Identification and comparative analysis of microRNAs in … · BYG plants and co-cultured...

Original Article

Identification and comparative analysis of microRNAs inbarnyardgrass ( Echinochloa crus-galli ) in response torice allelopathy

Changxun Fang1,2, Yingzhe Li1,2, Chengxun Li1,2, Biliang Li1, Yongjie Ren1, Haiping Zheng1, Xiaomei Zeng1, Lihua Shen1,2 &

Wenxiong Lin1,2

1Fujian Provincial Key Laboratory of Agroecological Processing and Safety Monitoring, College of Life Sciences, FujianAgriculture and Forestry University and 2Key Laboratory of Crop Ecology and Molecular Physiology of Fujian Universities,Fujian Agriculture and Forestry University, Fuzhou 35002, China

ABSTRACT

Rice allelopathy is a hot topic in the field of allelopathy, andbehaviour of donor allelopathic rice has been well docu-mented. However, few study addresses response of receiverbarnyardgrass (BYG). We found that expression of miRNAsrelevant to plant hormone signal transduction, nucleotideexcision repair and the peroxisome proliferator-activatedreceptor and p53 signalling pathways was enhanced in BYGco-cultured with the allelopathic rice cultivar PI312777, theexpression levels of these miRNAs in BYG plants were posi-tively correlated with allelopathic potential of theco-cultured rice varieties. Treatment of BYG plants with rice-produced phenolic acids also increased miRNA expression inBYG, while treatment with rice-produced terpenoids had noobvious effect on miRNA expression. In the hydroponicsystem, the largest number of Myxococcus sp. was foundin the growth medium containing rice with the highestallelopathic potential. The addition of phenolic acids inthe hydroponic medium also increased the number ofMyxococcus sp. More interestingly, inoculation withMyxococcus xanthus significantly increased miRNA expres-sion in the treated BYG. Jointed treatments of ferulic acidand M. xanthus led to strongest growth inhibition of BYG.The results suggest that there exist involvement ofMyxococcus sp. and mediation of miRNA expression in riceallelopathy against BYG.

Key-words: allelochemicals; miRNA; Myxococcus sp; phe-nolic compounds.

Abbreviations: AP, apurinic/apyrimidinic; ARF1, auxinresponse factor 1; ARF15, auxin response factor 15; BYG,barnyardgrass; C4H, cinnamate 4-hydroxylase; CA, cinnamicacid; FA, ferulic acid; GIF2, GRF interacting factor 2; HA,p-hydroxybenzoic acid; IAA, indole-3-acetic acid; OE,overexpression; PAL, phenylalanine ammonia-lyase; PPAR,peroxisome proliferator-activated receptor; qPCR, quantita-

tive PCR; RAD23, DNA repair protein RAD23; RNAi, RNAinterference; RT, reverse-transcription; sRNA, small RNA.

INTRODUCTION

Allelopathy is a biological phenomenon that refers to thebeneficial or harmful effects of one organism on another byinfluencing its growth, survival and reproduction through therelease of chemicals into the environment (Rice 1984), and ischaracteristic of certain plants, algae and microorganisms. Asone of the most important food crops, rice (Oryza sativa L.)provides over 21% of the calorific needs of the world’s popu-lation (Fitzgerald et al. 2009). However, weeds cause majorlosses in rice yields and also reduce the quality of the crop.Although barnyardgrass (BYG) is one of the most problem-atic weeds in transplanted and direct-seeded rice worldwide,some rice germplasms are allelopathic to BYG. Allelopathyin rice was first observed in the 1980s in seed production plotsthat were naturally infested with duck salad (Hetherantheralimosa) at the Arkansas Rice Research Institute, USA(Dilday et al. 1989).The phenomenon has also been reportedin many Asian and African countries (Fujii 1992; Olofsdotteret al. 1995; Hassan et al. 1998; Kim & Shin 1998; Weston &Duke 2003). Song et al. (2008) and He et al. (2012b) show thatallelopathic effect is predominant in allelopathic rice/BYGmixed-cultures according to the method of separatingallelopathic effect from resource competition in this system,and the underlying mechanism has been explored in recentyears (He et al. 2012b; Kato-Noguchi & Ino 2012; Xu et al.2012; Fang et al. 2013; Gealy et al. 2013a,b; Kato-Noguchi &Peters 2013).

Isolation and identification of responsible alleloche-micals are crucial for understanding rice allelopathy.However, there is disagreement among researchers aboutrice allelochemicals. Early studies have identified an array ofphenolic acids in the exudates of allelopathic rice cultivars(Seal et al. 2004; He et al. 2012a), and some phenolic acidswere phytotoxic to weeds (Rimando et al. 2001). However,Olofsdotter et al. (2002) claimed that phenolic acids releasedfrom living allelopathic rice roots were unlikely to reachphytotoxic levels. Later momilactone B was regarded as a

Correspondence: W. Lin. Fax: +86 591 83769440; e-mail: [email protected]

Plant, Cell and Environment (2015) doi: 10.1111/pce.12492

bs_bs_banner

© 2014 John Wiley & Sons Ltd 1

main allelochemical in rice because of its very highphytotoxicity (Kato-Noguchi & Ino 2012; Xu et al. 2012;Kato-Noguchi & Peters 2013). However, the evaluation ofthe phytotoxicity of these putative allelochemicals was con-ducted in the laboratory, and was not representative of actualenvironmental conditions. Plant allelochemicals releasedfrom roots can be transformed by surrounding microorgan-isms, and the degraded products may play a key role inallelopathic growth inhibition (Bertin et al. 2003; Inderjit2005). Phenolic acids, flavonoids and momilactones haveall been found to be degraded and transformed in the soil ofpaddy fields (Blum 2011; Kato-Noguchi & Peters 2013;Weston & Mathesius 2013). Our recent study found thatthe allelopathic rice variety PI312777 recruited more micro-bial populations than less allelopathic rice did, and themain genera of microbes in rice rhizosphere belongs tomyxobacteria (Xiong et al. 2012). Myxococcus sp. stronglyinhibits seed germination and seedling growth of BYG in theagar plate test (Supporting Information Fig. S1).

Despite these arguments, it is accepted that allelopathyincontrovertibly triggers the expression and regulation ofgenes in both the donor and receiver plants. Earlier studieshave documented that allelopathy is a quantitative trait(Dilday et al. 1998; Jensen et al. 2001), which is mediated byboth genetic effects and environment factors.Rice allelopathyis an inducible trait and can be induced by both biotic andabiotic stress conditions, such as higher accompanying weeddensities and nutrient deficiency (Song et al. 2008; He et al.2012a;Kato-Noguchi & Ino 2012).Previous studies have dem-onstrated that the expression of genes relevant tophenylpropanoid metabolism was enhanced whenallelopathic rice was exposed to stressful conditions. Bi et al.(2007) found that the transcription of the phenylalanineammonia-lyase (PAL) and cinnamate 4-hydroxylase (C4H)genes in allelopathic rice leaves was significantly increasedafter their exposure to methyl jasmonate and methyl salicy-late. Song et al. (2008) and Fang et al. (2009) also showed thatthe level of PAL expression was significantly increased inthe allelopathic rice cultivar PI312777 compared with thenon-allelopathic rice cultivar Lemont when the nitrogensupply was restricted or when the leaves were sprayedwith exogenous salicylic acid, which resulted in an enhancedinhibitory effect on target weeds. PAL plays a key regulatoryrole in controlling the biosynthesis of all phenylpropanoidproducts (Weisshaar & Jenkins 1998).Other genes involved inthe biosynthesis of phenolic compounds, including C4H,ferulic acid (FA) 5-hydroxylase and caffeic acid O-methyl-transferases, display higher induction in the allelopathic ricecultivar PI312777 than that in the non-allelopathic ricecultivar Lemont under different rice/weed ratio conditions,which was attributed to the higher phenolic acid content inPI312777 plants (He et al. 2012a).All these studies focused onthe response of allelopathic rice to weeds. However, few studyaddressed response of the target weed in the mixed system.

Recently, a class of transcriptional and post-transcriptionalregulators of gene expression was discovered, known asmicroRNAs (miRNAs), that essentially belong to thesmall non-coding RNA with 21–24 nucleotides (Llave et al.

2002; Lim et al. 2003). This class of non-coding RNAs plays aregulatory role in plant development (Reinhart et al. 2002)and responses to environmental signals (Sunkar & Zhu 2004;Sunkar et al. 2012; Shaik & Ramakrishna 2014).

In this study, we analysed the miRNA expression profile inBYG plants and co-cultured allelopathic rice, and comparedthe expression levels of miRNAs in BYG under varioustreatment conditions. The change in the expression levels oftarget genes and their relevant metabolic pathways wasdetermined to confirm the biological function of miRNAsfurther. Potential involvement of Myxococcus sp. in riceallelopathy against BYG and mediation of miRNA expres-sion in BYG were also evaluated.

MATERIALS AND METHODS

Plant materials

Rice (O. sativa L.) varieties PI312777 (high allelopathicpotential) and Lemont (low allelopathic potential) wereselected as the donor plants (Dilday et al. 1994b) and BYG(Echinochloa crus-galli L.) collected from a paddy field wasused as the receiver plant. In addition, two transgenic lines ofPI312777 – the PAL-RNA interference (PAL-RNAi) lineand PAL-overexpression (PAL-OE) line – were also used,because the RNAi line has lower allelopathic potential, whilethe OE line has a higher allelopathic potential thanPI312777. The generation of these transgenic rice lines wasfollowed by the method as described in our previous study(Fang et al. 2013).

Isolation of RNA for miRNA sequencing

Allelopathic rice PI312777 and BYG seeds were surface-sterilized with sodium hypochlorite for 30 min, placed inPetri dishes in a temperature-controlled growth chamber andthen sown in separate seedling plates. At the three-leaf stage,uniform seedlings from each plant species were selected andtransferred into a Styrofoam plate (with 24 holes distributedevenly). The seedlings were affixed to the plate by inserting acotton plug into each hole. The Styrofoam plate was allowedto float in a pot filled with 5 L of rice culture solution, whichwas a minor optimization of the composition of Yoshida et al.(1976), the contents of which are shown in Supporting Infor-mation Table S2. The culture solution was changed everyweek. The outside of each pot was painted black to preventthe growth of algae. The pH value of the solution was main-tained at 5.5 throughout the experiment, which was con-ducted in a greenhouse in which the temperature rangedfrom 25 °C (night minimum) to 32 °C (day maximum) andthe humidity ranged from 40% (day minimum) to 60% (nightmaximum).After 7 d, 12 rice seedlings and 12 BYG seedlingswere co-cultured in the same pot for 7 d. The root tip tissueof both the rice and BYG plants were sampled and thenimmediately frozen in liquid nitrogen and stored at −80 °Cfor isolation of the total RNA. The total RNA was isolatedusing the TRIzol method [Invitrogen Trading (Shanghai) Co.Ltd.] according to the manufacturer’s instructions and the

2 C. Fang et al.

© 2014 John Wiley & Sons Ltd, Plant, Cell and Environment

trace genomic DNA was digested using DNase I [TaKaRaBiotechnology (Dalian) Co., Ltd.].

Rice/BYG co-cultured for different periods

Two allelopathic rice varieties – PI312777 and Lemont – andreceiver BYG were used in this experiment. Both the riceand weed seeds were treated and grown under the sameconditions as those described earlier.When the rice and BYGseedlings had reached the four- or three-leaf stage, respec-tively, 12 rice seedlings and 12 BYG seedlings were trans-planted from the plate and placed in a 5 L pot. Both the riceand BYG root tips were sampled after co-culture for 3, 5, 7and 9 d, 24 uniform rice or BYG seedlings in independent5-L pots were used as the control, and all settings were rep-licated four times.The root samples were frozen immediatelyin liquid nitrogen and stored at −80 °C for the isolation ofmiRNAs.

Different co-culture densities in therice/weed system

In this experiment, the PI312777, the PAL-RNAi andPAL-OE transgenic lines, and the Lemont varieties wereused as the donor plants and BYG was used as the receiver.After carrying out the aforementioned procedures toprepare the pretreated plant seedlings, 12 seedlings of eitherPI312777 or Lemont were planted with four, six or 12 BYGseedlings, represented as rice : BYG = 3:1, 2:1 and 1:1. For thePAL-RNAi and PAL-OE transgenic lines, 12 rice seedlingsand 12 BYG seedlings were cultured in the same pot.Twenty-four uniform rice or BYG seedlings in independent 5 L potswere used as the control; all of the settings were replicatednine times. After 7 d, BYG root samples from the varioustreatments and the control were tested for the detection ofindole-3-acetic acid (IAA) and apurinic/apyrimidinic (AP)sites. The remaining BYG seedlings were also harvested andpacked in triplicate. These samples were all oven-dried at120 °C for 30 min, and at 80 °C for 48 h to measure dry plantweight and evaluate the rice allelopathic potential under dif-ferent rice : BYG co-culture conditions.

Treatment of BYG with different phenolic acidsand terpenoids

Three phenolic acids [p-hydroxybenzoic acid (HA), cinnamicacid (CA) and ferulic acid (FA) ] and four terpenoids [ (−)carveol (C10H16O), (+) carvone (C10H14O), (−) menthone(C10H18O) and (+) cedrol (C15H26O) ] were used as the testcompounds. Only BYG was used as a receiver. The BYGseedlings were prepared as described, and were exposed atthe four-leaf stage to different concentrations of phenolicacids (HA, 0.4, 0.5 and 0.6 mM; CA, 0.12, 0.20 and 0.28 mM;and FA, 0.02, 0.10 and 0.18 mM) or terpenoids [ (+) cedrol, 1,7 and 13 μM; (−) menthone, 5, 55 and 105 μM; (−) carveol, 30,130 and 230 μM; and (+) carvone, 3, 23 and 43 μM]. Theconcentrations were set following the bioassay results on

BYG in our previous study. Each treatment and a control ofBYG cultured in normal nutrient solution were repeatedfour times. After placing the BYG in a solution containingallelochemicals for 7 d, the root tips from the BYG in eachtreatment group were sampled, immediately frozen in liquidnitrogen and stored at −80 °C for the isolation of miRNAs.

Treatment of BYG with Myxococcus xanthus

The bacterium M. xanthus was cultured in CTT medium (1%Casitone, 10 mM of Tris-hydrochloride [pH 7.6], 1 mM ofpotassium phosphate, 8 mM of magnesium sulphate)(Hodgkin & Kaiser 1977) for 48 h. Forty millilitres of culturemedium that contained 0.3 g fresh weight of M. xanthus wasdiluted six times with sterile nutrient solution and used totreat the BYG; The control comprised sterile nutrient solu-tion with CTT at a ratio of 6:1. The BYG seedlings weretreated in the various solutions for 7 d, after which the roottips were sampled for the isolation of miRNAs.

Small RNA (sRNA) sequencing and identificationof miRNAs

DNA-free total RNA from PI312777 and BYG were used toconstruct sRNA libraries, using a TruSeq small RNA SamplePreparation Kit (Illumina, Inc. U. S. A.). The sRNA librarieswere sequenced in the Illumina HiSeq™ 2000 system, fol-lowed by sequenced analysis, the details of which are shownin Supporting Information Table S2. Finally, Gene Ontology(GO) enrichment goatools (https://github.com/tanghaibao/goatools) and Kyoto Encyclopedia of Genes and Genomes(KEGG) enrichment was carried out to annotate and classifythe candidate miRNAs.

Isolation of miRNAs from BYG root andreverse-transcription (RT)

BYG root miRNAs were extracted and isolated using amiRNA pure Mini Kit (Beijing ComWin Biotech Co., Ltd.)following the manual. A total of 1 μg of miRNAs wasreverse-transcripted following the stem-loop method todesign RT primers; 13 candidate miRNAs were chosen fromthe BYG miRNA library and U6 was used as the referencegene. Details of RT are shown in Supporting InformationTable S2.

Quantitative PCR (qPCR) analysis ofmiRNA expression

Stem-loop qPCR was performed with a realplex4 Real-TimePCR System (Eppendorf) using the SuperReal PreMix(SYBR Green) [Tiangen Biotech (Beijing) Co., Ltd.] andmiRNA-specific forward and general reverse primers (Sup-porting Information Table S1). The qPCR protocol detailsare given in Supporting Information Table S2. The relativeexpression level of the miRNAs was calculated using the−2ΔΔCt method (Livak & Schmittgen 2001), and the data werenormalized on the basis of U6 Ct values.

miRNAs expression in barnyardgrass 3


Detection of target gene expression

The total RNA of the BYG samples was extracted usingTRIzol [Invitrogen Trading (Shanghai) Co. Ltd.], and thegenomic DNA was digested using DNase I [TaKaRa Bio-technology (Dalian) Co., Ltd.]. A 1-μL aliquot of total RNAfrom each sample was reverse-transcripted into cDNA usinga TIANscript RT Kit [Tiangen Biotech (Beijing) Co., Ltd.].Three target genes involved in plant hormone signaltransduction and one gene relating to nucleotide excisionrepair were used to compare the relative expression levels inBYG that received various treatments. The primers for thefour target genes are listed in Supporting InformationTable S1. Semi-quantitative RT-PCR was performed todetect the target gene expression in BYG that receivedvarious treatments. The β-actin gene was used as the refer-ence gene. Details of the semi-quantitative RT-PCR areshown in Supporting Information Table S2. The PCR prod-ucts were detected in 1.2% (w/v) agarose gel.

DNA extraction from the hydroponic solution

To evaluate the relationship between the quantity of microbesand the rice allelopathic potential, replicates of each treatedplant were taken and the hydroponic solutions were uniformlymixed; 100 mL of the mixture were then filtered three timesusing filter paper to remove the sediment, the filtrate was thenfiltered again through a millipore filter (0.22 μm nominalpores), on which the microorganisms from the hydroponicsolution were collected and used to extract genomic DNAfollowing the freeze-melt method with sodium dodecyl sul-phate lysis. Details of the procedure are shown in SupportingInformation Table S2. These crude nucleic acid solutions werepurified using the EZNA™ Gel Extraction Kit (Omega Bio-Tek Inc. U. S.A) and re-dissolved in 30 μL of elution buffer.

Assessment of the size of the hydroponicMyxococcus sp. population using qPCR

The population of Myxococcus sp. was quantified using thespecific primers and the detection protocol of Martin & Bull(2006) with minor revisions. qPCR was performed using arealplex4 Real-Time PCR System (Eppendorf North AmericaInc.) with the SuperReal PreMix (SYBR Green) [TiangenBiotech (Beijing) Co., Ltd.].A linear standard curve of 5, 3, 2,1, 0.1 and 0.01 ng DNA standards with four replicate amplifi-cations at each dilution level was also produced to evaluatethe population in the sample.Ct values were recorded for boththe standard and test samples.The size of the Myxococcus sp.population was estimated using the standard curve.

Detection of IAA content and DNAdamage in BYG

To confirm the biological function of the miRNAs identified,the IAA content and DNA damage were determined inBYG, respectively. Details of the IAA extraction are shownin Supporting Information Table S2. The IAA content was

determined using an IAA enzyme-linked immunosorbentassay kit (GENTAUR Ltd. UK) following the manufactu-rer’s protocol. Rice and BYG root DNA was extracted usingthe cetyltrimethylammonium bromide (CTAB) method(Richards et al. 1994); the crude DNA was then purifiedusing an EZNA™ Gel Extraction Kit (Omega Bio-Tek Inc.U. S. A.) and re-dissolved in 30 μL of elution buffer. DNAdamage was evaluated according to the AP site using theDNA Damage Quantification Kit-AP Site Counting(Dojindo Molecular Technologies, Inc. China).

Effect of M. xanthus and FA on BYG growth

A 0.3 g fresh weight sample of M. xanthus with 10 mL of 0.1 mMFA was added to 100 g of sterile soil in a glass culture vessel;cultures of sterile soil with M. xanthus alone or with the additionof 0.1 mM FA were also carried out and sterile soil mixed with10 mL of sterile double-distilled water was used as a control.After3 d,30surface-sterilizedBYGseedsweresownintothesoilin the superclean bench, and the vessels were then kept in anincubator at 23 °C for 10 h (dark conditions) or 28 °C for 14 h(light conditions, 20 000 lux). The root length of the BYG fromeach vessel was measured,and the allelopathic inhibition of eachtreatment on BYG was calculated. The soil was sampled toextract DNA using an EZNA™ Soil DNA Kit (Omega Bio-TekInc. U. S.A.) and the number of M. xanthus was determined byqPCR following the protocol described earlier.

RESULTS

sRNA expression profiles in allelopathicrice and BYG

According to the sRNA sequence results, more sRNA wasexpressed in BYG than in allelopathic rice PI312777(Fig. 1a). In the rice/BYG co-culture system, sRNA expres-sion in the target weed was obviously affected by co-culturedPI312777; more than 62% of the total sRNA was expressed inBYG roots, which represented a twofold increase in compari-son with the rice (Table 1).The unique sRNA was then anno-tated by using Rfam (10.1; http://rfam.sanger.ac.uk/). Theresults showed that 0.48% of the sRNA in BYG and 2.44%of rice sRNA was regarded as positive miRNA (Fig. 1b).

Web Gene Ontology Annotation Plot (WEGO)annotation and KEGG pathway enrichmentanalysis of the identified miRNAs

The positive miRNAs from PI312777 and BYG were bothused to determine the rice genome from the Genbank data-base (http://www.ncbi.nih.gov/Genbank/) using Blastn. Theknown miRNAs were then investigated for target genes.According to WEGO analysis, miRNAs can generally bedivided into three categories: the cellular component,molecular function and biological process (Fig. 2).

A further KEGG enrichment analysis showed that in thisPI312777/BYG co-cultured system, the presence of PI312777dramatically influenced several vital pathways in BYG,

4 C. Fang et al.


including plant hormone signal transduction, nucleotideexcision repair and the peroxisome proliferator-activatedreceptor (PPAR) and p53 signalling pathways, while inPI312777 plants only tropane, piperidine and pyridine alka-loid biosynthesis were affected (Table 2). The negative effectof allelopathic rice on BYG was then further investigated inthe following studies.

miRNAs expression in BYG co-cultured with theallelopathic rice variety PI312777 and thenon-allelopathic rice variety Lemont

For plant hormone signal transduction, the relevantmiRNAs, including miRchromosome_5_50931 (miRchr5_50931), miR172b, miR393 and miR393b, were mostlyup-regulated in the BYG from the PI312777/BYG systemcompared with those in the monoculture BYG. In contrast,these miRNAs were mostly down-regulated in the BYGfrom the Lemont/BYG system, indicating that the high riceallelopathic potential resulted in increased expression levelsof the target miRNAs in BYG.

In addition, the abundance of miR437 and miR1871 inthe PPAR signalling pathway, of miR535, miR396a,miR396b and miR396c in the p53 signalling pathway and ofmiR529b in both pathways was also more strongly affectedby the allelopathic potential of PI312777. The relativeexpression levels of these miRNAs at different times afterthe initiation of co-culture were all higher in BYG in thePI312777/BYG than in the Lemont/BYG system. Theexpression levels of the miRNAs relevant to nucleotideexcision repair, including miR444b.1 and miR444c.1, werehigher in PI312777/BYG than those in Lemont/BYG, espe-cially on day 7, which confirmed the sRNA sequencingresults (Fig. 3a,b).

IAA concentration and DNA AP sites in BYG

The IAA content in the BYG roots was reduced when theplant was co-cultured with rice, especially with theallelopathic rice variety PI312777.The lowest IAA content inthe BYG roots was found in the system of rice/weed (1:1).Variations in the allelopathic potential among rice varietiesresulted in changes of the IAA concentrations in theco-cultured BYG, with increasing levels in plants co-culturedwith the PAL-RNAi lines and decreasing levels in thePAL-OE co-culture system. These results further confirmedthe changes in the expression of relevant miRNAs and thetarget genes (Fig. 4).

The BYG co-cultured with PAL-OE lines contained thelargest number of AP sites, followed by the BYG co-culturedwith PI312777 at a ratio of 1:1, then at a ratio of 1:2, the BYGco-cultured with PAL-RNAi and the BYG co-cultured withPI312777 at a ratio of 1:3. The smallest number of AP siteswas found in the BYG co-cultured with Lemont and thecontrol BYG (Fig. 4).

BYG growth in rice : BYG co-culture systems atdifferent rice : BYG ratios

A further evaluation of rice allelopathic inhibition inBYG showed that the inhibitory rates of rice on BYGdry weight increased as the number of BYG plantsincreased. However, the inhibitory rates of PI312777 onBYG were much higher than those of Lemont at the samerice : BYG ratios. The PAL-OE line of PI312777 showedthe highest inhibitory rate on BYG, while the allelopathicpotential of PAL-RNAi rice on BYG was significantlylower than that of PI312777 under the same conditions(Fig. 5).

(a)

(b)

Figure 1. Small RNAs (sRNAs) in allelopathic rice PI312777 andco-cultured barnyardgrass (Echinochloa crus-galli L.). Allelopathicrice PI312777 and barnyardgrass (BYG) were co-cultured for 7 d;the total RNA of rice or BYG root was extracted for microRNA(miRNA) sequencing. The miRNAs were sequenced by theIllumina HiSeq™ 2000 system. (a) Length distribution of the BYGand PI312777 sRNA libraries; (b) Annotation result of the sRNAfrom BYG and PI312777.

Table 1. Comparison of small RNA (sRNA) expression in allelopathic rice PI312777 and barnyardgrass

TypeUniquesRNA

Percentageof unique (%)

TotalsRNA

Percentageof total (%)

Barnyardgrass specific 2 997 165 62.44 4 972 406 31.56Allelopathic rice PI312777 specific 1 504 768 31.35 1 806 104 11.46Barnyardgrass and allelopathic rice PI312777 specific 297 760 6.20 8 975 645 56.97



Myxococcus sp. population in thehydroponic system

To determine the size of the Myxococcus sp. populations in thehydroponic system, standard curves related Ct values to thenumbers of Myxococcus sp. Each standard representedbetween 105 and 108 copies of Escherichia coli.The calibrationcurve for the bacterial copies was linear (R2 = 0.9938) andspread over four orders of magnitude (Supporting InformationFig. S2).The results showed that the allelopathic rice co-culturesystem contained larger numbers of Myxococcus sp. than thatof the non-allelopathic rice, and increased with weed density inthe PI312777 co-culture system. However, no significant differ-ence was found in the Lemont co-culture system with various

weed densities. In addition, the OE of PAL in PI312777 alsoresulted in an increased Myxococcus sp. population comparedwith the control PI312777 plant, while the PAL-RNAi trans-genic line of PI312777 contained a smaller Myxococcus sp.population than PI312777. The numbers were increased in therice/BYG co-culture system; the PAL-OE/BYG mixture con-tained the largest number of Myxococcus sp., followed by thePI312777 co-culture, and the smallest number was found in thePAL-RNAi co-culture (Fig. 6).

miRNA expression in M. xanthus-treated BYG

To investigate possible involvement of Myxococcus sp. inallelopathic effect of rice on BYG, the expression level of

Figure 2. WEGO analysis of identified microRNA (miRNA) target genes from PI312777 and barnyardgrass (BYG). According to WEGOanalysis, miRNAs can generally be divided into three categories: the cellular component, molecular function and biological process.

Table 2. KEGG pathway enrichment for microRNA (miRNA) target genes from PI312777 and barnyardgrass

Plants KEGG pathway ID P value Relevant miRNAs

Barnyardgrass Plant hormone signal transduction ko04075 7.26E–06 miRchromosome_5_50931, miR172b,miR393, miR393b

Barnyardgrass Peroxisome proliferator-activatedreceptor(PPAR) signalling pathway

ko03320 3.37E–03 miR529b, miR437, miR1871

Barnyardgrass p53 signalling pathway ko04115 4.78E–03 miR535, miR529b, miR396a, miR396b,miR396c

Barnyardgrass Nucleotide excision repair ko03420 3.45E–02 miR444b.1, miR444c.1PI312777 Tropane, piperidine and pyridine

alkaloid biosynthesisko00960 2.03E–04 os–miR812f, os–miR812g, os–miR812h

os–miR812i, os–miR812j

KEGG, Kyoto Encyclopedia of Genes and Genomes.

6 C. Fang et al.


miRNAs was determined in BYG treated with M. xanthus.The results showed that the expression level of miRNAs wasincreased in the M. xanthus-treated BYG, which indicatedthe allelopathic potential of M. xanthus on BYG (Fig. 7a).

A further comparison showed that the level of expressionof four target genes – auxin response factor 1 (ARF1), auxinresponse factor 15 (ARF15), GRF interacting factor 2 (GIF2)and DNA repair protein RAD23 (RAD23) – was obviouslydown-regulated in the treated BYG, and a correlationbetween miRNA and target gene expression was found(Fig. 7b).

miRNA expression in BYG treated with differentallelochemicals

Our previous studies showed that the selected phenolic acidsand terpenoids at the tested concentrations used in thehydroponic culture system have similar inhibitory effects on

BYG in bioassays under laboratory conditions. The resultsshowed that the phenolic acids greatly increased the level ofexpression of the miRNAs involved in plant hormone signaltransduction, especially miR172b and miR393, which weremore than 19-fold higher after each treatment. The expres-sion level of miR444b.1 and miR444c.1 also greatly increasedin all of the phenolic acid-treated BYG, indicating that treat-ment with phenolic acids reduced the DNA repair ability ofBYG, and was in turn harmful to BYG growth development.In addition, the expression level of the miRNAs related tothe PPAR and p53 signalling pathways also increased. All ofthese results demonstrated that phenolic acids significantlyinfluenced plant growth and development through theexpression of relevant miRNAs (Fig. 8a).

In contrast, both miRchr5_50931 and miR393 were down-regulated in all of the BYG treated with terpenoids com-pared with the normal control plants, and the expressionof miR172b and miR393b was enhanced after most of the

(a)

(b)

Figure 3. Comparison of identified enhanced miRNA expression in barnyardgrass co-cultured with PI312777 and Lemont, respectively.Validation by quantitative PCR of differentially expressed miRNAs obtained from high-throughput sequencing. U6 small RNA was used as areference, and the expression levels of each of the miRNAs in co-cultured barnyardgrass were then compared with those of mono-culturedbarnyardgrass, of which all miRNAs were set at 1.0. Error bars indicate the standard deviations of four replicates.



terpenoid treatments, but the increase was distinctly smallerthan that observed with phenolic acid treatment. Moreover,the change in the level of expression of miR444b.1 andmiR444c.1 also differed from that after phenolic acid treat-ment; these two miRNAs were either up-regulated or down-regulated after different terpenoid treatments (Fig. 8b).However, the change in expression level was no more thanthreefolds, which was significantly smaller than that afterphenolic acid treatment.

Myxococcus sp. populations in BYG treated withallelochemicals in the hydroponic system

The population of Myxococcus sp. was also determined byqPCR, which showed that their number was higher in thephenolic acid-treated BYG than in the control plants,except for those treated with 0.6 mM of HA, 0.12 mM ofCA and 0.18 mM of FA. The largest Myxococcus sp. popu-lation was found after treatment with 0.02 mM FA (about800 cells per mL of solution); the other five treatments thatenhanced the Myxococcus sp. population resulted in morethan 330 cells per mL of solution compared with the control(124 cells per mL of solution) (Fig. 9a). In contrast to treat-ment with phenolic acids, of the 12 different treatments withterpenoids, five reduced the population and resulted infewer than 100 cells/mL of solution. No significant differ-ence was found with three other treatments (230 μM ofcarveol, 55 μM of menthone and 3 μM of carvone),although the other four treatments (5 μM of menthone,

Figure 4. Comparison of indole-3-acetic acid (IAA) content andapurinic/apyrimidinic (AP) sites in the roots of barnyardgrass(BYG) at different co-culture ratios. Allelopathic rice PI312777and its PAL-RNAi and PAL-OE transgenic lines, and thenon-allelopathic rice Lemont varieties were used as the donorplants and BYG was used as the receiver. Twelve of each of thewild-type PI312777 and Lemont rice seedlings were planted with 4,6 or 12 BYG seedlings, represented as rice: BYG = 3:1, 2:1 and 1:1.For the PAL-RNAi and PAL-OE transgenic lines, 12 rice seedlingsand 12 BYG seedlings were cultured in the same pot. Twenty-fouruniform rice or BYG seedlings in independent 5-L pots were usedas the control. After 7 d, rice and BYG root samples from thevarious treatments and the control were tested for the detection ofIAA and AP sites. For each rice : BYG ratio, superscript lettersindicate statistical groups that are significantly different (P < 0.05).BCT, mono-culture barnyard grass control; PB31,PI312777:BYG = 3:1; PB21, PI312777:BYG = 2:1; PB11,PI312777:BYG = 1:1; LB31, Lemont : BYG = 3:1; LB21,Lemont : BYG = 2:1; LB11, Lemont : BYG = 1:1; OEB11,PAL-overexpression transgenic PI312777:BYG = 1:1; RNAiB11,PAL-RNA interference transgenic PI312777:BYG = 1:1.

Figure 5. Inhibitory rates (IR) of different types of rice on the dry weight of barnyard grass (BYG) plants.IR = (1 − treatment/control) × 100%. IR > 0 and IR < 0 indicate inhibitory and stimulatory effects, respectively. Each experiment wasperformed in triplicate. All of the experimental data are presented as the mean ± standard deviation. For each rice : BYG ratio, superscriptletters indicate statistical groups that are significantly different (P < 0.01).

8 C. Fang et al.


1 μM of cedrol, 13 μM of cedrol and 30 μM of carveol)resulted in an increased Myxococcus sp. population, but thenumbers were all less than 270 cells per mL of solution(Fig. 9b). This result demonstrated that phenolic acids andterpenoids had different effects on the multiplication ofMyxococcus sp.

Allelopathic inhibition of M. xanthus andFA on BYG

Compared with the control, M. xanthus with FA showed thestrongest allelopathic inhibition in BYG, with a 64.82% inhi-bition rate on BYG root length; treatment of BYG with

Figure 6. Myxococcus species populations in the hydroponic solution of different rice : BYG co-culture systems. The quantitative PCR wasused to determine the microbe population and establish standards for Myxococcus sp. Ct values were recorded for both the standard and testsamples. The size of the microbe population was estimated using the standard curve. For each rice : BYG ratio, superscript letters indicatestatistical groups that are significantly different (P < 0.05). PCT, mono-culture PI312777; LCT, mono–culture Lemont; OECT, mono-culturePAL-overexpression transgenic PI312777; RNAiCT, mono–culture PAL-RNA interference transgenic PI312777.

(a)

(b)

Figure 7. Comparison of identified target miRNA and target gene expression in barnyardgrass (BYG) treated with Myxococcus xanthus.Forty millilitres of culture medium that contained 0.3 g fresh weight sample of M. xanthus was diluted six times with sterile nutrient solutionand also mixed with 0.3 g fresh weight of M. xanthus as another treatment for the BYG. The control comprised sterile nutrient solution withCTT at a ratio of 6:1. The BYG seedlings were treated in the various solutions for 7 d. The expression levels of each of the miRNAs in BYGtreated with M. xanthus culture was then compared with those of untreated BYG, which were set at 1.0. Error bars indicate the standarddeviation (±SD) of four replicates. (a) The expression levels of each of the miRNAs in treated BYG; (b) the miRNA target gene expressionin treated BYG.



M. xanthus or FA alone resulted in 8.26 and 34.64% inhibi-tion rates, respectively (Fig. 10a,c). Moreover, the soil towhich M. xanthus and FA were added had the largestnumbers of M. xanthus, with 3025 cells per g of soil, andsignificantly larger numbers than that to which M. xanthus(640 cells per g of soil) or FA (278 cells per g of soil) wasadded alone; the control soil only contained 44 cells per gof soil after 7 d (Fig. 10b). These results indicated theallelopathic potential of M. xanthus with FA.

DISCUSSION

Allelopathy now receives more attention fromagroecologists (Fitter 2003; Gealy et al. 2013b). Anenvironment-friendly weed control based on the use ofchemical weapons from a donor plant against the targetweed, without the introduction of herbicides or pesticides, isappropriate for sustainable agriculture. About 3∼4% of the

rice accessions throughout the world have shown allelopathicpotential towards weeds (Dilday et al. 1994a), and specificvarieties have been used to investigate the underlyingmechanism. Previous studies mostly focused on exploring theresponse and reaction of the donor plants. However, a fewstudies have centred on the target weed, which is an equallyimportant field of research. According to our results, riceallelopathy significantly inhibited plant hormone signaltransduction in BYG; the level of four miRNAs –miRchr5_50931, miR172b, miR393 and miR393b – wasincreased in the BYG co-cultured with PI312777. Amongthese, miR172b and miRchr5_50931 are negative regulatorsof auxin response factors (ARF1 and ARF15), which aretranscription factors that bind with auxin response elementsand regulate the expression of auxin response genes(Guilfoyle & Hagen 2007), which can be induced by auxin,also known as IAA, and play a major role in the develop-mental and physiological responses of plants. Allelopathic

(a)

(b)

Figure 8. Expression levels of identified target miRNAs in barnyardgrass treated with phenolic acids (a) and terpenoids (b). Three phenolicacids – p-hydroxybenzoic acid (HA), cinnamic acid (CA) and ferulic acid (FA) – and four terpenoids – (−) carveol, (+) carvone, (−)menthone and (+) cedrol – were used as the test compounds. Only BYG was used as the receiver. The BYG seedlings at the four-leaf stagewere exposed to different concentrations of phenolic acids (a) (HA, 0.4, 0.5 and 0.6 mM; CA, 0.12, 0.20 and 0.28; and FA, 0.02, 0.10 and0.18 mM) or terpenoids (b) [ (+) cedrol, 1, 7 and 13 μM; (–) menthone 5, 55 and 105 μM; (–) carveol 30, 130 and 230 μM; and (+) carvone, 3,23 and 43 μM]. The expression levels of each of the miRNAs in phenolic acid- or terpenoid-treated BYG were then compared with those ofuntreated BYG, all of which were set at 1.0. Error bars indicate the standard deviations (±SD) of four replicates.

10 C. Fang et al.


inhibition by PI312777 resulted in increased expression ofmiR172b and miRchr5_50931 in the co-cultured BYG, andthe two miRNAs in BYG were clearly a response to therice allelopathic effect. Extension of the duration of the rice/weed co-culture resulted in enhanced miRNAs expression,which in turn led to the down-regulation of ARF expression,thereby preventing the growth development of the BYG.The IAA content was also lower in the co-cultured than inthe mono-cultured BYG, and the lower level of expression ofAFRs and auxin response genes has been attributed to areduction in IAA. The allelopathic potential of PI312777was enhanced by the OE of PAL in the rice, and BYGco-cultured with this transgenic rice showed higher levels ofexpression of miRNAs than that co-cultured with the wildtype, while the PAL-RNAi line of PI312777, a lessallelopathic transgenic rice, had less influence on the expres-sion level of these miRNAs. These results showed that riceallelopathy particularly suppressed the IAA signal in thetarget BYG. In contrast to PI312777, the co-culture of BYGwith the non-allelopathic rice accession Lemont had rela-tively little effect, due to its weak allelopathic potential.

Another negative effect of rice allelopathy is DNA damage,represented by AP sites in DNA locations that neither have apurine nor a pyrimidine base. Among the various treatments,more AP sites were detected in the PI312777/BYG co-culturesystem, especially under conditions of high BYG density; thefindings suggested that the allelopathic effect on BYG wasdue to damage to the genomic DNA, which prevented regularcell division.The expression of two miRNAs –miR444b.1 andmiR444c.1 – was increased in the BYG co-cultured withPI312777 for various periods and was enhanced BYG density.These two miRNAs degraded a gene that encodes the DNArepair protein, RAD23, which is relevant to DNA damagerepair.The decreased expression of RAD23 in the co-culturedBYG, especially that co-cultured with PI312777, resulted in a

decreased capacity to repair the DNA damage fromphytotoxicity. In addition, the degree of DNA damage wasgreater in the PAL-OE/BYG system, in correlation with theincreased expression of miRNAs and the down-regulation ofRAD23, but was lower in the BYG co-cultured with the lessallelopathic PAL-RNAi transgenic PI312777 rice. Severalsignal pathways, including the PPAR and p53 signalling path-ways,were also influenced in the co-cultured BYG.PPARs area group of nuclear receptor proteins that function as transcrip-tion factors that regulate the expression of genes (Michaliket al. 2006). In mammals, PPARs play an essential role in theregulation of cellular differentiation, development andmetabolism (Rosen & Spiegelman 2001). An increase inmiRNAs relevant to the PPAR signalling pathway impliedthat rice allelopathic inhibition of BYG growth developmentis correlated with signal transduction, and strong suppressionof the pathway was found in the most allelopathic riceco-culture system. For the other signalling pathway, the p53transcription factor regulates multiple biological functions,including DNA damage and oxidative stress (Finkel &Holbrook 2000; Maclaine & Hupp 2009).This gene is continu-ously degraded in the cell under normal conditions. Whenexternal or cellular stress causes DNA damage, p53 degrada-tion is inhibited, and p53 protein accumulates in the nucleus.In our study, increased expression of the miRNAs involved inthe p53 signalling pathway in the BYG co-cultured withallelopathic rice indicated a weakness in its DNA repaircapacity, which resulted in more serious DNA damage thanthat found in the control plant, based on the determination ofthe DNA AP sites.

Although some conflict remains over the main allelo-chemicals in rice, the investigation of the gene response ofthe BYG treated with exogenous putative allelochemicalshas supplied useful evidence. Our studies showed that theexpression of these miRNAs was greatly increased in BYGtreated with different phenolic acids. However, the treatmentof BYG with terpenoids produced no significant change in ordown-regulation of the expression of the miRNAs involvedin plant growth development and DNA repair.These findingsrevealed the phytotoxicity of phenolic acids on the targetweed.

Microorganisms, an important link between alleloche-micals and target plants, have received increasing attention(Bertin et al. 2003; Blum 2003; Inderjit 2005; Dayan et al.2010; Khan et al. 2013; Zuo et al. 2014), and thebiotransformation of allelochemicals by particular environ-mental microorganisms has been reported to be necessaryand to contribute to an increase in allelopathy (Wang et al.2013). In this study, a specific microbe, the Myxococcusspecies, showed a close relationship with rice allelopathicinhibition. Our previous studies found that the rhizosphericsoil of PI312777 contained a larger Myxococcus populationthan that of Lemont (Xiong et al. 2012). In this study, thepopulation of Myxococcus species was significantly higher inthe PI312777/BYG system than in the Lemont/BYG system,to which an increase in BYG density contributed. However,no significant difference in BYG density was seen in theLemont system. These results indicated the vital function of

(a)

(b)

Figure 9. Populations of Myxococcus species in the hydroponicsolution of barnyardgrass treated with different phenolic acids (a)and terpenoids (b).



Myxococcus during the rice allelopathic inhibitory process.The Myxococcus species belongs to the myxobacteria, agroup of proteobacteria that reside mainly in the soil(Reichenbach 2001; Velicer & Vos 2009) and produce a largenumber of natural products (Diez et al. 2012). The allelo-pathic effect may occur directly through the release ofnatural products by the particular allelopathic bacterium,and growth-inhibiting allelochemicals produced by plantgrowth-promoting bacteria have been isolated and character-ized previously (Duke et al. 2000; Barazani & Friedman2001). Our studies found that the addition of M. xanthus to

the BYG mono-culture system resulted in enhanced levels ofexpression of the miRNAs that are relevant for growth inhi-bition.The increase in the expression of these miRNAs mark-edly inhibited their target genes. These interesting outcomesfinally led to the allelopathic capacity of M. xanthus to sup-press BYG, which is an important link between the donorplant allelochemicals and the receiver plants, and thesespecific bacteria acted as transforming factors of plantallelochemical exudates.

Allelochemicals, especially phenolic acids, significantlypromoted the proliferation of Myxococcus sp. in the

(a)4 4000

16s rDNA

3000

2000800

600

400

200

0

3

2

1

0

c

c

d

bb

aa a

FA + M. xanthus M. xanthus M. xanthus

ii (M. xanthus)

FA FA

iii (FA)

FA +M. xanthus

Control Control

i (Control)

M. xanthusFA FA + M. xanthus

iv (FA + M. xanthus)

Control

Roo

t len

gth

(cm

)

Pop

ulat

ion

num

ber

(cel

ls p

er g

of

soil) (b)

(c)

Figure 10. Effects of application of ferulic acid (FA) and inoculation of Myxococcus xanthus on seedling growth of barnyardgrass and finalpopulation number of M. xanthus in soil. Pre-sterilized soil was used; M. xanthus and FA were added to the soil separately or together for3 d; pre-sterilized soil with same volume of sterilized water was used as the control. Thereafter, germinated BYG seeds were sown in the soil,the root length of the BYG was measured and the inhibitory rates (IR) were calculated as follows: IR = (1 − treatment/control) × 100%.IR > 0 and IR < 0 indicate inhibitory and stimulatory effects, respectively. Soils from various treatments were then sampled to extractgenomic DNA to determine the final population numbers of M. xanthus using the quantitative PCR. For each rice : BYG ratio, superscriptletters indicate statistical groups that are significantly different (P < 0.01). (a) The root length and the relevant IR of BYG; (b) theM. xanthus population numbers in different soils; (c) the phenomenon of BYG under different soil conditions.

12 C. Fang et al.


hydroponic system. Among the phenolic acids tested, lowconcentrations of FA and HA had a remarkable inductiveeffect, which further demonstrated their importance in theprocess of rice allelopathy. In contrast, the terpenoids testedshowed little positive effect on Myxococcus sp., the popula-tion numbers of which were much lower when terpenoidswere added to the hydroponic system rather than phenolicacids. A comparison of the growth-promoting effect ofdifferent allelochemicals on a characteristic bacterium repre-sents the cross-talk interaction between predominantallelochemicals and microorganisms, thereby clearly distin-guishing between the primary and secondary actions of dif-ferent allelochemicals.

In conclusion, phenolics and other compounds releasedfrom allelopathic rice plants can recruit myxobacteriaMyxococcus sp., and then, these allelochemicals andmyxobacteria jointly increase expression of relevantmiRNAs of the co-cultured BYG, which lead to inhibition ofplant hormone signal transduction and decreases of DNA

repair capacity in BYG, consequently inhibited BYG growth(Fig. 11).

ACKNOWLEDGMENTS

This work was supported by the National Natural ScienceFoundation of China (Nos. 31271670, 31300336) and theNational Research Foundation for the Doctoral Program ofHigher Education of China (No. 20133515130001) and Pro-vincial Natural Science Foundation of Fujian, China (No.2010J05045).

REFERENCES

Barazani O. & Friedman J. (2001) Allelopathic bacteria and their impact onhigher plants. Critical Reviews in Microbiology 27, 41–55.

Bertin C., Yang X. & Weston L.A. (2003) The role of root exudates andallelochemicals in the rhizosphere. Plant and Soil 256, 67–83.

Bi H.H., Zeng R.S., Su L.M., An M. & Luo S.M. (2007) Rice allelopathyinduced by methyl jasmonate and methyl salicylate. Journal of ChemicalEcology 33, 1089–1103.

Blum U. (2003) Fate of phenolic allelochemicals in soils – the role of soil andrhizosphere microorganisms. In Allelopathy: Chemistry and Mode of Actionof Allelochemicals (eds F.A. Macias, J.C.G. Galindo, J.M.G. Molinillo & H.G.Cutler), pp. 57–76. CRC Press, Boca Raton, FL.

Blum U. (2011) Plant–plant allelopathic interaction. Phase II: field/laboratoryexperiments. In Plant–Plant Allelopathic Interactions monograph, pp.85–149. Springer, the Netherlands.

Dayan F.E., Rimando A.M., Pan Z., Baerson S.R., Gimsing A.L. & Duke S.O.(2010) Sorgoleone. Phytochemistry 71, 1032–1039.

Diez J., Martinez J.P., Mestres J., Sasse F., Frank R. & Meyerhans A. (2012)Myxobacteria: natural pharmaceutical factories. Microbial Cell Factories 11,1–3.

Dilday R., Nastasi P. & Smith R. Jr. (1989) Allelopathic observations in rice(Oryza sativa L.) to duck salad (Heteranthera limosa). Proceedings of theArkansas Academy of Science 43, 21–22.

Dilday R., Lin J. & Yan W. (1994a) Identification of allelopathy in the USDA-ARS rice germplasm collection. Animal Production Science 34, 907–910.

Dilday R., Lin J. & Yan W. (1994b) Identification of allelopathy in USDA-ARS320 rice germplasm collection. Australian Journal of Experimental Agricul-ture 24, 907–910.

Dilday R., Yan W., Moldenhauer K. & Gravois K. (1998) Allelopathic activityin rice for controlling major aquatic weeds. In Allelopathy in Rice (ed. M.Olofsdotter), pp. 7–26. International Rice Research Institute, Los Baños,Laguna, Philippines.

Duke S., Dayan F., Romagni J. & Rimando A. (2000) Natural products assources of herbicides: current status and future trends. Weed Research 40,99–112.

Fang C., Zhuang Y., Xu T., Li Y., Li Y. & Lin W. (2013) Changes in riceallelopathy and rhizosphere microflora by inhibiting rice phenylalanineammonia-lyase gene expression. Journal of Chemical Ecology 39, 204–212.

Fang C.X., Xiong J., Qiu L., Wang H.B., Song B.Q., He H.B., . . . Lin W.X. (2009)Analysis of gene expressions associated with increased allelopathy in rice(Oryza sativa L.) induced by exogenous salicylic acid. Plant Growth Regu-lation 57, 163–172.

Finkel T. & Holbrook N.J. (2000) Oxidants, oxidative stress and the biology ofageing. Nature 408, 239–247.

Fitter A. (2003) Making allelopathy respectable. Science 301, 1337–1338.Fitzgerald M.A., McCouch S.R. & Hall R.D. (2009) Not just a grain of rice: the

quest for quality. Trends in Plant Science 14, 133–139.Fujii Y. (1992) The potential biological control of paddy weeds with

allelopathy: allelopathic effect of some rice varieties. In Proceedings Inter-national Symposium on Biological Control and Integrated Management ofPaddy and Aquatic Weeds in Asia, pp. 305–320. National AgriculturalResearch Centre of Japan, Tsukuba, Japan.

Gealy D.R., Moldenhauer K.A. & Duke S.O. (2013a) Root distribution andpotential interactions between allelopathic rice, sprangletop (Leptochloaspp.), and barnyardgrass (Echinochloa crus-galli) based on 13C isotope dis-crimination analysis. Journal of Chemical Ecology 39, 186–203.

Barnyardgrass

FA

HA

Allelopathic rice

O

O

OH

OHHO

HO

H3CO

PPAP signalling pathway

Myxococcus sp.

P53 signalling pathway

Plant hormone reduced Nucleotide excision repair unavailable

Cell division inhibition DNA damage

Suppress BYG growth

Figure 11. Schematic summary of model of allelopathicinteractions between barnyardgrass (BYG) and rice. The figureindicates that the phenolic acids released from allelopathic ricevariety promoted the proliferation of Myxococcus sp. in the ricerhizosphere; the Myxococcus sp. interacted with phenolic acid andaffected plant growth and DNA repair by altering the expressionof relevant microRNAs and their target genes in the BYG, whichin turn reduced the indole-3-acetic acid content and increasedapurinic/apyrimidinic sites, resulting in the inhibition on BYGgrowth and development. A detailed explanation is presented inthe text. Solid arrows point to the brief gene regulation network inBYG, based on the results of our study.



Gealy D.R., Moldenhauer K.A. & Jia M.H. (2013b) Field performance ofSTG06L-35-061, a new genetic resource developed from crosses betweenweed-suppressive indica rice and commercial southern US long-grains. Plantand Soil 370, 277–293.

Guilfoyle T.J. & Hagen G. (2007) Auxin response factors. Current Opinion inPlant Biology 10, 453–460.

Hassan S., Aidy I., Bastawisi A. & Draz A. (1998) Weed management usingallelopathic rice varieties in Egypt. In Allelopathy in Rice (ed. M.Olofsdotter), pp. 27–37. International Rice Research Institute, Los Baños,Laguna, Philippines.

He H., Wang H., Fang C., Wu H., Guo X., Liu C., . . . Lin W. (2012a) Barnyardgrass stress up regulates the biosynthesis of phenolic compounds inallelopathic rice. Journal of Plant Physiology 169, 1747–1753.

He H.B., Wang H.B., Fang C.X., Lin Z.H., Yu Z.M. & Lin W.X. (2012b)Separation of allelopathy from resource competition using rice/barnyardgrass mixed-cultures. PLoS ONE 7, e37201.

Hodgkin J. & Kaiser D. (1977) Cell-to-cell stimulation of movement in non-motile mutants of Myxococcus. Proceedings of the National Academy ofSciences of the United States of America 74, 2938–2942.

Inderjit I. (2005) Soil microorganisms: an important determinant ofallelopathic activity. Plant and Soil 274, 227–236.

Jensen L.B., Courtois B., Shen L., Li Z., Olofsdotter M. & Mauleon R.P. (2001)Locating genes controlling allelopathic effects against barnyardgrass inupland rice. Agronomy Journal 93, 21–26.

Kato-Noguchi H. & Ino T. (2012) The chemical-mediated allelopathic inter-action between rice and barnyard grass. Plant and Soil 370, 1–9.

Kato-Noguchi H. & Peters R.J. (2013) The role of momilactones in riceallelopathy. Journal of Chemical Ecology 39, 175–185.

Khan F., Kumari M. & Cameotra S.S. (2013) Biodegradation of the allelopathicchemical m-tyrosine by Bacillus aquimaris SSC5 involves the homogentisatecentral pathway. PLoS ONE 8, e75928.

Kim K. & Shin D. (1998) Rice allelopathy research in Korea. In Allelopathy inRice (ed. M. Olofsdotter), pp. 39–44. International Rice Research Institute,Los Baños, Laguna, Philippines.

Lim L.P., Glasner M.E., Yekta S., Burge C.B. & Bartel D.P. (2003) VertebratemicroRNA genes. Science 299, 1540.

Livak K.J. & Schmittgen T.D. (2001) Analysis of relative gene expression datausing real-time quantitative PCR and the 2−ΔΔCT method. Methods (SanDiego, Calif.) 25, 402–408.

Llave C., Kasschau K.D., Rector M.A. & Carrington J.C. (2002) Endogenousand silencing-associated small RNAs in plants. The Plant Cell 14, 1605–1619.

Maclaine N.J. & Hupp T.R. (2009) The regulation of p53 by phosphorylation:a model for how distinct signals integrate into the p53 pathway. Aging 1,490–502.

Martin K. & Bull C. (2006) Novel primers for detection and quantification ofMyxococcus species in situ. Molecular Ecology Notes 6, 773–775.

Michalik L., Auwerx J., Berger J.P., Chatterjee V.K., Glass C.K., Gonzalez F.J.,. . . Wahli W. (2006) International union of pharmacology. LXI. peroxisomeproliferator-activated receptors. Pharmacological Reviews 58, 726–741.

Olofsdotter M., Navarez D. & Moody K. (1995) Allelopathic potential in rice(Oryza sativa L.) germplasm. The Annals of Applied Biology 127, 543–560.

Olofsdotter M., Rebulanan M., Madrid A., Wang D.L., Navarez D. & Olk D.C.(2002) Why phenolic acids are unlikely primary allelochemicals in rice.Journal of Chemical Ecology 28, 229–242.

Reichenbach H. (2001) Myxobacteria, producers of novel bioactive sub-stances. Journal of Industrial Microbiology and Biotechnology 27, 149–156.

Reinhart B.J., Weinstein E.G., Rhoades M.W., Bartel B. & Bartel D.P. (2002)MicroRNAs in plants. Genes & Development 16, 1616–1626.

Rice E.L. (1984) Allelopathy, 2nd edn. Academic Press, Orlando, FL.Richards E., Reichardt M. & Rogers S. (1994) Preparation of genomic DNA

from plant tissue. Current Protocols in Molecular Biology Chapter 2:Unit2.3.

Rimando A.M., Olofsdotter M., Dayan F.E. & Duke S.O. (2001) Searching forrice allelochemicals. Agronomy Journal 93, 16–20.

Rosen E.D. & Spiegelman B.M. (2001) PPARγ: a nuclear regulator of metabo-lism, differentiation, and cell growth. The Journal of Biological Chemistry276, 37731–37734.

Seal A.N., Pratley J.E., Haig T. & An M. (2004) Identification and quantitationof compounds in a series of allelopathic and non-allelopathic rice rootexudates. Journal of Chemical Ecology 30, 1647–1662.

Shaik R. & Ramakrishna W. (2014) Machine learning approaches distin-guish multiple stress conditions using stress responsive genes and identifycandidate genes for broad resistance in rice. Plant Physiology 164,481–495.

Song B., Xiong J., Fang C., Qiu L., Lin R., Liang Y. & Lin W. (2008)Allelopathic enhancement and differential gene expression in rice under lownitrogen treatment. Journal of Chemical Ecology 34, 688–695.

Sunkar R. & Zhu J.-K. (2004) Novel and stress-regulated microRNAs andother small RNAs from Arabidopsis. The Plant Cell 16, 2001–2019.

Sunkar R., Li Y.-F. & Jagadeeswaran G. (2012) Functions of microRNAs inplant stress responses. Trends in Plant Science 17, 196–203.

Velicer G.J. & Vos M. (2009) Sociobiology of the myxobacteria. Annual Reviewof Microbiology 63, 599–623.

Wang C.M., Li T.C., Jhan Y.L., Weng J.H. & Chou C.H. (2013) The impact ofmicrobial biotransformation of catechin in enhancing the allelopathic effectsof Rhododendron formosanum. PLoS ONE 8, e85162.

Weisshaar B. & Jenkins G.I. (1998) Phenylpropanoid biosynthesis and itsregulation. Current Opinion in Plant Biology 1, 251–257.

Weston L.A. & Duke S.O. (2003) Weed and crop allelopathy. Critical Reviewsin Plant Sciences 22, 367–389.

Weston L.A. & Mathesius U. (2013) Flavonoids: their structure, biosynthesisand role in the rhizosphere, including allelopathy. Journal of ChemicalEcology 39, 1–15.

Xiong J., Lin H.F., Li Z.F., Fang C.X., Han Q.D. & Lin W.X. (2012) Analysis ofrhizosphere microbial community structure of weak and strong allelopathicrice varieties under dry paddy field. Acta Ecologica Sinica 32, 6100–6109. (inChinese with English abstract).

Xu M., Galhano R., Wiemann P., Bueno E., Tiernan M., Wu W., . . . Peters R.J.(2012) Genetic evidence for natural product-mediated plant-plantallelopathy in rice (Oryza sativa). The New Phytologist 193, 570–575.

Yoshida S., Forno D., Cock J. & Gomez K. (1976) Laboratory Manual forPhysiological Studies of Rice, 3rd edn. International Rice Research Institute,Los Baños, Laguna, Philippines.

Zuo S., Li X., Ma Y. & Yang S. (2014) Soil microbes are linked to theallelopathic potential of different wheat genotypes. Plant and Soil 378,49–58.

Received 13 August 2014; received in revised form 13 November 2014;accepted for publication 16 November 2014

SUPPORTING INFORMATION

Additional Supporting Information may be found in theonline version of this article at the publisher’s web-site:

Figure S1. Allelopathic inhibition of Myxococcus sp. onBYG.Figure S2. The standard curve of IAA concentration (I), thenumber of AP sites (II) and Myxococcus species populations(III).Table S1. The primers used in this study.Table S2. The details of methods.

14 C. Fang et al.


Identification and comparative analysis of microRNAs in … · BYG plants and co-cultured...

Documents

Transcript of Identification and comparative analysis of microRNAs in … · BYG plants and co-cultured...