A major vernalization-independent QTL for tiller angle on chromosome arm 2BL in bread wheat

Jijun Liu,Jiegung Zhou,Huping Tng,Yng Tu,Yng Mu,Lulu Gou,Qinto Jing,Yxi Liu,Guoyue Chen, Jirui Wng, Pengfei Qi, Wei Li, Yunfeng Jing, Zehong Yn, Houyng Kng,Yuming Wei, Xiujin Ln, Youling Zheng,*, Jin M,*

a State Key Laboratory of Crop Gene Exploration and Utilization in Southwest China, Triticeae Research Institute, Sichuan Agricultural University, Chengdu 611130, Sichuan, China

b College of Agronomy, Sichuan Agricultural University, Chengdu 611130, Sichuan, China

Keywords:Tiller angle QTL detection and validation Candidate genes Expression patterns

ABSTRACT Tiller angle (TA) strongly influences plant architecture and grain yield in cereals.However, the genetic basis of TA in wheat is largely unknown.We identified three TA-related quantitative trait loci (QTL).One of them was QTa.sau-2B-769, a major QTL localized on chromosome arm 2BL. QTa.sau-2B-769 was detected in seven environments,explaining 18.1%-51.1%of phenotypic variance.We developed a linked Kompetitive Allele-Specific Polymerase chain reaction(KASP)marker,KASP-AX-108792274,to further validate this locus in three additional populations in multiple environments.QTa.sau-2B-769 increased TA by up to 24.9% in these populations.There were significant and positive correlations between TA and flag leaf angle (FLANG).However, TA was not correlated with plant height or anthesis date, suggesting that expression of QTa.sau-2B-769 is independent of vernalization. TraesCS2B01G583800, a gene known to be involved in leaf angle regulation, was identified as the most likely candidate gene for QTa.sau-2B-769.These results enrich our understanding of the mechanisms regulating wheat TA at maturity and may support precise mapping and cloning of gene(s) underlying QTa.sau-2B-769.

1.Introduction

Bread wheat(Triticum aestivumL.)is a staple cereal crop worldwide and its planting area and annual production rank first among the cereal crops [1,2].A suitable plant architecture can greatly increase grain yield.Many traits affect plant structure.They include spike type, tiller number (TN), tiller angle (TA), leaf angle,and plant height(PH).TA,the angle between the main stem and its tillers, is a key agronomic trait in cereal crops for shaping plant structure and is a primary target for breeding an ideal plant architecture to improve grain yield [3,4].TA directly affects light absorption and ventilation between plants, thereby influencing crop yield and resistance to biotic stresses [5,6].Compact wheat plants with small TA typically have low photosynthetic efficiency and are more susceptible to pathogen attack.In contrast, wheat plants with large TA have a loose habit that inhibits weed growth and reduces water evaporation from soil during the juvenile growth stage (from prostrate to semi-prostrate and erect growth habit)[7].However,plants with large TA occupy more space,leading to decreased leaf photosynthetic efficiency and ventilation between plants, especially after the jointing stage.An appropriate TA will improve light capture and ventilation, reduce competition between neighboring plants, and minimize pest and disease threats to achieve optimum grain production.

Many TA-associated genes or quantitative trait loci (QTL) have been identified.Some of these genes have been cloned in rice.Yu et al.[8]identified a major TA QTL namedTAC1(Tiller Angle Control 1) and mapped it to a 35 kb region on rice chromosome 9.Loose Plant Architecture1(LPA1)regulates TA genes by affecting the gravitational response of rice [9].Sang et al.[10] reported that strigolactones attenuate shoot gravitropism by inhibiting auxin biosynthesis, thereby regulating TA in rice.ONAC106, a NAM/ATAF1/ATAF2/CUC2 transcription factor, inhibited leaf senescence and increased salt tolerance and TA in rice [11].Overexpression ofONAC106increased panicle length (PL), number of kernels per panicle (KP) and number of branches per panicle with no effect on 1000-kernel weight (TKW), TN, and grain yield per plant [11].The phytochrome-interacting factor-like protein OsPIL15 regulates TA by integrating light and gravitropism in rice [12].OsmiR167atargeted auxin response factors (OsARF12,OsARF17, andOsARF25)regulate TA by moderating auxin distribution in rice [13].These studies suggest that TA-associated genes affect agronomic traits such as KP, PH, and TN in cereals.

Some TA-related studies have been conducted in wheat.Studies on the genetic basis of wheat TA remain limited despite TA’s susceptibility to environmental conditions and variations among plants.TaTAC1, the ortholog of riceTAC1, positively regulates TA at the mRNA level by participating in the auxin polar transport process in wheat [13].Zhang [14] identified nine wheat TA QTL,explaining 1.8%-42.7% of phenotypic variations at maturity.Most recently, Marone et al.[7] analyzed the characteristic of prostrate/erect growth habit among 184 durum wheat (T.turgidumsubsp.durum) genotypes at their tillering age and identified several stably expressed QTL for TA.Zhao et al.[15] reported two major TA QTL(QTA.caas-1ALandQTA.caas-5DL)located on chromosome arms 1AL and 5DL,respectively,and suggestedTaTAC-D1as a candidate gene forQTA.caas-5DL.

This study aimed to(i)measure TA in a recombinant inbred line(RIL) population at maturity in multiple environments, (ii) genetically identify QTL for TA using a wheat 55 K single-nucleotide polymorphism (SNP) array-based linkage map, (iii) develop a linked KASP marker for the major TA QTL, (iv) validate the effects of QTL in three additional populations in various environments using the KASP marker, and (v) characterize the relationships between TA and other yield-related traits.

2.Materials and methods

2.1.Plant materials

Four wheat populations developed by single-seed descent were used.They included 20828/SY95-71 (2SY, 126 F7RILs), S849-8/SY95-71(SSY,214 F5RILs)[16],K13-868/SY95-71(KS,270 F5RILs)[17] and Shumai 1671 (SM1671)/SY95-71 (SY, 401 F2lines) [16].The 2SY population was used for QTL mapping, while the other three were used to validate the major QTL detected in the mapping population.

SY95-71 was derived from a cross between the hexaploid triticale Eronga 83 and a common wheat, Fan6 [18].It increased the QTL alleles, thus contributing to a large TA.Stable winter wheat breeding lines included 20828 (with a pedigree of G214-5/3/Chuanyu 19//Lang 9247/50788), SM1671, and S849-8.A new wheat-rye line K13-868,with high TKW and KP,was derived from progenies of wheat-ryePsathyrostachys huashanicatrigeneric hybrids[19].The four genotypes have genetic backgrounds distinct from that of SY95-71, and are characterized by small TA.

2.2.Phenotypic identification

Phenotypes of the 2SY population were measured in seven environments at four different field sites:Chongzhou (CZ, 103°38′E,30°32′N), Wenjiang (WJ, 103°51′E, 30°43′N), Ya’an (YA, 103°0′E,29°58′N) and Ma’erkang (MEK, 102°22′E, 31°92′E) in 2019 and 2020.The populations were planted in 2019CZ, 2019WJ, 2019YA,2020CZ, 2020WJ, 2020YA, and 2019MEK.SSY and KS populations were planted in 2019CZ and 2020CZ, while SY was planted in 2019WJ and 2020WJ.For 2SY, all field experiments were laid out in a randomized complete block design with two replications.Each line was planted in two rows in each environment.The planting method differed from that previously used [16,17].Seven seeds were planted in each row at a spacing of 20 cm within rows and 60 cm between rows.For the remaining populations, the planting method was the same as [16,17].Routine field management practices for wheat production were employed.TA from at least six plants of each line was measured during the heading stage following a previous method [20]:a protractor was used to measure the angle between the most distant tillers on the two sides of the culm base and half of the angle was assigned as the TA of the plant.Spikelet number per spike (SNS) was the mean of the main spikes of three independent plants from each line [21].PH was the mean height of three plants from the base to the tip of the spike(excluding awns).TKW was the mean weight of three different samples of 1000 seeds each [22].Effective tiller number (ETN), number of tillers that can ultimately produce spikes in a single wheat plant,was determined for at least three plants from each line before harvest and the mean was used for further analyses [16].Anthesis date(AD) was calculated as the time interval from the planting date to the date to flowering of the plants of a given line.Flag leaf angle(FLANG)was the angle between the straight section of the flag leaf and the internode immediately below the spike [17,23].Data sources for other traits are presented in Table S1.

2.3.Data analysis

SPSS Statistics 24(IBM,Armonk,NY,USA)was used to perform independent-samplet-test and calculate frequency distributions and Pearson correlation coefficients.SAS 8.0 (SAS Institute, Cary,NC, USA) was used to calculate best linear unbiased predictions(BLUP) and broad-sense heritability (H2) of TA among the investigated environments.Analysis of variance (ANOVA) of multienvironment trials was performed using QTL IciMapping 4.1 [24]to detect interaction between genotypes and environments.

2.4.QTL mapping

A whole-genome genetic map of the 2SY population constructed previously [16] was used for QTL mapping.The map had a length of 4273 cM with a marker density of 1.69 cM marker-1.The Biparental Populations module (BIP) of QTL IciMapping was used to detect QTL for TA by inclusive composite interval mapping.The BIP was also used to calculate the confidence interval (CI) of the QTL.A logarithm of odds (LOD) ≥3.57 based on 1000 permutations in QTL IciMapping was used as the threshold for claiming a significant QTL.The function of multi-environmental trials (METADD)of QTL IciMapping was used to indentify interaction between QTL and environments (Step = 1 cM,PIN = 0.001, and LOD= 3.57).Epistatic interactions were identified using the function of multienvironmental trials (MET-EPI) in QTL IciMapping.QTL detected in multiple environments were treated as major, and those with common and continuous CI or sharing common flanking markers were treated as a single QTL.QTL were named according to the International Rules of Genetic Nomenclature (http://wheat.pw.usda.gov/ggpages/wgc/98/Intro.htm).In the naming system,saurepresents Sichuan Agricultural University, and the trailing number represents the physical location of the QTL.

2.5.Evaluation of the effects of major QTL

SNPs located close to the LOD peak of the major QTL were selected to design KASP markers as previously [16,17] described.A linked KASP marker was used to trace the major QTL in 91 randomly selected lines from SY, 204 from SSY, and 70 from KS.Homozygous lines from each of the populations were divided into two groups:lines with homozygous alleles from SY95-71(group 1)and lines with homozygous alleles from parents other than SY95-71(group 2).Differences in mean TA between the two groups were evaluated in each population using Student’st-test (P< 0.05) to estimate the major QTL effect.

2.6.Gene annotations and comparison with previously reported QTL or genes

Sequences of flanking markers were aligned with the genomes of Chinese Spring(CS)(IWGSC RefSeq v1.0)and wild emmer(T.turgidumssp.dicoccoides)[25]to verify their physical positions.Genes between the flanking markers were subsequently retrieved from the coding sequences (IWGSC_RefSeq_Annotations_v1.0 for CS[26] and TRIDC_WEWseq_PGSB_20160501_CDS for wild emmer[25]).The genes were functionally annotated with UniProt(http://www.unipr ot.org/).

Physical intervals of QTL or genes were identified by anchoring sequences of their flanking markers on the IWGSC RefSeq v1.0 genome assembly.

2.7.Sequence analysis of TraesCS2B01G583800

Three sets of primers specific to chromosome 2B (Fig.S1;Table S2) were designed based on the genomic sequences of CS for isolation of the full-length genomic and promoter sequences of theTraesCS2B01G583800gene from CS gene annotations in 20828 and SY95-71.Amplification assays were performed in a 20-uL reaction mixture composed of 10 μL 2× Taq Master Mix(Vazyme Biotech Co., Ltd., Nanjing, China), 100 ng genomic DNA,0.5 μmol L-1of each primer, and 7.5 μL ddH2O.Polymerase chain reaction (PCR) was initiated with a pre-denaturation at 95 °C for 5 min,followed by 38 cycles of denaturation,annealing,and extension at 95°C for 30 s,56.5-58°C for 30 s,and 72°C for 45 s,respectively, and a final extension at 72 °C for 7 min.Purified PCR products were sent to Tsingke Biotech Co., Ltd., Beijing, China for gene cloning and sequencing.

3.Results

3.1.Phenotypic and correlation analyses between TA and yield traits

TA detected in 20828, SY95-71, and 2SY RIL ranged between 12.13°and 33.88°,23.90°and 38.78°,and 7.33°and 68.90°,respectively (Table 1).There were no significant (P> 0.05) differences in TA between 20828 and SY95-71 in any environment except 2019WJ and 2020CZ(Fig.1a,b;Table 1).The broad-sense heritability(H2=0.61;Table 1)and the frequency distribution(Fig.S2)suggested that TA was controlled by polygenes and affected by environmental factors.

TA was significantly and positively correlated (P< 0.05) in all environments (Table 2).However, the correlation coefficient was low because of the influence of environmental factors.ANOVA revealed that environment(E),genotype(G),and G×E interaction had a significant effect on TA (Table S3).There were no significant(P> 0.05) correlation between TA and SNS, AD, PH, TKW, and ETN(Fig.S3a-e).However,there was a significant correlation(P<0.05)between FLANG and TA (Fig.S3f).

3.2.QTL mapping for TA

Three QTL for TA were detected:one on chromosome 2B and two on 3D (Table 3).QTa.sau-2B-769was identified in all environments as well as by BLUP values and was accordingly designated as a major and stable QTL.It explained 18.1%-51.1%of the phenotypic variance (PVE).Its positive allele was derived from the parent SY95-71 (Table 3).QTa.sau-3D-603andQTa.sau-3D-607had high LOD (14.80 and 6.30) and PVE values (24.2% and 8.6%), but were detected in only one environment(Table 3).They were accordingly designated as minor QTL.Their positive alleles were derived from the parent SY95-71 and 20828, respectively (Table 3).

Of 15 QTL identified by a multi-environment QTL analysis,three(QTa.sau-2B-769,QTa.sau-3D-603, andQTa.sau-3D-607) were detected by both BIP (Table 3) and MET-ADD (Table S4).Most(80%) showed significant QTL × environment interactions and appeared to be environment-specific (Table S4).All QTL had low LOD and PVE values exceptQTa.sau-2B-769, which had a high LOD score (46.08) for the additive effect, explaining 37.5% of total phenotypic variation (44.0%) (Table S4).Three hundred epistatic QTL pairs associated with TA were identified.Their LOD and PVE values range from 5.00 to 12.23 and 0.6% to 3.4%, respectively(Table S5).Among them, 265 pairs showed negative interaction values and 35 pairs had positive interaction values (Table S5).An additive × additive (Add by Add) epistatic effect was detected betweenQTa.sau-2B-769and a locus on chromosome 5D (AX-108968976-AX-108881619, 2B/5D).The epistatic QTL pair 2B/5D had a LOD value of 5.43 and a PVE value of 1.8%.Its LOD value included that caused by the main epistasis effect (AA, 1.45) and that caused by the epistasis-by-environment interaction (AA by E,3.98).Similarly,the PVE value of epistatic QTL 2B/5D comprised AA-PVE, (0.4%) and AA by E-PVE (1.4%) (Table S5).The epistaticQTa.sau-3D-603interacted with QTL on chromosome 3D (AX-110941549-AX-108828486), 6D (AX-111368177-AX-109400957),and 7A (AX-108912046-AX-111592717).Each of the three pairs of epistatic QTL showed positive interaction(Table S5),suggesting that values of parental types were higher than those of the recombinant types.

A KASP marker (KASP-AX-108792274) was developed based on the SNPAX-108792274located between the flanking markersAX-111457622andAX-109521596(Fig.2a).The 2SY population was divided into two groups (lines with homozygous alleles from SY95-71 and 20828, respectively) based on the genotyping result from this marker.TA values of the two groups were significantly(P< 0.01) different in each environment (Fig.2b), indicating thatKASP-AX-108792274was linked withQTa.sau-2B-769.

3.3.Correlation between QTa.sau-2B-769 and other agronomic traits

The 2SY population was used to detect correlations betweenQTa.sau-2B-769and other yield-related traits.Lines of the population were divided into two groups (lines with homozygous alleles from SY95-71 (63 lines) or 20,828 (62 lines) based on genotyping withKASP-AX-108792274.There were no significant (P> 0.05) differences between the two groups for any of the yield-related traits(SNS, AD, PH, TKW, ETN, and FLANG) indicating that expression ofQTa.sau-2B-769was independent of other agronomic traits(Fig.3a-f).

3.4.Validation of the major QTL QTa.sau-2B-769

The effects ofQTa.sau-2B-769were further evaluated in three different populations (SSY, SY, and KS) using the newly designed KASP markerKASP-AX-108792274,linked toQTa.sau-2B-769.Genotyping was performed for respectively 120,204,and 96 lines of the SSY, SY, and KS population.Homozygous lines from each population were divided into two groups (groups 1 and 2) based on the genotyping results (Fig.4a, c, e).The mean TA in group 1 with the homozygous allele from the SY95-71 parent ranged between 27.15°and 34.87°,and that in group 2 with the homozygous allele from non-SY95-71 ranged between 22.34° and 31.04° (Table 4).Group 1 had significantly higher TA value ranging between 9.62%and 24.92%with a mean of 17.13%,than group 2 in the three populations (Fig.4b, d, f).

Table 1 Phenotypic variation, mean and heritability (H2) of tiller angle (TA) in the parents and 2SY population in multiple environments.

Table 2 Correlation coefficients for tiller angle (TA) in the 2SY population evaluated in multiple environments.

Table 3 Quantitative trait locus (QTL) mapping for tiller angle (TA) identified in multiple environments.

Fig.1.Phenotypesof20828,SY95-71andpartialRILs.Scalebars,10cm.

Fig.2.Partial genetic map of the major QTL QTa.sau-2B-769 and its effect.(a) genetic map of chromosome 2B containing the developed KASP marker.The black area is the QTa.sau-2B-769 interval.(b) a box plot that shows the effect of QTa.sau-2B-769 calculated after grouping the 2SY population into two groups based on the flanking marker.SY95-71 and 20,828 indicate lines with and without positive alleles of QTa.sau-2B-769.**,Significance at the 0.01 probability level.Differences in TA between the two groups are labeled below the environment names and BLUP.

3.5.Identification of candidate genes for QTa.sau-2B-769

The major and stable QTLQTa.sau-2B-769was mapped betweenAX-111457622andAX-109521596.It was located at a physical interval between 768.6 and 772.1 Mb in the deletion bin 2BL6-0.89-1.00 on chromosome arm 2BL in CS (Fig.5a, c).The genetic and physical distances separating the flanking markers were 3.22 cM and 3.5 Mb, respectively (Fig.5b, c).QTa.sau-2B-769was mapped between 766.8 and 769.6 Mb on chromosome arm 2BL in wild emmer(Fig.5c,d).There were 37 and 156 predicted genes in the corresponding intervals of CS and wild emmer,respectively.There were seven functional orthologs among the predicted genes(Fig.5; Table S6).Of the 37 annotated genes in CS,TraesCS2B01G583800is associated with the MYB transcription factor (TF, Table S6) and encodes an MYB TF that cooperates with bri1-ethylmethane sulphonate suppressor 1 (BES1) to regulate brassinosteroid-induced gene expression inArabidopsis[27].Brassinosteroids (BRs) play a critical role in regulating leaf angle[28,29].It was accordingly postulated thatTraesCS2B01G583800might be a candidate gene for TA.

4.Discussion

4.1.QTa.sau-2B-769 is a novel and major QTL

Fig.3.Student’s t-test comparing the means of two groups of lines from the 2SY RIL population carrying the allele from either 20,828 or SY95-71 at QTa.sau-2B-769 for agronomic traits.(a)Spikelet number per spike(SNS);(b)Anthesis date(AD);(c)Plant height(PH);(d)1000-kernel weight(TKW);(e)Effective tiller number(ETN);(f)Flag leaf angle(FLANG).No significant difference was detected between the two groups in any of the traits.Percentage differences in the various indexes between the two lines are indicated above the individual plots.

Previously reported QTL were collected and compared to determine whetherQTa.sau-2B-769is a new locus (Table S7).To date,only a few QTL or genes for TA have been reported in wheat.For example,Cao et al.[14]reported thatTaTAC1was located on chromosome 5D (414.1 Mb) (Table S7).Lian [30] identified two TAassociated QTL(Qta3B2andQta5B3)using a wheat 90 K SNP array.Qta3B2was mapped betweenKukri_c96747_274andKu_c663_1896(15.0-25.4 Mb) on chromosome arm 3BS andQta5B3was mapped betweenKukri_c6633_674andExcalibur_c26040_221(323.9-325.0 Mb) on chromosome arm 3BS (Table S7).Qta2Dx-1,Qta4A-1, andQta7A.3-1were mapped on chromosome arms 2DL(575.2-578.1 Mb), 4AL (566.2-566.7 Mb), and 7AL (644.2-644.8 Mb), respectively [14] (Table S7).Recently, a marker-trait association analysis (MTA) for the prostrate/erect habit identified a TA locus at the juvenile growth stage on chromosome arm 2BS[7].QTA.caas-1ALandQTA.caas-5DLhave been identified on chromosome arms 1AL and 5DL, respectively [15].Thus,QTa.sau-2B-769,mapped on chromosome arm 2BL,was different from the previously reported loci.

4.2.Expression of QTa.sau-2B-769 is unlikely to be dependent on vernalization

There was no significant correlation (P> 0.05) between TA and SNS, AD, PH, TKW, and ETN (Fig.S3).Similarly, there were no significant (P> 0.05) differences in SNS, AD, PH, TKW, and ETN between the lines with and without positive alleles ofQTa.sau-2B-769in the 2SY population(Fig.3).These results suggested that expression ofQTa.sau-2B-769is not affected by AD and PH.QTL mapping also showed that no QTL associated with SNS, PH, AD,ETN, or TKW was detected in theQTa.sau-2B-769interval [17,22].These results suggest thatQTa.sau-2B-769is not a pleiotropic locus and is not associated with any of the AD and PH-related traits.It was previously reported that TA during the juvenile growth stage was associated mainly with vernalization, dwarfing, and photoperiod genes[7].Our results suggested that expression ofQTa.sau-2B-769identified at maturity is likely independent of vernalization.There was no vernalization gene in the interval ofQTa.sau-2B-769.Rht4for PH (784.3 Mb) [31] andPpd2for photoperiod(293.7 Mb) [32] are located on chromosome 2B, but they were far away fromQTa.sau-2B-769, further confirming thatQTa.sau-2B-769may play a different regulatory role than the genes underlying TA at tillering stage.

FLANG and TA were positively and significantly correlated (P<0.05, Fig.S3f).Compact plants had upright flag leaves, whereas loose plants had a large FLANG.The transgenic plants ofAntiOsLICsimultaneously increased FLANG compared to their wild counterparts[33].BRs are involved in regulating both TA and FLANG in rice[34,35].These results suggest that TA and FLANG in wheat also have a similar regulatory mechanism.

4.3.Isolation and expression analysis of TraesCS2B01G583800

To confirm thatTraesCS2B01G583800is responsible for TA, we further isolated the promoter (2876 bp) and full-length genomic(1575 bp) sequences of this gene from the parents 20828 and SY95-71.Alignment results revealed no sequence difference in full-length genomic and promoter sequences between the two parental genotypes (Fig.S4a, b; Table S8).TraesCS2B01G583800was expressed in diverse tissues at various growth stages.It was upregulated in grain (z71) and spike (z32, z39) and expressed at relatively low levels in the (z75), leaf (z71), root (z10, z13, and z39),and stem(z65)(Fig.S4c).The role ofTraesCS2B01G583800in regulating TA in wheat awaits further investigation.Collectively, these results lay a foundation for fine mapping and gene cloning ofQTa.sau-2B-769.

Table 4 Phenotype of tiller angle (TA) in three validation populations.

4.4.Use of QTa.sau-2B-769 in wheat breeding

We identified a major QTL for TA (QTa.sau-2B-769).A linked KASP marker (KASP-AX-108792274) was developed for the QTL.Its positive alleles were from SY95-71.Wheat lines with large and small TA can thus be selected using the newly developed KASP marker.However,further field tests should be performed to determine whether wheat lines with large TA are more suitable for sunny fields and vice versa, as observed in rice [3,36].At present,we cannot breed wheat lines with moderate TA using this QTL or its linked marker.Understanding the regulatory mechanism of TA candidate genes can help us regulate TA and breed lines with moderate TA in wheat by regulating TA gene expression or plant hormone content [36-38].

Fig.4.Validation of QTa.sau-2B-769 in three populations.(a, c, e) Fluorescence PCR genotyping results of the KASP marker KASP-AX-108792274 in partial lines of the three populations(SSY,S849-8/SY95-71;SY,SM1671/SY95-71;KS,K13-868/SY95-71.Blue box(Hexachloro fluorescein)represents lines carrying the allele of S849-8(a),SM1671(c), and K13-868 (e), respectively; orange frame (6-carboxy-fluorescein) represents lines carrying the alternative alleles from SY95-71; green triangles represent lines with heterozygous alleles.(b,d,f)Effects of QTa.sau-2B-769 in the three validation populations(i.e.,SSY,SY,and KS)in different environment.**,Significance at the 0.01 probability level, *, Significance at the 0.05 probability level.The number of positive alleles from SSY, SY, and KS populations from SY95-71 are 42, 32, and 32, respectively.Percentage differences between the two groups are indicated below the environment names at the top of each plot.

Fig.5. The physical interval of QTa.sau-2B-769 and the predicted TA-related gene.(a) Deletion bin map of wheat chromosome 2B.(b) Partial genetic map of wheat chromosome 2B.(c)Physical map of CS chromosome 2B.(d)Physical map of T.turgidum ssp.dicoccoides chromosome 2B.Dotted lines indicates corresponding orthologs.The red area indicates the QTa.sau-2B-769 interval, and the black area indicates the centromere.CS, Chinese Spring.

5.Conclusions

We identified a novel, major, and stably expressed QTL for TA,and validated its effects in several populations.We identified a candidate gene in theQTa.sau-2B-769interval,and further isolated its promoter and full-length genomic sequences in two pairs of parents.The relationships between TA and other yield-related traits suggest that expression ofQTa.sau-2B-769is likely independent of vernalization.These findings shed light on the genetic relationships between TA and other agronomic traits.

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.

CRediT authorship contribution statement

Jiajun Liuperformed the entire study and participated in drafting the manuscript.Jieguang Zhou and Huaping Tangperformed phenotype measurement and data analysis.Yang Tu, Yang Mu,and Lulu Gouperformed field work and data analysis.Qiantao Jiang and Yaxi Liuhelped with data collection and analysis.Guoyue Chen,Jirui Wang,Pengfei Qi, Wei Li,and Yunfeng Jiangperformed QTL analysis and manuscript revision.Zehong Yan and Houyang Kangcreated populations.Yuming Wei and Xiujin Landiscussed the results and revised the manuscript.Youliang Zhengguided the study, and revised the manuscript.Jian Madesigned the experiments,guided the entire study,participated in data analysis, and wrote and extensively revised this manuscript.All authors participated in the research and approved the final manuscript.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (31971937 and 31970243), the International Science and Technology Cooperation and Exchanges Program of Science and Technology Department of Sichuan Province(2021YFH0083), the Applied Basic Research Programs of Science and Technology Department of Sichuan Province (2021YJ0503 and 2020YJ0140), the Key Projects of Scientific and Technological Activities for Overseas Students of Sichuan Province,and the Basic Research Project of Science and Technology Plan of Guizhou Province(ZK[2021]general 131).We appreciate Chunji Liu from Agriculture and Food in CSIRO and James C.Nelson from Kansas State University for reading and revising this manuscript.We thank the anonymous referees for critical reading and revising this manuscript.

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2021.02.013.