WO2023169490A1 - Key gene for controlling the transformation of dent corn to flint corn - Google Patents

Abstract

A method for improving and creating flint corn germplasm resources, comprising the following steps: enabling a coding gene function of an ARF transcription factor ZmARF17 in a corn genome to be lost or expression thereof to be downregulated, so that dent corn is transformed into flint corn or the flint phenotype is enhanced.

Description

Key genes controlling the transformation of dent corn into hard corn Technical field

The invention belongs to the technical field of corn breeding, and specifically relates to a key gene that controls the transformation of dent corn into hard corn and its application in improving and creating hard corn germplasm resources.

Background technique

Genetic germplasm improvement centered on utilizing hybrid vigor has always been the cornerstone of modern hybrid corn breeding. Flint and dent corn are two types of corn kernels well known to breeders and geneticists and everyone. They are an important agronomic trait in corn breeding, but this trait is caused by minor effects. Polygene control and complex molecular genetic mechanisms. Therefore, in the long history of human corn cultivation and genetic improvement, the molecular genetic mechanism of the formation of hard kernels and dent corns is still unclear, and the genes that control the formation of hard kernels and dent corns have not yet been cloned. Currently, in production, genetic improvement and germplasm innovation of durum and dent corn can only be carried out through traditional breeding methods, which is time-consuming, labor-intensive and low-efficiency. For a long time, there has been a lack of genes that can be directly applied in breeding practice to improve durum and dent corn quickly, efficiently and at low cost.

With the development of modern molecular biology, the genomes of multiple corn inbred lines have been sequenced and analyzed, including the classic dent corn B73, Mo17 and W22, tropical corn SK and K0326Y, and four important European durum corns F7 and EP1 , DK105 and PE0075 (Schnable, PS, et al. (2009). The b73 maize genome: complexity, diversity, and dynamics. Science, 326 (5956), 1112-1115. Hirsch, C., et al., (2016 ).Draft assembly of elite inbred line PH207provides insights into genomic and transcriptome diversity in maize.Plant Cell 28,2700–2714.Springer,NM,et al.,(2018)The maize W22genome provides a foundation for functional genomics and transposon biology. Nature Genetics,50,1282–1288.Yang,N.,et al.,(2019).Genome assembly of a tropical maize inbred line provides insights into structural variation and crop improvement.Nature Genetics,51,1052–1059.Li, C.,et al.,(2020).Long-read sequencing reveals genomic structural variations that underlie creation of quality protein maize.Nature Communications 11,17.Li 2020.Haberer,G.,et al.,(2020).European maize genomes highlight intraspecies variation in repeat and gene content. Nature Genetics, 52, 950–957.). This high-quality genome information reveals the secrets of maize genetic diversity and provides insights into maize genetics and breeding and important maize agriculture. Laying the foundation for genetic improvement of artistic traits. At the same time, in recent years, we have conducted research on agronomic traits related to corn kernels, including early kernel development (Yongcai, Huang, et al., (2019). Maize VKS1Regulates Mitosis and Cytokinesis During Early Endosperm Development. The Plant Cell, 31(6): 1238-1256.), corn hard mealy endosperm formation (Wang, H., et al., (2020). Carotenoids modulate kernel texture in maize by influencing amyloplast envelope integrity. Nature Communications 11,5346.), corn endosperm filling regulation A series of breakthroughs and accumulations in other aspects have laid the foundation for analyzing the molecular genetic mechanism of durum and dent corn formation.

Contents of the invention

In order to overcome major scientific issues such as the formation and evolution rules of corn germplasm resources and the molecular basis of the formation of important traits, reveal the molecular genetic mechanism of the formation of complex traits, quickly create new corn germplasm and improved inbred lines, and develop strong dominant hybrids, further To tap the potential of corn yield increase, we started in 2016. After unremitting efforts, arduous exploration and rigorous demonstration, we finally discovered and cloned a key gene Fln1 that controls the transformation of corn from dent type to hard kernel type, encoding the transcription factor ZmARF17. Its mutational inactivation will promote the accumulation of flavonoids in the seed coat and reduce the auxin content, thereby affecting the development of the seed coat and leading to the formation of hard corn, which can turn most corn into hard corn. This not only greatly increases the understanding of the formation of the important traits of hard kernels and dents, but also makes the genetic improvement of hard kernel corns and the innovation of germplasm resources more efficient and convenient.

Based on this important discovery, the first purpose of the present invention is to provide a method for improving and creating durum corn germplasm resources, including the following steps: making ARF (auxin response factor, auxin response factor) in the corn genome The function of the gene encoding the transcription factor ZmARF17 is lost or down-regulated, that is, the expression of the ARF transcription factor is inactivated or down-regulated), leading to the transformation of dent corn into hard-kernel corn, or the enhancement of the hard-kernel phenotype.

The amino acid sequence of the above-mentioned ARF transcription factor ZmARF17 (FLN1 for short) can be SEQ ID NO:1, and the NCBI accession number is Zm00001d014013; the nucleotide sequence of the encoding gene of the ARF transcription factor ZmARF17 is SEQ ID NO:2, which is named in this article Fln1.

SEQ ID NO:1 and SEQ ID NO:2 are respectively the amino acid sequence and coding gene sequence of ZmArf17 protein in dent corn B73.

The above method can be implemented in the following optional ways:

(1) Knock out the gene Fln1 encoding the ARF transcription factor in the maize genome, such as SEQ ID NO:2;

(2) Down-regulate the expression level of the ARF transcription factor encoding gene Fln1 in the maize genome, such as SEQ ID NO:2;

Or

(4) The gene encoding the ARF transcription factor and its three homologous genes are mutated at the same time, and their mutants cause the original function to be inactivated or down-regulated. The three homologous genes are SEQ ID NO: 7. ZmArf2 (NCBI accession number is Zm00001d032683, ZmArf2 gene CDS region), ZmArf19 with nucleotide SEQ ID NO:9 (NCBI accession number is Zm00001d014507, ZmArf19 gene CDS region) and ZmArf21 with nucleotide SEQ ID NO:9 (NCBI accession number is Zm00001d000358, ZmArf21 gene CDS region).

Among the above methods, the knockout of the ARF transcription factor encoding gene in method (1), the substitution of gene mutants in method (2), and the mutation of genes in method (3) can all be implemented through gene editing technology. The gene editing uses CRISPR-Cas9 system, CRISPR-Cas12 system, CRISPR-Cpf1 system, CRISPR-Cas related transposition system INTEGRATE system or CAST system.

In this article, the ARF transcription factor gene mutant is represented as fln1-m; the ZmArf2 gene is represented as arf2 ^WT , and the ZmArf2 gene mutant is represented as arf2m or arf2 ^cr ; the ZmArf19 gene is represented as arf19 ^WT , and the ZmArf19 gene mutant is represented as arf19m or arf19 ^cr ; ZmArf21 gene is represented as arf21 ^WT , and ZmArf21 gene mutant is represented as arf21m or arf21 ^cr , where "cr" is the abbreviation of the gene editing system CRISPR-Cas, which means gene mutation is achieved through gene editing technology.

Among them, method (2) can be selected from the following group:

(2-1) Mutation of the promoter region of the ARF transcription factor encoding gene Fln1 such as SEQ ID NO:2 results in down-regulation of the expression level of the ARF transcription factor encoding gene Fln1 such as SEQ ID NO:2;

(2-2) Mutation of the upstream regulatory factor of the ARF transcription factor encoding gene Fln1 such as SEQ ID NO:2 results in down-regulation of the expression level of the ARF transcription factor encoding gene Fln1 such as SEQ ID NO:2; or

(2-3) Introduce the interacting protein of ARF transcription factor into corn to change the function of the ARF transcription factor coding gene Fln1 such as SEQ ID NO:2.

Among them, the interacting protein of the ARF transcription factor described in (2-3) is Pericarp Color1, MYB40, the homologous gene of the transcription factor P1 that regulates zeaxanthin metabolism.

Preferably, the mutant of the ARF transcription factor gene in the above mode (3) and (4) is selected from the following group:

Nucleotide SEQ ID NO:3, represented as fln1-m1, the mutation method is that the mutation of G to A at position 995 in SEQ ID NO:2 causes Trp ³³² to become a stop codon;

Nucleotide SEQ ID NO:4, represented as fln1-m2, the mutation method is that the mutation of C to T at position 466 in SEQ ID NO:2 causes Gln ¹⁵⁶ to become a stop codon;

Nucleotide SEQ ID NO:5, expressed as fln1-m3, its mutation method is the deletion of base G at position 487 in SEQ ID NO:2, resulting in early termination;

Nucleotide SEQ ID NO:6, expressed as fln1-m4, its mutation method is the deletion of two bases CG at positions 486 and 487 in SEQ ID NO:2, resulting in early termination.

In one embodiment, the mutants of the three homologous genes in the above method (4) are selected from the following group:

arf2 ^cr , the mutation mode is that the deletion of base G at position 478 in SEQ ID NO:7 (NCBI accession number Zm00001d032683, ZmArf2 gene CDS region) leads to premature termination of translation;

arf19 ^cr , the mutation mode is that the deletion of base G at position 475 in SEQ ID NO:8 (NCBI accession number Zm00001d014507, ZmArf19 gene CDS region) leads to premature termination of translation;

arf21 ^cr , the mutation mode is the deletion of base G at position 496 in SEQ ID NO:9 (NCBI accession number Zm00001d000358, CDS region of ZmArf21 gene), which leads to premature termination of translation.

Preferably, the above-mentioned method of improving and creating durum corn germplasm resources can be implemented on the parents of hybrid corn, Zheng 58 (female parent) and Chang 7-2 (male parent).

For the detection of the mutants SEQ ID NOs:3-6 of the above-mentioned ARF transcription factor genes in the corn genome, and the mutants arf2 ^cr , arf19 ^cr and arf21 ^cr of the three homologous genes, the following mutant gene detection methods can be used:

1. When detecting SEQ ID NO:3, forward primer fln1-F: AAGGGTCCTGGTGGGTACAT; reverse primer fln1-R: TGCTTAGCGTGGGACTGAC. The PCR product is 638bp; the sequencing primer is fln1-F. PCR testing can be performed using ordinary PCR MIX and programs. The amplified sequence is SEQ ID NO:10, which is used as a fln1-m1 molecular marker.

2. When detecting SEQ ID NO:4, forward primer arf17-2-F: GTCCTTCAAGAACGCCGACA; reverse primer arf17-2-R: CGCTCCCGTGTCTACTACTTCC. The PCR product is 657bp; the sequencing primer is arf17-2-F. PCR testing can be performed using ordinary PCR MIX and programs. The amplified sequence is SEQ ID NO:11, which is used as a fln1-m2 molecular marker.

3. When detecting SEQ ID NO:5, forward primer arf17 ^cr -F: TACATTCCTACTCCGCTTTG; reverse primer arf17 ^cr -R: CCTGCCCTACCCTCATAC. The PCR product is 1798bp; the sequencing primer is arf17 ^cr -sequencing: CGCTCCCGTGTCTACTA; the target sequence is GTGCTCGCCAAGGACGTGCA. KOD FX Neo high-fidelity enzyme (KFX-201T, TOYOBO) and its PCR program were used for PCR detection. The sequence analysis website is: http://skl.scau.edu.cn/dsdecode/ ; the template sequence is SEQ ID NO:12, as fln1-m3 or fln1-m4 Molecular markers.

4. When detecting SEQ ID NO:6, it is the same as item "3" above, the forward primer arf17 ^cr -F: TACATTCCTACTCCGCTTTG; the reverse primer arf17 ^cr -R: CCTGCCCTACCCTCATAC. The PCR product is 1798bp; the sequencing primer is arf17 ^cr -sequencing: CGCTCCCGTGTCTACTA; the target sequence is GTGCTCGCCAAGGACGTGCA. PCR detection can use KOD FX Neo high-fidelity enzyme (KFX-201T, TOYOBO) and its PCR program. The sequence analysis website is: http://skl.scau.edu.cn/dsdecode/ ; the template sequence is SEQ ID NO:12, which is used as a fln1-m3 or fln1-m4 molecular marker.

5. When detecting arf2 ^cr , forward primer arf2 ^cr -F: ACGAAGAGGAGGAGGAGT; reverse primer arf2 ^cr -R: TCAGGTGGGCTACAAACA. The PCR product is 1349bp; the sequencing primer is arf2 ^cr -sequencing: CGCTCCCGTGTCTACTA; the target sequence is GTGCTTGCCAAGGACGTGCA. KOD FX Neo high-fidelity enzyme (KFX-201T, TOYOBO) and PCR program were used for PCR detection. The sequence analysis website is: http://skl.scau.edu.cn/dsdecode/ ; the template sequence is SEQ ID NO:13, which is used as the arf2 ^cr molecular marker.

6. When detecting arf19 ^cr , forward primer arf19 ^cr -F: CCCTTGCTTTCTTCTCAC; reverse primer arf19 ^cr -R: GCCGCCGTGGACGCCTTC. The PCR product is 1987bp; the sequencing primer is arf19 ^cr -sequencing: GTGCCTGGACCCGCAGCTGTGG; the target sequence is GTGCTCGCCAAGGACGTGCA. PCR detection can use KOD FX Neo high-fidelity enzyme (KFX-201T, TOYOBO) and PCR program. The sequence analysis website is: http://skl.scau.edu.cn/dsdecode/ ; the template sequence is SEQ ID NO:14, which is used as the arf19 ^cr molecular marker.

7. When detecting arf21 ^cr , forward primer arf21 ^cr -F: CCCGAATCCGTCGTTACCC; reverse primer arf21 ^cr -R: GCTTTGAGATGCCGTCCTT. The PCR product is 1599bp; the sequencing primer is arf21 ^cr -sequencing: CAATTCCCGCCGAAACCG; the target sequence is GTGCTCGCCAAGGACGTGCA. PCR detection can use KOD FX Neo high-fidelity enzyme (KFX-201T, TOYOBO) and PCR program. The sequence analysis website is: http://skl.scau.edu.cn/dsdecode/ ; the template sequence is SEQ ID NO:15, which is used as the arf21 ^cr molecular marker.

It is easy for those skilled in the art to understand that the above-mentioned mutation gene detection method can use a kit to perform DNA sequencing, thereby determining that the corn genome contains the above-mentioned gene mutant (mutation site). Therefore, another aspect of the present invention also provides a kit for implementing the above-mentioned mutation gene detection method, which includes the above-mentioned primers for detecting SEQ ID NOs: 3-6, arf2 ^cr , arf19 ^cr and arf21 ^cr , Or DNA/RNA probes, or DNA/RNA probe microarray chips.

The second object of the present invention is to provide a polynucleotide selected from:

(A) Any one of the mutant SEQ ID NOs: 3-6 of the above-mentioned ARF transcription factor gene;

(B) The homology between the nucleotide sequence and the nucleotide sequence shown in any one of SEQ ID NOs: 3-6 is ≥80%, preferably ≥85%, ≥90%, ≥95%, and more preferably ≥98 % polynucleotide;

(C) A nucleotide sequence complementary to the nucleotide sequence described in any one of (A) or (B).

The third aspect of the present invention is to provide the use of the above-mentioned polynucleotide, which is used to introduce into corn to replace the ARF transcription factor coding gene in the corn genome, causing the transformation of dent corn to hard kernel corn, or enhancing the hard kernel surface. type.

The present invention first discovered that the ARF transcription factor ZmARF17 (NCBI accession number Zm00001d014013) has the function of controlling the formation of hard kernel/dent corn, and the functional loss of its encoding gene Fln1, or the allelic variation fln1-m (selected from SEQ ID NOs: 3- 6) After the corn genome is introduced, the corn kernels can be changed into hard-kernel corn or the hard-kernel phenotype can be enhanced. Therefore, the regulation of gene Fln1 can be used to improve and create durum corn germplasm resources, greatly speeding up the breeding speed of durum corn genetic improvement and germplasm resource innovation, and avoiding the long-term, large-scale field exploration process with unclear prospects. Wide application prospects and huge economic value.

Description of the drawings

Figure 1 shows the phenotypic photos of the reciprocal cross of the hard-grained dent inbred line.

Figure 2 shows the mature grain phenotype and variation statistical results of dent corn B73 and EMS mutant (fln1-m1). Among them, A, E: B73; B, F: fln1-m1; C, G: B73x fln1-m1; D, H: fln1-m1x B73; I-K: The top angle I of the mature grains planted in Shanghai, the reciprocal mature grains The seed length is J, the seed width of the mature seeds of the reciprocal cross is K, the length-to-width ratio of the mature seeds of the reciprocal cross is L; M-P: the top angle of the mature seeds planted in Sanya is M; the length of the mature seeds of the reciprocal cross is N; the mature seeds of the reciprocal cross are length N. The seed width is O; the length-to-width ratio of the mature seeds of the reciprocal cross is P. Take 6 ears and no less than 50 seeds from each ear for seed testing. **,P<0.01 as determined by Student’s t test.

Figure 3 shows the grain phenotype and statistical results of the fln1-m1 mutant at different development stages. Among them A-D: B73 seeds A, fln1-m1 seeds B, longitudinal section C of B73 seeds, longitudinal section D of fln1-m1 seeds planted at 12, 15, 20, 25, 30 days after pollination (DAP) at Songjiang Farm in Shanghai. ; E-G: Seed length E, seed width F, and seed coat length G at different development stages planted in Shanghai; H-J: Seed length H, seed width I, and seed coat length J at different development stages planted in Sanya. Take 6 ears and no less than 50 seeds from each ear for seed testing. The length of the seed coat is measured on 6 ears, and no less than 6 kernels are measured on each ear. **,P<0.01as determined by Student’s t test.

Figure 4 shows the cytological changes in the testa and endosperm of mutant fln1-m1. Picture A: Transmission electron microscope section of the top testa of B73 seeds; B: Transmission electron microscope section of the top testa of fln1-m1 seeds; C: Semi-thin section of the distal embryo side of B73 seeds; D: Distal embryo of fln1-m1 seeds Semi-thin section from the side; E: Semi-thin section from the embryonic side of B73 seeds Section; F: semi-thin section of the near-embryo side of fln1-m1 seeds; G: length of top seed coat cells; H: length of seed coat cells on the far-embryo side; I: length of seed coat cells on the near-embryo side; J: whole Number of seed coat cells; K: Semi-thin and transmission electron microscope observations of the top endosperm of seeds at different development stages, showing different filling abilities. En, endosperm; Al, aleurone layer; Per, seed coat; S, amyloplast; P, protein body. **,P<0.01 as determined by Student's t test.

Figure 5 shows the BSA localization detection of gene Fln1. In the figure, A, F _2:3 localization map of B73 and fln1-m1; B, ED6 step-by-step on 10 chromosomes; C, ED6 fitting results.

Figure 6 shows the expression pattern analysis of the candidate gene ZmArf17. In the figure, A, the variation of ZmArf17 within the BSA localization interval; B, the expression level of ZmArf17 in the seed coat and endosperm at different development stages, three biological replicates; C, Western Blot analysis of ZmARF17 expression in the endosperm and seed coat at different development stages. Expression level; D, in situ hybridization analysis of ZmArf17 in 12DAP seed species; E, negative control of in situ hybridization; F, analysis of GFP transgenic maize driven by ZmArf17 promoter. fln1*, is fln1-m1; En, endosperm; Em, embryo; Per, testa; WT, wild type.

Figure 7 shows tissue expression pattern analysis of ZmArf17. In the figure, A, the expression level of ZmArf17 in different tissues of maize (relative expression level); B, the expression level of ZmArf17 in B73 and fln1-m1 seed coats at different development stages (relative expression level). Three biological replicates. En, endosperm; Em, embryo; Per, seed coat.

Figure 8 shows a combination of CRISPR knockout mutants of ZmArf17 and its three homologous genes. The figure shows the phenotypes of single gene mutants of arf2, arf17, arf19, and arf21, the phenotypes of different combinations of double mutations, triple mutations, and the phenotypes of quadruple mutations. The red line indicates the type of edit that deleted the base.

Figure 9 shows the detection results of mutant fln1-m1, arf17-2 and arf17 ^cr alleles. The bold part is the sequence of the mutation site in the wild type, and the dotted line shows the gene editing site.

Figure 10 shows the RNA-seq analysis profiles of the testa of dent corn B73 and EMS mutant fln1-m1. In the figure, A, the intersection of B73 and fln1-m1 at 12, 20, and 30DPA seed coat RNA-seq; B; KEGG analysis of the intersection of B73 and fln1-m1 at 12, 20, and 30DPA seed coat RNA-seq; C, B73 and Expression of phenylpropanoid metabolism pathway genes in seed coat RNA-seq of fln1-m1 at 12, 20, and 30 DPA; D, quantitative PCR verification of phenylpropanoid metabolism pathway genes under different seasonal planting conditions. Data are from three biological replicates.

Figure 11 shows the metabolome analysis of the testa of dent corn B73 and EMS mutant fln1-m1. The testa of 15 and 25 DAP was used for metabolome analysis, and the intersection analysis of differential metabolites in the two periods showed that the differential metabolites were specifically enriched in the flavonoid metabolic pathway. Data are from three biological replicates.

Figure 12 shows the changes in auxin content in the seed coat of mutant fln1-m1 seeds. PC, placental chalaza area; Ped, pedicel tissue; En, endosperm; Upper, seed coat tissue excluding basal tissue; Basal, seed basal tissue including PC and Ped. **,P<0.01 as determined by Student's t test.

Figure 13 shows the analysis spectrum of direct binding of FLN1 and P1 to inhibit its transcriptional activation activity. In the figure, A, BiFC analysis; B, Luciferase complementation experiment; C, Co-IP experiment; D, Luciferase transcription activation experiment of CHS promoter; E, Luciferase transcription activation experiment of DFR promoter; F, LUC transcription activation experiment in tobacco system; G, EMSA experiment proves that MYB40 can directly bind to the CHS promoter sequence, and FLN1 can enhance the binding function of MYB40; H, EMSA experiment proves that MYB40 can directly bind to the CHS promoter sequence, and FLN1 can Enhance the binding function of MYB40. **,P<0.01as determined by Student’s t test.

Figure 14 shows the phenotypes of wild-type dent corn B73 and its mutant fln1-m1 grown at the Experimental Base of Northeast Agricultural University in Harbin.

Figure 15 shows the creation of durum hybrids by introducing the gene fln1-m1 into Zheng 58 and Chang 7-2, the parents of the hybrid Zhengdan 958. In the picture A, the phenotype of the plants after introducing the gene fln1-m1 into the parents of the main cultivar Zhengdan 958 has no obvious changes compared to before improvement; B, the hybrid created after using fln1-m1 to improve the parents of Zhengdan 958. The hard-grain phenotype was significantly enhanced; C, 100-grain weight; D, moisture content; E, cob length; F, number of panicle rows, G, number of grains in a row.

Detailed ways

Hard-kernel and dent-shaped corns are important germplasm resources for modern hybrid corn breeding and are also important agronomic traits. Since hard-kernel dent corn is controlled by micro-effect polygenes and the genetic mechanism is complex, the QTL that controls its formation is difficult to clone and has not yet been reported; at the same time, the formation mechanism of hard-kernel corn is also unclear. Therefore, the genetic improvement of durum corn and the expansion of genetic germplasm through conventional breeding methods are not only time-consuming and labor-intensive, but also inefficient, and the progress is very slow and the direction is unclear.

For the first time, we discovered and cloned a key gene that controls the formation of hard/dent corn. Sequence analysis confirmed that it is the gene Fln1 encoding the ARF transcription factor ZmARF17. The NCBI accession number is Zm00001d014013. After introducing the allelic variant fln1-m (selected from SEQ ID NOs: 3-6, namely fln1-m1, fln1-m2, fln1-m3 and fln1-m4) of Fln1 function into conventional inbred lines, it was identified by PCR and sequencing The genotype of fln1-m, homozygous fln1-m can turn corn kernels into hard-kernel corn. Furthermore, we also found that the three homologous genes of ZmArf17, ZmArf2, ZmArf19, and ZmArf21, cannot cause the duplex phenotype, but the durum phenotype was more obvious after Fln1 and its homologous genes ZmArf2, ZmArf19, and ZmArf21 were simultaneously knocked out and mutated by CRISPR. .

In this article, for the sake of simplicity of description, sometimes the protein names of proteins such as the transcription factor ZmARF17 and the names of the genes encoding them (DNA) are mixed. Those skilled in the art should understand that they represent different meanings in different description situations. substance. For example, for ZmARF17 (gene), when used to describe the function or category of a transcription factor, it refers to the protein (SEQ ID NO: 1); when described as a gene, it refers to the gene encoding the transcription factor (SEQ ID NO. NO: 2), and so on, which are easily understood by those skilled in the art.

The basis for this invention includes the following processes and analysis.

Creation of durum corn EMS mutants

Starting in 2016, we created corn kernel development mutants through EMS (ethyl methane sulfonate) mutagenesis of dent-shaped corn inbred lines such as B73, A619, and W64A. The mutated seeds were selfed and the M2 generation obtained. It was found that the phenotype of the kernels on some ears was not isolated. However, when these seeds were selfed for the M3 generation, hard-grained ears would be isolated. In addition, when we used durum inbred lines and dent type A619 to create a near-isogenic line population, we found that the durum parent was used as the female parent. After crossing pollen from A619, the grains in the entire ear were durum type; conversely, A619 was used as the female parent. , the pollen of the hybrid hard-grain parent, the grains of the entire ear are dent-shaped; but the ears harvested from F1 are all dent-shaped. These genetic phenomena show that the durum type and dent type are determined by the maternal genotype, and the durum type is recessive relative to the dent type. We then sorted out and analyzed the EMS mutants accumulated over the years and obtained a series of stable genetic durum mutants. Among them, a durum mutant on the B73 background had the most obvious phenotype and the best genetic stability, and was named fln1-m1. The reciprocal crosses of maize mutants fln1-m1 and B73 are consistent with the genetic rules of hard kernels and dents mentioned above, that is, they are controlled by the maternal genotype. We then conducted a series of cytological and developmental biology studies on this mutant and found that compared with B73, fln1-m1 has shorter seed length, substantially unchanged seed width, shorter seed coat length, and shorter seed coat cells. The number of seed coat cells decreases, and the endosperm filling at the top of the seed becomes more substantial. The slowdown of fln1-m1 seed coat expansion and the enrichment of endosperm filling at the top of the seed lead to the formation of hard-grained maize.

BSA sequencing clones key genes controlling durum corn formation

We crossed fln1-m1 and B73 to obtain F1, and then selfed F1 to obtain F2. We planted more than 500 F2 seeds, numbered each individual plant and selfed, and counted the grain phenotype after harvest. The grain types of individual ears are consistent and not separated, but hard-grained and dent-shaped ears are separated, and the separation ratio is consistent with 3:1. This indicates that the durum mutant is controlled by a single gene. We selected the ears with the extreme dent phenotype and the ears with the extreme hard kernel phenotype, and then extracted DNA from the corresponding leaves for BSA mixed-pool sequencing analysis. The results showed that the only peak appeared on chromosome 5. Further analysis found that there was an auxin response factor (ARF) transcription factor ZmARF17 gene in the localization interval: NCBI accession number Zm00001d014013. A G to A base mutation occurred in the coding region. Trp ³³² becomes a stop gain codon, causing protein translation to terminate prematurely. The gene expression pattern shows that compared to roots, stems, leaves, endosperm and embryos, ZmArf17 is predominantly highly expressed in seed coat tissues at different developmental stages; and the expression level in the seed coat of the fln1-m1 mutant is significantly reduced; Western blot experiments also It shows that ZmARF17 protein is enriched in the testa and also in fln1-m1 Barely detectable in the seed coat. In addition, in situ hybridization experiments and fluorescence experiments of transgenic corn with ZmArf17 promoter driving GFP also showed that ZmArf17 is mainly highly expressed in the seed coat. Therefore, we used ZmArf17 as a candidate gene for subsequent research.

Genetic validation of candidate genes

ZmArf17 has three other homologous genes: ZmArf2, ZmArf19, and ZmArf21. In order to prevent functional redundancy, we used CRISPR to knock out these four genes simultaneously, and then isolated the single, double, triple and quadruple mutations of these four genes. Accurately identify the genotypes of different combinations of materials, and then strictly self-cross them to examine the phenotypes. It was found that in all combinations, as long as it contains a ZmArf17 loss-of-function mutation, it can lead to a durum phenotype; however, loss-of-function mutations in the other three genes, regardless of single, double or triple mutations, cannot lead to the formation of durum corn. Therefore, ZmArf17 is the key gene Fln1 that regulates hard corn formation. It was also found that when these four homologous genes ZmArf2, ZmArf17, ZmArf19, and ZmArf21 were mutated simultaneously, the hard-core phenotype was more obvious.

FLN1 affects auxin content and seed coat development by regulating the synthesis of flavonoids

In order to understand how FLN1 regulates the formation of durum corn, we conducted an in-depth analysis of the seed coat RNA-seq data of corn B73 and fln1-m1 at 12, 20 and 30 days after pollination (DAP, days after pollination) and found that three The intersection of differential genes in each developmental period is mainly enriched in the phenylpropanoid metabolism pathway. We performed RT-qPCR verification of the phenylpropanoid metabolism pathway genes expressed in the seed coat, and the results showed that the expression of these genes changed at different stages. Among them, the key enzymes of the flavonoid synthesis pathway are CHS (chalcone synthase gene), DFR (dihydro flavonol reductase gene) and two UGTs (UDP-glycosyltransferase gene, UDP). -glycosyltransferase genes) genes were significantly up-regulated in the fln1-m1 seed coat at these three developmental stages. We then conducted metabolome analysis using the seed coats of 15 and 25 DAP, and found that the intersection of differential metabolites in these two periods specifically enriched flavonoids and their derivatives. These results indicate that phenylpropanoid metabolism, especially flavonoid metabolites, plays an important role in the testa development of B73 and fln1-m1.

Studies have shown that flavonoids and flavonoid derivatives have a similar structure to the polar auxin transport inhibitor NPA (N-1-naphthylphthalamic acid), and have an inhibitory effect on auxin transport. Therefore, we further examined the changes in auxin content during the development of B73 and fln1-m1 seed coats. By measuring the auxin content, it was found that the auxin content in the entire seed coat, upper seed coat and seed base tissue of fln1-m1 decreased significantly. The decrease in auxin content inhibits the growth of seed coat cells. These results indicate that the loss of FLN1 function will promote the accumulation of flavonoids in the seed coat and reduce the auxin content, thereby affecting the development of the seed coat and leading to the formation of hard corn.

FLN1 and MYB40 combine to inhibit its transcriptional activation activity and regulate the flavonoid metabolism pathway.

In order to understand how FLN1 regulates the flavonoid metabolism pathway, we performed complementary experiments through BiFC and LUC. Experiments and pull-down experiments found that FLN1 can interact with the homologous gene MYB40 (Zm00001d040621) of P1 (Pericarp Color1), a key transcription factor that regulates zeaxanthin metabolism. Studies have shown that P1 can bind to the promoters of CHS and DFR, the key genes for flavonoid synthesis, and positively regulate the expression of these two genes to regulate the flavonoid metabolism pathway. Through LUC transcription activation experiments in maize leaf protoplasts and tobacco leaves, we proved that MYB40 can activate the CHS and DFR promoters, but FLN1 cannot activate the expression of these two genes; when FLN1 and MYB40 are co-transferred, P1 will be inhibited. Activating activity on CHS and DFR promoters. Further EMSA experiments were performed to prove that MYB40 can bind to CHS and DFR promoters in vitro, and adding FLN1 can enhance the DNA binding ability of MYB40. These results indicate that FLN1 negatively regulates the flavonoid metabolism pathway in corn seed coat by directly binding to MYB40 and inhibiting its transcriptional activation activity on flavonoid synthesis genes, thereby affecting the development of the seed coat.

Genetic Stability Test of Mutant fln1-m1 Durum Maize

We planted it at Songjiang Farm, the Center for Excellence in Molecular Plant Science of the Chinese Academy of Sciences in Shanghai, the Damao China Cotton Institute Base in Sanya, the Experimental Base of Northeast Agricultural University in Harbin, and Weifang, Shandong. The fln1-m1 durum phenotype was very stable in different years and locations. This indicates that the gene has strong genetic stability and can be applied to large-scale genetic improvement of hard corn and germplasm resource innovation.

Introducing fln1-m1 into both parents of the main cultivar Zhengdan 958 to create hard-grained hybrids

In order to test the application potential of the gene fln1-m1 mutant, we introduced fln1-m1 into Zheng 58 and Chang 7-2, the parents of the main cultivar Zhengdan 958, and then crossed them to create an improved version of Zhengdan 958. There were no obvious changes in the plant shape, 100-grain weight, cob length, number of panicle rows, and number of rows of kernels of the two improved parents and hybrids, but the kernels changed from dent type to hard kernel type, and the kernel content The amount of water dropped significantly. These results show that the fln1-m1 gene can improve traditional dent corn into hard corn and reduce the moisture content of the grain, which has important application value.

The advantages of the technical solution of the present invention are mainly reflected in the following aspects:

1. fln1-m has very strong function and good genetic stability. Introducing fln1-m hybridization into different corn inbred lines can turn their kernels into hard kernels. The theory is valid for most inbred lines, which will greatly expand the germplasm resources of hard kernels and dents. Moreover, the genetic stability of fln1-m is very good, and the phenotype is very stable in Northeast China, Huanghuaihai and Sanya.

2. The time required to improve durum corn using fln1-m is short. Since the genes and mechanisms that control the formation of durum corn are unclear, conventional genetic improvement of durum corn requires large-scale field investigations at multiple sites over many years to obtain stable materials. When using fln1-m to improve hard corn, you only need to cross it with other inbred lines, then self-cross for one generation, and obtain fln1-m homozygous plants through the molecular marker identification we developed; Cross or cross other corn pollen, the seeds produced are all hard-grained. Moreover, during the import process, at least one ear only needs to be retained for hybridization and selfing to obtain stable hard grain material, which can greatly save workload.

3. Using fln1-m to improve durum corn is simple and suitable for large-scale operations. The technology of corn hybridization and selfing is relatively simple and can be mastered by ordinary workers. We have developed molecular marker primers for fln1-m, which can easily identify the genotype of fln1-m. The corn leaf DNA extraction, PCR reaction and sequencing involved are all routine molecular experiments and can be completed in ordinary laboratories and sequencing companies.

4. The cost of using fln1-m to improve durum corn is low. Compared with traditional corn genetic improvement, using fln1-m to improve hard corn has strong effects, good stability, shortens time, and reduces workload, thus greatly saving costs. In addition, molecular experiments for fln1-m genotype identification are also routine and low-cost.

5. fln1-m can be used to create durum hybrid corn. We introduced the fln1-m genotype into the parents of the main corn varieties, and can quickly obtain hard corn hybrids while maintaining the original hybrid vigor. Promotion can create huge economic and social benefits.

The technical solutions of the present invention are explained and verified below through examples. It should be understood that the following examples are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

Example

The examples herein involve the addition amounts, contents and concentrations of various substances, and the percentages mentioned therein refer to mass percentages unless otherwise specified.

In the examples herein, if there is no specific description of the reaction temperature or operating temperature, the temperature generally refers to room temperature (15-30°C).

Materials and methods

Corn selfing, hybridization, transgenic operations, field breeding, etc. are carried out according to conventional breeding methods.

The primer synthesis and gene sequencing in the examples were completed by Shanghai Boshang Biotechnology Co., Ltd.

Molecular biology experiments in the examples include plasmid construction, enzyme digestion, ligation, competent cell preparation, transformation, culture medium preparation, etc., mainly refer to "Molecular Cloning Experiment Guide" (Third Edition), J. Sambrook, Edited by D.W. Russell (USA), translated by Huang Peitang et al., Science Press, Beijing, 2002). If necessary, specific experimental conditions can be determined through simple experiments.

PCR amplification experiments were performed according to the reaction conditions provided by the reagent supplier or the kit instructions. If necessary, it can be adjusted through simple experiments.

Example 1: Creation of genetic material

When we used hard-grain inbred lines and purdent A619 to create a near-isogenic genetic population, we found that the hard-grain parent was used as the female parent. After crossing A619 pollen, the grains in the entire ear were hard-grain type; conversely, A619 was used as the female parent. The pollen of the hybrid hard-grained parents and the grains in the entire ear are dent-shaped, see Figure 1. It shows that durum and dent are determined by the genotype of the female parent.

Example 2: Phenotypes of EMS mutants

EMS mutagenesis

(1) Ear preparation: Usually 50-100 corn plants are mutated each time, and the ears are trimmed one afternoon in advance to make the silk spinning neat the next day. When EMS is applied, the silk spinning is about 1-2cm.

(2) Preparation of EMS chemical mutagen (added in a fume hood): Add 40 μl of EMS reagent to 40 mL of mineral oil and use it to treat about 40 corn ears with good silking (plan in advance the number of plants required to be treated, Prepare reagent amounts in proportion).

(3) Prepare EMS/pollen solution: In the morning, first collect pollen in a large bag from male flowers (loose powder for more than half an hour). During this period, prepare EMS chemical mutagen (prepared in step 2), and then filter and collect pollen with a fine mesh bag. To 40mL of EMS mutagen solution, add 5-8mL of pollen to obtain an EMS/pollen solution. Mutagenize for 45 minutes. During this period, shake continuously or moderately to disperse the pollen evenly into the EMS mutagen.

(4) Pollen (smear) the ears of corn after EMS treatment: Use a brush dipped in the EMS/pollen solution and apply it to the ears of corn that spin well. After the ears are coated, put the bag in place immediately to avoid other pollen contamination. After completing the EMS mutagenesis, spray stopping agent on the body. The stopping agent is 10% sodium thiosulfate (containing 1% Tween20), which can quickly deactivate EMS into low/harmless substances.

Through the above EMS mutagenesis, we screened out a B73 background mutant with the most obvious and stable hard-core phenotype, named fln1-m1. B73 is a typical dent-shaped corn, with a concave top and a small concave angle (see A, E, and I in Figure 2); fln1-m1 is a hard-kernel type, with a convex top and a larger angle (B and F in Figure 2). , I); B73 is used as the female parent to cross fln1-m1, and the grain phenotype is similar to B73, which is dent (C, G, I in Figure 2); fln1-m1 is used as the female parent to cross B73, and the grain phenotype is Similar to fln1-m1, it is a duplex type (D, H, and I in Figure 2). It shows that the hard kernel phenotype is consistent with the genetic rules controlled by maternal genotype. At the same time, the seed length of the reciprocal hard grain is significantly shorter than that of the dent, the seed width does not change much, and the aspect ratio decreases (G-L in Figure 2). The phenotypes and test data of the reciprocal crosses grown in Sanya (M-P in Figure 2) were consistent with those grown in Shanghai (I-L in Figure 2).

Example 3: Study on seed coat development and cell biology phenotype of mutant fln1-m1

We conducted developmental biology research on this mutant fln1-m1 and found that compared with B73, the seed length of fln1-m1 was significantly shorter at different development stages after pollination, the seed width was basically unchanged, and the seed coat length was significantly shorter. and The phenotypes of plants grown in Sanya at different developmental stages were consistent with those grown in Shanghai (see Figure 3).

Cytological studies have found that the length of the seed coat cells on the top of fln1-m1 seeds is significantly shorter than that of B73 (A, B, and G in Figure 4). At the same time, the length of the seed coat cells on both sides of fln1-m1 seeds is also significantly shorter than that of B73 (Figure 4). C-D, H, I in Figure 4), and the number of seed coat cells was significantly reduced (J in Figure 4). These data indicate that fln1-m1 seed coat cells become shorter in length and the number of cells decreases, resulting in shorter seed coat length and smaller seeds. Changes in seed coat development can affect endosperm filling.

We found that starting from 20DAP, the filling of endosperm cells at the top of B73 seeds was obviously not as sufficient as that of fln1-m1; by the dehydration stage of grains at 30DAP, the amyloid bodies and protein bodies in the endosperm cells at the top of B73 began to dehydrate and aggregate together, resulting in cavities; The top seed coat cells are pulled inward, causing depressions, resulting in dent-shaped corn; while the top endosperm cells of fln1-m1 are relatively full. By 30 DAP, the endosperm cells are still not dehydrated, and the amyloid bodies and protein bodies are relatively filled, so that the top seed coat cells are fully filled. The skin is not easily sunken inwards, resulting in a hard-grained phenotype (K in Figure 4). Developmental and cell biology studies have shown that fln1-m1 seed coat expansion slows down, which is conducive to the filling of top endosperm cells and the formation of hard kernels.

Example 4: BSA sequencing and cloning of the key gene Fln1 that controls durum corn formation

We crossed fln1-m1 and B73 to create F1, and then selfed F1 to obtain F2. We planted more than 500 F2 seeds, and each individual plant was numbered and selfed. After harvest, we counted the grain phenotypes and found the grain type of individual ears. They are all consistent, and there is no separation of grain types on the ear. However, hard-grained and dent-shaped ears were separated, and the separation ratio was consistent with 3:1 (see A in Figure 5).

The ears with the extreme dent phenotype and the ears with the extreme hard kernel phenotype were selected, and DNA was extracted from the leaves corresponding to the numbers for BSA mixed pool sequencing analysis. DNA extraction, library construction, sequencing and bioinformatics analysis were entrusted to Shanghai Ouyi Biomedical Technology Co., Ltd. The BSA sequencing results showed that the only peak appeared on chromosome 5 (BD in Figure 5). Further analysis found that there was an ARF transcription factor ZmARF17 gene Zm00001d014013 in the localization interval, and a G to A (C to T) mutation in the coding region (fln1-m1 ), causing Trp ³³² to become a stop codon Stop gain, leading to premature termination of protein translation (A in Figure 6).

The gene expression pattern shows that ZmArf17 is predominantly highly expressed in seed coat tissues at different developmental stages compared to roots, stems, leaves, endosperm, embryos and seed coats (A in Figure 7); and in the seed coat of the fln1-m1 mutant The expression level decreased significantly (B in Figure 6; B in Figure 7); Western blot experiments also showed that ZmARF17 protein was enriched in the testa and was almost undetectable in the fln1-m1 testa (C in Figure 6). At the same time, in situ hybridization experiments and fluorescence experiments of transgenic corn with ZmArf17 promoter driving GFP also showed that ZmArf17 is mainly highly expressed in the seed coat (D-F in Figure 6). Therefore, we used ZmArf17 as a candidate gene for subsequent research.

Example 5: Genetic verification of whether ZmArf17 is the key gene Fln1 controlling durum corn formation

Because ZmArf17 has three other homologous genes: ZmArf2, ZmArf19, and ZmArf21. In order to prevent redundant functions We used CRISPR to knock out these four genes simultaneously, and then isolated the single, double, triple and quadruple mutations of these four genes.

Taking the CRISPR knockout of the homologous gene arf2 as an example, the CRISPR knockout caused the deletion of base G at position 478 in Zm00001d032683, ZmArf2 CDS region (i.e., SEQ ID NO: 7), resulting in early termination of translation. Includes the following steps:

5.1 CRISPR vector construction and transformation

(1) Use the online website http://cbi.hzau.edu.cn/CRISPR2/ to design the knockout target sequence, and use http://www.rgenome.net/cas-offinder/ for specificity verification. Select the homologous sequences of these four genes: GTGCTTGCCAAGGACGTGCA as the target sequence.

(2) Use primer U6Pro-F (TCGAGCTCGGTACCCGGGAAGTCGTAAAATAGTG) + ARF17-R (TTGCTATTTCTAGCTCTAAAACTGCACGTCCTTGGCGAGCAC) and primer gRNA-F (GTTTTAGAGCTAGAAATAGCAA) + U6Ter-R (GCTTGCATGCCTGCAGGCGAGGGCTAAATCGT) to introduce the target sequence into the U6 promoter and U6 terminator between Construct gRNA expression system;

(3) After recovering the two fragments respectively, use SmaI and PstI sites to homologously recombine the gRNA expression system into the Cas9 vector;

(4) The vector is transformed into Agrobacterium tumefaciens EHA105 strain after sequencing verification;

(5) The vector is used to transform corn high II B x A lines immature embryos using Agrobacterium-mediated method to obtain CRISPR transgenic corn.

5.2 Use CTAB method to extract corn DNA for subsequent identification

(1) Place corn leaves in a 2mL tube, add a steel column, treat with liquid nitrogen and then grind (60Hz, 60s).

(2) After grinding, add 0.6mL CTAB extraction buffer and mix well.

(3) Place in a 65°C oven for 60 minutes, mixing every 10-15 minutes.

(4) Take it out and leave it at room temperature for 5-10 minutes, add an equal volume of chloroform:isoamyl alcohol (24:1) to the centrifuge tube, seal it and shake it for 5 minutes.

(5) Centrifuge at room temperature at 13,000 rpm for 15 min, and pipet the supernatant into a new 1.5 mL centrifuge tube.

(6) Add an equal volume of isopropyl alcohol, mix by inverting back and forth, and place at minus 20 degrees for 20 minutes.

(7) Centrifuge at 12,000 rpm for 1 minute at room temperature and discard the supernatant. Wash the DNA pellet with 1 mL of 75% ethanol 1 to 2 times, centrifuge at 12,000 rpm for 1 min each time, and then pour out the ethanol.

(8) Centrifuge briefly, aspirate excess liquid, and dry the DNA pellet at room temperature.

(9) Add 0.3mL H2O to dissolve the DNA precipitate.

5.3 When detecting the arf2 ^cr mutation in the maize gene, the forward primer arf2 ^cr -F: ACGAAGAGGAGGAGGAGT; the reverse primer arf2 ^cr -R: TCAGGTGGGCTACAAACA. The PCR product is 1349bp; the sequencing primer is arf2 ^cr -sequencing: CGCTCCCGTGTCTACTA; the target sequence is GTGCTTGCCAAGGACGTGCA. KOD FX Neo high-fidelity enzyme (KFX-201T, TOYOBO) and PCR program were used for PCR detection. The sequence analysis website is: http://skl.scau.edu.cn/dsdecode/ ; the template sequence is SEQ ID NO:13.

According to the above method, ZmArf17, ZmArf19 and ZmArf21 mutants were constructed through CRISPR gene editing technology, and the maize mutant genes were identified. The ZmArf17 ^cr method includes CRISPR knockout, which causes the deletion of base G at position 487 of the B73 reference genome ZmArf17 (CDS region of Zm00001d014013, SEQ ID NO: 2), resulting in premature termination (fln1-m3); causing both positions 486 and 487 of the CDS region Deletion of CG bases leads to premature termination (fln1-m4). The ZmArf19 ^cr method is CRISPR knockout to make Zm00001d014507, and the deletion of base G at position 475 of the ZmArf19 CDS region (ie SEQ ID NO: 8) leads to premature termination of translation. The ZmArf21 ^cr method uses CRISPR to knock out Zm00001d000358, and the deletion of base G at position 496 in the ZmArf21 CDS region (ie SEQ ID NO: 9) leads to premature termination of translation.

Accurately identify the genotypes of corn experimental materials with different wild-type (WT) mutant (cr) combinations, and then strictly self-cross them to examine the corn kernel phenotype. It was found that in all combinations, as long as there is a loss-of-function mutation in ZmArf17, it can cause the durum phenotype; and the loss-of-function mutations in the other three genes, ZmArf2, ZmArf19, and ZmArf21, regardless of single, double, or triple mutations, cannot cause durum phenotypes. Formation. Therefore, ZmArf17 is the key gene Fln1 that regulates hard corn formation.

It was also found that when these four homologous genes ZmArf2/ZmArf17/ZmArf19/ZmArf21 were mutated simultaneously, the hard-core phenotype was more obvious (see Figure 8). At the same time, we also identified another mutation site arf17-2 (fln1-m2) in the EMS mutant fln1m. Due to the mutation from C to T, the 156th amino acid Gln becomes a stop codon. Allelic testing was performed using fln1-m1, arf17-2 (fln1-m2), and arf17 ^cr (CRISPR knockout material of ZmArf2), and it was found that different heterozygous mutation sites within this gene can cause hard corn (see Figure 9 ). It is further shown that ZmArf17 is the key gene Fln1 that controls the formation of durum corn.

Example 6: Investigating changes in gene expression in the seed coat of corn kernels

We conducted an in-depth analysis of the seed coat RNA-seq data and found that the intersection of differential genes in the three developmental stages was mainly enriched in the phenylpropanoid metabolism pathway (A and B in Figure 10). We performed RT-qPCR verification of the phenylpropanoid metabolism pathway genes expressed in the seed coat. Includes the following steps:

6.1 Extraction of RNA:

(1) Take the seed coat tissues of B73 and fln1-m1 at different development stages, freeze them quickly in liquid nitrogen, and grind them thoroughly with a prototype grinder. And quickly transferred to -80℃ ultra-low temperature refrigerator for storage.

(2) Put about 50 mg of sample into a 2 mL tube, add 400 μL of RNA extraction SDS buffer, shake thoroughly, and place on ice. See the table below for SDS buffer:

(3) Add 400 μL of phenol-chloroform (phenol is saturated in NaAc/HAc, mixed 1:1, phenol pH is 4.2, placed at 4°C, use the lower layer after layering), shake vigorously for 30 seconds, and place on ice for 5 minutes. Shake during the period, centrifuge at 12000 rpm at 4°C for 10 min, and pipet 350 μL of supernatant into a new 1.5 mL centrifuge tube.

(4) Add 1 mL of Trizol extraction solution to the supernatant tube, shake thoroughly, and place on ice for 5 minutes; then add 200 μL of chloroform to the mixed solution, shake thoroughly, place on ice for 5 minutes, centrifuge at 12,000 rpm at 4°C for 10 minutes, and aspirate 500 μL of the supernatant (do not suck into the middle layer and the organic phase below) was put into a new centrifuge tube.

(5) Add 500 μL of isopropyl alcohol to the supernatant, shake thoroughly, place on ice for 10 min, centrifuge at 12,000 rpm at 4°C for 10 min, and discard the supernatant.

(6) Add 1 mL of 70% ethanol solution, flick the precipitate to float, place on ice for 1 min, centrifuge at 12,000 rpm at 4°C for 5 min, and discard the supernatant.

(7) Centrifuge briefly and use a small pipette tip to absorb excess liquid. Dry the RNA pellet at room temperature for about 2 minutes, then add 100 μL of ddH ₂ O to dissolve the pellet.

(8) Use QIAGEN's RNeasy Plus Mini Kit to purify the above-mentioned primary RNA through the column according to the standard method of the kit, which involves using DNaseI to remove DNA, and finally dissolving it with 30 μL of RNase-free H ₂ O .

6.2 RNA-seq analysis

The library construction, sequencing and subsequent bioinformatics analysis during the RNA-seq analysis process were entrusted to Shanghai Ouyi Biomedical Technology Co., Ltd.

6.3 Quantitative RT-PCR identification of RNA-seq results

(1) Reverse transcription of RNA: Use Pomega's reverse transcription kit (ImProm-II ^TM Reverse Transcription System) to perform reverse transcription of RNA according to the standard method in the kit.

(2) Design specific primers through NCBI and perform specificity analysis, using Actin as the internal reference gene.

(3) Perform quantitative analysis according to the standard method of Takara's SYBR Green kit. Three technical replicates were performed for each sample. The above reverse-transcribed cDNA samples were uniformly diluted 8 times, and a 20 μL reaction system was used for quantitative analysis. Use SYBR Green Mix, according to a 20 μL reaction system, add 10 μL of mix, 0.5 μL of each primer, 3 μL of diluted cDNA, and 6 μL of ddH ₂ O.

(4) Use the BIO-RAD fluorescence quantitative analyzer CFX to detect gene expression using a two-step PCR amplification method. Reaction conditions: pre-denaturation 95°C, 30s; amplification: 95°C, 5s, 60°C, 35s, 40 Cycle; termination: 95℃, 15s; 60℃, 60s; 95℃, 15s.

(5) Use EXCELL 2010 and △△CT method to analyze quantitative data

6.4 Quantitative RT-PCR primers are shown in the table below

RT-qPCR results showed that the expression of these genes changed in different stages. Among them, the key enzymes of the flavonoid synthesis pathway: CHS (chalcone synthase gene), DFR (dihydro flavonol reductase gene) and two UGTs (UDP-glycosyltransferase genes) genes were significantly up-regulated in the fln1 seed coat at these three developmental stages (Figure 10 middle D).

Example 7: Investigation of flavonoids in corn seed coat

Then we commissioned Jiaxing Maiwei Metabolic Biotechnology Co., Ltd. to conduct metabolome analysis using the seed coats of 15 and 25 DAP, and found that the intersection of differential metabolites in these two periods specifically enriched flavonoids and their derivatives (Figure 11). These results indicate that phenylpropanoid metabolism, especially flavonoid molecules, plays an important role in the testa development of B73 and fln1.

Example 8: Detection of auxin content in corn seed coat

Weigh 50 mg of fresh seed coat sample, add 1 ml of 80% methanol, and extract for 4 hours; centrifuge at 15,000 rpm for 10 min, take the supernatant into a new centrifuge tube, and store in a refrigerator at -20°C overnight; centrifuge at 15,000 rpm for 10 min, take the supernatant into a new centrifuge tube, and keep at -20 ℃ refrigerator overnight; centrifuge at 15,000 rpm for 10 min, take 100 μL of the supernatant into a sample tube, and detect it with a quadrupole composite ion trap tandem mass spectrometer.

The experiment found that the content of IAA (auxin) in the entire testa, upper testa and seed base tissue of fln1-m1 was significantly reduced (F in Figure 12). The decrease in auxin content inhibits the growth of seed coat cells. These results indicate that the loss of FLN1 function will promote the accumulation of flavonoids in the seed coat and reduce the auxin content, thereby affecting the development of the seed coat and leading to the formation of hard corn.

Example 9: Study on proteins interacting with FLN1

9.1 Bimolecular fluorescence complementation (BiFC) detection:

9.1.1 Vector construction: Amplify the full-length CDS of ZmArf17 and ZmMyb40 and insert them into the pSAT4-nEYFP and pSAT4-cEYFP plasmids respectively.

9.1.2 Maize protoplasts are isolated and transformed: B73 seedlings are cultured in the dark at 28°C for 7 days, and then the leaves are cut into 1 mm size and incubated in enzymatic buffer (1.5% Cellulase R10, 0.5% Macerozyme R10, 0.4M Mannitol, 20mM KCl, 20mM MES pH 5.7, 10mM CaCl ₂ , 0.1% BSA, 5mM β-Mercaptoethanol) at 22°C for 3 hours. The protoplasts were terminated with W5 buffer (154mM NaCl, 125mM CaCl ₂ , 5mM KCl, 5mM Glucose, 0.03% MES, pH 5.7), and the protoplasts were washed and collected. Then MMG (0.4M Mannitol, 15mM MgCl2, 0.1% MES, pH 5.7) was added to adjust the protoplast concentration. Add 10 μg plasmid and 110 μL PEG (45% PEG4000, 0.2M Mannitol, 100mM CaCl ₂ ) to each 100 μL protoplast buffer for transformation. After reacting at room temperature for 15 minutes, add 440 μL W5 buffer to terminate. After removing the supernatant, add 1ml WI (20mM KCl, 0.6M Mannitol, 4mM MES pH5.7) and incubate the protoplasts for 16h.

9.1.3 Protoplasts were captured on a laser confocal scanning microscope LSM880 to capture the YFP signal.

9.2 Luciferase Complementation Imaging (LCI) Assay

9.2.1 Vector construction: Clone the CDS of ZmArf17 into JW771 (NLUC), and clone the CDS of ZmMyb40 into JW772 (CLUC). Each ORF is fused to the carboxyl-terminal half (CLUC-X) and N-terminal half of LUC, respectively. in the sequence of half (X-NLUC). Use empty 771 and 772 as negative controls.

9.2.2 Transfer the constructed vector into Agrobacterium tumefaciens GV3101.

9.2.3 Resuspend the transformed Agrobacterium cells in permeability buffer (10mM MgCl2, 10mM MES and 150mM acetosyringone), with an OD600 between 0.8-1.

9.2.4 Inject the suspension into the underside of the bentgrass leaves and culture for 2 days.

9.2.5 Inject the luciferin solution into the test leaves in the same area, and then use the Tanon-5200 chemiluminescence imaging system to detect the luciferase signal.

9.3 Dual-luciferase reporter gene (DLR) assay of maize protoplasts

9.3.1 Vector construction: Amplify the full-length CDS of ZmArf17 and ZmMyb40 and insert them into the 35S-driven effector plasmid 62SK respectively; amplify the promoter sequence and connect it to the upstream of the reporter vector pGreenII 0800LUC.

9.3.2 Then use the plasmid extraction kit NucleoBond Xtra Midi (50) to extract the plasmid, and adjust the plasmid concentration to 1μg/μL.

9.3.3 Transform corn protoplasts, see step 9.1.2.

9.3.4 After incubating the protoplasts for 16 hours, use The Reporter Assay System extracts the total protein and performs the reaction, and then analyzes the ratio of LUC and REN on a photometer (Promega 20/20).

9.4EMSA experiment

9.4.1 Construction of prokaryotic expression vector: The amplified ZmMyb40 coding sequence was cloned into the pCold vector and fused with the N-terminal GST tag. Amplify the ZmARF17 coding sequence, clone it into pET30a vector and fuse it with His tag

9.4.2 Expression and purification of recombinant protein: Transfer the prokaryotic expression vector into E. coli BL21 (DE3) (EC1002), and use 0.5mM isopropyl β-D-thiogalactopyranoside (IPTG) at 20°C. After induction for 20 h, bacterial cells were purified on Ni-NTA agarose (QIAGEN) according to the instructions.

9.4.3 Promoter probe preparation: Label the 100 bp region of the target promoter with the CAAC element with the 5' end of fluorescein FAM (primers synthesized by Shanghai Boshang Biotechnology Co., Ltd.).

9.4.4 EMSA experiment: 0.05pmol labeled probe and 100ng purified recombinant protein in 20μL reaction mixture (1μL of 1mg/ml salmon sperm DNA, 20mM Tris-HCl, pH 7.9, 5% glycerol, 0.04mg/mL bovine serum albumin, 2mM MgCl ₂ , 0.2mM dithiothreitol, 40mM KCl, ultrapure water (diluted to 20 μL) and react at room temperature for 30 minutes. Unlabeled DNA fragments were added to the reaction at 10x, 50x, and 100x for competition assays.

9.4.5 Electrophoresis: In a 4% polyacrylamide gel, use 0.5M Tris-borate-EDTA for electrophoresis at 4°C and 110V constant voltage for about 80 minutes.

9.4.6 Imaging: Fluorescence was detected using a Starion FLA-9000 instrument (FujiFilm, Japan).

9.5 Pull-down experiment:

9.5.1 Mix equal amounts of ZmARF17 recombinant protein and MYB40 recombinant protein in buffer (20mM Tris-HCl, pH 8.0, 150mM of NaCl, 0.2v/v% Triton X-100, 10v/v% glycerol, Roche (REF: 04693132001 )) Neutralized glutathione agarose beads (c650031-0010) were incubated overnight at 4°C.

9.5.2 Wash three times with washing solution (50mM Tris-HCl, pH 8.0, 140mM NaCl, 0.1% Triton X-100 and Roche (REF: 04693132001)), then remove unbound proteins, and use anti-gst (ab cloning) for bound proteins. Perform Western blotting with anti-flint (ab clone) antibody.

9.5.3 Western blotting: Separate the protein with 10% SDS-PAGE gel, then transfer it to PVDF membrane by electrophoresis, incubate with primary antibody at 4°C overnight, then incubate with secondary antibody at room temperature, and then use chemiluminescent substrate reagent (Invitrogen, catalog number WP20005), and then use the Tanon-5200 system to detect immunoreactive bands.

9.6 Primer list for vector construction:

We found through BiFC (bimolecular fluorescence complementation), LUC complementation experiment (luciferase complementation experiment) and Pull-down experiment (protein in vitro binding experiment, pull-down experiment) that FLN1 (ie ZmARF17) can interact with the key to regulating zeaxanthin metabolism. The homologous gene MYB40 of the transcription factor P1 (Pericarp Color1) interacts (A-C in Figure 13). Through LUC transcription activation experiments in maize leaf protoplasts and tobacco leaves, we have proven that MYB40 can activate the CHS and DFR promoters, but FLN1 cannot activate the expression of these two genes; when FLN1 and MYB40 are co-transferred, MYB40 will be inhibited Activating activity on CHS and DFR promoters (D-F in Figure 13). Further EMSA experiments demonstrated that MYB40 can bind to CHS and DFR promoters in vitro, and adding FLN1 can enhance the DNA binding ability of MYB40 (Figure 13, G). These results indicate that FLN1 negatively regulates the flavonoid metabolism pathway in corn seed coat by directly binding to MYB40 to inhibit its transcriptional activation activity on flavonoid synthesis genes, thereby affecting the development of the seed coat.

Example 10: Examining the performance of fln1m durum corn under different ecological conditions

Starting from 2018, we have planted wild-type dent corn B73 and its mutant fln1-m at Songjiang Farm, the Center for Excellence in Molecular Plant Science of the Chinese Academy of Sciences in Shanghai, the Damao China Cotton Institute Base in Sanya, and the Experimental Base of Northeast Agricultural University in Harbin (Figure 14). . After 4 years of investigation, it was found that the fln1-m dura phenotype was very stable in different years and locations. This shows that the gene fln1-m has strong genetic stability and can be applied to large-scale genetic improvement of hard corn and germplasm resource innovation.

Example 11: fln1m durum corn hybridization experiment

In order to test the application potential of gene Fln1 mutants, we introduced fln1-m1 into Zheng 58 and Chang 7-2, the parents of the main cultivar Zhengdan 958, and then hybridized to create an improved version of Zhengdan 958 (Figure 15). There were no obvious changes in the plant shape, 100-grain weight, cob length, number of panicle rows, and number of rows of kernels of the two improved parents and hybrids, but the kernels changed from dent type to hard kernel type, and the kernel content The amount of water dropped significantly. These results show that the fln1-m1 gene can improve traditional dent corn into hard corn and reduce the moisture content of the grain, which has important application value. Hard corn is an important type of germplasm resource utilization gene in hybrid breeding. The gene fln1-m will greatly promote the genetic improvement of hard corn and innovation of germplasm resources, and has very broad application prospects and huge economic value.

Although the above examples only use corn as an example to verify the function of the transcription factor ZmARF17, those skilled in the art should understand that without violating the idea of the present invention, those skilled in the art can make various Changes or modifications, such as application to other cereal improvements or creation of new varieties, and equivalent forms of various deformations or modifications made thereby, should also fall within the scope of the present invention.

Claims (11)

1. A method for improving and creating hard-grain corn germplasm resources, including the following steps: loss of function or down-regulation of expression of the gene encoding the ARF transcription factor ZmARF17 in the corn genome, resulting in the transformation of dent corn into hard-grain corn, or Enhanced durum phenotype.
2. The method of claim 1, wherein the amino acid sequence of the ARF transcription factor ZmARF17 is SEQ ID NO: 1, and the NCBI registration number is Zm00001d014013; the nucleotide sequence of the encoding gene of the ARF transcription factor ZmARF17 is SEQ ID NO. :2.
3. The method according to claim 1 or 2, characterized in that it is implemented in the following manner:

   (1) Knock out the gene encoding the ARF transcription factor in the maize genome;

   (2) Down-regulate the expression level of ARF transcription factor coding genes in the maize genome;

   (3) Replace the ARF transcription factor coding gene in the maize genome with a mutant of the ARF transcription factor gene that encodes a loss of function or is down-regulated; or

   (4) The gene encoding the ARF transcription factor and its three homologous genes are mutated simultaneously, and their mutants cause the original function to be inactivated or the expression to be down-regulated. The three homologous genes are SEQ ID NO: ZmArf2 of 7, ZmArf19 of SEQ ID NO:8 and ZmArf21 of SEQ ID NO:9.
4. The method according to claim 3, characterized in that mode (2) is selected from the following group:

   (2-1) Mutation of the promoter region of the gene encoding the ARF transcription factor leads to downregulation of the expression level of the gene encoding the ARF transcription factor;

   (2-2) Mutation of the upstream regulatory factor of the gene encoding the ARF transcription factor leads to downregulation of the expression level of the gene encoding the ARF transcription factor; or

   (2-3) Introduce the interacting protein of ARF transcription factor into maize to change the function of the gene encoding ARF transcription factor.
5. The method of claim 4, wherein the interacting protein of the ARF transcription factor in (2-3) is MYB40, a transcription factor that regulates zeaxanthin metabolism, Zm00001d040621.
6. The method according to claim 3, characterized in that the encoding gene for the ARF transcription factor is SEQ ID NO: 2, and the mutants of the ARF transcription factor gene in the manner (3) and (4) are selected from the following group:

   Nucleotide SEQ ID NO:3, represented as fln1-m1, the mutation method is that the mutation of G to A at position 995 in SEQ ID NO:2 causes Trp ³³² to become a stop codon;

   Nucleotide SEQ ID NO:4, represented as fln1-m2, its mutation mode is position 466 in SEQ ID NO:2 The C to T mutation causes Gln ¹⁵⁶ to become a stop codon;

   Nucleotide SEQ ID NO:5, expressed as fln1-m3, its mutation method is the deletion of base G at position 487 in SEQ ID NO:2, resulting in early termination;

   Nucleotide SEQ ID NO:6, expressed as fln1-m4, its mutation method is the deletion of two bases CG at positions 486 and 487 in SEQ ID NO:2, resulting in early termination.
7. The method of claim 3, wherein the mutants of the three homologous genes in mode (4) are selected from the following group:

   arf2 ^cr , the mutation mode is that the deletion of base G at position 478 in SEQ ID NO:7 leads to premature termination of translation;

   arf19 ^cr , the mutation mode is the deletion of base G at position 475 in SEQ ID NO:8, which leads to premature termination of translation;

   arf21 ^cr , the mutation mode is the deletion of base G at position 496 in SEQ ID NO:9, which leads to premature termination of translation.
8. A polynucleotide selected from:

   (A) Any one of the ARF transcription factor gene mutant SEQ ID NOs: 3-6 as described in claim 6;

   (B) The homology between the nucleotide sequence and the nucleotide sequence shown in any one of SEQ ID NOs: 3-6 is ≥80%, preferably ≥85%, ≥90%, ≥95%, and more preferably ≥98 % polynucleotide;

   (C) A nucleotide sequence complementary to the nucleotide sequence described in any one of (A) or (B).
9. The use of the polynucleotide as claimed in claim 8, characterized in that it is used to introduce into corn and replace the ARF transcription factor encoding gene in the corn genome.
10. A method for detecting mutant SEQ ID NOs: 3-6, arf2 ^cr , arf19 ^cr and arf21 ^cr of the genes described in claims 6 and 7 in the corn genome, comprising the following steps:

    When detecting SEQ ID NO:3, the forward primer fln1-F: AAGGGTCCTGGTGGGTACAT, the reverse primer fln1-R: TGCTTTAGCTGGGACTGAC, the PCR product is 638bp, the sequencing primer is fln1-F, PCR detection is performed with PCR MIX and program, and the amplified sequence For SEQ ID NO:10;

    When detecting SEQ ID NO:4, the forward primer arf17-2-F: GTCCTTCAAGAACGCCGACA, the reverse primer arf17-2-R: CGCTCCCGTGTCTACTACTTCC, the PCR product is 657bp; the sequencing primer is arf17-2-F, and PCR MIX is used for PCR detection. The program is carried out, and the amplified sequence is SEQ ID NO: 11;

    When detecting SEQ ID NO:5 or SEQ ID NO:6, the forward primer arf17 ^cr -F: TACATTCCTACTCCGCTTTG, the reverse primer arf17 ^cr -R: CCTGCCCTACCCTCATAC, the PCR product is 1798bp, the sequencing primer is arf17 ^cr -sequencing: CGCTCCCGTGTCTACTA, Target The sequence is GTGCTCGCCAAGGACGTGCA, and the template sequence is SEQ ID NO: 12;

    When detecting arf2 ^cr , the forward primer arf2 ^cr -F: ACGAAGAGGAGGAGGAGT, the reverse primer arf2 ^cr -R: TCAGGTGGGCTACAAACA, the PCR product is 1349bp; the sequencing primer is arf2 ^cr - sequencing: CGCTCCCGTGTCTACTA, the target sequence is GTGCTTGCCAAGGACGTGCA, and the template sequence is SEQ ID NO:13;

    When detecting arf19 ^cr , the forward primer arf19 ^cr -F: CCCCTTGCTTTCTTCTCAC, the reverse primer arf19 ^cr -R: GCCGCCGTGGACGCCTTC, the PCR product is 1987bp; the sequencing primer is arf19 ^cr - sequencing: GTGCCTGGACCCGCAGCTGTGG, the target sequence is GTGCTCGCCAAGGACGTGCA; the template sequence is SEQ ID NO:14;

    When detecting arf21 ^cr , the forward primer arf21 ^cr -F: CCCGAATCCGTCGTTACCC, the reverse primer arf21 ^cr -R: GCTTTGAGATGCCGTCCTT, the PCR product is 1599bp; the sequencing primer is arf21 ^cr - sequencing: CAATTCCCGCCGAAACCG, the target sequence is GTGCTCGCCAAGGACGTGCA; the template sequence is SEQ ID NO:15.
11. A kit for implementing the method according to claim 10, characterized in that it includes corresponding primers or DNA/RNA probes for detecting SEQ ID NOs: 3-6, arf2 ^cr , arf19 ^cr or arf21 ^cr , Or microarray chips for DNA/RNA probes.