*2.3. BSR1 Overexpression Enhances Oxidative Bursts in Suspension-Cultured Cells*

Contrary to the *BSR1* disruption phenotype, the overexpression of BSR1 is assumed to enhance defense responses. Whether the overexpression affects the robustness of the oxidative bursts and transcriptional activation was investigated. Rice plants overexpressing HA–PreScission–Biotin (HPB)-tagged BSR1 and GUS (BSR1-HPB:OX and GUS-HPB:OX, respectively) were generated. The GUS-HPB:OX line was used as a control. The integrities of the inserted constructs were confirmed by western analysis with an anti-HA antibody (Supplementary Materials Figure S1a). The overexpression of BSR1-HPB conferred resistance to rice blast, indicating that BSR1-HPB is functional (Supplementary Materials Figure S1b). Suspension-cultured cells were prepared from wild-type, GUS-HPB:OX, and two independent BSR1-HPB:OX lines. The transcript levels of *BSR1* and HPB-tagged transgenes in suspension-cultured cells were ascertained using qRT-PCR (Supplementary Materials Figure S1c).

In response to peptidoglycan treatments, suspension-cultured cells derived from two BSR1-HPB:OX lines produced H2O2 more rapidly than GUS-HPB:OX (Figure 3a; Supplementary Materials Figure S2a). At 60 min after treatment, the overexpression of BSR1 resulted in increased H2O2 concentrations to 1.6–2.0 times that of the control (Figure 3a). Transcript level of a defense-related gene *PAL1* was increased in BSR1-HPB:OX cells compared with GUS-HPB:OX control, while no significant changes in transcript levels of *PBZ1* and *KSL4* were detected (Figure 3b). Transcript levels in GUS-HPB:OX did not necessarily agree with those in WT, indicating that the responses would be slightly altered by overexpression of transgenes. BSR1-HPB:OX cells produced enhanced H2O2 bursts in response to LPS as well as peptidoglycan (Supplementary Materials Figure S3).

Interestingly, before the MAMP treatment, the overexpression of BSR1-HPB resulted in a slight but statistically significant increase in H2O2 concentrations compared with GUS-HPB in cell cultures. Comparisons between the untreated conditions (Figure 3a, 0 min) showed that there were significant differences between BSR1-HPB:OX lines and the GUS-HPB:OX line (*p* < 0.001 for BSR1-HPB:OX17 and BSR1-HPB:OX39, Student's *t*-test). These phenotypes were common to all the replicated experiments (Figure 4; Supplementary Materials Figure S3). BSR1 overexpression did not increase transcript levels of *RbohB*, encoding a NADPH oxidase related to ROS burst (Supplementary Materials Figure S4). These results suggest that an excess of BSR1 protein could constitutively promote NADPH oxidase activity of RBOH proteins but not their transcription.

Because of requirement of BSR1 in chitin oligomer-induced defense responses [39], we assessed the oxidative bursts after a chitin hexamer treatment. The amount of H2O2 produced by BSR1-HPB:OX cells significantly exceeded that of the control at each measured time point (Figure 4a; Supplementary Materials Figure S2b). At 60 min after treatment, BSR1-HPB:OX cells produced a 1.8–1.9-fold greater H2O2 concentration than GUS-HPB:OX cells (Figure 4a). The chitin-induced transcriptional activation of *PAL1*, but not *KSL4* and *PBZ1*, were enhanced by the overexpression of BSR1-HPB (Figure 4b). These comparisons of BSR1-HPBs with GUS-HPB clearly showed that BSR1 overexpression enhanced oxidative bursts and transcriptional activation of, at least, *PAL1* in response to multiple MAMPs.

**Figure 3.** The overexpression of BSR1-HPB enhanced peptidoglycan-induced oxidative bursts in suspension-cultured rice cells. Cells treated with peptidoglycan were analyzed for H2O2 production accompanying oxidative bursts (**a**) and the transcript levels of defense-related genes (**b**). Values are presented as the means ± standard deviations of three biological replicates. In (**a**), H2O2 concentrations were measured before treatment and at 20, 60, and 180 min after treatment. The statistical analysis was performed as shown in Figure S2a. Experiments were conducted three times with similar results. In (**b**), the *PBZ1*, *PAL1*, and *KSL4* transcript levels were normalized against the *RUBQ1* internal control levels. Experiments were conducted two times with similar results. Different letters indicate significant differences (Tukey's test; *p* < 0.05). PGN, peptidoglycan; OX, overexpressing line; HPB, HA–PreScission–Biotin; OX#17, BSR1-HPB:OX#17; OX#39, BSR1-HPB:OX#39; GUS, GUS-HPB:OX; WT, wild-type.

**Figure 4.** The overexpression of BSR1-HPB enhanced chitin-induced defense responses in suspension-cultured rice cells. Values are presented as the means ± standard deviations of three biological replicates. (**a**) H2O2 concentrations were measured before treatment and at 20, 60, and 180 min after treatment. The statistical analysis was performed as shown in Figure S2b. Experiments were conducted three times with similar results. (**b**) The *PBZ1*, *PAL1*, and *KSL4* transcript levels were normalized against the *RUBQ1* internal control levels. Experiments were conducted twice with similar results. Different letters indicate significant differences (Tukey's test; *p* < 0.05). CE, chitin elicitor; OX, overexpressing line; HPB, HA–PreScission–Biotin; OX#17, BSR1-HPB:OX#17; OX#39, BSR1-HPB:OX#39; GUS, GUS-HPB:OX; WT, wild-type.

#### *2.4. Oxidative Bursts against Blast Fungus Are Enhanced in Plants Overexpressing BSR1*

We speculated that H2O2 production in plant leaves as a response to pathogen challenges is increased by the overexpression of BSR1, as observed in suspension-cultured cells. To test the hypothesis, strips from leaf blades were quantitatively analyzed for H2O2 production after being treated with conidia of the blast fungus, which had been autoclaved to eliminate any biological activity. Before the treatment, the H2O2 concentration in water containing leaf strips of BSR1-HPB:OX#17 was slightly greater than that of GUS-HPB:OX (Figure 5). After exposure to autoclaved conidia, leaf strips of BSR1-HPB:OX plants produced far greater H2O2 concentrations than those of GUS-HPB:OX plants (Figure 5a). Taking into consideration the difference between untreated conditions, we calculated changes in H2O2 concentrations during the experiment. By 180 min after treatment, the overexpression of BSR1-HPB resulted in a ~4.2-fold increase in changes in H2O2 concentration, compared with GUS-HPB (Supplementary Materials Figure S5a). Autoclaved conidia-induced H2O2 hyperproduction was also

detected in leaf strips of BSR1-HPB:OX#39, an another BSR1-overexpressing line (Supplementary Materials Figure S6). Thus, *BSR1* overexpression enhanced oxidative bursts in leaf blades.

**Figure 5.** Rice leaf strips derived from BSR1-HPB:OX plants caused an enhanced burst of H2O2 when exposed to conidia of the compatible blast fungus. Leaf strips were cultivated with 8 <sup>×</sup> 10<sup>4</sup> mL−<sup>1</sup> autoclaved conidia (**a**) or 8 <sup>×</sup> <sup>10</sup><sup>3</sup> mL−<sup>1</sup> living conidia (**b**) in wells of a 12-well plate. H2O2 concentrations in wells were measured before treatment and at 60, 180, and 300 min after treatment. Values are presented as the means ± standard deviations of three biological replicates. Asterisks indicate significant differences between the untreated condition (0 min) values and the values at the indicated times in the same line (Student's *t*-test; \* *p* < 0.05, \*\* *p* < 0.01, and \*\*\* *p* < 0.001). Experiments were conducted twice with similar results. OX, overexpressing line; HPB, HA–PreScission–Biotin; OX#17, BSR1-HPB:OX#17; GUS, GUS-HPB:OX.

To assess the importance of the enhanced H2O2 bursts in host–microbe interactions, we also examined the oxidative bursts after a living conidia treatment. A treatment with 8 <sup>×</sup> 104 mL−<sup>1</sup> conidia depressed H2O2 levels in BSR1-HPB:OX leaves and GUS-HPB:OX leaves (Supplementary Materials Figure S7). This result corroborated a previous report that suspensions of conidia contain H2O2-degrading enzymes [19]. In order to avoid that abnormally strong ROS-degrading activity obscures the difference, we used lower concentration (8 <sup>×</sup> 103 mL−1) of conidia. Considering that the H2O2-degrading activity increased with the conidial concentration, comparisons between H2O2 levels were performed only under the same co-cultivation conditions. When co-cultivated with <sup>8</sup> <sup>×</sup> <sup>10</sup><sup>3</sup> mL−<sup>1</sup> conidia, no elevation in the H2O2 level was detected in GUS-HPB:OX leaves (Figure 5b; Supplementary Materials Figure S5b). Thus, the addition of this concentration of conidia completely suppressed MAMP-induced oxidative bursts in the control line. In contrast, when co-cultivated with BSR1-HPB:OX leaves, the H2O2 level significantly increased, compared with before the conidial inoculation (Figure 5b). These co-cultivation experiments revealed that rice plants overexpressing BSR1 produced large amounts of H2O2 that overwhelmed the ROS degradation caused by pathogens.
