p3 T-DNA A M0066754 p1 p2 OsDCL1a p3 5 Kb T-DNA p1 p2 M0040827 OsDCL1a 5 Kb B WT (TN67) WT (TN67) He (dcl1a) WT(Az) WT(Az) (dcl1a) M0066754 M0040827 Ho p1/p2 p1/p2 p3/p2 p1/p3 C WT(Az) dcl1a-Ac#1 WT(Az) dcl1a-Ac#1 21 d 60 d Relative expression (x103) D 0,016 Nascent polypeptide-associated complex 0,008 0 Fig. S1. Analysis of OsDCL1a-Ac mutant plants. (A) Schematic representation of the T-DNA insertion rice mutants (TRIM collection, M0066754 and M0040827 mutants). The position of primers used for the analysis of T-DNA integration is shown (p1, p2, p3). (B) PCR-amplified DNA fragments for homozygous (Ho), hemizygous (He) and azygous (Az, segregated) plants. The nucleotide sequence of the amplification fragments was confirmed. (C) Appearance of wild-type (segregated azygous plants) and dcl1a-Ac plants at the indicated developmental stages. (D) Accumulation of Nascent polypeptide- associated complex transcripts in dcl1a-Ac plants. Differences were not statistically significant (ANOVA test). A WT (TN67) WT(Az) dcl1a-Ac#2 B 12 ** Lesion area (%) 8 4 0 TN67 Az dcl1a-Ac#2 TN67 Az dcl1a-Ac#2 0,015 M. oryzae biomass (ng) ** 0,01 0,005 0 TN67 Az dcl1a-Ac#2 Fig. S2. Susceptibility of dcl1a-Ac#2 plants to infection by the fungal pathogen M. oryzae. (A) Disease symptoms at 7 days post-inoculation (dpi) with M. oryzae spores (1 x 105 spores/ml). (B) Lesion area of M. oryzae-infected leaves and quantification of M. oryzae DNA by qPCR, at 7 dpi. (**, p ≤ 0.01 by ANOVA). 180 ** Relative expression (x103) 130 80 30 16 WT(Az) 12 dcl1a-Ac 8 4 0 DCL1a DCL1a DCL1b DCL1b DCL2 DCL2a/b DCL3a DCL3a DCL3b DCL4 DCL4 DCL3b Fig. S3. Expression of OsDCL genes in wild-type and dcl1a-Ac plants grown under normal conditions. The expression of OsDCL1a (Os03g02970), OsDCL1b (Os06g25250), OsDCL2a/b (Os03g38740/Os09g14610), OsDCL3a (Os01g68120), OsDCL4 (Os04g43050) and OsDCL5 (Os10g34430) was examined by RT-qPCR. Data are mean ± SD of 3 biological replicates, each with 12 plants per genotype. (**, P ≤ 0.01 by ANOVA). Except for DCL1a, no significant differences were observed in the expression of other OsDCL genes between wild-type and dcl1a-Ac plants. Fig. S4. Distribution of differentially expressed genes in dcl1a-Ac plants (Osdcl1a-Ac#1 mutant) according to their molecular function determined by using AgriGO. The following parameters were used: (1) Fisher’s exact test with Bonferroni multiple comparison correction; (2) significance level α= 0.05, and the Oryza sativa NCBI database A NB dcl1a-IR 3000 * Chl a mg Chla+b /g FW) * Chl b 2000 Mock 1000 0 WT 0 dcl1a-IR 0 WT 0 dcl1a-IR 0 NB dcl1a-IR Mock MV 300 WT mg Car/g FW 200 dcl1a-IR MV 100 0 WT 1dcl1a-IR WT2 dcl1a-IR Mock MV B WT dcl1a-IR WT (NB) + H2O2 Fig. S5. Effect of methyl viologen on chlorophylls and carotenoids, and detection of O2•- in dcl1a-IR plants. (A) Leaves of 3-week-old knock-down DCL11a (dcl1a-IR) and wild- type (Nipponbare) plants were treated with methyl viologen (MV) for 72 h. Right panels: quantification of chorophyls (Chla+Chlb) and carotenoids (Car) in mock-inoculated and MV-treated wild-type and dcl1a-IR plants. Data shown correspond to wild-type and dcl1a- IR plants (*, P ≤ 0.05 by ANOVA). (B) Detection of superoxide ion (O2•-) by nitroblue tetrazolium (NBT) staining in leaves of wild-type and dcl1a-IR plants. As a control, leaves were treated with H2O2 for 6 h. Bars = 250 µm. A 0,12 pre-miR398 Relative expression (x103) 0,08 0,04 ** 0 WT(Az)24hdcl1a-Ac B Relative expression (x103) 80 SOD2 ** 60 40 20 0 WT(Az)24hdcl1a-Ac Fig. S6. Accumulation of miR398 precursor (A) and miR398-targeted OsSOD2 (Os07g46990) transcripts (B). Transcript levels were determined by RT-qPCR (**, P ≤ 0.01 by ANOVA).