Plant Specific Homeobox TaWUS-like Gene Associated with Thousand Grain Weight and Grain Number in Common Wheat

Abstract

WUSCHEL-related homeobox (WOX) gene is a plant-specific homeobox containing Transcription Factor (TF). WUSCHEL play critical roles in plant developmental events and crop yield. In present study, three copies of TaWUS-like were isolated from chromosomes 5A, 5B and 5D of common wheat. TaWUS-likes were investigated for sequence polymorphism to conduct Marker Trait Association (MTA). Twenty Single Nucleotide Polymorphisms (SNPs) variation and six Inserts and Deletions (InDels) were identified in the promoter (2.0 kb-3.0 Kb) and coding regions among the three copies of TaWUS-like homeologs. For high throughput genotyping, Kompetitive Allele-Specific PCR (KASP) markers were developed. Favored haplotypes were significantly associated with Thousand Grain Weight (TGW) (5B-HapI and 5D-HapI) and Grain Number (GN) (5A-HapI and 5B-HapI) in wheat population collected from China. More sequence polymorphisms presented in TaWUS-like-5B genome than those of TaWUS-like-5A and TaWUS-like-5D genome. Absence of sequence polymorphisms in studied diploid accessions for TaWUS-likes and presence in studied tetra and hexaploid wheat accessions along with polymorphic information content and genetic diversity values suggest that polyploidization events resulted in increased genetic diversity for TaWUS-like-5A while decreased diversity for TaWUS-like-5B and TaWUS-like-5D. Geographic distribution and allelic frequency indicated that the favored haplotypes experienced positive but less intensive selection point towards underutilization of TaWUS-likes in wheat breeding. The developed KASP markers can be instrumental in grain yield improvement by marker assisted selection in wheat breeding programs.

Keywords

Wheat; WOX; Haplotype; Allelic variation; MTA; Yield

Introduction

Breeding for higher grain yield has been a major objective in wheat breeding programs. A comprehensive understanding of genes conferring grain yield would benefit breeding for new cultivars. Wheat plays crucial role in ensuring global food security and doubling the wheat production by 2050 due to population surge is a big challenge [1,2]. Thousand Grain Weight (TGW) and Grain Number (GN) are considered as important yield contributing parameters. Variable correlations between TGW and GN have been reported, which mainly depends upon population [3,4]. In Chinese modern wheat cultivars, GN and TGW have positive correlation between them [5]. Several genes related to grain yield have been cloned in wheat and their Functional Markers (FMs) have been developed.

Marker Assisted Selection (MAS) has accelerated wheat breeding process [6]. As breeding science progressed, a better understanding of causal loci has increased breeders control over complex traits [7]. In comparison with conventional markers such as RFLP, AFLP, RAPD, ISSR and SSR, Single Nucleotide Polymorphism (SNP) based markers are more valuable due to its stability, cost effectiveness, high abundance and high-throughput scoring [8]. Moreover, SNP based markers are reported to be very effective in association analysis, genetic diversity, FMs development and MAS breeding [9-13]. Association analysis is an efficient tool used to identify marker trait association of target gene and has been widely reported in many plants including wheat and rice [11,14].

WUSCHEL-related homeobox gene is a plant specific clade of homeobox-containing Transcription Factors (TF). Increasing evidence reveal that WUSCHEL Transcription Factor plays an important role in the whole process of plant growth and development from maintenance of stem cells at meristem to the formation of embryo [15,16]. The role of WUSCHEL genes in plant development has been studied in detail in Arabidopsis thaliana, corn and rice. Till now, only few reports are available on the regulation of yield-related traits by WUSCHEL-related TF family in gramineous crops. Zhang, et al. found that two functionally redundant TF genes, WUSCHEL-related homeobox 6 (WOX6) and WOX11, are expressed asymmetrically in response to auxin to connect gravitropism responses with the control of rice tiller angle.

These approaches will identify genes to improve grain yields by facilitating the optimization of plant architecture [17]. Map-based cloning revealed that DWT1 encodes a WUSCHEL-related homeobox TF homologous to the Arabidopsis WOX8 and WOX9 and highly expressed in young panicles. The dwt1 mutant plants develop main shoots with normal height and larger panicles, but dwarf tillers bear smaller panicles as compared to those of the wild-type [18]. In case of wheat, gene sequence, structure and function of TaWUS-like genes are still limited.

In this study, we reported one WUSCHEL-related homeobox TF gene TaWUS-like, which were associated with TGW and GN in common wheat. The objectives of this study were to (i) to identify polymorphism sites and develop breeder friendly high throughput KASP marker(s) for TaWUS-like-5A, TaWUS-like-5B and TaWUS-like-5D, (ii) to identify favorable allelic variation(s), (iii) to study the frequency and geographic distribution of allelic variation(s) among Chinese landraces and Chinese modern cultivars, European, Pakistani, CIMMYT, Russian and North American wheat cultivars. The purpose of this study is to provide a scalable, flexible, high throughput and gel free molecular marker to assist ongoing marker assisted breeding in wheat.

Materials and Methods

Plant materials and growth conditions

Eight wheat germplasm populations were used as plant material to analyze the allelic variation in TaWUS-like-5A, TaWUS-like-5B and TaWUS-like-5D (Supplementary Table 1). A set of 348 Chinese modern wheat cultivars (Population-I) was phenotypes for grain-related traits including Thousand Grain Weight (TGW), Grain Thickness (GT), Grain Width (GW) and Grain Length (GL) and spike-related traits including Grain Number per spike (GN), Spikelet Number per spike (SN) and Spike Length (SL), Plant Height (PH), Heading Date (HD), Maturity Date (MD), Effective Tiller Number (ETN) were recorded in three environments in 2002, 2005 at Luoyong (112º45’E; 36º61’N) in Henan province and 2010 at Shunyi (116º56’E; 40º23’N) in Beijing. Randomized complete block design was followed in triplicates. Each accession was planted in four rows (plot size 2 meters). Forty seeds were planted in each row (row distance 25 centimetres). Ten plants from the middle of each plot were used to phenotype the aforementioned traits.

Table 1. Diversity at each locus in Chinese and non-Chinese wheat germplasm.

Another a set of 157 Chinese wheat landraces (Population-II) from Chinese Mini-core Collection (MCC), 336 European wheat cultivars (Population-III), 153 Pakistani wheat accessions (Population-IV), 53 CIMMYT wheat accessions (Population-V), 83 Russian wheat accessions (Population-VI) and 429 North American wheat accessions (Population-VII) were used to investigate the global geographic distribution of TaWUS-like-5A, TaWUS-like-5B and TaWUS-like-5D allelic variations and to evaluate the utilization value of the newly developed molecular markers in wheat breeding programs.

Chromosomal localization of TaWUS-likes gene

To identify the chromosomal location of target gene, the genomic sequences were used as queries to perform BLAST online (https://urgi.versailles.inra.fr/blast/blast.php). Chromosomal location was further confirmed by PCR amplification in diploid, tetraploid, hexaploid wheat species, nulli-tetrasomic, and ditelosomic lines of Chinese Spring (CS) wheat using TaWUS-like-5A-F/R, TaWUS-like-5B-F/R and TaWUS-like-5D-F/R primers (Supplementary Table 2).

A set of 16 diploid accessions for A subgenome, 12 diploid accessions for B subgenome, 11 diploid accessions for D subgenome and 21 tetraploid accessions for AABB genome were used to study genetic diversity for TaWUS-likes (Supplementary Table 3).

DNA isolation and sequencing of TaWUS-like-5A/TaWUS-like-5B/TaWUS-like-5D

DNA of 34 highly diverse polymorphic wheat accessions (Population-VIII) was extracted from young leaves with a DNA quick plant system (TIANGEN Biotech, Beijing, China) and dissolved in sterile water. A pair of primers for each TaWUS-like copy was selected to amplify the genomic sequences of TaWUS-like-5A, TaWUS-like-5B and TaWUS-like-5D (Supplementary Table 2).

KOD DNA polymerase was used for PCR amplification. PCR was performed in a total volume of 20 µL having 10 µL of KOD buffer, 1.2 µL of dNTPs (2.5 mM for each nucleotide), 0.2 µL of KOD FX enzyme, 0.4/0.4 µL forward and reverse primers (10 µM), 1 µL DNA (25 ng µl-1) and 6.8 µL of ddH2O. PCR conditions are given in Supplementary Table 2. PCR product was checked on agarose gel (1.2%) and desired bands were isolated and purified using AXYGEN DNA purifying kit, followed by cloning into pEASY®-Blunt cloning vector (TransGen Biotech, Beijing, China), then was transformed to 33 µL trans1-T1 phage resistant chemically competent cells (TransGen Biotech, Beijing, China) by heat shock. Ten positive clones for each sample were selected for sequencing by DNA analyzer 3730xl. To get full length desired sequences of TaWUS-like-5A, TaWUS-like-5B and TaWUS-like-5D, M13 forward/reverse primers and respective sequence walking primers were used (Supplementary Tables 2 and 3).

The sequence of each clone was obtained by assembling with SeqMan program in DNASTAR Lasergene V.7.1.0 software package. The genomic origin of each sequence was firstly confirmed by comparing with reference genomic sequence obtained from URGI (https://urgi.versailles.inra.fr/blast/blast.php). The gene structure of TaWUS-like-5A, TaWUS-like-5B and TaWUS-like-5D was determined using MegAlign (DNASTAR Lasergene 7.1.0) through aligning coding and genomic sequences.

Phylogenetic tree analysis

In order to investigate the evolutionary relationship of TaWUS-like gene, a BLAST search was performed in the ENSEMBL database (http://asia.ensembl.org/index.html) based on TaWUS-like gene ID. The amino acid sequences of TaWUS-like proteins of the targeted species were downloaded, followed by the construction of phylogenetic tree of TaWUS-like proteins from a complete alignment of 36 TaWUS-like protein sequences. Neighbor-joining method with 1000 bootstrap replicates and p-distance substitution model using MEGA 7.0 (https://www.megasoftware.net/) was used to construct the phylogenetic tree.

Functional marker development

Population-VIII was used to detect polymorphisms in TaWUS-like-5A, TaWUS-like-5B and TaWUS-like-5D. For KASP marker development and high throughput genotyping, five gel free KASP markers were developed on selected SNP sites by following standard KASP guidelines (http://www.lgcgenomics.com). KASP-5A-1126 at -1126 bp and KASP-5A-764 at -764 bp were developed on the SNP sites in TaWUS-like-5A upstream region. KASP-5B-883 at 883 bp and KASP-5B-1672 at 1672 bp were developed on the selected SNP sites in TaWUS-like-5B coding region. KASP-5D-226 at 226 bp was developed for solitary SNP in TaWUS-like-5D downstream region. Detailed information about KASP markers is given in Supplementary Table 4.

Moreover, the developed KASP markers were applied across the all seven wheat populations. In general, KASP mixture consisted of 30 µL common primers (100 µM), 12 µL of each tailed primer (100 µM) and 46 µL ddH₂O. KASP assays were tested in 5 µL PCR reaction mixture. Each reaction contains 2.4 µL DNA (25 ng/µL), 2.4 µL of 2X KASP master mixture, 0.04 µL MgCl₂/DMSO, 0.06 µL primer mixture and 0.1 µL ddH₂O. Fluorescence levels were detected by QuantStudioTM Flex (Applied Bio systems by Life Technologies) and data were visualized by using QuantStudioTM Real-time PCR software v.1.3 (Applied Biosystems by Life Technologies).

Statistical analysis

DNAMAN (9.0, Lynnon, Quebec, Canada) was used to compare amino acid sequences. Descriptive statistics, one-way ANOVA and Tukey test was performed for variance and significance analysis using SPSS (21.0, Chicago, USA).

Association analysis was used to check marker trait association in the recorded data set. Associations were considered statically significant at p<0.05. The effects of allelic variations on each trait were analysed by Student’s t-test at p<0.05 (even 0.01). Allelic frequencies were calculated in all seven wheat populations. Polymorphic Information Content (PIC) and gene diversity (He) were calculated using https://www.gene-calc.pl/pic website.

Results

Chromosomal location of TaWUS-likes

To investigate the function of TaWUS-like in wheat, three copies of genomic sequences of TaWUS-like obtained from Chinese Spring (CS). All three homoeologous gene of TaWUS-likes were composed of two exons and one intron located at base pairs 476 bp-1422 bp, 485 bp-1422 bp, 494 bp-1437 bp respectively, and encodes putative protein products of 318, 321, and 322 amino acids. These proteins had a high identity of 94.75% and all contained a homeobox domain. The amino acid sequence similarity of TaWUS-like-5A and TaWUS-like-5B is 89.10% and that between TaWUS-like-5A and TaWUS-like-5D is 92.90% and that between TaWUS-like-5B and TaWUS-like-5D is 91.36%. But the homology with rice, maize and Arabidopsis thaliana are low, about 29% (Figure 1).

A phylogenetic tree was constructed with the putative amino acid sequences of TaWUS-like members form wheat and other species. The results were in consistence with previous studies and as anticipated the three copies of TaWUS-like gene belong to the WUS branch of the WUSCHEL-related TF superfamily [19]. The deduced amino acid sequences showed a high homology with WUSCHEL-related TF family members of the monocotyledonous plants such as LOCOs01g62310, Zm0001d042920 and relatively lower homology with dicotyledonous species for example AT5G59340 (Figure 2A). Scanning site analysis indicated that TaWUS-like-5A/5B/5D have conserved homeobox domain (Figure 1).

BLAST, nulli-tetrasomic and ditelomsoic lines showed that TaWUS-likes were located on short arm of (chromosome 5A, 5B and 5D), hence were named as TaWUS-like-5A, TaWUS-like-5B and TaWUS-like-5D accordingly (Figure 2B). The amplified fragment sizes of TaWUS-like-5A/5B/5D were 3507 bp, 2983 bp and 3076 bp, respectively.

TaWUS-like-5A-HapI was significantly correlated with GN

Four allelic variations were identified in the upstream region of TaWUS-like-5A. Based on the two selected SNPs (-1126 bp and -764 bp) (Figure 3A), KASP markers named KASP-5A-1126 and KASP-5A-764 Figure 3B were developed to distinguish TaWUS-like-5A-HapI and TaWUS-like-5A-HapII.

Population-I was screened by the newly developed functional markers. Association analysis revealed that, 5A-HapI haplotype was significantly associated with higher GN, which is 3% more GN than 5A-HapII (p<0.05 or p<0.01) (Figure 3C). Wheat accession containing 5A-HapI also showed shorter PH as compared to 5A-HapII, which is 5.7% lower than 5A-HapII (p<0.01) (Figure 3D). In addition, 5A-HapII was the most frequent haplotype available in 80% accessions of Chinese modern cultivars.

TaWUS-like-5B-HapI associated with TGW and GN

Nineteen polymorphic sites were identified in the genomic region of TaWUS-like-5B (Figure 4A). Among the identified polymorphic sites, 14 are located in the promoter region, while four were identified in coding region and one in downstream region, forming two haplotypes. On the basis of the selected polymorphic site (at 883 bp and 1672 bp), KASP markers were developed to distinguish TaWUS-like-5B-HapI and TaWUS-like-5B-HapII (Figure 4B).

The newly developed KASP markers were then applied to Population-I. In Population-I, 5B-HapI was most frequently occurring (75%) haplotype, followed by 5B-HapII (25%). The association analysis showed that 5B-HapI was significantly associated with higher TGW (p<0.05) (Figure 4C) and GN (p<0.05 or p<0.01) (Figure 4D). Wheat accession possessing 5B-HapI had 3% more TGW than 5B-HapII, respectively.

TaWUS-like-5D-HapI associated with TGW

Two SSR loci in the upstream and a solitary SNP (226 bp) Figure 5A was identified in the downstream region of TGA (stop codon) of TaWUS-like-5D forming two haplotypes. KASP marker (5D-226) was developed on the SNP site (Figure 5B).

Population-I was genotyped by using the newly developed KASP marker. Significant differences were recorded between the two haplotypes in three environments. Association analyses showed that genotypes possessing 5D-HapI have higher TGW (3.4% higher) and lower PH (9.3% lower) as compared to genotypes possessing 5D-HapII (Figures 5C and 5D). In addition, 5D-HapI haplotypes showed early maturity compared with 5D-HapII (p<0.05) in two environments (Supplementary Figure 1).

Geographic distribution of TaWUS-like-5A/5B/5D haplotypes

To determine, whether the favored allelic variations of TaWUS-like-5A/5B/5D were selected in wheat breeding, we investigated the geographic distribution of favorable allelic variations in China, Pakistan, Europe, Russia, North America and CIMMYT wheat germplasm.

On the basis of wheat, China is divided into 10 agro-ecological zones [20,21]. In Chinese wheat landraces (Population-II), frequency of favored TaWUS-like-5A-HapI was very low (3.2%) and TaWUS-like-5A-HapII was the pre-dominant haplotype (96.8%) in all major wheat growing regions (Figure 6A). In Chinese modern wheat cultivars (Population-I), a significant increment (21%) in TaWUS-like-5A-HapI frequency took place (Figure 6B). Although positive selection of favored haplotype took place, the selection pressure was not as strong as we anticipated (Figures 6A and 6B). For TaWUS-like-5B, the frequencies of favored HapI were less (23.6%) in Population-II (Figure 6C). Remarkable increment (up to 75%) was observed in the Population-I Figure 6D, suggesting strong breeding selection of these allelic variations. For TaWUS-like-5D, the proportions of favored HapI were still lower (11.5%) in Chinese landraces (Figure 6E). However, different frequency changes occurred in Population-I (modern cultivars). As Figure 6F indicated, the frequency of the favored 5D-HapI showed an obvious increment in Chinese wheat zone-II, wheat zone-IV, wheat zone-VI, wheat zone-X regions, but a dramatic decline in zone-IX, a slight decrease in zone-VII and zone-III, and the others almost kept the same frequency. In China, wheat zone-I, wheat zone-II, wheat zone-III, wheat zone-IV and wheat zone-VIII are the major wheat production area, holding about 85% proportion of wheat production, especially for the wheat zone-II, almost holds 45%-50% proportion of the wheat production. In addition, the main breeding target of the modern cultivars in wheat zone-IV focus on grain size and high TGW. Thus, in these TGW targeted major wheat regions, breeders still prefer the favored haplotypes of TaWUS-like-5D-HapI, suggesting TaWUS-like-5D-HapI positively contribute to yield and thus has underwent breeding selections in China. However, as a whole, the favored haplotypes still remain very low in Chinese modern cultivars (10.6%) (Figure 6F).

Among non-Chinese regions i.e. Pakistan, Europe, Russia, North America and CIMMYT wheat germplasm, favored haplotypes of TaWUS-like-5A/TaWUS-like-5D remain in lower frequencies except for 5B-HapI which were apparently selected (Figures 7A-7C). However, the favored haplotypes of TaWUS-likes all showed a positive breeding pressure though the selection pressure on TaWUS-like-5A/TaWUS-like-5D haplotypes among Chinese and non-Chinese wheat germplasm differed greatly in degree. Overall, lower favorable allelic frequencies indicate underutilization of these favored haplotypes both in Chinese and non-Chinese wheat breeding programs.

Domestication and breeding evolution of TaWUS-likes in global wheat germplasm

To survey the evolutionary history of TaWUS-likes, we analyzed the TaWUS-likes polymorphism degree in wheat progenitor accessions, hexaploid landraces (CLR) and hexaploid modern cultivars (CMC). Results showed that the PIC of TaWUS-like-5B (0.620) in CLR was much higher than those in TaWUS-like-5D (0.183) and TaWUS-like-5A (0.06) Table 1, indicating a higher SNP polymorphism or gene diversity in TaWUS-like-5B genome than the ones in TaWUS-like-5D and TaWUS-like-5A genome. Similar results obtained in He analysis. Interestingly, during polyploidization events, gene diversity increased in TaWUS-like-5A (PIC from 0.06 in CLR to 0.277 in CMC), while decreased in TaWUS-like-5B (PIC from 0.620 in CLR to 0.577 in CMC) and TaWUS-like-5D (PIC from 0.183 in CLR to 0.172 in CMC) Table 1, showing a slight breeding pressure of TaWUS-like-5D and TaWUS-like-5A in Chinese breeding process. However, during domestication process, gene diversity of TaWUS-like-5A reduced dramatically (PIC from 0.269 in AABB to 0.06 in AABBDD), while the TaWUS-like-5B increased obviously (PIC from 0.164 in AABB to 0.620 in AABBDD), similar results also obtained in He analysis (Supplementary Table 5). These observations suggest that TaWUS-like-5A/TaWUS-like-5B were all subject to intense historic selection but at different evolution stages.

Genotype frequency of TaWUS-likes in Chinese wheat breeding

To evaluate the allelic variation frequencies of TaWUS-likes, we partitioned the studied wheat diversity panel into seven sub-groups by 10 years, namely 1930s, 1940s, 1950s, 1960s, 1970s, 1980s and 1990s. Results showed that the frequencies of TaWUS-like-5A/TaWUS-like-5B/TaWUS-like-5D favorable allelic variations showed slow but increasing trend from 1930s to 1990s (Figures 8A-8C), however, the frequency proportion of favorable allelic variations was still very low for TaWUS-like-5A/TaWUS-like-5D in Chinese modern cultivars (Population-I) compared that with the unfavorable haplotypes. According to Population-I, the average TGW and GN exhibited continuous increasing trend from pre-1960s to 1961-2000s, while PH exhibited decreasing trend from pre-1960s to 1961-2000s (Figure 8D). These outcomes suggest that favorable allelic variations of TaWUS-likes are valuable and could be selected to improve wheat grain yield.

Discussion

Resequencing small-scale core wheat accessions combined by large-scale KASP genotyping provides effective strategy for identifying superior alleles and marker-assisted breeding in wheat

Numerous studies have proved that Marker-Assisted Selection (MAS) is vital for speeding up genetic gain to get high prediction rate of economically important traits and develop new germplasmin crops, therefore, marker based breeding is of utmost important for global food security in upcoming times [22]. TaWUS-like-5D, an important TGW-related gene cloned by our lab, regulate sucrose metabolism and utilization in grain endosperm therefore affecting wheat yield greatly (unpublished data). Moreover, TaWOX-like-5D plays a vital role in wheat architecture due to its residual meristem functions [23]. However, the genome allelic variations and evolutionary history of TaWUS-like-5D and its homoeologs of TaWUS-like-5A and TaWUS-like-5B, were still not characterized yet, thus limited the wide application of TaWUS-likes gene in yield improvement and wheat breeding. In this study, we used resequencing combined by KASP strategy to systematically characterize the allelic variations of TaWUS-likes homoeologs.

Genotyping By Sequencing (GBS) has emerged as an effective and accurate breeding strategy to identify genomic sequence polymorphism in different crop plants such as rice, maize, sorghum, and wheat [24-28]. However, in wheat, conducting GBS in a larger wheat population (>50 accessions) still is cost and time-consuming and technologically challenging due to the wheat higher ploidy level (3 sub-genome) and high proportion of repeat sequences (>85%) [11]. Thus, exploit or combination of the other high-throughput cost-effective molecular markers is also vital for identifying or MAS of superior alleles in wheat [29,30]. KASP is such a uniplex and flexible genotyping platform that utilizes a unique form of competitive allele-specific PCR combined with a novel, homogeneous, fluorescence-based reporting system for identification and measurement of genetic variation occurring at the nucleotide level to detect SNPs or Inserts and Deletions (InDels), moreover, KASP enables high throughput with a time- and cost-effective way [31,32]. KASP assay has been proved to be successively utilized in the genetic improvement of tomato and field crops [33-35]. Therefore, in this study, we first used a diverse set of 34 wheat accessions from China representing 70% diversity in Chinese wheat germplasm and accurately identified twenty SNPs and six InDels across the three TaWUS-like homoeologous genes by GBS, then grouped these variations into haplotypes and conduct the Marker Traits Associations (MTAs), finally five pairs of KASP markers were developed to distinguish TaWUS-like allelic variations and haplotypes in a larger wheat population. We found these KASP markers can easily genotype thousands of wheat germplasm (1347 wheat germplasm) within very short periods (96-well plates, 3 days-5 days).

Moreover, the developed KASP markers can also be used to fingerprint that articulates TaWUS-like in any large breeding populations. Therefore, the application of KASP analysis in molecular breeding could significantly improve wheat breeding efficiency. We suggested that KASP markers can also be utilized in combination with other functional markers such as Cleaved Amplified Polymorphic Sequences (CAPS) and Simple Sequence Repeats (SSR) etc; developed for yield related traits. Thus, GBS (a small representative wheat set) combined by KASP genotyping assay could be a valuable method to be used in any large wheat breeding population.

Haplotype based MTAs are more informative than bi-allelic SNPs in wheat

In this study, we used haplotype not SNP to investigate MTAs. That because haplotype-based MTAs are more informative than bi-allelic SNPs [36]. Haplotype data can not only capture the associations that elude identification by solitary SNPs but also the epistatic interactions between SNPs. Hence, haplotype-based MTAs could increase prediction accuracies. We found the favored haplotypes of TaWUS-like-5A-HapI/TaWUS-like5B-HapI/TaWUS-like-5D-HapI were significantly associated with Thousand Grain Weight (TGW) (5B-HapI and 5D-HapI) and Grain Number (GN) (5A-HapI and 5B-HapI) in wheat population, accurately predict 61%-78% genotype among Chinese modern cultivars not considering the genetic effects of the other yield related genes. Spring wheat collection from Pakistan (153 accessions) also showed positive selection trend of favored haplotypes of TaWUS-likes, suggesting haplotype based association are more effective than SNP thus these haplotypes can be instrumental to improve wheat grain yield in breeding.

More sequence polymorphisms presented in TaWUS-like-5B genome than those of TaWUS-like-5A and TaWUS-like-5D genome

Higher PIC and He observed were in the TaWUS-like-5B genome both in hexaploid CLR and in hexaploid CMC, suggesting the presence of more frequent sequence polymorphisms in TaWUS-like-5B gene than those of TaWUS-like-5A and TaWUS-like-5D genes. Similar results were obtained by Hao, et al. through resequencing 145 wheat accessions and identifying the polymorphisms in a sub-genome scale [37]. It is well known that wheat B sub-genome has relatively broader genetic base compared to A sub-genome and D sub-genomes, therefore more sequence polymorphisms in TaWUS-like-5B is probably due to the diverse Aegilops speltoides (the cross-pollinator ancestors for Triticum aestivum L.) natures. In addition, Triticum aestivum A and Triticum aestivum D sub-genome donors were self-pollinators that might be another reason for less sequence polymorphism for TaWUS-like-5A/TaWUS-like-5D compared to TaWUS-like-5B [38].

Favored haplotypes of TaWUS-likes experienced positive but less intensive selection point towards underutilization of TaWUS-likes in wheat breeding

Before 1960s, wheat disease resistance was the breeder’s main choice trait in China. At 1960s-1980s, the breeding objective was changed to breed for yield traits such as Plant Height, TGW, GN and spike number etc. Post 1990s, the objective turned to high yield and quality [37]. The changes of Chinese breeding result in the selection of favorable allelic variations in different genes. In this study, from Chinese wheat landraces to modern cultivars, the proportion of favored haplotypes of TaWUS-like homoeologs showed slight increment (Figures 8A-8C). A rapid increase of TaWUS-like-5A-HapI occurred from 1940s-1950s, whereas TaWUS-like-5D-HapI/TaWUS-like-5B-HapI showed a gradually increasing trend since 1950s/1930s, indicating favored haplotypes of TaWUS-likes experienced positive breeding selection but at different breeding stages. We assumed that, based on the haplotype trait association analysis, the selected traits of TaWUS-like-5A, TaWUS-like-5B and TaWUS-like-5D genes might be lopsidedly correlated with Plant Height, GN and TGW respectively. This concept can be supported by the bias selection of TaWUS-likes at different evolution stage and by different selection degree according to the PIC and H_e analysis. In addition, we have transferred the TaWUS-like-5D into wheat background, and found that OE of TaWUS-like-5D resulted in less TGW and more serious disease symptom than the wild-type wheat receptor cultivar KN199 under field conditions (unpublished data), thus we do not exclude the possibility of TaWUS-like-5A was also contributed to adaptation traits such as disease resistance or flowering (meristem functions like TaWUS-like-5D) since their amino acid sequence was very conserved therefore be selected strongly at wheat domestication periods.

Interestingly, during polyploidization events (CLR to CMC), gene diversity increased dramatically in TaWUS-like-5A, while TaWUS-like-5B and TaWUS-like-5D decreased slightly, showing a less breeding pressure for TaWUS-like-5B and TaWUS-like-5D in Chinese breeding process except TaWUS-like-5A. Higher He for TaWUS-likes-5A in Chinese and global modern wheat cultivars may be illustrated that conventional breeding by artificial hybridization has increased H_e of TaWUS-like-5A. Since the adaptation trait of TaWUS-like-5A preferred in worldwide breeding, therefore the usage of wide geographic sources of germplasm by present-day breeders has not reduced the overall genetic diversity of TaWUS-like-5A. Zhao, et al. have reported that the diversity of functional gene increased gradually during successive decades of wheat breeding [13]. This concept, supported by our different association roles of TaWUS-likes homoeologs in Chinese modern cultivars, highlights the possibility to avert the narrowing of genetic diversity by introgression of diverse germplasm in wheat breeding program.

Accessions from Pakistan, Europe, North America, Russia and CIMMYT were also used to evaluate TaWUS-likes haplotypes in five different geographical regions. In these regions, favored haplotypes experienced positive selection but the selection intensity was not as strong as we anticipated except for 5B-HapI (Figures 7A-7C), indicating 5B-HapI has undergone more strong selections than 5A-HapI and 5D-HapI both in China and in global. Given that the favored haplotypes of TaWUS-likes, experienced less positive selection intensity in Chinese and non-Chinese studied wheat breeding history clearly indicates the underutilization in breeding. Although, the studied wheat germplasm has different population structures, still the unconscious selection of the favored haplotypes is likely due to the high linkage disequilibrium for major yield related genes selected during selection breeding. For example, we observed that both haplotypes of all TaWUS-like genes contributed equally for TGW trait in Pakistani wheat program. However, in China, the increase of favored haplotypes in 5A and 5B was more apparent than the one in 5D in major Chinese wheat production areas (Zone-I, Zone-II and Zone-III). Therefore, these observations suggest that TaWUS-like-5A/TaWUS-like-5B/TaWUS-like-5D were all subject to positive breeding selection but at different grain yield traits and by different degree among Chinese and global germplasm.

Conclusion

We identified allelic variations in TaWUS-like homoeologs among seven geographically different wheat germplasm. The favored haplotype of TaWUS-like-5B has been fully selected and utilized in global, whereas TaWUS-like-5A and TaWUS-like-5D could be further utilized in yield improvement breeding both in China and global. Association analysis showed that the favored haplotypes of TaWUS-like homoeologs were significantly associated with grain yield related traits. This evidence indicated that TaWUS-likes maybe involved in grain or spike development. These functional molecular markers can be used in future studies to assess their usefulness as selection criteria for improving grain yield related traits.

Authors Contribution

H.L. conceived the project, performed paper drafting (introduction and discussion section) and revised the whole manuscript; X.S., S.R. and Y.W. performed experiments, data interpretation and paper drafting (results section); J.H., T.L., C.H. and X.Z. provided valuable suggestions and comments. All authors have read and agreed to the published version of the manuscript.

Funding

The research was supported by the National Natural Science Foundation of China (No. 31471492 and 91935304), the National Key Research and Development Program of China (2016YFD0100402), Innovation Project of the Institute of Chinese Academy of Agricultural Sciences. and National Crop Genomics and Speed Breeding Centre for Agriculture Sustainability-ADP-2021-22 of Pakistan (Lo21002838).

Institutional Review Board Statement

Not applicable.

Informed Cosent Statement

Not applicable.

Data Availability Statement

Supporting information is available from the Wiley Online Library or from the author.

Ethical Approval and Cosent to Participate

We declare that these experiments comply with the current laws and ethical standards of China.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

1. Ju L, et al. JAZ proteins modulate seed germination through interaction with ABI5 in bread wheat and Arabidopsis. New Phytol. 2019;223:246-260.

   [Crossref][Google scholar][EBSCO]

2. Zhang L, et al. Identification of a novel ERF gene, TaERF8, associated with Plant Height and yield in wheat. BMC Plant Biol. 2020;20:1-12.

   [Crossref][Google scholar][EBSCO]

3. Liu H, et al. Identification and validation of quantitative trait loci for kernel traits in common wheat (Triticum aestivum L.). BMC Plant Biol. 2020;20:529.

   [Crossref][Google scholar][EBSCO]

4. Jamil M, et al. Genome-wide association studies of seven agronomic traits under two sowing conditions in bread wheat. BMC Plant Biol. 2019;19:1-18.

   [Crossref][Google scholar][EBSCO]

5. Zhang D, et al. Identifying loci influencing grain number by microsatellite screening in bread wheat (Triticum aestivum L.). Planta. 2012;236:1507-1517.

   [Crossref][Google scholar][EBSCO]

6. Rasheed A, et al. From markers to genome-based breeding in wheat. Theor Appl Genet. 2019;132:767-784.

   [Crossref][Google scholar][EBSCO]

7. Ramstein GP, et al. Breaking the curse of dimensionality to identify causal variants in breeding 4. Theor Appl Genet. 2019;132:559-567.

   [Crossref][Google scholar][EBSCO]

8. Tong J, et al. High resolution genome wide association studies reveal rich genetic architectures of grain zinc and iron in common wheat (Triticum aestivum L.). Front Plant Sci. 2022;13:840614.

   [Crossref][Google scholar][EBSCO]

9. Rehman SU, et al. Development and exploitation of KASP assays for genes underpinning drought tolerance among wheat cultivars from Pakistan. Front Genet. 2021;12:684702.

   [Crossref][Google scholar][EBSCO]

10. Zhuang M, et al. The wheat short root length 1 gene TaSRL1 controls root length in an auxin-dependent pathway. J Exp Bot. 2021;72:6977-6989.

    [Crossref][Google scholar][EBSCO]

11. Li A, et al. Wheat breeding history reveals synergistic selection of pleiotropic genomic sites for plant architecture and grain yield. Mol Plant. 2022;15:504-519.

    [Crossref][Google scholar][EBSCO]

12. Yang J, et al. Cloning, characterization of TaGS3 and identification of allelic variation associated with kernel traits in wheat (Triticum aestivum L.). BMC Genet. 2019;20:98.

    [Crossref][Google scholar][EBSCO]

13. Zhao J, et al. Global status of 47 major wheat loci controlling yield, quality, adaptation and stress resistance selected over the last century. BMC Plant Biol. 2019;19:5.

    [Crossref][Google scholar][EBSCO]

14. Liu J, et al. The conserved and unique genetic architecture of kernel size and weight in maize and rice. Plant Physiol. 2017;175:774-785.

    [Crossref][Google scholar][EBSCO]

15. Leibfried A, et al. WUSCHEL controls meristem function by direct regulation of cytokinin-inducible response regulators. Nature. 2005;438:1172-1175.

    [Crossref][Google scholar][EBSCO]

16. Jha P, et al. WUSCHEL: A master regulator in plant growth signaling. Plant Cell Rep. 2020;39:431-444.

    [Crossref][Google scholar][EBSCO]

17. Zhang N, et al. A core regulatory pathway controlling rice tiller angle mediated by the lazy1-dependent asymmetric distribution of auxin. Plant Cell. 2018;30:1461-1475.

    [Crossref][Google scholar][EBSCO]

18. Wang W, et al. Dwarf tiller1, a WUSCHEL-related homeobox transcription factor, is required for tiller growth in rice. PLoS Genet. 2014;10:e1004154.

    [Crossref][Google scholar][EBSCO]

19. Li M, et al. Genome-wide identification and analysis of the WUSCHEL-related homeobox (WOX) gene family in allotetraploid Brassica napus reveals changes in WOX genes during polyploidization. BMC Geno. 2019;20:317.

    [Crossref][Google scholar][EBSCO]

20. Hou J, et al. Global selection on sucrose synthase haplotypes during a century of wheat breeding. Plant Physiol. 2014;164:1918-1929.

    [Crossref][Google scholar][EBSCO]

21. He ZH, et al. A history of wheat breeding in China. J Comp Neurol. 2001;523:805-813.

    [Google scholar]

22. Zhang X, et al. TaCol-B5 modifies spike architecture and enhances grain yield in wheat. Science. 2022;376:180-183.

    [Crossref][Google scholar][EBSCO]

23. Si X, et al. A sheathed spike gene, TaWUS-like inhibits stem elongation in common wheat by regulating hormone levels. Int J Mol Sci. 2021;22:11210.

    [Crossref][Google scholar][EBSCO]

24. Spindel J, et al. Genomic selection and association mapping in rice (Oryza sativa): Effect of trait genetic architecture, training population composition, marker number and statistical model on accuracy of rice genomic selection in elite, tropical rice breeding lines. PLoS Genet. 2015;11:1004982.

    [Crossref][Google scholar]

25. Wang N, et al. Applications of Genotyping-By-Sequencing (GBS) in maize genetics and breeding. Sci Rep. 2020;10:16308.

    [Crossref][Google scholar][EBSCO]

26. Morris GP, et al. Population genomic and genome-wide association studies of agroclimatic traits in sorghum. Proc Natl Acad Sci. 2013;110:453-458.

    [Crossref][Google scholar][EBSCO]

27. Alipour H, et al. Genotyping-By-Sequencing (GBS) revealed molecular genetic diversity of Iranian wheat landraces and cultivars. Front Plant Sci. 2017;8:1293.

    [Crossref][Google scholar][EBSCO]

28. Singh N, et al. In-silico detection of aneuploidy and chromosomal deletions in wheat using genotyping-by-sequencing. Plant Meth. 2020;16:1-6.

    [Crossref][Google scholar][EBSCO]

29. Jang YR, et al. High-throughput analysis of high-molecular weight glutenin subunits in 665 wheat genotypes using an optimized MALDI-TOF-MS method. 3 Biotech. 2021;11:1-8.

    [Crossref][Google scholar][EBSCO]

30. Lee SB, et al. A rapid, reliable RP-UPLC method for large-scale analysis of wheat HMW-GS alleles. Molecul. 2021;26:6174.

    [Crossref][Google scholar][EBSCO]

31. Jiang P, et al. Linkage and association mapping and kompetitive allele-specific PCR marker development for improving grain protein content in wheat. Theor Appl Genet. 2021;134:3563-3575.

    [Crossref][Google scholar][EBSCO]

32. Makhoul M, et al. Overcoming polyploidy pitfalls: A user guide for effective SNP conversion into KASP markers in wheat. Theor Appl Genet. 2020;133:2413-2430.

    [Crossref][Google scholar][EBSCO]

33. Devran Z, et al. Development of molecular markers for the Mi-1 gene in tomato using the KASP genotyping assay. Hortic Environ Biotechnol. 2016;57:156-160.

    [Crossref][Google scholar][EBSCO]

34. Semagn K, et al. Single nucleotide polymorphism genotyping using Kompetitive Allele Specific PCR (KASP): Overview of the technology and its application in crop improvement. Mol Breed. 2014;33:1-14.

    [Crossref][Google scholar][EBSCO]

35. Rasheed A, et al. Development and validation of KASP assays for genes underpinning key economic traits in bread wheat. Theor Appl Genet. 2016;129:1843-1860.

    [Crossref][Google scholar][EBSCO]

36. Stephens JC, et al. Haplotype variation and linkage disequilibrium in 313 human genes. Science. 2001;293:489-493.

    [Crossref][Google scholar][EBSCO]

37. Hao C, et al. Resequencing of 145 landmark cultivars reveals asymmetric sub-genome selection and strong founder genotype effects on wheat breeding in China. Mol Plan. 2020;13:1733-1751.

    [Crossref][Google scholar][EBSCO]

38. Marcussen T, et al. Ancient hybridizations among the ancestral genomes of bread wheat. Science. 2014;345:1250092.

    [Crossref][Google scholar][EBSCO]