Jiling Lu ,Chuncho Wng ,Fn Zhng ,c,Dn Zeng ,Yongli Zhou ,*

a National Key Facility for Crop Gene Resources and Genetic Improvement,Institute of Crop Sciences,Chinese Academy of Agricultural Sciences,Beijing 100081,China

b College of Agronomy and Biotechnology,China Agricultural University,Beijing 100193,China

c Graduate School of the Chinese Academy of Agricultural Sciences,Beijing 100081,China

ABSTRACT Bacterial blight(BB),which is caused by Xanthomonas oryzae pv.oryzae(Xoo),is one of the most destructive bacterial diseases of rice(Oryza sativa L.).During plant defense responses,microRNAs(miRNAs)play important roles in regulating disease resistance.However,the functions of miRNAs in the interaction between rice and Xoo remain relatively uncharacterized.In this study,we compared the miRNA profiles of the BB resistant rice introgression line F329 and its susceptible recurrent parent Huang-Hua-Zhan(HHZ) at multiple time points after inoculation with Xoo. A total of 538 known and 312 novel miRNAs were identified,among which only 17 and 26 were responsive to Xoo infection in F329 and HHZ,respectively.Compared with the expression levels in HHZ,37 up-regulated and 53 down-regulated miRNAs were detected in F329.The predicted target genes for the miRNAs differentially expressed between F329 and HHZ were revealed to be associated with flavonoid synthesis,the reactive oxygen species regulatory pathway,plant hormone signal transduction,defense responses,and growth and development.Additionally,the patterns of interactions between osa-miR390-3p,novel_miR_104,novel_miR_238,osa-miR166k-5p,osa-miR529b,and osa-miR167h-3p and their target genes were further validated by quantitative real-time PCR.Furthermore,we overexpressed osa-miR167h-3p in transgenic plants and proved that this miRNA positively regulates the resistance of rice to BB.These results provide novel information regarding the miRNA-based molecular mechanisms underlying the disease resistance of rice.The data presented herein may be useful for engineering rice BB resistance via miRNAs.

Keywords:Rice Bacterial blight miRNA profiling osa-miR167h-3p

1.Introduction

Bacterial blight (BB),caused byXanthomonas oryzaepv.oryzae(Xoo),is one of the most devastating rice diseases in tropical Asia and Africa,where it can cause 20%–80% yield loss [1].An environmentally-friendly,cost-effective,and efficient strategy to control this disease involves breeding and deploying resistant cultivars [2].To date,44 BB resistance genes have been identified in rice [3–6].However,only some of these genes have been widely applied in rice breeding programs because most of the genes are associated with limited disease resistance and some are recessive.Therefore,only some dominant genes conferring broad-spectrum resistance (e.g.Xa3,Xa4,Xa21,andXa23) have been commonly applied in China and other Asian countries [7–10].The narrow genetic basis for resistance in rice cultivars leads to strong selection pressure on the pathogen population,resulting in a dramatic increase in the frequency of new virulent strains that can overcome the resistance [11].Thus,to ensure effective disease management strategies are applied to control BB,rice defense mechanisms against BB should be more thoroughly elucidated.

Accumulating data have revealed that microRNAs (miRNAs)help regulate plant disease resistance[12,13].Specifically,miRNAs,which are typically 20–24 nt long may bind to partially complementary sequences in target messenger RNAs(mRNAs)and repress gene expression at both the transcriptional and posttranscriptional levels via transcript cleavage and translation repression [14,15].A number of miRNAs contribute to defense responses inArabidopsis thaliana[16–18].In rice,the overexpression of miR7695 promotes the resistance to rice blast caused by the pathogenic fungusMagnaporthe oryzaeby suppressing the expression of the gene encoding natural resistance-associated macrophage protein 6 (OsNramp6) [19].The enhanced rice blast resistance of plants overexpressing miR160a or miR398b is reportedly associated with an increase in hydrogen peroxide accumulation and the upregulated expression of defense-related genes [20].Additionally,miR444 and miR528 are involved in rice immune responses to viruses [21,22].

Although some miRNAs have been characterized to regulate disease resistance,the roles of miRNAs involved in the interaction between rice andXooremain relatively unknown.Initially,several miRNAs responsive toXoowere identified following the analysis of dynamic expression-level changes to miRNAs in the susceptible rice variety Nipponbare.A previous study indicated that miR159a.1 might directly regulate rice responses to pathogens by targetingOsLRR-RLK2[23].Hong et al.[24] performed simultaneous genome-wide analyses of the expression of miRNAs and small interfering RNAs involved in the regulation ofXa3/Xa26-mediated resistance by comparing the small RNA (sRNA) between theXa3/Xa26transgenic rice line and its wild-type (WT).They uncovered some novel miRNAs potentially involved in the rice–Xoointeraction.Moreover,the overexpression of miR169o in transgenic rice plants promotes nitrogen use efficiency by regulatingNRT2level,while also increasing the susceptibility to BB by decreasing the transcription of defense genes [25].

In a previous study,we revealed that introgression line (IL)FF329 (F329),from a BC1F4population derived from a cross between the recipient Huang-Hua-Zhan (HHZ) and the donor PSBRC66,exhibits a typical hypersensitive response to BB [26].In the current study,to further clarify the potential F329 defense mechanism against BB,we performed a genome-wide expression analysis of miRNAs in F329 and HHZ via Illumina high-throughput miRNA sequencing.The differentially expressed miRNAs (DEMs) involved in the interaction between rice andXoowere identified,and correlations between the expression of miRNAs and their target genes were analyzed by quantitative real-time PCR (qRT-PCR).Additionally,transgenic analyses indicated that osa-miR167h-3p positively regulates the BB resistance of rice.

2.Materials and methods

2.1.Plant materials and sample collection

Rice lines F329 and HHZ,which differ significantly regarding their resistance to BB,were selected to identify the miRNAs associated with the rice–Xoointeraction.Seeds for both rice lines were sown in a seedling nursery,after which the resulting seedlings were transplanted to the field (three replicates) at the Institute of Crop Sciences,Chinese Academy of Agricultural Sciences,Beijing,China.The plants were cultivated in the field using routine practices without any germicide applications and in the absence of stress conditions until harvest.

TheXoostrain PXO99Awas used for inoculating rice varieties at the tillering stage(60-day-old plants).TheXoostrain was cultured on potato sucrose agar medium at 28 °C for 48 h.An established leaf-clipping method [27] was used to inoculate the uppermost fully expanded rice leaves with a bacterial suspension 108colony-forming units(CFUs)per milliliter prepared in sterile water.Approximately 1-cm-long leaf fragments adjacent to the inoculation sites were collected at 0,12,24,and 36 h post-inoculation(hpi)from 15 plants per replication for each variety,with two biological replications.The collected samples were immediately frozen in liquid nitrogen and stored at -80 °C.Samples for BBresistant line F329 were designated F329-0h-1,F329-0h-2,F329-12h-1,F329-12h-2,F329-24h-1,F329-24h-2,F329-36h-1,and F329-36h-2,whereas samples for BB-susceptible line HHZ were named HHZ-0h-1,HHZ-0h-2,HHZ-12h-1,HHZ-12h-2,HHZ-24h-1,HHZ-24h-2,HHZ-36h-1,and HHZ-36h-2.The numbers 1 and 2 at the end of sample names represent the first and second biological replications,respectively.

2.2.Small RNA library construction and high-throughput sequencing

A total of 16 RNA samples were extracted with the TRIzol reagent (Invitrogen,Carlsbad,CA,USA) and analyzed qualitatively with the Agilent 2100 Bioanalyzer system.The sRNAs were separated from the total RNA by polyacrylamide gel electrophoresis and ligated to 5′and 3′adapters.The sRNA libraries were prepared by reverse transcription and then sequenced with the Illumina HiSeq 2500 system by CapitalBio Corporation (Beijing,China).

2.3.Small RNA analysis and miRNA identification

Raw reads were filtered to remove low-quality reads,reads containing more than 10% unknown bases,null 5′or 3′adapters,and incorrect sequences shorter than 18 nt or longer than 30 nt.Unique sRNAs acquired by removing adapters were used to screen the Silva (https://www.arb-silva.de/),GtRNAdb (http://gtrnadb.ucsc.edu/),GenBank (https://www.ncbi.nlm.nih.gov/genbank/),Rfam(http://rfam.xfam.org/),and Repbase (https://www.girinst.org/)databases to eliminate rRNAs,tRNAs,snRNAs,snoRNAs,and other ncRNAs and repeats using Bowtie.The clean reads were subsequently aligned to the MSU v7.0 reference rice genome (http://rice.plantbiology.msu.edu/cgi-bin/gbrowse/rice/) using Bowtie,after which the clean reads mapped to exons,introns,or repeat sequences were trimmed.The remaining unique sequences were used as queries to search the miRBase v22.1 database (http://mirbase.org/index.shtml) to identify known miRNAs.Additionally,novel miRNAs were predicted based on the sequence locations in the rice genome and the distribution in precursor sequences as well as miRNA characteristics,including mature,star,and loop sequences,and the precursor structures predicted with RNAfold.

2.4.Analysis of differentially expressed miRNAs

Before analyzing the DEMs,the miRNA counts were normalized as transcripts per million (TPM) with the following formula [28]:TPM=read counts×1,000,000/mapped reads.In this formula,read counts were the number of reads aligned to a particular miRNA,whereas mapped reads were the number of reads aligned to all miRNAs.The DEMs between the two groups were analyzed with the DESeq2 package[29].The resultingP-value was adjusted using Benjamini and Hochberg’s approach to control the false discovery rate (FDR).The miRNAs with FDR <0.05 and |log2(Fold-change)|>1 were considered to be significantly differentially expressed.

2.5.Target gene prediction and functional annotation

The genes targeted by miRNAs were predicted with the psRNATarget server,using an expectation score of 3.0 as the cutoff[30].The gene ontology (GO) enrichment analysis was performed using an online analytical tool (GO classification and enrichment)of BMKCloud (https://international.biocloud.net).

2.6.Expression analysis of miRNAs and target genes via qRT-PCR

The miRNA cDNA was synthesized using the miRcute Plus miRNA First-Strand cDNA kit.The riceU6small nuclear RNA was selected as the reference gene (Table S1).To analyze target gene expression,cDNA was synthesized with the FastKing gDNA Dispelling RT Supermix.A rice ubiquitin geneUBQ5was used as an internal control.Details regarding the primers used are listed in Table S2.The qRT-PCR analysis of the miRNAs and target genes was respectively completed with the miRcute Plus miRNA qPCR kit (SYBR Green) and the SuperReal PreMix Plus (SYBR Green) kit by the Bio-Rad CFX-96 instrument.The 2-ΔΔCtmethod was used for quantifying relative expression levels.The above-mentioned kits were obtained from Tiangen Biotech(Beijing)Co.,Ltd.(Beijing,China).

2.7.Functional analysis of osa-miR167h-3p in transgenic lines

To develop an overexpression construct,the osa-miR167h-3p precursor sequence in the Nipponbare genome was cloned into the modified binary vector pCAMBIA3301 driven by maize ubiquitin promoter at theBamH I/Hind III sites.The construct was inserted into Nipponbare plants byAgrobacterium tumefaciensmediated transformation method.Details regarding the primers used for producing the constructs and examining the transgenic plants are listed in Table S3.The transgene-overexpression plants were inoculated with PXO99Aas described in Section 2.1,after which their phenotypes were compared.Lesion lengths(LLs)were measured for all inoculated leaves at 2 weeks after inoculation.

2.8.Analysis of bacterial growth

To compare the bacterial populations in F329 and HHZ plants,and the osa-miR167h-3p transgenic and WT plants,PXO99Agrowth curves were produced using a slightly modified version of a published procedure [31].Specifically,the whole inoculated leaves (three biological replications) were collected at different time points according to the responses of rice lines after inoculation.These leaf samples were immediately placed on ice and stored under-20°C until further manipulation.The leaves were surfacesterilized by briefly spraying 75% ethanol and then cut into fragments,placed in a mortar with sterilized silica sand and water,and ground to fine powder.The homogenates were transferred to 2-mL sterile Eppendorf tubes and serially diluted before plating on potato sucrose agar medium containing 200 μmol L-1azacytidine.Finally,the number of CFUs was calculated for the inoculated leaves by counting the bacterial colonies for the diluted suspensions.

3.Results

3.1.Phenotypic comparison of F329 and HHZ plants infected with Xoo

In this study,IL F329 and its recurrent parent HHZ were infected withXooto identify the miRNAs associated with the interaction between rice andXoo.At 48 h after the inoculation withXoostrain PXO99A,light brown edges were observed at the inoculation sites on F329 leaves,whereas water-soaked symptoms were visible on the inoculated HHZ leaves.At 2 weeks after the inoculation,F329 plants exhibited a typical hypersensitive response (Fig.1A),with brown necrotic lesions along the clipped sites of leaves and a LL of 0.4 ± 0.1 cm (Fig.1B).In contrast,the HHZ plants were highly susceptible to the infection (Fig.1A),with large withered areas on leaves and a LL of 19.6±1.3 cm(Fig.1B).Consistent with the LL,PXO99Agrew more slowly in F329 than in HHZ (Fig.1C).

3.2.Summary of sequencing data

Approximately 16.6 million raw reads were generated for each sequenced sample.After filtering reads,we obtained at least 8.5 million clean reads per sample.Screening against the MSU v7.0 rice genome revealed that the sRNA length in all samples ranged from 18 to 30 nt,with 21 and 24 nt being the most common lengths(Fig.2A).This result is consistent with previously reported sRNA lengths [24,32].Eliminating sRNAs annotated as exons,introns,repeats,rRNAs,tRNAs,snRNAs,snoRNAs,and ncRNAs enabled the detection of 850 miRNAs (538 known and 312 novel),which were distributed in 142 miRNA families based on their miRNA precursors (Fig.2B,Table S4).The correlation of miRNA abundance between the two biological replicates for each treatment for each variety ranged from 0.9659 to 0.9978 (Fig.S1),reflecting the high quality of the sequencing data.

Among the identified miRNAs,MIR818(comprising 44 miRNAs)was the most abundant miRNA family,followed by MIR812 and MIR395.Regarding the novel miRNAs,MIR818 was the most enriched miRNA family,followed by MIR167_2.A total of 831 and 812 miRNAs were identified in F329 and HHZ,respectively,of which 793 were common in both rice varieties(Fig.S2A).A comparison of the detected miRNAs in F329 and HHZ indicated there was no significant difference in the number of novel miRNAs,and there was only a small difference in the number of known miRNAs (Fig.S2B).A few miRNAs were specifically expressed in one rice line at different time-points after theXooinoculation.Specifically,13,7,6,and 10 miRNAs were expressed only in F329 at 0,12,24,and 36 hpi,respectively (Fig.S2C),whereas 4,14,12,and 7 miRNAs were expressed only in HHZ at the same time-points(Fig.S2D).

3.3.Differentially expressed miRNAs in the resistant response to BB

Compared with the non-inoculated samples collected at 0 hpi,17 (including 12 novel) and 26 (including 16 novel) unique miRNAs were differentially expressed in F329 (Fig.S3A) and HHZ(Fig.S3B)inoculated withXoo,respectively.Most of the DEMs were up-regulated in both incompatible and compatible interactions.Compared with the non-inoculated samples,only one and nine miRNAs were down-regulated in the inoculated F329 and HHZ samples,respectively.Notably,three DEMs (osa-miR1320-5p,osa-miR398a,and novel_miR47) were simultaneously detected in F329 and HHZ at one or more time-points.Of these DEMs,osamiR1320-5p expression was induced in both rice varieties,whereas osa-miR398a expression was down-regulated in HHZ,but up-regulated in F329.The expression of novel_miR47 was down-regulated at 12 and 36 hpi in HHZ and F329,respectively,but slightly up-regulated at 24 hpi in HHZ.

To identify the miRNAs regulating BB resistance,we compared the miRNA expression levels in F329 and HHZ.A total of 90 DEMs were detected between F329 and HHZ after the PXO99Ainfection,with 37 up-regulated and 53 down-regulated miRNAs in F329(Table S5).These DEMs included previously reported diseaserelated miRNAs,including miR167,miR169,miR171,and miR396 [13,25,33,34].Compared with non-inoculated HHZ,24 and 39 miRNAs were up-regulated and down-regulated,respectively,in non-inoculated F329,suggesting that the basal expression levels of these miRNAs differed between BB-resistant and BB-susceptible rice lines.After inoculation withXoo,26 and 16 miRNAs were up-regulated and down-regulated,respectively,in F329 relative to the corresponding expression levels in HHZ at 12 hpi,whereas 33 and 42 miRNAs were up-regulated and down-regulated,respectively,at 24 hpi.In contrast,28 and 41 miRNAs were up-regulated and down-regulated,respectively,at 36 hpi (Fig.3A,B).Of 90 DEMs,22 and 10 were commonly upregulated and down-regulated,respectively,at all four timepoints (Fig.3B).Twelve of the common up-regulated miRNAs belonged to the MIR167_1 family,whereas two of the common down-regulated miRNAs belonged to the MIR167_2 family,implying that members of the MIR167 family may play key roles in the rice response to BB.Interestingly,20%of the up-regulated miRNAs were novel,whereas more than 70% of the down-regulated miRNAs were novel (Fig.3C,D).

Fig.1.Disease symptoms in resistant line F329 and susceptible line HHZ at 2 weeks post-inoculation.(A)Representative leaves infected by Xoo strain PXO99A.(B)Statistical analysis of lesion length.Asterisks indicate the significance between two lines(**,P<0.01,two-tailed Student’s t-test).(C)PXO99A growth curve in F329 and HHZ at 1,2,3,4,and 5 days post-inoculation (dpi).CFU,colony forming units.

Fig.2.Length distribution and annotation of sRNAs.(A)Length distribution of 18–30 nt sRNAs.(B)Annotation of sRNAs.

3.4.Target gene prediction and GO enrichment

To clarify the biological functions of the DEMs in the rice–Xoointeraction,the target genes of DEMs (TDEMs) were predicted.A single miRNA generally had at least one target gene.For 37 upregulated and 53 down-regulated miRNAs,360 and 619 TDEMs,respectively,were predicted to be involved in the rice–Xoointeraction (Table S6).For example,novel_miR_120 was predicted to targetOsGRAS37.In contrast,novel_miR_101 was predicted to have 98 target genes.

GO enrichment analysis was completed to functionally annotate the target genes and identify the pathways associated with the rice–Xoointeraction.For the up-regulated miRNAs,360 target genes were assigned biological process GO terms,including the following:defense response,brassinosteroid mediated signaling pathway,response to gibberellin,and response to cadmium ion(Fig.4A).Moreover,the molecular function-related GO terms among these genes included the following:cysteine-type peptidase activity related to cell apoptosis,transferase activity,protein kinase,auxin binding,and inositol hexakisphosphate binding(Fig.4B).Regarding the down-regulated miRNAs,619 target genes were annotated with GO terms such as defense response to bacterium,cellular response to stress,L-ascorbic acid biosynthetic process,and cell wall organization,among others (Fig.4C).Additionally,these target genes were predicted to influence transferase activity,oxidoreductase activity,aspartic-type peptidase activity,phospholipid-hydroperoxide glutathione peroxidase activity,and other molecular functions (Fig.4D).

On the basis of the GO term enrichment and functional annotation of the TDEMs identified in this study,we formulated a putative working model for the miRNAs contributing to the rice–Xoointeraction (Fig.S4).Some down-regulated miRNAs,including osamiR398a,novel_miR_101,novel_miR_48,osa-miR1432-3p,and osa-miR1320-5p,may regulate flavonoid and reactive oxygen species (ROS) pathways,plant hormone signal transduction,defense responses,and growth and development.The up-regulated miRNAs osa-miR169i-5p.2,osa-miR169f.2,osa-miR167h-3p,and osamiR529b may affect plant hormone signal transduction,WRKY and receptor kinase-associated defense responses,growth and development,and the oxidoreductase-mediated pathway.

3.5.Validation of miRNA and target gene expression

Fig.3.miRNAs differentially expressed between F329 and HHZ at various time-points.(A)37 up-regulated and 53 down-regulated DEMs were detected in F329(relative to HHZ).(B)Venn diagram illustrating the up-regulated and down-regulated DEMs in F329 at four time-points after the Xoo infection.Heat maps of the up-regulated DEMs(C)and down-regulated DEMs(D)at various time-points after the Xoo infection.

Fig.4.Gene ontology(GO)enrichment of the target genes of miRNAs involved in the rice–Xoo interaction.Top 20 GO‘‘biological process”terms associated with the target genes of up-regulated(A)and down-regulated(C)miRNAs.Top 20 GO‘‘molecular function”terms associated with the target genes of up-regulated(B)and down-regulated(D)miRNAs.

To evaluate the accuracy of the sequencing data,the expression levels of 44 randomly selected miRNAs with TPM greater than 100 were analyzed by qRT-PCR.The correlation between the qRT-PCR results and the miRNA-seq data was reflected by a correlation coefficient of 0.8969(Fig.S5).We conducted qRT-PCR assays to analyze the expression levels of six DEMs between F329 and HHZ (osamiR390-3p,novel_miR_104,novel_miR_238,osa-miR166k-5p,osa-miR529b,and osa-miR167h-3p) and their target genes.The qRT-PCR results revealed that osa-miR390-3p,novel_miR_104,and novel_miR_238 expression levels were significantly downregulated in F329 compared with HHZ,whereas the expression levels of the corresponding target genesLOC_Os04g49960(glycosyl transferase domain-containing protein),LOC_Os08g40440(dihydroflavonol-4-reductase),andLOC_Os10g39520(OsMLO4)were up-regulated at 0,12,and 24 hpi,respectively.In contrast,osa-miR166k-5p,osa-miR529b,and osa-miR167h-3p expression levels were significantly up-regulated in F329 compared with HHZ,and the expression levels of their target genesLOC_Os10g23090(Oshox8),LOC_Os08g39890(OsSPL14),andLOC_Os01g74610(armadillo-type fold domain-containing protein)were down-regulated (Fig.5).These results were indicative of a negative correlation between the miRNA and target gene expression,and suggested that the above six miRNAs and their targets may affect the resistance of rice to BB.

3.6.Osa-miR167h-3p increased the BB resistance of the overexpression transgenic lines

Our data revealed that osa-miR167h-3p expression was higher in F329 than in HHZ at four time-points(Fig.3C).Accordingly,this miRNA was further analyzed to determine its regulatory function related to rice resistance to BB.Three independent transgenic lines overexpressing osa-miR167h-3p were analyzed in a conventional PCR using a vector primer and a gene-specific primer as well as in a miRNA qRT-PCR assay(Fig.S6A,B).The resistance of the transgenic lines and WT plants were evaluated using the strong virulentXoostrain PXO99A.The lesions of the osa-miR167-3p overexpression lines were more than 50%shorter than those of the WT plants(Fig.6A,B).Consistent with the observed resistance,the bacteria grew more slowly in the transgenic plants than in the WT plants(Fig.6C),suggesting that osa-miR167h-3p positively regulates the BB resistance of rice.Additionally,the height of the osamiR167h-3p -overexpressing lines was similar to that of the WT plants,but their seed setting rate was lower.The mechanism underlying the functions of osa-miR167h-3p related to rice disease resistance and development should be investigated in future studies.

4.Discussion

In the past few years,an increasing number of studies have proven that altered expressions of miRNA expression may regulate the biotic stress resistance of several plant species,includingA.thaliana[35],wheat[36],maize[37],and rice[20,23,24,38].Therefore,miRNAs can be used as regulators to increase crop stress resistance via genetic manipulations.Rice BB,which is one of the most devastating rice diseases worldwide,has served as a model system for elucidating the interactions between monocotyledonous crops and bacterial pathogens.Nevertheless,the involvement and functions of miRNAs in the rice–Xoointeraction are not well known[23–25,39].To the best of our knowledge,previous studies focused on profiling miRNA expression in a single cultivar or in comparison with transgenic lines[23,24].This may have been because of a lack of suitable genotypes with the same background but differing in BB resistance.The current study is the first to compare the miRNA profiles between F329,an IL with broad-spectrum resistance against BB,and its recurrent parent HHZ to examine the miRNAs involved in the rice–Xoointeraction.

Fig.5.Validation of the expression of miRNAs and their target genes by qRT-PCR.(A)Expression patterns of osa-miR390-3p and its target gene LOC_Os04g49960 in F329-0 h vs.HHZ-0 h.(B)Expression patterns of novel_miR_104 and its target gene LOC_Os08g40440 in F329-12 h vs.HHZ-12 h.(C)Expression patterns of novel_miR_238 and its target gene LOC_Os10g39520 in F329-24 h vs.HHZ-24 h.(D)Expression patterns of osa-miR166k-5p and its target gene LOC_Os10g23090 in F329-36 h vs.HHZ-36 h.(E)Expression patterns of osa-miR529b and its target gene LOC_Os08g39890 in F329-24 h vs.HHZ-24 h.(F)Expression patterns of osa-miR167h-3p and its target gene LOC_Os01g74610 in F329-36 h vs.HHZ-36 h.

Fig.6.Overexpression of osa-miR167h-3p enhances rice resistance to bacterial blight.(A) Disease symptoms of the overexpression transgenic lines and wild-type (WT)plants.(B) Statistical analysis of the lesion length at 2 weeks post inoculation.Asterisks indicate the significance of each line (**, P <0.01,two-tailed Student’s t-test).(C)Bacterial growth in the inoculated leaves of the transgenic lines and WT plants at 3,7,11,and 15 days post-inoculation (dpi).CFU,colony forming units.(**, P <0.01,twotailed Student’s t-test).

4.1.miRNAs involved in the interaction between rice and Xoo

In this study,compared with the non-inoculated samples,17 and 26 miRNAs were differentially expressed in F329 and HHZ,respectively.This suggests that miRNAs are involved in both compatible and incompatible interactions between rice andXoo.We also predicted and functionally annotated the genes targeted by the miRNAs that were responsive toXoo.For the incompatible interaction in F329,these genes were predicted to be related to auxin,jasmonic acid,ethylene,the brassinosteroid mediated signaling pathway,as well as defense and immune responses(Fig.S7A).They were also predicted to encode diverse proteins,including transcription factors,transferases,sugar transmembrane transporters,and superoxide dismutase (Fig.S7B).Regarding the compatible interaction in HHZ,the target genes were revealed to be involved in the auxin-activated signaling pathway and other biological processes such as the removal of superoxide radicals,anthocyanin and indoleacetic acid biosynthesis,metal ion transport,and growth and development (Fig.S7C).The encoded proteins included those binding metal ions as well as the enzymes superoxide dismutase,oxidoreductase,serine/threonine/tyrosine kinase,and indoleacetamide hydrolase (Fig.S7D).Notably,other isoforms of several miRNAs,including osamiR160e-5p (predicted to repress pathogen growth in rice by targeting auxin response factors),osa-miR166,osa-miR398a,and osa-miR169 (responsive toXooin HHZ),were also confirmed to participate in the compatible interaction of the susceptible rice variety Nipponbare [23].Additionally,highly induced osamiR166k-5p expression reportedly leads to rice blast resistance because it controls ethylene-insensitive 2 (EIN2) gene expression[40],which is consistent with the target genes predicted in this study.Previous studies demonstrated that osa-miR398b is responsive toXooin both Rb49 and MDJ8 [24] and targetsOsCSDgenes [20],which can protect plants from ROS damage induced by pathogen infections [41].In our study,osa-miR398a was also responsive toXooin both F329 and HHZ.These results suggest that certain pathways may be commonly activated in different rice lines in response toXoo.

Compared with HHZ,37 up-regulated and 53 down-regulated miRNAs were detected in F329 at time-points before and after theXooinoculation.In a previous study,202 miRNAs were detected as differentially expressed between disease-resistant transgenic rice line Rb49 (containingXa3/Xa26) and WT MDJ8 plants inoculated with PXO61 [24].Eighteen of these miRNAs were commonly expressed in F329 vs.HHZ and Rb49 vs.MDJ8,but few miRNAs had the same expression patterns in the two pairs of resistant and susceptible varieties.For example,osa-miR398a and osamiR398b expression levels were reportedly higher in Rb49 than in MDJ8 before theXooinfection,but in our study,they were lower in F329 than in HHZ at 0 and 36 hpi.These findings suggest that most miRNAs have similar roles in defense responses,but certain miRNAs may mediate race-specific resistance in diverse genetic backgrounds.Growing evidence exists that miRNAs in rice may activate defense responses to disease by regulating the expression of target genes.Specifically,miR159 might directly regulate responses to pathogens by targeting a leucine-rich repeat protein kinase gene(OsLRR-RLK2)[23].The susceptibility of plants overexpressing osa-miR171b to rice stripe virus decreases because of the effects of the miRNA on the expression of the negative regulatorGhd7[34].The overexpression of osa-miR169o and the subsequent targeted suppression ofNF-YA4expression decreases BB resistance in rice because of the resulting repressed transcription of defense-associated PR genes and the nitrogen use efficiency-associated geneNRT2[25].Another study proved that osa-miR164a negatively regulates the resistance to the rice blast fungus by targetingOsNAC60[42].In the present study,we identified miRNAs mediating the resistance of rice to BB,including the down-regulated novel_miR48 and the up-regulated osamiR169f.2,which probably targetLOC_Os07g28570(leucine-rich repeat protein) andLOC_Os11g31560to control LRR protein and protein kinase levels to enhance defense responses.Further clarifying the functions of DEMs will likely provide new insights into the rice–Xoointeraction,with implications for the development of new strategies to improve rice BB resistance.

4.2.A potential network underlying the miRNA-based regulation of rice BB resistance

The molecular mechanism underlying the plant miRNA-based regulation of disease resistance involves genes related to flavonoid biosynthesis,ROS metabolic pathways,plant hormones,defense responses,and growth and development.Down-regulated novel_miR_104 and novel_miR_238 expression may affect flavonoid synthesis and defense responses by suppressing the expression ofLOC_Os08g40440(dihydroflavonol-4-reductase) [43,44] andLOC_Os10g39520(OsMLO4) [45].The osmiR396–OsGRF8–OsF3H–f lavonoid pathway reportedly mediates rice resistance to the brown planthopper[38].Overexpressing osa-miR398b enhances rice blast resistance via multiple superoxide dismutase genes,including three genes involved in the ROS pathway:CSD1,CSD2,andSODX[46].In our study,we identified another member of the miR398 family,osa-miR398a,that likely functions as a negative regulator of BB resistance by modulatingCSD1andCSD2expression.Additionally,novel_miR_146 expression,which was induced in both F329 and HHZ,may enhance the scavenging of reactive oxygen through its effect onOsAKR3expression [47].

Plant hormones are critical modulators of pathways responsible for phenotypic responses to environmental stimuli.Jasmonic acid and its derivatives are especially important for responses to certain pathogens.The essential role of auxin in plant resistance to pathogens is also well known[48].The plant miRNA-based regulation of disease resistance also involves diverse hormone signaling pathways.A pleiotropic miRNA,miR319,suppressesOsTCP21expression to increase plant resistance to rice blast and rice ragged stunt virus via jasmonic acid synthesis [49,50].Inhibiting miR396 expression leads to enhanced resistance to rice blast and improves yield traits by inducing the expression of multipleOsGRFgenes encoding growth-regulating factors[33].On the basis of the results of the present study,we propose that inducing miR396 expression possibly influences the disease response through the auxin signaling pathway and growth-regulating factors.Moreover,regarding osa-miR167h-3p,the expression of its target auxin response factor genes (OsARF6andOsARF8) was negatively correlated with osamiR167h-3p expression,suggesting that the miRNA-mediated mechanism regulating resistance involves auxin signaling,with plants modulating growth and hormone signaling pathways to avoid,resist,or compensate forXooinfection.An important barrier exists that blocks linkages to undesirable agricultural traits such as growth retardation,male sterility and impedes the use of resistance genes to control diseases[51,52].Interestingly,appropriately controlling the expression of certain miRNAs and their target genes may positively affect pathogen resistance and the development of agronomic traits[33,39].Down-regulating miR156 expression and overexpressing its target geneIPA1with the pathogen-inducibleOsHEN1promoter can enhance both BB resistance and yieldrelated traits,thus indicating that miR156–IPA1is a novel regulator of the crosstalk between growth and defense through the GA signaling pathway [39].In our study,we predicted that novel_miR_6 may targetLOC_Os06g17390(auxin-independent growth promoter protein) andLOC_Os04g20400(cytokinin-O-glucosyltransferase 1,a key enzyme maintaining cytokinin homeostasis),which might regulate BB resistance and agronomic traits by modulating plant hormone signaling pathways.Further elucidating the functions of DEMs associated with biotic stress resistance and balanced plant growth will likely substantially improve crop breeding.

5.Conclusions

In this study,a comparison of the miRNA profiles of noninoculated resistant IL F329 and the susceptible recurrent parent HHZ plants as well as plants inoculated withXooprovided a comprehensive and dynamic overview of the miRNAs involved in the incompatible and compatible interactions between rice andXoo.Among 850 miRNAs identified in F329 and HHZ,17 and 26 miRNAs were respectively differentially expressed after theXooinfection.Compared with HHZ,37 up-regulated and 53 down-regulated DEMs were detected in F329 at four time-points.The subsequent investigation of the TDEMs revealed that miRNAs are involved in flavonoid synthesis,ROS pathway,plant hormone signal transduction,defense responses,and growth and development.Furthermore,analyses of transgenic plants proved that osa-miR167h-3p positively regulates rice immunity against BB.These findings will further clarify the mechanism mediating the interactions between rice andXooand provide new insights into the resistance of rice plants to BB.

Availability of data

The datasets generated in the current study are available in the NCBI SRA repository under SRA accession number PRJNA561728.

CRediT authorship contribution statement

Yongli Zhou:conceived and designed the experiments.Jialing Lu,Fan Zhang,and Dan Zeng:performed the experiments.Jialing Lu and Chunchao Wang:analyzed the data.Jialing Lu and Yongli Zhou:wrote the manuscript.All authors read and approved the manuscript.

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

This research was supported by grants from the National Natural Science Foundation of China(31571632 and 31661143009),the CAAS Innovative Team Award,and the Bill &Melinda Gates Foundation(OPP51587).The funders were not involved in designing the study,collecting,analyzing,and interpreting the data,or writing the manuscript.

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2020.08.009.