Department of Biology, University of North Carolina, Chapel Hill, NC, USA

Background and objectives
We are interested in genetic approaches to dissection of plant disease resistance. We use Arabidopsis to identify and isolate mutants of genes causally required for effective disease resistance against the bacterial pathogen Pseudomonas syringae and the oomycete parasite Peronospora parasitica. The first class of loci comprises classically defined disease resistance (R) genes. The second are genes whose mutant phenotype results in misregulated hypersensitive cell death. The third are regulatory loci controlling either, or both, of the first two classes.

Results and conclusions
We cloned the Arabidopsis RPM1 gene [1] which conditions resistance to P. syringae strains expressing either avrRpm1 or avrB. The deduced RPM1 protein is a prototypic member of the leucine zipper/nucleotide binding site/leucine-rich repeat (LZ-NBS-LRR) class of disease-resistance proteins. We have generated C-terminal myc epitope-tagged RPM1 and showed that this modification is functional in transgenic rpm1 mutant plants. This tool, in addition to antisera raised against the NBS and LRR domains, allowed us to localize RPM1 as a peripheral membrane protein. Similar studies are under way to localize the P. syringae encoded avrRpm1 and avrB proteins. These are apparently injected into the host cell, as both transient expression and inducible transgenic expression of these bacterial proteins leads to plant genotype-dependent responses. In the presence of RPM1, this response is rapid and necrotic; in some genetic backgrounds lacking the RPM1 gene, the response is either chlorotic or necrotic. The latter responses define new plant targets for bacterial avr protein function, and we will discuss both mutation and breeding strategies for the isolation of these important new genes.

We isolated the LSD1 gene, which encodes a novel cysteine zinc-finger protein possibly involved in transcription regulation [2]. LSD1 function is required for appropriate response to pathogen, and lsd1 null mutants can be driven into a runaway cell-death cycle which is superoxide-dependent [3]. Thus the wild-type LSD1 protein is required for correct interpretation of the plant oxidative burst leading to negative regulation of the defence response [4]. We are constructing double mutants between lsd1 and other components of the disease-resistance pathway such as eds1, ndr1, nim1/npr1 and nahG, in order to place LSD1 action with respect to other known components of the disease-resistance pathway. In addition, we have isolated a number of lsd1 suppressors whose function with respect to pathogen response is being analysed. These suppressor loci are candidates for either regulators or mediators of both cell death and disease resistance, and they may correspond to mutations in the genes encoding LSD1 interacting proteins, which we will also describe.

Using the lsd5 mutant as a starting point, we have isolated two classes of suppressors. One class defines loci broadly required for HR-like cells, that is, these mutations suppress other cell-death control mutations. Their response to the pathogen indicates that at least one new locus, PHX1, is a negative regulator of pathogen-driven HR as well as a common element in lsd-type cell death. Other lsd5 suppressors have altered responses to the pathogen, thus defining modifiers of disease resistance gene action.

1. Grant MR, Godiard L, Straube E et al., 1995. Science 269, 843-846.
2. Dietrich RA, et al., 1997. Cell 88, 685-694.
3. Jabs T, Dietrich RA, Dangl JL, 1996. Science 273, 1853-1856.
4. Dangl JL, Dietrich RA, Richberg MH, 1996. Plant Cell 8, 1793-1807.