CLONING OF TWO EXO-BETA-1,3-GLUCANASE ENCODING GENES OF PICHIA ANOMALA AND STUDY OF THEIR SEGREGATION IN RELATION TO P. ANOMALA ANTAGONISTIC PROPERTIES AGAINST BOTRYTIS CINEREA ON POSTHARVEST APPLES.
C GREVESSE1, MH JIJAKLI2 and P LEPOIVRE2.
1FNRS, 5, rue d'Egmont, B-1050 Bruxelles, Belgium; 2Unité de Phytopathologie, FUSAGX, 2, passage des Déportés, B-5030 Gembloux, Belgium (contact address).
Background and objectives
Botrytis cinerea (gray mold) is one of the most devastating pathogens of postharvest apples. Jijakli  isolated the yeast Pichia anomala strain K (PK) from the surface of an apple and demonstrated that it exhibits a high and reliable biocontrol activity against infection of B. cinerea on postharvest apples. Elucidation of the mechanisms of antagonism operative in biocontrol may be helpful in designing a screening test to enhance the antagonistic efficiency of biocontrol agents. An exo-beta-1,3-glucanase (exoglc1) purified from PK culture filtrate showed in vitro inhibitory effects on germinative tube growth of B. cinerea. An exo-beta-1,3-glucanase activity was detected on PK-treated apples and might be related to exoglc1 . However, experimental evidence of the possible role of exo-beta-1,3-glucanase activity is still missing. In this respect, we aim to study exo-beta-1,3-glucanase encoding genes in the genome of PK in relation to its biocontrol activity.
Materials and methods
Two PCR degenerate primers Pichia 1 (sens) and Pichia 3as (antisens) were designed from conserved amino acid regions found in exo-beta-1,3-glucanases of several fungi. The forward degenerate primer Pichia 7 corresponded to the N terminal sequence of exoglc1. Ascospores of the diploid PK were isolated by micromanipulation giving rise to haploid segregants. Exo-beta-1,3-glucanase activity was assayed in culture filtrates of the yeasts grown in presence of B. cinerea cell walls preparation by following the release of free glucose from laminarin with a glucose oxydase kit (Sigma). For biocontrol activity assays, wounds in apples were treated with each yeast, incubated for 24 h and inoculated with 50 µl of a B. cinerea conidial suspension.
Results and conclusions
Primers Pichia 1 and Pichia 3as gave rise to a 163 bp PCR fragment (PAEXG1a) from PK genomic DNA. A 390 bp DNA fragment (PAEXG2a) was also amplified from PK genomic DNA using the primers Pichia 7 and Pichia 3as. Both PCR products showed a significant similarity, at the deduced amino acid sequence level, with exo-beta-1,3-glucanases from other fungi. These results suggest that at least 2 exo-beta-1,3-glucanase encoding genes (PAEXG1 and PAEXG2) are present in the PK genome. This result corroborates the observation, with native gel detection, of two bands of exo-beta-1,3-glucanase activity of different intensity in the culture filtrate of PK grown with B. cinerea cell walls . The highest activity being produced by exoglc1 coded by PAEXG2, PAEXG1 could then code for an exo-beta-1,3-glucanase responsible for the less active band.
PAEXG1 and PAEXG2 were isolated by screening of the PK genomic library with P32 labelled PAEXG1a and PAEXG2a, and were sequenced. PAEXG1 and PAEXG2 comprise an open reading frame of 1494 pb and 1269 pb, respectively. At the deduced amino acid level, they share a 66 % similarity and present significant identities with exo-beta-1,3-glucanases from several yeasts.
Segregation of PAEXG1 and PAEXG2 in 10 haploid segregants was studied by Southern blots, using P32 labelled PAEXG1a and PAEXG2a as probes, in relation to their in vitro exo-beta-1,3-glucanase activity production and their in vivo protective activity against B. cinerea. All segregants showed an exo-beta-1,3-glucanase activity production equivalent to the production of the diploid strain (or even higher) and retains some significant biocontrol activity at a lower or equivalent level in comparison with the diploid strain. No relation was found between these properties and the segregation of PAEXG1 or PAEXG2 showing that either PK is homozygous at both loci and/or other genetic factors (genes or regulatory elements) are active in the protective effect. The implication of PAEXG1 and PAEXG2 in the antagonism will be studied in vivo via their disruption by integrative transformation in the genome of the haploid and/or diploid material.
 Jijakli M H and Lepoivre P, 1998. Phytopathology, in press.