Combined GWAS And MiRNA Target Prediction Reveal Gene Modules Underlying Quality Trait In Rice

Rice provides a fundamental guarantee for the global food supply.It is also one of the staple foods for Chinese people.Hybrid rice planting area accounts for 57% of China’s total rice planting area.With the improvement of people’s living standards,the requirements for rice quality are getting higher and higher.How to cultivate high-yield and high-quality rice is one of the important issues in rice breeding research.To meet this demand,the researchers mined many candidate gene loci associated with rice quality traits through genomic-wide association study(GWAS)analysis,and improved rice quality by mining associated genes and applied them in molecular design breeding.However,related SNP markers in GWAS only explained a small part of the genetic information.Secondary analysis of other accumulated statistics is likely to provide valuable information.Micro RNAs(miRNAs),as endogenous non-coding RNAs,play an important regulatory role in plant growth and development.This study attempted to combine genome-wide association analysis with miRNAs and their predicted target genes to mine candidate miRNAs and target genes associated with quality traits.Meanwhile,we explore and analyze the gene regulatory networks that affected rice grain quality,expect to provide references for rice quality trait breeding.In this study,113 rice varieties consisting of 29 three-line wild-abortive type hybrid rice restorer lines and 84 rice micro-core germplasms were used as male parents,and 5 sterile rice cultivars as female parents,were crossed to produce 565(113 × 5)testcross lines based on the incomplete diallel cross mating design(north Carolina design Ⅱ,NCⅡ).The phenotypic value of male parental varieties(V),general combining ability of male parental varieties(GCA),phenotypic value of testcross lines(TC)and mid-parental heterosis value of testcross lines(HMP)were saparately conducted GWAS analysis to detect significant association loci and mine candidate miRNAs associated with 12 rice quality traits.Then the miRNA-target gene modules associated with rice quality were confirmed by target gene prediction,association analysis results screening and degradome analysis.The main results of this study are as follows:(1)Through GWAS analysis,265 candidate miRNAs associated with 12 quality traits were detected.Some candidate miRNAs’ functions have been reported.Among them,osa-miR156 simultaneously detected in grain width,grain length-width ratio and whole rice rate,and osa-miR396 detected in the whole rice rate trait,have been confirmed by other studies they have effect on their corresponding trait.(2)Target gene prediction of miRNA was performed using target gene prediction software,and 20042 pieces of miRNA and corresponding target gene information were obtained,including 171 miRNAs and 4367 target genes.Annotation of target genes indicated that osa-miR156 may affect rice grain width and grain length-width ratio by regulating genes Os SPL13,Os SPL16 and Os SPL18.osa-miR396 may affect rice turning rate by regulating gene Os GRF4,which is consistent with other research results.(3)The target genes predicted by the software were screened by the significant SNP markers associated with the traits obtained from the GWAS analysis,and 487 pieces of miRNA-target gene module information that both miRNAs and target genes were associated with the corresponding traits were obtained.The six target genes are functionally annotated,namely the osa-miR159-Osmyb5 and osa-miR159-OSINV3 modules in the amylose content trait,the osa-miR2118-GRP162 module in the grain length trait,and the osa-miR156-Os SPL16 and osa-miR156-Os SPL18 modules,the osa-miR396-Os GRF9 module in the whole-milling rate trait.Among them,the three modules corresponding to osa-miR156 and osa-miR396 have been confirmed by others studies that their functions are related to the corresponding traits.While miR159 is a conserved and stable miRNA,it usually targets the gene family that regulates the MYB transcription factor.From the structural analysis,Osmyb5 may be the target gene of osa-miR159.(4)By using the miRNA-target gene module information of 487 miRNAs and target genes associated with the corresponding traits to draw a trait-miRNA-target gene regulatory network,the results showed that osa-miR5075 was linked to 19 target genes influenced grain length traits,which indicating that osa-miR5075 may affect grain length traits of rice through regulating these target genes.(5)The target genes of candidate miRNAs were verified by degradome analysis,and 94 of the target gene prediction results were confirmed.There are 5miRNA-target gene modules whose miRNAs and target genes are associated with corresponding traits,of which 3 modules have been studied by others.It was confirmed that their functions were associated with the corresponding trait.However,the osa-miR1436-LOC＿Os11g47160 module in the amylose content trait and the osa-miR5498-LOC＿Os07g07654 module in the transparency trait have not been investigated through experiments by others yet.(6)Through the phylogenetic tree analysis of miR159 and miR2118 described in part(3),the possible functions of target miRNAs in rice were deduced from similar cultivars,and it was found that miR159 in wheat was related to the thermostability of starch synthase.The functions of miR2118 in other cultivars were the same as miR2118 in rice,both of which regulate the target gene related to anther development.From the functional analysis results,Osmyb5 may be the target gene of osa-miR159.Through epistasis analysis of some unconfirmed miR1436,miR5498,miR5075 described in part(5)and the SNPs around the corresponding target gene,it was found that epistatic effects between the SNPs in three miRNAs and the SNPs around their corresponding target genes.The constructed epistatic effect model showed that there are differences in the trait phenotype of different SNP marker genotype combinations.In conclusion,this study combined genome-wide association analysis with miRNA information to mine candidate miRNAs associated with rice quality traits,and performed target gene prediction,functional annotation,GWAS result screening,and degradome analysis and verification for these miRNAs.Important candidate miRNAs were selected for phylogenetic tree construction and SNP epistasis analysis.It provides a theoretical basis for further understanding the genetic mechanism of rice quality traits and cultivating varieties with good quality traits and strong hybrid vigor.