A method for improving the editing efficiency of a single-base editor in dicotyledonous plants and its application

Abstract

The present invention provides a method and application for improving the editing efficiency of a single-base editor from cytosine to thymine in dicotyledons, and belongs to the field of biotechnology. The method is to increase the number of UGIs to 2 in a single-base editor system from cytosine to thymine in dicotyledons based on CRISPR / Cas9n, to rupture the fifth base T of the sgRNA scaffold into a C, and to extend its stem-loop structure, to replace the linker sequence between nCas9 and PmCDA1 with an ssDBD domain sequence, and to replace nCas9 with an nCas9 variant, thereby improving the editing efficiency. The method provided by the present invention is helpful for functional genomic research and molecular breeding of dicotyledons including soybeans.

Description

Technical Field

[0001] The present invention belongs to the field of biotechnology and relates to a method for improving the editing efficiency of a single-base editor from cytosine (C) to thymine (T) in dicotyledonous plants and its application in functional genomics research and molecular breeding. Background Art

[0002] Soybean (Glycinemax) is one of the most important crops. It has a long history of cultivation in my country and is rich in plant protein and oil. It plays an important role in the development of Chinese agriculture. Research on the functional analysis and regulatory expression of soybeans is of great significance to the improvement of soybean germplasm, stress resistance and yield increase, and food safety. Among them, obtaining mutant materials of relevant target genes is a key step in soybean functional research and new germplasm improvement and application. At present, traditional methods for obtaining soybean mutants include natural mutagenesis, chemical mutagenesis, physical mutagenesis, etc., but they have the disadvantages of long cycle, high investment, and low efficiency. With the innovation of science and technology, the emergence of gene editing has made it possible to achieve precise mutations, while greatly shortening the breeding period and providing new ideas for germplasm improvement.

[0003] Gene editing technology is a technology that uses sequence-specific nucleases expressed by genetic engineering to cut the genome and achieve site-specific editing of genes. It can produce insertions, knockouts, replacements, site-specific mutations, etc. It is now widely used in the treatment of medical diseases and the cultivation of multiple species such as rice, corn, and soybeans. The recently developed CRISPR-related tool, the base editing system, is a more precise gene modification technology that greatly expands the scope of genome editing. The base editing system does not require double-strand breaks and the provision of exogenous donor templates, and has wide adaptability and high efficiency. It fuses the inactivated dCas9 (D10A, H840A) or partially inactivated nCas9 (D10A) and deaminase into a protein complex, and then targets the protein complex to single-stranded DNA. It can accurately mutate one base pair to another base pair in the editing window. Currently, there are two main single-base editing systems, the cytosine base editing system (CBE) and the adenine base editing system (ABE). For a long time, most single-base editing systems have been based on SpCas9, which mainly recognizes NGGPAM sequences and cannot perform precise editing in areas rich in adenine and thymine. Without an adjacent, identifiable PAM sequence, the gene will not be able to be used as a target, and the Cas protein will not be able to recognize or successfully attach to and cut the required DNA fragment. This greatly limits the application of its system. In order to further broaden the scope of editing, researchers have developed many Cas9 protein variants.

[0004] Nowadays, crop breeding requires the cultivation of insect-resistant, disease-resistant and high-yield crops, and many agronomic traits are determined by gene point mutations or changes in a few bases, so single nucleotide mutations are the genetic basis for changes in crop agronomic traits. At present, the single-base editing system mediated by the CRISPR / Cas9 system has developed rapidly, providing new technical support for crop improvement and the creation of new germplasm. However, first, there is only one study in soybeans that successfully changed the flowering period of soybeans to adapt to a wider range of planting areas by using a single-base editing system to edit the GmFT2a and GmFT5a genes, but the CBE base editor used in it has not been optimized, and the highest single-base editing efficiency is only about 18%; second, in the optimization process of the CBE tool, the researchers mainly optimized cytosine deaminases such as A3A, while there were fewer optimizations for PmCDA1 cytosine deaminase.

[0005] It can be seen that there is no efficient CBE base editing tool in soybeans at present, so it is still necessary to further develop and establish more efficient and accurate single base editing technology in soybeans, broaden the application scope of single base editing in soybeans, and lay a technical foundation for soybean gene editing precision breeding. And apply the single base editing system to soybean breeding to create new germplasm. Summary of the invention

[0006] In view of this, the object of the present invention is to provide a method for improving the editing efficiency of single-base editors in dicotyledonous plants for use in functional genomics research and molecular breeding.

[0007] To achieve the above object, the present invention adopts the following technical solution:

[0008] A method for improving the editing efficiency of a dicotyledonous plant single base editor, the method comprising: modifying a uracil glycosylase inhibitor, a sgRNA scaffold, a connecting peptide and a Cas9 protein in a dicotyledonous plant single base editor system, thereby improving the editing efficiency;

[0009] The single-base editor system is a CRISPR / Cas9n-based cytosine to thymine single-base editor system, and its base sequence is shown in SEQ ID NO.7;

[0010] The modification of the uracil glycosylase inhibitor refers to the modification of the quantity of the uracil glycosylase inhibitor;

[0011] The modification of the sgRNA scaffold refers to the modification of the stem-loop structure and the base at position 5 of the sgRNA scaffold;

[0012] The modification of the connecting peptide refers to replacing the connecting peptide sequence between nCas9 and PmCDA1 with the ssDBD domain sequence;

[0013] The modification of Cas9 protein refers to replacing nCas9 with an nCas9 variant.

[0014] In the above-mentioned method for improving the editing efficiency of a single-base editor in dicotyledonous plants, the dicotyledonous plant is soybean.

[0015] In the above-mentioned method for improving the editing efficiency of a single-base editor in dicotyledons, the modification of the number of uracil glycosylase inhibitors is specifically to increase the number of uracil glycosylase inhibitors from 1 to 2.

[0016] In the above-mentioned method for improving the editing efficiency of a single-base editor in dicotyledonous plants, the modification of the stem-loop structure and the 5th base of the sgRNA scaffold is specifically to transform the 5th base T of the sgRNA scaffold into a C, and at the same time add "TGCTG" after the 12th base and add a "CAGCA" sequence after the 16th base to extend the stem-loop structure.

[0017] In the above method for improving the editing efficiency of a single-base editor in dicotyledonous plants, the ssDBD domain sequence is selected from the human ssDBD shown in SEQ ID NO.2 hsRAD51 Sequence, soybean endogenous ssDBD shown in SEQ ID NO.4 GmRAD51a Sequence, soybean endogenous ssDBD shown in SEQ ID NO.6 GmRAD51b sequence.

[0018] In the above method for improving the editing efficiency of a single base editor in dicotyledonous plants, the nCas9 variant comprises mutations at the following amino acid positions: D 10 A.R 1335 A.L 1111 R, D 1135 V.G. 1218 R, E 1219 F.A 1322 R and T 1337 R.

[0019] The above-mentioned method for improving the editing efficiency of a dicotyledonous plant single-base editor is used in constructing a dicotyledonous plant single-base editing vector, wherein the dicotyledonous plant is soybean.

[0020] A single base editing vector capable of improving the efficiency of single base editing from cytosine to thymine in dicotyledonous plants, wherein the single base editing vector is constructed by the above method, and the dicotyledonous plant is soybean.

[0021] The above-mentioned method for improving the editing efficiency of a single-base editor in dicotyledonous plants or the application of the above-mentioned single-base editing vector in single-base editing of the dicotyledonous plant genome, the dicotyledonous plant is soybean.

[0022] The significant advantages of the present invention are:

[0023] The present invention transforms the uracil glycosylase inhibitor, sgRNA scaffold, connecting peptide and Cas9 protein in the soybean cytosine to thymine single base editor system based on CRISPR / Cas9n, uses two copies of uracil glycosylase inhibitor (UGI) to fuse with PmCDA1 cytosine deaminase, transforms the fifth base T of sgRNA scaffold into C, and extends its stem-loop structure, creatively using human ssDBD hsRAD51 or soybean endogenous ssDBD GmRAD51a or endogenous ssDBD GmRAD51b The SH3 linker sequence between nCas9 and PmCDA1 was replaced by a Cas9 protein variant, which significantly improved the efficiency of single-base editing. Therefore, the present invention established an efficient CBE base editor in soybean, providing strong technical support for the research of functional genomes and molecular breeding in soybean or dicotyledonous plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 : Schematic diagram of the structures of CBE-V1 vector and SoyCBE-V2 vector.

[0025] Figure 2 : Schematic diagram of the sgRNA tandem sequence in Example 1.

[0026] Figure 3 : Statistical graph of editing efficiency of CBE-V1 vector and SoyCBE-V2 vector in soybean hairy roots.

[0027] Figure 4 : Schematic diagram of the vector structure of SoyCBE-V1 to SoyCBE-V5.

[0028] Figure 5 : Statistical diagram of editing efficiency of SoyCBE-V1 to SoyCBE-V5 vectors in soybean hairy roots.

[0029] Figure 6 : Schematic diagram of the structure of ChyBE-01 vector.

[0030] Figure 7 :Schematic diagram of the multi-sgRNA tandem strategy.

[0031] Figure 8 : Editing efficiency statistics of ChyBE-01 vector and SoyCBE-V2 vector.

[0032] Fig. 9 : Schematic diagram of the structure of ChyBE-02 vector and ChyBE-03 vector.

[0033] Fig.10 :Schematic diagram of the structure of ChyBE-02 vector and ChyBE-03 vector and their editing efficiency.

[0034] Fig.11 : Schematic diagram of the structures of NFR-sg1 and NFR-sg2 and the editing efficiency of ChyBE-01, ChyBE-02 and ChyBE-03 vectors. Dark and light colors represent homozygous editing and heterozygous editing, respectively.

[0035] Fig.12 : Representative plant and root phenotypes of ChyBE-03-mediated GmNFR1α mutation 21 days after rhizobium inoculation.

[0036] Fig.13 :Statistical analysis of single-base editing efficiency mediated by SoyCBE-V2 and ChyBE-03 in T0 generation soybean stably transformed plants.

[0037] Fig.14 :Statistics and analysis of CBE off-target in soybean T0 generation plants.

[0038] Fig.15 : Editing efficiency of ChyBE-NG-01 vector, ChyBE-NG-02 vector and ChyBE-NG-03 vector.

[0039] Fig.16 : Working model of ChyCBEs. DETAILED DESCRIPTION

[0040] In order to make the contents of the present invention easier to understand, the technical solution of the present invention is further described below in conjunction with specific implementation methods, but the present invention is not limited thereto.

[0041] It should be noted that the vectors obtained by the inventors of the present application during the research and development process were obtained by adjusting various cross combinations of gene sequences and fragments, and the specific acquisition process is a technical secret and will not be described in detail. Those skilled in the art can also directly synthesize according to the disclosure of the present invention, which does not affect the realization of the present invention, but the cost will increase.

[0042] The experimental materials involved in the present invention are as follows:

[0043] Plant material: Soybean (Williams 82);

[0044] sgRNA and its target sequence:

[0045] Table 1

[0046]

[0047]

[0048] Primers:

[0049] Table 2

[0050]

[0051]

[0052]

[0053] Example 1

[0054] The CBE-V1 vector was constructed using a single transcript CRISPR-spCas9n structure (STU single transcriptional uint) Figure 1 a), the sequence of CBE-V1 vector is shown in SEQ NO.7; based on CBE-V1 vector, one UGI is replaced with two UGIs, and the original sgRNA Scaffold is replaced with sgRNA eScaffold (the fifth base T in the sgRNA Scaffold sequence is transformed into C, and "TGCTG" is added after the 12th base and "CAGCA" sequence is added after the 16th base to extend the stem-loop structure and increase its stability), and the SoyCBE-V2 vector is constructed ( Figure 1 b).

[0055] Using CBE-V1 vector and SoyCEB-V2 vector as templates, PCR amplified sgRNA fragments, and the five sgRNA fragments (Target 1 to 5) in Table 1 were assembled in series into CBE-V1 and SoyCBE-V2 empty vectors by the BsaI-mediated Golden Gate method. The sgRNA fragment series scheme is shown in Figure 2 Hi-TOM sequencing confirmed that CBE-V1 and SoyCBE-V2 each contained five sgRNAs in series.

[0056] Using the Agrobacterium-mediated soybean hairy root transformation method, the CBE-V1 vector and SoyCBE-V2 vector after tandem assembly of sgRNA were transformed into soybeans respectively. Each vector corresponded to 15 hairy roots, and a total of approximately 30 soybean hairy roots were obtained, and three biological replicates were performed. The DNA of the hairy roots was extracted, the target region was amplified by PCR, and the PCR product was sequenced by HI-TOM second-generation sequencing. The editing efficiency of the target was calculated based on the sequencing results. The results showed that the single-base editing efficiency of the SoyCBE-V2 vector at the three targets of Traget 1, Traget 4 and Traget 5 was significantly higher than that of CBE-V1. The editing efficiency of the two targets of Traget 4 and Traget 5 was as high as 71.5% and 81.3%, respectively. Figure 3 The above results show that the number of UGIs and the structure of sgRNA Scaffold have a significant effect on the efficiency of single-base editing. The structure of two UGIs in series and sgRNA eScaffold can significantly improve the efficiency of single-base editing.

[0057] Example 2

[0058] In order to further improve the efficiency of CBE base editing in soybean, this example further established a plant single base editing vector (SBE-V2) for cytosine (C) to thymine (T) replacement mediated by different cytosine deaminases based on the SoyCBE-V2 vector. Figure 4 ), including rAPOBEC1-mediated cytosine (C) to thymine (T) replacement plant single base editing vector SoyCBE-V1, hA3A-mediated cytosine (C) to thymine (T) replacement plant single base editing vector SoyCBE-V3, hA3A / Y130F-mediated cytosine (C) to thymine (T) replacement plant single base editing vector SoyCBE-V4 and hA3A / N57G-mediated cytosine (C) to thymine (T) replacement plant single base editing vector SoyCBE-V5. The five sgRNA fragments (Target 1 to 5) in Table 1 were assembled in series in one step onto the plant single base editing vector by the BsaI-mediated Golden Gate method, and the sgRNA fragment series scheme was the same as in Example 1.

[0059] Using the Agrobacterium-mediated soybean hairy root transformation method, the SoyCBE-V1 vector, SoyCBE-V2 vector, SoyCBE-V3 vector, SoyCBE-V4 vector and SoyCBE-V5 vector after tandem assembly of sgRNA were transformed into soybeans respectively. Each vector corresponded to 15 hairy roots, and a total of approximately 75 soybean hairy roots were obtained, and three biological replicates were performed. The DNA of the hairy roots was extracted, the target region was amplified by PCR, and the PCR product was sent to HI-TOM second-generation sequencing. The editing efficiency of the target site was calculated based on the sequencing results. The results showed that SoyCBE-V2 had the highest editing efficiency, and the average editing efficiency of SoyCBE-V1, SoyCBE-V3, SoyCBE-V4 and SoyCBE-V5 at all target sites was lower than that of SoyCBE-V2 ( Figure 5 ), which indicated that PmCDA1 was more effective than hA3A and rAPOBEC1 in soybean or dicotyledonous plants.

[0060] Example 3

[0061] This example is based on the SoyCBE-V2 vector. In order to further optimize the vector structure and improve its gene editing efficiency, the SH3 linker sequence between nCas9 and PmCDA1 was optimized, and a human ssDBD was added to the C-terminus of the original SH3 linker. hsRAD51 Sequences (SEQ ID NO.1-2) were constructed to construct the ChyBE-01 vector ( Figure 6 ). Eight sgRNAs, Target 1 to 8, were selected from Table 1. The sgRNA fragments in Table 1 were assembled in series onto the plant single base editing vector by the BsaI-mediated Golden Gate method. Two groups of sgRNA series vectors were constructed, one of which was 5 sgRNAs, Target 1 to 5, and the other was 3 sgRNAs, Target 6 to 8. The sgRNA fragment series scheme is shown in Figure 7 .

[0062] Using the Agrobacterium-mediated soybean hairy root transformation method, the SoyCBE-V2 vector and ChyBE-01 vector after tandem assembly of sgRNA were transformed into soybeans respectively. Each vector corresponded to 15 hairy roots, and a total of about 30 soybean hairy roots were obtained, and three biological replicates were performed. The DNA of the hairy roots was extracted, the target region was amplified by PCR, and the PCR product was sent to HI-TOM second-generation sequencing. The editing efficiency of the target site was calculated based on the sequencing results. The results showed that compared with SoyCBE-V2, ChyBE-01 significantly improved the single-base editing efficiency of 3 sites among the 8 target sites, Target 1, Target 2, and Target 7, but did not significantly improve the single editing efficiency of the other 5 sites ( Figure 8 ).

[0063] The above results show that ssDBD hsRAD51 The structure has a certain effect in PmCDA1-mediated CBE single-base editing, but it does not have a significant effect in all tested sites. The reason may be that the linker sequence between nCas9 and PmCDA1 is too long. Therefore, based on the ChyBE-01 vector, the present invention deletes the SH3 linker between nCas9 and PmCDA1 and only retains 477bp of human ssDBD hsRAD51 domain sequence, and obtain the ChyBE-02 vector ( Fig. 9 ) and cloned soybean endogenous ssDBD GmRAD51a Sequence (SEQ ID NO.3-4) replaces ssDBD in ChyBE-2 vector hsRAD51 Sequence, construct ChyBE-03 vector ( Fig. 9 ). The sgRNA fragments in Table 1 were assembled in series onto the plant single base editing vector by the BsaI-mediated Golden Gate method. Two groups of sgRNA series vectors were constructed, one of which was 5 sgRNAs in series with Target 1 to 5, and the other was 3 sgRNAs in series with Target 6 to 8. The sgRNA fragment series scheme is shown in Figure 7 , and analyzed the efficiency of single-base editing.

[0064] Compared with ChyBE-01, the CBE editing efficiency of ChyBE-02 and ChyBE-03 was significantly improved at multiple sites; the single-base editing efficiency of ChyBE-02 at seven target sites, including Target1, Target2, Target4, Target5, Target6, Target7 and Target8, was significantly higher than that of ChyBE-01, while ChyCBE-03 was significantly higher than ChyBE-01 at all eight target sites tested. The editing efficiency of ChyBE-03 at Target2, Target3, Target7 and Target8 was further significantly improved compared with ChyBE-02, with the highest reaching 92.7% ( Fig.10 ). Therefore, the single-base editing method developed in this study replaces the original linker sequence (SH3 linker) between Cas9n and PmCDA1 with the human ssDBD hsRAD51 ssDBD GmRAD51a Domain sequences can significantly improve CBE base editing efficiency.

[0065] Example 4

[0066] In order to demonstrate the application of the efficient single-base editing method developed by the present invention in functional genomic research, this example further designed two sgRNAs (NFR-sg1 and NFR-sg2), in which the C in the sgRNA editing window is changed to T to produce a stop codon, and the mutation of the GmNFR1α gene can lead to a soybean non-nodulation phenotype. NFR-sg1 and NFR-sg2 were assembled into ChyBE-01, ChyBE-02 and ChyBE-03 empty vectors by the BsaI-mediated Golden Gate method, and tested by Hi-TOM sequencing. In the composite transgenic plants, the editing efficiencies of ChyBE-03 at the two targets of NFR-sg1 and NFR-sg2 were 43.4% and 75.9%, respectively, of which 13.3% and 27.8% of the lines were homozygous mutations ( Fig.11 Consistent with the signaling role of GmNFR1a in nodulation, no nodulation was observed in the homozygous mutant ( Fig.12 ).

[0067] In order to further test the application of the efficient CBE single-base editing method developed in this patent in the stable genetic transformation of soybeans, the present invention stably transformed soybeans with SoyBE-V2 and ChyBE-03 vectors carrying NFR-sg1 and NFR-sg2 target sgRNA. In the T0 generation, a total of 17 positive transgenic plants were obtained with the SoyBE-V2 vector, and a total of 19 positive transgenic plants were obtained with ChyBE-03. After extracting their DNA and PCR amplification, the single-base editing efficiency in soybean stable transformation was statistically analyzed by first-generation sequencing. The results showed that the single-base editing efficiency of the ChyBE-3 vector in the target NFR-sg1 reached 47.4%, and the single-base editing efficiency in NFR-sg2 reached 52.6%, both higher than the SoyBE-V2 vector ( Fig.13 ).

[0068] In order to verify the mutation in soybean T0 transgenic plants, the present invention predicted the off-target of NFR-sg1 and NFR-sg2 according to the online website CRISPR-GE (http: / / skl.scau.edu.cn / offtarget / ) and selected the first 5 sites as off-target detection sites. Two plants in the ChyBE-03 positive seedlings were used to detect the mutation of 10 off-target sites in NFR-sg1 and NFR-sg2 by first-generation sequencing. The results are as follows: Fig.14 As shown, no off-target effects were detected at any of the 10 off-target sites.

[0069] Example 5

[0070] To further expand the PAM sequence selection range of CBE editors, this example uses a variant of spCas9n, SpCas9nNG (the variant SpCas9n NG contains mutations at the following amino acid positions: 10 A.R 1335 A.L 1111 R, D 1135 V.G. 1218 R, E 1219 F.A 1322 R and T 1337 R) Replace spCas9n in ChyBE-01, ChyBE-02 and ChyBE-03 vectors to obtain vectors ChyBE-NG-01, ChyBE-NG-02 and ChyBE-NG-03.

[0071] Four sgRNAs with PAM sequences of NGA, NGT, NGC, and NGG were selected (Table 1) and assembled into ChyBE-NG-01, ChyBE-NG-02, and ChyBE-NG-03 empty vectors respectively by the BsaI-mediated GoldenGate method, and then tested by Hi-TOM sequencing and the editing efficiency was determined. Fig.15 As shown, under NGG PAM, the editing efficiency of ChyBE-NGs (ChyBE-NG-01, ChyBE-NG-02, and ChyBE-NG-03) was lower than that of ChyBE-03, while they showed significant activity at NGC, NGT, and NGA targets. In particular, the editing efficiency of ChyBE-NG-03 was 86.7% in Target NGT, 36.6% in Target NGC, and 19.4% in Target NGA. Therefore, the targeting range of CBE single-base editing in soybean was further expanded by fusing spCas9n-NG variants.

[0072] In summary, the present invention has developed a new set of efficient CBE tools and methods based on PmCDA1 ( Fig.16 ), which has high editing efficiency and expanded targeting range in soybean, will facilitate functional genomic research and molecular breeding of soybean and other dicotyledonous plants.

[0073] The above description is only a preferred embodiment of the present invention. All equivalent changes and modifications made according to the scope of the patent application of the present invention should fall within the scope of the present invention.

Claims

1. A method for improving the editing efficiency of a dicotyledonous plant single base editor, characterized in that: In the dicotyledonous plant single-base editor system, the uracil glycosylase inhibitor, sgRNA scaffold and connecting peptide were modified to improve the editing efficiency; The single-base editor system is a CRISPR / Cas9n-based cytosine to thymine single-base editor system, and the base sequence is shown in SEQ ID NO.7; Among them, modifying the uracil glycosylase inhibitor refers to modifying the number of the uracil glycosylase inhibitor; the modification of the number of the uracil glycosylase inhibitor specifically refers to increasing the number of the uracil glycosylase inhibitor from 1 to 2; Among them, modifying the sgRNA scaffold refers to modifying the stem-loop structure and the fifth base of the sgRNA scaffold; the modification of the stem-loop structure and the fifth base of the sgRNA scaffold specifically includes protruding the fifth base T of the sgRNA scaffold into a C, adding "TGCTG" after the 12th base, and adding a "CAGCA" sequence after the 16th base to extend the stem-loop structure; Among them, modifying the connecting peptide refers to replacing the SH3 linker sequence between nCas9 and PmCDA1 with a ssDBD domain sequence; the ssDBD domain sequence is selected from the human ssDBD shown in SEQ ID NO.2 hsRAD51 The endogenous ssDBD of soybean shown in the sequence or SEQ ID NO.4 GmRAD51a sequence; The dicotyledonous plant is soybean.

2. The method according to claim 1, characterized in that: The Cas9 protein in the single-base editor system is also modified; modifying the Cas9 protein means replacing nCas9 with an nCas9 variant; the mutations of the nCas9 variant consist of mutations at the following amino acid positions: D10A, R1335A, L1111R, D1135V, G1218R, E1219F, A1322R and T1337R.

3. Use of the method of claim 1 in constructing a single base editing vector for dicotyledonous plants, characterized in that: The dicotyledonous plant is soybean.

4. A single base editing vector for improving the efficiency of single base editing of cytosine to thymine in dicotyledonous plants, characterized in that: The single-base editing vector is constructed by the method described in claim 1, and the dicotyledonous plant is soybean.

5. Use of the method of claim 1 or the single base editing vector of claim 4 in single base editing of dicotyledonous plant genomes, characterized in that: The dicotyledonous plant is soybean.

Images