Identification of NADPH Oxidase Genes Crucial for Rice Multiple Disease Resistance and Yield Traits

Reactive oxygen species (ROS) act as a group of signaling molecules in rice functioning in regulation of development and stress responses. Respiratory burst oxidase homologues (Rbohs) are key enzymes in generation of ROS. However, the role of the nine Rboh family members was not fully understood in rice multiple disease resistance and yield traits. In this study, we constructed mutants of each Rboh genes and detected their requirement in rice multiple disease resistance and yield traits. Our results revealed that mutations of five Rboh genes (RbohA, RbohB, RbohE, RbohH, and RbohI) lead to compromised rice blast disease resistance in a disease nursery and lab conditions; mutations of five Rbohs (RbohA, RbohB, RbohC, RbohE, and RbohH) result in suppressed rice sheath blight resistance in a disease nursery and lab conditions; mutations of six Rbohs (RbohA, RbohB, RbohC, RbohE, RbohH and RbohI) lead to decreased rice leaf blight resistance in a paddy yard and ROS production induced by PAMPs and pathogen. Moreover, all Rboh genes participate in the regulation of rice yield traits, for all rboh mutants display one or more compromised yield traits, such as panicle number, grain number per panicle, seed setting rate, and grain weight, resulting in reduced yield per plant except rbohb and rbohf. Our results identified the Rboh family members involved in the regulation of rice resistance against multiple pathogens that caused the most serious diseases worldwide and provide theoretical supporting for breeding application of these Rbohs to coordinate rice disease resistance and yield traits. Supplementary Information The online version contains supplementary material available at 10.1186/s12284-023-00678-5.


Introduction
Plants employ pathogen-associated molecular pattern (PAMP)-triggered immunity (PTI) and effector-triggered immunity (ETI) to counterattack the invasion of pathogens (Jones and Dangl 2006).The PAMPs were perceived by cell surface-localized plant pattern recognition receptors (PRRs), which subsequently activate downstream early immune responses, such as activation of mitogen-activated protein kinase (MAPK) cascade, influx of intracellular calcium, reactive oxygen species (ROS) production, and callose deposition at infected sites (Boller and Felix 2009).Pathogen-secreted effectors were directly or indirectly recognized by its cognate resistance genes, which subsequently activate strong resistance resulting in hypersensitive responses accompanying ROS production and fast cell death at the infection sites (Yuan et al. 2021;Zhou and Zhang 2020).

−
), hydroxyl radicals (OH • ) and hydrogen peroxide (H 2 O 2 ), have been identified as a group of signaling molecules in plants functioning in regulation of growth, development, and stress responses (Chapman et al. 2019;Saxena et al. 2016).Respiratory burst oxidase homologues (Rbohs), known as Nicotinamide adenine dinucleotide phosphate (NADPH) oxidases (NOXs), are key enzymes in generation of O• 2 − by catalyzing the transfer of electrons from NADPH to oxygen (O 2 ) in plants (Kaur et al. 2014;Suzuki et al. 2011).
NOXs were first identified in human cells for catalyzing the production of ROS (Segal and Abo 1993).In plants, Rbohs were characterized as ROS synthetic enzymes based on their sequence similarity to the phagocyte gp91 phox /NOX2 in human (Keller et al. 1998).Rbohs contain six transmembrane domains, a N-terminal region with Ca 2+ -binding EF-hands, and a C-terminal NADPHand flavin adenine dinucleotide (FAD)-binding cytoplasmic region (Torres et al. 2002).In the transmembrane domains, two pairs of histidine residues were located on third and fifth transmembrane domains and attached by the heme groups to facilitate electron transport (Finegold et al. 1996).N-terminus contains two calcium-binding EF hand motifs and two phosphorylation sites required for the regulation of enzyme activity (Torres et al. 2002).C-terminus-contained FAD-and NADPH-binding domains are highly conserved in NADPH family members (Kaur et al. 2018).
Arabidopsis thaliana (Arabidopsis) Rboh family contain 10 genes and is the best characterized Rboh family in plant.The 10 Rbohs, from RbohA to RbohJ, display diverse expression patterns and play differential roles in development and stress responses (Chapman et al. 2019).For example, RbohE is involved in the formation and development of pollen grains.rbohe mutants showed reduced ROS accumulation in the anther, leading to delayed programmed cell death of the tapetum, which impairs pollen viability (Xie et al. 2014).RbohD is participated in biotic and abiotic stress responses (Lee et al. 2013;Torres et al. 2002) and is phosphorylated on its N-terminal cytoplasmic region by a variety of protein kinases, including receptor-like cytoplasmic kinases (RLCKs) such as BOTRYTIS INDUCED KINASE 1 (BIK1).Upon PAMP treatment, BIK1 directly phosphorylates RbohD at N-terminal specific sites to enhance ROS generation thus improves plant immunity (Kadota et al. 2014;Li et al. 2014).Moreover, RbohD and RbohF regulate ABA-and ethylene-induced stomatal responses (Sierla et al. 2016), as well as the complicated interaction between the ABA-, ethylene-, JA-and SA-signaling pathways (Chapman et al. 2019;Desikan et al. 2006;Maruta et al. 2011).
Rice Rboh family contains 9 members widely participating in rice growth, development, and stress responses (Kaur et al. 2018).OsRbohA, OsRbohD, and OsRbohE are involved in rice tolerance against salt stress because the transcription of these genes is induced in the roots of a tolerant cultivar (Saini et al. 2018).OsRbohB and OsR-bohE are participated in abscisic acid-induced H 2 O 2 production (Ni et al. 2019).OsRbohC is induced by arsenic stress and possibly participates in restoring root system architecture under arsenic stress (Ghate et al. 2022).OsR-bohD and OsRbohF collaborate with photoreceptor protein phytochrome B to regulate transcripts in response to light stress (Fichman et al. 2023).OsRbohF also improves salt tolerance in rice by boosting the accumulation of ROS, which is required for the activation of downstream stress-responsive genes and K + uptake transporters (Liu et al. 2020).OsRbohH and OsRbohI are involved in rice submergence resistance and highly induced by full submergence with increased ROS accumulation, indicating the important role of the two genes in maintaining homeostasis during submergence (Wu and Yang 2016).OsRbohH suppresses ethylene-induced aerenchyma formation in roots by mediating ROS production.osrbohh mutants displayed reduced ROS accumulation and inducible aerenchyma formation in rice roots; In contrast, the expression of OsRbohH is induced to greatly higher levels in cortical cells under oxygen-deficient conditions (Yamauchi et al. 2017).
Except abiotic stresses, rice is also threatened by multiple disease pathogens during the whole growth period.For example, the semi-necrotrophic fungal pathogen Magnaporthe oryzae (M.oryzae) causes blast disease occurred on leaves, panicles, and grains through the whole growth period (Talbot 2003); the necrotrophic fungal pathogen Rhizoctonia solani (R. solani) causes sheath blight on sheaths and leaves (Molla et al. 2020); the bacterial pathogen Xanthomonas oryzae pv.oryzae (Xoo) causes bacterial leave blight (Chukwu et al. 2019).OsRbohA RNAi plants were more susceptible to sheath blight (Chen et al. 2023).OsRbohB and OsRbohH possibly play an important role in rice fungal disease resistance because they are the major NADPH oxidases responsive to chitin.Upon M. oryzae infection, rbohb mutants displays greater lesion length and supports more extensive fungal growth than the wild type, indicating an essential role of OsRbohB in rice blast disease resistance (Nagano et al. 2016).Moreover, OsRbohB interacted with receptor like kinases RLK20, RLK21, and RLK20, which inhibits its enzyme activity to negatively regulate rice resistance to Xoo (Mei et al. 2022).
In a previous study, we found that OsRbohB and OsR-bohC (previously named as OsRbohF, (Jang et al. 2012) were responsive to M. oryzae (Li et al. 2019).A virulent strain Guy11 significantly suppressed the RNA amounts of the two genes in rice varieties Kasalath, NPB and TP309, implying a role of OsRbohs in rice-M.oryzae interaction (Li et al. 2019).However, the roles of the nine OsRbohs were not fully understood in rice multiple disease resistance and yield traits.In this study, we systemically explore the roles of all the nine OsRboh genes in the regulation of rice multiple disease resistance and yield traits.We detected the diverse expression patterns of all the nine Rbohs through the whole growth period and constructed the mutants of these OsRbohs.We examined the resistance of these mutants against M. oryzae, R. solani, and Xoo, respectively, and the production of ROS induced by PAMPs and pathogen.We also analyzed the agronomic traits of these mutants.Our results would shed light on the function of the Rboh family in rice, as well as provide experimental reference for breeding application of these genes to coordinate disease resistance and yield traits in rice.

Rice Rboh Family Members are Differentially Responsive to PAMPs
To make out the evolutional relationship of the Rbohs between Arabidopsis and rice, we constructed a phylogenetic tree via BLAST analysis.From this phylogenetic tree, the Rboh members can be classified into three clades, of which clade I contains OsRbohB, OsR-bohH, and OsRbohI belonging to the same clade close to AtRbohD; Clade II contains rice OsRbohF, OsRbohG, OsRbohA, and OsRbohC belonging to a clade close to AtRbohE and AtRbohF; Clade III contains OsRbohD and OsRbohE, and Arabidopsis AtRbohH and AtRbohJ (Fig. S1).These clades imply that rice OsRbohs may play conserved but differential functions in comparison with those of Arabidopsis in disease resistance and growth.
Although some Rbohs, including RbohA, RbohB, RbohC, and RbohH, have been characterized as regulators in rice disease resistance against Xoo or M. oryzae, the roles of the other Rbohs are unknown.To systemically explore the roles of the Rboh family members in rice disease resistance, we first tested their response to PAMPs by examining their expression pattern upon the treatment of chitin and flg22 in Lijiang xin Tuan Heigu (LTH) and Taipei 309 (TP309).LTH is a Japonica accession sensitive to over 1300 regional M. oryzae isolates worldwide (Lin et al. 2001).TP309 is another Japonica accession susceptible to many M. oryzae isolates and widely used in rice transformation.Except RbohD whose expression was below detectable level in leaves, the expression of six Rbohs (RbohA, RbohB, RbohC, RbohE, RbohH, and RbohI) was induced by chitin-or flg22-treatment more than 2-fold higher than that of the untreated samples in both accessions (Fig. 1), indicating the involvement of these Rbohs in rice responses to PAMPs that derived from fungal and bacterial pathogens.In contrast, although the expression of RbohF and RbohG were also enhanced by chitin or flg22, the enhancement levels were less than 2-fold of the untreated samples, suggesting a minor role in rice responses to PAMPs (Fig. 1).These data imply the differential roles of these Rbohs in disease resistance, of which six Rbohs may participate in the modulation of rice disease resistance.
To further explore the roles of the Rboh family members in the regulation of rice disease resistance and yield traits, we generated mutants for each of the nine Rbohs using the CRISPR/Cas9 technology in TP309 background.We obtained two or more homozygous mutants for each Rboh gene except RbohC, which we obtained two mutants with the same mutation of 7-base deletion (Fig. S2).Except rbohi-3, all mutants carried one or more base deletion or insertion creating a stop codon resulting in truncated proteins or changed amino acid sequence at N-terminus (Fig. S2A-I).Intriguingly, rbohi-3 carried a 3-base deletion leading to a serine deletion at A,a 338 site (Fig. S2I ), which was a possible key site for phosphorylation by upstream kinases.

Five OsRboh Family Members Regulates Rice Blast Disease at Different Levels
We first identified the Rbohs participated in rice blast disease resistance by examining the responses of these Rbohs to a virulent M. oryzae strain GZ8 in a susceptible accession LTH, a susceptible accession TP309, and a resistance accession Pyricularia-Kanto51-m-Tsuyuake (IRBLkm-Ts) carrying a resistance (R) locus Pi-km (Tsumematsu et al. 2000).As shown in Fig. S3, eight Rboh genes were suppressed by M. oryzae at one or more time points in both LTH and IRBLKm-Ts, indicating the involvement of these Rbohs in blast disease resistance.
We then examined the blast disease resistance of these rboh mutants in lab conditions and in a blast disease nursery.Plants of rboha, rbohb, rbohe, and rbohi displayed exacerbated susceptibility with bigger disease lesions than the TP309 control in the disease nursery (Fig. 2A-B) and punch-inoculated with a virulent strain GZ8 in lab conditions (Fig. 2C-D), implying a key role of the four Rbohs in rice blast disease resistance.Moreover, rbohh showed exacerbated susceptibility following GZ8 punch-inoculation in lab conditions whereas a similar disease symptom like TP309 when planted in disease nursery (Fig. 2A-D), implying the involvement of RbohH in post-invasive defense.In contrast, rbohc, rbohd, rbohf, and rbohg displayed a compatible disease symptom and fungal growth to the TP309 control, indicating the independence of these four Rbohs in rice resistance against M. oryzae.Altogether, these results indicate that five Rbohs (RbohA, RbohB, RbohE, RbohH, and RbohI) participated in and positively modulate rice blast disease resistance.

Five OsRboh Family Members Modulate Rice Sheath Blight Resistance
We next detected the resistance of these mutants to another fungal pathogen, R. solani, the causative agent of rice sheath blight.When planted in a disease nursery, rboha, rbohb, rbohc, rbohe, and rbohh showed bigger and remarkably longer disease lesions than that of the TP309 control (Fig. 3A-B).Consistently, similar results were obtained when these mutants were inoculated with a virulent R. solani strain AG-1-IA in lab conditions (Fig. 3C-D).These results indicate that the five Rbohs were involved in and positively regulated rice sheath blight resistance.In contrast, rbohd, rbohf, rbohg, and rbohi displayed disease lesions like that of TP309 (Fig. 3A-D), suggesting that these four Rbohs possibly not participated in rice disease resistance against R. solani.

Six OsRboh Family Members Regulate Rice leaf Blight Resistance
We also detected the resistance of these mutants to the bacterial pathogen, Xoo, which caused one of the most serious rice bacterial diseases, leaf blight disease.When inoculated with a virulent strain PXO99 in a paddy yard, rboha, rbohb, rbohc, rbohe, rbohh, and rbohi lines showed significantly longer disease lesions than that of the TP309 control, whereas rbohd, rbohf, and rbohg lines exhibited indistinguishable length of disease lesions (Fig. 4A-B).These results indicate that RbohA, RbohB, RbohC, RbohE, RbohH, and RbohI are required for rice leaf blight resistance.

Six Rboh Family Members Regulate ROS Production
Based on above disease analysis, we deduced that six Rbohs, including RbohA, RbohB, RbohC, RbohE, RbohH, and RbohI, participate and positively regulate rice disease resistance against one or more pathogens.These results are consistent with the induction of the six Rbohs by PAMPs.As Rbohs functioned as enzymes to catalyze the generation of O• 2 − , a key ROS molecule, we then examined the early ROS production triggered by PAMPs and the late O• 2 − and H 2 O 2 accumulation induced by M. oryzae in these mutants.
Consistent with the disease phenotypes in these mutants, both chitin-and flg22-triggered ROS production was significantly less in rboha, rbohb, rbohc, rbohe, rbohh, and rbohi than that in the TP309 control (Fig. 5A-B).Similarly, M. oryzae-induced accumulation of O• 2 − and H 2 O 2 was also remarkedly compromised in these mutants compared to that in TP309 control (Fig. 5C).These results implied that these six Rbohs play roles in rice disease resistance by boosting ROS production.In contrast, rbohd, rbohf, and rbohg showed a similar ROS burst and O• 2 − and H 2 O 2 accumulation like the TP309 control (Fig. 5A-C), suggesting that the three Rbohs were not involved in the regulation of rice leaf disease resistance.

Mutation of Rboh Family Members Lead to Diverse Changes of Yield Traits
Rbohs not only modulate rice resistance against multiple pathogens, but also control rice agronomic traits.We analyzed the yield traits of these rboh mutants that planted in a field during the rice growth season from middle April to middle September in 2022 and 2023 at Chengdu city, Sichuan Province, the southwest of China.We found that although all the mutants displayed a similar gross morphology like the TP309 control (Fig. 6A-C), they all displayed one or more defects in yield traits (Fig. 6D, Table S2).Plants of rboha mutants showed smaller panicles and significantly reduced grain numbers per panicles; rbohb lines displayed decreased seed setting rates but slightly increased panicle numbers; rbohc exhibited decreased plant height, significantly reduced grain numbers per panicle, seed setting rates, and grain weight; rbohd showed significantly increased panicle number but smaller panicles with significantly decreased grain numbers and seed setting rates; rbohe exhibited remarkably decreased seed setting rates and grain weight; rbohf displayed decreased seed setting rates whereas slightly increased panicle numbers, rbohg and rbohh showed decreased seed setting rates; rbohi displayed decreased plant height, smaller panicles with decreased grain numbers per panicle and seed setting rates (Fig. 6A-D, Table S2).As a result, rboha, rbohc, rbohd, rbohe, rbohg, rbohh, and rbohi exhibited significantly decreased grain yield, especially rbohc and rbohd lines, which showed the most reduction of yield (Fig. 6D, Table S2), indicating the requirement of the seven Rbohs in the development of panicle and the formation of grain yield.In contrast, although exhibited reduced seed setting rates, rbohb and rbohf showed slight but not

Rboh Family Members Show Diverse Spatio-Temporal Expression Patterns During Growth Period
As the function of Rbohs is associated with their expression levels, we then detect their spatio-temporal expression patterns by performing a time-course experiment to examine the relative mRNA levels (compared to that of Ubiquitin) of all the nine Rbohs in rice leaves at vegetative stage and in panicles at the reproductive stage.Based on their relative mRNA amounts, we classified the nine Rbohs into three groups at vegetative and panicle stages, respectively.
The group 1 members at vegetative stage showed relatively high expression levels, including RbohB, RbohC, and RbohE.RbohB displayed the highest RNA amounts in leaves among the three members, whereas RbohC and RbohE showed gradually increased RNA amounts in leaves at late vegetative stage (Fig. 7A), which were consistent with the key role of RbohB and RbohE in rice disease resistance.The members in group 2 exhibited lower RNA levels compared to those in group 1, including RbohA and RbohG, which were increased gradually in leaves (Fig. 7B).The members in group 3 exhibited the lowest RNA levels, including RbohF, RbohH, and RbohI.RbohH were decreased from 30 day whereas RbohI was decreased from 20 to 40 day and then increased gradually; RbohF was low-expressed during the whole growth period (Fig. 7C).RbohD was below to the detectable levels.The low expression of RbohD and RbohF partially explained their absence in rice immunity in leaves.
Similarly, we also classified the nine Rbohs into three groups at panicle stages based their expression levels.The group 1 members exhibited the highest expression levels, including RbohC and RbohE (Fig. 7D).The group 2 exhibited relatively lower RNA levels, including RbohA, RbohB, RbohD, and RbohG (Fig. 7E).The members in group 3 exhibited the lowest RNA levels and included RbohF, RbohH, and RbohI (Fig. 7F).Intriguingly, the expression of RhobA, RbohC, RbohE, and RbohG was gradually enhanced during panicle development, and the expression of RbohD and RbohI was enhanced during flowering (Fig. 7D-F).These results were consistent with the defects of yield traits of these rboh mutants (Fig. 6), implying a positive role of these genes in the development of panicle and flower organs.

Discussion
In this study, we systemically explored the functions of the Rboh family members in the regulation of multiple disease resistance and yield traits in rice (Fig. 8; Table S2).
Our results demonstrated that six Rboh family members, RbohA, RbohB, RbohC, RbohE, RbohH, and RbohI, modulate rice leaf disease resistance.RbohA, RbohB, RbohE and RbohH are involved in multiple disease resistance against fungal pathogen M. oryzae, R. solani, and bacterial pathogen Xoo; RbohI modulates resistance against M. oryzae and Xoo, RbohC participates in the regulation of resistance against R. solani and Xoo.Moreover, all Rboh family members regulate rice yield traits, such as panicle number, grain number per panicle, seed setting rate, and grain weight, and mutations of seven Rbohs result in reduced yield per plant associated with one or more compromised yield traits (Fig. 6).Intriguingly, although it is possible that these Rbohs play a redundant role each other, mutations in a single Rboh gene in this study led to compromised resistance against one or more pathogens or defected yield traits, implying essential function of each Rboh that cannot be replaced by any other Rboh in rice development or disease resistance.However, the preference and effect of these Rbohs on different disease resistance and yield traits are differential from each other.For example, compared to the TP309 control and other mutants, rbohb, rbohe, and rbohi showed the biggest lesions and supported the most fungal growth when inoculated with M. oryzae or planted in a blast nursery, indicating the key role of RbohB, RbohE, and RbohI in rice blast disease resistance.In contrast, rbohb displayed the most significant susceptibility to R. sonalo, whereas Rbohc and Rbohe exhibited the longest disease lesions following Xoo inoculation.RbohC and RbohD played crucial roles in the regulation of yield traits, for rbohc and rbohd displayed the most reduction of yield.Moreover, although RbohF and RbohG was expressed in both leaves and panicles, they were specifically functioned in panicles to regulate seed setting rate and had no effect on leaf disease resistance.Altogether, our results revealed the diverse preference of Rbohs in the regulation of rice different disease resistance and yield traits. ) induced the expression of six RBOHs, including RbohA, RbohB, RbohC, RbohE, RbohH, and RbohI, which subsequently triggered the ROS production.On the one side, the enhanced accumulation of ROS improves rice resistance against both fungal and bacterial pathogens.In the other side, the enhanced ROS accumulation impact the development of panicles and grains leading to compromised yield traits and reduced yield.As a results, all the nine Rbohs participated in the coordination of rice yield and multiple disease resistance The spatio-temporal expression patterns of Rbohs might be critical in determining their multiplicity of functions.For example, in 4-week-old Arabidopsis plants, AtRbohF was mainly expressed in the vascular tissue, while AtRbohD was expressed in disperse areas of the leaf lamina and stomatal cells, which was consistent with the higher ROS production in AtRbohD-overexpressed plants than that in AtRbohF-overexpressed plants triggered by PAMPs or pathogens (Morales et al. 2016).In this study, compared to the other Rbohs in rice, RbohB showed the most mRNA amounts in rice leaves, which was consistent with its key role in the regulation of leave disease resistance and defense responses (Figs. 2, 3, 4 and 5).Conversely, RbohD was low-expressed in leaves but expressed higher in panicles during flowering (Fig. 7), which was consistent with the lowest seed setting rate of rbohd in all tested rboh mutants (Fig. 6), indicating a key role of RbohD in the development of flowering organs.
Except for the spatio-temporal expression diversity, the preferences of Rbohs were probably caused by the different regulatory factors or hormone signalings functioning upstream or downstream.For example, RbohB protein was localized in microdomains and interacted with Rac1, a Rac/ROP small G protein acting as a molecular switch in defense signaling (Nagano et al. 2016).Disruption of the OsRac1-OsRbohB interaction resulted in compromised ROS production (Kosami et al. 2014).Whether other resistance-related Rboh family members also localized in microdomains and interacted with Rac1 to improve ROS production was unknown and need further study.
It was intriguingly that the mRNA amounts of most Rbohs were suppressed by M. oryzae in both susceptible and resistance varieties (Fig. S3).These results seemed be conflictive to their positive function in ROS production and indicated the existence of other regulation mechanism on translational level or enzyme activates to control the functions of Rbohs.It was reported that upon M. oryzae inoculation, the N terminus of RbohB was bound and phosphorylated by a receptor-like cytoplasmic kinase (RLCK)118, which was phosphorylated by another receptor like kinase SPL11 CELL-DEATH SUP-PRESSOR 2 (SDS2) (Fan et al. 2018).However, the phosphor-sites specifically modified by OsRLCK118 in RbohB need further investigation.A recent study demonstrated that RbohB is the main NADPH oxidase controlling ABA-induced H 2 O 2 production and related physiological responses in rice.DMI3, a Ca 2+ /calmodulin-dependent protein kinase acting as an important positive regulator of abscisic acid (ABA) signaling, directly interacts with and activates RbohB by phosphorylating RbohB at Ser-191 (Wang et al. 2023).The activated RbohB subsequently catalyzes the production of H 2 O 2 , which is in turn required for ABA-induced activation of OsDMI3 (Wang et al. 2023).Based on these results, we predicted that although the mRNA amounts of RbohB and other Rbohs were suppressed by M. oryzae, the phosphorylation of Rbohs by kinases at specific Ser/Thr sites was a crucial modification that required for the enzyme activation of Rboh proteins.However, whether the other Rbohs except for RbohB were phosphorylated by RLCK118 or OsDMI3 needs further investigation.

Conclusions
In this study, we functionally characterized rice Rbohs in resistance against the most serious disease worldwide and agronomic traits.Our data demonstrated that six Rbohs participated in rice multiple disease resistance and ROS production induced by PAMPs and pathogen; all of the nine Rbohs were involved in the regulation of yield traits and mutations of seven Rbohs led to compromised grain yield per plant.These results revealed the function of the Rbohs family in rice disease resistance and yield, as well as provided experimental reference for breeding application of these genes to coordinate disease resistance and yield traits in rice.

Plant Materials and Growth Conditions
Rice (Oryza sativa) japonica cultivars Taipei309 (TP309), a susceptible line Lijiang xin Tuan Heigu (LTH), and its resistant monogenic line IRBLkm-Ts were used in this study.For disease resistance analysis and yield trait analysis, rice plants were grown in a disease nursery and a paddy yard in Wenjiang, Chengdu, Southwest of China in the normal rice growing season from mid-April to mid-September, respectively.For resistance response analysis, rice plants were grown in a greenhouse (28 ± 2 ℃, 70-75% relative humidity) under 12-h/12-h light/dark cycles.

CRISPR/Cas9 Plasmid Construction and Mutant Screen
The CRISPR (clustered regularly interspaced short palindromic repeats)/Cas9 plasmids were constructed as previously reported (Xie et al. 2015).The guide RNA (gRNA) sequences were screened by the Cas-OFF inder system (http://www.rgenome.net/cas-offinder/)to avoid potential off-target sites with the screen parameters to allow less than three base-pair mismatches and one DNA/ RNA bulge.In brief, to construct the plasmids, the gRNA primers F and R were annealed to generate the gRNA fragment containing sticky ends for cloning.Then the gRNA fragment was inserted into the clone site of the pRGEB32 vector with T4 DNA ligase (NEB).The vector was transformed into Agrobacterium strain EHA105 for rice transformation.To identify the mutation sites of these mutants, the genomic DNA of transgenic plants was extracted and amplified using the primers flanking the designed target site (Table S3).The PCR products (300-500 bp) were sequenced and aligned with the wild type genome sequence to identify the mutation sites.

Reverse-Transcription Quantitative Polymerase Chain Reaction (RT-qPCR)
Rice leaves or panicles at indicated growth stage, or the leaves inoculated with or without M. oryzae were collected for RNA extraction.We extracted the total RNA using the TRIzol reagent (Invitrogen) and conducted the reverse transcription with 1 µg RNA to obtained cDNA using the PrimeScript™ RT reagent Kit with gDNA Eraser following the manufacturer's instruction (TaKaRa Biotechnology, Dalian, China).We performed RT-qPCR using SYBR Green mix (QuantiNova SYBR Green PCR Kit, Qiagen) with the indicated primers (Table S3).To normalize the mRNA amounts of the detected genes, rice Ubiquitin-1 (UBQ) gene was selected as the internal reference.Data were analyzed with the SPSS software using a one-way ANOVA followed by Tukey analysis or Student's t-test with significant differences.

Pathogen Infection and Microscopic Observation
The disease phenotype of blast disease and sheath blight disease were observed in a disease nursery.Blast disease were observed at tillering stage and the fungal lesion length was measured.Sheath blight disease were observed at heading stage and the lesion length on the sheathes was measured for statistics analysis.
The M. oryzae virulent isolate GZ8 were used to analyze rice blast disease resistance in lab conditions.GZ8 were cultivated in petri dish with rice flour medium (20 g L − 1 flour, 2 g L − 1 yeast extract, 15 g L − 1 agar) at 28 ℃ with 12-h/12-h light/ dark cycles.Seven days later, the hyphae growing on the medium were scraped off for sporulation.After 2 days, the spores were collected for punch-inoculation (1 × 10 5 spores/ml) or spray-inoculation (5 × 10 5 spores/ml).Disease lesions were photographed five days post-inoculation (dpi) and the lesion length was measured by image J software.
The R. solani isolate AG-1-IA was used to assess rice sheath blight disease resistance in lab conditions.AG-1-IA was incubated in petri dish with PDA medium for 2 days and cut into 5 mm-diameter circles and inoculated on the leaves at tillering stage.In brief, small lumps of mycelium were placed in the middle of the leaves at 26 ℃ with high-humidity conditions for 2 days.Lesion length on the inoculated leaves were measured for statistics analysis.
The Xoo isolate PXO99 was used to test rice leaf bright disease resistance in a paddy yard.PXO99 was incubated with TSA medium for 1 days and suspended in sterile water for inoculation at concentration O.D. 600 = 1.Leaves at tillering stage were inoculated with PXO99 via cutting at leaf tips by scissors covered with bacteria solution.After 14 days, all inoculated leaves were collected for lesion measurement and analysis.

ROS Burst Assay
We examined the production of ROS triggered by PAMPs flg22 and chitin, respectively.the leaves of the indicated mutant lines were cut into 3 × 3 mm 2 squares with six scrapes and incubated in 200 µl ddH 2 O in a 96-well plate overnight to eliminate basal ROS elicited by damage.Then, the strips were treated with 20 µg/ml chitin in 200 µl reaction buffer (20 mM luminol, 10 µg/ml horseradish peroxidase (Sigma-Aldrich Shanghai Trading Co Ltd, Shanghai, China)).Data were collected as relative luminescence units using a GloMax96 Microplate Luminometer (Promega Biotech Co., Ltd, Beijing, China) for more than 30 min.

O• 2 − and H 2 O 2 Staining
We collected leaves of three-leaf stage plants 48 h post spray-inoculation with M. oryzae for O• 2 − and H 2 O 2 staining.O• 2 − histochemical staining was performed using NBT (Nitrotetrazolium Blue chloride) following a previous report (Wu et al. 2017).Briefly, the leaf samples were spray-inoculated with GZ8 (1 × 10 5 mL − 1 spores) and submerged into 0.5 mg/ml NBT (Sigma).Then the stained leaves were incubated at 37 °C in the dark for 16 h and were de-stained with 95% alcohol at 65℃ to decolor the chlorophyll.The leaf sections were observed with a microscope (Zeiss Axio Imager A2).
DAB (Sigma Aldrich) was used as an indicator.In brief, leaves were collected and submerged into 0.5 mg/ ml 3,3′-diaminobenzidine (DAB) solution and stained overnight.Next, the samples were de-stained with 95% alcohol at 65℃ to decolor the chlorophyll.Then the decolored samples were observed and pictured by a fluorescence microscope (Zeiss Axio Imager A2) or a stereomicroscope (OLYMPUS SZX16).

Fig. 1
Fig. 1 Rbohs are responsive to pathogen-associated molecular patterns (PAMPs).A and B, Reverse-transcription quantitative polymerase chain reaction (RT-qPCR) data show the expression patterns of Rbohs in leaves of a susceptible cultivar Lijiang xin Tuan Heigu (LTH, A) and Taipei 309 (TP309, B) at 3-and 6-hours post treatment of PAMPs chitin and flg22, respectively.Data are shown as mean ± SD (n = 3 independent repeats).All the P values were determined by One-way ANOVA analysis, and the colored dots associated with the bars indicate the value of biologically independent samples

Fig. 4
Fig. 4 Mutations of RbohA, RbohB, RbohC, RbohE, RbohH, and RbohI lead to compromised leaf blight disease resistance in rice.(A) Disease symptoms of the rboh mutants and the TaPei309 (TP309) control 14 days post-inoculation of Xanthomonas oryzae pv.oryzae (Xoo) isolate PXO99 in a paddy yard at heading stage.Scale bars = 5 cm.(B) Quantitative lesion lengths measured as in (A).Data are shown as mean ± SD (n > 11).All the P values were determined by One-way ANOVA analysis, and the colored dots associated with the bars indicate the value of biologically independent samples

Fig. 5
Fig. 5 Mutations of RbohA, RbohB, RbohC, RbohE, RbohH, and RbohI suppress rice ROS production.(A, B) chitin-and flg22-induced reactive oxygen species (ROS) bursts in the leaves of the Rboh mutant lines and the TaPei309 (TP309) control.Data are shown as mean ± SD (n = 4).All the P values were determined by One-way ANOVA analysis, and the colored dots associated with the bars indicate the value of biologically independent samples.(C) O• 2 − and H 2 O 2 accumulation in the indicated mutants and the TP309 control at 48 h post-inoculation with M. oryzae.Nitrotetrazolium Blue chloride (NBT) was used as an indicator of O• 2 − amounts (blue).3, 3 N-diaminobezidin (DAB) was used as an indicator of H 2 O 2 amounts (reddish-brown).Scale bars = 5 mm

Fig. 6
Fig. 6 Mutations of Rbohs compromise rice yield traits. (A-C), The gross morphology and panicle morphology of the Rboh mutants and the Tapei309 (TP309) controls at the maturation stage.Scale bars, 50 cm for plants and 5 cm for panicles.(D) Panicle number, grain number per panicle, seed setting rates, grain weight, and yield per plant of the indicated lines planted in a paddy yard in the rice-growing season in Chengdu, Sichuan Province, Southwest of China.Data are shown as mean ± SD (n = 5 independent plants).All the P values were determined by One-way ANOVA analysis, and the colored dots associated with the bars indicate the value of biologically independent samples

Fig. 7
Fig. 7 Rboh family members displays diverse spatio-temporal expression patterns during the rice-growing period.(A-F) Reverse-transcription quantitative polymerase chain reaction (RT-qPCR) data show the expression patterns of Rbohs in leaves of the TaPei309 (TP309) plants during the vegetative stage (A-C) and in panicles during the reproductive stage (D-F).Data are shown as mean ± SD (n = 3).All the P values were determined by One-way ANOVA analysis, and the colored dots associated with the bars indicate the value of biologically independent samples

Fig. 8
Fig.8A work model depicting the differential roles of Rbohs in rice multiple disease resistance and growth.All the nine Rboh family members participated in the regulation of rice growth and development, and each Rboh preferred to differential agronomical traits.Pathogen-derived PAMPs (such as chitin and flg22) induced the expression of six RBOHs, including RbohA, RbohB, RbohC, RbohE, RbohH, and RbohI, which subsequently triggered the ROS production.On the one side, the enhanced accumulation of ROS improves rice resistance against both fungal and bacterial pathogens.In the other side, the enhanced ROS accumulation impact the development of panicles and grains leading to compromised yield traits and reduced yield.As a results, all the nine Rbohs participated in the coordination of rice yield and multiple disease resistance