Notice: Wiley Online Library will be unavailable on Saturday 27th February from 09:00-14:00 GMT / 04:00-09:00 EST / 17:00-22:00 SGT for essential maintenance. Apologies for the inconvenience.
•With the exception of root hair development, the role of the phytohormone ethylene is not clear in other aspects of plant responses to inorganic phosphate (Pi) starvation.
•The induction of AtPT2 was used as a marker to find novel signalling components involved in plant responses to Pi starvation. Using genetic and chemical approaches, we examined the role of ethylene in the regulation of plant responses to Pi starvation.
•hps2, an Arabidopsis mutant with enhanced sensitivity to Pi starvation, was identified and found to be a new allele of CTR1 that is a key negative regulator of ethylene responses. 1-aminocyclopropane-1-carboxylic acid (ACC), the precursor of ethylene, increases plant sensitivity to Pi starvation, whereas the ethylene perception inhibitor Ag+ suppresses this response. The Pi starvation-induced gene expression and acid phosphatase activity are also enhanced in the hps2 mutant, but suppressed in the ethylene-insensitive mutant ein2-5. By contrast, we found that ethylene signalling plays a negative role in Pi starvation-induced anthocyanin production.
•These findings extend the roles of ethylene in the regulation of plant responses to Pi starvation and will help us to gain a better understanding of the molecular mechanism underlying these responses.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Phosphorus is one of the most important macronutrients that are essential for plant growth and development. Although phosphorus is abundant in the biosphere, the concentration of inorganic phosphate (Pi), the only form of phosphorus that can be absorbed by plant roots, is almost always far below the amount needed for the optimal performance of crops (Schachtman et al., 1998; Raghothama, 1999). Plants have developed a host of adaptive strategies to cope with this nutritional stress, such as changes of root architecture, increased phosphate transport activity, induction and secretion of acid phosphatases, ribonucleases and organic acids, accumulation of starch and anthocyanin, as well as remobilization of the internal Pi (Abel et al., 2002; Ticconi & Abel, 2004; Schachtman & Shin, 2007). However, the signalling mechanisms underlying these responses remain largely unknown.
The gaseous plant hormone, ethylene, plays an important role in coordinating internal and external cues to regulate plant growth and development, as well as several stress responses (Guo & Ecker, 2004; Benavente & Alonso, 2006). Several reports have indicated that ethylene is involved in Pi starvation-mediated inhibition of primary root growth and root hair formation (He et al., 1992; Borch et al., 1999; López-Bucio et al., 2002; Ma et al., 2003; Zhang et al., 2003). In the common bean, phosphorus-deficient roots produced twice as much ethylene as phosphorus-sufficient roots (Borch et al., 1999), and in maize plants, phosphate deficiency increased the sensitivity of roots to ethylene (He et al., 1992). However, the role of ethylene in other plant responses to Pi starvation, such as expression of PSI genes, production of acid phosphatases and anthocyanin, remains unclear.
In Arabidopsis, ethylene is detected by a family of receptors in the endoplasmic reticulum (ER), including ETR1, ETR2, ERS1, ERS2 and EIN4 (Hua & Meyerowitz, 1998; Guo & Ecker, 2004). The ethylene receptors redundantly function as negative regulators of ethylene responses through another negative regulator, CTR1 (Kieber et al., 1993; Hua & Meyerowitz, 1998; Bleecker & Kende, 2000). CTR1 is a Raf-like serine/threonine kinase that is associated with the ER membrane (Kieber et al., 1993; Clark et al., 1998; Gao et al., 2003). In the absence of ethylene, the receptors that activate CTR1 are constitutively active. Activated CTR1 suppresses the downstream ethylene-signalling process. Loss-of-function ctr1 mutants exhibit constitutive ethylene responses. EIN2, an ER membrane-associated protein, acts downstream of CTR1 and regulates the availability of the transcription factor, EIN3, through an unknown mechanism (Chao et al., 1997; Bisson et al., 2009). Loss-of-function ein2 mutants are completely insensitive to ethylene treatment.
In this study, we report the identification of an Arabidopsis mutant, hps2, which displays enhanced responses to Pi starvation. Genetic and molecular analysis indicated that the mutant phenotype is caused by the insertion of T-DNA into the CTR1 gene. We found that ethylene signalling plays an important role in regulating PSI gene expression, induction of acid phosphatases and production of anthocyanin. These results revealed a broad role of ethylene in the response of the plant to Pi starvation that has not previously been reported and will help us to gain a better understanding of the signalling pathway of the plant response to Pi starvation.
Materials and Methods
Plant material and growth conditions
The Pi-sufficient medium (P+), used in this study, was Murashige & Skoog (MS) medium (Murashige & Skoog, 1962) containing 1% (w/v) sucrose and 1.2% (w/v) agar (Sigma A1296; batch no. 097k0027; Sigma-Aldrich Co., St Louis, MO, USA). To make Pi-deficient medium (P−), 1.25 mM KH2PO4 in P+ medium was replaced with 0.65 mM K2SO4. All Arabidopsis thaliana (L.) Heynh. plants used in this study, including mutants and transgenic plants, were on the Colombia background. Seeds were surface sterilized with 20% (v/v) bleach for 10 min. After three washes in sterile distilled water, seeds were sown on the Petri dishes containing P+ or P− medium. After 2 d of stratification at 4°C, plates were vertically placed in the growth room with a photoperiod of 16 h light : 8 h, dark at 22–24°C. The light intensity was 100 μmol m−2 s−1. The mutant seeds ctr1-1, eto1-1, etr1-1 and ein2-5 were kindly provided by Dr Hongwei Guo of Peking University.
Luciferase imaging and mutant isolation
The Arabidopsis T-DNA activation tagging library was made with the transgenic plants expressing luciferase reporter gene under the regulation of AtPT2 promoters (AtPT2-LUC) (Karthikeyan et al., 2002; Koiwa et al., 2006). Seeds of T2 progeny of original T-DNA lines were plated on P− medium and grown in the growth room for 8 d. Luciferin (100 mM), dissolved in 0.1% (v/v) Triton X-100, was sprayed uniformly on the surface of the seedlings. Luminescence images were captured by exposing the plants for 5 min to an Andor iXon CCD (charge-coupled device) camera (Andor Technology Ltd., South Windsor, CT, USA). Putative mutants with enhanced or decreased luminescence compared with the wild-type plants were selected and their mutant phenotypes were further confirmed in the next generation. All mutants were backcrossed to the original AtPT2-LUC line to eliminate undesired mutations.
Analysis of β-glucuronidase expression
The hps2 mutant plant was crossed with the AtPT2-β-glucuronidase (GUS) transgenic line, and resultant F3 plants carrying the homozygote hps2 mutation and the AtPT2-GUS transgene were used for GUS expression analysis. Eight-day-old seedlings grown on MS P− medium were incubated at 37°C for 4 h in GUS-staining buffer (2 mM 5-bromo-4-chloro-3-indolyl-β-D-glucuronide (X-Gluc) in 50 mM sodium phosphate buffer, pH 7.2) containing 0.1% Triton X-100, 2 mM K4Fe(CN)6, 2 mM K3Fe(CN)6 and 10 mM EDTA. The stained seedlings were transferred sequentially to 50, 70 and 100% (v/v) ethanol to remove chlorophyll. The photographs of whole seedlings were taken under a stereo-microscope (Olympus SZ61, Tokyo, Japan). Quantification of GUS activity was performed as described by Blázquez (2002).
Cloning of T-DNA flanking sequence
The flanking sequence of the T-DNA insertion in the hps2 mutant was cloned using the thermal asymmetric interlaced (TAIL)-PCR method (Liu et al., 1995). Positive clones were sequenced and confirmed as containing the mutant genomic DNA by PCR using a T-DNA left border primer (LB3: 5′-TTGACCATCATACTCATTGCTG-3′) and a gene-specific primer (CTR1_specific: 5′-AACGCCAATGCAAGCTG-3′).
Treatment of ethylene precursor and action inhibitor
To monitor the effects of the ethylene precursor, 1-aminocyclopropane-1-carboxylic acid (ACC), or the action inhibitor, AgNO3, on expression of the AtPT2 gene, seeds containing AtPT2-LUC or AtPT2-GUS constructs were sown directly on MS P+ medium and grown for 5 d. The seedlings were then transferred to MS P+ or P− medium supplemented with different amounts of ACC or AgNO3 and growth was continued for another 7 d before the LUC images were taken or the expression of GUS was analyzed.
Real-time PCR analysis
Total RNAs were extracted using the Qiagen RNAeasy kit (Qiagen, Germantown, MD, USA). One microgram of DNase-treated RNA was reverse transcribed using M-MLV Reverse Transcriptase (Promega, Madison, WI, USA) according to the manufacturer’s instruction manual. Complementary DNAs (cDNAs) were amplified using SYBR Premix Ex Taq (Takara Bio Inc., Kyoto, Japan) and gene-specific primers (Supporting Information Table S1) on an Applied Biosystems 7500 Real-time PCR detection system (Applied Biosystems, San Mateo, CA, USA). UBC21 was used as an internal control, and the relative expression level of each gene was calculated using the 2−ΔΔCt method (Livak & Schmittgen, 2001).
Analysis of acid phosphatase activity
Nine-day-old seedlings, vertically grown on P+ or P− medium, were stained in vivo with acid phosphatase (APase), as described by Zakhleniuk et al. (2001) and 0.5% (w/v) agar containing 0.01% (w/v) 5-bromo-4-chloro-3-indolyl phosphate p-toluidine salt (BCIP) was evenly overlaid on the root surface. After 8 h of colour development, photographs were taken using a camera attached to a stereo-microscope (Olympus SZ61; Olympus).
About 50 mg of seedlings were ground in liquid N2 and the total protein was extracted in protein extraction buffer (0.1 M potassium acetate (KAc), 20 mM CaCl2, 2 mM EDTA, 0.1 mM phenylmethylsulfonyl fluoride). The same amount of protein was separated on a 10% sodium dodecyl sulphate–polyacrylamide gel. After electrophoresis, the gel was washed in double distilled water (ddH2O) at 4°C six times (10 min each) to renature the proteins and equilibrated twice with sodium acetate buffer (50 mM sodium acetate, 10 mM MgCl2, pH 4.9). Then, the gel was stained at 37°C for 4 h with 0.5 mg ml−1 of Fast Black K coupled to 0.3 mg ml−1 of β-naphthyl phosphate dissolved in sodium acetate buffer (Trull & Deikman, 1998; Zakhleniuk et al., 2001).
Quantitative analysis of APase activity was carried out by mixing 10 μl of protein extract with 470 μl of 2-(N-morpholino) ethanesulfonic acid (MES) buffer (15 mM MES, pH 5.5; 0.5 mM CaCl2) and 20 μl of p-Nitrophenyl phosphate (pNPP), (Sigma-Aldrich Co., St Louis, MO, USA). After incubation at 27°C for 30 min, the reaction was terminated with 500 μl of 0.25 M NaOH. The colour development was determined spectrophotometrically at 412 nm and the APase activity was expressed as A412 mg−1 of protein (Richardson et al., 2001).
The anthocyanin content was measured using 50 mg of ground shoot tissue from 2-wk-old seedlings grown on MS P+ or P− medium. The tissues ground in liquid N2 were extracted with 1% HCl in methanol (v/v) overnight. ddH2O (0.5 volume) and 1 volume of chloroform were added to remove the chlorophyll. After centrifugation for 5 min at 13 000 g, the upper aqueous phase was used for spectrophotometric quantification at 530 nm. The relative anthocyanin contents were expressed as A530 g−1 FW (Bieza & Lois, 2001).
Identification and characterization of the hps2 mutant
To identify novel signalling components involved in plant responses to Pi starvation, we performed a large-scale screen for Arabidopsis mutants with altered Pi-starvation responses. A luciferase gene (LUC) was fused to the promoter of a high-affinity phosphate transporter gene (AtPT2 or Pht1;4) and transformed into Arabidopsis plants (Col-0 ecotype) (Karthikeyan et al., 2002). The AtPT2 promoter is specifically induced by Pi starvation, especially in root tissues. So, when AtPT2-LUC transgenic plants (herein referred to as wild-type (WT) plants) are grown on low-Pi medium, they emit luminescence after the application of luciferin, a substrate of luciferase, on the plant surface. The AtPT2-LUC plants were then transformed with an activation tagging plasmid, pSuperTag2, to generate a T-DNA insertion library (Koiwa et al., 2006). T2 plants from c. 10 000 independent T-DNA lines were screened under P+ or P− conditions for mutants with an altered LUC signal relative to WT plants. From the screen carried out on plants grown on P− medium, 54 mutants with either enhanced or reduced expression of the AtPT2-LUC gene were recovered, and a detailed characterization of one such mutant, hps2 (hypersensitive to phosphate starvation 2), is presented here.
When grown on P− medium for 8 d, the hps2 mutant emitted much stronger luminescent light in both its root and cotyledons compared with WT plants (Fig. 1a). In addition to enhanced AtPT2-LUC expression, the mutant also showed stronger inhibition of primary root growth and formed more root hairs compared with WT plants (Fig. 1b). These mutant phenotypes indicated that hps2 had hypersensitive responses to Pi starvation. On P+ medium, the hps2 mutant showed similar morphological phenotypes as on P− medium, that is, a shorter primary root and more root hairs, mimicking a constitutive Pi-starvation response (Figs 2b, S1). To ensure that the mutant phenotype of enhanced AtPT2-LUC expression was not caused by a mutation in the AtPT2-LUC transgene, we introduced the AtPT2-GUS reporter gene into the hps2 mutant through a genetic cross. Indeed, the expression of the AtPT2-GUS gene was also enhanced in the hps2 mutant (Fig. 1c). The enhanced GUS expression was confirmed by quantitative measurement of GUS activity (Fig. 1d). Furthermore, when grown on P− medium, the endogenous level of AtPT2 messenger RNA (mRNA) was higher in hps2 plants than in WT plants, as revealed by real-time PCR analysis (Fig. 3a). All of these results demonstrated that the HPS2 gene encodes a regulatory component involved in controlling AtPT2 gene expression.
To further compare the sensitivity of hps2 and WT plants to Pi starvation, we grew WT and hps2 plants on MS medium supplemented with different amounts of phosphate. Expression of the LUC gene in WT plants could only be observed when the Pi concentration in culture medium was reduced to 0.01 mM, whereas the expression of the LUC gene could already be detected when the Pi concentration was as high as 0.10 mM (Fig. S1).
hps2 is a new allele of the CTR1 gene
Analysis of F1 and F2 progeny derived from a cross between hps2 and WT plants indicated that mutant phenotypes of the enhanced AtPT2-LUC gene, a short primary root and more root hairs can be attributed to a single recessive mutation (Fig. S2, and data not shown). Further linkage analysis showed that the mutant phenotypes were associated with T-DNA insertion (data not shown). It was found that the T-DNA is inserted into the first exon of the CTR1 gene in hps2 plants, 50 bp downstream of the start codon, ATG (Fig. 2a). The CTR1 encodes a Raf1-like kinase that is a key negative regulator of the ethylene-signalling pathway (Kieber et al., 1993). Throughout the life cycle, the hps2 mutant displayed developmental characteristics of ctr1-1, a loss-of-function mutant of the CTR1 gene (Fig. 2b, Kieber et al., 1993). Furthermore, when the hps2 mutant was crossed to ctr1-1, the F1 progeny showed a phenotype similar to that of their parents, thus confirming that hps2 is a new allele of the CTR1 gene (Fig. S3).
To determine if other known ethylene-signalling components were involved in the regulation of AtPT2 gene expression, we examined mRNA level of AtPT2 using real-time PCR in ethylene signalling or biosynthesis mutants. These include etr1-1, which has a defect in an ethylene receptor, and ein2-5, which contains a mutation in an ER membrane protein, resulting in a complete insensitivity to ethylene. When under Pi starvation, the expression of AtPT2 was greatly reduced in these two mutants compared with WT plants (Fig. 3a). By contrast, the AtPT2 expression level was enhanced in the ethylene over-producing mutant, eto1-1, under P− conditions, similar to that of hps2 plants. Consistent with real-time PCR results, when AtPT2-GUS plants were introduced into ein2-5 or eto1-1 mutants, the expression of AtPT2-GUS was almost completely blocked or greatly enhanced in the respective background (Fig. 3b).
To further elucidate the role of ethylene in regulating AtPT2 gene expression, we grew AtPT2-LUC or AtPT2-GUS plants on P− medium for 5 d, and then transferred them to the same medium supplemented with different amounts of Ag+, an inhibitor of ethylene perception. Expression of both reporter genes (LUC and GUS) was suppressed by 5 μM Ag+ (Figs 4a, S4). We then grew AtPT2-LUC plants on P+ medium for 5 d, and transferred them to P+ or P− medium containing different concentrations of ACC, a precursor for ethylene biosynthesis, for another 7 d. On P+ medium, 0.5 μM ACC did not induce AtPT2-LUC expression at all (Fig. 4b), and only when the ACC concentration reached 25 μM, was there a noticeable induction of the AtPT2-LUC gene. By contrast, on P− medium, even 0.5 μM ACC could significantly enhance AtPT2-LUC expression. As the enhancement of AtPT2-LUC gene expression by the low concentration of ACC only occurred under P− conditions, this indicated that there is an interaction between ethylene and a low phosphate signal on AtPT2 gene expression. In fact, we observed that expression of three genes (ACS2, ACS4 and ACS6) involved in ethylene biosynthesis was increased under Pi starvation (Fig. S5), especially of ACS6 that was induced to three folds high. This implied that Pi starvation might increase ethylene biosynthesis in plants.
To determine if the ethylene-signalling pathway is also involved in regulating other PSI genes, the mRNA levels of another high-affinity phosphate transporter (AtPT1) (Muchhal et al., 1996), two noncoding transcripts (At4 and AtIPS1) (Burleigh & Harrison, 1999; Martín et al., 2000), an acid phosphatase (ACP5) (del Pozo et al., 1999), a ribonuclease (RNS1) (Bariola et al., 1994) and miR399D (Fujii et al., 2005) were examined in hps2 or ein2-5 mutants by real-time PCR analysis. The results indicated that the expression of all six PSI genes examined, except for AtIPS1, were greatly enhanced in hps2 mutants compared with WT plants under Pi starvation (Fig. 5). By contrast, the expression of five out of six PSI genes was significantly reduced in ein2-5 mutants. The mutation on the hps2 gene also enhanced the basal level of expression of AtPT1, ACP5, At4 and RNS1 on P+ medium, whereas a reduction in the basal level of expression of miRNA399D and RNS1 was observed in the ein2-5 mutant.
Ethylene signalling positively regulates Pi starvation-induced APase activity
Under Pi deficiency, plants increase the synthesis and secretion of APase to acquire internal or external organic phosphate. To determine if ethylene signalling plays any role in Pi starvation-induced production of APases, we grew WT, hps2 and ein2-5 plants on P+ or P− medium for 9 d, and then examined their APase activities using three methods. First, APase activities on the root surface were examined by overlaying agar solution containing the APase substrate, BCIP (Zakhleniuk et al., 2001). As shown in Fig. 6(b), on P− medium, the blue colour of BCIP staining of the hps2 root was stronger than that of WT plants, whereas the roots of ein2-5 plants were stained more weakly. For hps2, we could even detect the APase activity on P+ medium, which mimics a constitutive Pi-starvation response as its root morphological phenotype (Fig. 6a). Second, we extracted soluble proteins from the whole seedlings of these three types of plants. The proteins were separated by nonreducing sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE). The gel was then renatured and incubated with Fast Black K coupled with β-naphthyl phosphate, a substrate of acid phosphatases. As shown in Fig. 6(c), for the 95 and 250 kDa isoforms of APase, hps2 had stronger activity, while ein2-5 had weaker activity than WT under both P+ and P− conditions. Quantitative analyses further showed that hps2 plants had higher APase activities on both P+ and P− medium than did WT plants (Fig. 6d).
Ethylene signalling negatively regulates Pi starvation-induced anthocyanin accumulation
Although anthocyanin accumulation is known to be caused by various biotic and abiotic stresses, it is controlled by signalling pathways specific to each stress type (Rubio et al., 2001). When hps2 and WT plant seeds were sown on P− medium and grown for 2 wk, the two cotyledons of the WT plants turned purple, but their newly formed true leaves still stayed relatively green. At the same time, the whole seedlings of the hps2 mutant were not visibly purple (Fig. 7a). By contrast, the whole seedlings of ein2-5 mutants, including both cotyledons and two true leaves, turned dark purple. Quantitative analysis showed that, under Pi starvation, the anthocyanin content in the hps2 seedlings was only about half of that of WT plants, whereas ein2-5 plants accumulated 50% more anthocyanin than WT plants (Fig. 7b).
To further examine how the ethylene-signalling pathway affects Pi starvation-induced anthocyanin production, we compared the expression level of four genes, which are involved in anthocyanin metabolism (Holton & Cornish, 1995; Grotewold, 2006), in the WT, hps2 and ein2-5 plants. These four genes are PAP1 (or AtMYB75, a transcription factor), F3′H (flavone 3′-hydroxylase), UF3GT (UDP-Glc-flavonoid 3-O-glucosyltransferase) and LDOX (leuco-anthocyanidin dioxygenenase). Expression of these four genes was much higher in plants grown on P− medium compared with plants grown on P+ medium, regardless of their genotype (Fig. 7c). However, on the P− medium, the gene-expression levels of PAP1, UF3GT and LDOX in hps2 plants were significantly lower than those of WT plants. By contrast, the expression levels of all four genes examined in ein2-5 plants were at least double those of WT plants. These data indicate that the ethylene-signalling pathway negatively regulates the expression of anthocyanin metabolism genes to affect Pi starvation-induced anthocyanin accumulation.
In this work, we used a combination of molecular and genetic approaches to search for novel signalling components involved in plant responses to Pi starvation. We used the AtPT2 gene as a marker to identify the mutants with an altered response to Pi starvation. From a large-scale screen, an Arabidopsis mutant, hps2, which exhibits enhanced expression of the AtPT2 gene under Pi starvation, was recovered. Our genetic and molecular characterization of the hps2 mutant indicated that the enhanced AtPT2 expression phenotype is caused by insertion of a T-DNA in the CTR1 gene, which, when inactivated, render plants display constitutive ethylene responses (Kieber et al., 1993; Guo & Ecker, 2004). By contrast, AtPT2 expression was significantly reduced in the ethylene-insensitive mutants, ein2-5 and etr1-1, under conditions of low Pi (Fig. 3). Our data also showed that the expression of AtPT2-LUC in response to Pi deficiency was greatly enhanced by the ethylene biosynthesis precursor ACC, whereas the expression of AtPT2-LUC was almost completely blocked by Ag+, an inhibitor of ethylene perception (Fig. 4). Interestingly, it was found that on Pi-sufficient medium, at a concentration of 25 μM Pi, ACC has little effect on the induction of the AtPT2-LUC or AtPT2-GUS genes, whereas on Pi-deficient medium, even 0.5 μM ACC was sufficient to enhance expression of AtPT2-LUC. This result implies that the activation of ethylene signalling alone is not sufficient to induce expression of the PSI gene to the degree caused by Pi starvation. It suggests cross-talk between ethylene and low-Pi-induced signals, probably through some as-yet-unidentified factors, to regulate the expression of the PSI gene under low Pi. The existence of such a cross-talk is also supported by the fact that the expression of PSI genes is much higher in Pi-starved hps2/ctr1 plants than in nonstarved hps2/ctr1 plants, although the degree of activation of the ethylene signalling pathway in both type of plants is same, that is, constitutive. Ethylene signalling also affects the expression of other PSI genes that have been widely used as Pi starvation-induced markers (Fig. 5). The products of these PSI genes are thought to be directly involved in Pi uptake (AtPT1 and AtPT2), mobilization (ACP5 and RNS1) and control of Pi distribution between shoots and roots (At4 and miRNA399), which is critical for the plant to maintain Pi homeostasis under Pi starvation (Yuan & Liu, 2008). Similarly to AtPT2, the expression of those PSI genes was also largely enhanced in a constitutive ethylene-response mutant (hps2/ctr1) and significantly reduced in an ethylene-insensitive mutant (ein2-5). Taken together, our results provided the first genetic, molecular and biochemical evidence that ethylene is an important mediator for regulation of PSI genes by low Pi stress. At the same time, we also noticed that ein2-5, a complete ethylene-insensitive mutant, did not fully block the expression of PSI genes, suggesting that ethylene-independent pathways also exist to control PSI gene expression.
In some early reports, the function of ethylene in regulating the Pi starvation-induced inhibition of primary root growth and development of root hair has been extensively studied (Ma et al., 2001, 2003; Schmidt & Schikora, 2001; Zhang et al., 2003). However, so far, the role of ethylene in regulating other aspects of plant responses to Pi starvation has remained elusive. In this work, we showed that, besides control of PSI gene expression, the ethylene signal also regulates the production of acid phosphatases and anthocyanin, two other markers of Pi starvation-induced responses. In an hps2 mutant, even on Pi-sufficient medium, APase activity could be observed on the root surface (as revealed by BCIP staining) in addition to enhanced accumulation of several APase isoforms and an increase in total APase activity (Fig. 6). By contrast, the APase activities in ein2-5 mutants were lower on P− medium compared with wild-type plants (Fig. 6b). Thus, ethylene signalling has a similar effect on APase production as in regulating PSI gene expression in Arabidopsis. Furthermore, the hps2 mutant accumulated less anthocyanin, whereas the ein2-5 mutant accumulated more anthocyanin on P− medium compared with wild-type plants. These effects were correlated with changes in the transcript levels of several genes regulating anthocyanin metabolism (Fig. 7). Reduced anthocyanin may be an indication of alleviation of stress caused by activation of the ethylene-signalling pathway. The phenomenon that alteration of ethylene signalling could affect stress-induced symptoms has also been reported with respect to salt stress and potassium deprivation (Cao et al., 2007; Jung et al., 2009).
In conclusion, our work, along with the earlier studies on root system architecture by other groups (Ma et al., 2001; Schmidt & Schikora, 2001; Zhang et al., 2003), provides compelling evidence that ethylene signalling is involved in multiple responses of plants to Pi starvation. Here, we proposed a simple model to illustrate how ethylene affects plant adaptation to Pi starvation (Fig. 8). Inorganic phosphate starvation might induce ethylene production, as observed in common bean (Borch et al., 1999), or enhance plant sensitivity to ethylene, as seen in root tissues of Pi-starved maize plants (He et al., 1992). The increased ethylene production or ethylene responsiveness will then enhance root hair development and the expression of PSI genes. It is well known that the change of expression of some PSI genes could directly affect Pi uptake from soils, internal and external Pi remobilization, and Pi redistribution between roots and shoots, which together enable plants to maintain Pi homeostasis when grown under Pi starvation (Duff et al., 1994; Poirier & Bucher, 2002; Doerner, 2008; Pant et al., 2008; Liu et al., 2010). Thus, we believe that ethylene signalling is an important component that makes plants better able to adapt to low-Pi stress, possibly through its control of Pi homeostasis by affecting PSI gene expression. Besides, the interactions between ethylene and the low-Pi signalling pathway indicate that some unknown factors may serve as a converging point between ethylene and the low-Pi signal in the regulation of plant responses to Pi starvation. It would be of great interest to further identify these signalling components, which will definitely deepen our understanding of the molecular mechanism underlying plant responses to Pi starvation.
We thank the Arabidopsis Biological Resource Center and Dr Hongwei Guo of Peking University for providing the mutant seeds of etr1-1, eto1-1, ctr1-1 and ein2-5. Thanks are also due to Ms Hui Yuan for her technical assistance and to Bo Zhang and Liqiang Tang of Cold Spring Science Corporation for their help in setting up the CCD imaging system. This work was supported by the National Natural Science Foundation of China (grant no. 30670170), the National Key Basic Research Program of China (grant no. 2009CB119100) and Beijing Natural Science Foundation (grant no. 5082010).