Editing the CsDMR6 gene in citrus results in resistance to the bacterial disease citrus canker

Citrus is one of the most important fruit crops in the world. Citrus production worldwide faces challenges from devastating bacterial diseases, including citrus canker caused by Xanthomonas citri ssp. citri (Xcc). Improving citrus resistance to citrus canker and other major bacterial diseases has been a top priority in citrus biotechnology. Disabling disease susceptibility genes has emerged as a novel, promising approach to engineering disease resistance. The bottleneck for applying such an approach has been the identification of proper disease susceptibility-related genes in citrus. Here we show the first successful case of editing the CsDMR6 gene in citrus and obtaining strong resistance to citrus canker in six mutants in two citrus cultivars, ‘Duncan’ grapefruit and Carrizo citrange. Multiple types of deletions and insertions were induced in CsDMR6, resulting in frameshift of its coding region and presumably loss of gene function. The mutation frequency in most of the mutants reached 71.8% to 98.9%. The mutants showed 71.2% to 99.8% reduction in citrus canker lesion and greater than 99.7% or 2.45 to 4.95 Log10 unit reduction in Xcc bacterial cell population. Mutants also accumulated more salicylic acid and expressed much higher levels of the NPR1 gene than the wildtype with or without Xcc inoculation, which suggests potential resistance to other diseases in these mutants. The guide RNAs for targeting CsDMR6 were selected from highly conserved regions and have 100% nucleotide identity with DMR6 homologs in important citrus species; these guide RNAs are expected to work in other important citrus species and cultivars.


Dear Editor,
Citrus (Citrus L.) is one of the most important fruit crops in the world. Sweet oranges (C. × sinensis), grapefruit (C. × paradisi), pummelos (Citrus maxima), lemons (C. × limon), limes (Citrus aurantifolia), mandarins (Citrus reticulata), and tangerines (Citrus reticulata) are the major cultivated types of Citrus. Production worldwide faces challenges from devastating bacterial diseases, including citrus canker, citrus greening (Huanglongbing, HLB), and citrus variegated chlorosis [2]. Citrus canker is caused by Xanthomonas citri ssp. citri (Xcc); most citrus species and cultivars are susceptible to citrus canker. Improving resistance to citrus canker has been an important citrus breeding objective. Recent studies have shown that disabling disease susceptibility genes including the DOWNY MILDEW RESISTANCE 6 (DMR6) gene can be a promising approach to engineering resistance to diseases [1,3,5,6].
Two guide RNAs [dmr6-gRNA1 (CCTCGGGAATCCGGTA CACAAAC), and dmr6-gRNA2 (AGTGGAAAGAGTCTTA-GAAGTGG)] were designed to target CsDRM6. Both gRNAs have 100% identity with DMR6 in sweet orange, mandarins, limes, pummelo, and trifoliate orange (P. trifoliata). A plasmid vector ( Figure 1A) was constructed to express the gRNAs and the HypaCas9, GFP and NPTII genes, introduced into Agrobacterium tumefaciens EHA101, and then transferred into citrus through Agrobacteriummediated transformation (five and nine transformation experiments for "Duncan" and Carrizo, respectively). Cocultivation of 1900 "Duncan" and 5320 Carrizo epicotyl segments with Agrobacterium followed by kanamycin and green fluorescence protein (GFP) selection resulted in nine and 57 GFP-positive shoots, respectively. Based on the GFP expression, the transformation efficiency with "Duncan" and Carrizo was 0.47% and 1.07%, respectively, which were lower than those previously reported. One possible cause of the lower transformation efficiencies might be the much larger size of the T-DNA region in this gene-editing vector. Micro-grafting of GFP-positive shoots resulted in four "Duncan" and 16 Carrizo GFPpositive complete plants in soil. Out of these complete plants, two "Duncan" (DD9 and DD19) and four Carrizo lines (D4, D7, D10, and D12) were analyzed for induced mutations and resistance to Xcc.
Three rounds of deep amplicon sequencing analysis of multiple branches of each mutant over 18 months revealed similar types and frequencies of mutations in each mutant line but multiple types of mutations and mutation frequencies in different mutant lines ( Figure 1B). DD9 contained two primary types of mutations and a mutation frequency of 38.5%. DD19 contained three primary types of mutations and showed a mutation frequency of 74.2%. The primary mutations in D4 were one-base insertion and two-base deletion, and this line had a mutation frequency of 89.0%. The primary mutations in D7 were one-base deletion and one-base insertion, and the mutation frequency was 84.6%. The primary mutation in D10 was 10-base deletion, and this mutant had a mutation frequency of 91.8%. The primary mutations in D12 were four-base deletion and seven-base deletion, and D12 showed a mutation frequency of 100%.
Significant differences were observed in the two gRNAtargeted regions in mutation types and frequencies. Nine types of mutations were induced in the six mutants in the dmr6-gRNA1-targeted region, including the deletion of one, two, four, or 10-bases, or the insertion of one or two bases of different nucleotides ( Figure 1B). These deletions and insertions resulted in frameshifting of the coding region in CsDMR6. The mutation frequency in this region in five out of six mutants ranged from 71.8% (DD19) to 98.9% (D12). Fewer types of mutations (three) and much lower frequencies (2.5% to 12.2%) of mutation were induced in the dmr6-gRNA2-targeted region in most of the mutants.
The non-expressor of pathogenesis-related genes 1 (NPR1) acts as the master key to the plant defense signaling network and is essential for establishing systemic acquired resistance. Before Xcc inoculation (0 dpi), 15 and 20 dpi, the expression levels of NPR1 in wildtype Carrizo and mutants D10 and D12 were low, ranging from 0.021 to 0.053. However, by 10 dpi, NPR1 expression increased by 16-39 fold in both mutants whereas it increased by ∼2 fold in wildtype ( Figure 1H).
In summary, disruptive (frameshift) mutagenesis of CsDMR6 resulted in strong resistance to citrus canker, an important bacterial pathogen to the global citrus industry. HypaCas9 mediated high frequencies of mutations in citrus. The guide RNAs reported here can target DMR6 in multiple important citrus species and cultivars. Functional knocking down of CsDMR6 increased NPR1 expression in citrus, thus editing CsDMR6 may improve citrus resistance to other pathogens.