectively, these studies indicate that the transfer of the Dgenome chromosomes and/or chromosomal segments into hexaploid triticale is attractive for improving economically important traits[27].

Several types of crosses have been utilized to introduce the D-genome chromosomes from hexaploid wheat to hexaploid triticale[6-8,19,28].Hybridization between hexaploid triticale and hexaploid wheat is a routine procedure in hexaploid triticale breeding, which may result in chromosome substitutions between the R-genome of rye and the Dgenome of wheat[29].Another method is to cross octoploid triticale (AABBBDDRR) with tetraploid triticale (AARR or BBRR)[30].Furthermore,many D genome substitution lines of hexaploid triticale have been developed from the advanced generations of octoploid × hexaploid triticale crosses [31].

In this study,we characterized a 6D(6A)substitution line of hexaploid triticale that was derived from crossing of octoploid triticale × hexaploid wheat using GISH, FISH, and molecular marker analyses. In addition, a segregating F2population was obtained by crossing Lin 456 with an elite triticale cultivar 4100 (AABBRR). The genetic analysis showed that the F2individuals carrying wheat chromosome 6D conferred a higher grain weight and greater grain width in different environments, which is useful for diversifying triticale germplasms and breeding for new triticale cultivars.

2. Materials and methods

2.1. Plant materials and field trials

Lin 456 is a genetically stable triticale line with the preferable agronomic characteristics obtained from the hybridization of octoploid triticale H400(AABBDDRR)and hexaploid wheat Lin 56 (AABBDD) following 10 generations of consecutive selection for a triticale-like phenotype with a large grain size and high weight.The elite hexaploid triticale cultivar 4100 with an intact AABBRR genome was adapted to Henan province,China. Lin 456 and 4100 were grown with three replicates(two rows per replicate,30 seeds per row)in the experimental fields at three locations during the 2015-2016 growing season(N35°18′, E113°54′, Xinxiang, Henan province; N38°04′,E114°28′, Shijiazhuang, Hebei province; and N36°48′,E111°30′,Linfen,Shanxi province).Seeds were well distributed in rows that were 2.0 m long and 0.3 m apart, and the field trials were managed following the normal local practices.Furthermore, yield trials of Lin 456 and 4100 were conducted in a randomized complete design with three replicates (40 rows per replicate and 50 seeds per row) at Xinxiang, Henan province, during the 2015-2016 growing season.

The F2population from the cross between Lin 456 and 4100 was grown in experimental fields in the same three locations as the parents.The F2population in Henan,Hebei,and Shanxi sites contained 913, 568,and 665 individuals,respectively.

2.2. GISH and ND-FISH analyses

GISH was carried out in root tip cells using rye genomic DNA labelled with fluorescein-12-dUTP(PerkinElmer,Waltham,Massachusetts, USA, http://www.perkinelmer.com/) as a probe. The probe CRW from the PCR product of wheat was labelled with Texas-red-5-dUTP(PerkinElmer,Waltham,Massachusetts,USA)and applied to analyze the centromeres of the wheat chromosomes [32,33]. Genomic DNA from wheat cultivar CS was used for blocking.The preparation of the root tips and the hybridization procedure were carried out as previously described by Han et al. [34]. All of the cytological photographs were taken with a Nikon Eclipse E600 fluorescence microscope(Nikon Corporation,Tokyo,Japan)and were captured with a CCD camera(Diagnostic Instruments, Inc., Sterling Heights, MI, USA). At least five metaphase cells were observed in each slide, which was the same for the following FISH analysis.

ND-FISH analysis was performed according to the methods described by Xiao et al. [35]. Oligo-pTa535 and Oligo-pSc119.2 probes were used to identify the entire set of 42 wheat chromosomes. A combination of Oligo-1162, Oligo-pSc200,and Oligo-pSc250 probes unambiguously distinguished rye chromosomes from wheat chromosomes. Combined with Oligo-pSc119.2 (or Oligo-pSc200 and Oligo-pSc250), (AAC)6 was applied to precisely identify each rye chromosome. The probe sequences and amounts per slide were described in previous reports [36,37]. The synthetic oligonucleotides were 5′end-labelled with 6-carboxyfluorescein (6-FAM), 6-carboxytetramethylrhodamine (Tamra) or Cy5. The synthesized probes were diluted using 1× TE solution (10 mmol L-1Tris-HCl,1 mmol L-1EDTA,pH 7.0).The chromosome spreads of the materials were prepared with the methods described by Han et al.[38].

Fig.1-GISH pattern and ND-FISH pattern.(a)GISH and FISH pattern of Lin 456 showing the presence of 14 rye chromosomes(green)among the 42 chromosomes(2n = 42)using rye genomic DNA(green)and CRW(red)as probes.The red bands indicate centromeres. Chinese Spring(CS)was used for blocking.(b) Non-denaturing fluorescence in situ hybridization(ND-FISH)assays of Lin 456 using Oligo-1162(red),Oligo-pSc200(red),Oligo-pSc250(red),Oligo-pTa535(green),and Oligo-pSc119.2(yellow)as probes showing a pair of 6D substitution chromosomes (arrows)and the absence of chromosome 6A.The chromosomes were counterstained with DAPI (4′,6-diamidino-2-phenylindole).

2.3. Phenotypic evaluation

Nine agronomic traits were evaluated for the parental lines and their F2population, including PH, SL, TSN, FSN, SSN, SC,TGW, GL, and GW. The data of plant height and spike traits(SL, TSN, FSN, SSN, and SC) of the parental lines and their F2population were collected from the main tillers of each plant during the late grain-filling stage as described by Zhai et al.[39].After harvesting and threshing,the grain traits(TGW,GL,and GW)were measured using a SC-G grain appearance image analysis system (Hangzhou Wanshen Detection Technology Co., Ltd., Hangzhou, China; http://www.wseen.com/) according to the procedure described by Yin et al. [40]. The phenotypic evaluation of the PH and spike traits (SL, TSN,FSN, SSN, and SC) of the two parental lines were recorded as the average values of five representative plants per replicate.The grain traits(TGW,GL,and GW)of the parental lines were measured from 20 representative spikes of each replicate.For the yield trials,all the plants of each replicate were harvested and used for evaluation.

2.4. Molecular marker analysis

Genomic DNA was isolated from leaves of young seedlings using a modified CTAB method [41]. The PCR amplifications were performed as described by Zhai et al. [39]. The PCR products were electrophoresed on an 8% non-denatured polyacrylamide gel and visualized by staining with silver nitrate following the procedure of Marklund et al.[42].

A total of 13 chromosome-specific molecular markers on wheat 6D (12 SSRs and 1 STS) were used to determine the genomic substitution of Lin 456 (Table S1). In addition, six parental polymorphic wheat SSR markers were used to analyze the genetic constitutions of the F2population, and both chromosomes 6A and 6D contained three SSR markers (Table S2).Most of the selected wheat SSR primers are publicly available at the GrainGenes website(https://wheat.pw.usda.gov/GG3/).

2.5. Statistical analysis

The data preparation of the phenotypes and genotypes was performed with Microsoft Excel version 2016. The Pearson’s correlation coefficients between the parameters were calculated using SPSS version 23.0 (SPSS Inc., Chicago, USA). The frequency distributions and Shapiro-Wilk tests were performed with R software V. 3.3.1 (https://www.r-project.org/)to check departures from a normal distribution. The significance of the differences was assessed with the Student’s t-test at P ＜0.05.

3. Results

3.1. Chromosome constitution of Lin 456

GISH, ND-FISH, and PCR-based markers were employed to characterize the chromosome constitution of Lin 456.The GISH result showed that Lin 456 was a hexaploid triticale with 42 intact chromosomes, including 14 from rye and 28 form wheat(Fig.1-a).To clearly discern each chromosome of the wheat and rye genomes,we then analyzed this line using ND-FISH assays.As shown in Fig.1-b,the ND-FISH signal pattern of the wheat Bgenome chromosomes and rye chromosomes in Lin 456 were consistent with previous studies, whereas only six A-genome chromosome pairs, including 1A, 2A, 3A, 4A, 5A, and 7A, were detected.The signal pattern of Oligo-pTa535(green signal)on the pericentromere and the short arm of additional chromosome pair was similar to that of chromosome 6D,which indicates that Lin 456 is potentially a 6D(6A)substitution line(Figs.1-b and S1-a). Thus, the genetic constitution of Lin 456 was further confirmed by 13 molecular markers located on 6D flanking the entire chromosome.As expected,the target DNA bands of all the markers were amplified from the common wheat CS and Lin 456(Fig.2).The signal of Oligo-pSc119.2(yellow signal)was detected at the distal end of the long arm of chromosome 6D in Lin 456(Fig. 1-b). Collectively, the data indicate that Lin 456 is a 6D(6A)substitution line of hexaploid triticale.

Fig.2- Amplification patterns and genetic and physical positions of wheat chromosome 6D markers used for analyzing Lin 456.(a) PCR amplification patterns of 1 STS marker and 12 SSR markers physically mapped on wheat chromosome 6D in common wheat Chinese Spring(lane 1)and Lin 456(lane 2).M indicates the D2000 DNA marker(M122 Direct-load Star Marker,GenStar Inc.,Beijing,China)used as a marker reference.(b)Genetic positions(cM)of markers obtained from a previous report[64].(c)Physical positions(bp)of markers on wheat chromosome 6D(IWGSC RefSeq v1.0).(d)Deletion bin map of markers on chromosome 6D.SSR-119 was mapped on chromosome 6D of the Aegilops tauschii genome and a scaffold of wheat(TGACv1_scaffold_644465_U), and there was no hit on the wheat chromosome 6D(IWGSC RefSeq v1.0).The black triangles indicate centromeres in the genetic and deletion bin map.

3.2. Phenotypic evaluation

To evaluate the agronomic traits of Lin 456,an elite hexaploid triticale line 4100 with the entire and intact AABBRR genome was used as a control (Fig. S1-b-d). Lin 456 showed significantly higher TGW by at least 19.4 g,GL by 1.6 mm,and GW by 0.3 mm than 4100 in all environments. In contrast, SL, TSN,and FSN of Lin 456 were significantly lower than those of 4100.Consequently, SC of Lin 456 was much higher than that of 4100.No significant difference was observed between the two genotypes for SSN and PH(Table 1,Fig.3-a-e).Even with lower SL and FSN, the grain yield per plot of Lin 456 was 11.58%higher than that of 4100(Fig.3-f).

The means and ranges for the PH, spike traits (SL, TSN,FSN,SSN,and SC),and grain traits(TGW,GL,and GW)of the F2population derived from the cross between Lin 456 and 4100 are listed in Table 1.Wide variation for each trait(coefficients of variation of 5.04% to 19.17%) was observed in the F2population, and the variation of SSN in the population was much greater(132.73%to 199.91%)(Table 1).The Shapiro-Wilk tests revealed that the grain-related traits basically had normal distributions, whereas PH and the spike morphological traits departed significantly from normality(Table S3).

Pearson’s coefficients of correlation for different traits were similar in the three environments (Table S4). Plant height was positively correlated with SL and the grain traits,whereas SC was negatively correlated with SL and PH.Significantly positive correlations were also detected among different grain traits,including TGW,GL,and GW.

Table 1-Parental and population means,ranges, and coefficients of variation for plant height (PH),spike length (SL), total spikelet number per spike (TSN), fertile spikelet number per spike (FSN), sterile spikelet number per spike (SSN), spikelet compactness (SC),thousand-grain weight (TGW),grain length(GL),and grain width(GW).

3.3. Genetic effect of chromosome 6D

To distinguish chromosomes 6A and 6D in the hexaploid triticale, we used six chromosome-specific SSR markers (3 for 6A and 3 for 6D) in genotyping analysis. As expected, those chromosome-specific SSR markers amplified the diagnostic bands from 4100 and Lin 456 only (Fig. 4), which suggests that they can track chromosomes 6A and 6D. Therefore, these markers were further used to genotype the F2population derived from the 4100 × Lin 456 cross. For the F2population, the Chisquared test showed that the segregation ratios of the six markers deviated significantly from the expected ratio of 3:1,which provided genetic evidence for abnormal synapsis and segregation during meiosis in some F2individuals(Table S5).

To investigate the effect of chromosome 6D on various traits,the F2population were divided into substitution (with 6D) and non-substitution(without 6D)subpopulations according to their genotypes. The phenotypic distribution of FSN and grain traits was significantly different between the two subpopulations(P ＜0.05)(Figs.S2 and S3).To further evaluate the genetic effect of chromosomes 6A and 6D,we performed t-test analysis using the F2population under three environments and found that the average TGW and GW of the F2individuals with 6D were significantly higher than those without 6D,whereas the average SL,TSN,and FSN of the F2individuals with 6D were significantly lower than those without 6D(P ＜0.05)(Table 2).By contrast,PH,SL, TSN, and FSN of F2individuals with 6A were significantly higher than those without 6A under all environments. The chromosome 6A increased grain width but decreased grain length when compared the F2plants with and without 6A(Table 3).As a consequence,there is no significant difference for TGW between F2plants with and without chromosome 6A under the environments of Shanxi and Hebei.

4. Discussion

4.1. A 6D(6A) substitution line of hexaploid triticale revealed by GISH, FISH and PCR-based markers

Fig.3- Characterization of substitution line Lin 456 compared to 4100.(a) Phenotypes of triticale line 4100(left)and the substitution line Lin 456(right).Scale bar,10 cm.Comparisons between 4100(left)and Lin 456(right):(b)spikes,scale bar,3 cm;(c)spikelets in the middle of spikes,scale bar,1 cm;(d)grain widths and lengths,scale bar,1 cm;(e)thousand-grain weight in Xinxiang(2015-2016 growing season);(f)yields per plot in Xinxiang(2015-2016 growing season).Values are means with SD.*P ＜0.05, **P ＜0.01.

Most of the currently available triticale is hexaploid due to their superior vigour and reproductive stability[5,7,43-45].To enrich the genetic variability of triticale, wheat or wild relatives can be used to produce new triticale by replacing certain chromosomes of the hexaploid triticale(AABBRR)with the D-genome chromosomes of wheat [5]. Many modern triticale lines that have been developed from such crosses carry D(A),D(B),and D(R)whole chromosome substitutions or chromosome translocations that add valuable traits to triticale [1]. For more extensive exploitation of these modern triticale lines with genetic improvements, detailed information about their genomic constitution is needed.For example,a new T2DS.2DL-?R translocation triticale,ZH-1,with multiple resistances to diseases has been identified. The physical mapping of chromosome T2DS.2DL-?R showed that a minute chromosomal fragment derived from rye was attached to the distal end of 2DL [46]. In this study, we analyzed the genomic constitution of a hexaploid triticale Lin 456 that was derived from the cross of octoploid triticale× hexaploid wheat by GISH, ND-FISH, and PCR-based markers. The GISH results indicated that Lin 456 contained the complete RR genome. The ND-FISH results demonstrated that it is a 6D(6A) substitution line. Specifically, the signal of Oligo-pSc119.2 was detected at the distal end of 6DL in Lin 456,which is obviously different from the results of common wheat [36,37,47,48]. However, 13 molecular markers flanking the entire chromosome 6D further corroborated the 6D(6A) substitution by amplifying the target DNA bands from common wheat (CS) and Lin 456 (Fig. 2).The combined data of the GISH, FISH, and PCR-based markers show that Lin 456 is a 6D(6A) substitution line of hexaploid triticale.

To characterize the genomic constitution during triticale breeding programs,C-banding,in situ hybridization,and PCRbased markers have been exploited with great success.Although C-banding technology can distinguish parental genomes and analyze the genome organization of triticale, it is time consuming and requires a high level of skill[49].PCRbased markers are convenient for identifying the genomic constitution and variations of triticale; however, they cannot show the physical size of the translocations. Compared to Cbanding and PCR-based markers, FISH and GISH have been more useful for detecting alterations of wheat and rye chromosomes. Specifically, FISH can be used to investigate different morphologies of chromosomes and their organization [50], whereas GISH has the added advantage of determining the size and breakpoint of chromosome translocation[51,52]. Recently, oligonucleotide probes have provided an easier, faster, and more cost-effective method for ND-FISH analysis[37].We found that ND-FISH can identify the genome organization of triticale Lin 456 and can show the structural variation of chromosome 6D. When the chromosome constitution of Lin 456 was analyzed by molecular markers, it was evident that the supposed structural variation of chromosome 6D was only a signal alteration at the end of 6DL.Collectively,the combination of GISH, FISH and PCR-based markers is shown to be an ideal method to discriminate rye and wheat chromosomes in triticale.

4.2. Potential utilization of the 6D(6A) substitution line of hexaploid triticale in breeding

Fig.4- Amplification patterns of SSR markers for the F2 population.Lanes 1-3 are Chinese Spring(CS),4100,and Lin 456,respectively.Lanes 4-19 are plants of the F2 population.

Table 2-Comparison of yield traits between substitution and non-substitution subpopulations in the F2 segregating population.

Table 3-Comparison of yield traits between containing and non-containing chromosome 6A subpopulations in the F2 segregating population.

The results from the trials of the CIMMYT International Triticale Yield Nurseries suggest that a complete triticale carrying a 6D(6A) substitution has selective advantages,which may be caused by an intensive breeders’ selection pressure for a desired plant type[27].This could be potentially dangerous to breeding programs in CIMMYT because of narrowing of the gene basis toward lines that all contain 6D(6A)substitutions with similar origins.Here,we characterized a new 6D(6A)substitution of hexaploid triticale obtained from the hybridization of octoploid triticale×hexaploid wheat from China.It can be an important potential donor stock for genetic improvement during breeding due to its superior traits,especially for high grain weight.

The TGW is an important component of the grain yield in cereals. Over the past two decades, the successful application of quantitative genetic methodologies has facilitated the identification of numerous quantitative trait loci (QTL) for TGW in wheat [53-59]. Several major QTL for TGW have been identified on chromosome 6A[54,58,60]. TaGW2-6A, which is an orthologue of the rice OsGW2 gene on chromosome 6A, also controls grain development in wheat [61]. Moreover, in wheat relatives,the 6VS/6AL translocation line was identified to have effects on increasing the TGW and spike length [62]. The intercalary translocation line Pubing3035 with the Agropyron cristatum 6P segment had a higher TGW than common wheat[63].The TGW of Lin 456,which carries a 6D(6A) substitution, was higher than the triticale cultivar 4100, with a range of 19.4-25.5 g in different environments(Table 1). A genetic effect analysis of the F2population revealed that chromosome 6D from Lin 456 had a significantly positive effect on the grain weight (Table 2). By contrast, the F2plants with 6A increased grain width but decreased grain length as compared to those without 6A.As a result, no significant difference was detected for TGW between F2plants with and without chromosome 6A under two tested environments Shanxi and Hebei (Table 3).Collectively, these data indicated the high grain weight of Lin 456 could be partially attributed to the positive genetic effect of 6D as compared to 6A. We found that TGW was negatively correlated with the spikelet number. Consistently, chromosome 6D had a significantly negative effect on spikelet number(Table 2).The spikelet number per spike is one of the major components for grain number per spike.Theoretically, grain number per spike of 6D(6A) substitution should be decreased on account of the decreased spikelet number. As a consequence, the spike was able to provide more assimilate to the grains, which led to high TGW in turn.Therefore,the relationship between TGW and the spikelet/grain number of the 6D(6A) substitution is needed for further elucidation.

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2019.03.007.

Declaration of Competing Interest

The authors declare that they have no conflict of interest.

Acknowledgments

We thank Professor Zongxiang Tang and his postgraduate student Zhiqiang Xiao at Sichuan Agricultural University for their assistance in ND-FISH analysis.This study was supported by the National Key Research and Development Program of China (2017YFD0101004), the National Natural Science Foundation of China (91435204), and the Science and Technology Independent Innovation Ability Upgrading Project of Shanxi Academy of Agricultural Sciences(2017ZZCX-23).