Comparative analysis of drought-responsive transcriptome in Indica rice genotypes with contrasting drought tolerance

Plant Biotechnology Journal (2010), pp. 1–13 doi: 10.1111/j.1467-7652.2010.00560.x Comparative analysis of drought-responsive transcriptome in Indica rice genotypes with contrasting drought tolerance Sangram K. Lenka1, Amit Katiyar1, Viswanathan Chinnusamy2 and Kailash C. Bansal1,* 1 National Research Centre on Plant Biotechnology, Indian Agricultural Research Institute, New Delhi, India 2 Department of Botany and Plant Sciences, University of California, Riverside, CA, USA Received 24 March 2010; revised 24 May 2010; accepted 2 June 2010. *Correspondence (Tel 91 11 25843554; fax 91 11 25843984; email kailashbansal@hotmail.com) Summary Genetic improvement in drought tolerance in rice is the key to save water for sustainable agriculture. Drought tolerance is a complex trait and involves interplay of a vast array of genes. Several genotypes of rice have evolved features that impart tolerance to drought and other abiotic stresses. Comparative analysis of drought stressresponsive transcriptome between drought-tolerant (DT) landraces ⁄ genotypes and drought-sensitive modern rice cultivars will unravel novel genetic regulatory mechanisms involved in stress tolerance. Here, we report transcriptome analysis in a highly DT rice landrace, Nagina 22 (N22), versus a high-yielding but drought-susceptible rice variety IR64. Both genotypes exhibited a diverse global transcriptional response under normal and drought conditions. Gene ontology (GO) analysis suggested that drought tolerance of N22 was attributable to the enhanced expression of several enzymeencoding genes. Drought susceptibility of IR64 was attributable to significant downregulation of regulatory components that confer drought tolerance. Pathway analysis unravelled significant up-regulation of several components of carbon fixation, glycolysis ⁄ gluconeogenesis and flavonoid biosynthesis and down-regulation of starch and sucrose metabolism in both the cultivars under drought. However, significant upregulation of a-linolenic acid metabolic pathway observed in N22 under drought appears to be in good agreement with high drought tolerance of this genotype. Consensus cis-motif profiling of drought-induced co-expressed genes led to the identification of novel cis-motifs. Taken together, the results of the comparative Keywords: drought, rice, microarray, transcriptome analysis led to the identification of specific genotype-dependent genes pathway analysis, a-linolenic acid. responsible for drought tolerance in the rice landrace N22. Introduction Yield potential of rice is drastically reduced by drought across all agro-climatic regions of the globe. Hence, it is necessary to produce more ‘crop per drop’ to sustain our agricultural production (Witcombe et al., 2008). Rice has very low water-use efficiency, and about 50% of irrigated fresh water is used for rice cultivation alone. Therefore, genetic improvement in drought tolerance in rice assumes significance owing to dwindling water resources exacerbated by global climate change. Centuries of continuous selection and breeding efforts in different agro-climatic conditions led to the evolution of rice cultivars with wide range of drought and other abiotic stress tolerance. Some traditional cultivars and landraces of rice have evolved mechanisms that impart tolerance to various abiotic stresses such as drought, salinity, etc. These tolerant genotypes are excellent genetic resources for stress tolerance but are poor yielders. The Indian landrace selection Nagina 22 (N22) is one such example of a traditional genotype that is highly tolerant to drought and is a donor of drought tolerance traits in breeding programmes (Reddy Ch et al., 2009). On the contrary, most high-yielding modern rice varieties are developed for best performance under ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd 1 2 Sangram K. Lenka et al. irrigated conditions. One such widely grown lowland Indica variety IR64 is high yielding but susceptible to drought stress (Lafitte et al., 2007). Identification and characterization of novel genes for drought tolerance by comparative transcriptomic studies in a drought-tolerant viable option for enhancing abiotic stress tolerance of crop plants (Sreenivasulu et al., 2007). However, functional characterization of TFs in rice lags considerably behind the progress made in the model dicotyledonous species Arabidopsis thaliana. Application of modern high-throughput (DT) versus susceptible cultivar will help in understanding evolutionary basis of stress adaptation mechanisms, facilitating thereby crop improvement via genetic engineering and precision breeding. genomics tools is proving useful in understanding the molecular mechanisms responsible for the expression of the abiotic stress tolerance traits. Discovery of novel abiotic stress regulatory genes, identification of key pathways Drought tolerance is a complex trait that involves several metabolic and morphological adaptive pathways. Hence, deciphering genetic basis of drought tolerance mechanisms in plants still remains a challenging task (Price et al., that are altered in response to stress, and functional characterization of the genes involved are imperative to understand stress tolerance mechanisms. Over the past decade, various studies compared the expression profiling of 2002). Drought-responsive genes encode proteins involved in signalling, gene expression, stress damage control and repair (Valliyodan and Nguyen, 2006). Transcriptional regulation is an important regulatory mechanism of repro- stressed versus non-stressed plants and reported several abiotic stress–related genes (Reddy et al., 2002; Rabbani et al., 2003; Kushwaha et al., 2009). However, comparative expression profiling of DT versus susceptible geno- gramming of transcriptome in response to drought. Several genes induced by abiotic stresses, including those encoding transcription factors (TFs), have been identified, and some of them have been shown to be essential for stress tolerance (Hu et al., 2006; Guo et al., 2008; Oh types offers a way to identify novel genes and regulatory mechanisms with evolutionary adaptive significance. Here, we report genome-wide drought-responsive transcriptome changes in drought tolerant N22 (DT) versus drought-susceptible IR-64 (DS) Indica rice genotypes. GO et al., 2009). Genome-wide identification of droughtresponsive regulons in contrasting DT genotypes will help to unravel system-level interplay between different genetic pathways that impart drought tolerance. Although some and enrichment of conserved cis-regulatory elements within the 1- kb upstream sequences from translational start site (ATG) of the co-expressed drought-responsive genes were also investigated. Expression level of selected genes involved in stress response have been identified, the key water-deficit stress signalling components that are responsible for genotype-dependent drought tolerance have mostly remained unidentified (Wang et al., 2007; drought-responsive TFs was validated using quantitative real-time PCR (qRT-PCR). Pathway analysis was carried out to investigate the influence of genotype-dependent drought-responsive transcriptome on metabolic changes. Degenkolbe et al., 2009). Availability of diverse genetic resources for stress tolerance coupled with advanced genomics tools made rice a popular model system for abiotic stress research (IRGSP, Results and discussion 2005; Xu et al., 2006). Perception of drought stress followed by succession of signal transduction events to switch on molecular, cellular and whole plant adaptive processes are critical steps for stress tolerance. The spatiotemporal gene regulation in plants is governed by combi- Genotype-dependent drought-responsive transcriptome Comparison of the differential expression profiling of effector and master regulatory genes between a stresstolerant and a susceptible genotype of a species in natorial interactions of cis-acting DNA elements in the promoters with trans-acting protein factors. Genes encoding TFs represent a considerable fraction of the genomes of all eukaryotic organisms, including higher plants response to similar level of abiotic stress may help in identification of underlying metabolic pathways and regulatory mechanism(s) responsible for adaptation of plants to stress conditions. Two well-known rice genotypes with contrast- (Riechmann et al., 2000). Out of total annotated genes, TFs constitute approximately 2.6% of the rice genome (Gao et al., 2006). Molecular genetic and transgenic analysis of abiotic stress tolerance during the past one decade ing drought stress response were chosen for this study. N22 is a drought-tolerant variety (DT) adapted to upland conditions, whereas IR64 is a widely grown shallow-land variety with high susceptibility to drought (DS) stress revealed that transcriptome engineering (use of master switch genes such as signalling proteins and TFs) is a (Reddy et al., 2002; Lafitte et al., 2007; Reddy Ch et al., 2009). To investigate the intrinsic variation in drought ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd, Plant Biotechnology Journal, 1–13 Drought-responsive transcriptome analysis in rice 3 tolerance of these two genotypes, various indices of ‘stress-induced injury’ were scored in this study. Relative water content (RWC) under control condition was recorded as 96% in DT and 94.4% in DS, whereas under drought treatment, the RWC was reduced to 64.8% and Two rice genotypes N22 (DT) and IR64 (DS) were used for expression analysis under non-stress and water-deficit stress using GeneChip Rice Genome Array with three biological replications (Figure S2). We compared transcriptome response of whole rice seedlings to drought stress. 56.2% in DT and DS, respectively. Similarly, total leaf chlorophyll content was higher in DT (1.7 mg ⁄ g FW under control and 1.4 mg ⁄ g FW under drought) compared to DS (1.2 mg ⁄ g FW under control and 0.98 mg ⁄ g FW under The correlation coefficients of normalized samples between any two biological replications ranged from 0.960 to 0.998 indicating that microarray analyses were highly reproducible in this experiment (Table S1). Data on drought). The excised-leaf water content in DT was about 50% in 5- h duration from leaves of both control and drought-treated plants. In case of DS, the retention of excised-leaf water was 58% for control leaves but reduced expression profiling of probe sets showing more than twofold up or down-regulation compared with corresponding control condition were chosen for further analysis. These are illustrated in Venn diagrams and profile plot to 44% in leaves from drought treatment (Figure S1). Water retention in N22 (DT) excised leaves was always significantly higher than in IR64 (DS) leaves at different time intervals. This shows better ability of DT to conserve leaf (Figure S3a–e) and listed in the (Tables S2–S6). To evaluate the genome-wide response of DT and DS under control and drought conditions, linkage hierarchical clustering was performed (Figure 1). Genes that expressed under moisture when compared to DS in response to dehydration. In addition, DT exhibited better drought tolerance and recovery ability than the DS as observed by visual comparison of leaf senescence and wilting symptoms in the two cultivars. DT seedlings developed new leaves and control condition were clustered together and separated from the drought-responsive clusters for the two genotypes, which were also clustered together. This analysis clearly demonstrated that there was a genotype- and environment-dependent change in gene expression, and recovered faster than the DS. Hence, based on the previous reports and the various physiological parameters measured in this study, we confirmed that DT and DS differ significantly in their response to drought stress. we presume that the differential gene regulation observed here was attributable to genotype · environment interaction (G · E). Identification of significantly regulated target genes, which differed in their expression between DT and Previous studies reported a comparative analysis of large-scale expression profiling in a drought stress-treated and non-treated rice genotype as well in upland and lowland rice cultivars (Rabbani et al., 2003; Wang et al., DS under drought stress, might potentially serve as candidate alleles for genetic improvement in drought tolerance in rice. In DS under drought stress, 3097 and 2391 probes were up- and down-regulated, respectively, compared to 2007; Zhou et al., 2007; Degenkolbe et al., 2009). These transcript-profiling data suggested that rice plants perceive and respond to water stress quickly by altering gene expression. Thus, it is important to analyse tran- that under control conditions (Table S2), whereas 4513 probes were up-regulated and 2606 probes regulated under drought compared with control tions (Table S3). These probe sets represent script changes in parallel with biochemical pathway alterations. In the above-mentioned studies, several genes involved in drought stress response were identified; however, the key water stress signalling components that are responsible for genotype-dependent drought tolerance responsive genes as per their annotation and belong to different categories including TF-encoding genes. Many of these genes were shown previously to be involved in abiotic stress response (Reddy et al., 2002; Rabbani et al., 2003; Wang et al., 2007). Interestingly, comparison of the have mostly remained unidentified. In this study, transcriptome profiling of whole seedlings of Indica rice genotypes with contrasting drought tolerance was analysed comprehensively. We used transcriptome data for expression levels of genes between DT and DS genotypes revealed that 711 and 757 probes were up -and downregulated, respectively in DT compared with DS under control conditions (Table S4). These genes might be the identification of differences in drought-induced changes in biochemical pathways between DT and DS. Important cis-regulatory element profiling within the 1- kb upstream sequences from translational start site responsible for higher intrinsic tolerance to abiotic stresses in DT. Comparison of the expression levels of transcriptome under drought revealed that 1900 probes were up-regulated and 920 probes were down-regulated in DT (ATG) of the co-expressed drought-responsive genes was also investigated. when compared to DS (Table S5). In both basal and induced transcriptomes, expression levels of several ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd, Plant Biotechnology Journal, 1–13 in DT, downcondistress- 4 Sangram K. Lenka et al. of drought-responsive genes of DS (IR64 control vs. drought) and DT (N22 control vs. drought) was compared. In response to drought in the two genotypes, 1789 common probes showed up-regulation, whereas 1573 common probes showed down-regulation (Table S6). We considered these consensus set of drought-responsive transcripts across the two genotypes (DT and DS) as drought-responsive co-expressed genes for overrepresented motif discovery. Validation of expression of selected TF-encoding genes From 1900 probes that were up-regulated and 920 downregulated probes under drought in DT compared to DS (Figure 2, Table S5), 77 up-regulated probes and 14 down-regulated probes represented TFs and displayed a 3.5 versatile regulation pattern under different conditions (Table S7). The identified TFs belonged to diverse family of TFs such as ZFP, MADS-box, LZP, WRKY, HSF, NAC, NFY, etc. Interestingly, we found that the genes encoding proteins with zinc finger domains were the most enriched Figure 1 Hierarchical clustering. Hierarchical clustering of significantly expressed genes is displayed by average linkage and Euclidean distance as a measurement of similarity. stress-responsive genes showed differential expression between DT and DS. This clearly demonstrates that preferential gene regulation owing to G · E is crucial for determining transcriptome composition and thus stress responses of the genotypes suggesting thereby that the genomes of both the genotypes were preprogrammed to modulate expression of different sets of genes in response to both control and drought stress conditions. To identify drought-responsive probe sets that are regulated by drought, independent of the genotype, expression pattern –Log10 (corrected P value) –3.5 N22 drought IR64 drought N22 control IR64 control 0 4 3 2 1 –6 –4 –2 0 2 Log2 (fold change) 4 6 Figure 2 Volcano plot. Volcano plot illustrating the differentially expressed probes in DT (drought tolerant landrace N22) versus DS (high-yielding cultivar IR64) under drought stress. The X-axis represents the fold change in DT compared to DS (on a log2 scale), and the Y-axis represents the negative log10-transformed P-values (P < 0.05) of the t-test for finding differences between the samples. ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd, Plant Biotechnology Journal, 1–13 Drought-responsive transcriptome analysis in rice 5 (46.15%) among the up-regulated TFs in DT. The role of different members of zinc finger protein–encoding genes in abiotic stress tolerance is well documented (Liu et al., 2007; Huang et al., 2008; Xu et al., 2008). These TFs have also been reported to activate cascades of genes that act screening approach. Detailed characterization of the overrepresented drought-responsive cis-motif identified in this study (Figure S4a–e) might lead to the identification of sub-regulons that play a key role in drought tolerance in rice. There was no significant enrichment of a particular together in enhancing tolerance to multiple stresses (Bhatnagar-Mathur et al., 2008). Investigating the role of these TF-encoding candidate genes identified in this study (Table 1, Table S7) by over-expression in transgenic plants cis-motif obtained from the down-regulated genes. will further enhance our understanding of the mechanisms that impart abiotic stress tolerance to rice. As N22 is a well-known DT land race, we expect that N22 alleles of the identified candidate TFs (Table 1, Table S7) may per- Broad molecular functions of the differentially expressed genes were analysed in terms of GO enrichments (P £ 0.05) for molecular function. In DS control versus DS drought down-regulated probes, the significantly down- form better than the DS IR64 alleles. Similarly, functional characterization of down-regulated master regulatory genes (TFs) under drought in DT can be tested by RNAi approach (Table S7). regulated genes belonged to three important regulatory functional classes, namely nucleic acid binding (GO:0003676) (P = 0.04) representing 261 probes, DNA binding (GO:0003677) (P = 3.6e)04) with 212 probes and To validate the microarray results and quantify the expression of regulatory genes, 38 probe sets representing TF-encoding genes that showed differential regulation between DT and DS under drought stress were chosen. Although the microarray log2-fold change value of a probe transcription regulator activity (GO:0030528) (P = 2.9e)05) represented by 187 probes. The above regulatory GO classes notably include four probes of DREB, three probes of bZIP and ten probes of NAC TF family members among other important regulatory classes of genes that differed somewhat from the corresponding value of qRTPCR, the high correlation (R2 = 0.92, P < 0.05) between microarray and qRT-PCR expression values indicated that the expression analysis by both the approaches was in were earlier reported to confer drought tolerance in transgenic rice (Fang et al., 2008; Wang et al., 2008; Xiang et al., 2008). This analysis suggests that the susceptibility of IR64 is probably attributable to significant down- good agreement with each other (Figure 3). regulation of regulatory components that confer drought tolerance. Comparison of the transcriptome expression pattern under control condition between DT and DS led to the identification of 264 up-regulated probes that Consensus cis-motif pattern finding Gene ontology enrichment for molecular function Integration of co-expression profile data with cis-motif consensus pattern or promoter structure is crucial to understand the universal and organism-level molecular networking (Lenka et al., 2009). Promoter sequences of possessed catalytic activity (GO:0003824) and were significantly enriched (P = 8.9e)03) in DT. Comparative expression analysis between DT and DS under drought stress unravelled that 263 probes having catalytic activity the co-expressed genes were analysed to find the cis-motif consensus pattern. Top five significant cis-motif patterns were sampled from the drought up-regulated genes, and their features are given in Table 2. The motifs identified in this study were novel, and the respective TF family mem- (GO:0003824) were up-regulated significantly (P = 0.01) in DT. Similarly, hydrolase activity (GO:0016787) was also enriched (124 representative probes) considerably (P = 3.7e)04) in DT. The analyses suggest that drought tolerance in DT was attributable to enhanced enzymatic bers binding to them were unknown except CCA1 (TTTTTTTTHYW) to which the members of MYB-related TFs were reported to interact in Arabidopsis (Wang et al., 1997). Interestingly, CCA1 and the GC-rich SCGSCGSCG- activity in DT compared to DS. Hydrolases are known to impart stress tolerance to plants by participating in diverse physiological activities (Cho et al., 2006). Drought-induced catalytic components participate in the synthesis or catab- SCG motifs were found to be a common feature of rice genome, as reported in our previous study (Lenka et al., 2009). It will be useful to further characterize the above cis-elements by deletion analysis of promoters and linker olism of drought stress–associated metabolites, and some of these have been used to enhance stress tolerance in transgenic plants (Umezawa et al., 2006). Our results thus suggest that the preparedness of DT with higher enzy- scanning approach. More TFs could also be identified using these cis-elements as bait by yeast one-hybrid matic activity even under control conditions might be an evolutionarily acquired adaptive trait. ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd, Plant Biotechnology Journal, 1–13 6 Sangram K. Lenka et al. Table 1 Up-regulated genes in drought-tolerant rice landrace Nagina 22 (DT) identified by microarray analysis and validated by qRT-PCR Abiotic stress-responsive expression ⁄ phenotype of Probe set ID Protein function transgenic plants References Os.4174.1.S1_at AGL12 (AGAMOUS-LIKE 12); AGL12 induces multiple responses that are related to various Lee et al. (2008) transcription factor (TF) Os.4175.1.S1_at MADS-box protein stresses in transgenic rice and Arabidopsis Abiotic stress-induced response reported in MADS-box Arora et al. (2007) gene family Os.4706.1.S1_a_at MADS box-like protein Abiotic stress-induced response reported in MADS-box Arora et al. (2007) gene family Os.40018.1.S1_at Heat stress TF Imparts enhanced tolerance to heat and high salinity in Yokotani et al. (2008) transgenic Arabidopsis Os.7928.1.S1_at Putative CCAAT-binding TF Induction in transcription observed owing to drought stress This study Os.37813.1.S1_at MADS-box TF 56 Abiotic stress-induced response reported in MADS-box Arora et al. (2007) Os.15711.1.S1_at Putative bZIP TF Transgenic rice over-expressing OsbZIP23 showed gene family Xiang et al. (2008) significantly improved tolerance to drought and high-salinity stresses and sensitivity to abscisic acid NAC domain TF OsbZIP72 enhanced ability to tolerate drought stress Lu et al. (2009) ONAC063 imparts high-salinity and osmotic stress Yokotani et al. (2009) tolerance Transgenic rice plants over-expressing ONAC045 showed Zheng et al. (2009) enhanced tolerance to drought and salt treatments Os.23103.1.S1_at Ethylene-responsive element binding factor Enhances osmotic and drought tolerance in rice by Quan et al. (2010) modulating the increase in stress-responsive gene expression OsAffx.12789.1.S1_s_at TF MADS47 Abiotic stress-induced response reported in MADS-box Arora et al. (2007) gene family OsAffx.24128.1.S1_s_at Putative ethylene-responsive element binding factor Enhanced osmotic and drought tolerance in rice by Zheng et al. (2009) modulating the increase in stress-responsive gene expression Os.56298.1.S1_at WRKY TF 35 Enhanced heat and drought tolerance in transgenic Wu et al. (2009) rice seedlings OsAffx.14373.1.S1_s_at Ethylene-responsive element Enhanced osmotic and drought tolerance in rice by binding protein homolog modulating the increase in stress-responsive gene Quan et al. (2010) expression Os.3710.1.S1_at Predicted RING-H2 finger protein Confers drought tolerance upon induction Ko et al. (2006) ATL3F Os.11773.1.S1_at WRKY33; TF Responsive to abiotic stresses Ramamoorthy et al. (2008) Os.12032.1.S1_at TF OsWRKY71 Responsive to abiotic stresses Ramamoorthy et al. (2008) Os.2364.1.S1_at Homeodomain-leucine zipper TF Implicated in stress adaptation Agalou et al. (2008) Os.2363.1.S1_a_at Homeodomain TF HOX6 Implicated in stress adaptation Agalou et al. (2008) Os.7912.1.S1_at Zinc finger TF ZF1 Members of zinc finger TF family shown to confer Huang et al. (2009) drought tolerance in rice Os.27227.1.S1_at WRKY family TF-like Responsive to abiotic stresses Ramamoorthy et al. (2008) Os.11450.1.S1_at RING finger and CHY zinc finger Drought-induced transcription observed This study Os.11774.1.S1_s_at Zinc finger TF-like protein Members of zinc finger TF family shown to confer Huang et al. (2009) domain-containing protein drought tolerance in rice OsAffx.11050.2.S1_x_at OsWRKY1v2—Superfamily of TFs Responsive to abiotic stresses Ramamoorthy et al. (2008) Os.54569.1.S1_x_at Putative tuber-specific and Induction in transcription observed owing to drought This study sucrose-responsive element stress binding factor Os.19172.1.S1_at Putative zinc finger protein Members of zinc finger TF family shown to confer Huang et al. (2009) drought tolerance in rice Os.14823.1.S1_s_at P-type R2R3 Myb protein Over-expression of R2R3 Myb improves drought and Ding et al. (2009) salt tolerance Os.30568.1.S1_at WRKY TF 48-like protein Responsive to abiotic stresses Ramamoorthy et al. (2008) ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd, Plant Biotechnology Journal, 1–13 Drought-responsive transcriptome analysis in rice 7 Table 1 (Continued) Abiotic stress-responsive expression ⁄ phenotype of Probe set ID Protein function transgenic plants References Os.10172.1.S1_at Putative myb protein Members of MYB family well implicated in drought Dai et al. (2007) and other abiotic stress tolerance OsAffx.3135.1.S1_at Putative MYB TF Members of MYB family well implicated in drought Dai et al. (2007) and other abiotic stress tolerance Os.49023.1.S1_x_at Zinc finger (C3HC4-type RING finger) family protein Os.6261.1.S1_at Zinc finger (C3HC4-type RING finger) Abiotic stress-responsive regulation of C3HC4-type Putative TF WRKY5 Responsive to abiotic stresses R 2 = 0.9246 Log ratio (real-time RT PCR) 10 5 0 –5 0 5 10 Ramamoorthy et al. (2008) vars (Figure S5d). Induction of energy-related functions (glycolysis and gluconeogenesis) indicates the evolutionarily conserved process in both the genotypes to sustain the housekeeping function even under dehydration stress. Similar categories of ESTs were also abundantly induced in 15 –10 Ma et al. (2009) RING finger gene family observed in rice 20 –15 Ma et al. (2009) RING finger gene family observed in rice family protein Os.11945.1.S1_at Abiotic stress-responsive regulation of C3HC4-type 15 –5 –10 –15 Log ratio (microarray) Figure 3 Validation of the expression of selected probes representing transcription factor (TF) from microarray by qRT-PCR. Correlation analysis shows selected probes representing TF between microarray, and qRT-PCR experiments are in good agreement with each other. The fold changes in gene expression were transformed to log2 scale. The microarray data log2-values (X-axis) were plotted against the real-time RT-PCR log2 values (Y-axis). Drought-responsive pathway analysis The common drought-responsive probes that were up-regulated in both DT and DS under drought stress were analysed. Three important pathways were found to be significantly up-regulated (FDR = 0.05) in both the cultivars. These pathways included carbon fixation (17 enzymes, P = 9.52e)08) (Figure S5a), glycolysis ⁄ gluconeogenesis (13 enzymes, P = 3.94e)03) (Figure S5b) and flavonoid biosynthesis pathway (seven key enzymes, P = 2.63e)02) (Figure S5c). On the contrary, 11 vital enzymes of starch and sucrose metabolism (P = 2.28e)02) were considerably down-regulated under drought stress in both the rice culti- adaptive response to dehydration stress in Selaginella lepidophylla, an ancient lineage of vascular plants that can withstand complete desiccation for years and can be revived after only a few hours of rehydration (Iturriaga et al., 2006). Flavonoids are an important group of secondary metabolites with diverse molecular functions, including stress protection in plants (Winkel-Shirley, 2002). Previous reports have shown that flavonoid biosynthetic pathway is induced in salt-sensitive rice genotype IR29 during salt stress at the vegetative growth stage (Walia et al., 2005). Similarly, in cultivar N22 (DT), transcript level and ⁄ or transcript stability of components of the flavonoid pathway were reported to be significantly enhanced in seedlings treated with abscisic acid (ABA), dehydration or high salt stress (Ithal and Reddy, 2004). Induction of genes encoding important enzymes involved in the flavonoid biosynthetic pathway, as also observed in this study, appears to be a characteristic response of rice for protection against various stressinduced injuries. Reduction in starch and sucrose metabolism in both DT and DS (Figure S5d) suggests that drought stress negatively affects the export or the utilization of assimilates in the sink organs. Reduction in starch and sucrose metabolism also leads to reduction in starch and sucrose content—a welldocumented phenomenon in several higher plant species under drought conditions (Lawlor and Cornic, 2002). Interestingly, ten enzymes of the starch and sucrose metabolism pathway (P = 3.56e)02) were down-regulated in DT when compared to DS under drought (Figure S5e). This suggests that the level of starch and sucrose metabolism might be ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd, Plant Biotechnology Journal, 1–13 8 Sangram K. Lenka et al. Table 2 MEME-generated motifs sampled from the 1-kb upstream region (from ATG) within the common drought up-regulated regulons Occurrence of the Log likelihood MEME-generated motifs ratio (llr) E-value motif in drought Illustration as up-regulated WebLogo promoters (%) (Suppl. Fig. No.) Description CBCCKCCKCCKC 5414 2.0e)453 48.69 S4a Novel motif RRRRRRGARRRR 5757 1.7e)341 57.1 S4b Novel motif TTTTTTTTHYW 5722 2.0e)238 60.37 S4c CCA1 binding site (common feature of rice genome) SCGSCGSCGSCG 3466 1.4e)237 30 S4d Common feature of rice genome YYYCTCYYYYYC 5052 4.1e)264 49.62 S4e Novel motif reduced in DT (low yielding land race) under drought when compared to DS (high-yielding cultivar). Comparing the pathways between DS control and drought, it was evident that 19 enzymes of carbon fixation ids of rice leaves. Under well-watered conditions, the content of 18:3 lipids was reported to be the highest among the other membrane lipids. However, decreasing trend has been observed in the content of unsaturated fatty acids (P = 4.96e)06) and 19 enzymes of glycolysis ⁄ gluconeogenesis (P = 1.11e)04) were up-regulated, whereas 14 vital enzymes of starch and sucrose metabolism (P = 1.54e)02) were significantly down-regulated in response to drought with increasing water stress in plants (Chen et al., 2004). Constitutive antisense expression of an Arabidopsis omega3 fatty acid desaturase gene led to reduced salt ⁄ drought tolerance in transgenic tobacco plants (Im et al., 2002). On stress (pathway diagram not shown). Similar comparison in DT also revealed significant up-regulation of 20 enzymes of carbon fixation (P = 2.75e)06) and 17 enzymes of glycolysis ⁄ gluconeogenesis (P = 2.21e)02) (pathway diagram not shown). In addition to these two pathways, eight enzymes the other hand, over-expression of two fatty acid desaturase genes FAD3 or FAD8 resulted in increased tolerance to drought in tobacco plants and to osmotic stress in cultured cells suggesting thereby an inverse correlation between drought-responsive decreased levels of linolenic acid and of a-linolenic acid metabolism (P = 3.67e)02) were significantly induced under drought in DT (Figure 4). a-linolenic acid (18:3) is the major class of fatty acids in membrane lip- enhanced drought-induced damage to plant systems under stress (Zhang et al., 2005). Thus, an elevated a-linolenic acid metabolism in DT under drought appears to be in good C01226 C16324 5.3.99.6 1.3.1.42 C04672 C04780 C16328 C16329 C16332 C16336 C16330 4.2.1.74 4.2.1.17 1.1.1.35 1.1.1.211 1.3.3.6 C16333 C16337 C16327 C16334 C16338 C11512 2.3.1.16 C16335 C16331 2.1.1.141 C16339 C08491 4.2.1.92 C16321 C04785 1.13.11.12 C06427 3.1.1.32 3.1.1.4 C00157 Figure 4 Up-regulation of a-linolenic acid metabolism in drought-tolerant (DT) land race of rice (Nagina 22) under drought. Eight enzymes of a-linolenic acid metabolism (P = 3.67e)02) significantly induced under drought in DT are highlighted in yellow, and probes present in Affymetrix array and not induced significantly are highlighted in grey. Enzymes are given here as EC number, and carbon compounds are given as KEGG compound ID. ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd, Plant Biotechnology Journal, 1–13 Drought-responsive transcriptome analysis in rice 9 agreement with the inherent drought tolerance capacity of the rice landrace N22; it might also explain better recovery capacity from drought upon rehydration presumably because of efficient membrane reconstitution ability. Our results clearly demonstrate that N22 can serve as an excel- lings per pot was raised. The level of stress was quantified by measuring RWC of leaves of rice seedling. lent genetic resource for the identification of novel alleles encoding enzymes (Table S8) associated with drought tolerance for genetic improvement of rice and other cereals. To measure RWC, leaf samples were weighed immediately as fresh weight (FW), then chopped into 2- cm pieces and floated on distilled water for 4 h at 4 C. The turgid leaf pieces were then quickly blotted to remove extra surface water and weighed to record turgid weight (TW). The leaf samples were then packed in paper bags and oven-dried at 80 C for 24 h, and the dry weight (DW) of the samples were weighed. RWC was calculated as Conclusions Using expression profiling of Affymetrix Rice Genome Array, it was found that the drought-tolerant rice N22 (DT) and drought-susceptible rice IR64 (DS) exhibited a diverse global transcriptional response under both control and drought stress conditions. GO analysis suggested that drought tolerance of DT was found to be linked to enhanced enzymatic activity, whereas drought susceptibility of DS was governed by significant down-regulation of transcriptional regulatory protein–encoding genes, some of which were previously shown to confer drought tolerance in transgenic plants. Pathway analysis unravelled up-regulation of several components of carbon fixation, glycolysis ⁄ gluconeogenesis and flavonoid biosynthesis and down-regulation of starch and sucrose metabolism in both the cultivars under drought. Enhanced a-linolenic acid metabolism in DT under drought appears to be in accordance with its well-reported drought tolerance. The study revealed genotype-dependent drought tolerance mechanisms in DT versus DS. Functional validation of droughtresponsive genes identified in this study will help to dissect the complexity of drought tolerance at molecular level and subsequently enhance the pace of genetic improvement in drought tolerance in rice and other crop plants. Experimental procedures Plant materials, growth conditions and stress treatments Rice cultivar Nagina-22 (N-22) and IR-64 were grown side by side in plastic pots for 14 days at 28 ± 1 C with a daily photoperiodic cycle of 14 ⁄ 10 h light ⁄ dark provided by fluorescent tubes (Philips TL 40 W ⁄ 54, @ 100–125 lmol ⁄ m2 ⁄ s). Sterile absorbent cotton soaked with Hoagland’s solution was used as seed bed for growing rice seedlings. Water stress was given to the plants by withholding water supply till visible leaf rolling appeared in the plants. To impose equal level of stress, the amount of cotton used for seed bed, the pot size and the quantity of Hoagland’s solution were kept constant for all the pots, and same number of seed- Physiological analysis RWC ð%Þ ¼ ½FW  DWÞ=ðTW  DW  100: The chlorophyll content was measured using the method suggested by Arnon (1949) (Arnon, 1949). Initially, 0.05 g of leaf tissue was placed in 10 mL of DMSO in test tubes and incubated at 65 C for 4 h. Then, the test tubes were cooled to room temperature, and the absorbance was recorded at 645 and 663 nm, respectively. The excised-leaf water loss (ELWL) or retention of water in excised leaves was measured by recording the rate of water loss in leaf segments of 3 cm in length and by determining weight loss in samples periodically; i.e. the leaves were weighed immediately after sampling (FW), drying in an incubator at 28 C at 50% R.H. and weighing at each 1- h intervals up to 5 h and then oven-drying for 24 h at 70 C. ELWL was then calculated as ELWL ¼ Fresh weight  Weight after desire time interval=Fresh weight  Dry weight  100 Whole-genome expression analysis High-quality RNA was extracted from the whole seedlings (combined root and leaf samples) using TRI Reagent (Ambion, Inc., Austin, TX) and pooled from 12 independent stressed and non-stressed plant samples separately and treated with DNase-I (QIAGEN GmbH, Hilden, Germany). Subsequently, RNA clean-up was carried out using RNeasy Plant Mini Kit (QIAGEN GmbH), and 5 lg of total RNA from each sample with three biological replications was reverse-transcribed to double-stranded cDNA using the GeneChip One-Cycle cDNA Synthesis Kit. The biotin-labelled cRNA was made using the GeneChip IVT Labelling Kit (Affymetrix, Santa Clara, CA, USA). Twenty microgram of cRNA samples was fragmented, out of which 7.5 lg cRNA were hybridized for 16 h at 45 C to the Affymetrix GeneChip Rice Genome Array. After washing and staining with R-phycoerythrin streptavidin in a Fluidics Station, using the Genechip Fluidics Station 450, the arrays were scanned by the Genechip 3000 Scanner. The chip images were scanned and extracted using default settings, and the. CEL files were produced with the Affymetrix GeneChip Operating Software (GCOS 1.2). The resulting. CEL files were imported ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd, Plant Biotechnology Journal, 1–13 10 Sangram K. Lenka et al. into the GeneSpring GX 10 (Agilent Technologies Inc., Santa Clara, CA, USA) and normalized with the PLIER16 algorithm (Hubbell et al., 2002). The resulting expression values were log2-transformed. Average log signal intensity values of three biological replicates for each sample were used for advance analysis. Probes showing twofold up- or down-regulation compared to corresponding control condition were taken only into consideration for further analysis. All details of data are available for public access online at http://www.ebi.ac.uk/microarray/submissions. html (ArrayExpress accession: E-MEXP-2401). Differential gene expression was identified using significance analysis by unpaired Student’s t-test between appropriate pair wise comparison of different samples under consideration. Benjamini and Hochberg false discovery rate (FDR) multiple testing corrections was applied to the differentially expressed genes (P < 0.05). Hierarchical clustering of significantly expressed genes was carried out by average linkage and Euclidean distance as a measurement of similarity using GeneSpring GX 10. qRT-PCR analysis Similar growth conditions and level of stress were used for qPCR analysis as in the case of microarray experiments to validate the expression patterns of TF genes. First strand cDNA was synthesized using 2 lg of total RNA using Superscript-III RNase H– Reverse Transcriptase (Invitrogen, Carlsbad, CA) with oligo (dT) 20 primer following manufacturer’s instructions. qPCR of 20 lL each containing 50 ng of cDNA was conducted in an Eppendorf realplex-4 Mastercycler ep gradients machine. Rice actin gene was used as the endogenous control after testing positive for almost unaltered expression across different conditions tested here. For microarray data validation, QuantiFast SYBR Green PCR master mix (QIAGEN GmbH) was used according to manufactures instruction. The primer combinations used here for real-time RT-PCR analysis were specifically amplified only one desired band. The dissociation curve testing was carried out for each primer pair showing only one melting temperature. The efficiency test of three randomly selected primer pairs showed approximately the same efficiency as that of the normalizer actin at a series dilution of each group cDNA at rates of 1 ⁄ 10, 1 ⁄ 100 and 1 ⁄ 1000. The threshold cycles (CT) of each test target were averaged for triplicate reactions, and the values were normalized according to the CT of the control products (Os-actin). TFs expression data were normalized by subtracting the mean reference gene CT value from individual CT values of corresponding target genes (DCT). The fold change value was calculated using the expression, where DDCT represents DCT condition of interest -DCT control. The results obtained were transformed to log2 scale. The primer sets used in this study to validate microarray results are given in the Table S9. Over-represented cis-element finding Drought stress co-regulated probes were mapped onto the TIGR rice pseudomolecules, release 5.0 (http://www.tigr.org), and corresponding loci were listed after removing the duplicates. The file containing all the 1- kb upstream sequences (promoters) from translational start site ATG of rice genome was downloaded from TIGR and filtered to obtain sub-files with desired set of co-regulated promoters using Perl script. The widely used expectation maximization algorithm of MEME (Multiple EM for motif elicitation) (Version 4.0.0) was used to find over-represented cismotifs of a width of 8–12 nucleotides, on Linux x86_64 machine (Bailey et al., 2006). The relevance of discovered motifs was analysed using PLACE (http://www.dna.affrc.go.jp/PLACE/) (Higo et al., 1999). Molecular function enrichment and pathway analysis The broad molecular function of the differentially expressed probes was analysed in terms of significantly enriched GO categories for molecular function using EasyGO, GO enrichment analysis tool (http://bioinformatics.cau.edu.cn/easygo/) (Zhou and Su, 2007). Binomial statistical test and cut-off for FDR-adjusted P-value £0.05 was used to screen the GO term enrichment. Significantly influenced entire metabolic pathway analysis was carried out using PathExpress analysis tool (http://bioinfoserver.rsbs. anu.edu.au/utils/PathExpress/) with P-value threshold £0.05 where FDR was used for P-value adjustment (Goffard and Weiller, 2007). The list of unique enzymes (EC number) up- or down-regulated by different conditions within a corresponding pathway is given in Table S8. Authors’ contributions SKL performed all the wet-lab experiments, designed the study, analysed the data and drafted the manuscript; AK made Perl script and helped in data mining and management; VC and KCB facilitated in designing of the study and drafting the manuscript. All authors read and approved the final manuscript. Acknowledgements SKL gratefully acknowledges University Grants Commission (UGC) and Council of Scientific and Industrial Research (CSIR) for CSIR-UGC Junior Research Fellowship grant. This work was supported by the Indian Council of Agricultural Research (ICAR)-sponsored Network Project on Transgenics in Crops (NPTC). References Arnon, D.I. (1949) Copper enzymes in isolated chloroplasts. Polyphenoloxidase in beta vulgaris. Plant Physiol., 24, 1–15. Arora, R., Agarwal, P., Ray, S., Singh, A.K., Singh, V.P., Tyagi, A.K. and Kapoor, S. (2007) MADS-box gene family in rice: genome-wide identification, organization and expression profiling during reproductive development and stress. BMC Genomics, 8, 242. Agalou, A., Purwantomo, S., Overnas, E., Johannesson, H., Zhu, X., Estiati, A., de Kam, R.J., Engstrom, P., Slamet-Loedin, I.H., Zhu, Z., Wang, M., Xiong, L., Meijer, A.H. and Ouwerkerk, P.B. ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd, Plant Biotechnology Journal, 1–13 Drought-responsive transcriptome analysis in rice 11 (2008) A genome-wide survey of HD-Zip genes in rice and analysis of drought-responsive family members. Plant Mol Biol, 66, 87–103. Bailey, T.L., Williams, N., Misleh, C. and Li, W.W. (2006) MEME: discovering and analyzing DNA and protein sequence motifs. Nucleic Acids Res., 34, W369–W373. Bhatnagar-Mathur, P., Vadez, V. and Sharma, K.K. (2008) Transgenic approaches for abiotic stress tolerance in plants: retrospect and prospects. Plant Cell Rep., 27, 411–424. Chen, G., Liu, S.H., Zhang, C.J. and Lu, C.G. (2004) Effects of drought on photosynthetic characteristics of flag leaves of a newly-developed super-high-yield rice hybrid. Photosynthetica, 42, 573–578. Cho, S.K., Kim, J.E., Park, J.A., Eom, T.J. and Kim, W.T. (2006) Constitutive expression of abiotic stress-inducible hot pepper CaXTH3, which encodes a xyloglucan endotransglucosylase ⁄ hydrolase homolog, improves drought and salt tolerance in transgenic Arabidopsis plants. FEBS Lett., 580, 3136–3144. Dai, X., Xu, Y., Ma, Q., Xu, W., Wang, T., Xue, Y. and Chong, K. (2007) Overexpression of an R1R2R3 MYB gene, OsMYB3R-2, increases tolerance to freezing, drought, and salt stress in transgenic Arabidopsis. Plant Physiol., 143, 1739–1751. Degenkolbe, T., Do, P.T., Zuther, E., Repsilber, D., Walther, D., Hincha, D.K. and Kohl, K.I. (2009) Expression profiling of rice cultivars differing in their tolerance to long-term drought stress. Plant Mol. Biol., 69, 133–153. Ding, Z., Li, S., An, X., Liu, X., Qin, H. and Wang, D. (2009) Transgenic expression of MYB15 confers enhanced sensitivity to abscisic acid and improved drought tolerance in Arabidopsis thaliana. J Genet Genomics, 36, 17–29. Fang, Y., You, J., Xie, K., Xie, W. and Xiong, L. (2008) Systematic sequence analysis and identification of tissue-specific or stressresponsive genes of NAC transcription factor family in rice. Mol. Genet. Genomics, 280, 547–563. Gao, G., Zhong, Y., Guo, A., Zhu, Q., Tang, W., Zheng, W., Gu, X., Wei, L. and Luo, J. (2006) DRTF: a database of rice transcription factors. Bioinformatics, 22, 1286–1287. Goffard, N. and Weiller, G. (2007) PathExpress: a web-based tool to identify relevant pathways in gene expression data. Nucleic Acids Res., 35, W176–W181. Guo, J., Wu, J., Ji, Q., Wang, C., Luo, L., Yuan, Y., Wang, Y. and Wang, J. (2008) Genome-wide analysis of heat shock transcription factor families in rice and Arabidopsis. J. Genet. Genomics, 35, 105–118. Higo, K., Ugawa, Y., Iwamoto, M. and Korenaga, T. (1999) Plant cis-acting regulatory DNA elements (PLACE) database: 1999. Nucleic Acids Res., 27, 297–300. Hu, H., Dai, M., Yao, J., Xiao, B., Li, X., Zhang, Q. and Xiong, L. (2006) Overexpressing a NAM, ATAF, and CUC (NAC) transcription factor enhances drought resistance and salt tolerance in rice. Proc. Natl. Acad. Sci. U S A, 103, 12987– 12992. Huang, J., Wang, M.M., Jiang, Y., Bao, Y.M., Huang, X., Sun, H., Xu, D.Q., Lan, H.X. and Zhang, H.S. (2008) Expression analysis of rice A20 ⁄ AN1-type zinc finger genes and characterization of ZFP177 that contributes to temperature stress tolerance. Gene, 420, 135–144. Huang, X.Y., Chao, D.Y., Gao, J.P., Zhu, M.Z., Shi, M. and Lin, H.X. (2009) A previously unknown zinc finger protein, DST, regulates drought and salt tolerance in rice via stomatal aperture control. Genes Dev, 23, 1805–1817. Hubbell, E., Liu, W.M. and Mei, R. (2002) Robust estimators for expression analysis. Bioinformatics, 18, 1585–1592. Im, Y.J., Han, O., Chung, G.C. and Cho, B.H. (2002) Antisense expression of an Arabidopsis omega-3 fatty acid desaturase gene reduces salt ⁄ drought tolerance in transgenic tobacco plants. Mol. Cells, 13, 264–271. IRGSP (2005) The map-based sequence of the rice genome. Nature, 436, 793–800. Ithal, N. and Reddy, A.R. (2004) Rice flavonoid pathway genes, OsDfr and OsAns, are induced by dehydration, high salt and ABA, and contain stress responsive promoter elements that interact with the transcription activator, OsC1-MYB. Plant Sci., 166, 1505–1513. Iturriaga, G., Cushman, M.A.F. and Cushman, J.C. (2006) An EST catalogue from the resurrection plant Selaginella lepidophylla reveals abiotic stress-adaptive genes. Plant Sci., 170, 1173– 1184. Ko, J.H., Yang, S.H. and Han, K.H. (2006) Upregulation of an Arabidopsis RING-H2 gene, XERICO, confers drought tolerance through increased abscisic acid biosynthesis. Plant J, 47, 343– 355. Kushwaha, H.R., Singh, A.K., Sopory, S.K., Singla- Pareek, S.L. and Pareek, A. (2009) Genome wide expression analysis of CBS domain containing proteins in Arabidopsis thaliana (L.) Heynh and Oryza sativa L. reveals their developmental and stress regulation. BMC Genomics, 10, 200. Lafitte, H.R., Yongsheng, G., Yan, S. and Li, Z.K. (2007) Whole plant responses, key processes, and adaptation to drought stress: the case of rice. J. Exp. Bot., 58, 169–175. Lawlor, D.W. and Cornic, G. (2002) Photosynthetic carbon assimilation and associated metabolism in relation to water deficits in higher plants. Plant Cell Environ., 25, 275–294. Lee, S., Woo, Y.M., Ryu, S.I., Shin, Y.D., Kim, W.T., Park, K.Y., Lee, I.J. and An, G. (2008) Further characterization of a rice AGL12 group MADS-box gene, OsMADS26. Plant Physiol, 47, 156–168. Lenka, S.K., Lohia, B., Kumar, A., Chinnusamy, V. and Bansal, K.C. (2009) Genome-wide targeted prediction of ABA responsive genes in rice based on over-represented cis-motif in co-expressed genes. Plant Mol. Biol., 69, 261–271. Liu, K., Wang, L., Xu, Y., Chen, N., Ma, Q., Li, F. and Chong, K. (2007) Overexpression of OsCOIN, a putative cold inducible zinc finger protein, increased tolerance to chilling, salt and drought, and enhanced proline level in rice. Planta, 226, 1007–1016. Lu, G., Gao, C., Zheng, X. and Han, B. (2009) Identification of OsbZIP72 as a positive regulator of ABA response and drought tolerance in rice. Planta, 229, 605–615. Ma, K., Xiao, J., Li, X., Zhang, Q. and Lian, X. (2009) Sequence and expression analysis of the C3HC4-type RING finger gene family in rice. Gene, 444, 33–45. Oh, S.J., Kim, Y.S., Kwon, C.W., Park, H.K., Jeong, J.S. and Kim, J.K. (2009) Overexpression of the transcription factor AP37 in rice improves grain yield under drought conditions. Plant Physiol., 150, 1368–1379. Price, A.H., Cairns, J.E., Horton, P., Jones, H.G. and Griffiths, H. (2002) Linking drought-resistance mechanisms to drought avoidance in upland rice using a QTL approach: progress and ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd, Plant Biotechnology Journal, 1–13 12 Sangram K. Lenka et al. new opportunities to integrate stomatal and mesophyll responses. J. Exp. Bot., 53, 989–1004. Quan, R., Hu, S., Zhang, Z., Zhang, H. and Huang, R. (2010) Overexpression of an ERF transcription factor TSRF1 improves rice drought tolerance. Plant Biotechnol J, 8, 476–488. Rabbani, M.A., Maruyama, K., Abe, H., Khan, M.A., Katsura, K., Ito, Y., Yoshiwara, K., Seki, M., Shinozaki, K. and YamaguchiShinozaki, K. (2003) Monitoring expression profiles of rice genes under cold, drought, and high-salinity stresses and abscisic acid application using cDNA microarray and RNA gelblot analyses. Plant Physiol., 133, 1755–1767. Ramamoorthy, R., Jiang, S.Y., Kumar, N., Venkatesh, P.N. and Ramachandran, S. (2008) A comprehensive transcriptional profiling of the WRKY gene family in rice under various abiotic and phytohormone treatments. Plant Cell Physiol., 49, 865–879. Reddy Ch, S., Babu, A.P., Swamy, B.P., Kaladhar, K. and Sarla, N. (2009) ISSR markers based on GA and AG repeats reveal genetic relationship among rice varieties tolerant to drought, flood, or salinity. J. Zhejiang Univ. Sci. B, 10, 133–141. Reddy, A.R., Ramakrishna, W., Sekhar, A.C., Ithal, N., Babu, P.R., Bonaldo, M.F., Soares, M.B. and Bennetzen, J.L. (2002) Novel genes are enriched in normalized cDNA libraries from droughtstressed seedlings of rice (Oryza sativa L. subsp. indica cv. Nagina 22). Genome 45:204–211. Riechmann, J.L., Heard, J., Martin, G., Reuber, L., Jiang, C., Keddie, J., Adam, L., Pineda, O., Ratcliffe, O.J., Samaha, R.R., Creelman, R., Pilgrim, M., Broun, P., Zhang, J.Z., Ghandehari, D., Sherman, B.K. and Yu, G. (2000) Arabidopsis transcription factors: genome-wide comparative analysis among eukaryotes. Science, 290, 2105–2110. Sreenivasulu, N., Sopory, S.K. and Kavi Kishor, P.B. (2007) Deciphering the regulatory mechanisms of abiotic stress tolerance in plants by genomic approaches. Gene, 388, 1–13. Umezawa, T., Fujita, M., Fujita, Y., Yamaguchi-Shinozaki, K. and Shinozaki, K. (2006) Engineering drought tolerance in plants: discovering and tailoring genes to unlock the future. Curr. Opin. Biotechnol., 17, 113–122. Valliyodan, B. and Nguyen, H.T. (2006) Understanding regulatory networks and engineering for enhanced drought tolerance in plants. Curr. Opin. Plant Biol., 9, 189–195. Walia, H., Wilson, C., Condamine, P., Liu, X., Ismail, A.M., Zeng, L., Wanamaker, S.I., Mandal, J., Xu, J., Cui, X. and Close, T.J. (2005) Comparative transcriptional profiling of two contrasting rice genotypes under salinity stress during the vegetative growth stage. Plant Physiol., 139, 822–835. Wang, Z.Y., Kenigsbuch, D., Sun, L., Harel, E., Ong, M.S. and Tobin, E.M. (1997) A Myb-related transcription factor is involved in the phytochrome regulation of an Arabidopsis Lhcb gene. Plant Cell, 9, 491–507. Wang, H., Zhang, H., Gao, F., Li, J. and Li, Z. (2007) Comparison of gene expression between upland and lowland rice cultivars under water stress using cDNA microarray. Theor. Appl. Genet., 115, 1109–1126. Wang, Q., Guan, Y., Wu, Y., Chen, H., Chen, F. and Chu, C. (2008) Overexpression of a rice OsDREB1F gene increases salt, drought, and low temperature tolerance in both Arabidopsis and rice. Plant Mol. Biol., 67, 589–602. Winkel-Shirley, B. (2002) Biosynthesis of flavonoids and effects of stress. Curr. Opin. Plant Biol., 5, 218–223. Witcombe, J.R., Hollington, P.A., Howarth, C.J., Reader, S. and Steele, K.A. (2008) Breeding for abiotic stresses for sustainable agriculture. Philos. Trans. R. Soc. Lond. B Biol. Sci., 363, 703–716. Wu, X., Shiroto, Y., Kishitani, S., Ito, Y. and Toriyama, K. (2009) Enhanced heat and drought tolerance in transgenic rice seedlings overexpressing OsWRKY11 under the control of HSP101 promoter. Plant Cell Rep, 28, 21–30. Xiang, Y., Tang, N., Du, H., Ye, H. and Xiong, L. (2008) Characterization of OsbZIP23 as a key player of the basic leucine zipper transcription factor family for conferring abscisic acid sensitivity and salinity and drought tolerance in rice. Plant Physiol., 148, 1938–1952. Xu, K., Xu, X., Fukao, T., Canlas, P., Maghirang-Rodriguez, R., Heuer, S., Ismail, A.M., Bailey-Serres, J., Ronald, P.C. and Mackill, D.J. (2006) Sub1A is an ethylene-response-factor-like gene that confers submergence tolerance to rice. Nature, 442, 705–708. Xu, D.Q., Huang, J., Guo, S.Q., Yang, X., Bao, Y.M., Tang, H.J. and Zhang, H.S. (2008) Overexpression of a TFIIIA-type zinc finger protein gene ZFP252 enhances drought and salt tolerance in rice (Oryza sativa L.). FEBS Lett., 582, 1037–1043. Yokotani, N., Ichikawa, T., Kondou, Y., Matsui, M., Hirochika, H., Iwabuchi, M. and Oda, K. (2008) Expression of rice heat stress transcription factor OsHsfA2e enhances tolerance to environmental stresses in transgenic Arabidopsis. Planta, 227, 957–967. Yokotani, N., Ichikawa, T., Kondou, Y., Matsui, M., Hirochika, H., Iwabuchi, M. and Oda, K. (2009) Tolerance to various environmental stresses conferred by the salt-responsive rice gene ONAC063 in transgenic Arabidopsis. Planta, 229, 1065– 1075. Zhang, M., Barg, R., Yin, M., Gueta-Dahan, Y., Leikin-Frenkel, A., Salts, Y., Shabtai, S. and Ben-Hayyim, G. (2005) Modulated fatty acid desaturation via overexpression of two distinct omega-3 desaturases differentially alters tolerance to various abiotic stresses in transgenic tobacco cells and plants. Plant J., 44, 361–371. Zheng, X., Chen, B., Lu, G. and Han, B. (2009) Overexpression of a NAC transcription factor enhances rice drought and salt tolerance. Biochem Biophys Res Commun, 379, 985–989. Zhou, X. and Su, Z. (2007) EasyGO: gene ontology-based annotation and functional enrichment analysis tool for agronomical species. BMC Genomics, 8, 246. Zhou, J., Wang, X., Jiao, Y., Qin, Y., Liu, X., He, K., Chen, C., Ma, L., Wang, J., Xiong, L., Zhang, Q., Fan, L. and Deng, X.W. (2007) Global genome expression analysis of rice in response to drought and high-salinity stresses in shoot, flag leaf, and panicle. Plant Mol. Biol., 63, 591–608. Supporting information Additional Supporting Information may be found in the online version of this article: Figure S1 Retention of water (%) in excised leaf in DT and DS under control and drought conditions up to 5 h duration at 28 C and 50% R.H. ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd, Plant Biotechnology Journal, 1–13 Drought-responsive transcriptome analysis in rice 13 Figure S2 Flow-chart showing the strategic outline of microarray experiment. Figure S3 Venn diagram showing differential expression pattern of number of probes under different conditions in both DT and DS cultivars (a–d). (e) Profile plot showing Table S1 The correlation coefficients among the normalised samples of microarray experiment. Table S2 List of probe sets (both up and down regulated), fold change in expression and annotation of DS control vs. DS drought. the differential expression pattern of all the probes under control and drought stress conditions in DS and DT. The gene expression under different conditions (X-axis) is plotted against normalized intensities in log2-values Table S3 List of probe sets (both up and down regulated), fold change in expression and annotation of DT control vs. DT drought. Table S4 List of probe sets (both up and down regulated), (Y-axis). Figure S4 MEME sampled cis-elements discovered within the 1 kb up-stream sequence (from ATG) of drought upregulated regulons. fold change in expression and annotation of DS control vs. DT control. Table S5 List of probe sets (both up and down regulated), fold change in expression and annotation of DS drought Figure S5 Up-regulation of different enzymes of carbon fixation pathway (a), glycolysis/gluconeogenesis (b), and flavonoid biosynthesis pathway (c). The enzymes that were significantly induced in both DT and DS under drought vs. DT drought. Table S6 List of common drought responsive probe sets (both up and down regulated), fold change in expression and annotation of DS control vs. DS drought and DT con- stress are highlighted in yellow, and probes present in Affymetrix array and not induced significantly are highlighted in gray. Enzymes are given here as EC number and carbon compounds are given as KEGG compound ID. (d) Down regulation of different enzymes of starch and trol vs. DT drought. Table S7 List of all the common drought responsive probe sets representing transcription factors encoding genes and their regulation status across all the conditions. Table S8 The list of unique enzymes (EC number) asso- sucrose metabolism under drought stress in both DS and DT are highlighted in yellow, and probes present in Affymetrix array and not down regulated significantly are highlighted in gray. Enzymes are given here as EC number ciated with different pathways up or down regulated by different conditions. Table S9 List of probes sets, their level of expression in microarray, qRT-PCR and corresponding primers used for and carbon compounds are given as KEGG compound ID. (e) Down regulation of different enzymes of starch and sucrose metabolism pathway in DT as compared to DS under drought stress are highlighted in yellow, and probes expression analysis by qRT-PCR. present in Affymetrix array and not down regulated significantly are highlighted in gray. Enzymes are given here as EC number and carbon compounds are given as KEGG compound ID. Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article. ª 2010 The Authors Plant Biotechnology Journal ª 2010 Society for Experimental Biology and Blackwell Publishing Ltd, Plant Biotechnology Journal, 1–13