The function of OsDR8, a rice disease resistance-responsive gene, was studied. ... 2003), whereas most defense-responsive genes, .... mids/pMCG161.html).
Springer 2006
Plant Molecular Biology (2006) 60:437–449 DOI 10.1007/s11103-005-4770-x
Dual function of rice OsDR8 gene in disease resistance and thiamine accumulation Gongnan Wang , Xinhua Ding , Meng Yuan, Deyun Qiu, Xianghua Li, Caiguo Xu and Shiping Wang* National Key Laboratory of Crop Genetic Improvement, National Center of Plant Gene Research (Wuhan) Huazhong Agricultural University, Wuhan 430070, China (*author for correspondence; e-mail swang@ mail.hzau.edu.cn); These authors contributed equally to this work Received 20 May 2005; accepted in revised form 3 November 2005
Key words: bacterial blight, blast, defense, rice, thiamine
Abstract The function of OsDR8, a rice disease resistance-responsive gene, was studied. Silencing of OsDR8 using an RNA interference approach resulted in phenotypic alteration of the plants. The transgenic plants with repressed expression of OsDR8 showed reduced resistance or susceptibility to Xanthomonas oryzae pv. oryzae and Magnaporthe grisea causing bacterial blight and blast, which are two of the most devastating diseases in rice worldwide, respectively. The putative product of OsDR8 was highly homologous to an enzyme involved in the biosynthesis of the thiazole precursor of thiamine. Transgenic plants showing repressed expression of OsDR8 and reduced resistance had significantly lower levels of thiamine than the control plants. Exogenous application of thiamine could complement the compromised defense of the OsDR8-silenced plants. The expression level of several defense-responsive genes including the earlier functional genes of defense transduction pathway, OsPOX and OsPAL, and the downstream genes of the pathway, OsPR1a, OsPR1b, OsPR4, OsPR5 and OsPR10, was also decreased in the OsDR8-silenced plants. These results suggest that the impact of OsDR8 on disease resistance in rice may be through the regulation of expression of other defense-responsive genes and the site of OsDR8 function is on the upstream of the signal transduction pathway. In addition, the accumulation of thiamine may be essential for bacterial blight resistance and blast resistance. Abbreviations: PR, pathogenesis-related; QTLs, quantitative trait loci; R, resistance gene; RACE, rapid amplification of cDNA end; RNAi, RNA interference; RT, reverse transcription; SA, salicylic acid; UTR, untranslated region; Xoo, Xanthomonas oryzae pv. Oryzae
Introduction Plant disease resistance is regulated by two classes of genes: the major resistance (R) genes and the defense-responsive or defense-related genes. Most of the R genes that have been characterized are race-specific (Hammond-Kosack and Parker, 2003), whereas most defense-responsive genes,
which encode the components of the signal transduction pathways leading to defense responses of the host after the recognition event triggered by an R gene product, appear to not be race-specific (Maleck et al., 2000; Zhou et al., 2002; Wen et al., 2003). Therefore, defense-responsive genes may be valuable sources of broad-spectrum and durable resistance in breeding programs.
438 Bacterial blight disease, caused by Xanthomonas oryzae pv. oryzae (Xoo), and blast disease, caused by fungus Magnaporthe grisea, are two of the most serious diseases in rice worldwide. In recent years, large numbers of defense-responsive genes have been identified to be involved in the incompatible interaction between rice and Xoo or M. grisea (Zhou et al., 2002; Chu et al., 2004). However, the impact of each of these defenseresponsive genes on rice disease resistance is unknown, except for the knowledge that some of these genes colocalize with quantitative trait loci (QTLs) for rice disease resistance (Wang et al., 2001; Xiong et al., 2002; Ramalingam et al., 2003; Wen et al., 2003; Chu et al., 2004). Thus, the values of these genes in the improvement of disease resistance for rice breeding need to be explored. Our previous study identified that a rice cDNA, EI35I3, which is the fragment of a gene putatively encoding thiamine or a vitamin B-1 biosynthetic enzyme, showed increased expression level in the incompatible interaction between the rice cultivar Minghui 63 and M. grisea isolate V86013 (Zhou et al., 2002). Further analysis revealed that EI35I3 showed differential expression in several rice lines that carried different R genes conferring resistance for either Xoo or M. grisea, but no change in the expression of EI35I3 was observed in compatible interactions (Wen et al., 2003). In addition, EI35I3 colocalizes with a previous identified resistance QTLs against blast disease (Wang et al., 1994) in rice chromosome 7 (Wen et al., 2003). The information suggests that EI35I3 may be involved in disease resistance in a pathogen-nonspecific way, but the activation of R genes is the key to initiate its expression. This also implies that deletion of the gene represented by EI35I3 in resistant rice or expression of the gene represented by EI35I3 in a constitutive way without the existence of an R gene may cause a phenotypic change in rice, at least in terms of resistance to blast disease. We report here the isolation and functional characterization of the gene Oryza sativa defenseresponsive gene 8 (OsDR8), represented by the cDNA EI35I3, from the rice cultivar Minghui 63 (Chen et al., 2002, 2003; Yang et al., 2003), which carries R genes that confer resistance for both bacterial blight and fungal blast. Blocking the expression of OsDR8 by RNA interference (RNAi) technology in Minghui 63 reduced the ability of the plant against the two diseases and
also reduced the accumulation of thiamine and other defense-responsive gene transcripts. These results suggest that OsDR8 has a dual phenotypic impact on rice for both defense regulation and thiamine synthesis and that it could be used in rice improvement.
Materials and methods Isolation of OsDR8 and gene structure analysis The sequence of cDNA fragment (cDNA clone EI35I3, about 1.1 kb in length) of the OsDR8 gene from the rice cultivar Minghui 63 (Oryza sativa ssp. indica) was used to screen the nucleotide sequence database GenBank (http://www. ncbi.nlm.nih.gov) by use of the BLAST program (Altschul et al., 1997). The rice genomic sequence identified from BLAST analysis was analyzed using the GenScan program (Burge and Karlin, 1997) to predict the size and structure of the gene homologous to the cDNA sequence EI35I3. The genomic sequence of the gene that was homologous to EI35I3 and its flanking sequences were analyzed for identification of the restriction enzyme digestion sites using the computer program SEQUENCHER 4.1.4 (Gene Codes Corporation, Ann Arbor, MI, USA). The restriction enzyme sites flanking the homologous sequence of OsDR8 were used to isolate the DNA fragment containing OsDR8 from Minghui 63. A DNA fragment of about 1.9 kb in size was cloned into pCAMBIA1301 vector, and the plasmid containing the 1.9 kb fragment was named D55S. For sequencing analysis, the D55S plasmid was subcloned by digestion with a series of restriction enzymes – PstI, SacI, KpnI and BamHI/SacI – and the digested fragment was ligated into the pUC19 vector. The 5¢-untranslated region (UTR) of OsDR8 gene was analyzed by 5¢-rapid amplification of cDNA ends (RACE) assay using the 5¢ RACE System (Invitrogen Life Technologies, Carlsbad, California, USA), according to the manufacturer’s protocol. The OsDR8-specific primer GSP1 (5¢-GTAGGTGATCATGTCG-3¢) was used for reverse transcription. Another OsDR8-specific primer GSP2 (5¢-GGACTACCGCAACTCCA-3¢) was used in combination with 5¢ RACE Abridged Anchor primer in the 5¢ RACE System for PCR
439 amplification. The product of 5¢-RACE was cloned by use of pGEM-T Easy Vector Systems (Promega Corporation, Madison, WI, USA). Agrobacterium-mediated transformation To construct the OsDR8 RNAi vector, primers dsF (5¢-TAACTAGTGGCGCCTGCAGGTACCGGT CCG-3¢) and dsR (5¢-TAGAGCTCGCCTAGGT GCACGCGTACGTACGTAAGC-3¢) were used to amplify the cDNA fragment (cDNA clone EI35I3) of the OsDR8 gene cloned in the pSPORT1 vector. The dsF contained the digestion sites (underlined) of the restriction enzymes SpeI and AscI at the 5¢ end, and dsR contained the digestion sites (underlined) of the restriction enzymes SacI and AvrII at the 5¢ end. The sequences following the restriction enzyme sites of dsF and dsR were complementary to the sequences flanking the multicloning sites of pSPORT1 vector. Part of the PCR product of the cDNA clone EI35I3 was digested with AscI and AvrIIs and ligated into a AscI+AvrIIcleaved binary vector pMCG161 that contained a cassette designed for making inverted repeat transcripts of a gene, flanking a loop, which could efficiently produce double-stranded RNA (McGinnis et al., 2005, http://www.chromdb.org/info/plasmids/pMCG161.html). To clone the second repeat, the remaining PCR product was digested with SpeI and SacI and cloned into the SpeI+SacI-cleaved construct that contained the first repeat. The construct containing the inverted repeat transcripts of OsDR8 driven by 35S promoter was transferred into Agrobacterium tumefaciens strain EHA105 by electroporation. The clone was named D16R and was transferred by Agrobacteriummediated transformation into cultivar Minghui 63 according to the protocol for high efficiency transformation of indica rice (Lin and Zhang, 2005). The copy number of the OsDR8 RNAi construct in transgenic plants was determined by DNA gel blot analysis, with probes amplified using hygromycin phosphotransferase gene-specific primers Hpt-F (5¢-AGTCAATGACCGCTGTTA TGC-3¢) and Hpt-R (5¢-CTGATCGAAAA GTTC GACAGC-3¢). Pathogen inoculation To examine the resistance of plants to bacterial blight disease, plants at booting stages were
inoculated with Philippine Xoo strain PXO61 (race 1) by the leaf-clipping method (Kauffman et al., 1973). The disease was scored by measuring the lesion area (lesion length/leaf length) two weeks after inoculation. For the evaluation of blast disease, young leaves were inoculated in vitro with Philippine M. grisea isolate V86013 by the spotinoculation method (Jia et al., 2003). T0 transgenic plants and control plants were cut at about 12 cm above the ground level after seeds had been harvested. Fresh young leaves generated from the stubs of the plants were collected and inoculated with four to five droplets of inoculum per leaf and 10 ll of inoculum per droplet in Petri dishes. Inoculum preparation followed a method described previously (Chen et al., 2001). The disease was screened at about 4 days after inoculation. Thiamine treatment Seventeen transgenic plants of T1 generation (eight belonging to D16RMH9 family and nine belonging to D16RMH15 family) at 5- to 6-leaf stage were sprayed with a solution containing 50 mM thiamine and 0.02% Tween 20 at 4 h prior to pathogen inoculation. Eight and nine plants from the same two T1 families, respectively, were sprayed with 0.02% Tween 20 at 4 h prior to pathogen inoculation as the mock-treatment (control). Quantitative reverse transcription (RT)-PCR Total RNA was treated with DNase I (Invitrogen Life Technologies, Carlsbad, California, USA), to remove contaminating DNA, and used for RTPCR in a two-step reaction (Zhou et al. 2002). In brief, the RT step was performed in a 10 ll volume that contained 0.5–1 lg total RNA pretreated with DNase I, 50 ng oligo (dT)15 primer, 100 U MMLV reverse transcriptase (Promega Corporation, Madison, WI, USA), 1 first-strand buffer (50 mmol/l Tris-HCl (pH 8.3), 75 mmol/l KCl, and 3 mmol/l MgCl2), 20 mmol/l DTT, and 10 lmol/l each of dATP, dCTP, dGTP, and dTTP at 42 C for 1.5 h. The mixture was diluted by the addition of 40 ll of deionized water after the RT reaction, and 1 ll of the diluted mixture was used for quantitative PCR. Quantitative PCR was applied by use of Applied Biosystems 7500 RealTime PCR System (Applied Biosystems Corporation, USA). The PCR primers are listed in Table 1.
440 Table 1. Gene primers used for quantitative PCR. Gene name
GenBank accession number
Forward primer (5¢–3¢)
Reverse primer (5¢–3¢)
OsDR8 OsPR1a OsPR1b OsPR4 OsPR5 OsPR8 OsPR10 JIOsPR10 OsPOX OsPAL Actin
DQ176424 AJ278436 U89895 AY050642 X68179 AF296279 D38170 AF395880 X66125 X87946 X15865
TCCAGCCTCCTCAAGACCT CGTCTTCATCACCTGCAACTACTC GGCAACTTCGTCGGACAGA AGCGCATATTGTGCCACATG CAACAGCAACTACCAAGTCGTCTT GTTCATCTGGTCAGCGGATAGC CCCTGCCGAATACGCCTAA CCGGACGCTTACAACTAAATCG GCCATCATGGACGGTTCTGT AGCACATCTTGGAGGGAAGCT TGCTATGTACGTCGCCATCCAG
AGTCCTCCTGCTCGTCGTA CATGCATAAACACGTAGCATAGCA CCGTGGACCTGTTTACATTTTCA GGATACACTTGCCACACGAGTCT CAAGGTGTCGTTTTATTCATCAACTTT TCATAAGTATTATCACGACCGTTCGA CTCAAACGCCACGAGAATTTG CACTTCTCAATCACTGCTTGGAA TCTGGAGAAATTGCCGATAAGTTC GCGCGGATAACCTCAATTTG AATGAGTAACCACGCTCCGTCA
The expression of rice actin was used to standardize RNA samples for each RT-PCR. The expression level of the examined gene in each sample was calculated with the expression level of actin. The quantitative PCR was in a 25-ll volume that contained 1 ll of diluted RT product, 2.5 ll of 10 rTaq buffer (TaKaRa Biotechnology, Dalian, China), 3.2 mM MgCl2, 0.04 mM each of dATP, dCTP, dGTP, and dTTP, 0.24 lM each of primer 35I3F and primer 35I3R, 2 U rTaq polymerase (TaKaRa Biotechnology, Dalian, China), 5% glycerol, 5% DMSO, and 1 SYBR Green I.
produced by the thiochrome was detected at 365 nm excitation wavelength, 435 nm emission wavelength, a 5 nm excitation bandpass, and a 5 nm emission bandpass, using a spectrophotometer. The content of thiamine in each sample was calculated according to the standard curve of thiamine, measured by use of commercial thiamine.
Quantification of thiamine content
The cDNA clone EI35I3, from rice cultivar Minghui 63, a fragment of OsDR8, was completely sequenced. It was 1159 bp in length. Analysis of EI35I3 using the BLAST program identified a homologous rice genomic sequence (GenBank accession number AP004674) located on chromosome 7. The AP004674 had two restriction enzyme sites, SamI and BamHI, flanking the OsDR8 homolog. Using cDNA clone EI35I3 as a probe
The content of thiamine in the leaf tissue of rice plants was determined by use of the fluorescent method (Wang and Zhang, 1997). In brief, 4 g of ground leaf tissue from each plant were used. Thiamine was oxidized into thiochrome in a basic solution that contained 0.07% kalium ferricyanide and 14% sodium hydroxide. The fluorescence
Results Isolation of OsDR8 gene
Figure 1. The structure of OsDR8 gene from rice cultivar Minghui 63. The coding region (black boxes) of OsDR8 is interrupted by one intron (line). The positions of 5¢- and 3¢-untranslated regions (hatched boxes), the translation start codon (ATG), the translation stop codon (TGA), and primers (arrows) for the 5¢ RACE assay are also indicated. The numbers indicate the base pairs of each substructure. D55S, genomic clone; EI35I3, cDNA clone.
441 to screen Minghui 63 BAC library (Peng et al., 1998) identified a positive BAC clone 5F21. Digestion of 5F21 using SamI and BamHI and cloning of the DNA fragment obtained a subclone, D55S (GenBank accession number DQ176424). Sequence of D55S revealed that it was a fragment of 1980 bp in length that contained OsDR8 (Figure 1). Comparative analysis of the genomic sequence of OsDR8 (clone D55S) and its cDNA fragment sequence (clone EI35I3) identified an intron of 78 bp located at the 3¢ end of OsDR8 and the 3¢UTR of 64 bp (Figure 1). To determine whether OsDR8 contained other introns, the cDNA sequence of the gene at the 5¢ end was obtained by 5¢ RACE assay using the primers GSP1 and GSP2, which overlap with cDNA sequence EI35I3. Sequence analysis of the product of 5¢RACE assay revealed that OsDR8 contained only one intron and a 349-bp 5¢-UTR (Figure 1). Thus, OsDR8 had a size of 1550 bp and encoded a protein with 352 amino acids. OsDR8 from the indica rice cultivar Minghui 63 showed 97% sequence identity to both its allele (designated as OsDR8-93) in another indica rice cultivar, 93–11, and its allele (designated as OsDR8-Nip) in the japonica rice cultivar
Nipponbare. OsDR8-93 was harbored by a genomic sequence (GenBank accession number AAAA02022114) located at 5035–6587 bp. OsDR8-Nip was identified in a genomic sequence (GenBank accession number AP004674) located at 23468 to 25016 bp. More than one-half of the nucleotide polymorphism sites detected between OsDR8 and OsDR8-93 or between OsDR8 and OsDR8-Nip were located at the 3¢-UTR, 5¢-UTR or intron. The remaining nucleotide-site divergences were located in the coding region, which caused a 4% amino acid site divergence between the putative encoding products of OsDR8 and OsDR8-93 or between the putative encoding products of OsDR8 and OsDR8-Nip (Figure 2). The polymorphic amino acid sites between OsDR8 and OsDR893 or between OsDR8 and OsDR8-Nip were at the same positions. In other words, OsDR8-93 and OsDR8-Nip shared 99% sequence identity (Figure 2). The predicted encoding product of OsDR8 showed 81% sequence identity and 86% sequence similarity to the maize thiamine biosynthetic enzymes thil-1 (NCBI, http://www.ncbi.nlm.nih.gov; protein database accession number S61419) and thil-2 (protein database accession number S61420) (Belanger et al., 1995).
Figure 2. Comparison of thiamine biosynthetic enzyme and thiamine biosynthetic enzyme-like proteins. OsDR8, OsDR8-93 and OsDR8-Nip are thiamine biosynthetic enzyme-like proteins putatively encoded by the alleles, OsDR8, OsDR8–93 and OsDR8-Nip from rice cultivar Minghui 63, 93-11 and Nipponbare, respectively. Thil-1 and thil-2 are maize thiamine-biosynthetic enzymes (Belanger et al., 1995). The solid black shaded residues indicate the ones shared by all the proteins. The grey shaded residues indicate the ones shared by all the rice proteins.
442 Table 2. Performance of T0 transgenic plants (D16RMH). Material
Copy number
Disease Lesion area (%)
Minghui 63 D16RMH1 D16RMH2 D16RMH3 D16RMH5 D16RMH7 D16RMH8 D16RMH9 D16RMH11 D16RMH12 D16RMH13 D16RMH15 D16RMH16 D16RMH17 D16RMH18 D16RMH21 D16RMH23 D16RMH29 D16RMH24 D16RMH30
1 2 1 3 2 1 5 1 1 1 1 2 1 2 2 0 0
16.8 39.3 56.5 59.8 47.2 33.9 23.3 36.7 54.6 49.9 34.8 59.4 47.8 43.8 29.4 47.0 35.0 36.7 17.3 16.6
Thiamine Pa
0.079 0.022 0.000 0.008 0.113 0.035 0.023 0.032 0.114 0.00
Content (lg/g leaf)
P
0.128±0.004 0.062±0.003 0.073±0.007 0.068±0.002 0.062±0.003
0.000 0.001 0.000 0.000
0.065±0.001 0.060±0.003 0.086±0.004 0.064±0.001 0.050±0.001
0.001 0.000 0.000 0.001 0.001
0.073±0.002
0.000
0.070±0.002
0.000
0.120±0.009
0.261
0.007 0.169 0.033 0.926 0.955
a
The three to five uppermost fully expanded leaves of each plant were inoculated for most of the plants; the few plant without P values were inoculated on only one leaf each.
OsDR8 was involved in the regulation of resistance against both bacterial blight disease and fungal blast disease To determine whether OsDR8 had a phenotypic impact on rice disease resistance, the function of OsDR8 in the rice cultivar Minghui 63 was blocked using RNAi technology. In total, 30 independent transformants were generated by transformation with the OsDR8 RNAi plasmid (D16R). Minghui 63 was moderately resistant to Xoo strain PXO61 (Yang et al., 2003). All the transgenic plants of the T0 generation were inoculated with PXO61 at the booting stage. Seventeen of the T0 plants showed reduced resistance or susceptibility to PXO61, compared with the control plants. The lesion areas of the 17 plants ranged from 23% to 60%, compared with the approximately 17% measured for the controls of untransformed Minghui 63 and negative transgenic plants, D16RMH24 and D16RMH30 (Table 2). The expression levels of OsDR8 in 26 surviving plants of the 30 transgenic plants were examined by quantitative RT-PCR in the condition without pathogen challenge. The results showed that an extensive reduction of OsDR8 expression was
associated with the reduction of resistance in the examined plants, although a few of the plants had a level of P>0.05 for their phenotype variation, which indicates that OsDR8 was involved in the regulation of resistance against Xoo (Figure 3A and Table 2). The plants showing reduced resistance had approximate 13- to 30-fold reduction in OsDR8 expression, while those plants showing no distinct change of resistance had only less than 2fold reduction (T0 plants D16RMH8, 20, 22, 25, 27 and 28) or no reduction (T0 plants D16RMH4, 10, 14, 19, 24 and 30) in OsDR8 expression as compared with the control plants. Our previous study shows that incompatible pathogen can induce OsDR8 (cDNA EI35I3) expression (Wen et al., 2003). OsDR8 expression is also elevated upon incompatible-pathogen infection in transgenic plants showing reduced resistance (Figure 4A). However, the OsDR8-transcript level in pathogen-challenged transgenic plants was significantly (P
