WO2002020791A1 - Plant resistance gene - Google Patents

Abstract

The present invention relates in part to the identification, characterisation, cloning and sequencing of a new plant resitance gene herein illustruated as SED ID NOS. 1 or 2. More particularly the invention relates amongst other things to the provision of a new gene sequence and associated protein sequence that confer resistance to abiotic stress or pathogen attack, as well as products comprising such sequence, or obtainable using the sequence. Examples include vectors, cells or transgenic plants comprising said sequences or related sequences and uses thereof.

Description

PLANT RESISTANCE GENE

The present invention relates in part to the identification, characterisation, cloning and sequencing of a new plant resistance gene. More particularly the invention relates amongst other things to the provision of a new resistance gene sequence and associated protein sequence, as well as products comprising such sequence, or obtainable using the sequence. Examples include vectors, cells or transgenic plants comprising said sequences or related sequences and uses thereof.

Plants are constantly subject to attack by a plethora of microbial organisms, including fungi, bacteria, and viruses, and have evolved an array of sophisticated defence mechanisms to protect themselves against disease. Physical barriers, such as a waxy cuticle, may prevent pathogen ingress in the first instance, and preformed antimicrobial chemicals such as saponins may also inhibit the pathogens attempts at colonisation (Osbourn, 1996) . Assuming the pathogen is able to overcome these defences and the plant in turn can support its particular niche requirements, two principle outcomes are possible: successful colonisation resulting in disease, or the may plant may be resistant to infection. Such events are often referred to as virulent (compatible) and avirulent (or incompatible) interactions respectively.

Such plant-pathogen interactions are exquisitely specific and are thought to be the result of a single recognition event. In an incompatible interaction, an avirulence (avr) gene product encoded by the pathogen is thought to interact directly or indirectly with a resistance (R) gene product encoded by the plant. This event triggers a signal transduction pathway that unleashes a battery of defences both locally and systemically termed the hypersensitive response (HR) and systemic acquired resistance (SAR) respectively. If the plant is lacking in the relevant R gene, such defences are induced several days later after the development of disease symptoms .

Although over 20 i-genes from seven plant species have been cloned to date (Martin 1999) , surprisingly little has been elucidated about their function. Probably the best characterised is Pto of tomato, the first .R-gene to be "cloned, which recognises the avrPto gene of Pseudomonas syrlngae (Martin et al., 1993). Pto encodes a cytoplasmic serine-threonine kinase suggesting it plays a role in the phosphorylation cascade, and has been shown exhibit kinase activity in vitro (Loh & Martin 1995) . Direct physical interaction between the Pto and avrPto proteins has been demonstrated in vitro by means of a yeast two-hyrid assay (Scofield et al . 1996; Tang et al. 1996).

All other R genes share one common feature: a leucine rich repeat (LRR) region (Jones & Jones, 1997), and can be sub-grouped according to the additional structural domains they possess. LRRs consist of leucines and other hydrophobic residues at regularly spaced intervals, which are thought to specify protein-protein interactions. Analysis of the crystalline structure of another LRR, porcine ribonuclease inhibitor, has suggested that it is the interstitial residues between the conserved leucines that determine specificity of ligand binding which might equally apply to plant LRRs (Jones & Jones, 1997) .

Aside from Pto, only one other .R-gene has been shown to have a kinase domain, Xa21 of rice, which encodes a membrane-bound receptor kinase-like protein (Song et al . 1995). It has an external LRR domain which is postulated to interact with an Avr protein which may activate the cytoplasmic kinase domain to trigger a signal transduction cascade culminating in various defence responses. The Cf .R-genes of tomato (e.g. Cf-9, Cf-4) that convey resistance to the fungus Cladosporium fulvum also encode an extracellular LRR but lack any obvious domains that could effect downstream signal transduction (Jones et al . 1994; Dixon et al . 1996).

The most prolific domain within the LRR class of R genes is the nucleotide binding site (NBS) located at the N terminus of the protein. This element is common to a number of proteins of diverse organisms and is required for ATP and GTP binding (Saraste et al . 1990) though this has yet to be demonstrated biochemically. A number of conserved motifs have been identified within the NBS domain (Traut et al . 1994). Intriguingly, these motifs are also found in the pro-apoptotic genes apaf-1 and ced-4 of humans and C. elegans respectively, suggesting that regulation of these cell-death pathways might share conserved elements (van der Biezen & Jones, 1997) .

.R-genes containing NBS-LRRs can be further sub-divided according to whether they possess an amino terminus leucine zipper (LZ) (eg. rps2 (Bent et al . 1994) or region with homology to Toll/Interleukin-1 receptor (TIR) (eg. N of tobacco (Whitam et al. 1994) . Although Toll and interleukin receptor proteins are known to be involved in non-specific cell-immunity in animals, preliminary analysis of resistance to different flax rust isolates suggests this region may play a role in pathogen recognition (Ellis et al. 1999).

Leucine zippers are a sub-set of coiled-coil (CC) structures known to promote dimerization and facilitate protein-protein interactions (Alber, 1992), however little is known about their role in .R-gene signal transduction. Interestingly, whilst TIR-NBS-LRR genes comprise 75% of NBS-LRRs they have so far only been found in dicots despite exhaustive searching in cereals. In contrast, NBS-LRRs with a CC domain are widespread throughout the angiosperms (Pan et al . , 2000) .

Intriguingly, these two classes of NBS-LRR .R-genes appear to have different specificities in terms of their immediate downstream effectors. Pathogenicity analysis has revealed that the mutant edsl displays enhanced susceptibility to NBS-LRRs with a leucine zipper, while mediated by those with a TIR region is abrogated in the ndrl mutant. This finding therefore indicates that signalling following R-gene recognition is mediated by at least two distinct signalling pathways (Aarts et al., 1998).

Whilst the specific roles of R genes in the disease resistance signal transduction pathway remain elusive, some insights have been gleaned from studies in tomato. Pto and a related serine/threonine protein kinase Fen (conferring sensitivity to the herbicide Fenthion) have been shown to both be dependent on an additional gene, Prf, that encodes a LZ-NBS-LRR (Salmeron et al . 1996). This implicates both LRR-containing proteins and protein kinases as components of the same signalling pathway and this relationship is further underlined by the fact that the Xa21 protein contains both kinase and LRR domains . It is conceivable that other NBS-LRR genes may also have respective protein kinase partners. Furthermore, recent studies in protoplasts have also demonstrated that Rps2 co- immunoprecipitates with its respective avirulence gene product, AvrRpt2 in addition to an unknown plant protein (Leister et al . 2000) . This suggests that the formation of a complex of at least 3 proteins may be necessary for the elicitation of a resistant response.

It is an object of the present invention to provide a novel defence- related gene and uses thereof.

In a first aspect the present invention provides a polynucleotide fragment substantially identified as SEQ ID NOS. 1 or 2, or fragment or homologue thereof. The polynucleotide fragment identified as SEQ ID NO. 1 includes a gene sequence from bases 1-2787 of a plant defence-related gene, termed herein after as adrl . This sequence represents the genomic sequence and therefore includes both introns and exons . Thus the actual coding sequence corresponding to the mRNA, where T is substituted by U is predicted as bases 1-640; 710- 1044; 1131-1487; 1560-1765; 1961-2787. SEQ ID NO. 2 in fact shows the corresponding cDNA sequence of the adrl gene. The sequences may be used to confer resistance to abiotic stress or pathogen attack in plants, such as monocots or dicots .

"Polynucleotide fragment" as used herein refers to a chain of nucleotides such as deoxyribose nucleic acid (DNA) and transcription products thereof, such as RNA. Naturally, the skilled addressee will appreciate the whole naturally occurring Arabidopsis or other plant species genome is not included in the definition of polynucleotide fragment.

The polynucleotide fragment can be isolated in the sense that it is substantially free of biological material with which the whole genome is normally associated in vivo. The isolated polynucleotide fragment may be cloned to provide a recombinant molecule comprising the polynucleotide fragment. Thus, "polynucleotide fragment includes double and single stranded DNA, RNA and polynucleotide sequences derived therefrom, for example, subsequences (also referred to herein as sub-fragments) of said fragment and which are of any desirable length. Where a nucleic acid is single stranded then both a given strand and a sequence complementary thereto is within the scope of the present invention.

The polynucleotide fragment may be expressed in order to provide an expression product. In general, the term "expression product" refers to both transcription and translation products of said polynucleotide fragments. When the expression product is a "polypeptide" (i.e. a chain or sequence of a ino acids displaying a biological activity substantially similar to the biological activity of an essential protein, it does not refer to a specific length of the product as such. Thus, the skilled addressee will appreciate that "polypeptide" encompasses inter alia peptides, polypeptides and proteins. The polypeptide if required, can be modified in vivo and in vitro, for example by glycosylation, amidation, carboxylation, phosphorylation and/or post-translation cleavage .

using the information provided by the present invention, the DNA coding sequence for an ADR1 related polypeptide or gene sequence from any plant source may now be obtained using standard methods, for example, by employing consensus oligonucleotides and PCR. Furthermore, any promoter (s) associated with the adrl gene may also be identified using the information provided by the present invention.

"Homologue" as used herein refers to sequences, or fragments of sequences, which are similar to those disclosed herein.

The invention still further provides a polynucleotide sequence which is similar to the disclosed DNA sequences. By "similar" is meant a sequence which is capable of hybridising to a sequence which is complementary to the inventive nucleotide sequence. When the similar sequence and inventive sequence are double stranded the nucleic acid constituting the similar sequence preferably has a Tm within 20 °C of that of the inventive sequence. In the case that the similar and inventive sequences are mixed together and denatured simultaneously, the Tm values of the sequences are preferably within 10°C of each other. More preferably hybridisation may be performed under stringent conditions, with either the similar or inventive DNA preferably being supported. Thus for example either a denatured similar or inventive sequence is preferably first bound to a support and hybridisation may be effected for a specified period of time at a temperature of between 50 and 70 °C in double strength SSC (2xNaCl 17.5g/l and sodium citrate (SC) at 8.8g/l) buffered saline containing 0.1% sodium dodecyl sulphate (SDS) followed by rinsing of the support at the same temperature but with a buffer having a reduced SSC concentration. Depending upon the degree of stringency required, and thus the degree of similarity of the sequences, such reduced concentration buffers are typically single strength SSC containing 0.1% SDS, half strength SSC containing 0.1%SDS and one tenth strength SSC containing 0.1% SDS.

Sequences having the greatest degree of similarity are those the hybridisation of which is least affected by washing in buffers of reduced concentration. It is most preferred that the similar and inventive sequences are so similar that the hybridisation between them is substantially unaffected by washing or incubation at high stringency, for example, in one tenth strength sodium citrate buffer containing 0.1% SDS.

Therefore, the invention still further provides a nucleotide sequence which is complementary to one which hybridizes under stringent or moderately stringent conditions with the above disclosed nucleotide sequences. The present invention therefore provides nucleotide sequences which are 50%, 60%, 70%, 80%, 90%, 95% or 98% similar with the disclosed sequences.

As is well known in the art, the degeneracy of the genetic code permits substitution of bases in a codon resulting in a different codon which is still capable of coding for the same amino acid, eg. the codon for amino acid glutamic acid is both GAT and GAA. Consequently, it is clear that for the expression of polypeptides with the amino acid sequences shown in SEQ ID NOS. 1 or 2, or fragments or homologues thereof, use can be made of derivative nucleic acid sequences with such an alternative codon composition different from the nucleic acid sequences shown in SEQ ID NOS. 1 or 2.

The nucleotide sequences of the present invention may be isolated from another plant, such as a monocot or a dicot.

Additionally, it will be understood that for the particular polypeptides embraced herein, natural variations can exist between individuals or between members of the family. These variations may be demonstrated by (an) amino acid difference (s) in the overall sequence or by deletions, substitutions, insertions, inversions or additions of (an) amino acid(s) in said sequence. All such derivatives showing the recognised physiological activity are included within the scope of the invention. For example, for the purpose of the present invention conservative replacements may be made between amino acids within the following groups :

(I) Alanine, serine, threonine;

(II) Glutamic acid and aspartic acid;

(III) Arginine and leucine;

(IV) Asparagine and glutamine;

(V) Isoleucine, leucine and valine;

(VI) Phenylalanine, tyrosine and tryptophan.

Also encompassed are synthetic amino acids incorporated into the polypeptides of the present invention, either by way of addition or substitution of existing amino acids.

Polypeptides, or fragments thereof, modified as hereinbefore described, and which retain the physiological activity of the original, full-length peptide are referred to herein as "functional variants" or "functionally active variants".

Moreover, recombinant DNA technology may be used to prepare nucleic acid sequences encoding the various derivatives outlined above.

The amino acid sequence or sequence motifs may be used in homology searches in protein databases to find ADR1 related proteins from other plant species. In fact as will be described in more detail hereinafter, the present inventors have found two further ADR-like proteins in Arabidopsis as well as ADRl related proteins from a diverse number of plant species.

For recombinant production of the resistance protein in a host organism, the plant ADRl coding sequence may be inserted into an expression cassette to form a DNA construct designed for a chosen host and introduced into the host where it is recombinantly produced. The choice of specific regulatory sequences such as promoter, signal sequence, 5' and 3' untranslated sequences, enhancer and terminator appropriate for the chosen host is within the level of skill of the routine worker in the art. The resultant molecule, containing the individual elements linked in proper reading frame, may be introduced into the chosen cell, using techniques well known to those in the art, such as electroporation, biolistic introduction, Ti plasmid introduction etc. Suitable expression cassettes and vectors and methods for recombinant production of proteins are well known for host organisms such as E. coli (see, e.g. Studier and Moffatt, J. Mol . Biol . 189: 113 (1986); Brosius, DNA 8: 759 (1989)), yeast (see, e.g., Schneider and Guarente, Meth, Enzymol . 194: 373 (1991)) and insect cells (see, e.g., Luckow and Summers, Bio/Technol. 6: 47 (1988).

Examples of promoters suitable for use in DNA constructs of the present invention include viral, fungal, bacterial, animal and plant derived promoters capable of functioning in plant cells. The promoter may be selected from so-called constitutive promoters or inducible promoters .

Examples of suitable inducible or developmentally regulated promoters include the glutocorticoid-inducible transcription system, napin storage protein gene (induced during seed development) , the malate synthase gene (induced during seedling germination) , the small subunit RUBISCO gene (induced in photosynthetic tissue in response to light) , the patatin gene highly expressed in potato tubers, the cauliflower mosaic virus 35S (CaMV 35S) and 19S (CaMV 19S) promoters, the nopaline synthase promoter, octopine synthase promoter, heat shock 80 (hsp 80) promoter and the like. Alternatively, the promoter could be selected to express the DNA constitutively. Generally, in plants and plant cells of the invention inducible or developmentally regulated promoters are preferred.

A terminator is contemplated as a DNA sequence at the end of a transcriptional unit which signals termination of transcription. These elements are 3 '-non-translated sequences containing polyadenylation signals which act to cause the addition of polyadenylate sequences to the 3' end of primary transcripts. Sequences mentioned above may be isolated for example from fungi, bacteria, animals or plants.

Examples of terminators particularly suitable for use in nucleotide sequences and DNA constructs of the invention include the nopaline synthase polyadenylation signal of Agrobacterium tu efaciens, the 35S polyadenylation signal of CaMV, octopine synthase polyadenylation signal, the zein polyadenylation signal from Zea mays, and those found in plas ids such as pBluescript (Stratagene, La Jolla, CA) , pFLAG (International Biotechnologies, Inc., New Haven, CT) , pTricHis (Invitrogen, La Jolla, CA) , and baculovirus expression vectors, e.g., those derived from the genome of Autographica californica nuclear polyhedrosis virus (AcMNPV) . An example baculovirus/insect system is pV111392/Sf21 cells (Invitrogen, La Jolla, CA) .

Therefore, the invention further provides an expression cassette comprising a promoter operably linked to a DNA molecule encoding ADRl or functionally active variant thereof from a plant.

In a yet further aspect the present invention provides a nucleotide sequence comprising a transcriptional regulatory sequence, a sequence under the transcriptional control thereof which encodes an RNA sequence characterised in that the RNA sequence is anti-sense to an mRNA which codes for ADRl.

The nucleotide sequence encoding the antisense RNA molecule can be of any length provided that the antisense RNA molecule transcribable therefrom is sufficiently long so as to be able to form a complex with a sense mRNA molecule encoding for ADRl. Thus, without the intention of being bound by theory it is thought that the antisense RNA molecule complexes with the mRNA for the protein or proteins and prevents or substantially inhibits the synthesis of a functional ADRl . As a consequence of the interference by the antisense RNA, enzyme activity of ADRl is decreased or substantially eliminated.

The DNA encoding the antisense RNA can be from about 20 nucleotides in length up to the length of the relevant mRNA produced by the cell. Preferably, the length of the DNA encoding the antisense RNA will be from 50 to 1500 nucleotides in length. The preferred source of antisense RNA transcribed from DNA constructs of the present invention is DNA showing substantial identity or similarity to the genes or fragments of ADRl in plants .

Therefore, a further aspect of the present invention provides transgenic plants wherein the ADRl activity in the cells of the plants has been substantially reduced or eliminated. The plants may, for example, be monocots or dicots.

Transcriptional initiation sequences are commonly located upstream from the transcriptional initiation site and which contains all the regulatory regions required for transcription. Examples of such transcriptional initiation sequences (also known as protomers) are hereinbefore described.

It will be appreciated that the promoter employed should give rise to the transcription of a sufficient amount of the antisense RNA molecule at a rate sufficient to cause an inhibition of ADRl activity in plant cells. The required amount of antisense RNA to be transcribed may vary from plant to plant.

DNA constructs and nucleotide sequences of the invention may be used to transform cells of both monocotyledonous and dictoyledonous plants in various ways known in the art. For example, particle bombardment of embryogenic callus is the method of choice for production of transgenic monocotyledonous plants [Vasil (1994) Plant Mol. Biol. 25, 925-937]. In many cases transformed plant cells may be cultured to regenerate whole plants which can subsequently reproduce to give successive generations of genetically modified plants .

The invention also provides a biological vector comprising a DNA construct according to the present invention. The biological vector may be a virus or bacterium, such as Agrobacterium tumefaciens, for example, and the construct advantageously further encodes a marker protein, such as one having herbicide resistance, or anti-bacterial properties .

A further aspect of the invention is a recombinant biological vector comprising the said construct wherein said vector is capable of transforming a host cell. Also comprised is a host cell stably transformed with the said vector wherein said host cell is preferably a cell selected from the group consisting of a bacterial cell, a yeast cell, and an insect cell and is further capable of expressing the DNA molecule according to the invention.

The invention still further provides eukaryotic cells, such as plant cells (including protoplasts) for example, containing the said nucleotide sequence, DNA construct or vector.

The invention still further provides plant cell with gene "knockouts" wherein the gene encoding ADRl or functional fragment or homologue thereof has been mutated or removed to eliminate expression.

The invention still further provides transgenic plants comprising such plant cells, the progeny of such plants which contain the sequence stably incorporated and heritable in a Mendelian manner, and/or the seeds of such plants or such progeny. Expression of the ADRl gene in transgenic plants is shown herein to confer resistance to a broad variety of unrelated pathogens. Additionally, or alternatively, expression of the adrl gene in transgenic plants confers drought resistance to such plants.

The invention still further provides the use of the sequence according to the invention, whether "naked" or present in a DNA construct or biological vector - in the production of eukaryotic cells, particularly plant cells having a modified ADRl activity.

Other plant ADRl related coding sequences may be isolated according to well known techniques based on their sequence homology to the sequence as shown in SEQ ID NOS. 1 or 2, or fragment, or homologue thereof. In these techniques all or part of the known ADRl coding sequence may be used as a probe which selectively hybridizes to other ADRl coding sequences present in a population of cloned genomic DNA fragments or cDNA fragments (i.e. genomic or cDNA libraries) from a chosen organism. Such techniques include hybridization screening of plated DNA libraries (either plaques or colonies; see, e.g. Sambrook et al . , "Molecular Cloning", eds . , Cold Spring Harbor Laboratory Press. (1989)) and amplification by PCR using oligonucleotide primers, for example corresponding to sequence domains identified from the adrl gene sequence. For example, the present inventors have identified a number of domains: CDPK; coiled coil region, kinase; Leucine rich repeats which may be suitable for designing oligonucleotide primers to (see Figure 12) .

Therefore, a further embodiment of the invention is a method of isolating a polynucleotide fragment, said polynucleotide fragment comprising a sequence having at least 50%, 60%, 70%, 80%, 90%, 95% or 98% sequence similarity with the disclosed sequences comprising

(a) preparing a nucleotide probe capable of specifically hybridizing to a plant adrl related gene or mRNA, wherein said probe comprises a contiguous portion of the coding sequence for ADRl from Arabidopsis thaliana of at least 10 nucleotides in length;

(b) probing for other ADRl related coding sequences in populations of genomic DNA fragments or cDNA fragments from a chosen plant using the nucleotide probe prepared according to step (a) ; and

(c) isolating a polynucleotide fragment comprising a portion encoding a protein having ADRl-like activity.

The isolated plant ADRl and Arøl-related sequences taught by the present invention may be manipulated according to standard genetic engineering techniques to suit any desired purpose. For example, the entire ADRl coding sequence or portions thereof may be used as probes capable of specifically hybridizing to coding sequences and messenger RNAs . To achieve specific hybridization under a variety of conditions, such probes include sequences that are unique among ADRl coding sequences and are at least 10 nucleotides in length, preferably at least 20 nucleotides in length, and most preferably at least 50 nucleotides in length. Such probes may be used to amplify and/or analyse ADRl coding sequences from a chosen organism via the well known process of polymerase chain reaction (PCR) . This technique may be useful to isolate additional adrl related coding sequences from other plant species as hereinbefore described or as a screening assay to determine the presence of ADi?2-related coding sequences in a plant. Hybridisation probes may also be used to quantitate levels of ADRl mRNA in a plant using standard techniques such as Northern blot analysis. This technique may be useful as a diagnostic assay to detect altered levels of ADRl expression that may be associated with particular conditions such as abiotic stresses, pathogen attack, etc.

Therefore, the invention further provides methods for detecting the presence and form of the ADRl related gene and quantitating levels of ADRl related transcripts in an organism. These methods may be used to diagnose conditions, such as those described previously, which are associated with an altered form of the ADRl or altered levels of expression of ADRl .

ADRl specific hybridization probes may also be used, for example, to map the location of the native ADRl related gene(s) in the genome of a chosen plant using standard techniques based on the selective hybridization of the probe to genomic adrl sequences. These techniques include, but are not limited to, identification of DNA polymorphisms identified or contained within the ADRl probe sequence, and use of such polymorphisms to follow segregation of the ADRl gene relative to other markers of known map position in a mapping population derived from self fertilization of a hybrid of two polymorphic parental lines (see e.g. Helentjaris et al . , Plant Mol. Biol. 5: 109 (1985); Sommer et al . Biotechniques 12:82 (1992);. D'Ovidio et al . , Plant Mol. Biol. 15: 169 (1990)). While any plant ADRl related sequence is contemplated to be useful as a probe for mapping adrl related genes, preferred probes are those ADRl related sequences from plant species more closely related to the chosen plant species, and most preferred probes are those adrl related sequences from the chosen plant species. Mapping of adrl related genes in this manner is contemplated to be particularly useful for breeding purposes. For instance, by knowing the genetic map position of a mutant adrl related gene which could confer properties as described previously, flanking DNA markers can be identified from a reference genetic map (see, e.g., Helentjaris, Trends Genet. 3: 217 (1987) ) . During introgression of the mutant adrl related gene trait into a new breeding line, these markers can then be used to monitor the extent of ADRl related flanking chromosomal DNA still present in the recurrent parent after each round of back-crossing.

Recombinantly produced plant ADRl protein and/or ADRl related proteins or homologues can be isolated and purified using a variety of standard techniques . The actual techniques which may be used will vary depending upon the host organism used, whether the ADRl and/or ADRl related proteins or homologues is designed for secretion, and other such factors familiar to the skilled artisan (see, e.g. chapter 16 of Ausubel, F. et al . , "Current Protocols in Molecular Biology", pub. By John Wiley & Sons, Inc. (1994) .

Therefore, the present invention further provides the recombinant production of ADRl functionally active variant thereof, or homologue thereof. In particular, the invention relates to a method of producing a protein having ADRl or ADRl related activity in a host organism comprising

(a) inserting a DNA sequence encoding a protein having ADRl or ADRl related activity into a host cell;

(b) growing the said transformed host cell in a suitable culture medium;

(c) expressing said DNA sequence to produce said protein; and

(d) isolating the protein product either from the transformed host cell or the culture medium or both and purifying it.

The cloning and expression of a recombinant adrl or adrl related polynucleotide fragment or homologue also facilitates in producing anti-ADRl antibodies and fragments thereof (particularly monoclonal antibodies) and evaluation of in vitro and in vivo biological activity of recombinant ADRl or ADRl related polypeptides . The antibodies may be employed in diagnostic tests for native ADRl polypeptides .

These and other aspects of the present invention shall now be described, by way of example only, and with reference to the accompanying Figures and Examples, which show:

Figure 1: Activation tagging vector using pSKI015 vector. A) T-DNA insertion cassette demarcated by left border (LB) and right border (RB) containing Basta resistance gene, origin of replication of E. coll (pBstKS+) and tetramer of 35S enhancers 4x35S (from Weigel lab website: www. salk.edu/LABS/pbio-w/researchfs .html) , and B) endogenous expression of genes adjacent to the 4x35S enhancer region may be enhanced.

Figure 2 : adrl mutant consitutively expresses prl as shown by luciferase imaging. Prla : :luciferase expression is markedly upregulated in adrl mutant background compared with Col-0. Light intensity corresponds to prl gene expression.

Figure 3 : adrl in different mutant backgrounds . Top picture in each rectangle is homozygous with respect to adrl ; bottom picture is hemizygous .

Figure 4: Expression of defence-related transcripts in different mutant backgrounds. Northern analysis shows mRNA levels of prl , gstl , pdfl . 2, adrl , and r! 8, in mutant backgrounds as indicated above. All analyses carried out in 4 week old plants except "adrl seedlings". (H) - homozygous, (h) - hemizygous. Approximate size on right as indicated by RNA ladder (Pro ega #G3191) .

Figure 5: adrl conveys resistance to Pernospora parasl tlca (downy mildew) . Disease index scored for multiple replicates according to method devised by Cao et al . 1997. (No bar is visible for adrl as disease index score was 0 - no disease symptoms visible on any leaves . ) Figure 6: adrl conveys resistance to Erlsiphae ci choraceum (powdery mildew) . adrl plants are significantly more resistant to powdery mildew cf. wild-type col-0. The mean number of leaves infected/plant was recorded 10 days after infection.

Figure 7: Plasmid rescue and subsequent cloning of adrl .

A) The rescued plasmid containing mas5' promoter, pBluescript, 4x35S and approx. 650 bp of adrl genomic DNA.

B) A 2.3 kb EcoRI/BamHI fragment containing enhancers and rescued DNA is cloned to pBluescript. Sequencing is performed with this construct using the T7 primer.

C) A 650 bp Xbal fragment of rescued DNA is used to probe a cosmid genomic library and a single colony is isolated.

D) A 400 bp Spel /Xbal fragment was used to probe DNA from the positive cosmid previously digested with Spel and in conjunction with a number of different enzymes.

E) A 5.5 kb Spel/Pstl positive fragment is cloned to pBluescript. A nested set of deletions is generated which are sequenced using the T3 primer.

Abbreviations: B - BamHl ; E - EcoRI ; Sc - Sad ; S - SpeZ ; P - Pstl ;

X - Xbal.

Figure 8: The relative postion of adrl in the genome. Schematic shows genes adjacent to adrl , the point of T-DNA integration (depicted by 4 block arrows) and the 5.5 kb fragment cloned from the genomic library.

Figure 9: Structure of the adrl genomic clone. Schematic shows five exons (rectangle boxes) and four introns (diagonal lines) . Motifs are : ^ CDPK homology ϋjcoiled-coil region

g nucleotide binding site ^leucine rich repeat P P-Loop

Figure 10: Sequence of the adrl genomic clone. Numbers relate to position wrt translation start site; non-translated regions (ie promoter, introns and 3' UTR in lower case; motifs in promoter and amino acid sequence indicated in bold, LRRs underlined. Putative TATA box and two polyadenylation signals are shaded.

Figure 11: Conserved motifs in Adrl. Box shade was performed to highlight homologies between peptide residues of Adrl and other genes. Grey boxes indicate similar residues, black boxes indicate identical residues. GenBank accession numbers as follows: AtCDPK3 (BAA05918), CDPK ( Zea mays) (T03271) , Adrl-Ll (AL162972) , N (A54810), Rprl (BAA75812) , Apaf-1 (AF013263) . Number refers to amino acid residue position in translated sequence. An asterisk (*) indicates the K and E residues that are highly conserved in serine threonine kinases (Hanks and Quinn, 1991). N.B. translated sequence of ADRl -LI is based on computer prediction.

Figure 12: The ten imperfect LRRs of the Adrl peptide sequence. L stands for leucine though can include isoleucine and other aliphatic residues (F,V,M) and x stands for any other residue (Jones & Jones 1997). Number refers to amino acid residue position in peptide sequence .

Figure 13: The ADRl overexpression cassette. The ADRl genomic clone was ligated into the KpriΣ/BamHL site of the MCS of pART7 to generate the overexpression cassette CaMV35S: :ADRl: : OCS. This was then cloned into the jVotl site of the binary vector pGREEN, incorporating the BAR gene (promoter and terminator of BAR omitted) for Basta resistance between the left and right border.

Figure 14: Constitutive expression of the ADRl candidate gene recapitulates the adrl-O phenotype . Plants transformed with the construct CaMV 35S: : ADR1 show a spectrum of phenotypes which were grouped into four classes A, B, C, D according to morphology; A) indistinguishable from Col-) , B) slight dwarfing, leaf curling, lesions apparent under long-day conditions, C) severe dwarfing, visible lesions (equivalent to homozygous adrl mutant, and D) very severe dwarfing lesions, often non-viable (equivalent to homozygous adrl mutant) . Figure 15: Three classes of phenotype were observed in transgenic CaMV35S: : ADR1 plants: Col-0-like line 23 (A); adrl heterozygous-like line 36 (B) ; adrl homozygous-like line 10 (C) .

Figure 16: Northern blot analysis of CaMV35 S : : ADRl transgenic line tested with ADRl and PR1 probes. rl 8 was used to control equal loading. Lines on the right indicate transcript sizes as inferred from RNA ladder.

Figure 17: Plants were infected with a Perenospora parasl tica NOC02 conidiospores suspension and the number of conidiophores per plant were counted 10 days later. Col-0 wild type and immuno-compromised nahG plants were included as control.

Figure 18: The glucocorticoid-inducible system for expression of ADRl in transgenic plants ( TA: : ADR1 binary cassette). A cassette without ADRl was used as control ( TA: : ) .

Figure 19: TA: : ADR1 (A) and TA: : (B) plants 72 hours after treatment with lOμM dexamethasone.

Figure 20: TA : z ADRl and TA: : plants were detected for accumulation of ADRl and PR1 transcripts after treatment with lμM dexamethasone.

Figure 21: TA: : ADR1 (A) and TA: : (B) plants 50 hours after spraying with lμM dexamethasone showed different LUC activity level and PR-1 expression.

Figure 22: Adult plants were infected with a Perenospora parasltica NOC02 conidiospores suspension and the number of conidiophores per plant were counted 10 days later. TA: ; and TA: :ADR1 were treated with 1 and 0.4 M dexamethasone 50 hours previous the infection.

Figure 23: Northern blot analysis of sense and antisense ADRl transcript accumulation in Col-0 and ADRl antisense transgenic line Figure 24: Plants were infected with a P. parasl tica N0C02 conidiospores suspension and the number of conidiophores per plant were counted 10 days later.

Figure 25: ADRl northern blot analysis of Col-0 m-RNA plants under different treatments (L = local leaves; S = systemic leaves) .

Figure 26: Plants were left 10 days without water. They were then rescued by watering on day 11 and the percentage of drought resistance plants are shown in the picture.

Figure 27: Seeds were collected and weighted from plants which were grown in similar conditions. Values are means (±2SE) of measurements of 10 to 25 plants per line and bars labeled with different letters indict that the corresponding data are significantly different (P >0.95) .

EXAMPLES SECTION Abbreviations

Me-JA - Methyl-Jasmonate

Salicylic acid- SA

BTH- Benzothiadiazole nahG - Arabidopsis plants containing a SA hydroxylase transgene that degrades SA. etrl- ethylene insensitive Arabidopsis plant coil- Methyl-jasmonate insensitive Arabidopsis plant

PCR- polymerase chain reaction

CAPS- co-dominant amplified polymorphic sequences

RTPCR- Reverse transcription, polymerase chain reaction

MATERIALS & METHODS

Reagents

Unless otherwise stated, all reagents used were supplied by Sigma (Poole, Dorset, UK) .

Nucleic acid analysis Total RNA was extracted from 4-6 week old plants according to standard procedures (Reuber and Ausubel, 1996) . For Northern analysis, 12 μg total RNA samples were separated by electrophoresis through formaldehyde-agarose gels and transferred to a nylon membrane (Amersham) exactly according to manufacturers instructions (Hybond booklet) . 32P-labelled DNA probes were prepared using a Prime-a-Gene® labelling kit (Promega) . Hybridisation conditions and stringency washes (always at 65 °C) were as described by Ausubel et al, . 1996. Blot hybridizations were quantified with a Phosphorlmager (Molecular Dynamics Inc., Sunnyvale, CA) in conjunction with I ageQuant 3.3 software (Molecular Dynamics Inc., Sunnyvale, CA) and normalised with reference to rl8 hybridisation.

Sequencing reactions were prepared and run on a HYBAID Omnigene Thermocycler using the PERKIN-ELMER ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction Kit, according to the manufacturers instructions . Sequence data was transferred to the UNIX-based GCG package for further analysis.

Cross-pollination and selection of mutant backgrounds

Genetic crosses were undertaken by dissecting and emasculating unopened flower buds and then using the remaining pistils as recipients for pollen from 4 opened flowers. Transgenics with a selectable marker (i.e. gstl::luc and adrl) were always as the pollen donor and mutant lines as recipient, enabling kanamycin and Basta selection as appropriate, on MS plates or soil respectively. Successful crosses were allowed to self and homozygous transgenic plants were screened from their progeny.

Progeny containing nahG were identified by virtue of their brown deposits in root tissue when grown on MS media containing ImM salicylic acid (Bowling et al . , 1994). The ethylene insensitive mutants etrl and ein2-l were selected at 2-3 weeks on MS plates containing 50μM ACC by virtue of having significantly longer roots than wildtype plants. Mutants successfully introgressed into the coil background were selected by normal growth on MS plates containing 200μM Me-JA in comparison to stunted wild-type plants. The nprl-1 mutant background was selected by Nlalll digestion of PCR product of genomic DNA amplified by CAPS markers. Appropriate controls were always used to check selection method.

Treatment of Plants

Bacteria were maintained as previously described (Dangl et al . 1992) . Mutant hrp Pseudomonas strains were transformed with the pVBOl plasmid by electroporation as described by Keen et al. (1990) . Bacteria were inoculated into individual leaves by pressure infiltration using a 1 ml syringe; 10 μl of bacteria were infiltrated into the abaxial leaf surface.

Real time in planta imaging of LUC activity

Leaves of gstl::luc or Prla::luc transgenic plants were painted with a solution containing ImM Luciferin (Promega) and 0.01% triton X-100 and 0.03% Silwet (Union Carbide) in a ImM sodium citrate buffer (pH 5.8). All in planta LUC imaging was performed using an ultra low light imaging camera system (EG & G Berthold Luminograph 980) . Images were routinely collected over a Is { GST1 : : LUC) or 10s (PR1 : : LUC) time period. Microscopy imaging was carried out using Nikon Optiphot-2 microscope.

Pathogenicity assays

Perenospora parasi tica NOC02 (Parker et al . , 1992) was maintained on Col-0 seedlings grown in majenta jars. For the P. parasi tica disease resistance assays, conidiospores were harvested by vortexing infected seedlings in water. Spore concentration was determined using a haemocytometer, and resuspended in sterile distilled water to 1X105 spores per ml. Four-week old plants grown under short day conditions were sprayed with the conidiospore solution and placed in trays covered with SaranTM wrap to maintain a humid environment. Fungal growth on plant leaves (visualised as conidiophore growth) was scored 10 days post-infection using qualitative method adapted from Cao et al . (1997). Scoring was as follows: 0=no infection, l=less than 25% of one leaf with conidiophore growth, 2=25 to 50% of one or two leaves covered with condiophores, 3=25 to 50% of three or four leaves covered with conidiophore growth, 4=25 to 50% of all leaves covered with condiophore growth, 5=all leaves covered with conidiophore growth. Plants in different replicates .were assigned a disease index as follows: D.I.= iXj/n, where I=infection class, j=the number of plants scored for that infection class and n=the total number of plants in the replicate (based on Epple et al . 1997) .

Activation Tagging

Agrobacterium GV3101 transformed with the pSKI015 vector (Detlef et al . , 2000) was grown overnight at 30 C in LB media containing with 50 mg.l^-1 kanamycin (to select for the helper plasmid) and 50 mg.l^"1 ampicillin (to select for the pSKI015 binary vector) and used to transform PR1 : : LUC by floral dip method (Clough & Bent, 1998). Seed harvested from transformed plants was sown in flats and selected by lOOmg/1 Basta (Agrevo) two weeks after germination, then three times at four day intervals. Resistant plants were visually inspected for any abnormalities then sub-planted to pots and F2 seed collected for individual lines. Pools containing around 10 seed from 25 different lines were sown in pots and selected with Basta ten days post- germination. Five days later seedlings were painted with luciferin and imaged for constitutive luciferase activity. One mutant, later identified as adrl, was stunted and had a lesion mimic (LM) phenotype as a primary transformant and was therefore was also imaged as an adult plant.

Homology searches and sequence analysis

The following web-sites were used and instructions followed as detailed on web site. Coiled-coil structure prediction was performed with COILS (Lupas, 1997) by web server (http://www.ch.embnet.org/software/COILS_form.html). The scans were then made with variable window sizes of 14-28 residues using the MTIDK matrix. GCG10 and Genejockey were also used for miscellaneous functions .

Generation of ADRl transgenic line

The expression cassettes were created as outlined below, transformed into Agrobacterium by freeze-thaw method which was then used to transform Prla::luc lines by >floral dip= method.

i) CaMV35S: :ADR1 - PCR was used to amplify the ADRl genomic fragment with Kpnl and BamHI immediately adjacent to the start (ATG) and stop codon respectively using the following primers (bold indicates start/stop codons, italics shows RE recognition site) : Kpnl primer -5' -GCTTAGGTACCAAGATCGGTCTCGAT -3' BamHI primer -5' -GCGAAGGATCCAGAAGCCTAATCGTC -3'

This fragment was subsequently sub-cloned into Kpnl/BamHI sites of the MCS of pART7 (Gleave 1992) downstream of the CaMV35S promoter. The CaMV35S : :ADRl : :OCS cassette was sub-cloned by Notl digestion to pGreen (Hellens et al . , 2000) and this binary vector was then used to transform Agrobacterium containing the trans functioning plamid pJIC Sa_Rep. Transgenic plants generated were selected by spraying with lOOmg/1 Basta (Agrevo) two weeks after germination, then at three times at four day intervals.

Bacterial resistance assay

P. syringae pv. tomato DC3000 (Pst) was grown in King's broth (KB) liquid media (King 1954) supplemented with 50 mg.l^-1 rifampicin. Four week old plants were infected with a Pst suspension (OD600 = 0.0002) in 10 mM MgCl2 by pressure infiltration of the abaxial side of the leaf with a 1ml syringe. Three leaves per plant, and three plants per line were infiltrated. After three days, plants were inspected for development of symptoms. Leaves were also harvested at this time point for analysis of bacterial growth. Leaf discs of the same size (0.5cm²) were made from these samples using a cork borer. Three leaf discs from each plant were ground in 990μl 10 mM MgC12 in a pestle and mortar. Serial dilutions were made from the resulting bacterial suspension, and lOOμl of each dilution was plated onto KB plates containing 50 mg.l^-1 rifampicin. The plates were incubated at 30 °C for 2 days, and the number of bacterial colonies for each sample was recorded.

Abiotic stress assays i) Drought 20 seedlings (14 days old) of both adrl and Col-0 plants were transplanted into two separate halves of the same tray under short day conditions. Watering was stopped 11 days later and differences in terms of wilting or death was recorded as it became apparent. Dead plants were confirmed by failure to resuscitate on resumption of watering which was performed when all Col-0 plants appeared dead.

EXAMPLE 1: ISOLATION AND CHARACTERISATION OF ADRl: AN ACTIVATION- TAGGED DISEASE RESISTANCE MUTANT i) Background

Mutant screens have been used extensively as a means of dissecting genetic pathways in a wide array of different organisms. In plants this has taken the form of EMS mutagenesis, neutron particle bombardment and T-DNA tagging. EMS and irradiation are efficient methods of generating large numbers of mutations that effectively saturate the genome. However a significant drawback of these methods is the subsequent lengthy process of gene-cloning usually by means of chromosome walking. Cloning genes from T-DNA tagged plants is comparatively more straightforward though there are often associated problems such as chromosomal rearrangements that make plasmid rescue difficult (Feldmann, 1991) .

One major shortcoming of all three methods of mutant isolation is that they are largely confined to isolation of loss of function mutations. As such, genes whose function is essential during multiple stages of the plant life-cycle will not be uncovered if their disruption results in a lethal phenotype. In addition, sequencing of the Arabidopsis genome has revealed a high degree of apparent gene duplication (Bevan et al . 1998), so loss of function screens rarely identify genes that act redundantly. Activation tagging has recently been developed as a means of isolating gain of function mutations (Kaki oto 1996; Kardailsky et al . 1999; Weigel et al. 2000) and enabling facilitated cloning of the respective genes. The binary vector pSKI015 (Figure 1A) (Weigel et al . 2000) has been designed specifically for this purpose: between the left and right border is a tetramer of CaMV 35S enhancers (Fang et al . 1989) that is randomly integrated in the plant genome. Endogenous expression of genes immediately adjacent to the site of insertion should thus be significantly enhanced (Figure IB) . Also integrated into the transformation cassette is the BAR gene, thus facilitating large-scale screening by herbicide (Basta) spraying, and the oriV origin of replication of E. coli that enables regions of the plant genome flanking the T-DNA point of insertion to be cloned by plasmid rescue.

In addition to gain of function mutants, insertional loss of function knockouts can be uncovered in the T2 generation. This approach, therefore, is an efficient way of identifying genes involved in redox signalling, programmed cell death and downstream signalling pathways.

ii) Isolation of mutant displaying spontaneous lesion formation

Approximately 25,000 activation tagged Arabidopsis lines were generated in a genetic background containing a PRl : : LUC transgene (Murray and Loake, unpublished) using the pSKI015 vector. The firefly LUC gene was released from plasmid pSP-luc+ (Promega) via a Hind III, Xba I digest and the resulting fragment cloned into pART7 (Gleave 1992) digested with the same two enzymes. The PRl promoter was obtained via PCR using DNA isolated from wildtype Nicotiana tabacum cv. Xanthi plants as template (Payne et al . , 1988). The primers used were:

5' -GTAGGTACCGATATCCTCTTTACA-3' 5' -CTGAGCTCAGTTTGAGATCGAAGTGA-3'

The resulting PCR fragment was digested with Sac I and Nco I and cloned into the pART7 LUC derivative digested with the same two enzymes. The resulting pART7 derivative was digested with Not I, which released a chimeric gene consisting of the PRl promoter transcriptionally fused to the LUC coding sequence which was also fused to an octopine terminator sequence at its 3' -terminus. The Not I fragment containing the constructed chimeric gene was subsequently cloned into the Not I site of the binary transformation vector pART27 (Gleave 1992) . The resulting plasmid was transformed into A. tumefaciens, which was subsequently employed to generate transgenic Arabidopsis plants containing the PRl : : LUC: : OCS transgene. Approximately, 60 independent lines were screened for robust activation of LUC activity in response to attempted infection with Pst DC3000 expressing the avrB gene. One such line, which contained the transgene inserted at a single locus, was used for bulked seed collection when homozygous for the PRl : : LUC: : OCS transgene. The resulting seeds were transformed with A. tumefaciens containing the SKI15 activation tagging vector (Detlef 2000) . The resulting lines were screened in the first (Tl) generation for visual peculiarities and constitutive luciferase activity. One such mutant that showed constitutive luciferase activity as a seedling which was markedly enhanced in the adult (Figure 2), was isolated for further analysis.

The candidate mutant exhibited a "lesion mimic" (LM) phenotype characterised by spontaneous lesion formation in the absence of any stimuli. A number of mutants exhibiting a similar morphological phenotype have previously been isolated, most notably the accelerated cell death ( acd) mutants (Greenberg et al . 1994) and lesions simulating disease resistance response (Isd) mutants (Dietrich et al . 1994). Such mutants have been shown to be perturbed in redox signalling (Jabs et al . 1996) and often display elevated levels of SA and enhanced disease resistance (Greenberg 1994 et al.; Dietrich et al . 1994). Furthermore, no gain of function mutant, as opposed to the loss of a negative repressor, in this class has previously been reported making it an intriguing candidate for further analysis .

The mutant was allowed to self-pollinate and the F2 generation analysed. Basta resistance in the progeny always segregated with lesion formation and constitutive luciferase expression suggesting that the gene conveying this phenotype was tagged. Furthermore progeny segregated exactly 3:1 (BastaR.-BastaS) in 120 lines tested suggesting T-DNA insertion had occurred at one locus only.

Seedlings were indistinguishable from wild-type until 2-3 weeks following germination. After this point mutant plants became progressively stunted relative to wild-type, with more narrow curled leaves on which small lesions were apparent (Figure 3) . Interestingly, the phenotype appeared to be semi-dominant: in homozygotes (ie. two copies of 4x35S: : ADRl ) leaf curling and dwarfing were greatly exacerbated as compared with the hemizygote (Figure 3) . Fluorescent microscopy also revealed lesions in the hemizygote of around 10 cells in diameter, whereas in the homozygote the frequency and size of lesions in a given area was approximately double. Semi-dominance has also been observed in a number of other activation tagged mutants including another disease resistant mutant, CDR1 (Weigel et al . , 2000). Unlike lsdl , the phenotype was not strictly dictated by day-length (Jabs et al . 1996), though lesion development was enhanced when plants were grown under a highlight regime, probably as a consequence of greater photo-oxidative stress .

The adrl mutant was out-crossed to wild-type col-0 and progeny allowed to self-fertilise. The resulting progeny were selected for Basta resistance and scored for loss of prla::luc cassette by lack of constitutive luciferase activity, and these plants were crossed to transgenic gstl::luc plants. Basta resistant progeny of successful crosses showed constitutive GST1 : : LUC activity that also segregated with lesion formation and Basta resistance.

The mutant (subsequently named adrl for activated disease resistance) was crossed to a number of different mutant backgrounds, and progeny of successful crosses allowed to self-pollinate before selecting for homozygotes (with respect to mutant background) . The mutants employed were: nahG plants, which contain a salicylate hydrolase gene from Pseudomonas putida that degrades SA to inactive catechol (Gaffney et al . , 1993); etrl, (Blecker et al . , 1988) which is insensitive to ethylene; ein2-l , which is also ethylene insensitive by virtue of a mutation in an independent gene Guzman and Ecker (1990); coil , which is insensitive to Me-JA (Feys et al . , 1994) ; and, nprl , which does not establish SAR in response to SA (Cao et al . , 1994). These mutants were chosen for crossing purposes as they are belived to be defective in key disease resistance signalling pathways. The double mutants were characterised at 4 weeks in terms of severity of the adrl visible phenotype and accumulation of defence-related transcripts. In the case of adrlxnahG mutants, hydrogen peroxide accumulation and lesion proliferation was also measured, and bacterial and fungal pathogenicity assays performed.

iii) Morphological phenotype is mediated by SA, ethylene, and NPR1, but not by JA signalling pathway

Some lesion mimic mutants such as lsd6 and lsdl have previously been shown to lose the visible phenotype in the nahG background, suggesting that a SA-regulated positive feedback mechanism mediates lesion formation (Weymann et al., 1995). This phenomenon was also observed in plants heterozygous for 4x35S: :ADR_Z crossed to nahG which were indistinguishable from col-0 x nahG. Interestingly, adrl x nahG homozygotes displayed a significant degree of leaf-curling, slight reduction in size, and some lesion development visible under conditions of high light. This is the first report of a lesion mimic mutant able to partially overcome the ability of nahG to counteract the lesion phenotype.

Application of benzothiadiazole (BTH) , an analogue of SA, that is not a substrate for salicylate hydroxlase, has previously been shown to reiterate lesion development in nahG x lsd6 and nahG x lsd7 (Weymann et al . 1995). The commercial preparation of this compound, termed Bion (Novartis) ( Schweizer et al . 1999; Lawton et al.

1996) was used in a similar recapitulation experiment. Leaves of the nahG x adrl double mutant developed lesions approximately five days after Bion application, whilst no discernible difference was observed in nahG controls, nor in Col-0 painted with silwet alone, the surfactant previously added to the Bion solution. Intriguingly, new leaves and systemic leaves also showed lesion formation in nahGxadrl plants despite not having been directly treated. The formation of micro-lesions has previously been observed in naive leaves distal to those undergoing the hypersensitive response (Alvarez et al . , 1998), thus this observation in Bion-treated adr mutants may in effect be an amplified reiteration of lesions mediated by a similar enigmatic diffusible signal. Interestingly, strong PRl gene expression was found surrounding these lesions. Therefore, the local expression of ADRl engages a systemic signalling network leading to lesion formation, the engagement of PRl expression and presumably the establishment of SAR.

Lesion development, dwarfing and leaf-curling were partially attenuated in ein2, etrl , and nprl mutant backgrounds, though not to the same extent shown by nahG. adrlxnprl double mutants also exhibited a pronounced yellowing chlorosis which increased in severity towards the centre of the rosette. A similar effect has been reported in the lesion mimic mutants cpr5 and acdβ crossed with nprl (Bowling et al . , 1997; Rate et al., 1999), suggesting that nprl mutants are hypersensitive to SA accumulation. Lastly, coil did not appear to significantly block the LM phenotype though natural variation within the coilxadrl population was greater than observed for other double mutants.

Hydrogen peroxide accumulation was assayed in adrl in Col-0 and nahG backgrounds. Diaminobenzidine (DAB) staining, which detects hydrogen peroxide accumulation, was greatly enhanced in the adrl mutant. Staining was intensified in the homozygote but more homogenous in distribution. No staining was evident in the nahG background for the hemizygote, though the homozygote exhibited a light brown staining suggesting that excess hydrogen peroxide accumulation was not completely abrogated.

The adrl mutant exhibited both macro- and micro-lesions as indicated by trypan blue staining, though concentration and size of lesions was greater in the homozygote. Of the nahG x adrl double mutants, micro-lesions were only apparent in the nahG x adrl homozygote.

iv) Northern analysis directly implicates a role for ADRl in disease resistance

The proliferation of mRNA transcripts of five different genes was measured by northern analysis in adrl double mutants. GST1 , PRl , and PDF1.2 were chosen as markers of the HR, SAR and the SA- independent pathway respectively. Probing with ADRl was performed to assess if co-dominance was mediated at transcriptional level and whether adrl might be activated by a feedback mechanism. R18 was used as a loading control. The results of the northern analysis are summarised in Figure 4.

High levels of GST1 transcript were shown to accumulate in the adrl mutant with significantly greater amounts present in the homozygote than the hemizygote. adrl seedlings also over-express GST1 but not to the same extent as adult plants. Expression in adrl mutants was significantly reduced but not completely blocked in the nahG background. GST1 expression appeared to be slightly higher in adrlxcoil, but was unaffected in ein2, etrl , and nprl backgrounds. This apparent insensitivity and only partial reduction of GST1 in homozygous adrl x nahG is in contrast to PRl and may reflect different induction specificities of the respective genes.

The expression pattern of PRl was broadly similar to that of GST1 , however relative induction of PRl was significantly higher in adrl mutants than GST1 . This was not the case in adrl seedlings in which gstl but not prl induction was evident. nahG completely abolished prl gene expression in both the adrl homozygote and hemizygote. Expression also appeared slightly reduced in coil, ein2, and etrl , but significantly lower in nprl background. Relative induction of prl in seedlings was lower than for GSTl .

Transcript levels of pdfl.2 markedly contrasted with GSTl and PRl expression patterns, reflecting the different specificities of the SA-independent pathway. All mutants showed high expression of PDF1 . 2 except for adrl x coil in which it was entirely abolished. The ein2, etrl , and nprl mutant backgrounds appeared to partially attenuate expression of PDF1 . 2. The results for adrl x nahG are more ambiguous as expression appears to be reduced in the hemizygote but increased in the homozygote relative to adrl x Col-0. Surprisingly, expression in adrl seedlings was non-existent despite being evident in the adult col-0 suggesting that adrl-mediated PDF1 . 2 expression may be developmentally regulated.

v) adrl mutants display enhanced resistance to fungal pathogens

The overexpression of defence transcripts and hyperaccumulation of SA prompted the inventors to examine whether adrl mutants were enhanced in the ability to withstand pathogen infection. This was assessed by scoring disease symptoms in adrl in comparison to Col-0, nahG, hemizygous adrl x nahG, and the LM mutant cpr5 (Bowling et al . , 1997) following infection with the biotrophic oomycete Peronospora parasi tica . The scoring method took into account number of leaves infected and proportion of leaf covered.

The result of the P. parasitica resistance assay are summarised in Figure 5. There was no sign of fungal infection on any of the adrl plants but this resistance was dramatically abrogated in the nahG background which was the most susceptible of the lines. nahG was also significantly more susceptible than Col-0 with around three times the disease index score of the wild-type. Conversely, cpr5 was marginally more resistant than Col-0 with approximately half the disease index score of wild-type. adrl also exhibited enhanced resistance to Erisiphae cichoraceum as compared with Col-0 (Figure 6) .

EXAMPLE 2: STRUCTURE AND FUNCTION OF THE ADRl GENE i) Cloning of the ADRl gene

Plasmid rescue was carried out using EcoR I-digested adrl genomic DNA to obtain sequence downstream from the point of T-DNA insertion. 21 positive colonies were obtained and digestion with EcoRI/BamHI revealed that all contained a 650 bp fragment downstream from the 4x35S enhancers (Figure 7A) confirming that only one T-DNA insertion event had taken place. Two of these colonies also released 3 kb and 4 kb fragments respectively, following EcoR I digestion, possibly due to incomplete digestion of the genomic DNA during plasmid rescue . Digests with EcoR V and Xba I also confirmed the tetramer of 35S enhancers was intact (data not shown) . A 2.3 kb EcoRI/BamHI fragment containing the 4x35S enhancers and rescued DNA was cloned to pBluescript SK- for sequencing (Figure 7B) . Homology searches using this sequence revealed that this area had not already been sequenced by the Arabidopsis Genome Initiative (AGI) . Sequence immediately adjacent to the 4x35S enhancers was exactly homologous to EST AI995729 which encodes a gene with homology to serine carboxypeptidases . However, as the T-DNA appeared to have inserted into the coding region of this gene it was unlikely to be responsible for the phenotype of the mutant.

Sequence further downstream from the 4x35S enhancers was obtained by probing a Ws-0 cosmid genomic library {Arabidopsis Stock Centre, Ohio) . A 650 bp Xba I fragment (Figure 7C) was used to probe DNA isolated from individual library chapters using Southern blotting. The DNA from one chapter was found to hybridise with the 650 bp DNA probe. Subsequently, a single bacterial colony from this chapter containing the cross-hybridising cosmid was isolated via colony hybridisation using the 650 bp DNA probe. Cosmid DNA from this colony was digested with Spe I in tandem with 13 other restriction enzymes, the digests run on a gel and Southern hybridisation performed. This blot was probed with a 400 bp Spe I/Xba I fragment (Figure 7D) to isolate fragments downstream of the region already rescued. Autoradiography revealed positive bands ranging in size from 1 kb to 6 kb and one Spel/Pstl band of 5.5 kb was cloned to pBluecript SK- for further analysis (Figure 7E) .

The 5.5 kb fragment was sequenced to determine whether it contained any genes that might account for the adrl phenotype. The 5.5 kb fragment cloned into pBluecript SK- was digested with Sacl and Spel then incubated with Exonuclease III to generate a nested set of deletions . Timepoints for incubation with the exonuclease were chosen so as to generate fragments differing in size by 300-400 bp, which were subsequently transformed into E. coli DH5. Sequencing was performed using the T3 primer and a contiguous overlapping sequence assembled (Figure 9) .

The sequenced contig was analysed for the presence of open reading frames (ORFs) by Genscan on the web-server, and the results are summarised in the schematic in Figure 9. The 5.5 kb fragment was predicted to contain a gene designated C of 2787 bp approximately 700 bp downstream from the 4x35S (Figure 10), which was later shown to encode adrl. A 500 bp fragment of a gene was identified 1411 bp downstream from the stop codon of C, referred to as gene D. Homology searches indicated that C had homology to .R-genes whilst D was similar to genes encoding transposon-like proteins. Gene C was thus a suitable candidate for further analysis as it was immediately adjacent to the 35S enhancers and was possibly implicated in disease resistance signalling by virtue of its sequence. Moreover, gene C had very high homology (98-100%) at the nucleotide level to 3 ESTs (F19983; N96117; Z25604) in the database suggesting it encoded an expressed gene, while no ESTs were found for gene D.

During the course of this work AGI released the sequence of BAG F10C21 which was shown to contain the isolated 5.5 kb sequence. This BAC is located on the top arm of chromosome I near genetic marker mi423a. Although the ADRl and F10C21 sequences are from different ecotypes (Ws-0 and col-0 respectively) they differ by only 1 nucleotide in the 2787 bp gene sequence (G1413 to T1413 which changes a methionine to an isoleucine) which is possibly due to sequencing error . Sequence from this BAC was used to determine what genes were located upstream of ADRl. Gene B was predicted to encode a serine carboxypeptidase of 225 amino acid residues and further upstream was located a gene designated 'A' (Figure 10) with no homology to any known genes.

BlastX homology searches using the entire adrl nucleotide sequence revealed two distant homologues of adrl in Arabidopsis, AL162972 on chromosome 5 (62%) and AL161583 on chromosome 4 (56%), subsequently referred to as ADR1-LIKE1 (ADR1-L1) and ADR1-LIKE2 (ADR1-L2) respectively. These genes encode disease-resistance like proteins and a number of ESTs exist for each suggesting they both encode functional genes (Table 2) . The greater similarity for the ESTs of the two homologues is probably a more accurate reflection of the homology as genomic clones have no similarity in certain regions to ADRl which probably encode introns .

Gene C later named Activated Disease Resistance 1 (ADRl) , is predicted by Genscan to be 2787 bp in length from start to stop codon and contain five exons and four introns (Figure 10 and SEQ ID NO.l). A full length cDNA clone was isolated via RTPCR using adrl mRNA as template and the primers :

5'-GAGGTACCAAGATCGGTCTCGATGGCTTC (including start codon) 5'-GACCTAGGCTAATCGTCAAGCCAATCCA (including stop codon). Sequence analysis revealed that the full length transcript of ADRl is 787 aa residues, which is consistent with a transcript size of ~ 2.4 kb which was subsequently obtained by RTPCR using primers immediately adjacent to putative start and stop codons (see SEQ ID NOS. 2 and 3) .

ii) ADRl encodes a novel gene product with CDPK, CC, NBS, and LRR domains

ADRl is made up of four principle regions : the N terminus (nucleotides 1-561) containing a coiled-coil (CC) region (Lupas, 1996); a nucleotide binding site (562-1651) (NBS) (van der Biezen & Jones 1998) ; a region containing leucine rich repeats (1652-2629) (LRRs) (Jones & Jones 1997) (Figure 10), followed by small region at the C-terminus (2630-2787) containing no obvious domains.

The N terminus (residues 1-187) had no apparent homologues in the database except for a small region (residues 28-52) which shared significant homology (44%) to a domain highly conserved amongst plant calcium-dependent protein kinases (CDPKs) (Figure 11) . This homology was significantly lower in ADR1-L1 and ADR1 -L2 suggesting conservation is confined to ADRl . The genes shown selected for box- shade comparison were chosen as they represent closest homology to in this region: overall (ADR1-L1) ; to known genes {AtCDPK3, Urao et al. 1994); to monocot genes (maize CDPK) ; to non-plant genes (human leucine zipper-serine/threonine protein kinase.

A recent wealth of evidence has implicated CDPKs in .R-gene signalling and activation of the hypersensitive response (reviewed in Grant & Loake, 2000) . However, the region of similarity with CDPKs is relatively small (25 aa) and is confined to sub-domains II and III of the kinase (Hank S Quinn, 1991) . This region does not correspond with the catalytic site of kinases (sub-domain VII) and may be involved in substrate binding or necessary for proper conformation of the kinase (Harmon - pers . comm.) . The region of homology does however have a conserved lysine (K) and glutamic acid

(E) residues in sub-domains II and III respectively that are present in the same position in virtually all known serine-threonine kinases

(Hank & Quinn, 1991) . It is also interesting to note that in most serine-threonine kinases sub-domain I is a nucleotide binding site

(P-loop) and in ADRl a P-loop is also closely associated with this region, located downstream from sub-domain III. NBS-LRRs form a prolific class of genes in plants that includes many of the known

.R-genes. It is estimated that there are as many as 200-300 different NBS-LRRs in Arabidopsis comprising ~1% of the genome and tentative estimates in rice place this numbers as high as 1500

(Young, 2000) . The presence of a coiled-coil (CC) structure at the amino terminus (96-112) classifies ADRl as a Group I (or non-TIR)

NBS-LRR which is widespread throughout the angiosperms . Leucine does not predominate at position d of the heptad repeat so ADRl does not belong to the sub-class of CCs known as leucine zippers (LZ) (Pan et al. 2000) .

The domains that constitute the NBS were previously shown to be conserved between .R-genes, and genes encoding pro-apoptotic factors, APAF-1 & CED-4 (van der Biezen & Jones, 1998) . Conserved residues with other selected genes are highlighted in Figure 12. These genes were selected for comparison as they represent the closest homology to ADRl : overall, ADR1 -L1 ; of known genes, N (Whitam et al., 1994); of monocots RPR1 ; and of non-plant genes, APAF-1 (Zou et al., 1997).

These motifs are distributed between predicted exons 1-4 (Figure 12). A kinase la (P-loop) (consensus GXXXXGKT [T/S] ) is present which is common to an array of ATP/GTP binding genes (Saraste et al . , 1990) . In addition there is a kinase 2 domain and a HD motif (Fig. 11) located further downstream. The kinase 3a domain, and motifs 2, 4 and 5 are also present though these domains are less conserved.

In total, nine imperfect LRRs of 20 aa were found, all but one located in the third exon. The broad consensus (see Figure 12) (xLxxLxLxxCxxLxxLxxxx) is suggestive of cytoplasmic rather than extracytoplasmic proteins (Jones & Jones 1997) . The conserved cysteine residue found in seven of the LRRs in ADRl is frequently found in TIR-NBS-LRRs in the β-turn region after the LxxLxLxx motif (Botella et al . 1998). However, in comparison to other known R- genes (e.g. Cf-9) the LRRs in ADRl seemed more divergent from the general consensus sequence and the only consensus satisfying all ten LRRs is LxxLxLxx (where L is usually leucine but may be substituted for isoleucine and other aliphatic residues, and is x any other amino acid) .

RPM1, an .R-gene encoding a CC-NBS-LRR has been shown to have a glycosylation site and is localised at the cell membrane. No such glycosylation site was found in ADRl and the pattern of the LRR appears to suggest that the ADRl protein is localised in the cytoplasm, or signal sequence.

iii) ADRl is conserved in different plants and appears to have an ancient progenitor

TblastN homology searches were performed to compare the ADRl protein sequence against all plant DNA sequences translated in all three reading frames. In total seven different species besides Arabidopsis were shown to have transcribed sequences in the form of expressed sequence tags (ESTs) in the database. The distribution of hits reflects the fact that sequencing projects have concentrated on agronomically important crops or model species (e.g L. japonicus and M. truncula ta) . All of the ESTs with homology to ADRl were short sequences at the 3= end corresponding to the LRR which suggesting that the ESTs are not full length. The only ESTs from Arabidopsis with significant homology to ADRl were ADR1 -L2 (71%) and ADR1-L1 (68%) .

From an evolutionary perspective the high degree of conservation in a monocot, Sorghum, and gymnosperm, Loblolly pine, is particularly intriguing given that these species shared an evolutionary ancestor with Arabidopsis many millions of years ago. Furthermore the fact that the closest non-Arabidopsis homologue was shown to be a monocot suggests that genes with an even greater degree of homology may exist in the much less divergent dicot species. Interestingly, when a similar TBlastN search was performed for a cross-section of known .R-genes ( Prf, Rps2, Cf-9 and N) , the only ESTs with significant homology (>40%) in other species were confined to oilseed rape varieties (Rps2) , and potato (N) , and tobacco ( Cf-9) No homologues were found in monocots or gymnosperms .

Table 2: ESTs from other plant species with homology to ADRl. TblastN was performed against all Viridiplantae sequences in GenBank (www.arabidopsis.org/blast/). Identity refers percentage of identical residues as distinct from positives which are conserved residues. All hits are listed as obtained with default settings of Blast, unless more than one hit for a species in which case the hit with the highest homology is shown. ADR homologous region indicates the residue positions in ADRl with homology to the translated EST.

Homology searches against genes of known function (performed by BlastP search) ADRl was shown to have most similarity to NBS-LRR R- genes . This reflects the fact that with one exception, RPR1 , the only functionally characterised NBS-LRRs are £-genes . However, the most homologous .R-gene (N of tobacco) only shares 24% similarity which is largely confined to specific domains of the NBS (Figure 10) , and conserved leucines in the LRR.

iv) Motifs in the ADRl promoter

The 5' regulatory region of ADRl was analysed for regulatory motifs that might account for the pattern of inducible expression pattern previously observed. An ASF-1 motif (TGACG) was found at -592 relative to translation start site. This motif is thought to be involved in the transcriptional activation of several genes by auxin and/or salicylic acid (Terzaghi and Cashmore, 1995) . Additional support for SA-mediated transcription of ADRl comes from the presence of two TCA-elements (-32 & -93) containing the consensus TCATCTTCTT as previously found in the gstl promoter. Lastly, a motif (CACATG) found at -177 was shown to be necessary for drought/ABA-induction of the dehydration responsive gene RD22 in Arabidopsis, and binds a drought-regulated MYC transcription factor (Abe et al . , 1997) .

EXAMPLE 3: THE ROLE OF adrl IN DROUGHT RESISTANCE

Drought is second only to disease in terms of global loss of productivity, a problem that is likely to be exacerbated by global warming. Parallels have recently been drawn between the pathways governing abiotic stress and disease signalling. A number of signals are common to both pathways, such as an increase in cytostolic calcium, production of ROI, activation of MAPK cascades, and upregulation of antioxidant genes (reviewed in Bowler & Fluhr, 2000) . This fact coupled with the presence of a binding site for a drought responsive MYC transcription factor in the ADRl promoter prompted us to investigate whether adrl was altered in its ability to withstand drought conditions.

i) adrl is resistant to drought stress

A drought stress assay was carried out using adrl and Col-0 plants, with 20 of each grown in two separate halves of the same tray under short day conditions. Watering was stopped 25 days after germination and severe wilting was first evident in Col-0 plants 14 days later but adrl plants remained unaffected. At 16 days all wildtype plants were dead (Figure x) and adrl plants were showing severe wilting. After 17 days two of the 20 adrl plants had died. Visual evaluation of death was confirmed by complete resuscitation of all 18 of the 20 adrl plants but none of Col-0 (data not shown) . Drought resistance was also confirmed in 35S : : ADR1 transgenics suggesting that the phenotype is due to overexpression of ADRl and not some aberrant phenotype resulting from the activation tagging process .

Observations of plants grown in the greehouse under long day conditions suggest the drought tolerance exhibited by adrl plants is dramatically enhanced with mutants able to survive up to six days longer than Col-0 (data not shown) . This may be due to accelerated bolting and senescence of older leaves thus reducing water requirement and may be exacerbated by slight powdery mildew infection of wildtype but not mutant in this environment.

ii) SA, NPR1, JA, and ethylene are not key determinants of ADR1- mediated drought tolerance

The adrl plants lines introgressed into coil , nahG, etrl and nprl mutant backgrounds were deployed in drought assays to assess what signal transduction pathways might govern AQRl-mediated drought resistance. Another LM mutant, cpr5 was also used to determine whether drought resistance is commonly associated with the LM phenotype .

Between 6-8 lines of Col-0, adrl , adrl x nahG, adrl x etrl , and adrl x nprl were sub-planted at 14 days into different segments of the same tray and drought assay was then performed 11 days later exactly as described above. 16 days after watering was stopped all Col-0 plants but none of the other lines were dead although some exhibited severe wilting. After 17 days between 1-3 plants of the non-wild type lines except adrl x nprl were dead. Dead plants were confirmed by resumption of watering. In a separate experiment, adrl x coil mutants were shown to have greater resistance to drought than Col-0 which was not significantly different from adrl (data not shown) .

From this data we can conclude that neither SA, JA, ethylene, nor NPR1 significantly abrogate ADRl-mediated drought resistance. Furthermore, cpr5 also appeared to be have greater drought tolerance as compared with wildtype plants. As two of the four identified pathways governing drought resistance are mediated by abscisic acid (ABA) (Shinozaki & Yamaguchi-Shinozaki, 2000), ADRl has also been crossed into the ABA-insensitive mutant abil (Koornneef et al . , 1994) to determine whether ADi?l-mediated drought resistance is compromised in this background.

We anticipate the abiotic stress resistance shown by adrl plants will not be limited to drought. Thus, adrl plants are also likely to be resistant to freezing, salt, heat stress and potentially other abiotic stresses.

EXAMPLE 4: MANIPULATION OF ADRl EXPRESSION^'

AD.R1 overexpressing lines might help answer pertinent questions about ADi?l-mediated disease resistance. For example, whether it is possible to uncouple lesion development from disease resistance, and to what extent does severity of lesions correlate both with resistance and PR1 : : LUC expression. In addition, the disease resistant phenotype has potential for use in a biotechnological context as a means of conveying broad spectrum resistance to crop plants. Thus, it will be important to investigate whether the yield penalty associated with the dwarf phenotype can be attenuated without substantially compromising disease resistance. This could be achieved for e.g. by selecting the most desirable line among a large number of CaMV35S: : ADR1 transgenic plants. A complementary approach could also include an inducible gene expression system to transiently activate ADRl gene expression in a controlled fashion. To achieve these aims, the ADRl genomic clone was reintroduced into Arabidopsis under the control of the entire CaMV35S promoter, to convey constitutive ectopic expression of ADRl .

ii) Constitutive expression of the ADRl gene recapitulates the lesion mimic phenotype

The overexpression cassette was constructed by engineering Kpn and BamHI sites by PCR immediately adjacent to the respective start and stop codons of the ADRl genomic clone. This fragment was then cloned into the multiple cloning site (MCS) of pART7 (Gleave, 1992) to generate the overexpression cassette CaMV35S: : ADRl : :OCS. This construct was cloned into the jVotl site of the pGREEN (BastaR) (Hellens et al., 2000) which was transformed into an Agrobacterium strain containing the trans-acting plasmid, pJIC Sa_Rep (Hellens et al . , 2000) (see Figure 13). The resulting construct was then transformed into genetic background PRl : : LUC to facilitate subsequent screening of transgenics by means of luciferase imaging.

In total 110 CaMV35S : : ADRl transgenics were generated which were characterised with respect to morphology, luciferase activity and disease resistance. This was carried out in the FI generation due to time limitations but will be repeated in the progeny. A range of visible phenotypes was observed which were grouped into four distinct classes as depicted in Figure 14a) normal; B) slight dwarfing, leaf curling, (lesions apparent under long-day conditions) ; C) severe dwarfing, visible lesions (equivalent to hemizygous adrl mutant) ; D) very severe dwarfing, lesions, often non-viable (equivalent to homozygous adrl mutant) . In total 46% of transgenic plants generated exhibited some form of lesion mimic phenotype (classes B-D), as illustrated in Figure 14b.

iii) CaMV35S: :ADR1 transgenics have elevated PRl: : luc expression and enhanced resistance to a fungal pathogen

Two leaves of every plant were imaged for luciferase activity with reference to PRl : : LUC (Col-0) control plants. In total 46% of plants belonging to phenotypic classes B-D but none from class A had significantly elevated luciferase expression and the majority of such plants were from classes C and D with the more severe phenotype. Previous observations had indicated that luciferase imaging was much less sensitive than Northern analysis, thus it is likely some transgenic lines had elevated accumulation of prl transcripts that could not be detected by ultra-low light imaging camera .

Disease resistance in transgenic lines was recorded 14 days after transferring plants to an environment conducive to powdery mildew infection. Plants were scored as shown in Figure 1 follows: 0 - No infection; 1 - 1-2 leaves infected; 2 - mild infection on 3 or more leaves; 3 - heavy infection on 3 or more leaves. There was a significant correlation between the LM classes and resistance: 100% of plants of LM class A, but only 7% of B-D had infection scores of 2 or 3. Furthermore, resistance was shown to correlate strongly with severity of the phenotype, as D was the only class to exhibit zero infection. However it should be noted that plants were scored at 8 weeks by which stage it had previously been observed that senescing leaves of adrl mutants may become infected, as was the case in this instance in LM classes B & D. Drought resistance was also observed to be greater in classes B-D than A (results not shown) .

EXAMPLE 5: CaMV35S: :ADR1, transgenic lines constitutively expressing ADRl .

To confirm that AQR1 was responsible for enhanced disease resistance we generated a transgenic line carrying ADRl under the control of the constitutive promoter CaMV35S. Approximately, 100 T2 transgenic lines were screened and all the lines were recapitulated into three major classes.

The first class had the same phenotype as Col-0 wild type and was represented by line 23 (Figure 15A) . The second class showed an adrl hemizygotic phenotype and was represented by line 36 (Figure 15B) . Finally, line 10 represented the third class, which exhibited a phenotype very similar to that of adrl homozygotes, but of slightly larger size (Figure 15C) .

Northern analysis was then performed to investigate the accumulation of ADRl and PRl transcripts in these three different classes. The results clearly showed that plants manifesting an a rl-like phenotype constitutively expressed high levels of ADRl and PRl transcripts (Figure 16). In contrast, plants displaying the Col-0 wild type phenotype were unable to accumulate either AD.R2 nor PRl transcripts. Surprisingly, three different ADRl transcripts, of 2.8, 1.6 and 1.4 kb respectively, were visible, suggesting differential spicing .

Subsequently, a pathogenisity assay was performed infecting selected lines plants with Peronospora parasi tica (Figure 17) . Results showed that plants expressing high (line 10) and intermediate (line 36) levels of AD.R2 displayed high and medium disease resistance respectively, but plants that did not accumulate AQR2 (line 23) were as susceptible to P. parasi tica as wild type plants.

Thus all these results suggested that constitutive accumulation of ADRl drives the establishment of plant development, expression of defence-related genes and disease resistance, which consequentially impacts development in Arabidopsis .

EXAMPLE 6: Controlled expression of ADRl in TA::ADR1 transgenic line.

Since constitutive induction of AD.R2 led to disease resistance but simultaneously affected plant development, we then investigated whether the controlled expression of ADRl could orchestrate the establishment of resistance without affecting normal plant development and hence yield. This analysis was undertaken by employing a glucocorticoid-regulated gene trascriptional system. ADRl under the control of a dexamethasone-inducible promoter (Figure 18) , was transformed into wild type Ara idopsis plants carrying the PRl : : LUC reporter gene.

Analysis of T2 and T3 plants, identified a line homozygous for TA: :ADR1 which was used for further analysis. This line was treated with 10 μM dexamethasone and after a period of 48 hours chlorosis and HR-like lesions started to occur (Figure 19A) . After 6/7 days most TA: :ADR1 homozygous were dead, while control plants did not show any phenotypic irregularities (Figure 19B) .

Since it has been reported that a high level of dexamethasone treatment may trigger defence-related gene expression in plants, we sprayed plants with different dexamethasone concentrations in order to achieve effective ADRl and defence gene expression without potentially toxic consequences (Figure 20). TA: :ADR1 transgenic plants treated with lμM dexamethasone exhibited HR-like lesions after a 48 hours period; moreover, LUC activity was also detected via ultra low light imaging (Figure 21) .

We therefore investigated whether the controlled expression of ADRl orchestrated disease resistance by infecting plants with P. parasi tica 50 hours after lμM dexamethasone treatment. No pathogen growth could be detected on TA: :ADR1 plants following dexamethasone treatment (Figure 22) . The assay was then repeated treating plants with a lower concentration of dexamethasone (0.4 μM) and a similar result was obtained. Interestingly, younger plants treated with low dexamethasone concentration showed far fewer HR-like lesions and less LUC activity than older TA: :ADR1 plants, even though they all exhibited a high level of resistance to P. parasi tica .

Control plants containing the corresponding vector cassette devoid of ADRl did not exhibit either ADRl nor PRl transcript accumulation, increased LUC activity or HR-like lesion development.

EXAMPLE 7 : ADRl antisense transgenic line

In order to investigate whether ADRl expression is necessary to establish resistance against P. parasi tica and other pathogens, we generated an ADRl antisense transgenic Arabidopsis line. T2 generation plants were tested and one line was isolated that overexpressed ADRl antisense RNA and blocked the sense ADRl transcript accumulation (Figure 23) .

Two independent P. parasi tica infection assays were performed with similar results; ADP2 antisense plants were more susceptible compared to Col-0 wild type plants (Figure 24) .

EXAMPLE 8: Regulation of ADRl gene

In order to better understand the splicing mechanism of ADRl , northern blot analysis using polyA⁺ RNA was performed. The result confirmed ADRl accumulation following infection with Pst DC3000 (avrB) , BTH or wounding-like treatments (Figure 25). Moreover, ADRl induction was detected in systemic leaves 24 hours after infection with Pst DC3000 (avrB) , reaching maximum accumulation increase after 5 days. Furthermore northern blot analysis was performed on plants heavily infected with Erisiphae cicoraceum (powdery mildew) , a condition in which ADRl is highly expressed. Intriguingly, multiple splicing producing transcripts of 2.7 and 2.4 kb, was detected following wounding-like treatment and infection with powdery mildew; thus multiple splicing occurs. Moreover, the mechanism and the biological context of this alternative splicing remain to be established.

EXAMPLE 9 : Drought tolerance mediated by ADRl

It has been previously reported that adrl mutants show a higher dehydration tolerance compared with wild type plants. In order to confirm these data we repeated the same assay using plants constitutively expressing ADRl (Figure 26) . Remarkably, mutants that exhibited high (line 47) and intermediate (line 36) ADRl expression showed drought tolerance. In contrast, plants that did not accumulate ADRl transcript (line 23) were as susceptible as Col-0 wild type plants. To further investigate the effect of SA, JA, ET and NPRl on drought tolerance the same assay was carried out on four adrl double mutants, adrl : : coil , adrl : : ein2 and adrl : :nprl double mutants did not exhibit any significant variation on drought tolerance, yet adrl : : nahG showed only a partial resistance, cirl plants, accumulating high level of SA, was used as a control, but failed to show any drought tolerance. Therefore, drought tolerance mediated by ADRl appears to be independent of JA, ET or NPRl, but presumably partially dependent on SA. However, drought tolerance is generally independent of SA, since two mutants accumulating high levels of SA, cirl and cprl, did not exhibit any increased drought tolerance. Further assays will take place in order to confirm these data.

In addiction, ADRl expression after plant treatment with abscissic acid (ABA) was also investigated, revealing that ADRl transcript was accumulated in four folder higher levels than in control plants. Interestingly, under the same conditions, PRl transcript accumulation did not occur. Thus activation by exogenous factors, as ABA, may induce a specific ADRl expression leading to the establishment of distinct stress tolerance pathways.

EXAMPLE 10 : Seed production in different ADRl transgenic lines

In order to investigate whether disease resistance mediated by ADRl alters plant production, seeds from different transgenic lines were collected and weighted (Figure 27). adrl and 35S: :ADR1 plants that showed enhanced resistance exhibit a significant decrease in seed production. However, yield of TA: :ADR1 plants sprayed with 1 or 0.4 μM dexamethasone solution, treatments which confer disease resistance, was not significantly different to that of Col-0 wild type plants. In addition, yield of TA: : , TA : :ADR1 and TA: : plants sprayed with lμM dexamethasone solution was considered as controls and no differences compare to that of wild type plants were detected. Thus, the controlled expression of AQR1 can convey disease resistance in the absence of significant yield penalty.

References

Abe, H., Yamaguchi-Shinozaki, K., Urao, T., Iwasaki, T., Hosokawa, D. and Shinozaki, K. (1997) . Role of Arabidopsis MYC and MYB homologs in drought and abscisic acid-regulated gene expression. Plant Cell 9, 1859-1868.

Alber, T. Structure of the leucine zipper. Curr. Opin . Genet. Dev. 2:205-210, 1992. (Abstract)

Alvarez, M.E., Pennell, R.I., Meijer, P.J., Ishikawa, A., Dixon, R.A., and Lamb, C. Reactive oxygen intermediates mediate a systemic signal network in the establishment of plant immunity. Cell 92:773-784, 1998.

Aarts, M.g.M., Hekkert, B.T., Holub, E.B., Beynon, J. Stiekeme, W. J. and Pereira, A. (1998) . Identification of R-gene homologous DNA fragments genetically linked to disease resistance loci in Arabidopsis thaliana. Mol. Plant-Microbe Interact. 11, 251- 258.

Bent, A.F. Plant Disease Resistance Genes: Function Meets Structure. Plant CeJi 8:1757-1771, 1996.

Bevan et al . , (1998). Sequence of Arabidopsis chromosome 4. Nature 391, 485-488.

Bowler, C. and Fluhr, R. (2000) . The role of calcium and activated oxygen as signals for controlling cross-tolerance. Trends In Plant Sci . 5, 241-246.

Blecker et al . , (1988). Science 241, 1086-9.

Bowling, S.A., Guo, A., Cao, H., Gordon, A.S., Klessig, D.F., and Dong, X.I. A mutation in arabidopsis that leads to constitutive expression of systemic acquired-resistance. Plant Cell 6:1845-1857, 1994. Bowling, S.A., Clarke, J.D., Liu, Y.D., Klessig, D.F., and Dong, X.N. The cpr5 mutant of arabidopsis expresses both nprl- dependent and nprl-independent resistance. Plant Cell 9:1573-1584, 1997.

Cao, H., Glazebrook, J., Clarke, J.D., Volko, S., and Dong, X.N. The arabidopsis nprl gene that controls systemic acquired resistance encodes a novel protein containing ankyrin repeats. Ceil 88:57-63, 1997.

Clough, S.j. and Bent, A.F. (1998). Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidop[sis thaliana. Plant J. 16, 735-743.

Dietrich, R.A., Delaney, T.P., Uknes, S.J., Ward, E.R., Ryals, J.A., and Dangl, J.L. Arabidopsis mutants simulating disease resistance response. Cell 77:565-577, 1994.

Dixon, M.S., Jones, D.A., Keddie, J.S., Thomas, CM., Harrison, K., and Jones, J.D.G. The tomato cf-2 disease resistance locus comprises 2 functional genes encoding leucine-rich repeat proteins. Cell 84:451-459, 1996.

Ellis, J.G. Lawerence, G.J., Luck, J.E. and Dodds P.N. (1999). Plant Cell 11, 459-506.

Epple, P., Apel, K. and Bohlmann, H. (1997). Overexpression of an endogenous thionin enhances resistance of Arabidopsis against Fusarium oxysporum. Plant Cell 9, 509-520.

Feys et al . , (1994). Plant Cell 6, 751-9.

Gaffney, T., Friedrich, L., Vernooij, B., Negrotto, D., Nye, G., Uknes, S., Ward, E., Kessman, H., and Ryals, J. Requirement of salicylic acid for the induction of systemic acquired resistance. Science 261:754-756, 1993.

Gleave (1992). Plant Mol . Biol. 20, 1203-1207.

Greenberg, J.T., Guo, A.L., Klessig, D.F., and Ausubel, F.M. Programmed cell-death in plants - a pathogen-triggered response activated coordinately with multiple defense functions. Cell 77:551- 563, 1994.

Guzman and Ecker (1990). Plant Cell 2, 513-523.

Hanks, S.K. and Quinn, A.M. (1991). Protein kinase catalytic domain sequence database- identification of conserved features of primary structure and classification of family members. Methods in Enzy ol. 200, 38-62.

Hellens et al . , (2000). Plant Mol . Biol. 42, 819-832. Jabs, T., Dietrich, R.A., and Dangl, J.L. Initiation of runaway cell-death in an arabidopsis mutant by extracellular- superoxide. Science 273:1853-1856, 1996.

Jones, D.A. and Jones, J.D.G. (1997) . The role of leucine rich repeat proteins in plant defences, adv. Bot. Res. 24, 90-167.

Jones, D.A., Thomas, CM., Hammondkosack, K.E., Balintkurti, P.J., and Jones, J.D.G. Isolation of the tomato cf-9 gene for resistance to cladosporium-fulvum by transposon tagging. Science 266:789-793, 1994.

Keen, N.T. Gene-for-gene complementarity in plant-pathogen interactions. Annual Review Of Genetics 24:447-463, 1990.

Kiedrowski, S., Kawalleck, P., Hahlbrock, K. , Somssich, I. and Dangl, J. (1992) . Rapid activation of a novel plant defence gene is strictly dependent on the Arabidopsis RPM1 disease resistance locus. EMBO J. 11, 4677-4684.

Lawton, K.A., Friedrich, L., Hunt, M., Weymann, K., Delaney, T., Kessmann, H., Staub, T., and Ryals, J. Benzothiadiazole induces disease resistance in arabidopsis by activation of the systemic acquired-resistance signal-transduction pathway. Plant Journal 10:71-82, 1996.

Leister, R.T. and Katagiri, F. (2000) . A resistance gene product of the nucleotide binding site-leucine rich repeats class can form a complex with abacterial avirulence proteins in vivo. Plant J. 22, 345-354.

Loh, Y.T. and Martin, G.B. (1995). PTO bacterial resistance gene and the Fen insectide sensitivity gene encode functional protein-kinases with ser/thre specificity. Plant Physiol. 108, 1735- 1739.

Lupus, A. (1996). Coiled Coils: New structures and new functions. Trends Biochem Sci 21, 357-382.

Osbourn, A. (1996) . Preformed antimicrobial compounds and plant defence against fungal attack. Plant Cell 8, 1821-1831.

Pan, Q., Wendel, J. and Fluhr, R. (2000). divergent evolution of plant NBS-LRR resistance gene homologs in dicot and cereal genomes. J. Mol. Evol . 50, 203-213.

Parker, J.E., Holub, E.B., Frost, L.N., Falk, A., Gunn, N.D., and Daniels, M.J. Characterization of edsl, a mutation in arabidopsis suppressing resistance to peronospora-parasitica specified by several different rpp genes. Plant Cell 8:2033-2046, 1996.

Payne et al . , (1988). Plant Mol. Biol. 11, 89-94.

Rate, D., Cuenca, J., Bowman, G., Guttman, D. and Greenberg, J.T. (1999). Plant Cell 11, 1695-1708.

Salmeron, J.M., Oldroyd, G.E.D., Rommens, C.M.T., Scofield, S.R., Kim, H.S., Lavelle, D.T., Dahlbeck, D., and Staskawicz, B.J. Tomato-prf is a member of the leucine-rich repeat class of plant- disease resistance genes and lies embedded within the pto kinase gene-cluster. CeJI 86:123-133, 1996.

Saraste, M., Sibbald, P.R. and wittinghofer, A. (1990). The p-loop: Acommon motif in ATP- and GTP-binding proteins. Trends Bioche . Sci . 15, 430-434.

Scofield, S.R., Tobias, CM., Rathjen, J.P., Chang, J.H., Lavelle, D.T., Michelmore, R.W., and Staskawicz, B.J. Molecular- basis of gene-for-gene specificity in bacterial speck disease of tomato. Science 274:2063-2065, 1996.

Song, W.Y., Wang, G.L., Chen, L.L., Kim, H.S., Holsten, T., Wang, B., Ronald, P. (1995). A receptor kinase-like protein encoded by the rice disease resistance gene, Xa21. Science270, 1804-1806.

Tang, X.Y., Frederick, R.D., Zhou, J.M., Halterman, D.A., Jia, Y.L., and Martin, G.B. Initiation of plant-disease resistance by physical interaction of avrpto and pto kinase. Science 274:2060- 2063, 1996.

Traut, T. (1994). the functions and consensus motifs of nine types of peptide segments that form diffeent types of nucleotide- binding sites. Eur. J. Biochem. 222, 9-19.

Urao, T., Katagiri, T., mizoguchi, T., Yamaguchi, K., Shinozaki, K. (1994) . cDNA encoding calcium-dependent protein kinase. Plant Phsiol. 105, 1461-1462. van der Biezen, E. and Jones, J.D.G. (1998) . Current Biology 8, 226-227.

Weymann, K. , Hunt, M., Uknes, S., Neuenschwander, U., Lawton, K., Steiner, H.Y., and Ryals, J. Suppression and restoration of lesion formation in arabidopsis isd mutants. Plant Cell 7:2013-2022, 1995.

Whitham, S., Dinesh-Kumar S.P., Choi, D., Hehl, R., Coor, C and Baker, B. (1994). The product of the tobacco msaic virus gene N: Similarity to Toll and the interlukin-1 receptor. Cell 78, 1101- 1115 .

Weigel et al . , (2000). Plant Physiol. 122, 1003-1013.

Young, N.D. (2000). The genetic architecture of resistance. Curr. Opinion in Plant Biol. 3, 285-290.

Zou, H., Henzel, W.J., Liu, X., Lutschg, A. and Wang, X. (1997) . Apaf-1, a human protein homologous to Celegans CED-4, participates in cytochrome c-dependent activation of caspase-3. Cell 90, 405-413.

Claims

1. An isolated polynucleotide fragment which encodes a polypeptide, homologue or functional variant thereof from a plant, wherein said polypeptide comprises a coiled coil domain, a nucleotide binding site domain, a region containing leucine rich repeats, and a domain homologous to a calcium-dependent protein kinase domain, and wherein said polypeptide further confers resistance to abiotic stress or pathogen attack in said plant.

2. An isolated polynucleotide fragment according to claim 1, wherein said polynucleotide fragment comprises the sequence substantially as shown in SEQ ID No. 1, or SEQ ID No. 2, homologue or functional variant thereof, or sequence complimentary thereto.

3. An isolated polynucleotide fragment having at least 80% identity with said polynucleotide fragment (s) according to either of claims 1 or 2.

4. An isolated polynucleotide fragment which remains hybridised to said polynucleotide fragment (s) or complimentary sequences thereto according to either of claims 1 or 2 under stringent conditions.

5. The isolated polynucleotide fragment according to claim 4 which remains hybridised at a temperature of between 50°C and 70°C in single strength SSC containing 0.1% SDS.

6. A polypeptide encoded by a polynucleotide fragment (s) according to any one of claims 1 to 5.

7. A polypeptide comprising an amino acid sequence as substantially shown in SEQ ID No. 3, homologue or functional variant thereof.

8. A polypeptide having at least 80% similarity or 80% identity with the polypeptide according to claims 6 or 7.

9. An antibody specific to a polypeptide according to any one of claims 5 to 7.

10. An expression cassette comprising a promoter operably linked to a polynucleotide fragment according to any one of claims 1 to 5, or functional variant thereof from a plant.

11. A polynucleotide sequence comprising a transcriptional regulatory sequence, a sequence under the transcriptional control thereof which encodes an RNA sequence characterised in that the RNA sequence is antisense to a polynucleotide molecule according to any one of claims 1 to 5.

12. A polynucleotide sequence according to claim 11, wherein said antisense RNA is 20 - 2364 nucleotides in length.

13. A polynucleotide sequence according to claims 10 or 11, wherein said antisense RNA is from 50 - 500 nucleotides in length.

14. A biological vector comprising the expression cassette or polynucleotide sequence according to any one of claims 10 - 13.

15. A biological vector according to claim 14, wherein said vector is a recombinant biological vector which is capable of transforming a host cell.

16. A biological vector according to either of claims 14 or 15, wherein said vector is Agrobacterium tumefaciens.

17. A host cell stably transformed with the biological vector according to any one of claims 14 to 16.

18. A host cell according to claim 17, wherein said host cell is a bacterial cell or a eukaryotic cell

19. A host cell according to claim 18, wherein said eukaryotic cell is a plant cell or an insect cell.

20. A host cell according to any one of claims 17 to 19, wherein said host cell is further capable of expressing an isolated polynucleotide fragment according to any one of claims 1 to 5.

21. A transgenic plant comprising plant host cells as claimed in claim 19 or claim 20, wherein the progeny of said transgenic plant comprises the sequence stably incorporated and heritable in a Mendelian manner.

22. A transgenic plant comprising plant host cells, wherein said host cells contain an isolated polynucleotide fragment which encodes a polypeptide, homologue or functional variant thereof from a plant, wherein said fragment confers resistance to abiotic stress or pathogen attack in said plant

23. Use of any of the sequences substantially as shown in SEQ ID Nos . 1 or 2 in the production of transgenic eukaryotic cells .

24. Use according to claim 23, wherein said transgenic eukaryotic cells are resistant to abiotic stress or pathogen attack.

25. Use according to claim 24, wherein said transgenic eukaryotic cells are plant cells.

26. Use according to claim 25, wherein said transgenic plant cells are used in the production of transgenic plants.

27. Use according to claim 23, wherein said ADRl activity is substantially reduced through the use of a polynucleotide fragment which is antisense to a portion of the sequences substantially as shown in SEQ ID Nos. 1 or 2.

28. Use of an isolated polynucleotide fragment which encodes a polypeptide, homologue or functional variant thereof from a plant, for confering resistance to abiotic stress or pathogen attack in a second, different plant

29. A method of isolating a polynucleotide fragment, said polynucleotide fragment comprising a sequence having at least 50% similarity with SEQ ID Nos. 1 or 2, said method comprising:

(a) preparing a nucleotide probe capable of specifically hybridizing to a plant adrl related gene or mRNA, wherein said probe comprises a contiguous portion of the coding sequence for ADRl from Arabidopsis thaliana of at least 10 nucleotides in length;

(b) probing for other ADRl related coding sequences in populations of genomic DNA fragments or cDNA fragments from a chosen plant using the nucleotide probe prepared according to step (a) ; and

(c) isolating a polynucleotide fragment comprising a portion encoding a protein having ADRl-like activity.

30. A method of confering resistance to abiotic stress or pathogen attack in a plant, said method comprising:

(a) inserting a DNA sequence encoding a protein having ADRl or ADRl related activity into a plant host cell;

(b) growing the said transformed host cell in a suitable culture medium to form said plant; and

(c) expressing said DNA sequence to confer resistance to abiotic stress and pathogen attack in said plant.

31. A method for detecting the presence and form of an ADRl related gene and quantitating levels of ADRl related transcripts in an organism, said method comprising:

(a) making a polynucleotide probe having at least 10 nucleotides corresponding to part of the sequence substantially as shown as SEQ ID NO. 1 or 2, and

(b) amplifying and/or analysing ADRl coding sequences in said organism using the polymerase chain reaction.

32. A method according to claim 31, wherein said method is used to diagnose conditions which are associated with an altered form of ADRl or altered levels of expression in ADRl.

33. A method according to claim 32, wherein said method is used to diagnose conditions which are associated with abiotic stress or pathogen attack.

34. A method of producing a protein having ADRl or ADRl related activity in a host organism, said method comprising

(a) inserting a DNA sequence encoding a protein having ADRl or ADRl related activity into a host cell;

(b) growing the said transformed host cell in a suitable culture medium;

(c) expressing said DNA sequence to produce said protein; and

(d) isolating the protein product either from the transformed host cell or the culture medium or both and purifying it.

35. A method according to claim 34 wherein said host cell is a eukaryotic cell.

36. A method according to claim 36 wherein said host cell is a plant cell.