Inhibition of plant-pathogenic fungi by the barley cystatin Hv-CPI (gene Icy) is not associated with its cysteine-proteinase inhibitory properties.

The recombinant barley cystatin Hv-CPI inhibited the growth of three phytopathogenic fungi (Botrytis cinerea, Colletotrichum graminicola, and Plectosphaerella cucumerina) and the saprotrophic fungus Trichoderma viride. Several mutants of barley cystatin were generated by polymerase chain reaction approaches and both their antifungal and their cysteine-proteinase inhibitory properties investigated. Point mutants R38-->G, Q63-->L, and Q63-->P diminished their capacity for inhibiting papain and cathepsin B, retaining their antifungal properties. However, mutant C68-->G was more active for papain and cathepsin B than the wild type. These results indicate that in addition to the consensus cystatin-reactive site, Q63-V64-V65-A66-G67, the A37-R38-F39-A40-V41 region, common to all cereal cystatins, and the C68 residue are important for barley cystatin activity. On the other hand, the K92-->P mutant is inactive as a fungicide, but still retains measurable inhibitory activity for papain and cathepsin B. Against B. cinerea, the antifungal effect of Hv-CPI and of its derived mutants does not always correlate with their activities as proteinase inhibitors, because the Q63-->P mutant is inactive as a cystatin, while still inhibiting fungal growth, and the K92-->P mutant shows the reciprocal effects. These data indicate that inhibition of plant-pathogenic fungi by barley cystatin is not associated with its cysteine-proteinase inhibitory activity. Moreover, these results are corroborated by the absence of inhibition of intra- and extramycelia-proteinase activities by barley cystatin and by other well-known inhibitors of cysteine-proteinase activity in the fungal zymograms of B. cinerea.

Antibiotic activities of peptides, hydrogen peroxide and peroxynitrite in plant defence.

Genes encoding plant antibiotic peptides show expression patterns that are consistent with a defence role. Transgenic over-expression of defence peptide genes is potentially useful to engineer resistance of plants to relevant pathogens. Pathogen mutants that are sensitive to plant peptides in vitro have been obtained and a decrease of their virulence in planta has been observed, which is consistent with their hypothetical defence role. A similar approach has been followed to elucidate the potential direct anti-microbial role of hydrogen peroxide. Additionally, a scavenger of peroxynitrite has been used to investigate its involvement in plant defence.

Relative effects on virulence of mutations in the sap, pel, and hrp loci of Erwinia chrysanthemi.

We constructed strains of Erwinia chrysanthemi EC16 with multiple mutations involving three virulence systems in this bacterium, namely pel (coding for the major pectate lyases pelABCE), hrp (hypersensitive response and pathogenicity), and sap (sensitivity to antimicrobial peptides). The relative effects on virulence of those mutations have been analyzed on potato tubers and chicory leaves. In potato tubers, the sap mutation (BT105) had a greater effect in the reduction of the virulence than the pel (CUCPB5006) and hrp (CUCPB5039) mutations. This reduction was similar to that observed in the pel-hrp double mutant (CUCPB5037). The analysis of the strains affected in Pel-Sap (BT106), Hrp-Sap (BT107), and Pel-Hrp-Sap (BT108) suggested that the effects of these mutations are additive. In chicory leaves, the mutation in the sap locus appeared to have a greater effect than in potato tubers. The competitive indices of strains BT105, UM1005 (Pel-), CUCPB5039, and CUCPB5037 have been estimated in vivo and in vitro. These results indicate that the mutation in the hrp locus can be complemented in vivo by coinfection, whereas the mutations in pel and sap cannot.

Evidence against a direct antimicrobial role of H2O2 in the infection of plants by Erwinia chrysanthemi.

We have investigated the role of bacterial resistance to oxidative stress in pathogenesis. The oxyR gene from the pathogenic bacterium Erwinia chrysanthemi has been characterized. It is closely related to that found in Escherichia coli (88% overall amino acid identity). An E. chrysanthemi oxyR mutant strain was constructed by marker exchange. After induction with a sublethal dose of H2O2, this mutant was more sensitive to H2O2 and showed reduced levels of catalase and glutathione reductase activities, compared with the wild type. The oxyR mutant was unable to form individual colonies on agar plates unless catalase was added exogenously. However, it retained full virulence in potato tubers and tobacco leaves. These results suggest that the host-produced H2O2 has no direct antimicrobial effect on the interaction of E. chrysanthemi with the two plant species.

Inactivation of the sapA to sapF locus of Erwinia chrysanthemi reveals common features in plant and animal bacterial pathogenesis.

We investigated the role in pathogenesis of bacterial resistance to plant antimicrobial peptides. The sapA to sapF (for sensitive to antimicrobial peptides) operon from the pathogenic bacterium Erwinia chrysanthemi has been characterized. It has five open reading frames that are closely related (71% overall amino acid identity) and are in the same order as those of the sapA to sapF operon from Salmonella typhimurium. An E. chrysanthemi sap mutant strain was constructed by marker exchange. This mutant was more sensitive than was the wild type to wheat alpha-thionin and to snakin-1, which is the most abundant antimicrobial peptide from potato tubers. This mutant was also less virulent than was the wild-type strain in potato tubers: lesion area was 37% that of the control, and growth rate was two orders of magnitude lower. These results indicate that the interaction of antimicrobial peptides from the host with the sapA to sapF operon from the pathogen plays a similar role in animal and in plant bacterial pathogenesis.

Chromosomal location of PCR fragments as a source of DNA markers linked to aluminium tolerance genes in rye.

To identify and locate rye DNA sequences homologous to three wheat c-DNAs (wali1, wali2 and wali5) whose expression is induced by aluminium (Al) stress, we designed three pairs of specific primers. They were used in the amplification of genomic DNA from wheat-rye disomic addition lines. The wali2 pair of primers amplified a 878-bp rye DNA fragment (rali2) located on chromosomes 4R and 7R that showed 79.37% homology with the corresponding wheat c-DNA. RAPD fragments were also used as genetic markers. We located 22 different RAPDs distributed on 11 different rye chromosome arms using wheat-rye disomic and ditelocentric addition lines. Thirteen of these markers were located on the chromosomes 3R, 4R and 6R, which also carry aluminium-tolerance genes. The OPA08 415 and OPR01 600 RAPD markers, located on the 6RL and 6RS chromosome arms, respectively, were converted to SCAR markers (SCA08 415 and SCR01 600 ) and linked to Alt1 gene (SCR01 600 -2.1 cM-Alt1-33.5 cM-SCA08 415 ). We propose that the chromosomal location of RAPDs and SCARs using wheat-rye addition lines is a source of DNA markers linked to aluminium-tolerance loci and offers a valuable strategy in marker-assisted selection for the introgression of tolerance genes in wheat.

Mutants of Ralstonia (Pseudomonas) solanacearum sensitive to antimicrobial peptides are altered in their lipopolysaccharide structure and are avirulent in tobacco.

Ralstonia solanacearum K60 was mutagenized with the transposon Tn5, and two mutants, M2 and M88, were isolated. Both mutants were selected based on their increased sensitivity to thionins, and they had the Tn5 insertion in the same gene, 34 bp apart. Sequence analysis of the interrupted gene showed clear homology with the rfaF gene from Escherichia coli and Salmonella typhimurium (66% similarity), which encodes a heptosyltransferase involved in the synthesis of the lipopolysaccharide (LPS) core. Mutants M2 and M88 had an altered LPS electrophoretic pattern, consistent with synthesis of incomplete LPS cores. For these reasons, the R. solanacearum gene was designated rfaF. The mutants were also sensitive to purified lipid transfer proteins (LTPs) and to an LTP-enriched, cell wall extract from tobacco leaves. Mutants M2 and M88 died rapidly in planta and failed to produce necrosis when infiltrated in tobacco leaves or to cause wilting when injected in tobacco stems. Complemented strains M2* and M88* were respectively obtained from mutants M2 and M88 by transformation with a DNA fragment harboring gene rfaF. They had a different degree of wild-type reconstituted phenotype. Both strains retained the rough phenotype of the mutants, and their LPS electrophoretic patterns were intermediate between those of the wild type and those of the mutants.