Jieyun Li,Jindong Liu,Wei'e Wen,Pingzhi Zhng,Yingxiu Wn,Xinhun Xi,Yn Zhng,*,Zhonghu He,b,**

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

bInternational Maize and Wheat Improvement Center(CIMMYT)China Office,c/o CAAS,Beijing 100081,China

cCrop Research Institute,Anhui Academy of Agricultural Sciences,Hefei 230001,Anhui,China

1.Introduction

Vitamin B,one of the important microelements,is essential for maintaining normal life activities in human and animals.The vitamin B complex comprises eight water-soluble components,viz.thiamin(vitamin B1),riboflavin(vitamin B2),pantothenic acid(vitamin B3),nicotinic acid(vitamin B5),pyridoxine(vitamin B6),biotin(vitamin B7),folic acid(vitamin B9),and cobalamine(vitamin B12)that play important roles in the metabolism of carbohydrates,proteins and fats.Thiamin deficiency is associated with neurological problems,including Alzheimer's disease,cognitive deficit and encephalopathy[1–2].Riboflavin deficiency destroys mucosal membranes in the digestive system and can lead to cardiovascular disease and colorectal cancer[3–4].Instead of biosynthesizing these vitamins within their own bodies,humans and animals must obtain them from external sources in order to remain healthy.

Vitamins B1 and B2 often occur together in the same foods and were initially regarded as a single component.In cereals,the most important staple food sources,the complex vitamin B complex is concentrated in the bran and germ,with 32%to 64%of the vitamin B1 and 26%to 37%of the vitamin B2 being present in the aleurone layer and embryo,respectively[5–6].A number of studies of vitamin B1 and B2 contents in wheat have been reported.For example,Davis et al.[7]evaluated 231 cultivars grown at 49 locations over three years and determined variation in vitamins B1 and B2 levels ranging from 3.3 to 6.5 μg g?1and 1.0 to 1.7 μg g?1,respectively.Batifoulier et al.[8]determined the variation in vitamin B1(2.6–6.1 μg g?1)and vitamin B2(0.5–1.1 μg g?1)contents in 49 wheat cultivars.Shewry et al.[9]showed that there were large and significant variations in B1 and B2 contents among 24 wheat cultivars,ranging from 5.5 to 13.6 μg g?1for vitamin B1,and from 0.8 to 1.4 μg g?1for vitamin B2.Davis et al.[7]indicated that the total contents of vitamins B1 and B2 were influenced by genotype,environment and genotype×environment(G×E).However,there are no reports on QTL mapping and genome-wide association studies(GWAS)of the genetic bases of variation in vitamin B1 and B2 contents to date.

GWAS is an efficient approach to identify associations between genotypes and phenotypes in plants[10–11].For example,Rasheed etal.[12]identified 44marker-trait associations(MTAs)for nine yield and related traits using a GWAS of 123 wheat cultivars and 14,960 SNP markers.Dong et al.[13]detected 52 MTAs for stem water-soluble carbohydrate in 166 bread wheat cultivars using GWAS based on data obtained with the wheat 90 K SNP array.These studies were carried to determine genetic factors affecting complex traits.In the present study,a GWAS of vitamin B1 and B2 contents was performed using the same panel of 166 Chinese and foreign bread wheat cultivars and the wheat 90 K iSelect assay.The aim was to identify loci associated with vitamins B1 and B2 for quality improvement in bread wheat.

2.Materials and methods

2.1.Plant materials

A collection of 166 bread wheat cultivars and advanced lines from the Yellow and Huai Valley Facultative Region and foreign countries was used for the study(Table S1);144 of them were from China,nine from Italy,seven from Argentina,four from Japan,one from Australia and one from Turkey.Field trials were conducted in randomized complete blocks with three replicates in Anyang(Henan province)and Suixi(Anhui province)during the 2015–2016 cropping season,providing data for two environments.Each plot contained three 2 m rows spaced 20 cm apart.

2.2.Genotyping and quality control

Genomic DNA was extracted by a modified method according to Lagudah et al.[15];samples were sent to CapitalBio Corporation(Beijing,China;http://www.capitalbio.com/)for genotyping with the high-density illumina 90 K infinium SNP array[16].PowerMarker V3.2.5 was used to calculate gene diversity,minor allele frequency(MAF)and polymorphism information content(PIC).Genotyping and quality control was described in our previous study[13].Markers were removed if their locations in chromosomes were unknown,there were＞30%missing values,they showed a MAF of＜5%,or were represented by＞10%heterozygosis.

2.3.Milling

Thirty g kernel samples were milled using a Cyclotec 1093 Mill(Foss Tecator).The ground whole meal was stored at?20 °C prior to analysis.The water contents of the whole meal samples were measured in a drying oven at 130°C for 1 h after freezing.

2.4.Vitamin B1 and B2 extraction and determination

Vitamins B1 and B2 were extracted following Ndaw et al.[17]with minor modifications,in which the extraction solvent contents were reduced by 50%.A finely ground sample(2.5 g)was weighed into a 100 mL reagent bottle.Twenty-five ml of 0.05 mol L?1sodium acetate(pH 4.5)were added to the sample,followed by a mixture of papain(50 mg),1%glutathione(250 μL),acid phosphatase(10 mg)and α-amylase(5 mg).The sample was mixed completely and incubated in a shaker at 37°C for 20 h,then diluted with distilled water in a 50 mL volumetric flask.The supernatant was filtered through filter paper.The filtrate obtained after a second filtration through a cellulose acetate filter(0.2 μm)was used for chromatographic determination of vitamin B2.An aliquot of the first filtrate(2 mL)was added to a 10 mL tube with an alkaline solution of potassium ferricyanide(2 mL).The mixture was agitated and then left to stand for exactly 5 min.Two mL of butanol was added with vortexing,and the tube was stood for stratification.The supernatant fluid filtered through a cellulose acetate filter(0.2 μm)was used for the chromatographic determination of vitamin B1.

2.5.Chromatographic determination

A high performance liquid chromatography(HPLC)system(a 2010 Shimadzu Model)and an RF10Axl fluorescence detector(Shimadzu,Shimane,Japan)was used to determine vitamins B1 and B2 contents.Separation by HPLC was accomplished using a XTerra RP18 column(150.0 mm × 4.6 mm,5 μm,Waters Corporation,Massachusetts,USA)following Arella et al.[18].

2.6.Statistical analyses

Analysis of variance(ANOVA)was performed using PROC GLM in SAS software version 9.2(SAS Institute Inc.,Cary,NC,USA).Least square means were calculated for each parameter and used to test the significance of differences(P＜0.001)between samples.Broad-sense heritabilities(h2)were calculated following Lin and Allaire[19].

2.7.Population structure analysis

Population structure was described in our previous study[13].Briefly,Structure v.2.3.4 was used to estimate population structure based on 5624 SNP markers distributed evenly across the entire genome using Bayesian cluster analysis;markers were chosen on the basis of a MAF of over 5%and＜30%missing data,and showed heterozygosis of＜10%[20].Each K value was run repeatedly and independently to ensure the sampling variance of inferred population structure.A range of K from 1 to 10 was based on admixture and correlated allele frequencies models.Each run was carried out with 10,000 replicates for the burn-in period and 100,000 replicates during analysis.The optimum value of K was chosen by the highest ΔK[21].

Table 1–Analysis of variance of vitamin B1 and B2 contents in 166 cultivars grown in two environments.

2.8.Association analysis

Vitamin B1and B2 contents,genotype and populationstructure(Q-matrix)were implemented in TASSEL software version 5.0 using the mixed linear model(MLM)for association analysis.The significance of SNP markers was determined by a threshold P-value of 0.001[13,14],and MTAs within 5-cM intervals were declared to be the same loci according to Wang et al.[16].The distributions of observed and expected P-values.Manhattan plots were used to map SNP markers significantly associated with vitamin B1 and B2 contents.Both the Quantile-Qantile and Manhattan plots were drawn in R Language(R version 3.1.2;http://www.r-project.org/).

3.Results

3.1.Phenotypic variation

There was continuous variation for vitamin B1 and B2 contents in both environments(Fig.S1).The averaged vitamin B1 and B2 contents among cultivars ranged from 5.34 to 16.74 μg g?1and from 0.48 to 0.74 μg g?1,respectively.Cultivars Lumai 23,Xiaoyan 22,Zhengmai 366,Zhoumai 16,Nidera Baguette 10,and Nidera Baguette 20 had the highest vitamin B1 contents,whereas Aifeng 3,Xinmai 19,Xiaoyan 54,Bima 1,and Zhengzhou 3 had the highest vitamin B2 contents(Table S1).ANOVA showed significant differences for vitamins B1(P＜0.001)and B2 contents(P＜0.01)among genotypes(Table 1).No significant differences were identified among three sub-populations for vitamin B2 content,whereas the mean value of vitamin B1 content of sub-population 3 was significantly(P＜0.05)less than those of sub-populations 1 and 2 based on multiple comparison analysis(Table S2).The heritabilities of vitamin B1 and B2 contents were 0.83 and 0.77,respectively.

Table 2–SNP markers significantly associated with vitamin B1 content in the association panel.

Table 3–SNP markers significantly associated with vitamin B2 content in the association panel.

3.2.Analysis of SNP markers and population structure

Among the 81,587 SNP markers in the 90 K array,40,267(49.4%)were mapped to individual chromosomes[16].Finally,18,207(22.3%)markers were selected after a strict quality control in our association panel and were integrated into a linkage map involving all 21 wheat chromosomes.These markers covered a genetic distance of 3700 cM,with an average density of one marker per 0.2 cM.The marker density was much lower for the D genome(254.4 markers per chromosome)compared to the A(1007.7markersper chromosome)and B(1338.9 markers per chromosome)genomes.Among D genome chromosomes,and 4D had the lowest(50).The average SNP diversity(H)and PIC values were 0.35 and 0.29,respectively.

3.3.Marker-trait associations

Fig.1 –Manhattan plots from GWAS for vitamin B1 in two environments.The horizontal line depicts the 1E–03 threshold for significant association.A,Anyang;B,Suixi.

Fig.2 –Manhattan plots from GWAS for vitamin B2 in two environments.The horizontal line depicts the 1E–03 threshold for significant association.A,Anyang;B,Suixi.

Considering the criteria(P＜0.001),17 loci were significantly associated with vitamin B1 content,and 7 were associated with vitamin B2 content;these were distributed on 12 chromosomes(Tables 2 and 3).The largest numbers of MTAs were on chromosomes 4A and 6A,and no MTA was detected on chromosomes 1B,3A,3D,4D,5A,5D,6D,7A,and 7D(Figs.1 and 2).MTAs consistently identified in both environments were considered to be stable.Among them,multiple SNP markers associated with vitamin B1 were identified on chromosomes 6AS(0 cM)and 6AS(13 cM)in both environments,explaining 7.7%and 8.4%of the phenotypic variation(R2),respectively(Table 2).There were multiple SNP markers associated with vitamin B2 on chromosomes 1DS(68 cM),5BL(49 cM)and 6AS(59 cM).QQ plots for the distribution of expected and observed P-values of associated SNP markers are shown in Fig.S2.

3.4.Effects of favorable alleles on vitamins B1 and B2

Alleles with positive effects increasing vitamin B1 and B2 contents were considered to be favorable.Significantly positive correlations were observed between vitamin B1(r=0.97,P＜0.001)or vitamin B2(r=0.94,P＜0.001)contents and the number of favorable alleles(Fig.3).The numbers of favorable alleles present in a cultivar ranged from 5 to 16 for vitamin B1 content,and from 0 to 6 for vitamin B2(Table S1).

4.Discussion

4.1.Marker–trait associations for vitamins B1 and B2

The present results confirmed previous findings[8]that Vitamin B contents of cereal products are controlled mainly by genetic factors[8].Therefore,identification of QTL associated with vitamins B1 and B2 should be helpful for wheat improvement.Previously,work on vitamin B in cereals primarily focused on extraction[17],content determination[18]and synthesis pathways[22–24].In the present study,we used a GWAS approach to analyse a panel of 166 bread wheat cultivars by assaying 18,207 SNP markers to identify chromosomal regions associated with vitamin B1 and B2 contents.This is the first attempt to identify genes controlling vitamin B1 and B2 contents in wheat by GWAS.The results provide a basis for improving vitamin B1 and B2 contents.ANOVA indicated that the G×E contributed largely phenotypic variance in vitamin B1(24.53%)and vitamin B2(49.11%)contents(Table 1),resulting in some inconsistencies in MTA across environments.

Fig.3–Linear regression analysis for number of favorable alleles and vitamin B.A,vitamin B1;B,vitamin B2.

Markers IWB12483(1DL,100 cM),IWB6046(2AS,156 cM),IWB1795(2BL,146 cM),IWB11577(3B,33 cM),IWB48019(5BL,105 cM),IWB43809(6AS,0 cM)and IWB69903(6AS,13 cM)were significantly associated with vitamin B1 content.Among them,IWB12483(R2=10.6%),IWB6046(10.3%)and IWB1795(9.2%)loci explained the highest phenotypic variations.Notably,MTAs at IWB43809 and IWB69903 on chromosome 6AS were identified in both environments,indicating the QTL were stable.Markers IWB11044(1AS,51 cM),IWB23595(1DS,68 cM),IWB58793(3B,62 cM),IWB56921(4AL,75 cM),IWA8005(5BL,49 cM),IWB58995(6AS,59 cM)and IWB65016(7BL,159 cM)were significantly associated with vitamin B2 content.Among the IWB23595(R2=10.5%),IWA8005(9.0%)and IWB65016(9.7%)loci explained the highest phenotypic variations.

4.2.Putative candidate genes

The biosynthetic pathways of vitamins B1 and B2 have been well studied in prokaryotes Escherichia coli[25–26]and Bacillus subtilis[27–28],but the biosynthesis in eukaryotes is much less understood.The results of the present GWAS study provides a basis for searching for candidate genes involved in vitamin B1 and B2 biosynthesis in wheat.

In bacteria 12 genes involving 11 enzymatic steps are required for vitamin B1 biosynthesis[29–30];in yeast 19 genes are involved[31].The functions of several of these genes have been elucidated,e.g.,THI2,THI3,THI6,THI20 and THI21.One of the most important genes is THI3,which has an important role in pyrimidine biosynthesis and DNA repair[31–33],and has been isolated in Arabidopsis thaliana[22]and Solanum lycopersicum[34].However,the homologous gene in common wheat remains unidentified.The biosynthesis of one riboflavin molecule requires one molecule of GTP and two molecules of ribulose 5–phosphate.GTP cyclohydrolase II is the first committed step in biosynthesis of the key enzyme involved in riboflavin,catalyzing the opening of the imidazole ring of GTP and release of pyrophosphate[35–36].The encoding gene(rib A)has been isolated in bacteria[37]and yeast[38],but little information is available in plants.No QTL related to this gene has been identified.

4.3.Potential implications for wheat breeding

Wheat,with a total estimated production of 120 Mt.during 2015–2016(http://data.stats.gov.cn/),is one of the three most important crops in China.It is mainly consumed in human nutrition and is regarded as an important source of vitamins[39].Markers for MTAs explaining higher phenotypic variation,such as IWB12483,IWB6046,and IWB1795 identified in this study could be used for improvement of vitamin B1 in marker–assisted breeding.Markers IWB23595,IWA8005 and IWB65016 with high phenotypic variation explained could be used for improvement of vitamin B2.The cultivars Lumai 15,Jimai 19,Aifeng 3 and Bima 1 had higher contents of both vitamins B1 and B2,and therefore can be used as parents in breeding programs.There were also multiple MTAs associated with vitamin B1 on chromosomes 4A and 6A,implying that these regions are important for vitamin B1 content only.Germplasms of this type with higher numbers of favorable alleles included Zhoumai 31,Nidera Baguette 10 and Nidera Baguette 20(Table S1).Likewise,cultivars Lankao 2,Mantol and Funo have higher numbers of favorable alleles for vitamin B2 content,and these can be used as parents to improve vitamin B contents in breeding programs with the expectation of human health benefits.

Acknowledgments

This work was supported by National Key Research and Development Programs of China (2016YFD0101802,2016YFE0108600,2014CB1381050),Gene Transformation Projects(2016ZX08002003–003)and Core Research Budget of the Non–profit Governmental Research Institutions(Y2016XT06).

Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2017.08.002.

[1]G.E.Gibson,J.A.Hirsch,P.Fonzetti,B.D.Jordan,R.T.Cirio,J.Elder,Vitamin B1(thiamin)and dementia,J.Cereal Sci.1367(2016)21–30.

[2]J.L.Rodriguez,N.Qizilbash,J.Lopez-Arrieta,Thiamin for alzheimer's disease,Cochrane Database Syst.Rev.272(2001),CD001498.

[3]M.Buijssen,J.Eeuwijk,N.M.Vonk,Literature search and review related to specific preparatory work in the establishment of dietary reference values for riboflavin,EFSA Supporting Publications,11,2014,pp.245–258.

[4]Y.S.Yoon,S.Y.Jung,X.H.Zhang,S.J.Ogino,E.L.Giovannucci,E.Y.Cho,Vitamin B2 intake and colorectal cancer risk:results from the nurses'health study and the health professionals follow–up study cohort,Int.J.Cancer 139(2016)996–1008.

[5]P.M.Keagy,B.Borensterin,P.Ranum,M.A.Connor,K.Lorena,W.E.Hobbs,G.Hills,A.L.Bachman,W.A.Boyd,K.Kulp,Natural levels of nutrients in commercially milled wheat flours,Cereal Chem.57(1980)59–65.

[6]P.M.Ranum,F.F.Barrett,R.J.Loewe,K.Kulp,Nutrient levels in internationally milled wheat flours,Cereal Chem.57(1980)361–366.

[7]K.R.Davis,R.F.Cain,L.J.Peters,D.L.Tourneau,J.McGinnis,Evaluation of the nutrient composition of wheat II.Proximate analysis,thiamin,riboflavin,niacin and pyridoxine,Cereal Chem.58(1981)116–120.

[8]F.Batifoulier,M.A.Verny,E.Chanliaud,C.Rémésy,C.Demigné,Variability of B vitamin concentrations in wheat grain,milling fractions and bread products,Eur.J.Agron.25(2006)163–169.

[9]P.R.Shewry,F.V.Schaik,C.Ravel,G.Charmet,M.Rakszegi,Z.Bedo,J.L.Ward,Genotype and environment effects on the contents of vitamins B1,B2,B3,and B6 in wheat grain,J.Agric.Food Chem.59(2011)10564–10571.

[10]M.T.Hamblin,E.S.Buckler,J.K.Jannink,Population genetics of genomics–based crop improvement methods,Trends Genet.27(2011)98–106.

[11]C.S.Zhu,M.Gore,E.S.Buckler,J.M.Yu,Status and prospects of association mapping in plants,Plant Genome 1(2008)5–20.

[12]Q.Ain,A.Rasheed,A.Anwar,T.Mahmood,M.Imtiaz,T.Mahmood,X.C.Xia,Z.H.He,U.M.Quraishi,Genome–wide association for grain yield under rainfed conditions in historical wheat cultivars from Pakistan,Front.Plant Sci.6(2015)743–756.

[13]Y.Dong,Y.Zhang,A.Rasheed,Y.G.Xiao,L.P.Fu,J.Yan,J.D.Liu,W.E.Wen,Y.Zhang,R.L.Jing,X.C.Xia,Z.H.He,Genome–wide association for stem water soluble carbohydrates in bread wheat,PLoS One 11(2016),e0164293.

[14]M.Maccaferri,J.L.Zhang,P.bulli,Z.Abate,S.Chao,D.Cantu,E.Bossolini,X.M.Chen,M.Pumphrey,J.Dubcovsky,A genome–wide association study of resistance to stripe rust(Puccinia striiformis f.sp.tritici)in a worldwide collection of hexaploid spring wheat(Triticum aestivum L.),G3:Genes Genom.Genet.5(2015)449–465.

[15]E.S.Lagudah,R.Appels,A.H.D.Brown,The Nor–D3 locus of Triticum tauschii:natural variation and linkage to markers in chromosome 5,Genome 34(1991)387–395.

[16]S.C.Wang,D.B.Wong,K.Forrest,A.Allen,S.M.Chao,B.E.Huang,Characterization of polyploid wheat genomic diversity using a high–density 90,000 single nucleotide polymorphism array,Plant Biotechnol.J.12(2014)787–796.

[17]S.Ndaw,M.Bergaentzle,W.D.Aoude,C.Hasselmann,Extraction procedures for the liquid chromatographic determination of thiamin,riboflavin and vitamin B6 in foodstuffs,Food Chem.71(2000)129–138.

[18]F.Arella,S.Lahely,J.B.Bourguignon,C.Hasselmann,Liquid chromatographic determination of vitamins B1 and B2 in foods:a collaborative study,Food Chem.56(1996)81–86.

[19]C.Y.Lin,F.R.Allaire,Heritability of a linear combination of traits,Theor.Appl.Genet.51(1977)1–3.

[20]J.K.Pritchard,M.Stephens,P.Donnelly,Inference of population structure using multilocus genotype data,Genetics 155(2000)945–959.

[21]G.Evanno,S.Regnaut,J.Goudet,Detecting the number of clusters of individuals using the software structure:a simulation study,Mol.Ecol.14(2005)2611–2620.

[22]D.Y.Kong,Y.X.Zhu,H.L.Wu,X.D.Cheng,H.Liang,H.Q.Ling,AtTHIC,a gene involved in thiamin biosynthesis in Arabidopsis thaliana,Cell Res.18(2008)566–576.

[23]I.Ajjawi,Y.Tsegaye,D.Shintani,Determination of the genetic,molecular,and biochemical basis of the Arabidopsis thaliana thiamin auxotroph th1,Arch.Biochem.Biophys.459(2007)107–114.

[24]F.J.Sandoval,S.Roje,F.M.N.An,Hydrolase is fused to a riboflavin kinase homolog in plants,J.Biol.Chem.280(2005)38337–38345.

[25]M.Kriek,F.Martins,M.R.Challand,A.Croft,P.L.Roach,Thiamine biosynthesis in Escherichia coli:identification of the intermediate and by–products derived from tyrosine,Angew.Chem.-Int.Edit.46(2007)9223–9226.

[26]E.Settembre,T.P.Begley,S.E.Ealick,Structural biology of enzymes of the thiamin biosynthesis pathway,Curr.Opin.Struct.Biol.13(2003)739–747.

[27]Y.Zhang,S.V.Taylor,H.J.Chiu,T.P.Begley,Characterization of the Bacillus subtilis thiC operon involved in thiamine biosynthesis,J.Bacteriol.179(1997)3030–3035.

[28]G.Richter,M.Fischer,C.Krieger,S.Eberhardt,H.Luttgen,I.Gerstenschlager,A.Bacher,Biosynthesis of riboflavin:characterization of the bifunctional deaminase–reductase of Escherichia coli and Bacillus subtilis,J.Bacteriol.179(1997)2022–2028.

[29]T.P.Begley,D.M.Downs,S.E.Ealick,F.W.McLafferty,L.A.P.Van,S.Taylor,N.Campobasso,H.J.Chiu,C.Kinsland,J.J.Reddick,J.Xi,Thiamin biosynthesis in prokaryotes,Arch.Microbiol.171(1999)293–300.

[30]D.A.Rodionov,A.G.Vitreschak,A.A.Mironov,M.S.Gelfand,Comparative genomics of thiamin biosynthesis in prokaryotes,J.Biol.Chem.277(2002)48949–48959.

[31]K.Nosaka,Recent progress in understanding thiamin biosynthesis and its genetic regulation in Saccharomyces cerevisiae,Appl.Microbiol.Biotechnol.72(2006)30–40.

[32]P.C.Dorrestein,H.L.Zhai,S.V.Taylor,F.W.McLafferty,T.P.Begley,The biosynthesis of the thiazole phosphate moiety of thiamin(vitamin B1):the early steps catalyzed by thiazole synthase,J.Am.Chem.Soc.126(2004)3091–3096.

[33]M.Touchon,C.Hoede,O.Tenaillon,V.Barbe,S.Baeriswyl,P.Bidet,E.Bingen,S.Bonacorsi,C.Bouvet,A.Calteau,H.Calteau,O.Clermont,S.Cruveiller,A.Danchin,M.Diard,C.Dossat,M.E.Karoui,E.Frapy,L.Garry,J.M.Ghigo,A.M.Gilles,J.Johnson,C.L.Bouguenec,M.Lescat,S.Mangenot,V.Martinez-Jehanne,I.Matic,X.Nassif,S.Oztas,M.A.Petit,C.Pichon,Z.Rouy,C.S.Ruf,D.Scheider,J.Tourret,B.Vacherie,D.Vallenet,C.Medigue,E.P.Rocha,E.Denamur,Organised genome dynamics in the Escherichia coli species results in highly diverse adaptive paths,PLoS Genet.5(2009)e1000344.

[34]W.N.Zhao,X.D.Cheng,Z.A.Huang,H.J.Fan,H.L.Wu,H.Q.Ling,Tomato LeTHIC is an Fe–requiring HMP–P synthase involved in thiamin synthesis and regulated by multiple factors,Plant Cell Physiol.52(2011)967–982.

[35]M.Fischer,W.Romisch,B.IIIarionov,W.Eisenreich,A.Bacher,Structures and reaction mechanisms of riboflavin synthases of eubacterial and archaeal origin,Biochem.Soc.Trans.33(2005)780–784.

[36]J.S.Ren,M.Kotaka,M.Lockyer,H.K.Lamb,A.R.Hawkins,D.K.Stammers,GTP cyclohydrolase II structure and mechanism,J.Biol.Chem.280(2005)36912–36919.

[37]Y.S.Koh,J.Choih,J.H.Lee,J.H.Roe,Regulation of the ribA gene encoding GTP cyclohydrolase II by the soxRS locus in Escherichia coli,Mol Gen Genet 251(1996)591–598.

[38]O.Oltmanns,A.Bacher,Biosynthesis of riboflavine in Saccharomyces cerevisiae:the role of genes rib 1 and rib 7,J.Bacteriol.110(1972)818–822.

[39]H.S.Balyan,P.K.Gupta,S.Kumar,R.Dhariwal,V.Jaiswal,S.Tyagi,P.Agarwal,V.Gahlaut,S.Kumari,Genetic improvement of grain protein content and other health-related constituents of wheat grain,Plant Breed.132(2013)446–457.