Hormone Signalling Crosstalk in Plant Growth Regulation

Transcription

1 Current Biology 21, R365 R373, May 10, 2011 ª2011 Elsevier Ltd All rights reserved DOI /j.cub Hormone Signalling Crosstalk in Plant Growth Regulation Review Stephen Depuydt and Christian S. Hardtke* The remarkable plasticity of plant ontogeny is shaped by hormone pathways, which not only orchestrate intrinsic developmental programs, but also convey environmental inputs. Several classes of plant hormones exist, and among them auxin, brassinosteroid and gibberellin are central for the regulation of growth in general and of cell elongation in particular. Various growth phenomena can be modulated by each of the three hormones, in a sometimes synergistic fashion, suggesting physiological redundancy and/or crosstalk between the different pathways. Whether this means that they target a common and unique transcriptome module, or rather separate growth-promoting transcriptome modules, remains unclear, however. Nevertheless, while surprisingly few molecular mediators of direct crosstalk in the proper sense have been isolated, evidence is accumulating for complex cross-regulatory relations between hormone pathways at the level of transcription, as exemplified in root meristem growth. The growing number of available genome sequences from the green lineage offers first glimpses at the evolution of hormone pathways, which can aid in understanding the multiple relationships observed between these pathways in angiosperms. The available analyses suggest that auxin, gibberellin and brassinosteroid signalling arose during land plant evolution in this order, correlating with increased morphological complexity and possibly conferring increased developmental flexibility. Introduction A central feature of plant development is the largely postembryonic formation of the plant body. While in animals practically all organs are formed during embryogenesis and elaborated during the juvenile stage, plant embryogenesis results in the formation of a miniature plant. This socalled seedling possesses discrete stem cell pools in the shoot and root meristems, which form the vast majority of plant organs post-embryonically in a modular, reiterative fashion that also integrates environmental inputs. The shoot apical meristem and the lateral meristems derived from it give rise to the above-ground organs such as stems, leaves and flowers, while the root apical meristem forms the root system with its branches [1,2]. The advantage of this modular, post-embryonic mode of development is that it provides an appropriate response to environmental stimuli, which are of pivotal importance in plant development given that plants are sessile and thus have no choice but to adapt to their immediate environment. Such adaptation does not involve only physiological processes, for example to cope with given levels of soil salinity [3], but also significant morphological changes, for example the so-called shade Department of Plant Molecular Biology, University of Lausanne, Biophore Building, CH-1015 Lausanne, Switzerland. * christian.hardtke@unil.ch avoidance response that favours exaggerated elongation growth [4]. Another constraint on the development of plants compared with animals is their comparatively rigid cell walls [5], which consist mainly of cellulose and are attached to a pectin matrix that connects neighbouring cells, the socalled middle lamella. Generally, this scaffold restricts cell movements and, accordingly, the cell migrations frequently observed in animal development are extremely rare in plants. Perhaps this is why the differentiation of plant cells is generally dominated by position rather than lineage [6]. Because of the cell wall frame, growth processes have to be coordinated across neighbouring cells and morphology can only be elaborated by differentially modulating local growth rates. This mostly happens during organ growth, which relies on two basic processes cell proliferation and cell expansion. In many instances, the latter is anisotropic and thus referred to as cell elongation. A technical prerequisite for cells to elongate and hence change shape is the ability to alter cell wall extensibility [5]. This is achieved by the enzymatic rearrangement of existing cell-wall polymers and reorganization of their synthesis (not discussed in this review, but excellent summaries are available [5,7,8]). As meristematic cells proliferate and are eventually displaced towards the meristem periphery, differentiation and elongation start to define the shape and molecular identity of individual cells that ultimately form highly specialised tissues and organs. Plant body structure thus depends not only on the local rates of cell division, but, equally importantly, on the direction of cell elongation. All growth processes are under the pivotal influence of plant hormone pathways, and this review aims to summarize the current knowledge of the cross-regulatory pathway interactions that orchestrate morphological change. Multiple Hormones Are at Play Plant growth and development are controlled by both external cues and intrinsic growth regulators, such as hormones [9]. Mounting evidence suggests that environmental cues target the biosynthesis or perception of hormones, which therefore not only orchestrate intrinsic developmental programs, but also convey environmental inputs. Eight principal classes of plant hormones have been characterized: abscisic acid, auxin, brassinosteroids, cytokinins, ethylene, gibberellins, jasmonates and strigolactones. All of them have been linked to growth regulation in one way or another, sometimes in a context-specific manner [9 11]. Based on the phenotype of mutants with disrupted hormone biosynthesis or perception, only cytokinin, auxin, gibberellins and brassinosteroids are considered to be essential for growth. Among them, cytokinins regulate cell proliferation, while gibberellins promote cell elongation and auxin is involved in both processes. Moreover, brassinosteroids are essential for cell elongation, but might also have a role in cell proliferation [12,13]. Notably, all of these hormones can regulate many processes singlehandedly and independently, but cooperation and crosstalk between their signalling pathways appear to exist, as deduced from

2 Current Biology Vol 21 No 9 R366 Reproduction Internode elongation Embryo Embryogenesis Gibberellin Auxin Seed Germination Hypocotyl elongation Figure 1. Examples of auxin, gibberellin or brassinosteroid involvement in growth phenomena across the plant life cycle. Auxin signalling is essential for embryogenesis and defines the later growth axes of the plant. In the root meristem, auxin action is shaped through polar auxin transport and determines root system growth and its branching pattern. In the shoot apex, auxin accumulation at the sites of primordia is essential for lateral organ formation. Auxin activity is also observed in floral organ primordia, ovule primordia and zygotes. Gibberellins have a major influence on germination, plant growth in general (mainly via cell expansion), floral development and flowering time. s have been reported in almost all plant tissues, with highest levels found in seeds, pollen and young growing tissues. s act largely postembryonically with pronounced effects on general plant growth via cell elongation, vascular differentiation, and reproductive development. Leaf serration Reproductive growth Flowering Lateral organ formation Vegetative growth their overlapping influence on various cellular processes [9,14,15] (Figure 1). This is particularly evident for cell elongation, in part because it can be easily observed experimentally. In the following subsections, we will concentrate on the auxin, brassinosteroid and gibberellin pathways as the principal regulators of this process. The Auxin Response Pathway The cellular response to auxin is mediated by receptors, the F-box protein TRANSPORT INHIBITOR RESPONSE 1 (TIR1) and its homologs [16,17], which are integral components of SCF-type E3 ubiquitin ligases and can interact with a specific domain of transcriptional co-factors of the AUXIN/INDOLE ACETIC ACID (AUX/IAA) family [16,18,19]. Auxin can bind to TIR1-type auxin receptors and strongly enhances their affinity for AUX/IAA proteins [20], resulting in ubiquitination and proteasome-mediated degradation of the latter. AUX/ IAAs contain additional protein protein interaction domains that enable them to bridge the transcriptional co-repressor TOPLESS (TPL) and bona fide auxin response factors (ARFs) [21,22], a class of plant-specific B3-type transcription factors [23 25]. AUX/IAA degradation is the key event in auxin signalling, since it releases activating ARFs from TPL-mediated repression, thereby permitting activation of target genes [26]. Some responses to alterations in auxin conditions [27,28] are too rapid to be explained by transcriptional changes Root growth and might involve a second auxin Seedling receptor, AUXIN BINDING PROTEIN 1 (ABP1) [29]. Similar to comprehensive loss of function of the TIR1-type auxin receptors, loss of ABP1 leads to an embryonic lethal phenotype [30]. ABP1 overexpression or conditional inactivation leads to enhanced and impaired cell elongation, respectively [31,32]. Recently, it has been demonstrated that auxin binding to ABP1 at Current Biology the plasma membrane inhibits endocytosis of PIN-FORMED (PIN) auxin efflux carriers [33]. PINs are polar localized transmembrane proteins that are essential for directional cell-to-cell transport of auxin, so-called polar auxin transport (PAT) [34]. Similar to auxin signalling, PAT is essential and enables, for instance, the formation of auxin maxima in organogenesis, leading in turn to the formation of new growth axes [35,36]. The constitutive recycling of PIN proteins through endocytosis permits their dynamic intracellular reallocation in response to internal or external cues, and thereby a redistribution of auxin across tissues [37,38]. The Perception Pathway Compared with the auxin pathway, brassinosteroid perception represents a more classical signalling paradigm. Studies of exogenous brassinosteroid application, brassinosteroiddeficient and brassinosteroid-insensitive mutants have led to a detailed working model of brassinosteroid action in planta [39]. At the cell surface, the transmembrane protein BRI1 and its homologous receptors perceive brassinosteroid that binds to their extracellular domain [40 42] and thus activates the cytoplasmic serine/threonine kinase activity. Similar to animal receptor-like kinases, binding also results in ligand-induced homodimerization and heterodimerization with co-receptors [43 47]. Several signalling intermediates in a subsequent phosphorylation cascade have been identified through genetic screens, mostly because certain alleles suppress the dwarfism of the bri1

3 Special Issue R367 mutant. Other brassinosteroid-insensitive mutants with a dwarf phenotype include bin2 (brassinosteroid insensitive 2), which encodes a soluble kinase that acts downstream in the signalling cascade [48]. BIN2 negatively regulates BRASSINAZOLE RESISTANT 1 (BZR1) and bri1-ems- SUPPRESSOR 1 (BES1), the two highly similar principal transcription factor targets of brassinosteroid signaling, which regulate brassinosteroid-induced gene expression [49 51]. BIN2-mediated phosphorylation inhibits BZR1 and BES1 both by interfering with their DNA-binding activity and by stimulating their proteasome-dependent degradation [52]. BZR1 and BES1 are generally considered to be redundant transcriptional regulators that can bind so-called brassinosteroid-response elements in target gene promoters. However, the respective mutant phenotypes suggest some level of sub-functionalization, which might be contextdependent and involve the recruitment of accessory factors. The BZR1 and BES1 target genes include brassinosteroid biosynthetic genes, meaning that BIN2 initiates an immediate, negative-feedback loop to dampen excess signalling [53]. In summary, brassinosteroid perception by the receptors eventually inhibits BIN2 to increase the nuclear abundance of unphosphorylated BZR1 and BES1, thereby inducing transcriptional changes. The Gibberellin Signalling Pathway Similar to the auxin receptor, the gibberellin receptor, GIBBERELLIN INSENSITIVE DWARF 1 (GID1) is soluble. It was originally identified in a screen for insensitivity to gibberellin application in rice and has three redundantly acting Arabidopsis homologs, GID1A, B and C [54,55]. Several other gibberellin signalling components have been identified through screens in Arabidopsis. The respective mutants typically display a characteristic dwarf phenotype, which is strongest in the triple gid1abc mutants. These are severely affected in growth and development, i.e. very small and unable to reproduce [54,56]. Transcript profiles of gibberellin-deficient mutants resemble the transcriptome of gid1abc plants, suggesting that gibberellin signalling works exclusively through the GID receptors [56]. Upon binding of bioactive gibberellins, the predominantly nuclear-localized GID receptors [54,56] undergo an induced conformational change, which promotes their interaction with transcriptional repressors of the so-called DELLA family of transcription factors [57,58]. DELLAs, notably GIBBERELLIC ACID INSENSITIVE (GAI) and REPRESSOR OF ga1-3 (RGA), are the key downstream regulators in gibberellin signalling and have been identified through Arabidopsis mutants. Deletion of the so-called DELLA domain in some mutant alleles renders the respective proteins hyperactive and insensitive to gibberellin regulation, resulting in dwarf phenotypes [57,59,60]. By contrast, gai rga loss-of-function double mutants display a nearly wild-type phenotype and can suppress the dwarfism of gid1abc plants [54,56,61,62], suggesting that gibberellin signalling inactivates growth suppression by DELLA proteins. Indeed, the gibberellin GID1 complex stabilizes the interaction of DELLA proteins with the SLEEPY1 (SLY1) F-box protein, an SCF-type E3 ubiquitin ligase component [63,64] that targets DELLA proteins for degradation through the 26S proteasome pathway. Interestingly, DELLA proteins act at least in part through negative regulation of other transcription factors [65,66]. In conclusion, DELLA proteins are growth repressors that are degraded upon gibberellin perception through GIDs. Similar to brassinosteroid signalling, several studies have uncovered extensive feedback of gibberellin signalling on gibberellin biosynthesis, highlighting the importance of homeostasis. The Case For or Against a Central Growth Module In summary, the auxin, brassinosteroid and gibberellin signalling pathways ultimately target the activity of specific sets of transcription factors. Thus, their primary effect is a transcriptional reprogramming of the perceiving cells. Considering that all three hormones are involved in cell elongation, one might thus ask whether their pathways report different inputs, but ultimately target the same set of genes, at least those encoding enzymes needed for the structural rearrangements associated with cell elongation? The idea that such a central growth module could exist emerged from work on gibberellin-related mutants. For example, the essential role of auxin in root growth appears to involve a downstream effect on DELLA stability [67]. Also, stresstriggered growth restriction through the abscisic acid and ethylene pathways is likely mediated through DELLA protein stabilization [68 70]. Addressing the issue, a comparative study of microarraybased transcriptome surveys found surprisingly little overlap in the expression changes induced by exogenous application of each hormone [71]. Because a morphological growth response can be observed during the time of hormone application, this was taken as evidence that some elusive transcriptional core growth module does not exist. Since this also applied to the types of genes activated or inhibited, it appears that the regulation of distinct gene sets could lead to the same developmental response. However, one caveat of the study is that it applied rather stringent criteria to determine high-confidence targets from disparate microarray experiments. To some degree, such meta-analysis might have biased the results, since it surely identifies genes that reproducibly respond to a given hormone treatment, but also discards genes that did not respond above a certain threshold in each experiment, although their response might very well reflect a biological reality [72,73]. Common targets could also have been missed because of confounding factors: e.g. the use of not truly comparable doses and/or time points for the different treatments; tissue-specific competence to respond to a certain hormone; or status of the circadian clock, which has been shown to gate the amplitude of hormone responses [74 76]. In summary, however, standard approaches failed to identify a commonly targeted transcriptome for the three hormone pathways. Nevertheless, homologs of the highly redundant expansins, which are generally accepted to be key regulators of cell-wall extension, were regulated by each hormone, although to different degrees [71]. It would be interesting to determine whether this correlates with proteomic data, and whether different enzymatic activities that are correlated with cell elongation were induced [5]. Points of Hormone Crosstalk In summary, while many hormone pathway interactions have been described, it remains unclear whether these are hierarchically ordered and/or eventually channelled towards a set of transcriptional key targets. Given the physiological evidence for hormone interactions, surprisingly few molecular mediators of crosstalk in the proper sense [77,78] have been isolated. Regulation of common transcriptional targets

4 Current Biology Vol 21 No 9 R368 was generally considered a likely point of crosstalk, but the analyses described above have invalidated this idea to some degree. One problem associated with identifying crosstalk components might be that the actual existence of crosstalk has sometimes less support than it appears. For instance, whether reported changes in biosynthesis or catabolism of one hormone in response to stimulation by another truly reflect an intrinsic homeostatic network often remains unclear because of a mixing of correlations and causalities. This is problematic if a generic morphological change resulting from manipulation of the activity of one hormone necessarily results in seemingly altered activity of another, which could be aggravated by a lack of temporal and cellular resolution in the analyses of hormone responses. For example, if hormone A is involved in the differentiation of cell type X, and hormone B in the proliferation of the respective undifferentiated precursor cells Y, then a loss of activity in hormone B could lead to lower abundance of cell type X and possibly an associated quantitative reduction in observed overall activity of hormone A at the tissue level. Because hormone signalling cascades typically involve feedback regulation through adjustment of their own biosynthesis [9 11], this might thus be mis-interpreted as regulation of one pathway by the other at the level of transcription. Clearly, technically more advanced, mechanistic approaches are needed to substantiate the idea that the activities of hormone pathways can depend on each other. A recent example is the in vivo elucidation of BZR1 targets, which has revealed that BZR1 binds to the promoters of numerous, often unexpected targets, including genes involved in the metabolism of other hormones [79]. Equally comprehensive datasets for the transcription factors targeted by the auxin and gibberellin signalling modules would be needed to elucidate the degree to which the activities of the three hormones overlap. Combined with cell-sorting strategies, this could even reveal context-specific common targets of their pathways [80,81]. Such data might also ultimately determine whether a common transcriptional growth module, represented by common target genes, exists or not. The absence of a clearly identifiable transcriptional growth module might also indicate that a dwarf is not a dwarf is not a dwarf, i.e. that different hormones might interfere with growth at different stages of cell ontogeny. Nevertheless, instances in which multiple hormones act in the same process in time and space have been identified. This is, for instance, illustrated by the multiple levels of feedback and the complex interactions between different hormone pathways during root meristem growth in young Arabidopsis seedlings. A genuine point of crosstalk in this process is the AUX/IAA gene SHORT HYPOCOTYL 2 (SHY2), whose expression in the root meristem transition zone is controlled by B-type Arabidopsis response regulator (ARR) transcription factors (ARR1 and ARR12) [82,83], which are the end points of cytokinin signalling. In this manner, cytokinin exerts an indirect control on PAT, since PIN gene expression is inhibited by SHY2 activity due to SHY2 repression of the transcription activation potential of ARFs, such as MONPTEROS (MP) [83]. The resulting PAT downregulation favours the differentiation and elongation of cells and limits meristem growth. However, this effect only manifests several days after germination, because ARR1 expression is in turn suppressed by high gibberellin levels during early stages. This destabilizes the DELLA factor RGA, which, directly or indirectly, is required for high levels of ARR1 transcription [84]. Moreover, the process is refined by spatio-temporally regulated antagonism between SHY2 and a positive regulator of auxin signalling, BREVIS RADIX (BRX), which is necessary for the expression of both SHY2 and PIN3, the dominant PIN gene during meristem growth [85]. This network is further complicated by feedbacks on hormone biosynthesis. For instance, SHY2 action has a negative impact on the expression of cytokinin biosynthetic genes [83,86], thus dampening its own activity, while BRX action impinges on brassinosteroid levels [85,87,88], which could serve to propagate its non-cell-autonomous effect. In summary, it appears that auxin as well as brassinosteroid and gibberellin are involved in regulating the transition of root meristem cells from division to elongation, but in a highly intertwined rather than hierarchical manner (Figure 2). This example also demonstrates that a clear demarcation of the effects on cell division and elongation is not always possible, in particular during organ formation. Integration of Hormone Signalling with Parallel Physiological Conditions One assumption behind the idea of a central growth module for cell elongation is that such a morphological change can only be triggered in one particular way. However, although cell-wall loosening and/or remodelling is a prerequisite for cell elongation, it could be achieved in different ways by different hormones [5]. Moreover, it is conceivable that different, parallel physiological conditions could limit cell elongation. For instance, increasing cell-wall extensibility might not be sufficient to result in morphological change if other factors, e.g. insufficient turgor pressure, are limiting, and vice versa. In this respect, an interesting example is provided by recent reports that reactive oxygen species (ROS) limit root growth [89,90]. These studies were substantiated by the discovery of a regulatory factor, UPBEAT 1 (UBP1), which determines the ratio of particular ROS in the root transition zone [91]. In loss-of-function upb1 mutants, this balance is disturbed, leading to delayed cell differentiation and elongation, and thus to a larger meristem and increased root growth. This phenotype is likely related to the role of ROS in lignification and cross-linking of primary cell walls. The meristem-size increase in ubp1 mutants is quantitatively comparable to the effect of eliminating cytokinin control of SHY2 in the hormonal crosstalk described above. Nevertheless, the UBP1 pathway might act independently of the described cytokinin auxin crosstalk [91]. An interesting twist with respect to these supposedly separate root growth modules is that the gibberellin pathway has been implicated in regulating ROS levels [70], which could constitute a link between the two modules and explain the dominant role of stabilized DELLA proteins with respect to root growth. The Multiplicity of Hormone Pathways an Emerging Evolutionary Perspective As to why plants have evolved multiple hormone response pathways to regulate the same process remains largely a matter of speculation because of a sheer lack of data. Clearly, however, not all hormone pathways are equally important for development per se, as judged from the Arabidopsis mutant phenotypes of their signalling cascade endpoints. For instance, while auxin signalling is essential for embryogenesis [23], the phenotype of higher-order della loss-of-function mutants suggests that the gibberellin

5 Special Issue R369 BRX SHY2 Meristematic zone Differentiation / elongation zone MP/ARF Transition zone BRX PIN3 SHY2 PAT ARR1 ARR12 3 dag 3 dag 5 dag Differentiation BRX SHY2 ARR1 RGA Growth of the meristematic zone MP/ARF 5 dag BRX PIN3 SHY2 PAT Differentiation ARR1 ARR1 ARR12 RGA Stem cell niche Root cap/columella Current Biology Figure 2. Schematic working model of the spatio-temporal regulatory interactions between hormone pathways in root meristem growth, based on [82 85]. A complex network of regulatory interactions occurring across the transition zone balances cell division and differentiation/elongation. At early stages (3 days after germination, dag), the physiological conditions, notably high gibberellin levels required for germination, favour PIN gene expression and thereby polar auxin transport (PAT) and meristem growth. This process eventually comes to a halt around 5 dag as feedback regulation of hormone biosynthesis (not indicated) switches the regulatory interactions in favour of suppressing PAT and thereby promoting cell differentiation. See text for more details. pathway is dispensable for plant development [69]. The brassinosteroid pathway has a somewhat intermediate position, as brassinosteroid signalling has at most a minor role during embryogenesis [88,92]. The relative importance of the different hormone pathways during development might, to some degree, reflect their evolutionary history. The availability of an increasing number of genome sequences from the plant lineage has already shed some light on this aspect (Figure 3). In this context, much attention has been given to the study of the bryophyte Physcomitrella patens and the lycophyte Selaginella moellendorffii, which are two commonly used models for lower plant genera. Bryophytes are considered the ancestors of vascular plants and thus quite distant from modern flowering plants, while lycophytes are among the oldest lineages of vascular plants that gave rise to flowering plant species [93,94]. Evolution of the Auxin Response First analyses have already framed the evolution of auxin signalling, maybe aided by the fact that some of its components are plant-specific and can be unequivocally identified. For instance, auxin perception through TIR1 and subsequent signalling involving AUX/IAAs and ARFs is apparently absent from various green algae [95]. However, these algae exhibit a physiological response towards auxin, which is hypothesized to be channelled through ABP1 homologs [96]. Thus, ABP1 appears to precede TIR1 as the more ancient auxin receptor. Later in land plant evolution, and already in Physcomitrella, true AUX/IAA, TIR1 and ARF orthologs are present [96 98]. The notion that the basic auxin response machinery has been recruited from the moss lineage onwards is confirmed by the observation that there are no TPL homologs in algae, while their presence has been established in mosses and lycophytes [96]. Stepwise Acquisition of the Gibberellin Pathway Similar to auxin signalling, components of gibberellin perception, i.e. GID1 and DELLA homologs, are already present in Physcomitrella patens [99,100]. Interestingly however, the Physcomitrella GID1 homologs are unable to complement gibberellin receptor mutants of rice, indicating that their gibberellin dependence arose later in evolution [100]. Consistent with this observation, bioactive gibberellins appear to be absent from mosses [101], although gibberellin precursors might be physiologically relevant [102]. By contrast, GID1 homologs from Selaginella were fully able to complement rice gibberellin receptor mutants [100]. With respect to DELLA proteins, both Physcomitrella and

6 Current Biology Vol 21 No 9 R370 Gibberellin perception and signalling Auxin perception and signalling perception and signalling Dicotyledons Monocotyledons BRI1 receptor, active (+ID & LRR22) Vascular land plants Lycophytes Bryophytes non-vascular land plants Liverworts Gymnosperms Basal angiosperms Ferns Hornworts Mosses Growth Gibberellin response ARF + AUX/IAA Auxin binding TIR1 response BRI1 receptor homologs, inactive (-ID & LRR22) ( Charophyta algae Zygnematales Klebsomordiales Chlorokybales Choleochaetales Charales ( Gibberellin GID ABP1 Chlorophyta Ulvophyceae Chlorophyceae Prassinophyceae Auxin response Rhodophyta glaucophyta Trebouxihyceae DELLA GID precursor Auxin Current Biology Figure 3. Overview of the evolution of the auxin, gibberellin and brassinosteroid pathways on the basis of the available analyses. The occurrence of functional hormone receptor pathways is depicted according to their appearance during land plant evolution. Hormones are represented by circles, while the plant s ability to actively respond to a particular hormone is shown as a star. ID and LRR22 refer to the island domain and the leucine-rich repeat 22, respectively, of BRI1, which are essential for brassinosteroid binding. Selaginella homologs could replace their Arabidopsis counterparts [99,100]. However, although Selaginella seems to possess a functional gibberellin biosynthesis and response pathway, gibberellin application had no measurable effect on physiological responses in lycophytes [99,100]. Moreover, the primary function of early DELLA proteins was probably not growth inhibition [99]. Thus, it appears that the DELLA GID system only became functionally dependent on gibberellin during early vascular plant evolution, after divergence from the bryophytes, and acquired its growth inhibition capacity only later, after divergence from the lycophytes. The sometimes observed gibberellin-independent interaction between GIDs and DELLAs [63,103] might thus represent an evolutionary remnant. Evolutionary Footprints of the Pathway Compared with the auxin and gibberellin pathways, little is known about the evolution of brassinosteroid signalling components. To some degree, this might be related to the fact that they fall into generic classes of more ancient eukaryotic signalling molecules, which can also be found in animals [38,103]. Presence of the receptor could be taken as the central piece of evidence for the existence of a hormone pathway, and sequence homology searches across several available plant, moss and algae genomes indeed easily identify various BRI1 homologs, with reasonably well scoring alignments from Physcomitrella onwards. However, given the abundance of receptor-like protein kinases in the eukaroytes, it is hard to decide which hits are functionally meaningful. The same problem arises at the level of the biosynthetic genes, which mostly fall into the class of the equally abundant P450 enzymes, for which substrates and products can only be reliably determined experimentally. However, a wealth of structural information and its relation to function has been gathered for BRI1. The BRI1 brassinosteroid-binding domain has been well characterized and comprises the so-called island domain and the leucine-rich repeat 22 [104,105]. Within these domains, various aminoacid residues required for brassinosteroid binding have been mapped. We took advantage of this knowledge in homology searches by setting the conservation of the critical amino acids as a primary criterion. This revealed significant sequence homology hits only in flowering, vascular plants. Thus, similar to the auxin and gibberellin perception machineries, it appears likely that fully functional brassinosteroid signalling components are only present in vascular plants.

7 Special Issue R371 In summary, the available analyses suggest that the auxin signalling pathway is most ancient, followed by the gibberellin and then the brassinosteroid pathway. Increasing Hormonal Control Coincides with Increasing Complexity Notably, cellular anisotropy is already observed in some mosses, for instance in the caulonema cells of Physcomitrella patens. The transition to caulonema formation and thus anisotropic cell growth is controlled, at least in part, by auxin, which possibly represents a co-adoption of the auxin pathway for cell elongation on top of its more essential role in rhizoid formation [98]. This finding is also consistent with the presence of expansins in Physcomitrella, some of which are auxin inducible [106]. Thus, it appears that control of anisotropic growth by the gibberellin and brassinosteroid pathways is a later acquired feature of plant development. This is consistent with the idea that the latter pathways might have evolved to permit a more flexible development and a higher level of organization. In this context, the recruitment of the DELLA factors as growth repressors and their control by gibberellin must have been of particular evolutionary advantage. This is possibly related to the adoption of the gibberellin pathway as a major conveyor of environmental signals [68,69], adding a novel layer of control to maximize fitness. Among the various traits controlled by gibberellins, the optimal timing of germination might possibly have constituted a powerful initial selection pressure [107]. Conclusions The acquisition of multiple interacting hormone pathways in growth control has very likely dramatically widened the developmental spectrum of plants, including morphological variation, both at the intra- and inter-specific level [108,109]. A thorough investigation of the natural variation in hormone responses and interactions, combined with an understanding of the fitness relevance of morphological traits, might thus reveal why growth processes are frequently controlled by multiple hormones. Acknowledgments The hormone-related studies in our lab are supported by Swiss National Science Foundation grant 31003A_ We also would like to acknowledge support by a Marie-Curie postdoctoral fellowship to S.D. References 1. Barton, M.K. (2010). Twenty years on: the inner workings of the shoot apical meristem, a developmental dynamo. Dev. Biol. 341, Osmont, K.S., Sibout, R., and Hardtke, C.S. (2007). Hidden branches: developments in root system architecture. Annu. Rev. Plant. Biol. 58, Baxter, I., Brazelton, J.N., Yu, D., Huang, Y.S., Lahner, B., Yakubova, E., Li, Y., Bergelson, J., Borevitz, J.O., Nordborg, M., et al. (2010). A coastal cline in sodium accumulation in Arabidopsis thaliana is driven by natural variation of the sodium transporter AtHKT1;1. PLoS Genet. 6, e Kami, C., Lorrain, S., Hornitschek, P., and Fankhauser, C. (2010). Lightregulated plant growth and development. Curr. Top. Dev. Biol. 91, Sanchez-Rodriguez, C., Rubio-Somoza, I., Sibout, R., and Persson, S. (2010). Phytohormones and the cell wall in Arabidopsis during seedling growth. Trends Plant Sci. 15, van den Berg, C., Willemsen, V., Hage, W., Weisbeek, P., and Scheres, B. (1995). Cell fate in the Arabidopsis root meristem determined by directional signalling. Nature 378, Taylor, N.G. (2008). Cellulose biosynthesis and deposition in higher plants. New Phytol. 178, Darley, C.P., Forrester, A.M., and McQueen-Mason, S.J. (2001). The molecular basis of plant cell wall extension. Plant Mol. Biol. 47, Santner, A., Calderon-Villalobos, L.I., and Estelle, M. (2009). Plant hormones are versatile chemical regulators of plant growth. Nat. Chem. Biol. 5, Wolters, H., and Jurgens, G. (2009). Survival of the flexible: hormonal growth control and adaptation in plant development. Nat. Rev. Genet. 10, Santner, A., and Estelle, M. (2009). Recent advances and emerging trends in plant hormone signalling. Nature 459, Hardtke, C.S., Dorcey, E., Osmont, K.S., and Sibout, R. (2007). Phytohormone collaboration: zooming in on auxin-brassinosteroid interactions. Trends Cell Biol. 17, Nakaya, M., Tsukaya, H., Murakami, N., and Kato, M. (2002). s control the proliferation of leaf cells of Arabidopsis thaliana. Plant Cell Physiol. 43, Nemhauser, J.L., Mockler, T.C., and Chory, J. (2004). Interdependency of brassinosteroid and auxin signaling in Arabidopsis. PLoS Biol. 2, E Katsumi, M. (1985). Interaction of a with Iaa and Ga3 in the elongation of cucumber hypocotyl sections. Plant Cell Physiol. 26, Dharmasiri, N., Dharmasiri, S., and Estelle, M. (2005). The F-box protein TIR1 is an auxin receptor. Nature 435, Kepinski, S., and Leyser, O. (2005). The Arabidopsis F-box protein TIR1 is an auxin receptor. Nature 435, Gray, W.M., Kepinski, S., Rouse, D., Leyser, O., and Estelle, M. (2001). Auxin regulates SCF(TIR1)-dependent degradation of AUX/IAA proteins. Nature 414, Kepinski, S., and Leyser, O. (2004). Auxin-induced SCFTIR1-Aux/IAA interaction involves stable modification of the SCFTIR1 complex. Proc. Natl. Acad. Sci. USA 101, Tan, X., Calderon-Villalobos, L.I., Sharon, M., Zheng, C., Robinson, C.V., Estelle, M., and Zheng, N. (2007). Mechanism of auxin perception by the TIR1 ubiquitin ligase. Nature 446, Szemenyei, H., Hannon, M., and Long, J.A. (2008). TOPLESS mediates auxin-dependent transcriptional repression during Arabidopsis embryogenesis. Science 319, Tiwari, S.B., Hagen, G., and Guilfoyle, T. (2003). The roles of auxin response factor domains in auxin-responsive transcription. Plant Cell 15, Hardtke, C.S., and Berleth, T. (1998). The Arabidopsis gene MONOPTEROS encodes a transcription factor mediating embryo axis formation and vascular development. EMBO J. 17, Ulmasov, T., Hagen, G., and Guilfoyle, T.J. (1997). ARF1, a transcription factor that binds to auxin response elements. Science 276, Ulmasov, T., Hagen, G., and Guilfoyle, T.J. (1999). Activation and repression of transcription by auxin-response factors. Proc. Natl. Acad. Sci. USA 96, Guilfoyle, T.J., and Hagen, G. (2007). Auxin response factors. Curr. Opin. Plant Biol. 10, Shishova, M., and Lindberg, S. (2004). Auxin induces an increase of Ca2+ concentration in the cytosol of wheat leaf protoplasts. J. Plant Physiol. 161, Senn, A.P., and Goldsmith, M.H. (1988). Regulation of electrogenic proton pumping by auxin and fusicoccin as related to the growth of Avena coleoptiles. Plant Physiol. 88, Badescu, G.O., and Napier, R.M. (2006). Receptors for auxin: will it all end in TIRs? Trends Plant Sci. 11, Chen, J.G., Ullah, H., Young, J.C., Sussman, M.R., and Jones, A.M. (2001). ABP1 is required for organized cell elongation and division in Arabidopsis embryogenesis. Genes Dev. 15, Chen, J.G., Shimomura, S., Sitbon, F., Sandberg, G., and Jones, A.M. (2001). The role of auxin-binding protein 1 in the expansion of tobacco leaf cells. Plant J. 28, Jones, A.M., Im, K.H., Savka, M.A., Wu, M.J., DeWitt, N.G., Shillito, R., and Binns, A.N. (1998). Auxin-dependent cell expansion mediated by overexpressed auxin-binding protein 1. Science 282, Robert, S., Kleine-Vehn, J., Barbez, E., Sauer, M., Paciorek, T., Baster, P., Vanneste, S., Zhang, J., Simon, S., Covanova, M., et al. (2010). ABP1 mediates auxin inhibition of clathrin-dependent endocytosis in Arabidopsis. Cell 143, Leyser, O. (2005). Auxin distribution and plant pattern formation: how many angels can dance on the point of PIN? Cell 121, Benkova, E., Michniewicz, M., Sauer, M., Teichmann, T., Seifertova, D., Jurgens, G., and Friml, J. (2003). Local, efflux-dependent auxin gradients as a common module for plant organ formation. Cell 115, Blilou, I., Xu, J., Wildwater, M., Willemsen, V., Paponov, I., Friml, J., Heidstra, R., Aida, M., Palme, K., and Scheres, B. (2005). The PIN auxin efflux facilitator network controls growth and patterning in Arabidopsis roots. Nature 433, Dhonukshe, P., Aniento, F., Hwang, I., Robinson, D.G., Mravec, J., Stierhof, Y.D., and Friml, J. (2007). Clathrin-mediated constitutive endocytosis of PIN auxin efflux carriers in Arabidopsis. Curr. Biol. 17,

8 Current Biology Vol 21 No 9 R Kleine-Vehn, J., Dhonukshe, P., Sauer, M., Brewer, P.B., Wisniewska, J., Paciorek, T., Benkova, E., and Friml, J. (2008). ARF GEF-dependent transcytosis and polar delivery of PIN auxin carriers in Arabidopsis. Curr. Biol. 18, Kim, T.W., and Wang, Z.Y. (2010). signal transduction from receptor Kinases to transcription factors. Annu. Rev. Plant Biol. 61, Clouse, S.D., Langford, M., and McMorris, T.C. (1996). A brassinosteroidinsensitive mutant in Arabidopsis thaliana exhibits multiple defects in growth and development. Plant Physiol. 111, He, Z., Wang, Z.Y., Li, J., Zhu, Q., Lamb, C., Ronald, P., and Chory, J. (2000). Perception of brassinosteroids by the extracellular domain of the receptor kinase BRI1. Science 288, Li, J., and Chory, J. (1997). A putative leucine-rich repeat receptor kinase involved in brassinosteroid signal transduction. Cell 90, Albrecht, C., Russinova, E., Kemmerling, B., Kwaaitaal, M., and de Vries, S.C. (2008). Arabidopsis SOMATIC EMBRYOGENESIS RECEPTOR KINASE proteins serve brassinosteroid-dependent and -independent signaling pathways. Plant Physiol. 148, Karlova, R., Boeren, S., Russinova, E., Aker, J., Vervoort, J., and de Vries, S. (2006). The Arabidopsis SOMATIC EMBRYOGENESIS RECEPTOR-LIKE KINASE1 protein complex includes BRASSINOSTEROID-INSENSITIVE1. Plant Cell 18, Nam, K.H., and Li, J. (2002). BRI1/BAK1, a receptor kinase pair mediating brassinosteroid signaling. Cell 110, Russinova, E., Borst, J.W., Kwaaitaal, M., Cano-Delgado, A., Yin, Y., Chory, J., and de Vries, S.C. (2004). Heterodimerization and endocytosis of Arabidopsis brassinosteroid receptors BRI1 and AtSERK3 (BAK1). Plant Cell 16, Wang, X., Li, X., Meisenhelder, J., Hunter, T., Yoshida, S., Asami, T., and Chory, J. (2005). Autoregulation and homodimerization are involved in the activation of the plant steroid receptor BRI1. Dev. Cell 8, Li, J., and Nam, K.H. (2002). Regulation of brassinosteroid signaling by a GSK3/SHAGGY-like kinase. Science 295, He, J.X., Gendron, J.M., Yang, Y., Li, J., and Wang, Z.Y. (2002). The GSK3- like kinase BIN2 phosphorylates and destabilizes BZR1, a positive regulator of the brassinosteroid signaling pathway in Arabidopsis. Proc. Natl. Acad. Sci. USA 99, Wang, Z.Y., Nakano, T., Gendron, J., He, J., Chen, M., Vafeados, D., Yang, Y., Fujioka, S., Yoshida, S., Asami, T., et al. (2002). Nuclear-localized BZR1 mediates brassinosteroid-induced growth and feedback suppression of brassinosteroid biosynthesis. Dev. Cell 2, Yin, Y., Vafeados, D., Tao, Y., Yoshida, S., Asami, T., and Chory, J. (2005). A new class of transcription factors mediates brassinosteroid-regulated gene expression in Arabidopsis. Cell 120, Vert, G., and Chory, J. (2006). Downstream nuclear events in brassinosteroid signalling. Nature 441, He, J.X., Gendron, J.M., Sun, Y., Gampala, S.S., Gendron, N., Sun, C.Q., and Wang, Z.Y. (2005). BZR1 is a transcriptional repressor with dual roles in brassinosteroid homeostasis and growth responses. Science 307, Griffiths, J., Murase, K., Rieu, I., Zentella, R., Zhang, Z.L., Powers, S.J., Gong, F., Phillips, A.L., Hedden, P., Sun, T.P., et al. (2006). Genetic characterization and functional analysis of the GID1 gibberellin receptors in Arabidopsis. Plant Cell 18, Ueguchi-Tanaka, M., Ashikari, M., Nakajima, M., Itoh, H., Katoh, E., Kobayashi, M., Chow, T.Y., Hsing, Y.I., Kitano, H., Yamaguchi, I., et al. (2005). GIBBERELLIN INSENSITIVE DWARF1 encodes a soluble receptor for gibberellin. Nature 437, Willige, B.C., Ghosh, S., Nill, C., Zourelidou, M., Dohmann, E.M., Maier, A., and Schwechheimer, C. (2007). The DELLA domain of GA INSENSITIVE mediates the interaction with the GA INSENSITIVE DWARF1A gibberellin receptor of Arabidopsis. Plant Cell 19, Peng, J., Carol, P., Richards, D.E., King, K.E., Cowling, R.J., Murphy, G.P., and Harberd, N.P. (1997). The Arabidopsis GAI gene defines a signaling pathway that negatively regulates gibberellin responses. Genes Dev. 11, Silverstone, A.L., Jung, H.S., Dill, A., Kawaide, H., Kamiya, Y., and Sun, T.P. (2001). Repressing a repressor: gibberellin-induced rapid reduction of the RGA protein in Arabidopsis. Plant Cell 13, Peng, J., Richards, D.E., Hartley, N.M., Murphy, G.P., Devos, K.M., Flintham, J.E., Beales, J., Fish, L.J., Worland, A.J., Pelica, F., et al. (1999). Green revolution genes encode mutant gibberellin response modulators. Nature 400, Silverstone, A.L., Ciampaglio, C.N., and Sun, T. (1998). The Arabidopsis RGA gene encodes a transcriptional regulator repressing the gibberellin signal transduction pathway. Plant Cell 10, Iuchi, S., Suzuki, H., Kim, Y.C., Iuchi, A., Kuromori, T., Ueguchi-Tanaka, M., Asami, T., Yamaguchi, I., Matsuoka, M., Kobayashi, M., et al. (2007). Multiple loss-of-function of Arabidopsis gibberellin receptor AtGID1s completely shuts down a gibberellin signal. Plant J. 50, Nakajima, M., Shimada, A., Takashi, Y., Kim, Y.C., Park, S.H., Ueguchi- Tanaka, M., Suzuki, H., Katoh, E., Iuchi, S., Kobayashi, M., et al. (2006). Identification and characterization of Arabidopsis gibberellin receptors. Plant J. 46, Dill, A., Thomas, S.G., Hu, J., Steber, C.M., and Sun, T.P. (2004). The Arabidopsis F-box protein SLEEPY1 targets gibberellin signaling repressors for gibberellin-induced degradation. Plant Cell 16, McGinnis, K.M., Thomas, S.G., Soule, J.D., Strader, L.C., Zale, J.M., Sun, T.P., and Steber, C.M. (2003). The Arabidopsis SLEEPY1 gene encodes a putative F-box subunit of an SCF E3 ubiquitin ligase. Plant Cell 15, Feng, S., Martinez, C., Gusmaroli, G., Wang, Y., Zhou, J., Wang, F., Chen, L., Yu, L., Iglesias-Pedraz, J.M., Kircher, S., et al. (2008). Coordinated regulation of Arabidopsis thaliana development by light and gibberellins. Nature 451, de Lucas, M., Daviere, J.M., Rodriguez-Falcon, M., Pontin, M., Iglesias- Pedraz, J.M., Lorrain, S., Fankhauser, C., Blazquez, M.A., Titarenko, E., and Prat, S. (2008). A molecular framework for light and gibberellin control of cell elongation. Nature 451, Fu, X., and Harberd, N.P. (2003). Auxin promotes Arabidopsis root growth by modulating gibberellin response. Nature 421, Achard, P., Baghour, M., Chapple, A., Hedden, P., Van Der Straeten, D., Genschik, P., Moritz, T., and Harberd, N.P. (2007). The plant stress hormone ethylene controls floral transition via DELLA-dependent regulation of floral meristem-identity genes. Proc. Natl. Acad. Sci. USA 104, Achard, P., Cheng, H., De Grauwe, L., Decat, J., Schoutteten, H., Moritz, T., Van Der Straeten, D., Peng, J., and Harberd, N.P. (2006). Integration of plant responses to environmentally activated phytohormonal signals. Science 311, Achard, P., Renou, J.P., Berthome, R., Harberd, N.P., and Genschik, P. (2008). Plant DELLAs restrain growth and promote survival of adversity by reducing the levels of reactive oxygen species. Curr. Biol. 18, Nemhauser, J.L., Hong, F., and Chory, J. (2006). Different plant hormones regulate similar processes through largely nonoverlapping transcriptional responses. Cell 126, Marjoram, P., and Tavare, S. (2006). Modern computational approaches for analysing molecular genetic variation data. Nat. Rev. Genet. 7, Rockman, M.V., and Kruglyak, L. (2006). Genetics of global gene expression. Nat. Rev. Genet. 7, Covington, M.F., Maloof, J.N., Straume, M., Kay, S.A., and Harmer, S.L. (2008). Global transcriptome analysis reveals circadian regulation of key pathways in plant growth and development. Genome Biol. 9, R Covington, M.F., and Harmer, S.L. (2007). The circadian clock regulates auxin signaling and responses in Arabidopsis. PLoS Biol. 5, e Nozue, K., Covington, M.F., Duek, P.D., Lorrain, S., Fankhauser, C., Harmer, S.L., and Maloof, J.N. (2007). Rhythmic growth explained by coincidence between internal and external cues. Nature 448, Hardtke, C.S. (2007). Transcriptional auxin-brassinosteroid crosstalk: who s talking? Bioessays 29, Mundy, J., Nielsen, H.B., and Brodersen, P. (2006). Crosstalk. Trends Plant Sci. 11, Sun, Y., Fan, X.Y., Cao, D.M., Tang, W., He, K., Zhu, J.Y., He, J.X., Bai, M.Y., Zhu, S., Oh, E., et al. (2010). Integration of brassinosteroid signal transduction with the transcription network for plant growth regulation in Arabidopsis. Dev. Cell 19, Birnbaum, K., Shasha, D.E., Wang, J.Y., Jung, J.W., Lambert, G.M., Galbraith, D.W., and Benfey, P.N. (2003). A gene expression map of the Arabidopsis root. Science 302, Ubeda-Tomas, S., Federici, F., Casimiro, I., Beemster, G.T., Bhalerao, R., Swarup, R., Doerner, P., Haseloff, J., and Bennett, M.J. (2009). Gibberellin signaling in the endodermis controls Arabidopsis root meristem size. Curr. Biol. 19, Dello Ioio, R., Linhares, F.S., Scacchi, E., Casamitjana-Martinez, E., Heidstra, R., Costantino, P., and Sabatini, S. (2007). Cytokinins determine Arabidopsis root-meristem size by controlling cell differentiation. Curr. Biol. 17, Dello Ioio, R., Nakamura, K., Moubayidin, L., Perilli, S., Taniguchi, M., Morita, M.T., Aoyama, T., Costantino, P., and Sabatini, S. (2008). A genetic framework for the control of cell division and differentiation in the root meristem. Science 322, Moubayidin, L., Perilli, S., Dello Ioio, R., Di Mambro, R., Costantino, P., and Sabatini, S. (2010). The rate of cell differentiation controls the Arabidopsis root meristem growth phase. Curr. Biol. 20, Scacchi, E., Salinas, P., Gujas, B., Santuari, L., Krogan, N., Ragni, L., Berleth, T., and Hardtke, C.S. (2010). Spatio-temporal sequence of crossregulatory events in root meristem growth. Proc. Natl. Acad. Sci. USA 107, Jones, B., Gunneras, S.A., Petersson, S.V., Tarkowski, P., Graham, N., May, S., Dolezal, K., Sandberg, G., and Ljung, K. (2010). Cytokinin regulation of auxin synthesis in Arabidopsis involves a homeostatic feedback loop regulated via auxin and cytokinin signal transduction. Plant Cell 22,