Broad-spectrum suppression of innate immunity is required for colonization of Arabidopsis roots by the fungus Piriformospora indica

Abstract

Piriformospora indica is a root-colonizing basidiomycete that confers a wide range of beneficial traits to its host. The fungus shows a biotrophic growth phase in Arabidopsis (Arabidopsis thaliana) roots followed by a cell death-associated colonization phase, a colonization strategy that, to our knowledge, has not yet been reported for this plant. P. indica has evolved an extraordinary capacity for plant root colonization. Its broad host spectrum encompasses gymnosperms and monocotyledonous as well as dicotyledonous angiosperms, which suggests that it has an effective mechanism(s) for bypassing or suppressing host immunity. The results of our work argue that P. indica is confronted with a functional root immune system. Moreover, the fungus does not evade detection but rather suppresses immunity triggered by various microbe-associated molecular patterns. This ability to suppress host immunity is compromised in the jasmonate mutants jasmonate insensitive1-1 and jasmonate resistant1-1. A quintuple-DELLA mutant displaying constitutive gibberellin (GA) responses and the GA biosynthesis mutant ga1-6 (for GA requiring 1) showed higher and lower degrees of colonization, respectively, in the cell death-associated stage, suggesting that P. indica recruits GA signaling to help establish proapoptotic root cell colonization. Our study demonstrates that mutualists, like pathogens, are confronted with an effective innate immune system in roots and that colonization success essentially depends on the evolution of strategies for immunosuppression.

Figure 1.

Colonization of an Arabidopsis root by P. indica in the meristematic zone, elongation zone, and MZ. Colonization is restricted to rhizodermal, root hair, and cortical cells, while the root endodermis (brown cells) and root vasculature are not colonized. The letters and arrows in the middle panel point to interaction sites displayed in images A to C. A, Epifluorescence image of epidermal root cells that are penetrated (arrows) without specialized penetration organs. Intracellular hyphae (arrowheads) show a branched morphology. B, Epifluorescence image of intracellular hyphae that are characterized by a distinct globular structure (arrowheads). Penetration sites are indicated by arrows. C, Bright-field image of intracellular sporulation that starts around 14 dai in MZ II. Fungal hyphae were stained with WGA-AF488. Bars = 20 μm.

Figure 2.

Early biotrophic stages of the Arabidopsis-P. indica interaction. A and B, Transmission electron micrographs show intact root cells of Arabidopsis ecotype Col-0 colonized with fungal hyphae (H). Root cells with dense cytosol, intact plastids (P), mitochondria (M), vacuoles (V), ER (arrowheads in A), dictyosome (arrow in B), and cell walls (CW) are shown. A, Parts of a root cell with the plasma membrane closely surrounding the hyphae (arrows). B, Biotrophic colonization by P. indica in MZ II at 3 dai. C to E, Confocal microscopy of a living cortical cell (CC) from colonized Arabidopsis line GFP-Chi with GFP-tagged ER, ER bodies (+), and nucleus (*). See Supplemental Video S1 of this interaction site. Extracellular (arrows) but not intracellular (arrowheads) hyphae are stained with WGA-AF488 in C and with WGA-AF633 in E. Bars = 0.2 μm (A), 1 μm (B), and 15 μm (C–E).

Figure 3.

Suppression of flg22-triggered callose deposition by P. indica in Arabidopsis roots during biotrophic colonization (3 dai). Callose deposition in the MZ (A–D) and elongation/differentiation zone (E–G) of flg22- or mock-treated roots is shown. A to D, Noncolonized (A and C) or P. indica-colonized (3 dai; B and D) roots of 2-week-old seedlings were mock treated (A and B) or treated with 1 μm flg22 (C and D). Callose deposition was only observed in flg22-treated noncolonized roots (C) and almost absent in mock-treated (A and B) and flg22-treated (D) P. indica-colonized roots. E to G, Three types of callose deposition (strong, medium, no) were observed in root tips dependent on flg22 treatment and P. indica colonization. H, The occurrence of callose deposition in root tips in response to the various treatments was quantified. Bars = 200 μm. [See online article for color version of this figure.]

Figure 4.

P. indica suppresses MAMP-triggered growth retardation, oxidative burst, and gene transcription during biotrophic colonization (3 dai). For all analyses, MAMPs were applied to 2-week-old plants at 3 d after P. indica inoculation. A, Suppression of flg22-induced growth retardation by P. indica (se values are from 10 independent measurements of one biological experiment). Experiments were repeated three times with similar results. B and C, Suppression of flg22- and Glc8-induced root oxidative burst by P. indica. Values are given as relative light units (RLU) over time (se values are from four independent measurements per treatment in one experiment). Experiments were repeated three times with similar results. D, Suppression of flg22-induced gene transcription determined by qRT-PCR. Three days after P. indica inoculation or mock treatment, roots were treated with flg22 or mock and harvested 2, 24, and 72 h after treatments. Suppression of flg22-induced transcription was observed for WRKY22 (MTI marker), CBP60g (SA marker), and MYB51 (marker for antimicrobial glucosinolates). VSP2 (JA marker) was induced in P. indica-colonized roots after mock and flg22 treatment. See Supplemental Figure S5 for additional genes. The values represent means with se of one experiment. Experiments were repeated at least twice with similar results. Asterisks indicate significant differences at P < 0.05 (*), 0.01 (**), and 0.001 (***). For D, significant differences between individual time points of Col-0/mock and Col-0/P. indica or Col-0/flg22 and Col-0 + P. indica/flg22 were analyzed by Student’s t test. [See online article for color version of this figure.]

Figure 5.

MAMP-triggered immunity restricts colonization of Arabidopsis roots by P. indica. Three-week-old plants were inoculated with P. indica, and fungal biomass was determined during biotrophic (3 dai) and cell death-associated (7 dai) colonization stages by qRT-PCR. A, Col-0 roots were treated with 10 μm flg22 or mock treated 1 d prior to P. indica inoculation. flg22 pretreatment led to a reduced colonization at 3 and 7 dai (se values are from two independent experiments with 200 plants per treatment and time point). B, Reduced colonization of the MAMP-hyperresponsive triple mutant pub22/23/24 and enhanced root colonization of the chitin-insensitive mutant cerk1-2 [three independent experiments with 200 plants per mutant, wild type, and time point]. C, flg22-induced root oxidative burst in the pub22/23/24 mutant is not suppressed by P. indica. Values are given as relative light units (RLU) over time (se values are from four independent measurements per treatment in one experiment). The experiment was repeated three times with similar results. D, Failed suppression of root defense by P. indica in mutant pub22/23/24. Three-week-old Col-8 (wild type) or the pub22/23/24 mutant was inoculated with P. indica and analyzed at 0, 1, 3, and 7 dai with qRT-PCR for transcription of WRKY22 (MTI marker), CBP60g (SA marker), and MYB51 (marker for antimicrobial glucosinolates). Expression values were calculated by the 2^−ΔCt method by relating Ct values of candidates to those of the housekeeping gene AtUbiquitin5. The values represent means with se and are based on at least two independent biological experiments. Asterisks indicate significant differences at P < 0.05 (*), 0.01 (**), and 0.001 (***) between individual time points of Col-8/mock and Col-8/P. indica or pub22/23/24/mock and pub22/23/24/P. indica and analyzed by Student’s t test.

Figure 6.

SA and glucosinolate defense restrict colonization of Arabidopsis roots by P. indica. Three-week-old plants were inoculated with P. indica, and fungal biomass was determined during biotrophic (3 dai) and cell death-associated (7 dai) colonization by qRT-PCR. For all mutant experiments, the relative amount of fungal biomass was related to the wild type (wt; set to 1). Results shown are means of at least three independent experiments. For each experiment, 200 plants were analyzed per mutant or wild type and per time point. Asterisks indicate significant differences at P < 0.05 (*) analyzed by Student’s t test.

Figure 7.

JA is required for root MTI suppression by P. indica. For all analyses, MAMPs were applied to 2-week-old plants at 3 d after P. indica inoculation. A, Suppression of flg22-induced growth retardation by P. indica in jin1-1 and jar1-1 (se values are from 10 independent measurements of one biological experiment). Experiments were repeated twice with similar results. B, P. indica is unable to suppress flg22-induced root oxidative burst in jar1-1 and elevates the oxidative burst in jin1-1 roots. Values are given as relative light units (RLU) over time (se values are from four independent measurements per treatment of one experiment). Experiments were repeated three times with similar results. C, Three-week-old Arabidopsis plants (Col-0, jin1-1) were mock treated or inoculated with P. indica and harvested at 0, 3, and 7 d after treatments. CBP60g and MYB51 transcription were enhanced in jin1-1, while VSP2 expression was reduced in P. indica-colonized roots. Expression values were calculated by the 2^−ΔCt method by relating Ct values of candidates to those of the housekeeping gene AtUbiquitin5 (se values are from three technical replicates of one biological experiment). Experiments were repeated at least twice with similar results. Asterisks indicate significant differences at P < 0.05 (*), 0.01 (**), and 0.001 (***). For C, significant differences between individual time points of Col-0/mock and Col-0/P. indica or jin1-1/mock and jin1-1/P. indica were analyzed by Student’s t test.

Figure 8.

GA signaling is activated during P. indica colonization. A and B, Two-week-old plants expressing the GFP fusion of the DELLA protein RGA under the control of its native promoter (RGAp::RGA-GFP) were inoculated with P. indica. The degree of GFP fluorescence was determined at 3 and 7 dai, mock treatment, or GA₃ treatment. GFP fluorescence was determined in root tips (A) of short (less than 200 μm in length) and long (more than 200 μm in length) lateral roots and counted (B; se values are from 20–30 root tips per root and treatment, and four roots were analyzed). Experiments were repeated twice with similar results. Bars = 30 μm. C, The expression of the marker gene for GA signaling, Exp-PT1 (At2g45900), was quantified in P. indica-colonized and mock-treated roots at 0, 3, and 7 d after treatment (dat) by qRT-PCR. Expression values were calculated by the 2^−ΔCt method by relating Ct values of candidates to those of the housekeeping gene AtUbiquitin5 (se values are from technical replicates of one biological experiment). Experiments were repeated twice with similar results.

Figure 9.

Effects of GA on MAMP-triggered responses and root colonization. A, Three-week-old Arabidopsis plants (Ler, ga1-6, quintuple-DELLA) were inoculated with P. indica and harvested at 3 and 7 dai. Colonization of ga1-6 roots was reduced at 7 dai. Colonization of quintuple-DELLA mutant roots was reduced at 3 dai and enhanced at 7 dai. P. indica colonization of roots was determined by qRT-PCR (se values are from at least two biological experiments, and 200 plants were analyzed per mutant or wild type and per time point). Asterisks indicate significant differences at P < 0.01 (**) analyzed by Student’s t test. B and C, MAMPs were applied to 2-week-old plants at 3 dai with P. indica. B, Suppression of flg22-induced growth retardation by P. indica in ga1-6 and quintuple-DELLA mutants (se values are from 10 independent measurements of one biological experiment). Experiments were repeated twice with similar results. C, Suppression of flg22-induced root oxidative burst in ga1-6 and quintuple-DELLA mutants by P. indica. Values are given as relative light units (RLU) over time (se values are from four independent measurements per treatment of one experiment). Experiments were repeated three times with similar results. D, GA metabolism affects the expression of CBP60g (SA marker), VSP2 (JA marker), as well as BOI (negative cell death regulator). Expression values were calculated by the 2^−ΔCt method by relating Ct values of candidates to those of the housekeeping gene AtUbiquitin5 (se values are from technical replicates of one biological experiment). Experiments were repeated at least twice with similar results. Asterisks indicate significant differences at P < 0.05 (*), 0.01 (**), and 0.001 (***). For D, significant differences between individual time points of Ler/mock and Ler/P. indica, ga1-6/mock and ga1-6/P. indica, or quintuple-DELLA/mock and quintuple-DELLA/P. indica were analyzed by Student’s t test.

Figure 10.

Model of the spatiotemporal colonization pattern of Arabidopsis roots. Root colonization by P. indica can be divided into four stages. After germination of the spores and extracellular growth, hyphae penetrate epidermal or cortical cells and establish an early biotrophic colonization phase. Biotrophic stages can be preceded by intercellular colonization. The early and late biotrophic stages are characterized by complete intactness of the cell organelles (e.g. nucleus; blue) and plasma membrane invagination (dark gray lines inside cells). Biotrophically colonized cells die (light gray filling of cells) during subsequent cell death-associated colonization. Host cell death is indicated by organelle disruption, while the plasma membrane (dark gray lines inside cells) still surrounds intracellular hyphae. Intracellular sporulation takes place in epidermal and cortical cells at about 14 dai. Endodermis cells (brown) are not colonized. CC, Central cylinder; E, endodermis; C, cortex; RC, rhizodermal cells. MTI is restricting root colonization by P. indica from early through late interaction stages. The fungus achieves biotrophic root colonization by the suppression of early MTI. SA-mediated defense and antimicrobial indole glucosinolates (IGS) participate in MTI. P. indica recruits JA to suppress root oxidative burst. As indicated by mutant studies (Figs. 6 and 7) and gene expression profiles (Fig. 5D), SA and indole glucosinolates might take a dominant role at later colonization stages, at which P. indica might recruit JA signaling and other yet to be defined pathways to counteract SA-supported MTI. P. indica might further induce GA signaling to achieve DELLA protein degradation, thereby elevating the proapoptotic threshold in root cells and initiating cell death-associated colonization. ROS, Reactive oxygen species.