Introduction

Plant-parasitic nematodes (PPNs) inflict a heavy toll on global crop productivity that may amount to ~ 200 billion US$ annual economic loss [1, 2]. PPNs possess remarkable abilities to alter host development, physiology and immunity. This is facilitated by a ‘toolbox’ of effectors, which is secreted from PPN esophageal glands into the host cell apoplast or symplast during penetration till maintenance of a feeding site in the vascular tissue [3]. The hyper-metabolic feeding cells such as the giant cells (induced by the root-knot nematode Meloidogyne spp.) and syncytia (induced by cyst nematodes Heterodera and Globodera spp.), serve as the permanent nutrient source for the feeding nematodes [4, 5]. Because of the harmful effects of chemical nematicides on the environment and the lack of alternative management strategies, resistance breeding is emphasized as an efficient and sustainable nematode-management strategy. During the last decade, RNAi-based gene silencing strategies (plants expressing double-stranded RNA constructs that target vital nematode genes including effectors) have gained considerable momentum to manage nematode infection in different host plants [6]. However, in the future, targeting nematode effectors is unlikely to be a robust approach because they are already targets of the plant immune system and, therefore, would be under constant selection pressure to diversify and generate resistance-breaking phenotypes [7].

Notably, only a handful of PPN effectors were found to be involved in the zigzag model of plant immune responses by functioning as avirulence (Avr) proteins (SPRYSEC, RBP-1, GrUBCEP12) that interact with plant resistance (R) proteins (Gpa2, Hero, Mi1.2, and Gro1-4) [7,8,9]. In addition, R gene-mediated resistance in most cases is dependent on the recognition of a single pathogen-derived molecular pattern, and pathogens can readily overcome this type of narrow-spectrum resistance [7,8,9]. Thus, knocking out or mutating the function of a plant susceptibility (S) gene that critically facilitates disease compatibility would provide a broader spectrum and durable resistance [10]. To this end, CRISPR/Cas9-based genome editing technology (in which a single guide RNA or sgRNA directs Cas9 endonuclease to create a double-stranded break (DSB) at the targeted DNA site) has been utilized successfully to knock out a number of plant S genes (MLO, SWEET and eIF4E), that conferred notable resistance against fungal, bacterial and viral pathogens without incurring in any plant growth impairment [11,12,13].

However, deployment of CRISPR/Cas9 for achieving nematode resistance in plants is yet an underexplored research area, possibly because a limited number of S genes have been reported from the plant-nematode pathosystem. A W-box transcription factor (WRKY45) was found to be a key susceptibility factor that modulated the root-knot nematode (Meloidogyne javanica) disease progression in tomatoes [14]. A heavy metal-associated isoprenylated plant protein (HIPP27) was identified as a S gene candidate that supported the parasitism of the beet cyst nematode Heterodera schachtii in Arabidopsis. The loss of function of hipp27 in Arabidopsis (generated by T-DNA insertional mutagenesis) reduced plant susceptibility to H. schachtii infection without affecting the plant phenotype [15]. A spliceosomal protein, SmD1, that interacted with the M. incognita effector EFF18, played a crucial role in giant cell ontogenesis, suggesting the probable function of SmD1 as the susceptibility factor in Arabidopsis [16] and tomato [17]. Most recently, pantothenate synthetase (PANC, involved in the vitamin B5 biosynthetic pathway) was identified as an important susceptibility factor for H. schachtii infection in Arabidopsis. As nematode PANC sequences are quite dissimilar to plant orthologues, nematodes hijack this pathway to meet their nutritional requirements. Loss-of-function mutant lines of Arabidopsis for the AtPANB1 gene (involved in the penultimate step of the pathway) supported fewer nematodes with reduced fitness, and plant susceptibility to nematodes was restored by overexpressing AtPANB1 under the control of the 35S promoter in the mutant line [18].

Members of the HIPP family of proteins have different biological functions, such as heavy metal homeostasis and detoxification, transcriptional response to abiotic stresses (drought, cold, salt, heavy metal, abscisic acid, leaf senescence), and plant-pathogen interactions [19,20,21,22]. A number of HIPP proteins were shown to act as plant immunity hubs (HIPP26 in Nicotiana benthamiana and HIPP19 in Oryza sativa), which are targeted by multiple plant pathogens, including potato mop-top virus and the rice blast fungus Magnaporthe oryzae [23,24,25]. HIPP20 of O. sativa was suggested to be a S gene that promoted the infection of M. oryzae [26]. HIPP05 of O. sativa (alternatively known as Pi21) was depicted as a S gene, since its loss-of-function mutants conferred durable resistance to M. oryzae [27]. In addition, M. oryzae hyphal growth was documented in non-host Arabidopsis, which was overexpressing the OsHIPP05 gene, further verifying the role of HIPP05 as a susceptibility factor [28].

HIPPs are the largest metallochaperone family in A. thaliana and contain 45 HIPP genes, divided into seven major clusters [20]. The zinc-binding A. thaliana HIPP3 protein was localized to the cell nucleus, induced after infection with Pseudomonas syringae pv. tomato, and involved in the regulation of the salicylate-dependent pathogen response pathway, which was antagonistic to the jasmonate pathway [22]. Similarly, zinc-binding A. thaliana HIPP26 was localized to the cell nucleus and implicated in various abiotic stress responses [19]. A. thaliana HIPP27 was localized to the cell cytoplasm and promoted the infection of H. schachtii [15]. To further test the hypothesis that HIPP27 is a S gene, herein, we generated HIPP27 overexpressor (by using overexpression vectors) and knockout A. thaliana lines (by employing the CRISPR/Cas9 system) and investigated the role of HIPP27 in promoting the infection of the most damaging PPN, Meloidogyne incognita. A. thaliana-M. incognita is indeed considered as a model pathosystem to study plant-nematode interactions [29].

Results

Arabidopsis HIPP27 sequence is highly conserved in dicotyledonous plants

The nucleotide sequence of HIPP27 (Gene ID: AT5G66110) was retrieved from the A. thaliana TAIR genome assembly (https://plants.ensembl.org/Arabidopsis_thaliana/) and subjected to BLASTn analysis on the NCBI server using the programme MEGABLAST. Highly similar sequences (Percent identity: 72–100%, Query coverage: 38–100%, E-value: 0) were included in a Maximum Likelihood method-based phylogenetic analysis. The phylogenetic tree was rooted using a HIPP sequence from Oryza sativa as the out-group. As expected, HIPP27 from A. thaliana formed a distinct clade with HIPP27 genes from other members (A. lyrata, Capsella rubella, Camelina sativa, Raphanus sativus, Brassica rapa, B. oleracea, B. napus) from the Brassicaceae family (Fig. 1A). HIPP27 from other major dicot families (Malvaceae, Cucurbitaceae, Fabaceae) branched away from the Brassicaceae group. Most importantly, all these members belong to dicot plant families except for an entry from the monocot family Orchidaceae (Fig. 1A). Notably, the amino acid sequence of the A. thaliana HIPP27 protein showed a high degree of conservation with HIPP27 proteins of different cotton genotypes, including Gossypium hirsutum (Percent identity: 70.75%), G. arboreum (72.79%), and G. raimondii (72.11%) (Supplementary Figure S1).

Fig. 1
figure 1

Sequence conservation of the HIPP27 gene in Arabidopsis and other plants. A Evolutionary relationships of A. thaliana hipp27 (entry is indicated in bold font) with their corresponding homologues from other dicot plants. The phylogenetic tree was constructed in MEGA6 software using Maximum Likelihood method based on the Tamura 3-parameter model. Bootstrap consensus was inferred from 1000 replicates, and branches corresponding to less than 70% of replicates were collapsed. The NCBI accession numbers of 25 different entries are provided in parentheses. HIPP gene of O. sativa was used as the out-group (marked with ●). Entries in different colours represent different plant families, as written beside the clusters. B Phylogenetic analysis of different members of the HIPP gene family in A. thaliana. Gene IDs (retrieved from the TAIR10 genome assembly of A. thaliana) for 45 different entries are provided in parentheses. The branch corresponding to closely related homologues of HIPP27 is indicated in red colour. C Multiple sequence alignments indicated discontinuous sequence conservation between closely related homologues of HIPP27

Full size image

In order to find the highly homologous sequences of HIPP27 in A. thaliana, sequence conservation of HIPP27 across the known 45 HIPP sequences of A. thaliana was studied. A phylogenetic tree showed a distinct clade of HIPP27 that contained HIPP25 (nucleotide sequence identity with HIPP27 – 61.45%), HIPP26 (67.80%), HIPP21 (59.05%), HIPP22 (53.99%), HIPP20 (56.34%), HIPP24 (53.43%) and HIPP23 (54.63%) (Fig. 1B). Multiple sequence alignments of HIPP sequences (a highly homologous stretch is shown here) showed discontinuous sequence conservation (Fig. 1C), suggesting that HIPP27 is a unique gene in the A. thaliana genome.

In order to analyze the transcription pattern of HIPP27 in different developmental stages and plant parts of A. thaliana, RT-qPCR analysis was performed. No significant change (P > 0.05) in HIPP27 expression in different plant parts (roots, shoots, leaves, flowers, and seeds) and developmental stages (7, 14, 21, and 30 days) was observed (Supplementary Figure S2), indicating that HIPP27 is ubiquitously expressed in A. thaliana.

HIPP27 overexpression increased A. thaliana susceptibility to M. incognita

Using zero background TA-cloning vectors, we generated the HIPP27 overexpression lines in the A. thaliana Col-0 background (under the control of a constitutive promoter). Progeny plants of a homozygous line were inoculated with 500 M. incognita second-stage juveniles (J2s). At 3, 10, and 20 days post infection (dpi), approximately a 251-, 217-, and 158-fold increase (P < 0.0001) in HIPP27 transcripts was documented in overexpression plants, respectively, compared to control plants (Fig. 2A). The average number of galls, females, eggs per egg mass, and nematode multiplication factor (MF) ratio were significantly increased (P < 0.05) in overexpression plants compared to wild-type plants at 30 dpi (Fig. 2B), suggesting that increased HIPP27 expression might have caused increased plant susceptibility to nematodes.

Fig. 2
figure 2

Overexpression of HIPP27 increases the susceptibility of A. thaliana to M. incognita. A qPCR validation of an increase in HIPP27 transcript level (RNA was extracted from the root tissues) in an overexpression line upon M. incognita infection at 3, 10, and 20 days. Asterisks (*P < 0.05, **P < 0.01; paired t-test) indicate significant differential expression of HIPP27 mRNA in wild-type and overexpression plants compared to its baseline expression in non-infected control plants (fold change values were set at 1). Gene expression was normalized using two housekeeping genes of A. thaliana (ubiquitin and 18S rRNA). Each bar represents the mean fold change value ± standard error (SE) of qPCR runs in three biological and technical replicates. B The average number of galls, females, eggs per egg mass, and multiplication factor (MF) ratio per root system were increased in the overexpression line compared to the wild-type at 30 days after infection. Bars represent the mean ± SE (n = 10). Asterisks indicate a significant difference between two treatments (*P < 0.05, **P < 0.01; Tukey’s HSD test). Photomicrographs depict the expression of HIPP27 promoter:GUS fusion in A. thaliana root upon M. incognita infection. C Two-week-old root at 0 days post infection (dpi). M. incognita-infected root at 3 (D), 10 (E), and 20 (F) dpi. Scale bar = 100 μm

Full size image

To further investigate whether HIPP27 expression is correlated with nematode infection, A. thaliana Col-0 plants were transformed with a pHIPP27:GUS construct that contained the promoter region of HIPP27 (1 Kb upstream of the start codon) fused with the β-glucuronidase reporter gene. Progeny plants from a homozygous line were inoculated with M. incognita J2s. Roots were stained for GUS activity at different time points. No staining was detected in the root apex and elongation zone of the primary root (Fig. 2C). A strong GUS expression was detected at the nematode infection site (presumably the site of giant cell induction) at 3 dpi (Fig. 2D) and the staining became more intense at 10 dpi with root swelling at the infected site (Fig. 2E). However, GUS staining was weakly detected at 20 dpi in the galled roots (Fig. 2F), suggesting that HIPP27 expression is particularly induced in A. thaliana roots during the early infection stage of M. incognita.

CRISPR/Cas9-targeted mutagenesis of Athipp27 and generation of ‘transgene-free’ mutants

For targeted disruption of Athipp27 by CRISPR/Cas9, two sgRNAs targeting the second exon of Athipp27 were designed using different in silico tools. sgRNAs were targeted from the conserved regions of AT5G66110 transcript and its three transcript variants (AT5G66110.1, AT5G66110.2 and AT5G66110.3) in order to ensure knockout probabilities in all the splice variants of Athipp27 gene. The two sgRNAs and their protospacer adjacent motif (PAM) sequences are shown in Fig. 3A. The efficient selection of sgRNAs were initially validated by secondary structure prediction. Both the sgRNA structures contained the maximum free guide sequence (minimal internal base pairing in the guide sequence leads to greater target recognition), an intact tetra loop (repeat crRNA and anti-repeat tracrRNA), stem loop 2 and 3 (in tracrRNA) (Fig. 3A). The intact stem loop structures promote stable Cas9-sgRNA-DNA complex formation and improve in vivo editing efficacy. An edit vector pHEE401:Athipp27 expressing Cas9, two gRNAs, their scaffolds, and hygromycin antibiotic resistance gene were constructed, mobilized into Agrobacterium tumefaciens strain GV3101, and transformed into A. thaliana Col-0 via the floral dip method.

Fig. 3
figure 3

CRISPR/Cas9-targeted mutagenesis of the Athipp27 gene in A. thaliana. A Mutant identification in T0 plants. sgRNA-targeting sites were located on exon 2 of the Athipp27 gene. Predicted secondary structures of sgRNAs indicate ideal folding of gRNA scaffolds for efficient editing. Based on sequencing results, four different types of mutation (homozygous, heterozygous, bi-allelic and chimeric) were detected in different T0 events. Red hyphens and blue letters denote deletions and insertions, respectively. B Summarization of the editing efficiency and genotype of different T0 plants. C Alignment of the deduced amino acid sequences of the Athipp27 gene (codes for 147 aa) in wild-type and different mutant events. Asterisks indicate translation termination

Full size image

Twelve independent T0 transgenic events were obtained that were designated as Athipp27-cr-1 to Athipp27-cr-12. Sequencing-based genotyping showed that four different types of insertion-deletion (indel) mutations, i.e. homozygous, heterozygous, bi-allelic and chimeric, were detected in the progeny plants of different events (Fig. 3A). Overall, an editing efficiency of 68% was recorded in our study (Fig. 3B). Amino acid sequence analysis of the mutant events showed premature translation termination of HIPP proteins (Fig. 3C), suggesting the probable disruption of gene function. Notably, very high out-of-frame scores for targets 1 (73.55) and 2 (68.68) were recorded while sgRNA designing was performed.

A large deletion of 161 bp in the target gene was also found in event Athipp27-cr-12 (Fig. 4). Apart from sequencing (Fig. 4B), deletion alleles (homozygous and heterozygous) were detected in different individual plants by PCR using primers that flanked the two target sites. DNA fragments flanking the two target sites were 534 bp in the wild-type and 373 bp in the mutant (Fig. 4C). Amino acid sequence analysis of the mutant event showed premature translation termination of the HIPP protein (Fig. 4D). We hypothesize that the large fragment deletion in this case is the resultant effect of the microhomology-mediated end joining (MMEJ) repair mechanism. A very high microhomology score for targets 1 (7429.2) and 2 (5317.7) was recorded while sgRNA designing was performed (Fig. 4A). Our qPCR analysis showed that HIPP27 expression was significantly attenuated in the edited line Athipp27-cr-12 compared to the wild-type (P < 0.01), whereas expression of other closely-related genes, including HIPP20, HIPP21, HIPP22, HIPP23, HIPP24, HIPP25, and HIPP26, was unaltered in the edited and wild-type plants (P > 0.01; Supplementary Figure S3).

Fig. 4
figure 4

CRISPR/Cas9-induced large deletion of the Athipp27 gene in A. thaliana. A Schematic diagram of the sgRNA-targeting sites. sgRNA 1 and 2 sequences are indicated in blue and green, respectively. The PAM site is indicated in red colour. The Cas9 cleavage site is indicated by a red triangle. The use of two sgRNAs led to the deletion of a larger DNA fragment of 161 bp in the mutant T0 event Athipp27-cr-12. B A representative chromatogram of the DNA sequence of a mutant homozygote is given. C Agarose gel electrophoresis showing both homozygous (only deletion allele) and heterozygous (both deletion and wild-type allele) mutants in different clonal lines of Athipp27-cr-12. The DNA fragments flanking the two sgRNA sites were amplified by PCR and were 534 bp in the wild-type and 373 bp in the mutant. M – 100 bp DNA ladder. D Alignment of the deduced amino acid sequences of the Athipp27 gene in wild-type and event Athipp27-cr-12. Asterisk indicates translation termination

Full size image

Probable off-target sites of the two sgRNAs were predicted using Cas-OFFinder and CRISPR-P 2.0 web tools. Genomic DNA was isolated from T0 generation A. thaliana plants, subjected to PCR (using off-target site-specific primers) and sequencing. No induced mutations were detected in the off-target loci of mutant plants (Table 1).

Table 1 Analysis of the putative off-target effect at potential off-target sites of the Arabidopsisthaliana mutants. Mismatching bases are indicated in bold. Nucleotides in italics are PAM sequences. Off-target loci located in the intergenic region were not sequenced
Full size table

To identify ‘transgene-free’ homozygous Athipp27 knockout mutants, the presence or absence of Cas9 and sgRNA in transgenic A. thaliana plants of T1 generation (obtained via self-pollination of T0 plants) was determined by multiplex PCR (primers specific to a housekeeping gene, 18S rRNA, were included in the reaction mixture as a reference). Three mutant plants (plant numbers 5, 8, and 11) inherited from the Athipp27-cr-12 line did not generate Cas9- and sgRNA-specific amplicons (Fig. 5A, B). As these mutants did not contain any transgenic elements of hipp27-sgRNA/Cas9 vectors, we considered them as ‘transgene-free’ homozygous mutants.

Fig. 5
figure 5

Detection of ‘transgene-free’ homozygous Athipp27 knockout mutants in the T1 generation. A Detection of a partial Cas9 coding sequence (856 bp) in different mutant individuals. B Detection of sgRNA (639 bp), a region spanning from the sgRNA scaffold to downstream vector in different mutant individuals. A. thaliana 18S rRNA gene (506 bp) was used as the reference in multiplex PCR reactions. M – 100 bp DNA ladder, WT – wild-type plants, V – plasmid vector, N – negative control without template DNA

Full size image

Loss of function of HIPP27 increased A. thaliana resistance to M. incognita without impairing the host basal defense

By further self-pollinating ‘transgene-free’ homozygous Athipp27-cr-12 plants, a T2 generation was obtained that was subjected to growth phenotype analysis and ‘challenge inoculation’ with nematodes. No adverse effect on plant shoot and root morphology was documented due to the induced mutation in the hipp27 gene according to the comparative growth phenotypes of wild-type and mutant plants at 14 days after germination (Fig. 6A, B). No significant difference in average dry weight, root length, flowering time and plant height between wild-type and mutant plants was observed (P > 0.05; Supplementary Figure S4).

Fig. 6
figure 6

The loss of function of HIPP27 reduced A. thaliana susceptibility to M. incognita. Growth phenotypes of wild-type (A) and ‘transgene-free’ Athipp27 knockout mutants (B) showed no adverse effect on plant shoot and root morphology due to the induced mutation in the hipp27 gene. Scale bar = 3 cm. C The average number of galls, females, eggs per egg mass, and multiplication factor (MF) ratio per root system decreased in mutants compared to the wild-type at 30 days after infection. Bars represent the mean ± SE (n = 10). Asterisks indicate a significant difference between two treatments (*P < 0.05, **P < 0.01; Tukey’s HSD test). Photomicrographs demonstrate greater galling intensity in wild-type roots (D) compared to mutants (E). Scale bar = 500 μm

Full size image

Each T2 plant was infected with 500 J2s of M. incognita in a pot, and plants were harvested at 30 dpi to measure different susceptibility parameters. Compared to wild-type plants, mutant plants exhibited significantly lower susceptibility to M. incognita in terms of average number of galls (Percent reduction – 52.11%, P = 0.02), females (52.64%, P = 0.002), eggs per egg mass (26.53%, P = 0.032) and MF ratio (64.87%, P = 0.001) (Fig. 6C). As expected, the galling intensity was comparatively greater in wild-type roots compared to mutant ones (Fig. 6D, E). Overall, these results suggest that A. thaliana HIPP27 plays a crucial role in M. incognita parasitism.

To examine whether the reduced susceptibility of mutant plants to M. incognita is a result of altered plant defense responses, we carried out RT-qPCR analysis to quantify the expression of selective A. thaliana basal defense marker genes in the shoots and roots of nematode-infected wild-type and gene-edited plants at 2 days after inoculation. The marker genes include peroxidase, MPK4 (related to oxidative stress), EDS1, PAD4, PR1, PR2 (salicylic acid pathway), PDF1.2, HEL1 (jasmonic acid pathway), ERF6 and ACS2 (ethylene signaling). The transcript abundance of all 10 marker genes did not differ significantly between wild-type and edited plants (P > 0.05; Supplementary Figure S5), indicating the specific role of HIPP27 in conferring nematode susceptibility.

Discussion

Arabidopsis HIPP27 was previously shown to act as a S gene that facilitated the infection of the beet cyst nematode H. schachtii in A. thaliana; loss-of-function mutants were generated by T-DNA insertional mutagenesis [15]. The present study describes the role of HIPP27 as an important susceptibility factor for southern root-knot nematode M. incognita infection in A. thaliana. We generated the loss-of-function mutants using a CRISPR/Cas9-mediated targeted mutagenesis approach. To date, CRISPR/Cas9-based editing has been utilized to investigate the molecular basis of plant–nematode interactions rather than conducting research on S genes and creating mutants to achieve PPN resistance [30,31,32]. In a recent report, CRISPR/Cas9-based knockout of the S gene, OsHPP04, conferred rice (cv. Nipponbare) resistance to the root-knot nematode M. graminicola [33]. Our study provides further insights into the CRISPR/Cas9-mediated functional analysis of nematode-responsive S genes in plants.

Dong et al. [30] utilized the CRISPR/Cas9 system to knock out the syntaxin gene of the t-SNARE family in the nematode‐resistant soybean cultivar Peking. Intriguingly, Peking roots with mutations in the syntaxin gene exhibited increased susceptibility to H. glycines infection, suggesting the critical role of t-SNAREs in conferring H. glycines resistance. Using CRISPR/Cas9, a putative R gene (MG1) was functionally validated in rice. Compared to wild type, homozygous mutant lines of ZH11 (japonica variety), SL 22–620 and HKG 98 (aus rice) conferred susceptibility to M. graminicola, indicating that MG1 is required for M. graminicola resistance [31]. Similarly, CRISPR/Cas9-induced mutagenesis in the OsPR10/OsBet v1 gene conferred greater susceptibility of rice to M. graminicola, suggesting that OsBet v1 is involved in rice defense against nematodes [32]. A CRISPR/Cas9-induced mutation in the WRKY45 gene exhibited substantially lower M. incognita infection in tomatoes, thus revealing that SlWRKY45 may act as a negative regulator of tomato defense against M. incognita [34].

Arabidopsis HIPP proteins are the largest family of metallochaperones that contain a heavy metal-binding (HMA) domain [20]. A number of plant immune receptors possess HMA domains, implicating their putative role in host defense against pathogens [35]. Different HIPP orthologues, including HIPP26 in N. benthamiana and HIPP19 in O. sativa, act as plant immune response modulators that target multiple pathogen effectors [23,24,25]. HIPP05 and HIPP20 of O. sativa were reported as S genes that facilitated M. oryzae proliferation in the host tissue [26,27,28]. A. thaliana HIPP3 was localized to the cell nucleus, and was found to regulate the salicylate-dependent pathway of pathogen response via binding zinc [22]. Similarly, A. thaliana HIPP26 was localized to the cell nucleus, and by binding to a zinc finger transcription factor, it was found to regulate the drought stress response [19]. On the contrary, A. thaliana HIPP27 was localized to the cell cytoplasm only [15]. In this study, we showed that A. thaliana HIPP27 is ubiquitously expressed in different plant parts and across different developmental stages. Phylogenetic analysis indicated that HIPP27 is highly conserved across the dicotyledonous plants and most specifically in the Brassicaceae family. However, HIPP27 was found to be a unique member of the metallochaperone family when its sequence was compared with other HIPP genes from A. thaliana.

In the present study, we generated HIPP overexpression lines to assess whether increased transcription of HIPP27 conferred greater susceptibility of A. thaliana to M. incognita. As expected, the average number of galls, females, eggs per egg mass, and MF ratio (which determines nematode reproductive success in the host plant; [36]) were greatly increased in overexpression plants compared to the wild-type. In addition, HIPP27 fusion with a GUS reporter construct demonstrated that nematode infection potential is very much correlated with HIPP27 expression. A similar GUS assay was conducted to identify the nematode-responsive promoter elements in A. thaliana [37, 38].

Next, we generated a number of genome-edited lines that harboured the different knockout variants of the hipp27 gene. We designed two sgRNAs from the second exon of Athipp27 (as the first exon was quite smaller in size and did not contain conserved regions across the hipp27 transcript variants). The secondary structure of each sgRNA (which interacts with the Cas9 protein in vivo and thus improves editing efficiency) contained a fairly greater stretch of free guide sequence and intact stem loop structures, which was indicative of greater stability in the sgRNA structure [39]. We encountered a large deletion of 161 bp in an event called Athipp27-cr-12, probably because of the MMEJ-mediated DSB repair mechanism. We recorded greater microhomology scores for both sgRNAs, suggesting the proclivity of DSBs (at two target sites) to undergo the MMEJ repair pathway. Notably, a deletion of 248 bp was encountered in one of the transgenic rice events when the MG1 gene was knocked out via CRISPR/Cas9 [31]. Apart from non-homologous end joining (NHEJ) and homology-directed repair (HDR), the MMEJ repair mechanism has emerged as a major DSB repair pathway in plant genome editing applications [40].

We did not find any induced mutations in the off-target loci of the transgenic plants. Event number Athipp27-cr-12 was taken forward (as premature translational stoppage was documented in the HIPP amino acid sequences, presumably no functional HIPP protein was produced) for generating homozygous T2 mutants. Transgene-free mutants were ‘challenge inoculated’ with M. incognita J2s in the controlled environment. At 30 dpi, 52.11, 52.64, 26.53, and 64.87% reductions in gall number, nematode females, eggs per egg mass, and MF ratio, respectively, were documented in mutants compared to wild-type plants. Additionally, mutant plants did not show any pleiotropic effect on plant phenotypes. The expression of defense response genes did not significantly differ between wild-type and mutant plants, indicating that the reduced susceptibility of HIPP mutants to M. incognita is directly related with the loss of function of HIPP27 gene. Any possibility of CRISPR-induced positive or negative effect on plant defense gene expression can be excluded in our study. Notably, the loss of function (via T-DNA mutation) of HIPP27 did not impair plant basal defense in A. thaliana shoots treated with flg22 immunopeptide [15].

In conclusion, our CRISPR-Cas9-induced targeted mutagenesis of the HIPP27 gene established Athipp27 as a susceptibility factor for M. incognita infection in A. thaliana. As S genes are recessively inherited in host plants, induced mutations in S genes are likely to be durable and may confer broad-spectrum resistance [10]. The findings of this study can facilitate our understanding of the role of S genes in other plant-nematode pathosystems, and similar S genes may be targeted in crop plants (such as cotton) to generate nematode-resistant genome-edited lines.

Materials and methods

Culturing of nematodes

A pure culture of M. incognita race 1 was propagated in the roots of eggplant (Solanum melongena cv. Pusa Purple Long) in pots in a growth chamber at 28˚C with 16: 8 h light: dark photoperiod (light level – 300 μmol m−2 s−1). At 2 months after nematode inoculation, plants were harvested, egg masses were handpicked from the infected roots and hatched in sterile tap water at 28ºC. Freshly hatched J2s were used for all the experiments.

Plant growth conditions and nematode infection

Seeds of Arabidopsis thaliana ecotype Columbia-0 (Col-0) were surface sterilized with 70% ethanol (for 2 min), 0.1% HgCl2 and 0.1% SDS (for 5 min), followed by rinsing in sterile distilled water. Seeds were germinated in Petri plates containing the MS media. Plates were incubated at 21ºC with 16: 8 h light: dark photoperiod. Two-week-old seedlings were transferred to 500 ml pots containing soil rite (Keltech Energies Ltd., Bengaluru). Plants were grown in the growth chamber (at the National Phytotron Facility, ICAR-IARI) at 21ºC with 16: 8 h light: dark photoperiod. After 3 weeks, each plant was inoculated with 500 M. incognita J2s near the root zone using a sterilized pipette tip. Plants were harvested at different time points to examine different nematode infection parameters, i.e., number of galls, females (or egg mass), eggs per egg mass, and multiplication factor [(number of egg mass × number of eggs per egg mass) ÷ initial inoculum level]. An identical procedure was followed for analyzing the nematode infection level in transformed plants. At least 10 plants were included in each treatment, and the whole experiment was repeated at least three times.

Bioinformatics analyses

Sequences of different candidate genes were exported from the NCBI database and the A. thaliana TAIR genome assembly. Exon, intron, coding sequence, and 5´ and 3´ UTR in the sequence were analyzed in the FGENESH (https://www.softberry.com/) and Expasy (https://web.expasy.org/) webservers. Conserved domains and motif signatures were assessed in the NCBI conserved domain database (https://www.ncbi.nlm.nih.gov/Structure/cdd/) and the InterProScan database (https://www.ebi.ac.uk/interpro/). Sequences were aligned with their homologues from different plant species via the Clustal Omega tool. A phylogenetic tree was constructed using the MEGA6 tool by following the Maximum Likelihood method based on the Tamura 3-parameter model. The tree was generated with the highest log likelihood (-15,493.9900). A discrete Gamma distribution was used to model evolutionary rate differences among sites (5 categories (+ G, parameter = 1.5549)).

Generation of Athipp27-overexpression plants

pCXSN (a zero background TA cloning-based vector; [41]) was digested with XcmI (New England Biolabs) to create T-overhangs. In parallel, total RNA was extracted from the two-week-old Arabidopsis plants by using the NucleoSpin RNA Plant Kit (TaKaRa) in accordance with the manufacturer’s instructions and reverse-transcribed to cDNA by using the SuperScript VILO cDNA synthesis Kit (Invitrogen). The coding sequence of Athipp27 (444 bp) was PCR-amplified (primer detail is provided in Supplementary Table S1) by using proofread-capable Phusion DNA polymerase (Invitrogen) and A nucleotide overhang were created in PCR products by following the A-tailing protocol (https://www.promegaconnections.com/a-quick-method-for-a-tailing-pcr-products/). Gel-purified digested pCXSN vector and Athipp27 PCR product were ligated via T4 DNA ligase (Invitrogen) with standard vector to insert molar ratio of 1:3 [42]. pCXSN: Athipp27 was initially transformed into E. coli DH5α cells by electroporation, sequence verified, and subsequently transformed into A. tumefaciens strain GV3101 via the freeze–thaw method. The overexpression vector contained the CaMV 35S promoter.

Genomic DNA was extracted from two-week-old Arabidopsis plants using the NucleoSpin Plant II Kit (TaKaRa) in accordance with the manufacturer’s instructions. The promoter region of Athipp27 (1 Kb upstream of the start codon) was PCR-amplified (primer detail provided in Supplementary Table S1) from the genomic DNA and ligated into pCXGUS-P [41] by using the TA cloning method as described above. pCX-GUS-P: Athipp27 was transformed into A. tumefaciens strain GV3101 as described above.

Four- to six-week-old A. thaliana Col-0 wild-type plants (germinated in Petri plates containing MS media followed by transfer to pots containing soil rite) were transformed with A. tumefaciens by the floral dip method [43]. After drying the plants, T0 seeds were harvested and sterilized before being screened on MS media plates containing the antibiotic hygromycin (25 mg L−1). Transformants (3–4 leaf stage) that survived in antibiotics were transferred to pots containing soil rite in the growth chamber (Supplementary Figure S6). T3 homozygous plants were used for infection experiments.

GUS staining

Histochemical validation of GUS activity was analyzed using 5-bromo-4-chloro-3-indolyl-b-D glucuronide (XGluc) as the substrate. Nematode-infected roots at different stages (3, 10, and 20 days after infection) were carefully washed free of soil and submerged in a freshly-prepared GUS staining solution (0.5 mM X-Gluc, 0.1 M NaHPO4, 0.5 mM K3Fe(CN)6, 0.5 mM K4Fe(CN)6, 0.01 M EDTA, 20% methanol, 0.1% Triton X-100) for overnight at 37ºC. Root tissues were cleared by replacing the solution with 70% alcohol, and photomicrographs were taken with a Zeiss Axiocam MRm microscope.

Gene expression analysis

Total RNA was extracted from nematode-infected plants as described above. The integrity of the RNA molecule was checked by electrophoresis in a 1% (w/v) agarose gel. RNA purity and quantity were examined in a Nanodrop spectrophotometer (Thermo Fisher Scientific). Approximately 1 µg of RNA was reverse-transcribed to cDNA using the SuperScript VILO cDNA synthesis Kit (Invitrogen). A qPCR-based transcriptional profile of different candidate genes (primer details are provided in Supplementary Table S1) was carried out in a CFX96 thermal cycler (BioRad). The qPCR reaction volume (10 μL) contained 1.5 ng cDNA, 750 nM each of forward and reverse primers, and 5 μL SYBR Green PCR master-mix (BioRad). qPCR amplification conditions were as follows: a hot start phase of 95 °C for 30 s, 40 cycles of 95 °C for 10 s and 60 °C for 30 s. In addition, a melt curve programme (95 °C for 15 s, 60 °C for 15 s, then a slow ramp from 60 to 95 °C) was followed to visualize the qPCR amplification specificity. Quantification cycle (Cq) values were exported from CFX Maestro software (BioRad). Arabidopsis housekeeping genes (18S rRNA and ubiquitin) were used as internal references for normalizing the candidate gene expression data. The fold change in the target gene was calculated using the 2−ΔΔCq method. The qPCR run contained three biological and three technical replicates for each sample.

Designing of guide RNAs

Using the Athipp27 sequence as the query, gRNAs were designed via different in silico tools (RGEN—https://www.rgenome.net/; CHOPCHOP—https://chopchop.cbu.uib.no/; CRISPick—https://portals.broadinstitute.org/; MMEJ-KO—http://skl.scau.edu.cn/mmejko/). Based on greater microhomology predictor (repairs two DSBs using microhomology-mediated end joining (MMEJ)), out-of-frame score (ensures codon shifting), minimal self-complementarity, minimal off-target (by using Cas-OFFinder tool—http://www.rgenome.net/cas-offinder/) effect, and secondary structure prediction (RNAfold webserver (http://rna.tbi.univie.ac.at/cgi-bin/RNAWebSuite/RNAfold.cgi) predicts stem-loop and hairpin formation in the intended crRNA: tracrRNA hybrids), gRNAs were shortlisted.

Generation of CRISPR/Cas9 constructs

In order to assemble two gRNAs, two gRNA spacer or target sequences (20 bp each) along with the BsaI recognition site were incorporated into PCR forward and reverse primers, respectively. Four primers were used that contained overlapping sequences. Using the pCBC vector (Addgene) as a template, a single PCR fragment was amplified that contained target 1, gRNA scaffold, Arabidopsis U6 terminator, promoter, and target 2. Next, the purified PCR fragment was assembled into the multiple cloning site of the Cas9-expressing binary vector pHEE401 (Addgene) by using the Golden Gate cloning method [44, 45]. The recombinant pHEE401 (pHEE401:Athipp27-cr) contained two gRNA expression cassettes driven by U6 promoters and terminators and codon-optimized Cas9 expressed under the control of an egg cell-specific promoter and rbcS E9 terminator [45]. Recombinant pHEE401 was transformed into E. coli DH5α cells, followed by A. tumefaciens strain GV3101 via the freeze–thaw method. Construct length and orientation were confirmed by colony PCR and sequencing. Detailed information about the cloning procedure, gRNA module sequences, PCR primers, and colony PCR primers is provided in Supplementary Figure S7 and Table S1.

Plant transformation and mutation analysis

Four- to six-week-old A. thaliana Col-0 wild-type plants were transformed with A. tumefaciens as described above. T0 seeds were germinated in MS media supplemented with hygromycin (25 mg L−1). Antibiotic-resistant transformants were transferred to pots containing soil rite in the growth chamber. To detect mutagenesis at desired sites, genomic DNA was extracted from individual plants of the T0 generation using the NucleoSpin Plant II Kit (TaKaRa). Next, target regions were PCR amplified (forward and reverse primers flanked target 1 and target 2, respectively; primer details are provided in Supplementary Table S1) with specific primers and Sanger sequenced. Sequencing data were analyzed in the SnapGene viewer, and editing efficiency was calculated. Additionally, PCR products were resolved on an agarose gel to detect chromosomal fragment deletions. T0 plants were selfed to generate T1 followed by T2 seeds.

Off-target mutation analysis and identification of ‘transgene-free’ mutants

To analyze off-target effects, Cas-OFFinder (http://www.rgenome.net/cas-offinder/) and CRISPR-P 2.0 (http://crispr.hzau.edu.cn/CRISPR2/) were initially used to predict potential off-target sites for each target or gRNA spacer. Further, fragments of different potential off-target sites were amplified with their corresponding primers (listed in Supplementary Table S1) and sequenced to examine whether any mutation occurred or not.

To identify ‘transgene-free’ T1 generation plants, genomic DNA was extracted from T1 events using the NucleoSpin Plant II kit (TaKaRa). Subsequently, Cas9-specific primers, gRNA scaffold to downstream vector-specific primers, and Arabidopsis thaliana housekeeping gene (18S rRNA, NCBI accession number: X16077)-specific primers (listed in Supplementary Table S1) were used to amplify the transgenic elements from the genomic DNA by using the multiplex PCR technique.

Nematode resistance analysis of mutant plants

Homozygous mutant T2 plants were grown in Petri dishes containing MS medium. Two-week-old seedlings were transferred to pots as described above. Three-week-old plants were inoculated with 500 M. incognita J2s near the root zone. At 30 days after inoculation, plants were harvested, and different parameters, including numbers of adult females, egg mass, eggs per egg mass, and nematode multiplication factor, were studied. Different phenotypic characters such as plant length, weight, and flowering time were compared between wild-type and mutants to determine whether any growth penalty was incurred by the induced mutation.

Statistical analysis

The data are presented as the mean ± SE of at least three independent experiments. Data from different bioassays were initially checked for normality using the Shapiro–Wilk test and subjected to a one-way ANOVA test or a t-test (for pairwise comparisons between different treatments), followed by the Tukey’s honest significant difference test (for multiple comparisons) in SAS v. 14.1 software.