CRISPR/Cas9-driven double modification of grapevine MLO6-7 imparts powdery mildew resistance, while editing of NPR3 augments powdery and downy mildew tolerance

Abstract

The implementation of genome editing strategies in grapevine is the easiest way to improve sustainability and resilience while preserving the original genotype. Among others, the Mildew Locus-O (MLO) genes have already been reported as good candidates to develop powdery mildew-immune plants. A never-explored grapevine target is NPR3, a negative regulator of the systemic acquired resistance. We report the exploitation of a cisgenic approach with the Cre-lox recombinase technology to generate grapevine-edited plants with the potential to be transgene-free while preserving their original genetic background. The characterization of three edited lines for each target demonstrated immunity development against Erysiphe necator in MLO6-7-edited plants. Concomitantly, a significant improvement of resilience, associated with increased leaf thickness and specific biochemical responses, was observed in defective NPR3 lines against E. necator and Plasmopara viticola. Transcriptomic analysis revealed that both MLO6-7 and NPR3 defective lines modulated their gene expression profiles, pointing to distinct though partially overlapping responses. Furthermore, targeted metabolite analysis highlighted an overaccumulation of stilbenes coupled with an improved oxidative scavenging potential in both editing targets, likely protecting the MLO6-7 mutants from detrimental pleiotropic effects. Finally, the Cre-loxP approach allowed the recovery of one MLO6-7 edited plant with the complete removal of transgene. Taken together, our achievements provide a comprehensive understanding of the molecular and biochemical adjustments occurring in double MLO-defective grape plants. In parallel, the potential of NPR3 mutants for multiple purposes has been demonstrated, raising new questions on its wide role in orchestrating biotic stress responses.

Keywords: Erysiphe necator; New Plant Breeding Techniques (NPBTs); Plasmopara viticola; Vitis vinifera; stilbenes.

Figure 1

Alignment and editing frequencies. (A, C, E) Frequencies of editing percentages for each guide RNA (gRNA) and target gene. Numbers within the pie charts represent the number of plants belonging to the specific class. (B, D, F) For each target gene, three lines have been selected for subsequent experiments. The editing event for each line is reported for MLO6g7 (B), MLO7g25 (D), and NPR3g2 (F).

Figure 2

Evaluation of powdery mildew infection and leaf ecophysiological traits. Disease indices of powdery mildew‐infected plants were collected through the grape assess mobile application. (A, B) (A) Incidence and (B) severity percentages of leaf infections (n = 4, each biological replicate is composed by at least 50 observations). (C–E) Representative leaf of NPR3‐edited, MLO6‐7‐edited and WT‐plants, respectively. Black arrows indicate sites of infection. (F–H) Leaf ecophysiological traits, including: (F) Chlorophyll Content Index (CCI) (n = 10), (G) maximum quantum efficiency of PSII (F _v/F _m) (n = 10) and (H) leaf thickness (n = 10) were collected using a PhotosynQ apparatus. Lowercase letters above bars indicate significant differences among the samples, as determined by ANOVA followed by Tukey post hoc test. All data in bar charts represent mean values ± standard deviation bars. In box‐plot charts, box limits indicate the range of the central 50% of the data, with a central line marking the median value and lines extend from each box to capture the range of the remaining data.

Figure 3

Evaluation of downy mildew infection. Assessment of leaf resistance to Plasmopara viticola using the OIV 452‐1 descriptors (the higher the value, the higher the resistance). (A, B) Distribution of class frequencies recorded during the first trial 1 (A) and trial 2 (B) (n = 10 for each trial). The two trials were carried out 15 days one from the other. Asterisks (***) aside the legend denote highly significant differences as attested by the chi‐squared test on the class frequencies (P‐value ≤0.001). (C) representative leaf of NPR3‐edited, MLO6‐7‐edited and WT‐plants, respectively.

Figure 4

Analysis of the leaf‐associated fungal population in edited and WT plants. Fungal genera were identified through the qiime2‐ITSexpress pipeline and compared among samples (n = 4). Only the top 20 representative genera among the total classified amplicons were retained in the graphic representation.

Figure 5

Quantification of leaf stilbenes in edited and WT plants. (A–D) Contents of trans‐resveratrol (A), piceid (B), δ‐viniferin (C), and ε‐viniferin (D) quantified in leaf samples of Vitis vinifera wild type (WT), MLO6‐7‐edited, or NPR3‐like plants (n = 3 for each line and compound). Data are expressed as μmol g⁻¹ of dry weight (DW). Lowercase letters above bars indicate significant differences among the samples, as determined by ANOVA followed by Tukey post hoc test. All data in bar charts represent mean values ± standard deviation bars.

Figure 6

Overview of the main findings. Summary of the susceptibility degree to powdery mildew (PM) and of the molecular, physiological and biochemical adjustments observed in MLO6/7 edited (left side) and NPR3 edited (right side) plants. Metabarcoding analysis revealed that, while in MLO6/7 plants PM development was almost impaired, in NPR3 plants PM development significantly lower than in wild‐type plants (WT) but not that much as in MLO6/7. Additionally, while in MLO6/7 plants no influence was observed on the relative abundance of other pathogens, in NPR3 ones a significant reduction was observed. From the physiological point of view, leaf thickness, chlorophyll content index and photosystem II efficiency (F _v/F _m) were evaluated, highlighting a thicker leaf in NPR3 plants. From the molecular point of view, similarity between MLO6/7 and NPR3‐edited plants were observed. From the biochemical point of view we observed an enhancement of stilbenes in both type of plants but involving different type of molecules. In MLO6/7 plants increase in δ‐ and ε‐viniferin was observed while piceid and resveratrol increased more in NPR3 plants. Finally, the oxidative status was monitored revealing an increase of scavenging potential in both type of plants but with a larger extent in NPR3 plants.