•Here, we investigated the role of ethylene in high nitrate-induced change in root development in Arabidopsis thaliana using wild types and mutants defective in ethylene signaling (etr1, ein2) and nitrate transporters (chl1, nrt2.1).
•The length and number of visible lateral roots (LRs) were reduced upon exposure of wild-type seedlings grown on low (0.1 mm) to high nitrate concentration (10 mm). There was a rapid burst of ethylene production upon exposure to high nitrate concentration.
•Ethylene synthesis antagonists, cobalt (Co2+) and aminoethoxyvinylglycine (AVG), mitigated the inhibitory effect of high nitrate concentration on lateral root growth. The etr1-3 and ein2-1 mutants exhibited less reductions in LR length and number than wild-type plants in response to high nitrate concentration. Expression of nitrate transporters AtNRT1.1 and AtNRT2.1 was upregulated and downregulated in response to high nitrate concentration, respectively. A similar upregulation and downregulation of AtNRT1.1 and AtNRT2.1 was observed by ethylene synthesis precursor aminocyclopropane carboxylic acid (ACC) and AVG in low and high nitrate concentration, respectively. Expression of AtNRT1.1 and AtNRT2.1 became insensitive to high nitrate concentration in etr1-3 and ein2-1 plants.
•These findings highlight the regulatory role that ethylene plays in high nitrate concentration-regulated LR development by modulating nitrate transporters.
Root system architecture is an important trait that determines the efficiency for water and nutrients acquisition in plants. Plants have evolved an array of mechanisms to modify their root system architecture in response to environmental stimuli (Malamy, 2005; Osmont et al., 2007; Nibau et al., 2008). This developmental plasticity is regulated by nutrient availability in soils in general (Forde, 2002; Desnos, 2008) and soil nitrogen status in particular (Zhang et al., 1999; Zhang & Forde, 2000; Linkohr et al., 2002). A dual effect of external nitrate on lateral root (LR) development in Arabidopsis thaliana has been recognized; a localized stimulatory effect of external nitrate on LR elongation (Zhang & Forde, 1998; Zhang et al., 1999; Linkohr et al., 2002), and a systemic inhibitory effect of globally high external nitrate concentrations (> 5 mm) on LR growth (Zhang & Forde, 1998; Zhang et al., 1999; Linkohr et al., 2002). The localized stimulatory effect of nitrate on LR development is mediated by a putative MADS-box transcriptional factor, a nitrate-inducible Arabidopsis gene (ANR1) (Zhang & Forde, 1998). In addition to ANR1, nitrate transporter, AtNRT1.1 (also called CHL1, Tsay et al., 1993), a dual-affinity nitrate transporter that mediates both low- and high-affinity nitrate transport across the plasma membrane (Tsay et al., 1993) may function as a nitrate sensor to participate in the stimulatory effect of low nitrate concentration on LR growth (Remans et al., 2006a). However, how plants sense the nitrate signal and transmit it to regulate the AtNRT1.1-ANR1 signaling pathway remains to be deciphered.
The systemic inhibitory effect of high nitrate concentration on LR occurs immediately after emergence of the LR primordium from the parent root, leading to accumulation of short LRs (Zhang et al., 1999). The internal nitrate concentration rather than accumulation of the products of nitrate assimilation is likely to be a critical factor modulating the nitrate-dependent response as the inhibitory effect is enhanced in a nitrate reductase-deficient mutant (Zhang et al., 1999). By contrast, the same nitrate reductase-deficient mutant is more sensitive to high external nitrate concentration than wild-type plants in terms of LR development (Zhang et al., 1999). These observations suggest that the systemic inhibitory effect of high nitrate concentration on LR development may result from the accumulation of nitrate inside plant tissues.
Long-distance signaling molecules that originate from shoots and are transported to the roots have been suggested to be associated with the suppression of LR development by high nitrate concentration (Zhang & Forde, 1998; Zhang et al., 1999). The involvement of auxin in high nitrate concentration-induced inhibition of root development has been suggested (Caba et al., 2000; Walch-Liu et al., 2006; Tian et al., 2008). These studies revealed that plants exposed to high nitrate concentration exhibited reduced indoleacetic acid (IAA) contents in roots because of either inhibition of auxin synthesis and/or inhibition of auxin transport from shoots to roots. In addition, there is also evidence suggesting that abscisic acid (ABA) may play a role in high nitrate-induced inhibition of root development. For example, the high nitrate-induced systemic inhibition of LR development is less in ABA synthesis and ABA-insensitive Arabidopsis mutants than in wild-type plants (Signora et al., 2001), and exogenous application of ABA can mimic high external nitrate concentration in suppression of LR meristem (De Smet et al., 2003). In addition to auxin and ABA, cytokinin (Sakakibara, 2003; Tian et al., 2005) and nitric oxide (Zhao et al., 2007) were also reported to be involved in nitrate-dependent root growth and development. Recent studies also indicated that the two nitrate transporters of NRT1.1 and NRT2.1, which underpin low and high affinity, and high-affinity nitrate transport across plasma membranes, respectively (Forde, 2000), play a signaling role in LR development in response to the external nitrate supply (Little et al., 2005; Remans et al., 2006a,b). Given that expression of NRT1.1 and NRT2.1 is sensitive to auxin (Guo et al., 2002) and ethylene (Leblanc et al., 2008), it is conceivable that the interactions between these phytohormones and nitrate transporters are of importance in modulation of nitrate-dependent root development.
There are multilevel interactions between ethylene and auxin in the regulation of root development in Arabidopsis (Stepanova et al., 2005, 2007; Růžička et al., 2007; Swarup et al., 2007; Ivanchenko et al., 2008; Negi et al., 2008). Ethylene is closely associated with diverse physiological processes in plants, including root growth and development (Pierik et al., 2006). Ethylene is synthesized from methionine through S-adenosyl-l-methionine and 1-aminocyclopropane-1-carboxylic acid (ACC), which are catalysed by ACC synthase (ACS) and ACC oxidase (ACO) (Kende, 1993). Both ACS and ACO are encoded by multigene families and regulated by developmental and environmental factors (Wang et al., 2002). A number of key proteins involved in the ethylene signaling pathways have been identified (Alonso & Stepanova, 2004; Chen et al., 2005). The etr1 mutant has a defect in a membrane receptor for ethylene (Chen et al., 2005). The EIN2 protein acts downstream of ETR1 and resembles the NRAMP family of metal transporters (Alonso & Stepanova, 2004). The etr1 and ein2 mutants have been widely used in deciphering physiological functions of ethylene in the literature (Alonso et al., 1999; Binder et al., 2006; Buer et al., 2006; De Grauwe et al., 2007). There have been a number of reports demonstrating that ethylene is closely associated with responses of root development to nutritional deficiency, including phosphorus starvation, and deficiency in iron (Schmidt, 2001) and potassium (Shin & Schachtman, 2004). The possible involvement of ethylene in nitrate-dependent physiological processes has also been implicated, as shown by induction of ethylene evolution from alfalfa (Medicago sativa) (Caba et al., 1998) and chickpea (Cicer arietinum) (Nandwal et al., 2000) by high external nitrate concentration and increased sensitivity of maize to ethylene under nitrogen deficiency (Schmelz et al., 2003). However, there has been no detailed study to evaluate the role of ethylene in high nitrate concentration-induced suppression of root development.
To elucidate the mechanism underlying the systemic inhibition of lateral root development by high nitrate concentration, we examined the role of ethylene in responses of root development to high nitrate concentration in A. thaliana. More specifically, both Arabidopsis wild-types and mutants that are insensitive to ethylene (etr1-3 and ein2-1) and defective in nitrate transporters (chl1-5 and nrt2.1-1) were used to study the effects of high nitrate concentration on ethylene production and root development.
Materials and Methods
Plant materials and growth conditions
Arabidopsis thaliana ecotypes of Columbia (Col), Wassilewskija (Ws), ethylene-insensitive mutants etr1-3, ein2-1 and nitrate transporter mutants chl1-5, nrt2.1-1 were obtained from the Arabidopsis Biological Resource Centre, Columbus, OH, USA. The EBS-GUS reporter line was generously provided by Dr J. Alonso, and was originally generated by Dr Anna Stepanova. The END199 line was a kind gift of Dr P. Benfey. All seeds were surface-sterilized by incubation for 1 min in 75% ethanol, and rinsed with sterile distilled water, followed by 15 min in 10% (v : v) sodium hypochlorite, and then washed with sterile water. The sterilized seeds were germinated on half-strength Murashige and Skoog (1/2 MS) agar plates (0.6% agar (w : v), pH 5.8 for 4 d). Thereafter the germinated seedlings of wild-type, etr1-3 and ein2-1 were transferred to 1/2 MS agar medium that was deprived of nitrogen salts and contained varying nitrate concentrations (0.1, 1, 10 mm) with 0.8% sucrose and 0.8% (w/v) agar, pH 5.8 for 5 d. The full chemical composition of the 1/2 MS is (mm): 10.3 NH4NO3, 9.4 KNO3, 1.5 CaCl2, 0.75 MgSO4, 0.625 KH2PO4, 0.28 Inositol and (μm) 50 FeSO4, 50 H3BO3, 50 MnSO4, 15 ZnSO4, 0.5 NaMoO4, 0.02 CuSO4, 2.45 KI, 0.05 CoCl2, 3 Vitamin B1 (VB1), 1 Vitamin B6 (VB6), 2 Vitamin B5 (VB5).
Nitrogen was supplied in nutrient medium as KNO3. To exclude the possibility that potassium (K+) may play a role in the treatments, the concentrations of K+ in the low-nitrate treatments were supplemented to the same levels as those of the high-nitrate concentration using KCl. All seedlings were grown in 9-cm diameter glass dishes, oriented vertically, in a controlled environment with a temperature 20 : 23°C, 14 h : 10 h light cycle and photosynthetic photon flux density (PPFD) of 100–120 μmol m−2 s−1.
Root elongation assays
Arabidopsis seedlings were grown for 5 d in treatment medium containing either low (0.1 mm) or high (10 mm) nitrate in the absence and presence of 1 μm AVG or 3 μM CoCl2. The length of primary and lateral roots was measured with a ruler. The number of lateral roots that were greater than approx. ≥ 0.5 mm in length was recorded. At least 10 independent replicates were used for each treatment. All experiments were repeated at least three times.
Measurements of ethylene production
As collection of roots from agar grown Arabidopsis seedlings for determination of ethylene was technically difficult, hydroponically cultured Arabidopsis seedlings were used to determine the effect of high nitrate concentration on ethylene production from roots. Our preliminary results showed that wild-type Arabidopsis seedlings hydroponically cultured in high nitrate concentration exhibited inhibition of LR elongation similar to those grown in agar medium with high nitrate concentration (data not shown). Arabidopsis seedlings germinated on 1/2 MS agar were transferred to 70 ml sterile nutrient solutions containing either low nitrate (LN, 0.1 mm) or high nitrate (HN, 10 mm) on orbital shakers with constant (c. 50 rpm), uniform fluorescent light (approx. 120 μmol in the flask) and temperature (22°C) as described by Morcuende et al. (2007). Care was taken to prevent the seedlings from clumping during incubation. Arabidopsis seedlings hydroponically cultured in LN and HN solutions for 5 d were transferred to HN and LN, respectively, for varying times (0, 0.5, 1, 2, 6 and 24 h), and were used to assay ethylene production. The negative controls were prepared by transferring Arabidopsis seedlings incubated in LN and HN solutions to LN and HN solutions, respectively. Roots were excised from Arabidopsis seedlings that were exposed to different incubating regimes (high and low nitrate concentration). To minimize wounding effect, the excised roots were placed in 5-ml vials containing 1 ml agar medium (0.7% agar) for 1 h, and thereafter the vials were sealed with the gas-tight stopper. The excised roots were kept moist and did not dry out during the 1-h period. One milliliter of headspace gas was taken from the vials after 1 h collection time and injected into a gas chromatograph equipped with an alumina column (GDX 502) and a flame ionization detector (GC-7AG; Shimadzu, Kyoto, Japan).
Assays for ACS and ACO activity
The ACS activity was determined as described by Woeste et al. (1999). Briefly, c. 0.2 g roots were ground with liquid nitrogen, and then added to 700 μl buffer A (200 mm phosphate buffer, pH 8.0, 10 μm pyridoxal phosphate, 1 mm EDTA, 2 mm phenylmethylsulfonyl fluoride (PMSF), and 5 mm dithiothreitol (DTT)). The samples were centrifuged at 15 000 g for 15 min and the supernatant was respun at 15 000 g for 15 min at 4°C. Three-hundred microliters of extraction was transferred to 5-ml vials containing 100 μl 5 mm S-(5′-Adenosyl)-L-methioneine (AdoMet) (Sigma-Aldrich). The reaction was carried out for 1 h at 22°C. The ACC formed was converted to ethylene by adding 100 μl of 10 mm HgCl2, followed by 100 μl of 1 : 1 mixture of saturated NaOH: bleach. The reaction vials were then sealed with rubber serum stoppers and kept in ice for 20 min. One milliliter of gas samples were withdrawn for ethylene determination. All reactions were conducted with four replicates. The ACO activity determination was performed according to Sun et al. (2007). Briefly, c. 0.5 g roots were ground with liquid nitrogen and then resuspended in extraction buffer (0.1 m Tris–HCl, pH 7.2, 10% (w : v) glycerol, 30 mm sodium ascorbate and 5% (w : v) polyvinyl polypyrrolidine (PVPP). After centrifuging at 15 000 g for 20 min at 4°C, the supernatant was used for ACO determination. The activity of ACO was assayed immediately by mixing 0.2 ml of crude extraction with a 2 ml reaction mixture containing 1.7 ml of extraction buffer (without PVPP), 50 μm FeSO4, 2 mm ACC, and incubated at 30°C. Ethylene produced in the head space of 5 ml capped tubes after a 1-h incubation was determined as described earlier.
The β-Glucuuronidase (GUS) stain was done essentially as described by Jefferson et al. (1987), Malamy & Benfey (1997) and Stepanova et al. (2005), with some modifications. Briefly, 5-d-old seedlings were pulled out of agar, fixed in an ice-cold 90% acetone, washed once with the rinse buffer (100 mm NaPO4 buffer, pH 7.0, 1 mm K3Fe(CN)6, and 1 mm K4Fe(CN)6, and stained for 4 h in dark at 37°C. The staining buffer comprised 100 mm NaPO4 buffer, pH 7.0, 1 mm K3Fe(CN)6, 1 mm K4Fe(CN)6, 10 mm Na2EDTA, 0.1% (v : v) Triton X-100, 20% (v : v) methanol and 0.5 mg ml−1 X-Gluc. For observation of whole mounts, stained seedlings were transferred to small Petri dishes containing 0.24 n HCl in 20% methanol and incubated on a 57°C heat block for 15 min. This solution was replaced with 7% NaOH, 7% hydroxylamine–HCl in 60% ethanol for 15 min at room temperature. Roots were then rehydrated for 5 min in 40%, 20% and 10% ethanol, respectively, and infiltrated for 15 min in 5% ethanol–25% glycerol. Roots were mounted in 50% glycerol on glass microscope slides and individual representative seedlings were photographed as described by Malamy & Benfey (1997).
Gene expression analysis
Real-time reverse-transcription polymerase chain reaction (RT-PCR) was used to study the expression pattern of ACS2, ACS4, ACS5, ACS6, ACS7, ACS8, ACS9, ACS11, ACO1, ACO2, NRT2.1 and NRT1.1 genes in response to different treatments, including varying nitrate concentrations, ethylene precursors and ethylene synthesis inhibitors. Total RNAs were extracted from Arabidopsis roots with Trizol reagent (Invitrogen) and treated with RNase free DNase I (Promega). The total RNAs were reverse-transcribed into first-strand cDNA in a 20 μl volume with M-MLV reverse transcriptase (Promega). The samples were diluted to 100 μl with water and 5 μl of each sample (approx. 8 ng RNA equivalent) was amplified using SYBR GreenER qPCR SuperMix Universal (Invitrogen) in a 25-μl reaction, containing 5 μl diluted cDNA, 12.5 μl SYBR GreenER qPCR SuperMix Universal, 0.5 μl Rox Reference Dye, 1 μl of 10 μm forward primer, 1 μl of 10 μm reverse primer, and 5 μl of water. The Mx3000P machine was used to run quantitative RT-PCR (qRT-PCR) with the following primer pair combinations:
AtACS2 5′-TCATGGGAAAAGCTAGAGGTGGAAG-3′ and 5′-TCAACGGTTAATTTGAAATTGTCGG-3′, AtACS4 5′-TTGTCTTGCAGATCCCGGTGA-3′ and 5′-TTGAGTTCGGTTTGGGTTGTTGT-3′
AtACS5 5′-GTTTTAGCGGCTGGTTCGACATCT-3′ and 5′-CAACGCAGTGCCAAGTGGGTTA-3′
AtACS6 5′-AAACCGATGGCTGCAACAACTATGA T-3′ and 5′-TAAGTCTGTGCACGGACTAGCGGAG-3′,
AtACS7 5′-CCTGGGTTCCGTGAAAACGCATT-3′ and 5′-CGTCGTTAGGATCGGCGAGAATGA-3′
AtACS8 5′-TGGGGTGATTTACTCCAACGATGAT T-3′ and 5′-GACACTCGATGCCTGCAGCCTCTA G-3′
AtACS9 5′-CTTGAAATGGAGAACGGGAGCAGAGAT-3′ and 5′-TCAACATTGTGCCAAGAGGGTTAGA CG-3′
AtACS11 5′-CTGGTTTCGGGTCTAAAGGAAGCGG-3′ and 5′-AATGACACGATGAGCCTGGAGAGATGTT-3′
AtACO1 5′-CCGTGTAATGACAGTGAAGCATGGA AG-3′ and 5′-TCTCAAGTCTGGGGCCTTTGTCTCC-3′, AtACO2 5′-GGATGTCGGTTGCATCGTTTTA-3′ and 5′-TACGGCTGCTGTAGGATTCAGTTC-3′, AtNRT2.1 5′-CTGGAGGGAACTTTGGATCAGGG-3′ and 5′-GTCACAGGTAACGTGCAAGCGACTA-3′, AtNRT1.1 5′-ATCCAAGCCACGGGTGTTTCAAT-3′ and 5′-CGTCACAGCTAAAAGAGAGCCAACG-3′.
In addition, a housekeeping gene, AtActin11, was employed as a control: 5′-CCACATGCTATTCTGCGTTTGGACC-3′ and 5′-CATCCCTTACGATTTCACG-CTCTGC-3′.
Primers were designed across exon–exon junctions of cDNA to avoid potential problems caused by contaminating genomic DNA. The amplification efficiency for each primer pair was calculated using serial cDNA dilutions. The expression values of the 12 genes were normalized to the corresponding controls.
High external nitrate concentration inhibited LR root growth
The high nitrate concentration (HN) had little effect on length of primary root (PR), while both LR length and visible LR number were reduced when Arabidopsis seedlings pregrown in the low nitrate (LN) medium were transferred to the HN agar medium (Fig. 1a–c). Furthermore, the inhibitory effect of high nitrate concentration on LR length was recovered when seedlings pregrown in HN medium were transferred to LN solution (Fig. S1). By contrast, the reduction in LR number was not readily reversed after transferring seedlings from agar medium containing high to low nitrate concentration for 24 h (Fig. S1).
To further explore the effect of high nitrate concentration on LR development, the GUS marker line END199, an enhancer trap line which expressed GUS activity in developing LRs after LR initiation (Malamy & Benfey, 1997), was used to identify the stages of LR development in response to the high nitrate concentration. Arabidopsis seedlings grown in the LN and HN solution were stained for GUS activity and lateral roots of both emerged and unemerged were classified into the four developmental stages as used by Zhang et al. (1999). The distribution of unemerged LRs (stages A and B) along the primary roots did not differ between seedlings grown in the LN and HN solution (Fig. 1d), suggesting that high nitrate concentration has no effect on LR initiation. The frequency for both stages, particular for stage A, declined with increasing distance from the root tip as more and more LR primordial emerged from epidermis (Fig. 1d). The difference in distribution of LRs in stages C and D between seedlings grown in low and high nitrate concentration became evident such that most LRs in the LN medium rapidly progressed through stage C to stage D, while LRs growing in the HN medium mainly accumulated at stage C and few LRs progressed to stage D (Fig. 1d). These results indicate that high nitrate concentration mainly inhibits immature LRs during a discrete phase just after their emergence from the primary root.
Ethylene evolution was enhanced in high nitrate medium
To uncover the role of ethylene in the inhibitory effect of high nitrate concentration on LR growth, the effect of high nitrate concentration on ethylene evolution from Arabidopsis wild-type seedlings was studied. As shown in Fig. 2, there was a rapid burst of ethylene production upon exposure of 5-d-old seedlings grown in LN medium to HN medium. The ethylene production peaked after exposure to high nitrate concentration for approx. 1 h, and thereafter exhibited a gradual decline with time (Fig. 2). Nevertheless, the ethylene production in seedlings exposed to high nitrate concentration for 24 h was significantly greater than that at low nitrate concentration. Ethylene evolution from roots grown in the low nitrate solution was lower than from roots incubated in the HN solution, and there was no significant change in ethylene evolution when Arabidopsis seedlings grown in LN solution were transferred to LN solution (Fig. 2). This result suggests that ethylene burst evoked by shifting Arabidopsis seedlings from LN to HN solution is unlikely to result from artifacts associated with the experimental protocols used in our study. Similar to the inhibition of LR length induced by high nitrate concentration, the ethylene evolution elicited by high nitrate concentration was also reversible by transfer of seedlings from high to low nitrate concentration (Fig. S2). The effect of high external nitrate concentration on ethylene production was further investigated by monitoring the ethylene reporter EBS-GUS activity. An enhanced EBS-GUS activity was observed in LR primordia and initiating LRs in response to treatment with the HN solution (Fig. 3a). By contrast, no changes in the EBS-GUS activities in PRs and mature LRs were observed in response to the HN treatment (Fig. 3b). The EBS-GUS activity in stele of PRs was also enhanced by the HN treatment (Fig. 3a).
Both ACC synthase and ACC oxidase are key enzymes responsible for ethylene production in plants. To test whether activities of the two enzymes were changed in response to high nitrate concentration, the effect of high nitrate concentration on ACS and ACO activity was investigated. Both ACS and ACO activity was markedly enhanced upon transferring seedlings from low to high nitrate concentration, and the high nitrate-induced increase in ACS and ACO activity was transient such that the ACS and ACO activity in seedlings after a 24-h exposure to high nitrate concentration was not significantly higher than that in low nitrate concentrations (Fig. 4a,b).
The elevated ethylene production induced by high nitrate concentration was also investigated at transcriptional levels by studying the effects of high nitrate concentration on expression of genes encoding ACS and ACO (AtACS and AtACO) with qRT-PCR. The Arabidopsis genome contains 12 ACS genes that encode eight functional ACS proteins (ACS2, ACS4-9 and ACS11) (Tsuchisaka & Theologis, 2004). Transferring seedlings from LN to HN solution upregulated seven out of eight ACS genes (AtACS2, AtACS4-8, AtACS11), while the expression of AtACS9 was not responsive to the external nitrate concentration (Fig. 4c). Similar to the expression of AtACS, enhanced expression of AtACO1, AtACO2 was also found in Arabidopsis seedlings grown in low nitrate concentration upon exposure to high nitrate concentration (Fig. 4d). In addition, the upregulation of AtACS and AtACO was greater after HN treatment for 6 h than for 24 h (Fig. 4c,d).
High nitrate-induced inhibition of root growth was alleviated by CoCl2 and AVG
To determine whether the inhibitory effect of high nitrate concentration on LR length and number results from ethylene production, we then investigated the effects of CoCl2 and AVG, which are antagonists for ACO (Lau & Yang, 1976) and ACS (Satoh & Yang, 1989), respectively, on LR length and LR number under conditions of both high and low nitrate concentrations. Both CoCl2 and AVG had no effect on PR length of Arabidopsis seedlings grown in the LN and HN solutions (Fig. 5a). Neither LR length nor LR number were affected by the two inhibitors under conditions of low nitrate concentration (data not shown). However, CoCl2 and AVG effectively reversed the reductions in LR length and number induced by the high nitrate concentration (Fig. 5b,c). Given that high nitrate concentration mainly inhibited growth of immature LRs (cf. Fig. 1), we thus determined whether AVG and CoCl2 can alleviate the inhibitory effect. Figure 5d shows that AVG and CoCl2 had little effect on stages A, B and C of LR development, suggesting that AVG and Co2+ do not affect LR initiation. However, AVG and CoCl2 alleviated substantially the inhibitory effect of high nitrate concentration on LR development at stage D (Fig. 5c,d), indicating that ethylene may be responsible for the suppression of immature LR development under high nitrate conditions.
The etr1-3 and ein2-1 mutants were insensitive to high nitrate concentrations
To confirm the role of ethylene in the high nitrate-induced suppression of LR growth and development, we also adopted a genetic approach by using ethylene-insensitive mutants, etr1-3 and ein2-1. The etr1-3 mutant has reduced ethylene response because of the dominant-negative versions of membrane ethylene receptor (O’Malley et al., 2005). The ein2-1 mutant is also insensitive to ethylene, but the biochemical function of EIN2 remains unclear (Alonso & Stepanova, 2004). In contrast to wild-type seedlings (Fig. 1), the root system architecture of both etr1-3 and ein2-1 was less responsive to the external nitrate levels (Fig. 6a). Like wild-type plants, PR length in both etr1-3 and ein2-1 was independent of the external nitrate concentration between 0.1 and 10 mm (Fig. 6a). However, in contrast to wild-type plants, LR length and number of the two mutants were reduced less by the high nitrate concentration compared with those of wild-type plants. For example, LR length of wild-type plants was reduced by 73% when exposed to the HN solution, while the same treatment reduced LR length by 37% and 30% in etr1-3 and ein2-1 plants, respectively (Fig. 6b). Similarly, high nitrate concentration reduced LR number of wild-type, etr1-3 and ein2-1 plants by 65%, 32% and 28%, respectively (Fig. 6c).
Chl1-5 and nrt2.1-1 mutants were less sensitive to high nitrate than wild-type plants
To understand the function of nitrate transporters of NRT1 and NRT2 in the high nitrate-induced inhibition of LR growth, we studied the response of root growth in Arabidopsis mutants (chl1-5 and nrt2.1-1) defective in AtNRT1 and AtNRT2, respectively, to varying external nitrate concentrations. Because nrt2.1-1 mutant originated from the wild-type Wassilewskija (Ws), the wild-type of Ws was also used in our study. Primary root length of both wild-types (Col-0 and Ws) and mutants (chl1-5 and nrt2.1-1) did not differ when grown in the high and low nitrate concentration (Fig. 7a). Upon exposure to the high nitrate concentration, there were significant reductions of LR length in wild-types (Col-0 and Ws), nrt2.1-1and chl1-5 plants (Fig. 7b). However, the reductions of LR length in the two wild-type plants were greater than those in their corresponding mutants in response to high nitrate concentration (Fig. 7b). For example, LR length was reduced by 57% and 61% in Col-0 and Ws, respectively, when challenged by high nitrate concentration. By contrast, the same treatment led to reductions of 28% and 34% in chl1-5 and nrt2.1-1 plants, respectively (Fig. 7b). Further, LR number in nrt2.1-1 plants did not differ under low or high nitrate concentrations (Fig. 7c), whereas significant reductions in LR number were found in the two wild-type and chl1-5 plants grown in the HN solution compared with those grown in the LN solution (Fig. 7c).
Expression of AtNRT2.1 and AtNRT1.1 was regulated by ethylene
To determine the role of the two nitrate transporters in nitrate- and ethylene-dependent root growth and development, expression of the two genes in response to high and low nitrate concentrations was examined. Expression of NTR1.1 was strongly upregulated in seedlings grown in the low nitrate concentration upon exposure to the high nitrate concentration, and amounts of the high nitrate-induced NRT1.1 transcript were greater after 6 h in the high nitrate concentration than after 24 h (Fig. 8a). In contrast to NRT1.1, expression of NRT2.1 was downregulated in response to transferring seedlings from low to high nitrate concentration (Fig. 8a). Unlike the upregulation of expression of NRT1.1, the high nitrate-induced downregulation of NRT2.1 expression was comparable between treatments with high nitrate for 6 and 24 h (Fig. 8a).
To examine whether NRT2.1 and NRT1.1 are involved in inhibition of LR growth by ethylene, the effect of ACC and AVG on expression of the two genes in the LN and HN medium was investigated. The expression of AtNRT1.1 was markedly upregulated by ACC in the low nitrate concentration (Fig. 8b), while the same treatment downregulated AtNRT2.1 (Fig. 8b). The upregulation and downregulation of AtNRT1.1 and AtNRT2.1 by the high nitrate concentration was repressed and stimulated by AVG and ACC, respectively (Fig. 8b). These observations suggest that ethylene is an important modulator in the regulation of nitrate-dependent expression of AtNRT1.1 and AtNRT2.1 genes. To further test this possibility, the response of AtNRT1.1 and AtNRT2.1 expression to the high nitrate concentration was also investigated using the ethylene-insensitive mutants etr1-3 and ein2-1. In contrast to wild-type plants, expression of NRT1.1 and NRT2.1 in etr1-3 and ein2-1 plants was independent of high external nitrate concentration (Fig. 8c).
Previous studies have shown that nitrate in the external medium have dual effects on LR development; localized nitrate supply stimulates LR growth and globally high concentration (> 5 mm) of nitrate suppresses LR growth (Forde, 2002; Zhang et al., 2007). Phytohormones such as ABA (Signora et al., 2001), auxin (Walch-Liu et al., 2006) and cytokinin (Sakakibara, 2003) may play regulatory roles in the nitrate-dependent LR growth and development. Given that root growth and lateral root formation is sensitive to ethylene (Lopez-Bucio et al., 2002) and that auxin and ethylene synergistically regulate root development, particularly lateral root development (Ivanchenko et al., 2008), it may be informative to examine whether ethylene is involved in the nitrate-dependent root development. In the present study, we demonstrated that exposure of Arabidopsis seedlings grown in low (0.1 mm) to high nitrate concentration (10 mm) evoked a rapid burst of ethylene production owing to upregulation of expression of ACS and ACO genes and increases in activities of ACS and ACO. Similar results, that high nitrate concentration induced ethylene production, have been reported in alfalfa (Caba et al., 1998) and chickpea (Nandwal et al., 2000). However, unlike these previous studies, we further identified that expression of genes encoding ACS and ACO, the two key enzymes responsible for ethylene synthesis, was transcriptionally upregulated by high nitrate concentrations (Fig. 4c,d). Moreover, we demonstrated that the high nitrate concentration-induced ethylene production occurred predominantly in both unemerged and emerged lateral roots, while ethylene levels in mature LR and PR were little changed, as shown by changes in activities of ethylene reporter EBS-GUS (Fig. 3). The differences in EBS-GUS activity between lateral and primary roots may underlie the observed the difference in sensitivity of root growth to the external nitrate concentrations between lateral and primary roots. Furthermore, it is interesting that the observed ethylene burst in response to high nitrate concentration did not turn on the ethylene signal in the primary root, leading to inhibition of root growth (cf. Fig. 1). This result suggests that different mechanisms may be involved in the perception and signaling of ethylene in primary and lateral roots.
To establish the links between the high nitrate concentration-induced ethylene production in LRs and inhibition of LR growth by high nitrate concentration, both pharmacological and genetic approaches were employed. Suppression of ethylene production by inhibiting ACS and ACO with AVG and Co2+ markedly alleviated the high nitrate-induced reductions in LR length and number (Fig. 6). More specifically, we found that the ameliorative effect of AVG and Co2+ on LR growth and development lies in their effects on lateral roots immediately after emergence of LR primordia from the parental roots (Fig. 5). Our results (cf. Fig. 1d) and previous studies (Zhang et al., 1999) have shown that high nitrate concentration mainly inhibited growth of immature LRs emerging from the primary root, leading to reductions in numbers of visible LRs (<0.5 mm), while the high nitrate concentration had no effect on LR initiation (Fig. 1d). In addition, inhibition of LR length by the high nitrate concentration was reversible upon transferring the seedlings grown in HN medium to the LN medium (Fig. S1). This result confirms results reported by Zhang & Forde (2000). The ethylene production evoked by high nitrate concentration can also be reversed when seedlings were shifted from high to low nitrate concentration (Fig. S2). Together, these findings suggest that the inhibitory effect of high nitrate concentration on immature LRs is likely to result from elevated ethylene production. Furthermore, our observations that reductions in LR length and number in ethylene-insensitive mutants etr1-3 and ein2-1 were less than those in wild-type plants (Fig. 6) in response to high nitrate concentration highlight the important roles of ETR1- and EIN2-dependent ethylene signaling pathways played in nitrate-dependent LR development.
There is ample evidence that nitrate transporters of NRT1.1 and NRT2.1 are involved in the signaling cascades of the nitrate-dependent root growth and development (Orsel et al., 2004; Little et al., 2005; Remans et al., 2006a,b; Walch-Liu & Forde, 2008). For example, NRT1.1 is associated with the localized stimulatory effect of nitrate supply through putative interactions with ANR1 (Remans et al., 2006a). NRT2.1 has been shown to repress lateral root initiation (Little et al., 2005). However, previous studies mainly focus on the role of nitrate transporters in localized nitrate effect and their response to low nitrate concentration. No detailed study has been conducted to investigate the involvement of nitrate transporters in sensing and responding to high nitrate concentration. In the present study, we revealed that the transcript levels of AtNRT1.1 and AtNRT2.1 genes displayed contrasting responses to the high nitrate concentration (i.e. expression of AtNRT1.1 and AtNRT2.1 was upregulated and downregulated, respectively, in response to a shift from low to high nitrate concentration (Fig. 8a). These observations are in good agreement with those reported results in the literature (Orsel et al., 2004; Little et al., 2005; Remans et al., 2006a,b). More importantly, we found that the expression AtNRT1.1 and AtNRT2.1 was sensitive to ACC and AVG such that ACC and AVG rapidly upregulated and downregulated expression of AtNRT1.1 and AtNRT2.1 at low and high nitrate concentrations, respectively (Fig. 8b). These findings are indicative that the high nitrate concentration-induced AtNRT1.1 and AtNRT2.1 expression may occur via ethylene. This argument is strengthened by the findings that expression of AtNRT1.1 and AtNRT2.1 in ethylene-insensitive mutants of ert1-3 and ein2-1 was no longer responsive to the high nitrate concentration (Fig. 8c). Leblanc et al. (2008) recently reported similar effects of ACC and AVG on expression of BnNRT1.1 and BnNRT2.1 in oilseed rape. However, unlike our study, they did not compare the effects of ACC and AVG on expression of BnNRT1.1 and BnNRT2.1 in low and high nitrate concentrations; rather, they studied responses of BnNRT1.1 and BnNRT2.1 expression to ACC and AVG in nitrate concentration at 1 mm exclusively, and a longer duration of treatments with ACC and AVG (> 24 h) was used in their study.
In addition to the dependence of AtNRT1.1 and AtNRT2.1 expression on nitrate concentration, we also demonstrated that mutants defective in NRT1.1 (chl1-5) and NRT2.1 (nrt2.1-1) exhibited altered sensitivity to the high nitrate concentration in terms of LR length and number (Fig. 7b,c). In comparison with wild-type plants (Col-0 and Ws), LR length in chl1-5 and nrt2.1-1 plants was less reduced in response to the high nitrate concentration (Fig. 7). These observations may be explained by the fact that mutation of low-affinity NRT1.1 transporter renders the chl1-5 unable to accumulate nitrate inside cells as the accumulation of nitrate inside root cells is implicated in the high nitrate-induced LR growth (Zhang et al., 1999). We have no explanation for the less inhibitory effect of high nitrate concentration on LR length in nrt2.1-1 than in wild-type plants. However, it is possible that mutation of NRT2.1 may also affect function of NRT1.1, leading to inhibition of nitrate accumulation as a recent study showed close interaction between the two nitrate transporters at transcription level (Muños et al., 2004). Further, the observation that LR number in nrt2.1 plants was no longer responsive to high nitrate concentration (Fig. 7c) is in agreement with the role of NRT2.1 in repression of LR initiation (Little et al., 2005). The rapid burst of ethylene production in response to high nitrate concentration and transcriptional dependence of AtNRT1.1 and AtNRT2.1 on ethylene seems to imply that ethylene acts e upstream of the nitrate transporters. Alternatively, the lower responsiveness of chl1-5 and nrt2.1 to high nitrate concentration than that of wild-type plants in terms of LR length and number may result from the differences in ethylene evolution between the mutants and wild-type plants in response to high nitrate concentrations. However, our observation that the mutants (chl1-5 and nrt2.1) did not differ from their corresponding wild-type counterparts in terms of high nitrate-induced ethylene production (data not shown) seems to discount this possibility.
In our study, Arabidopsis seedlings were first grown in the 1/2 MS medium that contained c. 20 mm nitrate for 4 d and were then exposed to the treatment regimes of LN and HN media. This may cause a recovery of root growth because of the reduction in nitrate concentration. However, we found that LR length, LR number and ethylene production were not affected by transferring seedlings from 20 mm nitrate to 10 mm nitrate media (data not shown). Conversely, seedlings grown in LN medium following 4-d growth in the 1/2 MS medium and in HN medium showed comparable responses to the high nitrate treatment in terms of LR length, LR number and ethylene production. Therefore, the experimental protocols used in the present study might not affect the overall conclusions that ethylene plays a regulatory role in mediation of high nitrate concentration-induced lateral root growth and development in Arabidopsis.
In conclusion, we demonstrated that exposure of Arabidopsis seedlings to high nitrate concentration led to a burst of ethylene production that occurred very rapidly and predominately in immature lateral roots. The elevated ethylene was closely associated with reductions in LR length and LR number, as demonstrated by mitigation of the inhibitory effect on LR with antagonists of ACS and ACO, and the lower sensitivity of ethylene insensitivity mutants than that of wild-type plants to nitrate concentration. We further revealed that nitrate transporters AtNRT1.1 and AtNTR2.1 were regulated transcriptionally by high nitrate concentration, and that the effects were mimicked by ACC at low nitrate concentration, and the dependence of AtNRT1.1 and AtNTR2.1 on high nitrate concentration was abolished by treatment with AVG. These findings indicate that inhibition of immature LR growth by high nitrate concentration is likely to result from elevated ethylene production, which in turn regulates NRT1 and NRT2, leading to the observed changes in root branching of Arabidopsis grown in high nitrate concentration.
This study was supported by the National Natural Science Foundation of China (30788003, 90817011 and 30800706). We thank Dr P. Benfey at Duke University for kindly providing seeds of END199, and Dr J. Alonso at North Carolina State University for providing seeds of the EBS-GUS reporter lines, which were generated by Dr Anna Stepanova in Dr Joe Ecker’s laboratory.