Ustilaginoidea virens‐secreted effector Uv1809 suppresses rice immunity by enhancing OsSRT2‐mediated histone deacetylation

Summary Rice false smut caused by Ustilaginoidea virens is a devastating rice (Oryza sativa) disease worldwide. However, the molecular mechanisms underlying U. virens–rice interactions are largely unknown. In this study, we identified a secreted protein, Uv1809, as a key virulence factor. Heterologous expression of Uv1809 in rice enhanced susceptibility to rice false smut and bacterial blight. Host‐induced gene silencing of Uv1809 in rice enhanced resistance to U. virens, suggesting that Uv1809 inhibits rice immunity and promotes infection by U. virens. Uv1809 suppresses rice immunity by targeting and enhancing rice histone deacetylase OsSRT2‐mediated histone deacetylation, thereby reducing H4K5ac and H4K8ac levels and interfering with the transcriptional activation of defence genes. CRISPR‐Cas9 edited ossrt2 mutants showed no adverse effects in terms of growth and yield but displayed broad‐spectrum resistance to rice pathogens, revealing a potentially valuable genetic resource for breeding disease resistance. Our study provides insight into defence mechanisms against plant pathogens that inactivate plant immunity at the epigenetic level.


Introduction
Plants are constantly under attack from phytopathogens in the environment.In response to these pathogens, plants have evolved a unique innate immune system (Dou and Zhou, 2012).The plant immune system consists of two layers: PTI (PAMPtriggered immunity) triggered by PAMPs (pathogen-associated molecular patterns) and ETI (effector-triggered immunity) triggered by effector proteins (Dangl et al., 2013;Wang et al., 2020a).PTI is an early defence response in plants that can effectively prevent infection from a variety of phytopathogens (Dangl et al., 2013).PTI is activated by pattern recognition receptors (PRRs) that specifically recognize PAMPs (Couto and Zipfel, 2016).However, phytopathogens secrete virulence effectors into host cells that can block the recognition of PAMPs by PRRs, interfere with host PTI, and inhibit basal defence to successfully infect the host via effector-triggered susceptibility (Dodds and Rathjen, 2010;Jiang et al., 2020;Jones and Dangl, 2006).Several host protein kinases, transcription factors, enzymes, and protein complexes involved in plant basal immunity have been identified as targets of fungal effectors, which interfere with these immune-related target proteins and complexes using different strategies to suppress plant immunity and promote infection (Darino et al., 2021;Du et al., 2021;Fukada et al., 2021;Li et al., 2018;Tariqjaveed et al., 2021;Yang et al., 2021).
Epigenetic modifications such as histone methylation and acetylation affect the chromatin state and transcriptional regulation and are common in most organisms (Alvarez et al., 2010;Berger, 2007).Histone acetylation (Kac), one of the beststudied post-translational modifications (PTMs), plays an important role in regulating protein function and gene expression (Xu et al., 2021).Histone deacetylases (HDACs) and acetyltransferases (HATs) maintain the homeostasis of acetyl groups on lysine residues of histone proteins.HATs and HDACs are also involved in regulating plant immunity.In Arabidopsis thaliana (Arabidopsis), the histone acetyltransferase ELP3 (ELONGATOR COMPLEX SUBUNIT 3) positively regulates plant immunity by promoting the expression of defence-related genes (DeFraia et al., 2013).The histone deacetylase HDA19 activates the jasmonic acid/ ethylene (JA/ET) signalling pathway by promoting the expression of ERF1 (ETHYLENE RESPONSE FACTOR 1) to enhance resistance against black spot disease.In addition, HDA19 interacts with the transcription factors WRKY38 and WRKY62 and inhibits their transcriptional activity, thereby positively regulating salicylic acid (SA) synthesis and enhancing disease resistance (Choi et al., 2012;Kim et al., 2008).HDA19 regulates PR (pathogen resistance) gene expression by mediating histone acetylation (Zhou et al., 2005).Similarly, HDA6 mediates histone acetylation to repress the expression of PR genes and inhibit SA biosynthesis by directly controlling the expression of CBP60g (CAM-BINDING PROTEIN 60-LIKE G) and SARD1 (SAR DEFICIENT 1) during pathogen infection (Wang et al., 2017;Wu et al., 2021a).The NAD +dependent Sirtuin family histone deacetylase SRT2 (SIRTUIN 2) represses the expression of the SA biosynthesis genes SID2 (SALICYLIC ACID INDUCTION DEFICIENT 2), PAD4 (PHYTOALEXIN DEFICIENT 4) and EDS5 (ENHANCED DISEASE SUSCEPTIBILITY 5), and SRT2 negatively regulates plant resistance to Pseudomonas syringae (Wang et al., 2010).In the SA signalling pathway, HAC1/ 5 (HISTONE ACETYLTRANSFERASE OF THE CBP FAMILY 1/5) is recruited by a TGACG-binding (TGA) transcription factor to form a coactivator complex with NPR1 (NONEXPRESSER OF PR GENES 1) that activates PR gene transcription via histone acetylationmediated epigenetic reprogramming (Jin et al., 2018).In rice (Oryza sativa), HDT701 (Histone Deacetylase 701) negatively regulates resistance to rice blast by modulating histone H4 acetylation to affect the expression of disease-resistance genes (Ding et al., 2012).The histone deacetylases HDA705 and HDA701 negatively regulate resistance to rice blast, rice false smut (RFS), and bacterial blight (Chen et al., 2021(Chen et al., , 2022a)).These previous studies indicate that HATs and HDACs modulate histone acetylation to affect plant immunity.
RFS, one of the most devastating diseases affecting rice production worldwide, is caused by Ustilaginoidea virens infection (Fan et al., 2020;Sun et al., 2020).RFS not only reduces yield but also threatens animal and human health by producing cyclic peptide mycotoxins (Hu et al., 2019).Over 421 effectors are predicted to exist in U. virens based on genomic analysis (Zhang et al., 2014(Zhang et al., , 2021)).Recent studies have demonstrated that a number of these effectors play critical roles in infection, including SCRE1, SCRE4, SCRE6, SGP1, UvCBP1, UvPr1a, and UvSec117 (Chen et al., 2022a,b;Li et al., 2021a;Qiu et al., 2022;Song et al., 2021;Yang et al., 2022;Zhang et al., 2020;Zheng et al., 2022).For example, the effector SCRE6 acts as a tyrosine phosphatase that dephosphorylates the negative defence regulator OsMPK6 (MAP Kinase) to enhance its accumulation, thereby inhibiting rice immunity (Zheng et al., 2022).However, knowledge about the molecular mechanisms underlying the suppression of rice immunity by individual U. virens effectors is limited.
In this study, we identified Uv1809 as a key effector targeting the rice histone deacetylase OsSRT2, which is a negative regulator of innate immunity against rice pathogens.Uv1809 enhanced OsSRT2-mediated histone deacetylation, thereby reducing H4K5ac and H4K8ac levels in the host and interfering with the activation of defence gene transcription.Our study reveals a novel counter-defence mechanism whereby plant pathogens secrete effectors to interfere with host epigenetic modifications and thereby inactivate host defence responses.

Results
Uv1809 is required for full virulence of U. virens Based on transcriptome data from rice spikelets infected by U. virens (Tang et al., 2021), we found that Uv8b_ 1809 (renamed Uv1809 in this work) was significantly upregulated in the early stage of infection.RT-qPCR analysis suggested that Uv1809 expression is significantly upregulated at 3 days post inoculation (dpi) (Figure 1a), suggesting that Uv1809 plays an important role in the interaction between U. virens and rice.Uv1809 encodes a 391-aa protein, which contains a signal peptide (SP) and is functionally annotated as a hypothetical protein (Figure S1a).Phylogenetic tree analysis showed that Uv1809 homologues are widely distributed in fungi, but no homologous proteins were identified in animals or plants (Figure S1b).Notably, the C end of Uv1809 contains numerous repeat sequences (Figure S1c).Therefore, we generated Uv1809 knockout mutants (ΔUv1809-1 and ΔUv1809-3) and the complementation strain CΔUv1809-1 using Agrobacterium tumefaciens-mediated transformation (ATMT).Transformants were identified by PCR analysis and Southern blot (Figure S2ad).Pathogenicity in the susceptible rice cv.'Wanxian-98' was tested, and the virulence of ΔUv1809 mutants was significantly reduced compared to the wild-type strain HWD-2 and the complementation strain (Figure 1b, c).Compared with the wildtype strain HWD-2 and the complementation strain, ΔUv1809 mutants displayed no significant differences in mycelial growth rate and conidiation (Figure S3a-c).These results show that Uv1809 is required for full virulence of U. virens.

The signal peptide of Uv1809 is essential for its secretion and function
To investigate the functionality of the SP of Uv1809, we used a yeast secretion system to verify the secretion activity of the protein (Jacobs et al., 1997).The sequence encoding the 22-aa SP was ligated with the pSUC2 vector and then transformed into yeast strain YKT12.Uv1809SP and positive control Avr1bSP transformants could grow on YPRAA, but the negative control pSUC2 transformants and the untransformed YTK12 strain could not grow on YPRAA medium.The Uv1809SP and Avr1bSP transformants also secreted invertase and turned the TTC solution red, while the negative control pSUC2 transformants and the untransformed YTK12 strain could not change the colour of the TTC solution (Figure 1d).These results indicate that the SP of Uv1809 is a functional secretion signal, and Uv1809 is a secreted protein in U. virens.
To assess whether the SP of Uv1809 is related to the virulence of U. virens, we then generated a Uv1809 construct lacking the SP (Uv1809 DSP ) and introduced it into the ΔUv1809-1 mutant.The resulting CΔUv1809 DSP -1 transformant strain showed a similar virulence to that of the ΔUv1809-1 mutant and exhibited significantly reduced virulence compared to the wild-type strain HWD-2 and the complementation strain (Figure 1b, c).These results indicate that the SP of Uv1809 plays a key role in the virulence of U. virens.

Uv1809 suppresses Bax-induced cell death in Nicotiana benthamiana
To characterize the role of Uv1809 in host immune responses, a pVX-Uv1809 construct was generated and introduced into Agrobacterium strain EHA105.In N. benthamiana leaves infiltrated with Agrobacterium expressing Uv1809 alone, no cell death was observed (Figure 1e).Under the same conditions, cell death was observed on N. benthamiana leaves infiltrated with Agrobacterium expressing the mouse proapoptotic protein Bax.However, the cell death induced by Bax was suppressed when Uv1809 and Bax proteins were co-expressed (Figure 1e).As a negative control, expression of the empty vector pVX (GFP) and Bax did not abolish Bax-dependent cell death (Figure 1e).To delineate the region within Uv1809 that mediates the inhibition of Bax-induced cell death, we generated constructs expressing either the N-terminal (Uv1809 D246-391 ) or C-terminal (Uv1809 D23-245 ) portion of Uv1809 and co-infiltrated them with Bax in N. benthamiana.Only the construct encoding the C-terminal portion of Uv1809 (Uv1809 D23-245 ) showed an ability to suppress Bax-induced cell death comparable to full-length Uv1809 (Figure 1e).These results suggest that Uv1809 suppresses the cell death induced by Bax, possibly by interfering with the plant immune response.

Uv1809 is an intracellular effector
To further determine the subcellular localization of Uv1809 in N. benthamiana leaves, the pCNG-Uv1809 construct was generated and introduced into Agrobacterium strain EHA105.In N. benthamiana leaves infiltrated with Agrobacterium expressing the Uv1809-GFP fusion protein, fluorescence microscopy detected GFP fluorescence in the nucleus and the cytoplasm, and the fluorescence signal will gather in some positions (Figure 1f, Figure S4), which is different from the localization of Uv1809 DSP -GFP and empty GFP (Figure S4).After plasmolysis, the GFP fluorescence signal in the intracellular, not in apoplastic (Figure S4).These results suggests that Uv1809 is an intracellular effector.
To investigate whether Uv1809 is secreted and translocated into plant cells, Uv1809 carrying a nuclear localization signal (NLS) was ectopically expressed as a fusion with GFP in Magnaporthe oryzae strain P131 driven by the RP27 promoter.The engineered M. oryzae strains were inoculated onto detached barley leaves.Green fluorescence was detectable inside the invasive hyphae of GFP-labelled M. oryzae by microscopy at 30 h after inoculation (Figure S5).The majority of barley epidermal cells infected by Uv1809-GFP-NLS-transformed M. oryzae exhibited green fluorescence in the nuclei at 30 h after inoculation (Figure S5).When barley leaf sheath was inoculated with the engineered M. oryzae strain ectopically expressing Uv1809-GFP, green fluorescence was clearly observed to accumulate in the BIC at 30 h post inoculation (hpi).In contrast, green fluorescence was only discernible in invasive hyphae after barley sheaths were infected by GFPexpressing M. oryzae (Figure S5).These results indicate that Uv1809 can be secreted into plant cells.
Since Uv1809-GFP was present in the nucleus, we looked for nuclear localization sequences (NLSs) in Uv1809 using the tool NLStradamus (https://www.novopro.cn/tools/nls-signalprediction.html)(Nguyen et al., 2009).Uv1809 contained one potential NLS at aa 336-367 (KDSKDNKDSKANKDSKDNKDS-RAVNKPEGKAAR).Accordingly, we expressed the Uv1809 336-367 -GFP fusion protein in N. benthamiana leaves, the subcellular localization result showed the putative NLS was functional which succeeded in localizing GFP to the nucleus (Figure S4).We also tested the subcellular localization of Uv1809 D246-391 and Uv1809 D23-245 when transgenically expressed in N. benthamiana leaves.The Uv1809 D246-391 -GFP variant showed an almost complete loss of nuclear signal, while the Uv1809 D23-245 -GFP variant displayed strong GFP fluorescence in the nucleus (Figure S4).These results indicate that aa 246-391 of Uv1809 plays an important role in regulating the nuclear localization of Uv1809.

Heterologous expression of Uv1809 increases susceptibility to rice pathogens
To further determine how Uv1809 suppresses plant immune responses, we generated transgenic rice lines heterologously expressing Uv1809.These 35S-Uv1809 transgenic rice lines grew similarly to the empty vector control transgenic rice lines (35S-EV) (Figure S6).However, two independent 35S-Uv1809 transgenic rice lines had more rice smut balls than the 35S-EV transgenic rice lines after inoculation with U. virens strain HWD-2 (Figure 2a).Likewise, 35S-Uv1809 transgenic rice lines were more susceptible to infection by Xoo strain PXO99, with lesion lengths in 35S-Uv1809 transgenic rice lines being about two times longer than those in 35S-EV transgenic rice lines at 14 dpi (Figure 2b).We then evaluated the 35S-Uv1809 transgenic rice lines for resistance against the rice blast fungus M. oryzae strain ZB-25.The 35S-Uv1809 transgenic rice lines were more susceptible to M. oryzae than the 35S-EV transgenic rice lines at 7 dpi (Figure 2c).We also found that the expression of defence-related genes OsPR1b, OsPAL1, OsWRKY13, and OsAOS2 was significantly downregulated in the 35S-Uv1809 transgenic rice plants (Figure 2d), suggesting that Uv1809 leads to the suppression of the immune response.Taken together, these results indicate that heterologous expression of Uv1809 inhibits the host's immune response, thereby enhancing the susceptibility of rice plants to pathogens.
Engineering RFS-resistant rice via HIGS of Uv1809 We employed host-induced gene silencing (HIGS) by generating transgenic japonica rice cv.Nipponbare plants expressing an RNA interference (RNAi) construct against the Uv1809 transcript (Uv1809HIGS; Figure S7).At 25 dpi with U. virens, we scored an average of 7 smut balls per panicle on the spikelets of rice plants transformed with the empty vector (EV).By contrast, two   independent Uv1809HIGS transgenic rice lines displayed strong resistance to RFS, with an average of 1 to 3 smut balls per panicle (Figure 2e).To confirm that the resistance to RFS seen in infected Uv1809HIGS transgenic plants is caused by in planta silencing of Uv1809, we collected infected spikelets from T 2 transgenic rice lines (Uv1809HIGS and EV control) at 6 dpi and quantified Uv1809 and UvSec117 transcript levels by RT-qPCR.Indeed, relative Uv1809 transcript levels were much lower in the two Uv1809HIGS transgenic rice lines than in the EV control lines (Figure 2f), while the relative transcript level of the control effector encoding gene UvSec117 was no significant change; this result shows that the Uv1809HIGS transgenic plants resistance to RFS is caused by in planta silencing of Uv1809.
To ascertain whether silencing of Uv1809 in infecting U. virens is mediated by small interfering RNAs (siRNAs) generated by the Uv1809HIGS transgenic rice lines, we sequenced small RNAs in the Uv1809HIGS-L1 transgenic rice lines.The sequencing data indicated that siRNAs mapping to Uv1809 were highly abundant in Uv1809HIGS-L1 transgenic rice lines, accounting for 0.83% of all small RNAs detected in these lines.The siRNAs matching the Uv1809 transcript had a size distribution of between 18 and 30 nucleotides, with 21-nucleotide siRNAs being the most abundant (Figure 2g).In fluorescence in situ hybridization (FISH) assays, fluorescence signal was observed both in rice flower tissue and U. virens infection hyphae in the infected Uv1809HIGS-L1 transgenic rice lines at 6 dpi, whereas no fluorescence signal was detected in the EV control lines (Figure 2h).This result supports the notion that the Uv1809-RNAi vector produces effective siRNAs against Uv1809 in the Uv1809HIGS-L1 transgenic rice lines that are then translocated to fungal cells during infection to reduce Uv1809 transcript levels in invading U. virens hyphae.Together, these results demonstrate that silencing of Uv1809 in transgenic rice plants promotes resistance to U. virens, indicating that Uv1809 is a key virulence effector during infection.

Uv1809 interacts with the rice histone deacetylase OsSRT2
To further characterize its function during plant infection, we performed a yeast two-hybrid screen with Uv1809 as bait against a cDNA library constructed from RNA extracted from U. virensinfected rice spikelets, leading to the identification of 25 putative Uv1809-interacting proteins (Table S1).We showed that the rice histone deacetylase OsSRT2 is a target of Uv1809 and confirmed the interaction between Uv1809 and full-length OsSRT2 by yeast two-hybrid (Figure 3a).To validate the interaction in vivo, we next performed a co-immunoprecipitation (Co-IP) assay on N. benthamiana leaves transiently co-infiltrated with OsSRT2-Flag and Uv1809-GFP.In this assay, Uv1809 was immunoprecipitated by OsSRT2 (Figure 3b).Next, we performed in vitro pull-down assays with recombinant OsSRT2-GST (glutathione S-transferase) and Uv1809 DSP -His (His tag) proteins purified from E. coli.We detected Uv1809 DSP -His from protein samples pulled down with OsSRT2-GST loaded onto glutathione beads (Figure 3c), suggesting that Uv1809 DSP -His and OsSRT2-GST interact in vitro.Finally, we transiently co-infiltrated N. benthamiana leaves with Uv1809-cYFP and OsSRT2-nYFP fusion constructs for bimolecular fluorescence complementation (BiFC) analysis.We observed YFP fluorescence in the nucleus of N. benthamiana leaf epidermal cells (Figure 3d).We then attempted to delineate the interaction interface by testing two Uv1809 fragments (Uv1809 23-245 and Uv1809 246-391 ) for interaction with OsSRT2.Only Uv1809 246-391 fragments interacted with OsSRT2 in Y2H, Co-IP, and BiFC assays (Figure S8a-c).These results suggest that Uv1809 interacts with OsSRT2 in the nucleus of plant cells, Uv1809 246-391 is key region for interacting with OsSRT2.

OsSRT2 negatively regulates broad-spectrum resistance against rice pathogens
To explore the function of OsSRT2 in resistance against U. virens, we determined the transcript levels of OsSRT2 during infection by RT-qPCR.OsSRT2 transcript levels were significantly upregulated during infection, peaking at 6 dpi (Figure S9a).This result suggests that OsSRT2 expression is induced by U. virens infection and that OsSRT2 might negatively regulate rice resistance to U. virens.
To examine OsSRT2 localization in plant cells, we assessed the subcellular localization of an OsSRT2-GFP fusion protein in N. benthamiana leaves.We determined that OsSRT2 is localized to the nucleus (Figure S9b).To explore the function of OsSRT2 in resistance against rice pathogens, we generated OsSRT2 knockout mutants (ossrt2) using Clustered Regularly Interspaced Short Palindromic Repeats/CRISPR Associated Protein 9 (CRISPR/Cas9) gene editing (Figure S10).The morphology of ossrt2 rice lines was similar to that of wild-type ZH11 rice plants (Figure 4a, Figure S11).After inoculation with a mycelium/spore suspension of U. virens strain HWD-2, ossrt2 rice lines produced far fewer smut balls than ZH11 lines (Figure 4b).We then evaluated the ossrt2 rice lines for resistance against the rice blast fungus M. oryzae strain P131 or ZB-25.The ossrt2 rice lines were more resistant to M. oryzae than the ZH11 lines (Figure 4c-f).Inoculation of ossrt2 and ZH11 rice lines with Xoo strain PX099 using the scissor-clipping method produced similar results, with lesions being ~75% shorter on the ossrt2 rice lines than on the ZH11 lines at 14 dpi (Figure 4g).Likewise, ossrt2 rice lines were more resistant to infection by R. solani strain HG81 than ZH11 lines at 3 dpi (Figure 4h).Interestingly, we found that the transcription levels of defence-related genes (OsPR1b, OsPR10a, OsPAL1, and OsSGT1) were significantly increased in the ossrt2 plants after inoculation of U. virens at 1 dpi (Figure 4i), indicating that the upregulation of these genes is associated with OsSRT2.Overall, these results indicate that OsSRT2 negatively regulates rice resistance to multiple diseases.

Uv1809 increases the deacetylation activity of OsSRT2
Rice HDACs, including OsHDT701, regulate histone H3 or H4 acetylation levels (Ding et al., 2012).We therefore examined whether OsSRT2 affected histone deacetylation.Using histones isolated from the spikelets of ossrt2 and ZH11 rice lines, we performed an immunoblot analysis using various anti-H3ac and anti-H4ac antibodies.We established that levels of H4K5ac and H4K8ac are higher in the ossrt2 mutant than in ZH11, while levels of other lysine sites on H3 tested were not increased (Figure 5a).Thus, OsSRT2 mainly regulates histone acetylation levels of H4K5 and H4K8 in rice spikelets.We also used various anti-H3ac and anti-H4ac antibodies to detect the levels of H3ac and H4ac in 35S-EV and 35S-Uv1809 transgenic rice spikelets.H3K9ac, H4K5ac, and H4K8ac levels in 35S-Uv1809 transgenic rice lines were markedly lower than in 35S-EV transgenic rice lines and Nip rice plants (Figure 5b), this result indicated that Uv1809 might repress the H4K5ac and H4K8ac of rice by affecting OsSRT2.To test this hypothesis, we measured the activity of recombinant OsSRT2-GST alone or in the presence of Uv1809 DSP -His, Uv1809 Xiaoyang Chen et al. fluorometric assay.Notably, deacetylation activity in the reaction increased when OsSRT2-GST was incubated with Uv1809 DSP -His or Uv1809 246-391 -His, but not when it was incubated with Uv1809 23-245 -His or HDACi (Figure 5c).This rise in activity was not associated with the effector itself, as recombinant Uv1809 DSP -His, Uv1809 23-245 -His, and Uv1809 246-391 -His showed no deacetylation activity themselves (Figure 5c).We independently observed decreased levels of H4K5 and H4K8 deacetylation activity in vitro by immunoblot analysis of recombinant OsSRT2-GST incubated with purified rice histones.Furthermore, OsSRT2-mediated deacetylation of H4K5ac and H4K8ac was enhanced by co-incubation with recombinant Uv1809 DSP -His or Uv1809 246-391 -His (Figure 5d).In order to test the effect of Uv1809 on the histone deacetylase activity of OsSRT2 in planta, we transiently expressed OsSRT2-Flag, OsSRT2-Flag and Uv1809-GFP, OsSRT2-Flag and GFP, and Uv1809-GFP in N. benthamiana leaves.The nuclear proteins were then extracted and the H4K5 and H4K8 acetylation levels were evaluated.As shown in Figure 5e, OsSRT2-Flag and Uv1809-GFP co-expression reduced the acetylation levels of histones H4K5 and H4K8.In contrast, Uv1809-GFP or GFP alone did not reduce the acetylation levels of H4K5 and H4K8 (Figure 5e).Accordingly, OsSRT2-Flag and Uv1809-GFP co-expression increased the deacetylation activity of these nuclear proteins and were determined using a fluorometric assay (Figure 5f).Taken together, these results indicate that Uv1809 enhances OsSRT2mediated histone deacetylation in vitro and in vivo.
OsSRT2 suppresses rice immunity by modulating H4K5acand H4K8ac-marked genes To identify genes that could be regulated by OsSRT2 in rice spikelets, we performed RNA sequencing (RNA-Seq) to identify genes that were differentially expressed in ossrt2 mutant rice spikelets relative to wild-type ZH11 rice spikelets.Three biological replicates were performed per sample (Figure S12a).A total of 604 and 378 genes were upregulated and downregulated, respectively, to >2-fold (P < 0.05) in ossrt2 compared to wild type (Figure 6a; Table S2).
Gene set enrichment analysis (GSEA) was performed to investigate pathways associated with the ossrt2 mutation (Figure 6b).Regulation of defence response to fungus and regulation of defence response pathways were significantly enriched in our dataset, suggesting that genes in these pathways were upregulated in ossrt2.Furthermore, gene ontology (GO) pathway enrichment of significantly upregulated genes in ossrt2 also revealed that regulation of defence response, regulation of jasmonic acid-mediated signalling pathway, and regulation of response to stimulus pathways were enriched (Figure 6c).To confirm the gene expression patterns, we selected eight genes for RT-qPCR analysis.The expression pattern of each upregulated or downregulated gene was consistent with the RNA-Seq data (Figure S13).Collectively, these results indicate that mutation of OsSRT2 might promote defence pathways in rice spikelets and confer the mutants with resistance to pathogen infections.To further investigate the gene regulation mechanisms of OsSRT2 during infection, we performed ChIP-seq to compare genome-wide profiles of H4K5ac and H4K8ac in rice spikelets from wild-type and ossrt2 plants.For each ChIP-seq experiment, two biological replicates (Figures S12b, S14) were carried out.From the two replicates, 44 392 (38 761 marked genes) H4K5ac and 44 352 (38 628 marked genes) H4K8ac peaks were identified in wild-type rice spikelets.These two histone modifications showed a similar genomic distribution and were highly enriched at the transcription start site (Figure S15a, b).The genomic patterns of H4K5ac and H4K8ac were comparable (Figure S15c, d) and showed a moderately significant connection with gene transcription (Figure S15e).We identified a total of 657 peaks (458 genes) with elevated (>1.5-fold,P < 0.05) H4K5ac levels and 316 peaks (249 genes) with reduced H4K5ac levels (>1.5-fold,P < 0.05) in the mutant (Figure 6d, Table S3).
Comparative analysis of H4K8ac revealed that a total of 1236 peaks (932 genes) showed significant increase in H4K8ac depositions in ossrt2, whereas only 74 peaks (61 genes) showed a substantial decrease in H4K8ac depositions (>1.5-fold,P < 0.05) in srt2 (Figure 6d, Table S4).These findings were consistent with the immunoblotting estimates of H4K5ac and H4K8ac levels in ossrt2 shown in Figure 5a.
To investigate whether H4K5ac and H4K8ac co-regulate a subset of genes in the ossrt2 mutant, the overlap of H4K5ac and H4K8ac peaks (genes) was analysed.The results showed that 58.4% (61.2%) of the peaks (genes) with upregulated H4K5ac overlapped with those of H4K8ac in ossrt2 (Figure 6e), indicating that the ossrt2 mutation had a similar effect on both H4K5ac and H4K8ac deposition.To identify the metabolic pathways regulated by H4K5 and H4K8 hyper-acetylated genes (N = 282) in ossrt2, GO pathway enrichment analysis was performed.We identified 62 GO terms as being associated with H4K5 and H4K8 hyperacetylated genes, and these terms were involved with a variety of biological processes, including defence-related pathways such as plant-type hypersensitive response and modulation by virus of host process (Figure 6f, Table S5).
To investigate the relationship between the transcriptome and histone acetylation in rice spikelets, we first examined the correlation between the dynamic changes in H4K5/8ac and gene transcription in ossrt2 mutants.As expected, a moderate level of positive correlation (r = 0.30/0.31)between H4K5/8ac and gene expression changes was observed in the mutant (Figure 7a), which was in line with their moderately significant connection with gene transcription (Figure S15e).Furthermore, those upregulated genes displayed concurrent increases in both H4K5ac and H4K8ac in the ossrt2 mutant (Figure 7b) compared to the wild type.To identify the genes regulated by H4K5/8ac hyper-acetylation and transcriptional upregulation in the mutant, we analysed the overlap between H4K5/8ac hyper-acetylation and transcriptionally upregulated genes.As shown in our Venn diagram, 84 genes were both transcriptionally upregulated and H4K5/8ac hyper-acetylated in ossrt2 mutants (Figure 7c, Table S6).To establish the pathways in which these genes were enriched, we performed GO pathway analysis and identified 56 GO terms.Among the biological processes highlighted by GO analysis, plant defence-related pathways were found to be significant, including hyperosmotic response, activation of innate immune response, and plant-type hypersensitive response (Figure 7d, Table S7).We analysed precipitated chromatin fragments by ChIP-qPCR with primer sets designed to amplify the promoter regions of selected genes.Compared to the wild type, H4K5ac and H4K8ac levels were significantly increased at the promoters of these tested genes in the ossrt2 mutant.Therefore, increased H4K5ac or H4K8ac levels on histones at the promoter regions of rice defence-related genes lead to more active transcription (Figure 7e).Collectively, our results imply that the deacetylation of H4K5/8 by OsSRT2 suppresses the transcription of plant defence-related genes, leading to a decrease in rice immunity.

Discussion
Previous studies have revealed that acetylation plays an essential role in plant defences during plant-pathogen interactions (G omez-D ıaz et al., 2012;Parker et al., 2022).Recent studies have shown that fungal effectors can regulate plant immunity by interacting with HATs and HDACs, thereby affecting histone acetylation to regulate the expression of defence genes and interfere with plant immunity (Kong et al., 2017;Li et al., 2018).In this study, we showed that the secreted effector Uv1809 from U. virens physically interacts with histone deacetylase OsSRT2, enhancing its histone deacetylase activity and reducing H4K5 and H4K8 acetylation levels at the promoter of defencerelated genes.Subsequently, these lower H4K5 and H4K8 acetylation levels result in lower expression of defence genes and compromise rice resistance to pathogens (Figure 8).Interestingly, we established that Uv1809 246-391 localizes to the nucleus and contains numerous repetitive sequences.This region of Uv1809 is key for inhibiting Bax-induced cell death, interacting with OsSRT2, and enhancing its enzyme activity.Thus, the function of Uv1809 246-391 in the nucleus merits further analysis.The use of plant immunity to improve disease resistance is the most cost-effective and environmentally safe means of ensuring food security.In general, to exploit plant immunity, key genes involved in immune regulation need to be identified.In recent years, the development of cutting-edge omics analysis and CRISPR/Cas9-mediated gene editing technology has promoted the discovery of disease-resistance genes and allowed for the editing of disease-susceptibility genes, thus accelerating the development of sustainable and robust agricultural production.In rice, 18 HDACs have been identified, including 14 RPD3/ HDA1 (Reduced Potassium Dependence 3/Histone Deacetylase 1) members, two SIR2 (Silent Information Regulator 2) members, and two HD (Histone Deacetylase 2) members (Chen et al., 2020a).Among them, HDT701, HDA701, and HDA705 have been reported to negatively regulate rice blast, RFS, and/or bacterial blight disease resistance (Chen et al., 2021(Chen et al., , 2022a;;Ding et al., 2012).Sirtuin-like deacetylases are crucial regulators of a variety of biological processes, such as metabolism, organ development, and plant responses to oxidative and abiotic stress (Tang et al., 2022;Zhang et al., 2017Zhang et al., , 2018)).Here, we showed that this novel SRT2 negatively regulates broad-spectrum resistance against multiple rice pathogens, making it a prime candidate gene for breeding programs aimed at increasing resistance to rice pathogens.
In the osssrt2 mutants, some defence-related genes were upregulated (Figure S13), including those encoding transcription factors, receptor-like kinases, and disease-resistance proteins.Overexpression of the transcription factor OsMYB30 (Os09g26170) enhances resistance to rice blast, while knockdown of OsMYB30 decreases resistance to rice blast, indicating that OsMYB30 positively regulates immune responses (Li et al., 2020).Overexpression of the transcription factor OsWRKY89 (Os11g02520) led to increased salicylic acid levels, which enhanced resistance to rice blast (Wang et al., 2007).OsPR1b (Os01g28450), a key PR gene in rice, was significantly upregulated in the osssrt2 mutant (Mitsuhara et al., 2008).A respiratory burst oxidase homologue (Rboh) functions in the production of reactive oxygen species, and OsRbohD (Os05g38980) participates in the immune response process (Yoshiaki et al., 2005).Thus, genome editing of OsSRT2 directly or indirectly upregulates the expression of key defence-related genes (such as OsMYB30, OsWRKY89, OsRbohD, and OsPR1b) to enhance resistance to rice pathogens.
Chitin is a major component of the fungal cell wall and acts as a PAMP elicitor that triggers PTI in several plant species (Tariqjaveed et al., 2021).PAMP elicitors trigger a series of defence responses, including ROS generation and the expression of defence-related genes.Interestingly, we found that the ROS generation was significantly increased after chitin treatment in the ossrt2 plants relative to ZH11 plants (Figure S16), OsRbohD was significantly upregulated in the osssrt2 mutant, revealing that OsSRT2 affects ROS generation by regulating the expression of OsRbohD.OsSGT1 functions through the RAR1-SGT1-HSP90 cytosolic defensome complex and mediates accumulation of the ROS generated by NADPH oxidases (Azevedo et al., 2002).So, we speculate that the OsSRT2 function on key PTI-related components.Intriguingly, the expression of defence-related genes OsPR1b, OsPAL1, OsWRKY13, and OsAOS2 was significantly downregulated in the 35S-Uv1809 transgenic rice plants during infection by U. virens.Accordingly, the transcription levels of defence-related genes (OsPR1b, OsPR10a, OsPAL1, and OsSGT1) were significantly increased in the ossrt2 plants.Our data support the hypothesis that Uv1809 suppresses the activation of defencerelated genes by affecting OsSRT2 during infection of U. virens, thereby compromising plant immunity.
In this study, we established that OsSRT2 suppresses rice immunity by modulating H4K5ac-and H4K8ac-marked genes.Protein phosphorylation by protein kinases transmits defence signals to downstream proteins, such as WRKY transcription factors, which have been extensively characterized in plant defence responses (Eulgem and Somssich, 2007;Lin et al., 2022;Park et al., 2012;Wani et al., 2021).We also discovered that the transcription and histone acetylation levels of defencerelated genes, such as WRKY50 (Os11g02540) and protein kinases (Os11g10710 and Os08g28890), were increased in srt2 mutant plants, implying that SRT2 may regulate plant resistance to pathogens through a specific group of defence-related genes.The systemic acquired resistance (SAR) is effective against a broad range of pathogens, which is accompanied by the de-novo synthesis of PR proteins.In Arabidopsis, AtWRKY50 as an activator of PR1 gene expression, PR1 is a SA inducible marker gene for SAR (Hussain et al., 2018).These findings show that SRT2 negatively regulates innate immunity in rice by modulating the levels of histone H4 acetylation in defence-related genes.
Chromatin structure and transcriptional activity of genes are regulated by PTMs on histones such as methylation, acetylation, ubiquitination, and phosphorylation (Alvarez et al., 2010).Based on previous studies, PTMs of histones, including 2hydroxyisobutyrylation and crotonylation, also play important roles in gene expression (Chen et al., 2021).OsSRT2 is an "eraser" of these new PTMs, which might regulate the expression of defence-related genes.In Arabidopsis, the acetylation level of different histone sites was not significantly different between srt2-1 mutants and wild-type plants, but AtSRT2 can specifically deacetylate lysine residues at H4K8ac, H3K14ac, and H3K9ac loci in vitro (Tang et al., 2022).Interestingly, except H4K5ac and H4K18ac, heterologous expression of Uv1809 also strongly reduced the level of H3K9ac in rice (Figure 5b).Given that Sirtuin has been reported to possess H3K9ac deacetylation activity in other species (Tang et al., 2022), it's plausible that this activity might not be as robust in rice panicle.Nevertheless, it's not excluded that Uv1809 might potentially modulate other histone deacetylases to some extent.However, during infection by U. virens, OsSRT2 may regulate other histone acetylation sites to modulate the expression of defence-related genes, thereby negatively regulating rice immunity.Regardless, the insights gleaned from our study indicate that Uv1809 specifically enhances the activity of OsSRT2, which predominantly targets the H4K5ac and H4K8ac histone marks in rice panicle tissue.
The development of durable and broad-spectrum resistance is an economical and sustainable approach to control crop diseases for agricultural production.Recently, the RNAi-based approach host-induced gene silencing (HIGS) was developed as an alternative strategy to control fungal diseases.In this strategy, siRNAs targeting selected genes of the invading pathogen are produced by transgenic host plants to silence fungal genes during infection (Chen et al., 2022c;Li et al., 2021a,b;Machado et al., 2018;Nowara et al., 2010).For example, HIGS of FgChs3b, FgCYP51, FgSGE1, FgFGP1, or FgSTE12 conferred resistance to Fusarium head blight fungus Fusarium graminearum in wheat (Triticum aestivum) (Cheng et al., 2015;Koch et al., 2013;Wang et al., 2020a,b).Similarly, HIGS of PsCPK1 or PsFUZ7 enhanced wheat resistance to the stripe rust fungus Puccinia striiformis f. sp.tritici (Qi et al., 2018;Zhu et al., 2017).Other successful examples of engineered resistance have also been demonstrated in rice against rice blast or RFS (Chen et al., 2022c;Guo et al., 2019).
Our study now adds HIGS against Uv1809 as an effective strategy to reduce the severity of U. virens infection in rice, providing a reference for the development of stable transgenic plants using a HIGS-based strategy to enhance rice resistance.
In summary, we identified an important secreted effector, Uv1809, which interacts with the histone deacetylase OsSRT2.Uv1809 enhances OsSRT2-mediated histone deacetylation, thus modulating the expression of defence-related genes in the host.Our study demonstrates that a fungal effector from U. virens targets a HDAC to suppress plant immunity.Strikingly, stable transgenic rice lines expressing Uv1809-silencing RNAs and ossrt2 mutant rice lines showed increased broad-spectrum resistance to U. virens and other rice pathogens, pointing to novel strategies for protecting rice crops.

Experimental procedures
Plant materials and bacterial and fungal strains Oryza sativa cvs.' 'Zhonghua 11' [ZH11] and 'Nipponbare', and Nicotiana benthamiana plants were grown in a glasshouse.Transgenic rice plants with host-induced gene silencing (HIGS) of Uv1809 or heterologous expression of Uv1809 (35S-Uv1809) or the empty vector (35S-EV) were generated in the Nipponbare background.The ossrt2 knockout rice lines were generated in the ZH11 background using CRISPR/Cas9 technology.Rice transformation was performed by Wuhan Tianwen Biotechnology Co., Ltd (Wuhan, China).

Fungal inoculation assays
Rice plants at the booting stage were inoculated with U. virens HWD-2 mycelial/spore suspensions (1 9 10 6 conidia/mL) using a syringe, and the number of false smut balls was counted 21-25 dpi (Chen et al., 2020b).For spray-inoculations, the leaves of four-week-old rice plants were sprayed with M. oryzae P131 or ZB-25 conidial suspensions (1 9 10 6 conidia/mL) in 0.025% (v/v) Tween 20, and the disease lesions were examined at 7 dpi.For spot-inoculations, the leaves of 4-week-old plants were punctured with a needle, and a droplet of 10 lL M. oryzae P131 or ZB-25 conidial suspension (1 9 10 5 conidia/mL) was placed at each wound site.Rice plants were inoculated with Xoo PXO99 at the booting stage using the leaf clipping method, and the lesion lengths were scored at 14 dpi.The leaves of four-weekold rice plants were inoculated with R. solani HG81, and the lesion areas were scored at 3 dpi and calculated using ImageJ software.All inoculation experiments were repeated three times.

Gene deletion and complementation in Uv1809
U. virens transformation was carried out as previously described (Chen et al., 2020b).Briefly, about 1.2-kb of the downstream and upstream flanking sequences of Uv1809 were ligated with the knockout vector pGKO.An approximately 3.5-kb complementation fragment was cloned into the complementation vector pNeo3300III.The EHA105 Agrobacterium strain carrying the pGKO-Uv1809, pNeo3300III-Uv1809 ΔSP , or pNeo3300III-Uv1809 vectors was transformed with the ATMT method.Transformed strains were confirmed by PCR analysis and Southern blot.Southern blot was performed using the Amersham Gene Images Alkphos Direct Labelling and Detection System (GE Healthcare, Little Chalfont, UK).

Small RNA sequencing and FISH assays
Total RNA was extracted from the infected spikelets of Uv1809HIGS-L1 transgenic rice lines and then used for small RNA-Seq on an Illumina HiSeq platform by Wuhan IGENEBOOK Biotechnology Co., Ltd (Wuhan, China).Unique reads were mapped to the corresponding target fragments of Uv1809.For FISH assays, a 21-bp oligonucleotide probe (GTCAATCT-CAGCCCTGGCTCG), which was the most common siRNA sequence identified by small RNA-Seq, was labelled using digoxin ª 2023 The Authors.Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 22, 148-164 (DIG).Infected spikelets at 6 dpi were placed in fixative solution (Servicebio, Wuhan, China) for 6 h and embedded in paraffin.Paraffin sections were dewaxed with a xylene and rehydrated with a gradient alcohol, then digested with protease K (20 lg/ mL) at 37 °C for 30 min.After washing with PBS, the prehybridization solution was added to the mixture and incubated at 37 °C for 1 h.After incubation, sections were poured the prehybridization solution, dropped the hybridization solution containing DIG-labelled probe and hybridized at 37 °C for 16 h.After hybridization, it was washed and sealed, and then, anti-digoxin labelled peroxidase (anti-DIG-HRP) was added to incubate at 37 °C for 1 h, then wash three times with PBS and add Alexa Fluor 594 Tyramide (ThermoFisher Scientific, MA, USA).The fluorescence of the DIG-labelled siRNA was analysed with a Zeiss LSM 510 Meta confocal microscope (LSCM) (Carl Zeiss, Jena, Germany).DAPI (4 0 -6-diamidino-2-phenylindole) was used for nuclear staining.

Agrobacterium-mediated infiltration assays and confocal microscopy
The cDNA sequence of Uv1809 was ligated with pVX to obtain the recombinant vector pVX-Uv1809.pVX-Uv1809, pVX-Bax, and empty pVX were electroporated into Agrobacterium tumefaciens (Agrobacterium) strain EHA105.Agrobacterium cultures containing either the pVX-Uv1809 or pVX vectors were re-suspended in infiltration buffer (10 mM MgCl 2 , 0.5 mM MES and 0.2 mM acetylacetone) and infiltrated into the abaxial side of 4-week-old N. benthamiana leaves.24 h later, Agrobacterium carrying the pVX-Bax vector was infiltrated at the same position, and programmed cell death (PCD) was assessed at 4 dpi.For subcellular localization assays, Agrobacterium carrying the variant constructs pCNG-Uv1809 or pCNG-OsSRT2 was infiltrated into N. benthamiana leaves, and the fluorescence signal of the infiltrated areas was analysed with a LSCM at 2 dpi.After N. benthamiana leaves were treated with 1 M NaCl for 10 min, plasmolysis was observed.

Yeast two-hybrid (Y2H) assay
The cDNA sequence of Uv1809 ΔSP (without the SP) was ligated with pGBKT7 to generate the bait vector pGBKT7-Uv1809 ΔSP .The mRNA from rice spikelets was used to construct a cDNA library in the prey vector pGADT7.Library screening for Uv1809 interactors was carried out as described previously (Chen et al., 2020b).For interaction analysis of Uv1809 and OsSRT2 in yeast, the vectors pGBKT7-Uv1809 ΔSP and pGADT7-OsSRT2, pGBKT7-Uv1809 23-245 and pGADT7-OsSRT2, or pGBKT7-Uv1809 246-391 and pGADT7-OsSRT2 were co-transformed into the Y2H Gold strain.Positive clones were selected on SD-Trp-Leu-His medium, and further confirmed on SD-Trp-Leu-His-Ade medium containing X-a-Gal.

Co-immunoprecipitation (Co-IP) assays
The cDNA sequences of Uv1809 and OsSRT2 were ligated with the vectors pCNG (GFP tags) and pCNF (Flag tags) and then introduced individually into Agrobacterium strain EHA105.Agrobacterium carrying the pCNG-Uv1809 and pCNF-OsSRT2 vectors were co-infiltrated into N. benthamiana leaves.Total proteins were extracted from N. benthamiana leaves at 2 dpi and incubated with Anti-Flag M2 affinity gel (Yeasen Biotech, Shanghai, China).Proteins eluted from gels were analysed by immunoblotting with anti-GFP or anti-Flag antibodies (ABclonal, Wuhan, China).

GST pull-down assays
The cDNA sequences of Uv1809 ΔSP and OsSRT2 were ligated with the vectors pET32a (His tags) and pGEX4T-2 (GST tags) and then introduced individually into Escherichia coli BL21(DE3) cells.The OsSRT2-GST fusion protein was extracted from E. coli cells and incubated with 100 lL glutathione-agarose beads (Yeasen Biotech, Shanghai, China) at 4 °C for 4 h with shaking.After centrifugation at 4 °C, the beads were collected and washed with phosphate buffered saline (PBS) three times.Beads were then incubated with recombinant Uv1809 ΔSP -His protein at 4 °C for 2 h with shaking and then washed with PBS three times.Beads were boiled for 5 min at 100 °C in 40 lL SDS sample loading buffer, and the proteins were analysed by immunoblotting with anti-His and anti-GST antibodies (ABclonal).

Bimolecular fluorescence complementation assays
The cDNA sequences of Uv1809 and OsSRT2 were ligated with the BiFC vectors pCAMBIA1301-nYFP and pCAMBIA1301-cYFP, respectively.The resulting constructs encoding the Uv1809-cYFP and OsSRT2-nYFP fusion proteins were co-infiltrated into N. benthamiana leaves, and the fluorescence signal of the infiltrated areas was analysed with a LSCM at 2 dpi.

Immunoblot analysis
Proteins were separated by 12% SDS-PAGE and transferred onto a polyvinylidene fluoride (PVDF) membrane (Merck Millipore, Burlington, MA) using wet transfer at 80 V for 90 min with a BioRad electroblotting apparatus.Membranes were blocked in Tris buffered saline with 0.1% Tween 20 (TBST) containing 5% (w/v) non-fat dry milk at room temperature for 2 h.Primary antibodies used were anti-GFP, anti-Flag, anti-His, or anti-GST (ABclonal).The membrane was incubated with primary antibodies in TBST with 5% non-fat dry milk at room temperature for 2 h with shaking and washed six times (5 min each) with TBST.Next, the membrane was incubated with goat anti-mouse (ABclonal) secondary antibody in TBST with 5% non-fat dry milk at room temperature for 1.5 h with shaking.The membrane was washed six times (5 min each) with TBST, and the signals were detected using Pierce ECL Western blotting substrate (Thermo Fisher Scientific) in a ChemiDoc XRS+ system (BioRad, Hercules, USA).

ChIP-seq and data analysis
Chromatin immunoprecipitation (ChIP) was performed as described previously (Lu et al., 2018).Briefly, 5 g of ZH11 or ossrt2 mutant rice spikelets was crosslinked in 1% (v/v) formaldehyde.Chromatin was extracted and fragmented to 100-to 500-bp fragments by sonication, and ChIP was performed using the anti-H4K5ac and anti-H4K8ac antibodies (Abcam, ab51997 and ab45166).Sequencing libraries were constructed with DNA from ChIP following the Illumina TruSeqCHIP Sample Prep Set A protocol and sequenced on an Illumina HiSeq2000 as 150-bp paired-end reads by Wuhan IGENEBOOK Biotechnology Co., Ltd. FastP (v0.232) was used with the default parameters to trim low-quality bases and adapters.Bowtie2 (version 2.3.5.1) was used to map cleaned reads to the rice genome (MSU7.0)using the default settings (Langmead and Salzberg, 2012).Samtools (v1.9) was then used to remove duplicate reads (Li et al., 2009).The following parameters were used in MACS (Zhang et al., 2008) to identify histone modification peaks: -f BAMPE -B -q 0.05 -g 3.6e+8.DiffBind (v3.5) (Stark and Brown, 2011) was used to identify histone modification sites that differed between the wild-type and mutant rice plants.Peaks were annotated by Homer (v4.11) using the default parameters (Heinz et al., 2010).Metaplots were generated by ngsplot (v2.61) (Shen et al., 2014).

RT-qPCR and ChIP-qPCR
Total RNA was extracted using TRIzol reagent (Vazyme Biotech, Nanjing, China).First-strand cDNA synthesis was carried out using cDNA Synthesis SuperMix (TransGen Biotech, China).RT-qPCR was performed with TransStartâ Tip Green qPCR SuperMix (TransGen Biotech, Beijing, China).The precipitated and input DNA was analysed by ChIP-qPCR.Transcript levels were normalized using the rice ubiquitin gene (OsUBQ1) or U. virens b-tubulin gene (Uv8b_900).At least three biological replicates were tested per treatment.

Detection of ROS accumulation
Rice plants were grown on MS medium in the growth chamber for 12 days.Leaves of the seedlings were cut into discs (3 mm) and then submerged in distilled water in a 96-well plate under light overnight.The distilled water was pipetted out and replaced with 100 lL of mixed solution for each well (the solution contained 50 lΜ luminol, 10 lg/mL horseradish peroxidase and 8 nM chitin) using a multiple-channel pipette.Chemiluminescence was measured at 30 s intervals over a period of 20 min in a SPARK-10M microplate reader (TECAN, M€ annedorf, Switzerland).Ten biological replicates were used for each sample.Distilled water was used as the mock control.

Statistical analyses
Statistical analyses of each treatment were performed using SPSS version 14.0 software (SPSS, Chicago, IL), and when significant treatment effects (P ≤ 0.05) were found, the means were compared by the least significant difference (LSD) test at P = 0.05.Table S1 Putative Uv1809-interacting proteins identified by Y2H.
Table S2 Transcriptionally up and downregulated genes in ossrt2 mutant vs WT.Table S3 H4K5ac up and downregulated genes in ossrt2 mutant vs WT.Table S4 H4K8ac up and downregulated genes in ossrt2 mutant vs WT.Table S5 GO pathways of H4K5ac and H4K8ac upregulated genes in ossrt2 mutant vs WT.Table S6 H4K5ac, H4K8ac and transcriptionally upregulated genes in ossrt2 mutant vs WT.Table S7 GO pathways of H4K5ac, H4K8ac and transcriptionally upregulated genes in ossrt2 mutant vs WT.

Figure 1
Figure 1 Uv1809 is a key virulence effector.(a) RT-qPCR analysis of the expression of Uv1809 in different infection stages on rice spikelets (1-20 dpi).Data are presented as mean AE SD (n = 3).The P values were determined by unpaired t-tests compared with 0 dpi.(b) Virulence assays of the wildtype HWD-2, DUv1809 mutants, CDUv1809-1 and CΔUv1809 ΔSP -1 strains on rice cultivar Wanxian98 at 21 dpi.(c) Number of rice smut balls per panicle.Data were collected from three independent experiments for each treatment.The P values were determined by unpaired t-tests compared with the widetype strain HWD-2.(d) The yeast YTK12 and its transformants expressing the empty vector pSUC2 (negative control) or pSUC2-Uv1809SP and pSUC2-Avr1SP (positive control) were tested for growth on SD-Trp or YPRAA medium and invertase activity in TTC medium.(e) Transient expression of Uv1809 suppressed programmed cell death of Nicotiana benthamiana leaves triggered by Bax.Representative leaves were photographed at 4 dpi.(f) Subcellular localization of Uv1809-GFP in N. benthamiana leaves.N. benthamiana histone H2B protein was used as a nuclear localization marker protein.DIC, differential interference contrast; GFP, green fluorescent protein.Scale bar = 20 lm.

Figure 2
Figure 2 Heterologous overexpression of Uv1809 increases susceptibility to rice pathogens and HIGS of Uv1809 enhances rice resistance against Ustilaginoidea virens.(a) Left: Resistance assays of 35S-Uv1809 and 35S-EV transgenic rice lines to U. virens strain HWD-2 infection at 25 dpi.Right: Mean number of rice smut balls measured in resistance assays.(b) Left: Disease symptoms at 14 dpi of 35S-Uv1809 and 35S-EV transgenic rice lines after inoculation with Xoo PXO99.Right: Mean lesion lengths at 14 dpi on the leaves of 35S-Uv1809 and 35S-EV transgenic rice lines after inoculation with Xoo PXO99.(c) Disease symptoms (Left), leaf lesion area (Middle) and the relative fungal biomass (Right) of 35S-Uv1809 and 35S-EV transgenic rice lines after spot-inoculation with Magnaporthe oryzae ZB-25 at 7 dpi.Relative fungal biomass was determined using quantitative reverse transcription (RT-qPCR) for M. oryzae Pot2 and normalized to rice OsUBQ1.The leaf lesion area was measured using Image J software.(d) RT-qPCR analysis of defencerelated genes at 1 dpi in 35S-EV and 35S-Uv1809 transgenic rice lines inoculated with U. virens.(e) Left: Resistance assays of Uv1809HIGS and EV transgenic rice lines against U. virens strain HWD-2 at 25 dpi.Right: Mean number of rice smut balls measured in resistance assays.(f) Relative mRNA expression of Uv1809 of U. virens during infection in the T2 transgenic rice lines at 6 dpi.(g) Length distribution and abundance of siRNAs targeting Uv1809HIGS-L1 in T2 transgenic rice plants.(h) Visualization of siRNAs targeting Uv1809 in infected Uv1809HIGS-L1 rice spikelets at 6 dpi by FISH using a specific probe.An, anther; hy, U. virens hyphae.Scale bar = 20 lm.All data are presented as mean AE SD (n = 3 unless otherwise indicated) and analysed by Fisher's least significant difference (LSD) test.The P values were determined by unpaired t-tests compared with the 35-EV or EV.
Authors.Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 22, 148-164

Figure 3 Figure 4
Figure3Uv1809 physically interacts with OsSRT2.(a) Y2H analysis of the interaction between Uv1809 and OsSRT2.The interaction between BD-53 and AD-T was taken as the positive control, BD and AD-OsSRT2 was taken as the negative control.SD-3, SD-Trp-Leu-His; SD-4, SD-Trp-Leu-His-Ade; BD, pGBKT7; AD, pGADT7.(b) In vivo Co-IP of Uv1809 interacts with OsSRT2.Co-IP was performed on extracts of Nicotiana benthamiana leaves by coexpression of OsSRT2-Flag and Uv1809-GFP.Uv1809-GFP and OsSRT2-Flag were detected by western blotting using anti-GFP and anti-Flag antibodies, respectively.(c) A GST pull-down assay was used to detect the interaction between Uv1809 ΔSP -His and OsSRT2-GST.Uv1809 and OsSRT2 were fused to His and GST tags, respectively, and expressed in Escherichia coli.OsSRT2-GST or GST-bound resin was incubated with E. coli crude extracts containing Uv1809 ΔSP -His and analysed by western blotting.Uv1809 ΔSP -His and OsSRT2-GST were detected using anti-His and anti-GST antibodies, respectively.(d) BiFC assays for the interaction between Uv1809 and OsSRT2.N. benthamiana leaves were infiltrated with a mixture of Agrobacterium tumefaciens strains co-expressing the OsSRT2-nYFP and Uv1809-cYFP constructs.YFP signals were observed at 2 dpi.Infiltration with Agrobacterium co-expressing the OsSRT2-nYFP and cYFP, nYFP and Uv1809-cYFP constructs were used as the negative control.No YFP signals was observed in these negative controls.Scale bar = 20 lm.

Figure 6
Figure 6 RNA-seq and ChIP-seq data analysis of ossrt2 and wild-type rice spikelets.(a) Volcano plots of differential transcript levels in ossrt2 relative to wild type.Purple plots represent upregulated genes (Fold change > 2, P value < 0.05); cyan plots represent downregulated genes (Fold change > 2, P value < 0.05); grey plots represent genes with no significant difference.(b) Representative GSEA enriched pathways in ossrt2.(c) GO pathway analysis of the genes (n = 604) that were upregulated in ossrt2 mutants.(d) Peaks with H4K5ac and H4K8ac level changes in ossrt2 compared with wild type.Peaks with reduced H4K5ac or H4K8ac in the mutant are shown in cyan, and those with gained H4K5ac or H4K8ac are in pink (Fold change > 1.5, P < 0.05).(e) Venn diagrams of upregulated H4K5 and H4K8 acetylation peaks (left) or genes (right) in ossrt2.(f) GO pathways found in both H4K5ac and H4K8ac significantly upregulated genes in ossrt2 versus wild type.

Figure 7
Figure 7 Role of H4K5ac and H4K8ac in gene expression regulation in ossrt2 rice spikelets.(a) Correlation analysis of expression changes and H4K5ac or H4K8ac changes in ossrt2 versus wild type.Person Correlation Coefficient was shown.(b) Metaplots of H4K5 or H4K8 acetylation ChIP-seq reads in transcriptionally upregulated genes.TSS, transcriptional start site.TES, transcriptional end site, RPM, reads count per million mapped reads.(c) Venn diagrams of H4K5 and H4K8ac hyper-acetylated genes, and transcriptionally upregulated genes in ossrt2.(d) Representative GO pathways of H4K5ac, H4K8ac, and transcriptionally upregulated genes (N = 84) in ossrt2.(e) RT-qPCR and ChIP-qPCR analysis of H4K5ac-or H4K8ac-marked defence-related genes in ZH11 and ossrt2 rice spikelets.Asterisks indicate statistically significant differences compared to ZH11 at P < 0.05.

Figure 8
Figure 8 A working model illustrating how Uv1809 manipulates histone deacetylase OsSRT2 to suppress rice immunity during Ustilaginoidea virens infection.During infection, U. virens effector Uv1809 is secreted and translocated into host cells, and then physically interacts with OsSRT2.Uv1809 disrupts host immunity by recruiting enhancing OsSRT2-modulated deacetylation, thereby reducing the levels of H4K5ac and H4K8ac in rice plants and interfering with defence gene activation.

ª
2023 The Authors.Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 22, 148-164 (Millipore, MA, USA, 04-119).ImageJ software was used for quantification of the band intensities from the immunoblots.

Figure S1
Figure S1 Characterization of Uv1809 in U. virens.Figure S2 Gene deletion and complementation of Uv1809 in U. virens.Figure S3 Deletion of Uv1809 affects conidiation of U. virens.Figure S4 Subcellular localization of Uv1809 variants in N. benthamiana leaves.Figure S5 Uv1809 expressed in M. oryzae is translocated into rice cells during host infection.Figure S6 Expression and Agronomic traits of 5S-Uv1809 and 35S-EV transgenic rice plants.Figure S7 HIGS of Uv1809 in rice.Figure S8 Uv1809 246-391 interact with OsSRT2.Figure S9 Expression and subcellular localization of OsSRT2.Figure S10 Mutation identified within and around the target sites of OsSRT2 generated through CRISPR/Cas9-mediated genome editing in rice.

Figure S2
Figure S1 Characterization of Uv1809 in U. virens.Figure S2 Gene deletion and complementation of Uv1809 in U. virens.Figure S3 Deletion of Uv1809 affects conidiation of U. virens.Figure S4 Subcellular localization of Uv1809 variants in N. benthamiana leaves.Figure S5 Uv1809 expressed in M. oryzae is translocated into rice cells during host infection.Figure S6 Expression and Agronomic traits of 5S-Uv1809 and 35S-EV transgenic rice plants.Figure S7 HIGS of Uv1809 in rice.Figure S8 Uv1809 246-391 interact with OsSRT2.Figure S9 Expression and subcellular localization of OsSRT2.Figure S10 Mutation identified within and around the target sites of OsSRT2 generated through CRISPR/Cas9-mediated genome editing in rice.

Figure S4
Figure S1 Characterization of Uv1809 in U. virens.Figure S2 Gene deletion and complementation of Uv1809 in U. virens.Figure S3 Deletion of Uv1809 affects conidiation of U. virens.Figure S4 Subcellular localization of Uv1809 variants in N. benthamiana leaves.Figure S5 Uv1809 expressed in M. oryzae is translocated into rice cells during host infection.Figure S6 Expression and Agronomic traits of 5S-Uv1809 and 35S-EV transgenic rice plants.Figure S7 HIGS of Uv1809 in rice.Figure S8 Uv1809 246-391 interact with OsSRT2.Figure S9 Expression and subcellular localization of OsSRT2.Figure S10 Mutation identified within and around the target sites of OsSRT2 generated through CRISPR/Cas9-mediated genome editing in rice.

Figure S5
Figure S1 Characterization of Uv1809 in U. virens.Figure S2 Gene deletion and complementation of Uv1809 in U. virens.Figure S3 Deletion of Uv1809 affects conidiation of U. virens.Figure S4 Subcellular localization of Uv1809 variants in N. benthamiana leaves.Figure S5 Uv1809 expressed in M. oryzae is translocated into rice cells during host infection.Figure S6 Expression and Agronomic traits of 5S-Uv1809 and 35S-EV transgenic rice plants.Figure S7 HIGS of Uv1809 in rice.Figure S8 Uv1809 246-391 interact with OsSRT2.Figure S9 Expression and subcellular localization of OsSRT2.Figure S10 Mutation identified within and around the target sites of OsSRT2 generated through CRISPR/Cas9-mediated genome editing in rice.

Figure S6
Figure S1 Characterization of Uv1809 in U. virens.Figure S2 Gene deletion and complementation of Uv1809 in U. virens.Figure S3 Deletion of Uv1809 affects conidiation of U. virens.Figure S4 Subcellular localization of Uv1809 variants in N. benthamiana leaves.Figure S5 Uv1809 expressed in M. oryzae is translocated into rice cells during host infection.Figure S6 Expression and Agronomic traits of 5S-Uv1809 and 35S-EV transgenic rice plants.Figure S7 HIGS of Uv1809 in rice.Figure S8 Uv1809 246-391 interact with OsSRT2.Figure S9 Expression and subcellular localization of OsSRT2.Figure S10 Mutation identified within and around the target sites of OsSRT2 generated through CRISPR/Cas9-mediated genome editing in rice.
Figure S1 Characterization of Uv1809 in U. virens.Figure S2 Gene deletion and complementation of Uv1809 in U. virens.Figure S3 Deletion of Uv1809 affects conidiation of U. virens.Figure S4 Subcellular localization of Uv1809 variants in N. benthamiana leaves.Figure S5 Uv1809 expressed in M. oryzae is translocated into rice cells during host infection.Figure S6 Expression and Agronomic traits of 5S-Uv1809 and 35S-EV transgenic rice plants.Figure S7 HIGS of Uv1809 in rice.Figure S8 Uv1809 246-391 interact with OsSRT2.Figure S9 Expression and subcellular localization of OsSRT2.Figure S10 Mutation identified within and around the target sites of OsSRT2 generated through CRISPR/Cas9-mediated genome editing in rice.
Figure S1 Characterization of Uv1809 in U. virens.Figure S2 Gene deletion and complementation of Uv1809 in U. virens.Figure S3 Deletion of Uv1809 affects conidiation of U. virens.Figure S4 Subcellular localization of Uv1809 variants in N. benthamiana leaves.Figure S5 Uv1809 expressed in M. oryzae is translocated into rice cells during host infection.Figure S6 Expression and Agronomic traits of 5S-Uv1809 and 35S-EV transgenic rice plants.Figure S7 HIGS of Uv1809 in rice.Figure S8 Uv1809 246-391 interact with OsSRT2.Figure S9 Expression and subcellular localization of OsSRT2.Figure S10 Mutation identified within and around the target sites of OsSRT2 generated through CRISPR/Cas9-mediated genome editing in rice.
Figure S1 Characterization of Uv1809 in U. virens.Figure S2 Gene deletion and complementation of Uv1809 in U. virens.Figure S3 Deletion of Uv1809 affects conidiation of U. virens.Figure S4 Subcellular localization of Uv1809 variants in N. benthamiana leaves.Figure S5 Uv1809 expressed in M. oryzae is translocated into rice cells during host infection.Figure S6 Expression and Agronomic traits of 5S-Uv1809 and 35S-EV transgenic rice plants.Figure S7 HIGS of Uv1809 in rice.Figure S8 Uv1809 246-391 interact with OsSRT2.Figure S9 Expression and subcellular localization of OsSRT2.Figure S10 Mutation identified within and around the target sites of OsSRT2 generated through CRISPR/Cas9-mediated genome editing in rice.
Figure S11Agronomic traits of the wild-type ZH11 and ossrt2 mutants rice lines at mature stage following growth in field conditions.FigureS12Multiscatter plots of RNA-seq and ChIP-seq.FigureS13RT-qPCR to validate differentially expressed genes from RNA-seq. Figure S14 Pearson correlation coefficients of H4K5ac (a) and H4K8ac (b) ChIP-seq replicates.Figure S15 Analysis of leaf H4K5 and H4K8 acetylation ChIP-seq data of wild-type spikelets.Figure S16 ROS level in ZH11 and ossrt2 rice plants after chitin treatment.TableS1Putative Uv1809-interacting proteins identified by Y2H.

Figure S14
Figure S11Agronomic traits of the wild-type ZH11 and ossrt2 mutants rice lines at mature stage following growth in field conditions.FigureS12Multiscatter plots of RNA-seq and ChIP-seq.FigureS13RT-qPCR to validate differentially expressed genes from RNA-seq. Figure S14 Pearson correlation coefficients of H4K5ac (a) and H4K8ac (b) ChIP-seq replicates.Figure S15 Analysis of leaf H4K5 and H4K8 acetylation ChIP-seq data of wild-type spikelets.Figure S16 ROS level in ZH11 and ossrt2 rice plants after chitin treatment.TableS1Putative Uv1809-interacting proteins identified by Y2H.

Figure S15
Figure S11Agronomic traits of the wild-type ZH11 and ossrt2 mutants rice lines at mature stage following growth in field conditions.FigureS12Multiscatter plots of RNA-seq and ChIP-seq.FigureS13RT-qPCR to validate differentially expressed genes from RNA-seq. Figure S14 Pearson correlation coefficients of H4K5ac (a) and H4K8ac (b) ChIP-seq replicates.Figure S15 Analysis of leaf H4K5 and H4K8 acetylation ChIP-seq data of wild-type spikelets.Figure S16 ROS level in ZH11 and ossrt2 rice plants after chitin treatment.TableS1Putative Uv1809-interacting proteins identified by Y2H.
Figure S11Agronomic traits of the wild-type ZH11 and ossrt2 mutants rice lines at mature stage following growth in field conditions.FigureS12Multiscatter plots of RNA-seq and ChIP-seq.FigureS13RT-qPCR to validate differentially expressed genes from RNA-seq. Figure S14 Pearson correlation coefficients of H4K5ac (a) and H4K8ac (b) ChIP-seq replicates.Figure S15 Analysis of leaf H4K5 and H4K8 acetylation ChIP-seq data of wild-type spikelets.Figure S16 ROS level in ZH11 and ossrt2 rice plants after chitin treatment.TableS1Putative Uv1809-interacting proteins identified by Y2H.
23-245-His, or Uv1809 246-391 -His.First, we confirmed the in vitro deacetylation activity of OsSRT2-GST using a ª 2023 The Authors.Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 22, 148-164