**5. Genome editing provides the powerful tool for soybean design breeding**

Genome editing has emerged as more powerful approach for functional study and molecular design breeding compared with traditional genetics approaches, namely mutagenesis, transgenic RNAi, or overexpression in obtaining plant cultivars with predictable and desired traits [124, 125]. CRISPR (clustered regularly interspaced short palindromic repeat)/Cas (CRISPR-associated) is one of the most efficient genome editing systems and has been widely used in various plant species [126]. In 2015, the first knockout and DNA homology-directed recombination (HDR) in soybean plant was successfully obtained using CRISPR/Cas9 technology [127]. Du et al. discovered that altering the AtU6–26 promoter of the CRISPR/ Cas9 system to the GmU6-16g-1 promoter might considerably improve the efficacy of targeted mutagenesis in soybean [128]. Nearly 75% of the genes in soybean are duplicated, thus knocking out a single gene usually does not result in a mutant phenotype. It is critical to create a CRISPR/Cas9 system that can edit several homologous genes at the same time to obtain the desired phenotype. The CRISPR/ Cas9 system that can achieve multiplex mutagenesis with better efficiency was established by refining the phases of vector synthesis, sgRNA assessment, pooled transformation, and sgRNA identification [129]. Naturally, single-nucleotide polymorphism (SNP) variants account for a major portion of phenotypic variability in agronomic traits, in addition to alleles induced by loss-of-function mutations. When a gene function is disrupted by utilizing a gene-editing technique, mostly it results in undesirable phenotype that is difficult to optimize for agronomic trait improvement [27]. As a result, in molecular breeding, generating point mutations at specific locations impacting crucial agronomic properties is extremely valuable [130]. The CRISPR/Cas9 system has recently been used to develop "base editing," which changes single bases into another without the use of DNA DSBs or a donor template [131]. Cai et al. successfully used the technique to generate the point mutants of *GmFT2a* and *GmFT4* in soybean [132].

Currently, CRISPR/Cas9 is being frequently used in soybean functional studies [128, 133, 134]. For example, to identify the genes that are responsible for flowering time, frame shift mutations created by CRISPR/Cas9 revealed that *GmFT2a* functions primarily during short day (SD), while *GmFT5a* has more substantial impacts under long day (LD) [132, 135, 136]. Similarly, knockout of *GmPRR37* by the CRISPR system suggested that it can repress flowering under LD [137]. Male sterility was seen in two CRISPR/Cas9 gene-editing mutants of *Glyma.13G114200*, showing that it was the casual gene *GmMS1* responsible for male sterility [138–140].

#### **5.1 The advantages of genome editing for soybean molecular design breeding**

A key issue and important research goal for soybean researchers since the completion of the soybean genome sequencing project is to elucidate the function of 46–56 thousand identified genes [129]. Transgenic technology is a great tool for functional

#### *Soybean Molecular Design Breeding DOI: http://dx.doi.org/10.5772/intechopen.105422*

genomic research and crop genetic enhancement, but it has certain drawbacks when used in soybeans. *Agrobacterium-mediated transformation* and particle bombardment have been widely utilized to make transgenic soybean plants in recent decades. When considering its easy technique, low cost, single or low copy number of insertions, and relatively infrequent rearrangement, *Agrobacterium-mediated transformation* is a superior alternative [141]. However, because soybean is a refractory crop as far as the transformation and regeneration are considered, no sustained soybean genetic transformation has yet been developed, regardless of whatever technique is used. Furthermore, the efficacy of Agrobacterium-mediated transformation is affected by soybean tissue, cultivar, or species [142], resulting in a restricted number of soybeans types that may be enhanced directly by genetic modification. Transgenic technology is used to investigate gene function by integrating foreign DNA sequences into the plant genome, which results in either overexpression or silence of the target gene [142]. As a result of potential risks such as unintended gene insertions, endogenous gene disruption, and unpredictable gene expression that arise during transformation [127], transgenic plants frequently cause potential bio-safety issues and are subjected to regulatory restrictions on genetically modified organisms (GMOs).

Furthermore, being a diploid that developed from palaeotetraploid, soybean has a highly duplicated genome, with around 75% of projected genes possess multiple copies, resulting in substantial genetic redundancy and complicating the elucidation of soybean gene function. On the one hand, conventional random mutagenesis methods (physical, chemical mutagenesis, or T-DNA insertion) make it difficult to link genotype and phenotype because the loss of one homolog can be fully compensated by redundant homologous copies [143]. On the other hand, RNA-silencingbased technology frequently silences the entire gene family and is difficult to silence a single gene copy [144], and due to partial gene product depletion, the phenotype may be unstable [145]. To accelerate soybean gene function and breeding research, more accurate and efficient genetic engineering technology is required. Current genome editing technology has provided opportunity to overcome the aforesaid difficulties, at least in part. For starters, genome editing technology can not only alter a single gene without impacting other members of the gene family, but it can also edit many genes of interest at the same time using a single transformation, making it ideal for soybean and other polyploid crop studies [126].

Secondly, unlike transgenic technology, genome editing technology incorporates sequences that are genome editing components rather than foreign genes of interest into plant genomes. These genome editing components may be deleted once the target gene has been edited to yield transgene-free mutants, which are safer to employ in breeding and easier to commercialize under tight GMO rules. For example, in the United States, a gene-edited soybean oil with a different fatty acid profile was recently released (https://calyxt.com/ first-commercial-sale-of-calyxt-high-oleic-soybean-oil-on-the-usmarket/).

Finally, because the soybean genetic transformation is inefficient, one possible approach is to do direct genome editing without transformation or tissue culture. The standard procedure is to edit a variety with a high transformation efficiency to enhance one or more characteristics and then utilize the modified plant as a donor, which contains editing components. By hybridization with donor plants and subsequent backcrossing, the modified target gene or genome editing components can be introgressed into elite lines resistant to transformation. In conclusion, genome editing technology is a strong tool that has a lot of potential for speeding up soybean breeding [137, 146].

#### **5.2 Applications of genome editing in soybean improvement**

Hairy-root transformation mediated by *Agrobacterium rhizogenes* is a simple and rapid technique for studying soybean gene function, and it only takes a few weeks to get transgenic hairy root [142, 147]. When ZFNs, TALENs, and CRISPR/Cas9 systems were initially employed in soybean, researchers preferred to use the hairy-root transformation technique to quickly assess the efficiency of these genome editing tools before undertaking the time-consuming soybean whole-plant transformation [142, 147]. As a consequence, ZFNs technology delivered the first example of genome editing in soybean in 2011, producing heritable mutations in two homologous *DICER-LIKE genes,* namely *DCL4a* and *DCL4b* [144]. In 2015, five research groups successfully evaluated the mutation efficiency of the CRISPR/Cas9 system in soybean endogenous or exogenous genes in hairy roots, establishing a precedent for using CRISPR/Cas technology to study soybean genes [127, 148, 149]. One of these teams used NHEJ to accomplish about 76% targeted mutagenesis, and they also used HR to create mutant plants with targeted gene integration at the target location, as well as a chlorsulfuronresistant soybean with a mutated acetolactate synthase1 (ALS1) gene [127]. All of these early successes have highlighted the genome editing systems' remarkable potential to develop useful features in the near term through focused gene alterations.

Scientists evaluated the effectiveness of TALENs and CRISPR/ Cas9 in editing two phytoene desaturase genes in hairy roots, namely *GmPDS11* and *GmPDS18* [128]. The results demonstrated that CRISPR/Cas9, particularly CRISPR/Cas9 employing the GmU6-16g-1 promoter, was far more effective than TALENs at concurrently targeting two alleles. Another study was carried out to examine a variety of GmU6 promoters in soybean hairy roots and *Arabidopsis thaliana* to determine which ones were best for driving sgRNA production and discovered that GmU6-8 and GmU6-10 promoters had high activity, which improved editing effectiveness [150]. Both results are beneficial in the development of an effective CRISPR/Cas9 system for use in soybean research. In addition, the CRISPR/Cas9 method in combination with the hairy-root transformation technique has been used to edit and explore soybean storage protein genes [134], and the candidate gene governing nodulation specificity [151].

NHEJ has generally been used to delete target genes by tiny insertions or deletions, whereas HR is primarily utilized to replace or integrate targeted genes [152]. Using dual-sgRNA to cleave two neighboring loci on the same chromosome, two research studies recently showed that massive genomic deletions might be generated in soybeans [135, 153]. Due to the two editing chances, the dual-sgRNA design can not only boost gene mutation rates, but also construct substantial fragment deletion to ensure that the target gene is completely eliminated. Other fields of research will benefit from the substantial loss, such as understanding the role of regulatory elements or noncoding genes. Unlike prior HR-mediated donor integration, one study claimed that two large multigene donors (7.1 kb and 16.2 kb) were inserted into a target genomic location of soybean utilizing NHEJ and ZFNs, with donor heritability verified in T1 progeny plants [152]. This study found that NHEJ might be used instead of HR to induce accurate insertions of numerous transgenes in soybeans while avoiding the drawbacks of inefficient HR [154].

In addition to the abovementioned developments in genome editing systems to permit better application in soybean research, several researchers have used genome editing technology and entire plant transformation to investigate gene function or enhance agronomic features. The single and double mutants of the soybean genes, namely *DCL1a* and *DCL1b* using ZFNs-based mutagenesis techniques were created

to investigate their roles in the soybean miRNA system [143]. Only the *dcl1a/dcl1b* double mutants showed a morphological phenotype, which was characterized by shrunken and shriveled seed as well as seedling developmental abnormalities, whereas both *dcl1a* and *dcl1b* single mutants showed a normal phenotype, suggesting that *GmDCL1* homologs have functional redundancy. Besides, the two homologous genes, namely *GmPPD1* and *GmPPD2*, coding for Arabidopsis PEAPOD orthologs were edited with a single sgRNA [155]. As a result, double mutants with frame shift mutations displayed a severe phenotype of developmental abnormalities in leaf and pod development.

Previous results revealed that homozygous mutants produced by CRISPR/Cas9 mediated mutagenesis in *FLOWERING LOCUS T2a (GmFT2a)* delayed soybean flowering [133] and male sterile soybean lines [156]. They also used the CRISPR/Cas9 system to alter *GmFT5a* and then crossed it with *ft2a* mutants to get *ft2aft5a* double mutants. This double mutant bloomed 31.3 days later than wild-type plants and produced more pods as well as seeds under short day circumstances [136]. The researchers employed a multiplex genome editing method based on CRISPR/Cas9 to change four *SQUAMOSA PROMOTER BINDING PROTEIN-LIKE9* (*GmSPL9*) genes and generated several soybean mutants with different mutated locus combinations. These mutants revealed altered node number on the main stem and branch number [157]. CRISPR/Cas9 technology has also enhanced soybean seed oil profile [158], offensive beany flavor of soybean seed product [159], and isoflavone content and resistance to soybean mosaic virus [160]. The experiments mentioned above showed that genome editing technology has a lot of potential for improving soybeans.

Despite the fact that CRISPR/Cas9 technology has become the standard for genome editing, NGG PAM's fundamental restriction limits its use in highly precise genomic areas. ZFNs and TALENs have crucial tools as a complement to the CRISPR/ Cas9 system since their target range is infinite. The combined usage of these systems will aid the soybean breeding and functional genomics projects. For example, a group of genes were mutated that encode the main machinery proteins involved in small RNA processing in soybean and *Medicago truncatula* using CRISPR/Cas9 and TALENs [161]. Together with the previously reported mutants induced by ZFNs, the resultant mutant plants established a collection of mutant resources for future studies of short RNA biology in legume plants [143].

Many scientists used CRISPR/Cas9-based high-throughput mutagenesis to create a genome-wide mutant library or mutant collection related to certain functions for pooled functional screen, which provided a wealth of resource in mammalian cell, rice, and tomato [162–164]. The success of creating these CRISPR libraries, however, is strongly reliant on a reliable genetic transformation mechanism. The creation of a CRISPR library in soybean is not only a massive task, but it is also extremely important for research and breeding. A first soybean CRISPR library targeting over 100 candidate genes was recently developed. A collection of mutant soybean lines was also created utilizing an enhanced process at several important steps, demonstrating the feasibility and usefulness of using CRISPR/Cas9 technology to execute large-scale mutagenesis in soybeans [129]. Following the CRISPR/Cas9 system, new genomic editing tools such as the CRISPR/Cas12a system, BE systems, and other CRISPR/Cas variants have been developed. However, to our knowledge, most of these have yet to be employed in soybean, with the exception of one work [165] that used Cas12a-RNP in soybean protoplasts to induce gene editing. RNP-based genome editing may not be the best option for soybean research due to a lack of updated information on successful protoplast regeneration (**Figure 1**).

#### **Figure 1.**

*Genome editing platforms and editing outcomes. Each editing platform (arrow) and its outcomes (rectangular) are coded with the same color. ZFN, zinc-finger nuclease; TALEN, transcription activator-like effector nuclease; CRISPR, clustered regulatory interspaced short palindromic repeat; DSB, double-strand breaks; SSB, singlestrand breaks; outcomes of GE created by site-directed nucleases (SDN) include: SDN1 – The approach involves DNA breaks repair through DNA repair mechanisms in the host cellular without using an added repair template; SDN2 – The approach involves the break repair via HR using an added homologous repair template; and SDN3 – The approach involves DNA break repair via either HDR or NHEJ pathway using an added DNA template containing no homologous sequences but with homologous ends: Progress and application of genome editing in soybean improvement.*
