Genetic Improvement and Application Practices of Synthetic Hexaploid Wheat

Synthetic hexaploid wheat (SHW) is a useful genetic resource that can be used to improve the performance of common wheat by transferring favorable genes from a wide range of tetraploid or diploid donors. From the perspectives of physiology, cultivation, and molecular genetics, the use of SHW has the potential to increase wheat yield. Moreover, genomic variation and recombination were enhanced in newly formed SHW, which could generate more genovariation or new gene combinations compared to ancestral genomes. Accordingly, we presented a breeding strategy for the application of SHW—the ‘large population with limited backcrossing method’—and we pyramided stripe rust resistance and big-spike-related QTLs/genes from SHW into new high-yield cultivars, which represents an important genetic basis of big-spike wheat in southwestern China. For further breeding applications of SHW-derived cultivars, we used the ‘recombinant inbred line-based breeding method’ that combines both phenotypic and genotypic evaluations to pyramid multi-spike and pre-harvest sprouting resistance QTLs/genes from other germplasms to SHW-derived cultivars; consequently, we created record-breaking high-yield wheat in southwestern China. To meet upcoming environmental challenges and continuous global demand for wheat production, SHW with broad genetic resources from wild donor species will play a major role in wheat breeding.


Introduction
Even though wheat (Triticum aestivum L.) is the most widely consumed food in the world, its global production still needs to be enhanced to meet the growing demand in the coming decades [1]. To improve the genetic yield potential of wheat, the introduction of alien genomic components with superior alleles and elite genes from wild/related species is considered an effective conventional method that could broaden the narrow genetic basis of modern common wheat, especially the D genome [2]. However, homologous recombination between alien chromatin and wheat chromosomes is often repressed [3][4][5]; alien chromatin also leads to chromosomal aberrations [6], which is disadvantageous for breeding.
To simulate and study the origin process of the common wheat, synthetic hexaploid wheat (SHW) was created by crossing tetraploid T. turgidum with Aegilops tauschii and subsequently doubling chromosomes [7][8][9][10]. Fortunately, SHW makes it easy to transfer both T. turgidum and Ae. tauschii genes into common wheat, and homologous recombination could break the undesirable gene linkages when crossing SHW with common wheat and fix D genomes of SHW lines 'XX86', 'W7984', and 'SHW-L1'; the alleles are mostly involved in grain weight, grain number per spike, spike length, and tiller number. Wan et al. [37] detected a major QTL for leaf sheath hairiness (LSH) that was also associated with grain yield in the common wheat chromosome 4DL; the favorable QTL allele was introgressed from Ae. tauschii, which enhanced grain yield by increasing grain weight. This finding indicated that SHWs might carry yield-related QTL alleles superior to those of modern common wheat.

Enhanced Genomic Variation and Recombination in Synthetic Hexaploid Wheat
Approximately 9000 years ago, an accidental hybridization between domesticated emmer wheat (T. turgidum conv. turgidum, 2n = 4x = 28) and goat grass (Ae. tauschii spp. strangulate, 2n = 2x = 14) with chromosome doubling naturally resulted in the generation of a free-threshing hexaploid common wheat (T. aestivum, 2n = 6x = 42) (Figure 1). This hybrid accounts for approximately 95% of current global wheat production, with tetraploid durum wheat (T. turgidum ssp. durum) representing the remaining 5% [38]. Allohexaploidization added the Ae. tauschii D genome into tetraploid wheat, and the allohexaploid wheat was more adaptive to changing environments and then spreading more rapidly around the world than the tetraploid wheat.
However, the D genome of the first bread wheat originated from only a small number of wild Ae. tauschii ssp. strangulata plants [2,39], and the AB genome was suggested to be from a free-threshing form of tetraploid wheat [39,40]. As the AB and D genomes originated from limited sources, the individuals from tetraploid and diploid parents involved in the hexaploidization of wheat did not possess all superior characteristics in a few totipotent plants to make them sufficiently adaptable to changing environments worldwide. Therefore, in addition to the advantages of heterosis and gene redundancy [41], there are likely other forces in hexaploidization that accelerate wheat evolution and spread.
Genes 2023, 14, x FOR PEER REVIEW 3 of 13 totipotent plants to make them sufficiently adaptable to changing environments worldwide. Therefore, in addition to the advantages of heterosis and gene redundancy [41], there are likely other forces in hexaploidization that accelerate wheat evolution and spread. Our lab simulated the evolutionary hexaploidization process and generated SHW using different Ae. tauschii and tetraploid wheat with the ability to automatically double chromosomes, which occurs by unreduced gamete formation controlled by genetic factors (QTug.sau-3B [42]). With the 10× resequencing data of SHWs and their parents from nextgeneration sequencing (NGS) technology, we found that the sequences of the tetraploid AB and diploid D genomes were altered in SHW (Table 1: unpublished data provided by H.W.), and kept changing in subsequent generations of SHW. Genomic change by DNA elimination and interchromosomal exchange often occurs in newly formed hexaploid wheat [43][44][45]. Wan et al. [9] found that approximately 10% of the SNP loci of Ae. tauschii were eliminated in derived SHW using diversity array Our lab simulated the evolutionary hexaploidization process and generated SHW using different Ae. tauschii and tetraploid wheat with the ability to automatically double chromosomes, which occurs by unreduced gamete formation controlled by genetic factors (QTug.sau-3B [42]). With the 10× resequencing data of SHWs and their parents from nextgeneration sequencing (NGS) technology, we found that the sequences of the tetraploid AB and diploid D genomes were altered in SHW (Table 1: unpublished data provided by H.W.), and kept changing in subsequent generations of SHW. Genomic change by DNA elimination and interchromosomal exchange often occurs in newly formed hexaploid wheat [43][44][45]. Wan et al. [9] found that approximately 10% of the SNP loci of Ae. tauschii were eliminated in derived SHW using diversity array technology (DArT)-Seq technology. Moreover, in the co-dominant genotypes of F 2 individuals from a diploid population (SQ665 × SQ783, 2x, D 1 D 1 × D 2 D 2 ) and a new SHW-derived population (Langdon/SQ665 × Langdon/SQ783, 6x, AABBD 1 D 1 × AABBD 2 D 2 ), the recombination frequency of Ae. tauschii was found to be enhanced 2.3-fold by hexaploidization with T. turgidum [9]. The changes of ancestral genomes during hexaploidization could generate more genovariation or new alleles, and the gene redundancy that contributed to polyploidization could shield polyploids from the deleterious effect of unfavorable genomic variations. Additionally, the increased genetic recombination in new SHW-derived cultivars could produce more new allelic combinations subject to natural or artificial selection; the evolution of wheat could be accelerated via hexaploidization, and this could help wheat to rapidly spread and increase its role as a major global crop. Therefore, SHW have the potential to enhance variation and adaptive evolution of bread wheat in the breeding process [13].

A Case of Successful Direct Application of SHW: Chuanmai 42 from Southwestern China
SHW enhanced genomic variation and recombination, and had high breeding potential. However, primary SHWs also have many unfavorable traits, such as late maturity, taller plants, and difficulty in threshing, which made direct application of SHW difficult. Therefore, our team presented a breeding strategy using a large population with limited backcrossing to common wheat ( Figure 2). This breeding strategy involved three core aspects: (1) limited backcrossing with common wheat (2-3 times), reserving more favorable genes and genetic diversity from SHW; (2) using a population of more than 1000 individuals to select favorable gene recombination events between SHW and the backcrossed common wheat; and (3) selecting agronomic traits under multiple environments and testing the candidates' yield potential using high-yield cultivation methods.  [9].
The changes of ancestral genomes during hexaploidization could generate more genovariation or new alleles, and the gene redundancy that contributed to polyploidization could shield polyploids from the deleterious effect of unfavorable genomic variations. Additionally, the increased genetic recombination in new SHW-derived cultivars could produce more new allelic combinations subject to natural or artificial selection; the evolution of wheat could be accelerated via hexaploidization, and this could help wheat to rapidly spread and increase its role as a major global crop. Therefore, SHW have the potential to enhance variation and adaptive evolution of bread wheat in the breeding process [13].

A Case of Successful Direct Application of SHW: Chuanmai 42 from Southwestern China
SHW enhanced genomic variation and recombination, and had high breeding potential. However, primary SHWs also have many unfavorable traits, such as late maturity, taller plants, and difficulty in threshing, which made direct application of SHW difficult. Therefore, our team presented a breeding strategy using a large population with limited backcrossing to common wheat ( Figure 2). This breeding strategy involved three core aspects: (1) limited backcrossing with common wheat (2-3 times), reserving more favorable genes and genetic diversity from SHW; (2) using a population of more than 1000 individuals to select favorable gene recombination events between SHW and the backcrossed common wheat; and (3) selecting agronomic traits under multiple environments and testing the candidates' yield potential using high-yield cultivation methods.  Under this breeding strategy, our team used one SHW line introduced from the International Maize and Wheat Improvement Center (CIMMYT), Syn769, successively crossed with two local common wheat lines, SW3243 and Chuan 6415, and three cultivars were bred: Chuanmai 38, Chuanmai 42, and Chuanmai 43. Among them, Chuanmai 42 was released in 2003 and was the first commercial SHW derivative in the world. In the Sichuan regional trials of wheat cultivars in 2002 and 2003, the average grain yield of Chuanmai 42 increased by 70.2% and 28.3%, respectively, compared to the check cultivars Chuanmai 28 and Chuanmai 107 (Table 2). Overall, grain yield increased by 35%, which broke the yield record of commercial cultivars in southwestern China [46,47]. Since 2003, the use of Chuanmai 42 in wheat production has increased grain output by approximately 1,000,000,000 Kg compared to the old cultivars [47]. Moreover, as a leading wheat cultivar, Chuanmai 42 has become a foundation breeding parent for wheat improvement in southwestern China. From 2008 to 2021, a total of 26 cultivars were selected from crosses containing Chuanmai 42; among them, 17 cultivars were from the first generation of Chuanmai 42 crossed with another parental line with high-yield potential, such as Chuanmai 104 and Chuanmai 602 ( Table 3)  Since 2003, why is Chuanmai 42 so widely applied in wheat production and breeding in southwestern China? Additionally, what is the role of the SHW germplasm in Chuanmai 42 production?
Unlike the common hexaploid wheat with glabrous leaf sheaths, most SHW accessions have a hairy leaf sheath, which is mostly present in wild species of Triticeae, such as Ae. tauschii and T. turgidum var. dicoccoides. Chuanmai 42 has a hairy leaf sheath that was derived from the SHW line Syn769 (Figure 3). Genetic analysis showed that the LSH is from Ae. tauschii with its controlling gene on chromosome 4DL [37]. Interestingly, almost all cultivars derived from Chuanmai 42 inherited the hairy leaf sheath character, even in the 2nd generation, which was caused by the tight linkage between the LSH gene and the QTLs associated with grain weight and yield in Syn769 ( Figure 3E; [37]). In breeding programs, the leaf sheath hairiness can be used as a morphological marker for high grain yield QTL selection from Chuanmai 42 derivatives. Since 2003, why is Chuanmai 42 so widely applied in wheat production and breeding in southwestern China? Additionally, what is the role of the SHW germplasm in Chuanmai 42 production?
Unlike the common hexaploid wheat with glabrous leaf sheaths, most SHW accessions have a hairy leaf sheath, which is mostly present in wild species of Triticeae, such as Ae. tauschii and T. turgidum var. dicoccoides. Chuanmai 42 has a hairy leaf sheath that was derived from the SHW line Syn769 (Figure 3). Genetic analysis showed that the LSH is from Ae. tauschii with its controlling gene on chromosome 4DL [37]. Interestingly, almost all cultivars derived from Chuanmai 42 inherited the hairy leaf sheath character, even in the 2nd generation, which was caused by the tight linkage between the LSH gene and the QTLs associated with grain weight and yield in Syn769 ( Figure 3E; [37]). In breeding programs, the leaf sheath hairiness can be used as a morphological marker for high grain yield QTL selection from Chuanmai 42 derivatives. With a total of 1029 simple sequence repeat (SSR) and 2268 DArT markers detecting polymorphisms among three parents (Syn769, SW3243, and Chuan6415) of Chuanmai 42, the frequency of SHW alleles introgressed to Chuanmai 42 was 15.14%, which was significantly less than the expected 25% assuming random gene assortment. The distribution of introgressed alleles over the A, B, and D genomes was not uniform (B > A > D); introgression occurred most frequently on chromosomes 1A, 1B, 2B, 3A, 4D, 6A, and 6B, whereas none were detected on chromosomes 1D and 7A [13,48,49]. On chromosome 1B, the stripe rust resistance gene YrCH42 from Chuanmai 42 ( Figure 4A; [50]) and the QTL allele for  With a total of 1029 simple sequence repeat (SSR) and 2268 DArT markers detecting polymorphisms among three parents (Syn769, SW3243, and Chuan6415) of Chuanmai 42, the frequency of SHW alleles introgressed to Chuanmai 42 was 15.14%, which was significantly less than the expected 25% assuming random gene assortment. The distribution of introgressed alleles over the A, B, and D genomes was not uniform (B > A > D); introgression occurred most frequently on chromosomes 1A, 1B, 2B, 3A, 4D, 6A, and 6B, whereas none were detected on chromosomes 1D and 7A [13,48,49]. On chromosome 1B, the stripe rust resistance gene YrCH42 from Chuanmai 42 ( Figure 4A; [50]) and the QTL allele for increased grain number per spike ( Figure 4B; [51]) originated from its SHW parent Syn769 ( Figure 4C; [48]). increased grain number per spike ( Figure 4B; [51]) originated from its SHW parent Syn769 ( Figure 4C; [48]).

Further Application using SHW-derived Cultivars: Chuanmai 104
In 2020, the grain yield of Chuanmai 104 reached 729.8 Kg/mu in the high-yield cultivation field in Jiangyou, Sichuan, which was a record for the highest yield of wheat in southwestern China [52,53]. It was selected from the 127 F7 recombinant inbred lines (RILs) of an SHW-derived Chuanmai 42 crossed with Chuannong 16 (Figure 5), which increased the grain yield by approximately 8.42% more than Chuanmai 42 in 2010-2012 national regional cultivar trials.

Further Application Using SHW-Derived Cultivars: Chuanmai 104
In 2020, the grain yield of Chuanmai 104 reached 729.8 Kg/mu in the high-yield cultivation field in Jiangyou, Sichuan, which was a record for the highest yield of wheat in southwestern China [52,53]. It was selected from the 127 F 7 recombinant inbred lines (RILs) of an SHW-derived Chuanmai 42 crossed with Chuannong 16 (Figure 5), which increased the grain yield by approximately 8.42% more than Chuanmai 42 in 2010-2012 national regional cultivar trials.

Further Application using SHW-derived Cultivars: Chuanmai 104
In 2020, the grain yield of Chuanmai 104 reached 729.8 Kg/mu in the high-yield cultivation field in Jiangyou, Sichuan, which was a record for the highest yield of wheat in southwestern China [52,53]. It was selected from the 127 F7 recombinant inbred lines (RILs) of an SHW-derived Chuanmai 42 crossed with Chuannong 16 (Figure 5), which increased the grain yield by approximately 8.42% more than Chuanmai 42 in 2010-2012 national regional cultivar trials.  We genotyped each F 7 RIL using SSR and DArT markers, and evaluated yield-related traits of each RIL (10 m 2 /plot) in multiple environments with high-yield cultivation. The F 2 population of Chuanmai 42 × Chuannong 16 was exposed to different environmental stress to generate more gene recombination [54,55], and 127 RILs were finally obtained. Additionally, we combined phenotypic and genotypic evaluations to artificially select cultivars with high yield potential. Our team referred to this as the 'RIL-based breeding method' (Figure 5). Their obvious differences with the traditional phenotypic evaluation for QTL analysis are reflected in the following aspects: (1) replacing single-seed sowing with planting density of high-yield wheat production in China, considering the trade-off between yield components and interplant competition; (2) expanding the planting area of each plot from 2-3 rows to at least 10 m 2 , simulating field conditions of wheat production, with the aim to bridge the gaps between individual and population performances. In addition, evaluating the contributions of high-yielding QTLs to grain yield on the population level rather than in the individual plant or panicle level was also highly recommended by Xiong et al. [56] recently.
The yield potential of Chuanmai 104 was enhanced by its pre-harvest sprouting resistance and low-temperature tolerance at the flowering stage ( Figure 6: unpublished data provided by Y.T.); it is more resistant than its parent, Chuanmai 42 ( Figure 6B; [57]). Moreover, Chuanmai 104 had stripe rust and powdery mildew resistance, inheriting the resistance loci of YrCH42 [50], Qyr.saas-7B [58], and QPm.saas-4AS [59] from its parent Chuanmai 42 [51]. The actual grain yield over five continuous years' field production ranged from 650 kg/mu to 700 kg/mu. We genotyped each F7 RIL using SSR and DArT markers, and evaluated yield-related traits of each RIL (10 m 2 /plot) in multiple environments with high-yield cultivation. The F2 population of Chuanmai 42 × Chuannong 16 was exposed to different environmental stress to generate more gene recombination [54,55], and 127 RILs were finally obtained. Additionally, we combined phenotypic and genotypic evaluations to artificially select cultivars with high yield potential. Our team referred to this as the 'RIL-based breeding method' (Figure 5). Their obvious differences with the traditional phenotypic evaluation for QTL analysis are reflected in the following aspects: (1) replacing single-seed sowing with planting density of high-yield wheat production in China, considering the trade-off between yield components and interplant competition; (2) expanding the planting area of each plot from 2-3 rows to at least 10 m 2 , simulating field conditions of wheat production, with the aim to bridge the gaps between individual and population performances. In addition, evaluating the contributions of high-yielding QTLs to grain yield on the population level rather than in the individual plant or panicle level was also highly recommended by Xiong et al. [56] recently.
The yield potential of Chuanmai 104 was enhanced by its pre-harvest sprouting resistance and low-temperature tolerance at the flowering stage ( Figure 6: unpublished data provided by Y.T.); it is more resistant than its parent, Chuanmai 42 ( Figure 6B; [57]). Moreover, Chuanmai 104 had stripe rust and powdery mildew resistance, inheriting the resistance loci of YrCH42 [50], Qyr.saas-7B [58], and QPm.saas-4AS [59] from its parent Chuanmai 42 [51]. The actual grain yield over five continuous years' field production ranged from 650 kg/mu to 700 kg/mu. Chuanmai 104 was listed as the leading commercial cultivar for wheat production by the Sichuan Province government from 2013 to 2016 and the Ministry of Agriculture and Rural Affairs of China from 2015 to 2016. From 2015 to present, a total of 11 cultivars were selected from crosses involving Chuanmai 104, such as the big-spike cultivar Chuanmai 93 and tight-plant-type cultivar Chuanmai 98; generally, Chuanmai 104 has also been used as a foundation parent for recent wheat breeding (Table 4). Since 2012, the planting acreage of Chuanmai 104 has totaled 21.5 million mu and generated an additional economic output of 2.4 billion Chinese yuan [53]. The breeding and application of Chuanmai 104  Chuanmai 104 was listed as the leading commercial cultivar for wheat production by the Sichuan Province government from 2013 to 2016 and the Ministry of Agriculture and Rural Affairs of China from 2015 to 2016. From 2015 to present, a total of 11 cultivars were selected from crosses involving Chuanmai 104, such as the big-spike cultivar Chuanmai 93 and tight-plant-type cultivar Chuanmai 98; generally, Chuanmai 104 has also been used as a foundation parent for recent wheat breeding (Table 4). Since 2012, the planting acreage of Chuanmai 104 has totaled 21.5 million mu and generated an additional economic output of 2.4 billion Chinese yuan [53]. The breeding and application of Chuanmai 104 obtained the First-class Prize of the Sichuan Province Scientific and Technological Progress Award in 2020. The genetic components of Chuanmai 104 were dissected by QTL mapping for yieldrelated traits [51,60,61], pre-harvest sprouting resistance [57], and powdery mildew resistance [59]. On chromosome 1B, the majority of genomic regions of Chuanmai 104 were from the female parent Chuanmai 42, and among the detected QTLs or genes, the QTL allele of Chuanmai 42 increased the resistance of stripe rust and grain weight (Figure 7). On chromosome 1DS, the genomic region of Chuanmai 104 associated with increasing spike number per m 2 and grain yield was from Chuannong 16. Chromosome 2BS of Chuanmai 104 was from Chuannong 16, and is related to enhanced grain yield and pre-harvest sprouting resistance, whereas its chromosome 2BL was from Chuanmai 42 and was associated with increased thousand-kernel weight and plant height.  The genetic components of Chuanmai 104 were dissected by QTL mapping for yieldrelated traits [51,60,61], pre-harvest sprouting resistance [57], and powdery mildew resistance [59]. On chromosome 1B, the majority of genomic regions of Chuanmai 104 were from the female parent Chuanmai 42, and among the detected QTLs or genes, the QTL allele of Chuanmai 42 increased the resistance of stripe rust and grain weight (Figure 7). On chromosome 1DS, the genomic region of Chuanmai 104 associated with increasing spike number per m 2 and grain yield was from Chuannong 16. Chromosome 2BS of Chuanmai 104 was from Chuannong 16, and is related to enhanced grain yield and preharvest sprouting resistance, whereas its chromosome 2BL was from Chuanmai 42 and was associated with increased thousand-kernel weight and plant height.  Chuanmai 104 inherited the big-spike properties of higher grain number per spike and grain weight from SHW-derived Chuanmai 42, multiple spike properties including spike number per square meter from Chuannong 16, and grain yield from both parents contributed. Additionally, Chuanmai 42 provided stripe rust [50,58] and powdery mildew resistance [59], whereas Chuannong 16 provided pre-harvest sprouting resistance [57].

Conclusions and Perspectives
SHW is a useful genetic resource that can be used to improve the performance of common wheat by transferring favorable genes from a wide range of tetraploid or diploid donors because many QTLs/genes can be expressed under a genetic background of common hexaploid breeding lines [34][35][36][62][63][64][65]. The use of SHW could also bypass the disadvantages of using alien chromatin such as from rye, Agropyron elongatum, and Haynaldia villosa, which undergo reduced homologous recombination with wheat chromosomes. Importantly, genomic variation and recombination were enhanced in newly formed SHW, which could generate more genovariation or new gene combinations compared to ancestral genomes. Additionally, the genomic instability of newly formed SHW can also enhance the genetic variation and recombination when they are crossed with common breeding lines, as well as abundant variation in agronomic traits [66].
How to utilize primary SHW for wheat improvement effectively? Different geneticists and/or breeders summarized various advisable strategies from different perspectives [15,67]. Here, according to our application practices of SHW for decades, using the 'large population with limited backcrossing method' was suggested as an effective breeding strategy for direct breeding application of SHW to improve the breeding population and select elite cultivars. These improved SHW derivatives could be used as a basic genetic framework for the next round of pyramiding favorable genes from new SHW lines or other breeding lines. Additionally, for further QTL/gene pyramiding, the RIL-based breeding method could be beneficial by accurately evaluating their breeding value and identifying suitable new cultivars that should be selected.
The global demand for wheat production will keep continuously growing with the world's increasing population, especially as global warming becomes an increasing worldwide threat that enhances difficulty in crop cultivation. In this situation, a significant increase in yield like during the Green Revolution is desirable to break up the bottleneck of wheat yield and make wheat cultivars tolerant to drought, heat, cold, and flooding stress. SHW along with broad genetic resources from wild donor species will play a big role in the race to meet upcoming environmental challenges and the continuous global demand for wheat production.