SEARCH

SEARCH BY CITATION

Keywords:

  • Arabidopsis;
  • auxin redistribution;
  • lateral root;
  • mild salt stress;
  • SOS3;
  • the SOS signaling

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • The SOS signaling pathway plays an important role in plant salt tolerance. However, little is known about how the SOS pathway modulates organ development in response to salt stress. Here, the involvement of SOS signaling in NaCl-induced lateral root (LR) development in Arabidopsis was assessed.
  • Wild-type and sos3-1 mutant seedlings on iso-osmotic concentrations of NaCl and mannitol were analyzed. The marker lines for auxin accumulation, auxin transport, cell division activity and stem cells were also examined.
  • The results showed that ionic effect alleviates the inhibitory effects of osmotic stress on LR development. LR development of the sos3-1 mutant showed increased sensitivity specifically to low salt. Under low-salt conditions, auxin in cotyledons and LR primordia (LRP) of the sos3-1 mutant was markedly reduced. Decreases in auxin polar transport of mutant roots may cause insufficient auxin supply, resulting in defects not only in LR initiation but also in cell division activity in LRP.
  • Our data uncover a novel role of the SOS3 gene in modulation of LR developmental plasticity and adaptation in response to low salt stress, and reveal a new mechanism for plants to sense and adapt to small changes of salt.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Plants are immobile, and rely on very flexible plastic development of root system architecture (RSA) to cope with the changing environment. The central component of this adaptive system is lateral root (LR) initiation and development which is formed postembryonically. LR number, spacing, initiation and outgrowth are generally determined by environmental stimuli, so that they can reach mineral nutrients and water to meet plants’ needs (Malamy & Ryan, 2001; Malamy, 2005; Ditengou et al., 2008). Modulating LR development is also a very important survival strategy for plants to avoid damage in unfavorable conditions, such as high salinity, heavy metals, etc. (Ivanov et al., 2003; Hagemeyer, 2004; Sun et al., 2008). Therefore, plants have a very distinct RSA in response to different environmental cues, and produce different RSA when they encounter various degrees of the same stress.

The molecular basis of LR development has been studied in Arabidopsis and considerable progress has been made. Arabidopsis LR development has generally been divided into two stages: LR initiation and LR emergence (Laskowski et al., 1995). Recently, eight detailed chronological developmental stages of LR development have been defined (Volder et al., 2005). Clearly, LR development involves multiple processes and is controlled by a complex regulatory network. So far, many genes have been shown to relate to LR development, but the most exciting progress is the recognition of the central regulatory role of auxin and its polar transport in both LR initiation and emergence (Boerjan et al., 1995; Fukaki et al., 2002; Marchant et al., 2002). For example, auxin triggers the xylem pole pericycle cells to divide at a very early stage (De Smet et al., 2007). Some PIN transporters have been found to be essential for establishment of an auxin gradient during LR primordia (LRP) formation (Steinmann et al., 1999; Bhalerao & Bennett, 2003; Friml, 2003). After LRs are initiated, LR meristem activation and outgrowth also depend on auxin (Celenza et al., 1995; Wu et al., 2007). Recently, the roles of ethylene and abscisic acid on LR development have also been highlighted (Ivanchenko et al., 2008; Negi et al., 2008). One emerging question is how environmental cues constitute intrinsic signals to regulate LR development. To this end, the status of nutrients such as phosphate and nitrate has been shown to greatly affect LR development (Remans et al., 2006; Perez-Torres et al., 2008). Recent preliminary results start to shed light on the plastic development of LRs in response to osmotic and salt stress (Deak & Malamy, 2005; MacGregor et al., 2008; Wang et al., 2009; Zolla et al., 2010).

Soil salinity is a major abiotic stress in agriculture worldwide, which affects almost all aspects of plant development, including germination, shoot growth and RSA development (Lazof & Bernstein, 1998; Almansouri et al., 2001; Wang et al., 2009). In recent decades, physiological and biochemical evidence has demonstrated that excessive sodium (Na+) accumulated in the cells is the primary cause of inhibition of plant growth (Ungar, 1996; Ghoulam et al., 2002). In addition to ion toxicity, salinity also imposes osmotic stress and nutrient deficiency such as Ca2+ and K+ on plants (Zhu et al., 1998; Zhu, 2000). Therefore, maintenance of low cytosolic Na+ is critical for plant adaptation to salt stress. The strategies for protecting plants from high cytosolic ion accumulation include compartmentalization of Na+ into vacuoles, extrusion of Na+ and limitation of ion uptake (Zhu, 2002, 2003). Recent advances have revealed that plastic development, including primary root (PR) elongation, LR development, and growth direction, is also a very important strategy for plants to avoid high salinity damage (Sun et al., 2008; Wang et al., 2009; Zolla et al., 2010). It has been found that salinity affects LR number and the growth of both LRs and PRs in Arabidopsis through modulating auxin gradient and redistribution, shedding light on the interplay between stress signaling and internal developmental signal pathways (Wang et al., 2009; Zolla et al., 2010). However, the detailed molecular mechanisms underlying the salt stress-induced morphogenesis response in Arabidopsis roots remain elusive.

Molecular genetic evidence has identified the SOS signaling pathway through characterization of the salt overly sensitive (sos1, sos2 and sos3) Arabidopsis mutants, which specifically mediate ion homeostasis and salt tolerance in plants (Zhu, 2002, 2003; Chinnusamy et al., 2005). After perception of Na+ by an unknown sensor, the activated cytosolic calcium signals are transduced by SOS3, an EF-hand Ca2+-binding protein, which then interacts with SOS2, a Ser/Thr protein kinase, and activates it (Liu & Zhu, 1998; Halfter et al., 2000; Liu et al., 2000). The SOS3/SOS2 kinase complex phosphorylates and activates SOS1, encoding a Na+/H+ antiporter to extrude sodium (Quintero et al., 2002). Further results have shown that the SOS3/SOS2 complex plays a critical role in controlling the activities of SOS1, as well as AtNHX1, the first identified plant vacuolar protein transporting Na+ into vacuoles (Apse et al., 1999; Zhang & Blumwald, 2001; Shi et al., 2002; Qiu et al., 2004). Although the SOS genes and their biochemical functions have been extensively studied, their roles in regulation of root system development, in particular LR development and the interaction between SOS signaling and auxin, remain largely unknown.

In this study, we carried out a systematic analysis of the LR development of the wild-type (WT) seedlings of Arabidopsis under iso-osmotic potential NaCl and mannitol stress, and found that LR development was inhibited by both NaCl and mannitol, but the extent of inhibition of LR development caused by NaCl appears to be much lower than that caused by the iso-osmotic treatment of mannitol. By phenotypic analysis of the sos mutants with the emphasis on sos3-1, we provided evidence that SOS signaling plays an important role in LRP initiation and is essential for LRP emergence under low salt. Further, we reported that the specific regulatory role of SOS3 in LR development under low salt stress is through modulation of auxin gradient and auxin polar transport. Our findings provide evidence that SOS3 is required for sufficient auxin supply for LRP initiation and maintenance of cell division activity of the LRP cells during LR development under low salt stress.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Plant materials and growth conditions

Seeds of the sos1-1, sos2-1, and sos3-1 mutants in the gl1 background and the relative WT gl1 in Col-0 were obtained from Dr J-K Zhu’s laboratory. The marker lines in the Col-0 background used in the present study are as follows: DR5::GUS (Stepanova et al., 2005), CYCB1;1::GUS (Jain et al., 2007), WOX5::GUS (Sarkar et al., 2007), PIN1:YFP (Zhuang et al., 2006), and PIN2:GFP (Xu & Scheres, 2005). The seeds were sterilized with 50% (v/v) commercial bleach for 7–8 min, followed by five rinses with sterilized water. The seeds were planted on a control medium containing Murashige and Skoog (MS) nutrient mix (Sigma-Aldrich), 2% sucrose and 0.8% agar, pH 5.7. After 2 d stratification at 4°C, the seeds were germinated and grown at 23°C under 16 h light/8 h dark photoperiods.

Stress treatments

For the stress treatments, seeds of the WT were germinated on the control medium, with an osmotic potential of −0.05 Mpa, or the control medium containing NaCl or mannitol with osmotic potentials of −0.15, −0.25, −0.45, −0.7, −0.85 MPa for 10 d. For the low-salt treatment, seeds of the WT and sos3-1 were germinated on control medium with or without 30 mM NaCl. The water potential of the control and stress medium was determined by WP4-T Potentia Meter (Decagon Devices, Pullman, WA, USA). For the hormonal treatments, 10-d-old sos3-1 seedlings grown on MS medium containing 30 mM NaCl were transferred to MS medium containing 30 mM NaCl with and without 75 nM 1-naphthylacetic acid (NAA) (Sigma). The effects of 2,3,5-triiodobenzoic acid (TIBA, Sigma) on the LR development of sos3-1 were tested by transferring the 10-d-old seedlings of sos3-1 on 30 mM NaCl to MS medium ± 100 nM TIBA. The NAA and TIBA stock solutions were prepared in dimethyl sulfoxide (DMSO) and kept at 4°C, and the proper amount of the solution was added to the agar medium before it solidified. For the control cultures, an equivalent volume of DMSO was included in the medium.

Assays for PRs and LRs

Arabidopsis seedlings were scanned on the agar plates at 600 dpi using a desktop scanner (Epson Perfection 1670). The PR lengths were measured using the scanned images. Visible LRs were observed under a stereo microscope (Olympus, SZX7, Tokyo, Japan), and those LRs that penetrated the root epidermis were counted. These experiments were repeated, and data are means ± SD for three replicates.

GUS-histochemical analysis

The DR5::GUS, CYCB1;1::GUS and WOX5::GUS seedlings were harvested after the treatments, and the freshly prepared buffer containing 5-bromo-4-chloro-3-indolyl-b-d-glucuronic acid was added to the test tubes. The GUS reactions were performed for 4 h at 37°C in the dark, and at least 20 seedlings for each treatment were analyzed under a stereo microscope; the representative images were photographed.

Microscopic analysis of LR initiation and development

The seedling roots were treated as described previously (Deak & Malamy, 2005), and then LR initiation numbers were analyzed under a compound microscope (Zeiss Axioskop). To mark LRP in growing seedlings, plants were visualized using a Zeiss Axioskop dissecting microscope. Marks were scratched on to the back of the Petri dish to identify the location of the primordia.

Determination of PIN1 and PIN2 distribution

For visualization of yellow and green fluorescent protein (YFP and GFP) from jellyfish Aequorea victoria, 10-d-old transgenic seedlings expressing PIN1:YFP and PIN2:GFP were harvested from MS with or without 30 mM NaCl medium. The excised roots were mounted immediately and examined with a Zeiss LSM 510 confocal laser scanning microscope (Carl Zeiss MicroImaging) with a 488 nm excitation line and a 530 nm emission filter. All images were taken under the same conditions.

Statistical analysis

Data were analyzed by ANOVA using Excel. The significance of differences was determined by t-test. A P-value < 0.05 was considered statistically significant.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

The ionic effect of Na+ alleviates the LR growth inhibition caused by osmotic stress

It was previously reported that salt stress remodels RSA (Wang et al., 2009; Zolla et al., 2010). Because salt imposes both osmotic and ionic stresses, we first analyzed the morphogenetic responses of roots in Arabidopsis WT plants treated with the iso-osmotic concentrations of NaCl and mannitol (Fig. 1a) to differentiate the effects of ionic stress on RSA from equivalent osmotic stress. The seeds were germinated on the stress media; the growth of PR and development of LR were assessed 10 d after germination. We found that at external osmotic potentials higher than −0.7 MPa, PR length was not significantly different for both NaCl and mannitol compared with those of the unstressed controls (Fig. 1b). Lower external water potentials (< −0.7 MPa) substantially reduced PR growth of the stressed plants treated by NaCl and mannitol, and no significant difference in PR growth was observed between the equivalent osmotic pressures generated by NaCl and mannitol. By contrast, LR development displayed great sensitivity to both salt and mannitol stresses (Fig. 1b). Numbers of LRs were markedly decreased with increasing concentrations of mannitol. At an external osmotic potential of −0.25 MPa, the number of LRs on mannitol-treated seedlings was reduced by c. 50%, and LR development was completely inhibited at the −0.85 MPa generated by the mannitol treatment. However, the effect of low osmotic potential imposed by NaCl on Arabidopsis LR development was different from that caused by mannitol. As shown in Fig. 1(b), LR development was not markedly affected at the −0.15 MPa osmotic potential caused by NaCl compared with that at the control value (−0.05 MPa). At osmotic potentials < −0.15 MPa, the number of LR in salt-treated plants was also substantially reduced, but to a much lesser extent than those grown at the equivalent osmotic potentials arising from mannitol. At the osmotic potential of −0.45 MPa, LR numbers of NaCl- and mannitol-stressed plants were reduced by c. 30 and 80%, respectively. Twenty per cent of LRs still formed on the plants treated with the iso-osmotic concentration (−0.85 MPa) of NaCl at which LR formation was completely inhibited when treated with mannitol. These results suggest that inhibition of LR development in the presence of NaCl and mannitol was caused by the osmotic pressure, whereas the differential response of LR development to the equivalent osmotic potentials simulated by NaCl and mannitol was the result of the ionic effects. It is obvious that the sodium ion plays an important role in alleviating the negative effect on LR development of osmotic stress, and PR growth is less sensitive to Na+ than LR development.

image

Figure 1.  Dose–response assessment of mannitol and NaCl at the equivalent osmotic potentials on root system architecture in Arabidopsis. The wild-type seedlings were germinated on the treated medium for 10 d. (a) Comparison of equivalent osmotic potentials induced by the isotonic iso-osmotic medium of NaCl and mannitol. (b) The effect of different osmotic potentials on the length of the primary root (PR) and the number of visible lateral roots (LRs). Closed squares, NaCl; open squares, mannitol. Osmotic potential = −0.05 represents Murashige and Skoog (MS) medium. The data presented are combined from three experimental replicates. Error bars represent the average of 30 or more plants ± SE.

Download figure to PowerPoint

LR development under low NaCl stress depends on the SOS signaling pathway

The SOS signaling pathway is a well-characterized pathway for mediating plant responses to ionic stress (Zhu, 2002; Chinnusamy et al., 2005). To examine whether SOS signaling modulates the ionic effects on root development, in particular LR development under salt stress, we analyzed the root responses of the sos mutants. Because the sos mutants are hypersensitive to salt stress, we treated the sos mutants with low concentrations of salt which did not inhibit PR growth. As shown in Fig. 2, the seedlings of WT (Col-0 gl1) and sos3-1 mutants produced a similar root system, including PR length and LR number at 10 d after germination under the nonstress medium (Fig. 2). However, in the presence of low concentrations of NaCl (30 mM, water potential −0.15 MPa), at which PRs of the WT were not affected, the LR number of sos3-1 was dramatically reduced (Fig. 2a,d). No visible LRs in the seedlings of sos3-1 were observed when they were grown on the medium, although their above-ground growth was similar to that of WT. The sos1-1 and sos2-1 mutant seedlings displayed a similarly increased sensitivity of LR development to sos3-1 under lower concentrations of NaCl, at which PR growth was not affected (data not shown). In contrast, the sos mutants did not display any differences in LR development in response to the same iso-osmotic concentration of mannitol (Fig. 2c). The results indicate that the ionic effect of NaCl on LR development is specifically regulated by the SOS signaling pathway. We hereafter studied the mechanism of LR development under mild salt stress using sos3-1, since the sos1-1, sos2-1 and sos3-1 showed similar LR phenotypes under the stress.

image

Figure 2.  Effect of low NaCl and mannitol stress on primary root (PR) and lateral root (LR) development in wild-type (WT) and sos3-1 seedlings. The seedlings germinated on control (Murashige and Skoog (MS)) or stress medium for 10 d. (a) Comparisons of root system architecture between the WT and sos3-1 seedlings on the MS medium, MS with 25 mM mannitol or 30 mM NaCl. The representative seedlings from each genotype and treatment are presented. (b) The effect of the low NaCl concentration on PR length in WT and sos3-1 plants. (c) The LR number of WT and sos3-1 plants under the low concentration of mannitol. (d) The LR number of WT and sos3-1 plants under the low-salt condition. Black bars, WT; shaded bars, sos3-1. Data are means of 20 or more plants ± SE, and are representative of similar results in three independent experiments. Different letters indicate a significant difference at P < 0.05.

Download figure to PowerPoint

SOS3 regulates both LR initiation and emergence under low salt stress

Lateral root development in general can be divided into two stages: an initiation phase and an emergence phase (Bhalerao et al., 2002). To examine which developmental stage of LR formation was defective in the sos3-1 mutant, the initiation and emergence of LRs in 10-d-old seedlings of the WT and sos3-1 grown on the medium containing a low concentration of NaCl (30 mM NaCl, water potential −0.15 MPa) were visualized under a microscope. The LRs that broke the epidermis were counted. Under mild NaCl stress, there were, on average, 10 LRs per root of the WT plant; however, no LRs per sos3-1 seedling root were observed (Fig. 3b). The number of LRP per WT and sos3-1 seedling under the stress conditions was also examined. We found that the number of LRP per sos3-1 salt-treated seedling was c. 50% of that of the WT, and the majority of the LRP of sos3-1 were arrested at various stages of LR development before LR emergence (Fig. 3a,b). These findings suggest that the sos3-1 mutant is defective in both LR initiation and LRP emergence under low-salt conditions.

image

Figure 3.  Effects of low salt on lateral root (LR) initiation and emergence. (a, b) The wild-type (WT) and sos3-1 plants were grown under 30 mM NaCl conditions for 10 d. (a) Lateral root primordia (LRP) development of sos3-1 and WT plants. The LRP of WT plants could penetrate through the epidermis layer while the LRP was arrested at the stage just before emergence in sos3-1 plants. Bar, 100 μm. (b) Comparison of the number of LR and LRP between WT (black bars) and sos3-1 plants (hatched bars); the LR represents the visible LR. Seven-day-old seedlings of WT and sos3-1 germinated on MS medium (MS-7d in c) were transferred on MS without (MS-7+MS-9d in c) or with 30 mM NaCl (MS-7+NaCl-9d in c) for 9 d, and then LR number were counted. Black bars (d), LR number of the 7-day-old seedlings; hatched and white bars (d), LR numbers of 7-day-old plants not treated or treated with 30 mM NaCl for another 9 d, respectively. Data presented are combined from three experimental replicates. Error bars represent averages of 10 or more seedlings ± SE. Different letters indicate a significant difference at P < 0.05.

Download figure to PowerPoint

To further verify the role of the SOS3 gene in LR emergence, 7-d-old seedlings of both the WT and the sos3-1 grown on MS were transferred to MS and MS containing the low concentration of NaCl, respectively. Because the LRP are already initiated in the roots of 7-d-old seedlings of both genotypes under normal conditions (Supporting Information, Fig. S1), which is consistent with previous observations (Deak & Malamy, 2005), the primordia development in the region of the root already formed at the time of transfer was assessed (Fig. 3c,d). When the seedlings germinated on the MS medium were transferred on to the MS medium for another 9 d, both the WT and sos3-1 seedlings displayed normal and similar LR growth, with about nine LRs formed. However, when they were transferred on to the medium containing the low concentration of NaCl, the WT seedlings had, on average, 10 visible LRs in the region with LRP after 9 d; in contrast, only two emerged LRs were observed in the same region per mutant root (Fig. 3c,d). Interestingly, the mutant seedlings produced more and longer adventitious roots at the root–shoot junction sites (Fig. 3c). These results suggest that the SOS3 gene is essential for LR emergence under low-salt stress conditions and it also negatively regulates adventitious root development induced by mild salt stress.

SOS3 facilitates LR emergence by promoting shoot-derived auxin production and acropetal transport under low salt stress

Lateral root emergence is dependent on auxin and auxin gradient. It has been proposed that auxin sources derived from leaves determine outgrowth of LRs (Bhalerao et al., 2002; De Smet et al., 2006). To test whether the defects in LR emergence of the sos3-1 mutant are the result of disruption of auxin maxima and auxin distribution pattern, GUS staining of DR5::GUS in the sos3-1 and the WT background was conducted. In order to determine the time point for shoot-derived auxin movement, we first performed an assay by removal of aerial tissues at different times after germination. We found that auxin synthesized by aerial tissues began to be transported to the root tip c. 5 d after germination to promote LR emergence (Fig. S1), which is consistent with the report by Bhalerao et al. (2002). We then checked the DR5::GUS activity in both WT and sos3-1 seedlings 5 d after germination. On the nonstress medium, no significant differences in the GUS activity and distribution pattern were detected between the WT and the mutant seedlings. DR5::GUS was mainly expressed at the tips of the cotyledons and the tips of PRs at the regions just below the quiescent center (QC) and provascular tissues (Fig. 4a). Surprisingly, DR5::GUS activity in the NaCl-stressed seedlings of different backgrounds exhibited very different patterns. GUS activity disappeared in the tips of cotyledons of both WT and sos3-1 plants. GUS staining was detected in the cotyledon margins of the WT seedlings, whereas GUS activity was hardly detectable in the cotyledon of the sos3-1 seedlings (Fig. 4a). In comparison, the DR5::GUS activity in the root apex was similar between the untreated and treated WT seedlings. However, no DR5::GUS staining occurred in the provascular tissues of the salt-treated sos3-1 root tips, and the GUS intensity in the region below the QC was increased (Fig. 4a). These observations suggest that loss of function in the SOS3 gene reduces auxin accumulation in aerial tissues, which may result in altered auxin polar transport to root tips and thus subsequent LRP emergence in response to mild salt stress.

image

Figure 4.  The role of auxin in lateral root (LR) development in wild-type (WT) and sos3-1 plants under low salt stress. (a) DR5::GUS-staining of cotyledon and primary root (PR) tip in 5-d-old WT and sos3-1 plants on Murashige and Skoog (MS) medium with or without 30 mM NaCl. Bar, 100 μm. (b) DR5::GUS-staining of the whole root in 10-d-old WT and sos3-1 plants on MS with or without 30 mM NaCl. (c) DR5::GUS-staining of the LR primordia (LRP) in 10-d-old WT and sos3-1 plants on MS with or without 30 mM NaCl in successive time points. Bar, 100 μm. (d) The effect of exogenous auxin on LR arrest of sos3-1 under MS supplemented with 30 mM NaCl. Ten-day-old sos3-1 seedlings grown on MS containing 30 mM NaCl (NaCl-10 d) were transferred to the same medium for 10 d (+NaCl-10 d) or the MS supplemented with 75 nM NAA for 10 d (+NaCl+NAA-10 d). (e) The LR number of sos3-1 was restored on MS plus 30 mM NaCl after imposing on exogenous NAA. Data presented are combined from three experimental replicates (total n > 60). Black bars, LR number of 10-day-old seedlings of sos3-1 grown on MS plus 30 mM NaCl; hatched bars, LR number of the seedlings after transfer to the same medium or medium with NAA for another 10 d. Error bars represent ± SE values. Different letters indicate a significant difference at P < 0.05. (f) DR5::GUS-staining of the LRP in sos3-1 seedlings on MS plus 30 mM NaCl with (NaCl+NAA) or without (NaCl) exogenous NAA in successive time points. Bar, 100 μm.

Download figure to PowerPoint

To further investigate the auxin pattern and gradient during LR emergence, the DR5::GUS activity of 10-d-old seedlings was analyzed. In the absence of NaCl, intensity and distribution pattern of DR5::GUS in sos3-1 was quite similar to the WT (Fig. 4b). In the presence of NaCl (30 mM, water potential −0.15 MPa), GUS staining was only detected in the tips of PRs and adventitious roots in the sos3-1 mutant seedlings, which is in sharp contrast to the auxin distribution pattern in the WT plants. When we observed LRP closely, the intensity of DR5::GUS in the LRP of the WT seedlings remained unchanged once LRP were activated. These LRP can form LR apical meristems to maintain the outgrowth (Fig. 4c). However, among the reduced number of activated LRP in the sos3-1 plants, DR5::GUS activity was substantially reduced compared with WT plants (Fig. 4c). Only a very tiny amount of DR5::GUS activity was detected in the outer layers and the cells behind the cap cells in LRP. The DR5::GUS expression faded and eventually disappeared in the LRP with prolonged NaCl treatment and left the newly formed LRP arrested at various stages before LR emergence (Figs 3b, 4b,c). We therefore speculate that LR emergence defects in sos3-1 mutants under low salt stress are mainly caused by insufficient amounts of auxin, which is essential for LRP development and LR apical meristem formation during LR development.

To test this hypothesis, we transferred the stressed 10-d-old mutant seedlings to the same stress medium or the stress medium supplemented with 75 nM NAA (De Smet et al., 2003; Deak & Malamy, 2005). After 10 d culture, the sos3-1 plants still displayed very few LRs and long adventitious roots on the stress medium without NAA; in sharp contrast, LR phenotypes of the sos3-1 seedlings on the stress medium with NAA were completely restored and the LRs developed rapidly (Fig. 4d,e). Microscopic analysis revealed that when the sos3-1 seedlings were grown on the low concentration of NaCl supplemented with NAA, the LR growth defect can be fully compensated for (Fig. 4d–f), and the DR5::GUS intensity and maximum were also restored by application of exogenous NAA (Fig. 4f). These results demonstrate that disruption of LR outgrowth in the sos3-1 mutant is caused by insufficient shoot-derived auxin for establishment and maintenance of LR meristem under low salt stress.

It has been shown previously that auxin polar transport from shoot to root by the auxin efflux facilitators (e.g. PIN1) mediates LR emergence (Benkováet al., 2003). To test whether PIN1 mediated auxin transport in the sos3-1 was also affected under the mild salt stress condition, the WT and sos3-1 plants expressing PIN1:YFP (Zhuang et al., 2006) were subjected to the stress response experiment. First, we inspected PR root tips of 10-d-old WT and mutant plants grown on medium with or without the low-salt concentration. As shown in Fig. 5(a), localization of PIN1 exhibited a similar pattern in the WT and the sos3-1 mutant in the absence of salt (Fig. 5a). In the presence of the low-salt concentration, the PIN1:YFP signal in the WT plants did not show much difference; however, very weak PIN1:YFP fluorescence in the root tips of the sos3-1 plants was detected, particularly in the region below the elongation zone. Next we analyzed PIN1:YFP during LR activation and emergence in both the WT and the mutant. We found that during LR development, PIN1:YFP signals were similar in the WT and mutant seedling roots under control conditions (Fig. 5b). When the plants were exposed to mild salt stress, PIN1:YFP intensity remained unchanged in the WT, whereas the PIN1:YFP signal in the mutant became very diffuse and so weak as to be hardly detectable (Fig. 5b). This result reveals that alteration of PIN1 activity may be involved in the defect in LR emergence of the sos3-1 plants in response to mild salt stress.

image

Figure 5.  SOS3 affects auxin transport and cell division activity of the lateral root primordia (LRP). (a) PIN1 intensity at the tips of the primary root (PR) in both wild-type (WT) and sos3-1 plants on Murashige and Skoog (MS) medium with or without a low concentration of NaCl. (b) PIN1 expression at the LRP in both WT and sos3-1 plants under MS and low-salt treatments. (c) PIN2 expression at the PR in both WT and sos3-1 plants on MS with or without a low concentration of NaCl. (d) The length of PIN2 expression domain in both WT (black bars) and sos3-1 (hatched bars) plants under MS and low-salt treatments. Data presented are combined from three experimental replicates (total n > 30). Error bars represent ± SE values. Different letters indicate a significant difference at P < 0.05. (e) The effect of auxin inhibitor 2,3,5-triiodobenzoic acid (TIBA ) on lateral root (LR) arrest of sos3-1 under MS medium. Ten-day-old sos3-1 seedlings on MS containing 30 mM NaCl (NaCl-10d) were transferred to MS with (+TIBA-5d) or without (+MS-5d) 2 × 10−6 M TIBA for 5 d. (f) WOX5::GUS staining of the LRP in 10-d-old WT and sos3-1 plants on MS with or without 30 mM NaCl in successive time points. Bar, 100 μm. (g) CYCB1;1::GUS staining of the LRP in 10-d-old WT and sos3-1 plants on MS with or without 30 mM NaCl in successive time points. Bar, 100 μm. The representative seedlings from each genotype and treatment are presented.

Download figure to PowerPoint

The SOS3 gene affects auxin basipetal transport and subsequent activation of LRP under low salt stress

Lateral root primordia initiation is dependent on auxin sourced from root tips through basipetal transport according to the phase model of LR development (Bhalerao & Bennett, 2003). Because LRP in the sos3-1 mutant seedlings were dramatically reduced under low salt stress (Fig. 3b), we then tested whether auxin basipetal transport from the root tip is also defective in sos3-1 mutants, resulting in failure to activate pericycle founder cells in LR initiation under low salt stress. It is well known that the auxin efflux carrier PIN2 acts in basipetal auxin transport in root tips. To check whether the expression and distribution of the PIN2 protein were altered under low-salt treatment in the sos3-1 roots, the sos3-1 mutant expressing PIN2:GFP was subjected to the treatment. Under normal conditions, the intensity and localization of PIN2:GFP in the sos3-1 root tips were similar to those in the WT (Fig. 5c). After exposure to the low concentration of NaCl (30 mM, −0.15 MPa), the expression level of PIN2:GFP in the mutant sos3-1 seedlings decreased, in particular in the elongation zone of the root tips (Fig. 5c,d). In comparison, the amount and localization of PIN2 protein in the WT root tips remained fairly stable regardless of whether or not they were stressed (Fig. 5c,d). These results indicate that disruption of LRP initiation in the sos3-1 seedlings under mild salt stress is likely caused by failure in auxin transport from root tip to the elongation zone and subsequent formation of auxin maxima for onset of LR initiation. To verify the results, we transferred the 10-d-old sos3-1 seedlings germinated on low-salt medium to MS medium in the presence or absence of auxin polar transport inhibitor TIBA (Deak & Malamy, 2005). As shown in Fig. 5(e), after 5 d culture, LR development was completely recovered in both the PR and the adventitious root when the sos3-1 seedlings were transferred on to the MS medium without auxin polar transport inhibitors. In contrast, the inhibitory effect of salt stress on LR development of the sos3-1 seedlings remained when TIBA was included in the MS medium. A LR phenotype similar to that induced by mild salt stress was seen with TIBA and NPA (N-1-naphthylphthalamic acid, another auxin transport inhibitor) when they were applied in shoots of sos3-1 (Fig. S2). The data confirm that the SOS3 gene and SOS signaling play a pivotal role in regulating auxin polar transport and LR initiation in Arabidopsis plants in response to low salt stress.

Local auxin gradients and maxima determine the onset of LR initiation (Benkováet al., 2003). Because about 50% of LRP are formed in the presence of the mild salt stress, we then investigated whether the early events during LRP initiation and development in the sos3-1 seedlings are abnormal. As we showed earlier, during LRP initiation, normal DR5::GUS activity at the mutant pericycle cells was detected (Fig. 4c). Among the very early events during LRP initiation, DR5::GUS and the homeobox gene WOX5 required for stem cell maintenance are simultaneously expressed (Sarkar et al., 2007; Ding & Friml, 2010). To further investigate the effect of the SOS3 mutation on LR initiation and development, the sos3-1 mutant expressing WOX5::GUS was analyzed under normal and stressed conditions. As shown in Fig. 5(f), WOX5 was expressed at the site of LRP initiation in both stressed and unstressed mutant plants, which is similar to the expression pattern of WOX5::GUS in the WT (Fig. 5f). This result suggests that the initiation process of the LRP in the sos3-1 mutant is not affected. When analyzing WOX5::GUS in the LRP at various stages before emergence, similar expression patterns of WOX5::GUS were also observed in the WT and mutant plant roots under both control and stress conditions. The activity and localization of WOX5::GUS remained unchanged in the arrested LRP of the stressed mutant as compared with those LRP of the WT at equivalent developmental stages (Fig. 5f). This is in sharp contrast to DR5::GUS, expression of which was markedly reduced after LRP initiation took place in the stressed mutant (Fig. 4c). In the very few emerged LRs of the mutant, the expression pattern of WOX5::GUS remained similar at later stages of LR development. WOX5::GUS was expressed in the QC in both WT and mutant seedlings (data not shown). The observation revealed that the stem cell fate and QC formation of the LR in the sos3-1 mutant is not disturbed in response to mild salt stress.

Periclinal and anticlinal divisions of LRP cells have key roles in LRP activation and development (Malamy & Benfey, 1997). To test whether arrest of the mutant LRP in the presence of mild salt stress is caused by abnormal mitotic activity of LRP cells, the mutant seedlings expressing CYCB1;1::GUS were characterized. As shown in Fig. 5(g), under normal conditions, the mutant LRP exhibited normal mitotic activity as visualized by CYCB1;1::GUS staining (Fig. 5g). At the early developmental stages of LRP, the mutant and WT showed similar patterns of CYCB1;1::GUS activity, although the GUS blue spots were not uniformly distributed. CYCB1;1::GUS activity became stronger and was mainly localized in the central region of the LRP at the stage before emergence. Under low salt stress, the mutant exhibited normal CYCB1;1::GUS activity during LRP initiation and at the early stages; however, at the later stages, the CYCB1;1::GUS activity reduced and eventually became hardly detectable, especially in the LRP before emergence (Fig. 5g). These findings suggest that the SOS3 gene is required for sufficient auxin supply for LRP activation and maintenance of cell division activity of stem cells and their daughter cells in LRP once the auxin signal is perceived by the pericycle cells under mild salt stress.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

It is well known that Na+ is the predominant ion in salt stress which is detrimental to plant cell metabolisms and the subsequent root and shoot growth when ion accumulation reaches toxic amounts (Russell, 1917; Munns, 1993; Neumann, 1997). In addition to this ion effect, osmotic stress caused by salt stress also negatively affects plant growth and development (Mart nez-Ballesta et al., 2004; Wahid, 2004). In this study, we found that PR elongation of Arabidopsis plants was not sensitive to low and moderate osmotic stress or the iso-osmotic salt solutions. In contrast, LR development was very sensitive to osmotic stress, and surprisingly the degree of LR development inhibition was markedly alleviated under the iso-osmotic salt solutions. These results provide evidence not only that PR and LR development have distinct stress sensitivities, but also that osmotic and ionic stress have contrasting effects on LR development (Fig. 1). Alleviation of LR inhibition by NaCl may be the result of more water uptake caused by influx ions, which changes water potentials of the plants as well as osmotic regulation activity. We found that 30 mM NaCl (water potential −0.15 MPa) showed no remarkable effect on LR development and that higher concentrations of NaCl inhibited LR initiation and emergence. These results are different from the recent report showing that low concentrations of NaCl promote LR proliferation (Zolla et al., 2010). The different results obtained were mainly due to the different experimental designs. Zolla et al. (2010) germinated their seeds on vertical MS medium and then transferred the 4- to 5-d-old seedlings to the stress medium and kept them vertically for another 21 d. At the time they transferred the seedlings, the LRP had begun to form on the control media. In our experiments however, seeds were germinated directly on the control or various stress media. The different results suggest that the effects of NaCl on LR development are dependent on the developmental stage. It is conceivable that plants may activate different molecular and physiological mechanisms at various developmental stages to adapt to salt stress. When subjected to salt stress, plants can also instantly reprogram their growth at organ levels by largely unknown mechanisms to cope with salt stress. This notion is supported by recent results showing that the different cell types of Arabidopsis roots or leaves of different ages exhibited different transcriptional responses to high salinity and mild osmotic stress (Dinneny et al., 2008; Skirycz et al., 2010).

The ion homeostasis of plant cells and the plant responses specific to ion toxicity are regulated by the SOS genes and the corresponding signaling pathway (Hasegawa et al., 2000; Zhu, 2003). The extensive phenotypic analyses of the sos mutants in response to salt stress have been carried out almost exclusively using 4- or 6-d-old seedlings before LR emergence under moderate or high concentrations of salt (Liu & Zhu, 1998; Zhu et al., 1998). Therefore, their roles in seedling growth and primary root elongation under moderate and high concentrations of NaCl stress are well documented; however, whether the SOS genes modulate LR development under low salt concentrations is still not known. Our data demonstrate that the SOS3 gene and SOS signaling play a key role in LR initiation and are essential for LR emergence in response to low salt stress. sos mutants plants exhibited strong LR initiation and progression defects with complete inhibition of LR emergence under a degree of salt stress which does not affect PR growth for both WT and sos mutants (Fig. 2). This hypothesis is supported by the finding that the transgenic lines overexpressing SOS1, SOS2 and SOS3 genes conferred enhanced LR number under salt stress, whereas they showed similar LR growth to the WT plants under normal conditions (Yang et al., 2008). Thus, we identified the SOS3 gene and the SOS pathway as a new positive regulator that specifically regulates the plastic development of LR in Arabidopsis under low salt stress. We also revealed the SOS pathway as a negative regulator in development of adventitious roots; the molecular mechanism underlying adventitious root regulation in response to salt stress is under investigation.

Auxin has been identified as the most important intrinsic signal that controls organ formation, including LR and leaf primordia (Benkováet al., 2003; Friml et al., 2003; Reinhardt et al., 2003). In the case of LR development, auxin gradient and maxima in the root apex regulate all developmental stages (Casimiro et al., 2001; Fukaki et al., 2007; Osmont et al., 2007; Péret et al., 2009b). Shoot-derived auxin entering the root is mainly transported through phloem and polar transport (Ljung et al., 2005), and this auxin, together with apically produced auxin, is redistributed basipetally to activate LRP initiation and emergence (Swarup & Bennett, 2003). Auxin derived from shoots is essential for LR emergence, and basipetal transport of auxin in roots is a key signal triggering the initiation of LRP in the pericycle tissue (Benkováet al., 2003). LR initiation and emergence are differentially and coordinately regulated (Péret et al., 2009a). However, the molecular mechanism of regulation of LR development specifically in response to ionic stress remains elusive. Given the known fact that salt stress affects auxin distribution patterns in WT roots (Wang et al., 2009), we hypothesized that the SOS pathway may modulate the reprogramming of auxin signals and redistribution, resulting in altered LR initiation, progression and emergence. The evidence obtained in this study supports this hypothesis. Our data revealed that mild salt stress induces ectopic auxin accumulation in the margin of cotyledons in germinating WT seedlings of Arabidopsis, which may be required for maintenance of normal development of LR (Fig. 4a). The SOS3 gene is essential for this change in auxin distribution pattern because no auxin was detected in a loss of function mutant of the SOS3 gene, as visualized by DR5::GUS activity. Deficient auxin production in aerial tissues of the mutant under mild salt stress may be the primary reason for defective LR emergence, because exogenous addition of NAA in the stress medium completely restored auxin maxima in the LRP and the subsequent LRP progression and emergence, whereas application of TIBA and NPA in shoots of sos3-1 blocked LR development (Figs 4d–f, S2). The fact that PIN1 activity is substantially reduced in the stressed mutant root suggests that reduced acropetal auxin transport from shoot to root tip also contributes to developmental arrest of LRP progression and outgrowth (Fig. 5a,b). Therefore, the SOS3 gene controls LR emergence through increasing auxin production in shoots of stressed plants and subsequent acropetal auxin transport from shoot apex to root apex, as well as basipetal auxin transport to root tips.

In our hypothesis, the SOS signaling pathway is also required for LR initiation of Arabidopsis plants in response to mild salt stress. Recent work has shown that auxin acts as the local instructive signal to trigger LR organogenesis, and its local accumulation in root pericycle cells is necessary and sufficient to specify the founder cell identity (Dubrovsky et al., 2008). Under mild salt stress, the SOS3 gene participates in maintenance of local auxin maxima through increasing shoot derived auxin (Figs 4a, 5a) and basipetal auxin transport activity (Fig. 5c,d). The evidence that loss of function in SOS3 resulted in a sharp reduction in DR5::GUS spots along the primary roots (Fig. 4b) and PIN2 expression especially at the root elongation zone in sos3-1 under low-salt conditions supports this notion (Fig. 5c,d). The observation that reduced auxin transport resembled the effect of low salt stress on LR initiation of the sos3-1 mutant (Fig. 5e) further demonstrates the important role of the SOS3 gene in maintenance of basipetal auxin transport necessary for local auxin accumulation to trigger LR initiation. But we do not exclude the possibility that the SOS3 gene also plays a role in auxin synthesis in the root apex of stressed plants.

Our findings revealed that SOS3 does not affect the very early events of LR development once identity of the founder cells is determined. When exposed to mild salt stress, the pericycle cells of the sos3-1 mutant seedlings, like that of the WT, can generate a double layer of pericycle-derived cells by periclinal division and primordia cells, which begin to differentiate immediately after initiation, as evidenced by DR5::GUS and WOX::GUS expression at the sites of the initiation and LRP (Figs 4c, 5f). The SOS3 gene also does not influence the cell specification of stem cells and QC, but it is involved in mitotic cell division of LRP of stressed plants (Fig. 5g). Loss of function mutation in the SOS3 gene caused cell division defects of the LRP of the stressed plants, resulting in growth arrest before LR emergence. Such cell division failure is likely due to defect in auxin transport and insufficient maxima in mutant root tip under mild stress condition (Fig. 4c, Fig. 5a–c). The findings also suggest that cell cycle regulation could possibly play a role in the interplay between salt and LR development.

Salt stress is a key environmental signal affecting plant growth and development. As sessile organisms, plants have evolved the ability to accurately monitor the ion content of their ambient environment and adapt their developmental program in response to environmental signals. We have previously reported that, in response to salt stress, Arabidopsis plants adjust their developmental programs, for example delaying flowering time and altering RSA (Li et al., 2007; Wang et al., 2008). In this study, we report for the first time that Arabidopsis plants can perceive small changes of salt in their growth environment and can accurately reprogram development of PR and LR through a highly regulated network in which the auxin and SOS signaling pathway are integrated possibly via the SOS3 gene. Cross-talk between auxin and salt stress signaling can remodel LR developmental processes through modulation of spatiotemporal accumulation and distribution of auxin. Our results provide a novel insight into how LRs are regulated in response to salt stress. Further research into the mechanisms of ion perception, root-to-shoot communication and physiological responses of plants will enable us to further understand how plants respond to various degrees of salt stress and develop strategies to improve salt tolerance of crops.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank Drs J-K Zhu, J. M. Alonso and K. G. Raghothama for providing the sos mutants, and the DR5::GUS and CycB1;1::GUS seeds, respectively. We are grateful to Drs B. Scheres and T. Laux for kindly providing us with the lines expressing PIN1:YFP, PIN2:GFP, and WOX5::GUS, respectively. This work was supported by grants from the Ministry of Agriculture of the People’s Republic of China (2008ZX08002-002), the Ministry of Sciences and Technology of China (2009CB118305), and the State Key Laboratory of Plant Cell and Chromosome Engineering (PCCE-2008-TD-02).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Fig. S1DR5::GUS staining of lateral root primordia with above ground removal treatments.

Fig. S2 Effects of TIBA and NPA on LR development of sos3-1.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

FilenameFormatSizeDescription
NPH_3545_sm_FigS1-S2.doc3245KSupporting info item