JXB Advance Access published online on May 23, 2008
Journal of Experimental Botany, doi:10.1093/jxb/ern115
RESEARCH PAPER |
Transcript profiling of Zea mays roots reveals gene responses to phosphate deficiency at the plant- and species-specific levels
1Departamento de Ingeniería Genética de Plantas, Centro de Investigación y de Estudios Avanzados, Campus Guanajuato, PO BOX 629, Irapuato Guanajuato, México 36821
2Laboratorio Nacional de Genómica para la Biodiversidad, Centro de Investigación y de Estudios Avanzados, Campus Guanajuato, PO Box 629, Irapuato Guanajuato, México 36821
* To whom correspondence should be addressed. E-mail: lherrera{at}ira.cinvestav.mx
Received 24 January 2008; Revised 4 March 2008 Accepted 27 March 2008
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Maize (Zea mays) is the most widely cultivated crop around the world; however, it is commonly affected by phosphate (Pi) deficiency in many regions, particularly in acid and alkaline soils of developing countries. To cope with Pi deficiency, plants have evolved a large number of developmental and biochemical adaptations; however, for maize, the underlying molecular basis of these responses is still unknown. In this work, the transcriptional response of maize roots to Pi starvation at 1, 3, 6, and 10 d after the onset of Pi deprivation was assessed. The investigation revealed a total of 1179 Pi-responsive genes, of which 820 and 363 genes were found to be either up- or down-regulated, respectively, by 2-fold or more. Pi-responsive genes were found to be involved in various metabolic, signal transduction, and developmental gene networks. A large set of transcription factors, which may be potential targets for crop breeding, was identified. In addition, gene expression profiles and changes in specific metabolites were also correlated. The results show that several dicotyledonous plant responses to Pi starvation are conserved in maize, but that some genetic responses appear to be more specific and that Pi deficiency leads to a shift in the recycling of internal Pi in maize roots. Ultimately, this work provides a more comprehensive view of Pi-responses in a model for economically important cereals and also sets a framework to produce Pi-specific maize microarrays to study the changes in global gene expression between Pi-efficient and Pi-inefficient maize genotypes.
Key words: Abiotic stress, maize, microarrays, phosphate, root
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Maize (Zea mays) is the most widely cultivated crop constituting a major component in the diet of many developing countries and considered as one of the crops with most biotechnological potential for energy production and other industrial applications (McLaren, 2005).
Low Pi availability is one of the major constraints for maize production worldwide, since this crop is largely grown in areas in which soils with low Pi availability are predominant, such as the acid soils of tropical and subtropical regions and the calcareous soils of temperate regions. These soils account for more than half of the area cultivated with maize worldwide (Fairhurst et al., 1999; Fixen, 2002; Leakey et al., 2006; Pingali and Pandey, 2001). Pi availability is critical in the early developmental stages (Barry and Miller, 1989; Hajabbasi and Schumacher, 1994; Plénet et al., 2000) and, therefore a direct constraint for maize production, particularly under agricultural conditions where intensive fertilization is not affordable (Lynch, 1995).
Plants have evolved a wide array of molecular and biochemical adaptive strategies to optimize Pi uptake and utilization that include the release of soil Pi from organic and inorganic sources that are not readily available for plant uptake, the employment of high affinity Pi transporters (Raghothama, 1999), the modification of root system architecture to increase the exploratory capacity and reach Pi-rich patches in the soil (Hodge, 2004; López-Bucio et al., 2002), the establishment of symbiotic relations with arbuscular-mycorrhizal fungi (Bucher, 2007), and the recycling and mobilization of internal Pi as well as the optimization of the exploitation of a wide range of structural and metabolic compounds (Duff et al., 1994; Essigmann et al., 1998; Theodorou and Plaxton, 1993; Usuda and Shimogawara, 1991).
Although macro/microarray studies have contributed enormously to demonstrate the transcriptional regulation of some genes related to biochemical adaptations and an integral reprogramming of major metabolism under Pi-starvation, i.e. carbohydrate mobilization, nitrate assimilation, lipid recycling, and secondary metabolism in plants (Wasaki et al., 2003; Misson et al., 2005; Hernandez et al., 2007; Morcuende et al., 2007), the impact of Pi-starvation on the expression of genes encoding proteins that mediate such pathways in roots remains to be determined. Although previous analyses have also shown that nitrogen metabolism is modified mainly through the down-regulation of nitrate assimilation, protein synthesis (Misson et al., 2005; Morcuende et al., 2007) and changes in particular metabolite contents (including some amino acids and sugars; Hernandez et al., 2007; Morcuende et al., 2007), little is known about the impact of Pi starvation on the equilibrium between carbohydrate and nitrate pools as mediated by amino acid metabolism. Moreover, the knowledge of Pi-starvation responses in economically important crops is still limited since most studies have been done using Arabidopsis thaliana (Arabidopsis) as a model system. In the case of monocotyledonous, only a partial characterization of transcriptional responses to Pi starvation in rice (Oryza sativa) has been carried out (Wasaki et al., 2003, 2006). Despite the economic and nutritional importance of maize, little is known about the molecular adaptations to Pi starvation in this species. Li et al. (2007), using a proteomic approach with root tips, reported changes in some specific proteins in response to long-term Pi-starvation treatment, indicating modifications in the alteration of the balance of carbohydrate [glycolisis and tricarboxylic acid (TCA) cycle mainly], protein, nucleotide, and secondary metabolism (Li et al., 2007). Nevertheless, a detailed analysis of the effects of Pi-starvation on the expression of maize genes involved in different metabolic pathways is still lacking.
The wide variety of modifications in metabolic, developmental, and global gene expression observed in Pi-deprived plants show that the Pi-deficiency response in plants is quite complex and probably mediated by several local and systemic signalling pathways (Franco-Zorrilla et al., 2004). Among the transcription factors (TFs) involved in low Pi-responses in Arabidopsis, the MYB transcription factor PHR1 stands out as a central regulator of downstream starvation-induced genes (Rubio et al., 2001). Further analyses in Arabidopsis have contributed to identify additional TFs involved in the Arabidopsis response to Pi starvation (Hammond et al., 2003; Wu et al., 2003; Misson et al., 2005; Devaiah et al., 2007a, b; Morcuende et al., 2007). For monocotyledons, the current information of TFs that mediate the mechanisms behind Pi starvation responses is scarce. To date, only the gene orthologues to SCARECROW 8 (SCR 8; Wasaki et al., 2003) and OsPTF1 (Yi et al., 2005) have been identified in rice as TF genes responsive to Pi starvation. Interestingly, OsPTF1 overexpression conferred enhanced tolerance to Pi-starvation (Yi et al., 2005). These latter results highlighted the importance of research carried out directly on economically important crop species since Arabidopsis genes do not always represent the respective orthologue genes in cereal genomes. Although the identification of TFs involved in the Pi-deficiency response could be important targets for breeding, no TFs responsive to Pi-starvation have been identified in maize.
In this work, a genomic-level approach was undertaken using a Pi-starvation-tolerant genotype of maize L3x228-3 in order to identify Pi-responsive genes and to obtain an overview of molecular modifications of different biochemical, cellular, and developmental processes which occur as a result of Pi-starvation in maize roots. Transcriptional profiling of maize roots was carried out by analysing global gene expression alterations, at four time points to identify both early and late Pi responsive genes by employing an oligonucleotide microarray platform representing about 56 600 maize genes. As a result, a comprehensive view of Pi starvation responses in maize roots was obtained. The robustness of this study permitted us to identify (i) genes involved in biochemical processes temporarily affected by Pi starvation and a large set of TFs differentially regulated by Pi availability, and (ii) a comprehensive catalogue of differentially expressed genes for detailing novel and conserved virtual metabolic pathways affected by Pi starvation in plants. In addition, the impact of gene expression modifications as a result of Pi-starvation on metabolite profiles was corroborated by determining the changes in lipid composition and anthocyanin content.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Plant material and growth conditions
For all experiments, the P-utilization efficient hybrid Zea mays L3x228-3 (kindly supplied by EMBRAPA, Brazil) was used. Sterile seeds without endosperm were germinated in 0.1x MS medium in a growth room at 28 °C with a 16/8 h light/dark cycle for 3 d under sterile conditions followed by 2 d of growth in pots with perlite as a solid support. Seedlings were then carefully transferred into a hydroponic system for 5 d in individual pots with 3.0 l of a nutrient solution at pH of 5.5 [NH4NO3 0.3 mM, Ca(NO3)2 0.75 mM, CaCl2 0.25 mM, MgSO4 1 mM, K2SO4 0.5 mM, H3BO3 46 µM, MnSO4 9 µM, ZnSO4 0.8 µM, Na2MoO4 0.8 µM, CuSO4 0.3 µM, FeEDTA 75 µM, and 200 µM Ca(H2PO4)2 as a source of P. After 5 d of adaptation, seedlings were transferred to a fresh nutrient solution containing optimal (0.2 mM) or limiting (0.002 mM) P. Calcium in low-P solutions was added as CaCl2 to a final concentration of 0.45 mM. Roots and shoots were collected separately at 1, 3, 6, and 10 d after the onset of stress (AOS).
Experimental design and microarray platform
For microarray analyses, a dye balanced modified loop design was implemented. Four biological replicates representing each sampling point were obtained by pooling the whole root system of eight randomly chosen plants. This experiment involved a total of 16 sets of microarray hybridizations, including direct and dye swap comparisons between treatments as well as across time points for the same treatment. This design allowed us to determine differences in gene expression between P-depleted and control roots (P availability effect) and whether the differences were time dependent (Pxtime effect). The Maize Oligonucleotide Array (MOA) from www.maizearray.org was used to carry out this study. The MOA contains about 57 000 individual spots on two slides (A and B) and putatively contains all maize genes identified to date. Array annotation and composition is available at www.maizearray.org.
RNA isolation, labelling, hybridization, and image processing
Total RNA was isolated from roots using the Trizol reagent (Invitrogen) and re-purified with the RNeasy kit (Qiagen) following the manufacturer's instructions. Purified total RNA was then labelled according to the protocols recommended at www.maizearray.org. Briefly, for each treatmentxtime combination, four biological replicates were used for probe synthesis. For each sample, 1.5 µg total RNA were amplified in the presence of aminoallyl-dUTP (Ambion) using the Aminoallyl Message Amp II kit (Ambion). Resulting amplified RNA probes were further labelled with fluorescent Cy3 and Cy5 dyes (Amersham). The fluorescent dye-labelled probes were then purified using RNeasy columns (Qiagen). For hybridization, probes were mixed, concentrated by precipitation, and resuspended in the hybridization solution (50% formamide, 5x SSC, 0.1% SDS, 0.4 µg µl–1 tRNA, and 0.2 µg µl–1 Salmon Sperm DNA) for 14 h. Slides were washed for 5 min in each of the following solutions: (i) 2x SSC, 0.1% SDS/42 °C, (ii) 0.1x SSC/RT, and two final washes with 0.05x SSC/RT.
Slides were scanned with an Axon GenePix 4100 scanner at a resolution of 10 µm adjusting the laser and gain parameters to obtain similar levels of fluorescence intensity in both channels. Spot intensities were quantified using Axon GenePix Pro 5.1 image analysis software. The mean of the signals and the median of backgrounds were used for further analysis. The design, protocols, and microarray data can be found at the Zea mays Microarray Gene Expression Database (ZEAMAGE), www.maizearray.org/maize_study.shtml.
Normalization and data analysis
Raw data were imported into the R 2.2.1 software (http://www.R-project.org). Background correction was done using the RMA algorithm (Irizarry et al., 2003) and normalization of the signal intensities within slides was carried out using the ‘printtiploess’ method (Yang et al., 2002) using the LIMMA package (Smyth et al., 2003, at www.bioconductor.org). Normalized data were log2 transformed and then fitted into mixed model ANOVAs (Wolfinger et al., 2001; Gibson and Wolfinger, 2004) using the Mixed procedure (SAS 9.0 software, SAS Institute Inc., Cary, NC, USA) with two sequenced linear models considering as fixed effects the dye, time, Pi-treatment, and timexPi-treatment. Array and arrayxdye were considered as random effects. The Type 3 F-tests and P-values of the timexP-treatment and P-treatment model terms were explored and significance levels for those terms were adjusted for by the False Discovery Rate (FDR) method (Benjamini and Hochberg, 1995). Estimates of the expression differences were calculated using the mixed model. Based on these statistical analyses, the spots with tests with an FDR less than or equal to 5% and with changes in signal intensity between Pi-depleted and control roots of 2.0-fold or higher were considered as differentially expressed.
Real-time quantitative RT-PCR (qRT-PCR)
Genes whose expression was considered as regulated by Pi-deficiency in the microarray analysis were selected with the aims of both validating the expression patterns found and also to gain further biological information through selecting only annotated genes. Genes known to be Pi-deficiency-regulated in other plant species were also included. Primer design (Tm, 60–65 °C) was performed according the guidelines recommended in the Primer Express Software, Version 3 (Applied Biosystems) using as template the original target sequences from which the oligonucleotides printed in the array were designed. Oligonucleotide sequences for qRT-PCR are shown in Supplementary Table S1 at JXB online.
Total RNA for qRT-PCR was isolated from 1 g frozen root tissue based on the protocol of the Trizol reagent (Invitrogen) and further purified using Qiagen RNeasy columns according to the manufacturer's protocol (Qiagen). cDNA was first synthesized using 10 µg total RNA with SuperscriptIII reverse transcriptase (Invitrogen), according to the manufacturer's instructions and used for performing qRT-PCR (7500 Real Time PCR System, Applied Biosystems). qRT-PCR of POLIUBIQUITIN2 (UBQ2, TC305418) was performed for normalization. SYBR Green PCR Master Mix was used for the PCRs according to the manufacturer's protocol. Gene expression was normalized to that of the control UBQ2 gene by subtracting the CT value of UBQ2 from the CT value of the gene of interest. –P to +P average expression ratios were obtained from the equation (1+E)2
CT where CT represents CT(–P)–CT(+P), and E is the PCR efficiency according to protocol reported by (Czechowski et al., 2004).
Metabolite determinations
Total anthocyanin content was measured from about 300 mg of root tissue from both low-Pi and optimal growth conditions at 6 d and 10 d AOS. The optical density from acidified methanolic extracts (5% HCl) was measured and corrected by using OD530nm–0.25 OD657nm as described in (Pietrini et al., 2002) using a Beckman DU® 650 spectrophotometer. Anthocyanin content was calculated as cyanidine-3-glucoside by using 29 600 (L mol–1 cm–1) as the molar extinction coefficient and 449.2 as the molecular weight.
Lipids from root tissues were extracted by homogenization in a chloroform solution and isolated by the TLC technique by employing a solvent system of acetone/toluene/water (91:30:7, by vol.; Welti et al., 2002). For quantification, individual lipids were isolated from TLC plates and used to prepare fatty acid methyl esters. The corresponding methyl esters were quantified by GC-MS with myristic acid as an internal standard control (Hartel et al., 2000). Two independent replicates of each sample were analysed with similar results (R2=0.966).
Clustering, functional annotation, and metabolic pathway analysis
For comparison of transcriptome profiles, the significant genes according to the selected parameters (FDR <0.05 and Fold ±2) were clustered and visualized with the standard correlation measure using Genespring 7.0 software (Silicon Genetics, Redwood City, CA). To visualize the gene lists in the figures, the Excel (Microsoft) FiRe 2.2 macro (Garcion et al., 2006) was used. To gain further information of the biological relevance of the differentially expressed transcripts, BLAST alignments of the consensus sequences were performed from which the oligonucleotide array was designed against the TIGR Plant Transcript Assemblies, release 02/06/2007 (Childs et al., 2006) and the NCBI non-redundant database release (03/08/2007). These BLAST results provided us with a new annotation for each sequence based on protein similarity. If additional information was found using an E-value 
1.0E–10 as a threshold, the original annotation was replaced.
As the functional annotation of maize sequences is still limited, the functional classification implemented in the mapping files that structure the Arabidopsis genes from the Affymetrix ATH1 array into distinct metabolic and cellular processes from the MapMan program (Thimm et al., 2004) was used. To associate the maize differentially expressed genes with a functional annotation, a BLAST alignment was performed against the TAIR Arabidopsis database release 6.0 (www.arabidopsis.org), the annotations of the mapping files for the best match to the TAIR protein database (with at least an E-value of 1.0E–10) were applied to the corresponding maize orthologue, in the case of double or more category assignments, the most informative was chosen. When possible, the genes without a metabolic or cellular annotation were further assigned a putative function based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway database (Ogata et al., 1999).
To show the comparisons between the maize and Arabidopsis P-responsive genes (see Supplementary Fig. S2 at JXB online), the MapMan software was used (Thimm et al., 2004). The averaged signals of each time point for differentially expressed sequences were expressed relative to the control for Arabidopsis –P to +P (see Supplementary Table S4 in Morcuende et al., 2007), Arabidopsis root (see Supplementary Table 7 in Misson et al., 2005), and maize root (day 6 AOS) experiments. Ratios were expressed in a log2 scale for importing into the software and then drawn into the diagrams with a false colour scale for each gene in an experiment. Each square represents a single gene.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
The Pi deficiency treatment consisted in growing maize plants on media containing sufficient (0.2 mM) or limiting (0.002 mM) Pi levels. In order to examine changes in global gene expression that may precede the appearance of Pi-starvation symptoms but also to identify late responsive genes, roots were collected from plants grown for 1, 3, 6, and 10 d AOS. Seedlings at the onset of the treatment had one fully developed leaf and the primary root system was already completely developed, with shoot-borne roots at the first and second phytomers. Phenotypic differences between Pi sufficient and depleted plants were visible from day 6 AOS (Fig. 1). Pi-depleted plants presented anthocyanin pigmentation in the oldest leaves with decreased shoot weight (Fig. 1A) and leaf area. A delay in the appearance of shoot-borne roots was also observed (Fig. 1B). By day 10 AOS, Pi-depleted plants also presented longer shoot-borne roots (Fig. 1B) and a higher anthocyanin content in roots and leaves. Despite the decrease in shoot weight in Pi-depleted plants, a relative maintenance of C translocation to roots was observed, as reflected by the increase of the root-to-shoot weight ratio, as compared to Pi-sufficient plants (Fig. 1C). Pi deficiency also led to a significant decrease in P-accumulation in both shoot and roots starting in the first days of Pi starvation (Fig. 1D).
|
Transcriptome profiling of root response to Pi deficiency
A microarray platform containing about 56 600 maize gene oligonucleotides was used, so that the transcriptome profiling of maize roots under Pi starvation using most of the currently available genes and potentially covering most of the maize transcriptome was determined. The differences in gene expression between Pi-depleted versus Pi-sufficient roots (the overall P availability effect) were identified and also the differences caused by the P availability by time interaction (TimexP effect). According to the stringency levels (FDR
0.05 and Fold ±2), a total of 1179 genes showed differential expression in at least one of the four sampled time points (see Supplementary Table S2 at JXB online). Over time, for both induced and repressed genes, the fold-change in expression of Pi-responsive genes gradually increased from the first to the sixth day AOS. At the sixth and tenth days AOS, the greatest fold changes in gene expression were observed (Fig. 2A). Principal component analysis showed that this was the main trend and accounted for 86% of the differentially expressed genes. Common differentially expressed genes were found mainly between days 6 and 10 AOS, whereas only nine induced and four repressed genes were found to be differentially expressed over all four sampled time points (Fig. 2B) indicating that at early time points (days 1 and 3 AOS) the response is represented by a distinct set of differentially expressed genes compared with those identified at later time points.
|
Annotation and functional classification
As the functional annotation of maize sequences is still limited, functional classification was implemented by comparison (using an expected value of 1E-10 as threshold) with the Arabidopsis genomic data implemented in the MapMan program (Thimm et al., 2004). Remaining sequences without annotation were further classified based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway database (Ogata et al., 1999). Five hundred and sixty genes were automatically annotated with MapMan and 214 sequences were further manually annotated using the KEGG database, however, 407 sequences had no significant correlation to any annotated database (see Supplementary Table S2 at JXB online). Among the annotated genes, the signalling, hormone, and transcription related genes accounted for about 10% of the Pi-responsive genes, whereas both N metabolism (including that of amino acids and proteins) and transport systems were the most affected processes since they each accounted for around 7% of the Pi-responsive genes (see Supplementary Fig. S1 at JXB online).
Table 1 shows the 50 most strongly differentially expressed genes. Among them, a strong induction of Pi-starvation-related genes was observed (encoding Pi transporters, a putative acid phosphatase, and a SPX domain protein), for which differential expression was initiated at day 3 AOS. These results suggest that maize roots sense Pi-starvation prior to the development of visible symptoms and developmental modifications. Interestingly, four genes related to hormone/signalling showed strong differential expression at day 1 AOS (Table 1), suggesting that genes involved in sensing/signal transduction mechanisms are activated rapidly after the onset of the stress.
|
Expression analyses by real-time quantitative RT-PCR (qRT-PCR)
Differential expression was corroborated for 18 genes using qRT-PCR. Ten orthologues of genes previously reported as Pi-responsive in other plant species and eight that were not previously known to be regulated by Pi were included. The expression pattern observed in the microarray experiments was confirmed for the genes analysed (Fig. 3A); however, several genes presented a higher level of induction when determined by qRT-PCR than estimated by microarray analyses (Fig. 3B). Notably, genes encoding a sulpholipid synthase and a putative acid phosphatase showed the highest values of expression (fold values of 455 and 541, respectively). This suggests that microarray experiments are underestimating the levels of changes in expression. Similar quantitative differences have also been reported previously (Morcuende et al., 2007).
|
Metabolite determinations
To evaluate the degree to which changes in gene expression in maize roots are reflected in the corresponding metabolite products, lipid composition and anthocyanin content was measured at 6 d and 10 d AOS, because these metabolic pathways were among the most affected in both the level of differential expression and the number of affected genes. Pi-deprived maize roots showed a decrease of 30–50% in phospholipid content [phosphatidylglycerol (PG), phosphatidylethanolamine (PE), and phosphatidylcholine (PC)]. By contrast, an increase of 2–4-fold was determined for sulphoquinovosyldiacylglycerol (SQDG), monogalactosyldiacylglycerol (MGDG) and digalactosyldiacylglycerol (DGDG). Total anthocyanin content was also significantly increased in the roots of Pi-deprived maize plants (Table 2). These determinations demonstrate that the changes in expression are truly reflected at the metabolite level.
|
Metabolic pathway analysis
Analysis of the kinetics of gene induction and repression during Pi-deprivation led to the identification of genes involved in a large number of metabolic, developmental and signalling pathways (see Supplementary Fig. S1 at JXB online). We will focus on describing the most prominent specific pathways and categories affected by Pi starvation in maize roots. In addition, those responses are illustrated through drawing virtual metabolic pathways.
Carbon metabolism
In addition to previously reported Pi-starvation-responsive pathways, the list of differentially expressed genes was enriched in genes involved in carbohydrate, amino acid, and lipid metabolism (see Supplementary Fig. S2 at JXB online).
In terms of genes related to photosynthate partitioning it was found that the sucrose synthase 2 gene (SUS2) was up-regulated at day 1 AOS suggesting that an increase in the C supply toward the roots is one of the first responses of maize to Pi-deprivation (see Supplementary Fig. S3 at JXB online). Starch synthesis appears to be enhanced since two genes encoding ADP-glucose pyrophosphorylase (AGPase) and one encoding a glucose 6P/P translocator are induced at days 6 and 10 AOS, although a gene encoding the enzyme that participates in the last step of starch degradation, 
-glucosidase 1, is also induced. Glycolysis is modified through bypassing reactions that require ATP, as reflected by the increased expression of genes encoding phosphoglycerate mutase (PGM), phosphoenol pyruvate carboxylase (PEPCase), and PEPcase kinase (PEPK), probably to supply the carbon skeletons necessary for the next intermediary reactions in C metabolism (i.e. the TCA cycle). Interestingly, eight PEPcase/PEPK encoding genes were induced at day 6 AOS, indicating that these enzymes may serve as key check points to direct carbon flow under Pi limitation in maize (see Supplementary Fig. S3 at JXB online). In addition, it was found that two glyoxysomal malate synthase (MS) encoding genes were significantly repressed; indicating that the carbon flow from β-oxidation (Cornah et al., 2004) is attenuated in maize roots (see Supplementary Fig. S3 at JXB online).
Nitrogen metabolism
A significant repression of three nitrate reductase (NR) genes and one glutamine synthetase (GS1) encoding gene starting at day 3 AOS was observed in maize roots (Fig. 4). The effect of Pi-starvation on N metabolism was also reflected in a dynamic modification of both amino acid catabolism and anabolism. A degree of specificity for synthesis and/or degradation of amino acids during the Pi-deprivation progress was observed. Ten genes encoding enzymes involved in degradation of Pro, Tyr, Leu, Val, Trp, Gly, and Ser were found to be down-regulated, in parallel with an increase in the transcript levels of genes involved in the synthesis of Trp, Phe, Pro, Ser, and Cys. In addition, a constant down-regulation of genes involved in the synthesis of Leu, Ile, Val, Tyr, Hys, and Gly was observed (Fig. 4).
|
10 is shown in red, a ratio Protein synthesis and activation was less affected by Pi limitation: only one gene encoding the translation initiation factor 2B and one for a proline-tRNA ligase were down-regulated starting at day 1 AOS, whereas four genes (encoding a tRNA synthetase, a translation factor-like protein, and two 60S acidic P2 and S6 kinase-homolog ribosomal proteins) were up-regulated at day 6 AOS. The expression pattern of several genes encoding proteins related to post-translational modifications were also altered (see Supplementary Fig. S4 at JXB online).
Lipid metabolism
Figure 5 shows changes in the expression level of genes related to the lipid metabolism. In Arabidopsis, recycling of internal Pi involves sulpho/galactolipid synthesis and phospholipid degradation (Hammond et al., 2004). Such degradation is mediated by a combination of phospholipases C (PLC) and D (PLD; Cruz-Ramirez et al., 2006). Despite the fact that the microarray contained 36 oligos targeting PLD genes and five targeting PLC genes, no accumulation of these transcripts was observed. Instead, a strong induction was found of one gene encoding a phospholipase A2 and six glycerophosphodiesterases (GPDEs), all involved in phospholipid degradation (Fig. 5).
|
The impact of Pi limitation on lipid metabolism in maize roots is further reflected by the altered expression of 30 additional sequences corresponding to another set of unspecified lipases/hydrolases as well as other proteins related to fatty acid (FA) synthesis, β-oxidation, PA synthesis, TAG degradation, and lipid transfer activities (Fig. 5A). In total, the modification in expression of 59 putative lipid-related genes was identified. Under long-term Pi starvation, maize roots seem also to be using P-inositol as a source of Pi (Fig. 5B). An increase in the transcript abundance of genes encoding inositol monophosphatase, phosphoinositide phosphatase, inositol-1,4, 5-trisphosphate-5-phosphatase-like proteins, and two myo-inositol-3-phosphate synthase enzymes was observed (Fig. 5B). The increased expression of these genes suggests that P-inositol could be degraded to release Pi to support other metabolic activities.
Transport and other enzyme families
Further metabolic modifications under Pi limitation are reflected by the activation and repression of several genes encoding members of protein families involved in secondary metabolism, such as several members of the large gene families encoding cytocrome P450, UDP-glucosyltransferase, peroxidases, and different oxidases (see Supplementary Table S2 at JXB online). These alterations are also observed in Arabidopsis, white lupin, and rice plants subjected to Pi starvation, thus indicating that their response may be ubiquitous in higher plants. Similarly, transport systems were strongly affected under Pi deficiency. As shown in Table S2 (see Supplementary Table S2 at JXB online), alterations in the transcript level of phosphate, sulphate, Fe, and ABC transporters as well as phytosiderophores, sugars, oligo-peptides, and aquaporin encoding genes was identified.
Secondary metabolism
A significant modulation of anthocyanin-related genes occurred under Pi starvation in maize roots, including the induction of Bronze 2 (Marrs et al., 1995), a gene encoding a transporter of anthocyanins to the vacuole (Fig. 6A). In addition, several putative genes related to the phenylpropanoid pathway and thus lignin biosynthesis were identified as up- or down-regulated (Fig. 6B). Differential changes were identified in the transcript level of genes encoding phenylalanine amonnia-lyase (PAL), 4-coumarate-CoA ligase (4CL), N-hydroxycinnamoyl/benzoyl transferase (HCT), caffeoyl-CoA 3-O-methyltransferase (CCoAOMT), caffeic acid O-methyltransferase (COMT), cinnamoyl-CoA reductase (CCR), cinnamyl alcohol dehydrogenase (CAD), and a member of the laccase gene family. Most of these were up-regulated, although, three HCT, two COMT, one CCR-encoding genes and one laccase-encoding gene were down-regulated (Fig. 6B). In agreement with our results, Uhde-Stone et al. (2003) also reported an increase in the transcript level of the gene encoding CCoAOMT and of the laccase in proteoid roots of white lupin under Pi starvation (Uhde-Stone et al., 2003). These results allow us to draw a comprehensive panorama of lignin synthesis modifications in plants under Pi starvation (Fig. 6B).
|
Pi starvation and other stress responses
Among the responses identified in this work, the up-regulation of Pi-transport and Pi-recycling related genes was the most robust, extensive, and constant adaptation to Pi limitation (Fig. 7). Eight putative Pi-transporters, 26 different phosphatases, four ribonucleases, and 12 putative Mt4-like/SPX-domain encoding genes were induced almost throughout the whole experiment. In addition, 18 differentially regulated genes related to biotic and 13 to abiotic stress were identified. Other genes significantly modulated by Pi-starvation were those of the nodulin family (20 genes; see Supplementary Table S2 at JXB online). Taken together, these results suggest that intrinsic and common interactions among cell response pathways protect cells against toxic effects produced by different stress conditions.
|
Signalling/transcription related responsive genes
Of great interest is the finding that a large set of genes related to transcriptional regulation and hormone signalling was identified. The transcriptional regulation of the Pi-starvation responses in maize roots is reflected by the altered expression of 42 TFs (Table 3), of which 28 are induced and 14 repressed. These include bHLH, zinc finger, and leucine zipper families. An orthologue of SCR and one belonging to the MYB-family were the earliest TFs showing differential expression, starting at day 3 AOS. Likewise, significant up- or down-regulation of calcium, phosphoinositol, G-class and light signalling-proteins was observed. Eleven genes encoding protein kinases and receptor protein kinases were also differentially expressed. Among the hormone-related genes, a set of transcripts related to brassinosteroids, gibberellins, ethylene and cytokinins, abscisic acid and auxins were also up- or down-regulated (Fig. 8).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Several reports on the identification of maize Pi uptake/use efficient genotypes were published several years ago (DaSilva and Gabelman, 1992; Hajabbasi and Schumacher, 1994; Gaume et al., 2001; Fan et al., 2003; Zhu and Lynch, 2004; Corrales et al., 2007), however, the exploration of maize biodiversity as a potential source of target genes for crop breeding remained to be done. A previous study suggested that L3x228-3 possess an important P-uptake capacity (Corrales et al., 2007). The results presented here show that L3x228-3 is able to maintain root mass and increase root length even under low internal P concentrations (Fig. 1). In addition, it is shown that L3x228-3 displays a wide set of biochemical and transcriptome responses to P deprivation in roots.
The robustness of the experimental strategy permitted the expression kinetics of 1179 genes over four time points to be identified. Of these, 774 genes are potentially associated with or have a potential role in responses to P-status. In addition, we have uncovered 407 P-responsive maize genes with no significant hit to any annotated database, whose characterization could lead to the discovery of novel pathways, some of which could be monocotyledonous-specific, involved in mechanisms to cope with P starvation. An overall comparison of the proteome results reported by Li et al. (2007) with the results presented here shows that members of around 35% of the gene families identified by Li et al. (2007) as phosphate-responsive are also regulated at the transcript level (see Supplementary Table S3 at JXB online), however, for the other differentially accumulated proteins no statistically significant difference in transcript level between the controls and Pi-deprived maize plants were determined. Post-transcriptional regulation, as well as technical, genotypical, and experimental variations, may explain those differences. Such contrasts have previously been highlighted when comparing transcript expression levels between different Pi-related microarrays in Arabidopsis (Morcuende et al., 2007).
The results presented here also show that about 33% of the genes identified as Pi-responsive in maize do not have a significant match with an orthologue in the Arabidopsis genome, emphasizing the importance of global gene expression studies directly on crops of economic importance. However, 210 of the maize differentially regulated transcripts matched with 148 Arabidopsis orthologues that were also reported as Pi-starvation-regulated (see Supplementary Table S4 at JXB online), showing that there are many conserved molecular responses to P availability between maize and Arabidopsis.
Pi starvation integrates P, C, and N metabolism in maize roots
Under Pi limitation, L3x228-3 line showed higher Pi-uptake compared with inefficient lines (HS 2841x5046; Corrales et al., 2007). The results presented here suggest that the enhanced Pi uptake may be related to the increased transcription of genes encoding Pi transporter, phosphatase and ribonuclease (Fig. 7).
In addition, the identification of altered expression of several sugar-related genes, including those involved in photosynthate partition and glycolysis in maize roots under Pi starvation, supports the evidence of direct cross-talk between sugar metabolism and P stress in plants, as reported by Tesfaye et al. (2007) who noted a dark/light-directed expression of several sugar-sensing and metabolism genes in response to Pi stress in proteoid roots of white lupin (Tesfaye et al., 2007).
Organic acid synthesis and excretion have been documented in maize as a response to Pi starvation (Gaume et al., 2001). A significant induction of several genes encoding PEPCase was observed (see Supplementary Fig. S3 at JXB online), whose activity is needed to synthesize malate and citrate. However, no alterations in the transcript level of either malate dehydrogenase or malate/citrate transporters was observed, suggesting that in L3x228-3 roots PEPcase is the limiting step for organic acid synthesis or that the increased PEPcase expression may instead be supporting other C needs in the cell such as providing carbon to increase citrate/malate exudation. In this sense, it is possible that maize roots may be redirecting PEPcase products and other C pathways to provide C skeletons for amino acid synthesis. In fact, six genes involved in Trp, Phe, Pro, Ser, and Cys synthesis were induced (Fig. 4). Besides their structural importance, amino acid synthesis may control either directly or indirectly various aspects of plant growth: Trp is a precursor in auxin synthesis (Ljung et al., 2005), Phe is the key substrate for many secondary plant metabolites including anthocyanins and lignin (Fig. 6; Noel et al., 2005), Pro participates directly in cellular osmotic protection (Ashraf and Foolad, 2007), and Ser and Cys synthesis is part of the sulphate assimilation pathway (Kopriva, 2006). Although further research is needed in maize roots, synthesis of such amino acids fits into the framework of Pi responses, given that, under Pi starvation, alterations in root system architecture are probably mediated by auxin (Franco-Zorrilla et al., 2004), a substantial increase of anthocyanins and lignin occurs, and sulpholipid synthesis may demand an increase in sulphate assimilation. The latter observation is further supported by the induction of at least one member of the sulphate transporter gene family (see Supplementary Table S2 at JXB online).
The ability of maize roots to preserve N- and C-containing metabolites under Pi starvation is notable. An extensive repression of genes involved in amino acid degradation was found together with modifications in both protein synthesis and degradation (Fig. 4; and see Supplementary Fig. S4 at JXB online). These results reflect a contrasting response between maize and Arabidopsis roots. In the latter species, modifications in the transcript abundance of genes involved in amino acid metabolism are limited, whereas general decreases in the expression of genes involved in amino acid and protein synthesis in concert with increases in the expression of amino acid and protein degradation genes were reported (see Supplementary Fig. S2 at JXB online; Wu et al., 2003; Misson et al., 2005; Morcuende et al., 2007).
It was also found that genes involved in TAG breakdown and β-oxidation are induced upon Pi starvation. Induction of β-oxidation has been observed in sugar-starved detached maize root tips (Dieuaide et al., 1992), presumably to provide C for gluconeogenesis (mediated by the glyoxylate cycle) and also substrates for energy production in mitochondria (Baker et al., 2006). However, in maize roots under Pi starvation, the carbon flow from lipid catabolism towards gluconeogenesis may be attenuated, since the transcript levels of MS, a key enzyme of the glyoxylate cycle are significantly decreased (see Supplementary Fig. S3 at JXB online). The apparently contrasting regulation of TAG breakdown and β-oxidation with MS may indicate the turnover of membrane lipids in conjunction with the induction of FA synthesis (Fig. 5), avoiding C loss from the membrane lipid pool or to provide C intermediates to the TCA cycle for energy production during Pi starvation. These observations suggest that Pi starvation in maize induces changes in gene expression that promote flexible use of C skeletons through integrating carbohydrate, glycolysis, and TCA cycle modifications with those found in lipid metabolism. It is also worthwhile mentioning that β-oxidation mediates the activation of signalling molecules including the auxin indole acetic acid, as confirmed by the root phenotype found in mutants with β-oxidation defects (Baker et al., 2006), thus implying that such modifications may produce signals to modify root architecture under Pi starvation.
P recycling
Despite the fact that Pi recycling in maize roots involves the substitution of phospholipids by non-phospholipids, the recycling in maize is distinct from that previously reported in Arabidopsis (Misson et al., 2005; Cruz-Ramirez et al., 2006). In maize, phospholipid degradation correlates with the increased transcription of GPDE and PLA coding genes while in Arabidopsis it has been shown to be mediated mainly by PLC and D. Moreover, two phosphatidic acid phosphatase 2 encoding genes were repressed, suggesting that the acyl moiety required for glycolipid synthesis is probably produced by TAG degradation (Fig. 5) and not by the direct degradation of phosphatidic acid (PA). In fact, PA synthesis could increase under Pi starvation in maize roots, as a result of the induction of genes encoding G3PDH and G3PAT (Fig. 5). The role of PA as a signalling molecule in root hair elongation and auxin sensitivity has been demonstrated (Li and Xue, 2007), therefore an increase in PA could be influencing changes in maize root architecture in response to Pi-deprivation.
Secondary metabolism
An interesting observation from this work is that anthocyanin accumulation in roots may be co-ordinated through the regulatory leaf colour (LC) gene, a TF known to regulate the expression of genes encoding chalcone synthase and dihydroflavonol 4-reductase (Dooner et al., 1991). LC was among the most highly induced genes in Pi-deprived maize roots and is an orthologue of the rice transcription factor OsPTF1, which was reported to be involved in the tolerance to Pi starvation in rice (Yi et al., 2005). Therefore, it is possible that LC might also regulate the expression of other genes involved in Pi-deficiency responses in maize.
The observed changes in lignin biosynthesis (Fig. 6) together with the differential expression of genes encoding other cell wall-degrading proteins such as glucanases, mannosidases, callose synthase, and polygaracturonidases (see Supplementary Table S2 at JXB online) suggest a dynamic rearrangement of cell wall structure during Pi starvation in maize roots. In addition, the larger number of Pi-responsive genes of the phenylpropanoid pathway in maize roots indicates that not only lignin synthesis but also the production of a wide range of metabolites derived from this pathway might be significantly altered. Interestingly, ZRP4-like genes (encoding O-methyltransferases) were found to be either induced or repressed. The development-driven expression of ZRP4-like genes occurs preferentially in cell elongation and maturation zones of young roots (Held et al., 1993). This suggests that besides their specific metabolic role these genes may also participate in changes in root architecture, possibly to enhance nutrient conductivity in vascular cylinders.
Signalling and transcription related genes
Research in Arabidopsis has shown that the regulation of Pi-starvation responses involves a large set of signalling genes including PHR1 (Rubio et al., 2001), the At4/Mt4 family (Shin et al., 2006), miR399 (Franco-Zorrilla et al., 2007), and PHO2 (Bari et al., 2006), but also PHO/SPX/EXS domain-containing proteins (Wykoff and O'Shea, 2001; Giots et al., 2003). The presence of three PHR1-like genes in the maize genome and the up-regulation of two Mt4-like genes and eight PHO/SPX/EXS domain encoding genes in Pi-deprived maize roots (Fig. 7; Table 1), suggests that some of the mechanisms controlling Pi responses are conserved in maize and Arabidopsis. However, although previous studies have suggested that Pi-deficiency responses are well conserved in plants (Franco-Zorrilla et al., 2004), it is striking that only 10 out of 42 maize Pi-responsive TF genes (Table 3) are also differentially regulated in Arabidopsis. These results suggest that some of the common responses to Pi-deprivation between Arabidopsis and maize are mediated by different TFs that activate a similar set of genes or that the response to Pi deprivation is significantly different between these plants species, as reflected by the higher number of differential genes identified in maize in comparison to Arabidopsis.
Changes in root system architecture
The genotype selected for this study showed marked developmental changes in its root architecture under Pi limitation, including an increase in shoot-borne and lateral root length but also changes in root-to-shoot weight ratios indicating its plasticity (Fig. 1). In this context, it is notable that several hormone-related genes were differentially regulated in the root of this maize genotype, including several encoding ARF and AUX/IAA TFs. In addition, orthologues to SHORT-ROOT and SCR TFs involved in determining meristem identity and thus in root morphology in both dicotyledons and monocotyledons were also differentially regulated (Nakajima and Benfey, 2002; Lim et al., 2005). Similarly, several homologues were found of ENHANCER OF GLABRA3, TRANSPARENT TESTA1, NAC, AP1, and AP2, all reported as being involved in developmental processes such as lateral root emergence (Xie et al., 2000, 2002; Hardtke, 2006). All of these genes may be acting as intermediates in the developmental responses to Pi-starvation and the task remains to clarify their roles, considering the remarkable differences between the developmental programs in maize and Arabidopsis root systems (Hochholdinger et al., 2004).
Trehalose metabolism plays a key role in plant development (Ramon and Rolland, 2007). In Arabidopsis, crucial roles in embryo and vegetative development as well as in floral transition have been reported (Ramon et al., 2007). The maize RAMOSA3 (RA3), gene that encodes a trehalose-6-phosphate phosphatase (TPP) controls inflorescence architecture, and a mutation in this gene presents an altered branching organization (Satoh-Nagasawa et al., 2006). The differential expression in response to Pi-deprivation of three TPP encoding genes, including two related to RA3, in maize roots suggests that trehalose-related enzymes may be regulators in root developmental responses, including those triggered by P starvation in maize. Interestingly, three TPP genes are also differentially regulated in Arabidopsis roots under Pi-starvation (Misson et al., 2005).
The genes identified here and their proposed role in Pi adaptation, supported by an integral analysis of the data available from existing transcriptome, protein, and metabolite profiling experiments conducted in other species provide an opportunity to identify and functionally characterize different alleles involved in the adaptive response to Pi-deficiency and to identify candidate genes and processes whose manipulation may improve Pi-starvation tolerance in maize but also in other cereal crops. This work also provides the framework to produce Pi-specific maize arrays to study the changes in global gene expression between Pi-efficient and inefficient maize genotypes.
|
Supplementary data |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Supplementary data are available at JXB online.
Table S1. List of PCR primers used for qRT-PCR and corresponding CT values.
Table S2. Transcript levels of significant genes in the maize oligonucleotide microarray.
Table S3. Comparison of protein specificity under Pi deficit (Li et al., 2007) with the gene expression changes for the respective gene family members.
Table S4. Comparison of changes in expression levels of maize transcripts with those found for the corresponding Arabidopsis orthologues under Pi limitation.
Fig. S1. Functional classification of genes responsive to Pi-deficiency.
Fig. S2. Comparison of overall metabolic responses to Pi-deficiency reported for Arabidopsis with those found in maize roots.
Fig. S3. Effects of Pi-deficiency on the transcript accumulation of genes encoding enzymes involved in primary pathways of Carbon metabolism in maize roots.
Fig. S4. Effects of Pi-deficiency on the transcript accumulation of genes encoding enzymes involved in protein metabolism in maize roots.
|
Acknowledgements |
|---|
We would like to thank June Simpson and José Lopez-Bucio for valuable help in reviewing the manuscript. We also thank Enrique Ramirez-Chavez for lipid analyses; Susana ML Fuentes-Guerra and Flor MX Zamudio-Hernandez for qRT-PCR analysis; Liu Jia from TIGR for microarray image analysis; EMBRAPA for maize lines; David Galbraith, Vicki Chandler, and Jack Gardiner at the BIO5 Institute at the University of Arizona who produced the arrays by the Microarray Resources for Maize Research Project supported by NSF DBI 0321663. Cheryl Vanier and J Burgueño for statistical advice. This work was supported in part by grants from SAGARPA (Zea-2006) and HHMI (Grant 55003677) to LH-E.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Ashraf M, Foolad MR. Roles of glycine betaine and proline in improving plant abiotic stress resistance. Environmental and Experimental Botany (2007) 59:206–216.[CrossRef][Web of Science]
Baker A, Graham IA, Holdsworth M, Smith SM, Theodoulou FL. Chewing the fat: beta-oxidation in signalling and development. Trends in Plant Science (2006) 11:124–132.[CrossRef][Web of Science][Medline]
Bari R, Datt Pant B, Stitt M, Scheible WR. PHO2, microRNA399, and PHR1 define a phosphate-signaling pathway in plants. Plant Physiology (2006) 141:988–999.
Barry D, Miller M. Phosphorus nutritional requirement of maize seedlings for maximum yield. Agronomy Journal (1989) 81:95–99.
Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society B (1995) 57:289–300.
Bucher M. Functional biology of plant phosphate uptake at root and mycorrhiza interfaces. New Phytologist (2007) 173:11–26.[CrossRef][Web of Science][Medline]
Childs KL, Hamilton JP, Zhu W, Ly E, Cheung F, Wu H, Rabinowicz PD, Town CD, Buell CR, Chan AP. The TIGR Plant Transcript Assemblies database. Nucleic Acids Research (2006) 35:D846–D851.[CrossRef][Web of Science][Medline]
Cornah JE, Germain V, Ward JL, Beale MH, Smith SM. Lipid utilization, gluconeogenesis, and seedling growth in Arabidopsis mutants lacking the glyoxylate cycle enzyme malate synthase. Journal of Biological Chemistry (2004) 279:42916–42923.
Corrales I, Amenos M, Poschenrieder C, Barcelo J. Phosphorus efficiency and root exudates in two contrasting tropical maize varieties. Journal of Plant Nutrition (2007) 30:887–900.[CrossRef][Web of Science]
Cruz-Ramirez A, Oropeza-Aburto A, Razo-Hernandez F, Ramirez-Chavez E, Herrera-Estrella L. Phospholipase DZ2 plays an important role in extraplastidic galactolipid biosynthesis and phosphate recycling in Arabidopsis roots. Proceedings of the National Academy of Sciences, USA (2006) 103:6765–6770.
Czechowski T, Bari RP, Stitt M, Scheible WR, Udvardi MK. Real-time RT-PCR profiling of over 1400 Arabidopsis transcription factors: unprecedented sensitivity reveals novel root- and shoot-specific genes. The Plant Journal (2004) 38:366–379.[CrossRef][Web of Science][Medline]
DaSilva AE, Gabelman WH. Screening maize inbred lines for tolerance to low-P stress condition. Plant and Soil (1992) 146:181–187.[CrossRef][Web of Science]
Devaiah BN, Karthikeyan AS, Raghothama KG. WRKY75 transcription factor is a modulator of phosphate acquisition and root development in Arabidopsis. Plant Physiology (2007a) 143:1789–1801.
Devaiah BN, Nagarajan VK, Raghothama KG. Phosphate homeostasis and root development in Arabidopsis is synchronized by the zinc finger transcription factor ZAT6. Plant Physiology (2007b) 145:147–159.
Dieuaide M, Brouquisse R, Pradet A, Raymond P. Increased fatty acid beta-oxidation after glucose starvation in maize root tips. Plant Physiology (1992) 99:595–600.
Dooner HK, Robbins TP, Jorgensen RA. Genetic and developmental control of anthocyanin biosynthesis. Annual Review of Genetics (1991) 25:173–199.[CrossRef][Web of Science][Medline]
Duff SMG, Sarath G, Plaxton WC. The role of acid phosphatase in plant phosphorus metabolism. Physiologia Plantarum (1994) 90:791–800.[CrossRef]
Essigmann B, Guler S, Narang RA, Linke D, Benning C. Phosphate availability affects the thylakoid lipid composition and the expression of SQD1, a gene required for sulfolipid biosynthesis in Arabidopsis thaliana. Proceedings of the National Academy of Sciences, USA (1998) 95:1950–1955.
Fairhurst T, Lefroy R, Mutert E, Batjes N. The importance, distribution and causes of phosphorus deficiency as a constraint to crop production in the tropics. Agroforestry Forum (1999) 9:2–8.
Fan M, Zhu J, Richards C, Brown KM, Lynch JP. Physiological roles for aerenchyma in phosphorus-stressed roots. Functional Plant Biology (2003) 30:493–506.[CrossRef][Web of Science]
Fixen P. Soil test levels in North America. Better Crops (2002) 86:12–15.
Franco-Zorrilla JM, Gonzalez E, Bustos R, Linhares F, Leyva A, Paz-Ares J. The transcriptional control of plant responses to phosphate limitation. Journal of Experimental Botany (2004) 55:285–293.
Franco-Zorrilla JM, Valli A, Todesco M, Mateos I, Puga MI, Rubio-Somoza I, Leyva A, Weigel D, Garcia JA, Paz-Ares J. Target mimicry provides a new mechanism for regulation of microRNA activity. Nature Genetics (2007) 39:1033–1037.[CrossRef][Web of Science][Medline]
Garcion C, Applimath FRI, Métraux JP. FiRe and microarrays: a fast answer to burning questions. Trends in Plant Science (2006) 11:320–322.[CrossRef][Web of Science][Medline]
Gaume A, Mächler F, De Leon C, Narro L, Frossard E. Low-P tolerance by maize (Zea mays L.) genotypes: significance of root growth, and organic acids and acid phosphatase root exudation. Plant and Soil (2001) 228:253–264.[CrossRef][Web of Science]
Gibson G, Wolfinger R. Gene expression profiling using mixed models. In: Genetic analysis of complex traits using SAS—Saxton A, ed. (2004) Cary, NC, USA: SAS Press. 251–278.
Giots F, Donaton MC, Thevelein JM. Inorganic phosphate is sensed by specific phosphate carriers and acts in concert with glucose as a nutrient signal for activation of the protein kinase A pathway in the yeast Saccharomyces cerevisiae. Molecular Microbiology (2003) 47:1163–1181.[CrossRef][Web of Science][Medline]
Hajabbasi MA, Schumacher TE. Phosphorus effects on root growth and development in two maize genotypes. Plant and Soil (1994) 158:39–46.[CrossRef][Web of Science]
Hammond JP, Bennett MJ, Bowen HC, Broadley MR, Eastwood DC, May ST, Rahn C, Swarup R, Woolaway KE, White PJ. Changes in gene expression in Arabidopsis shoots during phosphate starvation and the potential for developing smart plants. Plant Physiology (2003) 132:578–596.
Hammond JP, Broadley MR, White PJ. Genetic responses to phosphorus deficiency. Annals of Botany (2004) 94:323–332.
Hardtke CS. Root development: branching into novel spheres. Current Opinion in Plant Biology (2006) 9:66–71.[CrossRef][Web of Science][Medline]
Hartel H, Dormann P, Benning C. DGD1-independent biosynthesis of extraplastidic galactolipids after phosphate deprivation in Arabidopsis. Proceedings of the National Academy of Sciences, USA (2000) 97:10649–10654.
Held BM, Wang H, John I, Wurtele ES, Colbert JT. An mRNA putatively coding for an O-methyltransferase accumulates preferentially in maize roots and is located predominantly in the region of the endodermis. Plant Physiology (1993) 102:1001–1008.[Abstract]
Hernandez G, Ramirez M, Valdes-Lopez O, et al. Phosphorus stress in common bean: root transcript and metabolic responses. Plant Physiology (2007) 144:752–767.
Hochholdinger F, Woll K, Sauer M, Dembinsky D. Genetic dissection of root formation in maize (Zea mays) reveals root-type specific developmental programmes. Annals of Botany (2004) 93:359–368.
Hodge A. The plastic plant: root responses to heterogeneous supplies of nutrients. New Phytologist (2004) 162:9–24.[CrossRef][Web of Science]
Irizarry R, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, Speed TP. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics (2003) 4:249–264.[Abstract]
Kopriva S. Regulation of sulfate assimilation in Arabidopsis and beyond. Annals of Botany (2006) 97:479–495.
Leakey AD, Uribelarrea M, Ainsworth EA, Naidu SL, Rogers A, Ort DR, Long SP. Photosynthesis, productivity, and yield of maize are not affected by open-air elevation of CO2 concentration in the absence of drought. Plant Physiology (2006) 140:779–790.
Li G, Xue HW. Arabidopsis PLDzeta2 regulates vesicle trafficking and is required for auxin response. The Plant Cell (2007) 19:281–295.
Li K, Xu C, Zhang K, Yang A, Zhang J. Proteomic analysis of roots growth and metabolic changes under phosphorus deficit in maize (Zea mays L.) plants. Proteomics (2007) 7:1501–1512.[CrossRef][Web of Science][Medline]
Lim J, Jung JW, Lim CE, Lee MH, Kim BJ, Kim M, Bruce WB, Benfey PN. Conservation and diversification of SCARECROW in maize. Plant Molecular Biology (2005) 59:619–630.[CrossRef][Web of Science][Medline]
Ljung K, Hull AK, Celenza J, Yamada M, Estelle M, Normanly J, Sandberg G. Sites and regulation of auxin biosynthesis in Arabidopsis roots. The Plant Cell (2005) 17:1090–1104.
López-Bucio J, Hernández-Abreu E, Sánchez-Calderón L, Nieto-Jacobo M, Simpson J, Herrera-Estrella L. Phosphate availability alters architecture and causes changes in hormone sensitivity in the Arabidopsis root system. Plant Physiology (2002) 129:244–256.
Lynch J. Root architecture and plant productivity. Plant Physiology (1995) 109:7–13.[CrossRef][Web of Science][Medline]
Marrs KA, Alfenito MR, Lloyd AM, Walbot V. A glutathione S-transferase involved in vacuolar transfer encoded by the maize gene Bronze-2. Nature (1995) 375:397–400.[CrossRef][Medline]
McLaren JS. Crop biotechnology provides an opportunity to develop a sustainable future. Trends in Biotechnology (2005) 23:339–342.[CrossRef][Web of Science][Medline]
Misson J, Raghothama KG, Jain A, et al. A genome-wide transcriptional analysis using Arabidopsis thaliana Affymetrix gene chips determined plant responses to phosphate deprivation. Proceedings of the National Academy of Sciences, USA (2005) 102:11934–11939.
Morcuende R, Bari R, Gibon Y, et al. Genome-wide reprogramming of metabolism and regulatory networks of Arabidopsis in response to phosphorus. Plant, Cell and Environment (2007) 30:85–112.[CrossRef][Medline]
Nakajima K, Benfey PN. Signaling in and out: control of cell division and differentiation in the shoot and root. The Plant Cell (2002) 14:S265–276.
Noel JP, Austin MB, Bomati EK. Structure–function relationships in plant phenylpropanoid biosynthesis. Current Opinion in Plant Biology (2005) 8:249–253.[CrossRef][Web of Science][Medline]
Ogata H, Goto S, Sato K, Fujibuchi W, Bono H, Kanehisa M. KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Research (1999) 27:29–34.
Pietrini F, Iannelli MA, Massacci A. Anthocyanin accumulation in the illuminated surface of maize leaves enhances protection from photo-inhibitory risks at low temperature, without further limitation to photosynthesis. Plant, Cell and Environment (2002) 25:1251–1259.[CrossRef]
Pingali PL, Pandey S. Part 1: Meeting world maize needs: technological opportunities and priorities for the public sector. In: CIMMYT 1999–2000 world maize facts and trends. meeting world maize needs: technological opportunities and priorities for the public sector—Pingali PL, ed. (2001) Mexico, D.F: CIMMYT. 1–24.
Plénet D, Etchebest S, Mollier A, Pellerin S. Growth analysis of maize field crops under phosphorus deficiency. I. Leaf growth. Plant and Soil (2000) 223:117–130.[Web of Science]
Raghothama KG. Phosphate acquisition. Annual Review of Plant Physiology and Plant Molecular Biology (1999) 50:665–693.[CrossRef][Web of Science]
Ramon M, Rolland F. Plant development: introducing trehalose metabolism. Trends in Plant Science (2007) 12:185–188.[CrossRef][Medline]
Ramon M, Rolland F, Thevelein JM, Van Dijck P, Leyman B. ABI4 mediates the effects of exogenous trehalose on Arabidopsis growth and starch breakdown. Plant Molecular Biology (2007) 63:195–206.[CrossRef][Web of Science][Medline]
Rubio V, Linhares F, Solano R, Martin AC, Iglesias J, Leyva A, Paz-Ares J. A conserved MYB transcription factor involved in phosphate starvation signaling both in vascular plants and in unicellular algae. Genes and Development (2001) 15:2122–2133.
Satoh-Nagasawa N, Nagasawa N, Malcomber S, Sakai H, Jackson D. A trehalose metabolic enzyme controls inflorescence architecture in maize. Nature (2006) 441:227–230.[CrossRef][Medline]
Shin H, Shin HS, Chen R, Harrison MJ. Loss of At4 function impacts phosphate distribution between the roots and the shoots during phosphate starvation. The Plant Journal (2006) 45:712–726.[CrossRef][Web of Science][Medline]
Tesfaye M, Liu J, Allan DL, Vance CP. Genomic and genetic control of phosphate stress in legumes. Plant Physiology (2007) 144:594–603.
Theodorou ME, Plaxton WC. Metabolic adaptations of plant respiration to nutritional phosphate deprivation. Plant Physiology (1993) 101:339–344.[Abstract]
Thimm O, Blasing O, Gibon Y, Nagel A, Meyer S, Kruger P, Selbig J, Muller LA, Rhee SY, Stitt M. MAPMAN: a user-driven tool to display genomics data sets onto diagrams of metabolic pathways and other biological processes. The Plant Journal (2004) 37:914–939.[CrossRef][Web of Science][Medline]
Uhde-Stone C, Zinn KE, Ramirez-Yanez M, Li A, Vance CP, Allan DL. Nylon filter arrays reveal differential gene expression in proteoid roots of white lupin in response to phosphorus deficiency. Plant Physiology (2003) 131:1064–1079.
Usuda H, Shimogawara K. Phosphate deficiency in maize. II. Enzyme activities. Plant and Cell Physiology (1991) 32:1313–1317.
Wasaki J, Shinano T, Onishi K, et al. Transcriptomic analysis indicates putative metabolic changes caused by manipulation of phosphorus availability in rice leaves. Journal of Experimental Botany (2006) 57:2049–2059.
Wasaki J, Yonetani R, Kuroda S, et al. Transcriptomic analysis of metabolic changes by phosphorus stress in rice plant roots. Plant, Cell and Environment (2003) 26:1515–1523.[CrossRef]
Welti R, Li W, Li M, Sang Y, Biesiada H, Zhou HE, Rajashekar CB, Williams TD, Wang X. Profiling membrane lipids in plant stress responses. Role of phospholipase D alpha in freezing-induced lipid changes in Arabidopsis. Journal of Biological Chemistry (2002) 277:31994–32002.
Wolfinger RD, Gibson G, Wolfinger ED. Assessing gene significance from cDNA microarray expression data via mixed models. Journal of Computational Biology (2001) 8:625–637.[CrossRef][Web of Science][Medline]
Wu P, Ma L, Hou X, Wang M, Wu Y, Liu F, Deng XW. Phosphate starvation triggers distinct alterations of genome expression in Arabidopsis roots and leaves. Plant Physiology (2003) 132:1260–1271.
Wykoff DD, O'Shea EK. Phosphate transport and sensing in Saccharomyces cerevisiae. Genetics (2001) 159:1491–1499.
Xie Q, Frugis G, Colgan D, Chua NH. Arabidopsis NAC1 transduces auxin signal downstream of TIR1 to promote lateral root development. Genes and Development (2000) 14:3024–3036.
Xie Q, Guo HS, Dallman G, Fang S, Weissman AM, Chua NH. SINAT5 promotes ubiquitin-related degradation of NAC1 to attenuate auxin signals. Nature (2002) 419:167–170.[CrossRef][Medline]
Yang YH, Dudoit S, Luu P, Lin DM, Peng V, Ngai J, Speed TP. Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation. Nucleic Acids Research (2002) 30:e15.
Yi K, Wu Z, Zhou J, Du L, Guo L, Wu Y, Wu P. OsPTF1, a novel transcription factor involved in tolerance to phosphate starvation in rice. Plant Physiology (2005) 138:2087–2096.
Zhu J, Lynch JP. The contribution of lateral rooting to phosphorus acquisition efficiency in maize (Zea mays) seedlings. Functional Plant Biology (2004) 31:949–958.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  W.-Y. Lin, S.-I Lin, and T.-J. Chiou Molecular regulators of phosphate homeostasis in plants J. Exp. Bot., April 1, 2009; 60(5): 1427 - 1438. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||