Plant-Microbe Interactions: Chemical Diversity in Plant Defense

See allHide authors and affiliations

Science  08 May 2009:
Vol. 324, Issue 5928, pp. 746-748
DOI: 10.1126/science.1171661


The chemical diversity within the plant kingdom is likely to be a consequence of niche colonization and adaptive evolution. Plant-derived natural products have important functions in defense. They also have broader ecological roles and may in addition participate in plant growth and development. Recent data suggest that some antimicrobial phytochemicals may not serve simply as chemical barriers but could also have functions in defense-related signaling processes. It is important, therefore, that we should not to be too reductionist in our thinking when endeavoring to understand the forces and mechanisms that drive chemical diversification in plants.

Collectively, plants are a tremendous resource of structurally diverse metabolites. Examples of the structures of some of these are shown in Fig. 1. It is widely accepted that these compounds have important functions in influencing interactions between plants and other organisms. Deciphering the chemical signaling processes that mediate these interactions represents a substantial challenge for plant science.

Fig. 1

Examples of phytochemicals with roles in plant-microbe interactions.

The importance of synthesis and accumulation of antimicrobial metabolites for plant defense has intrigued researchers for the best part of a century (1). Antimicrobial compounds can be produced as part of normal plant growth and development and are usually stored in specialized organs or tissues (e.g., trichomes, oil glands, or epidermal cell layers). These constitutive or preformed antimicrobial chemicals are sometimes also referred to as phytoanticipins (2). In addition, antimicrobial compounds can be synthesized de novo in response to microbial attack by transcriptional activation of genes for biosynthetic pathways. Such compounds are known as phytoalexins.

Our understanding of the role of phytochemicals in plant defense is still incomplete. Knowledge and interpretation of chemical diversity in plants depends very much on the ability to detect, analyze, and measure compounds, as well as their metabolic precursors and derivatives in plant tissues. Despite recent progress in the development of higher-resolution multidimensional separation/detection systems, many compounds are likely to be present in trace amounts that are below the current levels of detection, at least in wild-type plants (3). A shortage of reference standards makes comprehensive analysis of phytochemicals in plant extracts even more challenging. Use of mutant, RNA interference, and overexpression lines in which the expression of genes encoding regulatory and biosynthetic components of metabolic pathways has been altered can greatly facilitate both identification of new metabolites and pathway discovery (46).

There is evidence to indicate that preformed antimicrobial chemicals confer protection against disease. For example, oat roots produce an antimicrobial triterpene glycoside known as avenacin (1). Genetic analysis of a fungal pathogen of oat, Gaeumannomyces graminis var. avenae, has shown that infection of oat roots depends on production of a fungal avenacin hydrolase (1). Complementary experiments involving isolation and characterization of avenacin-deficient mutants of diploid oat have provided further evidence that this compound confers broad-spectrum disease resistance (7). Another class of constitutively produced secondary metabolites is the steroidal glycoalkaloids. The tomato steroidal glycoalkaloid α-tomatine has long been implicated in plant defense (1). The ability to degrade this compound contributes to the pathogenicity of various microbes to tomato (1, 8, 9). Superficially, these degradation processes represent simple detoxification events. However, α-tomatine hydrolysis products are able to suppress induced plant defenses (10, 11). The aglycone of α-tomatine, tomatidine, has recently been shown to inhibit sterol biosynthesis in yeast (12). It is not yet known whether suppression of induced defenses in tomato by steroidal alkaloid hydrolysis products is due to interference with plant sterol metabolism.

Induced defense responses in plants usually involve cell polarization, reorganization of the actin cytoskeleton, directed movement of particular organelles, targeted secretion, and deposition of the glucan polymer callose at the site of pathogen contact (13). This may also include trafficking and secretion of antimicrobial compounds to the infection site (14). Visualization of delivery of vesicles containing red-pigmented flavonoids to pathogen challenge sites in sorghum leaves provides a clear demonstration of such trafficking (15). Recent evidence suggests that glucosinolates, amino acid–derived thioglucosides that are commonly synthesized and stored in cells of healthy crucifer plants, may also be mobilized to pathogen challenge sites (16, 17). Upon tissue disruption, glucosinolates are converted to biologically active compounds by myrosinases (plant glycosyl hydrolases). Although best known as insect deterrents, glucosinolate breakdown products have potent antimicrobial activity (1).

Research drawing on the extensive array of mutants and tools available for the model crucifer Arabidopsis thaliana has recently implicated indole glucosinolates and their breakdown products (compounds distinct from those considered as insect deterrents) in induced broad-spectrum disease resistance to microbial challenge (16, 17). Unlike the passive mode of glycoside activation proposed for other phytoanticipins, pathogen-triggered glucosinolate metabolism is an active process involving directed movement and the concentration of a hydrolase to the cell periphery at fungal penetration sites (18), which likely generates high local concentrations of end product(s). Of note, the same genetic and metabolic components are required for the extracellular accumulation of callose in response to treatment with a microbe-associated molecular pattern derived from bacterial flagellin (16). Localized accumulation of glucosinolate metabolism products may serve as the signal for this deposition, or alternatively may trigger callose deposition directly as a consequence of phytotoxicity. Accumulation of late avenacin pathway intermediates in oat roots also results in callose accumulation (19), raising the question of whether this is a straightforward response to localized accumulation of a toxic compound or part of a more sophisticated defense response. More broadly, the contribution of phytoalexins to disease resistance has been the subject of intensive investigation in a range of different plant species [e.g., for camalexin in Arabidopsis and pisatin in pea (1, 20, 21)]. These compounds have, in at least some cases, been shown to contribute to disease resistance, possibly by serving as disinfectants that assist in isolating infected cells from healthy tissue.

Some phytochemicals are known to have multiple functions in ecological interactions. For example, glucosinolates serve as antimicrobial defense compounds and as attractants and repellents for insects and predators of feeding insects. They are also important in determining palatability of brassicas for animals and humans (1, 22). These multiple roles present challenges when considering how to distil knowledge of the biological functions of particular phytochemicals into a framework that will enable the importance of these chemicals for survival in nature to be understood, particularly given temporal and spatial environmental heterogeneity.

Phytochemicals can also have functions in plant growth and development, for example in auxin transport and in regulation of seed longevity and dormancy (14). Strigolactones, terpenoid lactones produced by many plants, are not known to be involved in defense against microbes but nevertheless merit attention here because they provide an excellent example of the multiple roles that phytochemicals can serve. Strigolactones stimulate seed germination of parasitic plants such as Striga (23). More recently, they have also been implicated as signals for the establishment of mycorrhizal symbioses (24, 25), which suggests that parasitic plants have learned to eavesdrop on a signal that plants normally produce to favor the establishment of beneficial interactions. A further twist to this story is the finding that strigolactones serve as plant growth hormones and can suppress subapical shoot outgrowth (26, 27).

Genes functioning in secondary metabolism are generally more divergent than those coding for proteins involved in primary metabolism. Investigations of chemical diversification in plants will therefore benefit from focusing on genes and regions of genomes that are divergent when related producing and nonproducing species are compared, rather than those that are conserved. Although genes for most metabolic pathways in plants are generally thought to be unclustered, an increasing number of operon-like gene clusters have been identified that are required for synthesis of plant defense compounds (5, 2830). It is not clear why genes for some metabolic pathways are clustered and others are not. There may be epistatic selection for the maintenance of a gene cluster because the pathway end-product confers a selective advantage and, in at least some cases, because disruption of the pathway can lead to the accumulation of toxic intermediates (5, 19). Clustering may in addition facilitate coordinated regulation at the level of nuclear organization/chromatin. It may be that these gene clusters, which tend to be in subtelomeric regions, are of recent origin and have not yet been fully integrated into the genome.

Clustered or not, natural product pathways serve as read-outs for adaptive evolution and offer a means of understanding the mechanisms that drive metabolic diversification. Computational biology and genomics provide new and powerful tools with which to understand the molecular basis of complex traits and to integrate these with knowledge of the natural products that plants produce (31, 32). Connecting the catalytic landscape of secondary metabolism to fitness landscapes of organisms represents a major challenge, particularly as it is becoming increasingly evident that secondary metabolites are likely to have multiple functions both in plant development (in at least some cases) and chemical ecology (33). Real progress will require a shift from reductionism to more comprehensive analyses of natural and agronomic systems (34), and the development of sophisticated ecological and evolutionary models that integrate plant metabolism and biotic interactions.

References and Notes

  1. The authors acknowledge their sources of funding (A.O., Biotechnology and Biological Sciences Research Council UK; P.B., Max Planck Chemical Genomics Center) and thank their colleagues for helpful discussions in the preparation of this article. A.O. has the following patents awarded/pending: A.O. and K. Haralampidis (2000), Plant gene PCT/GB00/04908 (WO 01/46391); U.S. patent awarded 6 March 2007 (US7,186,884,B2); Mexico 25 July 2008 (259077); Canada pending (2,392,435); Australia issued (783739); Europe pending (EP1240312). X. Qi and A.O. (2004), Enzymes involved in triterpene synthesis (WO/06044508); U.S. pending (11/248,986); Canada pending (CA2581099); Australia pending (AU5295733); Europe pending (EP18709752). A.O., K. Haralampidis, R. Melton, S. Bakht, and X. Qi (2006), Root-specific promoters (U.S. patent application no. 11/940,638, pending). A. O. and X. Qi (2007), Enzymes involved in triterpene synthesis (WO/2009/041932, pending).
View Abstract

Navigate This Article