Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Identification of drought-responsive genes in roots of upland rice (Oryza sativa L)

Identification of drought-responsive genes in roots of upland rice (Oryza sativa L) Background: Rice (Oryza sativa L.) germplasm represents an extraordinary source of genes that control traits of agronomic importance such as drought tolerance. This diversity is the basis for the development of new cultivars better adapted to water restriction conditions, in particular for upland rice, which is grown under rainfall. The analyses of subtractive cDNA libraries and differential protein expression of drought tolerant and susceptible genotypes can contribute to the understanding of the genetic control of water use efficiency in rice. Results: Two subtractive libraries were constructed using cDNA of drought susceptible and tolerant genotypes submitted to stress against cDNA of well-watered plants. In silico analysis revealed 463 reads, which were grouped into 282 clusters. Several genes expressed exclusively in the tolerant or susceptible genotypes were identified. Additionally, proteome analysis of roots from stressed plants was performed and 22 proteins putatively associated to drought tolerance were identified by mass spectrometry. Conclusion: Several genes and proteins involved in drought-response, as well as genes with no described homologs were identified. Genes exclusively expressed in the tolerant genotype were, in general, related to maintenance of turgor and cell integrity. In contrast, in the susceptible genotype, expression of genes involved in protection against cell damage was not detected. Several protein families identified in the proteomic analysis were not detected in the cDNA analysis. There is an indication that the mechanisms of susceptibility to drought in upland rice are similar to those of lowland varieties. Page 1 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 sites for stress signal perception in which a signaling Background Rice (Oryza sativa L.) is a cereal of high economic and mechanism initiates a cascade of gene expression social value, which is used as a staple food by more than responses to drought. These transcriptional changes can half of the world's population. It is the only cereal which result in successful adaptations leading to stress tolerance is solely produced for human consumption. The produc- by regulating gene expression and signal transduction in tion of rice must increase 20% in the next 15 years in order the stress response (regulatory proteins) or directly pro- to keep pace with population growth. One of the main tecting the plant against environmental stress (functional constraints that affect yield in rice production is water def- proteins) [11]. icit. The increasing worldwide water shortage and uneven rainfall distribution limit the use of irrigated agriculture, Several functional genomic studies of rice have been per- typical of rice production. Irrigation costs are increasingly formed using different approaches such as macro and high worldwide. There is, therefore, a need to develop rice microarray [12,13], RT-qPCR, SAGE (Serial Analysis of varieties, which are more efficient in the use of water [1,2]. Gene Expression), MPSS (Massive Parallel Signature Sequenc- A major challenge for the research community is the rela- ing) and more recently oligoarray using the transcriptome tively limited progress made so far in improving the of rice to evaluate responses to abiotic stresses [14]. Pro- drought tolerance of high yielding rice varieties [3]. teome analyses have also been increasingly employed to complement genomic studies [15-18], however in a lower Rice is a highly diverse species, which can be grown in rate. Although numerous genes and proteins, which many types of soil moisture regimes, ranging from aerobic potentially contribute to drought tolerance in rice, have upland to permanently flooded lowland. Although been reported [19-22], most of these studies have focused upland rice constitutes a relatively small proportion of the on lowland rice genotypes. Currently, very little is known total rice area worldwide, it is the predominant method of about gene and protein expression in upland rice [22-25]. rice cultivation in Latin America and West Africa (about Moreover, most ESTs from drought stressed plants availa- 75% and 50% of rice area, respectively) [4]. In Brazil, ble were obtained from libraries constructed using seed- upland rice responds for approximately 40% of the total lings [26]. There are very few reports on gene expression rice production. In some areas of the country, upland rice of drought-stressed plants in the reproductive stage and is a subsistence crop planted by farmers who apply limited using root tissue of plants growing under defined field inputs to their crops. The cultivation of upland rice in capacity. marginal areas with low soil fertility and threatened by severe abiotic stresses, such as periods of drought during The comprehension of drought responses in upland rice is the cropping season, has a significant impact on rice pro- important for designing breeding strategies to develop duction [5,6]. Due to exposure to many environmental varieties more tolerant to water constraints. Recently, the constraints, some local varieties of the tropical japonica tolerance of ten traditional upland varieties of rice sub- rice developed high adaptability to drought stress, hot and mitted to drought stress has been evaluated as part of an dry climatic conditions of regions in Latin America and effort to identify new sources of drought tolerance in rice Africa. Therefore, these varieties may show high levels of [27]. Concomitantly, the root system of two of the above water usage efficiency and constitute an excellent material mentioned upland rice genotypes, characterized as sus- for studying drought tolerance mechanisms in rice. In Bra- ceptible and tolerant to drought stress, have been ana- zil, for example, EMBRAPA maintains a germplasm bank lyzed at the reproductive stage using genomic and enriched with traditional upland rice landraces collected proteomic approaches. Several genes and proteins were in areas where cultivated rice has been grown since its identified, which may play important roles in drought tol- introduction in the country, centuries ago, and may repre- erance. sent an extraordinary source of genes that control traits of economic importance such as drought tolerance [7]. Methods 1. Plant material and phenotypic evaluation The determination of the mechanisms directly involved in Plants of traditional upland rice (O. sativa L. var. japonica) drought tolerance remains a challenging task since varieties were grown on PVC pipe columns (25 cm of drought is a complex trait that involves several metabolic diameter; 80 cm of height) filled with fertilized Oxisol pathways [3]. The identification and isolation of genes under screenhouse conditions [27]. The experimental associated with drought tolerance is of major importance design was a split-plot design with two watering regimes in order to better understand this trait and increase the as main plots, ten traditional upland varieties as subplots efficiency in developing drought tolerant varieties [8-10]. and three replications. The watering regimes were (a) con- At the molecular level, the response of roots to water lim- trol, consisting of a main plot of well-watered plants iting conditions seems to be crucial to trigger drought tol- throughout the experiment, which received 100% reposi- erance mechanisms, since roots are one of the primary tion of the water lost daily and a minimum soil humidity Page 2 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 of -0,025 MPa at 15 cm of depth, and (b) drought stress, and sequence homologues were identified using the Blast which consisted of 50% reposition of the water lost daily program [29]. An in silico subtraction was performed by from anthesis on. Water reposition was calculated based clustering all sequences from both cDNA libraries accord- on daily weighting of columns with a mechanical scale. ing to the methodology described by Telles and da Silva Twenty-one days after initiating the drought stress treat- [30], allowing the identification of genes exclusively ment (at anthesis), roots of each treatment (control and found in each library. drought stress) were collected from each rice variety. All root samples were immediately frozen in liquid nitrogen 3. Protein extraction and 2-DGE and maintained at -80°C until their use for RNA and pro- Total protein was extracted from roots of the drought tol- tein extractions. At harvest, grain yield and yield compo- erant (Prata Ligeiro) and susceptible (IRAT20) genotypes nents of each genotype were evaluated, including root and according to procedures described by de Mot and Vander- shoot dry weight, harvest index, spikelet sterility, grains leyden [31] Plant material of the three replications were per panicle and weight of 100 grains. Drought tolerance pooled, pulverized and mixed with extraction buffer (0.7 parameters were estimated based on calculations of M sucrose, 0.5 M TrisHCl, 30 mM HCl, 50 mM EDTA, 0.1 drought severity, drought tolerance index and drought M KCl and 40 mM DTT) and phenol (100%) in the same susceptibility index [28]. The genotypes submitted to the volume (750 μl). Proteins were precipitated with ammo- drought stress showed differences in most of the yield nium acetate 0.1 M in methanol, washed with acetone parameters analyzed, which were significantly influenced 80% (v/v), dried and stored at -20°C. Protein quantifica- by the drought severity applied to the experiment [27]. tion was performed using the Bradford Reagent (Invitro- These parameters were then used to classify the genotypes gen, USA). Isoelectric focusing was conducted using 11- according to their reaction to stress. Among them, two cm immobilized pH gradient (IPG) strips with a pH range contrasting genotypes for drought stressing conditions of 4–7 and a Multiphor II electrophoresis system (GE). were selected for the present study: Prata Ligeiro, as the Strips containing approximately 220 μg of protein were tolerant, and IRAT20, as the susceptible variety. The RNA rehydrated with 2% (v/v) CHAPS, 8 M urea, 7 mg dithio- and protein analyses proceeded only with root tissue threitol (DTT) and 2% IPG buffer. Second dimension extracted from these two varieties. analysis was performed in 10% gels by SDS-PAGE as described by Laemmli [32] and at least five replications of 2. RNA extraction and subtractive library construction each genotype were performed. Protein spots were visual- For each genotype, a bulk of approximately 250 mg of ized after silver [33] or Comassie blue staining. plant roots from the three replications were homogenized 4. Image analysis in liquid nitrogen and total RNA was extracted using the Concert™ Plant RNA Reagent (Invitrogen, USA), accord- The 2D gel images were evaluated using the Platinum soft- ing to manufacturer's instructions. This procedure was fol- ware (GE Healthcare, UK) and three high quality gels lowed for roots harvested from drought stressed as well as obtained for both genotypes were analyzed. First, a cali- unstressed plants. mRNA was then isolated from total bration with a grey scale was performed to transform grey RNA by using PolyATtract mRNA Isolation System levels into OD values for each pixel (px) of the gel image. (Promega, USA). Quantity and quality of the isolated The wizard detection method proposed by the software mRNA was evaluated by spectrophotometry and electro- was used to detect the spots with the following parame- phoresis in agarose gel 1%, respectively. ters: 15 px for estimated spot size, 50 px for minimum spot size and a spot contrast enhancement of 75%. Auto- Isolated mRNAs were used for cDNA synthesis and sup- matically detected spots were checked and some of them pression subtractive hybridization (SSH) library construc- were manually added or removed. Following the detec- tion by using the PCR Select Subtraction Kit (Clontech, tion procedure, the normalization step was carried out to USA). Subtractive hybridizations were performed using attribute a common spot identity for the same spots cDNA from stressed plant roots (as tester) against cDNA derived from different images utilizing the reference gel from well-watered unstressed plant roots (as driver) of construct and automatically matching options. A syn- each genotype, in order to identify genes involved in thetic gel from each genotype was constructed by using drought response. The subtractive PCR products obtained the mean value of volume percentage of each protein spot were cloned into pGEM T-Easy (Promega, USA) and present in the three replicates, according to the Platinum sequenced in ABI Prism 3700 DNA Analyser (Applied Bio- software's (GE Healthcare, UK) instructions. The two systems Inc., USA). A minimum insert size of 30 bp and at obtained synthetic gels were then overlapped using the least 20 bp with quality of phred > 20 were considered for molecular marker as well as several protein spots present the analysis. Sequences were deposited in GenBank under in both profiles as landmarks. The overlapped images the accession numbers of FG124418 through FG124880 were based on landmark spots showing same pI and Mw. Page 3 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 5. Trypsin digestion and mass spectrometry analysis Results and discussion Protein spots were excised manually from 2D gels and in- 1. Experimental design and sampling Plants were submitted to drought stress after anthesis for gel digested with sequencing grade trypsin (Promega, Madison, WI) according to Schevchenko et al. [34]. twenty-one days. Flowering is the period in which the Briefly, each protein spot was placed in a 0.5 mL polypro- plant is most sensitive to water deficit and several toler- pylene (Eppendorf) tube and destained by washing 5–8 ance mechanisms need to be activated at this stage in times with 200 μL of 50% (v/v) acetonitrile/10 mM order to guarantee grain filling and production [6]. Dur- ammonium bicarbonate solution. The gel pieces were ing root sampling, a clear visual difference in Prata Ligeiro subsequently dehydrated by washing with 200 μL of and IRAT20 plants could be observed. An intense leaf roll- 100% acetonitrile and completely dried in a Speedvac ing was noticed in the susceptible genotype as opposed to concentrator. Ten microliters of 50 mM ammonium bicar- the tolerant. In addition, a more pronounced aerial bio- bonate/10% (v/v) acetonitrile solution containing 100 ng mass loss could be visualized in IRAT20. At harvest, yield of trypsin were added, and the sample incubated at 37°C and yield component parameters were measured [27]. The for 16 h. Aliquots of each tryptic digest (1 μL) were mixed variety IRAT20, a high yielding variety under irrigated with a saturated solution of α-cyano-4-hydroxycinnamic controlled conditions, showed a 51% reduction in grain acid, spotted onto a MALDI target plate, and allowed to yield when submitted to drought stress. On the other air dry. hand, Prata Ligeiro, a low yielding variety under well watering conditions, had a 23% reduction in grain yield Mass spectra were acquired using a MALDI-TOF/TOF under drought stress. The drought susceptibility index Autoflex II spectrometer (Bruker Daltonics, Bremen, Ger- based on yield was estimated as 0.73 for Prata Ligeiro (tol- many) operating at a laser frequency of 50 Hz. MS analysis erant) and 1.57 for IRAT20 (susceptible). were performed in a positive ion reflection mode. Voltage parameters were set as IS1 19 kV, IS2 16.8 kV, Lens 8 kV, Collected roots of both genotypes were then used for Reflector 20 kV, Reflector2 9.54 kV. The delay time was 70 cDNA library construction and proteome studies. In the ns and acquisition mass range 700–3200 Da. External cal- cDNA library study, stressed plants were contrasted with ibration was performed using a peptide mix contaning well-watered plants, whereas in the proteome analysis, ACTH (1–24), ACTH (18–39), Somatostatin, Angiotensin stressed plants from both genotypes were compared. I and Angiotensin II, all from Sigma. MS/MS analysis were performed in a positive ion LIFT reflection mode. Voltage Water reposition, based on the evapotranspiration rate, parameters used were IS1 6 kV, IS2 5.3 kV, lens 3.15 kV, has been used to determine an impartial and consistent Reflector 23.5 kV, Reflector2 9.7 kV, LIFT1 19 kV and response of plants to drought stress, during long periods LIFT2 4 kV. The delay time was set as zero and acquisition of drought in the soil [35]. Several studies have tried to mass range 40–2400 Da. define the critical limit of water in the soil after which crop development and production are significantly affected Peak lists were generated using the FlexAnalysis 3.0 soft- [36]. According to Rosenthal et al. [37], the symptoms of ware (Bruker Daltonics). The sophisticated numerical water deficit occur when water availability is around 50% annotation procedure (SNAP) algorithm was used to of the field capacity. detect the monoisotopic peak values, with a quality factor threshold of 30 and 6 as S/N threshold. Database searches The response of plants to drought stress is also dependent were performed in February 2008 using the MASCOT on the extension and rate of water loss [38]. Fukai et al. search engine (Matrix Science, UK) with the NCBInr pro- [39] reported that when a rapid water deficit occurs, the tein database and Oryza sativa taxonomy. The mass toler- morpho-physiological mechanisms are severely affected. ance was 100 ppm and one missed cleavage was allowed. When the deficit is prolonged for a few days, plants are Carbamidomethylation of cysteines, oxidation of methio- allowed to adapt to the stress, enabling the identification nine, and acrylamide-modified cysteines were considered of variability in drought tolerance within different geno- for PMF searches. For accepting the identification, the cut- types, since plants can respond differently to the same off value for the Probability Based Mowse score calculated stress condition [38]. Therefore, the sampling time used by MASCOT (at p < 0.05) was used. For MS/MS data, the in this study (21 days of drought stress) may have allowed peptide mass tolerance was 0.5 Da, MS/MS ion mass tol- the analysis of adaptive responses of the plant to tolerate erance at 0.5 Da, allowance of 1 missed cleavage, and water deficit. charge state +1. When the pI and MW of matched proteins were not available, these values were calculated using Several studies reported the response of rice seedlings to ExPASy Compute pI/Mw tool http://ca.expasy.org/tools/ drought stress [13,26,40] however, little attention has pi_tool.html. been given to the expression of genes in water-stressed Page 4 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 plants at the reproductive stage (flowering, grain filling) previously reported as associated to drought stress and in which a higher yield impact is observed [6]. some of them are discussed below. 2. cDNA library analysis Genes involved in signaling routes were exclusively iden- Roots are one of the primary sites responsive to restrictive tified in Prata Ligeiro and include serine/threonine kinase, conditions of water availability and, as a result, synthesize ethylene-responsive factor and calcium-transporting chemical signals for a rapid response of the plant to ATPase/calmodulin binding sequences. Serine/threonine 2+ drought stress [41]. This occurs since the response in kinases are Ca dependent proteins kinase (CDPKs), leaves must be stimulated rapidly to avoid irreversible involved in the phosphorylation cascade of proteins. Sev- damage to the photosynthetic machinery. In this work, eral studies have shown that CDPKs are induced or acti- two subtractive cDNA libraries were constructed using vated by abiotic stresses, suggesting that they may be mRNA from roots of tolerant and susceptible upland rice involved in drought signaling [42-45]. Another identified genotypes subtracted from their respective unstressed gene associated to signal transduction was an ethylene- well-watered controls. The subtracted PCR products responsive factor. Ethylene is a well characterized phyto- obtained after primary and secondary PCR ranged from hormone that may act alone or in combination with ABA 0,1 – 1,5 kb. in regulating gene expression under abiotic stress [46]. Calcium-transporting ATPase/calmodulin binding are The SSH libraries of the tolerant (Prata Ligeiro) and sus- also stress-signaling proteins and are responsible for regu- ceptible (IRAT20) genotypes were concluded with a nov- lation of the osmotic potential of the cell. elty index of 66% and 55%, respectively. The general analysis of the two libraries revealed a total of 463 valid Some genes that participate in metabolism alterations as sequences (230 from Prata Ligeiro and 233 from IRAT20) a result of the limitation caused by low levels of intracel- and the average fragment size was of 300 bp. Several genes lular CO observed during drought stress were also identi- commonly expressed in both genotypes were identified fied only in Prata Ligeiro. Among these genes are those and are probably not directly involved in drought toler- coding for Phosphoenolpyruvate carboxykinase, an ance. enzyme that has a key role in nocturnal fixation of CO ; malato dehydrogenase, which is an enzyme particularly In order to determine the genes exclusively expressed in important for the assimilation of carbon in C4 plants; the tolerant and susceptible genotypes, an in silico subtrac- Glutamate-1-semialdehyde aminotransferase and glu- tion was performed using sequences of both libraries. The cose-1-fosfato adenililtransferase [47-49], both involved results for the in silico subtraction revealed that the 463 in carbohydrate metabolism. sequences represented 282 different transcripts: 127 were found in both genotypes, 84 were exclusively expressed in It has been proposed that the mechanism involved in the Prata Ligeiro library (Table 1) and 71 were observed drought tolerance in upland rice is a result of a higher only in the IRAT20 library (Table 2). expression of genes involved in oxidative stress protection [23]. Indeed, in the present study some genes associated 2.1. Putative drought-tolerance genes identified in Prata Ligeiro to the protection of the cell were expressed only in the tol- Drought tolerance is a complex trait and involves mecha- erant genotype. Among them, we found a Methionine sul- nisms that act in isolation or combined to avoid or toler- foxide reductase A and a Respiratory burst oxidase ate periods of water deficit. It is expected that genotypes homolog, which act in the recognition of reactive oxygen responding differently to drought stress show differences species (ROS) in biotic and abiotic stresses [50]. Other in gene expression, and that a portion of the differences is interesting genes identified are Metallothionein, a super- related to drought tolerance. Therefore, the analysis of the family of low molecular weight proteins involved in metal genes found exclusively in the tolerant genotype is of detoxification [51] and scavenging of oxygen-free radicals, interest to identify genes associated with water usage effi- which can decrease injury in oxidative tissue, and Ferre- ciency. doxin, regulated by different environmental stresses including biotic and abiotic conditions. Among the 84 transcripts uniquely reported in the toler- ant genotype, 14 did not present known homologs (no Genes associated to maintenance of cell turgor were also hits) and 17 showed similarities to proteins with identified such as IQ calmodulin-binding and Calcium- unknown function (hypothetical proteins). Three transporting ATPase/calmodulin binding. These genes sequences showed similarity to non-plant proteins and were previously reported to participate in typical defense probably represent contaminating sequences (Table 1). mechanisms in upland varieties [23]. The other transcripts showed similarity to several proteins Page 5 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 Table 1: Genes detected exclusively in roots of the tolerant genotype (Prata Ligeiro) SSH library Encoded protein Homologous organism Accession number Proteins of known function Glutamate-1-semialdehyde 2,1 aminomutase Oryza sativa NM_001068872 Metallothionein-like protein Oryza sativa NM_001056317 Malate dehydrogenase Oryza sativa NM_001062924 Methionine sulfoxide reductase A Oryza sativa NM_001063272.1 Phosphatidylinosytol 3 and 4 kinase Oryza sativa NM_001060732 Ubiquitin-conjugating enzyme Oryza sativa NM_001048429 Nuclear protein SET domain containing protein Oryza sativa NM_001067672 Splicing factor 3B subunit 5-like protein Oryza sativa dbj|BAD10044.1| PEP carboxikinase Oryza sativa gb|ABF95034.1 Putative malate dehydrogenase Oryza sativa gb|AAT69584.1| Eukaryotic translation initiation factor 5A-2 (eIF-5A) (eIF-4D) Oryza sativa NC_008405 Metallothionein-like protein type 1 Oryza sativa NP_001068544.1 ADP glucose pyrophosphorylase Oryza sativa EF122437 CBL-interacting protein kinase 1 Oryza sativa NM_001049327 ADP-ribosylation factor Oryza sativa NM_001051134 DSS1/SEM1 family protein Oryza sativa NC_008394 Ankyrin repeat containing protein Oryza sativa NM_001054582 Pathogenesis-related transcriptional factor and ERF domain containing protein Oryza sativa NC_008402 E-class P450, group I family protein Oryza sativa NM_001074239 FAR1 domain containing protein Oryza sativa NM_001057341 Tubulin alpha-1 chain Oryza sativa NM_001074145 Putative ubiquitin conjugating enzyme Oryza sativa dbj|BAB89662.1| DEAD/DEAH box helicase domain containing protein Oryza sativa NM_001069156 Putative pollen specific protein C13 precursor Oryza sativa gb|AAM08621.1| IQ calmodulin-binding Oryza sativa NM_001061046 HAD superfamily hydrolase 5' nucleotidase protein Oryza sativa NM_001057956 SAM biding motif domain containing protein Oryza sativa NM_001070787 Peptidase aspartic family protein Oryza sativa NM_001063168 Nonaspanin (TM9SF) family protein Oryza sativa NM_001056027 Ethylene responsive element binding factor 5 Oryza sativa NM_001063579 TMS membrane protein Oryza sativa NM_001054899 Heat shock protein DnaJ family protein Oryza sativa NM_001060020 Ferredoxin III, chloroplast precursor (Fd III) Oryza sativa NC_008396 Anther ethylene-upregulated protein ER1 (Fragment) Oryza sativa NM_001055765 Chaperone protein DNA-J-related like Oryza sativa dbj|BAD27799.1| Isoflavone reductase family protein Oryza sativa NM_001068997 U box domain containing protein Oryza sativa NM_001071339 Ribossomal protein L Curculio glandium AM049038 Short chain dehydrogenase tic32 Oryza sativa NM_001048577 Arabinogalactan protein Oryza sativa NC_008394 Ribonuclease T2 family protein Oryza sativa NM_001070328 HvB12D protein (B12Dg1 protein) Oryza sativa NM_001063815 Respiratory burst oxidase homolog Oryza sativa NM_001049555 Phosphatidylinositol-4-phosphate 5-kinase family protein Oryza sativa NM_001068386 Nodulin-like Oryza sativa NM_001070322 Cathepsin B-like cysteine protease form 2 Ixodes ricinus gb|ABO26563.1| Cathepsin L-like cysteine proteinase precursor Acanthoscelides obtectus gb|AAQ22984.1| Calcium-transporting ATPase/calmodulin binding Arabidopsis thaliana NP_188931.1 Myb, DNA biding domain containing protein Oryza sativa NM_001062445 TGA-type basic leucine zipper protein Phaseolus vulgaris gb|AF402607.1| Tocopherol O-methyltransferase, choroplast precursor Oryza sativa NM_001054379 ATP-dependent Clp protease ATPbiding subunit Clpx-like mitochondrial precursor Oryza sativa dbj|BAD15818.1| HvB12D protein (B12Dg1 protein) Oryza sativa NM_001063815 Uncharacterized protein family containing protein Oryza sativa gb|ABA91393.1| Protein of unknown function Protein of unknown function Oryza sativa NC_008397 Protein of unknown function Oryza sativa NC_008403 Page 6 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 Table 1: Genes detected exclusively in roots of the tolerant genotype (Prata Ligeiro) SSH library (Continued) Unknow function Oryza sativa NM_001067277 Hypothetical protein Oryza sativa AP008208 Hypothetical protein Oryza sativa gb|EAY93896.1| Conserved hypothetical protein Oryza sativa NM_001065538 Hypothetical protein Oryza sativa gb|EAY84091.1| Hypothetical protein Oryza sativa CT836006 Hypothetical protein Oryza sativa NC_008394.1 Hypothetical protein Oryza sativa NC_008394.1 Hypothetical protein Oryza sativa AP008208 Hypothetical protein Oryza sativa NM_001057688 Hypothetical protein Oryza sativa NM_001066910 Hypothetical protein Oryza sativa NM_001053573 Hypothetical protein Oryza sativa CT829595 Hypothetical protein Oryza sativa CT834076 In this study we have also identified genes which have not 3. Proteome analysis yet been directly related to drought tolerance, such as In order to complement the genomic studies, protein B12Dg1 protein, Nuclear protein SET domain containing maps of roots from water-stressed plants of the suscepti- protein and Putative pollen specific protein C13 precur- ble (Figure 1A) and tolerant (Figure 1B) genotypes were sor, as well as genes with unknown function. Further stud- compared. Triplicates of the gels from each genotype were ies need to be performed in order to assign biological compared and revealed a total of 463 proteins in the Prata function, since these genes may play important roles in Ligeiro profile and 522 in IRAT20. The two obtained syn- plant adaptation during drought stress conditions. thetic gels were overlapped and this procedure allowed the identification of 307 overlapped spots, 156 proteins 2.2. Drought-responsive genes identified in IRAT20 exclusive to the tolerant genotype and 215 proteins exclu- Regarding the response of the susceptible genotype to sive to the susceptible genotype. These results show a drought stress, 71 transcripts were exclusively expressed in higher diversity in the protein pattern of the susceptible this genotype. As in Prata Ligeiro, a high number of genes genotype. (14) with no known homologs (no hits) were identified (Table 2). Moreover, a total of 23 genes encoding hypo- A total of 50 intense proteins observed in the tolerant gen- thetical or unknown proteins were also observed. Further otype profile after Coomassie blue staining was excised expression studies of these genes may reveal important from the gel, digested and analyzed by mass spectrometry. genes associated to drought stress response, which have By using the Mascot program, 22 proteins could be iden- not been explored so far. This information may contribute tified with a significant score (Table 3), including 16 up- to a better understanding of the mechanisms related to and 4 down-regulated, 1 new and 1 equally expressed in drought susceptibility in upland rice varieties. both genotypes (Figure 2). The other proteins were in insufficient amounts for the identification analysis or did As in Prata Ligeiro, three transcripts showed similarity to not return reliable matches when using the Mascot pro- non-plant proteins and were not considered in the analy- gram. This probably occurs due to a low protein quantity sis since they probably represent contaminating and/or low ionization capacity of molecular components sequences (Table 2). The other transcripts showed similar- present in the samples analyzed. It is also possible that, ity to genes associated to different functions including the considering the high amount of "no hits" obtained in the transport of small molecules or inorganic ions, such as genomic analysis, protein sequences matching the pep- HCO -transporter and Vacuolar H+ pyrophosphatase. tides searched were not available in public databases. The The expression of these genes was previously reported by peptide sequences obtained were also analyzed using the Wang et al. [23] in a lowland variety. These results suggest Blastp program. that upland genotypes susceptible to drought may present similar responses to those of lowland varieties, which are Spots PL1 and PL2 (up-regulated in Prata Ligeiro) were naturally more susceptible to water deficit. identified as hypothetical proteins which contain Ricin B- related lectin domain. Other up-regulated hypothetical Interestingly, the well-known transcription factor WRKY proteins were also identified and include protein spots was uniquely identified in IRAT20. WRKY mediates plant PL34, PL45 and PL51. Spot PL45 and PL51 were expressed stress responses [52-54] and the increased expression of 2.6 and 4.5 fold, respectively, in the tolerant genotype this protein has been frequently associated to drought (Figure 2), indicating that these proteins may play an stress response in rice [23,55]. important role in drought tolerance. Spot PL57 was Page 7 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 Table 2: Genes detected exclusively in roots of the susceptible genotype (IRAT20) SSH library Encoded protein Homologous organism Accession number Proteins of known function T complex 11 family protein Oryza sativa NM_001059402 Protein kinase domain containing protein Oryza sativa NM_001071926 Protein disulphide isomerase family protein Oryza sativa AP008208 TPR-like domain containing protein Oryza sativa NM_001058028 Protein kinase Oryza sativa NM_001074788 Pinoresinol-lariciresinol reductase TH1 Oryza sativa NM_001073059 Smr protein; MutS2 c- terminal domain containing protein Oryza sativa NM_001048992 SIPL protein (Membrane-type 1 matrix metalloproteinase cytoplasmic tail binding protein-1) Oryza sativa NM_001055581 Similar to CG 9092- PA Tribolium castanum XP_967647.1 Putative ATP-dependent Clp protease ATP-binding subunit ClpX1 (CLPX) Oryza sativa dbj|BAD15818.1| Cytocrome P450 family protein Oryza sativa NM_001071591 Preprotein translocase subunit sec Y, chloroplast precursor Oryza sativa NM_001067916 Vacuolar H+ pyrophosphatase Oryza sativa NM_001063501 Similar to UPF 0139 protein CGI-140 Tribolium castaneum XP_971064.1| 60 kDa inner membrane insertion protein family protein Oryza sativa NM_001055291 Glyceraldehyde-3-phosphate dehydrogenase (Fragment) Oryza sativa NM_001055382 Similar to splicing coativator subunit SRm 300 Monodelphis domestica XP_001371550.1| Cysteine synthase, mitocondrial precursor Oryza sativa NM_001052112 TPR-like domain containing protein Oryza sativa NM_001056953 HCO3-transporter Oryza sativa NM_001073581 Banched chain amino-acid aminotransferase-like protein 3 Oryza sativa NM_001049072 Beta tubulin (fragment) Oryza sativa NM_001049296 HAT dimerisation domain containing protein Oryza sativa NC_008402 Urease accessory protein G Oryza sativa NM_001062872 Glycoside hydrolase, family 47 protein Oryza sativa NM_001054615 WRKY transcription factor 82 Oryza sativa DQ298186 Tubby family protein Oryza sativa NM_001062568 Ribosomal protein L41 family protein Oryza sativa NC_008400 Granule-bound starch synthase I, chloroplast precursor Oryza sativa NM_001065985 Putative RNA polymerase I transcription factor RRN3 Oryza sativa dbj|BAD45608.1| Aconitate hydratase, cytoplasmic (Citrate hydro-lyase) (Aconitase) Oryza sativa NM_001055433 Short chain alcohol dehydrogenase-like Oryza sativa NM_001056212 Putative ubiquitin-conjugating enzyme E2 Oryza sativa dbj|BAD25096.1| Peptidase s26A signal peptidase I family protein Oryza sativa NM_001074823 Protein of unknown function Unknown protein Oryza sativa NM_001068742 Hypothetical protein Oryza sativa AC119292 Hypothetical protein Oryza sativa AP008208 Hypothetical protein Oryza sativa AK243578 Hypothetical protein Oryza sativa NC_008395.1 Hypothetical protein Oryza sativa AP008208 Hypothetical protein Oryza sativa NM_001057104 Hypothetical protein Oryza sativa NC_008395 Hypothetical protein Oryza sativa NM_001074804 Hypothetical protein Oryza sativa NM_001057688 Hypothetical protein Oryza sativa NC_008401.1 Hypothetical protein Oryza sativa NC_008395.1 Hypothetical protein Oryza sativa CR855113 Hypothetical protein Oryza sativa AC145477 Hypothetical protein Oryza sativa AC092556 Hypothetical protein Oryza sativa AK242616 Hypothetical protein Oryza sativa AP008209 Hypothetical protein Oryza sativa NC_008398.1 Hypothetical protein Oryza sativa AC099401 Hypothetical protein Oryza sativa NM_001050487 Hypothetical protein Oryza sativa CT831698 Hypothetical protein Oryza sativa CT828847 Hypothetical protein Oryza sativa CT832865 Page 8 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 another protein identified as hypothetical and was exclu- (Table 3), involved in carbohydrate metabolism. Accord- sively expressed in Prata Ligeiro. These proteins are inter- ing to Wang et al. [23], genes related to metabolism are esting candidates for futures studies aiming at the more expressed in lowland than in upland genotypes. It is determination of biological function. possible that susceptibility to drought in upland rice may occur in a similar way as in lowland rice. Spots PL3 and PL60 were identified as the same protein chitinase and spot PL11 as a Chain A, Crystal Structure of Spots PL43 and PL46 were both identified as enolase, a Class I Chitinase. Chitinases are pathogenesis-related pro- glycolytic enzyme, which participates in metabolic proc- teins expressed in response to biotic and abiotic stresses esses. The up-regulation of enolase has been previously and have been studied in grasses such as rye in response reported in rice roots in response to salt stress [61] and to to cold and drought stress [56]. Spot PL60 was highly PEG treatment [24]. Unexpectedly, PL46 was equally induced in the tolerant genotype, which confirms the up- expressed in Prata Ligeiro and IRAT20, while spot PL43 regulation of this protein during drought stress. Chiti- was up-regulated in Prata Ligeiro. The existence of multi- nases have also been reported as being induced in tomato ple enolase isoforms in plants has been reported [62] and plants tolerant to drought when compared to the suscep- it is possible that the enolases identified in this study rep- tible genotype [57]. resent different isoforms, which respond differently to drought stress conditions. Indeed, difference in the Two other pathogenesis-related proteins were identified: expression of enolase isoforms was observed in maize in one was up-regulated (spot PL33) and the other repressed response to anaerobiosis [63]. (PL30) in the tolerant genotype (Figure 2). The expression of these proteins has been previously reported in roots of A highly induced protein (15 fold) in the tolerant geno- rice in drought stress conditions and although the role of type (PL40) showed identity to a hypothetical protein as proteins of this family is not well established, they have well as a salt stress induced protein (Table 3). Similarly, been associated to hypersensitive reaction in response to spot 27 (2.6 fold higher in Prata Ligeiro) also presented biotic and abiotic factors [58]. In drought stress condi- identity to the salt stress induced protein. It is possible tions, pathogenesis-related proteins as well as the salt that these spots represent new rice proteins, not identified stress-responsive SalT protein have been reported in rice so far that contain a conserved region present in both roots [59]. matching proteins. The induction of proteins involved in tolerance to salt stress, during water deficit conditions, As observed in the constructed cDNA libraries, several shows that osmotic stress is an important aspect during proteins involved in oxidative stress protection were drought. Similar mechanisms are activated in response to induced in the tolerant genotype and were identified as a different abiotic stresses, as previously reported [10]. superoxide dismutase [Cu-Zn] (PL20), L- ascorbate perox- idase 1 (PL23), ascorbate peroxidase (PL38) and cytosolic Conclusion malate dehydrogenase (PL63) (Table 3). Peroxidases are Several genes and proteins involved in drought-response anti-oxidative enzymes, described in varieties of rice toler- as well as genes with no described homologs were identi- ant to high salinity conditions [25,60] and in upland rice fied in this work. Genes exclusively expressed in the toler- roots in response to osmotic stress [24]. These proteins are ant genotype were, in general, related to maintenance of involved in cellular detoxification and it is possible that turgor and cell integrity. In contrast, in the susceptible this is a general defense mechanism in response to water genotype, expression of genes involved in protection deficit in upland rice. According to Wang et al. [23,24] tol- against cell damage was not detected, indicating that there erance to drought stress observed in upland varieties may be a higher degradation of cellular components in includes detoxification mechanisms, limiting the accumu- these genotypes. Similar results were obtained by Wang et lation of reactive oxygen species. These authors reported al. [23] when comparing tolerant upland and lowland that these proteins were up-regulated in upland cultivars varieties. These results indicate that the mechanisms of when comparing tolerant lowland and upland rice. Unex- susceptibility in upland rice are similar to those of low- pectedly, proteins identified as superoxide dismutase land varieties, considering that the upland rice is naturally (PL7) and GSH-dependent dehydroascorbate reductase more tolerant to drought stress. (PL13) were down-regulated in the tolerant genotype. These proteins were not identified in the genomic analy- The proteomic analyses were complementary to the sis, highlighting the importance of proteomics studies to genomic data obtained. The expression of genes associ- complement the results obtained. ated with cell protection against oxidative damage is con- sidered important to cope with water deficit in upland Another down-regulated protein (PL24) identified in the rice. In this study, genes and proteins related to this func- Prata Ligeiro genotype was triosephosphate isomerase tion showed a higher expression in the tolerant genotype. Page 9 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 Root protein pr Figure 1 ofiles by 2-DGE of the susceptible (A) and tolerant (B) genotypes Root protein profiles by 2-DGE of the susceptible (A) and tolerant (B) genotypes. Total soluble protein (ca. 220 μg) was separated by 2-DGE and the spots were visualized after silver staining. Numbers indicate the protein spots successfully identified by mass spectrometry. Benchmark Protein Ladder (Invitrogen, USA) was used to estimate the molecular mass of the proteins visualized. Page 10 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 H ceptible (IRA Figure 2 istogram reT20) g presenting enotypes, as det expression levels of ermined by up- and the Platinu downm -regulated protei software (GE Healthcare, UK) ns identified in the tolerant (Prata Ligeiro) and sus- Histogram representing expression levels of up- and down-regulated proteins identified in the tolerant (Prata Ligeiro) and susceptible (IRAT20) genotypes, as determined by the Platinum software (GE Healthcare, UK). Interestingly, in the proteomics analysis, the susceptible Overall, due to the low amount of information regarding genotype showed a higher diversity in the protein profile, upland rice gene and protein expression in response to revealing more uniquely expressed proteins than the tol- water deficit, this study sheds some light over the compre- erant genotype. On the other hand, in the genomic study, hension of this complex mechanism. However, the high the number of exclusively expressed transcripts in the sus- amount of transcripts and proteins with unknown func- ceptible genotype was lower. It is well known that tran- tion obtained is still intriguing. These genes and proteins script levels do not always reflect protein amounts need to be further investigated in order to assign their bio- [64,65]. Therefore, it is possible that the transcripts related logical function and advance our knowledge regarding to the proteins exclusively present in IRAT20 2D maps drought tolerance in upland rice. were in low amounts, and not detected by the genomic analysis, or they were subtracted from the control condi- Authors' contributions tion in the hybridization process. Differences in transla- AM, ARR, CMG, MEF and PHNR designed and performed tion efficiency may have occurred, resulting in a higher the research. FRS analyzed the sequence data and EMS and amount of the corresponding proteins, further detected by DS analyzed the mass spectrometry data. ARR and AM 2-DGE. These results clearly show that proteomics studies drafted the manuscript. ACMB and CRS critically revised can reveal important additional information and that the the article. All authors approved the final version. use of complementary approaches is useful for a better understanding of complex biological traits, such as Acknowledgements This research was supported by Embrapa, CNPq and Embrapa Recursos drought tolerance. Genéticos e Biotecnologia. Page 11 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 Table 3: Proteins identified by peptide mass fingerprinting or de novo sequencing Spot n° Peptide sequence Protein identification Accession # Score Mr (gel) pI (gel) Mr (cal) pI (cal) PL 1 Hypothetical protein gi|115452789 138 38.0 6,7 39 6.3 PL2 Hypothetical protein gi|115452789 65 39.0 6.6 39 6.3 PL 3 WAPSPADAAAGR Chitinase gi|407472 56 36.0 6.6 35.5 7.3 PL 7 EHGAPQDENR Zinc-superoxide dismutase gi|22296339 26 15.0 6.4 14.7 5.9 PL 11 GPIQLSFNFNYGPAGR Chain A, Crystal Structure Of pdb|2DKV|A 30 37.0 6.2 32.6 5.8 Class I Chitinase PL 13 AAVGHPDTLGDCPFSQR GSH-dependent gi|6939839 43 26.0 6.1 23.5 5.6 dehydroascorbate reductase 1 PL 20 GTSQVEGVVTLTQDDQGP Putative superoxide dismutase gi|42408425 72 17.0 5.5 20.5 5.7 TTVNVR [Cu-Zn] PL 23 L-ascorbate peroxidase 1, P93404 99 28.0 5.5 27 5.4 PL 24 VATPDQAQEVHDGLR Triosephosphate isomerase gi|553107 49 28.0 5.4 27.5 6.6 PL 27 EFSIPLQDSGHVVGFFGR Salt stress-induced protein gi|158513205 88 11.0 5.0 15.1 5.1 PL 30 MIEDYLVAHPAEYA Pathogenesis-related protein Bet gi|9230755 55 18.0 4.9 16.6 4.9 v I PL 33 ADVGVGPVSWDDTVAAY Acidic PR-1 type pathogenesis- gi|12005673 182 17.5 4.2 17.5 4.5 AESYAAQR related protein PR-1 PL34 WWDTFPANVDGAR Hypothetical protein gi|115461070 87 29.0 4.7 27.2 5.0 PL 38 Ascorbate peroxidase NP_001060741 74 32.0 5.2 27 5.2 PL 43 MTAEIGEQVQIVGDDLLVT Enolase gi|780372 88 60.0 5.4 47.9 5.4 NPTR PL 46 Enolase Q42971 74 50.0 5.4 47.9 5.4 PL 45 Hypothetical protein gi|115465323 98 60.0 5.2 58.8 5.9 PL 51 KADATVAGDDR Hypothetical protein gi|125557770 37 45.0 5.7 95.7 8.0 PL 57 AGYAPPHWVQPGQGDR Hypothetical protein gi|125532459| 73 25.0 4.2 24.5 4.6 PL 60 ELFEQLLLHR Chitinase gi|561873 51 36.5 4.2 34.3 4.4 PL 63 ELVADDEWLNTEFISTVQQ Cytosolic malate dehydrogenase gi|115482534 66 37.5 5.9 35.5 5.75 PL 40 EFSIPLQDSGHVVGFFGR Salt stress-induced protein gi|158513205 104 39 5.3 15.1 5.1 Hypothetical protein EAY73933 40.6 8.6 insights into the genomics of rice root adaptive develop- ment. In Rice Genetics V Edited by: Brar DS, Mackill DJ, Hardy B. Sin- References gapore: World Scientific; 2007:117-141. 1. Luo LJ, Zhang QF: The status and strategy on drought resist- 12. Kawasaki S, Borchert C, Deyholos M, Wang H, Brazille S, Kawai K, ance of rice (Oryza sativa L.). Chin J Rice Sci 2001, 15:209-214. Galbraith D, Bohnert HJ: Gene expression profiles during the 2. Liu H, Mei H, Yu X, Zou G, Liu G, Luo L: Towards improving the initial phase of salt stress in rice. The Plant Cell 2001, drought tolerance of rice in China. Plant Genet Resour 2007, 13(4):889-905. 4(01):47-53. 13. Rabbani MA, Maruyama K, Abe H, Khan MA, Katsura K, Ito Y, Yoshi- 3. Price AH, Cairns JE, Horton P, Jones HG, Griffiths H: Linking wara K, Seki M, Shinozaki K, Yamaguchi-Shinozaki K: Monitoring drought-resistance mechanisms to drought avoidance in expression profiles of rice genes under cold, drought, and upland rice using a QTL approach: progress and new oppor- high-salinity stresses and abscisic acid application using tunities to integrate stomatal and mesophyll responses. J Exp cDNA microarray and RNA gel-blot analyses. Plant Physiol Bot 2002, 53(371):989-1004. 2003, 133(4):1755-1767. 4. Gupta PC, O'Toole JC: Upland rice: a global perspective. Los 14. Yamaguchi-Shinozaki K, Shinozaki K: Transcriptional regulatory Baños, Laguna: IRRI; 1986. networks in cellular responses and tolerance to dehydration 5. Pinheiro B da S, de Castro E da M, Guimarães CM: Sustainability and cold stresses. Annu Rev Plant Biol 2006, 57:781-803. and profitability of aerobic rice production in Brazil. Field 15. Salekdeh GH, Siopongco J, Wade LJ, Ghareyazie B, Bennett J: Pro- Crops Res 2006, 97(1):34-42. teomic analysis of rice leaves during drought stress and 6. Heinemann A, Dingkuhn M, Luquet D, Combres J, Chapman S: Char- recovery. Proteomics 2002, 2(9):1131-1145. acterization of drought stress environments for upland rice 16. Salekdeh GH, Siopongco J, Wade LJ, Ghareyazie B, Bennett J: A pro- and maize in central Brazil. Euphytica 2007, 162(3):395-410. teomic approach to analyzing drought- and salt-responsive- 7. Pessoa-Filho M, Beló A, Alcochete AAN, Rangel PHN, Ferreira ME: A ness in rice. Field Crops Res 2002, 76(2–3):199-219. set of multiplex panels of microsatellite markers for rapid 17. Agrawal GK, Jwa NS, Iwahashi Y, Yonekura M, Iwahashi H, Rakwal R: molecular characterization of rice accessions. BMC Plant Biol Rejuvenating rice proteomics: Facts, challenges, and visions. 2007, 7:23-23. Proteomics 2006, 6(20):5549-5576. 8. Tuberosa R, Salvi S: Genomics-based approaches to improve 18. Agrawal GK, Rakwal R: Rice proteomics: A cornerstone for drought tolerance of crops. Trends Plant Sci 2006, 11(8):405-412. cereal food crop proteomes. Mass Spectrom Rev 2006, 9. Lafitte HR, Yongsheng G, Yan S, Li ZK: Whole plant responses, 25(1):1-53. key processes, and adaptation to drought stress: the case of 19. Yang L, Zheng B, Mao C, Qi X, Liu F, Wu P: Analysis of transcripts rice. J Exp Bot 2007, 58(2):169-175. that are differentially expressed in three sectors of the rice 10. Sreenivasulu N, Sopory SK, Kavi Kishor PB: Deciphering the regu- root system under water deficit. Mol Genet Genomics 2004, latory mechanisms of abiotic stress tolerance in plants by 272(4):433-442. genomic approaches. Gene 2007, 388(1–2):1-13. 11. Périn C, Rebouillat J, Brasileiro ACM, Diévart A, Gantet P, Breitler JC, Johnson AAT, Courtois B, Ahmadi N, de Raissac M, et al.: Novel Page 12 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 20. Fu BY, Xiong JH, Zhu LH, Zhao XQ, Xu HX, Gao YM, Li YS, Xu JL, located in the endoplasmic reticulum of Arabidopsis. Proc Li ZK: Identification of functional candidate genes for drought Natl Acad Sci USA 2000, 97(11):6224-6229. tolerance in rice. Mol Genet Genomics 2007, 278(6):599-609. 43. Pei ZM, Ward JM, Harper JF, Schroeder JI: A novel chloride chan- 21. Gorantla M, Babu PR, Lachagari VBR, Reddy AMM, Wusirika R, Ben- nel in Vicia faba guard cell vacuoles activated by the serine/ netzen JL, Reddy AR: Identification of stress-responsive genes threonine kinase, CDPK. EMBO J 1996, 15(23):6564-6574. in an indica rice (Oryza sativa L.) using ESTs generated from 44. Tähtiharju S, Sangwan V, Monroy AF, Dhindsa RS, Borg M: The drought-stressed seedlings. J Exp Bot 2007, 58(2):253-265. induction of kin genes in cold-acclimating Arabidopsis thal- 22. Wu C-Q, Hu H-H, Zeng Y, Liang D-C, Xie K-B, Zhang J-W, Chu Z- iana. Evidence of a role for calcium. Planta 1997, H, Xiong L-Z: Identification of Novel Stress-responsive Tran- 203(4):442-447. scription Factor Genes in Rice by cDNA Array Analysis. J 45. Urao T, Katagiri T, Mizoguchi T, Yamaguchi-Shinozaki K, Hayashida Integr Plant Biol 2006, 48(10):1216-1224. N, Shinozaki K: Two genes that encode Ca(2+)-dependent pro- 23. Wang H, Zhang H, Gao F, Li J, Li Z: Comparison of gene expres- tein kinases are induced by drought and high-salt stresses in sion between upland and lowland rice cultivars under water Arabidopsis thaliana. Mol Gen Genet 1994, 244(4):331-340. stress using cDNA microarray. Theor Appl Genet 2007, 46. Xiong L, Schumaker KS, Zhu J-K: Cell signaling during cold, 115(8):1109-1126. drought, and salt stress. Plant Cell 2002, 14(Suppl):S165-S183. 24. Wang H, Zhang H, Li Z: Analysis of Gene Expression Profile 47. Chollet R, Vidal J, O'Leary MH: Phosphoenolpyruvate Carboxy- Induced by Water Stress in Upland Rice (Oryza sativa L. var. lase: A ubiquitous, highly regulated enzyme in plants. Annu IRAT109) Seedlings using Subtractive Expressed Sequence Rev Plant Physiol Plant Mol Biol 1996, 47:273-298. Tags Library. J Integr Plant Biol 2007, 49(10):1455-1463. 48. Nimmo GA, Nimmo HG, Hamilton ID, Fewson CA, Wilkins MB: 25. Wu Y, Wang Q, Ma Y, Chu C: Isolation and expression analysis Purification of the phosphorylated night form and dephos- of salt up-regulated ESTs in upland rice using PCR-based phorylated day form of phosphoenolpyruvate carboxylase subtractive suppression hybridization method. Plant Sci 2005, from Bryophyllum fedtschenkoi. Biochem J 1986, 168(3):847-853. 239(1):213-220. 26. Reddy AR, Ramakrishna W, Sekhar AC, Ithal N, Babu PR, Bonaldo 49. Vidal J, Chollet R: Regulatory phosphorylation of C4 PEP car- MF, Soares MB, Bennetzen JL: Novel genes are enriched in nor- boxylase. Trends in Plant Sci 1997, 2(6):230-237. malized cDNA libraries from drought-stressed seedlings of 50. Torres MA, Dangl JL: Functions of the respiratory burst oxidase rice (Oryza sativa L. subsp. indica cv. Nagina 22). Genome in biotic interactions, abiotic stress and development. Curr 2002, 45(1):204-211. Opin Plant Biol 2005, 8(4):397-403. 27. Guimarães CM, Stone LF, Rangel PHN, Ferreira ME, Rodrigues CAP: 51. Kägi JH, Schäffer A: Biochemistry of metallothionein. Biochemis- Resistência à seca: III. Avaliação do sistema radicular do try 1988, 27(23):8509-8515. arroz de terras em condições controladas. In Anais do II Con- 52. Eulgem T, Rushton PJ, Robatzek S, Somssich IE: The WRKY super- gresso da Cadeira Produtiva de Arroz e Reunião Nacional de Pesquisa de family of plant transcription factors. Trends Plant Sci 2000, Arroz Brasília, DF. Embrapa, SPI; 2006. 5(5):199-206. 28. Fischer RA, Maurer R: Drought resistance in spring wheat cul- 53. Ross CA, Liu Y, Shen QJ: The WRKY Gene Family in Rice tivars. I. Grain yield responses. Aust J Agric Res 1978, 29:897-912. (Oryza sativa). J Integr Plant Biol 2007, 49(6):827-842. 29. Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lip- 54. Ulker B, Somssich IE: WRKY transcription factors: from DNA man DJ: Gapped BLAST and PSI-BLAST: a new generation of binding towards biological function. Curr Opin Plant Biol 2004, protein database search programs. Nucleic Acids Res 1997, 7(5):491-498. 25(17):3389-3402. 55. Qiu YP, Jing SJ, Fu J, Li L, Yu DQ: Cloning and analysis of expres- 30. Telles GP, Silva FR: Trimming and clustering sugarcane ESTs. sion profile of 13 WRKY genes in rice. Chin Sci Bull 2004, Genet Mol Biol 2001, 24(1–4):17-23. 49(20):2159-2168. 31. De Mot R, Vanderleyden J: Application of two-dimensional pro- 56. Yu X-M, Griffith M, Wiseman SB: Ethylene Induces Antifreeze tein analysis of strain fingerprinting and mutant analysis of Activity in Winter Rye Leaves. Plant Physiol 2001, Azospirillum species. Can J Microbiol 1989, 35(10):960-967. 126(3):1232-1240. 32. Laemmli UK: Cleavage of structural proteins during the 57. Yu LX, Djebrouni M, Chamberland H, Lafontaine JG, Tabaeizadeh Z: assembly of the head of bacteriophage T4. Nature 1970, Chitinase: Differential induction of gene expression and 227(5259):680-685. enzyme activity by drought stress in the wild (Lycopersicon 33. Blum H, Beier H, Gross HJ: Improved silver staining of plant chilense Dun.) and cultivated (L. esculentum Mill.) tomatoes. proteins, RNA and DNA in polyacrylamide gels. Electrophoresis J Plant Physiol 1998, 153(5–6):745-753. 1987, 8(2):93-99. 58. Liu Q, Xue Q: Molecular Phylogeny, Evolution, and Functional 34. Shevchenko A, Wilm M, Vorm O, Mann M: Mass spectrometric Divergence of the LSD1-Like Gene Family: Inference from sequencing of proteins silver-stained polyacrylamide gels. the Rice Genome. J Mol Evol 2007, 64(3):354-363. Anal Chem 1996, 68:850-858. 59. Moons A, Prinsen E, Bauw G, Montagu MV: Antagonistic effects of 35. Sinclair TR, Ludlow MM: Influence of Soil Water Supply on the abscisic acid and jasmonates on salt stress-inducible tran- Plant Water Balance of Four Tropical Grain Legumes. Funct scripts in rice roots. Plant Cell 1997, 9(12):. Plant Biol 1986, 13(3):329-341. 60. Mittal R, Dubey RS: Behaviour of peroxidases in rice: changes 36. Grant RF, Jackson BS, Kiniry JR, Arkin GF: Water Deficit Timing in enzyme activity and isoforms in relation to salt tolerance. Effects on Yield Components in Maize. Agron J 1989, Plant Physiol Biochem 1991, 29(1):31-40. 81(1):61-65. 61. Yan S, Tang Z, Su W, Sun W: Proteomic analysis of salt stress- 37. Rosenthal WD, Arkin GF, Shouse PJ, Jordan WR: Water Deficit responsive proteins in rice root. Proteomics 2005, 5(1):235-244. Effects on Transpiration and Leaf Growth. Agron J 1987, 62. van der Straeten D, Rodrigues-Pousada RA, Goodman HM, van Mon- 79(6):1019-1026. tagu M: Plant enolase: gene structure, expression, and evolu- 38. Bray EA: Plant responses to water deficit. Trends Plant Sci 1997, tion. Plant Cell 1991, 3:719-735. 2(2):48-54. 63. Lal SK, Lee C, Sachs MM: Differential regulation of enolase dur- 39. Fukai S, Cooper M: Development of drought-resistant cultivars ing anaerobiosis in maize. Plant Physiol 1998, 118(4):1285-1293. using physiomorphological traits in rice. Field Crops Res 1995, 64. Jones AME, Thomas V, Truman B, Lilley K, Mansfield J, Grant M: Spe- 40(2):67-86. cific changes in the Arabidopsis proteome in response to 40. Agarwal S, Grover A: Isolation and transcription profiling of bacterial challenge: differentiating basal and R-gene medi- low-O2 stress-associated cDNA clones from the flooding- ated resistance. Phytochemistry 2004, 65(12):1805-1816. stress-tolerant FR13A rice genotype. Ann Bot 2005, 65. Lian H-L, Yu X, Lane D, Sun W-N, Tang Z-C, Su W-A: Upland rice 96(5):831-844. and lowland rice exhibited different PIP expression under 41. Wilkinson S, Davies WJ: ABA-based chemical signalling: the co- water deficit and ABA treatment. Cell Res 2006, 16(7):651-660. ordination of responses to stress in plants. Plant Cell Environ 2002, 25(2):195-210. 42. Hwang I, Sze H, Harper JF: A calcium-dependent protein kinase can inhibit a calmodulin-stimulated Ca2+ pump (ACA2) Page 13 of 13 (page number not for citation purposes) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Genomics Springer Journals

Identification of drought-responsive genes in roots of upland rice (Oryza sativa L)

Download PDF
13 pages

Loading next page...
 
/lp/springer-journals/identification-of-drought-responsive-genes-in-roots-of-upland-rice-2gXTRWbaOB

References (69)

Publisher
Springer Journals
Copyright
2008 Rabello et al; licensee BioMed Central Ltd.
eISSN
1471-2164
DOI
10.1186/1471-2164-9-485
Publisher site
See Article on Publisher Site

Abstract

Background: Rice (Oryza sativa L.) germplasm represents an extraordinary source of genes that control traits of agronomic importance such as drought tolerance. This diversity is the basis for the development of new cultivars better adapted to water restriction conditions, in particular for upland rice, which is grown under rainfall. The analyses of subtractive cDNA libraries and differential protein expression of drought tolerant and susceptible genotypes can contribute to the understanding of the genetic control of water use efficiency in rice. Results: Two subtractive libraries were constructed using cDNA of drought susceptible and tolerant genotypes submitted to stress against cDNA of well-watered plants. In silico analysis revealed 463 reads, which were grouped into 282 clusters. Several genes expressed exclusively in the tolerant or susceptible genotypes were identified. Additionally, proteome analysis of roots from stressed plants was performed and 22 proteins putatively associated to drought tolerance were identified by mass spectrometry. Conclusion: Several genes and proteins involved in drought-response, as well as genes with no described homologs were identified. Genes exclusively expressed in the tolerant genotype were, in general, related to maintenance of turgor and cell integrity. In contrast, in the susceptible genotype, expression of genes involved in protection against cell damage was not detected. Several protein families identified in the proteomic analysis were not detected in the cDNA analysis. There is an indication that the mechanisms of susceptibility to drought in upland rice are similar to those of lowland varieties. Page 1 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 sites for stress signal perception in which a signaling Background Rice (Oryza sativa L.) is a cereal of high economic and mechanism initiates a cascade of gene expression social value, which is used as a staple food by more than responses to drought. These transcriptional changes can half of the world's population. It is the only cereal which result in successful adaptations leading to stress tolerance is solely produced for human consumption. The produc- by regulating gene expression and signal transduction in tion of rice must increase 20% in the next 15 years in order the stress response (regulatory proteins) or directly pro- to keep pace with population growth. One of the main tecting the plant against environmental stress (functional constraints that affect yield in rice production is water def- proteins) [11]. icit. The increasing worldwide water shortage and uneven rainfall distribution limit the use of irrigated agriculture, Several functional genomic studies of rice have been per- typical of rice production. Irrigation costs are increasingly formed using different approaches such as macro and high worldwide. There is, therefore, a need to develop rice microarray [12,13], RT-qPCR, SAGE (Serial Analysis of varieties, which are more efficient in the use of water [1,2]. Gene Expression), MPSS (Massive Parallel Signature Sequenc- A major challenge for the research community is the rela- ing) and more recently oligoarray using the transcriptome tively limited progress made so far in improving the of rice to evaluate responses to abiotic stresses [14]. Pro- drought tolerance of high yielding rice varieties [3]. teome analyses have also been increasingly employed to complement genomic studies [15-18], however in a lower Rice is a highly diverse species, which can be grown in rate. Although numerous genes and proteins, which many types of soil moisture regimes, ranging from aerobic potentially contribute to drought tolerance in rice, have upland to permanently flooded lowland. Although been reported [19-22], most of these studies have focused upland rice constitutes a relatively small proportion of the on lowland rice genotypes. Currently, very little is known total rice area worldwide, it is the predominant method of about gene and protein expression in upland rice [22-25]. rice cultivation in Latin America and West Africa (about Moreover, most ESTs from drought stressed plants availa- 75% and 50% of rice area, respectively) [4]. In Brazil, ble were obtained from libraries constructed using seed- upland rice responds for approximately 40% of the total lings [26]. There are very few reports on gene expression rice production. In some areas of the country, upland rice of drought-stressed plants in the reproductive stage and is a subsistence crop planted by farmers who apply limited using root tissue of plants growing under defined field inputs to their crops. The cultivation of upland rice in capacity. marginal areas with low soil fertility and threatened by severe abiotic stresses, such as periods of drought during The comprehension of drought responses in upland rice is the cropping season, has a significant impact on rice pro- important for designing breeding strategies to develop duction [5,6]. Due to exposure to many environmental varieties more tolerant to water constraints. Recently, the constraints, some local varieties of the tropical japonica tolerance of ten traditional upland varieties of rice sub- rice developed high adaptability to drought stress, hot and mitted to drought stress has been evaluated as part of an dry climatic conditions of regions in Latin America and effort to identify new sources of drought tolerance in rice Africa. Therefore, these varieties may show high levels of [27]. Concomitantly, the root system of two of the above water usage efficiency and constitute an excellent material mentioned upland rice genotypes, characterized as sus- for studying drought tolerance mechanisms in rice. In Bra- ceptible and tolerant to drought stress, have been ana- zil, for example, EMBRAPA maintains a germplasm bank lyzed at the reproductive stage using genomic and enriched with traditional upland rice landraces collected proteomic approaches. Several genes and proteins were in areas where cultivated rice has been grown since its identified, which may play important roles in drought tol- introduction in the country, centuries ago, and may repre- erance. sent an extraordinary source of genes that control traits of economic importance such as drought tolerance [7]. Methods 1. Plant material and phenotypic evaluation The determination of the mechanisms directly involved in Plants of traditional upland rice (O. sativa L. var. japonica) drought tolerance remains a challenging task since varieties were grown on PVC pipe columns (25 cm of drought is a complex trait that involves several metabolic diameter; 80 cm of height) filled with fertilized Oxisol pathways [3]. The identification and isolation of genes under screenhouse conditions [27]. The experimental associated with drought tolerance is of major importance design was a split-plot design with two watering regimes in order to better understand this trait and increase the as main plots, ten traditional upland varieties as subplots efficiency in developing drought tolerant varieties [8-10]. and three replications. The watering regimes were (a) con- At the molecular level, the response of roots to water lim- trol, consisting of a main plot of well-watered plants iting conditions seems to be crucial to trigger drought tol- throughout the experiment, which received 100% reposi- erance mechanisms, since roots are one of the primary tion of the water lost daily and a minimum soil humidity Page 2 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 of -0,025 MPa at 15 cm of depth, and (b) drought stress, and sequence homologues were identified using the Blast which consisted of 50% reposition of the water lost daily program [29]. An in silico subtraction was performed by from anthesis on. Water reposition was calculated based clustering all sequences from both cDNA libraries accord- on daily weighting of columns with a mechanical scale. ing to the methodology described by Telles and da Silva Twenty-one days after initiating the drought stress treat- [30], allowing the identification of genes exclusively ment (at anthesis), roots of each treatment (control and found in each library. drought stress) were collected from each rice variety. All root samples were immediately frozen in liquid nitrogen 3. Protein extraction and 2-DGE and maintained at -80°C until their use for RNA and pro- Total protein was extracted from roots of the drought tol- tein extractions. At harvest, grain yield and yield compo- erant (Prata Ligeiro) and susceptible (IRAT20) genotypes nents of each genotype were evaluated, including root and according to procedures described by de Mot and Vander- shoot dry weight, harvest index, spikelet sterility, grains leyden [31] Plant material of the three replications were per panicle and weight of 100 grains. Drought tolerance pooled, pulverized and mixed with extraction buffer (0.7 parameters were estimated based on calculations of M sucrose, 0.5 M TrisHCl, 30 mM HCl, 50 mM EDTA, 0.1 drought severity, drought tolerance index and drought M KCl and 40 mM DTT) and phenol (100%) in the same susceptibility index [28]. The genotypes submitted to the volume (750 μl). Proteins were precipitated with ammo- drought stress showed differences in most of the yield nium acetate 0.1 M in methanol, washed with acetone parameters analyzed, which were significantly influenced 80% (v/v), dried and stored at -20°C. Protein quantifica- by the drought severity applied to the experiment [27]. tion was performed using the Bradford Reagent (Invitro- These parameters were then used to classify the genotypes gen, USA). Isoelectric focusing was conducted using 11- according to their reaction to stress. Among them, two cm immobilized pH gradient (IPG) strips with a pH range contrasting genotypes for drought stressing conditions of 4–7 and a Multiphor II electrophoresis system (GE). were selected for the present study: Prata Ligeiro, as the Strips containing approximately 220 μg of protein were tolerant, and IRAT20, as the susceptible variety. The RNA rehydrated with 2% (v/v) CHAPS, 8 M urea, 7 mg dithio- and protein analyses proceeded only with root tissue threitol (DTT) and 2% IPG buffer. Second dimension extracted from these two varieties. analysis was performed in 10% gels by SDS-PAGE as described by Laemmli [32] and at least five replications of 2. RNA extraction and subtractive library construction each genotype were performed. Protein spots were visual- For each genotype, a bulk of approximately 250 mg of ized after silver [33] or Comassie blue staining. plant roots from the three replications were homogenized 4. Image analysis in liquid nitrogen and total RNA was extracted using the Concert™ Plant RNA Reagent (Invitrogen, USA), accord- The 2D gel images were evaluated using the Platinum soft- ing to manufacturer's instructions. This procedure was fol- ware (GE Healthcare, UK) and three high quality gels lowed for roots harvested from drought stressed as well as obtained for both genotypes were analyzed. First, a cali- unstressed plants. mRNA was then isolated from total bration with a grey scale was performed to transform grey RNA by using PolyATtract mRNA Isolation System levels into OD values for each pixel (px) of the gel image. (Promega, USA). Quantity and quality of the isolated The wizard detection method proposed by the software mRNA was evaluated by spectrophotometry and electro- was used to detect the spots with the following parame- phoresis in agarose gel 1%, respectively. ters: 15 px for estimated spot size, 50 px for minimum spot size and a spot contrast enhancement of 75%. Auto- Isolated mRNAs were used for cDNA synthesis and sup- matically detected spots were checked and some of them pression subtractive hybridization (SSH) library construc- were manually added or removed. Following the detec- tion by using the PCR Select Subtraction Kit (Clontech, tion procedure, the normalization step was carried out to USA). Subtractive hybridizations were performed using attribute a common spot identity for the same spots cDNA from stressed plant roots (as tester) against cDNA derived from different images utilizing the reference gel from well-watered unstressed plant roots (as driver) of construct and automatically matching options. A syn- each genotype, in order to identify genes involved in thetic gel from each genotype was constructed by using drought response. The subtractive PCR products obtained the mean value of volume percentage of each protein spot were cloned into pGEM T-Easy (Promega, USA) and present in the three replicates, according to the Platinum sequenced in ABI Prism 3700 DNA Analyser (Applied Bio- software's (GE Healthcare, UK) instructions. The two systems Inc., USA). A minimum insert size of 30 bp and at obtained synthetic gels were then overlapped using the least 20 bp with quality of phred > 20 were considered for molecular marker as well as several protein spots present the analysis. Sequences were deposited in GenBank under in both profiles as landmarks. The overlapped images the accession numbers of FG124418 through FG124880 were based on landmark spots showing same pI and Mw. Page 3 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 5. Trypsin digestion and mass spectrometry analysis Results and discussion Protein spots were excised manually from 2D gels and in- 1. Experimental design and sampling Plants were submitted to drought stress after anthesis for gel digested with sequencing grade trypsin (Promega, Madison, WI) according to Schevchenko et al. [34]. twenty-one days. Flowering is the period in which the Briefly, each protein spot was placed in a 0.5 mL polypro- plant is most sensitive to water deficit and several toler- pylene (Eppendorf) tube and destained by washing 5–8 ance mechanisms need to be activated at this stage in times with 200 μL of 50% (v/v) acetonitrile/10 mM order to guarantee grain filling and production [6]. Dur- ammonium bicarbonate solution. The gel pieces were ing root sampling, a clear visual difference in Prata Ligeiro subsequently dehydrated by washing with 200 μL of and IRAT20 plants could be observed. An intense leaf roll- 100% acetonitrile and completely dried in a Speedvac ing was noticed in the susceptible genotype as opposed to concentrator. Ten microliters of 50 mM ammonium bicar- the tolerant. In addition, a more pronounced aerial bio- bonate/10% (v/v) acetonitrile solution containing 100 ng mass loss could be visualized in IRAT20. At harvest, yield of trypsin were added, and the sample incubated at 37°C and yield component parameters were measured [27]. The for 16 h. Aliquots of each tryptic digest (1 μL) were mixed variety IRAT20, a high yielding variety under irrigated with a saturated solution of α-cyano-4-hydroxycinnamic controlled conditions, showed a 51% reduction in grain acid, spotted onto a MALDI target plate, and allowed to yield when submitted to drought stress. On the other air dry. hand, Prata Ligeiro, a low yielding variety under well watering conditions, had a 23% reduction in grain yield Mass spectra were acquired using a MALDI-TOF/TOF under drought stress. The drought susceptibility index Autoflex II spectrometer (Bruker Daltonics, Bremen, Ger- based on yield was estimated as 0.73 for Prata Ligeiro (tol- many) operating at a laser frequency of 50 Hz. MS analysis erant) and 1.57 for IRAT20 (susceptible). were performed in a positive ion reflection mode. Voltage parameters were set as IS1 19 kV, IS2 16.8 kV, Lens 8 kV, Collected roots of both genotypes were then used for Reflector 20 kV, Reflector2 9.54 kV. The delay time was 70 cDNA library construction and proteome studies. In the ns and acquisition mass range 700–3200 Da. External cal- cDNA library study, stressed plants were contrasted with ibration was performed using a peptide mix contaning well-watered plants, whereas in the proteome analysis, ACTH (1–24), ACTH (18–39), Somatostatin, Angiotensin stressed plants from both genotypes were compared. I and Angiotensin II, all from Sigma. MS/MS analysis were performed in a positive ion LIFT reflection mode. Voltage Water reposition, based on the evapotranspiration rate, parameters used were IS1 6 kV, IS2 5.3 kV, lens 3.15 kV, has been used to determine an impartial and consistent Reflector 23.5 kV, Reflector2 9.7 kV, LIFT1 19 kV and response of plants to drought stress, during long periods LIFT2 4 kV. The delay time was set as zero and acquisition of drought in the soil [35]. Several studies have tried to mass range 40–2400 Da. define the critical limit of water in the soil after which crop development and production are significantly affected Peak lists were generated using the FlexAnalysis 3.0 soft- [36]. According to Rosenthal et al. [37], the symptoms of ware (Bruker Daltonics). The sophisticated numerical water deficit occur when water availability is around 50% annotation procedure (SNAP) algorithm was used to of the field capacity. detect the monoisotopic peak values, with a quality factor threshold of 30 and 6 as S/N threshold. Database searches The response of plants to drought stress is also dependent were performed in February 2008 using the MASCOT on the extension and rate of water loss [38]. Fukai et al. search engine (Matrix Science, UK) with the NCBInr pro- [39] reported that when a rapid water deficit occurs, the tein database and Oryza sativa taxonomy. The mass toler- morpho-physiological mechanisms are severely affected. ance was 100 ppm and one missed cleavage was allowed. When the deficit is prolonged for a few days, plants are Carbamidomethylation of cysteines, oxidation of methio- allowed to adapt to the stress, enabling the identification nine, and acrylamide-modified cysteines were considered of variability in drought tolerance within different geno- for PMF searches. For accepting the identification, the cut- types, since plants can respond differently to the same off value for the Probability Based Mowse score calculated stress condition [38]. Therefore, the sampling time used by MASCOT (at p < 0.05) was used. For MS/MS data, the in this study (21 days of drought stress) may have allowed peptide mass tolerance was 0.5 Da, MS/MS ion mass tol- the analysis of adaptive responses of the plant to tolerate erance at 0.5 Da, allowance of 1 missed cleavage, and water deficit. charge state +1. When the pI and MW of matched proteins were not available, these values were calculated using Several studies reported the response of rice seedlings to ExPASy Compute pI/Mw tool http://ca.expasy.org/tools/ drought stress [13,26,40] however, little attention has pi_tool.html. been given to the expression of genes in water-stressed Page 4 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 plants at the reproductive stage (flowering, grain filling) previously reported as associated to drought stress and in which a higher yield impact is observed [6]. some of them are discussed below. 2. cDNA library analysis Genes involved in signaling routes were exclusively iden- Roots are one of the primary sites responsive to restrictive tified in Prata Ligeiro and include serine/threonine kinase, conditions of water availability and, as a result, synthesize ethylene-responsive factor and calcium-transporting chemical signals for a rapid response of the plant to ATPase/calmodulin binding sequences. Serine/threonine 2+ drought stress [41]. This occurs since the response in kinases are Ca dependent proteins kinase (CDPKs), leaves must be stimulated rapidly to avoid irreversible involved in the phosphorylation cascade of proteins. Sev- damage to the photosynthetic machinery. In this work, eral studies have shown that CDPKs are induced or acti- two subtractive cDNA libraries were constructed using vated by abiotic stresses, suggesting that they may be mRNA from roots of tolerant and susceptible upland rice involved in drought signaling [42-45]. Another identified genotypes subtracted from their respective unstressed gene associated to signal transduction was an ethylene- well-watered controls. The subtracted PCR products responsive factor. Ethylene is a well characterized phyto- obtained after primary and secondary PCR ranged from hormone that may act alone or in combination with ABA 0,1 – 1,5 kb. in regulating gene expression under abiotic stress [46]. Calcium-transporting ATPase/calmodulin binding are The SSH libraries of the tolerant (Prata Ligeiro) and sus- also stress-signaling proteins and are responsible for regu- ceptible (IRAT20) genotypes were concluded with a nov- lation of the osmotic potential of the cell. elty index of 66% and 55%, respectively. The general analysis of the two libraries revealed a total of 463 valid Some genes that participate in metabolism alterations as sequences (230 from Prata Ligeiro and 233 from IRAT20) a result of the limitation caused by low levels of intracel- and the average fragment size was of 300 bp. Several genes lular CO observed during drought stress were also identi- commonly expressed in both genotypes were identified fied only in Prata Ligeiro. Among these genes are those and are probably not directly involved in drought toler- coding for Phosphoenolpyruvate carboxykinase, an ance. enzyme that has a key role in nocturnal fixation of CO ; malato dehydrogenase, which is an enzyme particularly In order to determine the genes exclusively expressed in important for the assimilation of carbon in C4 plants; the tolerant and susceptible genotypes, an in silico subtrac- Glutamate-1-semialdehyde aminotransferase and glu- tion was performed using sequences of both libraries. The cose-1-fosfato adenililtransferase [47-49], both involved results for the in silico subtraction revealed that the 463 in carbohydrate metabolism. sequences represented 282 different transcripts: 127 were found in both genotypes, 84 were exclusively expressed in It has been proposed that the mechanism involved in the Prata Ligeiro library (Table 1) and 71 were observed drought tolerance in upland rice is a result of a higher only in the IRAT20 library (Table 2). expression of genes involved in oxidative stress protection [23]. Indeed, in the present study some genes associated 2.1. Putative drought-tolerance genes identified in Prata Ligeiro to the protection of the cell were expressed only in the tol- Drought tolerance is a complex trait and involves mecha- erant genotype. Among them, we found a Methionine sul- nisms that act in isolation or combined to avoid or toler- foxide reductase A and a Respiratory burst oxidase ate periods of water deficit. It is expected that genotypes homolog, which act in the recognition of reactive oxygen responding differently to drought stress show differences species (ROS) in biotic and abiotic stresses [50]. Other in gene expression, and that a portion of the differences is interesting genes identified are Metallothionein, a super- related to drought tolerance. Therefore, the analysis of the family of low molecular weight proteins involved in metal genes found exclusively in the tolerant genotype is of detoxification [51] and scavenging of oxygen-free radicals, interest to identify genes associated with water usage effi- which can decrease injury in oxidative tissue, and Ferre- ciency. doxin, regulated by different environmental stresses including biotic and abiotic conditions. Among the 84 transcripts uniquely reported in the toler- ant genotype, 14 did not present known homologs (no Genes associated to maintenance of cell turgor were also hits) and 17 showed similarities to proteins with identified such as IQ calmodulin-binding and Calcium- unknown function (hypothetical proteins). Three transporting ATPase/calmodulin binding. These genes sequences showed similarity to non-plant proteins and were previously reported to participate in typical defense probably represent contaminating sequences (Table 1). mechanisms in upland varieties [23]. The other transcripts showed similarity to several proteins Page 5 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 Table 1: Genes detected exclusively in roots of the tolerant genotype (Prata Ligeiro) SSH library Encoded protein Homologous organism Accession number Proteins of known function Glutamate-1-semialdehyde 2,1 aminomutase Oryza sativa NM_001068872 Metallothionein-like protein Oryza sativa NM_001056317 Malate dehydrogenase Oryza sativa NM_001062924 Methionine sulfoxide reductase A Oryza sativa NM_001063272.1 Phosphatidylinosytol 3 and 4 kinase Oryza sativa NM_001060732 Ubiquitin-conjugating enzyme Oryza sativa NM_001048429 Nuclear protein SET domain containing protein Oryza sativa NM_001067672 Splicing factor 3B subunit 5-like protein Oryza sativa dbj|BAD10044.1| PEP carboxikinase Oryza sativa gb|ABF95034.1 Putative malate dehydrogenase Oryza sativa gb|AAT69584.1| Eukaryotic translation initiation factor 5A-2 (eIF-5A) (eIF-4D) Oryza sativa NC_008405 Metallothionein-like protein type 1 Oryza sativa NP_001068544.1 ADP glucose pyrophosphorylase Oryza sativa EF122437 CBL-interacting protein kinase 1 Oryza sativa NM_001049327 ADP-ribosylation factor Oryza sativa NM_001051134 DSS1/SEM1 family protein Oryza sativa NC_008394 Ankyrin repeat containing protein Oryza sativa NM_001054582 Pathogenesis-related transcriptional factor and ERF domain containing protein Oryza sativa NC_008402 E-class P450, group I family protein Oryza sativa NM_001074239 FAR1 domain containing protein Oryza sativa NM_001057341 Tubulin alpha-1 chain Oryza sativa NM_001074145 Putative ubiquitin conjugating enzyme Oryza sativa dbj|BAB89662.1| DEAD/DEAH box helicase domain containing protein Oryza sativa NM_001069156 Putative pollen specific protein C13 precursor Oryza sativa gb|AAM08621.1| IQ calmodulin-binding Oryza sativa NM_001061046 HAD superfamily hydrolase 5' nucleotidase protein Oryza sativa NM_001057956 SAM biding motif domain containing protein Oryza sativa NM_001070787 Peptidase aspartic family protein Oryza sativa NM_001063168 Nonaspanin (TM9SF) family protein Oryza sativa NM_001056027 Ethylene responsive element binding factor 5 Oryza sativa NM_001063579 TMS membrane protein Oryza sativa NM_001054899 Heat shock protein DnaJ family protein Oryza sativa NM_001060020 Ferredoxin III, chloroplast precursor (Fd III) Oryza sativa NC_008396 Anther ethylene-upregulated protein ER1 (Fragment) Oryza sativa NM_001055765 Chaperone protein DNA-J-related like Oryza sativa dbj|BAD27799.1| Isoflavone reductase family protein Oryza sativa NM_001068997 U box domain containing protein Oryza sativa NM_001071339 Ribossomal protein L Curculio glandium AM049038 Short chain dehydrogenase tic32 Oryza sativa NM_001048577 Arabinogalactan protein Oryza sativa NC_008394 Ribonuclease T2 family protein Oryza sativa NM_001070328 HvB12D protein (B12Dg1 protein) Oryza sativa NM_001063815 Respiratory burst oxidase homolog Oryza sativa NM_001049555 Phosphatidylinositol-4-phosphate 5-kinase family protein Oryza sativa NM_001068386 Nodulin-like Oryza sativa NM_001070322 Cathepsin B-like cysteine protease form 2 Ixodes ricinus gb|ABO26563.1| Cathepsin L-like cysteine proteinase precursor Acanthoscelides obtectus gb|AAQ22984.1| Calcium-transporting ATPase/calmodulin binding Arabidopsis thaliana NP_188931.1 Myb, DNA biding domain containing protein Oryza sativa NM_001062445 TGA-type basic leucine zipper protein Phaseolus vulgaris gb|AF402607.1| Tocopherol O-methyltransferase, choroplast precursor Oryza sativa NM_001054379 ATP-dependent Clp protease ATPbiding subunit Clpx-like mitochondrial precursor Oryza sativa dbj|BAD15818.1| HvB12D protein (B12Dg1 protein) Oryza sativa NM_001063815 Uncharacterized protein family containing protein Oryza sativa gb|ABA91393.1| Protein of unknown function Protein of unknown function Oryza sativa NC_008397 Protein of unknown function Oryza sativa NC_008403 Page 6 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 Table 1: Genes detected exclusively in roots of the tolerant genotype (Prata Ligeiro) SSH library (Continued) Unknow function Oryza sativa NM_001067277 Hypothetical protein Oryza sativa AP008208 Hypothetical protein Oryza sativa gb|EAY93896.1| Conserved hypothetical protein Oryza sativa NM_001065538 Hypothetical protein Oryza sativa gb|EAY84091.1| Hypothetical protein Oryza sativa CT836006 Hypothetical protein Oryza sativa NC_008394.1 Hypothetical protein Oryza sativa NC_008394.1 Hypothetical protein Oryza sativa AP008208 Hypothetical protein Oryza sativa NM_001057688 Hypothetical protein Oryza sativa NM_001066910 Hypothetical protein Oryza sativa NM_001053573 Hypothetical protein Oryza sativa CT829595 Hypothetical protein Oryza sativa CT834076 In this study we have also identified genes which have not 3. Proteome analysis yet been directly related to drought tolerance, such as In order to complement the genomic studies, protein B12Dg1 protein, Nuclear protein SET domain containing maps of roots from water-stressed plants of the suscepti- protein and Putative pollen specific protein C13 precur- ble (Figure 1A) and tolerant (Figure 1B) genotypes were sor, as well as genes with unknown function. Further stud- compared. Triplicates of the gels from each genotype were ies need to be performed in order to assign biological compared and revealed a total of 463 proteins in the Prata function, since these genes may play important roles in Ligeiro profile and 522 in IRAT20. The two obtained syn- plant adaptation during drought stress conditions. thetic gels were overlapped and this procedure allowed the identification of 307 overlapped spots, 156 proteins 2.2. Drought-responsive genes identified in IRAT20 exclusive to the tolerant genotype and 215 proteins exclu- Regarding the response of the susceptible genotype to sive to the susceptible genotype. These results show a drought stress, 71 transcripts were exclusively expressed in higher diversity in the protein pattern of the susceptible this genotype. As in Prata Ligeiro, a high number of genes genotype. (14) with no known homologs (no hits) were identified (Table 2). Moreover, a total of 23 genes encoding hypo- A total of 50 intense proteins observed in the tolerant gen- thetical or unknown proteins were also observed. Further otype profile after Coomassie blue staining was excised expression studies of these genes may reveal important from the gel, digested and analyzed by mass spectrometry. genes associated to drought stress response, which have By using the Mascot program, 22 proteins could be iden- not been explored so far. This information may contribute tified with a significant score (Table 3), including 16 up- to a better understanding of the mechanisms related to and 4 down-regulated, 1 new and 1 equally expressed in drought susceptibility in upland rice varieties. both genotypes (Figure 2). The other proteins were in insufficient amounts for the identification analysis or did As in Prata Ligeiro, three transcripts showed similarity to not return reliable matches when using the Mascot pro- non-plant proteins and were not considered in the analy- gram. This probably occurs due to a low protein quantity sis since they probably represent contaminating and/or low ionization capacity of molecular components sequences (Table 2). The other transcripts showed similar- present in the samples analyzed. It is also possible that, ity to genes associated to different functions including the considering the high amount of "no hits" obtained in the transport of small molecules or inorganic ions, such as genomic analysis, protein sequences matching the pep- HCO -transporter and Vacuolar H+ pyrophosphatase. tides searched were not available in public databases. The The expression of these genes was previously reported by peptide sequences obtained were also analyzed using the Wang et al. [23] in a lowland variety. These results suggest Blastp program. that upland genotypes susceptible to drought may present similar responses to those of lowland varieties, which are Spots PL1 and PL2 (up-regulated in Prata Ligeiro) were naturally more susceptible to water deficit. identified as hypothetical proteins which contain Ricin B- related lectin domain. Other up-regulated hypothetical Interestingly, the well-known transcription factor WRKY proteins were also identified and include protein spots was uniquely identified in IRAT20. WRKY mediates plant PL34, PL45 and PL51. Spot PL45 and PL51 were expressed stress responses [52-54] and the increased expression of 2.6 and 4.5 fold, respectively, in the tolerant genotype this protein has been frequently associated to drought (Figure 2), indicating that these proteins may play an stress response in rice [23,55]. important role in drought tolerance. Spot PL57 was Page 7 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 Table 2: Genes detected exclusively in roots of the susceptible genotype (IRAT20) SSH library Encoded protein Homologous organism Accession number Proteins of known function T complex 11 family protein Oryza sativa NM_001059402 Protein kinase domain containing protein Oryza sativa NM_001071926 Protein disulphide isomerase family protein Oryza sativa AP008208 TPR-like domain containing protein Oryza sativa NM_001058028 Protein kinase Oryza sativa NM_001074788 Pinoresinol-lariciresinol reductase TH1 Oryza sativa NM_001073059 Smr protein; MutS2 c- terminal domain containing protein Oryza sativa NM_001048992 SIPL protein (Membrane-type 1 matrix metalloproteinase cytoplasmic tail binding protein-1) Oryza sativa NM_001055581 Similar to CG 9092- PA Tribolium castanum XP_967647.1 Putative ATP-dependent Clp protease ATP-binding subunit ClpX1 (CLPX) Oryza sativa dbj|BAD15818.1| Cytocrome P450 family protein Oryza sativa NM_001071591 Preprotein translocase subunit sec Y, chloroplast precursor Oryza sativa NM_001067916 Vacuolar H+ pyrophosphatase Oryza sativa NM_001063501 Similar to UPF 0139 protein CGI-140 Tribolium castaneum XP_971064.1| 60 kDa inner membrane insertion protein family protein Oryza sativa NM_001055291 Glyceraldehyde-3-phosphate dehydrogenase (Fragment) Oryza sativa NM_001055382 Similar to splicing coativator subunit SRm 300 Monodelphis domestica XP_001371550.1| Cysteine synthase, mitocondrial precursor Oryza sativa NM_001052112 TPR-like domain containing protein Oryza sativa NM_001056953 HCO3-transporter Oryza sativa NM_001073581 Banched chain amino-acid aminotransferase-like protein 3 Oryza sativa NM_001049072 Beta tubulin (fragment) Oryza sativa NM_001049296 HAT dimerisation domain containing protein Oryza sativa NC_008402 Urease accessory protein G Oryza sativa NM_001062872 Glycoside hydrolase, family 47 protein Oryza sativa NM_001054615 WRKY transcription factor 82 Oryza sativa DQ298186 Tubby family protein Oryza sativa NM_001062568 Ribosomal protein L41 family protein Oryza sativa NC_008400 Granule-bound starch synthase I, chloroplast precursor Oryza sativa NM_001065985 Putative RNA polymerase I transcription factor RRN3 Oryza sativa dbj|BAD45608.1| Aconitate hydratase, cytoplasmic (Citrate hydro-lyase) (Aconitase) Oryza sativa NM_001055433 Short chain alcohol dehydrogenase-like Oryza sativa NM_001056212 Putative ubiquitin-conjugating enzyme E2 Oryza sativa dbj|BAD25096.1| Peptidase s26A signal peptidase I family protein Oryza sativa NM_001074823 Protein of unknown function Unknown protein Oryza sativa NM_001068742 Hypothetical protein Oryza sativa AC119292 Hypothetical protein Oryza sativa AP008208 Hypothetical protein Oryza sativa AK243578 Hypothetical protein Oryza sativa NC_008395.1 Hypothetical protein Oryza sativa AP008208 Hypothetical protein Oryza sativa NM_001057104 Hypothetical protein Oryza sativa NC_008395 Hypothetical protein Oryza sativa NM_001074804 Hypothetical protein Oryza sativa NM_001057688 Hypothetical protein Oryza sativa NC_008401.1 Hypothetical protein Oryza sativa NC_008395.1 Hypothetical protein Oryza sativa CR855113 Hypothetical protein Oryza sativa AC145477 Hypothetical protein Oryza sativa AC092556 Hypothetical protein Oryza sativa AK242616 Hypothetical protein Oryza sativa AP008209 Hypothetical protein Oryza sativa NC_008398.1 Hypothetical protein Oryza sativa AC099401 Hypothetical protein Oryza sativa NM_001050487 Hypothetical protein Oryza sativa CT831698 Hypothetical protein Oryza sativa CT828847 Hypothetical protein Oryza sativa CT832865 Page 8 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 another protein identified as hypothetical and was exclu- (Table 3), involved in carbohydrate metabolism. Accord- sively expressed in Prata Ligeiro. These proteins are inter- ing to Wang et al. [23], genes related to metabolism are esting candidates for futures studies aiming at the more expressed in lowland than in upland genotypes. It is determination of biological function. possible that susceptibility to drought in upland rice may occur in a similar way as in lowland rice. Spots PL3 and PL60 were identified as the same protein chitinase and spot PL11 as a Chain A, Crystal Structure of Spots PL43 and PL46 were both identified as enolase, a Class I Chitinase. Chitinases are pathogenesis-related pro- glycolytic enzyme, which participates in metabolic proc- teins expressed in response to biotic and abiotic stresses esses. The up-regulation of enolase has been previously and have been studied in grasses such as rye in response reported in rice roots in response to salt stress [61] and to to cold and drought stress [56]. Spot PL60 was highly PEG treatment [24]. Unexpectedly, PL46 was equally induced in the tolerant genotype, which confirms the up- expressed in Prata Ligeiro and IRAT20, while spot PL43 regulation of this protein during drought stress. Chiti- was up-regulated in Prata Ligeiro. The existence of multi- nases have also been reported as being induced in tomato ple enolase isoforms in plants has been reported [62] and plants tolerant to drought when compared to the suscep- it is possible that the enolases identified in this study rep- tible genotype [57]. resent different isoforms, which respond differently to drought stress conditions. Indeed, difference in the Two other pathogenesis-related proteins were identified: expression of enolase isoforms was observed in maize in one was up-regulated (spot PL33) and the other repressed response to anaerobiosis [63]. (PL30) in the tolerant genotype (Figure 2). The expression of these proteins has been previously reported in roots of A highly induced protein (15 fold) in the tolerant geno- rice in drought stress conditions and although the role of type (PL40) showed identity to a hypothetical protein as proteins of this family is not well established, they have well as a salt stress induced protein (Table 3). Similarly, been associated to hypersensitive reaction in response to spot 27 (2.6 fold higher in Prata Ligeiro) also presented biotic and abiotic factors [58]. In drought stress condi- identity to the salt stress induced protein. It is possible tions, pathogenesis-related proteins as well as the salt that these spots represent new rice proteins, not identified stress-responsive SalT protein have been reported in rice so far that contain a conserved region present in both roots [59]. matching proteins. The induction of proteins involved in tolerance to salt stress, during water deficit conditions, As observed in the constructed cDNA libraries, several shows that osmotic stress is an important aspect during proteins involved in oxidative stress protection were drought. Similar mechanisms are activated in response to induced in the tolerant genotype and were identified as a different abiotic stresses, as previously reported [10]. superoxide dismutase [Cu-Zn] (PL20), L- ascorbate perox- idase 1 (PL23), ascorbate peroxidase (PL38) and cytosolic Conclusion malate dehydrogenase (PL63) (Table 3). Peroxidases are Several genes and proteins involved in drought-response anti-oxidative enzymes, described in varieties of rice toler- as well as genes with no described homologs were identi- ant to high salinity conditions [25,60] and in upland rice fied in this work. Genes exclusively expressed in the toler- roots in response to osmotic stress [24]. These proteins are ant genotype were, in general, related to maintenance of involved in cellular detoxification and it is possible that turgor and cell integrity. In contrast, in the susceptible this is a general defense mechanism in response to water genotype, expression of genes involved in protection deficit in upland rice. According to Wang et al. [23,24] tol- against cell damage was not detected, indicating that there erance to drought stress observed in upland varieties may be a higher degradation of cellular components in includes detoxification mechanisms, limiting the accumu- these genotypes. Similar results were obtained by Wang et lation of reactive oxygen species. These authors reported al. [23] when comparing tolerant upland and lowland that these proteins were up-regulated in upland cultivars varieties. These results indicate that the mechanisms of when comparing tolerant lowland and upland rice. Unex- susceptibility in upland rice are similar to those of low- pectedly, proteins identified as superoxide dismutase land varieties, considering that the upland rice is naturally (PL7) and GSH-dependent dehydroascorbate reductase more tolerant to drought stress. (PL13) were down-regulated in the tolerant genotype. These proteins were not identified in the genomic analy- The proteomic analyses were complementary to the sis, highlighting the importance of proteomics studies to genomic data obtained. The expression of genes associ- complement the results obtained. ated with cell protection against oxidative damage is con- sidered important to cope with water deficit in upland Another down-regulated protein (PL24) identified in the rice. In this study, genes and proteins related to this func- Prata Ligeiro genotype was triosephosphate isomerase tion showed a higher expression in the tolerant genotype. Page 9 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 Root protein pr Figure 1 ofiles by 2-DGE of the susceptible (A) and tolerant (B) genotypes Root protein profiles by 2-DGE of the susceptible (A) and tolerant (B) genotypes. Total soluble protein (ca. 220 μg) was separated by 2-DGE and the spots were visualized after silver staining. Numbers indicate the protein spots successfully identified by mass spectrometry. Benchmark Protein Ladder (Invitrogen, USA) was used to estimate the molecular mass of the proteins visualized. Page 10 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 H ceptible (IRA Figure 2 istogram reT20) g presenting enotypes, as det expression levels of ermined by up- and the Platinu downm -regulated protei software (GE Healthcare, UK) ns identified in the tolerant (Prata Ligeiro) and sus- Histogram representing expression levels of up- and down-regulated proteins identified in the tolerant (Prata Ligeiro) and susceptible (IRAT20) genotypes, as determined by the Platinum software (GE Healthcare, UK). Interestingly, in the proteomics analysis, the susceptible Overall, due to the low amount of information regarding genotype showed a higher diversity in the protein profile, upland rice gene and protein expression in response to revealing more uniquely expressed proteins than the tol- water deficit, this study sheds some light over the compre- erant genotype. On the other hand, in the genomic study, hension of this complex mechanism. However, the high the number of exclusively expressed transcripts in the sus- amount of transcripts and proteins with unknown func- ceptible genotype was lower. It is well known that tran- tion obtained is still intriguing. These genes and proteins script levels do not always reflect protein amounts need to be further investigated in order to assign their bio- [64,65]. Therefore, it is possible that the transcripts related logical function and advance our knowledge regarding to the proteins exclusively present in IRAT20 2D maps drought tolerance in upland rice. were in low amounts, and not detected by the genomic analysis, or they were subtracted from the control condi- Authors' contributions tion in the hybridization process. Differences in transla- AM, ARR, CMG, MEF and PHNR designed and performed tion efficiency may have occurred, resulting in a higher the research. FRS analyzed the sequence data and EMS and amount of the corresponding proteins, further detected by DS analyzed the mass spectrometry data. ARR and AM 2-DGE. These results clearly show that proteomics studies drafted the manuscript. ACMB and CRS critically revised can reveal important additional information and that the the article. All authors approved the final version. use of complementary approaches is useful for a better understanding of complex biological traits, such as Acknowledgements This research was supported by Embrapa, CNPq and Embrapa Recursos drought tolerance. Genéticos e Biotecnologia. Page 11 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 Table 3: Proteins identified by peptide mass fingerprinting or de novo sequencing Spot n° Peptide sequence Protein identification Accession # Score Mr (gel) pI (gel) Mr (cal) pI (cal) PL 1 Hypothetical protein gi|115452789 138 38.0 6,7 39 6.3 PL2 Hypothetical protein gi|115452789 65 39.0 6.6 39 6.3 PL 3 WAPSPADAAAGR Chitinase gi|407472 56 36.0 6.6 35.5 7.3 PL 7 EHGAPQDENR Zinc-superoxide dismutase gi|22296339 26 15.0 6.4 14.7 5.9 PL 11 GPIQLSFNFNYGPAGR Chain A, Crystal Structure Of pdb|2DKV|A 30 37.0 6.2 32.6 5.8 Class I Chitinase PL 13 AAVGHPDTLGDCPFSQR GSH-dependent gi|6939839 43 26.0 6.1 23.5 5.6 dehydroascorbate reductase 1 PL 20 GTSQVEGVVTLTQDDQGP Putative superoxide dismutase gi|42408425 72 17.0 5.5 20.5 5.7 TTVNVR [Cu-Zn] PL 23 L-ascorbate peroxidase 1, P93404 99 28.0 5.5 27 5.4 PL 24 VATPDQAQEVHDGLR Triosephosphate isomerase gi|553107 49 28.0 5.4 27.5 6.6 PL 27 EFSIPLQDSGHVVGFFGR Salt stress-induced protein gi|158513205 88 11.0 5.0 15.1 5.1 PL 30 MIEDYLVAHPAEYA Pathogenesis-related protein Bet gi|9230755 55 18.0 4.9 16.6 4.9 v I PL 33 ADVGVGPVSWDDTVAAY Acidic PR-1 type pathogenesis- gi|12005673 182 17.5 4.2 17.5 4.5 AESYAAQR related protein PR-1 PL34 WWDTFPANVDGAR Hypothetical protein gi|115461070 87 29.0 4.7 27.2 5.0 PL 38 Ascorbate peroxidase NP_001060741 74 32.0 5.2 27 5.2 PL 43 MTAEIGEQVQIVGDDLLVT Enolase gi|780372 88 60.0 5.4 47.9 5.4 NPTR PL 46 Enolase Q42971 74 50.0 5.4 47.9 5.4 PL 45 Hypothetical protein gi|115465323 98 60.0 5.2 58.8 5.9 PL 51 KADATVAGDDR Hypothetical protein gi|125557770 37 45.0 5.7 95.7 8.0 PL 57 AGYAPPHWVQPGQGDR Hypothetical protein gi|125532459| 73 25.0 4.2 24.5 4.6 PL 60 ELFEQLLLHR Chitinase gi|561873 51 36.5 4.2 34.3 4.4 PL 63 ELVADDEWLNTEFISTVQQ Cytosolic malate dehydrogenase gi|115482534 66 37.5 5.9 35.5 5.75 PL 40 EFSIPLQDSGHVVGFFGR Salt stress-induced protein gi|158513205 104 39 5.3 15.1 5.1 Hypothetical protein EAY73933 40.6 8.6 insights into the genomics of rice root adaptive develop- ment. In Rice Genetics V Edited by: Brar DS, Mackill DJ, Hardy B. Sin- References gapore: World Scientific; 2007:117-141. 1. Luo LJ, Zhang QF: The status and strategy on drought resist- 12. Kawasaki S, Borchert C, Deyholos M, Wang H, Brazille S, Kawai K, ance of rice (Oryza sativa L.). Chin J Rice Sci 2001, 15:209-214. Galbraith D, Bohnert HJ: Gene expression profiles during the 2. Liu H, Mei H, Yu X, Zou G, Liu G, Luo L: Towards improving the initial phase of salt stress in rice. The Plant Cell 2001, drought tolerance of rice in China. Plant Genet Resour 2007, 13(4):889-905. 4(01):47-53. 13. Rabbani MA, Maruyama K, Abe H, Khan MA, Katsura K, Ito Y, Yoshi- 3. Price AH, Cairns JE, Horton P, Jones HG, Griffiths H: Linking wara K, Seki M, Shinozaki K, Yamaguchi-Shinozaki K: Monitoring drought-resistance mechanisms to drought avoidance in expression profiles of rice genes under cold, drought, and upland rice using a QTL approach: progress and new oppor- high-salinity stresses and abscisic acid application using tunities to integrate stomatal and mesophyll responses. J Exp cDNA microarray and RNA gel-blot analyses. Plant Physiol Bot 2002, 53(371):989-1004. 2003, 133(4):1755-1767. 4. Gupta PC, O'Toole JC: Upland rice: a global perspective. Los 14. Yamaguchi-Shinozaki K, Shinozaki K: Transcriptional regulatory Baños, Laguna: IRRI; 1986. networks in cellular responses and tolerance to dehydration 5. Pinheiro B da S, de Castro E da M, Guimarães CM: Sustainability and cold stresses. Annu Rev Plant Biol 2006, 57:781-803. and profitability of aerobic rice production in Brazil. Field 15. Salekdeh GH, Siopongco J, Wade LJ, Ghareyazie B, Bennett J: Pro- Crops Res 2006, 97(1):34-42. teomic analysis of rice leaves during drought stress and 6. Heinemann A, Dingkuhn M, Luquet D, Combres J, Chapman S: Char- recovery. Proteomics 2002, 2(9):1131-1145. acterization of drought stress environments for upland rice 16. Salekdeh GH, Siopongco J, Wade LJ, Ghareyazie B, Bennett J: A pro- and maize in central Brazil. Euphytica 2007, 162(3):395-410. teomic approach to analyzing drought- and salt-responsive- 7. Pessoa-Filho M, Beló A, Alcochete AAN, Rangel PHN, Ferreira ME: A ness in rice. Field Crops Res 2002, 76(2–3):199-219. set of multiplex panels of microsatellite markers for rapid 17. Agrawal GK, Jwa NS, Iwahashi Y, Yonekura M, Iwahashi H, Rakwal R: molecular characterization of rice accessions. BMC Plant Biol Rejuvenating rice proteomics: Facts, challenges, and visions. 2007, 7:23-23. Proteomics 2006, 6(20):5549-5576. 8. Tuberosa R, Salvi S: Genomics-based approaches to improve 18. Agrawal GK, Rakwal R: Rice proteomics: A cornerstone for drought tolerance of crops. Trends Plant Sci 2006, 11(8):405-412. cereal food crop proteomes. Mass Spectrom Rev 2006, 9. Lafitte HR, Yongsheng G, Yan S, Li ZK: Whole plant responses, 25(1):1-53. key processes, and adaptation to drought stress: the case of 19. Yang L, Zheng B, Mao C, Qi X, Liu F, Wu P: Analysis of transcripts rice. J Exp Bot 2007, 58(2):169-175. that are differentially expressed in three sectors of the rice 10. Sreenivasulu N, Sopory SK, Kavi Kishor PB: Deciphering the regu- root system under water deficit. Mol Genet Genomics 2004, latory mechanisms of abiotic stress tolerance in plants by 272(4):433-442. genomic approaches. Gene 2007, 388(1–2):1-13. 11. Périn C, Rebouillat J, Brasileiro ACM, Diévart A, Gantet P, Breitler JC, Johnson AAT, Courtois B, Ahmadi N, de Raissac M, et al.: Novel Page 12 of 13 (page number not for citation purposes) BMC Genomics 2008, 9:485 http://www.biomedcentral.com/1471-2164/9/485 20. Fu BY, Xiong JH, Zhu LH, Zhao XQ, Xu HX, Gao YM, Li YS, Xu JL, located in the endoplasmic reticulum of Arabidopsis. Proc Li ZK: Identification of functional candidate genes for drought Natl Acad Sci USA 2000, 97(11):6224-6229. tolerance in rice. Mol Genet Genomics 2007, 278(6):599-609. 43. Pei ZM, Ward JM, Harper JF, Schroeder JI: A novel chloride chan- 21. Gorantla M, Babu PR, Lachagari VBR, Reddy AMM, Wusirika R, Ben- nel in Vicia faba guard cell vacuoles activated by the serine/ netzen JL, Reddy AR: Identification of stress-responsive genes threonine kinase, CDPK. EMBO J 1996, 15(23):6564-6574. in an indica rice (Oryza sativa L.) using ESTs generated from 44. Tähtiharju S, Sangwan V, Monroy AF, Dhindsa RS, Borg M: The drought-stressed seedlings. J Exp Bot 2007, 58(2):253-265. induction of kin genes in cold-acclimating Arabidopsis thal- 22. Wu C-Q, Hu H-H, Zeng Y, Liang D-C, Xie K-B, Zhang J-W, Chu Z- iana. Evidence of a role for calcium. Planta 1997, H, Xiong L-Z: Identification of Novel Stress-responsive Tran- 203(4):442-447. scription Factor Genes in Rice by cDNA Array Analysis. J 45. Urao T, Katagiri T, Mizoguchi T, Yamaguchi-Shinozaki K, Hayashida Integr Plant Biol 2006, 48(10):1216-1224. N, Shinozaki K: Two genes that encode Ca(2+)-dependent pro- 23. Wang H, Zhang H, Gao F, Li J, Li Z: Comparison of gene expres- tein kinases are induced by drought and high-salt stresses in sion between upland and lowland rice cultivars under water Arabidopsis thaliana. Mol Gen Genet 1994, 244(4):331-340. stress using cDNA microarray. Theor Appl Genet 2007, 46. Xiong L, Schumaker KS, Zhu J-K: Cell signaling during cold, 115(8):1109-1126. drought, and salt stress. Plant Cell 2002, 14(Suppl):S165-S183. 24. Wang H, Zhang H, Li Z: Analysis of Gene Expression Profile 47. Chollet R, Vidal J, O'Leary MH: Phosphoenolpyruvate Carboxy- Induced by Water Stress in Upland Rice (Oryza sativa L. var. lase: A ubiquitous, highly regulated enzyme in plants. Annu IRAT109) Seedlings using Subtractive Expressed Sequence Rev Plant Physiol Plant Mol Biol 1996, 47:273-298. Tags Library. J Integr Plant Biol 2007, 49(10):1455-1463. 48. Nimmo GA, Nimmo HG, Hamilton ID, Fewson CA, Wilkins MB: 25. Wu Y, Wang Q, Ma Y, Chu C: Isolation and expression analysis Purification of the phosphorylated night form and dephos- of salt up-regulated ESTs in upland rice using PCR-based phorylated day form of phosphoenolpyruvate carboxylase subtractive suppression hybridization method. Plant Sci 2005, from Bryophyllum fedtschenkoi. Biochem J 1986, 168(3):847-853. 239(1):213-220. 26. Reddy AR, Ramakrishna W, Sekhar AC, Ithal N, Babu PR, Bonaldo 49. Vidal J, Chollet R: Regulatory phosphorylation of C4 PEP car- MF, Soares MB, Bennetzen JL: Novel genes are enriched in nor- boxylase. Trends in Plant Sci 1997, 2(6):230-237. malized cDNA libraries from drought-stressed seedlings of 50. Torres MA, Dangl JL: Functions of the respiratory burst oxidase rice (Oryza sativa L. subsp. indica cv. Nagina 22). Genome in biotic interactions, abiotic stress and development. Curr 2002, 45(1):204-211. Opin Plant Biol 2005, 8(4):397-403. 27. Guimarães CM, Stone LF, Rangel PHN, Ferreira ME, Rodrigues CAP: 51. Kägi JH, Schäffer A: Biochemistry of metallothionein. Biochemis- Resistência à seca: III. Avaliação do sistema radicular do try 1988, 27(23):8509-8515. arroz de terras em condições controladas. In Anais do II Con- 52. Eulgem T, Rushton PJ, Robatzek S, Somssich IE: The WRKY super- gresso da Cadeira Produtiva de Arroz e Reunião Nacional de Pesquisa de family of plant transcription factors. Trends Plant Sci 2000, Arroz Brasília, DF. Embrapa, SPI; 2006. 5(5):199-206. 28. Fischer RA, Maurer R: Drought resistance in spring wheat cul- 53. Ross CA, Liu Y, Shen QJ: The WRKY Gene Family in Rice tivars. I. Grain yield responses. Aust J Agric Res 1978, 29:897-912. (Oryza sativa). J Integr Plant Biol 2007, 49(6):827-842. 29. Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lip- 54. Ulker B, Somssich IE: WRKY transcription factors: from DNA man DJ: Gapped BLAST and PSI-BLAST: a new generation of binding towards biological function. Curr Opin Plant Biol 2004, protein database search programs. Nucleic Acids Res 1997, 7(5):491-498. 25(17):3389-3402. 55. Qiu YP, Jing SJ, Fu J, Li L, Yu DQ: Cloning and analysis of expres- 30. Telles GP, Silva FR: Trimming and clustering sugarcane ESTs. sion profile of 13 WRKY genes in rice. Chin Sci Bull 2004, Genet Mol Biol 2001, 24(1–4):17-23. 49(20):2159-2168. 31. De Mot R, Vanderleyden J: Application of two-dimensional pro- 56. Yu X-M, Griffith M, Wiseman SB: Ethylene Induces Antifreeze tein analysis of strain fingerprinting and mutant analysis of Activity in Winter Rye Leaves. Plant Physiol 2001, Azospirillum species. Can J Microbiol 1989, 35(10):960-967. 126(3):1232-1240. 32. Laemmli UK: Cleavage of structural proteins during the 57. Yu LX, Djebrouni M, Chamberland H, Lafontaine JG, Tabaeizadeh Z: assembly of the head of bacteriophage T4. Nature 1970, Chitinase: Differential induction of gene expression and 227(5259):680-685. enzyme activity by drought stress in the wild (Lycopersicon 33. Blum H, Beier H, Gross HJ: Improved silver staining of plant chilense Dun.) and cultivated (L. esculentum Mill.) tomatoes. proteins, RNA and DNA in polyacrylamide gels. Electrophoresis J Plant Physiol 1998, 153(5–6):745-753. 1987, 8(2):93-99. 58. Liu Q, Xue Q: Molecular Phylogeny, Evolution, and Functional 34. Shevchenko A, Wilm M, Vorm O, Mann M: Mass spectrometric Divergence of the LSD1-Like Gene Family: Inference from sequencing of proteins silver-stained polyacrylamide gels. the Rice Genome. J Mol Evol 2007, 64(3):354-363. Anal Chem 1996, 68:850-858. 59. Moons A, Prinsen E, Bauw G, Montagu MV: Antagonistic effects of 35. Sinclair TR, Ludlow MM: Influence of Soil Water Supply on the abscisic acid and jasmonates on salt stress-inducible tran- Plant Water Balance of Four Tropical Grain Legumes. Funct scripts in rice roots. Plant Cell 1997, 9(12):. Plant Biol 1986, 13(3):329-341. 60. Mittal R, Dubey RS: Behaviour of peroxidases in rice: changes 36. Grant RF, Jackson BS, Kiniry JR, Arkin GF: Water Deficit Timing in enzyme activity and isoforms in relation to salt tolerance. Effects on Yield Components in Maize. Agron J 1989, Plant Physiol Biochem 1991, 29(1):31-40. 81(1):61-65. 61. Yan S, Tang Z, Su W, Sun W: Proteomic analysis of salt stress- 37. Rosenthal WD, Arkin GF, Shouse PJ, Jordan WR: Water Deficit responsive proteins in rice root. Proteomics 2005, 5(1):235-244. Effects on Transpiration and Leaf Growth. Agron J 1987, 62. van der Straeten D, Rodrigues-Pousada RA, Goodman HM, van Mon- 79(6):1019-1026. tagu M: Plant enolase: gene structure, expression, and evolu- 38. Bray EA: Plant responses to water deficit. Trends Plant Sci 1997, tion. Plant Cell 1991, 3:719-735. 2(2):48-54. 63. Lal SK, Lee C, Sachs MM: Differential regulation of enolase dur- 39. Fukai S, Cooper M: Development of drought-resistant cultivars ing anaerobiosis in maize. Plant Physiol 1998, 118(4):1285-1293. using physiomorphological traits in rice. Field Crops Res 1995, 64. Jones AME, Thomas V, Truman B, Lilley K, Mansfield J, Grant M: Spe- 40(2):67-86. cific changes in the Arabidopsis proteome in response to 40. Agarwal S, Grover A: Isolation and transcription profiling of bacterial challenge: differentiating basal and R-gene medi- low-O2 stress-associated cDNA clones from the flooding- ated resistance. Phytochemistry 2004, 65(12):1805-1816. stress-tolerant FR13A rice genotype. Ann Bot 2005, 65. Lian H-L, Yu X, Lane D, Sun W-N, Tang Z-C, Su W-A: Upland rice 96(5):831-844. and lowland rice exhibited different PIP expression under 41. Wilkinson S, Davies WJ: ABA-based chemical signalling: the co- water deficit and ABA treatment. Cell Res 2006, 16(7):651-660. ordination of responses to stress in plants. Plant Cell Environ 2002, 25(2):195-210. 42. Hwang I, Sze H, Harper JF: A calcium-dependent protein kinase can inhibit a calmodulin-stimulated Ca2+ pump (ACA2) Page 13 of 13 (page number not for citation purposes)

Journal

BMC Genomics – Springer Journals

Published: Dec 1, 2008

Keywords: life sciences, general; microarrays; proteomics; animal genetics and genomics; microbial genetics and genomics; plant genetics and genomics

There are no references for this article.