The Small Molecule 2-Furylacrylic Acid Inhibits Auxin-Mediated Responses in Arabidopsis thaliana

Transcription

1 Plant Cell Physiol. 48(12): (27) doi:1.193/pcp/pcm141, available online at ß The uthor 27. Published by xford University Press on behalf of Japanese Society of Plant Physiologists. ll rights reserved. For permissions, please The Small Molecule 2-Furylacrylic cid Inhibits uxin-mediated Responses in rabidopsis thaliana Can Sungur 1, Sarah Miller 1, 2, Johann Bergholz 1, 2, Rebecca C. Hoye 2, Ronald G. Brisbois 2 and Paul vervoorde 1, * 1 Department of Biology, Macalester College, St Paul, M 5515, US 2 Department of Chemistry, Macalester College, St Paul, M 5515, US uxins, typified by I, are a class of plant hormones involved in a wide array of processes including cell division, cell elongation, tissue patterning, phototropism, gravitropism and root development. Despite recent descriptions of the machinery involved in auxin transport and perception, additional regulatory mechanisms and components remain to be identified and characterized. Chemical genetics has proven to be a valuable means by which to investigate the auxin response pathway in rabidopsis thaliana. screen for small molecules that block auxin signaling was performed previously, leading to the characterization of four compounds with this inhibitory activity. Here, we have synthesized various analogs of one of these molecules, compound, a furyl acrylate ester of a thiadiazole heterocycle. The biological activity of these derivatives was initially assessed based on their ability to inhibit the auxin-inducible expression of the B3-GUS reporter gene and indicated that the active portion of the molecule was 2-furylacrylic acid (2-F). In the micromolar range, 2-F attenuates the auxin-inducible expression of I5, fails to alter the interaction of I7/XR2 with the SCF TIR1 complex and inhibits both root and hypocotyl elongation of wild-type seedlings. Based on our structure function analysis of compound, we conclude that 2-furylacrylic acid is liberated by hydrolysis of an ester linkage. Identification of the cellular target of this molecule will add to our understanding of auxinmediated events. Keywords: uxin Chemical genetics 2-Furylacrylic acid. bbreviations: TS, rabidopsis thaliana salts; Cy-FE, cyclohexyl-furylacrylate ester; DCP-F,3-[5-(3,4-dicholorphenyl)-2-furyl]acrylic acid; DMS,dimethylsulfoxide; 2-F,2- furylacrylic acid; 3-F,3-furylacrylic acid; FE,-(2-furfurylideneacetyl)-glycine methyl ester; GST,glutathione S-transferase; GUS,b-glucuronidase; HT,2-hydroxy--(5-propyl-1,3,4-thiadiazol-2-yl)acetamide; 1-,1-naphthalene acetic acid; P, 1-naphthylphthalamic acid; TIB,2,3,5-triiodobenzoic acid; TIR1,transport inhibitor 1. Introduction uxins are phytohormones that facilitate a diverse array of cellular and molecular events during plant growth and development. The primary free auxin in most plants is I. During development or in response to environmental stimuli, I is redistributed, leading to a number of cellular changes, including altered rates of protein degradation, changes in membrane potential and varying patterns of gene expression (Woodward and Bartel 25). cascade of gene expression changes underlies many of the phenotypic responses associated with auxin effects. Results from a range of studies indicate that Rs corresponding to members of three large gene families, GH3, ux/i and SUR, accumulate rapidly in response to auxin (Woodward and Bartel 25). subset of proteins encoded by the 19-member GH3 and 29-member ux/i gene families have been assigned biochemical functions, while the function of the47-member SUR protein family remains unknown. The GH3 proteins function as Iconjugating enzymes involved in auxin homeostasis (Staswick et al. 25, Park et al. 27) and the ux/i proteins contribute a critical role to auxin signaling by serving as transcriptional repressors (Ulmasov et al. 1997). t least one way that the ux/i proteins contribute their repressive activity is by dimerizing with members of the auxin response factor (RF) protein family, which have the capacity to interact with the promoters of auxinregulated genes (Kim et al. 1997, Liscum and Reed 22, Woodward and Bartel 25). combination of genetic and biochemical studies established that control of ux/i levels by regulated proteolysis is critical for proper auxinmediated gene expression and the concomitant changes in plant development (Dharmasiri and Estelle 24, Dreher and Callis 27). The E3 ubiquitin-ligase complex, called SCF TIR1, controls ux/i protein levels through an auxin-stabilized interaction (Gray et al. 21). TIR1 encodes an F-box protein that associates with Skp1 and a Cullin protein to form a functional E3-ligase complex. uxin stabilizes the interaction between TIR1 and ux/i proteins, leading to an Downloaded from by guest on 1 ctober 218 *Corresponding author: , overvoorde@macalester.edu; Fax, þ

2 Furylacrylic acid blocks auxin responses enhanced rate of ux/i turnover. The TIR1 ux/i interface serves as the auxin receptor involved in auxininduced proteolysis of ux/i proteins (Gray et al. 21, Dharmasiri et al. 25a, Kepinski and Leyser 25). The recent description of TIR1 auxin ux/i crystal structures demonstrates that auxin serves as molecular glue to stabilize the interaction between TIR1 and the ux/ I proteins (Tan et al. 27). Interestingly, other auxin-resistant mutants have revealed that modification of the cullin CUL1 by RUB1/ EDD8, a ubiquitin-related protein, regulates SCF TIR1 activity. Components of this regulatory pathway include XR1, ECR1 and RCE1 (reviewed by Dharmasiri and Estelle 24, Woodward and Bartel 25). lthough the molecular mechanisms that control SCF TIR1 assembly and activity have not been defined, loss-of-function mutations in these components or other related factors (e.g. CD1 or SGT1b) result in auxin-resistant phenotypes (Gray et al. 22, Cheng et al. 24, Chuang et al. 24, Quint et al. 25). t the cellular level, much has been learned about the importance of the PI and PGP/MDR auxin transport proteins for creating auxin gradients required for patterning and development (Blakeslee et al. 25, Tanaka et al. 26). The capacity of plants to reorganize auxin gradients in response to environmental stimuli results from the recycling of transport proteins from the plasma membrane to endocytic vesicles and then back to the membrane (Paciorek et al. 25, Sauer et al. 26). The mechanism responsible for localization is not well understood, although it appears that the auxin transport inhibitor 1-naphthylphthalamic acid (P) blocks some aspect of the endocytic step in PI protein cycling, while the auxin transport inhibitor 2,3,5-triiodobenzoic acid (TIB) leads to an intracellular accumulation of PI protein associated with endosomes (bas et al. 26). Despite the advances in understanding both the distribution and response to auxin, additional factors that control subcellular localization, protein complex formation and additional signaling mechanism have not been described. Chemical genetics represents a useful tool to elucidate additional components or provide insight into the role of auxin during plant growth and development. Several small molecules affecting auxin-related processes have been purified from microbial extracts or other synthetic sources (Hayashi et al. 21, Hayashi et al. 23, Dai et al. 25, Yamazoe et al. 25). In addition, large-scale screens of combinatorial libraries for novel molecules affecting this pathway have led to the identification of additional inhibitory compounds (rmstrong et al. 24, Surpin et al. 25). Defining the active core moieties and describing their mode of action through identification of target cellular components B S S H H D E F G H H Fig. 1 The chemical structures and biological activity of compound, HT and furylacrylic acid. The structures of compound () and its two constituents, a 2-hydroxy-- (5-propyl-1,3,4-thiadiazol-2-yl) acetamide, HT (B), and 2-furylacrylic acid, 2-F (C), are shown. B3 GUS seedlings were either mock treated (D) or treated for 2 h with 5 mm I alone (E), 5 mm I and 5 mm compound (F), 5 mm I and 1 mm HT (G), or 5 mm I and 5 mm 2-F (H) before being stained for GUS activity for 12 h. Scale bar ¼ 1 mm. remains a challenge. In 24, rmstrong et al. reported the initial identification of compound, a furyl acrylate ester of a thiadiazole heterocycle, and three additional molecules as inhibitors of auxin-related signaling. Using compound derivatives, we show that 2-furylacrylic acid (2-F) is the active component of this molecule by demonstrating that 2-F attenuates auxin-responsive gene expression and decreases the rate of auxin-stimulated XR17 b-glucuronidase (GUS) protein turnover. Further, physiological assays show that 2-F alters auxin responses in the root. Taken together, these data show that 2-F is a potent inhibitor of auxin-mediated events. Results 2-F is the active component of compound The ability of compound (Fig. 1) to alter the auxin signaling pathway (rmstrong et al. 24) prompted us to define the active core moiety of this synthetically accessible molecule. The auxin-responsive B promoter element has been extensively characterized (Ballas et al. 1995), and a transgenic rabidopsis line harboring this promoter (B3) fused to the GUS gene was used in the original chemical C H Downloaded from by guest on 1 ctober 218

3 2-Furylacrylic acid blocks auxin responses 1695 genetic screen that identified compound as a smallmolecule inhibitor of auxin-mediated gene expression (rmstrong et al. 24). s a qualitative approach to determine the activity of compound (Fig. 1) or its derivatives, 2-hydroxy--(5-propyl-1,3,4-thiadiazol-2-yl)acetamide (HT; Fig. 1B) and 2-F (Fig. 1C), B3 seedlings were mock treated or treated with 5 mm I alone or in the presence of one of these compounds followed by histochemical staining for GUS activity (Fig. 1D H). s previously reported (ono et al. 1998), I induces the expression of B3-GUS in the distal cell elongation zone of roots, and the addition of compound abolishes this auxin-inducible accumulation of GUS activity (rmstrong et al. 24). Interestingly, HT fails to affect the auxininduced accumulation of enzyme activity, while 2-F blocks GUS accumulation, indicating that this portion of the molecule is the active core moiety. In order to characterize the effect of 2-F on plant growth, we transferred 4-day-old seedlings to TS medium supplemented with increasing concentrations of this compound and monitored root growth. s previously described by rmstrong et al. (24), compound inhibits primary root growth in the 5 2 mm range (Fig. 2). In a similar manner, 2-F also inhibits root growth while HT has no effect on the rate of root elongation (Fig. 2). To analyze directly the effects of compound and its derivatives on altering auxin-inducible gene expression, we used real-time reverse transcription PCR (RT PCR) to monitor the steady-state levels of an endogenous, primary auxin-response gene, I5 (t1g1558), which belongs to the ux/i gene family (bel et al. 1995). uxin induces the expression of this gene, and both compound and 2-F strongly attenuate the auxin induction of I5 expression, while HT is ineffective at blocking I5 expression (Fig. 2B). The slight reduction of I5 induction in the presence of HT was reproduced in each of the three biological replicates, but in each of the bioassays, HT did not alter plant growth. In contrast, the altered expression patterns seen in the presence of compound and 2-F are consistent with the phenotypic data and corroborate the notion that 2-F is the active moiety of compound. 2-F attenuates the degradation of an ux/i repressor Exogenous auxin stimulates the degradation of ux/ I proteins, which serve as transcriptional repressors (uellet et al. 21, Zenser et al. 21). ne mechanism to explain the F-mediated changes in auxin responses described above could be that this small molecule alters the stability of ux/i proteins. The rabidopsis HS::XR3T-GUS transgenic line contains a transgene that mediates the heat shock-inducible production of an I17/XR3 GUS translational fusion protein Root growth (% of untreated control) B Relative R level Compound concentration (mm) I I + Comp. I + HT I + F Fig. 2 Quantitative analysis of the inhibitory effects of compound, HT or F. () rabidopsis (Col- ecotype) seedlings were grown for 4 d on TS and then transferred to fresh TS or TS supplemented with the indicated concentration of 2-F (open circle), compound (filled square) or HT (open triangle) for an additional 4 d. Root growth is plotted as a percentage of the root length of seedlings transferred to TS in the same experiment. Individual data points represent the mean SD of at least three independent experiments (total n ¼ 36 94). (B) The auxin inducibility of the ux/i gene I5 is attenuated in the presence of either compound or 2-F, but is unaffected by HT. rabidopsis seedlings were grown in an aseptic liquid culture for 7 d before being mock treated or treated for 1 h with 5 mm I, 5 mm I and 1 mm compound, 5 mm I and 1 mm HT, or 5 mm I and 5 mm 2-F. Total R was isolated and used as template in a one-step semi-quantitative real-time PCR analysis. The constitutively expressed gene t5g655 (Czechowski et al. 25) was used as a reference and the data (SD) are expressed as relative R levels (I5/t5g655). These data are from one experiment and are representative of three biological replicates of this experiment. (Gray et al. 21). fter stimulating the production of the reporter protein at 378C for 2 h, treatment of seedling with 5 mm I plus either compound or 2-F slowed the disappearance of the fusion protein, when compared with Downloaded from by guest on 1 ctober 218

4 Furylacrylic acid blocks auxin responses GUS activity (nmol MG 5 mg 1 hr 1 ) Root growth (% of untreated control) mock +Comp. + 1 mm 1- t min DMS Comp. HT F B C D DMS I I + Comp. +HT +F I + F +FE Comp. F F Concentration of 2-F (mm) Fig. 3 Effects of 2-F and compound on SCF TIR1 -mediated protein turnover. The ability of compound or 2-F to alter the rate of I17/XR3 GUS fusion protein degradation was determined using 7-day-old light-grown HS::XR3T-GUS rabidopsis, which were heat-shocked (378C) for 2 h to induce expression of the fusion protein. The seedlings (n ¼ 15 in each of three experiments) were then incubated with 1 mm 1- plus either DMS, 1 mm compound (Comp.), 1 mm HT or 5 mm 2-F for 5 min at 228C. The amount of GUS activity remaining was assessed qualitatively by staining for GUS activity followed by light seedlings that were incubated with 5 mm I plus either dimethylsulfoxide (DMS) or HT, which led to the rapid removal of this protein (Fig. 3, B). These data are consistent with previous observations and indicate that compound attenuates the rate of ux/i protein degradation. The degradation of ux/i proteins is dependent on their interaction with the SCF TIR1 complex (Gray et al. 21). Immunoprecipitation assays using glutathione S-transferase (GST) I7/XR2 and plant extracts from seedlings expressing myc-tagged TIR1 have allowed the identification of an auxin receptor (Dharmasiri et al. 25a, Kepinski and Leyser 25). In order to test whether 2-F or compound is capable of blocking the interaction of TIR1 with the ux/i proteins, we performed a series of immunoprecipitation assays. either compound nor 2-F blocks the auxin-stabilized interaction of GST I7/XR2 with myc-tir1 (Fig. 3C). Furthermore, these assays show that neither 2-F nor compound by themselves is able to stimulate the interaction of myc- TIR1 with GST I7/XR2. These biochemical assays are supported by the observation that the roots of tir1-1 and axr2-1 seedlings respond in the same way as the wild type to increasing concentrations of 2-F (Fig. 3D). These results indicate that 2-F inhibits the auxin signaling pathway either by altering a component upstream of the SCF TIR1 complex or by serving as a general proteosome inhibitor. Structure activity analysis of F derivatives We synthesized or obtained a series of F derivatives to explore the structural requirements for its activity. Fig. 4 E shows the chemical structure of five of the 13 microscopy () or quantitatively using using 4-MUG as a substrate (B). For (B), extracts from seedlings were made 1 h after transfer to TS at 238C containing the indicated compounds. (C) The interaction of the ux/i protein I7/XR2 with the TIR1was assessed by co-immunoprecipitation assays using protein extracts made from 7-day-old light-grown rabidopsis seedlings expressing myc-tir and bacterially produced GST I7/XR2. These mixtures were mock treated (DMS) or supplemented with 5 mm I, 5 mm I and 1 mm compound (Comp.), 5 mm I and 5 mm 2-F, or either 5 mm compound or 2-F alone. The amount of myc-tir1 pulled down using glutathione agarose was detected by immunoblotting using an anti-myc monoclonal antibody. Molecular mass standards are indicated. (D) Root growth of Col (open square), axr2-1 (open triangle) and tir-1 (open circle) seedlings. Seedlings were grown for 4 d on TS and then transferred to fresh TS or TS supplemented with the indicated concentration of F for an additional 4 d. Root growth is plotted as a percentage of the root length of seedlings transferred to TS in the same experiment. The mean (SD) length of the main root was determined for seedlings of each line in three independent experiments. The length of the control roots grown on TS were: Col, mm; tir1-1, mm; and axr2-1, mm). Downloaded from by guest on 1 ctober 218

5 2-Furylacrylic acid blocks auxin responses 1697 D F H Root growth (% of untreated control) 2-F B 3-F C Cy-FE FE H H molecules we examined. We first determined whether the position of the acrylic acid would affect the activity of F. 2-F and 3-F inhibit root growth with similar potency, suggesting that the position of the acrylic acid does not affect activity (Fig. 4F). Similarly these molecules are equally effective at blocking auxin-induced B3-GUS or DR5-GUS expression (data not shown). In compound, 2-F is connected to the aminothiadiazole through an ester linkage. Because of the prevalence and diversity of esterases, it is possible that compound is hydrolyzed into 2-F and HT. To explore this possibility, we synthesized cyclohexylfurylacrylate ester (Cy-FE; Fig. 4C) in which the HT portion of compound is replaced by a cyclohexyl ester. This Cy-FE molecule inhibits root growth, similarly to compound, but it is less effective than either 2- or 3-F (Fig. 4F). ext we tested whether the formation of the acrylic acid is an important structural requirement by examining the activity of -(2-furfurylideneacetyl)-glycine methyl ester (FE; Fig. 4D), in which the ester linkage Cl Cl DCP-F FE 3-F Cy-FE DCPF 2-F Compound concentration (mm) Fig. 4 The chemical structures and effects of compound and F analogs on root growth. The structures of 2-furylacrylic acid, 2-F (), 3-furylacrylic acid, 3-F (B), cyclohexylfurylacrylate ester, Cy-FE (C), -(2-furrurylideneacetyle)-glycine methyl ester, FE (D) and dichlorophenoxy-furylacrylic acid, DCP-F (E) are shown. (F) Root growth of Col- seedlings on TS supplemented with varying concentrations of F analogs. Seedlings were grown for 4 d on TS and then transferred to fresh TS or TS supplemented with the indicated concentration of compound for an additional 4 d. Root growth is plotted as a percentage of the root length of seedlings transferred to TS in the same experiment. Individual data points represent the mean SD from 3 4 independent experiments (n ¼ 36 48). E H is replaced with an amide that is much less susceptible to hydrolysis. In contrast to 2-F alone or Cy-FE, FE was ineffective at inhibiting root growth (Fig. 4F), altering the rate of I17/XR3 GUS turnover (Fig. 3, B), and was unable to block auxin-induced B3-GUS or DR5-GUS expression (data not shown). lthough we cannot rule out differences in permeability of these molecules, taken together the analyses of these molecules strongly indicate that the ester linkage of compound is hydrolyzed to release 2-F, which in turn is capable of altering auxinmediated responses. Finally, we compared the biological activity of 3-[5-(3,4-dichlorophenyl)-2-furyl]acrylic acid (DCP-F; Fig. 4E). DCP-F, molecule in the Chembridge DIVERSet library, was identified in a chemical genetic screen aimed at identifying small molecules that affect the endomembrane system during the gravitropic response (Surpin et al. 25). We found that while DCP-F inhibits root elongation, it is less effective than compound, Cy-F, or either 2-F or 3-F. 2-F attenuates the inhibitory effect on root growth by 2,4-D and 1- B3-GUS seedlings treated with either 2,4-D or 1-naphthaleneacetic acid (1-) show strong GUS staining in the distal root elongation zone (Fig. 5, C) and 2-F blocks this molecular response to these auxins. ne potential mechanism to explain the altered auxin responses conferred by 2-F is that this molecule alters the transport of I. The natural auxin I is a substrate for both influx and efflux carriers. Previous reports show that the synthetic auxin, 1- enters the cell primarily through diffusion and is a substrate for the efflux carrier, while 2,4-D enters the cell by carrier-mediated influx and is a less efficient substrate for the efflux carrier (Delbarre et al. 1996, Parry et al. 21). s a means to address whether 2-F is altering auxin transport or auxin response, we examined the effect on root growth of seedlings transferred either to media containing 2,4-D or 1- alone or in the presence of increasing concentrations of 2-F. Based on dose response assays using 2,4-D or 1- alone, we determined that.6 mm 2,4-D and.5 mm 1- inhibited root growth by 9% (data not shown). Interestingly, root growth was restored when seedlings were transferred to media containing both 2-F and the inhibitory quantities of either 2,4-D or 1- (Fig. 5E) suggesting that transport into or out of the cell is not affected, but rather that 2-F alters another aspect of the cellular response to exogenous auxin. Finally, growth of seedlings at 288C stimulates I production and leads to longer hypocotyls (Gray et al. 1998). Because we wanted to determine whether the effects of compound and 2-F were root specific, we analyzed the length of seedling hypocotyls grown at 288C in the Downloaded from by guest on 1 ctober 218

6 Furylacrylic acid blocks auxin responses B C D absence or presence of these molecules (Fig. 5F). These analyses show that 2-F also attenuates the temperatureinduced increases in hypocotyl length that have been linked to higher levels of endogenous auxin. E F Root growth (% of untreated control) Hypocotyl length (mm) Concentration of F (mm) Concentration of F (mm) Fig. 5 2-F suppresses the induction of B3-GUS expression by 1- or 2,4-D, blocks the inhibition of root growth by these auxins and attenuates hypocotyl elongation mediated by elevated levels of endogenous I. Four-day-old seedlings were incubated in TS supplemented with ().5 mm 1-, (B).5 mm 1- and 1 mm 2-F (C).6 mm 2,4-D or (D).6 mm 2,4-D and 1 mm 2-F for 4 h and then stained for GUS activity for 16 h. Representatives of 2 25 seedlings are shown. (E) Four-day-old Discussion We have reported the identification of 2-F as the active core component of compound, which rmstrong et al. (24) identified. Using a combination of molecular, biochemical and physiological approaches, we have characterized the response of rabidopsis seedlings to this molecule. The role of the F-box protein TIR1 as the recognition component of the auxin-responsive SCF complex that interacts with ux/i proteins to target them for proteolysis has been convincingly demonstrated (Gray et al. 21, Dharmasiri et al. 25a, Kepinski and Leyser 25, Tan et al. 27). The use of a chemical genetics approach by several groups reveals that there is chemical specificity in auxin signaling mediated by this system (ono et al. 23, Rahman et al. 26, Walsh et al. 26). The three auxin signaling F-box (FBs) proteins that share the highest level of sequence similarity with TIR1 (FB1 3) play an overlapping role in mediating plant response to I and 2,4-D, as combinations of mutations in these four genes give rise to plants with severe developmental phenotypes and increasing 2,4-D resistance (Dharmasiri et al. 25b). Plants with mutations in R1, which encodes a small acidic protein, show resistance specifically to 2,4-D and not I or 1-, suggesting that the responses to 2,4-D and I can be at least partially separated by this regulatory component (Rahman et al. 26). Finally, mutations in the more distantly related FB5 confer resistance to the synthetic auxin, picoloram, while providing very limited or no cross-resistance to 2,4-D or I (Walsh et al. 26). Collectively, these reports demonstrate the benefits of performing genetic screens using structurally diverse auxin-related molecules to identify novel components and regulatory pathways. Previous studies have generalized the structures of natural and synthetic auxins as having a carboxylic acid moiety separated from an aromatic ring by approximately seedlings were transferred to TS medium supplemented with 2-F alone (open triangles),.6 mm 2,4-D and increasing concentrations of 2-F (filled triangles) or.5 mm 1- and increasing concentrations of 2-F (filled squares). fter 4 d, the length of the roots was determined. The data represent the mean SD of three independent experiments, each with n ¼ (F) Col- seeds were plated on TS medium containing increasing concentrations of 2-F and then grown at 288C with constant light for 7 d. The error bars represent the mean SD of the hypocotyl lengths from three independent experiments, each with an n ¼ Scale bars ¼ 1 mm. Downloaded from by guest on 1 ctober 218

7 2-Furylacrylic acid blocks auxin responses nm (Porter and Thimann 1965). nalogs of compound that lack the ester linkage are ineffective at inhibiting auxin-induced gene expression or primary root elongation, while those that contain the carboxylic acid group or can form a carboxylic acid group by hydrolysis are effective at altering auxin responses. Thus, although 2-F satisfies the structural requirements of an auxin, it has distinct antiauxin effects including blocking auxin-induced gene expression, slowing the rate of I17/XR3 GUS protein turnover, and attenuating root elongation. Furthermore, although 2-F does block primary root growth, it does not induce the expression of synthetic or auxin-regulated genes. Identifying the cellular target of 2-F remains an important challenge. Small molecules that block auxin influx or mutants that are defective in auxin influx carriers inhibit primary root growth and alter gravitropic responses (Marchant et al. 1999, Parry et al. 21). These transport-dependent responses can be suppressed by supplying 1-, which enters the cell in a protein-independent fashion (Delbarre et al. 1996). In contrast, while 2-F alters root growth and gravitropic responses, it blocks the induction of B3-GUS expression in responses to I and 2,4-D, as well as 1-. Furthermore, 2-F relieves the inhibitory effects of both 2,4-D and 1- on root elongation (Fig. 5E). Small molecules that block auxin efflux or mutants that are defective in auxin efflux lead to an accumulation of auxin, alter root growth and change the expression of the DR5-GUS region (Friml et al. 22, Geisler et al. 25). If 2-F inhibits the efflux of auxin, similar changes in the pattern of reporter gene expression would be expected, but are not observed. Taken together, these data suggest that 2-F does not block the uptake or efflux of auxin into the cell, but rather alters an aspect of the signaling machinery. Small molecules that alter cellular activity are powerful tools for defining and manipulating protein function in vivo and represent major classes of drugs used in humans (Walsh and Chang 26). ew synthetic approaches create novel small molecules that fill a wide range of chemical space and lead to the possibility of addressing two problems that arise when trying to work with proteins that co-exist as members of a multigene family. First, molecules that interact with all members of a protein family might be identified, creating a tool to disrupt, or activate, this entire suite of proteins at once. lternatively, small molecules that selectively inhibit or stimulate individual isoforms can be identified. Selective modulators might be identified during an initial large-scale screen or they might arise during a detailed structure function analysis of a more general inhibitor (Kim et al. 2, Hirao et al. 23). While we have presented evidence that 2-F on its own does not stimulate or alter the auxin-mediated interaction between myc-tir1 and GST I7/XR2 in vitro, it is possible that 2-F could alter the interaction of other FB proteins with ux/i proteins or that 2-F could be metabolized in vivo to another form that could alter TIR1 ux/i interactions. The wild-type sensitivity of afb1-1, afb2-1 and afb3-1 mutants to increasing concentrations of 2-F (data not shown) makes it unlikely that these closely related proteins are targets of 2-F. In a screen aimed at uncovering molecules that could provide insight into the link between endomembrane trafficking and gravitropism, Surpin et al. (25) identified molecule , i.e. 3-[5-(3,4-dicholorphenyl)-2-furyl]acrylic acid, DCP-F. While this molecule has a 2-F component, the dichlorophenyl group at the 5 position of the furan ring introduces unique chemical and structural characteristics that confer a distinct capacity for affecting root elongation (Fig. 3). These differences might be due to common binding proteins with unique affinities for each molecule, differences in the pharmokinetics of the molecules or the interaction of each compound with discrete protein target(s). Future experiments with 2-For DCP-2-F-resistant mutants or the use of additional 2-F analogs should clarify these issues. The identification of 2-F as the active portion of compound defines a readily available starting point for biochemical or genetic approaches aimed at identifying its cellular target. We anticipate that a combination of chemical and genetic approaches will provide insight into novel components or regulatory mechanisms in auxin signal transduction. Materials and Methods Chemicals DCP-F (258247) was purchased from Chembridge (San Diego, C, US). I, 2,4-D, 1-, 2-F, 3-F and FE were purchased from Sigma-ldrich. ll other molecules were synthesized and characterized as described in Supplementary methods S1. Plant material and growth conditions Wild-type Col-, Ws, B3-GUS, DR5-GUS and tir1 rabidopsis seeds were surface sterilized using 3% bleach solution containing.15% Tween-2. fter rinsing with sterile water, seeds were either placed into liquid rabidopsis thaliana salts (TS) medium or plated on TS medium containing 1.5% agar (Gibco- BRL) (Estelle and Somerville 1987). fter stratification at 48C for 48 h, the seedlings were grown at 218C under constant light (1 mmol h 1 m 2 ) with the liquid cultures maintained at 15 r.p.m. Care was taken to ensure that manipulations were performed at the same time of day. For R experiments, liquid cultures were grown for 7 d before seedlings were treated as described in the text. R isolation and real-time PCR Total R was isolated from tissue samples using TRI Reagent (Sigma, St Louis, M, US) and an additional ethanol precipitation. Following treatment with Dase I (mersham) to Downloaded from by guest on 1 ctober 218

8 17 2-Furylacrylic acid blocks auxin responses remove trace amounts of contaminating genomic D, the R was ethanol precipitated and then converted into cd using the SuperScript II First-strand Synthesis System (Invitrogen, Carlsbad, C, US). Semi-quantitative real-time PCR was performed using iq SYBR Green Supermix (Bio-Rad, Hercules, C, US) in an icycler System according to the manufacturer s directions. Melt-curve analyses were run to ensure amplification of a single product. Primers used were obtained from Integrated D Technology Inc. (Coralville, I, US). The C T values for I5 were normalized using the constitutive Expressed protein (t5g655) described by Czechowski et al. (25), and changes in gene expression were determined using the -C T calculations (Pfaffl 21). Root elongation assays Seeds that were going to be used in experimentation were surface sterilized for 8 min in a solution of.15% Tween in bleach. Seeds were plated onto TS medium and grown vertically in a growth chamber with continuous light at 218C for 4 d. For root inhibition studies, the seedlings were transferred onto plates containing compounds at the indicated concentrations, and root lengths determined after an additional 4 d using a digital caliper. Seedling compound treatments and GUS staining (HS-XR3::GUS lines) HS-XR3::GUS seedlings were grown for 5 d vertically as described in the previous section. fter 5 d, seedlings were placed into pre-warmed (378C) TS. fter 2 h, seedlings where transferred to 228C TS containing DMS, I alone or with compounds. GUS histochemistry and fluorescent quantification were measured as previously described (ono et al. 1998). Co-immunoprecipitation assays with GST-I7/XR2 and myc-tagged TIR1 Co-immunoprecipitation assays with bacterially produced GST I7/XR2 were performed as described previously (Gray et al., 21). Briefly, 1 mm I and/or 5 mm of 2-F were added directly to a suspension containing tir1-1[myctir1] seedling extract and 4. mg of GST I7/XR2 immobilized on glutathione agarose beads (Sigma-ldrich). fter a 1 h incubation at 48C, the agarose beads were collected by a brief centrifugation, washed three times and suspended in SDS PGE sample buffer. The bound proteins were separated by SDS PGE, and interacting myc-tir1 was detected by immunoblotting with anti-myc antibodies. Supplementary material Supplementary material mentioned in the article is available to online subscribers at the journal website cknowledgments We thank Dr. Josh rmstrong for helpful discussions, Dr. Mark Estelle for sharing tir1-1, afb2-1, afb2-1/afb3-1 and afb2/afb3/afb4 seeds, and Dr. Bill Gray for tir1-1, HS::XR3T GUS and myc-tir1 seeds and the GST I7/XR2-expressing E. coli. This research was funded by the Macalester ffice for Research and Experiential Learning and grants from the Merck-merican ssociation for the dvancement of Science Undergraduate Science Research Program and the Beckman Foundation to Macalester College. References bas, L., Benjamins, R., Malenica,., Paciorek, T., Wisniewska, J., Moulinier-nzola, J.C., Sieberer, T., Friml, J. and Luschnig, C. (26) Intracellular trafficking and proteolysis of the rabidopsis auxin-efflux facilitator PI2 are involved in root gravitropism. at. Cell Biol. 8: bel, S., guyen, M.D. and Theologis,. (1995) The PS-I4/5-like family of early auxin-inducible mrs in rabidopsis thaliana. J. Mol. Biol. 251: rmstrong, J.I., Yuan, S., Dale, J.M., Tanner, V.. and Theologis,. (24) Identification of inhibitors of auxin transcriptional activation by means of chemical genetics in rabidopsis. Proc. atl cad. Sci. US 11: Ballas,., Wong, L.M., Ke, M. and Theologis,. (1995) Two auxinresponsive domains interact positively to induce expression of the early indoleacetic acid-inducible gene PS-I4/5. Proc. atl cad. Sci. US 92: Blakeslee, J.J., Peer, W.. and Murphy,.S. (25) uxin transport. Curr. pin. Plant Biol. 8: Cheng, Y., Dai, X. and Zhao, Y. (24) tcd1, a HET-repeat protein that participates in auxin signaling in rabidopsis. Plant Physiol. 135: Chuang, H.W., Zhang, W. and Gray, W.M. (24) rabidopsis ET2, an apparent ortholog of the human cullin-interacting protein CD1, is required for auxin responses mediated by the SCF(TIR1) ubiquitin ligase. Plant Cell 16: Czechowski, T., Stitt, M., ltmann, T., Udvardi, M.K. and Scheible, W.R. (25) Genome-wide identification and testing of superior reference genes for transcript normalization in rabidopsis. Plant Physiol. 139: Dai, X., Hayashi, K., ozaki, H., Cheng, Y. and Zhao, Y. (25) Genetic and chemical analyses of the action mechanisms of sirtinol in rabidopsis. Proc. atl cad. Sci. US 12: Delbarre,., Muller, P., Imhoff, V. and Guern, J. (1996) Comparison of mechanisms controlling uptake and accumulation of 2,4-dichlorophenoxy acetic acid, napthalene-1-acetic acid, and indole-3-acetic acid in suspension-cultured tobacco cells. Planta 198: Dharmasiri,., Dharmasiri, S. and Estelle, M. (25a) The F-box protein TIR1 is an auxin receptor. ature 435: Dharmasiri,., Dharmasiri, S., Weijers, D., Lechner, E., Yamada, M., Hobbie, L., Ehrismann, J.S., Jurgens, G. and Estelle, M. (25b) Plant development is regulated by a family of auxin receptor F box proteins. Dev. Cell 9: Dharmasiri,. and Estelle, M. (24) uxin signaling and regulated protein degradation. Trends Plant Sci. 9: Dreher, K. and Callis, J. (27) Ubiquitin, hormones and biotic stress in plants. nn. Bot. 99: Estelle, M.. and Somerville, C. (1987) uxin-resistant mutants of rabidopsis thaliana with an altered morphology. Mol. Gen. Genet 26: Friml, J., Benkova, E., Blilou, I., Wisniewska, J., Hamann, T., Ljung, K., Woody, S., Sandberg, G., Scheres, B., Jurgens, G. and Palme, K. (22) tpi4 mediates sink-driven auxin gradients and root patterning in rabidopsis. Cell 18: Geisler, M., Blakeslee, J.J., Bouchard, R., Lee,.R., Vincenzetti, V., et al. (25) Cellular efflux of auxin catalyzed by the rabidopsis MDR/PGP transporter tpgp1. Plant J. 44: Gray, W.M., Hellmann, H., Dharmasiri, S. and Estelle, M. (22) Role of the rabidopsis RIG-H2 protein RBX1 in RUB modification and SCF function. Plant Cell 14: Gray, W.M., Kepinski, S., Rouse, D., Leyser,. and Estelle, M. (21) uxin regulates SCF(TIR1)-dependent degradation of UX/I proteins. ature 414: Gray, W.M., stin,., Sandberg, G., Romano, C.P. and Estelle, M. (1998) High temperature promotes auxin-mediated hypocotyl elongation in rabidopsis. Proc. atl cad. Sci. US 95: Hayashi, K., Jones,.M., gino, K., Yamazoe,., ono, Y., Inoguchi, M., Kondo, H. and ozaki, H. (23) Yokonolide B, Downloaded from by guest on 1 ctober 218

9 2-Furylacrylic acid blocks auxin responses 171 a novel inhibitor of auxin action, blocks degradation of UX/I factors. J. Biol. Chem. 278: Hayashi, K., gino, K., ono, Y., Uchimiya, H. and ozaki, H. (21) Yokonolide, a new inhibitor of auxin signal transduction, from Streptomyces diastatochromogenes B59. J. ntibiot 54: Hirao, M., Posakony, J., elson, M., Hruby, H., Jung, M., Simon, J.. and Bedalov,. (23) Identification of selective inhibitors of D þ - dependent deacetylases using phenotypic screens in yeast. J. Biol. Chem. 278: Kepinski, S. and Leyser,. (25) The rabidopsis F-box protein TIR1 is an auxin receptor. ature 435: Kim, J., Harter, K. and Theologis,. (1997) Protein protein interactions among the ux/i proteins. Proc. atl cad. Sci. US 94: Kim, T., Kim, T.Y., Lee, W.G., Yim, J. and Kim, T.K. (2) Signaling pathways to the assembly of an interferon-beta enhanceosome. Chemical genetic studies with a small molecule. J. Biol. Chem. 275: Liscum, E. and Reed, J.W. (22) Genetics of ux/i and RF action in plant growth and development. Plant Mol. Biol. 49: Marchant,., Kargul, J., May, S.T., Muller, P., Delbarre,., Perrot- Rechenmann, C. and Bennett, M.J. (1999) UX1 regulates root gravitropism in rabidopsis by facilitating auxin uptake within root apical tissues. EMB J. 18: ono, Y., Chen, Q.G., vervoorde, P.J., Kohler, C. and Theologis,. (1998) age mutants of rabidopsis exhibit altered auxin-regulated gene expression. Plant Cell 1: ono, Y., oura, C., Rahman,., spuria, E.T., Hayashi, K., Tanaka,. and Uchimiya, H. (23) p-chlorophenoxyisobutyric acid impairs auxin response in rabidopsis root. Plant Physiol. 133: uellet, F., vervoorde, P.J. and Theologis,. (21) I17/XR3: biochemical insight into an auxin mutant phenotype. Plant Cell 13: Paciorek, T., Zazimalova, E., Ruthardt,., Petrasek, J., Stierhof, Y.D., Kleine-Vehn, J., Morris, D.., Emans,., Jurgens, G., Geldner,. and Friml, J. (25) uxin inhibits endocytosis and promotes its own efflux from cells. ature 435: Park, J.E., Park, J.Y., Kim, Y.S., Staswick, P.E., Jeon, J., Yun, J., Kim, S.Y., Kim, J., Lee, Y.H. and Park, C.M. (27) GH3-mediated auxin homeostasis links growth regulation with stress adaptation response in rabidopsis. J. Biol. Chem. 282: Parry, G., Delbarre,., Marchant,., Swarup, R., apier, R., Perrot- Rechenmann, C. and Bennett, M.J. (21) ovel auxin transport inhibitors phenocopy the auxin influx carrier mutation aux1. Plant J. 25: Pfaffl, M.W. (21) new mathematical model for relative quantification in real-time RT PCR. ucleic cids Res. 29: e45. Porter, W.L. and Thimann, K.V. (1965) Molecular requirements for auxin action. Phytochemistry 4: Quint, M., Ito, H., Zhang, W. and Gray, W.M. (25) Characterization of a novel temperature-sensitive allele of the CUL1/XR6 subunit of SCF ubiquitin-ligases. Plant J. 43: Rahman,., akasone,., Chhun, T., oura, C., Biswas, K.K., Uchimiya, H., Tsurumi, S., Baskin, T.I., Tanaka,. and ono, Y. (26) small acidic protein 1 (SMP1) mediates responses of the rabidopsis root to the synthetic auxin 2,4-dichlorophenoxyacetic acid. Plant J. 47: Sauer, M., Balla, J., Luschnig, C., Wisniewska, J., Reinohl, V., Friml, J. and Benkova, E. (26) Canalization of auxin flow by ux/i-rfdependent feedback regulation of PI polarity. Genes Dev. 2: Staswick, P.E., Serban, B., Rowe, M., Tiryaki, I., Maldonado, M.T., Maldonado, M.C. and Suza, W. (25) Characterization of an rabidopsis enzyme family that conjugates amino acids to indole-3- acetic acid. Plant Cell 17: Surpin, M., Rojas-Pierce, M., Carter, C., Hicks, G.R., Vasquez, J. and Raikhel,.V. (25) The power of chemical genomics to study the link between endomembrane system components and the gravitropic response. Proc. atl cad. Sci. US 12: Tan, X., Calderon-Villalobos, L.I., Sharon, M., Zheng, C., Robinson, C.V., Estelle, M. and Zheng,. (27) Mechanism of auxin perception by the TIR1 ubiquitin ligase. ature 446: Tanaka, H., Dhonukshe, P., Brewer, P.B. and Friml, J. (26) Spatiotemporal asymmetric auxin distribution: a means to coordinate plant development. Cell Mol. Life Sci. 63: Ulmasov, T., Murfett, J., Hagen, G. and Guilfoyle, T.J. (1997) ux/i proteins repress expression of reporter genes containing natural and highly active synthetic auxin response elements. Plant Cell 9: Walsh, D.P. and Chang, Y.T. (26) Chemical genetics. Chem. Rev. 16: Walsh, T.., eal, R., Merlo,.., Honma, M., Hicks, G.R., Wolff, K., Matsumura, W. and Davies, J.P. (26) Mutations in an auxin receptor homolog FB5 and in SGT1b confer resistance to synthetic picolinate auxins and not to 2,4-dichlorophenoxyacetic acid or indole-3-acetic acid in rabidopsis. Plant Physiol. 142: Woodward,.W. and Bartel, B. (25) uxin: regulation, action, and interaction. nn. Bot. 95: Yamazoe,., Hayashi, K., Kepinski, S., Leyser,. and ozaki, H. (25) Characterization of terfestatin, a new specific inhibitor for auxin signaling. Plant Physiol. 139: Zenser,., Ellsmore,., Leasure, C. and Callis, J. (21) uxin modulates the degradation rate of ux/i proteins. Proc. atl cad. Sci. US 98: (Received July 26, 27; ccepted ctober 11, 27) Downloaded from by guest on 1 ctober 218