|Home | About | Journals | Submit | Contact Us | Français|
We undertook a chemical genetics screen to identify chemical inhibitors of brassinosteroid (BR) action. From a chemical library of 10,000 small molecules, one compound was found to inhibit hypocotyl length and activate the expression of a BR-repressed reportergene (CPD-GUS) in Arabidopsis, and it was named brassinopride (BRP). These effects of BRP could be reversed by co-treatment with brassinolide, suggesting that BRP either directly or indirectly inhibits BR biosynthesis. Interestingly, the compound causes exaggerated apical hooks, similar to that caused by ethylene treatment. The BRP-induced apical hook phenotype can be blocked by a chemical inhibitor of ethylene perception or an ethylene insensitive mutant, suggesting that, in addition to inhibiting BR, BRP activates ethylene response. Analysis of BRP analogs provided clues about structural features important for its effects on two separate targets in the BR and ethylene pathways. Analyses of the responses of various BR and ethylene mutants to BRP, ethylene, and BR treatments revealed modes of crosstalk between ethylene and BR in dark-grown seedlings. Our results suggest that active downstream BR signaling, but not BR synthesis or a BR gradient, is required for ethylene-induced apical hook formation. The BRP-related compounds can be useful tools for manipulating plant growth and studying hormone interactions.
Brassinosteroids (BRs) are a family of poly-hydroxylated steroid hormones that are involved in many aspects of plant growth and development. The biosynthesis of biologically active brassinosteroids involves a series of cytochrome P450 and steroid 5α-reductase enzymes. BRs bind to the extracellular domain of the leucine-rich-repeat receptor like kinase (LRR-RLK) BRI1 (Brassinosteroid Insensitive 1) (Li and Chory, 1997; Kinoshita et al., 2005) and activate its kinase function. BRI1 and its co-receptor BAK1, another LRR-RLK (Li et al., 2002; Nam and Li, 2002), transduce the BR signal through an unknown mechanism to soluble downstream components. Two of these components, BIN2 (BR-Insensitive 2) (Li et al., 2001; Li and Nam, 2002) and BSU1 (bri1 Suppressor 1) (Mora-Garcia et al., 2004), a GSK3 kinase and a Ser-Thr phosphatase, respectively, control the phosphorylation states of a family of nuclear transcription factors that include BZR1 (Brassinazole Resistant 1) (Wang et al., 2002; He et al., 2005) and BES1 (bri1 EMS Suppressor 1), also known as BZR2 (Yin et al., 2002; Yin et al., 2005). BZR1 and BZR2/BES1 were identified as dominant gain-of-function mutants that are resistant to brassinazole (BRZ), a BR biosynthesis inhibitor, and suppress the dwarf phenotype of bri1 mutants. Phosphorylation of BZR1 and BZR2/BES1 inhibits their activity through multiple mechanisms, including proteasome degradation, cytoplasmic retention, and abolishment of DNA binding (Gampala et al., 2007; Gendron and Wang, 2007). BR-induced dephosphorylation activates BZR1 and BZR2/BES1 proteins, which directly regulate the transcription of BR-responsive genes.
Ethylene is a gaseous hormone that regulates various developmental processes in plants. It is perceived by a small family of membrane-bound histidine kinase receptors that are similar to bacterial two-component histidine kinases. These include ETR1 (Ethylene Response 1), ERS1 (Ethylene Resistant 1), ETR2 (Ethylene Response 2), ERS2 (Ethylene Resistant 2), and EIN4 (Ethylene Insensitive 4) (Chang et al., 1993; Hua et al., 1995; Roman et al., 1995). Downstream of the receptors, a Ser-Thr kinase, CTR1 (Constitutive Triple Response 1), negatively regulates ethylene signaling in the absence of ethylene, however, the substrates of CTR1 remain unknown (Kieber et al., 1993). Further downstream, EIN2 (Ethylene Insensitive 2), which contains a domain similar to an NRAMP type metal ion transporter, is essential for ethylene response, yet its biochemical role in the pathway remains unclear (Alonso et al., 1999). Finally, ethylene signaling controls transcriptional responses through the regulation of a family of transcription factors called EIN3 (Ethylene Insensitive 3) and the five EILs (EIN3 Like) (Chao et al., 1997). Interestingly, small molecule inhibitors of ethylene biosynthesis and perception have been identified. AgNO3 is one compound that can block the perception of ethylene by the receptors and is useful for chemical genetic studies of ethylene deficiency (Beyer, 1976).
Chemical inhibitors of hormone synthesis or signaling are powerful tools widely used for studying hormone function. BRZ was used to identify BZR1 (Wang et al., 2002) and has been used to study the effects of BR on growth and development in various plant species (Asami et al., 2000). Other BR biosynthesis inhibitors have been discovered, including triadimefon (Asami et al., 2003), propiconazole (Sekimata et al., 2002), tebconazole (Sekimata et al., 2002), BRZ2001 (Sekimata et al., 2001), and spironolactone (Asami et al., 2004). All of these compounds contain a triazole functional group except for spironolactone, which is a steroid like compound that presumably mimics BR and competes for association with BR biosynthesis enzymes.
Chemical genetics has led to identification of components of several hormonal pathways, such as auxin signaling and transport, and vesicle trafficking. Sirtinol was used in a screen for mutants that alter the expression pattern of the auxinreporter, DR5-GUS, and enhance the phenotype of the yucca mutant (Grozinger et al., 2001; Zhao et al., 2003). Yokonolides A & B were identified as natural products of the soil organism Streptomyces diastatochromogenes that inhibit auxin-responsive gene expression (Hayashi et al., 2001; Hayashi et al., 2003). A screen for compounds that affect gravitropic responses identified multiple compounds that affect membrane trafficking in Auxin-dependent and independent manners (Surpin et al., 2005). A screen of a chemical library identified several compounds that block auxin induced gene expression (Armstrong et al., 2004). These studies demonstrate the importance of small molecule screens in plant biology.
We screened 10,000 small molecules to find inhibitors of BR action. Based on hypocotyl growth inhibition in the dark and activation of the BR-responsive reporter gene (CPD-GUS), we identified one compound that we named brassinopride (BRP). Physiological experiments using combinations of BRP, brassinolide, and BR- and ethylene-signaling mutants show that BRP inhibits BR action and promotes ethylene action. These experiments also provide new insights into BR and ethylene crosstalk in seedling development.
Inhibition of BR action causes dwarfism and increased expression of BR-repressed genes. To identify small molecule inhibitors of BR action we screened a diverse set of 10,000 synthesized chemicals using the CPD-GUS transgenic line, which contains a BR-repressed CPD promoter driving the expression of β-glucuronidase (Mathur et al., 1998) (Fig. 1A). The seeds were placed in 96-well plates containing half-strength MS (MS) agar medium and individual chemicals at various concentrations in each well. Plants were grown in the dark for five days at room temperature and examined visually for decreased hypocotyl length. Seedlings showing hypocotyls shorter than wild type grown on MS alone were placed in GUS substrate medium overnight. The histochemical staining pattern and intensity of GUS were examined and those that exhibited darker staining than the control were marked as positive hits for inhibitors of BR.
Ninety chemicals caused short hypocotyls similar to BR inhibitors. To confirm the effects of the chemicals and determine if they inhibit BR signaling or at a step upstream of BR signaling we grew wild type and bzr1-1D mutant plants on medium containing the chemicals and BR. If the wild type plants were rescued by BR, we classified the chemical as a potential biosynthesis inhibitor. If the wild type plants were not rescued by application of BR but were rescued by the bzr1-1D mutant, then the compound was classified as a possible BR signaling inhibitor.
After retesting, one compound caused effects similar to a BR biosynthesis inhibitor (Compound 5141662, Fig. 1B). The hypocotyls of wild type plants grown on the compound at 33.6 μM (the screening concentration) were one third the height of those grown on MS medium, while the hypocotyls of bzr1-1D plants were two thirds the height of those grown on MS medium, indicating that bzr1-1D is resistant to the effects of the compound. The GUS staining in the cotyledons of CPD-GUS plants treated with the compound was darker than the MS control (Fig. 1A). These observations indicate that the compound inhibits BR action.
The IUPAC name for compound 5141662 is N-benzyl-N-(1-cyclopropylethyl)-4-fluorobenzamide. Due to the BR inhibition caused by the compound and the presence of a cyclopropyl side group and a di-benzyl amide backbone we named the compound Brassinopride [Brassinosteroid inhibitor + cyclopropyl + amide = Brassinopride(BRP)].
To determine at which concentration BRP is most effective, we grew wildtype and bzr1-1D plants on BRP concentrations ranging from 0 to 500 μM in the dark (Fig. 2A). The Figure 2A inset shows that the largest ratio of bzr1-1D to wildtype hypocotyl lengths occurs at 100μM, but this is close to the saturation point for BRP and causes the plants to die. The concentration of BRP that causes 50% inhibition of hypocotyl length (IC50) is around 17μM. Since the IC50 of BRP is around 17μM and BRZ is around 1μM (Asami et al., 2000) it seems that BRP is less active than BRZ. On 40μM BRP, hypocotyl lengths of bzr1-1D and WT showa ratio of approximately two, which is similar to that on 2uM BRZ (hypocotyl lengths on BRZ marked by arrows in fig. 2A).
Additionally, we tested the effects of BRP on light-grown wildtype and bzr1-1D plants (Fig. 2B). The effects of BRP on plant size were similar to those of BRZ. Increasing concentrations of BRP caused smaller rosettes, shorter hypocotyls and petioles, and shorter roots; all typical dwarf phenotypes caused by BR deficiency. The bzr1-1D plants are less sensitive on BRP compared to wild type as shown by larger rosettes and longer roots (Fig. 2B). These data suggest that BRP has similar effects in both light- and dark-grown plants.
We tested whether BRP inhibits BR biosynthesis or signaling by co-treating wildtype and bzr1-1D with either 40μM BRP or 1μM BRZ and various concentrations of BL (Fig. 3A–D). Increasing concentrations of BL rescued the BRP-treated plants similarly to BRZ-treated plants (Fig. 3B and 3C), suggesting that BRP inhibits BR biosynthesis. Additionally, we examined the expression of the BR marker gene, CPD, using the CPD::GUS transgenic line (Fig. 3D). BRP increased CPD::GUS expression in the cotyledons, and this was reversed by co-treatment with BL, providing further evidence that BRP inhibits BR biosynthesis. The optimal BL concentration for rescue of both BRP and BRZ is 5nM. As expected, bzr1-1D was not rescued by BR treatment and was slightly hypersensitive to BL, as shown by its shorter hypocotyls in BL treated plants (Fig. 3A–C). These results show that BRP has similar effects on hypocotyl elongation and BR-responsive gene expression as BRZ, and thus BRP most likely inhibits BR biosynthesis rather than signaling.
To test if BRP specifically inhibits BR or also affects other growth-promoting hormones such as gibberellic acid (GA), we co-treated wildtype and bzr1-1D with 40μM BRP, 1μM Paclobutrazol (PAC, a GA inhibitor) (Wang et al., 1986), or 1μM BRZ with various concentrations of GA (Fig. 3E–H). Plants grown in the dark on regular medium showed no further hypocotyl elongation in response to treatment with increasing concentrations of GA (Fig. 3E). Similarly, plants grown on BRP medium showed little response to GA (Fig. 3F). In contrast, GA increased the hypocotyl length of both PAC-treated plants and BRZ-treated plants (Fig. 3G and 3H), indicating that PAC and BRZ inhibit GA synthesis but BRP does not. As a constitutive BR-response mutant, bzr1-1D responded similarly to GA as wild type plants (compare Fig 3A–C with G–H).
To further characterize BRP and the differences between BRP and BRZ we tested the reversibility and inducibility of BRP and BRZ (Fig. 4). To test reversibility we grew wild type plants on either 40μM BRP or 1μM BRZ in the dark for three days and then transferred half of the plants to MS medium and allowed them to grow for an additional two days in the dark. After transfer, plants that were initially grown on BRP had longer hypocotyls than those that were initially grown on BRZ, indicating that they recovered better than those grown on BRZ (Fig. 4A). The increased reversibility of BRP suggests that it is metabolized faster than BRZ or that it has a lower affinity for its target, as suggested by its higher physiological IC50. To test the inducibility, we grew wild type plants on MS medium for two days in the dark and then transferred 30 seedlings to MS containing either 40μM BRP or 1μM BRZ for an additional three days. In this case, the plants transferred to either BRP or BRZ had similar hypocotyl lengths. Thus, both compounds can affect brassinosteroid action after germination.
One characteristic of BRZ-treated or BR-deficient plants is the lack of apical hook formation (Chory et al., 1991; Li et al., 1996; Asami et al., 2000). Surprisingly, BRP shows the opposite effect on apical hook formation to BRZ. BRP causes anenhanced apical hook similar to that caused by ethylene or the ethylene precursor ACC(Fig. 5A). In Arabidopsis, ethylene is known to cause the triple response phenotype in dark grown seedlings: short hypocotyl, short roots, and enhanced apical hook (Guo and Ecker, 2004). Therefore, we tested whether part of the BRP phenotype is due to enhanced ethylene action (Fig. 5). Wild type and ein2-1 (an ethylene insensitive mutant) (Guzman and Ecker, 1990)were grown on BRP, ACC (an ethylene precursor), and BRZ (Fig. 5A), and hypocotyl length and apical hook angle were measured. Interestingly, ein2-1 showed no apical hook formation but increased hypocotyl length on BRP medium. Similar to the ein2-1 mutant, AgNO3 significantly decreased the effects of BRP on the apical hook and partially reversed its inhibition of hypocotyl length (Fig. 5B). In contrast, ein2-1 and AgNO3 have little effect on hypocotyl lengths of plants grown on BRZ (Fig. 5A and data not shown). These results suggest that enhanced ethylene action is responsible for the exaggerated apical hook and contributes partly to the short hypocotyl phenotype caused by BRP.
To address the role of BR in formation of the apical hook in BRP treated plants, wildtype plants were grown on MS, BRP, and BRZ with various concentrations of BL. BL was able to partially suppress the enhanced apical hook phenotype of BRP treated plants suggesting that BR does have a role in apical hook formation (Fig. 5C). This is consistent with previous reports showing that ACC controls BR activity to induce apical hook formation (De Grauwe et al., 2005).
To further test if BRP functions by activating ethylene and inhibiting BR, wildtype, bzr1-1D, bzr1-1D;bes1-D, and ein2-1 plants were co-treated with 1μM BRZ and various concentrations of ACC (Fig. 6) in an attempt to mimic the BRP phenotype. BRZ and ACC have similar inhibitory effect on hypocotyl elongation and opposite effect on apical hook formation. Their effects on hypocotyl elongation are additive when both are applied to plants (Fig 6A). In contrast, the inhibitory effect of BRZ on apical hook formation can be cancelled by increasing concentration of ACC. As such, a combination of 1 μM BRZ and 5 μM ACC yielded similar phenotypes as 40 μM BRP, supporting that the phenotypes caused by BRP can be due to inhibition of BR action and promotion of ethylene action.
The ein2-1 mutant showed no response to ACC, but similar response as wild type to BRZ, consistent with EIN2’s specific function in ethylene signaling. In contrast, the bzr1-1D;bes1-D double mutant and bes1-D single mutant (Fig. 7A) were not only insensitive to BRZ and BRP but also less sensitive to ACC in hypocotyl inhibition than wild type and bzr1-1D (Fig. 6A), suggesting that BES1 mediates ethylene inhibition of hypocotyl elongation. These results support that ethylene-induced hypocotyl shortening is at least in part due to a reduction in BR action. This reduction in BR action can be suppressed by activating the BR pathway through stabilization of BES1 by the bes1-D mutation. In contrast, BR action on the hypocotyl is not mediated by ethylene since ein2-1 shows normal sensitivity to BRZ treatment (Fig. 6A).
To further address the relationship between ethylene and BR in dark grown seedlings, WT (Col-0 and WS), bzr1-1D, bes1-D (WS background), bzr1-1D;bes1-D, bzr1-1D;bri1-116 (a null allele of bri1), and ein2 were grown on MS, BRZ, ACC, and BRP (Fig. 7). As expected bzr1-1D, bes1-D, bzr1-1D;bes1-D, and bzr1-1D;bri1-116 were all resistant to BRZ and BRP in hypocotyl length and apical hook formation (Fig. 7A and 7B). On ACC, only ein2-1, bes1-D and bzr1-1D;bes1-D had longer hypocotyls than the wildtype controls suggesting that ethylene may function through BES1, but not BZR1, to control cell expansion in the hypocotyl (Fig. 7A). Interestingly, bzr1-1D, bes1-D, bzr1-1D;bes1-D, and bzr1-1D;bri1-116 were all slightly resistant to ACC in apical hook formation compared to wildtype (Col-0 and WS) confirming that proper BR action is necessary for ethylene induced hook formation (Fig. 7B). Together these data suggest that ethylene functions partly through BR to regulate both hypocotyl length and apical hook formation, and it is likely that ethylene functions through the BES1 dependent branch of the BR signaling pathway to control hypocotyl elongation.
Previous reports of CPD-GUS expression on the outer side of the apical hook suggested that ethylene establishes a BR gradient in the apical hook that contributes to hook formation (De Grauwe et al., 2005). To test if a gradient of BR is necessary to form an apical hook, the effect of a bri1 null mutant was examined. In the bri1-116;bzr1-1D double mutant, BR perception is defective but downstream BR responses are activated by the increased stability of BZR1. Although the double mutant has no BR receptor for perception of a BR gradient, it still forms an apical hook similar to wildtype plants (Fig. 7B), suggesting that perception of a BR gradient is not required for apical hook formation. The bzr1-1D mutation stabilizes BZR1 and activates the downstream BR signaling pathway, which is sufficient for hook formation in the absence of BR perception. Therefore, the gradient of CPD expression observed previously (De Grauwe et al., 2005) is unlikely to contribute to the apical hook formation. Ethylene may function through modulating downstream BR signaling rather than BR biosynthesis to regulate apical hook formation. Alternatively, an active BR pathway might be required for sufficient ethylene synthesis or signaling, which is supported by the rescue of apical hook formation of BRZ-treated seedlings by increasing concentration of ACC (Fig. 6B). Interestingly, the bzr1-1D;bri1-116 double mutant shows severe hypocotyl bending below the apical hook when the plants are grown on ACC (Fig. 7C), suggesting BR perception is required for maintaining a straight hypocotyl when the ethylene pathway is activated. Proper BR synthesis and signaling appears necessary for maintaining a straight hypocotyl in the presence of ethylene. These results support a complex interaction between BR and ethylene.
It is possible that different structural domains of BRP are important for its effect on BR and ethylene pathways. We used the program ChemMine (Girke et al., 2005) to search for analogs of BRP (Fig. 8A). We grew wild type and bzr1-1D plants on twelve analogs of BRP to test if any analogs are specific to either BR or ethylene and to identify functional groups of interest on the BRP molecule (Fig. 8).
One of the analogs tested (a3) showed the same level of biological activity as BRP in both hypocotyl inhibition and apical hook formation (Fig. 8A and 8B). Interestingly, a3 does not contain the cyclopropyl group present in BRP, suggesting that the cyclopropyl side group on BRP is not necessary for its biological activity. Furthermore, changes in the backbone and phenyl-based side group of the molecule appear to affect the specificity and effectiveness of these compounds. All compounds missing the alkane located next to the amide (a4, a5, a7, a9), an isopropyl or cyclopropyl side chain (a2, a5, a12), or one of the phenyl rings (a10, a11) are less effective. These observations indicate that the backbone, size, and shape of the molecule are important for the effectiveness of BRP. Compounds with nitro groups replacing the halogen on the phenyl ring (a4 & a8) have decreased effectiveness (Fig. 8A and 8B), likely due to the decreased lipophilic nature of the nitro compounds, which could reduce membrane permeability.
It would be of interest to find BRP-like molecules that specifically affect either BR or ethylene independently. None of the BRP analogs affected BR specifically, but one of them, a6, seems to have reduced effect on BR action. Wild type plants grown on compound a6 showed reduced hypocotyl length and enhanced apical hooks (Fig. 8A and 8B). However, the bzr1-1D mutation was less effective in suppressing the effect of a6 than that of BRP or a3, suggesting that a6 may be more specific to the ethylene pathway and less effective in BR inhibition than BRP. The a6 compound contains a urea group in place of the amide group in BRP. These results suggest that modifying the side-groups of BRP can have specific effects on the activity of BRP in BR and ethylene functions. The results support the hypothesis that BRP has two targets in plants, one involved in BR action and the other involved in ethylene action.
Based on inhibition of cell elongation and activation of BR-repressed gene expression, we identified BRP as a chemical that inhibits BR action. Abolishment of BRP’s effects on hypocotyl inhibition and CPD-GUS expression by co-treatment with BR and by bzr1-1D mutation suggests that BRP inhibits BR biosynthesis rather than BR signaling. Interestingly, BRP also causes exaggerated apical hooks in dark-grown seedlings, an effect similar to that of ethylene. Physiological experiments using ethylene mutants and treatment with ethylene (ACC) and an ethylene perception inhibitor support that BRP promotes ethylene action at a step of or upstream of ethylene perception, possibly ethylene synthesis. These experiments establish BRP as a chemical with dual activities of inhibiting BR action and promoting ethylene action. Our experiments using a combination of BRP with BR- or ethylene mutants further elucidate the mode of crosstalk between the BR and ethylene pathways.
Most of the compounds known to inhibit BR biosynthesis are triazole type cytochrome P450 inhibitors, or BR structural mimics such as spirinolactone. BRP is likely to represent a unique class of BR biosynthesis inhibitors that does not contain the traditional triazole structural component or structural similarities to BRs. It is interesting to note that compounds containing cyclopropyl groups have been found to interact with or be catalyzed by cytochrome p450 enzymes (Shaffer et al., 2002; Cryle et al., 2005), but our results suggest that the cyclopropyl group is not necessary for BRP’s function in BR inhibition or ethylene agonism (Fig. 8A, compound a3). Instead, the common factors necessary for BRP to inhibit BR are the amide backbone, the two benzene rings, and the presence of side chains on the amide backbone and the benzene ring. Interestingly, BRP has been patented (Patent # WO 2004 089470) (Andersen et al., 2004) as an inhibitor of 11β-hydroxysteroid dehydrogenase in animals. It is possible that BRP also directly inhibits an enzyme involved in steroid modification in Arabidopsis.
Our data show that BRP not only inhibits BR action but also activates the ethylene pathway. BRP is likely to increase ethylene synthesis because the ethylene-like effects of BRP are abolished by AgNO3, which blocks ethylene perception. Furthermore, the BRP-like effects can be mimicked when BRZ is combined with ACC treatment. BRP does contain a cyclopropyl side group that is structurally similar to ACC, but analyses of analogs suggest that the cyclopropyl group is not necessary for BRP’s function in ethylene agonism. One possibility is that the alkane linked to the amide backbone and the cyclopropyl can be metabolized either to ACC or ethylene. Alternatively, BRP may directly affect an enzyme in ethylene synthesis. Genetic studies of mutants in the ethylene biosynthesis pathway or direct measurements of ethylene and its precursors in BRP-treated plants will elucidate how BRP acts on ethylene pathway. Further studies of more varied analogs of BRP could provide more information about the functional groups of BRP and lead to identification of compounds that specifically affect BR or ethylene.
The interactions between hormone pathways have become a focus for many labs studying hormone signaling, and the identification of small molecules affecting multiple pathways will aid in the dissection of hormone crosstalk. Paclobutrazol is an example of one small molecule that inhibits both GA and ABA synthesis in plants (Wang et al., 1986; Wang et al., 1987) and ABA synthesis in Cercospora rosicola, a fungal plant pathogen (Norman et al., 1986). BRZ inhibits the synthesis of both BR and GA. BRP is a unique compound that acts on both BR and ethylene in plants. The dual role of BRP in ethylene and BR function makes BRP a useful tool for investigating BR-ethylene crosstalk.
Recent studies have shown that BR and ethylene have overlapping functions in hypocotyl elongation and apical hook formation (De Grauwe et al., 2005). It was suggested that ethylene controls the biosynthesis of BRs and establishes a gradient of BR in the apical hook region that contributes to the hook formation. Our results refine this conclusion by showing that a null bri1 mutant crossed to the bzr1-1D gain of function mutant forms an apical hook similar to wildtype. In this case BR perception is negated by the lack of BRI1, but the downstream pathway is activated by the bzr1-1D mutation, which stabilizes the BZR1 protein. Therefore, perception of a gradient of BR appears to be not necessary to form a proper apical hook; rather it is likely the activation of the BR signaling pathway that is required. Ethylene may control the gradient of activity of BR signaling downstream of BRI1, or activation of certain BR responsive genes by BZR1 is required for ethylene promoted apical hook formation.
Our results showing that the bzr1-1D;bes1-D double mutant is partially resistant to ACC in inhibiting hypocotyl elongation suggests that ethylene inhibition of hypocotyls requires normal BR response. Yet, BR does not seem to act through ethylene in the hypocotyl since ein2-1 is not resistant to BRZ. Therefore, our results support that ethylene reduces BR action to inhibit hypocotyl elongation, which supports that ethylene, at least in part, acts by recruiting the functions of other hormones to control hypocotyl elongation and hook formation (De Grauwe et al., 2005). It is possible though that BR also regulates ethylene action, as BR has been shown to regulate the level of an ACC oxidase (Deng et al., 2007). Further chemical genetic studies using BRP as a tool will shed light on the molecular mechanisms of BR-ethylene crosstalk in seedling development.
Arabidopsis thaliana ecotype Columbia was used for all experiments (except bes1-D which is in WS background). Seeds were sterilized in 75% Ethanol with 0.01% Triton X-100. Seeds were sown on 0.5X Murashige and Skoog (MS) media containing 0.8% Phytoblend (Caisson, Utah, USA). All seeds were cold treated for 48 hours, placed in the light overnight and then grown on vertical plates (except screen plates and light grown plants) in the dark at room temperature for four days. Light grown plants were transferred from the cold to the light and grown on horizontal plates for 10 days.
The DIVERSet library (ChemBridge, San Diego, CA), containing 10,000 organic molecules in 96 well plate format, was used for the screen. 0.1mg of each compound was dissolved in 20μl of DMSO and then diluted 5 fold in water to approximate concentrations of 2–4mM in 20% DMSO. 1μl of the diluted compound was then aliquotted to new 96 well plates and diluted in 99μl of warm 0.5X MS and 0.8% Phytoblend which was allowed to solidify before seeds were sown. Approximately 10 CPD::GUS transgenic seeds were placed in each well, and part of each plate was used for MS, BL, and BRZ controls. The plants were screened visually for hypocotyl length, apical hook formation, and cotyledon opening.
Histochemical staining was used for the secondary screen. Approximately one half of the seedlings from the wells that had plants with short hypocotyls were placed directly in GUS substrate media. These were incubated at 72°C for approximately 12 hours and visualized under a dissecting microscope. Those lines showing darker histochemical staining than the MS grown controls were scored as positive hits.
Pictures of each plate were taken using a digital camera and then analyzed using ImageJ (National Institutes of Health) software. Hypocotyl and apical hook measurements were done for at least 20 seedlings and error bars represent standard error. The angle tool was used to measure apical hook angles.
We thank Dr. Chris Somerville for providing the chemical library. This work was supported by a research grant (R01GM66258, Z-Y. W) and a training grant (5T32GM007276, J.M.G.) from NIH.