The modes of action of the antagonistic yeast Pichia anomala (strain K) have been studied; however, thus far, there has been no clear demonstration of the involvement of exo-beta-1,3-glucanase in determining the level of protection against Botrytis cinerea afforded by this biocontrol agent on apple. In the present study, the exo-beta-1,3-glucanase-encoding genes PAEXG1 and PAEXG2, previously sequenced from the strain K genome, were separately and sequentially disrupted. Transfer of the URA3-Blaster technique to strain K, allowing multiple use of URA3 marker gene, first was validated by efficient inactivation of the PaTRP1 gene and recovery of a double auxotrophic strain (uracil and tryptophan). The PAEXG1 and PAEXG2 genes then were inactivated separately and sequentially with the unique URA3 marker gene. The resulting mutant strains showed a significantly reduced efficiency of biocontrol of B. cinerea when applied to wounded apple fruit, the calculated protection level dropping from 71% (parental strain) to 8% (mutated strain) under some experimental conditions. This suggests that exo-beta-1,3-glucanases play a role in the biological control of B. cinerea on apple. Furthermore, biological control experiments carried out in this study underline the complexity of the host-antagonist-pathogen interaction. Two experimental parameters (yeast inoculum concentration and physiological stage of the fruit) were found to influence dramatically the protection level. Results also suggest that, under some conditions, the contribution of exo-beta-1,3-glucanase to biological control may be masked by other modes of action, such as competition.

This study should contribute towards improving the biocontrol efficacy of T. asperellum strains used against P. megakarya. Integrated into a broader study of the impact of environmental factors on the biocontrol agent-pathogen system, this work should help to build a more rational control strategy, possibly involving the use of a compatible adjuvant protecting T. asperellum against desiccation.

The objective of this work was to develop validated models predicting the 'in vitro' effect of a w and temperature on the radial growth of Botrytis cinerea. The growth rate (g, mm d− 1 ) of B. cinerea was calculated at three incubation temperatures (25°C, 15°C, 5°C) and six water activities (ranging from 0.995 to 0.890). The water activity was adjusted with glucose, NaCl, glycerol, or sorbitol. Statistical analysis showed a significant effect of temperature, solute, a w , and their two-and three-way interactions on the growth rate. No growth was observed at a w = 0.93 in the presence of NaCl or at 0.89 in the presence of a non-ionic solute. The maximum colony growth rate decreased when the incubation temperature and water activity was lowered. Secondary models, relating the colony growth rate with a w or a w and temperature were developed. Optimum a w values for growth ranged from 0.981 to 0.987 in glycerol-, sorbitol-, or glucose-modified medium and were close to 1 in NaClmodified medium. A quadratic polynomial equation was used to describe the combined effects of temperature and a w on g (mm d − 1 ) in the presence of each solute. The highest and lowest radial growth rates were observed in models based on glucose and NaCl respectively, whatever the incubation temperature. All models prove to be good predictors of the growth rates of B. cinerea within the limits of experiments. The quadratic polynomial equation has bias factors of 0.957, 1.036, 0.950, and 0.860 and accuracy factors of 1.089, 1.070, 1.120 and 1.260 in media supplemented with glucose, NaCl, glycerol and sorbitol respectively. The results from modelling confirm the general finding that a w has a greater influence on fungal growth than temperature.

In vitro effects of water activity, temperature and solutes on the growth rate of P. italicum Wehmer and P. digitatum Sacc. IntroductionCitrus fruits are generally stored in temperatures ranging from 0 to 7°C and relative humidities from 85% to 90%. During their storage, fruits are affected by several decays (Brown and Miller 1999). Penicillium digitatum Wehmer and P. italicum Sacc. are the main postharvest pathogens of citrus fruits, responsible, respectively, for green and blue decay. In turn massive spore production by these two pathogens ensures their presence wherever fruit is produced, handled and stored. Contamination takes place only through wounds where nutriments are available to AbstractAims: To evaluate the effect of water activity (a w 0AE98-0AE89, adjusted with glycerol, sorbitol, glucose, or NaCl) and temperature (5-25°C) on the lag phase and radial growth rate (mm day )1 ) of the important citrus spoilage fungi, such as Penicillium italicum and Penicillium digitatum grown in potato dextrose agar (PDA) medium. To select, among models based on the use of different solutes, a model fitting accurately the growth of these species in relation to a w and temperature. Methods and Results: Extensive data analyses showed for both Penicillium species a highly significant effect of a w , temperature, solutes and their interactions on radial growth rate (P < 0AE0001). Radial growth rate was inhibited and the lag phase (i.e. the time required for growth) lengthened as the a w of the medium decreased. NaCl appeared to causes the greatest stress on growth when compared with other nonionic solutes. Penicillium italicum stopped growing at 0AE96 a w and P. digitatum at 0AE93 a w . Under the dry conditions where growth was observed, P. italicum grew faster than P. digitatum at low temperature and P. digitatum remained more active at ambient temperature. Multiple regression analysis applied to the square roots of the growth rates observed in the presence of each solute showed that both the 'glycerol model' and the 'sorbitol model' yielded a good prediction of P. italicum growth and the 'sorbitol model' gave an accurate fit for P. digitatum growth, offering high-quality prediction within the experimental limits described. Conclusions: Mathematical models describing and predicting, as a function of a w and temperature, the square root of the radial growth rate of the agents responsible for blue and green decays are important tools for understanding the behaviour of these fungi under natural conditions and for predicting citrus fruit spoilage. Significance and Impact of the Study: Implementation of these results should contribute towards a more rational control strategy against citrus spoilage fungi. spore germination. Fruit decay thus begins at the site of injury. Blue decay develops less rapidly than green decay under ambient conditions. Green decay is often observed in mixed infections. Blue decay is more common in fruit held in cold storage during the summer; it can spread through packed cartons more easily than ...

Pichia anomala (strain K) is an efficient biocontrol agent against post-harvest diseases affecting apples. To study the role of strain K genes in biocontrol activity, it is useful to identify selectable markers on which to base a gene disruption strategy. The Pichia anomala TRP1 gene (

Aims:To obtain information about the genomic organization of Pichia anomala (strain K) and about its genomic diversity at species and intraspecies level. Methods and Results: The PFGE karyotype of strain K was composed of four bands ranging in size from 1AE1 to 3AE2 Mb. The number of chromosomes was estimated at six since bands 2 and 3 seemed to result from the comigration of two chromosomes with similar size. A comparison of strain K and Hansenula wingeii migration profiles led to the estimate of K strain genome size at 11AE7 Mb. Comparison with isogenic strains, resulting from the sporulation of strain K, highlighted some major karyotypic differences. Two segregants (KH6 and KH7) showed supernumerary chromosomes and one (KH9) displayed chromosomal length polymorphism. This genomic instability was confirmed by molecular hybridization with four probes, consisting of URA3, LEU2, PAEXG1 and PAEXG2 genes of P. anomala. URA3 and LEU2 probes showed second hybridization signals on supernumerary chromosomes of strain KH7 and on chromosome 6 of strain K for LEU2 only. Karyotypic comparison of seven nonisogenic P. anomala strains revealed chromosomal length polymorphism, a sign of intraspecies variation. Conclusions: This work has supplied information about genome size and chromosome number of strain K of P. anomala. The strain seems to be aneuploid because of the presence of supernumerary chromosomes and additional hybridization signals for URA3 and LEU2 probes in the chromosomal profile of some segregants. The work also highlighted genomic diversity within the P. anomala species. Significance and Impact of the Study: Results obtained here increase information about the aneuploidy of P. anomala (strain K). Information about the genomic diversity of the segregants will be of great interest for further studies on strain K mode of action. The genome size and chromosomal profile of P. anomala presented here are different from the results obtained elsewere for Hansenula anomala, while Hansenula is included as a synonym of Pichia. This warrants further studies to investigate this taxonomic relationship.

a b s t r a c tAureobasidium pullulans strains Ach 1-1 and 1113-5 are two effective biocontrol agents against Botrytis cinerea and Penicillium expansum on stored apples. In the present work, a monitoring system allowing their identification and quantification was developed. The methodology used consisted of the development of both molecular markers and a semi-selective medium. The random amplified polymorphic DNA (RAPD) technique was applied to a collection of 15 strains of A. pullulans, including Ach 1-1 and 1113-5. Five specific RAPD fragments were amplified for strain Ach 1-1 and three others for strain 1113-5. Among them, a fragment of 528 bp specific to strain Ach 1-1 (generated with the OPR-13 RAPD primer) and another one of 431 bp specific to strain 1113-5 (amplified with the OPQ-03 RAPD primer) were selected, cloned, sequenced, and used to design sequence-characterized amplified region (SCAR) primers. Three different SCAR markers were amplified: two specific to strain Ach 1-1 (189 bp and 387 bp) and one specific to strain 1113-5 (431 bp). These SCAR primers can clearly identify strains Ach 1-1 and 1113-5 among 14 strains of A. pullulans and among eight yeast strains commonly present on apple fruit surfaces. Their selectivity was also tested using DNA extracted from epiphytic microflora of the apple surface. As a semi-selective medium, PDA medium supplemented with 0.5 mg L −1 euparen, 1 mg L −1 sumico, 2.5 mg L −1 hygromycin B, 30 mg L −1 streptomycin sulphate, and 1 mg L −1 cycloheximide was selected. It inhibited the development of the air microflora and appeared highly toxic for the epiphytic microflora of apple surface without altering the growth of the targeted strains Ach 1-1 and 1113-5. The combination of the semi-selective medium and SCAR markers provides a valuable monitoring tool to specifically identify and quantify A. pullulans strain Ach 1-1 and strain 1113-5 and could be used in future studies to evaluate their population dynamics under various laboratory and practical conditions.