LysM receptors recognize friend and foe

Chitin, a β(1→ 4)-linked polymer of N-acetylglucosamine, is the second-most abundant polysaccharide in nature after cellulose and serves as a major structural component in the exoskeleton of arthropods and in the cell walls of fungi. The latter include some of the most important pathogens of plants. Upon microbial attack, plants normally mount a multicomponent defense response that efficiently stops the invasion. This important and durable type of plant disease resistance is called species or nonhost resistance. To become pathogenic, a microbe needs to “learn” to suppress the defense response of a plant species, thereby turning it into a host species. Nonhost resistance is based on a non-self recognition system that perceives at the site of attempted penetration typical microbe-associated molecular patterns (MAMPs), which normally do not occur in plants (1). Prominent MAMPs recognized by plant cells are chitin fragments (chitooligosaccharides) released from fungal cell walls during pathogen attack, which in many plants elicit the plant defense response (oxidative burst, protein phosphorylation, transcriptional activation of defense-related genes, phytoalexin biosynthesis, etc.). High-affinity binding sites were found in suspension-cultured rice and tomato cells (2, 3), and a 75-kDa chitooligosaccharide-binding protein was identified in rice plasma membranes by affinity labeling and cross-linking (4). Results from inhibitor studies using different oligosaccharides were in good agreement with the activities of these oligosaccharides in the induction of phytoalexin biosynthesis and of other cellular responses. Although this suggested the 75-kDa protein to be the functional receptor for the chitooligosaccharide elicitor, the receptor protein at first remained elusive. In this issue of PNAS, Kaku et al. (5) report on the purification of the chitin oligomer-binding protein (CEBiP) from rice and the cloning of its gene. It encodes a 356-aa protein with a 28-aa secretory signal sequence. The mature protein of 328 aa has a calculated molecular weight of 34,640, the higher molecular mass of 75 kDa obtained in previous experiments being due to glycosylation. In CEBiP-RNAi lines, where expression of the gene was knocked down, specific binding of chitooligosaccharides, chitooligosaccharide-elicited oxidative burst, and defense-related reprogramming of gene expression were strongly reduced. In contrast, these cell lines were fully responsive to another typical MAMP, bacterial lipopolysaccharide, indicating that specific, receptor-mediated perception of chitooligosaccharides was affected but not that of other structurally unrelated MAMPs.

Chitooligosaccharides as MAMPs

MAMP recognition is also part of the vertebrate immune system, which comprises an innate and an acquired component, the former being required for the activation of the latter. Acquired immunity is exerted by mobile cells specialized on pathogen defense (lymphocytes). It provides the capability to recognize a huge variety of antigens through antibodies targeted to specific pathogen epitopes and generated after somatic recombination of antibody genes. However, innate immunity is based on the recognition of MAMPs such as lipopolysaccharides, peptidoglycans, bacterial flagellin, or chitin fragments. Although cell-surface Toll-like receptors and intracellular NOD (nuclear oligomerization domain) receptors have been found to be involved in MAMP recognition in animals, no receptor of chitin fragments has been identified so far (6).

Like the animal immune system, the plant immune system is built up of two components, both of which are, however, inherited. In recent years, it has been shown that plant genomes contain a large number of resistance genes, with the majority encoding intracellular immune receptors with some structural similarity to Toll-like and NOD receptors (6). These proteins, typically built up of a leucine-rich repeat (LRR), a NOD-NBS domain, and either a coiled-coil or a TIR domain, recognize pathogen-specific proteins either directly or by their impact on plant virulence targets. Hence, they provide a similar pathogen-specific immunity (cultivar-specific resistance) as the adaptive immune system in animals. In addition, several cultivar-specific immune receptors are located in the plasma membrane, exhibiting extracellular LRR domains (1). Because of the lack of a substantial cytoplasmic domain, functional signaling is presumed to require additional proteins. However, the majority of encounters with pathogenic microbes are handled by the nonhost resistance system, which is like animal innate immunity based on the recognition of MAMPs such as bacterial flagellin again or fungal cell wall structures including branched glycans or chitin fragments. The plant flagellin receptor FLS2 is a typical receptor-like kinase with an extracellular LRR domain and cytoplasmic kinase domain (7). The two receptors for microbial cell wall components characterized to date are members of different protein families. The hepta-β-glucoside receptor of legumes constitutes a novel class of glucan-binding proteins lacking recognizable functional domains (8). However, the newly identified chitooligosaccharide receptor belongs to yet another type of cell-surface receptor carrying LysM domains (5). This finding brings another role of chitooligosaccharides into focus.

Chitooligosaccharides in Development and Symbiosis

In addition to their role as MAMPs in defense activation, chitooligosaccharides have been associated with regulatory signaling during developmental processes in plants and animals. Recently, a receptor-like kinase, CHRK1, was identified in tobacco that contains an extracellular chitinase-like domain lacking any catalytic activity and a functional cytoplasmic kinase domain (9). Transgenic plants in which the CHRK1 gene was silenced showed pleiotropic developmental abnormalities such as shooty callus formation and profuse shoot formation. At the cellular level, ectopic cell proliferation, reduced cell specificity, and aberrant chloroplast development were noted. These observations are reminiscent of cytokinin-overproducing plants (10), and the elevated cytokinin levels in the CHRK1-silenced tobacco plants appear to be responsible for the majority of observed phenotypes. However, among >600 receptor-like kinases encoded in the genome of the model plant Arabidopsis thaliana, none was found to contain a chitinase-like domain (11). Instead, the gene AtCTL1 encodes a ubiquitously expressed extracellular chitinase-like protein, whose mutation caused pleiotropic effects on plant growth and development such as ectopic deposition of lignin, aberrant cell shapes, and overproduction of ethylene (12). It remains to be determined whether signaling by the secreted AtCTL1 protein across the plasma membrane is mediated by a membrane-anchored receptor-like kinase. However, for both CHRK1 and AtCTL1, it remains unclear how they regulate hormone levels. Furthermore, the long-standing question as to the substrate/ligand of these chitinase-like proteins needs to be answered.

In contrast to plants, chitooligosaccharides have been shown to be produced in animals such as zebrafish and Xenopus laevis (13), and genes encoding proteins with sequence similarity to the rhizobial nodC protein have been isolated (developmental gene DG42; ref. 14). NodC catalyzes a step in the biosynthesis of nodulation (nod) factors. These lipochitooligosaccharides are organogenesis-inducing signal molecules produced by Rhizobium spp. to establish the formation of nitrogen-fixing root nodules in leguminous plants. This finding raises the intriguing possibility that a bacterium-plant type of lipochitooligosaccharide signaling system may operate during early stages of vertebrate embryonic development and raises issues about the use of chitin synthase inhibitors as fungal-specific drugs.

The specific recognition of rhizobial lipochitooligosaccharide signals in legumes involves serine/threonine receptor kinases (NRF1 and NRF5 in Lotus japonicus, LYK3 in Medicago truncatula) that contain LysM motifs in their extracellular domains (15). These LysM domains are found in a variety of peptidoglycan- and chitin-binding proteins, suggesting that they may be directly involved in the perception of rhizobial lipochitooligosaccharide signals. Mutational analysis revealed that the LysM domain receptors appear to function in the direct recognition of nod factors. In L. japonicus, they are crucial for the early responses to rhizobial infection, whereas in M. truncatula, they are also involved in specific recognition during later stages of bacterial infection (16). Another type of receptor, SYMRK in L. japonicus and NORK in M. truncatula, belongs to the LRR receptor kinases and appears to be involved in the early perception of signals from both symbiotic rhizobia and mycorrhizal fungi. The intracellular kinase domains of both types of receptors are structurally related, suggesting possible similarities in downstream signaling of these LRR receptors and LysM receptors (17).

Interestingly, the now identified rice CEBiP also harbors two extracellular LysM motifs, thus linking signal perception during development, symbiosis formation, and defense. CEBiP lacks any cytosolic domain, such as the kinase domains of the putative nod factor receptors. Most importantly, however, Kaku et al. (5) have clearly demonstrated that CEBiP represents the ligand-binding protein within the plasma membrane-localized chitooligosaccharide receptor, something that still needs to be shown for the other candidate receptors of chitooligosaccharide derivatives. However, it is evident that additional proteins must be present within the receptor complex, which is responsible for initiating the intricate defense signaling network. It will be very exciting to view the unraveling of the molecular basis of chitooligosaccharide signaling within and between different organisms and to understand their possible evolutionary relationships.