Engineering plant immune circuit: walking to the bright future with a novel toolbox

Summary

Plant pathogens destroy crops and cause severe yield losses, leading to an insufficient food supply to sustain the human population. Apart from relying on natural plant immune systems to combat biological agents or waiting for the appropriate evolutionary steps to occur over time, researchers are currently seeking new breakthrough methods to boost disease resistance in plants through genetic engineering. Here, we summarize the past two decades of research in disease resistance engineering against an assortment of pathogens through modifying the plant immune components (internal and external) with several biotechnological techniques. We also discuss potential strategies and provide perspectives on engineering plant immune systems for enhanced pathogen resistance and plant fitness.

Generating and developing disease‐resistant plant varieties are required to maintain a sustainable food stockpile and the environment. The ability to engineer immune regulatory components through several approaches, such as genetic transformation, CRISPR/Cas‐mediated gene knockdown/knockout, decoy engineering, pathogen effector‐based strategy, RNAi‐based biocontrol, and VIGS‐mediated gene suppression, can activate diverse adaptive and defence responses. This review summarizes and discusses the recent advances toward engineering disease resistance and future perspectives. Created with www.BioRender.com.

Introduction

Plants have evolved a sophisticated immune system to protect against pathogen invasion. Preformed immunity, including physical and chemical barriers such as leaf trichomes, cell walls, and surface pH, is utilized early in the attack of the pathogen. However, some pathogenic organisms can overcome this immunity and cause diseases in plants. To survive, plants have evolved innate immunity as a barrier, comprising complex pathogen‐interfering patterns called induced immunity, which is based on the recognition and activation of different immune receptors that sense the presence of pathogens (Deslandes and Rivas, 2012; Jones and Dangl, 2006; Zipfel and Robatzek, 2010). Transmembrane pattern recognition receptors (PRRs) are typically utilized in the first layer of this system to recognize virulence factors such as pathogen‐/microbe‐/damage‐associated molecular patterns (PAMPs/MAMPs/DAMPs, hereafter MAMPs) to induce a basal response known as pattern‐triggered immunity (PTI; Zipfel, 2009, 2014). Once PTI is activated, a series of defence responses is triggered to suppress the colonization of the pathogen (Bigeard et al., 2015; Torres et al., 2006). However, many pathogens can suppress PTI by introducing interfering molecules (called effectors) into the cell (Iswanto et al., 2021b; Stergiopoulos and de Wit, 2009; Tampakaki et al., 2010). Effectors target various host proteins to facilitate pathogen replication, colonization, and pathogenicity in plants to cause effector‐triggered susceptibility (ETS; Deslandes and Rivas, 2012; Raffaele and Kamoun, 2012). In turn, plants have developed a strategy to recognize specific effectors using intracellular nucleotide‐binding and leucine‐rich repeat receptors (NLRs; Bonardi et al., 2012; Nguyen et al., 2021; Su et al., 2018; Takken et al., 2006). Once the plant cell has sensed the effectors, it undergoes a robust and timely effector‐triggered immunity (ETI) response. ETI is often associated with programmed cell death and typically involves a hypersensitive response (HR) at the site of infection and systemic acquired resistance (SAR) in distal plant tissue to limit pathogen growth and proliferation (Cui et al., 2015; Durrant and Dong, 2004; Fu et al., 2012; Greenberg and Yao, 2004). Recent studies revealed the consequential relationship between PRR‐mediated PTI and NLR‐mediated ETI during bacterial infection (Ngou et al., 2021; Nguyen et al., 2021; Yuan et al., 2021). The continuous adaptation and evolution of PTI, ETS, and ETI have further strengthened the improvisation of both the host and pathogen to simultaneously diversify effector and resistance proteins (Delaux and Schornack, 2021; Deslandes and Rivas, 2012).

The plant immune system comprises a complicated system of immune genes with positive or negative effects. Immune receptors and positive regulators of plant immunity contribute to the recognition of intruders and help transmit signals to activate a series of downstream events after sensing the presence of the pathogen. The activation of these signalling cascades often enhances immune responses (Durrant and Dong, 2004; Fu et al., 2012; Greenberg and Yao, 2004; Kachroo and Robin, 2013). By contrast, genes that encode proteins that facilitate the compatibility of pathogens and plants are defined as susceptibility (S) genes (Koseoglou et al., 2022). S genes encode negative regulators of plant defence and facilitate pathogen invasion and susceptibility. Some negative regulators encoded by S genes also control the activation of positive regulators of plant defence under normal conditions to avoid the abnormal plant growth and autoimmune responses (Kim et al., 2010; Mackey et al., 2003).

Crop improvement has continuously aimed to achieve higher yields and more substantial tolerance against various pathogens. Traditional breeding methods have been used to select desirable traits within crossbreeding plants. However, conventional methods have some disadvantages, such as a long production time, the need for large populations, labor‐intensive procedures, and a limited gene pool. Therefore, new genome editing techniques have been developed to facilitate plant breeding and increase its efficiency, such as oligonucleotide‐directed mutagenesis (ODM; Sauer et al., 2016), transcription activator‐like effector nucleases (TALENs; Joung and Sander, 2013), zinc finger nucleases (ZFNs; Petolino, 2015), the clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR‐associated (Cas) system (Cong et al., 2013; Iswanto et al., 2021a; Jinek et al., 2012; Shmakov et al., 2017; Zetsche et al., 2015), and modified versions on the CRISPR/Cas system, such as base editor and prime editor, according to the needs of the scientist (Anzalone et al., 2019; Gaudelli et al., 2017; Kang et al., 2018; Lin et al., 2020; Zou et al., 2022). Compared to traditional breeding, these techniques allow researchers to rapidly make more precise genome changes to generate plants with desirable traits (Abdallah et al., 2015).

Remarkable advancements in elucidating plant immune components have been utilized for the genetic engineering of crop plants. Several reviews have discussed some aspects of engineering plant resistance, but a thorough summary of the engineering of plant innate immunity from a different point of view is needed. In this review, we summarize recent advances in plant immune engineering based on important publications over the past two decades. We divide the engineering targets into three types: PTI‐related receptors, ETI‐related receptors, and other downstream components. We also discuss the novel concept of transferring pathogen genes into plants to counter their aggressiveness.

Engineering PTI receptors to confer broad‐spectrum resistance

Overview of pattern recognition receptors

PRRs are membrane‐localized receptor proteins that are divided into two types: receptor‐like kinases (RLKs) and receptor‐like proteins (RLPs; Boutrot and Zipfel, 2017; Couto and Zipfel, 2016; Zipfel, 2014). RLKs possess an ectodomain in the extracellular space, a transmembrane domain, and an endokinase domain in the cytoplasm. The ectodomain functions in ligand perception, while the endokinase domain is required for signal transduction and to activate defence responses in the cell (Boutrot and Zipfel, 2017). Unlike RLKs, RLPs lack an endokinase domain. Therefore, RLPs normally associate with other RLKs to sense the presence of MAMPs or DAMPs. The leucine‐rich repeat (LRR) domain, epidermal growth factor‐like domain, and lysin motif (LysM) have been identified as common ectodomains that guard the apoplast environment following successful infection by intruders and the onset of plant disease. The recognition signals are transduced into the cytoplasm through the endokinase domain, triggering downstream events such as protein oligomerization, auto‐ and transphosphorylation, and the release of negative regulators of immunity, thereby inducing PTI (Boutrot and Zipfel, 2017; de Vries, 2015). PTI responses confer broad‐spectrum resistance in plants. However, in many cases, the plant lacks PRRs to recognize particular pathogens, or the current immune system is insufficient to fend off the invasion. Here, we summarize two strategies of PTI‐related engineering to make more durable and sustainable resistance in the field.

Transformation‐based engineering to transfer PRRs across species

The bacterial peptide elf18, derived from the well‐known MAMP elongation factor thermo unstable (EF‐Tu), is recognized by the RLK EF‐Tu receptor (EFR) in Arabidopsis (Arabidopsis thaliana). EFR binds to EF‐Tu to prevent the transmission of bacterial genetic material and protein synthesis, triggering basal defence responses against the bacterial pathogen (Kunze et al., 2004; Zipfel et al., 2006). Since EFR is only present in the plant family Brassicaceae, the ability of plants from other families to recognize EF‐Tu is limited. In 2010, Lacombe et al. successfully transferred EFR from the cruciferous plant Arabidopsis (AtEFR) into the solanaceous plants Nicotiana benthamiana and tomato (Solanum lycopersicum) to induce broad‐spectrum resistance to different phytopathogens (Lacombe et al., 2010). Moreover, potato (Solanum tuberosum) plants exogenously overexpressing AtEFR showed significant resistance to bacterial wilt caused by the bacterium Ralstonia solanacearum (Boschi et al., 2017). AtEFR was also recently introduced into barrel clover (Medicago truncatula) and orange (Citrus sinensis) to prevent infection by R. solanacearum and Xanthomonas citri subsp. citri (Xcc) and Xylella fastidiosa subsp. pauca (Xfp), respectively (Mitre et al., 2021; Pfeilmeier et al., 2019). Notably, interspecies transfer of AtEFR into apple (Malus domestica) significantly reduced the extent of tissue necrosis associated with Erwinia amylovora infection, which causes fire blight disease (Piazza et al., 2021). The success of this study demonstrates the feasibility of using biotechnological strategies to tackle the devastating fire blight disease in apple trees, which causes substantial losses to cultivated apple production worldwide. Several studies have demonstrated the power of constitutively expressing AtEFR to increase plant immunity in both dicotyledons and monocotyledons such as rice (Oryza sativa) and common wheat (Triticum aestivum) in response to Xanthomonas oryzae pv. oryzae and Pseudomonas syringae pv. oryzae, respectively (Lu et al., 2015; Schoonbeek et al., 2015).

Another well‐studied PRR in plants is FLAGELLIN‐SENSITIVE 2 (FLS2), an RLK that recognizes the most highly conserved domain of the bacterial flagellin flg22 (Chinchilla et al., 2006; Gomez‐Gomez et al., 2001). Unlike EFR, FLS2 is conserved in several plant species. However, flg22 perception by FLS2 orthologs differs among species (Chinchilla et al., 2006). Notably, overexpressing grapevine (Vitis vinifera) FLS2 (VvFLS2) in Arabidopsis conferred resistance against Burkholderia phytofirmans (Trda et al., 2014). Overexpressing frost grape (V. riparia) FLS2 ^XL in N. benthamiana induced resistance to Agrobacterium tumefaciens (Furst et al., 2020), and overexpressing NbFLS2 in citrus reduced susceptibility to X. citri (Hao et al., 2016).

Xa21 is a PRR that confers robust resistance to X. oryzae pv. oryzae (Xoo; Ronald et al., 1992; Song et al., 1995; Wang et al., 1996). A RaxX protein sulfated by the tyrosine sulfotransferase RaxST from Xoo is required to activate Xa21‐mediated immunity (Pruitt et al., 2015). Moreover, chimeric PRRs generated using an extracellular domain of AtEFR and a kinase domain of Xa21 conferred elf18‐induced signalling and quantitative immunity in Arabidopsis and N. benthamiana, revealing the essential role of Xa21 kinase activity in plant resistance (Holton et al., 2015). Indeed, introducing Xa21 into four sweet orange cultivars conferred resistance to Xanthomonas axonopodis pv. citri, the causal agent of leaf‐spotting and fruit rind‐blemishing disease. In addition, the introduction of Xa21 via genetic transformation enhanced resistance to pathogens in other crops, such as tomato (Afroz et al., 2010) and banana (Musa acuminata; Tripathi et al., 2014), including against bacterial wilt pathogens and Xanthomonas campestris pv. musacearum, respectively.

Domain swapping‐based engineering to enhance resistance

Engineering PRRs by domain swapping represents another promising approach for achieving disease resistance (Figure 1). Chitin‐elicitor binding protein (CEBiP), a plasma membrane receptor in rice that perceives chitin from the fungal cell wall, is a glycoprotein with two LysM domains in the extracellular portion and a putative transmembrane domain (Kaku et al., 2006; Shinya et al., 2012). Xa21, a member of the RLK family, harbours a leucine‐rich repeat motif and a serine–threonine kinase‐like domain (Song et al., 1995). Two LysM domains of CEBiP and the kinase‐like domain of Xa21 were combined to generate the new chimeric receptor CRXA (Figure 1). Transgenic rice plants accumulating CRXA showed more robust resistance to Magnaporthe oryzae than plants harbouring wild‐type CEBiP, producing an HR‐like response and strong disease resistance to rice blast (Kishimoto et al., 2010). This enhanced immune response suggests that the LysM domains of CEBiP are required for the early perception of fungal invasion (Kishimoto et al., 2011).

Figure 1.

Biotechnological strategies to improve disease resistance in crops. PRR engineering: PRRs (e.g., EFR and FLS2) can be transformed across species for broadening pathogen recognition. Besides, a new chimeric of PRR can be generated through domain swapping to enhance resistance. Chimera receptor CRXA, made of two LysM domains from CEBiP and kinase domain from Xa21, induces a more robust immune response to Magnaporthe oryzae. The combination of the LRR domain from ReMax and JM‐TM‐CT domain from LeEix can resist Xanthomonas in Nicotiana benthamiana. NLR engineering: NLRs (e.g., MR5, Bs2, and RRS1/RPS4) can be introduced inter‐/intra‐species through transformation to gain resistance against specific pathogens. Besides, random mutagenesis in NLR (e.g., R3a and I2) can be applied to expand effector recognition. Secondary mutation in NLR (e.g., Rx) increases its activation sensitivity to overcome the cost of over‐activated NLR through primary mutagenesis. Decoy engineering: Arabidopsis RPS5 specifically recognizes protease effector AvrPphB by the cleavage of PBS1 and promotes a defensive response. The recognition specificity of RPS5 could be expanded by substituting AvrPphB cleavage sequence in AtPBS1 with other new proteases (e.g., AvrRpt2 from bacteria, TEV NIa protease, and TuMV NIa protease from viruses). Pathogens‐based engineering: Pathogen effectors (e.g., SDT1, Crh1, SM1, CRN115, and SCR1) can be utilized as alternative genetic resources for a transgenic‐based approach to crop disease management. Regulatory component engineering: Downstream regulatory components involved in immune signalling positively or negatively regulate the resistance to pathogen attack. Overexpression of positive regulators (e.g., SnRK1 and Hsp18) increases disease resistance. Alternatively, transcriptional suppression of negative regulators (e.g., DMR6 and SRFR1) through CRISPR/Cas9, VIGS, RNAi, and T‐DNA can boost plant immunity. Created with www.BioRender.com.

Enigmatic MAMP of Xanthomonas (eMax) is recognized by the receptor‐like protein RECEPTOR OF eMax (ReMAX) in Arabidopsis (Jehle et al., 2013b). In addition, the PAMP xylanase provokes a defence response in tomato via the activity of Lycopersicon esculentum ethylene‐inducing xylanase (LeEix). ReMAX and LeEix are RLPs containing large extracellular LRR domains, a juxtamembrane domain (JM), a transmembrane domain (TM), and a cytoplasmic tail (CT). N. benthamiana lacks an endogenous perception system for both eMax and xylanase. However, transgenic N. benthamiana plants accumulating a hybrid protein containing the ReMAX LRR and Eix2 JM‐TM‐CT domains successfully induced defence responses upon eMax recognition (Figure 1; Jehle et al., 2013a,b). One possible strategy for enhancing the basal response layer may therefore take advantage of each domain of different RLKs and RLPs to engineer upgraded versions of PRRs.

Engineering ETI receptors to enhance specific resistance

Overview of NLR receptors

Some NLRs can recognize their corresponding effectors directly via physical interactions. However, other NLRs recognize the presence of effectors via an indirect pathway (Martin et al., 2003). In detail, NLRs are in company with other host components targeted by effectors, thus recognizing effector presence through the modification of the host proteins. The other host components can be a decoy or a guardee (Cesari, 2018). Once the guardee/decoy undergoes a conformational change caused by pathogen effectors, the vigilant NLR is activated, thereby inducing ETI responses. In some cases, NLRs work in pairs in effector recognition, in which one NLR functions as a sensor to trap the effector while the other NLR acts as an executor to induce signalling (Cesari, 2018). Besides the effector recognition function, some “helper” NLRs commonly contribute to the signal transduction downstream after the vigilant NLR activation (Castel et al., 2019; Dong et al., 2016; Wu et al., 2019b).

Altering NLR structure or effector recognition mechanisms is an excellent approach for engineering ETI components to boost resistance to a specific pathogen effector. The NLR itself represents a popular target for engineering. The guardee/decoy in the indirect recognition system may also be targeted (Kim et al., 2016; Kourelis et al., 2016). A guardee is unlikely to tolerate a change due to its specific functions in plant signalling in addition to effector recognition (Cesari, 2018), whereas a decoy is a preferred engineering target by taking advantage of a mimic host target protein without any role in the plant. In the following sections, we summarize strategies and provide examples of increased plant resistance engineering that targets ETI components.

Transformation‐based engineering to transfer NLRs within species

Many studies using crops or model plants have demonstrated the successful transformation of resistance genes to increase resistance against pathogens. For instance, in 2014, Broggini et al. proposed a transformation strategy to increase the resistance of apple to fire blight disease (Broggini et al., 2014). This devastating disease can destroy an entire apple/pear (Pyrus communis) orchard if just one plant is infected. E. amylovora, the causal agent of fire blight disease, contains the effector AvrPpt2^EA, which is recognized by the R protein FB_MR5 from apple accession Mr5. To generate resistance to E. amylovora in apple cultivar Gala, FB_MR5 was transformed from apple accession Malus × robusta 5 (Mr5; Figure 1). The transgenic lines expressing Mr5 showed significantly less severe disease symptoms compared to susceptible wild‐type ‘Gala’. Therefore, it is possible to isolate a single resistance‐determinant gene from the native genome of apple and transfer it to a susceptible cultivar to protect it from fire blight. Although FB_MR5 is an endogenous apple gene, exogenous T‐DNA from the transformation construct is still present in transgenic apple plants, prompting researchers to design ways to remove it. In 2015, Kost et al. generated the first cisgenic ‘Gala’ apple using the same method involving A. tumefaciens‐mediated transformation of FB_MR5 from wild apple accession Mr5 (Kost et al., 2015). Interestingly, the cisgenic apple line, displaying markedly fewer fire blight disease symptoms, carried only the cisgene FB_MR5 controlled by a copy of the MR5 promoter and no transgene due to the removal of the T‐DNA by heat‐induced flippase.

Transformation‐based engineering can also be used for the intraspecies transfer of NLRs when the effector from the pathogen and the NLR from the resistant cultivar are known. However, in some cases, even when the cognate effector is unknown, NLR transfer can still be exploited to generate resistant cultivars. Indeed, soybean (Glycine max) lines overexpressing GmKR3, a soybean gene encoding a TIR‐NB‐LRR (TNL) similar to tobacco mosaic virus (TMV) resistance protein N, showed decreased susceptibility to several soybean mosaic virus strains, bean common mosaic virus, watermelon mosaic virus, and bean pod mottle virus (Xun et al., 2019; Table 1). Remarkably, the enhanced resistance in GmKR3 overexpression lines did not sacrifice plant growth, including seed production and quality. In addition, overexpressing the RPM1 (RESISTANCE TO P. SYRINGAE PV MACULICOLA 1)‐like resistance gene OsRLR1 enhanced the resistance of rice to the fungus Pyricularia oryzae and the bacterium Xoo (Table 1; Du et al., 2021).

Table 1.

Representative of positive immune regulators that were identified in crop plants

No.	Gene	Pathogen/pest	Hormone signalling	Reference
Monocotyledon
Brachypodium distachyon *
1	AtSAG101	Puccinia brachypodii	SA	Wang et al. (2020c)
		M. oryzae
2	BdWRKY38	R. solani	SA	Kouzai et al. (2020)
Haynaldia villosa *
3	Stpk‐V	Bgt, Bgh	SA, JA	Qian et al. (2017)
4	HvSERK2	Bgh	nd	Li et al. (2018)
5	RLK‐V1.1	Bgt	nd	Hu et al. (2018)
6	LecRK‐V	Bgt	SA	Wang et al. (2018c)
Hordeum vulgare *
7	SnRK1	Bgh	nd	Han et al. (2020)
8	Sr22, Sr33, Sr35 and Sr45	Pgt	nd	Hatta et al. (2021)
Oryza sativa *
9	SDS2	M. oryzae	nd	Fan et al. (2018)
10	OsWRKY67	M. oryzae, Xoo	nd	Vo et al. (2017)
11	OsMYC2	Xoo	JA	Uji et al. (2016)
12	OsCIPK30	RSV	nd	Liu et al. (2017c)
13	OsHsp18.0	Xoo, Xoc	SA, JA	Ju et al. (2017), Kuang et al. (2017)
14	OsSnRK1a	P. oryzae, C. miyabeanus, R. solani	JA and SA	Filipe et al. (2018)
		Xoo
15	WRKY45	Magnaporthe grisea	SA	Shimono et al. (2012)
		Xoo
16	OsRSR1	R. solani	nd	Wang et al. (2021a)
17	OsRLCK5	R. solani	nd	Wang et al. (2021a)
18	SWEET14	R. solani	nd	Kim et al. (2021)
19	DOF11	R. solani	nd	Kim et al. (2021)
20	OsbHLH034	Xoo	JA	Onohata and Gomi (2020)
21	Osa‐miR162a	M. oryzae	nd	Li et al. (2020b)
22	Os6PGDH1	Nilaparvata lugens (BPH)	JA and ET	Chen et al. (2020b)
23	OsPDR1	Xoo	JA	Zhang et al. (2020a)
24	MIM156‐3p	M. oryzae	nd	Zhang et al. (2020c)
25	OsMIM167d	M. oryzae	IAA and JA	Zhao et al. (2020)
26	FBL55	R. solani	Auxin	Qiao et al. (2020)
27	ALEX1	Xoo, Xoc	JA	Yu et al. (2020)
28	OsHsfB4d	Xoo, Xoc	nd	Yang et al. (2020)
29	OsCM	Xoo	SA and JA	Jan et al. (2020)
30	OsWRKY114	Xoo	nd	Son et al. (2020)
31	IPA1	Xoo	Gibberellin (GA)	Liu et al. (2019)
32	LPA1	R. solani	nd	Sun et al. (2019)
33	OsFWL5	Xoo	JA	Li et al. (2019a)
34	OsGRF8	BPH	Flavonoid componds	Dai et al. (2019)
35	OsMKK3	BPH	JA, JA‐Ile and ABA	Zhou et al. (2019b)
36	OsPGIP1	Xoc	JA	Wu et al. (2019a)
37	HIR3	RSV	SA	Li et al. (2019c)
		Pto DC3000
		Xoo
38	AtRPS2 and AtRPM1(D505V)	M. oryzae	nd	Li et al. (2019e)
		Xoo
		BPH
39	OsERF83	M. oryzae	SA, MeJA and Ethephon (ETH)	Tezuka et al. (2019)
40	OsRRK1	BPH	nd	Ma et al. (2017)
41	BAG2	Xoo	nd	You et al. (2016)
42	OsRLR1	P. oryzae	nd	Du et al. (2021)
		Xoo
Saccharum spp. hybrid *
43	ScPR10	Sporisorium scitamineum, Sorghum mosaic virus	SA and MeJA	Peng et al. (2017)
Triticum aestivum *
44	TaRCR1	Rhizoctonia cerealis	nd	Zhu et al. (2017)
45	ERF1‐V	Bgt	ET and ABA	Xing et al. (2017)
46	TaPIMP2	Bipolaris sorokiniana	nd	Wei et al. (2017)
47	TaSnRK1α	F. graminearum	nd	Jiang et al. (2020)
48	WFhb1‐1	F. graminearum	nd	Paudel et al. (2020)
49	TaUGT6	F. graminearum	nd	He et al. (2020)
50	TaJAZ1	B. graminis	JA	Jing et al. (2019)
51	AtLTP4.4	F. graminearum	nd	McLaughlin et al. (2021)
Triticum urartu *
52	TuACO3	Bgt	ET	Zheng et al. (2020)
Phalaenopsis aphrodite sub sp. Formosana *
53	PaAGO5s	Cymbidium mosaic virus, Odontoglossum ringspot virus	nd	Kuo et al. (2021)
Zea mays *
54	ZmERF105	Exserohilum turcicum	nd	Zang et al. (2020)
55	LcCHI2	E. turcicum	nd	Liu et al. (2020b)
		Curvularia lunata
56	ZmDEF1	Sitophilus zeamais	nd	Vi et al. (2019)
Dicotyledons
Brassica napus *
57	AtGDSL1	Sclerotinia sclerotiorum	SA	Ding et al. (2020)
Capsicum annuum *
58	Can‐miRn37a	Colletotrichum truncatum	nd	Mishra et al. (2018)
59	CaHDZ27	R. solanacearum	SA, JA, Ethephon (ETH)	Mou et al. (2017)
60	CaLRR51	R. solanacearum	SA, JA, Ethephon	Cheng et al. (2017)
Chrysanthemum morifolium *
61	CmMYB19	Macrosiphoniella sanborni	nd	Wang et al. (2017e)
62	CmMYB15	M. sanborni	nd	An et al. (2019)
Citrus sinensis *
63	EFR	Xcc, Xfp	nd	Mitre et al. (2021)
64	CsPrx25	Xcc	nd	Li et al. (2020a)
65	CsGH3.1 and CsGH3.1L	Xcc	SA, ET	Zou et al. (2019)
Cucumis sativus *
66	CsWRKY10	Corynespora cassiicola	nd	Liu et al. (2020a)
67	CsERF004	Pseudoperonospora cubensis, C. cassiicola	SA, ET	Liu et al. (2017a)
68	CsRSF1 and CsRSF2	Sphaerotheca fuliginea	ABA, GA	Wang et al. (2021c)
Glycine max *
69	GmSN1	SMV	nd	He et al. (2017)
70	Rsv3	SMV	nd	Tran et al. (2018)
71	GmKR3	SMV, Bean common mosaic virus (BCMV), Watermelon mosaic virus, Bean pod mottle virus (BPMV)	ABA	Xun et al. (2019)
72	hrpZm	P. syringae pv. tabaci, P. sojae	nd	Du et al. (2018)
73	GmPAL2.1	P. sojae	SA and Genistein	Zhang et al. (2017a)
74	GmERF113	P. sojae	nd	Zhao et al. (2017)
75	OXO	S. sclerotiorum	nd	Yang et al. (2019)
76	GmSnRK1.1	P. sojae	SA	Wang et al. (2019b)
77	GmDR1	Fusarium virguliforme	SA and JA	Ngaki et al. (2021)
		Tetranychus urticae
		A. glycines, Matsumura
		H. glycines
78	GmPI4L	P. sojae	Daidzein and genistein	Chen et al. (2019)
79	AtQQS	H. glycines	nd	Qi et al. (2019)
		BPMV
		PsgR4
		Myzus persicae
		F. virguliforme
80	AtNF‐YC4, GmNF‐YC4‐1	H. glycines	nd	Qi et al. (2019)
		TuMV
		Pto DC3000
		M. persicae
		F. virguliforme
Gossypium hirsutum *
81	GhERF‐IIb3	Xanthomonas citri pv. malvacearum (Xcm)	JA	Cacas et al. (2017)
82	miR398, miR2950	Cotton leaf curl Multan virus (CLCuMuV)	nd	Akmal et al. (2017)
83	GhLAC15	Verticillium dahliae	nd	Zhang et al. (2019)
84	ghr‐miR164	V. dahliae	nd	Hu et al. (2020)
85	GhMORG1	F. oxysporum	nd	Wang et al. (2020a)
86	ZmASN	Bemisia tabaci	nd	Gul et al. (2020)
87	GhPAP1D	Helicoverpa armigera	Anthocyanin	Li et al. (2019d)
		Tetranychus cinnabarinus
Ipomoea batatas *
88	IbBBX24	Fusarium oxysporum f. sp batatas	JA	Zhang et al. (2020b)
Malus domestica *
89	MdERF11	Botryosphaeria dothidea	SA	Wang et al. (2020b)
90	MdMYB88, MdMYB124	Alternaria alternata	nd	Geng et al. (2020)
		Valsa mali
91	MdMYB73	B. dothidea	SA	Gu et al. (2021)
92	MdUGT88F1	V. mali	SA	Zhou et al. (2019a)
93	MdATG18a	Diplocarpon mali	SA	Sun et al. (2018)
Manihot esculenta *
94	MeNR1, MeNR2	Xanthomonas axonopodis pv. manihotis	nd	Yan et al. (2021)
Populus tremula *
95	PtDefensin	Septotis populiperda	JA	Wei et al. (2020)
Populus alba var. pyramidalis *
96	PalbHLH1, PalMYB90	B. cinerea, Dothiorella gregaria	Flavonoid compounds	Bai et al. (2019)
Solanum lycopersicum *
97	SlMAPK3	TYLCV	SA, JA	Li et al. (2017)
98	Csl	TYLCV	nd	Choe et al. (2021)
99	SlMYC1	T. urticae	JA	Hua et al. (2021)
100	SlMYB75	B. cinerea	JA	Liu et al. (2021b)
101	lncRNA39026	P. infestans	nd	Hou et al. (2020)
102	Pti4/5/6	Pto DC3000	nd	Wang et al. (2021d)
103	SlWRKY8	Pto DC3000	SA, ABA	Gao et al. (2020)
Solanum tuberosum *
104	StMPK7	P. infestans, Phytophthora parasitica	SA	Zhang et al. (2021)
105	StRac1	P. infestans	nd	Zhang et al. (2020d)
Vitis pseudoreticulata *
106	VpRH2	Uncinula necator	nd	Wang et al. (2017c)
107	VpEIFP1	Golovinomyces cichoracearum	nd	Wang et al. (2017b)
108	VpSTS29/STS2	U. necator	SA	Xu et al. (2019)

*Indicated for the applied plant.

Transformation‐based engineering to transfer NLRs from one species to another

Following the pioneering study of 2010 in which the PRR protein EFR from Arabidopsis was expressed in N. benthamiana and tomato (Lacombe et al., 2010), an increasing number of studies have reported the transfer of NLRs across species. Some studies have focused on the transfer of NLRs to increase resistance against bacteria. For example, the bacterial spot 2 (Bs2) resistance gene from pepper (Capsicum annuum) was transferred into tomato, which is phylogenetically close to pepper, to control bacterial spot disease (Horvath et al., 2012). Pepper Bs2 is an R protein that senses the effector AvrBs2, which is conserved in bacterial spot‐inducing Xanthomonas strains. Tomato plants harbouring the Bs2 transgene exhibited a dramatic decrease in disease severity (Figure 1). Moreover, the fruit yield from tomato plants expressing Bs2 was more than double that of non‐transgenic tomatoes. A later study demonstrated that pepper Bs2‐related immunity could also be engineered in a plant family other than Solanaceae (Sendin et al., 2017). Indeed, the Bs2 gene was used to develop transgenic sweet oranges (from the Rutaceae family) with increased resistance to citrus canker disease, which is caused by Xcc harbouring the conserved avrBs2.

Other studies have focused on genetic transformation‐based engineering for resistance to fungal pathogens. For instance, Arabidopsis plants harbouring the Mildew A (MLA) NLR gene from barley (Hordeum vulgare) were fully resistant to the barley powdery mildew fungus Blumeria graminis f. sp. hordei (Bgh) containing the effector AVR_A1 (Maekawa et al., 2012), indicating that AVR_A1‐mediated resistance via MLA in monocot barley was recapitulated in the transgenic dicot Arabidopsis. Whereas wheat contains many Stem rust (Sr) resistance genes against the fungus Puccinia graminis f. sp. tritici (Pgt), barley contains a limited number of genetically determined resistance genes. The transformation of the wheat NLR genes Sr22, Sr33, Sr35, and Sr45 successfully increased resistance to stem rust in barley (Table 1; Hatta et al., 2021). The transgenic barley plants showed no agronomically negative effects in the absence of disease. Another example is the transfer of the NLR gene Ve1 from tomato to tobacco (Nicotiana tabacum) and cotton (Gossypium hirsutum; Song et al., 2018b). In detail, Song et al. successfully increased the resistance of tobacco and cotton against Verticillium spp. strains via the effector Ave1 recognition by tomato Ve1. In Arabidopsis, the NLR proteins RESISTANT TO P. SYRINGAE 4 (RPS4) and RESISTANT TO RALSTONIA SOLANACEARUM 1 (RRS1) recognize three distinct effectors: PopP2, AvrRps4, and an unknown Colletotrichum effector (Birker et al., 2009; Narusaka et al., 2009). Transferring the RPS4/RRS1 pair from Arabidopsis conferred resistance to the fungal pathogen Colletotrichum higginsianum in Brassicaceae and protected cucumber (Cucumis sativus, a member of the Cucurbitaceae) against C. orbiculare (Figure 1; Narusaka et al., 2013). Transgenic tomato plants expressing RPS4/RRS1 also specifically recognized the PopP2 and AvrRps4 effectors (Narusaka et al., 2013). These findings expand our knowledge of the roles of NLR‐type genes in recognizing and conferring resistance to distinct pathogens. The specific recognizing mechanism is highly conserved, as vigilant NLR protein pairs were formed between different species. Therefore, transferring NLR genes to other species represents a powerful strategy for genetic engineering for enhanced plant immunity.

Random mutagenesis of NLRs enhances plant responses to effectors

Each NLR can recognize only a few effectors. Therefore, the limitation of ETI is that the plant must develop a new NLR to recognize a newly evolved effector secreted from the pathogen. To address this issue, scientists have performed random mutagenesis to broaden the effector recognition capacity of an NLR. For example, the potato R protein R3a was subjected to random mutagenesis (Segretin et al., 2014) to alter its response to the pathogenic effector AVR3a from the oomycete Phytophthora infestans. In detail, wild‐type R3a responds effectively to AVR3a^KI but weakly to its variant AVR3a^EM (Armstrong et al., 2005; Bos et al., 2009), which causes virulence in R3a‐containing potato plants. Through a random mutagenesis screening of R3a, Segretin et al. obtained eight mutant R3a proteins with single amino acid mutations that showed sufficient recognition of virulent AVR3a^EM and even other isoforms. These eight mutations were located across the R3a protein, but mainly in LRR domains (Figure 1). Interestingly, R3a^N336Y, harbouring a mutation in a nearby pocket of the NB, conferred resistance to the effector PcAVR3a4 from the vegetable pathogen P. capsici.

Furthermore, based on the R3a case study in potato, the authors applied the previously identified mutations in R3a to its ortholog in tomato, NLR I2, which confers resistance to the fungal pathogen Fusarium oxysporum f. sp. lycopersici (Figure 1; Giannakopoulou et al., 2015). Whereas wild‐type I2 conferred a weak response to AVR3a, the I2^I141N mutant significantly enhanced the response to AVR3a. Moreover, I2^I141N partially mediated resistance against P. infestans and showed a broad‐spectrum response to effector Avr2 variants from F. oxysporum f. sp. lycopersici. This finding suggests that it is possible to engineer NLR via mutation to confer resistance to distantly related pathogens. The knowledge obtained from studying an NLR in one plant may thus be applied to improve the resistance conferred by orthologous NLRs in other plants.

In some cases, broadening the effector recognition of an NLR through mutagenesis comes at a cost, i.e., yield‐compromising and trailing necrosis caused by autoimmunity. For instance, in the case of potato Rx protein, which mediates resistance to potato virus X (PVX), mutations in the LRR domain of Rx confer resistance to both PVX and the phylogenetically unrelated poplar mosaic virus (PopMV). In response to PopMV, one Rx mutant showed a trailing necrosis phenotype that annihilated the plant, while plants carrying wild‐type Rx displayed mild disease symptoms. However, the researchers overcame this cost by performing secondary mutagenesis (Harris et al., 2013). Specifically, the authors conducted random mutagenesis in the N terminus of the Rx mutant with impaired broad‐host recognition. Four mutations located close to the NB pocket were identified, conferring more robust resistance to PopMV. Moreover, stable transgenic plants expressing one of these secondary Rx mutants showed resistance to PVX and PopMV, which previously induced trailing necrosis in transgenic plants harbouring the primary Rx mutant (Figure 1). These findings demonstrate that the recognition of an NLR can be improved by stepwise mutagenesis. The stepwise mutagenesis has been recently applied for Sw‐5b, an NLR recognizing the movement protein NSm from Tomato spotted wilt virus (TSWV), to confer resistance against several TSWV variants and American‐type tospoviruses (Huang et al., 2021). However, since the gain‐of‐function mutations arose from random mutagenesis, no rational design strategy has thus far been proposed to engineer NLRs with new recognition specificities, which represents an obstacle to NLR engineering.

Utilizing a decoy system to expand the effector recognition specificity of an NLR

The classical gene‐for‐gene hypothesis posits that a single R protein recognizes only one avirulence effector. The concept that NLRs could be engineered to recognize several unrelated effectors had been challenging until the decoy engineering of Arabidopsis AVRPPHB SUSCEPTIBLE 1 (PBS1) was proposed (Kim et al., 2016). In nature, the Arabidopsis NLR RPS5 perceives the cleavage products of the decoy host protein PBS1 following its cleavage by the protease type III effector AvrPphB to trigger a defence response (DeYoung et al., 2012; Qi et al., 2014). Targeting PBS1, Kim and colleagues replaced the cleavage sequence targeted by AvrPphB with those of AvrRpt2, tobacco etch virus (TEV) NIa protease, or turnip mosaic virus (TuMV) NIa protease. In response to each tested protease, the corresponding engineered PBS1 successfully activated RPS5‐mediated ETI (Kim et al., 2016; Figure 1). Notably, unlike engineering to target an R protein, decoy engineering provides a safe and predictable method for protein modification. Indeed, except for effector recognition, decoy PBS1 has no other biological function in the plant; therefore, PBS1 decoy engineering will not have any additional side effects on plants.

The decoy PBS1‐based engineering strategy is suitable for plant species containing Arabidopsis PBS1 orthologs and an endogenous R protein that recognizes AvrPphB. Alternatively, the Arabidopsis RPS5‐engineered PBS1 cassette can be transferred into crop plants. In addition, the target pathogens must be able to take advantage of the host protease function for their virulence activity to use this strategy. PBS1 orthologs have been characterized in several crops such as wheat, barley, soybean, and potato (Bai et al., 2022; Carter et al., 2019; Helm et al., 2019; Sun et al., 2017). To date, the PBS1 decoy approach was successfully used to limit soybean mosaic virus (SMV) and potato virus Y (PVY) in soybean and potato, respectively (Bai et al., 2022; Helm et al., 2019). The successful engineering of soybean PBS1 (GmPBS1) and potato PBS1 (StPBS1) did not require the identification of the corresponding R protein. Therefore, PBS1‐based decoy engineering is expected to be widely applied to crops without the need to identify the responsible R protein for protease effector‐mediated resistance.

Unlike a decoy, a guardee is not considered to be a good target for engineering due to its other functions besides effector recognition. For example, a T‐DNA knockout mutant of the guardee gene RPM1 INTERACTING PROTEIN 4 (RIN4) caused seedling lethality due to the failure to suppress RPS2 autoactivation (Mackey et al., 2003). Therefore, engineering the guardee RIN4 could alter its function, posing a challenge for researchers. However, a recent study of natural RIN4 variants created an opportunity to overcome that challenge (Kim et al., 2022). In detail, natural RIN4 variants require RIN4‐specific motifs, asparagine/tyrosine (NY) or aspartic acid/phenylalanine (DF), to regulate distinct NLRs. In general, while NY‐type RIN4s suppress the autoactivation of Arabidopsis RPM1 and RPS2, DF‐type RIN4s activate apple FB_MR5‐mediated resistance. More importantly, the conserved H167 residue in various RIN4s plays a key role in regulating multiple NLRs. The idea of expanding NLR recognition specificity by combining two motifs proved to be fruitful, since three of six engineered chimeric RIN4 proteins designed using two full‐length NY‐type RIN4s and three C‐terminal DF‐type RIN4s suppressed RPM1/RPS2 autoactivation and activated the R protein FB_MR5. When co‐expressed with RPM1 and RPS2, the constructs encoding the tested RIN4 chimeric proteins triggered AvrRpm1‐ and AvrRpt2‐mediated resistance, respectively. Since RIN4 is conserved among species, but with several polymorphisms (Kim et al., 2022), it might be possible to engineer chimeric RIN4 to broaden the specificity of NLR recognition in plants of interest.

Engineering crop resistance by targeting the regulators of plant innate immunity

Enhancing the expression of positive regulators of plant innate immunity for broad‐spectrum resistance

Genetic transformation is the most commonly employed strategy compared to several other biotechnological approaches, as it permits researchers to insert specific gene sequences into a host plant and to enhance the expression of positive regulatory components in plant defence. During defence responses, phytohormone signalling, particularly jasmonic acid (JA), ethylene (ET), salicylic acid (SA), abscisic acid (ABA), auxin, cytokinin (CK), gibberellin (GA), and brassinosteroid (BR) signalling, is somehow induced, indicating that these plant hormones play significant roles in plant defence responses (Bari and Jones, 2009; Berens et al., 2017).

JA and ET primarily control resistance against necrotrophic pathogens. By contrast, SA is a primary regulator of defence against biotrophic and hemibiotrophic pathogens and is also required to establish SAR (Durrant and Dong, 2004). Hence, many researchers have utilized various regulatory components from these phytohormone signalling networks to reduce host susceptibility to specific pathogens. Table 1 lists positive immune regulators that were engineered in crops over the past 5 years and conferred disease resistance against several pathogens. For instance, overexpressing SNF1‐related protein kinase 1 (SnRK1), encoding a key regulator of cellular metabolism, positively affected SA and JA signalling pathways and enhanced resistance against necrotrophic and biotrophic fungal pathogens such as Fusarium graminearum, Bgh, M. oryzae, Cochliobolus miyabeanus, Rhizoctonia solani, and Plasmodiophora brassicae in various crops (Table 1).

The activation of incompatible plant‐pathogen interactions mediated by SA and JA signalling is also applicable for many bacteria. SnRK1 has been utilized to generate broad‐spectrum disease resistance. SnRK1 was also reported to decrease susceptibility to bacterial leaf blight Xoo in various rice cultivars. In addition to SnRK1, many regulatory components in the SA signalling pathway are involved in the plant defence response against bacteria (Filipe et al., 2018). Overexpressing Heat shock protein 18.0 (Hsp18.0) activated SA signalling and positively regulated resistance to Xoo and X. oryzae pv. oryzicola (Xoc), the causal agent of bacterial leaf streak in rice. Conversely, suppressing this gene increased susceptibility to Xoo and Xoc and decreased free SA levels, suggesting that OsHsp18.0‐mediated resistance functions via an SA‐dependent signalling pathway (Ju et al., 2017; Kuang et al., 2017). The decreased susceptibility of the transgenic rice lines to Xoo and Xoc was not solely due to increased SA levels but also to increased JA and GA levels (Table 1; Onohata and Gomi, 2020; Zhang et al., 2020a). Consistent with these findings, the enhanced transcriptional regulatory activity of SnRK1 that confers resistance to Xoo involves three phytohormone signalling pathways. Thus, defence responses against the leaf blight pathogens seem to be coordinately regulated by SA, JA and GA signalling pathways.

Engineering of S genes to enhance crop resistance

The engineering of durable, broad‐spectrum resistance involves the loss of function of S genes. Pathogens target and exploit S genes to establish a compatible interaction with the host. Therefore, the mutation or loss of function of S genes can limit the ability of the pathogen to cause infection and disease (Moniruzzaman et al., 2020). Recent advances in crop protection have led to the identification and targeting of an assortment of S genes in vegetables and other crops that play critical roles in plant susceptibility to a wide range of pathogens and pests. Table 2 shows a summary of representative S genes identified in crops over the past decade. One significant constraint to yield production in rice is blast disease caused by the hemibiotrophic fungus M. oryzae. Over the past decade, powerful biotechnology approaches aimed at exploring the rice blast disease circuitry have led to the discovery of various S genes involved in M. oryzae infection. Interestingly, some S genes of M. oryzae were also designated as S genes of the biotrophic bacterium Xoo (Table 2). Even though numerous rice genes have been identified and defined as S genes to Xoo, their roles in M. oryzae infection have not yet been tested (Table 2). Thus, future studies on the pathogenicity of Xoo and M. oryzae could help elucidate the connection between M. oryzae and Xoo S genes. Besides interacting with bacteria and fungi, S genes play roles in plant‐virus interactions. For example, knockout of ARGONAUTE 2 (OsAGO2) and SQUAMOSA PROMOTER BINDING PROTEIN‐LIKE 9 (OsSPL9) conferred resistance against rice black‐streaked dwarf virus (RBSDV) and rice stripe virus (RSV), respectively (Wang et al., 2021e; Yao et al., 2019). Soybean MITOGEN‐ACTIVATED PROTEIN KINASE 6 (MPK6), tomato eIF4E2, and Pelota genes also function as S genes that interact with plant viruses. Suppression of MPK6 expression decreased susceptibility to SMV (Liu et al., 2014), whereas knockout mutations of eIF4E2 and Pelo conferred resistance against pepper veinal mottle virus (PVMV) and tomato yellow leaf curl virus (TYLCV), respectively (Moury et al., 2020; Pramanik et al., 2021).

Table 2.

Representative of S genes identified in crop plants (2011–2021)

No.	Gene	Method	Pathogen/pest	Reference
Monocotyledon
Hordeum vulgare *
1	HvCRK1	RNAi	Bgh	Rayapuram et al. (2012)
2	HvMORC1, HvMORC6a	CRISPR/Cas9	Bgh	Galli et al. (2021)
			F. graminearum
Musa spp.*
3	MusaDMR6	CRISPR/Cas9	Xcm	Tripathi et al. (2021)
Oryza sativa *
4	OsAGO2	CRISPR/Cas9, transposon insertion	RBSDV	Wang et al. (2021e)
5	OsMADS26	RNAi	M. oryzae	Khong et al. (2015)
			Xoo
6	OsDCL1	RNAi	M. oryzae	Zhang et al. (2015a)
7	GF14e	RNAi	Xoo	Manosalva et al. (2011)
8	OsGLIP1, OsGLIP2	RNAi	Xoo	Gao et al. (2017)
			M. oryzae
9	OsNramp6	T‐DNA	M. oryzae	Peris‐Peris et al. (2017)
10	SPL33	EMS	M. oryzae	Wang et al. (2017d)
			Xoo
11	OsDjA6	RNAi	M. oryzae	Zhong et al. (2018)
12	OsCUL3a	EMS	M. oryzae	Liu et al. (2017b)
			Xoo
13	OsCPK4	T‐DNA	M. oryzae	Wang et al. (2018b)
			Xoo
14	OsWRKY28	T‐DNA	M. oryzae	Delteil et al. (2012)
15	OsEDR1	RNAi, T‐DNA	Xoo	Shen et al. (2011)
16	OsERF922	RNAi	M. oryzae	Liu et al. (2012)
17	OsNPR1	Antisense expression	C. suppressalis	Li et al. (2013)
18	CPK18	RNAi	M. oryzae	Xie et al. (2014)
19	LMR	RNAi and EMS	M. oryzae	Fekih et al. (2015)
			Xoo
20	OsHDT701	RNAi	M. oryzae	Ding et al. (2012)
			Xoo
21	OsMESL	T‐DNA, CRISPR/Cas9, RNAi	Xoo	Hu et al. (2021)
			R. solani
22	OsTrxm	CRISPR/Cas9	Xoo	Hu et al. (2021)
			R. solani
23	OsSULTR3;6	CRISPR/Cas9	Xoc	Xu et al. (2021)
24	DEP1	RNAi, T‐DNA insertion	R. solani	Miao Liu et al. (2021)
25	Osa‐miR1873	miR1873	M. oryzae	Zhou et al. (2020)
26	OsSPL9	CRISPR/Cas9	RSV	Yao et al. (2019)
27	OsMPK15	CRISPR/Cas9	M. oryzae	Hong et al. (2019)
			Xoo
28	Os8N3	CRISPR/Cas9	Xoo	Kim et al. (2019)
29	OsHCAR	CRISPR/Cas9	Xoo	Kampire et al. (2021)
30	OsPG1	CRISPR/Cas9	Xoo	Cao et al. (2021)
31	OsVOZ1/OsVOZ2	CRISPR/Cas9, RNAi	M. oryzae	Wang et al. (2021b)
Triticum aestivum *
32	TaNAC1	VIGS	Pst	Wang et al. (2015)
33	Ta‐A/N‐Inv1	VIGS	Pst	Liu et al. (2015)
34	TaBON1 and TaBON3	VIGS	Bgt	Zou et al. (2018)
35	TaWRKY49	VIGS	Pst	Wang et al. (2017a)
36	TaEDR1	RNAi, VIGS, CRISPR/Cas9	Bgt	Zhang et al. (2017b)
37	TaLSD1	VIGS	Pst	Guo et al. (2013)
38	TaEIL1	VIGS	Pst	Duan et al. (2013)
39	TaNAC21/22	VIGS	Pst	Feng et al. (2014)
40	TaDIR‐B1	VIGS, EMS	Fusarium pseudograminearum	Yang et al. (2021)
41	TaSTP13	VIGS	Pst	Huai et al. (2020)
42	TaHRC	RNAi, CRISPR/Cas9	F. graminearum	Su et al. (2019)
43	TaCSN5	RNAi	Pst	Bai et al. (2021)
44	TaADF3	VIGS	Pst	Tang et al. (2015)
45	DRM2	VIGS	Bgt	Geng et al. (2019)
Zea mays *
46	ZmLOX3	Mu‐transposable element‐insertional mutagenesis	Colletotrichum graminicola	Constantino et al. (2013)
47	ZmFBL41	Transposon insertion	R. solani	Li et al. (2019b)
Dicotyledons
Brassica napus *
48	BnCRT1a	EMS, CRISPR/Cas9	Verticillium longisporum	Probsting et al. (2020)
Capsicum annuum *
49	CaMLO2	VIGS	Xanthomonas campestris pv. vesicatoria (Xcv)	Kim et al. (2014)
50	CaWRKY58	VIGS	R. solanacearum	Wang et al. (2013)
51	CaGRP1	VIGS	Xcv	Kim et al. (2015)
Citrus sinensis *
52	CsWRKY22	RNAi	Xcc	Long et al. (2021)
53	CsDMR6	CRISPR/Cas9	Xcc	Parajuli et al. (2022)
54	CsTCTP1 or CsTCTP2	VIGS	S. fuliginea	Meng et al. (2018)
55	CsMLO1 or CsMLO2	VIGS	C. cassiicola	Yu et al. (2019a,b)
56	CYP6CY14, CYP6CY22, CYP6UN1	RNAi	Aphis gossypii	Chen et al. (2020a)
Fragaria × ananassa *
57	FaWRKY25	RNAi	B. cinerea	Jia et al. (2020)
Glycine max *
58	Gm‐NDR1‐1	RNAi	Meloidogyne incognita	McNeece et al. (2017)
59	GmMPK6	VIGS	P. manshurica	Liu et al. (2014)
			SMV
Gossypium barbadense *
60	GbWRKY1	VIGS	V. dahliae	Li et al. (2014a)
			B. cinerea
61	GbMPK3	RNAi	V. dahliae	Long et al. (2020)
Gossypium hirsutum *
62	HDTF1	VIGS	V. dahliae	Gao et al. (2016)
			B. cinerea
63	GhSSI2	VIGS	V. dahliae	Mo et al. (2021)
			F. oxysporum
64	GhADF6	VIGS	V. dahliae	Sun et al. (2021)
65	GhNAC100	VIGS	V. dahliae	Hu et al. (2020)
66	GhBsr‐k1	VIGS	V. dahliae	Li et al. (2021)
			F. oxysporum
Malus domestica *
67	MdMLO19	RNAi	Podosphaera leucotricha	Pessina et al. (2016)
Solanum lycopersicum *
68	SlWRKY70	VIGS	Macrosiphum euphorbiae	Atamian et al. (2012)
			Meloidogyne javanica
69	SlSR1, SlSR3L	VIGS	B. cinerea	Li et al. (2014b)
			Pto DC3000
70	eIF4E2	EMS	PVMV	Moury et al. (2020)
71	SlSRFR1	CRISPR/Cas9	Pto DC3000	Son et al. (2021)
72	SlPelo	CRISPR/Cas9	TYLCV	Pramanik et al. (2021)
73	SlMlo1	CRISPR/Cas9	Oidium neolycopersici	Pramanik et al. (2021)
74	SlDMR1	RNAi	O. neolycopersici	Huibers et al. (2013)
75	SlDMR6	CRISPR/Cas9	Pto DC3000	Thomazella et al. (2021)
			X. gardneri
			X. perforans
			P. capsici
			P. neolycopersici
Solanum lycopersicum, Solanum tuberosum *
76	StDND1	RNAi	P. infestans	Sun et al. (2016a)
			O. neolycopersici
			Golovinomyces orontii
Solanum tuberosum *
77	StDND1	CRISPR/Cas9	P. infestans	Kieu et al. (2021)
78	StCHL1	CRISPR/Cas9	P. infestans	Kieu et al. (2021)
79	StDMR6‐1	RNAi	P. infestans	Sun et al. (2016b)
		CRISPR/Cas9	P. infestans	Kieu et al. (2021)
80	StERF3	RNAi	P. infestans	Tian et al. (2015)

*Indicated for the applied plant.

The biotrophic fungus B. graminis causes powdery mildew, one of the most destructive foliar diseases of cereals, thereby significantly reducing crop productivity. Three S genes from barley, cysteine‐rich RLK 1 (HvCRK1), Microrchidia 1 (HvMORC1), and HvMORC6a, were recently shown to contribute to susceptibility to Bgh, as transcriptional suppression of HvCRK1 and knockout of HvMORC1/MvMORC6a enhanced resistance to Bgh (Table 2; Galli et al., 2021; Rayapuram et al., 2012). S genes to B. graminis f. sp. Tritici (Bgt) from wheat have also been characterized, including BONZAI 1 (BON1), BON3, ENHANCED DISEASE RESISTANCE 1 (EDR1), and DOMAINS REARRANGED METHYLASE 2 (DRM2); the downregulation of any of these four genes led to significantly reduced powdery mildew symptoms (Table 2; Geng et al., 2019; Zhang et al., 2017b; Zou et al., 2018). Therefore, S genes play a role in disease susceptibility to a wide range of pathogens.

Translating S gene engineering from Arabidopsis to crops

Initial research to increase disease resistance in the model plant Arabidopsis by downregulating S genes laid the foundation for crop improvement. For example, DOWNY MILDEW‐RESISTANT 6 (DMR6) was first identified as a negative regulator of plant innate immunity via EMS mutagenesis of the highly susceptible Arabidopsis mutant enhanced disease susceptibility1‐2 (eds1‐2) in the accession Landsberg erecta. The growth of Hyaloperonospora parasitica was highly constrained in the resulting dmr6 mutants (Van Damme et al., 2005). Moreover, overexpressing DMR6 in the Columbia‐0 accession enhanced susceptibility to P. syringae pv. tomato DC3000 (Pto DC3000) and Hyaloperonospora arabidopsidis. Remarkably, the mutation in DMR6 strongly enhanced resistance to H. arabidopsidis without any growth defects in the plants (Zeilmaker et al., 2015). DMR6 is a SA 5‐hydroxylase (S5H) that catalyses the formation of 2,5‐DHBA (an aromatic compound in green plants) from SA (Zhang et al., 2017c). Since DMR6 plays an important role in regulating plant defence responses in Arabidopsis, its orthologs have been successfully targeted in many different crops (Table 2). For example, a loss‐of‐function mutant of SlDMR6 in tomato generated by CRISPR/Cas9‐mediated mutagenesis showed broad‐spectrum resistance to pathogens, including Pto DC3000, X. gardneri, X. perforans, P. capsici, and Pseudoidium neolycopersici (Thomazella et al., 2021). Knockout mutants of sweet basil (Ocimum basilicum) ObDMR6 using CRISPR/Cas9‐mediated mutagenesis exhibited enhanced resistance to downy mildew caused by Peronospora belbahrii (Hasley et al., 2021). Knockdown of StDMR6 in potato by RNAi (Sun et al., 2016b) and knockout of this gene by CRISPR/Cas9 (Kieu et al., 2021) conferred resistance to P. infestans, which causes late blight disease. A mutation in the banana ortholog MusaDMR6 generated by CRISPR/Cas9 enhanced resistance to banana xanthomonas wilt (BXW) caused by Xanthomonas campestris pv. musacearum (Xcm; Tripathi et al., 2021). In two citrus cultivars, ‘Duncan’ grapefruit and Carrizo citrange, the mutation of CsDMR6 via CRISPR/Cas9 led to significantly increased resistance to the bacterial disease citrus canker caused by X. citri ssp. citri (Xcc; Parajuli et al., 2022). These findings highlight the benefits of engineering disease resistance by disabling disease S genes using genome‐editing techniques in which CRISPR/Cas is a dominant tool.

SUPPRESSOR OF rps4‐RLD 1 (SRFR1) is a negative regulator of ETI‐associated transcriptional immune responses in Arabidopsis (Kwon et al., 2004, 2009). SRFR1 functions as a scaffold protein in association with the defence regulator EDS1 and other TNL proteins, such as RPS4, RPS6, and SUPPRESSOR OF NPR1‐1 CONSTITUTIVE 1 (SNC1), to regulate downstream plant immune signalling (Bhattacharjee et al., 2011; Kim et al., 2010). Mutations in SRFR1 enhanced resistance to the Pto DC3000‐expressing effectors avrRps4 or hopA1 in Arabidopsis in the absence of functional RPS4 or RPS6, respectively (Kim et al., 2009b,c; Kwon et al., 2009), as well as resistance to the generalist chewing insect beet armyworm (Spodoptera exigua) and sugar beet cyst nematode (Heterodera schachtii; Figure 3; Nguyen et al., 2016). Knowledge of Arabidopsis SRFR1 prompted the editing of tomato SRFR1 (SlSRFR1) with the CRISPR/Cas9 system. Interestingly, this approach not only enhanced disease resistance against Pto DC3000, but it also revealed the antagonistic functions of Arabidopsis SRFR1 and tomato SRFR1 as negative regulators of the response to hemibiotrophic pathogens (Pto DC3000) and positive regulators of the response to necrotrophic pathogens (F. oxysporum f. sp. lycopersici and Botrytis cinerea), respectively (Table 2 and Figure 3; Son et al., 2021). Targeting SRFR1 orthologs in other crop plants could offer a new means of engineering enhanced disease resistance.

Figure 3.

SRFR1 engineering as a case study of modifying gene encoding a negative regulator/S protein. (a) Summary of defence responses in the loss‐of‐function mutant srfr1 in Arabidopsis and tomato. Mutation of SRFR1 in Arabidopsis induces disease resistance to Heterodera schachtii, Spodoptera exigua, and Pto DC3000 expressing AvrRps4. However, srfr1 enhanced disease susceptibility to Fusarium oxysporum f. sp. lycopersici (FOL) and Botrytis cinerea. Consistently, mutation of SlSRFR1 enhances resistance to Pto DC3000 and enhances susceptibility to FOL and B. cinerea. Enhancement of resistance and susceptibility are indicated by green and red arrows, respectively. (b) Neighbour‐joining phylogenetic tree based on the amino acid alignment of full‐length products of Arabidopsis SRFR1 (AtSRFR1) and its orthologs in other crops. AtSRFR1 sequence was obtained from The Arabidopsis Information Resource (TAIR10, arabidopsis.org), and SRFR1‐related protein sequences were obtained from the Dicots PLAZA 4.0 and Monocots PLAZA 5.0. Eleven different SRFR1‐like homologous were shown in the figure. The phylogenetic tree with 12 amino acid sequences was made using MEGAX. Amino acid sequences were aligned using the MUSCLE alignment method, and the result was collected to generate a phylogenetic tree with the neighbour‐joining model. Clades were assessed using 1000 bootstrap repeats. (c) Transcriptional reduction of SRFR1 homologous in crops to generate new disease resistance by using CRISPR/Cas9 and CRISPR/dCas9 systems. Created with www.BioRender.com.

As demonstrated by the above examples, successful studies in Arabidopsis have paved the way for the editing of S genes to enhance disease resistance in crops. With this in mind, we compiled a list of S genes that were mutated to improve plant resistance in Arabidopsis but that have not yet been targeted in any crop (Table 3). We categorized these S genes into seven groups, based on the pathogen to which the mutation in the gene confers resistance: bacterium, fungus, virus, oomycete, nematode, arthropod, and multiple pathogens. Notably, the mutation of many S genes causes a broad range of resistance to multiple pathogens (Table 3). For instance, Mediator Complex Subunit 18 (MED18) is a multifunctional protein that regulates plant immunity, flowering time, and responses to plant hormones (Lai et al., 2014). Mutating Arabidopsis MED18 via T‐DNA insertion strongly enhanced plant immunity in response to the fungus F. oxysporum (Fallath et al., 2017) and to many viruses such as TuMV, cauliflower mosaic virus (CaMV), Alternanthera mosaic virus (AltMV), and CMV (Table 3; Hussein et al., 2020). Therefore, orthologs of Arabidopsis MED18 in other crops that are generally affected by fungi and viruses represent potential targets for engineering to reduce crop losses in the future. The mutation of LHP‐INTERACTING FACTOR 2 (LIF2), which controls flowering time and cell fate in Arabidopsis (Latrasse et al., 2011), also enhanced resistance to both the hemibiotrophic bacterium Pto DC3000 and the necrotrophic ascomycete B. cinerea (Table 3; Le Roux et al., 2014). Based on the success of generating new disease resistance traits by knocking down/out the S genes DMR6 and SRFR1 from model plants into crop plants (Table 2 and Figure 3; Son et al., 2021; Thomazella et al., 2021), the information in Table 3 could serve as a guide for improving productivity via plant immune engineering simply by targeting S genes.

Table 3.

List of S genes from Arabidopsis thaliana for potential resistance improvement in crops

No.	Gene name	Gene ID	Method	Pathogen/pest	Reference
Pathogen: Bacteria
1	AtHMAD1	AT1G51090	T‐DNA insertion	Pto DC3000	Imran et al. (2016)
2	ORM1	AT1G01230	T‐DNA insertion	P. syringae	Li et al. (2016)
3	ORM2	AT5G42000	artificial microRNA	P. syringae
4	CBP60a	AT5G62570	T‐DNA insertion	P. syringae pv. maculicola	Truman et al. (2013)
5	LECRK‐I.7	AT5G60270	T‐DNA insertion	P. brassicae, P. capsici	Wang et al. (2014)
6	LecRK‐V.5	AT3G59700	T‐DNA insertion, OX	Pto DC3000	Desclos‐Theveniau et al. (2012), Wang et al. (2014)
7	AtLIK1	AT3G14840	T‐DNA insertion	Pto DC3000	Le et al. (2014)
8	PAT1	AT1G79090	T‐DNA insertion	Pto DC3000	Roux et al. (2015)
9	PBL13	AT5G35580	T‐DNA insertion	Pto DC3000	Lin et al. (2015)
10	PICC	AT2G32240	Knockout	Pto DC3000	Venkatakrishnan et al. (2013)
11	EIJ1	AT2G24860	T‐DNA insertion	Pto DC3000	Liu et al. (2021a)
12	EIL1	AT2G27050	PCR‐based Screen knockout	Pto DC3000	Chen et al. (2009)
13	AtG3BP‐LIKE	AT5G48650	T‐DNA insertion	Pto DC3000	Abulfaraj et al. (2018)
14	IAN9	AT1G33970	CRISPR/Cas9	Pto DC3000	Wang et al. (2019c)
15	IAP1	AT1G18660	T‐DNA insertion	Pto DC3000
16	ATG5	AT5G17290	Autophagy knock‐out	Pto DC3000	Lenz et al. (2011)
17	ATG10	AT3G07525	Autophagy knock‐out	Pto DC3000
18	ATG18a	AT3G62770	RNAi, T‐DNA insertion	Pto DC3000
19	MIEL1	AT5G18650	T‐DNA insertion	Pto DC3000	Marino et al. (2013)
20	MOM1	AT1G08060	TGS mutants	Pto DC3000	Cambiagno et al. (2018)
21	MKP2	AT3G06110	T‐DNA insertion	R. solanacearum	Lumbreras et al. (2010)
22	At NFXL1	AT5G50440	T‐DNA insertion	Pto DC3000	Zhang et al. (2011)
23	AtMEMB12	AT1G10170	T‐DNA insertion	Pto DC3000	Asano et al. (2008)
24	AtNUDT7	AT4G12720	T‐DNA insertion	Pto DC3000	Ge et al. (2007)
25	AtPLA2‐α	AT2G26560	T‐DNA insertion	Pto DC3000	Froidure et al. (2010)
26	AtPRN2	AT2G43120	T‐DNA insertion	R. solanacearum	Zhang et al. (2014)
27	PROSCOOP12	AT5G44585	T‐DNA insertion	E. amylovora	Gully et al. (2019)
28	AtPUB13	AT3G46510	T‐DNA insertion	Pto DC3000	Antignani et al. (2015)
29	AtCBRLK1	AT1G11350	T‐DNA insertion	Pto DC3000	Kim et al. (2009a)
30	SEF	AT5G37055	T‐DNA insertion	Pto DC3000	March‐Diaz et al. (2008)
31	PIE1	AT3G12810
Pathogen: Fungus
32	ABA2	AT1G52340	Amino acid substitution	G. cichoracearum	Xiao et al. (2017)
33	ABA3	AT1G16540	Amino acid substitution	G. cichoracearum	Xiao et al. (2017)
34	ATG2	AT3G19190	EMS, T‐DNA insertion	G. cichoracearum	Wang et al. (2011)
35	RWA2	AT3G06550	T‐DNA insertion	B. cinerea	Manabe et al. (2011)
36	EXO70B1	AT5G58430	T‐DNA insertion	G. cichoracearum	Zhao et al. (2015)
37	EDR2	AT4G19040	EMS	Erysiphe cichoracearum	Tang et al. (2005)
38	EDR4	AT5G05190	EMS	G. cichoracearum	Wu et al. (2015)
39	AtGRXS13	AT1G03850	T‐DNA insertion	B. cinerea	La Camera et al. (2011)
40	MED20	AT2G28230	T‐DNA insertion	F. oxysporum	Fallath et al. (2017)
41	MED8	AT2G03070	T‐DNA insertion	F. oxysporum	Kidd et al. (2009)
42	MED25, PFT1	AT1G25540	T‐DNA insertion	F. oxysporum	Kidd et al. (2009)
43	AtMLO2	AT1G11310	T‐DNA insertion	B. cinerea	Consonni et al. (2010)
44	MYB3R4	AT5G11510	T‐DNA insertion	G. orontii	Chandran et al. (2010)
45	AtMYB46	AT5G12870	T‐DNA insertion	B. cinerea	Ramirez et al. (2011)
46	AtMYB44	AT5G67300	T‐DNA insertion	Alternaria brassicicola	Shim et al. (2013)
47	OCP3	AT5G11270	EMS	B. cinerea, Plectosphaerella cucumerina	Coego et al. (2005)
48	PAD4	AT3G52430	EMS	G. cichoracearum	Neubauer et al. (2020)
49	PMR6	AT3G54920	T‐DNA insertion	E. cichoracearum	Vogel et al. (2002)
50	PUB25	AT3G11840	T‐DNA insertion	B. cinerea	Wang et al. (2018a)
51	PUB26	AT3G19380	T‐DNA insertion	B. cinerea
52	PUX2	AT2G01650	T‐DNA insertion	G. cichoracearum	Chandran et al. (2009)
53	AtSEX1	AT1G10760	T‐DNA insertion	E. cruciferarum	Engelsdorf et al. (2013)
54	PGM	AT1G78050
55	ADG1	AT5G48300
56	At2OGO	ND	CRISPR/Cas9	F. graminearum	Low et al. (2020)
Pathogen: Virus
57	ACS6	AT4G11280	T‐DNA insertion	TMV	Chen et al. (2013b)
58	AtTOR	AT1G50030	RNAi	CaMV	Schepetilnikov et al. (2011)
59	BIR1	AT5G48380	T‐DNA insertion	Tobacco rattle virus	Guzman‐Benito et al. (2019)
60	CBP20	AT5G44200	T‐DNA insertion	Plum pox virus (PPV)	Pasin et al. (2020)
61	CBP80, ABH1	AT2G13540	T‐DNA insertion	PPV
62	CDKC;2	AT5G64960	T‐DNA insertion	CaMV	Cui et al. (2007)
63	CYCT1;5	AT5G45190	T‐DNA insertion	CaMV	Cui et al. (2007)
64	AtDBP1	AT2G25620	T‐DNA insertion	PPV, TuMV	Castello et al. (2010)
65	PCaP1, MDP25	AT4G20260	T‐DNA insertion	TuMV	Cheng et al. (2020)
66	NISP	AT4G30240	T‐DNA insertion	Begomovirus	Gouveia‐Mageste et al. (2021)
67	PAP85	AT3G22640	RNAi	TMV	Chen et al. (2013a)
Pathogen: Oomycete
68	AtERF019	AT1G22810	CRISPR/Cas9	P. parasitica	Lu et al. (2020)
69	IOS1	AT1G51800	T‐DNA insertion	H. arabidopsidis	Hok et al. (2011)
70	AtOBE1	AT3G07780	T‐DNA insertion	H. arabidopsidis	Mukhtar et al. (2011)
71	NPR3	AT5G45110	T‐DNA insertion	H. parasitica	Zhang et al. (2006)
Pathogen: Nematode
72	AtWRKY23	AT2G47260	RNAi	H. schachtii	Grunewald et al. (2008)
73	STP12	AT4G21480	T‐DNA insertion	H. schachtii	Hofmann et al. (2009)
74	PME3	AT3G14310	T‐DNA insertion	H. schachtii	Hewezi et al. (2008)
75	RPE	AT5G61410	T‐DNA insertion	M. incognita	Favery et al. (1998)
Pests: Arthropods
76	AtWSCP	AT1G72290	T‐DNA insertion	Porcellio scaber, Armadillidium vulgare	Boex‐Fontvieille et al. (2015)
77	AGO1	AT1G48410	EMS	M. persicae	Kettles et al. (2013)
78	DCL1	AT1G01040	T‐DNA insertion	M. persicae	Kettles et al. (2013)
79	HRL1	AT4G23660	EMS	S. exigua	Mewis et al. (2005)
80	ETR1	AT1G66340	EMS	S. exigua	Mewis et al. (2005)
81	LOX5	AT3G22400	T‐DNA insertion	M. persicae	Nalam et al. (2012)
Multiple pathogens
82	LIF2	AT4G00830	T‐DNA insertion	Pto DC3000	Le Roux et al. (2014)
				B. cinerea
83	AtLYK3	AT1G51940	T‐DNA insertion	Pectobacterium carotovorum	Paparella et al. (2014)
				B. cinerea
84	RTP1	AT1G70260	T‐DNA insertion, RNAi	G. cichoracearum	Pan et al. (2016)
				P. parasitica
85	AtMED18	AT2G22370	T‐DNA insertion	F. oxysporum	Fallath et al. (2017)
				TuMV, CaMV, AltMV, Cytomegalovirus (CMV)	Hussein et al. (2020)
86	AtSSI2	AT2G43710	EMS	CMV	Sekine et al. (2004)
				M. persicae	Louis et al. (2010)
87	PGL3	AT5G24400	T‐DNA insertion	P. syringae pv. maculicola	Xiong et al. (2009)
				H. arabidopsidis

Alternative ways to engineer disease resistance: introducing pathogen molecules into plants

Plant disease management in the agricultural sector principally relies on breeding disease‐resistant varieties, chemical applications, biological control and cultural practices. Identifying new disease‐resistance resources and selecting new resistant cultivars with broad‐spectrum resistance are the most economical, practical, and effective methods for disease control and prevention. However, the disease resistance of plant varieties against specific pathogens is often limited by the existence of regulatory components that suppress defence responses. Intriguingly, recent studies have identified an assortment of effector proteins that conferred improved disease resistance when expressed in transgenic plants. For instance, overexpressing Phytophthora sojae crinkling and necrosis (CRN) effector (PsCRN115; Zhang et al., 2015b), small cysteine‐rich effector (PstSCR1) from Puccinia striiformis f. sp. tritici (Pst; Dagvadorj et al., 2017), B. cinerea Congo red hypersensitivity effector (BcCrh1; Bi et al., 2021), and M. oryzae SM1 and systemic defence trigger 1 effectors (MoSM1 and MoSDT1; Hong et al., 2017; Wang et al., 2019a) significantly improved disease resistance in various crops (Table 4). Thus, pathogen effectors could be used as genetic resources for a transgenic‐based approach to disease management in crops.

Table 4.

List of effector genes and RNA molecules from pathogens that enhanced resistance in crops

No.	Gene	Gene source	Method	Applied plant	Pathogen	Reference
Effector genes
1	MoSDT1	M. oryzae	OE	O. sativa	M. oryzae	Wang et al. (2019a)
2	MoSM1	M. oryzae	OE	O. sativa	M. oryzae	Hong et al. (2017)
					Xoo
3	PsCRN115	P. sojae	OE	N. benthamiana	Phytophthora capsica, P. parasitica	Zhang et al. (2015b)
4	BcCrh1	B. cinerea	OE	N. benthamiana, A. thaliana, S. lycopersicum, P. vulgaris	B. cinerea	Bi et al. (2021)
5	PstSCR1	Pst	Transiently expressed	N. benthamiana	P. infentans, P. hyoscyami f. sp. tabacina	Dagvadorj et al. (2017)
RNA molecules
6	Bmp3	B. cinerea	dsRNA‐based control	L. sativa	B. cinerea	Spada et al. (2021)
7	CYP51A, CYP51B, CYP51C	F. graminearum	dsRNA‐based control	H. vulgare	F. graminearum	Koch et al. (2016)
8	Myo5	F. asiaticum	dsRNA‐based control	T. aestivum	F. asiaticum	Song et al. (2018a)
9	Bc‐DCL1, Bc‐DCL2	B. cinerea	OE, sRNAs‐and dsRNA‐based control	A. thaliana, S. lycopersicum, Fragaria × ananassa, L. sativa, A. cepa and Rosa hybrida	B. cinerea	Wang et al. (2016)
10	Amino acyl tRNA ligase (SS1G_01703) thioredoxin reductase (SS1G_05899), TIM44 (SS1G_06487)	S. sclerotiorum	dsRNA‐based control	B. napus, A. thaliana	S. sclerotiorum, B. cinerea	McLoughlin et al. (2018)
11	Faβ2Tub‐3	F. asiaticum	dsRNA‐based control	T. aestivum	F. asiaticum	Gu et al. (2019)
				C. sativus	B. cinerea
				H. vulgare	M. oryzae
				G. max	C. truncatum
12	BCMVNIb, BCMVCP	BCMV	dsRNA‐based control	N. benthamiana, V. unguiculata	BCMV	Worrall et al. (2019)
13	CsCYP15C1	C. suppressalis	dsRNA‐based control	O. sativa	C. suppressalis	Sun et al. (2020)

However, the approval of transgenic plants or genetically modified (GM) plants is controversial. The most important consideration for GM plant cultivation is their impact on human health and environmental sustainability. Consequently, GM plants and their products must undergo adequate screening and extensive safety testing before commercialization. Thus, transgenic‐based breeding is relatively time‐consuming and costly. One alternative method that does not produce GM plants involves spray‐induced gene silencing (SIGS) using exogenously applied double‐stranded RNA (dsRNA), short interfering RNA (siRNA), or hairpin RNA (hpRNA). This technique has been applied to enhance resistance against pathogens and pests in several plants (Table 4). For example, the application of dsRNAs or siRNAs to target F. graminearum genes, such as Cytochrome P450 monooxygenase 51A (CYP51A), CYP51B, and CYP51C, suppressed fungal growth in barley (Koch et al., 2016). Similarly, exogenously applied dsRNA targeting myosin5 of Fusarium asiaticum reduced fungal virulence in wheat (Song et al., 2018a). The exogenous application of dsRNAs targeting numerous genes of B. cinerea, including Bmp3, DCL1, DCL2, amino acyl tRNA ligase, thioredoxin reductase, and TIM44, significantly decreased the severity of grey mould disease in various fruits and vegetables (McLoughlin et al., 2018; Spada et al., 2021; Wang et al., 2016). SIGS is not only compatible with fungi, but it has also been used to target a gene from the moth Chilo suppressalis (striped rice stemborer) and several viral components (Sun et al., 2020; Worrall et al., 2019). Therefore, RNAi‐based biocontrol holds great potential for managing devastating diseases and engineering plant innate immunity.

Perspectives

Strategies for targeting PRR and NLR receptors: from broad‐spectrum perception to specific resistance

In general, most plants are resistant to most pathogens, so disease is the exception. A susceptible plant may not contain the proper PRR found in resistant plants to recognize the pathogen. Taking advantage of gene transfer techniques, scientists have introgressed PRR genes from resistant to susceptible plants (Figure 1; Lacombe et al., 2010; Piazza et al., 2021). The exogenous PRRs functioned well in the previously susceptible plants, providing these plants with resistance against a range of pathogens. To enhance the response of a PRR to a PAMP of interest, domain swapping can be performed. The idea of combining a PAMP recognition ectodomain from one PRR with an endokinase domain from another PRR has proven to be successful, since the chimeric PRRs exhibited more robust resistance to the target pathogen compared to the original receptor lacking a kinase domain (Figure 1; Kishimoto et al., 2010). Therefore, PRRs can be transferred from one plant to another or engineered by domain swapping based on the domains of other receptors.

Notably, NLRs are mainly engineered to perceive a specific effector. An NLR normally recognizes a specific cognate effector: recognizing multiple effectors is the exception. If an NLR that is responsible for perceiving a particular effector is known, the NLR could be introduced from resistant plants into susceptible ones. The transfer of NLR genes has been carried out within and across species (Figure 1). Although overexpressing an NLR gene can induce an autoimmune response (Li et al., 2007; Stokes et al., 2002), in some cases, an NLR gene can be overexpressed in a plant to obtain resistance without any side effects (Du et al., 2021; Xun et al., 2019). Because NLRs recognize specific effectors, researchers have tried to expand the recognition specificity of NLRs using several methods. The first such method was random mutagenesis of the LRR domain (responsible for effector recognition; Giannakopoulou et al., 2015; Segretin et al., 2014). Some randomly mutated NLRs conferred enhanced immunity and resistance to another pathogen. Although a mutagenized NLR can lead to a growth defect in the plant, secondary stepwise mutagenesis can be utilized to overcome this issue. However, information about the effects of mutated residues from a mutagenized NLR might not be applicable to another NLR due to the random mutagenesis strategy itself. Secondly, the domain‐swapping strategy for PRRs prompted the idea of domain engineering in NLRs. In detail, in the cases that NLRs work in pairs to recognize effectors, some NLRs that act as sensors but not executors contain an integrated decoy domain for effector recognition. Promising approaches for NLR engineering are introducing the integrated decoy domain into an executor NLR or replacing/incorporating the integrated decoy domain with other decoys/effector targets in a sensor NLR. Indeed, a recent study of the sensor/executer NLR pair RGA5/RGA4 revealed that this strategy is achievable (Cesari et al., 2022). After all, scientists need to consider the partners of NLR in effector recognition and downstream components of NLRs‐mediated resistance to apply NLR engineering in crops. Some NLR introductions could become successful due to the identity between signalling components in the target plants and those in the studied plants. Otherwise, NLR introductions could not be accomplished.

NLRs are not the only targets of ETI‐based engineering. Based on an indirect recognition system, a guardee or decoy protein targeted by a pathogenic effector could be engineered to expand the recognition specificity of a corresponding NLR. In this review, we discussed the engineering of the novel guardee RIN4 and the decoy PBS1. When RIN4 was engineered, the chimeras generated from RIN4s with two functional motifs regulated multiple NLRs (Kim et al., 2022). Crops such as soybean and grape contain RIN4 homologues with variable RIN4‐specific motifs (Kim et al., 2022); these motifs might function in NLR‐mediated resistance that has not yet been characterized. It would be fascinating to explore their functions and engineer a chimeric RIN4 for the regulation of multiple NLRs. When PBS1 was engineered, swapping the PBS1 cleavage site of the effector AvrPphB with other sites cleaved by other effectors expanded the recognition capacity of RPS5 (Figure 1; Kim et al., 2016). This finding suggests that we can engineer crop resistance against any pathogen that takes advantage of a protease as part of its effector repertoire as long as the target crop exhibits PBS1‐mediated immunity.

Two examples of applications for decoy engineering (Bai et al., 2022; Helm et al., 2019) revealed the challenge of translating this strategy into crops: (i) crops might harbour more than one PBS1 orthologs that impede the identification of the actual decoy protein for engineering (Helm et al., 2019); (ii) modifying endogenous decoy genes raises a technical tricky in crops. Therefore, to apply PBS1 decoy engineering, first, the actual native PBS1 decoy must be carefully identified. Second, validated genome editing tools should be considered to generate desirable modifications efficiently. Nowadays, new plant genome editing techniques detailed in Engineering PRRs, NLRs, and regulators of plant innate immunity by CRISPR/Cas‐mediated gene editing section allow the installation of precise mutation. Everything has its challenges; however, every challenge is possible to solve.

Moreover, the successful engineering of PBS1 sheds light on the concept that other target proteins could perhaps be engineered to be cleaved by protease effectors. We suspect that if negative immune regulators are engineered to be cleaved by a specific protease effector, they will become dysfunctional in immunity suppression in the presence of pathogens that harbour the protease. Therefore, depending on the pathogens and protease effectors they contain, engineering negative immune regulators by adding a cleavage site represents a possible approach for resistance engineering based on the case study of PBS1.

How can the expression levels of plant immune regulators be modified as needed?

Besides immune receptors, downstream immune regulators represent potential targets for resistance engineering. Immune regulators positively or negatively regulate disease resistance upon pathogen attack. One basic strategy is to transcriptionally and translationally boost the activities of positive immune regulators. Overexpression systems are commonly used in this approach (Table 1). However, overexpressing genes in plants sometimes leads to slower development due to over‐active immunity (Stokes et al., 2002; Tong et al., 2017). To modify the expression levels and abundance of positive immune regulators, the use of native or pathogen‐specific promoters is recommended. The translation of TL1‐BINDING TRANSCRIPTION FACTOR 1 (TBF1), a transcription factor that functions in the switch from plant growth to defence upon defence activation in Arabidopsis, was shown to be modulated by an upstream open reading frame (uORF; Figure 2a; Xu et al., 2017). Generally, an uORF located in the 5′ untranslated region of a major open reading frame of a gene serves as a translational control factor to precisely fine‐tune the translation of the encoded protein. Based on this finding, Xu et al. developed a ‘TBF1‐cassette’ consisting of the immune‐inducible promoter and two pathogen‐responsive uORFs from the TBF1 promoter region. This cassette constrained the translation of the autoactivated immune receptor snc1‐1 in Arabidopsis and the positive immune regulator AtNPR1 in rice. The characterization of more genes exhibiting a similar type of translation regulation is expected to increase the available repertoire of uORFs. The combination of uORF cassettes with positive immune regulators might function efficiently in crops, resulting in enhanced resistance and normal plant growth.

Figure 2.

Multiple strategies to engineer disease resistance by targeting the promoter region of immune‐related genes. (a) Using dCas9 to control the transcriptional expression of the target genes. dCas9 is fused with a transcriptional activator or repressor. Specific gRNA(s) specifically guide(s) the complex of dCas9 and the activator or repressor to the promoter region of the target gene to up‐regulate or down‐regulate its transcriptional level, leading to disease resistance enhancement in the plant. (b) Using uORF of TBF1 gene to overcome growth defect issue in plant disease engineering. Overexpressing some positive regulators of immunity somehow causes an autoimmune response, which affects plant growth. Incorporating uORF sequences of the TBF gene enables to compromise of normal phenotype without changing disease resistance traits of the plant. (c) Using CRISPR/Cas9 to remove uORF sequences of a positive regulator gene efficiently increases the translational level, leading to disease resistance enhancement. Created with www.BioRender.com.

Another method for resistance engineering is to modify effector targets or negative regulators of the immune system. Specifically, pathogenic effectors interact with and modify host proteins for their virulence function (Jones and Dangl, 2006), and negative regulators suppress immunity during pathogen infection. Therefore, effector targets or negative regulators of immunity could be repressed to boost plant defence. To date, several methods have been used to accomplish this type of suppression, such as T‐DNA insertion to knockout target genes and VIGS or RNAi to knockdown these genes (Table 2). In some cases, knockout/knockdown of negative regulators leads to defective growth and abnormal phenotype by the uncontrolled positive immune regulators. For example, in the Arabidopsis Col‐0 ecotype, a T‐DNA knockout in AtSRFR1 led to defective development but not in the RLD ecotype. It is explained that SRFR1 suppresses the autoimmunity induced by the NLR SNC1 (that does not function in RLD; Kim et al., 2010). Interestingly, unlike the T‐DNA knockout of AtSRFR1/Col‐0, the mutation of SlSRFR1 enhanced defence against a bacterial pathogen with mild growth defects in tomato (Son et al., 2021). These studies revealed that the challenge of engineering immune regulators in one plant due to the growth‐defence tradeoff could be overcome in other plant species. Therefore, identifying and engineering the orthologs of immune regulators remains a promising strategy (Figure 3).

Plant pathologists have long focused on fighting enemies (pathogens) and cherishing allies (plants). Surprisingly, however, using a pathogen component to improve plant resistance can sometimes be successful. Table 4 lists several immune regulators originating from pathogen species that have been used to enhance plant defence responses. Some effectors may be used as critical genetic resources for the transgenic improvement of plant disease resistance. However, the process of genetically modifying crops remains controversial, not only due to technical limitations but also because many consumers are apprehensive about consuming GM products. The emergence of RNAi technology utilizing exogenous dsRNAs, siRNAs, and hpRNAs could be viewed as a viable alternative, as it is more eco‐friendly, sustainable, and broadly acceptable than genetic engineering/transformation. Additionally, RNAi‐based biocontrol could be utilized for pre‐ and post‐harvest disease management in vegetables and other crops. However, the number of target genes that could be used for RNAi‐based biocontrol and our understanding of how plants or pathogens absorb exogenous dsRNA are still limited. These issues pose challenges that could hinder the use of RNA molecules on a large scale or in open‐field conditions. In summary, RNAi‐based biocontrol represents a promising approach to managing devastating plant diseases, but additional studies are needed to overcome the limitations of this technology.

Engineering PRRs, NLRs, and regulators of plant innate immunity by CRISPR/Cas‐mediated gene editing

Plants that were engineered using either transgenic or cisgenic strategies still contain exogenous genetic material, precluding their use as non‐GM crops. The CRISPR/Cas system has recently emerged as a powerful engineering method to generate T‐DNA‐free plants. This system was developed based on the bacterial immune system, using the Cas nuclease to bind to and cleave exogenous DNA sequences from viruses (Terns and Terns, 2011; Wiedenheft et al., 2012). A single guide RNA (sgRNA) accompanies a Cas nuclease to a specific target sequence to generate DNA double‐strand breaks. This event triggers a natural repair mechanism of the host cell via two pathways: homology‐directed repair (HDR) and non‐homologous end‐joining (Malzahn et al., 2017), resulting in a modified host DNA sequence. The versatility of the CRISPR/Cas system allows users to create knockout mutants, insert donor DNA, edit the bases of a target sequence, or control the expression of target genes. This section focuses on the CRISPR/Cas‐based genome editing approach and its potential for disease resistance engineering.

CRISPR‐based tools open the door to immune receptor engineering, which was previously impossible. We discussed LRR domain swapping and NLR mutagenesis in a previous section. These methods could theoretically be performed using CRISPR/Cas. For LRR domain swapping, CRISPR/Cas could be used to insert donor DNA encoding the desired additional domain. CRISPR/Cas‐mediated insertion is challenging, depending on the length of donor template DNA. Recently, the improvement of HDR efficiency in the plant has been validated (Vu et al., 2020, 2021). Therefore, it might be possible to perform LRR domain swapping using an improved CRISPR/Cas system in the future. For NLR mutagenesis, CRISPR/Cas could be used to precisely edit specific target bases to obtain the amino acid sequence of interest. However, more studies of NLR mutagenesis or NLR crystallization are needed to provide references for CRISPR/Cas‐mediated NLR editing. In addition, two recent successful applications of PBS1 decoy engineering are overexpression of modified PBS1 in transgenic soybean and potato (Bai et al., 2022; Helm et al., 2019), which are regarded as GM organisms. This disadvantage inhibits the approval of engineered plants in global markets. CRISPR/Cas‐based modification can produce T‐DNA free in the engineered plants providing the desired resistance traits without foreign DNA.

CRISPR/Cas9 has been widely used to engineer immune regulators, especially negative immune regulators. CRISPR/Cas9 can be used to induce mutations or a premature stop codon to knock out a gene of interest (Tables 2 and 3). SRFR1 and DMR6 are two recent examples of negative regulator genes that were knocked out to gain resistance in tomato (Son et al., 2021; Thomazella et al., 2021). Using CRISPR/Cas9 to knockout genes of interest and applying information about the orthologs of negative regulators will facilitate disease resistance engineering.

The DNA‐binding activity of the gRNA‐Cas9 complex provides an excellent system for altering the expression of genes of interest, especially genes encoding immune regulators. A transcriptional repressor or activator could be combined with dead Cas9 (dCas9) to suppress or enhance the gene expression of a target negative/positive regulator of immunity without introducing DNA double‐strand breaks (DSBs; Figure 2b; Moradpour and Abdulah, 2020; Selma et al., 2019). Based on this knowledge, to enhance the expression level of a positive immune regulator in the plant, targeting the promoter region of the gene using dCas9 represents a promising approach. When the gene encoding the positive regulator contains a uORF(s), standard CRISPR/Cas9‐induced mutation could be utilized to repress uORF‐mediated translational suppression, resulting in the enhanced production of the positive immunity regulator (Figure 2c). Notably, compared to the traditional CRISPR/Cas system, base‐editor and prime editor are the most recent evolution of this technology to generate desired mutations without DSBs (Anzalone et al., 2019; Kang et al., 2018; Komor et al., 2016; Lin et al., 2020; Nishida et al., 2016). Combining engineering strategies mentioned in this manuscript and upgraded versions of new genome editing techniques opens the new door to engineering plant innate immunity wisely.

In summary, our present understanding of plant immune components and current engineering strategies is being utilized to enhance plant resistance. Although every strategy has its advantages and disadvantages, the choice of a suitable strategy could greatly facilitate crop engineering for improved disease resistance. With rapid advances in our understanding of plant immunity, we predict that new engineering strategies will be developed, promising a bright future for crop protection.

Conflict of interest

The authors declare no conflict of interest.

Author contributions

U.T.V. and S.H.K. designed the manuscript structure. U.T.V., A.B.B.I., Q.‐M.N., and S.H.K. wrote the manuscript. U.T.V., A.B.B.I., Q.‐M.N., H.K., J.L., and J.M. generated the figures and tables. All authors read and approved the manuscript.

Funding

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (No. 2021R1I1A3054417), the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT; No. 2022R1A5A1031361), and grants from the New Breeding Technologies Development Program (project no. PJ01653202) and the BioGreen21 Agri‐Tech Innovation Program (project no. PJ015756), Rural Development Administration, Republic of Korea.