Expression of Putative Defense Responses in Cannabis Primed by Pseudomonas and/or Bacillus Strains and Infected by Botrytis cinerea

Abstract

Cannabis (Cannabis sativa L.) offers many industrial, agricultural, and medicinal applications, but is commonly threatened by the gray mold disease caused by the fungus Botrytis cinerea. With few effective control measures currently available, the use of beneficial rhizobacteria represents a promising biocontrol avenue for cannabis. To counter disease development, plants rely on a complex network of inducible defense pathways, allowing them to respond locally and systemically to pathogens attacks. In this study, we present the first attempt to control gray mold in cannabis using beneficial rhizobacteria, and the first investigation of cannabis defense responses at the molecular level. Four promising Pseudomonas (LBUM223 and WCS417r) and Bacillus strains (LBUM279 and LBUM979) were applied as single or combined root treatments to cannabis seedlings, which were subsequently infected by B. cinerea. Symptoms were recorded and the expression of eight putative defense genes was monitored in leaves by reverse transcription quantitative polymerase chain reaction. The rhizobacteria did not significantly control gray mold and all infected leaves were necrotic after a week, regardless of the treatment. Similarly, no systemic activation of putative cannabis defense genes was reported, neither triggered by the pathogen nor by the rhizobacteria. However, this work identified five putative defense genes (ERF1, HEL, PAL, PR1, and PR2) that were strongly and sustainably induced locally at B. cinerea's infection sites, as well as two stably expressed reference genes (TIP41 and APT1) in cannabis. These markers will be useful in future researches exploring cannabis defense pathways.

Keywords: Bacillus; Botrytis cinerea; Cannabis sativa; Pseudomonas; gray mold; induced systemic resistance; plant growth promoting rhizobacteria; systemic acquired resistance.

FIGURE 1

Symptoms severity on cannabis leaves 2, 4, and 7 days after infection by B. cinerea. Cannabis seedlings were primed with single-bacteria root treatments (Pseudomonas strains WCS417r and LBUM223; Bacillus strains LBUM279 and LBUM979), or consortium root treatments (LBUM279 + WCS417r, LBUM279 + LBUM223, LBUM979 + WCS417r, LBUM979 + LBUM223), or mock-primed with water (control). After one week, plants were infected with B. cinerea (2 droplets containing 10³ conidia/mL) and kept under high humidity for 2, 4, and 7 days. Symptoms were assigned into 4 severity classes: I, no visible symptoms (blue); II, chlorotic tissues forming a yellow halo (yellow); III, necrotic localized lesions smaller than the original droplets size (orange); IV, large spreading lesions with tissue maceration expanding beyond the original droplets and/or sporulating (gray). Stacked barplots represent the number of leaves assigned into each class per bacterial treatment, with 4 independent biological replicates at each harvest time (each leaf comes from a different plant). No statistically significant biocontrol protection was reported when comparing bacteria-treated plants to control plants, at each harvest time (Kruskal-Wallis rank sum test with Dunn pairwise comparisons and Benjamini-Hochberg correction, α = 0.05).

FIGURE 2

Determination of the optimal set of reference genes during a pilot study. The expression stability of seven candidate genes (TIP41, APT1, AP2M, EF1A, YLS8, MON1, and DRH1) was assessed during an independent pilot study. Samples harvest, total RNA extraction and RT-qPCR assays were performed similarly to the defense genes expression study, with 20 biological replicates and 3 technical replicates. (A) Cq values distribution per gene are presented on boxplots with the interquartile range as a box, the lowest and highest values as whiskers, the median as an inner line and the mean as a dot. (B) The determination of the optimal number of reference genes is based on GeNorm pairwise variation values (V_{n/n + 1}). As V_2/3 (highlighted in orange) dropped below the commonly used 0.15 threshold (dashed line), two reference genes should suffice. (C) Expression stability values from BestKeeper (r), GeNorm (M) and NormFinder (SV) are presented for each gene, ranked from the most stable gene (1st, top) to the least stable (7th, bottom). Lower GeNorm and NormFinder values and higher BestKeeper value indicate more stable expression. (D) The final determination of the most stable reference genes is based on their low cumulative stability score, which is the sum of their ranks obtained from the three methods used (rank values inside bars, from left to right: BestKeeper, GeNorm, and NormFinder). The optimal set of reference genes (a pair formed by TIP41 and APT1) is highlighted in orange across all panels.

FIGURE 3

Heatmap showing the expression pattern of putative JA/ET- and SA-mediated genes in cannabis plants primed by Pseudomonas and/or Bacillus strains and infected by B. cinerea. Cannabis seedlings were primed with one of four single-bacteria root treatments, or one of four consortium root treatments, or mock-primed with water (control). After one week, plants were infected with B. cinerea (diseased plants) or left untreated (uninfected plants). After 2, 4, and 7 days (2, 4, and 7 dpi), leaf tissues were sampled from uninfected plants (uninfected leaves) and from diseased plants (systemic leaves remote from infection site and local infected leaves). Total RNA was extracted and the expression level of eight putative defense genes was analyzed by RT-qPCR accounting for primer amplification efficiency. Data were normalized with the reference genes TIP41 and APT1. Fold changes of expression were scaled per gene relatively to the average expression in uninfected leaves at 2 dpi (set as black on color scale). The heatmap represents the log₁₀-transformed mean expression from 4 independent biological replicates and 3 technical replicates; blue indicates lower gene expression than the average expression in uninfected leaves at 2 dpi (downregulation) while yellow indicates higher gene expression (upregulation). Missing values are grayed out.

FIGURE 4

Effects of B. cinerea treatment on the expression of putative JA/ET- and SA-mediated genes at different harvest times in cannabis leaves. Fourteen-day-old cannabis plants were infected with B. cinerea (diseased plants) or left untreated (uninfected plants). Leaf tissues were sampled after (A) 2 days, (B) 4 days, and (C) 7 days. Total RNA was extracted and the expression level of eight putative defense genes was analyzed by RT-qPCR in uninfected plants (uninfected leaves, white) and diseased plants (systemic leaves remote from infection site, light gray, and local infected leaves, dark gray). Fold changes of expression were scaled per gene relatively to the average expression in uninfected leaves at 2 days after infection (expression level of 1). Means and standard errors of 36 independent biological replicates and 3 technical replicates are displayed on a log₁₀ scale. Treatments sharing the same letter are not significantly different according to Wald chi-square test with Benjamini-Hochberg correction, α = 0.05.