Skip to main content

Genome-wide identification and analysis of Japonica and Indica cultivar-preferred transcripts in rice using 983 Affymetrix array data

Abstract

Background

Accumulation of genome-wide transcriptome data provides new insight on a genomic scale which cannot be gained by analyses of individual data. The majority of rice (O. sativa) species are japonica and indica cultivars. Genome-wide identification of genes differentially expressed between japonica and indica cultivars will be very useful in understanding the domestication and evolution of rice species.

Results

In this study, we analyzed 983 of the 1866 entries in the Affymetrix array data in the public database: 595 generated from indica and 388 from japonica rice cultivars. To discover differentially expressed genes in each cultivar, we performed significance analysis of microarrays for normalized data, and identified 490 genes preferentially expressed in japonica and 104 genes in indica. Gene Ontology analyses revealed that defense response-related genes are significantly enriched in both cultivars, indicating that japonica and indica might be under strong selection pressure for these traits during domestication. In addition, 36 (34.6%) of 104 genes preferentially expressed in indica and 256 (52.2%) of 490 genes preferentially expressed in japonica were annotated as genes of unknown function. Biotic stress overview in the MapMan toolkit revealed key elements of the signaling pathway for defense response in japonica or indica eQTLs.

Conclusions

The percentage of screened genes preferentially expressed in indica was 4-fold higher (34.6%) and that in japonica was 5-fold (52.2%) higher than expected (11.1%), suggesting that genes of unknown function are responsible for the novel traits that distinguish japonica and indica cultivars. The identification of 10 functionally characterized genes expressed preferentially in either japonica or indica highlights the significance of our candidate genes during the domestication of rice species. Functional analysis of the roles of individual components of stress-mediated signaling pathways will shed light on potential molecular mechanisms to improve disease resistance in rice.

Background

Oryza sativa, Asian cultivated rice, is grown all over the world (Khush 1997). Japonica and indica are representative subspecies of O. sativa. Japonica and indica rice evolved from different ancestors and diverged about 0.2 ~ 0.44 million years ago (Sang and Ge 2007; Wei et al. 2012). Genome-wide analysis to elucidate the differences between japonica and indica will be useful to explain the evolutionary events that led to their distinct features. During cultivation, these subspecies have developed unique morphologies and characteristic agronomic traits. Although several studies have tried to explain the differences between japonica and indica at a certain developmental stage or under experimental conditions, data from these studies are quite limited in their ability to explain general differences between japonica and indica. For example, transcriptome analysis of the light response in leaf tissues from 3 japonica varieties (Nipponbare, TP309, and Kitaake) and an indica variety (IR64) revealed that about 10% of light-responsive rice genes differed between subspecies (Jung et al. 2008b). Affymetrix microarrays were used to compare 93–11 (indica) and Nipponbare (japonica) seedlings and identify gene expression-level polymorphisms between them (Liu et al. (2010). RNA-seq data analyses recently revealed genome-wide transcriptome data for 93–11 (indica)/Nipponbare (japonica) or Gla4 (indica)/Nipponbare (japonica) seedlings (Lu et al. 2010). A significant amount of expression data is also available for the Agilent platform for both indica and japonica. The RiceXPro database in particular provides a large collection of expression data on japonica (Sato et al. 2013). This database provides diverse views for expression analysis of rice genes in terms of anatomy, development, diurnal regulation, hormone response and laser-captured root cell types (Sato et al. 2013).

Expression quantitative trait loci (eQTLs) regulate expression of mRNA or protein and originate from transcript-level polymorphisms of a single gene with a specific chromosomal location. Recently, tb1 and tag1, QTLs with key roles in maize domestication, have been cloned and analyzed (Tsiantis 2011; Kim et al. 2012). An exciting finding was that transcription factors are cloned in both QTLs. The tb1 QTL has variable promoter sequences and expression levels in maize and teosinte, while a single amino acid difference in tga1 explains the phenotypic variations in kernel traits. Thus, differential gene expression and genetic sequences have been selected during domestication. High-throughput array-based methods to measure mRNA abundance have enabled the identification of numerous eQTLs in plants, animals, and humans. Microarray analysis of 110 recombinant inbred lines (RILs) derived from a cross between Zhenshan 97 and Minghui 63 revealed 26,051 eQTLs in rice shoots 72 hours after germination (Wang et al. 2010). The study revealed correlations between QTLs for shoot dry weight and eQTLs, indicating possible candidate genes for the trait. More interestingly, chemotherapeutic drug susceptibility-associated SNPs are more likely to be eQTLs than a random set of SNPs in the genome, suggesting that eQTLs could also explain phenotypic differences between japonica and indica. Since the whole rice genome was sequenced, rice researchers have made efforts to characterize individual genes. The O verview of Functionally Characterized G enes in R ice O nline database (OGRO) (http://qtaro.abr.affrc.go.jp/ogro) provides information on 702 functionally-characterized genes (Yamamoto et al. 2012). This information will be very useful to evaluate the significance of candidate genes identified from high-throughput analysis for the discovery or enhancement of agronomic traits.

In this study, we collected 983 Affymetrix microarray data entries from the NCBI gene expression omnibus or ArrayExpress and normalized all slides by the same normalization method. The data were then sorted by cultivar—595 from indica and 388 from japonica. Using significant microarray analysis (SAM) in the TIGR multi experiment viewer (MEV) toolkit, we identified 104 genes with preferential expression in indica (indica eQTLs) and 490 genes preferentially expressed in japonica (japonica eQTLs). Here, we present the identification and analyses of these eQTLs.

Results and discussion

Japonica or indica eQTLs identified from rice Affymetrix microarray data

To identify eQTLs between japonica and indica, we collected Affymetrix microarray data from the NCBI gene expression omnibus (NCBI GEO, http://www.ncbi.nlm.nih.gov/geo/) or ArrayExpress (http://www.ebi.ac.uk/arrayexpress/) (Barrett et al. 2011; Parkinson et al. 2011). The total number of Affymetrix array entries was 983, as described in Additional file 1: Table S1; 388 entries were derived from japonica, and 595 were derived from indica. Array data associated with other species and those for which the subspecies were not identified were excluded. We normalized whole microarray data with the Affy package encoded using the R language (Gautier et al. 2004) and sorted by subspecies. We identified 4,685 probes with at least 2-fold differences in expression between japonica and indica. After performing significance analysis of microarrays (SAM) on the data installed in MultiExperiment Viewer (MEV) (Saeed et al. 2006), we identified 699 probes with preferential expression in japonica and 118 probes with preferential expression in indica. This resulted in 490 japonica eQTLs from 609 probes and 104 indica eQTLs from 118 probes. The number of eQTLs is less than that of corresponding probes because multiple probes target a single locus and some probes are unmapped to the chromosome. Therefore, we present expression profiles for the 490 japonica eQTLs and 104 indica eQTLs (Figure 1). The probes on the Affymetrix array platform are largely based on the Nipponbare genome sequence; thus, mRNAs from japonica might have higher affinity for the probes on this array platform. This could introduce bias in favor of japonica eQTLs. RNA-seq based on next-generation sequencing technology is expected to overcome the fixed-genome limitations of microarray technology. The expression patterns of japonica and indica samples were compared in 15 major categories of anatomical samples collected from 983 affymetrix arrays (Figure 1). Most candidate genes were differentially regulated between japonica and indica samples. Detailed information about the samples used in this figure is shown in Additional file 1: Table S1. In addition, we prepared the mapping data of 490 genes and 104 genes that are preferentially expressed in japonica and in indica, respectively, onto the 12 rice chromosomes (Additional file 2: Figure S1). Before carrying out further functional analysis, we confirmed the expression of 10 japonica and 7 indica eQTLs by reverse transcriptase (RT)-PCR (Additional file 3: Figure S2).

Figure 1
figure 1

Expression patterns of japonica and indica eQTLs identified using 983 affymetrix array data points. From the rice Affymetrix microarray data available in the NCBI gene expression omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/), we used 983 arrays consisting of 595 indica and 388 japonica samples. Using a 2-fold change cut-off of japonica over indica expression and SAM analysis, we identified 490 loci showing preferential expression in japonica and 104 loci in indica. Blue color indicates low level of log2 intensity and yellow color indicates high level of log2 intensity. Green bar indicates indica samples and gray bar indicates japonica samples.

Recently, comparative analysis of genome expression in the heading-stage panicle from the six cultivated and wild rice lineages Oryza sativa subsp. indica, japonica and javanica, O. nivara, O. rufipogon and O. glaberrima was carried out. 5,116 genes differentially expressed in the heading-stage panicle of japonica and indica were identified (Peng et al. 2009). The large difference in the number of candidate genes identified in this analysis and in ours might come from differences in the range of analyzed samples and statistical criteria: we used 388 japonica and 595 indica samples, while Peng et al. (2009) used two biological samples prepared from the heading-stage panicle; we used SAM installed in MEV software, while Peng et al. (2009) used p-value > 0.7. Liu et al. (2010) compared 93–11 (indica) and Nipponbare (japonica) seedlings using methyl viologen (MV) as a reactive oxygen species agent in affymetrix microarrays. 232 probesets (213 genes) were identified with gene expression level polymorphisms between the two rice cultivars regardless of MV treatment through analysis using the supplemental data. Of these, we identified 55 probes (51 genes) showing more than 4-fold upregulation in indica compared to japonica, while 177 probes (162 genes) were more than 4-fold upregulated in japonica compared to indica (Additional file 4: Table S3). In this study, the number of genes preferentially expressed in japonica in the seedling stage is 3-fold more than the number of genes preferentially express in indica, identifying again the bias in favor of japonica-preferred genes. Of 162 genes preferentially expressed in japonica identified by Liu et al. (2010), 41 were also more than 4-fold upregulated in japonica samples when compared to indica samples from our analysis, while 5 of 51 genes preferentially expressed in indica had similar feature in our analysis (Additional file 4: Table S3). This data indicates that data on differential expression in japonica might be more stable than those in indica. In total, the differential expression patterns identified in the two cultivars in Liu et al. (2010) were much less than half repeated in our analysis, and the remaining significant genes might retain developmental stage-specific or stress-specific features. In summary, Peng et al. (2009) focused on analyzing genes differentially expressed in the heading-stage panicle between cultivated rice and wild rice, and Liu et al. (2010) provides genome-wide comparison between japonica and indica under stress (MV treatment) in the seedling stage. Compared to previous analyses, our analysis focused on identifying genes differentially expressed between japonica and indica through the whole life-cycle. Therefore, our data might be useful to determine general differences between japonica and indica. The differential expression patterns can be explained by deletion of japonica eQTLs in indica genome, suppression of japonica or indica eQTLs by defects in promoter or epigenetic regulation, and mismatches between japonica an indica sequences as indicated in Additional file 5: Table S2.

Unknown genes and transposable elements are overrepresented in japonica and indica eQTLs

Of 490 japonica eQTLs, 181 entries are annotated with a gene function; the remainder are 256 unknown (expressed) genes, 9 hypothetical proteins, and 44 transposons/retro-transposons. Forty-nine of the indica eQTLs are annotated with a gene function and the others are 36 unknown genes, 9 hypothetical proteins, and 10 transposons/retro-transposons. Therefore, 36 (34.6%) indica eQTLs and 256 (52.2%) japonica eQTLs were unknown genes (Figure 2). Recently, the Rice Genome Annotation Project (RGAP) released annotations of 56,798 non-redundant loci, and 6311 (11.1%) of them are annotated as unknown genes (Figure 2). Therefore, we expect the japonica or indica eQTLs identified in this study to have novel functions. We also analyzed the relative ratio of transposable element-related genes (TEs, transposons, or retro transposons) and identified 10 (9.6%) in the indica eQTLs (104) and 44 (9.0%) in the japonica eQTLs (490). The ratios of TEs in japonica and indica eQTLs were much lower than those of the whole genome (15588/56798, 27.4%). However, due to redundancy among TEs, only 3579 were printed on the Affymetrix array (Jung et al. 2008a). If we multiply 3579/15588 and 27.4%, we find that only 6.3% of TEs are covered in the Affymetrix array. Therefore, TEs in both indica and japonica eQTLs were enriched about 1.5-fold more than expected; suggesting TEs might have significant evolutionary roles in traits that distinguish japonica from indica. Transcription of the TEs was generally weaker than that of their non-TE counterparts (Jiao and Deng 2007). We recently reported on light-responsive genes in a comparison of light- vs. dark-grown rice seedlings. Using criteria of at least a 2-fold upregulation in the light-grown seedlings and less than 10-6 false discovery rate p-value, we identified 7 TEs (0.1%) among 1108 light-responsive transcripts. This is a significantly smaller proportion than that found in indica and japonica eQTLs. Among 295 genes with at least a 4-fold higher drought-inducible expression (Plexdb, http://www.plexdb.org/), we identified 11 TE genes, a ratio (3.7%) about one-third of that found in japonica or indica eQTLs. These results indicate that the portion of active TEs under stress is significantly lower than that in japonica and indica eQTLs, suggesting the active role of TEs in the domestication of japonica and indica.

Figure 2
figure 2

Over-representation of unknown genes in japonica and indica eQTLs. X axis indicates the gene classes with different features and Y axis indicates the relative ratio in each class. Dark gray bars indicate the relative ratios of TEs and weak gray bars indicate the relative ratios of unknown genes.

Defense response is enriched in both japonica and indica eQTLs

GO enrichment analysis for japonica eQTLs identified at least a 2-fold greater enrichment of apoptosis, defense response, transmembrane transport, protein folding, protein amino acid phosphorylation, carbohydrate metabolism, and type I hypersensitivity genes (Figure 3a). GO enrichment analysis of indica eQTLs revealed that glycolysis, oxidative stress response, defense response, apoptosis, and protein amino acid phosphorylation genes are significantly enriched (Figure 3b). Genes involved in the defense response, apoptosis, and protein amino acid phosphorylation are significantly enriched in both japonica and indica eQTLs.

Figure 3
figure 3

Gene Ontology enrichment analysis of japonica and indica eQTLs. The gene Ontology enrichment analysis tool installed in rice oligonucleotide array database (ROAD, http://www.ricearray.org/) was used to identify GO terms enriched in japonica (a) and indica (b) eQTLs. For each GO term, the GO enrichment value was the ratio of the number of observed genes in the gene list divided by the expected number, given the size in the gene list compared with the whole genome. X axis indicates the GO term and Y axis indicates GO enrichment value. The GO enrichment values were represented numerically in this figure.

We found significant overlap between the defense response and apoptosis: 26 japonica eQTLs and 6 indica eQTLs were involved in these functions (Additional file 6: Table S4). More interestingly, 15 japonica eQTLs associated with the defense response are located on chromosome 11, suggesting that this chromosome may be a hotspot for enhanced disease resistance. Our findings correlate well with the finding that the sequences of rice chromosomes 11 and 12 are rich in disease resistance genes and recent gene duplications (Consortia. 2005). However, we cannot ignore the possibility that some of the selected genes can function negatively to disease resistance. Among the indica eQTLs associated with the defense response, we identified an NBS-LRR, 2 disease resistance protein RPM1s, a terpene synthase, a thionin family protein, and a transposon (Additional file 6: Table S4). NBS-LRR proteins are important for the resistance response to biotic challenges, especially for rice blast fungus (Jeung et al. 2007; Lin et al. 2008; Okuyama et al. 2011). Identification of NBS-LRR and RPM1 in both japonica and indica eQTLs indicates that both cultivars might have evolved unique NBS-LRR and RPM1 genes for their defense responses.

We identified 19 japonica eQTLs and 5 indica eQTLs associated with protein phosphorylation. Of the 19 japonica eQTLs, there are 11 receptor-like kinases, 2 wall-associated kinases, a DUF26 kinase, an O-methyltransferase, and an expressed protein. Of the indica eQTLs, there are 2 calcium-dependent protein kinases (CAMKs), 2 protein kinases, and a casein kinase. Receptor-like kinases and wall-associated kinases are characteristic of japonica eQTLs and CAMKs are unique to indica eQTLs.

Transcription regulation is generally affected by upstream phosphorylation cascades. We identified 4 japonica eQTLs associated with transcriptional regulation (i.e., Os04g35010 encoding bHLH, Os05g25770/OsWRKY45 and Os11g45850/OsWRKY81 encoding WRKY TF, and Os12g07950 encoding transcriptional regulator Sir2 family protein), and 3 indica eQTLs (i.e., Os06g30090 encoding bHLH, Os08g18000 encoding a F-box domain-containing protein, and Os10g26940 encoding a BURP domain-containing protein) (Additional file 6: Table S4). OsWRKY45 confers enhanced resistance to fungal pathogens (Shimono et al. 2007). OsWRKY45-1 (japonica eQTL) - overexpressing plants showed increased susceptibility, and OsWRKY45-1 - knockout plants showed enhanced resistance to bacterial pathogens Xanthomonas oryzae pv oryzae ( Xoo) and Xanthomonas oryzae pv oryzicola (Xoc). OsWRKY45-2 (indica eQTL)-overexpressing plants showed enhanced resistance and OsWRKY45-2-suppressed plants showed increased susceptibility to Xoo and Xoc. This suggests that OsWRKY45-1 and OsWRKY45-2 confer distinct disease resistance to bacterial pathogens (Tao et al. 2009).

Protein folding, transmembrane transport, carbohydrate metabolism, and type I hypersensitivity genes are exclusively enriched in japonica eQTLs (Additional file 6: Table S4). In contrast, glycolysis and the oxidative stress response are uniquely enriched in indica eQTLs. The smaller population of indica eQTLs identified in this study may explain the smaller number of GO terms enriched in indica eQTLs. Johns and Mao (2007) carried out gene ontology analysis for polymorphism levels between putative homologues in japonica and indica, identifying four functional classes: genes involved in production of defense-related compounds, cell wall synthesis, cell signaling, and transcription factors. At least 6% of the japonica and indica genomes are unusually divergent (Tang et al. 2006). Of GO terms associated with biological processes in highly-divergent regions of japonica and indica, response to biotic stimulus is the most significantly overrepresented (Tang et al. 2006). These results further highlight the significance of the defense response GO term enriched in both japonica eQTLs and indica eQTLs.

Biotic stress overview in MapMan toolkit revealed key elements of the signaling pathway for the defense response in japonica and indica eQTLs

MapMan is a useful tool for visualizing high-throughput omics data such as genome-wide microarray gene expression data or RNA-seq technologies in the context of metabolic pathways or cellular processes. Since GO enrichment analysis revealed that “defense response” genes were significantly overrepresented between both genes that were preferentially expressed in japonica and those in indica, we employed the biotic stress overview installed in the MapMan toolkit. To do this, we uploaded log2-fold change data of the average expression level of japonica samples over the average expression level of indica samples in Additional file 5: Table S2 to the cellular process overview and biotic stress overview installed in MapMan (Figure 4). Then, we mapped the fold-change data to MapMan using the Affymetrix probeset_ids in Additional file 5: Table S2. It is also possible to map rice genome annotation project (RGAP) locus ids in the MapMan tool. We mapped 594 genes on the Affymetrix array to MapMan terms (Additional file 7: Table S5).

Figure 4
figure 4

MapMan analysis of japonica and indica eQTLs. Biotic stress overview installed in the MapMan toolkit after integration of log2 fold change data of average japonica intensity over average indica intensity. Red color indicates upregulation in japonica cultivar and green color indicates upregulation in indica cultivar. Detailed information of japonica and indica eQTLs integrated in MapMan overviews is prepared in Additional file 7: Table S5.

Two hundred and forty-four genes have MapMan terms with assigned functional classifications but the others (350 genes) do not, which further demonstrates the high frequency of unknown genes in japonica and indica eQTLs (Figure 2). Figure 4 suggests detailed components of the signaling pathway in the biotic stress overview. The biotic stress overview feature begins with pathogen recognition by R genes, followed by respiratory burst, heat shock proteins, miscellaneous function, signaling, MAPK cascades, transcription factors, defense response by pathogen-related (PR) proteins, protein degradation, and secondary metabolites (Figure 4).

Pathogen recognition by R genes

MapMan analysis for japonica and indica eQTLs revealed 31 elements relating to pathogen recognition by R genes, and these elements include 26 NB-ARC/NBS-LRR proteins, 2 disease resistance RPP13-like proteins, 2 Yr10 stripe rust resistance proteins, OsWRKY125, etc. (Additional file 7: Table S5). Four (Os09g09750, Os06g22460, Os06g4939 0, and Os11g35490) of them are indica eQTLs, and the others are japonica eQTLs. Pia (Os11g11810) encodes a NBS-LRR disease resistance protein preferentially expressed in indica and Pi25/Pid3 (Os06g22460) encodes a NB-ARC disease resistance protein preferentially expressed in japonica that are known to be involved in blast tolerance (Shang et al. 2009; Chen et al. 2011; Okuyama et al. 2011). These results suggest that rice cultivars might develop unique defense mechanisms by selecting unique NB-ARC/NBS-LRR proteins. We expect that all remaining preferentially-expressed japonica and indica genes encoding NB-ARC/NBS-LRR proteins might play key roles in enhanced defense responses against blast or other pathogens.

Hormone signaling

We identified 5 elements related to auxin (Os07g38890 and Os03g62060), jasmonate (Os02g12690), abscisic acid (ABA) (Os07g18162), and ethylene (Os09g27820) signaling (Additional file 7: Table S5). The three genes relating to auxin and jasmonate signaling are indica eQTLs, and those relating to ABA and ethylene signaling are japonica eQTLs. We identified 3 elements related to redox reactions and 7 elements related to miscellaneous functions, including glutathione S-transferase and peroxidase (Additional file 7: Table S5). Two of them are indica eQTLs, and the others are japonica eQTLs.

Signaling pathways

Signaling pathways in general are understood as events occurring after recognition by R genes, although there are receptor-like kinases such as Xa21 and Xa26 that play roles as defense genes (Song et al. 1995; Sun et al. 2004). We identified 32 elements, including 25 receptor-like kinases (RLKs), 5 non-receptor kinases, a G-protein, and a calcium-binding molecule (Additional file 7: Table S5). Receptor-like kinases might have roles in the recognition of pathogen infection, as well as in signal transduction. Among the non-receptor kinases, we identified 2 Ca2+/calmodulin-dependent protein kinases (CAMKs; Os02g41580 and Os08g34240), a case in kinase II subunit alpha-1 (Os03g55389), a serine/threonine-protein kinase Arabidopsis Fus3-complementating gene 2 (AFC2; Os12g27520), and a protein kinase (Os03g53410). Calcineurin B may be either directly or indirectly linked with CAMKs. The ras family domain protein associated-G protein might have roles in the upstream signaling of diverse non-receptor kinases.

Transcription factors

Transcription factors (TFs) have roles in signaling pathways downstream of kinases. We identified 3 WRKY genes (OsWRKY45, OsWRKY81, and OsWRKY125) showing preferred expression in japonica. The roles of OsWRKY45 in the defense response against pathogens have already been discussed. Phylogenetic comparison of rice and Arabidopsis WRKY families revealed that OsWRKY81 was clustered with AtWRKY18 in WRKY subgroup IIa, which functions in the positive regulation of basal defense and systemic acquired resistance (SAR) operating downstream from NPR1 in Arabidopsis (Wang et al. 2006). OsWRKY125 contains both a NB-ARC domain, which is important in the recognition of pathogen attack, and a WRKY domain, which is important in transcriptional regulation of defense genes (Pandey and Somssich 2009). Arabidopsis RRS1-R, a chimera of a TIR-NB-LRR protein and a WRKY-type transcription factor, conferred resistance against multiple pathogens by the cooperation with the nuclear R protein RPS4 (Slootweg et al. 2010). Functions of several R genes such as MLA, RPS4, and snc1 are activated after nuclear localization. These findings suggest multiple roles for the OsWRKY125 gene, including recognition of pathogens, signal transduction, and stimulation of defense gene functions.

Protein/cell wall degradation, secondary metabolism, and PR proteins

We identified 29 elements related to protein degradation (proteolysis), 16 related to secondary metabolism, 2 related to cell wall degradation, and a PR protein (Additional file 7: Table S5). Protein degradation is mainly performed by 2 methods: one depending on proteases such as cysteine protease (5 elements), aspartate protease (1 element), subtilase (1 element) and 2 other elements, and the other depending on proteasome cooperation with Skp, Cullin, and the F-box-containing complex (SCF). The substrate specificity of the latter is determined by a distinct F-box protein (Spencer et al. 1999). Twenty F-box proteins identified in this study belong to the SCF E3 ligase complex. 14 of these are japonica QTLs, and the others are indica eQTLs. Each QTL plays a unique function associated with biological processes or cellular events in the respective cultivars. In addition, we identified an Skp subunit (Os12g40300) of the SCF E3 ligase complex, 1 subunit (Os12g14840) of the Ring E3 ligase complex, 1 subunit (Os08g13130) of the BTB/POZ E3 ligase complex, and 1 subunit (Os03g37950) of a proteasome (Additional file 7: Table S5). The detailed features of ubiquitin- and autophagy-dependent degradation are presented in Additional file 8: Figure S3 and Additional file 7: Table S5. Secondary metabolism also plays significant roles in the defense response (Degenhardt 2009). Five types of secondary metabolites were identified: terpenoid (7 elements), phenylpropanoid (6 elements), flavonoid (1 element), carotenoid (1 element), and another (1 element). Cell wall degradation is considered as a mechanism to protect plants from biotrophs (Sun et al. 2011); 2 elements in our analysis were found to be associated with cell wall degradation (Additional file 7: Table S5). PR genes are well-known marker genes of pathogenesis or defense responses, and we identified 1 such PR gene (Os08g28670). The elements identified in this section might indicate candidate genes involved in defense signaling pathways in rice.

Identification of functionally-characterized genes expressed preferentially in japonica or indica

To evaluate the functional significance of genes expressed preferentially in japonica or indica, we queried the OGRO database with 594 genes to see if any of them were functionally characterized. We identified 10 genes (Table 1). A blast resistance gene, Pid3, was identified from a cross between a rice blast-resistant indica variety and a susceptible japonica variety (Shang et al. 2009). The allelic Pid3 loci in most of the tested japonica varieties were identified as pseudogenes due to a nonsense mutation at nucleotide position 2,208 starting from the translation initiation site. In addition, the expression of this gene was significantly repressed in japonica plants, suggesting another reason for the non-functionality of this gene in japonica. It was recently reported that the resistance of Pia originated from japonica rice (Cho et al. 2009), explaining its preferred expression in japonica. Tiller Angle Control 1 (TAC1) controls tiller angle for dense planting during rice cultivation and was identified from crosses between an indica rice, IR24, which displays a relatively spread-out plant architecture, and an introgressed line, IL55, derived from japonica rice Asominori, which displays a compact plant architecture with extremely erect tillers (Yu et al. 2007). Heading date 6 (Hd6) encoding the alpha subunit of casein kinase II is a quantitative trait locus involved in rice photoperiod sensitivity and was identified from a cross between the japonica variety, Nipponbare and the indica variety Kasalath (Takahashi et al. 2001). A pair of allelic genes, OsWRKY45-1 and OsWRKY45-2, with a 10-amino acid difference, play opposite roles in rice resistance against bacterial pathogens and preferential expression of OsWRKY45 in japonica provides another example of functional diversity (Tao et al. 2009). Pia, Pid3/Pi25, OsWRKY45, TAC1, and Hd6 are good examples for evaluating the significance of genes expressed preferentially in japonica or indica during domestication.

Table 1 Summary of functionally characterized japonica or indica preferred genes from OGRO ( http://qtaro.abr.affrc.go.jp/ogro )

Unlike the above five genes, the functions of the other genes were identified through transgenic loss of function studies. Rice brassinolide-enhanced gene 2 (OsBLE2) encoding fructose-bisphosphate aldolase is involved in the brassinolide-regulated growth and development processes in rice (Yang et al. 2003). This gene has been reported to regulate root development (Konishi et al. 2004). Histone deacetylase 710 (HDA710) in rice regulates vegetative growth (Hu et al. 2009). Rice beta-1,3-glucanase gene 1 (Osg1) is required for callose degradation in pollen development (Wan et al. 2011). Rice yellow stripe 1-like gene 2 (OsYSL2) is required for the long-distance transport of iron and manganese and also contributes to eating quality (Ishimaru et al. 2010). Knock-down of rice S-adenosyl-l-methionine synthetase 1, 2, and 3 (OsSAMS1, 2, and 3) showed pleiotropic phenotypes, including dwarfism, reduced fertility, delayed germination, and late flowering (Li et al. 2011). However, elucidation of the unique function of OsSAMS3 may require gain of function analysis or the correlation to natural variation. Although the functions of the five genes in this section were not evaluated based on agronomic traits, these genes may have the potential to play key roles in domestication process.

Conclusion

Meta-analysis based on the collection of genome-wide transcriptome data is useful to identify valuable information that is otherwise difficult to obtain from individual transcriptome data. In this study, we identified a large number of genes preferentially expressed in japonica and indica by analyzing Affymetrix array data comprising 595 indica and 388 japonica samples. Various types of epigenetic regulation mediated by miRNA, siRNA, histone modification, and insertion patterns of transposons might result in the differential expression of the identified eQTLs in japonica and indica (Slotkin and Martienssen 2007; Berdasco et al. 2008; Mallory and Bouche 2008; Byun et al. 2012). The most striking feature of this collection of genes is the dominance of genes with unknown function and TEs. It is interesting that there is a possible association between agronomic traits distinguishing japonica and indica cultivars and these novel genes. Recently, Gramene developed a web-based QTL database. From this database, 8646 QTLs were identified in rice. These data might be useful in association studies of candidate genes identified from genome-wide analysis and related agronomic traits. Even if the resolution of most QTLs is very low, the linkage information can help to connect the phenotype of gene-indexed mutants to agronomic traits. Functional analysis using T-DNA insertional mutants for genes preferentially expressed in japonica or indica is currently underway and will provide new insights into the evolution and domestication of japonica and indica cultivars.

Methods

Identification of japonica and indica eQTLs using 983 Affymetrix microarrays

The Affymetrix raw data for 983 arrays made with japonica and indica samples were downloaded from NCBI GEO (platform Accession Number is GPL2025) (Additional file 1: Table S1). Additional file 1: Table S1 summarized the accession number, cultivar name, variety name, and tissue/organ type of 983 slides based on the information in NCBI GEO or Arrayexpress. We used MAS 5.0 provided by the affy R package for Affymetrix arrays to convert probe-level data to expression values and normalize the expression values. The data were log2 transformed. We classified our data into 2 subgroups, namely, japonica (388) and indica (595), based on the cultivar name of samples used for the microarray experiments. We generated average log2 fold change of japonica samples over indica samples. SAM analysis in the TIGR Multi Experiment Viewer (MeV; http://www.tm4.org/mev.html) between japonica and indicia cultivars revealed 699 probes showing preferential expression in japonica and 118 probes showing preferential expression in indica. We identified 490 japonica eQTLs from 609 probes and 104 indica eQTLs from 118 probes (Additional file 5: Table S2). Selected genes in a cultivar showed at least 2.7 fold higher expression than in the other (Additional file 5: Table S2).

Heatmap analysis

We used MeV to generate Heatmap expression patterns of 594 genes in Additional file 5: Table S2.

Gene ontology term enrichment analysis

We evaluated the enrichment of GO (Gene Ontology) terms for preferentially expressed genes in japonica and indica (eQTL) in the biological process category. We calculated fold-enrichment for each GO Slim term in a querying gene list and identified GO terms and related gene entries with more than 2-fold GO enrichment and a hyper geometric p-value of < 0.05 (Additional file 6: Table S4). For each term, fold-enrichment is the observed number of genes in the gene list divided by the expected number of genes, given the size of the gene list compared with the whole genome. Additional file 9: Table S6 contains data of GO Slim terms selected by >2-fold enrichment and a hyper geometric p-value of < 0.05.

MapMan analysis

Thirty-six MapMan BINs are currently used for the Rice MapMan classification, and these BINs can be extended in a hierarchical manner into subBINs (Usadel et al. 2005; Urbanczyk-Wochniak et al. 2006) (Additional file 7: Table S5). To integrate significant gene expression data from our transcriptome analysis into the diverse MapMan tools, we generated a dataset including Affymetrix probe_ids and average fold-change data of japonica over indica. For Figure 4, we used Cellular_response_overview and Biotic_stress_overview, and for Additional file 8: Figure S3, we used Proteasome Detail and Autophagy installed in the MapMan toolkit. The detailed procedure is described in a recent study by our group (Jung et al. 2012).

RNA extraction and RT-PCR analysis

To evaluate the expression patterns of eQTLs identified in this study, we carried out RT-PCR for 17 candidates (10 japonica and 7 indica eQTLs). We grew seeds of Nipponbare and Dongjin (japonica cultivars) and IR8 and IR64 (indica cultivars) for a week and harvested the leaves of the plants for RNA expression analysis. RT-PCR was carried out as in a previous study (Jung et al. 2006). The reaction included an initial 5-min denaturation at 94°C; followed by 27 cycles of 94°C for 45 minutes, 60°C for 45 minutes, and 72°C for 1 minute; and a final 10 - minute extension at 72°C. Next, 20 μL of the reaction mixture was separated on a 1.2% agarose gel. The primers used for RT-PCR analysis are listed in Additional file 10: Table S7 and Additional file 11: Table S8.

Abbreviations

ABA:

Abscisic acid

CAMKs:

Calcium-dependent protein kinases

eQTLs:

Expression quantitative trait loci

GEO:

Gene expression omnibus

GO:

Gene ontology

ids:

Identifiers

MeV:

Multi experiment viewer

PR:

Pathogen-related

RLKs:

Receptor-like kinases

RT-PCR:

Reverse transcriptase polymerase chain reaction

RILs:

Recombinant inbred lines

ROAD:

Rice oligonucleotide array database

SAM:

Significant microarray analysis

SAR:

Systemic acquired resistance

SCF:

F-box-containing complex

TEs:

Transposable elements

TFs:

Transcription factors

References

  • Barrett T, Troup DB, Wilhite SE, Ledoux P, Evangelista C, Kim IF, Tomashevsky M, Marshall KA, Phillippy KH, Sherman PM, Muertter RN, Holko M, Ayanbule O, Yefanov A, Soboleva A: NCBI GEO: archive for functional genomics data sets–10 years on. Nucleic Acids Res 2011, 39: D1005-D1010. 10.1093/nar/gkq1184

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Berdasco M, Alcazar R, Garcia-Ortiz MV, Ballestar E, Fernandez AF, Roldan-Arjona T, Tiburcio AF, Altabella T, Buisine N, Quesneville H, Baudry A, Lepiniec L, Alaminos M, Rodriguez R, Lloyd A, Colot V, Bender J, Canal MJ, Esteller M, Fraga MF: Promoter DNA hypermethylation and gene repression in undifferentiated Arabidopsis cells. PLoS ONE 2008, 3: e3306. 10.1371/journal.pone.0003306

    Article  PubMed Central  PubMed  Google Scholar 

  • Byun HM, Heo K, Mitchell KJ, Yang AS: Mono-allelic retrotransposon insertion addresses epigenetic transcriptional repression in human genome. J Biomed Sci 2012, 19: 13. 10.1186/1423-0127-19-13

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Chen J, Shi Y, Liu W, Chai R, Fu Y, Zhuang J, Wu J: A Pid3 allele from rice cultivar Gumei2 confers resistance to Magnaporthe oryzae. J Genet Genomics 2011, 38: 209–216. 10.1016/j.jgg.2011.03.010

    Article  CAS  PubMed  Google Scholar 

  • Cho Y-C, Suh J-P, Jeung J-U, Roh J-H, Yang C-I, Oh M-K, Jeon J-S, Choi I-S, Hwang H-G, Yang S-J, Kim Y-G: Resistance genes and their effects to blast in korean rice varieties (Oryza sativa L.). In Advances in genetics. Edited by: Wang G-L, Valent B. Springer Netherlands: Genomics and Control of Rice Blast Disease; 2009:291–304.

    Google Scholar 

  • The Rice Chromosomes 11 and 12 Sequencing Consortia: The sequence of rice chromosomes 11 and 12, rich in disease resistance genes and recent gene duplications. BMC biol 2005, 3: 20.

    Article  PubMed Central  Google Scholar 

  • Degenhardt J: Indirect defense responses to herbivory in grasses. Plant Physiol 2009, 149: 96–102. 10.1104/pp.108.128975

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Gautier L, Cope L, Bolstad BM, Irizarry RA: affy–analysis of Affymetrix GeneChip data at the probe level. Bioinformatics 2004, 20: 307–315. 10.1093/bioinformatics/btg405

    Article  CAS  PubMed  Google Scholar 

  • Hu Y, Qin F, Huang L, Sun Q, Li C, Zhao Y, Zhou DX: Rice histone deacetylase genes display specific expression patterns and developmental functions. Biochem Biophys Res Commun 2009, 388: 266–271. 10.1016/j.bbrc.2009.07.162

    Article  CAS  PubMed  Google Scholar 

  • Ishimaru Y, Masuda H, Bashir K, Inoue H, Tsukamoto T, Takahashi M, Nakanishi H, Aoki N, Hirose T, Ohsugi R, Nishizawa NK: Rice metal-nicotianamine transporter, OsYSL2, is required for the long-distance transport of iron and manganese. Plant J 2010, 62: 379–390. 10.1111/j.1365-313X.2010.04158.x

    Article  CAS  PubMed  Google Scholar 

  • Jeung JU, Kim BR, Cho YC, Han SS, Moon HP, Lee YT, Jena KK: A novel gene, Pi40(t), linked to the DNA markers derived from NBS-LRR motifs confers broad spectrum of blast resistance in rice. Theor Appl Genet 2007, 115: 1163–1177. 10.1007/s00122-007-0642-x

    Article  CAS  PubMed  Google Scholar 

  • Jiao Y, Deng XW: A genome-wide transcriptional activity survey of rice transposable element-related genes. Genome Biol 2007, 8: R28. 10.1186/gb-2007-8-2-r28

    Article  PubMed Central  PubMed  Google Scholar 

  • Johns MA, Mao L: Differentiation of the two rice subspecies indica and japonica: a Gene Ontology perspective. Funct Integr Genomics 2007, 7: 135–151. 10.1007/s10142-006-0036-1

    Article  CAS  PubMed  Google Scholar 

  • Jung KH, An G: Application of MapMan and RiceNet drives systematic analyses of the early reat stress transcriptome in rice seedlings. J Plant Biol 2012, 55: 436–449. 10.1007/s12374-012-0270-0

    Article  CAS  Google Scholar 

  • Jung KH, An G, Ronald PC: Towards a better bowl of rice: assigning function to tens of thousands of rice genes. Nat Rev Genet 2008, 9: 91–101.

    CAS  PubMed  Google Scholar 

  • Jung KH, Dardick C, Bartley LE, Cao P, Phetsom J, Canlas P, Seo YS, Shultz M, Ouyang S, Yuan Q, Frank BC, Ly E, Zheng L, Jia Y, Hsia AP, An K, Chou HH, Rocke D, Lee GC, Schnable PS, An G, Buell CR, Ronald PC: Refinement of light-responsive transcript lists using rice oligonucleotide arrays: evaluation of gene-redundancy. PLoS ONE 2008, 3: e3337. 10.1371/journal.pone.0003337

    Article  PubMed Central  PubMed  Google Scholar 

  • Jung KH, Han MJ, Lee DY, Lee YS, Schreiber L, Franke R, Faust A, Yephremov A, Saedler H, Kim YW, Hwang I, An G: Wax-deficient anther1 is involved in cuticle and wax production in rice anther walls and is required for pollen development. Plant Cell 2006, 18: 3015–3032. 10.1105/tpc.106.042044

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Kim MY, Van K, Kang YJ, Kim KH, Lee SH: Tracing soybean domestication history: From nucleotide to genome. Breed Sci 2012, 61: 445–452. 10.1270/jsbbs.61.445

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Khush GS: Origin, dispersal, cultivation and variation of rice. Plant Mol Biol 1997, 35: 25–34. 10.1023/A:1005810616885

    Article  CAS  PubMed  Google Scholar 

  • Lin F, Chen S, Que Z, Wang L, Liu X, Pan Q: The blast resistance gene Pi37 encodes an NBS-LRR protein and is a member of a resistance gene cluster on rice chromosome 1. Phytopathol 2008, 98: S90-S90.

    Google Scholar 

  • Konishi H, Yamane H, Maeshima M, Komatsu S: Characterization of fructose-bisphosphate aldolase regulated by gibberellin in roots of rice seedling. Plant Mol Biol 2004, 56: 839–848. 10.1007/s11103-004-5920-2

    Article  CAS  PubMed  Google Scholar 

  • Li W, Han Y, Tao F, Chong K: Knockdown of SAMS genes encoding S-adenosyl-l-methionine synthetases causes methylation alterations of DNAs and histones and leads to late flowering in rice. J Plant Physiol 2011, 168: 1837–1843. 10.1016/j.jplph.2011.05.020

    Article  CAS  PubMed  Google Scholar 

  • Liu FX, Xu WY, Wei Q, Zhang ZH, Xing Z, Tan LB, Di C, Yao DX, Wang CC, Tan YJ, Yan H, Ling Y, Sun CQ, Xue YB, Su Z: Gene Expression Profiles Deciphering Rice Phenotypic Variation between Nipponbare (Japonica) and 93–11 (Indica) during Oxidative Stress. PLoS ONE 2010, 5: e8632. 10.1371/journal.pone.0008632

    Article  PubMed Central  PubMed  Google Scholar 

  • Lu TT, Lu GJ, Fan DL, Zhu CR, Li W, Zhao QA, Feng Q, Zhao Y, Guo YL, Li WJ, Huang XH, Han B: Function annotation of the rice transcriptome at single-nucleotide resolution by RNA-seq. Genome Res 2010, 20: 1238–1249. 10.1101/gr.106120.110

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Mallory AC, Bouche N: MicroRNA-directed regulation: to cleave or not to cleave. Trends Plant Sci 2008, 13: 359–367. 10.1016/j.tplants.2008.03.007

    Article  CAS  PubMed  Google Scholar 

  • Okuyama Y, Kanzaki H, Abe A, Yoshida K, Tamiru M, Saitoh H, Fujibe T, Matsumura H, Shenton M, Galam DC, Undan J, Ito A, Sone T, Terauchi R: A multifaceted genomics approach allows the isolation of the rice Pia-blast resistance gene consisting of two adjacent NBS-LRR protein genes. Plant J 2011, 66: 467–479. 10.1111/j.1365-313X.2011.04502.x

    Article  CAS  PubMed  Google Scholar 

  • Pandey SP, Somssich IE: The Role of WRKY transcription factors in plant immunity. Plant Physiol 2009, 150: 1648–1655. 10.1104/pp.109.138990

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Parkinson H, Sarkans U, Kolesnikov N, Abeygunawardena N, Burdett T, Dylag M, Emam I, Farne A, Hastings E, Holloway E, Kurbatova N, Lukk M, Malone J, Mani R, Pilicheva E, Rustici G, Sharma A, Williams E, Adamusiak T, Brandizi M, Sklyar N, Brazma A: ArrayExpress update–an archive of microarray and high-throughput sequencing-based functional genomics experiments. Nucleic Acids Res 2011, 39: D1002-D1004. 10.1093/nar/gkq1040

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Peng ZY, Zhang H, Liu T, Dzikiewicz KM, Li S, Wang X, Hu G, Zhu Z, Wei X, Zhu QH, Sun Z, Ge S, Ma L, Li L, Deng XW: Characterization of the genome expression trends in the heading-stage panicle of six rice lineages. Genomics 2009, 93: 169–178. 10.1016/j.ygeno.2008.10.005

    Article  CAS  PubMed  Google Scholar 

  • Saeed AI, Bhagabati NK, Braisted JC, Liang W, Sharov V, Howe EA, Li J, Thiagarajan M, White JA, Quackenbush J: TM4 microarray software suite. Methods Enzymol 2006, 411: 134–193.

    Article  CAS  PubMed  Google Scholar 

  • Sang T, Ge S: The puzzle of rice domestication. J Integr Plant Biol 2007, 49: 760–768. 10.1111/j.1744-7909.2007.00510.x

    Article  CAS  Google Scholar 

  • Sato Y, Takehisa H, Kamatsuki K, Minami H, Namiki N, Ikawa H, Ohyanagi H, Sugimoto K, Antonio BA, Nagamura Y: RiceXPro version 3.0: expanding the informatics resource for rice transcriptome. Nucleic Acids Res 2013, 41: D1206-D1213. 10.1093/nar/gks1125

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Shang J, Tao Y, Chen X, Zou Y, Lei C, Wang J, Li X, Zhao X, Zhang M, Lu Z, Xu J, Cheng Z, Wan J, Zhu L: Identification of a new rice blast resistance gene, Pid3, by genomewide comparison of paired nucleotide-binding site–leucine-rich repeat genes and their pseudogene alleles between the two sequenced rice genomes. Genetics 2009, 182: 1303–1311. 10.1534/genetics.109.102871

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Shimono M, Sugano S, Nakayama A, Jiang CJ, Ono K, Toki S, Takatsuji H: Rice WRKY45 plays a crucial role in benzothiadiazole-inducible blast resistance. Plant Cell 2007, 19: 2064–2076. 10.1105/tpc.106.046250

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Shimono M, Koga H, Akagi A, Hayashi N, Goto S, Sawada M, Kurihara T, Matsushita A, Sugano S, Jiang CJ, Kaku H, Inoue H, Takatsuji H: Rice WRKY45 plays important roles in fungal and bacterial disease resistance. Mol Plant Pathol 2012, 13: 83–94. 10.1111/j.1364-3703.2011.00732.x

    Article  CAS  PubMed  Google Scholar 

  • Slootweg E, Roosien J, Spiridon LN, Petrescu AJ, Tameling W, Joosten M, Pomp R, van Schaik C, Dees R, Borst JW, Smant G, Schots A, Bakker J, Goverse A: Nucleocytoplasmic distribution is required for activation of resistance by the potato NB-LRR receptor Rx1 and is balanced by its functional domains. Plant Cell 2010, 22: 4195–4215. 10.1105/tpc.110.077537

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Slotkin RK, Martienssen R: Transposable elements and the epigenetic regulation of the genome. Nat Rev Genet 2007, 8: 272–285.

    Article  CAS  PubMed  Google Scholar 

  • Song WY, Wang GL, Chen LL, Kim HS, Pi LY, Holsten T, Gardner J, Wang B, Zhai WX, Zhu LH, Fauquet C, Ronald P: A receptor kinase-like protein encoded by the rice disease resistance gene, Xa21. Science 1995, 270: 1804–1806. 10.1126/science.270.5243.1804

    Article  CAS  PubMed  Google Scholar 

  • Spencer E, Jiang J, Chen ZJ: Signal-induced ubiquitination of IkappaBalpha by the F-box protein Slimb/beta-TrCP. Genes Dev 1999, 13: 284–294. 10.1101/gad.13.3.284

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Sun Q, Greve LC, Labavitch JM: Polysaccharide compositions of intervessel pit membranes contribute to Pierce’s disease resistance of grapevines. Plant Physiol 2011, 155: 1976–1987. 10.1104/pp.110.168807

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Sun X, Cao Y, Yang Z, Xu C, Li X, Wang S, Zhang Q: Xa26, a gene conferring resistance to Xanthomonas oryzae pv. oryzae in rice, encodes an LRR receptor kinase-like protein. Plant J 2004, 37: 517–527. 10.1046/j.1365-313X.2003.01976.x

    Article  CAS  PubMed  Google Scholar 

  • Takahashi Y, Shomura A, Sasaki T, Yano M: Hd6, a rice quantitative trait locus involved in photoperiod sensitivity, encodes the alpha subunit of protein kinase CK2. Proc Natl Acad Sci U S A 2001, 98: 7922–7927. 10.1073/pnas.111136798

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Tang T, Lu J, Huang J, He J, McCouch SR, Shen Y, Kai Z, Purugganan MD, Shi S, Wu CI: Genomic variation in rice: genesis of highly polymorphic linkage blocks during domestication. PLoS Genet 2006, 2: e199. 10.1371/journal.pgen.0020199

    Article  PubMed Central  PubMed  Google Scholar 

  • Tao Z, Liu H, Qiu D, Zhou Y, Li X, Xu C, Wang S: A pair of allelic WRKY genes play opposite roles in rice-bacteria interactions. Plant Physiol 2009, 151: 936–948. 10.1104/pp.109.145623

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Tao Z, Kou Y, Liu H, Li X, Xiao J, Wang S: OsWRKY45 alleles play different roles in abscisic acid signalling and salt stress tolerance but similar roles in drought and cold tolerance in rice. J Exp Bot 2011, 62: 4863–4874. 10.1093/jxb/err144

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Tsiantis M: A transposon in tb1 drove maize domestication. Nat Genet 2011, 43: 1048–1050. 10.1038/ng.986

    Article  CAS  PubMed  Google Scholar 

  • Urbanczyk-Wochniak E, Usadel B, Thimm O, Nunes-Nesi A, Carrari F, Davy M, Blasing O, Kowalczyk M, Weicht D, Polinceusz A, Meyer S, Stitt M, Fernie AR: Conversion of MapMan to allow the analysis of transcript data from Solanaceous species: effects of genetic and environmental alterations in energy metabolism in the leaf. Plant Mol Biol 2006, 60: 773–792. 10.1007/s11103-005-5772-4

    Article  CAS  PubMed  Google Scholar 

  • Usadel B, Nagel A, Thimm O, Redestig H, Blaesing OE, Palacios-Rojas N, Selbig J, Hannemann J, Piques MC, Steinhauser D, Scheible WR, Gibon Y, Morcuende R, Weicht D, Meyer S, Stitt M: Extension of the visualization tool MapMan to allow statistical analysis of arrays, display of corresponding genes, and comparison with known responses. Plant Physiol 2005, 138: 1195–1204. 10.1104/pp.105.060459

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Wan L, Zha W, Cheng X, Liu C, Lv L, Wang Z, Du B, Chen R, Zhu L, He G: A rice beta-1,3-glucanase gene Osg1 is required for callose degradation in pollen development. Planta 2011, 233: 309–323. 10.1007/s00425-010-1301-z

    Article  CAS  PubMed  Google Scholar 

  • Wang D, Amornsiripanitch N, Dong XN: A genomic approach to identify regulatory nodes in the transcriptional network of systemic acquired resistance in plants. PLoS Pathog 2006, 2: 1042–1050.

    Article  CAS  Google Scholar 

  • Wang J, Yu H, Xie W, Xing Y, Yu S, Xu C, Li X, Xiao J, Zhang Q: A global analysis of QTLs for expression variations in rice shoots at the early seedling stage. Plant J 2010, 63: 1063–1074. 10.1111/j.1365-313X.2010.04303.x

    Article  CAS  PubMed  Google Scholar 

  • Wei X, Wang R, Cao L, Yuan N, Huang J, Qiao W, Zhang W, Zeng H, Yang Q: Origin of Oryza sativa in China inferred by nucleotide polymorphisms of organelle DNA. PLoS ONE 2012, 7: e49546. 10.1371/journal.pone.0049546

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  • Yamamoto E, Yonemaru J, Yamamoto T, Yano M: OGRO: The overview of functionally characterized genes in rice online database. Rice 2012, 5: 26. 10.1186/1939-8433-5-26

    Article  Google Scholar 

  • Yang G, Matsuoka M, Iwasaki Y, Komatsu S: A novel brassinolide-enhanced gene identified by cDNA microarray is involved in the growth of rice. Plant Mol Biol 2003, 52: 843–854. 10.1023/A:1025001304994

    Article  CAS  PubMed  Google Scholar 

  • Yu S, Zhang X, Guan Q, Takano T, Liu S: Expression of a carbonic anhydrase gene is induced by environmental stresses in Rice (Oryza sativa L.). Biotechol Letters 2007, 29: 89–94.

    Article  Google Scholar 

Download references

Acknowledgements

We thank Dr. Pamela C. Ronald and Dr. Peijian Cao (Zhengzhou Tobacco Research Institute, China) for developing rice meta-profiling data with Affymetrix array data from NCBI gene expression omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/). This research was funded by the Next-Generation BioGreen 21 Program of South Korea (PJ008079 and PJ009514 to KHJ).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Ki-Hong Jung.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

KHJ, WW, HKC and GA proposed the project idea. KHJ, HJG, HBC, NCAK, HKG, TQNN and TZ did the microarray data analysis and RT-PCR analysis. KHJ and GA wrote the manuscript. All authors read and approved the final manuscript.

Electronic supplementary material

Additional file 1: Table S1: Summary of 983 Affymetrix microarray data points used in this study. (XLSX 36 KB)

Additional file 2: Figure S1: Chromosomal distribution of japonica and indica eQTLs. (JPEG 99 KB)

Additional file 3: Figure S2: RT-PCR analysis of 17 eQTLs to validate microarray data. (JPEG 81 KB)

12284_2013_56_MOESM4_ESM.xlsx

Additional file 4: Table S3: List of genes preferentially expressed in japonica and indica that were identified in Liu et al. (2010) and comparison with our japonica and indica eQTLs. (XLSX 75 KB)

12284_2013_56_MOESM5_ESM.xlsx

Additional file 5: Table S2: Average log2 fold change data of eQTLs used in Figure 1. (XLSX 66 KB)

Additional file 6: Table S4: Detailed information of genes identified by GO enrichment analysis. (XLSX 14 KB)

Additional file 7: Table S5: Detailed information of japonica and indica eQTLs used for MapMan analysis. (XLSX 43 KB)

12284_2013_56_MOESM8_ESM.jpeg

Additional file 8: Figure S3: Ubiquitin and autophagy-dependent degradation overview with integration of japonica and indica eQTLs. (JPEG 941 KB)

12284_2013_56_MOESM9_ESM.xlsx

Additional file 9: Table S6: Data used for GO enrichment analysis in Figure 3. (XLSX 12 KB)

12284_2013_56_MOESM10_ESM.xlsx

Additional file 10: Table S7: Information on primers used to validate expression patterns of 17 selected eQTLs. (XLSX 11 KB)

Additional file 11: Table S8: Blast search result of primers in Table S7 using BGI rice genome sequence. (XLSX 12 KB)

Authors’ original submitted files for images

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 2.0 International License (https://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and permissions

About this article

Cite this article

Jung, KH., Gho, HJ., Giong, HK. et al. Genome-wide identification and analysis of Japonica and Indica cultivar-preferred transcripts in rice using 983 Affymetrix array data. Rice 6, 19 (2013). https://doi.org/10.1186/1939-8433-6-19

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/1939-8433-6-19

Keywords