Transcriptional profiling of genes responsive to abscisic acid and gibberellin in rice: phenotyping and comparative analysis between rice and Arabidopsis

Junshi Yazaki1, Zenpei Shimatani2, Akiko Hashimoto2, Yuko Nagata2, Fumiko Fujii1, Keiichi Kojima3, Kohji Suzuki3, Toshiki Taya4, Mio Tonouchi4, Charles Nelson5, Allen Nakagawa5, Yasuhiro Otomo6, Kazuo Murakami6, Kenichi Matsubara6,7, Jun Kawai8, Piero Carninci9, Yoshihide Hayashizaki8,9 and Shoshi Kikuchi1

1 Department of Molecular Genetics, National Institute of Agrobiological Sciences, Tsukuba, Ibaraki 305-8602, Japan
2 Institute of the Society for Techno-innovation of Agriculture, Forestry and Fisheries, Tsukuba, Ibaraki 305-0854, Japan
3 Hitachi Software Engineering, Tokyo 140-0002, Japan
4 Agilent Technologies, Tokyo 190-0033, Japan
5 Agilent Technologies, Palo Alto, California 94304
6 Laboratory of Genome Sequencing and Analysis Group, Foundation for the Advancement of International Science, Tsukuba, Ibaraki 305-0062, Japan
7 Nara Institute of Science and Technology, Ikoma, Nara 630-0101, Japan
8 Laboratory for Genome Exploration Research Group, Institute of Physical and Chemical Research (RIKEN) Genomic Sciences Center, RIKEN Yokohama Institute, Yokohama, Kanagawa 230-0045, Japan
9 Genome Science Laboratory, RIKEN, Hirosawa, Wako, Saitama 351-0198, Japan


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 GRANTS
 REFERENCES
 
We collected and completely sequenced 32,127 full-length complementary DNA clones from Oryza sativa L. ssp. japonica cv. "Nipponbare." Mapping of these clones to genomic DNA revealed ~20,500 transcriptional units (TUs) in the rice genome. For each TU, we selected 60-mers using an algorithm that took into account some DNA conditions such as base composition and sequence complexity. Using in situ synthesis technology, we constructed oligonucleotide arrays with these TUs on glass slides. We targeted RNAs prepared from normally grown rice callus and from callus treated with abscisic acid (ABA) or gibberellin (GA). We identified 200 ABA-responsive and 301 GA-responsive genes, many of which had never before been annotated as ABA or GA responsive in other expression analysis. Comparison of these genes revealed antagonistic regulation of almost all by both hormones; these had previously been annotated as being responsible for protein storage and defense against pathogens. Comparison of the cis-elements of genes responsive to one or antagonistic to both hormones revealed that the antagonistic genes had cis-elements related to ABA and GA responses. The genes responsive to only one hormone were rich in cis-elements that supported ABA and GA responses. In a search for the phenotypes of mutants in which a retrotransposon was inserted in these hormone-responsive genes, we identified phenotypes related to seed formation or plant height, including sterility, vivipary, and dwarfism. In comparison of cis-elements for hormone response genes between rice and Arabidopsis thaliana, we identified cis-elements for dehydration-stress response as Arabidopsis specific and for protein storage as rice specific.

genome-wide expression analysis; germination; dormancy; comparative genomics; cis-element


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 GRANTS
 REFERENCES
 
RICE (Oryza sativa) is an important food crop that is useful as an experimental model because of its small genome size, extensive genetic map, relative ease of transformation, and synteny with other cereal crops. Draft sequences of the O. sativa L. ssp. indica (41) and japonica (8) genomes, obtained by the whole-genome shotgun sequencing method, have been published. The National Institute of Agrobiological Sciences (NIAS) at Tsukuba, Japan, and its collaborators have constructed useful tools for functional genomics through the Rice Genome Project of Japan. These tools consist of 32,127 full-length cDNA (FL-cDNA) clones (18), over 600 expression profiles developed by using an 8,987-cDNA microarray (39), a high-quality genomic sequence that has 99.99% accuracy (5, 32), a genetic map with 3,267 DNA markers (9), rice material for genetic analysis, and about 50,000 transposon insertion lines (24). All the information from these tools can be accessed through the Rice PIPELINE system (40). These tools may be important for improving not only rice but also other cereals, because functionally important sequences are conserved and may be identified by their synteny. Of these tools, the FL-cDNA clones, in particular, are necessary for the identification of exon-intron boundaries and gene-coding regions within genomic sequences and for comprehensive gene function analysis at transcriptional and translational levels. On the basis of the results of our large-scale FL-cDNA analysis (18), we have constructed a monitoring system that uses an oligonucleotide array to monitor gene transcriptional levels and to develop genome-wide functional analysis of rice. The array was composed of 21,938 probes with 60-mer oligonucleotides synthesized at a gene-specific region (2, 13, 34) from 32,127 FL-cDNAs. Mapping of these cDNA clones to genomic DNA revealed that there are about 20,500 transcriptional units (TUs), and clustering of these cDNA clones revealed a unique clone set. Of the ~20,500 TUs located on genomic DNA, single clones on TU are ~14,000 and multiple clones on TU are ~16,000 (~6,500 TUs). About 2,000 clones were unmapped according to incomplete genome sequences. The probes were selected from these results.

We used the arrays as probes to hybridize target RNAs prepared from normally grown callus and from callus treated with abscisic acid (ABA) or gibberellin (GA). The interaction between ABA and GA is an important factor controlling the transition from embryogenesis to germination in seeds. The effects of these hormones are competitive, in that ABA promotes seed dormancy, whereas GA promotes seed germination. Cereals are excellent plants in which to explore the molecular mechanisms involved in hormonally regulated gene expression, particularly the antagonism between ABA and GA (14). Reports of such explorations (10, 22) have suggested the presence of cross talk, but these studies have investigated only part of the relationship between dormancy and germination in plants.

In a previous study we elucidated the mechanisms of interaction between ABA and GA signaling in rice by using the above tools for rice functional genomics, including rice 8,987-cDNA microarray (38). Here we describe a comprehensive transcriptional profiling of phytohormone response genes in rice, phenotype analysis, and comparative analysis of transcriptional regulators between rice and Arabidopsis, for which we used a more comprehensive and specific 21 938-oligonucleotide array.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 GRANTS
 REFERENCES
 
Rice FL-cDNA Clones
All 32,127 FL-cDNA clones were generated by large-scale FL-cDNA analysis as part of the Rice Full-length cDNA Project, which is a collaborative effort of NIAS, Foundation for the Advancement of International Science, and Institute of Physical and Chemical Research (RIKEN) under the supervision of the Bio-oriented Technology Research Advancement Institution (BRAIN) (18). A homology search was performed with BLASTN and BLASTX, and by homology searches of publicly available sequence data we assigned tentative protein functions. We used 21,938 TUs that were equivalent to ~32,000 FL-cDNA clones; to maximize the effective use of these ~32,000 published clones, in accordance with the recommendations of the manufacturer of oligonucleotide array (Agilent Technologies, Tokyo, Japan), the array format was 22,000 spots per glass slide. The functional annotations and identities (including accession numbers and related reference papers) of all ~32,000 rice FL-cDNA clones are listed in the Knowledge-based Oryza Molecular Biological Encyclopedia (KOME; http://cdna01.dna.affrc.go.jp/cDNA/).

Rice High-Density Oligonucleotide Array Construction
The array was composed of 21,938 probes with 60-mer oligonucleotides synthesized at a gene-specific region from 28,469 FL-cDNAs. Mapping of these cDNA clones to genomic DNA revealed that there are ~20,500 TU, and clustering of these cDNA clones revealed a unique clone set. The probes were selected by means of these results. For each of the clones, we selected the top 60-mers by using an algorithm that took into account binding energy, base composition, sequence complexity, cross-hybridization binding energy, and secondary structure (2, 13, 34). We constructed the oligonucleotide arrays with these TUs on glass slides by using in situ synthesis technology under a customized contract with Agilent Technologies (2, 13, 34).

Plant Material and RNA Preparation
The callus used for total RNA extraction was derived from the scutellum of the japonica rice cultivar "Nipponbare" and was cultivated in Murashige and Skoog medium (27) containing 10 µM 2,4-dichlorophenoxyacetic acid. Such callus maintains the ability to develop roots and leaves. After the calluses had been cultured in the medium for 30 days, they were transferred to a medium containing the plant hormones ABA or GA and cultured for 3 days. The concentration of each plant hormone was adjusted to 50 µM. After culturing, we used an RNeasy Plant Mini Kit (Qiagen, Tokyo, Japan) to extract total RNA from the hormone-treated calluses and from the control calluses (which were not treated with either hormone for the 3-day period). mRNA was isolated with an Oligotex-dt30 (Super) mRNA purification kit (Takara, Shiga, Japan). Purified mRNA was amplified, labeled, and hybridized to the rice 22,000-oligonucleotide array according to the manufacturer’s protocols (Agilent Technologies). For each experiment described in this study, the data presented represent averaged results from hybridization to four oligonucleotide arrays in two-dye swap experiments.

Data Scanning, Quantification, and Processing
The hybridized and washed material on each glass slide was scanned with an Agilent DNA microarray scanner (model G2565BA; Agilent Technologies). Feature Extraction and Image Analysis software (version A.6.1.1, Agilent Technologies), was used to locate and delineate every spot in the array and to integrate each spot’s intensity, filtering, and normalization by using the LOWESS (also know as "LOESS") method (36, 37). From the four replications of each experiment we calculated the average ratio of expression of each spot after dividing the signal intensity of mRNA from hormone-treated callus by the signal intensity of that from untreated callus; these data were then subjected to LOWESS normalization. All of the expression profiles are available as gene series 661 (GSE661), including gene sample 9853 (GSM9853), GSM9854, GSM9855, GSM9856, GSM9857, GSM9858, GSM9859, and GSM9860 on gene platform 477 (GPL477) in the Gene Expression Omnibus (GEO) at the National Center for Biotechnology Information (NCBI; http://www.ncbi.nlm.nih.gov/geo/).

Cis-Element Search in Upstream Region of Rice Genes
After we had mapped the FL-cDNA links with the 22,000-oligonucleotide array on the rice genome, we performed a cis-element search. From the 5'-end sequences of the FL-cDNAs, the promoter sequences were obtained by comparison with the rice genomic sequences. We selected 1,000 bp of genomic sequences upstream from the 5' terminus of each FL-cDNA clone by using the data from the TIGR Rice Genome Project (http://www.tigr.org/tdb/e2k1/osa1/BACmapping/description.shtml) and searched for about total 300 cis-elements known in plants by using the PLACE cis-element database (11) (http://www.dna.affrc.go.jp/htdocs/PLACE/).

Specification of Cis-Elements for Hormone-Responsive Genes in Rice
The cis-elements for hormone-responsive genes were compared among the gene groups in comparisons 1 and 2 below, and the types of cis-element of each gene group were characterized.

Comparison 1.
We compared cis-elements among genes that were upregulated by ABA, downregulated by GA, and both upregulated by ABA and downregulated by GA. For each of these three groups of genes, we divided the total number of each type of cis-element in each group by the number of genes in each group. We then compared the numbers of each type of cis-element per gene among the three groups.

Comparison 2.
We compared cis-elements among genes that were upregulated by GA, downregulated by ABA, and both upregulated by GA and downregulated by ABA. By the same method used in comparison 1, we compared the numbers of cis-elements per gene among the three groups. We selected cis-elements that were present in at least two genes, and the cis-elements of each group were characterized.

Cis-Element Search for Genes of Arabidopsis thaliana and Specification of Cis-Element for Rice and A. thaliana
The identifiers (IDs) of the Munich Information Center for Protein Sequences (MIPS ID; http://www.mips.biochem.mpg.de/) for A. thaliana genes corresponding to the rice FL-cDNA obtained were searched by means of KOME. Arabidopsis FL-cDNAs corresponding to the MIPS IDs of Arabidopsis genes obtained were searched using via the data from the RIKEN Arabidopsis Genome Encyclopedia (RARGE; http://rarge.gsc.riken.go.jp/). From the 5'-end sequences of the FL-cDNAs of Arabidopsis (http://cdna01.dna.affrc.go.jp/cDNA/), the promoter sequences were obtained by comparison with the Arabidopsis genomic sequences on RARGE. We compared the numbers of cis-elements between genes that respond to ABA or GA in rice and corresponding genes in Arabidopsis and characterized the types of cis-element in each plant. To characterize cis-elements in each species, we divided the total number of each type of cis-element in each species by the number of genes in each species. We then compared the numbers of each type of cis-element per gene between the two kinds of species.

Quantitative Real-Time PCR
Two micrograms of mRNA from each callus was denatured at 65°C for 5 min and then transferred to a bath at 37°C and incubated for 5 min. At the same time, the first-strand reaction mix containing mouse reverse transcriptase and d(T)18 primer (Amersham Pharmacia, Tokyo, Japan) was held at 37°C for 5 min. The RNA solution was transferred to the first-strand reaction mix for synthesis of cDNA and incubated at 37°C for 60 min. Specific primers for real-time PCR were designed to work under the same experimental conditions (95°C for 10 min followed by 45 cycles of 95°C for 1 min and 60°C for 30 s), generating products of about 200–250 bp at the 3'-untranslated region (3'-UTR) of each expressed sequence tag (EST) clone on the rice 8,987-cDNA microarray (34). Genes encoding heat shock protein (AU062924) and polyubiquitin (AU108666, D22109) were tested as references for the hormonally treated and untreated samples. We chose the polyubiquitin gene (AU108666) for the standard lines because it was amplified by PCR faster than the other two genes were. Quantitative RT-PCR was performed with an iCycler (Bio-Rad, Tsukuba, Japan) and SYBR Green reagent (Qiagen). For each reaction, standard lines for both the treated and untreated samples were made by six fivefold serial dilutions. The relative amounts of the target gene were calculated by comparison with standard lines of the polyubiquitin gene (AU108666).


    RESULTS AND DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 GRANTS
 REFERENCES
 
Genes with Altered Expression Levels in Callus Cultured with ABA for 3 days
The standard deviation (SD) of the [log2(average ratio)] of four expression replications of the experiment was 0.270 after the removal of flagged data, and all the selected genes in Supplemental Table S1 (available at the Physiological Genomics web site)1 were contained within the area of the median (0.00) ±3SD over a normal distribution. The changes in gene expression were contained in the area of the median (0.00) ±3SD over a normal distribution. Of the 200 genes selected, 110 were upregulated and 90 were downregulated by ABA; altered expression levels were identified from the expression profile of the analysis of the four oligonucleotide array analysis in two-dye swap experiments. Sixty-two of the 200 genes had functional annotations in a BLASTN search of the National Center for Biotechnology Information (NCBI; >e-100) (Supplemental Table S1). All of the expression profiles for the four replications are available as GSE661, including GSM9853, GSM9854, GSM9855 and GSM9856, on GPL477 in the NCBI GEO database. The upregulated genes included gene homologs already reported as responding to ABA (see "hormone, ABA" category in Supplemental Table S1) (21). Four clones [gene homologs for aldose-reductase-related protein (AK066733), glucose and ribitol dehydrogenase (AK110652), lipid transfer protein (AK062506), and the gene for globulin-like protein (AK105347)] have been reported as responding not only to ABA but also to GA (34). Gene homologs for lipid transfer protein are not only ABA inducible but also salt and salicylic acid inducible (7). Furthermore, it has been reported that, for genes in the "storage protein" and "stress" categories in Supplemental Table S1, there is cross talk between responses to such factors as low-temperature stress and ABA response (20). The levels of expression of these genes verified the accuracy of the experimental conditions. The gene for PBZ1 (AK071613; "defense" category in Supplemental Table S1), which was upregulated in our experiments, has been reported as a pathogen-related-protein gene (23). Upregulation of genes and gene homologs for development, transcriptional factors, and generation and differentiation in Supplemental Table S1 was also detected with ABA treatment, although there were few compared with the GA-responsive genes (described later). This result suggests that the gene expression profiles in ABA-treated callus look very similar to that in cells in the dormancy process. Downregulated genes included those functionally annotated as responding to ABA ("hormone, ABA" category in Supplemental Table S1) (38). Also, downregulation of a gene homolog for auxin response related to gravitropism (AK101504) was detected with ABA treatment ("hormone, others" category in Supplemental Table S1). Genes and gene homologs that were related to cell division and differentiation of plant cells and that had not previously been reported as ABA responsive, such as senescence-associated protein (AK061848), root cap protein (AK108077), OsNAC3 (AK073667), and MADS-box-like protein (AK111859), were downregulated by ABA treatment ("development", "generation, differentiation", and "transcriptional factor" categories in Supplemental Table S1). Downregulation of these genes by ABA treatment, which demonstrated seed dormancy, are consistent with the fact that plant cells are at a decreased level of growth during seed dormancy. Fourteen defense-related genes and gene homologs were identified as belonging to distinct gene groups among those genes downregulated by ABA treatment that demonstrated seed dormancy ("defense" category in Supplemental Table S1). Eighty genes with unknown functions were upregulated by ABA treatment, and 58 were downregulated by ABA treatment. These unknown genes were newly classified as ABA-responsive genes ("unclassified" category in Supplemental Table S1).

We compared ABA response genes in plants among our result and other results of comprehensive expression analysis (12, 29, 33) using MIPS code and accession ID (Table 1). Among them, 200 (this report), 1,401 (12), 43 (29), and 493 (33) genes are candidates for ABA-inducible genes, respectively. These inconsistencies of genes among these results may be attributed to the biological differences among plant species used, organ/tissues, ABA treatment condition, their response to ABA, and/or detection methodology.


View this table:
[in this window]
[in a new window]
 
Table 1. Comparison of identified ABA-inducible genes in rice with those of Arabidopsis and other report of rice

 
Genes with Altered Expression Levels in Callus Cultured with GA for 3 days
The SD of the [log2(average ratio)] of four replications of the experiment was 0.311 after removal of flagged data, and all the selected genes in Supplemental Table S2 were contained in the area of the median (0.00) ±3SD over a normal distribution. The changes in gene expression were contained within the area of the median (0.00) ±3SD over a normal distribution. Of the 301 genes selected, 206 were upregulated and 95 were downregulated; altered expression levels were identified by expression profile from analysis of the four oligonucleotide arrays in two-dye swap experiments. Of these genes, 103 had functional annotations in a BLASTN search on NCBI (>e-100) (Supplemental Table S2). All the expression profiles for the analysis of the four oligonucleotide arrays in dye swap experiments were available as GSE661, including GSM9857, GSM9858, GSM9859, and GSM9860 on GPL477 in GEO at NCBI (http://www.ncbi.nlm.nih.gov/geo/). The upregulated genes included genes reported as responding to GA ("hormone, GA" category in Supplemental Table S2) (38). Also, upregulation of genes for other hormone responses, including auxin-responsive genes related to response to light (AK059838) and gravitropism (AK101191), was detected after GA treatment ("hormone, others" category in Supplemental Table S2). Thirty-one genes for development, cell division, generation, and differentiation in Supplemental Table S2 that had previously not been reported as GA responsive were identified as being upregulated. Such cell-division-related genes for plant growth were not detected by ABA treatment. Twenty defense-related genes were identified as belonging to distinct functional categories among those genes upregulated by GA treatment ("defense" category in Supplemental Table S2). The ABA and GA treatments revealed genes responsive to various physiological events in addition to their already known response to ABA or GA. In particular, many genes for defense-related proteins were responsive to ABA and GA treatment. In expression analysis using the same oligonucleotide array, genes for defense-related proteins, including peroxidase, were also upregulated in germinating seeds (H. Yamada, NIAS, personal communication). The detection of many kinds of genes for defense-related proteins suggests more strongly than our other report (38) that the ABA and GA response pathways have cross talk with pathogen-related pathways in rice callus. Besides, salicylic acid (SA), a hormone known to mediate disease response, has recently been shown to positively or negatively regulate cell enlargement and division (35), two physiological processes known to be controlled by GA. Therefore, expression profiles of defense-related genes in ABA and GA treatment callus show that may revealed the cross talk of SA, GA and ABA. Genes for stress response, such as the ABA-independent OsDREB1B (AK062422; a drought-inducible gene) and a gene homolog for osmotic-inducible ankyrin kinase (AK100268), were upregulated by GA treatment ("stress" category in Supplemental Table S2). Such genes had not previously been reported as GA responsive, except for metallothionein-related gene homologs (AK062653, AK103445) (38). Some of the downregulated genes included had already been functionally annotated as responsive to GA (38) ("hormone, GA" and "storage protein" categories in Supplemental Table S2). These genes included homologs of genes for hydrophobic LEA-like protein (AK102039) and RAB24 (AK102982), which was reported as not only GA inducible but also antagonistically ABA inducible (upregulated) (38). Eight genes related to development had not previously been reported as GA responsive ("development" category in Supplemental Table S2). The gene homolog for calcium-binding protein, CaBP1, in barley kernels (AK063625), which carries a single calcium-binding EF-hand motif and one transmembrane helix and is upregulated by endogenous ABA (15), was downregulated by our GA treatment. The gene homolog of CaBP1 has shown that calcium-binding protein carrying a single EF-hand motif has Ca2+-binding activity (6). Also, the gene homolog of CaBP1 was upregulated by our ABA treatment in two of the four replicated experiments (data not shown in Supplemental Table S1; see deposited data for GSM9853 and GSM9854 of GSL661 in GEO at NCBI). The data from the other two replications (data not shown in Supplemental Table S1; see GSM9855 and GSM9856 of GSL661) on the DNA spotting on our microarray of the gene homolog were removed as flagged data of saturated signal intensity. The gene for CaBP1 of barley was continuously detected in the vascular tissues in barley kernels during development (15), and the gene homolog was upregulated by our ABA treatment and downregulated by our GA treatment. These facts suggest that the gene homolog’s product might play key roles as a calcium receptor in highly secretory organs during kernel development. In contrast, the gene for calmodulin-binding protein, glutamate decarboxylase (AK061977; OsGAD1), which is expressed in various tissues, including maturing seeds (1), was not only downregulated by our GA treatment but also upregulated by our ABA treatment (Supplemental Table S1). We can speculate that the relationship of this gene’s activity to seed formation is demonstrated by its upregulation by ABA treatment and downregulation by GA treatment. Some genes for stress response, such as gene homologs responsive to low-temperature stress (AK070872, AK073109) and genes responsive to submergence stress (AK058296), which had not previously been reported as GA responsive, were downregulated by GA ("stress" category in Supplemental Table S2). A gene (OsNAC4; AK073848) and gene homologs (OsNAC6; AK068606, AK107746) of the NAC family, related to development of plant tissue, were downregulated by our GA treatment ("transcriptional factor" category in Supplemental Table S2). The NAC family is a group of transcriptional factors expressed in specific tissues such as mature leaf (OsNAC3, 4, 6, 7, and 8), stem (OsNAC4, 6, 7, and 8), root (OsNAC4, 5, 6, 7, and 8), embryos after pollination (OsNAC5, 6, 7, and 8), and callus (OsNAC4, 5, 6, 7, and 8) (17). However, there has been no report of the induction of expression of these genes by GA treatment in tissues at the germination stage. Other homologs of the OsNAC family, including OsNAC3, 5, 7, and 8 on our array, were expressed at almost the same level (average ratio = 1.0384) in GA-treated and untreated callus. (See deposited data, GSE661 including GSM9857 through GSM9860, at the NCBI GEO database.) The gene homolog (AK063685) for the homeodomain leucine zipper protein family, which has a role as a developmental regulator, was newly found to be downregulated by GA. We found 132 genes with unknown functions that were upregulated by GA treatment and 66 that were downregulated. These unknown genes were newly classified as GA responsive.

We performed a comprehensive analysis of gene expression using callus tissue treated with ABA or GA. Genes for embryogenesis, germination, seed dormancy, and grain filling were included among those with altered expression levels. The fact that the experiments identified many already known ABA- and GA-responsive genes in other tissues shows that callus tissue can be mimic the mechanisms of germination and dormancy. The high number of responsive genes detected showed that this analysis, which used our oligonucleotide array, was more efficient than those used in our other report (38) for genome-wide functional analysis of rice.

Genes Responsive to Both Culture with GA and Culture with ABA
To elucidate the interaction between ABA and GA, we selected 68 genes that had shown a response to both hormone treatments (Supplemental Table S3). Sixty-six of the 68 genes showed antagonistic responses to the 2 hormones. Twenty-five of these genes had functional annotations in a BLASTN search at NCBI (>e-100). Two genes were either upregulated or downregulated by both hormones. Of the 66 antagonistically regulated genes, 34 were upregulated by ABA and downregulated by GA, and 32 were downregulated by ABA and upregulated by GA. The former 34 genes included LEA family genes (AK102039, AK102982) previously reported as antagonistically responding to both hormones ("hormone" category in Supplemental Table S3) (38). Although there have been no previous reports of a gene for cysteine protease inhibitor (AK105278) having antagonistic responses to the two hormones, our result was not surprising, in that one would expect ABA to promote such a gene for inhibiting hydrolysis of storage protein and GA to inhibit it. The pattern of expression of a gene for cysteine proteinase (AK106508), which was downregulated by ABA and upregulated by GA, verifies this result. Genes and gene homologs for stress response, including response to low temperature (AK073109, AK070872), have cross talk with the ABA-responsive genes (20) that were detected in both our treatments ("stress" category in Supplemental Table S3). Gene homologs for homocysteine S-methyltransferase-4 (AK073362) and homeodomain leucine zipper protein (AK063685), which have various known functions, including roles as developmental regulators, were assigned new functions related to the ABA and GA response, which modeled seed dormancy and germination. The 32 genes in Supplemental Table S3 that were downregulated by ABA and upregulated by GA included 8 clones of defense-related genes and gene homologs that had not previously been reported as antagonistically responsive (38) ("defense" category in Supplemental Table S3) and gene homologs for development and stress ("development" and "stress" categories in Supplemental Table S3). Detection of such changes in the expression of these defense-related genes and gene homologs by ABA and GA treatment suggests that the plant seed does not require high levels of protection from threats such as pathogens in the external environment, but that at germination the plant is more vulnerable and the defense-related genes are therefore upregulated. The fact that a gene for MADS-box-like protein (AK111859), which functions in the construction of the reproductive organs (30), was regulated antagonistically by ABA and GA suggests that the gene might have a new function related to seed dormancy and germination. Forty-two clones of genes with unknown functions were newly assigned functions as genes antagonistically responsive to ABA and GA treatment. Fifteen genes were reported as representing genes responsive to both hormones in our previous study, which used the rice 8,987-cDNA microarray and quantitative RT-PCR (38). In contrast, we detected 66 genes responsive to both culture with GA and culture with ABA, which is about 4 times as many as in previous reports (38), from an analysis of our oligonucleotide array. We hope that our results will add to the currently incomplete research data on the interaction between the ABA and GA responses in plant dormancy and germination.

Specification of Cis-Elements of ABA- and GA-Responsive Genes
To elucidate the mechanism of transcriptional regulation of the hormone-response systems of plants, we analyzed the cis-elements of genes that were responsive to our hormone treatment. We assigned our FL-cDNAs on the rice genome constructed by bacterial artificial chromosome (BAC) contigs from the results of the TIGR Rice Genome Project (http://www.tigr.org/tdb/e2k1/osa1/BACmapping/description.shtml) and searched the cis-elements up to 1,000 bp upstream from the cDNA from the 5' terminus in the rice genomic sequence. The results of this cis-element analysis of all the clones on our oligonucleotide array are available on the KOME web site. The numbers of all the types of cis-elements of genes responsive to our treatments are shown in Supplemental Tables S4 and S5. The results of comparison 1 (see MATERIALS AND METHODS) among 76 genes upregulated by ABA, 61 genes downregulated by GA, and 34 genes both upregulated by ABA and downregulated by GA are summarized in Table 2A. Two kinds of cis-element for light response (IBOX, BOXIIPCCHS) and seven for protein storage response were remarkably rich in the upstream regions of genes upregulated by ABA and downregulated by GA. Cis-elements for two kinds of storage protein (RAV1BAT, CANBNNAPA) were specified as elements of genes upregulated only by ABA in the comparison. Cis-elements for defense (SEBFCONSSTPR10A), ethylene (ERELEE4), and development (ACGTCBOX) responses were specified as elements of genes downregulated only by GA in the comparison. Cis-elements for dehydration stress response (MYCATRD22) and light response (TBOXATGAPB) were specified as elements of both the gene group upregulated only by ABA and the group downregulated only by GA. The results of comparison 2 (see MATERIALS AND METHODS) among 173 genes upregulated by GA, 57 genes downregulated by ABA, and 32 genes both upregulated by GA and downregulated by ABA are summarized in Table 2B. Two kinds of cis-element for amylase (AMYBOX1, TATCCAOSAMY) were rich in the upstream regions of genes that were both upregulated by GA and downregulated by ABA. Genes downregulated by ABA and genes upregulated by GA were rich in cis-elements for stress response (MYBPLANT, TATABOXOSPAL, upregulated by GA; LTRECOREATCOR15, PALBOXAPC, downregulated by ABA) in the upstream regions. Cis-element for the defense response (SEBFCONSSTPR10A) was specified as an element common to both the gene group upregulated by GA and the group downregulated by ABA. These results suggest that antagonistic-response genes regulated by ABA treatment (Table 2A) or GA treatment (Table 2B) have cis-elements such as RYREPEATVFLEB4 ("protein storage" category in Table 2A) and AMYBOX1 ("amylase" category in Table 2B), which are related to seed dormancy and germination, respectively. In contrast, genes for response to only one of the two hormones are rich in elements such as MYCATRD22 ["stress (dehydration)" category in Table 2A] and SEBFCONSSTPR10A ("defense" category in Table 2B), which promote seed dormancy and germination, respectively.


View this table:
[in this window]
[in a new window]
 
Table 2. Specification of cis-elements of ABA- and GA-responsive genes

 
Cis-elements for stress response were richer in the gene groups of Table 2B than those of Table 2A. There are five elements for stress-response factor (MYBST1, MYBPLANT, TATABOXOSPAL, LTRECOREATCOR15, and PALBOXAPC) in Table 2B. MYBST1, for responding to various stresses, was included in the gene group both upregulated by GA and downregulated by ABA, which demonstrates that it is related to stress response in seed germination. In contrast, one element for dehydration stress response (MYCATRD22) was rich in genes upregulated by ABA and genes downregulated by GA ("stress" category in Table 2A). We speculate that plant cells are particularly susceptible to receive stress from the external environment during germination, or that germination is in fact a phenomenon of self-stress. In addition, these results suggest that plant cells are resistant to receive stress from the external environment during seed dormancy. We compared the cis-elements for tissue-specific response between parts A and B of Table 2. There are nine kinds of cis-element for seed protein storage in Table 2A. Seven kinds of element for protein storage were included in the gene group both upregulated by ABA and downregulated by GA, demonstrating that this group is related to seed dormancy. In contrast, two elements (AMYBOX1, TATCCAOSAMY) for seed germination response were included in the gene group both upregulated by GA and downregulated by ABA (Table 2B) and were thus related to seed germination. The variety of cis-elements for protein storage in the gene group both upregulated by ABA and downregulated by GA (Table 2A), which mimics seed dormancy, is larger than those for amylase in the gene group both upregulated by GA and downregulated by ABA (Table 2B), which mimics germination. We do not know whether the cis-elements of the latter group have multiple functions or whether they are poorly represented on the PLACE database.

In promoter sequences search of the ABA response genes obtained, we detected that several genes (AK064966, AK073100, AK073380, AK073777, AK073833, AK102307, AK103170, AK105316, AK106508, AK108159, AK110259, and AK110912 in Supplemental Table S4) did not contain any ABA-responsive element in their promoters as well as other report (29). These results suggest the existence of novel cis-acting elements involved in ABA-inducible gene expression in their promoters.

Phenotypes of Transposon Insertion Mutants
To elucidate the functions of the genes responsive to our treatments, we investigated the phenotypes of mutants in which the rice retrotransposon Tos17 was inserted in the regions of these responsive genes with the aid of a Tos17 mutant panel database (http://tos.nias.affrc.go.jp/). Tos17, which is highly activated by tissue culture, has been used for insertion mutagenesis of rice (24). The Tos17 mutant panel database enables one to link flanking sequences with phenotype information by using the BLAST program. With this database, one can perform gene function analysis by computer and reverse genetics. We obtained 12 insertion mutants from the 200 ABA-responsive genes (Table 3). Gene homologs for ankyrin kinase (AK100268), protein phosphatase type 2C (AK068272), and zinc transporter protein (AK105258) had functional annotations in a BLASTX search on NCBI (identities >40%) and in published reports. The phenotypes of the insertion mutants of the gene homolog for ankyrin kinase (AK100268) of alfalfa, induced by not only ABA but also by GA, showed dwarfism, brittleness, and lethality. The gene, containing an NH2-terminal region with an ankyrin domain, plays a role in signal transduction in various developmental processes, particularly in legumes at the onset of germination and during the early stage of nodule development (3). The phenotypes of dwarfism, brittleness, and lethality indicate growth delay and suggest that disruption of this gene led to reduced protein synthesis for plant growth using nitrogen because of disruption of nitrogen metabolism pathways. Because, unlike legumes, rice does not form root nodules, we speculate that this gene homolog may play a role in nitrogen metabolism with regard to plant growth dependent on ABA and GA signal transduction. The phenotype of the insertion mutant of the gene homolog for protein phosphatase (AK068272) of Lotus japonicus, which has a function in nodules that are in the process of initiating nitrogen fixation (16), showed sterility. Loss of function of the gene homolog in response to Tos17 insertion might reduce protein storage during seed formation by disruption of a nitrogen metabolism pathway. We speculate that the gene homolog may function partly in nitrogen metabolism with respect to seed formation dependent on ABA signal transduction. The phenotypes of insertion mutants of the gene homolog for zinc transporter protein (AK105258), which plays a housekeeping role in zinc metabolism of soybean nodules (26), showed abnormalities, kernel with a white base, of seed formation in rice. Although of course rice does not form root nodules, the effects of zinc deficiency on nitrogen metabolism of meristematic tissues with regard to protein synthesis in rice have been reported (19). The phenotypes suggest that loss of function of the gene homolog of zinc transporter protein in response to Tos17 insertion might suppress protein synthesis. We speculate that this gene homolog’s product may function farther upstream in the nitrogen metabolism pathway than does protein phosphatase with regard to protein synthesis dependent on ABA signal transduction. We obtained 29 kinds of insertion mutant from the 301 GA-responsive genes (Table 4). Gene homologs for ß-mannosidase (AK068499), cysteine synthase (AK072457), ankyrin kinase (AK100268), and seven-transmembrane protein, Mlo8 (AK072733) had functional annotations in a BLASTX search of NCBI and in published reports. The phenotypes of tomato containing insertion mutants of the gene homolog for ß-mannosidase (AK068499), which is involved in the mobilization of galactomannans in the cell wall of the lateral endosperm in the early stages after germination (25), showed growth delay. Loss of function of the gene homolog in response to Tos17 insertion in rice might reduce the synthesis of proteins for plant growth by disrupting a carbohydrate metabolism pathway; accordingly, the phenotypes included growth delay. We speculate that the gene homolog may play a role in carbohydrate metabolism with regard to plant growth dependent on GA signal transduction. The phenotype of rice carrying the insertion mutant of the gene homolog for cysteine synthase (AK072457), which functions in the regulation of sulfur and nitrogen availability (28), included sterility. Loss of function of the gene homolog in response to Tos17 insertion might reduce protein storage ability during seed formation by disruption of a nitrogen metabolism pathway. We speculate that its protein product may function in nitrogen and sulfur metabolism with regard to seed formation dependent on GA signal transduction. The phenotypes of alfalfa carrying an insertion mutant of the gene homolog for ankyrin kinase (AK100268), induced not only by GA but also by ABA in our experiments, included dwarfism, brittleness, and lethality. We speculate that the gene functions as described at phenotyping of ABA response genes, earlier. The phenotype of rice insertion mutant of the gene homolog for Mlo8 (AK072733), which is involved in defense, response to biotic stress, and leaf senescence (4), included dwarfism and sterility. Loss of function of the gene homolog in response to Tos17 insertion might cause suppression of the cell development in plant height and seed formation, because of reduced stress response.


View this table:
[in this window]
[in a new window]
 
Table 3. Results of a phenotype search of insertion mutants of ABA-responsive genes, using Tos17

 

View this table:
[in this window]
[in a new window]
 
Table 4. Results of a phenotype search of insertion mutants of GA-responsive genes, using Tos17

 
From these phenotypic results and published reports, we suggest that ABA and GA signal transduction interacts with three signal-transduction pathways responsible for nutrient metabolism, cell wall metabolism, and biotic-stress defense. First, the phenotypic results in Tables 3 and 4 suggest that ABA and GA signal transduction interacts with nutrient metabolic pathways, especially in nitrogen metabolism, through gene homologs for ankyrin kinase, protein phosphatase, zinc transporter protein, and cysteine synthase. We can speculate that the gene homolog for ankyrin kinase, which has a transmembrane domain, has a wide effect on plant growth and may function further upstream in the nitrogen metabolism pathway than the products of the other two genes (protein phosphatase, cysteine synthase). And we suggest that protein phosphatase and cysteine synthase have an effect on tissue-specific development (seed formation) and may function farther downstream in the nitrogen metabolism pathway than the ankyrin kinase. However, because these two genes are regulated by different hormones, and one gene homolog (for cysteine synthase) is related to the sulfate metabolism pathway, the nitrogen metabolism pathway based on ABA metabolism might be different from that based on GA metabolism, and the two pathways of nitrogen metabolism may interact on ankyrin kinase (for ABA, ankyrin kinase -> protein phosphatase; for GA, ankyrin-kinase -> cysteine synthase). Also, the zinc transporter, which has a transmembrane domain, has an effect on tissue-specific development (seed formation) and may function farther upstream in the nitrogen metabolism pathway than does protein phosphatase without interaction of ankyrin kinase with a transmembrane domain (zinc transporter -> protein phosphatase).

Second, the results in Table 4 suggest that GA signal transduction interacts with the cell wall metabolic pathway through its effects on mannosidase and invertase. We can speculate that mannosidase is relevant to polysaccharide metabolism, which has a wide effect on plant growth, and may function farther upstream in the cell wall metabolism pathway than invertase, which is relevant to disaccharide metabolism. Third, the results in Table 4 suggest that GA signal transduction interacts with signal transduction of biotic stress through Mlo8. We can speculate that other genes such as ankyrin-kinase (plant height) and cysteine synthase (sterile) downstream of the protein of Mlo8 have a wide effect (plant height and sterility) on plant growth. These results and speculations, derived from phenotype analysis by molecular simulation using genome-wide expression profiles and a huge mutant collection, represent new knowledge of the systematic cataloguing of ABA and GA responses in rice.

However, the phenotypes of transposon insertion mutants in our data are preliminary data from the Tos17 mutant panel database. The studies are underway not only to analyze whether they are a homozygous line or not but also to analyze consistent of the phenotype among different lines of the same gene and to analyze whether the mutant phenotypes are rescuable.

Specification of Cis-Elements of ABA/GA-Responsive Genes in Rice Callus and Arabidopsis Genes Corresponding to Genes for ABA/GA Response of Rice Callus
To characterize the mechanism of transcriptional regulation of the hormone response systems in rice, we analyzed cis-elements of genes of Arabidopsis corresponding to genes for ABA response of rice and compared the types of cis-element in each species. In genes with cis-element more than one element in total number of each type of element, we showed cis-element profile of 116 (rice) and 94 (Arabidopsis) clones in Supplemental Tables S6 and S7, respectively. The results of comparison between 116 clones responsive to ABA in rice and 94 clones of corresponding Arabidopsis genes are summarized in Table 5. From all detected cis-elements, we selected cis-elements for determination of specific elements that were present in at least two genes in either species and specific element were more than ±2-fold in the ratio of (rice/Arabidopsis). Also, we selected cis-elements for determination of common elements that were present in at least one gene in both species.


View this table:
[in this window]
[in a new window]
 
Table 5. Specification of cis-element of Arabidopsis and rice gene for ABA response

 
Six kinds of cis-elements for dehydration-stress response (ACGTATERD1, MYB1AT, ABRELATERD1, MYB2CONSENSUSAT, MYCATERD1, MYCCONSENSUSAT) were specified as elements in Arabidopsis (italic characters in "cis-element" column in Table 5). The specificities of Arabidopsis cis-elements for dehydration stress might be derived from differences in the growth environments of each species. These six kinds of elements do not exist in rice. However, the number of cis-elements for protein storage was remarkably rich in both species ("common" in "specific or common" column in Table 5), and the number of other elements for protein storage (–300ELEMENT, RYREPEATGMGY2, RYREPEATLEGUMINBOX, RYREPEATBNNAPA) were richer in the upstream regions of genes of rice than in those of Arabidopsis. We suggest that Arabidopsis might use these conserved elements for protein storage. The specificities of rice in protein storage might be derived so that structure of rice seed is different from that of Arabidopsis. Also, that the cis-element for expression of the amylase gene (CGACGOSAMY3) was specified as an element in rice may suggest the view that the difference in these elements is derived from differences in organization between rice and Arabidopsis. Although the cis-element MYBCORE for water-stress response appears in both species as a common cis-element in Table 5, the number of elements in rice was 1.34 times that in Arabidopsis. These results suggest that rice might use a slightly different mechanism for response to water stress. Two kinds of ABA-responsive element (RAV1AAT, DPBFCOREDCDC3) and one GA-responsive element (DOFCOREZM) were rich in cis-elements conserved between rice and Arabidopsis ("common" in "specific or common" column in Table 5). We suggest that each species might have a common pathway for ABA or GA metabolism. The numbers of all types of cis-elements of GA-responsive genes in rice and Arabidopsis are shown in Supplemental Tables S8 (rice) and S9 (Arabidopsis). The results of comparison between 151 clones responsive to GA in rice and 117 clones of corresponding Arabidopsis genes are summarized in Supplemental Table S10. The comparison of cis-elements for GA-responsive genes between rice and Arabidopsis gives a result similar to that for ABA. These differences in cis-elements for protein storage and dehydration-stress response between rice and Arabidopsis may have accumulated through differences in the organization of each plant or through evolutionary responses to the growth environment. The comparative analysis of cis-elements among various species for the detection of characteristic in plants may become a useful indicator for the more efficient creation of transgenic plants. The comparison also may support the prediction of the physiological functions of the products of "unknown" genes. In promoter sequences search of the ABA response genes obtained, we detected that several rice genes (AK064966, AK073100, AK073380, AK073777, AK073833, AK102307, AK103170, AK105316, AK106508, AK108159, AK110259, and AK110912 in Supplemental Table S6) did not contain any ABA-responsive element in their promoters on rice genome as well as Arabidopsis report (33). These results suggest the existence of novel cis-acting elements involved in ABA-inducible gene expression in their promoters of rice and Arabidopsis. In our study, WRKY transcription factor (AK073100, AK110912 in Supplemental Table S6), a protein known to mediate the pathogen-induced defense program, were induced by ABA treatment in rice callus. The data suggest that these not only reveal the cross talk between ABA response pathway and pathogen-induced defense program but also support the suggestion that W-box binding motif sequences, most of the WRKY proteins showed to bind to the sequence, are a novel cis-acting element involved in ABA-inducible gene expression (33).

Comparison of ABA-Responsive Genes Among 60-mer Oligonucleotide Array, cDNA Array, and Quantitative RT-PCR
We previously identified a set of transcripts that were more abundant in ABA-treated callus than in untreated callus by using rice 8,987-cDNA microarray (38). To assess the utility of our oligonucleotide array in a practical context (i.e., are the same genes identified?), we compared expression ratios determined by cDNA microarray and oligonucleotide array for ABA-responsive transcripts (Table 6).


View this table:
[in this window]
[in a new window]
 
Table 6. Comparison of cDNA array, 60-mer oligo array and Q-RT-PCR relative expression results for ABA responsive transcripts

 
The sequences of EST clones on rice 8,987-cDNA microarray were searched for in the rice FL-cDNA database, KOME (http://cdna01.dna.affrc.go.jp/cDNA/), which includes a collection of about 32,000 unique FL-cDNAs. We obtained 36 FL-cDNAs linking with the oligonucleotide array corresponding to the 40 EST clones selected as responsive to ABA treatment in the cDNA microarray analysis (38) by BLAST search (>e-10).

Of 40 transcripts with significant expression differences of median ± 2SD over a normal distribution in cDNA microarray analysis, 26 (65%) were positively correlated at the level of expression differences of median ± 2SD over a normal distribution in the oligonucleotide array analysis. Twenty (50%) clones in the results of the cDNA microarray analysis showed an ABA-responsive:control ratio >2.0 in the oligonucleotide array system. Quantitative RT-PCR measurements of 35 transcripts using a specific primer designed from EST clones on the cDNA microarray showed agreement of 30 (86%) with the cDNA array and 26 (74%) with the oligonucleotide array. PCR-based ratios agreed slightly better with the cDNA microarray than with the oligonucleotide array because of the use of the primer designed from the EST clones.

Conclusions
On the basis of our huge FL-cDNA collection, we constructed a monitoring system that uses the 22,000-cDNA oligonucleotide array to monitor gene transcription levels and expanded the genome-wide functional analysis of rice. We constructed a comprehensive profile of gene expression for ABA and GA response in rice. We identified genes that had never before been annotated as ABA or GA responsive in other reports of comprehensive expression analysis of ABA-responsive genes in plant (12, 29, 33), detected new interactions between genes responsive to the two hormones, comprehensively characterized cis-elements of hormone-responsive genes, obtained new putative gene functions from phenotypic results, and characterized cis-elements of rice and Arabidopsis. The results revealed that our tools and methods of functional genomics can identify genes that control particular phenotypes and can identify the transcriptional regulator (cis-element) of rice faster and more accurately than ever before. For the most effective functional analysis of the genome, all available information needs to be integrated. Systematically connecting powerful tools and information for functional genomics of rice will allow researchers in the life sciences, especially crop science, to change the direction of research. All of the information on rice functional genomics used in this report can be accessed through the Rice PIPELINE (http://cdna01.dna.affrc.go.jp/PIPE/) (40), including the databases KOME, PLACE, Tos17, Integrated Rice Genome Explorer (31) (INE: http://rgp.dna.affrc.go.jp/giot/INE.html), and the Rice Expression Database (RED; http://red.dna.affrc.go.jp/RED/) (39). Details of all of our materials, including FL-cDNA and mutant lines, can be obtained from the Rice Genome Resource Center (http://www.rgrc.dna.affrc.go.jp).


    GRANTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 GRANTS
 REFERENCES
 
This work was supported by a grant from the Ministry of Agriculture, Forestry, and Fisheries of Japan (Rice Genome Project MA-1000).


    ACKNOWLEDGMENTS
 
We thank Dr. Hisako Ooka, Dr. Toshifumi Nagata, Dr. Hitomi Yamada, and Masaki Shimono (NIAS) for helpful comments. We also thank Kanako Shimbo, Yumiko Yoshida (Institute of the Society for Techno-innovation of Agriculture, Forestry and Fisheries), Dr. Naoki Kishimoto, Sachiko Honda, Ayano Endo, Yuki Sato, Chikako Miyamoto, Kazuko Toyoshima, and Keiko Takeuchi (NIAS) for helpful support. We also thank Chao Jie Li and Makoto Yamamoto (Hitachi Software Engineering Co. Ltd.) for technical support.


    FOOTNOTES
 
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).

Address for reprint requests and other correspondence: S. Kikuchi, National Institute of Agrobiological Sciences, 2-1-2, Kannondai, Tsukuba, Ibaraki 305-8602, Japan (E-mail: skikuchi{at}nias.affrc.go.jp).

10.1152/physiolgenomics.00201.2003

1 The Supplementary Material for this article (Supplemental Tables S1 through S10) is available online at http://physiolgenomics.physiology.org/cgi/content/full/00201.2003/DC1. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 GRANTS
 REFERENCES
 

  1. Akama K, Akihiro T, Kitagawa M, and Takaiwa F. Rice (Oryza sativa) contains a novel isoform of glutamate decarboxylase that lacks an authentic calmodulin-binding domain at the C-terminus. Biochim Biophys Acta 1522: 143–150, 2001.[ISI][Medline]
  2. Carter MG, Hamatani T, Sharov AA, Carmack CE, Qian Y, Aiba K, Ko NT, Dudekula DB, Brzoska PM, Hwang SS, and Ko MSH. In situ-synthesized novel microarray optimized for mouse stem cell and early developmental expression profiling. Genome Res 13: 1011–1021, 2003.[Abstract/Free Full Text]
  3. Chinchilla D, Merchan F, Megias M, Kondorosi A, and Sousa C. Ankyrin protein kinases: a novel type of plant kinase gene whose expression is induced by osmotic stress in alfalfa. Plant Mol Biol 51: 555–566, 2003.[CrossRef][ISI][Medline]
  4. Devoto A, Hartmann HA, Piffanelli P, Elliott C, Simmons C, Taramino G, Goh CS, Cohen EF, Emerson CB, Schulze-Lefert P, and Panstruga R. Molecular phylogeny and evolution of the plant-specific seven-transmembrane MLO family. J Mol Evol 56: 77–88, 2003.[CrossRef][ISI][Medline]
  5. Feng Q, Zhang Y, Hao P, Wang S, Fu G, Huang Y, Li Y, Zhu J, Liu Y, Hu X, Jia P, Zhang Y, Zhao Q, Ying K, Yu S, Tang Y, Weng Q, Zhang L, Lu Y, Mu J, Lu Y, Zhang LS, Yu Z, Fan D, Liu X, Lu T, Li C, Wu Y, Sun T, Lei H, Li T, Hu H, Guan J, Wu M, Zhang R, Zhou B, Chen Z, Chen L, Jin Z, Wang R, Yin H, Cai Z, Ren S, Lv G, Gu W, Zhu G, Tu Y, Jia J, Zhang Y, Chen J, Kang H, Chen X, Shao C, Sun Y, Hu Q, Zhang X, Zhang W, Wang L, Ding C, Sheng H, Gu J, Chen S, Ni L, Zhu F, Chen W, Lan L, Lai Y, Cheng Z, Gu M, Jiang J, Li J, Hong G, Xue Y, and Han B. Sequence and analysis of rice chromosome 4. Nature 420: 316–320, 2002.[CrossRef][ISI][Medline]
  6. Frandsen G, Muller-Uri F, Nielsen M, Mundy J, and Skriver K. Novel plant Ca2+-binding protein expressed in response to abscisic acid and osmotic stress. J Biol Chem 271: 343–348, 1996.[Abstract/Free Full Text]
  7. Garcia-Garrido JM, Menossi M, Puigdomenech P, Martinez-Izquierdo JA, and Delseny M. Characterization of a gene encoding an abscisic acid-inducible type-2 lipid transfer protein from rice. FEBS Lett 428: 193–199, 1998.[CrossRef][ISI][Medline]
  8. Goff SA, Ricke D, Lan TH, Presting G, Wang R, Dunn M, Glazebrook J, Sessions A, Oeller P, Varma H, Hadley D, Hutchison D, Martin C, Katagiri F, Lange BM, Moughamer T, Xia Y, Budworth P, Zhong J, Miguel T, Paszkowski U, Zhang S, Colbert M, Sun Wl Chen L, Cooper B, Park S, Wood TC, Mao L, Quail P, Wing R, Dean R, Yu Y, Zharkikh A, Shen R, Sahasrabudhe S, Thomas A, Cannings R, Gutin A, Pruss D, Reid J, Tavtigian S, Mitchell J, Eldredge G, Scholl T, Miller RM, Bhatnagar S, Adey N, Rubano T, Tusneem N, Robinson R, Feldhaus J, Macalma T, Oliphant A, and Briggs S. A draft sequence of the rice genome (Oryza sativa L. ssp. japonica). Science 296: 92–100, 2002.[Abstract/Free Full Text]
  9. Harushima Y, Yano M, Shomura A, Sato M, Shimano T, Kuboki Y, Yamamoto T, Lin SY, Antonio BA, Parco A, Kajiya H, Huang N, Yamamoto K, Nagamura Y, Kurata N, Khush GS, and Sasaki T. A high-density rice genetic linkage map with 2275 markers using a single F2 population. Genetics 148: 479–494, 1998.[Abstract/Free Full Text]
  10. Heck GR, Chamberlain AK, and Ho DTH. Barley embryo globulin 1 gene, Beg1: characterization of cDNA, chromosome mapping and regulation of expression. Mol Gen Genet 239: 209–218, 1993.[ISI][Medline]
  11. Higo K, Ugawa Y, Iwamoto M, and Higo H. PLACE: a database of plant cis-acting regulatory DNA elements. Nucleic Acids Res 26: 358–359, 1998.[Abstract/Free Full Text]
  12. Hoth S, Morgante M, Sanchez JP, Hanafey MK, Tingey SV, and Chua NH. Genome-wide gene expression profiling in Arabidopsis thaliana reveals new targets of abscisic acid and largely impaired gene regulation in the abi1–1 mutant. J Cell Sci 115: 4891–4900, 2002.[CrossRef][ISI][Medline]
  13. Hughes TR, Mao M, Jones AR, Burchard J, Marton MJ, Shannon KW, Lefkowitz SM, Ziman M, Schelter JM, Meyer MR, Kobayashi S, Davis C, Dai H, He YD, Stephaniants SB, Cavet G, Walker WL, West A, Coffey E, Shoemaker DD, Stoughton R, Blanchard AP, Friend SH, and Linsley PS. Expression profiling using microarrays fabricated by an ink-jet oligonucleotide synthesizer. Nat Biotechnol 19: 342–347, 2001.[CrossRef][ISI][Medline]
  14. Jacobsen SE and Olszewski NE. Gibberellins regulate the abundance of RNAs with sequence similarity to proteinase inhibitors, dioxygenases and dehydrogenases. Planta 198: 78–86, 2003.
  15. Jang CS, Lee MS, Kim JY, and Seo YW. Molecular characterization of a cDNA encoding putative calcium binding protein, HvCaBP1, induced during kernel development in barley (Hordeum vulgare L.). Plant Cell Rep 22: 64–70, 2003.[CrossRef][ISI][Medline]
  16. Kapranov P, Jensen TJ, Poulsen C, de Bruijn FJ, and Szczyglowski K. A protein phosphatase 2C gene, LjNPP2C1, from Lotus japonicus induced during root nodule development. Proc Natl Acad Sci USA 96: 1738–1743, 1999.[Abstract/Free Full Text]
  17. Kikuchi K, Ueguchi-Tanaka M, Yoshida KT, Nagato Y, Matsusoka M, and Hirano HY. Molecular analysis of the NAC gene family in rice. Mol Gen Genet 262: 1047–1051, 2000.[CrossRef][ISI][Medline]
  18. Kikuchi S et al. (Rice Full-Length cDNA Consortium). Collection, mapping, and annotation of over 28,000 cDNA clones from japonica rice. Science 301: 376–379, 2003.[Abstract/Free Full Text]
  19. Kitagishi K and Obata H. Effects of zinc deficiency on the nitrogen metabolism of meristematic tissues of rice plant with reference to protein synthesis. Soil Sci Plant Nutr 32: 397–406, 1986.[ISI]
  20. Leung J and Giraudat J. Abscisic acid signal transduction. Annu Rev Plant Physiol Plant Mol Biol 49: 199–222, 1998.[CrossRef][ISI]
  21. Li B and Foley ME. Cloning and characterization of differentially expressed genes in imbibed dormant and afterripened Avena fatua embryos. Plant Mol Biol 29: 823–831, 1995.[ISI][Medline]
  22. Lovegrove A and Hooley R. Gibberellin and abscisic acid signalling in aleurone. Trends Plant Sci 5: 102–110, 2000.[CrossRef][ISI][Medline]
  23. Midoh N and Iwata M. Cloning and characterization of a probenazole-inducible gene for an intracellular pathogenesis-related protein in rice. Plant Cell Physiol 37: 9–18, 1996.[ISI][Medline]
  24. Miyao A, Tanaka K, Murata K, Sawaki H, Takeda S, Abe K, Shinozuka Y, Onosato K, and Hirochika H. Target site specificity of the Tos17 retrotransposon shows a preference for insertion within genes and against insertion in retrotransposon-rich regions of the genome. Plant Cell 15: 1771–1780, 2003.[Abstract/Free Full Text]
  25. Mo B and Bewley JD. Beta-mannosidase (EC 3.2.1.25) activity during and following germination of tomato (Lycopersicon esculentum Mill.) seeds. Purification, cloning and characterization. Planta 215: 141–152, 2002.[CrossRef][ISI][Medline]
  26. Moreau S, Thomson RM, Kaiser BN, Trevaskis B, Guerinot ML, Udvardi MK, Puppo A, and Day DA. GmZIP1 encodes a symbiosis-specific zinc transporter in soybean. J Biol Chem 277: 4738–4746, 2002.[Abstract/Free Full Text]
  27. Murashige T and Skoog F. A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiol Plant 15: 473–497, 1962.[ISI]
  28. Nakamura T, Yamaguchi Y, and Sano H. Four rice genes encoding cysteine synthase: isolation and differential responses to sulfur, nitrogen and light. Gene 229: 155–161, 1999.[CrossRef][ISI][Medline]
  29. Rabbani MA, Maruyama K, Abe H, Khan MA, Katsura K, Ito Y, Yoshiwara K, Seki M, Shinozaki K, and Yamaguchi-Shinozaki K. Monitoring expression profiles of rice genes under cold, drought, and high-salinity stresses and abscisic acid application using cDNA microarray and RNA gel-blot analyses. Plant Physiol 133: 1755–1767, 2003.[Abstract/Free Full Text]
  30. Riechmann JL and Meyerowitz EM. MADS domain proteins in plant development. Biol Chem 378: 1079–1101, 1997.[ISI][Medline]
  31. Sakata K, Antonio BA, Mukai Y, Nagasaki H, Sakai Y, Makino K, and Sasaki T. INE: a rice genome database with an integrated map view. Nucleic Acids Res 28: 97–101, 2000.[Abstract/Free Full Text]
  32. Sasaki T, Matsumoto T, Yamamoto K, Sakata K, Baba T, Katayose Y, Wu J, Niimura Y, Cheng Z, Nagamura Y, Antonio BA, Kanamori H, Hosokawa S, Masukawa M, Arikawa K, Chiden Y, Hayashi M, Okamoto M, Ando T, Aoki H, Arita K, Hamada M, Harada C, Hijishita S, Honda M, Ichikawa Y, Idonuma A, Iijima M, Ikeda M, Ikeno M, Ito S, Ito T, Ito Y, Ito Y, Iwabuchi A, Kamiya K, Karasawa W, Katagiri S, Kikuta A, Kobayashi N, Kono I, Machita K, Maehara T, Mizuno H, Mizubayashi T, Mukai Y, Nagasaki H, Nakashima M, Nakama Y, Nakamichi Y, Nakamura M, Namiki N, Negishi M, Ohta I, Ono N, Saji S, Sakai K, Shibata M, Shimokawa T, Shomura A, Song J, Takazaki Y, Terasawa K, Tsuji K, Waki K, Yamagata H, Yamane H, Yoshiki S, Yoshihara R, Yukawa K, Zhong H, Iwama H, Endo T, Ito H, Hahn JH, Kim HI, Eun MY, Yano M, Jiang J, and Gojobori T. The genome sequence and structure of rice chromosome 1. Nature 420: 312–316, 2002.[CrossRef][ISI][Medline]
  33. Seki M, Ishida J, Narusaka M, Fujita M, Nanjo T, Umezawa T, Kamiya A, Nakajima M, Enju A, Sakurai T, Satou M, Akiyama K, Yamaguchi-Shinozaki K, Carninchi PKJ, Hayashizaki Y, and Shinozaki K. Monitoring the expression pattern of around 7,000 Arabidopsis genes under ABA treatments using a full-length cDNA microarray. Funct Integr Genomics 2: 282–291, 2002.[CrossRef][Medline]
  34. Shoemaker DD, Schadt EE, Armour CD, He YD, Garrett-Engele P, McDonagh PD, Loerch PM, and Leonardson A. Experimental annotation of the human genome using microarray technology. Nature 409: 922–927, 2001.[CrossRef][ISI][Medline]
  35. Vanacker H, Lu H, Rate DN, and Greenberg JT. A role for salicylic acid and NPR1 in regulating cell growth in Arabidopsis. Plant J 28: 209–216, 2001.[CrossRef][ISI][Medline]
  36. Yang IV, Chen E, Hasseman JP, Liang W, Frank BC, Wang Sharov V, Saeed AI, White J, Li J, Lee NH, Yeatman TJ, and Quackenbush J. Within the fold: assessing differential expression measures and reproducibility in microarray assays. Genome Biol 3: research0062. 1, 2002.
  37. Yang YH, Dudoit S, Luu P, Lin DM, Peng V, Ngai J, and Speed TP. Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation. Nucleic Acids Res 30: e15, 2002.[Abstract/Free Full Text]
  38. Yazaki J, Kishimoto N, Fujii F, Nagata Y, Hashimoto A, Shimbo K, Shimatani Z, Kojima K, Kouji S, Makoto Y, Honda S, Endo A, Yoshida Y, Sato Y, Takeuchi K, Toyoshima K, Miyamoto C, Wu J, Sasaki T, Sakata K, Yamamoto K, Iba K, Oda T, Ootomo Y, Murakami K, Matsubara K, Kawai J, Carninci P, Hayashizaki Y, and Kikuchi S. Genomics approach to abscisic-acid and gibberellin-responsive genes in rice. DNA Res 10: 249–261, 2003.[ISI][Medline]
  39. Yazaki J, Kishimoto N, Ishikawa M, and Kikuchi S. Rice Expression Database: the gateway to rice functional genomics. Trends Plant Sci 7: 563–564, 2002.[CrossRef][ISI]
  40. Yazaki J, Kojima K, Suzuki K, Kishimoto N, and Kikuchi S. The Rice PIPELINE: a unification tool for plant functional genomics. Nucleic Acids Res 1: D383–D387, 2004.[CrossRef]
  41. Yu J et al. (Beijing Genomics Institute). A draft sequence of the rice genome (Oryza sativa L. ssp indica). Science 296: 79–92, 2002.[Abstract/Free Full Text]