Strigolactones (SLs) are plant hormones and regulators of root development, including lateral root (LR) formation, root hair (RH) elongation and meristem cell number, in a MORE AXILLARY GROWTH 2 (MAX2)-dependent way. However, whether SL signaling is acting cell-autonomously or in a non-cell-autonomous way in roots is unclear.
We analyzed root phenotype, hormonal responses and gene expression in multiple lines of Arabidopsis thaliana max2-1 mutants expressing MAX2 under various tissue-specific promoters and shy2 mutants.
The results demonstrate for the first time that expression of MAX2 under the SCARECROW (SCR) promoter, expressed mainly in the root endodermis, is sufficient to confer SL sensitivity in the root for RH, LR and meristem cell number. Moreover, loss of function mutation of SHORT HYPOCOTYL 2 (SHY2), a key component in auxin and cytokinin regulation of meristem size, has been found to be insensitive to SLs in relation to LR formation and meristem cell number.
Endodermal SL signaling, mediated by MAX2, is sufficient to confer SL sensitivity in root, and SHY2 may participate in SL signaling to regulate meristem size and LR formation. These SL signaling pathways thus may act through modulation of auxin flux in the root tip, and may indicate a root-specific, yet non-cell-autonomous regulatory mode of action.
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.
Strigolactones (SLs) are plant hormones that have been shown to act as long-distance branching inhibitors, suppressing the outgrowth of preformed axillary shoot buds (Gomez-Roldan et al., 2008; Umehara et al., 2008). Present in a wide variety of plant species, they are also involved in plant communication in the rhizosphere, and act as stimulants of parasitic plant (Striga and Orobanche) seed germination (Cook et al., 1966; reviewed by Xie et al., 2010) and as stimulants of hyphal branching of arbuscular mycorrhizal fungi (reviewed by Koltai et al., 2012).
SLs are derived from carotenoids (Matusova et al., 2005) and are biosynthesized via several steps, including those carried out by a cytochrome P450, two carotenoid cleavage dioxygenase (CCD) enzymes (CCD7/MORE AXILLARY GROWTH 3 (MAX3) and CCD8/MORE AXILLARY GROWTH 4 (MAX4); reviewed by e.g. Dun et al., 2009; Brewer et al., 2013) and DWARF27 (D27), a β-carotene isomerase (Lin et al., 2009; Alder et al., 2012). As for the signal transduction pathway of SLs, only a few components have been identified to date. One of them is MAX2 (Umehara et al., 2008), which encodes an F-box protein (Stirnberg et al., 2007). Another is DWARF14 (D14), an α/β hydrolase (Arite et al., 2009), which may interact with MAX2 in the presence of the synthetic SL analog GR24, to confer SL perception and response (Hamiaux et al., 2012).
SLs are synthesized in a few different plant parts, with roots being the main site of SL biosynthesis (reviewed by Xie et al., 2010; Liu et al., 2011; Proust et al., 2011; Delaux et al., 2012). The SL orobanchol can be found in the xylem sap of Arabidopsis thaliana (Kohlen et al., 2011), suggesting that root-derived SLs are transported in the root-to-shoot direction to confer inhibition of axillary bud outgrowth in the shoots (reviewed by Dun et al., 2009). Furthermore, the ATP-binding cassette (ABC) transporter PDR1 in petunia (Petunia hybrida) was suggested to function as a cellular SL exporter facilitating delivery of SLs to their site of action (Kretzschmar et al., 2012). It seems that SLs are transported in the plant and may thus act at a distance.
However, whether SL signaling is acting cell-autonomously or in a non-cell-autonomous way is unclear. A possible answer to this question, relating to the effect of SL activities on shoot secondary growth and adventitious root formation, came from two studies that used lines expressing MAX2 under different tissue-specific promoters in max2-1 mutants (Agusti et al., 2011; Rasmussen et al., 2012). These max2-1 lines express MAX2 under the control of the WUSCHEL-RELATED HOMEOBOX4 (WOX4), SCARECROW (SCR), ALTERED PHLOEM DEVELOPMENT (APL) or NAC SECONDARY WALL THICKENING PROMOTING FACTOR3 (NST3) promoter (Agusti et al., 2011). WOX4 is expressed mainly in the (pro)cambium (Hirakawa et al., 2010); SCR is expressed in the root endodermis and the quiescence center, and in the starch sheath of the shoot, the shoot apical meristem and the leaves (Di Laurenzio et al., 1996; Wysocka-Diller et al., 2000; Sabatini et al., 2003). APL is expressed mainly in the shoot and root phloem (Bonke et al., 2003; Birnbaum & Benfey, 2004). NST3 is expressed mainly in the xylem (Mitsuda et al., 2007). One study examined secondary growth associated with SL activity. It was found that only lines expressing MAX2 in the vascular cambium of max2-1 mutants displayed secondary growth at wild-type (WT)-like levels, suggesting that, in this case, MAX2-dependent SL regulation occurs in a cell-autonomous fashion (Agusti et al., 2011).
In another case concerning the effect of SL activity on adventitious root formation, a non-cell-autonomous mode of action was proposed for MAX2. In this case, expression of MAX2 under a xylem-specific promoter led to the restoration of SL repression of adventitious root formation. As adventitious roots of A. thaliana hypocotyls develop from pericycle cells that are aligned with the xylem poles, it was suggested that the SL signaling in the xylem acts non-cell-autonomously, by short-range signaling (Rasmussen et al., 2012).
Previously, we found that root development of seedlings is regulated by SL activity. In a MAX2-dependent fashion, SLs are positive regulators of root hair (RH) length and negative regulators of lateral root (LR) formation (Kapulnik et al., 2011a). Also, SLs were shown to be positive regulators of meristem cell number in the primary root meristem (Ruyter-Spira et al., 2011). As in seedlings MAX2 is expressed ubiquitously (Shen et al., 2007), SL activity regulating meristem size, LR formation and RH elongation may be either cell-autonomous or non-cell-autonomous.
To gain a better insight into SL signal transduction in roots, we determined the spatial requirements of MAX2 expression for root response to SLs by analyzing root phenotype and hormonal response of multiple lines expressing MAX2 under various tissue-specific promoters. The results suggested that expression of MAX2 under the SCR promoter, expressed mainly in the root endodermis, is sufficient to confer GR24 sensitivity for RH elongation, LR formation and meristem cell number. Moreover, loss-of-function mutation of SHY2, a key component of hormonal regulation of mersitem size, was found to be insensitive to GR24 in relation to LR formation and meristem cell number. Together, these results demonstrate that endodermal SL signaling, mediated by MAX2, is sufficient to confer SL sensitivity in roots and that SHY2 might participate in SL signaling. These SL signaling pathways may act through changes in PIN-FORMED (PIN) level, and may indicate a local yet non-cell-autonomous regulatory mode of action.
Materials and Methods
Arabidopsis thaliana strains and growth conditions
Seeds of Arabidopsis thaliana (L.) Heynh. used in this study included WT (Columbia (Col-0) and Landsberg erecta (Ler)), the Col-0 homozygous line max2-1 (obtained from the Arabidopsis Biological Resource Center (ABRC) stock center http://abrc.osu.edu/), max2-1 lines expressing MAX2 under the control of the WOX4 (two separate lines designated herein 89-3 and 92-2), SCR (two separate lines designated herein 40-5 and 90-1), APL (two separate lines designated herein 37-3 and 37-1), and NST3 (two separate lines designated herein 73-2 and 39-2) promoters (Agusti et al., 2011), and Ler homozygous lines shy2-31 and shy2-2 (SHY2 loss and gain of function, respectively; kindly provided by Dr S. Sabatini, University of Rome, Rome, Italy). Seeds were surface-sterilized and germinated on 1/2 Murashige and Skoog (MS) plates supplemented with 1.5% sucrose and solidified with 1% agar. Plates contained various concentrations of hormones: GR24 as a mixture of four diastereomers ((±)-GR24 and (±)-2′-epi-GR24; Johnson et al., 1981), kindly provided by Dr Koichi Yoneyama (Utsunomiya University, Utsunomiya, Japan), and benzyl amino purine (BAP; Sigma). GR24 treatments were conducted at a concentration of 3 × 10−6 M. GR24 was initially dissolved in acetone (5 mg GR24 per 3 ml of acetone) to give a 4.5 mM solution; this solution was then further diluted with double-distilled sterile water. Hence, in addition to nontreated roots, the experimental control included roots treated with acetone at the concentrations used in the respective GR24 treatments. BAP treatments were conducted at a concentration of 5 × 10−8 M dissolved in water. In each of the experiments, hormone-treated roots were compared with the respective acetone control. No significant differences were found between acetone and water controls (results not shown).
Plates were incubated vertically in the dark at 4°C for 2 d to synchronize germination. Plates remained unsealed to prevent accumulation of gases (e.g. ethylene), and were positioned in an upright 45° position, and incubated at 22°C with a light intensity of 50–60 mol photons m−2 s−1 provided by white fluorescent tubes and under a photoperiod of 16 h light : 8 h dark for 6–12 d.
Determination of lateral root density
To determine primary root length and number of LRs, seedlings were grown on GR24 and/or BAP and control plates as described in the previous section. The length of the primary root and the number of LRs were recorded at 8 d post germination (dpg). LRs were counted under a stereoscope (Leica MZFLIII; Leica Microsystems GmbH, Wetzlar, Germany) from the stage of emergence (Malamy & Benfey, 1997) and primary root length was measured from pictures of the roots at 8 dpg, with a Nikon DS-Fi1 camera, for 10 separate roots per treatment. Experiments were repeated four times; each treatment within each experiment included three replicates, with 10 germinated seedlings per replicate (n =30). Means of replicates were subjected to statistical analysis by one-way analysis of variance (ANOVA) with Tukey–Kramer multiple comparison test (P <0.01) and ANOVA pair-wise Student's t-tests (P ≤0.05) using the jmp (SAS Institute Inc., Cary, NC, USA) statistical package.
Determination of root-hair length
For examination of RH length, roots were grown on GR24 and control plates as described in the section ‘Arabidopsis thaliana strains and growth conditions’. The root tip location on the plate was marked at 48 h post germination (hpg). At 72 hpg, roots were examined on the plates using a stereomicroscope (Leica MZFLIII; Leica Microsystems GmbH) and pictures were taken of root segments that had grown on the plates for 48 h with a Nikon DS-Fi1 camera. Experiments were repeated four times; each treatment within each experiment included three replicates, with 10 germinated seedlings per replicate. Measurements of RH length were taken from 10 pictures per treatment; between 15 and 20 different RHs were measured per picture, using imagej (http://rsbweb.nih.gov/ij/; n =200). Means of replicates were subjected to statistical analysis by one-way analysis of variance (ANOVA) with Tukey–Kramer multiple comparison test (P <0.01) and ANOVA pair-wise Student's t-tests (P ≤0.05) using the jmp statistical package.
Determination of meristem cell number
Seedlings of the different genotypes, treated as described in the section ‘Arabidopsis thaliana strains and growth conditions’, were taken at 72 hpg for determination of root meristem size, which is expressed as the number of cortex cells in a file of the proximal meristem extending from the quiescent center to the first elongated cortex cell, as described in Dello Ioio et al. (2008). To visualize the cell files, seedlings were stained with a solution containing propidium iodide (20 μg ml−1; Sigma), aniline blue (5 mg ml−1 in 0.1 M K3PO4; Sigma) and calcofluor-white satin (1.5 ml; FLUKA; Sigma-Aldrich). Roots of stained seedlings were then examined and pictures were taken using Olympus IX 81 (Olympus, Tokyo, Japan) inverted laser scanning confocal microscope supplemented with FLUOVIEW 500 software (Olympus) and equipped with a 405-nm diod laser and 40X0.9 NA PlanApo with 405 nm light; emissions were collected through a BA 430-460 filter. For each experiment, a minimum of 50 plants were analyzed. Means of replicates were subjected to statistical analysis by one-way analysis of variance (ANOVA) with Tukey–Kramer multiple comparison test (P < 0.01) and ANOVA pair-wise Student's t-tests (P ≤0.05) using the jmp statistical package.
Determination of shoot branching
WT and SCR::MAX2 max2-1 lines were grown in pots and subject to a long-day photoperiod (16 h light : 8 h dark). Plants of 3–4 wk old had approximately six to eight mature leaves and a primary bolt of 5–10 cm. Lateral shoot development at consecutive node positions was determined by dissecting leaves and associated axillary shoots from the shoot axis, attaching them to black paper sheets via double-sided sticky tape and laying them out in the order of emergence. The number of rosette leaves and outgrown buds was then determined. The branching of 40 plants was determined per line (n =40). Means of replicates were subjected to statistical analysis by one-way analysis of variance (ANOVA) with Tukey–Kramer multiple comparison test (P <0.01) using the jmp statistical package.
RNA extraction and preparation of cDNA
For RNA extraction from shoots, 2–5 mm of hypocotyl and four initial rosette leaves grown on 1/2 MS plates solidified with 1% agar and placed horizontally were taken and snap-frozen in liquid nitrogen. Each replicate consisted of 20 plants. Total RNA was extracted using TRI reagent (MRC, Cincinnati, OH, USA) and digested with Turbo DNase enzyme (Ambion Life Technologies, Grand Island, NY, USA) as per the manufacturer's instructions. Complimentary-DNA reverse transcription was performed as described by Koltai et al. (2010b).
Quantitative real-time PCR
Quantitative real-time (qPCR) was performed on cDNA reverse-transcribed from RNA extracted from rosette leaves and hypocotyl, as described in the previous section. Arabidopsis thaliana 15S ribosomal RNA (GenBank accession no. AT1G04270.1) served as the reference gene for the amount of RNA, and was amplified using specific primers (forward) 5′-CAAAGGAGTTGATCTCGATGCTCTT-3′ and (reverse) 5′-GCCTCCCTTTTCGCTTTCC-3′. To determine MAX3 (GenBank accession no. AT2G44990) and MAX4 (GenBank accession no. AT4G32810) gene expression, the following primers were used: (forward) 5′-GATTGGGCATTCACGGATAC-3′ and (reverse) 5′-CGTTGCTTGGGTTTAACGAT -3′ for MAX3 and (forward) 5′-GAAAAGCGACGGTCATCATT-3′ and (reverse) 5′-ATCCTGAATCTCCCGATCCT-3′ for MAX4. qPCR amplification was performed as described by Mayzlish-Gati et al. (2010), using the method (Livak & Schmittgen, 2001). The experiment was performed in five biological replicates, for which three technical repeats were performed. Means and standard errors were calculated for all biological replicates. Values above or below 1 represent an increase or decrease, respectively, in the steady-state level of gene transcripts for the examined conditions (i.e. that of the nominator vs that of the denominator). Means ± SD were calculated for at least three biological replicates for each examined treatment. Means of replicates were subjected to statistical analysis by ANOVA pair-wise Student's t-tests (P ≤0.05) using the jmp statistical package.
Expression of MAX2 under the SCR promoter in max2-1 is sufficient to confer the RH elongation response to GR24
RH elongation is positively affected by GR24 in a MAX2-dependent manner (Kapulnik et al., 2011a). Hence, we sought to examine where SL signaling is important to confer SL sensitivity in max2-1 with regard to RH elongation. max2-1 lines expressing MAX2 under the control of the WOX4 (89-3 and 92-2; Hirakawa et al., 2010), SCR (40-5 and 90-1; Wysocka-Diller et al., 2000), APL (37-3 and 37-1; Bonke et al., 2003), and NST3 (73-2 and 39-2; Mitsuda et al., 2007) promoters (two lines for each specific promoter; Agusti et al., 2011) were examined for GR24 sensitivity in terms of RH elongation. RH length was determined at 48 hpg in seedling roots grown under GR24 treatments. As expected, WT and max2-1 were not significantly different in terms of RH length (Fig. 1; Kapulnik et al., 2011a). Also, WT responded to GR24 with increased RH length, whereas max2-1 was insensitive to GR24 in this regard (Fig. 1; Kapulnik et al., 2011a). Expression of MAX2 solely under the SCR promoter, in both examined max2-1 lines (40-5 and 90-1), restored the ability of max2-1 to respond to GR24, to a level similar to that of WT (Fig. 1). Expression of MAX2 under the APL or NST3 promoter did not lead to an RH elongation response to GR24 in max2-1 mutants (Fig. 1). Only one max2-1 line that expressed MAX2 under the WOX4 promoter (89-3), out of the two lines examined, responded to GR24 in terms of RH elongation (Fig. 1).
Expression of MAX2 under the SCR promoter in the max2-1 mutant is sufficient to confer the LR formation response to GR24
SLs, under suffice phosphate conditions, are negative regulators of LR development in a MAX2-signaling-dependent fashion (Kapulnik et al., 2011a; Ruyter-Spira et al., 2011). Hence, we sought to examine where SL signaling is important, so as to induce in max2-1 a WT phenotype and SL sensitivity with regard to LR formation. LR density was determined in seedlings of WT, max2-1 and the different promoter–MAX2-expressing lines treated with GR24 and control. As previously reported, WT seedlings had a reduced density of LR formation in comparison to max2-1 (Kapulnik et al., 2011a). Among the promoter–MAX2-expressing max2-1 lines, only one line expressing MAX2 under the NST3 promoter (39-2) and one line expressing MAX2 under the APL promoter (37-1) had a significantly reduced density of LR in comparison to max2-1 (Fig. 2).
GR24 application led, in WT, but not in max2-1, to a reduction in LR formation (Fig. 2; Kapulnik et al., 2011a). Expression of MAX2 under the SCR promoter was sufficient to confer sensitivity to SLs in both examined max2-1 lines (40-5 and 90-1): GR24 application in these lines led to significantly reduced LR densities, as in WT (Fig. 2).
Expression of MAX2 under the SCR promoter represses shoot branching to some extent
SLs regulate shoot development by repressing lateral bud outgrowth (Gomez-Roldan et al., 2008; Umehara et al., 2008). SCR is expressed in the starch sheath of the shoot, the shoot apical meristem and leaves (Wysocka-Diller et al., 2000). Hence, we investigated whether expression of MAX2 under the SCR promoter, which led to GR24 sensitivity in terms of RH and LR, would also lead to inhibition of shoot branching in the SCR::MAX2 max2-1 lines. For that purpose we determined the number of branched buds versus the number of rosette leaves in the SCR::MAX2 max2-1 lines. In both lines, shoot branching was inhibited to some extent in comparison to max2-1; in line 40-5 this inhibition was significant. However, in both the SCR::MAX2 max2-1 lines branching was not repressed to the WT level (Fig. 3). These results suggest that, although SCR::MAX2 expression leads to RH and LR sensitivity to GR24, and although SCR is expressed in the shoot, expression of SCR::MAX2 is not sufficient to repress shoot branching in max2-1 to WT levels.
Feedback regulation is not restored in SCR::MAX2 lines
SL levels are regulated by SLs themselves. It was shown in several plant species that SL mutants show higher levels of SL biosynthesis gene expression (Hayward et al., 2009; Umehara et al., 2010). Hence, we investigated whether expression of MAX2 under the SCR promoter in max2-1, which led to GR24 sensitivity in terms of RH length and LR density, would also lead to inhibition of the expression of MAX3 and MAX4, two of the A. thaliana SL biosynthesis genes. As expected, levels of MAX3 and MAX4 gene expression were higher in max2-1 than in WT (Hayward et al., 2009; Umehara et al., 2010; Fig. 4). In the SCR::MAX2 max2-1 lines, similarly to max2-1, both MAX3 and MAX4 gene expression levels were significantly higher than in WT (Fig. 4), suggesting that SCR::MAX2 expression in max2-1 does not lead to a significant reduction of SL biosynthesis gene expression in the shoot.
Expression of MAX2 under the SCR promoter is sufficient to increase meristem cell number in max2-1 mutants
SLs are positive regulators of meristem cell number in a MAX2-signaling-dependent manner (Ruyter-Spira et al., 2011). Hence, we investigated whether expression of MAX2 under the SCR promoter in max2-1, which led to GR24 sensitivity in terms of RH and LR, would also lead to a WT phenotype and/or SL sensitivity with regard to meristem cell number. Meristem cell number was determined in seedlings of WT, max2-1 and the different promoter–MAX2-expressing lines treated with GR24 and control. Under the examined conditions (of 1.5% sucrose), no significant differences were found between WT and max2-1 for meristem cell number (Fig. 5), in agreement with Ruyter-Spira et al. (2011).
GR24 application led in WT to an increase in meristem cell number (Fig. 5), in accordance with the findings of Ruyter-Spira et al. (2011). Also, under these conditions, GR24 application to max2-1 seedlings did not lead to changes in meristem cell number. Expression of MAX2 under the SCR promoter was sufficient to confer sensitivity to SLs in the max2-1 lines and to produce a significant increase in meristem cell number, with numbers being similar to those found in WT (Fig. 5).
SHY2 loss of function mutant is insensitive to GR24 regarding LR formation
SHY2 is the central switch that controls meristem size by regulation of PIN auxin transporters in the root tip, and is involved in LR development (Moubayidin et al., 2010; Goh et al., 2012). On the one hand, the results reported by Ruyter-Spira et al. (2011) suggest that SLs positively regulate meristem cell number. On the other hand, we found that, in max2-1 lines, expression of MAX2 under the promoter of SCR, which is expressed in the meristematic endodermis (Sabatini et al., 2003), was sufficient for restoration of GR24 sensitivity in terms of meristem size and LR development. Also, the endodermis was found to play a major role in regulation of LR initiation via modulation of auxin flux (Marhavy et al., 2012). Therefore, we sought to examine possible connections between SLs and SHY2. Thus, shy2-31, a loss-of-function mutant, and shy2-2, a gain-of-function mutant, were examined (Dello Ioio et al., 2008). RHs were excessively long in both mutant lines (Knox et al., 2003; not shown), preventing examination of the effect of GR24 on RH length.
When LR density was examined, shy2-31 was found to have reduced sensitivity to GR24, similarly to max2-8. In both shy2-31 and max2-8, application of GR24 did not lead to a significant reduction in LR density, whereas in both WT and shy2-2, GR24 significantly reduced LR density (Fig. 6).
The fact that loss of action of SHY2 leads to reduced sensitivity to SLs positions SHY2 as a possible component of SL signaling with regard to LR density. The fact that the gain-of-function shy2-2 mutant (which has lost its sensitivity to auxin-mediated protein degradation; Colon-Carmona et al., 2000) retained its GR24 sensitivity for LR density suggests that SL regulation of SHY2 is not through auxin-mediated SHY2 degradation.
Reduced sensitivity of max2, similarly to shy2-31, to cytokinin treatments for lateral root density
The loss-of-function mutant shy2-31 has reduced sensitivity to cytokinin (Dello Ioio et al., 2008), and cytokinins are negative regulators of LR formation (Laplaze et al., 2007). As both shy2-31 and max2-8 loss-of-function mutants are insensitive to GR24 for LR formation, we also examined max2-8 sensitivity to cytokinin (BAP) in terms of LR density. In WT (Ler) and the shy2-2 gain-of-function mutant, application of cytokinin led to a marked reduction in LR formation (results for BAP treatment are shown in Fig. 6), in agreement with other studies (Laplaze et al., 2007). However, both shy2-31 and max2-8 showed reduced sensitivity to BAP (Fig. 6), further suggesting that SHY2 loss of function and MAX2 loss of function share similar hormone sensitivities. Notably, in WT and shy2-2, the effect of BAP on LR density dominated that of GR24: application of BAP reduced LR density to the same extent, either with or without GR24 application.
Loss-of-function mutant shy2-31 is insensitive to GR24 for meristem cell number
To further examine the suggestion that SHY2 loss-of-function and MAX2 loss-of-function mutants share similar hormone sensitivities, we examined the effect of GR24 on meristem cell number in shy2-31 mutants. As expected, GR24 application led to an increase in meristem cell number in WT (Ler) and in shy2-2 gain-of-function mutants (Fig. 7). However, meristem cell number was not significantly increased by GR24 application in either max2-8 or shy2-31 mutants (Fig. 7). Notably, shy2-31 meristem cell number was significantly higher than that of max2-8 (Fig. 7), suggesting that either SLs differently regulate meristem cell number in these mutants or that these differences between the two lines reflect the activity of other factors (e.g. other plant hormones) affecting meristem size regulation.
These results further suggest that shy2-31 is similar to max2-8 in relation to SL sensitivity, and that SLs are regulators of meristem cell number at least partially in a SHY2- and/or MAX2-dependent way. The GR24 sensitivity of shy2-2 gain of function for meristem cell number further supports the suggestion that SL regulation of SHY2 is not through auxin-mediated SHY2 degradation.
SLs are regulators of different developmental processes in roots, depending on MAX2 activity. They regulate LR formation and RH elongation during early seedling development (Kapulnik et al., 2011a; Ruyter-Spira et al., 2011). Also, SLs are positive regulators of meristem cell number. However, the exact mechanism by which SLs and MAX2 exert this regulation of root development is still unclear. Here, we have examined the spatial requirements of MAX2 expression for root response to SLs. The results suggested that expression of MAX2 under the SCR promoter, which is expressed mainly in the root endodermis and quiescence center (Di Laurenzio et al., 1996; Sabatini et al., 2003), is sufficient to confer root GR24 sensitivity. Moreover, our results indicate that SHY2 may also be involved in SL signaling. Together, the results suggest that endodermal SL signaling, mediated by MAX2, is sufficient to confer root sensitivity to SLs, and that SHY2 may be another component of SL signaling, acting in the root tip.
The formation of LR involves generation of dynamic gradients of auxin; an auxin maximum is formed at the tip of the primordia, leading to LR formation. This gradient is formed as a result of asymmetrical localization of the auxin efflux carrier PIN1 (Benkova et al., 2003). In accordance with the involvement of SLs in LR formation, GR24 application led to a decrease in PIN1-GFP intensity in LR primordia, suggesting that PIN1 is involved in the SL-mediated reduction of LR development (Ruyter-Spira et al., 2011). Supporting the suggestion that SLs affect auxin flux in roots is the study of Koltai et al. (2010a), in which only 2,4-d (a synthetic auxin, i.e. not secreted by efflux carriers) led to reversion of the GR24 root effect, suggesting a functional interaction between SLs and auxin efflux in the root. Moreover, the endodermis was shown to play an active role in the regulation of auxin accumulation, being needed for LR founder cells to progress during the LR initiation phase. PIN3 was shown to have functional importance in the auxin reflux between overlying endodermal and pericycle founder cells. It was suggested that the endodermis has a regulatory function role in LR initiation following founder cell specification (Marhavy et al., 2012). The fact that endodermal SL signaling, mediated by MAX2, is sufficient to confer SL sensitivity for LR formation suggests that endodermal SL signaling may affect auxin flux in the root tip, and thereby LR formation.
RH tip elongation has been suggested to be a function of a hormonal balance in the epidermal cell layer. Ethylene directs auxin in the epidermal cell layer to promote RH elongation (Strader et al., 2010) by promoting auxin biosynthesis (Swarup et al., 2007) and/or auxin transport (Ruzicka et al., 2007). Previously we found that SLs positively affect RH elongation in young seedlings (Kapulnik et al., 2011a). Both auxin and ethylene were shown to be involved in the SL regulation of RH elongation (Kapulnik et al., 2011b). As the regulation of RH elongation takes place in the epidermis, the sufficiency of the endodermal expression of MAX2-dependent SL signaling indicates a non-cell-autonomous effect of SLs on RH elongation.
SLs were shown by Ruyter-Spira et al. (2011) and our study to regulate meristem cell number; GR24 treatments have a positive effect on meristem cell number, in a MAX2-dependent manner. Following seed germination, the root apical meristem grows as cell division overcomes differentiation. Final meristem size is reached when division and differentiation are balanced; this balance results from the interaction between cytokinin (promoting differentiation) and auxin (promoting division; Moubayidin et al., 2010 and references therein). Endodermal SL signaling, mediated by MAX2, was sufficient to confer SL sensitivity to meristem size. Furthermore, PIN auxin transport in the root tip was shown to be closely associated with determining meristem size (Dello Ioio et al., 2008). Hence, these results further support the suggested activity of the endodermal MAX2-dependent SL signaling effect on auxin flux in the root tip, and thereby on meristem size.
In the above instances, endodermal SL signaling was effective in restoring GR24 sensitivity in max2-1, but not LR density and meristem size phenotypes to WT levels. It may be that SCR::MAX2 expression is either spatially or temporally insufficient to confer a WT phenotype. For example, it may be necessary to express MAX2 at the early embryonic stages of development and/or in certain tissues to confer a WT phenotype.
The primary root meristem is regulated by means of a regulatory circuit in which cytokinin and auxin interplay in opposite ways to balance cell differentiation with cell division so as to determine root meristem size. SHY2 is the central mediator in this interplay (Dello Ioio et al., 2008; Moubayidin et al., 2009; Perilli et al., 2010). SHY2 transcription is positively regulated by the cytokinin-responsive transcription factors ARABIDOPSIS RESPONSE REGULATOR 12 (ARR12) and ARABIDOPSIS RESPONSE REGULATOR 1 (ARR1). SHY2 acts to repress PIN genes, and therefore negatively affects auxin transport in the root tip, promoting cell differentiation. In contrast, auxin promotes SHY2 protein degradation, and therefore sustains PIN activity and cell division in the proximal meristem (reviewed by Perilli et al., 2012). SHY2 is expressed in roots, particularly in the meristem, at the vascular tissue transition zone (Dello Ioio et al., 2008). Also, the SHY2–ARF-signaling module was suggested to positively regulate LR development and negatively regulate LR initiation, by affecting auxin homeostasis (Goh et al., 2012).
Several lines of evidence suggest that SHY2 is a candidate for perceiving the SL signal for regulation of meristem size and LR development. One is the involvement of SHY2 as a key regulator in meristem size determination by integrating signals from different hormones and its involvement in LR development. A second line of evidence is the regulatory effect of MAX2-dependent SL signaling on meristem size and LR development. A third is that our present results suggest that shy2-31 loss of function is insensitive to GR24 application, for both LR formation and meristem cell number. Moreover, both shy2-31 and max2 were shown here to have reduced sensitivity to cytokinin treatments. Taken together, these results suggest that both MAX2 and SHY2 are involved in SL and cytokinin signaling in the root. As SL signaling was suggested to modulate auxin flux and both SHY2 and MAX2 were shown to regulate PIN-dependent auxin transport in the root, it may be that SL signaling leads to local changes in PIN level via MAX2 and/or SHY2 activity and thereby regulates meristem size and LR formation. Hence, in these cases the mode of SL regulation is probably regional, localized to the root tip, and non-cell-autonomous. Furthermore, although the manner in which SLs regulate SHY2 is unknown, the sensitivity to SLs of the shy2-2 mutant, resistant to auxin-mediated protein degradation, suggests that SLs exert their effect on SHY2 via a mechanism that does not involve SHY2 auxin degradation.
However, the suggested regulatory effect of SLs on auxin flux in the root, via MAX2 and/or SHY2, to regulate meristem size and LR formation is probably root-specific. Despite the fact that SCR is expressed in shoot tissues, including the shoot apical meristem and young and fully expanded leaves (Wysocka-Diller et al., 2000), from examination of shoot branching and SL biosynthesis gene expression it seems that the endodermal effect of SL signaling is restricted to roots and does not confer changes in SL homeostasis in the shoot. In this regard, interestingly, SHY2 mutants branch normally (P. Brewer, pers. comm.), thus further supporting different regulation schemes of SLs for root and shoot.