Rice transcription factor bHLH25 confers resistance to multiple diseases by sensing H2O2

Abstract

Hydrogen peroxide (H₂O₂) is a ubiquitous signal regulating many biological processes, including innate immunity, in all eukaryotes. However, it remains largely unknown that how transcription factors directly sense H₂O₂ in eukaryotes. Here, we report that rice basic/helix-loop-helix transcription factor bHLH25 directly senses H₂O₂ to confer resistance to multiple diseases caused by fungi or bacteria. Upon pathogen attack, rice plants increase the production of H₂O₂, which directly oxidizes bHLH25 at methionine 256 in the nucleus. Oxidized bHLH25 represses miR397b expression to activate lignin biosynthesis for plant cell wall reinforcement, preventing pathogens from penetrating plant cells. Lignin biosynthesis consumes H₂O₂ causing accumulation of non-oxidized bHLH25. Non-oxidized bHLH25 switches to promote the expression of Copalyl Diphosphate Synthase 2 (CPS2), which increases phytoalexin biosynthesis to inhibit expansion of pathogens that escape into plants. This oxidization/non-oxidation status change of bHLH25 allows plants to maintain H₂O₂, lignin and phytoalexin at optimized levels to effectively fight against pathogens and prevents these three molecules from over-accumulation that harms plants. Thus, our discovery reveals a novel mechanism by which a single protein promotes two independent defense pathways against pathogens. Importantly, the bHLH25 orthologues from available plant genomes all contain a conserved M256-like methionine suggesting the broad existence of this mechanism in the plant kingdom. Moreover, this Met-oxidation mechanism may also be employed by other eukaryotic transcription factors to sense H₂O₂ to change functions.

Fig. 1. H₂O₂ promotes OsLAC7/28/29-mediated lignin biosynthesis and disease resistance by repressing miR397b expression.

a Three-week-old Kitaake plants were pre-treated with or without 1 mM H₂O₂ on roots for 72 h, then their leaves were inoculated with M. oryzae Zhong10-8-14. Lesion length (n ≥ 18 lesions) and fungal growth (n = 3 technical replicates) at 7 days post inoculation (dpi) are shown. Mock treatment indicates treatment without H₂O₂. b OsLAC7/28/29 expression levels in three-week-old Kitaake leaves at 0–72 hours post treatment (hpt) with or without 1 mM H₂O₂ on roots (n = 3 technical replicates). c Lesion length (n = 9 lesions) and fungal growth (n = 3 technical replicates) of three-week-old Kitaake, Oslac7-KO and Oslac7/28/29-KO plants at 7 dpi with Zhong10-8-14. d Lignin contents in four-week-old Kitaake, Oslac7-KO and Oslac7/28/29-KO plants (n = 3 biological replicates). e Three-week-old triple KO of OsLAC7/28/29 and Kitaake plants were pre-treated with or without 1 mM H₂O₂ on roots for 72 h, then their leaves were inoculated with Zhong10-8-14. Lesion lengths (means ± SD, n = 30 lesions) at 7 dpi are shown. f OsLAC7/28/29 RNA levels in three-week-old Kitaake, miR397b-KO and miR397b-OE plants (n = 3 technical replicates). g miR397b levels in three-week-old Kitaake 0, 60 and 72 hpt with or without 1 mM H₂O₂ on roots (n = 3 technical replicates). h Lesion length (n = 9 lesions) of three-week-old Kitaake, miR397b-KO and miR397b-OE plants at 7 dpi with Zhong10-8-14. i Lesion lengths (n = 20 lesions) of three-week-old Kitaake, miR397b-KO, Oslac7/28/29-KO and double mutant Oslac7/28/29-KO/miR397b-KO plants at 7 dpi with Zhong10-8-14. Data are means ± SD and analyzed by two-tailed Student’s t-test (a, b, g), one-way ANOVA with Least Significant Difference (LSD) test (c, d, f, h, i), and two-way ANOVA with Tukey’s test at ***P < 0.001; ns, not significant (e). Scale bar, 1 cm (a, c, e, h, i). Experiments were done with three biologically independent replications.

Fig. 2. bHLH25 represses miR397b expression to enhance lignin biosynthesis and disease resistance.

a Yeast one-hybrid assay for binding of bHLH25 to pmiR397b. 3-amino-1,2,4-triazole (3-AT) was used to inhibit leaky reporter expression. b Subcellular localization of bHLH25 in rice protoplasts. RFP-NLS, red fluorescence protein with a nuclear localization signal. c The G-box-like motifs and seven fragments of pmiR397b are shown. d Fold enrichment of pmiR397b in fragmented rice DNA pulled down by GST-bHLH25 in DAP-qPCR assay (n = 3 technical replicates). GST was used as a negative control. e Fold enrichment of fragmented DNA of pmiR397b pulled down by bHLH25-YFP in ChIP-qPCR assay (n = 3 technical replicates). bHLH25-YFP was immunoprecipitated from plants overexpressing bHLH25-YFP in bhlh25-KO background by protein A-magnetic beads coupled with anti-GFP or without antibodies (No Abs, negative control). f EMSA for examining the binding of bHLH25 to pmiR397b. GST-bHLH25 and GST (negative control) were incubated with biotin-labeled probe (bio-pmiR397b-G-box) and unlabeled competitors containing wild-type G-box-like-2 motif (in blue) of pmiR397b (com-pmiR397b-G-box) or mutant motif (in red, mut-pmiR397b-G-box). The red arrow indicates biotin-labeled probes bound to GST-bHLH25. FP, free probes. g miR397b levels (n = 3 technical replicates) in three-week-old Kitaake, bhlh25-KO and bHLH25-OE plants. h Lesion length (n = 9 lesions) of 3-week-old Kitaake, bhlh25-KO and bHLH25-OE plants at 7 dpi with Zhong10-8-14. i Lignin contents (n = 4 biological replicates) in three-week-old Kitaake, bhlh25-KO and bHLH25-OE plants. j Histochemical staining of cross-sectioned leaves with phloroglucinol-HCl and thickness of sclerenchyma cells in three-week-old Kitaake, bhlh25-KO and bHLH25-OE plants (n = 21 cells). k Lesion length (n = 20 lesions) of three-week-old Kitaake, bHLH25-OE, miR397b-OE, and double mutant miR397b-OE/bHLH25-OE plants at 7 dpi with Zhong10-8-14. Data are means ± SD and analyzed by two-tailed Student’s t-test (d, e) and one-way ANOVA with LSD test (g–k). Scale bars, 10 μm (b), 50 μm (j) and 1 cm (h, k). Experiments were done with three biologically independent replications.

Fig. 3. bHLH25 promotes CPS2-mediated phytoalexin biosynthesis and disease resistance in rice in a manner independent of miR397b.

a Representative leaves of Kitaake, bHLH25-OE, miR397b-KO and OsLAC7-OE plants at the tillering stage. b Binding of bHLH25 to the N-box-like motif in pCPS2 in EMSA. GST-bHLH25 and GST (negative control) were incubated with biotin-labeled probe (bio-pCPS2-N-box) with or without unlabeled wild-type (com-pCPS2-N-box) or mutant competitor (mut-pCPS2-N-box). Wild-type and mutant N-box-like sequences are highlighted in blue and red, respectively. The red arrow indicates biotin-labeled probes bound to GST-bHLH25. c CPS2 RNA levels in three-week-old Kitaake, bhlh25-KO and bHLH25-OE plants (n = 3 technical replicates). d Phytocassane C contents in three-week-old Kitaake, cps2-KO and CPS2-OE plants (n = 3 biological replicates). e Phytocassane C contents in three-week-old Kitaake, bhlh25-KO and bHLH25-OE plants (n = 3 biological replicates). f Representative leaves of Kitaake and CPS2-OE plants at the tillering stage. g Lesion length (n = 9 lesions) and fungal growth (n = 3 technical replicates) of three-week-old Kitaake, cps2-KO and CPS2-OE plants 7 dpi with Zhong10-8-14. h Lesion lengths (n = 20 lesions) of three-week-old Kitaake, bHLH25-OE, cps2-KO, and double mutant cps2-KO/bHLH25-OE plants at 7 dpi with Zhong10-8-14. i Lignin contents in three-week-old Kitaake, cps2-KO and CPS2-OE plants (n = 3 biological replicates). j Phytocassane C contents in three-week-old Kitaake, miR397b-KO and miR397b-OE plants (n = 3 biological replicates). Data are means ± SD and analyzed by one-way ANOVA with LSD test (d, g, h–j) or Dunnett’s test (c, e). Scale bar, 1 cm (a, f–h). Experiments were done with three biologically independent replications.

Fig. 4. bHLH25 directly senses H₂O₂ to regulate miR397b- and CPS2-mediated defense pathways in rice.

a Schematic drawing of oxidized methionine residues of bHLH25. M256 of bHLH25 is in the basic domain of bHLH25. Mass-to-charge ratio (m/z) is shown. b Representative secondary peaks of mass spectrometry spectrum of oxidized M256 of bHLH25. c Alteration of bHLH25 DNA-binding specificity by H₂O₂ treatment in EMSA. GST-bHLH25 pre-treated with H₂O₂ was incubated with probe bio-pmiR397b-G-box (left panel) or bio-pCPS2-N-box (right panel). The red arrow indicates biotin-labeled probes bound to the GST-bHLH25. d H₂O₂ enhances the inhibitory effect of bHLH25 on pmiR397b (left panels) but weakens the activation effect of bHLH25 on pCPS2 (right panels) in plants. YFP RNA levels (n = 3 technical replicates) were determined by RT-qPCR analysis. YFP protein levels, bHLH25-HA and RLUC-HA (negative control) were detected by immunoblot analysis. Signal levels were quantified based on gray-scale values (c, d). e–j Endogenous H₂O₂ content (e, n = 3 biological replicates), miR397b expression (f, n = 3 technical replicates), OsLAC7 expression (g, n = 3 technical replicates), lignin content (h, n = 3 biological replicates), CPS2 expression (i, n = 3 technical replicates), and phytocassane C content (j, n = 3 biological replicates) were measured for three-week-old Kitaake leaves 0–96 hpi with or without Zhong10-8-14. k Three-week-old Kitaake, bhlh25-KO and bHLH25-OE plants were pre-treated with or without 1 mM H₂O₂ on roots for 72 h, then their leaves were inoculated with Zhong10-8-14. Lesion length (n = 30 lesions) at 7 dpi is shown. Scale bar, 1 cm. Data are means ± SD and analyzed by one-way ANOVA with Dunnett’s test (d), two-tailed Student’s t-test (e–j) and two-way ANOVA with Tukey’s test at **P < 0.01, ***P < 0.001; ns, not significant (k). Experiments were done with three biologically independent replications.

Fig. 5. M256 is essential for bHLH25 to sense H₂O₂ and promote disease resistance.

a In silico analysis of the predicted homodimer and the basic domain of bHLH25. The purple circle indicates the basic domain contributing to DNA binding specificity. b Representative lesions, lesion length (n = 18 lesions) and fungal growth (n = 3 technical replicates) of three-week-old Kitaake, bhlh25-KO#6, bHLH25-complemented and bHLH25^M256V-complemented plants 7 dpi with Zhong10-8-14. Scale bar, 1 cm. c In vitro oxidation of M256 in bHLH25 in the presence of H₂O₂. GST-bHLH25 and GST-bHLH25^M256V proteins were treated with 0–1 mM H₂O₂ before immunoblotting. Anti-GST antibody indicates GST-bHLH25 and GST-bHLH25^M256V levels. d In vivo oxidation levels of M256 in bHLH25 in nuclei extracted from three-week-old bHLH25-OE#4 plants at different hours post treatment with or without H₂O₂. e, f In vivo oxidation levels of M256 in bHLH25 in nuclei extracted from rice protoplasts (e) or leaf clippings (f) at different hours post treatment with or without H₂O₂. g In vivo oxidation levels of M256 in bHLH25 in protoplasts of two-week-old Kitaake and Osrboha-KO#1 plants expressing bHLH25-GFP. The endogenous H₂O₂ levels in protoplasts were measured (n = 6 biological replicates). h In vivo oxidation levels of M256 in bHLH25 in three-week-old bHLH25-OE#4 plants at different hours post inoculation with or without Zhong10-8-14. Relative oxidation value was calculated by dividing the oxidation value of M256 in bHLH25 treated with M. oryzae by that of bHLH25 treated with mock. Anti-GFP detects bHLH25-YFP protein levels immunoprecipitated from rice protein extracts (f–h). Data are means ± SD and analyzed by one-way ANOVA with LSD test (b) and two-tailed Student’s t-test (g). Experiments were done with three biologically independent replications.

Fig. 6. bHLH25 confers resistance to multiple diseases and its M256 is highly conserved among bHLH25 homologs in plants.

a M256 and neighboring residues in the basic DNA-binding domain of bHLH25 are conserved across different plant species. The red box and arrow highlight the conserved M256-like residues in bHLH25 orthologues, and the black box highlights the target recognized by the oM256 antibody. b Immunoblot shows that H₂O₂ directly oxidizes the M291 residue of AtbHLH25. GST-AtbHLH25 and GST-AtbHLH25^M291V were pre-treated with or without H₂O₂ and probed on an immunoblot. Anti-GST indicates a loading control. c, d Representative lesions and lesion length of tillering-stage Kitaake, and bHLH25^M256V- and bHLH25-overexpressing plants in bhlh25-KO background at 14 dpi with Xanthomonas oryzae pv. oryzae PXO99A (c, n = 20 lesions) and at 2 dpi with Rhizoctonia solani AG-1-IA (d, n = 10 lesions). Data are means ± SD and analyzed by one-way ANOVA with LSD test; scale bar is 1 cm (c, d). e Working model for bHLH25 which confers resistance to pathogens by directly sensing H₂O₂. H₂O₂ induced in rice upon pathogen infection oxidizes bHLH25 at M256 to promote OsLAC7/28/29 expression, leading to lignin accumulation for cell wall reinforcement to prevent pathogen penetration. Laccases in turn consume H₂O₂ in lignin biosynthesis, leading to accumulation of non-oxidized bHLH25. Non-oxidized bHLH25 promotes CPS2 expression leading to phytoalexin accumulation to inhibit hyphae expansion of pathogens that escape into plant cells. Experiments were done with three biologically independent replications.