The power and versatility of genome editing tools in crop improvement

Increasing world population, global climate change, decreased farmland, environmental pollution and ecological deterioration represent unprecedent challenges for crop production to ensure global food security (Hickey et al., 2019; Li et al., 2021a). It is estimated that by the year 2050, 50% more food is needed to feed the increasing population (Bailey-Serres et al., 2019). Thus, it is urgent to boost crop production by using cutting-edge technologies. Genome editing technologies have revolutionized the plant research field and offer great potential in crop improvement (Ma et al., 2015; Li et al., 2021b; Zhan et al., 2021). Of the several genome editing technologies, clustered regularly interspaced palindromic repeats (CRISPR) and its associated Cas protein (CRISPR/Cas), has become the dominant one in the past several years because it is simple, efficient and cost-effective. The CRISPR/Cas nuclease generates a double-stranded DNA break (DSB) at the target site in a specific gene locus in the crop genome, causing random indel mutations via the error-prone non-homologous end joining (NHEJ) DNA repair pathway in cells (Ma et al., 2015; Li et al., 2021b; Zhan et al., 2021) or DNA sequence replacement or insertion through the homology-directed repair (HDR) pathway when a donor repair template (DRT) is available (Sun et al., 2016; Li et al., 2019; Lu et al., 2020). In addition, several nucleic acid deaminases have been fused with CRISPR/Cas nickase (nCas) (e.g., nCas9 (D10A) or nCas9 (H840A)) to achieve base editing. For example, cytidine deaminases and adenine deaminases can be fused with CRISPR/nCas9 (D10A) to generate cytosine and adenine base editors (CBEs and ABEs), which enable C∗G to T∗A and A∗T to G∗C transitions, respectively, in the target region of a specific gene (Li et al., 2017; Lu and Zhu, 2017; Ren et al., 2018; Yan et al., 2018). On the other hand, prime editor, which is engineered by fusing a mutated M-MLV-RT (Moloney murine leukemia virus reverse transcriptase) to the C-terminus of a catalytically impaired Cas9 (H840A), and programmed with a prime editing guide RNA (pegRNA) composed of a single guide RNA (sgRNA) targeting the specific site, a reverse transcription (RT) template encoding the desired edit, and a primer-binding site (PBS), can introduce all 12 classes of single nucleotide substitutions as well as predefined small indels without the need for DSB or DRT (Li et al., 2020a, 2020b; Lin et al., 2020; Tang et al., 2020; Xu et al., 2020a). Improving the editing efficiency, expanding the CRISPR toolkits and utilizing the CRISPR/Cas tools to improve crops are important for addressing the diverse challenges in crop production. In this Special Issue, we collected 10 articles to demonstrate the latest progress in crop genome editing, including improved base editing systems, expanded CRISPR toolbox, multiplex gene editing and the creation of novel rice, wheat, and maize germplasms. Many important agronomic traits are dictated by mutations of one or a few nucleotides. The CRISPR/Cas base editors such as the CBEs and ABEs generate base transitions and are therefore powerful tools for rapid and efficient crop improvements and directed evolution of agriculturally important genes in planta (for review, please see Bharat et al., 2020 and Zhan et al., 2021). However, the efficiencies of base editors, especially the ABEs, are generally not efficient enough to generate the homozygous base substitutions in one generation needed for crop improvement. In this issue, Wei et al. (2021) developed a high-efficiency adenine base editor, ABE8e in rice (rABE8e), by combining monomeric TadA8e, bis-bpNLS and codon optimization. The rABE8e editor substantially increased the editing efficiencies at NG-protospacer adjacent motif (PAM) and NGG-PAM target sequences compared with ABEmax (Koblan et al., 2018). For most targets, rABE8e exhibited nearly 100% editing efficiency and high homozygous substitution rates within the editing window, especially at positions A5 and A6. Xu et al. (2021) developed a novel plant dual-base editor version 1 (pDuBE1) by integrating an eCDAL (Japanese lamprey cytidine deaminase CDA1-like 4) into TadA8e to generate an efficient plant dual cytosine and adenine editor in order to simultaneously introduce C-to-T and A-to-G substitutions. The editing efficiency of pDuBE1 reached at 87.6%, with frequencies of concurrent A-to-G and C-to-T conversions as high as 49.7% in stably transformed plant cells. Development of these highly efficient rABE8e and pDuBE1 systems will greatly facilitate the application of base editing in crop improvement and directed evolution of agriculturally important genes. In general, genome editing is constrained by PAM and/or the editing window. Development of Cas9 variants with broader PAM ranges is an important aspect in expanding the editing scope of the CRISPR/Cas system. The original Cas9 from Streptococcus canis (ScCas9) can recognize simple NNG-PAM targets, thus expanding the target range, but its editing efficiency is low in plants (Xu et al., 2020b). To expand the CRISPR toolbox, Liu et al. (2021) systematically investigated the targeted gene mutation frequencies and cytidine base editing efficiencies of ScCas9 and two evolved ScCas9 variants, ScCas9+ and ScCas9++ in rice (Oryza sativa). The authors generated a new CBE editor, PevoCDA1-ScCas9n++ with the evoBE4max structure and ScCas9n++, and achieved stable and efficient multiplex base editing efficiencies at NNG-PAM with wider editing windows without target sequence context preference. They found that ScCas9++ was more efficient than the original ScCas9 and ScCas9+. Using ScCas9++, they generated new rice germplasms by simultaneously editing OsWx and Badh2. Ma et al. (2021) also demonstrated that ScCas9++ enables broader genome editing compared to the original ScCas9, and ScCas9++-based ABE showed an editing efficiency comparable with that of Cas9-based ABE in transgenic rice plants. Furthermore, they generated many new OsGS1 alleles using the ScCas9++-based ABE, which will benefit future screens of rice germplasms with potential herbicide resistance. Cpf1 (also known as Cas12a) is a class 2 type V CRISPR/Cas DNA endonuclease that is smaller than Cas9 (Fonfara et al., 2016; Zetsche et al., 2017). Cpf1 recognizes T-rich PAM sequences such as the TTTN and can process Cpf1-associated CRISPR repeats into mature crRNAs that function as guides without requiring the trans-activating crRNA (tracrRNA) needed for the CRISPR/Cas9 system (Zetsche et al., 2017). Although CRISPR/LbCpf1 system (Cpf1 from Lachnospiraceae bacterium ND2006) has been successfully applied in rice, maize and cotton (Wang et al., 2017, 2020; Li et al., 2018, 2019), large chromosomal segment deletions and multiplex gene editing by CRISPR/Cpf1 have not been documented in soybean. In this issue, Duan et al. (2021) reported the development of an efficient CRISPR/LbCpf1 system for multiplex gene editing and large chromosomal segment deletion in soybean (Glycine max). They demonstrated that their optimized system could reach an editing efficiency of 91.7% with a crRNA array comprising of a 21-nt direct repeat (DR) fused with a 22-nt guide RNA target. Eight gene targets could be edited simultaneously in one step using the system with eight guide RNA targets in a single crRNA array. Deletions of both large chromosomal segments with lengths of 10 Kb to 1 Mb and small fragments (<1 Kb) could be efficiently achieved in soybean using this multiplex gene editing system. These data suggest that the CRISPR/LbCpf1 system is robust and can provide significant benefits to soybean research and genetic improvement through multiplex gene editing or chromosome engineering. Rice is a major staple food crop consumed by nearly half of the world population. Each year, about 4.5 million hectares of rice are grown worldwide. Amylose content (AC), which is mainly regulated by the Waxy (Wx) gene, is a major indicator of rice eating and cooking quality (ECQ). Huang et al. (2021) generated a series of nucleotide substitutions in the middle region of the Wx gene using CBE and ABE. The ACs of the germplasm with novel Wx alleles ranged 0.3%–29.43%, whereas that of the wild-type is around 19.87%. Importantly, the waxyabe2 allele showed improved ECQ and favorable appearance without obvious undesirable agronomic traits. Besides, rice blast and bacterial blight are important diseases of rice (Oryza sativa) which are caused by the fungus Magnaporthe oryzae and the bacterium Xanthomonas oryzae pv. oryzae (Xoo), respectively. Breeding rice varieties with broad-spectrum resistance is an effective strategy for sustainable rice production. Although dominant resistance genes have been extensively used in rice breeding, generating disease-resistant varieties by manipulating susceptibility (S) genes that facilitate pathogen compatibility has not been documented. In this issue, Tao et al. (2021) used CRISPR/Cas9 to generate a series of loss-of-function mutants of the S genes Pi21 and Bsr-d1, which showed increased resistance to M. oryzae. They also generated a knockout mutant of the S gene Xa5 that showed increased resistance to Xoo. In particular, a triple mutant of all three S genes had significantly enhanced resistance to both M. oryzae and Xoo. Moreover, the triple mutant lines showed similar growth phenotypes and agronomic performance compared to the wild-type. Thus, simultaneous editing of multiple S genes is an effective strategy for generating new rice varieties with broad-spectrum resistance. Wheat (Triticum aestivum L.) is a major staple food crop consumed by more than 30% of the world population. Nitrogen (N) fertilizer has been applied extensively in agriculture to improve wheat yields. However, undue N fertilizer application and the low N use efficiency (NUE) of modern wheat varieties are exacerbating the problem of environmental pollution and ecological deterioration. An abnormal cytokinin response1 repressor1 (are1) mutant in rice exhibits increased NUE, and delayed senescence and consequently increased grain yields under N-limiting conditions (Wang et al., 2018). However, the function of ARE1 ortholog in wheat remains unknown. In this issue, Zhang et al. (2021) isolated and characterized three TaARE1 homeologs from an elite Chinese winter wheat cultivar ZhengMai 7698. The authors then used CRISPR/Cas9-mediated targeted mutagenesis to generate a series of transgene-free mutant lines either with partial or triple-null taare1 alleles. All transgene-free mutant lines showed enhanced tolerance to N starvation, delayed senescence and increased grain yields under normal field conditions. In particular, the AABBdd and aabbDD mutant lines exhibited delayed senescence and significantly increased grain yields without growth defects compared to the wild-type control. These results demonstrate the potential to edit ARE1 orthologs to breed high-yield wheat, which may also be extended to other important cereal crops. Maize (Zea mays L.) is a major staple food and feed crop worldwide. Aroma is an important quality parameter for breeding in rice. 2-acetyl-1-pyrroline (2AP) is a key scent compound identified in fragrant rice. Lower activity of BETAINE ALDEHYDE DEHYDROGENASE 2 (BADH2) is associated with fragrance in natural germplasms of diverse plant species. However, no aromatic germplasm has been documented in maize. In this issue, Wang et al. (2021) characterized the two maize BADH2 homologs, ZmBADH2a and ZmBADH2b. They then generated zmbadh2a and zmbadh2b single mutants and the zmbadh2a-zmbadh2b double mutant in four inbred maize lines using CRISPR/Cas. A popcorn-like scent was noticeable in seeds from the double mutant but not from either the single mutant or wild type. They detected 2AP in fresh kernels and dried mature seeds from the double mutant lines at levels of 0.028 to 0.723 mg/kg. These results suggest that ZmBADH2a and ZmBADH2b function redundantly in 2AP biosynthesis in maize. The authors thus generated the world's first aromatic maize germplasm by simultaneously editing the two BADH2 genes. Finally, reporter-assisted CRISPR/Cas systems can facilitate the selection of transgene-free edited plants following genetic segregation. Yan et al. (2021) established the seed fluorescence reporter (SFR)-assisted CRISPR/Cas9 systems in maize by using the red fluorescent DsRED protein expressed in the endosperm (En-SFR/Cas9), embryos (Em-SFR/Cas9), or both tissues (Em/En-SFR/Cas9). These three SFRs showed distinct fluorescent patterns in the seed endosperm and embryo, allowing easy identification of transgene-free edited seeds following segregation. In summary, impressive progresses have been made in crop genome editing, from improving the editing efficiencies, expanding CRISPR toolkits and applying these tools to generate novel germplasms. This Special Issue covers a selected range of topics in this research area. Expanding the CRISPR toolboxes and improving editing efficiencies would make genome editing more useful in crop plants. The generation of novel crop germplasms, which is difficult or impossible through conventional breeding, demonstrates the power and versatility of genome editing in crop improvement for better food, increased yield, enhanced disease resistance, and higher resource use efficiency that are critical for sustainable agriculture and global food security.