Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, UK.

Background and objectives
Evidence for the presence, in pear fruit, of a proteinaceous inhibitor of fungal pectinases was first published in 1953. In recent years, the ability to clone genes from a range of fruit species has led to the isolation of genes encoding polygalacturonase-inhibiting proteins (PGIPs). The importance of these proteins in host-pathogen interactions is not fully understood, but it is known that endo-polygalacturonases (endo-PGs) released by some pathogenic fungi are involved in the breakdown of pectin, a major component of plant cell walls. Botrytis cinerea, the causal agent of grey mould in raspberry and several other fruit crops, releases two constitutive endo-PGs which are strongly inhibited by raspberry PGIP. A PGIP cDNA has been cloned from raspberry (Rubus idaeus) in this laboratory [1], and in common with the other PGIP genes that have been cloned, the bulk of the mature protein consists of 10 loosely conserved leucine-rich repeats (LRRs), and a number of potential glycosylation sites. The LRR units are thought to be involved in specific protein-protein interactions and are also found in other plant disease resistance genes [2]. The aim of this work was to express the raspberry PGIP cDNA in Escherichia coli to obtain PGIP that could be used to investigate binding and inhibition of the fungal endo-PG by unglycosylated protein. Such a system could be used to compare the efficacy of diverse PGIPs prior to transformation of crop plants to enhance disease resistance.

Materials and methods
Two expression systems for E. coli were used: the Qiagen Type IV system and the maltose binding protein (MBP) fusion system (New England Biolabs). Inserts for the expression constructs were made by PCR using pfu polymerase (Stratagene). Protein products from the Qiagen system incorporated the 6xHIS tag; these proteins were purified using HisTrap columns (Pharmacia). The MBP-PGIP fusion product was purified using amylose resin (New England Biolabs). Binding and inhibition of fungal endo-PGs was measured using published methods.

Results and conclusions
A series of deletions to the N terminus of the PGIP were engineered for expression using the Qiagen system. Four deletions were made, each resulting in a protein devoid of the 61-residue N-terminal portion and minus the first 1, 2, 3 or 4 of the LRRs. Significant accumulation of an induced protein product was obtained with each of these four constructs, but not when the full-length protein was expressed. Crude extracts of these cells were probed in Western blots using a polyclonal antibody raised using glycosylated raspberry PGIP. In each case, the induced protein product was recognized by this antibody, confirming expression of the mutant PGIP protein.

Stability of heterologous proteins in E. coli can be improved by fusion to an indigenous protein. Using the MBP system, two types of construct were made that localized the expression product in the cell cytoplasm or in the periplasmic space. Accumulation of an MBP-PGIP fusion product was apparent in both of the plasmid types. Protein levels were much higher in the cytoplasmic construct, an inducible product being visible in Coomassie-stained SDS-PAGE gels. As before, this induced product was recognized by the anti-PGIP polyclonal antibody.

This fusion product was purified in addition to the mutant forms of the PGIP, and we are attempting to investigate inhibition of fungal PGs, and the binding affinity between inhibitor and enzyme. These experiments may clarify the role of the carbohydrate groups in the inhibition process and also the effect of the removal of small parts of the mature PGIP, including some of the LRRS.

1. Ramanathan V, Simpson CG, Thow G et al., 1997. Journal of Experimental Botany 48, 1185-1193.
2. Jones DA, Jones JDG, 1997. Advances in Botanical Research (incorporating Advances in Plant Pathology), 89-167.