RESEARCH PAPER |
Early PLD
-mediated events in response to progressive drought stress in Arabidopsis: a transcriptome analysis
1Department of Plant Pathology, Physiology, and Weed Science, Virginia Tech, Blacksburg, VA 24061, USA
2Department of Computer Science, Virginia Tech, Blacksburg, VA 24061, USA
To whom correspondence should be addressed. E-mail: grene{at}vt.edu
Received 20 September 2006; Accepted 6 November 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionConclusionSupplementary dataReferences |
|---|
Phospholipase D (PLD) has been implicated in a variety of stresses including osmotic stress and wounding. PLD
1-derived phosphatidic acid interacts with ABI1 phosphatase 2C and promotes abscisic acid signalling. It has also been shown to regulate proline biosynthesis negatively. Plants with abrogated PLD show insensitivity to abscisic acid (ABA) and impaired stomatal conductance. The goal in the present study was to identify early PLD-mediated events in response to progressive drought stress in Arabidopsis. Water was withheld from 7-week-old Arabidopsis thaliana (Col-0) and antisense-PLD1 (anti-PLD) in a controlled environment chamber. Diurnal leaf water potential (LWP) and photosynthesis measurements were recorded five and three times a day, respectively. Quantitative reverse transcription–polymerase chain reaction (qRT–PCR) and microarray analyses were conducted using RNA from shoots collected at the fourth LWP time point on the ninth day after stress imposition. Anti-PLD experienced severe water stress (–1.28 MPa) at the same time period that Col-0 experienced less water stress (–0.31 MPa). Diurnal LWP measurements showed that anti-PLD had a lower LWP than Col-0 in both control and drought-stress conditions. Photosynthesis was also more affected in anti-PLD than in Col-0. Anti-PLD plants recovered fully following rehydration after 10 d of stress. qRT–PCR revealed up to 18-fold lower values for PLD transcripts in stressed anti-PLD plants when compared with stressed Col-0. Microarray expression profiles revealed distinct gene expression patterns in Col-0 and anti-PLD. No differences in gene expression were detected between the two genotypes in the absence of drought stress. ROP8, PLD
, and lipid transfer proteins were among the differentially expressed genes between the two genotypes. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionSupplementary dataReferences |
|---|
Plants perceive and respond to drought stress by altering transcriptional, biochemical, and physiological processes. The mechanism of molecular responses of plants to drought stress has been extensively studied (Ingram and Bartels, 1996; Seki et al., 2002; Rabbani et al., 2003; Rizhsky et al., 2004), and phospholipid-based signalling, involving phospholipase C (PLC) and phospholipase (PLD), has emerged as a component of drought-responsive signal transduction pathways (Hirayama et al., 1995; Sang et al., 2001b).
PLD genes play diverse roles in plant cells in cell signalling and in metabolism by hydrolysing phospholipids to phosphatidic acid (PA). PLDs have been shown to be involved in programmed cell death, cell growth, cell patterning, vesicular trafficking, cytoskeletal changes, root growth, abiotic stress responses, and the oxidative burst (Sang et al., 2001b; Lee et al., 2003; Ohashi et al., 2003; Potocky et al., 2003; Li et al., 2004; Wang, 2005). The various PLDs have different requirements for Ca2+ and phosphoinositides, and varied substrate preferences. Depression of PLD1 activity by antisense suppression or its ablation by knockout mutation resulted in alteration in several stress-related physiological processes such as production of reactive oxygen species (Sang et al., 2001a), abscisic acid (ABA) signalling (Fan et al., 1997), wounding response (Wang et al., 2000), freezing tolerance (Welti et al., 2002), and increased water loss (Sang et al., 2001b).
PLD1 interacts directly with G, the only canonical -subunit of the heterotrimeric G protein in Arabidopsis (Zhao and Wang, 2003). PLD1–G interaction inhibits PLD activity, stimulating the intrinsic GTPase activity of G (Zhang et al., 2004a), and regulates stomatal movement (Mishra et al., 2006). ABA-independent activation of PLD by the G
subunit has also been reported in barley aleurone (Ritchie and Gilroy, 2000).
In mammalian cells, several PA targets have been identified (Ktistakis et al., 2003). These include protein kinases, protein phosphastases, phosphodiesterases, ATPase, small GTPases, sphingosine kinase, and protein kinase C (Jenkins et al., 1994; Ghosh et al., 1996; Manifava et al., 2001; Lopez-Andreo et al., 2003; Delon et al., 2004). In plant systems, less is known. In Arabidopsis, PLD1-derived PA has been shown to interact with ABI1, a protein phosphatase 2 C (PP2C) which is a negative regulator of ABA responses (Zhang et al., 2004a). PA binds to ABI1 and reduces its translocation to the nucleus by tethering it to the plasma membrane, thus promoting ABA responses. PLD1 and PA interact with PP2C to signal ABA-promoted stomatal closure, whereas PLD1 and PA interact with the G-subunit to mediate ABA inhibition of stomatal opening (Mishra et al., 2006). PA also binds to PDK1, a protein kinase which acts upstream of a mitogen-acivated protein kinase (MAPK) pathway (Anthony et al., 2004; Rentel et al., 2004). PA has been shown to bind MAPK6-related protein, an important mediator in stress and ethylene signalling, in soybean cells (Liu and Zhang, 2004). Other PA-binding targets have been identified in a proteome-wide study (Testernik et al., 2004). However, an overall understanding of any one PLD-mediated pathway is still lacking for plants.
Given this diverse role of PLD in plant function, it is possible that additional targets of PA exist. No transcriptome studies have been carried out on PLD1-depleted plants as yet, and it still yet unclear which downstream pathways are regulated by PLD1. To identify these downstream targets of PLD, the effect of progressive drought stress on global transcriptional profiles obtained from non-transformed and antisense-PLD (anti-PLD) plants, and also on physiological responses, were compared. Antisense suppression of PLD1 significantly altered drought responses in Arabidopsis, as was to be expected. Microarrays with 
26 000 elements consisting of 70-mer gene-specific oligonucleotides were used to profile gene expression under drought conditions. A total of 920 genes (4.4% of the transcriptome) that were significantly regulated by drought stress were identified in Col-0. In anti-PLD, 765 genes (3.7% of the transcriptome) were drought regulated, with a modest (15%) overlap between the two genotypes. The results suggest that PLD1-mediated signalling in response to drought stress leads to changes at many levels of plant processes including transcription, metabolism, and signalling. These include ROP8, PLD, and lipid transfer proteins (LTPs). Small G-proteins, 14-3-3 proteins, genes involved in lipid signalling, and several transcription factors may also be regulated by PLD1 under drought stress.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionSupplementary dataReferences |
|---|
Anti-PLD
1 plants in the Columbia background were obtained from the laboratory of Dr X Wang, Kansas State University. Wild-type (Col-0) and anti-PLD seeds were sown in pots of moist soil (Sunshine Mix 1) in a growth chamber (Conviron 4030, Winnipeg, MB, Canada). The growth chamber was set to 8 h light at 22 °C, 16 h dark at 20 °C. The plants were watered at regular intervals. After 4 weeks, the plants were transplanted to trays with a spacing of 2 inches. Water was withheld just prior to flowering, 7 weeks after sowing. Sampling began on the eighth day after imposition of drought stress. Samples were taken five times a day (one pre-dawn, three during the day, and one after dusk). Relative to dawn, the samples were taken at the following times: –0.5, 1, 2, 4, and 9 h. Two leaves from each plant were taken for leaf water potential (LWP) measurements using a pressure bomb (Soil Moisture Equipment Corp., Santa Barbara, CA, USA). Photosynthesis measurements were taken using a Li-6400 portable Photosynthesis System (Li-Cor Inc., Lincoln, NE, USA) at 1, 2, and 4 h after dawn. All measurements were made on two replicates. Aerial portions of the plant were collected and flash frozen in liquid nitrogen, and were stored at –70 °C for further studies.
RNA extraction and microarray hybridization
Total RNA was obtained by using a method developed by Jaakola et al. (2001) and further purified using RNeasy columns (Qiagen, Foster City, CA, USA). Total RNA, 100 µg per sample, was labelled using the indirect labelling procedure as described by Hegde et al. (2000). RNA was mixed with 0.8 mM dATP, dCTP, and dGTP, 0.5 mM dTTP, 0.3 mM 5-(3-aminoallyl)-2'-deoxyuridine-5'-triphosphate (Sigma, St Louis, MO, USA), and 2 µg of oligo(dT) (Invitrogen, Carlsbad, CA, USA), followed by first-strand cDNA synthesis (SuperscriptII RT; Invitrogen). After incubation for 2 h at 42 °C, the cDNA was treated with 2 U of RNase H (Invitrogen) for 15 min at 37 °C and purified (Qiagen). After binding the cDNA, columns were washed with phosphate–ethanol buffer (5 mM potassium phosphate, pH 8.0 and 80% ethanol) and the cDNA was eluted with phosphate buffer (4 mM K-phosphate, pH 8.5).
Purified cDNA was dried, and subjected to a coupling reaction with Cy3-dUTP or Cy5-dUTP dye ester (Amersham-Pharmacia, St Louis, MO, USA) and the labelled cDNA was purified (Qiagen). Glass slides were pre-treated and hybridized as described (Kawasaki et al., 2001). Hybridizations were done using the two-dye (Cy3 and Cy5) method. In total, four hybridizations were performed with two hybridizations per treatment (dye swaps with biological replicate).
Quantitative RT–PCR
DNase treatment, cDNA synthesis, primer design, and SYBR Green I reverse transcription polymerase chain reaction (RT–PCR) were carried out as described by Vandesompele et al. (2002). In brief, 2 µg of each total RNA sample was treated with RNase-free rDNase I according to the manufacturer's instructions (Ambion). Treated RNA samples were purified before cDNA synthesis using Quia columns (Qiagen). First-strand cDNA was synthesized using oligo(dT) (18-mer) and SuperscriptII reverse transcriptase according to the manufacturer's instructions (Invitrogen), and subsequently diluted with nuclease-free water (Sigma) to 12.5 ng ml–1 cDNA. RT–PCR amplification mixtures (25 ml) contained 25 ng of template cDNA, 2x SYBR Green I Master Mix buffer (12.5 ml) (Applied Biosystems), and 300 nM forward and reverse primer. While designing primers, care was taken to avoid amplification of the antisense region of PLD1. The following genes were amplified: PLD [AT4G35790: forward primer (FP), ACGATCCATGTGTTTGGGTT; reverse primer (RP), GCCTCAGCCTCATCTTCATATT], ROP8/ARAC9 (AT2G44690: FP, CTTATTTCCTACACCAGCAACAC; RP, GTATCCCAGAGACCCAGATTGAC), glutathione S-transferase (GST) (AT5G17220: FP, TTGGTCGAGGATCTCAAAGTGAAG; RP, GCATGTGCGTCAAATCAGCC), ankyrin-repeat protein (AT1G62050: FP, TGGCGAGAAACCCGATGCTC; RP, GCGTTTGTTGCTCCTCCTCTTC), Tu-GTP (AT5G13650: FP, GTTTGTAGGTTCTGGAGTGG; RP, TTATGTTTGTCGCTGCCTTC), phosphatidylinositol-4-phosphate 5-kinase (PIP5K) (AT1G21920: FP, GATTCAGGGGTTCTTGTGAC; RP, ATTCACTGCCTTATTCGCTG), and glyceraldehyde phosphate dehydrogenase (GAPDH) (AT3G26650: FP, CGTGATCTAAGGAGAGCAAGA; RP, TTCCCTTGAGGTTAGGGAGC). The reactions were run on an ABI 7300 Real-time PCR System (Applied Biosystems). The cycling conditions comprised 10 min denaturation at 95 °C and 35 cycles at 95 °C for 15 s, 56 °C for 30 s 72 °C for 30 s. Each assay included a standard curve of seven serial dilution points of PCR fragments (500–850 bp) of each gene, a no-template control, and 25 ng of each test cDNA. GAPDH was used a control gene. Each assay was performed in triplicate. All PCR efficiencies were >95%. Sequence Detection Software (version 1.2.3) (Applied Biosystems) was used for data analysis.
Microarray data analyses
Microarray slides including 26 000 elements consisting of 70-mer gene-specific oligonucleotides were used (see: http://www.ag.arizona.edu/microarray/). Expresso, a system of tools and databases for integration of computation and experimentation in the context of gene expression experiments, was used for analysis of the data (Watkinson et al., 2003; Li et al., 2006). The sensitivity of individual genes to the experimental treatments is estimated using a two-stage statistical analysis (Wolfinger et al., 2001), as modified by Li et al. (2006). Current methodology (Brinker et al., 2004; Byrne et al., 2005; Kirst et al., 2005) and stringent statistical models (Wolfinger et al., 2001; Chu et al., 2002; Cui and Churchill, 2003; Allison et al., 2006) include factors not previously considered, such as dye and random block effects. These analyses have now replaced the focus on fold changes that did not incorporate variance or probability considerations. Thus, probability can be assessed of changes that previously were considered too close to call. The details of analyses of the data are thoroughly described and their advantages discussed elsewhere (Sioson et al., 2006). Data mining was carried out using MIPS Functional categorical (FunCat) analysis (http://mips.gsf.de/proj/funcatDB/search_main_frame.html).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionSupplementary dataReferences |
|---|
Drought stress and physiological measurements: phenotypic differences observable only under stress
The physiological drought responses of wild-type (Col-0) and anti-PLD
plants were compared. Under control conditions, there was no observable phenotypic difference between Col-0 and anti-PLD plants (Fig. 1b). LWP measurements began at the eighth day after withholding water. LWP measurements were taken 30 min before dawn, 1, 2, and 4 h during the day, and 1 h after dusk. Previous experiments under similar conditions showed no significant phenotypic difference between control and stressed plants until after 7 d of withholding water (R Grene et al., unpublished data). Diurnal LWP measurements were similar in unstressed Col-0 and anti-PLD plants (Fig. 1a). Differences in LWP became apparent on day 8 after withholding water. Anti-PLD experienced severe water stress at the same time that Col-0 experienced less water stress (–0.31 MPa in Col-0 and –1.28 MPa in anti-PLD at peak drought stress) (Fig. 1b). Diurnal LWP measurements showed that anti-PLD had significantly lower LWP than Col-0 as the drought stress progressed. However, anti-PLD plants were able to recover in the dark. Net photosynthesis was also more affected in anti-PLD than in Col-0 (data not shown). Both the untransformed and the anti-PLD plants recovered fully (0.11 MPa in Col-0 and 0.15 MPa in anti-PLD) upon rehydration at 10 d after the end of stress imposition (Fig. 1a, c).
|
Transcriptome profiling
Plants from day 8, the fourth time point of drought stress, were chosen for gene transcription profiling in order to identify drought-mediated transcript changes associated with PLD
function at an early stage of drought stress. RNAs from two biological replicates were hybridized on two microarray slides with dye swap. To determine drought-regulated genes, a two-step analysis of variance (ANOVA) was performed using Expresso (Sioson et al., 2006). To determine the effect of antisense suppression of PLD1 under control conditions, gene expression patterns of unstressed Col-0 and anti-PLD plants were compared. A total of only 39 (0.15% of the transcriptome) genes were differentially expressed between unstressed Col-0 and anti-PLD (see Supplementary Table S1 at JXB online).
Genes that were identified as significantly up- or downregulated under drought stress at P <0.05 are shown in Fig. 2; see Supplementary Table S2 at JXB online. A total of 920 genes (4.4% of the transcriptome) were drought regulated in Col-0, with 460 upregulated and 460 downregulated. In anti-PLD, 765 genes (3.7 of the transcriptome) were drought regulated, with 431 upregulated and 334 downregulated. Only 14.67% and 16.25% of the drought up- and downregulated genes, respectively were common to both Col-0 and anti-PLD. Significantly regulated genes were classified according to the Munich Information Center for Protein Sequences (MIPS) categorization. In both Col-0 and anti-PLD, the most represented category was transcription, followed by metabolism, signalling, and subcellular localization (Table 1; see Supplementay Table S3 at JXB online). In most categories, barring transcription, a greater number of genes were downregulated in Col-0 than in anti-PLD.
|
|
Transcription:
Forty-seven transcription-associated genes were upregulated and 35 were downregulated by drought in Col-0. In anti-PLD
, 43 genes were upregulated and 24 were downregulated (Table 4). Among the drought-regulated transcription factors, only 10 genes were common to both genotypes. The transcription factors were classified by their domains into 14 groups. Zinc finger proteins were the most regulated, followed by MYB and AP2/ERF (APETALA 2/ETHYLENE-RESPONSIVE ELEMENT-BINDING FACTOR). One AP2 gene (AT4G27950) was upregulated in both Col-0 and anti-PLD, while another gene, AT1G36060, was upregulated only in Col-0. Three genes encoding bZIP transcription factors were upregulated in anti-PLD, out of which one gene (AT1G53490) was downregulated in Col-0. A CCAAT-box-binding transcription factor and two homeodomain/lipid-binding START domain-containing (HD-START) genes were upregulated in Col-0. Out of the two MADS box genes downregulated in Col-0, one gene (AT5G65330) was upregulated in anti-PLD. All significantly regulated genes belonging to MYB and NAC were differentially regulated in both Col-0 and anti-PLD. Not only were the zinc finger proteins the most regulated group of transcription factors but they were also the most differentially regulated between Col-0 and anti-PLD. Out of 16 upregulated genes in Col-0, only two genes were upregulated in anti-PLD and out of 13 downregulated genes, only one gene was downregulated in anti-PLD.
|
Metabolism:
The second most drought-responsive group of genes was that associated with metabolism. Among metabolism genes, 28 genes were upregulated in Col-0 while 34 genes were upregulated in anti-PLD
, with nine genes common to both, and 37 genes were downregulated in Col-0 while 19 genes were downregulated in anti-PLD, with eight genes common to both (see Supplementary Table S4 at JXB online). More genes belonging to ‘Carbon (C)-compounds and carbohydrate metabolism’ were downregulated in wild-type (17 genes) than in anti-PLD (seven genes), with three genes common to both. On the other hand, more genes associated with C-compounds and carbohydrate metabolism were upregulated in anti-PLD (21 genes) than in Col-0 (12 genes), with five genes common to both.
Signal transduction genes responsive to drought:
Calcium plays an important role in early signal transduction events of many stresses. In Col-0, two 14-3-3 proteins (AT1G34760 and AT5G38480) and a calmodulin-binding protein were upregulated, and two calmodulin-binding proteins were downregulated, while in anti-PLD, two 14-3-3 proteins and three calmodulin proteins were upregulated and one 14-3-3 protein (AT5G10450) was downregulated. Only the 14-3-3 protein GF14 omicron (grf11) (AT1G34760) response was common to both genotypes (Table 2). Eight protein kinases including five putative protein kinases, a receptor-like kinase, a serine/threonine kinase, and a leucine-rich repeat (LRR)-transmembrane kinase were upregulated in Col-0, while two putative protein kinases, two receptor-like kinases, two serine/threonine kinases, and an LRR-transmembrane kinase were upregulated in anti-PLD. Five phosphatases including two PP2Cs and one PP2A were upregulated, and three phosphatases were downregulated in Col-0, while two PP2Cs (including ABI1) were upregulated in anti-PLD. Two response regulators (including a pseudoresponse regulator, APRR5) were upregulated in Col-0. Nine GTPases were downregulated in Col-0, but were either upregulated (AT5G64990 and AT2G44690) or unchanged in anti-PLD (AT1G43890, AT5G13650, AT3G05310, and AT5G60860). ROP8 was downregulated in Col-0 but upregulated in anti-PLD.
|
Eleven and seven genes involved in phospholipid signalling were regulated by drought (Table 6) in Col-0 and anti-PLD
, respectively. PIP5K (AT1G21920) and LTP (AT2G10940) were upregulated, and lipid-binding START domain-containing protein was downregulated in both Col-0 and anti-PLD. Genes upregulated only in Col-0 are inositol-1,4,5-trisphosphate 5-phosphatase, scramblase-like, and phospholipase-like protein. PLD and an LTP were downregulated in Col-0 and upregulated in anti-PLD (Table 2).
|
Protein fate:
Genes belonging to ‘protein fate’ showed differential expression in Col-0 and anti-PLD
. These included proteins involved in protein modification, degradation, and folding. In Col-0, HSP60 (AT2G33210) was increased and three genes involved in protein degradation were downregulated (AT3G62250, AT5G53300, and AT5G42990). None of these were drought regulated in the antisense genotype.
Hormone-related genes:
Five genes related to ABA signalling were upregulated in Col-0 while two of them were downregulated and one unchanged in anti-PLD (Table 5). Three cytokinin genes (IPT2 and two UDP-glucosyl transferase genes) were upregulated in Col-0 but downregulated in anti-PLD. Several other UDP-glucosyl transferase genes were differentially regulated in Col-0 and anti-PLD. Ethylene-responsive element-binding factor family (EREBP) was downregulated in anti-PLD. Gibberellin response modulator (GAI) (RGA2) was downregulated in Col-0.
|
Quantitative RT–PCR (qRT–PCR)
PLD
transcript levels were 16-fold less in drought-stressed anti-PLD than in Col-0. To validate the microarray data, qRT–PCR analyses were performed on six genes that responded on the microarrays. qRT–PCR results confirmed that these genes are regulated by drought stress (Table 7). The qRT–PCR data correlated with the microarray results with a correlation coefficient of 0.96 and 0.95 in Col-0 and anti-PLD, respectively. qRT–PCR was also performed on samples taken from other time points to determine the kinetics of PLD expression in both genotypes. At the initial time point, PLD was repressed, while at later time points PLD was upregulated in Col-0. PLD was upregulated at all time points in anti-PLD (Fig. 3).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionSupplementary dataReferences |
|---|
Significant work has been done to try to understand the role of PLD in drought signalling, using short-term ‘shock’ treatments. However, little or no work has been done to study its role in natural progressive drying. The aim in this study was to identify PLD
1-regulated genes at early stages of progressive drought. To study the role of PLD1 in response to drought stress, Arabidopsis wild-type (Col-0) and anti-PLD were subjected to progressive drought by withholding water. LWP was measured five times a day to monitor diurnal changes in plant moisture status. LWP was reduced in both Col-0 and anti-PLD in response to drought. Anti-PLD plants are less sensitive to ABA and have impaired stomatal closure during water deficits (Sang et al., 2001). Antisense suppression of PLD1 did not show any observable phenotypic differences when compared with the wild type under control conditions. During drought, anti-PLD plants were more wilted than Col-0, indicating a greater transpirational loss of water in anti-PLD (Fig. 1). However, anti-PLD plants were able to recover to Col-0 moisture status during the night, suggesting that depletion of PLD does not affect nocturnal stomatal closure. Moreover, reduction in PLD1 did not affect the plants' ability to recover fully after rehydration. Photosynthesis was also more affected in the antisense plants as a result of reduced plant moisture content, and it also recovered fully upon rehydration.
In order to understand more about the role of PLD1 in drought signalling, eighth day–fourth time point (initial stage of drought) was chosen for gene expression analysis in response to progressive drought. This was when anti-PLD just began to experience higher water loss than Col-0. The goal was to capture early PLD-mediated signalling events in response to progressive drought stress. The effect of antisense suppression of PLD1 on global gene expression was minimal under unstressed conditions. About 0.15% of the transcriptome was differentially regulated between Col-0 and anti-PLD. Lipid profiling studies have shown that PLD1 deletion does not significantly alter basal PA levels in leaves (Devaiah et al., 2006). Monitoring changes in drought-induced gene expression revealed significant differences between drought-stressed Col-0 and anti-PLD. The expression of very few (15%) drought-regulated genes was common to both Col-0 and anti-PLD (Fig. 2; see Supplementary Table S1 at JXB online). This result suggests that changes in plant PA levels via the PLD pathway significantly alter global gene expression under progressive drought stress.
In wild-type plants, stomata are closed for increasingly longer periods during the day under progressive drought, starting mid-morning (Tenhunen et al., 1987). Due to stomatal closure in response to drought, there is a reduction in water loss as well as gas exchange, leading to reduced carbon assimilation and to a near optimization of carbon assimilation in relation to water supply (Cowan, 1981; Jones, 1992). This partly explains why Col-0 has more downregulation of genes involved in metabolism (C-compound and carbohydrate metabolism) than anti-PLD (12 genes up- and 17 downregulated, and 21 up- and seven downregulated in Col-0 and anti-PLD, respectively).
Responses of transcription factors to drought stress
The relatively large numbers of up- and downregulated genes (9% of the drought-regulated genes) involved in transcription suggest that the plants are undergoing significant changes in transcription in response to water loss. The number of upregulated genes encoding transcription factors was higher than that of downregulated genes, suggesting that drought stress leads mainly to activation of transcription factors (Table 4; see Supplementary Table S5 at JXB online). In general, there was little overlap of responsive genes (7%) between Col-0 and anti-PLD in this category. AP2 transcription factors play an important role in cold and drought-stress responses. RAP2 (AT1G36060), a part of the CBF regulon and possibly controlling a subregulon (Fowler and Thomashow, 2002; Zhang et al., 2004b), was upregulated in Col-0, but not in antisense plants.
Evidence for PLD-dependent and independent drought-responsive pathways
Several transcription factors involved in plant development were differentially regulated in Col-0 and anti-PLD by drought. These include homeodomain (HD-START), NAC, and MADS. NAC transcription factors have been reported to be involved in abiotic stress signalling (Fujita et al., 2004; Tran et al., 2004). In the present experiment, these two groups were downregulated in both Col-0 and anti-PLD, suggesting the existence of a drought-responsive pathway that is independent of PLD. MYB and basic helix–loop–helix (bHLH) transcription factors showed diverse responses. HD-START domain proteins were upregulated in Col-0. These proteins bind to phosphatidylcholine via the START domain and modulate the activity and transcription of the gene via the homeodomain (Schrick et al., 2004). By analogy with animal systems, this provides a mechanism whereby lipid content in the cytosol regulates transport/sequestration of the transcription factor to the nucleus. These data suggest that the HD-START mechanism is on the same pathway as PLD.
Zinc finger domain-containing transcription factors were by far the largest group of transcription factors regulated by drought (29 in Col-0 and 19 in anti-PLD). C2H2-type zinc fingers have been reported to be involved in stress regulation (Sakamoto et al., 2004). Several C3HC4-type and CHP-type zinc finger proteins (four and three genes, respectively) were upregulated in anti-PLD but downregulated or unchanged in Col-0 (ST-2). Three CCCH-type zinc fingers were upregulated in Col-0, out of which only one was upregulated in anti-PLD.
Drought signal transduction
PLD1 may be involved in upstream activation of 14-3-3 and ankyrin repeat-containing proteins:
Calcium is a critical component in stress signal transduction pathways. Consistent with the central role of calcium, more calcium-binding proteins (three calmodulin-binding proteins and two 14-3-3 proteins) were upregulated in anti-PLD (Table 2; see Supplementary Table S6 at JXB online). One of them, rare cold-inducible (RCI1; AT5G38480), was upregulated only in Col-0. Interestingly, 14-3-3 protein GF14 
(grf6/AFT1) was downregulated in anti-PLD. Overexpression of the Arabidopsis 14-3-3 protein GF14 in cotton leads to a ‘Stay-Green’ phenotype and improves tolerance to moderate drought stress (Yan et al., 2004). 14-3-3 protein GF14 interacts with ankyrin repeat-containing protein 2 (AKR2). AKR2 is involved in both disease resistance and antioxidation metabolism (Yan et al., 2002). Four ankyrin-repeat proteins were found to be upregulated in Col-0, out of which one gene (AT3G24210) was upregulated and two (AT4G19660 and AT1G62050) genes were downregulated in anti-PLD. These results suggest that PLD1 may be involved in upstream activation of 14-3-3 and ankyrin-repeat containing proteins.
A possible interaction between cytokinin and drought/cold signalling:
Phosphorylation and dephosphorylation play an important role in stress signal transduction. The number of protein kinases and phosphatases was the highest (56) out of a total of 99 drought signalling genes (Table 2). Fifty per cent of the downregulated genes were common to Col-0 and anti-PLD. These include LRR-transmembrane kinase, putative protein kinases, and receptor and receptor-like protein kinases, suggesting that these genes act upstream of the PLD signalling pathway, or are on another pathway altogether. ABI1 was downregulated in anti-PLD. ABI1, a PP2C, interacts with PLD1-derived PA and negatively regulates ABA signalling (Zhang et al., 2004a). Two ARR (Arabidopsis response regulator) genes were upregulated by PLD1. ARR5 has been implicated in circadian rhythms (Fujimori et al., 2005). ARR15 was downregulated in anti-PLD. ARR15, a type-A response regulator, acts as a negative regulator in cytokinin-mediated signal transduction in Arabidopsis and is also implicated in ethylene signalling (Kiba et al., 2003). In their study, Lee et al. (2005) found that other response regulators (ARR10 and ARR16) are downregulated during cold stress. These genes are also involved in cytokinin signalling. This suggests an interaction between cytokinin and drought/cold signalling.
Involvement of other phospholipases in drought signalling:
PIP5K (AT1G21920) was upregulated in both Col-0 and anti-PLD. Increased phosphatidylinositol 4,5-bisphosphate (PIP2) production is associated with upregulation of PIP5K during osmotic stress (Mikami et al., 1998). Two ß-tubulin (AT5G62700 and AT2G29550) and three calmodulin genes were upregulated in anti-PLD. In animals, ß-tubulin binds to PIP2 and enhances PLC activity (Chang et al., 2005). PIP2 also plays a critical role in PLD activation in mammals and plants. PIP2 is required for activities of Arabidopsis PLDß and PLD (Qin et al., 1997). This suggests that plants channel drought signalling through other members of the PLD or PLC pathway (possibly signal amplification via inositol trisphosphate-mediated Ca2+ release) and calmodulin signalling when PLD availability is greatly reduced. However, no changes in gene expression of PLC family members were seen, suggesting the possibility of increased post-translational activation of the PLC and PLD pathway.
PLD may act as ‘damage control’ and possibly function in late drought stress signalling:
PLD was highly upregulated in anti-PLD when compared with Col-0. In earlier studies, anti-PLD plants did not show any change in activity or expression of PLDß, PLD (Pappan et al., 1997; Wang et al., 2000), or PLD (Sang et al., 2001b) under control conditions. Drought conditions were not used in these experiments. PLD is known to be activated by hydrogen peroxide (H2O2), and the resulting PA functions to decrease H2O2-promoted programmed cell death (Zhang et al., 2003). However, a recent study showed that abrogation of PLD does not affect tolerance to drought stress (Li et al., 2005). Studies have revealed that PLD1 and PLD have different, and sometimes opposite, roles in signalling (Welti et al., 2002; Li et al., 2004). Although PLD1 and PLD produce PA, they may prefer phospholipid substrates with different fatty acid side chains. This, in turn, would change the PA species pool, thereby activating distinct intracellular signals. The results suggest that PLD acts as ‘damage control’ and has a role in late drought stress signalling.
PLDa1 negatively regulates small GTPases:
Nine GTPases were downregulated in Col-0 but were either upregulated or unchanged in anti-PLD (Table 3). A Rho-like GTPase (ROP8) was downregulated in Col-0 but upregulated in anti-PLDa. Nothing is known about the function of ROP8, but other members of the ROP gene family (ROP6 and ROP10) are involved in negative regulation of ABA responses. The results are consistent with the hypothesis that PLD1 may negatively regulate small GTPases.
|
Hormone signalling
Much information has accumulated on the role of ABA in stomatal movement and its importance in dehydrating roots and ABA movement in the plants. There is a very limited knowledge about the exact relationship between water deficit and ABA long-distance signalling in field drought conditions. Under rapid dehydration, ABA is synthesized mainly in roots and transported to the shoots. Contrary to most dehydration studies, very few genes involved in ABA signalling were found under the present conditions of progressive drought stress. There are two possible reasons why induction of ABA biosynthesis genes was not seen. First, root tissue was not included in the microarray experiment. Secondly, in leaves, ABA biosynthesis is increased only when LWP falls to the point where turgor approaches zero. In this study, leaf samples were collected when the LWP just began to drop. Five ABA-responsive genes were upregulated in Col-0, while only one gene (ABI1) was upregulated in anti-PLD
, suggesting a role for PLD1 in ABA signalling (Table 5). |
Conclusion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionSupplementary dataReferences |
|---|
In this study, responses of an antisense plant for PLD
1 and wild-type (Col-0) to progressive drought were compared. Although under control conditions abrogation of PLD1 has little/no effect on phenotype, drought stress shows significant differences in physiological behaviour. Using transcriptome analysis, several new genes regulated by progressive drought were identified and genes whose expression may be part of the PLD1 pathway were also found. ROP8, PLD, and LTPs were among the differentially expressed genes between the two genotypes. The results provide a basis to explore further the processes regulated by PLD in response to drought using transgenic studies and molecular biology. |
Supplementary data |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionSupplementary dataReferences |
|---|
The following supplementary data are available at JXB online.
Table S1. Significantly regulated genes in Anti-PLD vs Col-0 control.
Table S2. Significantly differential expressed genes in Wildtype (Col-0) and Anti-PLD at alpha = 0.05.
Table S3. Functional characterization of drought regulated genes in wildtype (Col-0) and antisense-PLD.
Table S4. Genes involved in metabolism.
Table S5. Differentially regulated transcription factors.
Table S6. Differentially regulated signalling genes.
|
Acknowledgements |
|---|
We thank Dr Xuemin Wang for sharing anti-sense PLD
plants. The work has been supported by NSF grant # BIO/IBN-0219322 to LSH and RG, and VPI&SU institutional funds. |
Footnotes |
|---|
* Present address: Ateneo de Naga University, Naga City, The Philippines
|
Abbreviations |
|---|
ABA, abscisic acid; ARR, Arabidopsis response regulator; GAPDH, glyceraldehyde phosphate dehydrogenase; LRR, leucine-rich repeat; LTP, lipid transfer protein; LWP, leaf water potential; PA, phosphatidic acid; PIP2, phosphatidylinositol 4,5-bisphosphate; PIP5K, phosphatidylinositol-4-phosphate 5-kinase; PLC, phospholipase C; PLD, phospholipase D; PP2C, protein phosphatase 2 C; qRT–PCR, quantitative reverse transcription–polymerase chain reaction.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionSupplementary dataReferences |
|---|
Allison DB, Cui X, Page GP, Sabripour M. (2006) Microarray data analysis: from disarray to consolidation and consensus. Nature Reviews Genetics 7 55–65.[CrossRef][ISI][Medline]
Anthony RG, Henriques R, Helfer A, Meszaros T, Rios G, Testerink C, Munnik T, Deak M, Koncz C, Bogre L. (2004) A protein kinase target of a PDK1 signalling pathway is involved in root hair growth in Arabidopsis. EMBO Journal 23 572–581.[CrossRef][ISI][Medline]
Brinker M, van Zyl L, Liu W, Craig D, Sederoff RR, Clapham DH, von Arnold S. (2004) Microarray analyses of gene expression during adventitious root development in Pinus contorta. Plant Physiology 135 1526–1539.
Byrne KA, Wang YH, Lehnert SA, Harper GS, McWilliam SM, Bruce HL, Reverter A. (2005) Gene expression profiling of muscle tissue in Brahman steers during nutritional restriction. Journal of Animal Science 83 1–12.
Chang J-S, Kim S-K, Kwon T-K, Bae SS, Min DS, Lee YH, Kim S-O, Seo J-K, Choi JH, Suh P-G. (2005) Pleckstrin homology domains of phospholipase C-1 directly interact with ß-tubulin for activation of phospholipase C-1 and reciprocal modulation of ß-tubulin function in microtubule assembly. Journal of Biological Chemistry 280 6897–6905.
Chu T-M, Weir B, Wolfinger R. (2002) A systematic statistical linear modeling approach to oligonucleotide array experiments. Mathematical Biosciences 176 35–51.[CrossRef][ISI][Medline]
Cowan IR. (1981) Regulation of water use in relation to carbon gain in higher plantsBerlin Springer.
Cui X and Churchill G. (2003) Statistical tests for differential expression in cDNA microarray experiments. Genome Biology 4 210.[CrossRef][Medline]
Delon C, Manifava M, Wood E, Thompson D, Krugmann S, Pyne S, Ktistakis NT. (2004) Sphingosine kinase 1 is an intracellular effector of phosphatidic acid. Journal of Biological Chemistry 279 44763–44774.
Devaiah SP, Roth MR, Baughman E, Li MY, Tamura P, Jeannotte R, Welti R, Wang XM. (2006) Quantitative profiling of polar glycerolipid species from organs of wild-type Arabidopsis and a phospholipase D 1 knockout mutant. Phytochemistry 67 1907–1924.[CrossRef][ISI][Medline]
Fan L, Zheng S, Wang X. (1997) Antisense suppression of phospholipase D retards abscisic acid and ethylene-promoted senescence of postharvest Arabidopsis leaves. The Plant Cell 9 2183–2196.[Abstract]
Fowler S and Thomashow MF. (2002) Arabidopsis transcriptome profiling indicates that multiple regulatory pathways are activated during cold acclimation in addition to the CBF cold response pathway. The Plant Cell 14 1675–1690.
Fujimori T, Sato E, Yamashino T, Mizuno T. (2005) PRR5 (PSEUDO-RESPONSE REGULATOR 5) plays antagonistic roles to CCA1 (CIRCADIAN CLOCK-ASSOCIATED-1) in Arabidopsis thaliana. Bioscience, Biotechnology and Biochemistry 69 426–430.[CrossRef][Medline]
Fujita M, Fujita Y, Maruyama K, Seki M, Hiratsu K, Ohme-Takagi M, Tran L-SP, Yamaguchi-Shinozaki K, Shinozaki K. (2004) A dehydration-induced NAC protein, RD26, is involved in a novel ABA-dependent stress-signaling pathway. The Plant Journal 39 863–876.[CrossRef][ISI][Medline]
Ghosh S, Strum JC, Sciorra VA, Daniel L, Bell RM. (1996) Raf-1 kinase possesses distinct binding domains for phosphatidylserine and phosphatidic acid. Journal of Biological Chemistry 271 8472–8480.
Hegde P, Qi R, Abernathy K, Gay C, Dharap S, Gaspard R, Hughes JE, Snesrud E, Lee N, Quackenbush J. (2000) A concise guide to cDNA microarray analysis. Biotechniques 29 548–550.[ISI][Medline]
Hirayama T, Ohto C, Mizoguchi T, Shinozaki K. (1995) A gene encoding a phosphatidylinositol-specific phospholipase C is induced by dehydration and salt stress in Arabidopsis thaliana. Proceedings of the National Academy of Sciences, USA 92 3903–3907.
Ingram J and Bartels D. (1996) The molecular basis of dehydration tolerance in plants. Annual Review of Plant Physiology and Plant Molecular Biology 47 377–403.[CrossRef][ISI][Medline]
Jaakola L, Pirttilä AM, Halonen M, Hohtola A. (2001) Isolation of high quality RNA from bilberry (Vaccinium myrtillus L.) fruit. Molecular Biotechnology 19 201–204.[CrossRef][ISI][Medline]
Jenkins AL, Bootman MD, Berridge MJ, Stone SR. (1994) Differences in intracellular calcium signaling after activation of the thrombin receptor by thrombin and agonist peptide in osteoblast-like cells. Journal of Biological Chemistry 269 17104–17110.
Jones HG. (1992) Plants and microclimateCambridge Cambridge University Press.
Kawasaki S, Borchert C, Deyholos M, Wang H, Brazille S, Kawai K, Galbraith D, Bohnert HJ. (2001) Gene expression profiles during the initial phase of salt stress in rice. The Plant Cell 13 889–906.
Kiba T, Yamada H, Sato S, Kato T, Tabata S, Yamashino T, Mizuno T. (2003) The type-A response regulator, ARR15, acts as a negative regulator in the cytokinin-mediated signal transduction in Arabidopsis thaliana. Plant and Cell Physiology 44 868–874.
Kirst M, Basten CJ, Myburg AA, Zeng Z-B, Sederoff RR. (2005) Genetic architecture of transcript-level variation in differentiating xylem of a eucalyptus hybrid. Genetics 169 2295–2303.
Ktistakis NT, Delon C, Manifava M, Wood E, Ganley I, Sugars JM. (2003) Phospholipase D1 and potential targets of its hydrolysis product, phosphatidic acid. Biochemical Society Transactions 31 94–97.[ISI][Medline]
Lee B-h, Henderson DA, Zhu J-K. (2005) The Arabidopsis cold-responsive transcriptome and its regulation by ICE1. The Plant Cell 17 3155–3175.
Lee S, Park J, Lee Y. (2003) Phosphatidic acid induces actin polymerization by activating protein kinases in soybean cells. Molecules and Cells 15 313–319.[ISI][Medline]
Li P, Mane SP, Sioson AS, Vasquez-Robinet C, Heath LS, Bohnert HJ, Grene R. (2006) Effects of chronic ozone exposure on gene expression in Arabidopsis thaliana ecotypes and in Thellungiella halophila. Plant, Cell and Environment 29 854–868.[CrossRef][Medline]
Li W, Li M, Zhang W, Welti R, Wang X. (2004) The plasma membrane-bound phospholipase D enhances freezing tolerance in Arabidopsis thaliana. Nature Biotechnology 22 427–433.[CrossRef][ISI][Medline]
Li W, Welti R, Wang X. (2005) The different roles of phopsholipase D1 and in water-deficit stress signaling in Arabidopsis thaliana. Proceedings of the international conference on plant lipid-mediated signaling: building connectionsRaleigh, NC.
Liu Y and Zhang S. (2004) Phosphorylation of 1-aminocyclopropane-1-carboxylic acid synthase by MPK6, a stress-responsive mitogen-activated protein kinase, induces ethylene biosynthesis in Arabidopsis. The Plant Cell 16 3386–3399.
Lopez-Andreo MJ, Gomez-Fernandez JC, Corbalan-Garcia S. (2003) The simultaneous production of phosphatidic acid and diacylglycerol is essential for the translocation of protein kinase C
to the plasma membrane in RBL-2H3 cells. Molecular Biology of the Cell 14 4885–4895.
Manifava M, Thuring JWJF, Lim Z-Y, Packman L, Holmes AB, Ktistakis NT. (2001) Differential binding of traffic-related proteins to phosphatidic acid- or phosphatidylinositol (4,5)-bisphosphate-coupled affinity reagents. Journal of Biological Chemistry 276 8987–8994.
Mikami K, Katagiri T, Iuchi S, Yamaguchi-Shinozaki K, Shinozaki K. (1998) A gene encoding phosphatidylinositol-4-phosphate 5-kinase is induced by water stress and abscisic acid in Arabidopsis thaliana. The Plant Journal 15 563–568.[CrossRef][ISI][Medline]
Mishra G, Zhang W, Deng F, Zhao J, Wang X. (2006) A bifurcating pathway directs abscisic acid effects on stomatal closure and opening in Arabidopsis. Science 312 264–266.
Ohashi Y, Oka A, Rodrigues-Pousada R, Possenti M, Ruberti I, Morelli G, Aoyama T. (2003) Modulation of phospholipid signaling by GLABRA2 in root-hair pattern formation. Science 300 1427–1430.
Pappan K, Zheng S, Wang X. (1997) Identification and characterization of a novel phospholipase D that requires polyphosphoinositide and submicromolar calcium for activity in Arabidopsis. Journal of Biological Chemistry 272 7048–7054.
Potocky M, Elias M, Profotova B, Novotna Z, Valentova O, Zarsky V. (2003) Phosphatidic acid produced by phospholipase D is required for tobacco pollen tube growth. Planta 217 122–130.[ISI][Medline]
Qin W, Pappan K, Wang X. (1997) Molecular heterogeneity of phospholipase D (PLD). Cloning of PLD and regulation of plant PLD, -ß, and - by polyphosphoinositides and calcium. Journal of Biological Chemistry 272 28267–28273.
Rabbani MA, Maruyama K, Abe H, et al. (2003) Monitoring expression profiles of rice genes under cold, drought, and high-salinity stresses and abscisic acid application using cDNA microarray and RNA gel-blot analyses. Plant Physiology 133 1755–1767.
Ramsay NA and Glover BJ. (2005) MYB–bHLH–WD40 protein complex and the evolution of cellular diversity. Trends in Plant Science 10 63–70.[CrossRef][ISI][Medline]
Rentel MC, Lecourieux D, Ouaked F, et al. (2004) OXI1 kinase is necessary for oxidative burst-mediated signalling in Arabidopsis. Nature 427 858–861.[CrossRef][Medline]
Ritchie S and Gilroy S. (2000) Abscisic acid stimulation of phospholipase D in the barley aleurone is G-protein-mediated and localized to the plasma membrane. Plant Physiology 124 693–702.
Rizhsky L, Liang H, Shuman J, Shulaev V, Davletova S, Mittler R. (2004) When defense pathways collide. The response of Arabidopsis to a combination of drought and heat stress. Plant Physiology 134 1683–1696.
Sakamoto H, Maruyama K, Sakuma Y, Meshi T, Iwabuchi M, Shinozaki K, Yamaguchi-Shinozaki K. (2004) Arabidopsis Cys2/His2-type zinc-finger proteins function as transcription repressors under drought, cold, and high-salinity stress conditions. Plant Physiology 136 2734–2746.
Sang Y, Cui D, Wang X. (2001a) Phospholipase D and phosphatidic acid-mediated generation of superoxide in Arabidopsis. Plant Physiology 126 1449–1458.
Sang Y, Zheng S, Li W, Huang B, Wang X. (2001b) Regulation of plant water loss by manipulating the expression of phospholipase D. The Plant Journal 28 135–144.[CrossRef][ISI][Medline]
Schrick K, Nguyen D, Karlowski WM, Mayer KFX. (2004) START lipid/sterol-binding domains are amplified in plants and are predominantly associated with homeodomain transcription factors. Genome Biology 5 R41.[CrossRef][Medline]
Seki M, Narusaka M, Ishida J, et al. (2002) Monitoring the expression profiles of 7000 Arabidopsis genes under drought, cold and high-salinity stresses using a full-length cDNA microarray. The Plant Journal 31 279–292.[CrossRef][ISI][Medline]
Sioson AA, Mane SP, Li P, Sha W, Heath LS, Bohnert HJ, Grene R. (2006) The statistics of identifying significantly expressed genes in Expresso and TM4: a comparison. BMC Bioinformatics 7 215.[CrossRef][Medline]
Stracke R, Werber M, Weisshaar B. (2001) The R2R3-MYB gene family in Arabidopsis thaliana. Current Opinion in Plant Biology 4 447–456.[CrossRef][ISI][Medline]