Pathogen resistance ( R ) genes of the NBS-LRR class (for nucleotide binding site and leucine-rich repeat) are found in many plant species and confer resistance to a diverse spectrum of pathogens. Little is known about the mechanisms that drive NBS-LRR gene evolution in the host-pathogen arms race. We cloned the RPP8 gene (for resistance to Peronospora parasitica ) and compared the structure of alleles at this locus in resistant Landsberg erecta (L er -0) and susceptible Columbia (Col-0) accessions. RPP8-L er encodes an NBS-LRR protein with a putative N-terminal leucine zipper and is more closely related to previously cloned R genes that confer resistance to bacterial pathogens than it is to other known RPP genes. The RPP8 haplotype in L er -0 contains the functional RPP8-L er gene and a nonfunctional homolog, RPH8A. In contrast, the rpp8 locus in Col-0 contains a single chimeric gene, which was likely derived from unequal crossing over between RPP8-L er and RPH8A ancestors within a L er -like haplotype. Sequence divergence among RPP8 family members has been accelerated by positive selection on the putative ligand binding region in the LRRs. These observations indicate that NBS-LRR molecular evolution is driven by the same mechanisms that promote rapid sequence diversification among other genes involved in non-self-recognition. INTRODUCTIONA broad range of microorganisms have evolved the ability to use plants as a nutritional resource, and plants in turn have evolved multiple lines of defense against pathogen invasion (Hammond-Kosack and Jones, 1996a). Inducible defenses are mediated through gene-for-gene systems in which the plant carrying a particular resistance ( R ) gene allele responds to pathogens carrying a matching avirulence ( avr ) gene (Flor, 1971). Most plants contain large collections of highly specific R genes, which are thought to encode specialized receptors that recognize avr gene-dependent elicitors (Keen, 1990). If the R gene or the corresponding avr gene is not functional, then recognition does not occur, defenses are not activated, and the plant is susceptible to infection. Thus, pathogens can circumvent gene-for-gene resistance by alteration or loss of avr genes. This places the host under selective pressure to evolve new recognition capabilities. avr gene mutations and deletions occur at high frequency in nature (van Kan et al., 1991;Rohe et al., 1995;Sweigard et al., 1995;Joosten et al., 1997), but the host's response in this evolutionary arms race is not well understood.Two themes have emerged from recent molecular characterization of R genes. R genes are often members of tightly linked multigene families, which can be functionally diversified (Hammond-Kosack and Jones, 1996b). A second, somewhat unexpected generality is that all R genes characterized to date, with one exception (Martin et al., 1993), encode proteins with long stretches of leucine-rich repeats (LRRs) . LRRs are present in a wide variety of proteins and participate in protein-protein interactions and ligand binding (Kobe ...

Pathogen resistance ( R ) genes of the NBS-LRR class (for nucleotide binding site and leucine-rich repeat) are found in many plant species and confer resistance to a diverse spectrum of pathogens. Little is known about the mechanisms that drive NBS-LRR gene evolution in the host-pathogen arms race. We cloned the RPP8 gene (for resistance to Peronospora parasitica ) and compared the structure of alleles at this locus in resistant Landsberg erecta (L er -0) and susceptible Columbia (Col-0) accessions. RPP8-L er encodes an NBS-LRR protein with a putative N-terminal leucine zipper and is more closely related to previously cloned R genes that confer resistance to bacterial pathogens than it is to other known RPP genes. The RPP8 haplotype in L er -0 contains the functional RPP8-L er gene and a nonfunctional homolog, RPH8A. In contrast, the rpp8 locus in Col-0 contains a single chimeric gene, which was likely derived from unequal crossing over between RPP8-L er and RPH8A ancestors within a L er -like haplotype. Sequence divergence among RPP8 family members has been accelerated by positive selection on the putative ligand binding region in the LRRs. These observations indicate that NBS-LRR molecular evolution is driven by the same mechanisms that promote rapid sequence diversification among other genes involved in non-self-recognition. INTRODUCTIONA broad range of microorganisms have evolved the ability to use plants as a nutritional resource, and plants in turn have evolved multiple lines of defense against pathogen invasion (Hammond-Kosack and Jones, 1996a). Inducible defenses are mediated through gene-for-gene systems in which the plant carrying a particular resistance ( R ) gene allele responds to pathogens carrying a matching avirulence ( avr ) gene (Flor, 1971). Most plants contain large collections of highly specific R genes, which are thought to encode specialized receptors that recognize avr gene-dependent elicitors (Keen, 1990). If the R gene or the corresponding avr gene is not functional, then recognition does not occur, defenses are not activated, and the plant is susceptible to infection. Thus, pathogens can circumvent gene-for-gene resistance by alteration or loss of avr genes. This places the host under selective pressure to evolve new recognition capabilities. avr gene mutations and deletions occur at high frequency in nature (van Kan et al., 1991;Rohe et al., 1995;Sweigard et al., 1995;Joosten et al., 1997), but the host's response in this evolutionary arms race is not well understood.Two themes have emerged from recent molecular characterization of R genes. R genes are often members of tightly linked multigene families, which can be functionally diversified (Hammond-Kosack and Jones, 1996b). A second, somewhat unexpected generality is that all R genes characterized to date, with one exception (Martin et al., 1993), encode proteins with long stretches of leucine-rich repeats (LRRs) . LRRs are present in a wide variety of proteins and participate in protein-protein interactions and ligand binding (Kobe ...

Root-knot nematodes and rhizobia establish interactions with roots characterized by the de novo induction of host structures, termed giant cells and nodules, respectively. Two transcription regulators, PHAN and KNOX, required for the establishment of meristems were previously shown to be expressed in tomato giant cells. We isolated the orthologues of PHAN and KNOX (Mt-phan and Mt-knox-1) from the model legume Medicago truncatula, and established the spatial distribution of their expression in situ. We confirmed that Mt-phan and Mt-knox-1 are expressed in lateral root initials and in nematode-induced giant cells and showed that they are expressed in nodules induced by Sinorhizobium meliloti. Expression of both genes becomes spatially restricted as the nodules develop. We further examined nematode feeding sites for the expression of two genes involved in nodule formation, ccs52 (encodes a mitotic inhibitor) and ENOD40 (encodes an early, nodulation mitogen), and found transcripts of both genes to be present in and around giant cells induced in Medicago. Collectively, these results reveal common elements of host responses to mutualistic and parasitic plant endosymbionts and imply that overlapping regulatory pathways lead to giant cells and nodules. We discuss these pathways in the context of phytohormones and parallels between beneficial symbiosis and disease.

SUMMARYLeucine-rich repeat receptor-like kinases (LRR RLKs) form a large family of plant signaling proteins consisting of an extracellular domain connected by a single-pass transmembrane sequence to a cytoplasmic kinase domain. Autophosphorylation on specific Ser and/or Thr residues in the cytoplasmic domain is often critical for the activation of several LRR RLK family members with proven functional roles in plant growth regulation, morphogenesis, disease resistance, and stress responses. While identification and functional characterization of in vivo phosphorylation sites is ultimately required for a full understanding of LRR RLK biology and function, bacterial expression of recombinant LRR RLK cytoplasmic catalytic domains for identification of in vitro autophosphorylation sites provides a useful resource for further targeted identification and functional analysis of in vivo sites. In this study we employed high-throughput cloning and a variety of mass spectrometry approaches to generate an autophosphorylation site database representative of more than 30% of the approximately 223 LRR RLKs in Arabidopsis thaliana. We used His-tagged constructs of complete cytoplasmic domains to identify a total of 592 phosphorylation events across 73 LRR RLKs, with 497 sites uniquely assigned to specific Ser (268 sites) or Thr (229 sites) residues in 68 LRR RLKs. Multiple autophosphorylation sites per LRR RLK were the norm, with an average of seven sites per cytoplasmic domain, while some proteins showed more than 20 unique autophosphorylation sites. The database was used to analyze trends in the localization of phosphorylation sites across cytoplasmic kinase subdomains and to derive a statistically significant sequence motif for phospho-Ser autophosphorylation.

No abstract