In silico curation of QTL-rich clusters and candidate gene identification for plant height of bread wheat

Dengn Xu,Chenfei Ji,Xinru Lyu,Tingzhi Yng,Huimin Qin,Ylin Wng,Qinlin Ho,Wenxing Liu,Xuehun Di,Jinin Zeng,Hongsheng Zhng,Xinchun Xi,Zhonghu He,Shunghe Co,Wujun M,

a College of Agronomy,Qingdao Agricultural University,Qingdao 266109,Shandong,China

b National Wheat Improvement Center,Institute of Crop Science,Chinese Academy of Agricultural Sciences,Beijing 100081,China

Keywords: QTL-rich clusters Plant height Semi-dwarfism Reduced height genes Candidate genes

ABSTRACT Many genetic loci for wheat plant height(PH)have been reported,and 26 dwarfing genes have been catalogued.To identify major and stable genetic loci for PH,here we thoroughly summarized these functionally or genetic verified dwarfing loci from QTL linkage analysis and genome-wide association study published from 2003 to 2022.A total of 332 QTL,270 GWAS loci and 83 genes for PH were integrated onto chromosomes according to their locations in the IWGSC RefSeq v2.1 and 65 QTL-rich clusters(QRC)were defined.Candidate genes in each QRC were predicted based on IWGSC Annotation v2.1 and the information on functional validation of homologous genes in other species.A total of 38 candidate genes were predicted for 65 QRC including three GA2ox genes in QRC-4B-IV,QRC-5A-VIII and QRC-6A-II (Rht24) as well as GA 20-oxidase 2 (TaSD1-3A) in QRC-3A-IV.These outcomes lay concrete foundations for mapbased cloning of wheat dwarfing genes and application in breeding.

1.Introduction

Wheat is a highly adaptable and widely cultivated food crop in the world,providing about 21%of human food calories and 20%of protein[1].It has high nutritional value and unique gluten characteristics that can be used to produce a variety of foods,making it the most important food for trade [2].Wheat plant height is an important agronomic trait that affects the plant architecture and yield.As the aboveground supporting organ,wheat stem is the intermediate bridge connecting wheat leaves and grains through various physiological activities [3].High wheat plant height can cause plant lodging,which directly leads to lower yield,whereas too short plants can also cause crowding of canopy leaves,affect photosynthetic rate and cause too low biomass to provide sufficient ‘‘source”,leading to a low yield [4].Therefore,unless a lodging resistance gene is available,wheat plant height should be appropriately reduced to increase plant lodging resistance and harvest index.

To date,26 reduced height genes (Rht1-Rht26) have been named in wheat [5-7].Six of them are homologs ofRht-1locus on group 4 chromosomes,includingRht-B1b(Rht1),Rht-D1b(Rht2),Rht-B1c(Rht3),Rht-D1c(Rht10),Rht-B1e(Rht11) andRht-B1p(Rht17) [8-10].Rht9,14,15,16,18,19 and 22 are from tetraploid wheat(Triticum turgidum)and the others are from hexaploid wheat (T.aestivum) [11].Rht4,5,7,11,12,13,14,15,16,17,18,19,20 and 23 are mutated by mutagens including fast neutron radiation,X-ray,γ-radiation,ethyl methanesulfonate,N-Nitroso-N-methylurea,and the other 11 Rht genes were detected from breeding materials [6,11].

Among the 26Rhtgenes catalogued,the reduction of plant height in breeding was dominated by the utilization ofRht-B1b(Rht1),Rht-D1b(Rht2),Rht8,Rht24[12-15].The‘‘Green Revolution”genes,Rht-B1b(Rht1) andRht-D1b(Rht2) alleles,have increased wheat yield through reducing height,improving lodging resistance,partitioning more assimilates to grains and increasing spike numbers per plant [4].However,most modern semi-dwarf wheat varieties have eitherRht-B1borRht-D1b,showing short coleoptile,weak seedling vigor,low nitrogen fertilizer use efficiency,frost susceptible,and undesirable performance in shallow soils and dryland farming [16-19].Both ofRht-B1andRht-D1encode DELLA proteins,which are the key repressor of GA-responsive growth[20].Rht8 encodes a protein with a zinc finger BED-type motif and an RNase H-like domain that reduces plant height via regulating bioactive GA biosynthesis [13,15].Rht24 encodes a gibberellin 2-oxidase (TaGA2ox-A9) contributing to the reduction of bioactive GA in stems [14].Both Rht8 and Rht24 are sensitive to exogenous GA and have no negative effects on coleoptile length,seedling vigor and grain yield [13-15,21].While Rht8 and Rht24 have been widely used in wheat breeding,continuous efforts have been made to search for new Rht genes for modern breeding needs.Rht13 was an induced mutant originated from Magnif,which reduces final plant height between 17% and 34% and improved stem strength[22-26].Recently,Rht13 was reported to encode an autoactive allele of a nucleotide-binding site/leucine-rich repeat (NB-LRR)gene on chromosome 7BL [27].The GA-sensitive dwarfing locus Rht25 encodes a plant-specific AT-rich sequence-and zincbinding protein(PLATZ,TraesCS6A02G156600),which can interact with DELLA to modulate the wheat plant height [28].In addition,many other new genetic loci related to plant height have been reported,providing highly effective and simple way to identify targets for semi-dwarf wheat breeding.In the current article,we surveyed the genetic loci of plant height reported in the last 20 years and curated 65 QTL-rich clusters (QRC).

2.Materials and methods

2.1.Identification of PH-related QTL and QTL-rich clusters

A total of 151 published articles of QTL linkage analysis and genome-wide association study (GWAS) as well as cloned genes for PH in wheat from 2003 to 2022 were surveyed to curate QTLrich clusters(QRC).For each QTL,its associated markers,LOD(Logarithm of the odds ratio) value,R2or PVE (phenotypic variance explained) value were collected.In order to ensure the validity of QTL,only the QTL detected in at least half of the environments withR2values higher than 5%in individual studies were investigated in this study.Markers from at least three independent studies that physically located in 10 Mb range were considered as a QTL-rich cluster(QRC)based on the wheat reference genome assembly from cv.Chinese Spring (IWGSC RefSeq v2.1).

2.2.Mapping markers or genes on the genome

All collected QTL,GWAS loci and cloned genes were mapped to the wheat reference genome IWGSC RefSeq v2.1 [29,30].The closest linked markers or two flanking markers on both sides of the QTL confidence interval were manually searched for genomic positions.The flanking markers or primer sequences were obtained from wheat 90K [31],660K [32],GrainGenes (https://wheat.pw.usda.gov/GG3/),and DArT (https://www.diversityarrays.com).Based on the BLASTN,the genomic positions of flanking marker and primer sequences on the IWGSC RefSeq v2.1 were obtained(https://urgi.versailles.inra.fr/blast_iwgsc/blast.php).If the physical locations of markers were not found,it will be replaced by the linked marker on the genetic linkage map.

2.3.Candidate genes prediction

All genes from IWGSC RefSeq Annotations v2.1 (https://urgi.versailles.inrae.fr/download/iwgsc/IWGSC_RefSeq_Annotations/v2.1/) were annotated for theKnumber using GhostKOALA tool(https://www.kegg.jp/ghostkoala/).We enriched wheat genes on the KEGG (Kyoto Encyclopedia of Genes and Genomes) database and extract genes in the GA synthesis pathway (map00904: diterpenoid biosynthesis)using TBtools[33].Since there are many studies on rice genes for plant height,the strategy of homology comparison between wheat and rice was used to explore key candidate genes in QRC region.All published basic information of functionally validated rice PH-related genes was downloaded from the China Rice Data Center with site tools of ontology (https://www.ricedata.cn/),and wheat-rice orthologous comparison were extracted from EnsemblPlants database with site tools of Biomart(https://plants.ensembl.org/index.html).Genes located in the QRC region were considered to be candidate genes affecting PHrelated traits in wheat.The transcriptomic data of developmental time-course of Chinese Spring,Azhurnaya,and synthetic hexaploidy deposited in the expression Visualization and Integration Platform (expVIP,https://www.wheat-expression.com) were used to explore the tissue expression patterns of candidate genes[34,35].

2.4. Plant materials

We previously conducted the overexpressed and CRISPR/Cas9 knockout ofRht-B1bin the cultivar of fielder which is a bread wheat cultivar withRht-A1a,Rht-B1bandRht-D1a[36].TheRht-B1boverexpression plants (Rht-B1b-OE),loss-of-function plants(Rht-B1b-KO) and transgenic null plants were grown in the greenhouse for Quantitative Real-time PCR (qPCR) experiments.The uppermost internodes were collected at the stage of 1 d and 8 d after heading fromRht-B1b-OE,Rht-B1b-KO and TNL,respectively.

2.5. RNA extraction and qPCR

Total RNA of three biological replicates were extracted using the RNAprep Pure Plant Plus Kit (TIANGEN,https://www.tiangen.com/)following the manufacturer’s instructions.cDNA was prepared using a FastKing RT kit (with gDNase) (TIANGEN).Gene-specific qPCR primer pairs were designed according to gene annotation from RefSeq v2.1 [29,30].qPCR assays were performed on the ABI QuantStudio3 system with the 2X Universal SYBR Green Fast qPCR Mix (ABclonal,https://abclonal.com.cn/).Relative gene expression pattern was normalized to EF1α gene using the 2-ΔΔCTequation [37].

3.Results

3.1. QTL for plant height from linkage analysis

About 90 PH-related QTL linkage analysis in wheat from 2003 to 2022 were indexed (Table S1),resulting in a collection of 332 PHrelated QTL that met the predefined criteria.These QTL were labelled on the IWGSC RefSeq v2.1 genome based on the genomic positions of linked markers or mid-points of their genetic intervals(Table S1).All 21 wheat chromosomes contain varying number of QTL for PH.Chromosomes 2D,3A,4B,4D,5A,and 6A each harbor more than 20 QTL for PH,whereas Chr.5D contains a minimum number of three QTL (Fig.1;Table S1).

Fig.1.Distribution of genetic loci for plant height (PH) on wheat genome.Genetic loci for PH were collected from QTL linkage analysis,genome-wide association study(GWAS)and cloned genes in wheat.Some major genes or loci for vernalization,flowering time and domestication are also integrated into this map.QTL and GWAS loci are shown in black and blue colors,respectively.Cloned and named Rht genes are highlighted in pink.The black and cyan bars in chromosomes indicate the positions of centromeres and QTL-rich clusters (QRC),respectively.

3.2. GWAS loci for plant height

Among the 22 PH-related GWAS studies,most of them were based on SNP arrays,and a few were conducted with DArT and SSR markers [12,38-58].A total of 270 GWAS loci associated with PH were identified (Table S2).As the QTL studies mentioned above,the 21 wheat chromosomes harbor QTL for PH and the locus distribution also showed chromosome bias.Each of Chr.5A and Chr.5B chromosomes harbors more than 20 loci for PH,representing the chromosomes with the highest number of GWAS loci,whereas only two loci on Chr.3D were reported (Fig.1;Table S2),representing the lowest number.Interestingly,only 19 loci each were detected on Chr.4B and 4D,less than the Chr.5A and 5B thoughRht1andRht2are located on the Chr.4B and Chr.4D,respectively.

3.3.Cloned genes for plant height

A total of 83Rhtgenes have been cloned or catalogued(Table S3).Of them,threeRht-1homologous genes were the firstly cloned and their wild-type alleles were designated asRht-A1a,Rht-B1a,andRht-D1a[20].The gain-of-function mutation inRht-B1b(Rht1) andRht-D1b(Rht2) generates a truncated protein without DELLA domain that confers semi-dwarf phenotype [20,59].Rht8andRht24were map-based cloned,reducing plant height via influencing bioactive GA biosynthesis[13-15].The GA-sensitive dwarfing locusRht25was also map-based cloned,encoding a PLATZ transcription factor that interacts with DELLA (RHT1) to influence the plant height [28].Rht13is a NB-LRR gene and an amino acid change S240F mutation in the dwarfingRht-B13ballele that contributes to the up-regulation transcription of pathogenesisrelated genes such as class III peroxidases [27].The domesticated geneQon wheat Chr.5A (5AQ) affecting plant height,spike type,threshing ability and other important agronomic traits encodes an AP2 transcription factor,which might be the candidate gene ofRht23[60,61].Rht12was mapped into a 10.73 Mb fragment deletion on the terminal of Chr.5AL,which affects GA synthesis and reduces stem cell length [62].

In addition,there are many other PH related genes that have pleiotropic effect on wheat growth and development.TB1affects multiple agronomic traits including tillering number,inflorescence architecture,and plant height [63,64].TaNAC100 was reported as an integrator of seed protein and starch synthesis;TaNAC100overexpression repressed PH,increased heading date,and promoted seed size and thousand kernel weight[65].TaSEP3-D1overexpression plants exhibited delayed heading and reduced PH [66].The blue light photoreceptor TaCRY1a could repress GA signaling and reduce PH by competitively attenuating the GID1-DELLA interaction [67].TaDEP1is an important regulator of wheat growth and development and the aabbdd loss-of-function mutant plants exhibited an extremely dwarf phenotype [68].Knock-down ofTaARF12(TraesCS2A01G547800) by RNAi reduced plant height by up to 20.7% [57].TaWUS-likemay inhibit the synthesis of GA3 and/or BR,thus caused stem shortening and plant dwarfing in wheat[69].SQUAMOSA-SVP complexes promoted stem elongation during the early reproductive phase,and wheatvrt2 svp1mutants showed later flowering and shorter stems [70].Three MADS-box genes VRN1,FUL2 and FUL3 was reported to play key and redundant roles in spikelet and spike development,and also affect PH[71].BothPpd-D1and its paralog geneTaPRR37affected heading time and PH[72,73].Although these genes can also reduce PH,they were barely used in breeding due to side or pleiotropic effects on wheat growth and development.

3.4.Prediction of candidate genes for QTL-rich clusters

After deep analysing the collected 332 QTL,270 GWAS loci and 83 genes for PH,a total of 65 stable QRC were curated (Table 1).Among them,50 QRC can be found or validated by combining QTL linkage analysis with GWAS (Table S4).The candidate genes of the QRC were predicted according to IWGSC Annotation v2.1 enriched on the gibberellin metabolism pathway and PH-related genes reported in other species.The expression patterns of the 38 candidate genes were also retrieved from the expVIP database(Fig.S1).

Table 1Detailed information and candidate genes for QTL-rich clusters associated with plant height.

Many cloned genes related to wheat PH were found in QRC regions,includingPpd-A1in QRC-2A-I [72,74,75],FUL3-2A [71] in QRC-2A-III,TaCOLD1-2Ain QRC-2A-V [76],TaARF12-2Ain QRC-2A-VI [57],RNHL-B1in QRC-2B-I [13,15],TaARF12-2Bin QRC-2BV [57],Rht8/RNHL-D1in QRC-2D-I [13,15],Ppd-D1in QRC-2D-II[72,74,75],Rht-B1in QRC-4B-I [20],TaPRR73-4Bin QRC-4B-III[73],Rht-D1in QRC-4D-I [20],TaWUS-like-5Ain QRC-5A-I [69],Vrn1-5Ain QRC-5A-VI [71,77],Vrn2-5A/ZCCT1-5Ain QRC-5A-VIII[78],Vrn1-5Bin QRC-5B-IV [71,77],Vrn1-5Din QRC-5D-I [71,77],Rht24in QRC-6A-II [14],SVP1-6Ain QRC-6A-III [70],TaPRR1-D1in QRC-6D-I [79],Vrn3-7Bin QRC-7B-I,andVrn3-7Din QRC-7D-I[80] (Table 1).In addition,four genes involved in GA metabolism were located in the QRC region,including asd1homologous geneTraesCS3A03G0950800and three GA 2-oxidase-encoding genesTraesCS4B03G0723300,TraesCS5A03G1269600andTraesCS6A03G0611100.SD1encodes a GA20ox2 that controls PH and mutations in this locus,leading to different degrees of dwarfing in rice[81-83].GA 2-oxidases are key enzymes in the GA synthesis pathway and catalyze the conversion of GA53 to GA20[84].To further explore the candidate genes affecting wheat PH,a detailed search was carried out for the cloned genes related to PH in rice,and 371 functional genes were obtained.Based on BLASTP,wheat orthologs of the 371 rice genes influencing PH were obtained.Among them,16 wheat orthologs of rice PH genes were found in the QRC regions.Interestingly,two homologs ofsui1,TraesCS3A03G0043500andTraesCS3B03G0056000,were located in QRC-3A-I and QRC-3B-I,respectively.SUI1,a putative phosphatidyl serine synthase (PSS) family protein,is conserved in plant kingdoms and plays a key role in regulating uppermost internode length in rice [85,86].TraesCS4B03G0903600andTraesCS5A03G1208400are the homologs ofOsDCL1,located in QRC-4D-I and QRC-5A-VII,respectively.OsDCL1plays an important role in miRNA processing and loss-of-function ofOsDCL1that causes dwarfism [87].These genes have been well studied in rice.Thus,their homologous genes in wheat are likely the candidate genes for the QRC.

3.5.Release the potential for reducing plant height of TaSD1 in wheat

The‘Green Revolution’in wheat and rice was due to the using of wheatRht-B1b/Rht-D1bgenes and the ricesd1gene,respectively[4].BothRht-B1bandRht-D1bencode the N-terminally truncated DELLA proteins that are insensitive to the GA-induced degradation[20].The wild-typeSD1gene encodes a biosynthesis enzyme,GA20ox-2,that catalyses the conversion of GA53 to GA20,andsd1mutant leads to a short stature of rice without penalty to yield[81-83].Three homologues ofsd1,TraesCS3A03G0950800(TaSD-3A),TraesCS3B03G1087000(TaSD-3B) andTraesCS3D03G0884400(TaSD-3D),were identified in wheat and onlyTaSD1-3Ais located in QRC-3A-IV (Table 1).We detected the sequence variation ofTaSD1in 677 bread wheat varieties based on the resequencing data[100].OnlyTaSD1-3Acontained sequence variation in the gene region(Table S5)in that five missense mutations formed two haplotypes(Fig.2A).The Hap I ofTaSD1-3Awas the dominant allele in the regions of Europe,America and Australia,especially in China,which indicates thatTaSD1had not been widely utilized in wheat due to lack of alleles(Fig.2B).In fact,the height of tiller displayed an average reduction of 5 cm in the knockout ofTaSD1-3Dplants compared with the wild-type at the anthesis stage[101].The weak height reduction of theTaSD1-3Dmutation may be caused by its homeologues on the A and B genomes with similar expression patterns that compensate the function loss ofTaSD1-3D(Fig.S1).

Fig.2.Gene variation,distribution and expression of TaSD1-3A.(A)Gene structure and five missense mutations in the coding region of TaSD1-3A.(B)Worldwide distribution of two major haplotypes of TaSD1-3A in wheat.The Hap I and Hap II of TaSD1-3A is represented in blue and orange on the pie chart,respectively.The map content approval number is GS(2016)1565(https://bzdt.ch.mnr.gov.cn/).(C)The expression of three homologues of TaSD1 in the transgenic null plants(TNL),the overexpression of Rht-B1b(OE) and CRISPR-Cas9 knockout of Rht-B1b (KO) under the background of Fielder.

Negative feedback regulation of GA-biosynthetic enzyme GA 20-oxidase by DELLAs contributes to homeostasis of the GA level in Arabidopsis (REF).DELLA proteins are the major negative regulators of GA signal transduction and could be rapidly degraded via the ubiquitin-proteasome pathway in the presence of GAs.In wheat,both semi-dwarfingRht-B1bandRht-D1bencoding truncated DELLA proteins orthologous to the gibberellin-insensitive(GAI)protein inArabidopsis[20].Therefore,it is important to check whetherTaSD1was regulated by DELLA in wheat.Our qPCR resultsshowed that the expression of three homeologues ofTaSD1have no differences in the stem among TNL,Rht-B1b-OE andRht-B1b-KO lines,which allow breeders breeding semi-dwarf wheat combinedTaSD1withRht-B1b(Fig.2C;Table S6).Taking together,gene editing or mutation manipulation of the threeTaSD1homeologues can create new allele combinations to fine tune PH for various breeding needs.

4.Discussion

4.1.Cloned dwarfing genes and future ideal plant height

Wheat has a large (～16 Gb) and complex genome with more than 85% repetitive DNA that make cloning of individual genes challenging [29].Up to now,only several dwarfing genes were cloned includingRht-1,Rht8,Rht13,Rht14/Rht18,Rht23,Rht24andRht25.Rht-1(Rht-B1bandRht-D1b,encoding DELLA protein)was cloned by homology cloning based on the sequence ofArabidopsis Gibberellin Insensitive(GAI) gene[20].Rht8gene was identified almost simultaneously by two groups via map-based cloning based on a heterozygous recombinant inbred line derived from the cross Yumai 8679 × Jing 411 and EMS-induced mutants in the wild-type background of Jing 411,respectively [13,15].Rht8encodes a protein containing a zinc finger BED-type motif and an RNase H-like domain that reduces plant height via controlling bioactive GA biosynthesis.Rht23was identified from an ethyl methane sulphonate (EMS)mutant with dwarf and compact spike phenotypes [102].It was mapped to the distal of the long arm of Chr.5D linked to the domesticated geneQon Chr.5D (5Dq).Sequence comparison,linkage analysis and expression analysis showed that a mutation within themiR172target site in the 5Dq led to the dwarf and compact spike phenotype of theRht23mutant[61].The cloning ofRht13gene allows breeders to use autoactive NB-LRR genes in wheat breeding programs.AlthoughRht13dwarfing mutation reduced PH without reduction on coleoptile length,further tests are needed to validate the effect ofRht-B13bon yields[27].

The causal gene ofRht24was also isolated through map-based cloning,which encodes a gibberellin (GA) 2-oxidase (TaGA2ox-A9).The dwarf allele(Rht24b)ofRht24generated higher expression ofTaGA2ox-A9in stems,leading to a reduction of bioactive GA in stems as well as PH reduction but no yield penalty [14].TaGA2ox-A9was also predicted to be the causal gene ofRht18by chromosome flow sorting and sequencing of the Chr.6A from the wildtype (Icaro) and 5 mutants [103].Rht14andRht18mutation were developed by fast neutron radiation in Durum,genetic studies have shown thatRht14is probably allelic toRht18[104,105].However,no sequence difference was observed in the gene ofGA2oxA9while obvious DNA methylation difference was found in its promoter leading to a differential expression pattern ofGA2oxA9betweenRht14dwarf mutant and tall parent.It was proposed thatRht14might regulate the expression ofGA2oxA9to reduce plant height and thus another candidate gene may exist forRht14[106].Further study was needed for the candidate gene ofRht14andRht18though the genomic regions ofRht14,Rht18,andRht24are overlapped [106].

Rht24bwas originated from wild emmer and is an evolutionarily olderRhtgene thanRht-B1b,which had been through both natural and artificial selection into most modern varieties [14].Rht8has been widely introgressed into modern wheat varieties across the world because it can complement the disadvantages ofRht-B1bandRht-D1b[13,15].Both Rht8 and Rht24 are sensitive to exogenous GA and have no negative effects on coleoptile length,seedling vigor and grain yield [13-15,21].Rht25 is GA-sensitive dwarfing locus encoding a PLATZ transcription factor which can interact with DELLA (a key repressor of GA signal transduction)to modulate the wheat plant height.Five platz-A1 mutants(Rht25b,Rht25c,Rht25d,Rht25e,Rht25f) have variable effects,which reduces plant height by～10%to 14%,respectively.Although a small reduction in grain size is linked with the platz-A1 mutation,platz-A1 mutant combined with Rht24b,Rht8 and another QRC loci in this study can be used to fine-tune wheat plant height in wheat breeding programs.These genes are already present in adapted germplasm,which may have good breeding value for‘‘Post-green Revolution”.

4.2.Plant height QRC contributes to ideal plant height

Since 1990,most wheat cultivars were upright and compact,the plant height decreased from 120 to 130 cm in early 1950 to about 75 cm,and the 1000-seed weight increased from about 35 g to 48-50 g in Chinese wheat cultivars [107].The utilization ofRht-B1b(Rht1),Rht-B1b(Rht2),Rht8,Rht24and other dwarf genes together with 1BL.1RS translocation lines played a key role in reducing PH,increasing yield and improving disease resistance[12-15].Despite their widespread adaptation in the world,semidwarf varieties containingRht-B1borRht-D1bhave shorter coleoptiles,low grain weight and fail to produce higher grain yield than wildtype plants in dryland [18,108].In addition,although semidwarf green revolution varieties of cereals (GRVs) boosted crop yields,they are associated with a reduced nitrogen-use efficiency[17,19].Therefore,in future,wheat plant may be optimized to a moderate height but with improved drought tolerance and nitrogen-use efficiency without yield penalty [18].In the current study,a total of 65 QRC controlling PH were identified,which provides a suite of target loci for map-based cloning of wheat dwarfing genes.Due to the availability of abundant resequencing and highquality genome resources such as Chinese Spring,Kenong 9204,Fielder and wheat pan-genome,map-based gene cloning has become easier in wheat [29,30,109-113].It is expected that more dwarfing genes will be cloned in the future,as several PH QTL have been fine-mapped such asRht5,Rht22andQPh.cau-4B-1.2[114-116].It will make breeders easier to obtain proper plant height via targeted modification of genes within specific germplasm using modern gene technology [117].

4.3.Gibberellin system might create ideal plant height type

GA signaling pathway plays an important role in promoting PH and regulating flowering[118].In plants,GA is catalyzed by a complex multi-step pathway.Active GA is recognized by Gibberellin Insensitive Dwarf1 (GID1),which mediates degradation of a growth inhibitory factor DELLA [119-121].When the synthesis of gibberellin is disrupted,DELLA accumulates and inhibits plant growth,resulting in a dwarfing phenotype [122].In modern agricultural practice,fertilizer is used extensively,which makes the crops grow in vain and the stems are thin.In stormy weather,it is easy to produce lodging and lead to yield loss [123].

As gibberellin plays an important role in plant height regulation,genes in GA synthesis pathway are likely to play a role in plant height regulation [124].Several cloned genes are also involved in the GA metabolism and signal transduction includingRht-1(Rht1,Rht2,Rht3,Rht10,Rht11,Rht17),Rht8,Rht14/Rht18,Rht24andRht25(Fig.3).In the current study,wheat genes on the KEGG database were enriched and these in the GA synthesis pathway were extracted (map00904: diterpenoid biosynthesis).Four genes in the GA synthesis pathway were found in the QRC,includingTraesCS4B03G0723300(GA2ox) in QRC-4B-IV,TraesCS5A03G1269600(GA2ox) in QRC-5A-VIII,TraesCS3A03G0950800(GA 20-oxidase 2,sd1) in QRC-3A-IV,andTraesCS6A03G0611100(TaGA2ox-A9,Rht24) in QRC-6A-II.TraesCS1A03G0753800in QRC-1A-I is the orthologs of riceEUI1which encode a cytochrome P450 monooxygenase for GA deactivation [88,89].In addition toRht24that has been cloned and widely used in wheat breeding,the other three genes may also have important potential breeding values.

Fig.3.A simplified model for plant height involved in the GA pathway.Black letters in the box show non-differentially expressed genes.Red and blue letters show cloned and candidate genes in the QRC predicted by this study,respectively.

5.Conclusions

Although many genetic loci for wheat plant height have been reported and 25 dwarfing genes have been catalogued,onlyRht-B1b,Rht-D1b,Rht8,Rht24andRht25were widely used in breeding(Linked markers in the Table S7).In order to explore dominant semi-dwarf allele that can reduce PH and increase or do not reduce yield,we compiled 65 QTL-rich clusters (QRC) from 332 QTL,270 GWAS loci and 83 genes for wheat plant height.A total of 38 candidate genes were predicted for 65 QRC including four genes involved in the gibberellin biosynthesis.These candidate genes provide many targets for us to study the regulation of plant height and molecular assisted breeding for wheat.

CRediT authorship contribution statement

Dengan Xu:Project administration,Investigation,Visualization,Writing -Original Draft,Funding acquisition.Chenfei Jia:Investigation,Data curation,Writing -review &editing.Xinru Lyu:Investigation,Visualization,Writing-Original Draft.Tingzhi Yang:Investigation,Visualization,Writing -Original Draft.Huimin Qin:Investigation,Data curation.Yalin Wang:Investigation,Data curation.Qianlin Hao:Investigation,Data curation.Wenxing Liu:Investigation.Xuehuan Dai:Investigation.Jianbin Zeng:Investigation.Hongsheng Zhang:Investigation.Xianchun Xia:Writing-Review &Editing.Zhonghu He:Writing -Review &Editing.Shuanghe Cao:Writing -Review &Editing.Wujun Ma:Supervision,Project administration,Investigation,Visualization,Writing-Review &Editing,Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was funded by the National Natural Science Foundation of China (32101733),Shandong Provincial Natural Science Foundation (ZR202103020229),the High-Level Talents Project of Qingdao Agricultural University(663/1122023),and National Natural Science Foundation of China Regional Innovation and Development Joint Fund Project (U22A20457).

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2023.05.007.