The Arabidopsis transgenic and mutant plants BRI1OX, DWF4OX, det2, bes1-D, BES1-RNAi, and gsl8-1 are in Col-0 background, and the DET2OX is in Wassilewskija (WS) background. GSL8 T-DNA insertion mutant gsl8-1 (SALK_111094) was obtained from Arabidopsis Biological Resource Center (ABRC), and the details were described previously (Chen et al., 2009). Arabidopsis plants used in the study were sterilized using 70% (v/v) ethanol and 0.1% (v/v) Triton X-100, plated on 1/2 Murashige and Skoog (1/2 MS) medium, vernalized at 4°C for 2 days in the dark, were incubated for 6 h in the light (150 μmol m^–2 s^–1) at 22°C for germination, and then grown under a long-day condition (22°C, 16-h light/8-h dark).

The Arabidopsis leaves were pretreated by BL or BRZ (1 μM with 0.02% Tween 20) at 12 h before infection. Pst DC3000 or Pst hrcC was cultured on the solid King’s B (KB) medium (peptone 20 g/L, glycerin 10 ml/L, K₂HPO₄ 1.5 g/L, MgSO₄⋅7H₂O 1.5 g/L, and rifampicin 50 mg/L) at 28°C for 24–48 h. Bacteria were scraped off the plates and suspended in 10 mM MgCl₂ to OD600 of 0.02, inoculating 10 mM MgCl₂ without bacteria as mock (Choi et al., 2010), photographed at 3 and 5 days postinoculation (dpi). Pathogen growth analyzed at 1 dpi, and 0.8 cm² leaf disks were taken by puncher, washed by 15% H₂O₂ for 3 min, and then washed with sterile distilled H₂O. The leaf disks were then continuously diluted by water and plated onto the KB medium.

The analysis of chlorophyll fluorescence was described previously (Deng et al., 2015). In brief, chlorophyll fluorescence was determined with an imaging pulse amplitude-modulated fluorometer (IMAG-MINI, Heinz Walz, Germany). For the measurement of Fv/Fm, plants were dark-adapted for 30 min. Minimal fluorescence (Fo) was measured during the weak measuring pulses, and maximal fluorescence (Fm) was measured by a 0.8-s pulse of light at about 4,000 μmol m^–2 s^–1. An actinic light source was then applied to obtain steady-state fluorescence yield (Fs), after which a second saturation pulse was applied for 0.7 s to obtain light-adapted maximum fluorescence (Fm′). Fv/Fm and non-photochemical quenching (NPQ) were calculated as Fm−Fo/Fm and (Fm/Fm′)^–1, respectively.

For superoxide staining, leaves were stained by nitroblue tetrazolium (NBT; 0.5 mg/ml) for 2 h and then soaked in boiling ethanol (95%) until the green color of the leaves faded. The Micro Superoxide Anion Assay Kit (BC1295, Solarbio, Beijing, China) was used to measure the superoxide content. The Catalase (CAT) Activity Assay Kit (BC0205, Solarbio), the Superoxide Dismutase (SOD) Activity Detection Kit (BC0170, Solarbio), the Peroxidase (POD) Activity Detection Kit (BC0090, Solarbio), and the Ascorbate Peroxidase (APX) Activity Assay Kit (BC0220, Solarbio) were used to detect the activity of CAT, SOD, POD, and APX, respectively.

Callose deposition staining was observed at 1 dpi; leaves were cleared by decolorizing solution (acetic acid:ethanol = 1:3) for 12 h, then washed by water, and stained by aniline blue (150 mM K₂HPO₄ (pH 9.5) supplemented with 0.01% aniline blue). Callose deposition was observed by fluorescence microscope with DAPI filter (MDG41, Leica). ImageJ software¹ was used to count the number of callose depositions (Choi et al., 2010).

Total RNA was extracted by the Total RNA Extraction Kit (Solarbio). For quantitative reverse-transcription PCR (qRT-PCR), cDNA was prepared using PrimeScript™ RT Reagent Kit (Takara). Gene expression was performed using the SYBR Green PCR Master Mix (Invitrogen). The CFX Connect Real-Time System (Bio-Rad, Hercules, CA, United States) was used for the qRT-PCR analysis. For each sample, three replicates were performed, and the expression levels were normalized to those of ACTIN2. The primers used for qRT-PCR are listed in Supplementary Table 1.

Chromatin immunoprecipitation (ChIP) assays were performed as previously described (Zou et al., 2019). In brief, 4-week-old Col-0 was cross-linked with formaldehyde, and 125 mM glycine stopped the reaction. Chromatin was sonicated to produce approximately 0.3 kbp DNA fragments. The sonicated protein-DNA complexes were precipitated with an anti-BES1 antibody. After incubation with protein A beads, the beads were further washed with low salt and high salt buffer and reverse cross-linked with 200 mM NaCl. After removing proteins with proteinase K, DNA fragments were purified by phenol-chloroform extraction and ethanol precipitation. The DNA fragments were dissolved in TE buffer (10 mM Tris–HCl pH 8.0, 1 mM EDTA) and used as qPCR templates for the real-time system. TA3 fragment served as a normalization for the qPCR analysis.

Arabidopsis mesophyll cell protoplasts were prepared and transformed as described previously (Li et al., 2020). For luciferase (LUC) assays, we cloned the promoters of GSL8 into the pGreen II 0800 vector and the full-length coding sequence (CDS) of BES1 into the pCAMBIA1307. The constructs used in this study were mentioned in the previous studies (Li et al., 2020; Zhang et al., 2021). Plasmids were singly or co-transformed into Arabidopsis protoplasts. LUC activities were measured using a Luciferase Assay System (Promega) after 16 h, and the data were normalized to REN activity. The experiments were repeated three times with similar results.

Sequence data from this study can be found in the Arabidopsis Genome Initiative database under the following accession numbers: GSL8 (AT2G36850), BES1 (AT1G19350), BRI1 (AT4G39400), DWF4 (AT3G50660), and DET2 (AT2G38050).

The experimental data were statistically analyzed using three or more averages, using one-way ANOVA, and considered significant when P < 0.05.

To study the roles of BRs in plant–pathogen interactions, we examined the effect of brassinolide (BL; the most active BRs) or brassinazole (BRZ, a specific BR biosynthesis inhibitor) treatments on Pst DC3000 resistance in Arabidopsis. At 3 and 5 dpi, plants presented typical disease symptoms and chlorotic leaves (Ishiga et al., 2009). It spread more rapidly in treatment with 1 μM BRZ, while the application of 1 μM BL significantly enhanced Pst DC3000 resistance (Figures 1A,B). Then, we detected the pathogen accumulation in infected leaves at 1 dpi, and BL-treated plants showed less bacterial counts than mock and BRZ-treated plants (Figure 1C). Then, we used various BR biosynthesis genes and transgenic and mutant plants, including BR biosynthesis gene DWF4 overexpression transgenic line (DWF4OX), BR biosynthesis gene DET2 knock out mutant (det2), and DET2 overexpression transgenic line (DET2OX), to investigate the role of BR biosynthesis in plant immunity. BR biosynthesis-enhanced transgenic plants DWF4OX and DET2OX showed higher resistance to Pst DC3000 infection, while BR biosynthesis-deficient mutant det2 showed susceptibility to infection (Figure 1). It demonstrated that BRs enhance plant resistance to Pst DC3000.

Two typical indicators of photosystem II (PSII) photochemistry activity, namely, Fv/Fm (the maximal quantum efficiency of PSII) and NPQ, were detected to test the degree of damage to the light system caused by bacterial inoculation. As shown in Figure 2, there were no significant differences in Fv/Fm (Figures 2A,B) and NPQ (Figures 2C,D) in unchallenged or Pst hrcC inoculated plants. On Pst DC3000 infection, compared with mock-treated wild-type plants, both Fv/Fm (Figures 2A,B) and NPQ (Figures 2C,D) decreased in all of the plants, but BRZ-treated and det2 plants decreased more. It indicated that BRs played a critical role in protecting plant photosystem against Pst DC3000 infection.

Pathogen infection promotes the accumulation of ROS in plants (Deng et al., 2015). Then, we explored the effects of BRs on antioxidant systems when plants were incubated with Pst DC3000. We detected the accumulation of superoxide by NBT staining (Figure 3A) and quantified it by biochemical testing (Figure 3B). The accumulation of superoxide had no significant difference in unchallenged plants but increased after Pst DC3000 infection. The accumulation of superoxide was higher in BL-treated, DWF4OX, and DET2OX plants but lower in BRZ-treated and det2 plants. The enzyme activity of several antioxidative enzymes, such as SOD, POD, CAT, and APX, and the relative expression levels of defense-related genes (PR1 and PR2) were also detected. Pst DC3000 infection increased the activities of all these antioxidative enzymes (Figures 3C–F) and relative expression levels of defense-related genes (Figures 3G,H), and the increase was higher in BL-treated and BR biosynthesis-enhanced transgenic plants and lower in BRZ-treated and BR biosynthesis-deficient mutant plants. All these results illustrated that BRs-induced defense of Pst DC3000 was related to the antioxidant system.

The induction of callose deposition indicates the activation of basal defenses. As Pst DC3000 suppressed callose deposition, to understand the nature of resistance induced by BL, callose deposition was observed at the Pst hrcC infection leaves. When infected with Pst hrcC, compared with wild-type plants, BRZ-treated and BR biosynthesis-deficient mutant showed significantly lower levels of callose deposition, while BL-treated and BR biosynthesis-enhanced transgenic plants accumulated a higher number of callose deposition in leaves (Figures 4A,B). It indicated that BRs induced callose accumulation against Pst DC3000 infection.

To further investigate the roles of BRs in plant disease resistance, we analyzed the effects of different BR signaling components and transgenic and mutant plants in Pst DC3000 resistance. BR receptor BRI1 overexpression transgenic line (BRI1OX), BES1 gain-of-function mutant (bes1-D), and BES1 RNA interference transgenic line (BES1-RNAi) were used in the future study. BRI1OX and bes1-D displayed increased Pst DC3000 resistance, including fewer disease symptoms (Figures 5A,B), less bacterial accumulation (Figure 5C), alleviated photosystem damage (Figures 6A–D), enhanced antioxidant system (Figures 7A–F), and increased defense-related gene expression (Figures 7G,H), while BES1-RNAi displayed the opposite. Then, we analyzed callose deposition in different transgenic and mutant plants after infection. As shown in Figures 8A,B, after infecting with Pst hrcC, BES1-RNAi showed significantly lower levels of callose deposition, while more callose accumulated in the leaves of BRI1OX and bes1-D. These results indicated that BR signaling positively regulated plant resistance to Pst DC3000.

Previous ChIP-chip studies have shown that GSL8 was a direct target of BES1 and induced by BRs (Yu et al., 2011) (Arabidopsis eFP Browser²), and GSL8-deficient mutants had short hypocotyls (Chen et al., 2009) which are typical BR-deficient phenotypes. Thus, we hypothesized that GSL8 was a direct target of BES1 in BR-induced plant immunity. To confirm the result, qRT-PCR experiments were performed. The expression of GSL8 was increased in Col-0 seedlings after BL treatment. In addition, the expression of GSL8 increased to 294% without exogenous BL in bes1-D and even more increased with BL treatment (Figure 9A). Then, the ChIP experiments were performed using an anti-BES1 antibody to confirm whether GSL8 is a direct target of BES1. TA3, a retrotransposable element, was used as the internal control. In the promoter of GSL8, there are two putative BES1 binding fragments at the promoter of GSL8 (Figure 9B). Results of ChIP-qPCR showed that BES1 was enriched significantly at the A1 and A2 regions of GSL8 which contain a typical E-box (CANNTG). We then expressed GSL8 promoter:LUC reporter gene in tobacco leaves. When co-expressed with BES1, GSL8 pro:LUC gene expression was induced (Figures 9D,E). Taken together, our results demonstrated that GSL8 was a direct target of BES1, and its expression was induced by BES1.

To further investigate the connection of BR-induced disease resistance and GSL8, a GSL8 T-DNA insertion mutant gsl8-1 was used for the follow-up experiments. We analyzed the effects of BL on Pst DC3000 resistance in Col-0, BL, gsl8-1, and BL + gsl8-1. As shown in Figures 10A,B, gsl8-1 mutant showed more obvious disease symptoms, after being treated with BL, the symptoms were relieved but still severer than BL-treated wild-type plants. Then, we detected bacterial growth in infected leaves. As the same as symptoms, bacterial counts in gsl8-1 were higher than wild-type. After being treated with BL, bacterial counts in gsl8-1 were obviously relieved but still higher than BL-treated wild-type plants (Figure 10C). We found BR-induced callose accumulation against Pst DC3000, and GSL8 is one of the callose synthases, whether the susceptibility of gsl8-1 is due to callose synthesis blocked? Thus, we detected the callose accumulation in Col-0 and gsl8-1 with/without BL treatment after being infected with Pst hrcC. Compared with Col-0, callose accumulation in gsl8-1 was decreased 38%, although increased 28% in BL + gsl8-1, and increased 58% in BL-treated wild-type plants (Figures 10D,E). Then, we detected the superoxide contents, antioxidative enzyme activities, and defense-related gene expressions in Col-0 and gsl8-1 with/without BL treatment after infection. Compared with Col-0, superoxide contents in gsl8-1 were significantly decreased to 19% (Figure 10F). Also, the antioxidative enzyme activities (Figures 10G–J) and defense-related gene expressions (Figures 10K,L) showed similar trends. These results suggested that BR-induced disease resistance partially depends on GSL8.