Metabolome-based genome-wide association study of maize kernel leads to novel biochemical insights

Abstract

Plants produce a variety of metabolites that have a critical role in growth and development. Here we present a comprehensive study of maize metabolism, combining genetic, metabolite and expression profiling methodologies to dissect the genetic basis of metabolic diversity in maize kernels. We quantify 983 metabolite features in 702 maize genotypes planted at multiple locations. We identify 1,459 significant locus-trait associations (P≤1.8 × 10(-6)) across three environments through metabolite-based genome-wide association mapping. Most (58.5%) of the identified loci are supported by expression QTLs, and some (14.7%) are validated through linkage mapping. Re-sequencing and candidate gene association analysis identifies potential causal variants for five candidate genes involved in metabolic traits. Two of these genes were further validated by mutant and transgenic analysis. Metabolite features associated with kernel weight could be used as biomarkers to facilitate genetic improvement of maize.

Figure 1. Casual sites identification and functional validation of putrescinehydroxycinnamoyltransferase.

(a) Structure of N, N-Di-feruloylputrescine (DFP or DiFer-Put) in the polyamine pathway. (b) LC/MS fragmentation of DFP. Possible structures of the major fragments are shown. (c) Manhattan plot displaying the GWAS result of the content of DFP (MLM, N=339). (d) Regional association plot for locus PHT. The significant sites identified by re-sequencing PHT (GRMZM2G030436), show in red (MLM, N=230~320). The bigger red points show the putative functional polymorphisms, an insertion/deletion at the site InDel_17/15/0 and a SNP at Chr1.S_140321605. (e) Gene model of PHT. Filled blue boxes represent exons and UTRs. The dashed boxes mark the re-sequenced region, and the stars represent the significant sites identified by re-sequencing, the bigger stars represent the proposed functional sites. (f) A representation of the pair-wise r² value among all polymorphic sites in PHT, where the colour of each box corresponds to the r² value according to the legend. (g) Manhattan plot shows the association between expression level of PHT and genome-wide SNPs. Significant signals are mapped to SNPs within PHT, indicating a cis transcriptional regulation of this gene (MLM, N=368). The presence of the proposed functional site, InDel_17/15/0, is associated with both the expression level and the content of DFP (h,i), implying that the changed expression level is partially responsible for the change in DFP content. (h) Plot of correlation between the content of DFP and the normalized expression level of the PHT gene. Maize inbred lines with different genotypes at the InDel_17/15/0 site were shown in red, sky blue and midnight blue, respectively. The r value is based on a Pearson correlation coefficient. The P value is calculated using the t approximation. (i) Box plot for DFP content (red) and expression of PHT (sky blue); plotted as a function of genotypes at the site InDel_17/15/0. (j) Box plot for DFP content (red) and expression of PHT (sky blue), plotted as a function of genotypes at the site Chr1.S_140321605. Horizontal line represents the mean and vertical lines mark the range from 5th and 95th percentile of the total data (i,j), respectively. (k) Bar plot for DFP content and PHT expression level in rice transgenic individuals (T0). The content of DFP and expression level of PHT in the leaves of each transgenic individual is shown in red and sky blue, respectively. Vertical lines represent the s.e. (N=3).

Figure 2. Proposed pathway of polyamine conjugates biosynthesis.

The common conjugates are indicated in blue and new candidate genes in red (confirmed) and golden (not verified). ADC, arginine decarboxylase; AIH, agmatineiminohydrolase; CPA, N-carbamoylputrescineamidohydrolase; DAO, diamine oxidase; SPDS, spermidinesynthase; SPMS, spermine synthase; PAO, polyamine oxidase; PHT, putrescine: hydroxycinnamoyltransferase; GT, glycosyltransferase; CCoAOMT1, caffeoyl-CoA O- methyltransferase 1. Candidate gene revealed by the association analysis was put in the bracket under the corresponding metabolite.

Figure 3. Proposed pathway of flavonoid biosynthesis in maize kernel.

Candidate genes identified by GWAS are shown in orange, under the corresponding associated metabolites. Api, Apigenin; Chr, chrysoeriol; Lut, Luteolin; cafpen, caffeoylpentoside; couhex, coumaroylhexoside; Cya, Cyanidin; der, derivative; glc, glucose; hes, hesperetin; hex, hexose; MethylChr, Methylchrysoeriol; Mal, Malvidin; pen, pentose; rha, rhamnose; Sel, Selgin; Tri, trincin; 3′,4′,5′-Tri, 3′,4′,5′-tricetin,(eβg)eth, (erythro-β-guaiacylglyceryl)ether; (tβg)eth, (threo-β-guaiacylglyceryl)ether; 4CL, 4-coumarate-CoA ligase; CHS, chalcone synthase; CHI, chalconeisomerase; FNS, flavone synthase; F3’H, flavonoid 3'-hydroxylase; FOMT, flavonoid O-methyltransferase; bHLH, basic helix-loop-helix (GRMZM2G162382); CPSF, cleavage and polyadenylation specificity factor 73-I(GRMZM2G422649); HAD, haloaciddehalogenase-like hydrolase superfamily(GRMZM2G035651); OMT, O-methyltransferase (GRMZM2G104710); RSP, ribosomal protein (GRMZM2G344279); MADS, MADS-box family protein (GRMZM2G129034); ELMO, ELMO/CED-12 family protein (GRMZM2G031952); BAM, beta-amylase (GRMZM2G069486); DB, DNA-binding (GRMZM2G478370); ADS, AMP-dependent synthetase and ligase (GRMZM2G019746); IMI, plant invertase/pectin methylesterase inhibitor superfamily (GRMZM2G054225); P1, MYB R2R3type transcription factor (GRMZM2G084799); AQU, aquaporin NIP-type (GRMZM2G126582); GRD2, glucose/ribitol dehydrogenase (GRMZM2G170013); GRD3, glucose/ribitol dehydrogenase (GRMZM2G059361); BTF, basic transcription factor(GRMZM2G110116); ATD, acetamidase/formamidasefamily protein (GRMZM2G424857); HIS, histone superfamily protein (GRMZM2G176358); UGT88A1, UDP-glycosyltransferase 88A1 (GRMZM2G122072); ABCT, ABC transporter (GRMZM2G018074); PDHE, erythronate-4-phosphate dehydrogenase family protein (GRMZM2G177982); RSP, 60S ribosomal protein (GRMZM2G344279); OBG, GTP1/OBG family protein (GRMZM2G077632); TPR, tetratricopeptide repeat (TPR)-like superfamily protein (GRMZM2G177072); SDH, succinate dehydrogenase (GRMZM2G134134); ABCB2, ABC transporter group B2 (GRMZM2G156145); PK, pyruvate kinase (GRMZM2G119175); UGT73B5: UDP-glycosyltransferase 73B5 (GRMZM5G888620); ZF, RING/U-box superfamily protein zinc finger (GRMZM2G145104); SAMDC, S-adenosylmethionine decarboxylase proenzyme Precursor (GRMZM2G154397); NMT, histone-lysine N-methyltransferase (GRMZM2G025924); RHC1A, RING-H2 finger C1A (GRMZM2G176028); GRAM, GRAM domain family protein (GRMZM2G106622); HLY, hemolysin-III homologue (GRMZM2G114650);GH35, glycoside hydrolase, family 35 (GRMZM2G153200); GRD1, glucose/ribitol dehydrogenase (GRMZM2G076981); DPB, DNA binding and protein binding (GRMZM2G393471); HCT, hydroxycinnamoyl-CoA shikimate/quinatehydroxycinnamoyltransferase (GRMZM2G156816); CYP86A35, cytochrome P450 family 86, subfamily A, polypeptide 35 (GRMZM2G062151); UGT, UDP glycosyltransferases (GRMZM2G383404); WRKY53, superfamily of transcriptional factors having WRKY and zinc finger domains (GRMZM2G449681); RMVB, regulator of Vps4 activity in the MVB pathway protein (GRMZM2G059590); bx8, benzoxazinone synthesis 8 (GRMZM2G085054); ZFR, zinc finger, RING-CH-type (GRMZM2G358987); SDR, short-chain dehydrogenase/reductase (GRMZM2G000586); OXY, 2OG-Fe(II) oxygenase superfamily (GRMZM5G843555); PPR, PPR repeat domain containing protein (GRMZM2G325019).