CRISPR/Cas9-Mediated Multiple Knockouts in Abscisic Acid Receptor Genes Reduced the Sensitivity to ABA during Soybean Seed Germination

Abscisic acid (ABA) is an important plant hormone that regulates numerous functions in plant growth, development, and stress responses. Several proteins regulate the ABA signal transduction mechanism in response to environmental stress. Among them, the PYR1/PYL/RCAR family act as ABA receptors. This study used the CRISPR/Cas9 gene-editing system with a single gRNA to knock out three soybean PYL genes: GmPYL17, GmPYL18, and GmPYL19. The gRNA may efficiently cause varying degrees of deletion of GmPYL17, GmPYL18, and GmPYL19 gene target sequences, according to the genotyping results of T0 plants. A subset of induced alleles was successfully transferred to progeny. In the T2 generation, we obtained double and triple mutant genotypes. At the seed germination stage, CRISPR/Cas9-created GmPYL gene knockout mutants, particularly gmpyl17/19 double mutants, are less susceptible to ABA than the wild type. RNA-Seq was used to investigate the differentially expressed genes related to the ABA response from germinated seedlings under diverse treatments using three biological replicates. The gmpyl17/19-1 double mutant was less susceptible to ABA during seed germination, and mutant plant height and branch number were higher than the wild type. Under ABA stress, the GO enrichment analysis showed that certain positive germination regulators were activated, which reduced ABA sensitivity and enhanced seed germination. This research gives a theoretical basis for a better understanding of the ABA signaling pathway and the participation of the key component at their molecular level, which helps enhance soybean abiotic stress tolerance. Furthermore, this research will aid breeders in regulating and improving soybean production and quality under various stress conditions.


Introduction
Soybean (Glycine max) is an important economic crop; it has high nutritional value and is commonly used as a prime source of vegetable oils and proteins for human consumption [1][2][3]. In many plants, especially annual seed plants, seed germination plays a key role in the survival and spread of plants under suitable environmental conditions [4,5].
Abscisic acid (ABA) is an essential plant hormone; which governs various aspects of plant growth and development, including seed development, bud dormancy, and is a negative regulator of seed germination [6][7][8][9]. There is an antagonistic relationship between dormancy and seed germination of plants [10]. During germination, the ABA content dropped, indicating that the ABA signaling pathway was inhibited. The seeds of typical ABA defective mutants germinated faster than the wild types [11]. The seeds of transgenic plants constitutively expressing ABA biosynthetic genes maintained deep dormancy [12,13]. the species in the tree. Three soybean homolog genes Glyma.13G041800, Glyma.03G066200, and Glyma.06G126100, were named GmPYL17, GmPYL18, and GmPYL19 and displayed homology to AtPYL8 from Arabidopsis. The 15 homolog genes with AtPYL8 clustered into two groups, dicotyledons, and monocotyledons. According to the phylogenetic analysis, legumes were grouped within the dicotyledonous group, and these three soybean PYL8 proteins were closely related to those of alfalfa and kidney beans ( Figure 1A-C). These PYL proteins contain a common polyketide cyclase domain 2 (Pfam: Polyketide_cyc2). It is a member of the superfamily of the START domain and the key domain of PYL as an ABA receptor [40]. There are six conserved motifs (motif1~motif6) which were identified in six species, with lengths of 50,50,50,21,11, and 29 amino acids, respectively (Table S1). The motif shared by all PYLs proteins is motif1, motif2, and motif3, which are consistent with the multiple sequence alignment analysis ( Figures 1D and S1). The FPKM level of GmPYL17, GmPYL18, and GmPYL19 genes in seeds was significantly higher than that in aboveground parts, suggesting that these three genes may play an important role in seeds ( Figure 1E). To explore the function of GmPYL17, GmPYL18, and GmPYL19 in seeds, we used the CRISPR-Cas9 approach to create the loss of function mutants of these three genes in Dongnong 50 (DN50) background.

GmPYL17, GmPYL18, and GmPYL19 Are Homologous with AtPYL8 and Expressed Specifically in Dry Seeds
Protein sequence alignment was used to identify PYLs protein homologs in legumes with other plant species. After amino acid sequence alignment and verification on the NCBI database BLASTp, three genes in soybean, two in alfalfa, one in common bean, four in rice, and four in tomato were revealed to be homologous to AtPYL8 (At5g53160). In phylogenetic tree construction, the Marchantia polymorpha is the root node, the common ancestor of all the species in the tree. Three soybean homolog genes Glyma.13G041800, Glyma.03G066200, and Glyma.06G126100, were named GmPYL17, GmPYL18, and GmPYL19 and displayed homology to AtPYL8 from Arabidopsis. The 15 homolog genes with AtPYL8 clustered into two groups, dicotyledons, and monocotyledons. According to the phylogenetic analysis, legumes were grouped within the dicotyledonous group, and these three soybean PYL8 proteins were closely related to those of alfalfa and kidney beans ( Figure 1A-C). These PYL proteins contain a common polyketide cyclase domain 2 (Pfam: Polyketide_cyc2). It is a member of the superfamily of the START domain and the key domain of PYL as an ABA receptor [40]. There are six conserved motifs (motif1~mo-tif6) which were identified in six species, with lengths of 50, 50, 50, 21, 11, and 29 amino acids, respectively (Table S1). The motif shared by all PYLs proteins is motif1, motif2, and motif3, which are consistent with the multiple sequence alignment analysis ( Figures 1D  and S1). The FPKM level of GmPYL17, GmPYL18, and GmPYL19 genes in seeds was significantly higher than that in aboveground parts, suggesting that these three genes may play an important role in seeds ( Figure 1E). To explore the function of GmPYL17, GmPYL18, and GmPYL19 in seeds, we used the CRISPR-Cas9 approach to create the loss of function mutants of these three genes in Dongnong 50 (DN50) background.  Values are means ± SD (n = 3). Student's t-tests indicate significant differences in mean values relative to the mean value of seeds (** p < 0.01).

Generation of Transgenic Soybean Plants Harboring the CRISPR/Cas9 Expression Module
We used CRISPR-cas9 to create multiple mutants in three lines. In the T 2 generation, the homozygous mutant materials were screened. Among them, five bases were deleted at 1700-1705 bp of the GmPYL17 gene in gmpyl17/19-1 mutant line and two bases at 348-350 bp of the GmPYL19 gene. In the gmpyl17/18/19-1 mutant line, five bases at 1700-1705 bp of the GmPYL17 gene were deleted, seven bases at 1676-1683 bp of the GmPYL18 gene were deleted, and three bases at 347-350 bp of the GmPYL19 gene were deleted. gmpyl17/18/19-2 mutant line has five bases deleted at 1700-1705 bp of GmPYL17 gene, four bases deleted at 1680-1684 bp of GmPYL18 gene, and nine bases deleted at 342-351 bp of GmPYL19 gene. (Figure 2A-C). and blue, respectively. (D) Amino acid sequence of motif1 (green box), motif2 (blue box), and motif3 (red box) of PYLs in six species. (E) FPKM of GmPYL17, GmPYL18, and GmPYL19 in different tissues were determined from the transcriptome. The tissues are dry seed, shoot apical meristem of DN50, and Williams 82 (W82) plants. Values are means ± SD (n = 3). Student's t-tests indicate significant differences in mean values relative to the mean value of seeds (** p < 0.01).

Generation of Transgenic Soybean Plants Harboring the CRISPR/Cas9 Expression Module
We used CRISPR-cas9 to create multiple mutants in three lines. In the T2 generation, the homozygous mutant materials were screened. Among them, five bases were deleted at 1700-1705 bp of the GmPYL17 gene in gmpyl17/19-1 mutant line and two bases at 348-350 bp of the GmPYL19 gene. In the gmpyl17/18/19-1 mutant line, five bases at 1700-1705 bp of the GmPYL17 gene were deleted, seven bases at 1676-1683 bp of the GmPYL18 gene were deleted, and three bases at 347-350 bp of the GmPYL19 gene were deleted.  The homozygous mutant genes that have been sequenced were further verified by cleaved amplified polymorphism sequences (CAPS). The PCR products of wild type, heterozygous mutant, and homozygous mutant were digested by restriction endonuclease according to different mutation types of different genes (Table S3). The results showed homozygous mutations in all mutant lines ( Figure 3A-C). The homozygous mutant of the GmPYL17 gene was cut into two bands after PvuI digestion, the heterozygous mutant had The homozygous mutant genes that have been sequenced were further verified by cleaved amplified polymorphism sequences (CAPS). The PCR products of wild type, heterozygous mutant, and homozygous mutant were digested by restriction endonuclease according to different mutation types of different genes (Table S3). The results showed homozygous mutations in all mutant lines ( Figure 3A-C). The homozygous mutant of the GmPYL17 gene was cut into two bands after PvuI digestion, the heterozygous mutant had three bands, and the wild type could not be cut into a single band. The homozygous mutant of the GmPYL18 gene could not be cut into one band after SmII digestion, but the wild type could be cut into two bands, while the homozygous mutant of the GmPYL19 gene could not be cut into one band after SmII digestion (the size of the band was similar) ( Figure 3D).

The gmpyl17/19-1 Mutant Has Decreased ABA Sensitivity in Seed Germination
The MS medium with 0 µM (CK) and 10µM ABA was used to analyze the germination rate of gmpyl17/19-1, gmpyl17/18/19-1 mutant, and wild type. We define the seed hypocotyl to break through the seed coat as the boundary to determine whether the seed germinates or not. After 36 h, there was no significant difference in the seed germination rate of WT, gmpyl17/19-1, and gmpyl17/18/19-1 mutant treated with 0 µM ( Figure 5A,B). However, under the treatment of 10 µM ABA, the seed germination rate of gmpyl17/19-1 mutant was significantly higher than that of the wild type. The root length was also significantly higher than that of WT ( Figure 5C). After 72 h, the germination rate of WT, gmpyl17/19-1, and gmpyl17/18/19-1 mutant treated with 0 µM reached 100%. However, under the treatment of 10 µM ABA, the germination rate of gmpyl17/19-1 mutant was close to 80%, and the germination rate of WT was less than 50%. The root length of gmpyl17/19-1 mutant was still significantly higher than that of WT ( Figure 5D,E). These indicated that 10µM ABA inhibited the seed germination of WT and gmpyl17/18/19-1 mutant but had less effect on the gmpyl17/19-1 mutant. We previously found that the gmpyl17/18/19-2 mutant also showed that seed germination was inhibited by ABA, which was similar to that of the gmpyl17/18/19-1 mutant (not shown). According to the amino acid sequence analysis, we found that the GmPYL19 gene of gmpyl17/19-1 mutant was terminated prematurely, and the amino acid deletion of the GmPYL19 gene occurred in gmpyl17/18/19-1 and gmpyl17/18/19-2 mutants ( Figure 4C). Therefore, we think the GmPYL19 gene may be more critical in responding to ABA stress during seed germination.

The gmpyl17/19-1 Mutant Has Decreased ABA Sensitivity in Seed Germination
The MS medium with 0 μM (CK) and 10μM ABA was used to analyze the germination rate of gmpyl17/19-1, gmpyl17/18/19-1 mutant, and wild type. We define the seed hypocotyl to break through the seed coat as the boundary to determine whether the seed germinates or not. After 36 h, there was no significant difference in the seed germination rate of WT, gmpyl17/19-1, and gmpyl17/18/19-1 mutant treated with 0 μM (Figure 5A,B). However, under the treatment of 10 μM ABA, the seed germination rate of gmpyl17/19-1 mutant was significantly higher than that of the wild type. The root length was also significantly higher than that of WT ( Figure 5C). After 72 h, the germination rate of WT, gmpyl17/19-1, and gmpyl17/18/19-1 mutant treated with 0 μM reached 100%. However, under the treatment of 10 μM ABA, the germination rate of gmpyl17/19-1 mutant was close to 80%, and the germination rate of WT was less than 50%. The root length of gmpyl17/19-1 mutant was still significantly higher than that of WT ( Figure 5D,E). These indicated that 10μM ABA inhibited the seed germination of WT and gmpyl17/18/19-1 mutant but had less effect on the gmpyl17/19-1 mutant. We previously found that the gmpyl17/18/19-2 mutant also showed that seed germination was inhibited by ABA, which was similar to that of the gmpyl17/18/19-1 mutant (not shown). According to the amino acid sequence analysis, we found that the GmPYL19 gene of gmpyl17/19-1 mutant was terminated prematurely, and the amino acid deletion of the GmPYL19 gene occurred in gmpyl17/18/19-1 and gmpyl17/18/19-2 mutants ( Figure 4C). Therefore, we think the GmPYL19 gene may be more critical in responding to ABA stress during seed germination.

The gmpyl17/19-1 Mutant Induce Positive Regulators of Seed Germination
The sensitivity to ABA was reduced during seed germination due to GmPYL17 and GmPYL19 gene mutations. The RNA-seq was performed from the mutant seeds of gmpyl17/19-1 line after 36 h of treatment with 10 μM ABA and control. When the differentially expressed genes (DEGs) were calculated (absolute log2 fold value > 1 and adjusted p < 0.05), the gmpyl17/19-1 mutant had fewer DEGs (5250 DEGs, with 1515 upregulated and 3735 downregulated) compared to WT (10,369 DEGs, with 4733 upregulated and 5636 downregulated) in response to ABA stress. A lesser proportion of ABA stress-repressed DEGs (71.14%) in the gmpyl17/19-1 mutant than the 54.35% in WT, whereas the upregulated DEGs showed a reverse change trend ( Figure 6A). Under normal conditions, 2195

The gmpyl17/19-1 Mutant Induce Positive Regulators of Seed Germination
The sensitivity to ABA was reduced during seed germination due to GmPYL17 and GmPYL19 gene mutations. The RNA-seq was performed from the mutant seeds of gmpyl17/19-1 line after 36 h of treatment with 10 µM ABA and control. When the differentially expressed genes (DEGs) were calculated (absolute log2 fold value > 1 and adjusted p < 0.05), the gmpyl17/19-1 mutant had fewer DEGs (5250 DEGs, with 1515 upregulated and 3735 downregulated) compared to WT (10,369 DEGs, with 4733 upregulated and 5636 downregulated) in response to ABA stress. A lesser proportion of ABA stress-repressed DEGs (71.14%) in the gmpyl17/19-1 mutant than the 54.35% in WT, whereas the upregulated DEGs showed a reverse change trend ( Figure 6A). Under normal conditions, 2195 DEGs had higher expression levels in the gmpyl17/19-1 mutant than WT, and 2033 DEGs had lower expression levels in the gmpyl17/19-1 mutant than WT; and when subjected to ABA stress treatment, more DEGs were present between the genotypes ( Figure 6B). the expression of GmAHG3 and GmHONSU was upregulated, and the seed germination rate increased, which was consistent with the previous research conclusions.
The qRT-PCR results showed no significant difference in the expression level of the AHG3 gene between WT and gmpyl17/19-1 mutant without ABA treatment, as did HONSU. However, it is worth noting that neither AHG3 nor HONSU genes were significantly changed in WT after 10 μM ABA treatment, but their expression was significantly increased in gmpyl17/19-1 mutant ( Figure 6E,F). These results indicated that the gmpyl17/19-1 mutant promoted AHG3 and HONSU expression under 10 μM ABA treatment and improved the germination rate of soybean seeds. Primers used for qRT-PCR in this study are shown in Table S5. Based on the expression patterns, the DEGs associated with both ABA stress and GmPYL17/19 regulation can be grouped into six clades ( Figure 6C,D). To test which clade contributes more to the different growth and stress response of the gmpyl17/19-1 mutant and WT, we performed a Gene Ontology (GO) enrichment analysis of the six clade DEGs ( Figure S4, Table S4). Among them, Glyma.18g035000 (homologous gene of AHG3 ( Figure S5A,C) and Glyma.19g069200 (homologous gene of HONSU ( Figure S5B,D) were included in clade 3. Previous studies have reported that these two genes are positive regulators of seed germination. We knocked out GmPYL17 and GmPYL19, and under ABA treatment, the expression of GmAHG3 and GmHONSU was upregulated, and the seed germination rate increased, which was consistent with the previous research conclusions.
The qRT-PCR results showed no significant difference in the expression level of the AHG3 gene between WT and gmpyl17/19-1 mutant without ABA treatment, as did HONSU. However, it is worth noting that neither AHG3 nor HONSU genes were significantly changed in WT after 10 µM ABA treatment, but their expression was significantly increased in gmpyl17/19-1 mutant ( Figure 6E,F). These results indicated that the gmpyl17/19-1 mutant promoted AHG3 and HONSU expression under 10 µM ABA treatment and improved the germination rate of soybean seeds. Primers used for qRT-PCR in this study are shown in Table S5.

gmpyl17/19-1 Mutant Enhanced Morphological Traits but Inconsequential Effect on 100-Seeds Weight
To further explore the agronomic characteristics of the mutant lines, the plants were sown in the greenhouse until the flowering stage, the number of branches and plant height were recorded, as well as the 100-seed weight at maturity. At the R2 flowering stage, the number of branches in the mutant lines was significantly higher than in the wild type ( Figure 7A-C). The plant height of the R8 mature wild type and mutant was measured, and it was discovered that all mutants were taller than the wild type, with the gmpyl17/19-1 mutant being substantially higher than the wild type. There was no substantial difference between the mutant and the wild type in terms of 100-seeds weight ( Figure 7D).

gmpyl17/19-1 Mutant Enhanced Morphological Traits but Inconsequential Effect on 100-Seeds Weight
To further explore the agronomic characteristics of the mutant lines, the plants were sown in the greenhouse until the flowering stage, the number of branches and plant height were recorded, as well as the 100-seed weight at maturity. At the R2 flowering stage, the number of branches in the mutant lines was significantly higher than in the wild type ( Figure 7A-C). The plant height of the R8 mature wild type and mutant was measured, and it was discovered that all mutants were taller than the wild type, with the gmpyl17/19-1 mutant being substantially higher than the wild type. There was no substantial difference between the mutant and the wild type in terms of 100-seeds weight ( Figure 7D).

Discussion
Soybean provides more than half of the global oilseed production and a quarter of the protein for human food. Soybean seeds contain 40% protein and 20% oil [41]. To guarantee a sufficient food supply for the human population, its source should be increased by 50% by 2050 [42][43][44]. In turn, more effective plant breeding of soybean varieties is urgently needed. A comprehensive evaluation and utilization of genetically diverse germplasm are essential for crop improvement. CRISPR/Cas9 technology has the advantages of complete knockout expression, high specificity, low miss rate, heredity, functional research, etc. It has been successfully used for gene editing in plant species, including Arabidopsis, tomato, rice, wheat, and soybean [45][46][47][48][49][50][51].
In the construction of CRISPR/Cas9 multi-gene editing vector, the design principles of sgRNA still follow the following points: the length of sgRNA is about 20 bp, neither too long nor too short, too long can easily lead to vector construction failure, too short can easily increase the probability of miss events; the position of sgRNA should be before the PAM (NGG) sequence; the GC content of sgRNA should be about 50% [52,53]. However, the experiment shows that the GC content of 70% is also effective; in the design of sgRNA, there is a 1-2 base mismatch far away from the PAM sequence, which can also successfully target genes [54,55]. We also counted the transformation efficiency. Concerning the design of a vector with multiple genes corresponding to a single gRNA, it was found that the efficiency of simultaneous editing of three genes was 3.3%. Although the efficiency of triple editing was relatively low, it was still feasible ( Figure S3).
Most flowering plants reproduce through sexual reproduction and seed production. Seed germination not only determines the reproduction and survival of plant populations but also determines the time for plants to enter the natural and agroecosystem, which directly affects the yield of crops. As an important plant hormone, ABA plays an important role in seed germination [28].
Previous studies have shown that rice PYLs multi-gene editing materials edited by CRISPR/Cas9 system are not sensitive to ABA during seed germination [47]. Among the rice Subclass-I and -II SnRK2s, osmotic stress/ABA-activated protein kinase 2 (SAPK2) may be the primary mediator of ABA signaling [56]. Rice mutants with OsSARK2 deficiency also showed low sensitivity to ABA during seed germination [57]. This study also showed that the gmpyl17/19-1 mutant was less sensitive to ABA than the wild type under the 10 µM ABA treatment. The GO enrichment analysis of the clade1 also enriched many transcription factors, such as Glyma.07G156200 (DREB2C), which regulates seed germination and responds to drought stress [18]. Previous studies have carried out a lot of research on PYL-mediated drought resistance. AtPYL5 can reduce the transpiration rate of plants under drought stress, minimize water loss, reduce oxidative stress damage, and improve photosynthesis efficiency [58]. PYL4 significantly reduced the water consumption of transgenic wheat throughout its life cycle and maintained wheat yield under drought conditions [59]. Ectopic expression of AtPYL13 in Arabidopsis thaliana leads to ABA hypersensitivity; affects seed germination and stomatal closure [60]. AtPYL4 can reduce stomatal conductivity and improve water use efficiency [61]. AtPYR1, AtPYL1, AtPYL2, and AtPYL3 can improve the drought resistance and water use efficiency of Arabidopsis thaliana. AtPYL5 enhances the drought resistance of plants by directly inhibiting type A PP2Cs and activating the ABA signal [62]. The mutants we created may also affect soybean drought or flooding stress, and further experiments are needed to explore these functions.
In the gmpyl17/19-1 mutant, the mutation type of the GmPYL19 allele is early termination of translation, but in gmpyl17/18/19-1 mutant, the mutation type of GmPYL19 allele is amino acid deletion. Therefore, we believe GmPYL19 may play a key role in soybean seed germination. In gmpyl17/19-1 mutant, the mutation type of the GmPYL19 allele is early termination of translation, whereas, in gmpyl17/18/19-1 mutant, the mutation type of GmPYL19 allele is an amino acid deletion. However, the agronomic characters and the mutant lines with early termination of translation during seed germination showed greater differences than those of the wild type, indicating that the early termination of the GmPYL19 gene may be involved in more regulation of soybean growth, development, and germination. Therefore, GmPYL17, 18, 19, and especially GmPYL19 genes play a key role in seed germination and affect agronomic traits such as plant height and the number of branches. Our results reveal an important new component in soybean seed germination.

CRISPR/Cas9 Vector Construction and Target Selection
The nucleotide sequence of 14 AtPYLs and their corresponding encoded proteins were searched from the TAIR database (https://www.arabidopsis.org/, accessed on 17 December 2022). According to the AtPYLs amino acid sequence of Arabidopsis thaliana, the homologous genes of other species were found from the JGI database (https://phytozome-next.jgi.doe. gov/, accessed on 17 December 2022), the candidate GmPYLs proteins were selected.
The NCBI database identified the members of the GmPYLs gene family by BLASTp comparison. The amino acid sequences of proteins encoded by members of the OsPYLs gene family were obtained from the NCBI database (http://plants.ensembl.org/index.html, accessed on 17 December 2022). The phylogenetic tree was constructed using the Neighborjoining method by MEGA 10, and 3000 confidence tests were carried out by the bootstrap method to analyze the phylogenetic relationship between the GmPYLs gene family and AtPYLs and OsPYLs gene family [63]. Finally, the homozygous genes with the AtPYL8 gene in soybean were determined.
The skeleton of the vector is pSC1, sgRNA was promoted by the GmU6 promoter, and the Cas9 gene was expressed by the GmUbi3 promoter (provided by Yu Deyue, Nanjing Agricultural University); Escherichia coli, DH5α (purchased from Transgen Biotech Company); Agrobacterium tumefaciens, EHA105 (Provided by Feng Xianzhong, Northeast Institute of Geography and Agricultural Ecology, Chinese Academy of Sciences). The soybean cultivar used in the experiment was Dongnong50 (DN50) (Provided by Jiang Zhenfeng, Northeast Agricultural University). DNA oligonucleotides and gRNA are provided by Qingke Biology Company (Table S2). LgUI digested the empty pSC1 vector; using T4 ligase, the digested vector was ligated with dimerized oligonucleotides, and finally, the transgenic vector was constructed. Sanger sequencing was performed to validate the transgenic vector's sequences and orientation of different components. The constructed plasmid was transferred into Agrobacterium tumefaciens EHA105 for the stable transformation of the soybean.

Stable Soybean Transformation and Transgene Confirmation
The cotyledon Agrobacterium tumefaciens-mediated genetic transformation was carried out according to the established procedure, and the recipient was Dongnong 50 soybean variety [64]. When the first triple compound leaf grew out of the tissue culture seedling (V1 stage), a compound leaf was cut, and the DNA of soybean leaves was extracted by the CTAB method (Table S3). The PCR reaction was carried out with the specific primers of the pSC1-Cas9 gene and gRNA, which was used to verify the successful transfer of the vector into the tissue culture plant.

Acquisition and Identification of GmPYLs Mutants
The single sgRNA was designed to target GmPYL17, GmPYL18, and GmPYL19 based on the recognition and cutting principles of Cas9 protein in the CRISPR/Cas9 vector and referenced to whole-genome resequencing data of DN50 ( Figure S6A). In the three GmPYLs genes, the gRNA can target the second exon in the GmPYL17, GmPYL18, and GmPYL19 genes, the gRNA targets the same sequence. In previous studies, the SNP at the distal end of the PAM sequence did not affect the recognition of the Cas9 protein [47]. This remarkable approach, which directs one gRNA to three unique genes simultaneously, greatly improves the efficiency of gene editing mutant generation. The designed gRNA can knock out PYL17, PYL18, and PYL19 genes in soybean. The GmUbi3 promoter drives the Cas9 protein in the CRISPR vector, and gRNA is driven by the GmU6 promoter ( Figure S6B). The DN50 was selected as an explant for genetic transformation and directed mutagenesis because of its high transformation efficiency, short growth cycle, and suitability for growth chamber culture.
The Agrobacterium-mediated genetic transformation with soybean cotyledon node method was used, and 28 T 0 -transgenic lines were obtained ( Figure S2). Firstly, the method of applying herbicide on the leaves (200 mg/L glyphosate) was used to identify successful transformation. Then, the existence of the T-DNA sequence was confirmed by PCR screening using specific primers of the cas9 gene and specific primers of the gRNA-pSC1 crossing region in transgene (Table S2).
To identify CRISPR/Cas9-induced mutant genes, primers G13F/G13R, G03F/G03R, G06F/G06R were used to amplify DNA fragments covering the target regions of GmPYL17, GmPYL18, and GmPYL19 genes from the genomic DNA of transgenic plants, respectively. Sanger sequencing revealed the amplified fragments, as well as two forms of mutation in the gene target region (Table S3). We planted T 0 plants in a greenhouse, collected seeds, and continued sowing the harvested seeds (T 1 generation). When the unifoliolate leaves were fully unrolled, DNA was extracted, and three pairs of specific primers (G13F/G13R, G03F/G03R, G06F/G06R) were used to amplify the target fragments from the genomic DNA of transgenic plants and Sanger sequencing.

Genetic Markers Development
The sequence information of GmPYL17, GmPYL18, and GmPYL19 mutation sites was obtained by Sanger sequencing (Beijing Tsingke Biology Co., Ltd., Beijing, China). Primers with an amplification product of about 800 bp were designed. PCR products with a single clear band and accurate band size detected by electrophoresis were sequenced. Further, the cleaved amplified polymorphic sequence (CAPS) primers were designed for more confirmation. The CAPS marker was developed based on the GmPYL17, GmPYL18, and GmPYL19 sequence variations. After 37 • C for 4 h of enzymatic digestion, 4% agarose gel was used for electrophoresis.

Seed Treatment
Soybean seeds of wild type and mutants were soaked in 58 • C warm water and dried for 2 days. The well-dried seeds were selected for the germination experiment. Put two layers of filter paper at the bottom of the petri dish and add 25 soybean seeds to each plate. For the control experiment, we added 20 mL aseptic water; for the treatment group, we added 20 mL ABA solution (10 µM) and put it in dark culture at 25 • C, 60% relative humidity. The total number of seeds in each group (control and treatment) is more than 75 (3 Petri dishes per group). Take pictures every 6 h and record the number of germinated seeds in each group. After 60 h, calculate the germination rate of the treatment and control groups [65].

Analysis of Agronomic Characters of Mutant Plants
The T 2 generation mutants and wild type plants were planted in the experimental field of Northeast Forestry University. The T 3 strain was planted in the greenhouse of the College of Life Sciences of Northeast Forestry University (26 • C, light 14 h/dark 10 h). Place the pots in a non-porous sink and water from the bottom. Keep the surface of the soil slightly moist. Half of the soil used for planting was dug from the field, plus half of the grass charcoal soil. The phenotypes of T 3 mutants and wild type were observed, and the flowering time was counted. The plant height and branch number of mutants and wild type plants at full maturity (R8) were recorded. After seed threshing, the 100-seed weight of mutant and wild type seeds was counted, and Student's t-test was used to analyze the difference.

RNA Extraction, Library Construction, and Transcriptome Sequencing
Total mRNA was extracted from whole soybean seeds treated with water or ABA solution (10 µM) by the Trizol method. The RNA-seq data were obtained by high-throughput transcriptome sequencing, and the sequencing platform was Illumina Hiseq2500 at Shanghai Biozeron Biological Technology Co., Ltd. (Shanghai, China). Each sample produces 6.0-Gb sequencing data, sequenced by double-terminal (PE) sequencing with a reading length of 150 bp. Original data was obtained by sequencing to obtain high-quality clean reads by Fastp [66]. The clean reads were compared with the soybean reference genome (Phytozome v13, https://phytozome-next.jgi.doe.gov, accessed on 17 December 2022) using hisat2 [67]. After that, the gene's FPKM value was quantitatively obtained using Stringtie [68]. The reads number of genes in each sample was obtained by Samtools [69]. Detection of differentially expressed genes (DEGs) using DEseq2 packages in the R package [70]. The criteria for screening differential genes were p < 0.05 and the fold change >2 (up-regulated or down-regulated). Gene Ontology was annotated by SoyBase (https://www.soybase.org/, accessed on 17 December 2022) [71]. The enrichment test was performed by the TBtools [72].

Conclusions
In the present study, three genes, GmPYL17, GmPYL18, and GmPYL19, were altered using the CRISPR/Cas9 system, resulting in three homozygous mutants. The morphological and genetic effects on plant growth and development were observed. The mutant plants showed significant changes in the height and number of branches compared to wild types. Furthermore, at the seed germination stage, the mutants (gmpyl17/19-1) were insensitive to ABA treatment, which confirmed that the GmAHG3 and GmHONSU were upregulated in the mutant lines and acted as positive regulators of seed germination. These findings provide a basis for the in-depth understanding of the ABA signal regulation network in soybean, especially the regulatory mechanism of seed germination.