Non-host defense response in a novel Arabidopsis-Xanthomonas citri subsp. citri pathosystem

Abstract

Citrus canker, caused by Xanthomonas citri subsp. citri (Xcc), is one of the most destructive diseases of citrus. Progress of breeding citrus canker-resistant varieties is modest due to limited resistant germplasm resources and lack of candidate genes for genetic manipulation. The objective of this study is to establish a novel heterologous pathosystem between Xcc and the well-established model plant Arabidopsis thaliana for defense mechanism dissection and resistance gene identification. Our results indicate that Xcc bacteria neither grow nor decline in Arabidopsis, but induce multiple defense responses including callose deposition, reactive oxygen species and salicylic aicd (SA) production, and defense gene expression, indicating that Xcc activates non-host resistance in Arabidopsis. Moreover, Xcc-induced defense gene expression is suppressed or attenuated in several well-characterized SA signaling mutants including eds1, pad4, eds5, sid2, and npr1. Interestingly, resistance to Xcc is compromised only in eds1, pad4, and eds5, but not in sid2 and npr1. However, combining sid2 and npr1 in the sid2npr1 double mutant compromises resistance to Xcc, suggesting genetic interactions likely exist between SID2 and NPR1 in the non-host resistance against Xcc in Arabidopsis. These results demonstrate that the SA signaling pathway plays a critical role in regulating non-host defense against Xcc in Arabidopsis and suggest that the SA signaling pathway genes may hold great potential for breeding citrus canker-resistant varieties through modern gene transfer technology.

Figure 1. Xcc does not grow in Arabidopsis.

(A) Growth of syringe-infiltrated Xcc in Col-0. (B) Growth of syringe-infiltrated Xcc in Ler. (C) Growth of syringe-infiltrated Xcc in Ws. (D) Growth of syringe-infiltrated Xcc in RLD. (E) Growth of dip-inoculated Xcc in Col-0 and Ler. (F) Growth of spray-inoculated Xcc in Col-0 and Ler. Four-week-old plants were inoculated with Xcc. Bacterial suspensions with an OD₆₀₀ of 0.002 and 0.02 were used for syringe infiltration and dip/spray inoculation, respectively. The in planta bacterial titers were determined on day 0, 3, 6, 9, 12, and 15 post-inoculation for syringe infiltration, and on day 3, 6, and 9 post-inoculation for dip and spray inoculation (cfu, colony-forming units). Data represent the mean of eight independent samples with standard deviation. The experiment was repeated twice with similar results.

Figure 2. Xcc induces ROS accumulation in Arabidopsis.

Four-week-old Arabidopsis plants were syringe-infiltrated with Xcc (OD₆₀₀ = 0.02) or mock control (10 mM MgCl₂). At 4 hpi, infiltrated leaves were excised and stained with DAB (3,3′-diaminobenzidine tetrahydrochloride). The presence of ROS (mainly hydrogen peroxide) caused polymerization of DAB, yielding a reddish-brown color. Tissue was examined under a Leica MEIJI scope. Representative images shown here came from 24 leaves from 12 independent plants. Bars represent 1 cm and 1 mm in images magnified 0.7 and 4 folds, respectively.

Figure 3. Xcc activates both ROS- and flg22-inducible early response genes in Arabidopsis.

(A) Expression of GST1. (B) Expression of FRK1. (C) Expression of NHO1. (D) Expression of WRKY29. Four-week-old Col-0 plants were inoculated with Xcc (OD₆₀₀ = 0.02) or mock-treated with 10 mM MgCl₂. Leaf samples were collected at different time points (0, 4, 8, 12, and 24 hpi) for total RNA isolation and gene expression analysis using RT-qPCR. Expression levels were normalized against constitutively expressed UBQ5. GST1 is a marker gene for the engagement of ROS-dependent defense. FRK1, NHO1, and WRKY29 are flg22-inducible genes. Data represent the mean of three biological replicates with standard deviation. The experiment was repeated twice with similar results.

Figure 4. Xcc induces SA production in Arabidopsis.

(A) Free SA levels. (B) Total SA (SA+SAG) levels. Leaves of wild-type Col-0 plants were inoculated with Xcc (OD₆₀₀ = 0.02) or treated with 10 mM MgCl₂. The inoculated leaves were collected at different time points (0, 4, 8, 16, 24, 48, 72, and 96 hpi) for SA measurement by HPLC. Data represent the mean of four independent samples with standard deviation. The experiment was repeated twice with similar results.

Figure 5. Xcc induces PR gene expression in Arabidopsis.

(A) Expression of PR1. (B) Expression of PR2. (C) Expression of PR5. Four-week-old Col-0 leaves were inoculated with Xcc (OD₆₀₀ = 0.02) or mock-treated with 10 mM MgCl₂. Leaf samples were collected at different time points (0, 4, 8, 12, and 24 hpi) for total RNA isolation and gene expression analysis using RT-qPCR. Expression levels were normalized against constitutively expressed UBQ5. Data represent the mean of three biological replicates with standard deviation. The experiment was repeated twice with similar results.

Figure 6. Growth of Xcc in several Arabidopsis SA, JA, and ET singling mutants.

Leaves of four-week-old plants were inoculated with Xcc (OD₆₀₀ = 0.002). The in planta bacterial titers were determined immediately (day 0) or on day 5 post-inoculation (cfu, colony-forming units). Data represent the mean of eight independent samples with standard deviation. Xcc grew significantly more in eds5, pad4, nho1, eds5npr1, and sid2npr1 than in the wild-type Col-0 plants (*P<0.01, 0.001, 0.001, 0.001, and 0.05, respectively). Similarly, Xcc grew significantly more in eds1 than in the wild-type Ler plants (*P<0.01). The experiment was repeated three times with similar results.

Figure 7. Xcc-induced callose deposition is not changed in the SA signaling mutants.

Four-week-old Arabidopsis plants were inoculated with Xcc (OD₆₀₀ = 0.2) or mock-treated with 10 mM MgCl₂. At 9 and 15 hpi, inoculated leaves were excised and stained with aniline blue. Fluorescence was observed using an Olympus BH-2 epifluorescent microscope. No significant differences were detected among wild type (Col-0 and Ler) and the mutant plants. Representative images shown here came from 24 leaves from 12 independent plants. Bars represent 100 µm.

Figure 8. Expression of early response genes in the Xcc susceptible mutants.

(A to D) Expression of GST1, FRK1, NHO1, and WRKY29 in npr1, eds5, sid2, eds5npr1, and sid2npr1. (E to H) Expression of GST1, FRK1, NHO1, and WRKY29 in eds1. (I to L) Expression of GST1, FRK1, NHO1, and WRKY29 in nho1 and pad4. Four-week-old plants were inoculated with Xcc (OD₆₀₀ = 0.02). Leaf tissues were collected at different time points (0, 4, 8, and 12 hpi) for total RNA isolation and gene expression analysis using RT-qPCR. Expression levels were normalized against constitutively expressed UBQ5. Data represent the mean of three biological replicates with standard deviation. Mutant eds1 is in Ler genetic background, whereas others (nho1, eds5, pad4, sid2, npr1, eds5npr1, and sid2npr1) are in Col-0 genetic background. The experiment was repeated twice with similar results.

Figure 9. Expression of PR genes in the Xcc susceptible mutants.

(A to C) Expression of PR1, PR2, and PR5 in npr1, eds5, sid2, eds5npr1, and sid2npr1. (D to F) Expression of PR1, PR2, and PR5 in eds1. (G to I) Expression of PR1, PR2, and PR5 in nho1 and pad4. The experiment was performed as in Figure 7. Expression was normalized against constitutively expressed UBQ5. Data represent the mean of three biological replicates with standard deviation. Mutant eds1 is in Ler genetic background, whereas others (nho1, eds5, pad4, sid2, npr1, eds5npr1, and sid2npr1) are in Col-0 genetic background. Xcc-induced expression of PR1, PR2, and PR5 was dramatically inhibited in all the tested mutants except nho1. The experiment was repeated twice with similar results.