The rice miR171b-SCL6-IIs module controls blast resistance, grain yield,and flowering

Yan Li, Ying Tong, Xiaorong He, Yong Zhu, Tingting Li, Xiaoyu Lin, Wei Mao,Zeeshan Ghulam Nabi Gishkori, Zhixue Zhao, Jiwei Zhang, Yanyan Huang, Mei Pu, Jing Fan,Jing Wang, Wenming Wang,*

a State Key Laboratory of Crop Gene Exploration and Utilization in Southwest China, Sichuan Agricultural University, Chengdu 611130, Sichuan, China

b Rice Research Institute and Key Lab for Major Crop Diseases, Sichuan Agricultural University, Chengdu 611130, Sichuan, China

Keywords:miR171b SCL6-IIs Blast disease resistance Yield trait Flowering

ABSTRACT MicroRNAs (miRNAs) act as regulators of plant development and multiple stress responses.Here we demonstrate that the rice miR171b-SCL6-IIs module regulates the balance between blast resistance,grain yield, and flowering.miR171b-overexpressing rice plants (OX171b) displayed increased rice blast resistance accompanied with enhanced defense responses and late heading, whereas blocking miR171b expression in rice (MIM171) led to greater susceptibility to blast disease, associated with compromised defense responses and early heading.Either overexpressing or silencing of miR171b significantly affected plant height and number of filled seeds per panicle(seed-setting rate),resulting in decreased grain yield.miR171b targets SCL6-IIa, SCL6-IIb, and SCL6-IIc, whose expression was suppressed in OX171b but increased in MIM171.Mutants of SCL6-IIa, SCL6-IIb, and SCL6-IIc all displayed phenotypes like that of OX171b, including markedly increased blast disease resistance, slightly decreased grain yield, and delayed flowering.Amounts of miR171b increased gradually in leaves during the vegetative stage but decreased gradually in panicles during the reproductive stage, whereas SCL6-IIs displayed the reverse expression pattern.Together, these results suggest that the expression of miR171b was time- and space-dependent during the rice growth period and regulated the balance between rice blast disease resistance,grain yield,and flowering via SCL6-IIs,and that appropriate accumulation of miR171b is essential for rice development.

1.Introduction

Rice blast disease, caused by the fungal pathogenMagnaporthe oryzae(anamorphPyricularia oryzaeCavara), is one of the most serious disease threats to global production of rice (Oryza sativaL.), and occurs over the entire rice growth period.The most economical strategy for controlling the disease is breeding rice cultivars with high blast resistance.However, resistance is often associated with reduced plant growth [1].Promisingly, recent studies [2-4] have revealed several genes or loci that can balance rice blast resistance and yield, includingBroad-Spectrum Resistance-Digu 1(bsr-d1),Ideal Plant Architecture 1(IPA1), andPigm.

Increasing evidence supports the notion that microRNAs (miRNAs) act as key regulators of plant stress responses, growth, and development [5].InBrassica rapa, miR1885 directly targets theTIR-NBS-LRRclass of disease resistanceRgeneBraTNL1and suppresses the expression of a photosynthesis-related geneBraCP24,to maintain a balance between growth/development and resistance [41].In rice, several miRNAs function to control resistance toM.oryzaeby regulating the expression of their target genes[6].Whereas miR159, miR160, miR162, miR166k-miR166h,miR7695, and miR398b [7-12] contribute positively to rice blast disease resistance, miR156, miR164a, miR167, miR168, miR169,miR319b, miR396, and miR1873 negatively regulate resistance[13-20].Many miRNAs have also been identified as regulators of rice development and yield components.For example, miR156 controls grain size,grain yield,grain quality,panicle branching,tillering, and plant height [21-23].miR172 regulates spikelet determinacy, floral organ development, and flowering time [24-26].miR393 is involved in the regulation of tillering and flowering[27].miR396 regulates panicle branching, grain size, and grain yield[28,29].miR397 controls grain size, grain number, and grain yield[30].miR444 is involved in tiller development [31].Several miRNAs regulate the tradeoff between blast disease resistance and growth.Overexpression of miR162 increases resistance but compromises grain yield, being associated with reduced numbers of filled kernels per plant, whereas silencing miR162 increases grain yield but reduces blast resistance [8].Silencing miR396 increases both rice yield and resistance,and overexpression of miR396 target genes leads to increased blast disease resistance without yield penalty[18].Silencing miR156fhl-3p also increases resistance without yield penalty [13], and silencing miR168 increases grain yield and resistance, as well as accelerating rice flowering [20].

miR171 is a conserved miRNA family in plant involved in development and biotic/abiotic stress responses via downregulation of the expression ofGRASfamily genes.GRAS is a large family of transcription factors, named after GA INSENSITIVE (GAI), REPRESSOR OF GA1-3 (RGA), and SCARECROW (SCR), the first three members functionally identified in this family [32-34].In rice, overexpression of miR171b also increases resistance to rice stripe virus(RSV) associated with suppressed expression ofSCL6-IIs, whereas silencing of miR171b facilitates RSV infection, accompanied by increased expression ofSCL6-IIs [35].miR171 is also responsive to stress from the heavy metal chromium (Cr) in rice.In response to short-term (less than 24 h) Cr stress, the accumulation of miR171 was markedly decreased, whereas, after prolonged Cr stress, from 24 h onward, miR171 expression was increased [36].Overexpression of miR171b also leads to a longer vegetative growing period and larger panicles with more spikelets [35], indicating that miR171b functions in the regulation of grain yield and growth period.However, it is still unclear whether miR171 regulates rice resistance toM.oryzaeor whether miR171 regulates rice resistance, yield traits, and growth period viaSCL6-IIs.

In a previous RNA-seq analysis[7],we found thatmiR171b/c/d/e/fwere differentially responsive toM.oryzaestrains in a susceptible rice accession, Lijiangxin Tuan Heigu (LTH), and a resistant accession, International Rice Blast Line Pyricularia-Kanto51-m-Tsuyuake(IRBLkm-Ts).IRBLkm-Ts contains a single resistance gene locus,Pyricularia-Kanto 51-m(Pikm).In the present study,we try to achieve functional characterization of miR171 and its target genes in rice blast disease resistance and yield traits.We constructed transgenic lines overexpressing miR171b or expressing a target mimic of miR171, as well as knockout mutants of three miR171 target genes.We then performed rice blast disease assays and the yield traits measurement in these lines to analyze the roles of miR171b-SCL6-IIs model in regulating rice resistance and yield.We also examined the expression pattern of miR171b and the target genes throughout the whole growth period of rice plants to reveal the relationship between miR171b-SCL6-IIs model and rice resistance and growth.

2.Materials and methods

2.1.Plant materials and growth conditions

The rice (Oryza sativa japonica) accession Nipponbare was used to construct transgenic lines OX171b and MIM171.Thejaponicaaccession Zhonghua 11 (ZH11) was used to construct knockout mutant lines of miR171b target genes.All controls and transgenic and mutant lines were grown in a greenhouse at 28±2 °C with 70% relative humidity and 12-h/12-h light/dark cycles for resistance and defense response assays.For yield traits assays, all the controls, mutants, and transgenic lines were planted in a paddy field during the rice growing season in Wenjiang district, Sichuan province, China, using standard agronomic inputs and conditions usually employed by local farmers.

2.2.Plasmid construction and transformation

The genomic sequence ofMIR171bwas amplified from Nipponbare genomic DNA with primers miR171b-KpnI-F and miR171b-SalI-R, containing 431-bp upstream and 405-bp downstream sequences of the miR171b mature sequence(Table S1).The amplified fragments were digested and cloned into theKpnI-SalI sites of the binary vector 35S-pCAMBIA1300,resulting in a construct overexpressingmiR171b(OX171b).

The MIM171 construct was constructed as described previously[16,37].A target mimic double-stranded fragment of miR171(GATATTGGCACACAGGCGCAATCA) was cloned into theKpnI-SalI sites of the binary vector 35S-pCAMBIA1300-IPS1, to substitute the target site of miR399.The target mimic fragments were obtained by annealing with primers MIM171-BamHI-F and MIM171-BglII-R (Table S1).

The mutants of the target genes were constructed using clustered regularly interspaced short palindromic repeats(CRISPR)/Cas9 plasmids as described in a previous report[3].The guide RNA sequences of target gene, listed in Table S1, were screened with the Cas-OFFinder system (http://www.rgenome.net/cas-offinder/) to avoid potential off-target sites, with screening parameters including fewer than three base-pair mismatches and one DNA/RNA bulge.

TheAgrobacterium tumefaciensstrain EHA105 was used for rice genetic transformation.Hygromycin B was used for selecting transgenic plants for hygromycin resistance.To identify mutations,genomic DNA was extracted from the transgenic lines and the primers flanking the designated target sites (Table S1) were used for PCR amplification.The PCR products(500-600 bp)were sequenced and aligned against the wild-type genome sequence to identify the mutation sites.

2.3.Pathogen cultivation

TheM.oryzaestrains Guy11, GFP-tagged Zhong8-10-14 (GZ8),CRB1, and 97-27-2 were used to test rice resistance and defense responses.Strain Guy11 is virulent tojaponicaaccessions and is used to assess the resistance of rice accessions worldwide.Strains 97-27-2 and CRB1 are virulent tojaponicaaccessions and were isolated from a paddy in north China.Strain GZ8 is a green fluorescent protein (GFP)-tagged Zhong10-8-14, which is virulent tojaponicaaccessions and was isolated from a paddy in north China.Strains YC32, DZ108, and DZ67 are virulent tojaponicaaccessions and were isolated from a paddy in Sichuan province, southwest China.Strains were cultured on solid medium plates containing oatmeal and tomato(1 L of medium contained 150 g tomato,50 g oatmeal,0.6 g CaCO3, and 15 g agarose)at 28 °C with 12-h/12-h light/ dark cycles.Ten days later, the hyphae growing on the medium were scraped and the plates were then cultured at 28°C under continuous 24-h light to induce sporulation.After five days, the spores were collected and suspended (1 × 105spores mL-1) in ddH2O for resistance and defense response assays.

2.4.Inoculation and microscopy assay

Three-to five-leaf-stage seedlings of the controls,mutants,and transgenic lines were inoculated with or without the indicatedM.oryzaestrains.For punch inoculation, the leaves of seedlings (10 seedlings for each line) were slightly wounded with a mouse-ear punch, and 5 μL of spore suspension (1 × 105spores mL-1) was punch-inoculated on the wound.For spray-inoculation,more than 20 seedlings of each line were spray-inoculated with spore suspension (1 × 105spores mL-1) and covered with a thin film to moisturize the seedlings and thereby improve infection.The inoculation assays were repeated at least two times.For resistance assays, symptoms on infected leaves were recorded at 5-7 days post-inoculation (dpi).The relative fungal biomass on the infected leaves was quantified as previously described [38].Briefly, inoculated leaves(three repeats per line)were collected for DNA extraction.The relative fungal biomass was determined using the ratio of DNA levels of theM.oryzae MoPot2gene to that of the riceubiquitin(UBQ) gene by quantitative real-time polymerase chain reaction.

For detection of H2O2amounts,spray-inoculated leaves were collected 48 h post-inoculation (hpi) and placed in 3, 3′-diaminobenzidine(DAB)(1 mg mL-1;Sigma-Aldrich,USA)for 12 h staining at room temperature in darkness.The stained leaves were decolored in 90%ethyl alcohol at 65°C several times[39].H2O2accumulation was photographed with a stereomicroscope(Zeiss Imager A2,Zeiss,Germany).For detection and quantification of the expression of defense-associated genes,the spray-inoculated leaves were collected at indicated time points for reverse-transcription quantitative polymerase chain reaction(RT-qPCR)assay.

2.5.RT-qPCR assay

Total RNA of the examined samples was extracted using TRIzol reagent(Thermo-Fisher,USA)and was reverse transcribed to cDNA using the PrimeScriptTM RT Reagent Kit with gDNA Eraser(Takara Biotechnology, China).RT-qPCR was performed using SYBR Green Mix (QuantiNova SYBR Green PCR Kit, Qiagen, China) and the appropriate primers (Table S1).The riceUBQgene was selected as an internal reference by which to normalize the expression of genes.For detection of miR171b levels, total RNA was used for reverse transcription (RT) with miR171b-specific stem-loop RT primers (Table S1).The RT product was used as a template for real-time PCR to detect the amounts of miR171b.The small nuclear RNA(snRNA)U6 was selected as an internal reference by which to normalize the amounts of miR171b.Expression values were compared by one-way ANOVA, followed by post-hoc Tukey HSD pairwise multiple comparisons at a significance level ofP< 0.01.

2.6.Yield trait measurements

The controls (Nipponbare and ZH11), the transgenic lines(OX171b and MIM171), and the mutants (scl6-IIa,scl6-IIb, andscl6-IIc) were planted in a paddy field during the regular sowing seasons in 2019 and 2020 at Wenjiang district, Sichuan province, and Lingshui, Hainan province, China.Each line was represented by 30 plants.The agronomic traits of plants in the Nipponbare background were evaluated as the mean of five selected plants and those in the ZH11 background as the means of 10 plants.The yield trait measurements were repeated in 2019 and 2020.Plant morphology was recorded at the heading and full-maturity stages.Yield components from the selected 10 plants of each line were measured, including plant height, panicle number per plant, seed setting rate (the ratio of filled to total kernels per plant), kernel number per panicle,1000-kernel weight, and yield per plant.The 1000-kernel weight and grain yield per plant were measured with an SC-A grain analysis system (Wanshen Ltd., Hangzhou, China) after the filled kernels were dried in a 42 °C oven for one week.All yield trait data were analyzed by one-way ANOVA, followed by post-hoc Tukey HSD pairwise multiple comparisons at a significance level ofP< 0.05.

3.Results

3.1.Rice miR171b is responsive to challenge by M.oryzae

We first studied the responses of miR171 to the challenge of the rice blast fungus.LTH was highly sensitive to Guy11, exhibiting large disease lesions, whereas IRBLKm-Ts proved to be resistant to Guy11,exhibiting small lesions(Fig.1A).The expression pattern of miR171b was different between LTH and IRBLkm-Ts.In LTH,miR171b abundance was significantly decreased 12 hpi with Guy11, then recovered to being not significantly different from the mock sample 24 hpi(Fig.1B).In IRBLkm-Ts,the expression pattern of miR171b was opposite to that in LTH:miR171b abundance was significantly increased in IRBLkm-Ts 12 hpi and then decreased to a similar level 24 hpi in comparison with the mock sample (Fig.1B).These results suggested that miR171b might be involved in rice resistance againstM.oryzae.

3.2.Overexpression of miR171b increased rice resistance to M.oryzae

Transgenic lines overexpressing miR171b (OX171b) showed significantly greater miR171b abundance than the Nipponbare control (Fig.2A).We examined the sensitivity of OX171b to three virulentM.oryzaestrains, namely 97-27-2, GZ8, and Guy11.The OX171b lines were less susceptible than the Nipponbare control to these strains, exhibiting smaller disease lesions and decreased fungal biomass, following either punch (Fig.2B, C) or spray(Fig.S1) inoculation.In response to inoculation with YC32,DZ108, and DZ67, the OX171b lines again showed smaller disease lesions than did the control (Fig.S2).Thus, overexpression of miR171b increased rice blast disease resistance.

3.3.Silencing of miR171b compromised resistance to M.oryzae in rice

To further analyze the roles of miR171b in rice resistance, we examined the resistance of the transgenic lines silencing miR171b by expressing a target mimic of miR171(MIM171)(Fig.S3).Unlike OX171b, MIM171 showed significantly lower miR171 abundance than the Nipponbare control (Fig.3A), and compromised resistance,resulting in larger disease lesions and greater fungal biomass following inoculation with 97-27-2, GZ8, and Guy11 (Figs.3B, C,S1), as well as the strains isolated from southwest China(Fig.S2).These results demonstrated that miR171b positively regulates rice blast disease resistance.

In agreement with the disease phenotypes, the infection process of GZ8 was delayed in OX171b but accelerated in MIM171 in comparison with the Nipponbare control.The spores had developed into invasive hyphae and invaded the local cells in MIM171 24 hpi, and the percentage of spores invading cells was significantly higher than in either the Nipponbare control or OX171b(Fig.S4A, B).The spores had invaded the second and neighboring cells 36 and 48 hpi, and the percentage was significantly higher in MIM171 but lower in OX171b than in the Nipponbare control(Fig.S4A, B).Thus, miR171 delayed the invasion process of the blast fungus.

3.4.miR171b positively regulates M.oryzae-triggered defense responses in rice

To investigate how miR171b positively contributed to rice resistance againstM.oryzae, we examined the induction of defenserelated genes and H2O2accumulation in OX171b and MIM171 following inoculation of GZ8.NAC DOMAIN-CONTAINING PROTEIN 4(NAC4), andPATHOGENESIS-RELATED GENE 1(PR1), areM.oryzaeinduced genes [8].The mRNA levels ofNAC4were significantly greater in OX171b 12 hpi in comparison with that in the Nipponbare control, but significantly lower in MIM171 12, 24, and 48 hpi(Fig.4A).Similarly,the expression ofPR1in response toM.oryzaewas significantly increased in OX171b in comparison with that in the Nipponbare control 24 and 48 hpi but markedly decreased in MIM171 12 and 48 hpi (Fig.4A).

Fig 1.miR171b is responsive to the rice blast fungus Magnaporthe oryzae.(A) The disease phenotype of Lijiangxin Tuan Heigu (LTH) and International Rice Blast Line Pyricularia-Kanto51-m-Tsuyuake(IRBLkm-Ts)following inoculation with M.oryzae.(B)The miR171b levels in LTH and IRBLKm-Ts with or without M.oryzae inoculation.The miRNA levels were determined by reverse transcription quantitative polymerase chain reaction.Values are means±SD(n=3).Hpi,hours post-inoculation.Pairs of samples with shared letters above the column were not different at P > 0.01 as determined by one-way ANOVA followed by Tukey HSD test.

Fig 2.miR171b increases rice resistance against Magnaporthe oryzae.(A) The miR171b levels in the transgenic lines overexpressing MIR171b gene (line OX171b) and the Nipponbare(NPB)control.The miR171b levels were measured by reverse transcription quantitative polymerase chain reaction.Total RNA was used for reverse transcription(RT)with miR171b-specific stem-loop RT primers(Table S1).The RT product was used as a template for real-time PCR to detect the amounts of miR171b.The amount of the small nuclear RNA (snRNA) U6 was used as an internal reference for normalizing expression.(B) Disease symptoms on leaves of miR171b-overexpressing transgenic line OX171b, following inoculation with M.oryzae strains GZ8 or 97-27-2.The photo was taken six days post-inoculation (dpi).Scale bars, 5 mm.(C)Relative fungal biomass of indicated strains on the Nipponbare control and OX171b.The fungal biomass was presented as the ratio of DNA levels of M.oryzae MoPot2 to the DNA levels of the rice ubiquitin gene.Error bars indicate SD(n =3).Two samples with shared letters above the column are not different (P>0.01) as determined by one-way ANOVA followed by Tukey HSD test.

Accumulation of H2O2is a key defense response againstM.oryzae.In the Nipponbare control, H2O2accumulated in the cell following the invasion ofM.oryzae, whereas, in OX171b, H2O2accumulated not only locally in the cell invaded, but also abundantly in adjacent cells.In MIM171, H2O2was detected only in cells invaded by the blast fungus (Fig.4B).These results indicated that miR171b boosted the resistance responses triggered byM.oryzae.

3.5.miR171b regulates rice growth and grain yield

We next investigated the roles of miR171b in the regulation of rice growth and yield traits.We measured the plant height,flowering time, and yield components of these lines, including panicle number, kernel number per panicle, seed setting rate (SSR),1000-kernel weight,and grain yield per plant.OX171b plants were taller with later heading and more kernels per panicle than the Nipponbare control when planted in a paddy field during the regular sowing season.In contrast, MIM171 plants were shorter with earlier heading,and fewer kernels per panicle(Fig.5A-C;Table S2).These results resemble those from a previous study [35].Both OX171b and MIM171 showed panicle number and grain weight like those of the Nipponbare control, but markedly lower SSR,resulting in significantly reduced grain yield per plant (Fig.5DG;Table S2).Thus,overexpression of miR171b delayed rice flowering and compromised rice yield, whereas blocking miR171 fastened flowering but suppressed grain yield.

3.6.Mutations in SCL6-IIs increase rice resistance to M.oryzae

miRNAs regulate plant stress responses and development via the suppression of the expression of target genes.At least nine rice genes have been identified as targets of miR171b via degradome analysis, including several genes encoding SCARECROW-like (SCL)proteins:Os02g44370(SCL6-IIa),Os02g44360(SCL6-IIb), andOs04g46860(SCL6-IIc) [40].TheseSCL6-IIswere further confirmed as target genes of miR171b by 5′-rapid amplification of cDNA ends(RACE) analysis (Fig.S5, [35]).The mRNA levels of theseSCL6-IIswere significantly lower in OX171b lines than in the Nipponbare control but significantly higher in MIM171 (Fig.6A, B), confirming their by miR171b.

Fig 3.Silencing miR171 increases rice susceptibility to Magnaporthe oryzae.(A) The miR171b levels in the transgenic lines overexpressing a target mimic of miR171b(MIM171) and the Nipponbare (NPB) control.The miR171b levels were measured by reverse transcription quantitative polymerase chain reaction.Total RNA was used for reverse transcription (RT) with miR171b-specific stem-loop RT primers (Table S1).The RT product was used as a template for real-time PCR to detect and quantify the amounts of miR171b.The small nuclear RNA(snRNA)U6 was used as an internal reference.(B)Disease symptoms of MIM171 upon inoculation with M.oryzae strains GZ8 or 97-27-2.The photos were taken five days post-inoculation(dpi).Scale bars,5 mm.(C) Relative fungal biomass on the Nipponbare control and MIM171.Fungal biomass was expressed as a ratio of DNA levels of M.oryzae MoPot2 to DNA levels of the rice ubiquitin.Error bars indicate SD (n = 3).Pairs of samples with a common letter above the column are not different (P > 0.01) as determined by one-way ANOVA followed by Tukey HSD test.

Fig 4.miR171b enhances rice defense responses.(A)The mRNA levels of resistance-associated genes NAC4 and PR1 in the Nipponbare (NPB)control, OX171b and MIM171,following inoculation with M.oryzae.The rice ubiquitin gene was used as an internal reference.The mRNA levels were normalized to those in the Nipponbare control 0 h postinoculation(hpi).Error bars indicate SD(n=3).Pairs of samples with a common letter above the column were not different(P>0.01)as determined by one-way ANOVA and Tukey HSD test.(B)Accumulation of hydrogen peroxide(H2O2)in the Nipponbare control,OX171b,and MIM171 48 hpi with M.oryzae strain GZ8.The amount of H2O2 was revealed by staining with 3,3′-diaminobenzidine(DAB)and quantified by the intensity of the brown stain.Black arrows indicate appressoria and red arrows indicate spores.The photos in the upper portion were taken with a stereomicroscope.Scale bars,1 mm.The photos in the lower portion were taken with a microscope(Zeiss Imager A2).Scale bars, 40 μm.

To further study the roles ofSCL6-IIa/b/cin the regulation of rice resistance and yield, we constructed transgenic lines with knockout mutations in these genes, using CRISPR/Cas9 gene-editing technology.We identified one homozygous mutant forSCL6-IIa(scl6-IIa-1), two forSCL6-IIb(scl6-IIb-1andscl6-IIb-2), and two forSCL6-IIc(scl6-IIc-1andscl6-IIc-2).All five mutants carried insertions or deletions causing protein truncations (Figs.S6-S8).We then assessed the sensitivity of these mutants toM.oryzae.Like OX171b, all mutants displayed increased resistance, with smaller disease lesions and less fungal biomass following inoculation withM.oryzaestrains GZ8, CRB1, and Guy11 than the wild-type ZH11 control (Figs.6C-E, S9), indicating negative roles forSCL6-IIa/b/cin resistance toM.oryzaein rice.These results indicated that miR171b regulated rice blast disease resistance viaSCL6-IIs.

Fig 5.miR171b regulates rice grain yield traits.(A)Gross morphology of the Nipponbare(NPB)control,OX171b,and MIM171 at heading and full-maturity stages.Scale bars,50 cm.(B-G)Plant height,kernel number per panicle,number of panicles,1000-kernel weight,seed setting rate,and grain yield per plant of the Nipponbare control,OX171b and MIM171.Values are means±SD(n=5 independent plants).Pairs of samples with a common letter above the bars are not different(P>0.05)as determined by one-way ANOVA and Tukey HSD test.The experiments were performed twice in 2019 and 2020 with similar results.

3.7.Mutations in SCL6-IIs affect yield traits and delay rice flowering

We also studied the roles ofSCL6-IIa/b/cin regulating rice growth and yield traits using the mutants.When planted in a paddy field during the regular sowing season,all five mutants displayed gross morphology like that of the ZH11 control, although plant heights varied(Fig.7A).Althoughscl6-IIaplants were slightly taller than ZH11 plants,scl6-IIb, andscl6-IIcwere significantly shorter than ZH11(Fig.7A,B;Table S3).We further compared their yield components and flowering time.Like OX171b,scl6-IIa,scl6-IIb, andscl6-IIcshowed panicle number and grain weight similar to the corresponding values in the ZH11 control (Fig.7C;Table S3), but showed increased kernel number per panicle but decreased SSR leading to reduced yield per plant (Fig.7D-F;Table S3).

All the mutants flowered later than the ZH11 control,indicating longer vegetative growth (Fig.7G).Their delayed flowering was consistent with the delayed heading of miR171b-overexpressing lines in this study (Fig.5A) and a previous study [35], indicating that miR171b regulated the rice growth period viaSCLs.

3.8.miR171b and SCL6-IIs are dynamically expressed during the growth period

Grain yield, growth duration, and disease resistance are three key factors in rice production that compete with one another.Our results suggested that miR171b regulates these factors viaSCL6-IIs.To determine how the ‘‘miR171b-SCLIIs” module coordinated the three key factors, we performed a time-course experiment and examined the expression of miR171b andSCL6-IIsin rice leaves and panicles.RT-qPCR analysis revealed that the accumulation of miR171b in leaves increased gradually as the seedlingsgrew, whereas the mRNA levels of theSCL6-IIsdecreased correspondingly(Fig.8A).In contrast,after the conversion of the reproductive stage from the vegetative stage, the accumulation of miR171b in panicles decreased gradually, whereas the mRNA levels of theSCL6-IIsincreased (Fig.8B).It may be that during the vegetative stage, the increased accumulation of miR171b in leaves increases resistance by suppressing the expression ofSCL6-IIs,which reduces blast disease resistance.In contrast,during the reproductive stage, the decreased accumulation of miR171b increased the expression ofSCL6-IIs, the appropriate levels of whose expression were required for normal development of panicles and flowering(Fig.8C).Thus,miR171b-SCL6-IIsmodule acts in regulating resistance, grain yield, and flowering viaSCL6-IIs.

Fig 6. SCL6-II genes are suppressed by miR171b and negatively regulate rice blast disease resistance.(A)Relative mRNA levels of SCL6-II genes in OX71b lines in comparison with those in the Nipponbare (NPB) control.(B) Relative mRNA levels of SCL6-II genes in MIM171 lines in comparison with those in the Nipponbare control.(C) Disease phenotypes of indicated mutants and the Zhonghua 11(ZH11) control at five days post-inoculation (dpi) with M.oryzae strains GZ8 and CRB1.Bar,5 mm.(D, E)The fungal biomass of GZ8 and CRB1 in(C).Relative fungal biomass is shown as a ratio of the DNA level of the M.oryzae MoPot2 gene to that of the rice ubiquitin gene.For(A),(B),(D),and(E), error bars indicate SD(n=3).Two samples with a common letter above the bars are not different (P >0.01)as determined by one-way ANOVA and Tukey HSD test.All experiments were performed twice, with similar results.

Fig 7. SCL6-IIs regulate rice yield components and flowering time.(A) Gross morphology of the Zhonghua 11 (ZH11) control and the scl6-II mutants at the maturity stage.Scale bars,50 cm.(B-G)The plant height,number of panicles,kernel number per panicle,seed setting rate,grain yield per plant,and flowering time of the ZH11 control and the scl6-II mutants.Values are means±SD (n = 10 independent plants).For any two samples, a shared letter above the bars indicates no significant difference (P > 0.05) as determined by one-way ANOVA and Tukey HSD test.The experiments were performed twice in 2019 and 2020 with similar results.

4.Discussion

As fine-tuning agents,miRNAs can regulate the defense-growth tradeoff in plants.In this study,we revealed that the miR171-SCL6-IIs module regulates the balance between blast disease resistance and yield, and controls flowering in rice, and identified potential candidate genes for rice breeding to increase rice blast resistance with negligible yield penalty.As a result, miR171b controls the trade-off between rice resistance and yield viaSCL6-IIs.

A previous study[35]had shown that miR171b was involved in plant responses to rice stripe virus infection and the regulation of disease symptoms, suggesting that miR171b was a regulator of plant resistance.Although that study identifiedSCL6-IIa,SCL6-IIb,andSCL6-IIcas targets of miR171b in rice, no genetic evidence was provided to demonstrate the involvement of these target genes in rice disease resistance.In this study, we demonstrated that we could increase rice blast resistance by blocking the expression ofSCL6-IIa,SCL6-IIb,orSCL6-IIc(Figs.6,S9),indicating the negative roles of these genes in plant resistance.SCL proteins such as SCL27(SCL6-II)and SCL3 have been shown to interact with DELLA proteins, the master repressors in gibberellin (GA) responses inArabidopsis[42,43].DELLAs could release the transcription factor MYC2 to activate jasmonate (JA) responses by binding to JA ZIM domains(JAZ)[44].SCL6-IIa/b/c in rice may also interact with DELLAs and interfere with the DELLA-JAZ association and JA responses.JA synthesis and signaling components have been suggested[17]to be involved in rice blast resistance.

Fig 8. miR171b-SCL6-IIs module is dynamically expressed during the rice growing period.(A)Time course of the accumulation of miR171b and the mRNA levels of SCL6-IIs in leaves in Nipponbare plants during the vegetative stage.(B) Time course of the accumulation of miR171b and the mRNA levels of SCL6-IIs in panicles in Nipponbare plants during the reproductive stage.For A and B,error bars indicate SD(n=3).Any two samples with a common letter above the bars are not different(P>0.01)as determined by one-way ANOVA and Tukey HSD test.(C) A model of miR171b regulation of the expression of target genes during the rice growth period.During the vegetative stage,miR171b is upregulated and suppresses SCL6-IIs to increase disease resistance.During the reproductive stage, miR171b is downregulated and the expression of SCL6-IIs is upregulated to manipulate the normal development and flowering of panicles.

It is well known thatSCLsare involved in plant growth and development processes such as seed germination, hypocotyl and root elongation, chlorophyll biosynthesis in leaves inArabidopsis[42,43], and flowering and fruit set in tomato [45].In the present study, mutants ofSCL6-IIsshowed increased rice resistance,decreased yield, and delayed flowering (Figs.6, 7), suggesting the multiple roles of SCL family members in rice development and resistance.The mutants ofSCL6-IIsexhibited a phenotype like those of the transgenic lines overexpressing miR171b (Figs.2-5),indicating that miR171b regulated these phenotypes via the three target genes.However, the mutants displayed slightly decreased SSR and grain yield, but this decrease was less than that of the OX171b lines.Moreover, the decreased plant height ofscl6-IIbandscl6-IIcmutants was inconsistent with that of OX171b, which was taller than the Nipponbare control.These results suggested the presence of other target genes involved in the regulation of rice yield components and plant height, or miR171 may coordinately control rice yield and plant height via the three target genes and some other target genes not identified in this study.For example,several other genes have been predicted to be targets of miR171[40], includingOs10g40390andOs06g01620encoding other SCL6-II proteins,Os03g04300encoding an ankyrin repeat protein,Os05g34460encoding a Do-like 4 protease, and so on.Whether these genes play roles in rice resistance and yield awaits further study.

Although only onescl6-IIcmutant showed a significant yield loss, allscl6-IIsshowed increased rice blast resistance.These results provide a promising target for rice breeding to improve rice resistance without a yield penalty (Figs.6, 7).We hypothesized that,similar toArabidopsisSCL27(SCL6-II),rice SCL6-IIa/b/c associates with DELLAs, regulating DELLA-EDS1 and/or DELLA-JAZ modules that can control the plant growth-defense tradeoff [44,46].The DELLA-EDS1 module can avoid excessive defense or growth via a feedback regulatory loop, modulating salicylic acid (SA)-mediated defense and GA-mediated plant growth.The DELLA-JAZ module can regulate GA-JA hormone crosstalk, thus balancing growth and JA-mediated defense.It will be desirable to determine the associations between rice SCL6-IIa/b/c and DELLAs and the effects of SCL6-IIa/b/c on DELLA-EDS1 and DELLA-JAZ interactions.

CRediT authorship contribution statement

Yan Li, Jing Fan, andWenming Wangconceived the experiment, and together withYing Tong, Xiaorong He, Yong Zhu,Xiaoyu Lin, Tingting Li, Zeeshan Ghulam Nabi Gishkori, andWei Maocarried it out;Zhixue Zhao, Jing Wang, andYanyan Huanganalyzed the data;Jiwei ZhangandMei Pucarried out the field trial;Yan LiandWenming Wangwrote the paper.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We thank Dr.Cailin Lei (Institute of Crop Sciences, Chinese Academy of Agricultural Sciences) for providing the monogenic resistant line IRBLkm-Ts.This work was supported by the National Natural Science Foundation of China (U19A2033, 31672090, and 31430072) and the Sichuan Applied Fundamental Research Foundation (2020YJ0332) to Wenming Wang.

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2021.05.004.