Essay 1.1

Exploring Chemical Space in the Plant World

Natasha Raikhel, Plant Cell Biology, Department of Botany and Plant Sciences, University of California, Riverside, Riverside, CA; Glenn R. Hicks, Exelixis, South San Francisco, CA

August, 2006

Bioactive chemicals have a long and important history in helping plant scientists to unravel physiological processes. Such chemicals include inhibitors of essential plant hormone biosynthesis, transport and action, disruptors of cytoskeleton, and inhibitors of important intracellular signaling proteins such as GDP–GTP exchange proteins, among many others. Some of these chemicals occur naturally in plants, whereas others are synthetic. With the advent of advanced methods of molecular biology and genetics, the use of small molecules to uncover basic processes in plant biology received little attention in the past few decades. However, as we enter the twenty-first century, interest in bioactive chemicals as experimental tools to understand basic biological processes is reawakening (Shogren-Knaak et al. 2001; Blackwell and Zhao 2003). What has motivated biologists to revisit small molecules?

While perhaps ten million compounds are documented in the chemical literature, the potential chemical diversity of organic compounds that have a molecular mass less than 500 daltons (similar to many natural compounds) may exceed 10⁶⁰ (Dobson 2004). This presents an enormous opportunity to use diverse chemicals to probe the biology of living systems. Several important advances have allowed scientists to exploit this vast chemical diversity (also referred to as "chemical space") and rapidly discover novel bioactive chemicals. The development of combinatorial and automated techniques for synthesizing chemical structures has significantly enhanced the productivity of chemists to the point that chemical libraries containing thousands—or even millions—of potentially bioactive compounds can be synthesized (Smukste and Stockwell 2005). These advances permit the identification of chemicals that disrupt specific processes or the functions of particular proteins. Once these bioactive chemicals are identified, we can use powerful genetic screens to identify genes within the affected pathways (Blackwell and Zhao 2003). This scientific approach is referred to as "chemical genomics" (Figure 1). We are developing chemical genomics approaches to understand how the endomembrane system contributes to signal transduction and development.

Figure 1 The use of chemical genomics to uncover new pathways. (a) In a screen for bioactive chemicals, seeds of Arabidopsis are germinated in multiwell plates. Each well contains one of thousands of different compounds from a diverse chemical library. When the seeds germinate, the seedlings can be examined for phenotypes of interest, which can range from visual developmental defects such as slower root growth to intracellular phenotypes that require screening by automated microscopy. The number of different screens possible is enormous. In our laboratory, we have used this approach to identify chemicals that interfere with gravitropic bending and the normal development of vacuoles. (b) Once specific bioactive chemicals are identified from the library then genetic approaches are used to identify potential molecular sites of chemical action. Typically, seeds treated with a mutagen such as EMS are grown on media containing specific bioactive chemicals. The resulting seedlings are then scored for resistance or hypersensitivity to the chemical. For example, the vacuoles of mutant seedlings may be more sensitive than wild-type seedlings to drugs that disrupt vacuole biogenesis (see Figure 2). The mutated genes can then be identified using standard mapping and cloning methods. Identifying the gene products impacted by bioactive chemicals can lead to the identification of new biological pathways and mechanisms. (Reproduced from Blackwell and Zhao 2003, with kind permission of the authors and the publisher. © American Society of Plant Biologists.) (Click image to enlarge.)

The endomembrane system functions to segregate biochemical reactions within membrane-limited compartments and as a pathway for the transport of materials both intracellularly and across the plasma membrane via secretion or endocytosis. Components of the endomembrane system include organelles such as the endoplasmic reticulum (ER), Golgi apparatus, trans-Golgi network, pre-vacuolar compartment, vacuoles, and endosomes. In plants, there are at least two kinds of vacuoles. Lytic vacuoles are similar in function to lysosomes in yeast and mammalian cells, whereas protein storage vacuoles are specific to plants (Bassham and Raikhel 2000). Vesicles provide for the transport of cargo between the different compartments. In Arabidopsis, the isolation and characterization of genes that encode proteins that participate in vesicular trafficking and vacuolar biogenesis reveal that important developmental processes and signal transduction pathways rely on normal cargo protein movement within the cell (Surpin and Raikhel 2004). Research from our laboratory and others unequivocally demonstrate that a functional secretory system is necessary for normal development, but we do not yet understand the mechanistic details of this association. One important developmental process linked to the endomembrane system is the ability of plants to bend in response to gravity (see Chapter 19). This ensures that the shoot grows away from the direction of gravity (negative response) and is thus positioned to maximize light-harvesting, whereas the roots grow downward toward gravity (positive response) to maximize water and nutrient uptake. When plants are repositioned relative to the gravity vector, by wind or hail for example, they quickly reorient themselves. The molecular mechanisms underlying the gravitropic response are complex and mutants identified from screens for defects in gravitropic responses have been determined to contain lesions in genes encoding endomembrane system components (Surpin and Raikhel 2004). Although such classical genetic screens have been quite useful in the study of the plant endomembrane system, they have clear limitations.

The Arabidopsis genome has been fully sequenced and contains around 30,000 genes (The Arabidopsis Genome Initiative, 2000; Borevitz and Ecker 2004). About one-third of these genes are single copy; the remaining two-thirds have at least one homolog. Significantly, one-third of the total genes belong to families containing more than five members. Genes within a given family probably share similar biological functions. Thus, a mutation in a single family member may not result in a discernible phenotype that could indicate the function of the gene product. Unfortunately, it is generally not feasible to construct a mutant in which all the genes within a large family have been eliminated. Another limitation of classical genetics is gene lethality, and it is not restricted to single-copy genes. For example, loss-of-function mutants have been isolated for genes involved in endomembrane trafficking and many result in embyro or gametophyte lethality (Rojo et al. 2001; Sanderfoot et al. 2001). These issues of redundancy and lethality have limited our ability to assign functions to individual components of the plant endomembrane system, but they can be addressed using chemical genomics. A single chemical can affect all members of a protein family circumventing gene redundancy, and the severity of chemically-induced phenotypes—including lethality—can be modulated by adjusting the dosage. Moreover, chemicals can be applied at any point during development and to specific tissues, permitting the study of proteins with multiple or complex functions at different stages of development.

Using diverse chemical libraries containing thousands of low-mass chemicals and Arabidopsis genomic resources, we are working to understand the biology of the endomembrane system. In one approach, we have identified novel synthetic chemicals that affect the ability of plants to respond normally to gravity; these chemicals also disrupt the proper function of the endomembrane system allowing us to understand the link between the secretory system and gravitropism (Surpin et al. 2005). In another screen (Zouhar et al. 2004), we identified a novel chemical that interferes with the normal biogenesis of vacuoles in Arabidopsis (Figure 2a,b). There are several other examples of successful chemical screens in Arabidopsis performed in other laboratories (Zhao et al. 2003; Armstrong et al. 2004). A major challenge to any chemical genomics screen is target identification. Arabidopsis offers unique opportunities for target identification including a fully annotated genome, extensive knockout and activation-tagged mutant collections, and a collection of cloned full-length cDNAs. An abundance of resources for the mapping of genes has also greatly enhanced our ability to identify genes whose products are the targets of bioactive chemicals. Target identification typically involves screening for mutants that are resistant or hypersensitive to a specific bioactive chemical and the subsequent mapping and cloning of the corresponding gene (Figure 2c–f). The products of such genes may not necessarily interact directly with a chemical of interest; instead, they may be upstream or downstream components within a pathway. For example, it is possible to identify a transporter that is necessary for the intracellular accumulation of a specific chemical rather than a component directly in a pathway of interest; however, there are strategies to eliminate such targets if they are not desired. The essential point is that a combination of classical and chemical genomic methods can be used to overcome the disadvantages of either approach individually. An important practical benefit of using synthetic chemicals as investigative tools is that knowledge gained from the characterization of chemicals and the identification of their cognate targets and pathways in Arabidopsis can be applied directly to economically important crop species to increase yield and improve other agronomic traits.

Figure 2 Vacuole biogenesis of mutants that are hypersensitive to the novel vacuole-disrupting drug Sortin1. The hypocotyl cells of Arabidopsis seedlings expressing green fluorescent protein (GFP) fused to the tonoplast-specific protein dTIP (GFP:dTIP) are shown. GFP:dTIP allows the visualization of the vacuole membrane in living seedling stems using confocal microscopy. (a) In control wild-type seedlings grown in the absence of drug, the vacuoles outline the periphery of the cells due to the large volume of the cell occupied by this organelle. (b) However, when exposed to 57 mM Sortin1 the morphology of the vacuoles is disrupted with the appearance of invaginations, small vesicles and tonoplast-lined strands traversing the vacuoles. (c) Mutant s1h40-10 grown in the absence of Sortin1 appears to have normal vacuoles. (d) However, the mutant is much more sensitive to Sortin1 and displays defective vacuoles when grown in the presence of only 1 mM Sortin1. (e) Similarly, s1h21-2 has normal vacuoles in the absence of drug. (f) However, this mutant has aberrant vacuole biogenesis at only 11 mM Sortin1. Using modern genetics, the mutated genes leading to hypersensitivity can be identified. These genes could be previously unknown components of the endomembrane system and may provide insights into the specific mechanisms of vesicular transport and vacuole biogenesis. Bars represent 40 mm.

References

Armstrong, J. I., Yuan, S., Dale, J. M., Tanner, N. V., and Theologis, A. (2004) Identification of inhibitors of auxin transcriptional activation by means of chemical genetics in Arabidopsis. Proc. Natl. Acad. Sci. USA 101(41): 14978–14983.

Bassham, D. C., and Raikhel, N. V. (2000) Unique features of the plant vacuole sorting machinery. Curr. Opin. Cell Biol. 12: 491–495.

Blackwell, H. E., and Zhao, Y. (2003) Chemical genetic approaches in plant biology. Plant Physiol. 133(2): 448–455.

Borevitz, J. O., and Ecker, J. R. (2004) Plant genomics: The third wave. Annu. Rev. Genomics Hum. Genet. 5: 443–477.

Dobson, C. M. (2004) Chemical space and biology. Nature 432: 824–828.

Rojo, E., Gillmor, C. S., Kovaleva, V., Somerville, C. R., and Raikhel, N. V. (2001) VACUOLESS1 is an essential gene required for vacuole formation and morphogenesis in Arabidopsis. Dev. Cell 1: 303–310.

Sanderfoot, A. A., Pilgrim, M., Adam, L., and Raikhel, N. V. (2001) Disruption of individual members of Arabidopsis syntaxin gene families indicates each has essential functions. Plant Cell 13(3): 659–666.

Shogren-Knaak, M. A., Alaimo, P. J., and Shokat, K. M. (2001) Recent advances in chemical approaches to the study of biological systems. Annu. Rev. Cell Dev. Biol. 17: 405–433.

Smukste, I., and Stockwell, B. R. (2005) Advances in chemical genetics. Annu. Rev. Genomics Hum. Genet. 6: 261–286.

Surpin, M., and Raikhel, N. V. (2004) Traffic jams affect plant development and signal transduction. Nat. Rev. Mol. Cell Biol. 5(2): 100–109.

Surpin, M., Rojas-Pierce, M., Cater, C., Hicks, G. R., Vasquez, J., and Raikhel, N. V. (2005) The power of chemical genomics to study the link between endomembrane system components and the gravitropic response. Proc. Natl. Acad. Sci. USA 102(3): 14978–14983.

The Arabidopsis Genome Initiative. (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408: 796–815.

Zhao, Y., Dai, X., Blackwell, H. E., Schreiber, S. L., and Chory, J. (2003) SIR1, an upstream component in auxin signaling identified by chemical genetics. Science 301: 1107–1110.

Zouhar, J., Hicks, G. R., and Raikhel, N. V. (2004) Sorting inhibitors (Sortins): Chemical compounds to study vacuolar sorting in Arabidopsis. Proc. Natl. Acad. Sci. USA 101(25): 9497–9501.