Rice lateral floret regulation gene LF2 as well as mutant gene and application thereof

Abstract

The invention relates to the technical field of plant genetic engineering, and particularly discloses a rice lateral floret regulation gene LF2 as well as a mutant gene and application thereof. Wherein the nucleotide sequence of the rice lateral floret regulation gene LF2 is as shown in SEQ ID No.1, the amino acid sequence of the encoded protein is as shown in SEQ ID No.2, the gene interacts with OsNF-YA3, HAP3J, HAP5A, HAP5B, HAP5D and HAP5E, and the development of rice glume protection is regulated by regulating the expression of G1; the mutant gene of the rice lateral floret regulation gene LF2 is characterized in that the first basic group of the fourth intron of the LOCOs07g44210 gene is subjected to T-to-A single basic group substitution, so that mRNA (messenger ribonucleic acid) shearing is disordered to form various new transcripts, and the transcript sequences are as shown in SEQ ID No.3-6. The rice lateral floret regulation gene LF2 and the mutant gene thereof disclosed by the invention provide more evidence and basis for a'three-flower spikelet 'hypothesis and'three-flower spikelet' breeding, provide a new gene for breeding for increasing the number of grains per ear, and have very important significance for further improving the rice yield.

Description

Technical Field

[0001] The invention relates to the technical field of plant genetic engineering, and in particular to a rice lateral floret regulating gene LF2 and a mutant gene and application thereof. Background Art

[0002] Rice is a model for monocotyledonous plant research and an important food crop. Nearly half of the world's population relies on rice as their staple food. Rice yield is mainly affected by three factors: the number of effective panicles per unit area, the number of grains per panicle, and the thousand-grain weight. Among them, the number of grains per panicle is the key factor affecting rice yield. The main way to increase the number of grains per panicle is to increase the number, length, and grain density of stalks. The cloned yield-increasing genes such as Gn1a, DEP1, Ghd7, and OsSPL14 are mainly used to increase the number of stalks to achieve rice yield increase. In addition, the number of florets in a spikelet is actually an important factor affecting rice yield. Normally, a rice spikelet contains only one normal floret and two pairs of glume organs (guard glume and secondary guard glume), which eventually form a grain. In 1937, scientists proposed the "three-flower spikelet" hypothesis, believing that rice spikelets may be composed of three florets, including one terminal and two lateral florets, and the guard glume is the relic of the two lateral florets in the evolutionary process. At present, existing studies have provided evidence for this hypothesis. Related studies on some reported genes such as EG1, EG2, G1, OsMADS34 / PAP2, ASP1, NSG1 and OsMADS1 / LHS1 have revealed that the guard lemma and the lemma are homologous organs, which may be the lemma of the lateral florets that gradually degenerate. The spikelets of the gain-of-function mutant lf1 develop a new lateral floret with internal floral organs between the guard lemma and the terminal floret, providing direct evidence for the "three-flowered spikelet" hypothesis and theoretically revealing the possibility of a breeding approach to increase the total number of grains per spikelet by increasing the number of florets in the spikelet. However, the gain-of-function mutant lf1 produces a higher proportion of lateral florets but the structure is not complete and cannot bear fruit, so it cannot be directly used to breed "three-flowered spikelets". Therefore, discovering more lateral floret mutants and their regulatory genes, deeply analyzing the molecular mechanism of the production of "three-flower spikelets", and "restoring" the lateral florets to normal fertile florets through molecular design will greatly increase rice yield and will be of great significance in rice breeding research. Summary of the invention

[0003] The purpose of the present invention is to provide a rice lateral floret regulating gene LF2 and a mutant gene and application thereof.

[0004] In order to achieve the above object, the present invention provides the following technical solutions:

[0005] The present invention provides a rice lateral floret regulatory gene LF2, the nucleotide sequence of which is shown as SEQ ID No.1, and the amino acid sequence of the encoded protein is shown as SEQ ID No.2.

[0006] The present invention also provides a protein encoded by the rice lateral floret regulatory gene LF2, and its amino acid sequence is shown as SEQ ID No.2.

[0007] The present invention also provides the use of the rice lateral floret regulatory gene LF2 and the protein encoded by the rice lateral floret regulatory gene LF2 in improving rice yield.

[0008] The present invention provides a mutant gene of the rice lateral floret regulatory gene LF2, in which the first base of the fourth intron of the LOC_Os07g44210 gene undergoes a single base substitution from T to A, resulting in disordered mRNA splicing to form multiple new transcripts, and the transcript sequences are shown in SEQ ID No.3-6.

[0009] The present invention also provides transcripts of the mutant gene of the rice lateral floret regulatory gene LF2, and the transcript sequences are shown in SEQ ID No. 3-6.

[0010] The present invention also provides the use of the mutant gene of the rice lateral floret regulatory gene LF2 and the transcript of the mutant gene of the rice lateral floret regulatory gene LF2 in improving rice yield.

[0011] The beneficial effects of the present invention are as follows: the present invention discloses a gain-of-function mutant lf2, in which the spikelet auxiliary guard glume is elongated, the guard glume is elongated and lemma-like, and lateral florets or lateral spikelets are produced in the axils of the auxiliary guard glume or guard glume of some spikelets, and the lateral florets contain normal floral organ structures such as inner and outer palea, pistil, lodicule and stamen. The present invention also discloses the control gene LF2 gene of the lf2 mutant trait and its mutant gene, wherein the cDNA sequence of the LF2 gene is shown in SEQ ID No. 1, the amino acid sequence of the encoded protein is shown in SEQ ID No. 2, and the mutant gene of the LF2 gene is LOC_Os07g44210 gene, in which the first base of the fourth intron undergoes a single base substitution from T to A, and its transcript sequence is shown in SEQ ID No. 3-6. The mutant lf2, mutant trait controlling gene LF2 and LF2 mutant gene disclosed in the present invention provide more evidence and basis for the "three-flower spikelet" hypothesis and "three-flower spikelet" breeding, provide a new gene for breeding to increase the number of grains per spike, and are of great significance for further improving rice yield. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1Phenotypic analysis of class I spikelets of lf2 mutants; AB: wild-type spikelets; CF: surface structures of guard glume, lemma, palea, and secondary guard glume of wild-type; G: paraffin section of cross section of wild-type spikelet; HI: class I spikelet of lf2; JL: surface structures of elongated guard glume, lemmaized guard glume, and elongated secondary guard glume of lf2 spikelet; M: paraffin section of cross section of lf2 spikelet; NQ: expression analysis of characteristic genes of lemma and lemma. rg: secondary guard glume; sl: guard glume; le: lemma; pa: palea; st: stamen; pi: pistil; esl: elongated guard glume; lesl: lemmaized guard glume; erg: elongated secondary guard glume. Scale bars: AB and HI are 1000 μm, CF and JL are 100 μm, and G and M are 500 μm.

[0013] Figure 2 The phenotype analysis of type II spikelets of lf2 mutants; AE: wild-type spikelet; FJ: lf2 spikelet with incomplete lateral florets developed in the axils of the guard glume; KO: lf2 spikelet with complete lateral florets / spikes developed in the axils of the guard glume; PT: lf2 spikelet with lateral florets / spikes developed in the axils of the secondary guard glume. rg: secondary guard glume; sl: guard glume; le: lemma; pa: palea; lo: lodicule; st: stamen; pi: pistil; tf: terminal floret; lf: lateral floret; ls: lateral spikelet; Scale bar: 500 μm for E, J, O and T, and 1000 μm for the rest.

[0014] Figure 3 This is an analysis of the floral organ characteristics of lateral florets, where AB: stereomicroscopic observation of floral organs in wild-type florets and mutant lateral florets; CF: histological observation of floral organs in wild-type florets and mutant lateral florets; GH: pollen fertility detection of wild-type and lf2 mutants; I: expression analysis of floral organ characteristic genes in wild-type guard glumes, terminal florets and lf2 lateral florets; Scale bar: AB is 1000μm, CF is 500μm, GH is 200μm.

[0015] Figure 4 The early scanning electron microscopy observations of wild-type and lf2 mutant spikelets; AC: wild-type spikelet; DF: lf2Ⅰ spikelet, GI: lf2Ⅱ spikelet with lateral florets developed from the axils of the guard glume; JL: lf2Ⅱ spikelet with lateral florets developed from the axils of the secondary guard glume; A, G and J are the Sp3-4 period; B, H and K are the Sp5-6 period; C, I and L are the Sp7-8 period. rg: secondary guard glume; sl: guard glume; le: lemma; pa: palea; lf: lateral floret; bar = 100 μm.

[0016] Figure 5The expression pattern analysis of meristem characteristic genes OSH1 and OsMADS15 in wild type and lf2 mutant; AB: expression analysis of OSH1 gene in wild type and lf2 spikelets; CD: expression analysis of OsMADS15 gene in wild type and lf2 spikelets; black arrows indicate lateral floral meristems. fm: floral meristem; pi: pistil; lo: lodicule; scale bar = 100 μm.

[0017] Figure 6 This is the map-based cloning of the LF2 gene; A: the location and mutation site of the LF2 gene; B: transcript analysis of the wild type and lf2 mutant.

[0018] Figure 7 For the identification of complementary plants and phenotypic observation; A: Schematic diagram of complementary vector; B: Sequencing peak diagram of endogenous mutation sites; C: GFP fluorescence observation of transgenic plants; D: Phenotypic analysis of transgenic plants, scale bar = 1000μm.

[0019] Figure 8 CRISPR-cas9 knockout verification of LF2 gene; A: LF2 gene editing site and effect; B: expression analysis of LF2 gene in transgenic plants; C: phenotypic analysis of transgenic plants; Scale bar = 1000 μm.

[0020] Fig. 9 Overexpression analysis of LF2 gene; A: qPCR analysis of LF2 gene in overexpression plants; B: spikelet phenotype of overexpression plants; scale bar = 1000 μm.

[0021] Fig.10 Expression pattern analysis of LF2 gene; A: expression of LF2 gene in various rice tissues; B: in situ hybridization analysis of LF2 gene; scale: BC is 500μm, DI is 100μm.

[0022] Fig.11 This is the sequence and structure analysis of LF2 protein; A: Schematic diagram of LF2 protein domain; B: Sequence analysis of LF2 homologous proteins in different species.

[0023] Fig.12 Phylogenetic analysis of the LF2 gene.

[0024] Fig.13 Subcellular localization of LF2 protein; A: fluorescence signal observation in LF2 overexpressing plants; B: subcellular localization of LF2 protein in rice protoplasts; scale bar = 50 μm.

[0025] Fig.14Analysis of the transcriptional activity of LF2 protein; A: LF2 protein self-activation detection; B: dual luciferase system detection of LF2 protein transcriptional activity.

[0026] Fig.15 Screening of LF2 interacting proteins from cDNA library; A: screening results of rice transcription factor library; B: yeast two-hybrid interaction verification, 53+T7-T: pGBKT7-53+pGADT7-T, as positive control, 53+lam: pGBKT7-53+pGADT7-lam, as negative control; C: BiFC verification of the interaction between LF2 and HAP2H; scale bar = 50μm.

[0027] Fig.16 Phenotypic analysis of OsNF-YA3 knockout plants.

[0028] Fig.17 To screen the interaction between LF2 protein and rice nuclear factor; among them, 53+T7-T: pGBKT7-53+pGADT7-T, as a positive control, 53+lam: pGBKT7-53+pGADT7-lam, as a negative control.

[0029] Fig.18 lf2 transcriptome sequencing analysis; A: transcriptome sequencing differential gene enrichment; B: differential gene qPCR verification.

[0030] Fig.19 GO and KEGG analysis of transcriptome data; A: GO enrichment analysis; B: KEGG analysis.

[0031] Fig. 20 This is the expression analysis of genes related to the auxin signaling pathway; A: Determination of auxin content in the wild type and lf2 mutant; B: Expression analysis of auxin synthesis genes; C: Expression analysis of auxin transport related genes; D: Expression analysis of auxin response inhibitors; E: Expression analysis of auxin response factors.

[0032] Fig.21 Binding analysis of LF2 and G1; A: expression analysis of G1 in lf2; B: detection of H3K27me3 level of G1 gene; C: phenotypic analysis of g1+lf2 double mutant; scale bar = 1000 μm. DETAILED DESCRIPTION

[0033] The technical solution of the present invention will be further described in detail below in conjunction with specific embodiments. It should be noted that the experimental materials in the embodiments of the present invention can all be obtained commercially, and the experimental methods in the embodiments of the present invention that do not specify specific conditions are usually carried out under conventional conditions or under conditions recommended by the experimental material manufacturer. The experimental materials involved in the present invention are specifically described as follows:

[0034] 1.1 Plant materials

[0035] The lateral flower development mutant lf2 and the wild-type maintainer indica rice Xida 1B plant in the present invention are both derived from the Rice Research Institute of Southwest University. Among them, the lateral flower development mutant lf2 is obtained by mutagenesis of the maintainer indica rice Xida 1B (Oryza sativaL.ssp.indica) through ethyl methane sulfonate (EMS), and the lf2 mutant trait can be stably inherited after multiple generations of self-pollination. Among them, the maintainer indica rice Xida 1B plant is used as a wild-type control for mutant trait investigation and other related experiments. The rice materials used in the present invention are all planted in the Xiema planting base of the Rice Research Institute of Southwest University in Chongqing.

[0036] 1.2 Main strains and vectors

[0037] The strains involved in the present invention mainly include Escherichia coli strain DH5α, Agrobacterium tumefaciens strain GV3101 and GAL4 system yeast two-hybrid experiment strain Y2HGold, all of which are provided by the Rice Research Institute of the College of Agriculture and Biotechnology, Southwest University. The vectors involved mainly include: TA cloning vector PMD19-T purchased from TaKaRa Bioengineering Co., Ltd. (Dalian); plant expression vectors pCAMBIA1301-GFP(S65T)-NOS, pCAMBIA1300-2X35S-GFP-NOS, pCAMBIA1300-GFP-NOS, subcellular localization vector pAN580 and yeast two-hybrid expression vectors pGBKT7 and pGADT7 were preserved and used by the Rice Research Institute, College of Agriculture and Biotechnology, Southwest University; bimolecular fluorescence complementation (BiFC) vectors pSCYNE and pSCYCE were donated by Professor Luo Keming's laboratory, School of Life Sciences, Southwest University; and dual luciferase reporter vector CaMV35S-Ω-GAL4BD-NOS was donated by Qian Qian's laboratory, Chinese Academy of Rice Research.

[0038] 1.3 Kit

[0039] The plasmid mini-extraction kit, endotoxin-free medium-volume plasmid extraction kit, and universal gel DNA recovery kit involved in the present invention were purchased from Beijing Tiangen Biochemical Technology Co., Ltd.; the plant total RNA extraction kit and the dual luciferase detection kit were purchased from Beijing Promega Biotechnology Co., Ltd.; the reverse transcription kit and the fluorescence quantification kit were purchased from TaKaRa Bioengineering Co., Ltd.; the DNA cleaning and recovery kit was purchased from AXYGEN Biotechnology Co., Ltd.; the digoxin RNA labeling (SP6 / T7) kit and the digoxin detection kit were purchased from Roche, Germany; and the plant chromatin extraction and immunoprecipitation (ChIP) kit was purchased from Beijing Qiwei Yicheng Technology Co., Ltd.

[0040] 1.4 Various enzymes and molecular biology reagents

[0041] The high-fidelity enzymes KOD FX and KOD FX neo involved in the present invention were purchased from Toyobo Biotechnology Co., Ltd. of Japan, and the high-fidelity enzyme PrimerStar MAX and various restriction endonucleases, DNaseI digestion enzyme and T4 rapid ligase were purchased from TaKaRa Biotechnology Co., Ltd. The EasyTaq enzyme for ordinary PCR amplification and seamless cloning recombinase t were purchased from Beijing Quanshijin Biotechnology Co., Ltd. The antibody anti-H3K27me3 used in the ChIP experiment was purchased from abcom, USA, and anti-LF2 was synthesized by Shanghai Youke Biotechnology Co., Ltd. Tryptone, peptone, sodium chloride, yeast extract, glucose, agar powder used for the preparation of various culture media; various antibiotics and CTAB, Tris-HCl and other molecular biological reagents were purchased from Shenggong Biotechnology (Shanghai) Co., Ltd.; xylene, formaldehyde, anhydrous ethanol and other reagents were purchased from Chengdu Kelong Chemical Reagent Factory; solid and liquid reagents related to in situ hybridization experiments and protoplast extraction experiments were purchased from Sigma, USA.

[0042] 1.5 Primers used in the present invention

[0043] 1.5.1 Gene mapping primers

[0044]

[0045] 1.5.2 Candidate gene sequencing primers

[0046]

[0047]

[0048] 1.5.3 Construction of complementary vector and primers for genetic transformation

[0049]

[0050] 1.5.4 Overexpression vector construction and transformation primers

[0051]

[0052] 1.5.5 Primers for gene expression analysis (qPCR)

[0053]

[0054]

[0055]

[0056] 1.5.6 Primers for in situ hybridization

[0057]

[0058] 1.5.7 Subcellular localization primers

[0059]

[0060] 1.5.8 Yeast two-hybrid primers

[0061]

[0062]

[0063] 1.5.9 Primers for Bimolecular Fluorescence Complementation Assay

[0064]

[0065] 1.5.10 Dual luciferase reporter assay primers

[0066]

[0067] Example 1 Acquisition and phenotypic analysis of lf2 mutants

[0068] Using ethyl methanesulfonate (EMS) to induce the indica rice maintainer variety Xida 1B, a genetically stable lateral floret development mutant was discovered, named lf2, and the lateral floret mutant trait regulatory gene was named LF2. Compared with the wild-type spikelet, the mutant only developed one floret containing four whorls of floral organs. The secondary guard glumes of the spikelets in the mutant were elongated, the guard glumes were elongated and the lemmas were transformed, and lateral florets or lateral spikelets were produced in the secondary guard glumes or guard glumes of some spikelets. The lateral florets contained normal floral organ structures such as inner and outer palea, pistils, lodicules and stamens.

[0069] A wild rice spikelet contains a terminal floret, which is followed by a pair of guard glumes and a pair of secondary guard glumes. The terminal floret consists of four whorls of floral organs: palea, lemma, lodicule, stamens, and pistil. The palea and lemma are similar in shape and closed together to wrap the three whorls of floral organs inside. Two "lung"-shaped transparent lodicules are located on one side of the lemma, and six stamens grow around the central pistil ( Figure 1 , AB). The palea is hard, with silicified cells and a lot of burrs on the surface. The difference is that the palea also contains a smooth and transparent edge structure ( Figure 1 , DE). The guard glumes are located on both sides of the base of the terminal floret. They are two transparent, thin, lamellar organs with a very smooth surface and no silicified cells ( Figure 1 , C). The secondary guard glume is extremely degenerate and located below the guard glume, with a rough surface structure and full of burrs and pores ( Figure 1 , F).

[0070] There was no significant difference between the lf2 mutant and the wild-type plants in the vegetative growth stage. In the reproductive growth stage, the lf2 mutant showed obvious defects in spikelet development, and its phenotypes were mainly divided into two categories: elongation of the lemma (type I) and formation of lateral florets (type II).

[0071] In type I spikelets, the secondary glume was significantly elongated, and the surface had a large number of burrs and stomata like the wild type secondary glume ( Figure 1 , HI, L), the guard glumes showed different degrees of elongation, and a large number of silicified cells appeared on the surface ( Figure 1 , HK), some guard glumes are consistent with the lemma in terms of shape, size, and degree of surface silicification. Histological analysis found that the wild-type lemma contains 5 vascular bundles, the palea contains 3 vascular bundles, the palea body and the lemma both contain four layers of cell structure, including silicified upper epidermal cells, thick-walled cells, thin-walled cells and non-silicified lower epidermal cells, the palea edge is mainly thin-walled cells and a small number of thick-walled cells, and the guard glumes contain a large number of thin-walled cells and only one vascular bundle ( Figure 1 , G); in type I spikelets, the lemma-like guard glume develops a four-layer cell structure consistent with the lemma ( Figure 1 , M). In addition, expression analysis of the guard glume characteristic gene OsMADS34 and the inner and outer lemma characteristic genes OsMADS1, OsMADS6, and DL in different guard glumes and inner and outer lemmas revealed that the expression patterns of these genes in the unelongated guard glume of lf2 were basically consistent with those of the wild-type guard glume, while the expression of OsMADS34 in the elongated guard glume was significantly reduced, and OsMADS6 and DL were almost not expressed or expressed at very low levels, which was consistent with the wild-type guard glume. In the lemma guard glume, the expression of OsMADS34 was extremely significantly reduced, and there was a strong ectopic expression of the outer lemma characteristic gene DL, while the inner lemma characteristic gene OsMADS6 was not expressed ( Figure 1The above results indicate that in the lf2 mutant, the lemma characteristics are transformed to different degrees.

[0072] In type II spikelets, the terminal floret contains a complete number of palea, lodicules, stamens, pistils and other organs, and in a few cases, the secondary lemma is elongated ( Figure 2 , FT). Unlike wild-type mature pollen grains, lf2 pollen grains cannot be stained in iodine–potassium iodide (I2–KI) solution and appear sterile ( Figure 3 , GH). Compared with the wild type, 17.6% of lf2 spikelets developed lateral florets / spike ( Figure 3 , J), in these lf2 spikelets, most of the spikelets developed partial floral organs (lacking lemma) or complete lateral florets in the axils of the lower lemma. Figure 2 , FM), spikelet-like organs developed under the axils of some spikelets, including palea organs and tissues similar to early tender spikelets ( Figure 2 , NO), in addition, a very small number of lf2 spikelets developed lateral florets in the axils of the secondary guard glume, usually accompanied by the elongation of the guard glume and the other side of the secondary guard glume, and even developed lateral florets / spikelets in the axils of the guard glume and the secondary guard glume at the same time ( Figure 2 , RT). Morphological and histological observations of the lateral florets revealed that the morphology and cell structure of the floral organs, such as the palea, lodicule, stamens, and pistils, were consistent with those of the wild type, but their number was usually reduced compared to the wild type. For example, the lemma was missing and the number of stamens was only 2-5 ( Figure 3 , AF). qPCR was further used to detect the expression levels of related floral organ characteristic genes including palea characteristic gene OsMADS1, lodicule and stamen characteristic genes OsMADS2, OsMADS4 and OsMADS16, palea characteristic gene OsMADS6, lemma and carpel characteristic gene DL, stamen and carpel characteristic genes OsMADS3, OsMADS58, and ovule characteristic gene OsMADS13 in wild-type guard glumes, terminal flowers and lf2 lateral florets. The results showed that these genes were highly expressed in lateral florets, while these genes were almost not expressed in guard glumes, indicating that the lateral florets developed by the lf2 mutant have complete floral organ characteristics ( Figure 3 , I).

[0073] The early development of spikelets in wild-type and mutant plants was observed using scanning electron microscopy (SEM). In the wild type, the spikelet primordium first differentiated into the secondary glume primordium and the glume primordium and transformed into floral meristems. During the Sp3-Sp8 period, the primordiums of floral organs such as the lemma, palea, lodicule, stamen, and pistil (including ovules and carpels) were differentiated in sequence. By the Sp8 period, the palea primordium and the lemma primordium were closed together ( Figure 4, AC). In the lf2 mutant type I spikelets, the development of the guard glumes was observed to be significantly accelerated. The development of some guard glumes was similar to that of the lemma primordium, and their size and shape were very similar to those of the lemma primordium ( Figure 4 DF). In the lf2 mutant type II spikelets, additional lateral meristems were observed in the axils of the guard glume primordium or the secondary guard glume primordium from about the Sp4 period, and various internal floral organ primordia gradually differentiated with the progress of spikelet development. However, the lateral florets were significantly smaller than the terminal florets in general, and the development of each floral organ primordium was also delayed by 1-2 periods compared with the terminal florets and the wild-type floral organ primordium ( Figure 4 , GI). In addition, some lf2 spikelets differentiated into lateral meristems in the axils of the secondary glume primordium and gradually differentiated into complete internal floral organ primordia. The development process of each organ primordium lagged behind the terminal floret and wild-type floral organ primordia by 2-3 periods ( Figure 4 , JL). The results showed that the development process of the spikelet guard primordium of the lf2 mutant was similar to that of the lemma primordium, and lateral floral meristems were differentiated from the early spikelet development, which was consistent with the phenotype of the mature spikelet.

[0074] The expression of meristem characteristic genes OSH1 and OsMADS15 in wild-type and lf2 spikelets was further analyzed by in situ hybridization. The results showed that the OSH1 gene was strongly expressed in the floral meristem of wild-type spikelets ( Figure 5 , A1-A2), when the pistil primordium is formed, the expression of OSH1 gene is significantly weakened ( Figure 5 , A3-A4). In the terminal florets of the lf2 mutant, the expression pattern of the OSH1 gene was consistent with that of the wild-type spikelets, and from the Sp4 period, obvious OSH1 expression signals were detected in the glume primordium ( Figure 3-5 , B1-B4). OsMADS15 is strongly expressed in wild-type floral meristems, and is expressed in floral organ primordia such as lodicules and pistils as they differentiate. Figure 5 , C1-C4). In the lf2 mutant, OsMADS15 was strongly expressed in the floral meristems of both the terminal and lateral flowers ( Figure 5 , D1-D4). The above results indicate that the ectopic expression of OSH1 and OsMADS15 genes in the lf2 mutant induces the formation of lateral florets.

[0075] Example 2 Gene Identification of LF2 Gene

[0076] 1. Genetic localization of LF2 gene and candidate gene sequencing analysis

[0077] The LF2 gene was located between SSR markers R-20 and R-39 at the end of the long arm of chromosome 7 using the map-based cloning method, with a physical distance of approximately 180 kb ( Figure 6 , A). There are 31 annotated genes in this interval, including eight expressed proteins, four putative proteins, three cytochrome proteins, two MYB proteins, two transposon proteins, one transcriptional regulator, one SNF2 family transcription factor and some functional proteins (Table 1). No genes that have been reported to regulate rice flower / panicle development have been found.

[0078] Sequencing analysis of genes within the interval revealed that in the lf2 mutant, the first base of the fourth intron of the LOC_Os07g44210 gene underwent a single base substitution from T to A, resulting in disordered mRNA splicing and the formation of multiple new transcripts ( Figure 6 , B, transcript sequences are shown in SEQ ID No.3-6), resulting in serious changes in protein function. According to the annotation information provided on the website, the LOC_Os07g44210 gene contains 24 exons, the coding frame genomic sequence is 12318-bp in length, the cDNA is 2244-bp in length (the sequence is shown in SEQ ID No.1), and encodes a protein containing 747 amino acids (the sequence is shown in SEQ ID No.2). Therefore, the LOC_Os07g44210 gene is positioned as a candidate gene of the LF2 gene.

[0079] Table 1 Gene annotation information within the localization interval

[0080]

[0081] 2. Transgenic verification of LF2 gene

[0082] 2.1 Complementary Verification

[0083] Construction of a complementary vector containing the LOC_Os07g44210 gene (including a 2995-bp promoter and full-length CDS sequence) fused to GFP ( Figure 7 , A). Since the lf2 mutant is sterile, the complementary vector was transformed into the callus induced by lf2 heterozygous seeds by Agrobacterium infection, and 25 transgenic plants were obtained and planted at the Xiema planting base of the Rice Research Institute of Southwest University. When the plants grew up, the DNA of each individual plant was extracted, and exogenous GFP amplification primers were designed for amplification. The results showed that all transgenic plants were positive plants. After further design of specific primers for the endogenous LF2 gene (including the mutation site), the amplified fragments were sequenced and analyzed, and the GFP fluorescence signal in the transgenic plants was observed at the same time. The results showed that 7 strains contained heterozygous mutation sites and the plant phenotype was normal, 4 strains contained pure and lf2 mutation sites, of which 3 strains had the same phenotype as the wild type, and 1 strain showed a mutant phenotype ( Figure 7, B, D), the rest of the plants contained wild-type sites and had normal phenotypes. In addition, obvious GFP fluorescence signals were observed in all 11 transgenic plants containing homozygous / heterozygous mutation sites ( Figure 7 , C), indicating that the transfer of the complete CDS sequence of the LOC_Os07g44210 gene can restore the lf2 mutant phenotype, that is, the LOC_Os07g44210 gene is the LF2 gene.

[0084] 2.2 CRISPR-cas9 knockout verification

[0085] The genome sequence of LOC_Os07g44210 was analyzed on the CRISPR-P website and two editing target sites were designed on the first exon (a: TGATCCGTTGGGTACAAGGA; b: CCAGGAGTCTTGCTTCGGCA). Wuhan Boyuan Biotechnology Co., Ltd. completed the construction of the knockout vector and transformed the japonica rice variety ZH11, obtaining a total of 16 transgenic plants. The genome sequence analysis of the transgenic plants showed that 10 of the lines had an editing effect, with a T base or GATCCGT or CTGAATCCCATGATCCGTTGGGTACAA short fragment deletion at site a, and a G base or TTCG deletion at site b ( Figure 8 , A). Plant RNA was extracted and two pairs of primers were designed to detect the expression of LF2 gene. The results showed that the expression level of LF2 gene in cas9 knockout plants was significantly lower than that in wild-type plants ( Figure 8 , B). Phenotypic observation of transgenic plants with editing effects during flowering period revealed that the spikelet morphology of some plants was not significantly different from that of the wild type, but the fertility was significantly reduced. In addition, spikelets of 5 strains showed various abnormal developments, similar to the lf2 mutant traits, such as elongated / lemmaized guard glumes, degenerated palea, and the formation of lateral florets ( Figure 8 , C). The above results further proved that LOC_Os07g44210 is the LF2 gene.

[0086] 2.3 Overexpression analysis

[0087] Using wild-type plant cDNA as a template, the CDS sequence of the LOC_Os07g44210 gene was amplified and fused with GFP to form a fusion protein. The 2ⅹ35S promoter was used to drive the LF2-GFP fusion protein and transform the japonica rice ZH11 callus. Ten overexpressing plants were obtained and planted at the Xiema Base of the Rice Institute of Southwest University. After the plants grew up, RNA from individual plant tissues was extracted for expression analysis. In these plants, the expression level of the LOC_Os07g44210 gene increased significantly ( Fig. 9, A). However, the spikelets of the overexpressing plants were consistent with the wild-type spikelets phenotype throughout the growth and development process. The transgenic plant seeds harvested in the second year were planted again, and the spikelets of the overexpressing plants detected were still no different from the wild-type, indicating that the overexpression of the LOC_Os07g44210 gene did not cause abnormal development of rice plants ( Fig. 9 , B).

[0088] Example 3 Analysis of expression pattern of LF2 gene

[0089] In order to study the function of LF2 gene, the expression pattern of LF2 gene was analyzed. Firstly, the expression of LF2 gene in different tissues was detected by qPCR. The results showed that the expression of LF2 gene was relatively low in roots, and it was strongly expressed in other tissues such as leaves, sheaths, buds and inflorescences, and the expression in inflorescences gradually decreased with growth and development. Among the organs of spikelets, LF2 gene was mainly strongly expressed in the glumes, and it was also expressed in other flower organs of each round, but relatively weakly ( Fig.10 , A). Further in situ hybridization was used to detect the expression of LF2 gene in inflorescence and various floral organs, and it was found that strong LF2 expression signals were detected in inflorescence, especially in spikelet meristem and floral meristem ( Fig.10 , BD), with the differentiation and development of various organs, obvious LF2 expression signals were detected in the secondary guard glumes, guard glumes and various floral organs ( Fig.10 , EI). The above results indicate that LF2 gene is a non-specific expression gene, which is highly expressed in both vegetative and reproductive growth organs.

[0090] Example 4 The LF2 gene encodes a chromatin remodeling factor localized in the cell nucleus

[0091] 1. Phylogenetic tree and protein sequence analysis of LF2 protein

[0092] The LF2 gene encodes a SNF2 family chromatin remodeling factor, which belongs to the SMARCAL1 subfamily (Hu et al., 2013). The SMART protein analysis website predicts that the LF2 protein contains a DEXDc domain and a HELECc domain ( Fig.11 , A). The DEXDc domain and the HELECc domain contain ATP binding sites, which can bind and hydrolyze ATP to release energy. In addition, the HELECc domain contains nucleic acid binding sites, which can bind to the target gene sequence to affect the transcriptional activation or inhibition of the target gene. LF2 and its homologous proteins in different species contain DEXDc and HELECc domains, and the protein sequences, especially the parts where the two domains are located, are highly conserved, with more than 70% similarity ( Fig.11 , B).

[0093] Studies have shown that 41 SNF2 family chromatin remodeling factors have been identified in Arabidopsis, which can be divided into 18 subfamilies. These proteins play an important role in plant development regulation, stress response and epigenetic inheritance (Knizewski et al., 2008). According to the information listed in the chromDB database, rice contains 42 SNF2 family chromatin remodeling factors, of which the SMARCAL1 subfamily contains two members, LF2 and CHR726 (LOC_Os07g40730), which are homologous to Arabidopsis AtCHR18 and AtCHR14, respectively (Hu et al., 2013). Further selection of LF2 homologous genes in different species for phylogenetic tree analysis showed that the homology between LF2 proteins was high, and LF2 and CHR726 formed independent branches with their respective homologous proteins, and the LF2 homologous proteins from monocots and dicots each formed a small branch. In the monocotyledonous plant branch, proteins from grasses such as rice, corn, sorghum, wheat, and Brachypodium distichum form a small branch with a closer relationship ( Fig.12 ).

[0094] 2. Subcellular localization of LF2 protein

[0095] In order to clarify the localization of LF2 protein in cells, we first took the young panicles of LF2 overexpressing plants and placed them under a ZEISS LSM800 confocal microscope to observe the fluorescence of LF2-GFP fusion protein. The results showed that there was a strong fluorescence signal in the young panicles, and it was specifically concentrated in the cell nucleus ( Fig.13 , A). The subcellular localization of LF2 protein was further verified by transiently expressing LF2-GFP fusion protein in rice protoplasts. Rice protoplasts were transiently transformed with pAN580-GFP empty vector and pAN580-LF2-GFP fusion expression vector, respectively. After cultivation, the fluorescence signal in the protoplasts was observed under a ZEISS LSM 800 confocal microscope. The results showed that in the pAN580-GFP control, there were strong green fluorescence signals in all parts of the cells (except the vacuole), while in the cells expressing LF2-GFP fusion protein, only the cell nucleus was observed to have obvious fluorescence signals ( Fig.13 , B), The above results indicate that LF2 protein is localized in the cell nucleus.

[0096] 3. Analysis of transcriptional activity of LF2 protein

[0097] The transcriptional activity of LF2 protein was detected by the yeast expression system nuclear dual luciferase reporter gene system. First, the pGBKT7-LF2 expression vector was constructed, and pGBKT7-LF1 (LF1 protein is known to have self-activation activity) was used as a positive control, and the pGBKT7 empty vector was used as a negative control. It and the pGBKT7-LF2 fusion expression vector were transformed into Y2HGold yeast cells, and the self-activation activity of LF2 protein was screened and verified using SD / -Trp and SD / -Trp-His-Ade defective medium. The results showed that yeast cells transformed with pGBKT7-LF1 could grow on SD / -Trp and SD / -Trp-His-Ade medium, and yeast cells with pGBKT7 empty vector and pGBKT7-LF2 fusion vector could grow normally on the medium but could not grow on SD / -Trp-His-Ade medium, indicating that LF2 protein does not have self-activation activity ( Fig.14 , A). Next, the dual luciferase (LUC) reporter gene was used to further study its transcriptional activity (the LUC gene has 5 copies of binding sites that can act on GAL4), and the activity of LUC in rice protoplasts was detected. The LF2 cDNA was fused with the GAL4 DNA binding domain (BD) and driven by the 35S promoter of the cauliflower mosaic virus. The transcription activator VP16 was used as a positive control and GAL4-BD was used as a negative control (Ren et al., 2016). GAL4BD-VP16, GAL4BD-LF2 and GAL4-BD effectors were transiently expressed in rice protoplasts. The results showed that the positive control GAL4BD-VP16 had strong luciferase activity (Firefly / Rellinaratio), and the luciferase activity of the GAL4BD-LF2 protein was slightly higher than that of the negative control GAL4BD, but the difference in activity was not significant, further indicating that the LF2 protein did not have obvious transcriptional activation activity ( Fig.14 , B).

[0098] Implementation of 5 LF2 interacting protein screening and verification

[0099] 1. cDNA library screening

[0100] In order to clarify the mechanism of LF2 gene inducing lateral florets and regulating floral organ development, the pGBKT7-LF2 vector with no self-activation activity was first used as bait to provide to Shanghai Ouyi Biotechnology Co., Ltd. for one-to-one screening of rice transcription factor AD library to find the interacting factors of LF2 protein. pGBKT7-LF2 was mated with rice transcription factor library. All clones grew normally on the non-screening medium SD / -Trp / -Leu. A total of 6 transcription factors interacting with LF2 were screened on the screening medium SD / -Trp-Leu-His-Ade / X-α-gal / AbA ( Fig.15 , A). These six transcription factors were then cloned and connected to the pGADT7 vector, and their interaction with the LF2 protein was verified in yeast. The results showed that except for LOC_Os11g08210, which did not grow on the screening medium SD / -Trp-Leu-His-Ade, the other five transcription factors could grow on SD / -Trp-Leu-His-Ade ( Fig.15 , B), indicating that the five transcription factors except LOC_Os11g08210 interact with LF2 protein. The gene information of these five transcription factors was queried and analyzed, and it was found that the LOC_Os03g44540 gene encodes the A subunit of nuclear factor Y (NF-Y), with the gene symbol OsNF-YA3. Nuclear factor Y, also known as CCAAT-binding factor (CBF) or heme activator protein (HAP), is widely present in eukaryotic organisms such as yeast, animals and plants. HAP is a heterotrimeric complex containing three subunits: HAP2 / NF-YA / CBF-B, HAP3 / NF-YB / CBF-A and HAP5 / NF-YC / CBF-C. The HAP complex can bind to the CCAAT sequence in the promoter and regulate the expression of the target gene (Thirumurugan et al., 2008). In rice, nuclear factors play important roles in embryonic development, regulation of organ characteristics, flowering regulation, leaf development, seed germination, resistance, and many other aspects (Alam et al., 2015; Ito et al., 2011; Miyoshi et al., 2003; Sun et al., 2014). Therefore, the interaction between LF2 and OsNF-YA3 was further verified in tobacco epidermal cells by bimolecular fluorescence complementation (BiFC). Obvious yellow fluorescence was observed in tobacco leaf cells co-expressing YN-LF2 and YC-OsNF-YA3, and it overlapped with the blue fluorescence in the nucleus stained with DAPI, proving that there is an interaction relationship between LF2 and OsNF-YA3 proteins ( Fig.15 , C).

[0101] To determine the role of nuclear factor subunit OsNF-YA3 in rice panicle development, CRISPR technology was used to knock out OsNF-YA3 in ZH11. Transgenic plants were grown in the greenhouse and OsNF-YA3 was expressed at the flowering stage. cas9 Elongated secondary guard glumes and lemma-like guard glumes were observed in the spikelets of the plants, similar to the lf2 mutant. In addition, additional lemma-like organs and glumes were observed in the spikelets ( Fig.16 ), indicating that OsNF-YA3 plays an important regulatory role in rice spikelet development, and its regulation of guard glume development is similar to that of the LF2 gene.

[0102] 2. Screening of interactions between LF2 and rice nuclear factor

[0103] Since the three subunits of the nuclear factor need to form a heterotrimer, the complex binds to the CCAAT sequence in the promoter of the target gene to regulate the expression of the target gene. A total of 11 HAP2, 11 HAP3 and 12 HAP5 genes have been reported in rice. According to the gene expression information provided on the website (http: / / rice.plantbiology.msu.edu / ), 9 HAP3 and 7 HAP5 genes expressed in rice inflorescences were cloned into the pGADT7 vector, and the interaction was verified by yeast two-hybrid with pGBKT7-LF2. Finally, 1 HAP3 subunit: HAP3J and 4 HAP5 subunits: HAP5A, HAP5B, HAP5D, HAP5E were screened on SD / -Trp-Leu-His-Ade medium, indicating that in addition to HAP2H, LF2 protein also interacts with these five nuclear factors. These proteins may form heterotrimers to interact with LF2 and thus affect the expression of downstream genes ( Fig.17 ).

[0104] Example 6 Transcriptome Analysis of lf2 and Wild Type

[0105] In order to identify the downstream target genes of the LF2 gene and the pathways involved, young inflorescences (less than 2 cm) of the lf2 mutant and wild type were taken for transcriptome sequencing analysis at the booting stage. In the mutant, 835 upregulated genes and 832 downregulated genes were obtained. Fig.18, A). Among these genes, there are several genes related to flower development, such as genes OsMADS1, OsMADS6, OsMADS14, and OsMADS15 that regulate palea development; genes OsMADS2 and OsMADS4 that regulate lodicule development; genes OsMADS7, OsMADS16, and DL that regulate stamen and pistil development; genes OsMADS18 and OSH1 that regulate meristem; and genes OsMADS34 that regulate lemma and peduncles development. qPCR verification of these genes revealed that the expression of OsMADS1, OsMADS6, OsMADS7, OsMADS16, and DL genes was significantly downregulated in inflorescence, while the expression of OsMADS2, OsMADS4, OsMADS14, OsMADS15, OsMADS18, OsMADS34, and OSH1 was significantly upregulated ( Fig.18 , B), which is consistent with the transcriptome sequencing results and lf2 mutation traits.

[0106] In addition, GO and KEGG enrichment analysis of all differentially expressed genes obtained by transcriptome sequencing analysis revealed that almost all differentially expressed genes were enriched in various biological processes such as transcription, translation, and molecular synthesis, and only a few were enriched in molecular components and molecular functions, mainly playing a role in nucleic acid binding and transcription factor activity, indicating that the LF2 gene has important functions in various biological processes of plant development ( Fig.19 , A). KEGG signaling pathway analysis found that the most genes were enriched in the plant hormone signaling pathway, among which the most genes were involved in the auxin signaling pathway ( Fig.19 , B).

[0107] Studies have shown that auxin plays an important role in flower development. In Arabidopsis, the auxin response factor MONOPTEROS (MP) recruits the SWI / SNF chromatin remodeling complex to increase DNA accessibility, thereby activating key regulatory factors that induce flower primordium initiation (Wu et al., 2015). Liquid chromatography was used to determine the auxin IAA content in the young panicles of wild type and lf2 mutants. The results showed that the IAA content in the mutant was significantly lower than that in the wild type ( Fig. 20 , A).

[0108] qPCR expression analysis of key genes OsYUCCAs in the auxin synthesis pathway, key genes OsPINs in polar transport, and auxin response factors Aux / IAA (OsIAAs) and OsARFs revealed that the expression levels of key genes OsYUCCA2 and OsYUCCA4 in auxin synthesis, and polar transport-related genes OsPIN1, OsPIN2, OsPIN5a, OsPIN6, and OsPIN10b were significantly decreased ( Fig. 20, BC). In the auxin response pathway, when the auxin concentration is low, the inhibitory factor Aux / IAA binds to the auxin response factor ARFs to form a dimer, inhibiting the transcriptional activity of ARFs and thus inhibiting the expression of auxin response genes. When the intracellular auxin increases, Aux / IAA binds to the SCF ubiquitin complex (SCF TIR / AFB The binding is ubiquitinated and degraded by the 26S proteasome. ARFs are released to activate or inhibit the expression of downstream genes. qPCR results showed that the expression of OsIAA5, 10, 13, 24, and 26 was significantly upregulated in the lf2 mutant, while the expression of auxin response factors ARF3, 4, 6, 7, 8, 9, 10, 11, 14, 18, 19, 20, 22, 23, and 24 was significantly downregulated, indicating that the auxin response in the mutant was also weakened ( Fig.19 , DE), indicating that the LF2 gene may be involved in the auxin biosynthesis pathway.

[0109] Example 7 ChIP-Seq analysis of If2 and wild type

[0110] Chromatin remodeling factors can usually affect the methylation level of target genes and thus affect their expression levels. Therefore, RNA was extracted from wild-type and lf2 mutant panicles, and ChIP-Seq was performed after precipitation with H3K27me3 antibody. The results were correlated with transcriptome data to find possible downstream target genes of LF2 gene. The correlation data were analyzed and some genes that may be involved in the growth and development of rice were screened out, including creeping growth gene PROG1; RING-H2 zinc finger protein ATL5A; glume characteristic development regulatory gene G1 / ELE; hormone pathway related genes: BR receptor kinase OsBRLv1, strigolactone secondary response gene OsRR5, auxin polar transport gene OsPIN9 and an auxin repressor gene; branch and glume development regulatory gene OsMADS34 / PAP2; tapetum development regulatory gene OsTDL1A and a DNA binding protein (Table 2). Among these genes, the rice long glume gene G1 / ELE gene is involved in the fate determination of the glume, inhibiting the development of the glume into the lemma, and the glume homeotic transformation into the lemma in the g1 mutant (Yoshida et al., 2009), which is largely similar to the trait of the lf2 mutant. qPCR results showed that the expression of the G1 / ELE gene was significantly downregulated in the lf2 mutant ( Fig.21, A), ChIP-Seq data showed that there were two obvious H3K27me3 binding peaks on the G1 / ELE gene and the H3K27me3 level at the G1 / ELE site increased in lf2. Chromatin from wild-type and lf2 mutant spikelets was extracted and precipitated with H3K27me3 antibody, and then ChIP-qPCR was performed for verification. The results showed that the H3K27me3 level in lf2 at the G1 / ELE promoter and P1 site on CDS was significantly increased, corresponding to the downregulation of G1 / ELE gene expression ( Fig.21 , B). In addition, lf2 and g1 were combined to form double mutants to observe the phenotype of spikelets. In the g1 mutant, the two guard glumes were elongated and tended to be consistent with the lemma in morphology and size, and the internal floral organs developed normally; in the lf2 mutant, the secondary guard glumes were elongated, the guard glumes were elongated or lemmaized to varying degrees, and the number of stamens in some spikelets was reduced; in the g1+lf2 double mutant, the secondary guard glumes were observed to be elongated, and both guard glumes were elongated and lemmaized, and the degree of elongation was higher than that of the g1 and lf2 single mutants, and the number of stamens was significantly reduced, indicating that the LF2 gene acts upstream of the G1 gene and regulates the development of rice guard glumes by affecting the H3K27me3 level of the G1 gene to inhibit the expression of G1 ( Fig.21 , C).

[0111] Table 2 Genes screened by ChIP-Seq and transcriptome association analysis

[0112]

[0113] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. All modifications, equivalent substitutions, improvements, etc. within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A rice lateral floret regulatory gene LF2, the nucleotide sequence of which is shown in SEQ ID No.1, and the amino acid sequence of the encoded protein is shown in SEQ ID No.

2.

2. The protein encoded by the rice lateral floret regulating gene LF2 according to claim 1, whose amino acid sequence is shown in SEQ ID No.

2.

3. Use of the rice lateral floret regulatory gene LF2 according to claim 1 and the protein according to claim 2 in increasing rice yield.

4. The mutant gene of the rice lateral floret regulatory gene LF2 according to claim 1, wherein the first base of the fourth intron of the LOC_Os07g44210 gene undergoes a single base substitution from T to A, and the transcript sequence thereof is shown in SEQ ID No.3-6.

5. The transcript of the mutant gene of the rice lateral floret regulatory gene LF2 according to claim 4, wherein the transcript sequence is shown as SEQ ID No. 3-6.

6. Use of the mutant gene of the rice lateral floret regulatory gene LF2 according to claim 4 and the transcript of the mutant gene of the rice lateral floret regulatory gene LF2 according to claim 5 in improving rice yield.

Images