Mycopathologia (2005) 159: 53–63 © Springer 2005 53 Paecilomyces fumosoroseus (Deuteromycotina: Hyphomycetes) as a potential mycoparasite on Sphaerotheca fuliginea (Ascomycotina: Erysiphales) Miloslava Kavková1 & Vladislav Čurn2 1 University of South Bohemia, Faculty of Biology, Dept. of Botany, Branišovská 31, CZ-370 05 České Budějovice, Czech Republic; 2 University of South Bohemia, Faculty of Agriculture, Dept. of Genetic, Studentská 13, CZ-370 05 České Budějovice, Czech Republic Received 13 March 2003; accepted in final form 8 September 2003 Abstract Hyphomycete Paecilomyces fumosoroseus that is well known as saprophytic and entomopatogenic fungus was investigated for its mycoparasitism on the cucumber powdery mildew pathogen. Mycoparasitism was documented by using standard bioassay and SEM. Effects of mycoparasitism were evaluated in three types of experiments. Paecilomyces fumosoroseus was applied in the form of graded suspensions into a colony of powdery mildew on a leaf segment. Interaction between both fungi was observed as the percentage of colonized area vs. experimental time. In the second experiment, young cucumber plants were sprayed with a suspension of Paecilomyces fumosoroseus 24 h before inoculation of Sphaerotheca fuliginea. Pre-treatment with P. fumosoroseus reduced development and spreading of powdery mildew infection significantly 15 days post-inoculation in contrast to pre-treatments with sulfur fungicide and distilled water. The development of pure culture powdery mildew under determined experimental conditions was observed and compared with treated variants. In the third experiment, mildewed plants were treated with a suspension of P. fumosoroseus. The control treatments with sulfur fungicide and distilled water were tested. Effects of P. fumosoroseus on the dispersion of powdery mildew during a 21-day period were observed. P. fumosoroseus suppressed the development and spread of cucumber powdery mildew significantly during the time of the experiment. The mechanical and physical damages and disruptions of vegetative and fruiting structures of powdery mildew were recorded under light microscopy and S.E.M. Results were concluded in pursuance to differences between the natural behaviour and development of S. fuliginea on cucumber plants treated with P. fumosoroseus and non-treated plants. Key words: cucumber powdery mildew, mycoparasitism, Paecilomyces fumosoroseus, Sphaerotheca fuliginea Introduction Paecilomyces fumosoroseus (WIZE) BBROWN & SMITH (Deuteromycotina: Hyphomycetes) is a widespread species able to parasitize many different insect and non-insect hosts. It survives as a saprophyte and a decomposer of organic matter in soil [1, 2]. Neither the complete life cycle nor sexual form is well understood. Because strains of Paecilomyces fumosoroseus can parazitize species of insect such as whiteflies (Homoptera: Aleyrodidae), thrips (Thysanoptera: Thrip- idae), aphids (Homoptera: Aphididae) and continually reduce insect populations, it has been characterized as an entomopathogenic fungi [3–5]. Cucumber powdery mildew caused by FR) Sphaerotheca fuliginea (SCHLECTEND: POLLACCI (Erysiphales: Ascomycetes) is a widespread biotrophic mycoorganism causing diseases in the field and greenhouse crops from the family Cucurbitaceae. The infection process of host plants by S. fuliginea can be accomplished either by conidia or ascospores. Conidia cause extensive infections during 54 the growing season repeatedly. The germ tube and appressorium emerges from the conidium after the conidia have been in contact with the host surface for several hours. The penetration peg develops from the underside of the appressorium and penetrates the host cuticule and epidermal layer of the leaf. The haustoria as the infection structure are developed soon. The fine myceliar web of white threads spreads out of the infection site. Intensive production of appressoria and haustoria is allied with the growth and spreading of mycelia. Presently, the threads bear spores (conidia) that stand up on short chains. Conidia are one celled and appear hyaline, thin walled, ellipsoid to barrel shaped 25–35 × 14–20 µm [6]. When the leaf is disturbed (air circulation, rain . . . etc.), the chains of conidia disrupt and conidia spread onto other plants. Infected leaves become discoloured, brown and shrivelled. An optimal day/night temperature regime of 23◦ /17 ◦ C, a relative humidity of 75%, a photoperiod of 12 h are causal limits for the spreading of powdery mildew in greenhouse conditions [7–12]. S. fuliginea reduces significantly the assimilation area of the leaf and causes a disorder in water regime of epidermal cell. Mycoparasitism is common between all groups of fungi from simple chytrids to higher basidiomycetes. Many mycoparasites have been reported as potential biological control agents on the basis of laboratory experiments, but only a few have been exploited successfully for the biological control of diseases under greenhouse conditions. Several significant fungal organisms were documented as able to parazitize powdery mildews. Tilletiopsis spp. is a common phylloplane yeast belonging to the family Sporobolomycetaceae. Inoculation with Tilletiopsis cell suspension on detached mildew-infected cucumber leaves destroyed the superficial hyphae and conidial inoculum of the powdery mildew. The pathogenic microorganism Tilletiopsis spp. does not apparently penetrate the host fungus, Sphaerotheca fuliginea, but may produce fungistatic antibiotics [8, 13]. Epiphytic yeast-like fungi, Pseudozyna flocculosa (anamorph: Sporothrix flocculosa), is the parasite on the rose powdery mildew (Sphaerotheca pannosa rosae) [14]. Mycoparasitism of coleomycete Ampelomyces quisqualis on many hosts of order Erysiphales should be an intermediate point between unspecialised and direct parasitism. Ampelomyces quisqualis penetrates from cell to cell through the septal pores of the powdery mildew and continues to grow during the gradual degeneration of the infected cells [15, 16]. Polyphagous fungus, Verticillium lecanii, was also tested many times as potential mycoparasitic agent affecting the sporulation and development of cucumber powdery mildew [17–19]. The objective of this study was to determine relationships between P. fumosoroseus and S. fuliginea. The stages of life cycle of S. fuliginea as germination, mycelia net, beginning of sporulation, sporulation, spreading on inoculated leaf and spreading on plants finally was established as criterions simplified by transformation to the arbitrary index scale. Deviations in criterions caused by application of P. fumosoroseus and/or sulfur fungicide was compared with natural growth of S. fuliginea under determined conditions. The observation of interactions between both fungi was subjected by using light microscopy and scanning electron microscopy. Materials and methods Plant material Seeds of cucumber (Cucumis sativus) cv. Stela were soaked 72 h on wet filter paper in sterile plastic boxes and kept in a growth cabinet, maintained at 25 ◦ C with a 12/12 h dark/light regime, 145 E.m−2 .s−1 . Pregerminated seeds in the phase of cotyledon leaves were planted into pots (10 cm in diameter) containing peat potting mix. Seedlings were kept in abovementioned conditions until occurrence of first real leaf. Plants were watered until run-off every second day. During experiments, the temperature and humidity was monitored by small digital thermohygrometers (Tinytalk, Alfatronic Ltd., U.K.) placed in the growth cabinet (accuracy ± 1 ◦ C and ±5% RH). Mycoorganisms Sphaerotheca fuliginea was isolated from cucumber leaves in greenhouses where natural infection occurred. Pathogen was cultured on cotyledon leaves of cucumber cv. Stella in sterile petri dishes containing 2% agar and benzimidazole [10, 20]. Isolated S. fuliginea was re-cultured every tenth day by overprint on new sterile cotyledon leaves in sterile conditions. Petri dishes with inoculated leaves were stored under 12 h photoperiod at a temperature between 18◦ – 20 ◦ C. Viability of conidia S. fuliginea was checked as percentage of germinated conidia per 100 observed conidia. Conidia from fresh culture were tapped onto the glass slides covered with 1% water agar thin layer 55 and maintained in a sterile, glass humidity chamber (at relative humidity 85 ± 5%) in a climate room at 20 ◦ C [21]. Viability was checked after 24 and 48 h. Experimental plants were inoculated by overprint of cotyledon leaves with fully developed colony of S. fuliginea (10 days old colony). Paecilomyces fumosoroseus was applied as a suspension of conidia. The P. fumosoroseus isolate used in our study – PFR 97-2B-originated from Collection of Entomopathogenic Fungi, University of Florida, CFREC, Apopka. The conidia were obtained from P. fumosoroseus alginate prills. Alginate pellets of strain PFR 97-2B were activated on petri dishes with sterile filter paper moistened with sterile water (100 ml of water per 100 g of prills) and kept in a growing chamber (25 ◦ C, 24 h light) for 5–7 days until sporulation. Conidia, obtained from “sporulating” prills, were used to inoculate petri dishes with PDA (potatodextrose agar, Difco agar) medium. After 7 days fully sporulating culture suitable for experimental work was obtained. The suspension of conidia in sterile water with 0.01% Tween-20 was used in experiments. The growth development and quality (sporulation) of P. fumosoroseus was assessed in fungus development and growth index (FDGI) bioassay continually with every experiment [3, 22–24]. The concentration of conidial suspension was determined using an Improved Neubauer heamocytometer. The final concentration was adjusted properly before use. Chemicals SULIKOL K (Spolana Neratovice, Czech Republic) fungicide on the base of colloid sulfur, powder formulation and with recommended dosage 0.5% in the laboratory or 3 kg per 600 litres of water per 10.000 m2 in field conditions. Methods of observation Germinability and viability of conidia was controlled under Olympus microscope (magnification × 400). Olympus binocular was used for observation of interactions between both fungi on the leaf surface. The described bioassay for evaluation of mycoparazitic effect was visible under lens magnification from × 40 to × 70. Scanning electron microscopy Leaf segments from the leaf disk experiment were viewed using a SEM. Representative samples were vapor-fixed with 2% (wt/vol) osmium tetraoxide in distilled water for 20 h at room temperature, dried, and sputter-coated with gold palladium. Samples were kept in a desiccator until examination with a JEOL JSM-35CF scanning electron microscope. Leaf disk experiment Cucumber leaves infected by S. fuliginea were used for this experiment. Plants were inoculated with isolate of S. fuliginea 20 days before the start of experiment. Potted plants were maintained at 20 ◦ C in climate room required 12/12 h photoperiod of approximately 145 E.m−2 .s−1 . Leaf segments with one fully sporulating colony of powdery mildew in central position were cut by cork borer (20 mm in diameter). Leaf discs were put on petri dishes lined with 2% agar with benzimidazole. Five leaf discs were placed in each petri dish. Petri dishes were kept in the climate room at 20 ◦ C under 12/12 h day/ night conditions. Three replications of ten petri dishes were done. The drops (0.02 ml) of P. fumosoroseus suspensions (1 × 105 , 106 , 107 , 108 conidia/ml) were put in the middle of mildew colony on each leaf disc. P. fumosoroseus colonized areas of powdery mildew by were recorded daily under Olympus binocular. Colonized areas were evaluated using a categorical scale: 0 = 0% represented leaf segments with colony of powdery mildew without presence of P. fumosoroseus, 1 = 0–25% of mildewed area was colonized by P. fumosoroseus, 2 = 25–50%, 3 = 50–75%, 4 = 75– 100%. Mildewed leaf segments treated with a drop of sterile distilled water were used as a control variant. Bioassay on plants pre-treated with P. fumosoroseus The aim of this experiment was to evaluate the influence of P. fumosoroseus on the development and spreading of S. fuliginea on plants. Interactions between P. fumosoroseus and S. fuliginea were evaluated according to an index scale (Table 1). The index scale was established on base of the asexual life cycle. The asexual life cycle of S. fuliginea, from conidia via conidiogenesis to dispersion on the whole plant, was a principle of established index scale. The course of infection and dispersion of S. fuliginea on plants were simplified under several limiting points described in Table 1. Suspensions of conidia (107 conidia in 1 ml) 56 were sprayed using a laboratory standard sprayer on set of ten plants with fully expanded first real leaf until run-off. A set of control plants was sprayed with distilled water. Fungicide based on inorganic sulfur SULIKOL K (0.5%) was applied on another ten plants for comparison of final effect. The trial was repeated three times. Inoculation by cucumber powdery mildew followed 24 h after the application of P. fumosoroseus. Inoculum, pure culture of S. fuliginea, was obtained by method described above. Fully sporulating colony of powdery mildew on cotyledons was tapped onto the adaxial surface of the first true leaf. Each plant was inoculated by new colony of powdery mildew. Conidia were detached in the same manner to 20 ml of sterile water containing 0.01% (vol/vol) Tween 80. Following 50 replicate counts of 0.5 ml drops using an Improved Neubauer heamocytometer (Sigma) was possible to ascertain that the inoculum contained 5×103 ±7×102 conidia in 1 ml of suspension per 1 cm2 [25]. The first observation was assessed 24 h after application of S. fuliginea and continued daily for 20 days. Development of S. fuliginea on plants was evaluated according to established bioassay (Table1). The supplement information about behaviour of pathogens on leaves of treated plants provided microscopic analysis of leaf segments. Experiments were performed three times with a minimum of three replicates treatments. Data were tested for normality and heterogeneity of variance. The percentage of parasitized S. fuliginea in leaf disk experiment was analysed by arcsine-square root transformation before analysis of variance (ANOVA) to improve homogeneity of variance. Data from each experiment were subjected to ANOVA. Based on consistency of the sequential experiments, factorial analyses of variance [26] were structured over the replicate experiment (indicated as block). The effect of pre-treatment on development of S. fuliginea was analysed by ANOVA ANOVA – factorial design, (STATISTICA for Windows 6.0) followed by Tukey’s HSD mean separation test α ≤ 0.05) to declare the difference and dependency on developmental scale of S. fuliginea on pre-treated plants by P. fumosoroseus, sulfur and distilled water in the time of experiment (20 days). A decision to analysed the data by using analyses of variance was inspired by works of Yang [27], Newton et al. [28], Askary et al. [12], and Verhaar et al. [29]. The effect of post-application of P. fumosoroseus, sulfur and combined treatment was subjected to ANOVA followed by Tukey HSD mean separation test too. Bioassay on mildewed plants post-treated by P. fumosoroseus Results The cucumber plants with first real leaf were inoculated by S. fuliginea by inoculum from cotyledon leaves. Plants were potted in containers (25 cm in diameter) and kept under conditions mentioned above. When the infection site was fully sporulating, the suspension of P. fumosoroseus (107 conidia/1 mL) was sprayed on plants until run-off. One set concluded ten plants. Treatments with inorganic sulfur (0.5%) SULIKOL K, and sulfur combined with P. fumosoroseus suspension were used in comparison with P. fumosoroseus variant. Plants infected by pure culture of S. fuliginea represented the control variant. Development and spreading of S. fuliginea on plants was observed during a 21 days period. The ability of P. fumosoroseus to affect development and spreading of powdery mildew on plants was expressed as total number of leaves and number of infected leaves. In the experimental design ten plants were used per set, and trials were repeated three times. Experimental design and Statistical Analyses Leaf disks experiment All of the concentrations tested caused total colonization of mildewed leaf segments without significant difference on the last day of the experiment. In spite of this, during the course of the experiment some significant difference among tested concentrations became evident between the sixth to the eighth day. Development of colonization is present in Figure 1. Data are modified by ArcSin transformation to improve its homogeneity and subjected to ANOVA, Statistica 6 Software. The percentage of colonized area vs. experimental time increased according to enhanced concentration of used suspension, although significant difference was noted between the 105 conidia per 1 ml of suspension, 106 conidia per 1 ml of suspension and the rest of the tested variants (F(3,587) = 43, 59; p ≤ 0.0E, Post Hoc Tukey HSD). 57 Table 1. The index scale of S. fuliginea development 2–3 Dispersion on plant 1–2 Sporulation on infection spot 0–1 Formation of mycelia net 3 2.5 2 2 1.5 1 1 0.5 0 Dispersion on plant New sides of infection on infected leaf Sporulation – conidia Sporulation – conidia Formation of conidiophores Net of mycelia Net of mycelia Initial formation of mycelia Conidia Figure 1. The effect of scale dilution Paecilomyces fumosoroseus (leaf disk experiment) on colonization of Sphaerotheca fuliginea at 20 ◦ C. Points represent means of values scored on an arbitrary scale (see methods) modified by ArcSin data transformation. (ANOVA, Tukeys HSD, p ≤ 0.05). Colony of powdery mildew was parazitized by P. fumosoroseus from 75 to 100% after the eighth day of the experiment. Pre-treatment of plants by P. fumosoroseus Pure culture of S. fuliginea developed according to an assessed bioassay from index 1 to index 3. The inoculated area of each leaf was fully sporulated from day six under the described conditions. New sites of infec- tion and dispersion of diseases were noted from day nine until the end of experiment. The course of infection of S. fuliginea resulted from values of data from ten observations including three replicates. Replicates did not have a significant effect on obtained values either in Table 2. or in Figure 2. (F(2,324) = 2.12; p ≤ 0.123 ≥ α). Formation of conidiophores was visible after four days of observation. The colony of S. fuliginea sporulated fully from day six. Infection 58 Table 2. Effect of pre-treatment on development of S. fuliginea on cucumber plants. Averages of scale index with standard error of means (Tukey’s HSD test for mean separation, p ≤ 0.05) describes development of S. fuliginea during 20 days on plants pre-treated with distilled water (CPM), sulfur (SULIKOL) and P. fumosoroseus at 20 ◦ C. Values followed by the same letters are not significantly different (homogenous group) (ANOVA, F(6,324) = 10.63; p ≤ 0.0 E ≥ α Post Hoc Tukey’s HSD, p ≤ 0.05) Tested agents No. of days CPM 5th day 10th day 15th day 20th day 1.50 ± 0.066b 2.40 ± 0.066cde 2.68 ± 0.066e 2.95 ± 0.100e SULIKOL P. fumosoroseus 0.55 ± 0.208a 2.00 ± 0.106c 2.40 ± 0.076cde 2.93 ± 0.100e 0.57 ± 0.076a 1.80 ± 0.105c 2.17 ± 0.100cd 2.50 ± 0.076e was spreading throughout the plants from day ten of experiment. Development of S. fuliginea on plants pre-treated by P. fumosoroseus was limited by treatment from the beginning of experiment. In comparison, the net of mycelia (1 on arbitrary scale) was observed on day six whereas pure culture of S. fuliginea sporulated fully. The extension of time period was noted between germinating (formation of parasitic structure) and formation of mycelia net from the beginning of infection. Discontinuance of development was noted from day 12 to 14. There was achieved index 2 – sporulation of inoculated area whereas natural development of S. fuliginea was 2.5 and 3 – spread on whole plant. Development of S. fuliginea attained index value 2.5 on the end of experiment repeatedly meaning that spreading of pathogen on inoculated leaf. Pre-treatment by sulfur influenced development of S. fuliginea like to pre-treatment with P. fumosoroseus till day nine approximately. Next progress of infection was similar to growth of pure culture of S. fuliginea. The critical values for development of S. fuliginea were established on day five, ten, fifteen and twenty. Table 2 presents overview of situation on critical days. Pre-treatment with sulfur and P. fumosoroseus affected development of S. fuliginea negatively on the same level of significance from the beginning until tenth day of the experiment. On day fifteen, the extent of infection achieved similar values on plants pretreated with sulfur and control plants. P. fumosoroseus reduced development of S. fuliginea significantly. Development of S. fuliginea was noted to be almost the same on the last day of the experiment irrespectively of various pre-treatments. Table 3. Development of cucumber powdery mildew on treated plants at 20 ◦ C. Mean number of leaves per plant (F(3,107) = 11.21, p ≤ 0.00E ≤ 0.05) and mean number of infected leaf (F(3,107) = 21.72, p ≤ 0.00E ≤ 0.05) are given with ± s.e.m. Control agents No. of leaves per plant No. of infected leaves S. fuliginea control P. fumosoroseus SULIKOL K SULIKOL K P. fumosoroseus 5.13 (± 0.67)a 4.83 (± 0.73)a 5.63 (± 0.91)b 5.60 (± 0.71)b 3.41 (± 0.43)b 2.13 (± 0.70)a 3.83 (± 0.49)b 2.79 (± 0.50)c Values followed by the same letters are not significantly different (homogenous group). Pre-treatment with sulfur showed suppressive effects on S. fuliginea until the tenth day. In the third experiment, P. fumosoroseus suspension, sulfur and P. fumosoroseus combined with sulfur was applied on inoculated plants when the site of infection fully sporulated. P. fumosoroseus and P. fumosoroseus with sulfur suppressed infection of S. fuliginea on plants significantly opposite to control variant and pure sulfur. There was no significant difference found between three replications of experiment F(8.107) = 1.074, p ≤ 0, 387 ≤ 0.05. The number of emerged leaves during these 21 days differed significantly. Plants treated with P. fumosoroseus and control plants produced less number of leaves in average than plants treated with sulfur and P. fumosoroseus with sulfur (Table 3). The number of infected leaves on plants treated with P. fumosoroseus and P. fumosoroseus with sulfur was significantly less, opposite to control and treatment with pure sulfur. The percentage of infected leaves to total number of leaves was found 68% in variant treated with sulfur and 66% in control variant. Treatments with P. fumosoroseus constrained infection on the level of 44% and 46% when combined with sulfur. Discussion Results showed that P. fumosoroseus should be suitable agents for suppression of cucumber powdery mildew considering environmental conditions and behaviour of isolate. Set of concentrations of P. fumosoroseus provided the same effect at least although larger area of powdery mildew was colonized soon when 107 , 108 conidia in 1 ml of suspension were used. Compare to practical use of mycoparazite as for example Verticillium lecanii concentrations moved from 59 Figure 2. The effect of pre-treatment on development of Sphaerotheca fuliginea at 20 ◦ C. The curves show development of S. fuliginea within 20 days under different pre-treatment. Particular points represent means of scale index value correspond with day of observation. Statistica 6 Software: ANOVA (F(2.324) = 2.12; p ≤ 0.123 ≥ α). 106 to 107 [19]. Environmental conditions and the origin of isolate also play important role. P. fumosoroseus grows and multiplies at temperatures between 15 ◦ C and 30 ◦ C, with colony growth optimal between 23 ◦ C and 25 ◦ C. Germination of conidia and growth of mycelium decline above 25 ◦ C and cease above 32 ◦ C [23]. Development of cucumber powdery mildew as a biotrophic organism is limited by several factors including quality of host plant and environmental conditions. The test of germination and fungal development and growth index was done for P. fumosoroseus in conditions, that are specific for powdery mildew −20 ◦ C, >75% of relative humidity and light/dark conditions 12/12 h. According to these conditions, the germination of P. fumosoroseus conidia on 2% agar slide attained 98.5% in 24 h and conidiogenesis was achieved within the fourth day. Germination of powdery mildew conidia on a 2% agar slide achieved 85% within 24 h and 88.5% within 48 h; on the leaf it was 89% during 24 h and 92% during 48 h. One pre-treatment with P. fumosoroseus did not eliminate infection of powdery mildew desirably, but the reduction was opposite to control significant during limited period of experiment. We can suppose that repeated treatment with P. fumosoroseus during critical phase of powdery mildew development (index scale 1.5–2) could provide favourable results. The increase of temperature to optimal level 25 ◦ C after application of P. fumosoroseus can also support parasitic behaviour against S. fuliginea. In comparison, appropriate timing of biocontrol treatment application of Verticillium lecanii was important to achieve good control [19]. Formulation of P. fumosoroseus spores in oil medium can improve the situation [30]. The development of S. fuliginea was retarded on plants 60 Figure 3A–F. 61 Figure 3G, I, J, K Figure 3. Scanning electron micrographs of interactions between S. fuliginea and P. fumosoroseus. A, Conidia of S. fuliginea colonized by mycelia of P. fumosoroseus. Hyphae of P. fumosoroseus creates loops and deforms the conidia of S. fuliginea. Magnification × 5,000, bar = 5 µm. B, Hyphae of P. fumosoroseus adjacent to conidia of S. fuliginea. Magnification × 4,000; bar = 5 µm. C, Chain of conidia attached by hyphae of P. fumosoroseus. Magnification × 3,500; bar = 10 µm. D, P. fumosoroseus sporulates on collapsed conidia of S. fuliginea. Magnification × 3,500; bar = 10 µm. E, Conidia of P. fumosoroseus adjacent to the hyphae of S. fuliginea. Magnification × 4,500; bar = 5 µm. F, Hyphae of P. fumosoroseus creates a fine web of filaments around collapsed conidia of S. fuliginea. Magnification × 4,000; bar = 5 µm. G, Hyphae of S. fuliginea are turgid with well-developed conodiophores. Magnification × 900; bar = 50 µm. I. The crust, P. fumosoroseus covers the colony of powdery mildew by thick myceliar layer when presence of free water occurred or humidity increased. In this case, P. fumosoroseus never sporulates. Magnification × 1,000; bar = 20 µm. J, Sporulation of P. fumosoroseus on conidia of S. fuliginea. Magnification × 1,200; bar = 20 µm. K, Conidia of S. fuliginea attached by hyphae of P. fumosoroseus. Magnification × 4,000; bar = 5 µm. 62 inoculated by P. fumosoroseus but in spite of it the cucumber powdery mildew was able to disperse and colonize new space on plant. These new colonies were always contaminated by P. fumosoroseus. There are two explanations. Dispersed conidia can transport propagules of P. fumosoroseus on its surface and (or) P. fumosoroseus can survive and disperse on the plant. Both possibilities were also examined, but it is another chapter of this study. Post- application of P. fumosoroseus on infected plants showed significant retardation of spreading and development of powdery mildew continually with development of plants. Repeated treatment in period shorter than 20 days can increase the control effect of P. fumosoroseus against S. fuliginea. In conclusion, the pre-application of P. fumosoroseus showed that development of S. fuliginea after treatment was retarded. The pathogen keeps the basic character of development, but deviates from index values depending on the form of treatment. The most important index value of bioassay is the interval 1.5–2.0, start of sporulation and full sporulation into inoculated spot. In the case of spraying plants with P. fumosoroseus suspension, the critical level was reached on the seventh day and it culminated on the ninth day. The spread of conidia of P. fumosoroseus on the leaf surface could have a straightforward relationship with conidia of powdery mildew. There is some unknown mechanism, which kept the powdery mildew on a low level necessary for reproduction and limited spreading on the host. P. fumosoroseus colonizes mycelium and conidia of S. fuliginea and it is able to sporulate repeatedly on host fungi. Nevertheless a small part of the parazitized powdery mildew is able to re-infect new sites on plants. Detailed microscopic observation showed that P. fumosoroseus colonized myceliar structures and detached conidia of S. fuliginea on the leaf surface as a first. The mycelia of P. fumosoroseus created “tent” like structures of aerial mycelia when they contacted higher structure as upright chains of conidia and leaf trichomes. Hyphae of P. fumosoroseus twisted the host structures and branched freely. Mycelia of P. fumosoroseus showed to be very adaptive to host surface. The different thickness of hyphae is noted in Figures 3B and 3F. The structures of S. fuliginea attached by P. fumosoroseus collapsed and lost the structural integrity. Disrupted conidia of S. fuliginea were covered by sporulating mycelia of P. fumosoroseus (Figure 3J). In the case that water had condensed in the wet chamber, P. fumosoroseus covered structures of S. fuliginea by crust (Figure 3I). This phenomenon is also undermined by conditions (humidity, free water on the surface of leaves, temperature) and its combination. The exact effect of conditions on behaviour of fungi is not clear. We observed that in case of creating of a crust on alive plants the powdery mildew never dispersed on a leaf or plant opposite to a sporulating P. fumosoroseus on colony of powdery mildew (Figure 1D, J). Even though the phenomena of mycoparazitism was mentioned the principle of mycoparazitical relationship between both of fungi is not clear. There are many questions such as chemical attraction, nutritional dependence, ecto- or endo-parasitism and others aspects couplet with phenomena of mycoparasitism. The mycoparazitical effect depends also on source of fungal isolates. It entails much research and using special techniques to the study in this diverse system, plantobligate biotrophic fungus and polyphagous fungus, as mycoparasitical agents. In comparison with other studies we observed some similarity with well-known Verticillium lecanii. We hope that this study can open new approaches and ideas to the study of Hyphomycetes and its ecology, life cycles, and practical use in biological control. Acknowledgements The authors thank Dr. Jana Nebesářová and her colleagues for producing the scanning electron micrographs. References 1. Samson R. Paecilomyces and some allied hyphomycetes. Studies in Mycology 1974; 6: 1–43. 2. Samson RA, Rombach MC. Biology of the Verticillium and Aschersonia. In: Hussez, N.V., scopes, N.E.A., Biological pest control – the glasshouse experience. Cornell University Ithaca Press, N.Y. 1985; 32–42. 3. Osborne LS, Landa Z. Biological control of whiteflies with entomopathogenic fungi. Florida entomologist 1992, 75(4): 456–471. 4. Bolckmans K, Sterk G, Eyal J., Sels B. Stepman, W. PreFeRal, (Paecilomyces fumosoroseus strain Apopka 97), a new microbial insecticide for the biological control of whiteflies in greenhouses, Mededelingen – Faculteit Landbouwkundige en Toegepaste Biologishe Wetenschappen, Universiteit Gent. 1995; 60: 707–711. 5. Lacey LA, Kirk AA, Millar L, Mercardier G, Vidal C. Ovicidal and larvicidal activity of conidia and blastospores of Paecilomyces fumosoroseus (Deuteromycotina: Hyphomycetes) against Bemisisa argentifolii (Homoptera:Aleyrodidae) with description of a bioassay system allowing prolonged survival of control insects. Biocontrol Sci. Techn. 1999; 9: 9–18. 63 6. Ellis MB, Ellis JP. Fungi specific to various herbaceous plants. In: Microfungi on land plants. The Richmond Publishing Co. Ltd., 1997: p. 867. 7. Sundheim L. and Admundsen T. Fungicide tolerance in the hyperparasite Ampelomyces quisqualis and integrated control of cucumber powdery mildew, Acta Agri Scand. 1982; 32: 349. 8. Hijwegen T. Effect of culture filtrates seventeen fungicolous fungi on sporulation of cucumber powdery mildew. Neth. J. Pl. Path. 1989; 95: 95–98. 9. Hijwegen, T. Biological control of cucumber powdery mildew with Tilletiopsis minor under greenhouse conditions. Neth. J. Pl. Path. 1992; 98: 221–225. 10. Lebeda A. The genera and Species spectrum of cucumber powdery mildew in Czechoslovakia. Phytopath. Z. 1983; 108: 71–79. 11. Daayf F, Schmitt A, Bélanger R. Evidence of phytoalexins in cucumber leaves infected with powdery mildew following treatment with leaf extracts of Reynoutria sachalinensis. Plant Physiol. 1997; 113: 719–727. 12. Askary H, Benhamou N, Brodeur J. Ultrastructural and cytochemical investigations of the antagonistic effect of Verticillium lecanii on cucumber powdery mildew. Phytopathology 1997; 87: 359–368. 13. Hoch HC, Prowidenti R. Mycoparasitic relationships 5., Cytology of the Sphaerotheca fuliginea – Tilletiopsis sp. interaction, (Cause of powdery mildew of cucumber, biological control), ill. 15 ref. Phytopathology (USA) 1979; v.69 (4): 359–362. 14. Bélango RR, Deacon JW. Interaction specificity of the biocontrol agent Sporothrix flacculosa: a video microscopy study. Phytopathology 1996; 86: 1317–1323. 15. Elad Y, Kirshner B, Sztejnberg A. Management of powdery mildew and gray mold of cucumber by Trichoderna harzianum T39 and Ampelomyces quisqualis AQ10. Biocontrol 1998; 43: (2) 241–251. 16. Rotem Y, Yarden O, Sztejnberg A, The mycoparasite Ampelomyces quisqualis expresses exgA encoding and exo-beta- 1,3 glucanase in culture and during mycoparasitism Phytopathology 1999; 89: (8) 631–638. 17. Verhaar MA, Hijwegen T, Zadoks JC. Glasshouse experiments on biocontrol of cucumber powdery mildew (Sphaerotheca fuliginea) by mycoparasites Verticillium lecanii and Sporothrix rugulosa. Biol Control 1996; 6: 353–360. 18. Askary H, Carrière Y, Bélanger R.R, Brodeur J. Pathogenicity of the fungus Verticillium lecanii to aphids and powdery mildew Biocontrol Science and technology 1998; 8: 23–32. 19. Verhaar MA, Hijwegen T. Efficient production of phialoconidia of Verticillium lecanii for biocontrol of cucumber powdery mildew, Sphaerotheca fuliginea. Neth. J. Pl. Path. 1997; 99(2): 101–103. 20. Lebeda A. Screening of wild Cucumis species for resistance cucumber powdery mildew (Erysiphe cichoracearum and Sphaerotheca fuliginea). Sci. Hort. 1984; 24: 241–249. 21. Celio GJ, Hausbeck MK. Conidial germination, infection structure formation, and early colony development of powdery mildew on Poisettia. Phytopathology1998; 88: 105–113. 22. Landa Z, Osborne L, Lopez F, Eyal J. A bioassay for determining pathogenicity of entomopathogenous fungi on whiteflies. References Biological Control – theory and applications in pest management 1994; 4: 341–350. 23. Vidal C, Lacey LA, Fargues J. Intraspecific variability of Paecilomyces fumosoroseus; effect of temperature on vegetative growth. J Invertebr Pathol 1997; 70: 18–26. 24. Vidal C, Fergues J, Lacey LA, Jackson MA. Effect of various liquid culture media on morphology, growth, propagule production, and pathogenic activity to Bemisia argentifolii of the entomopathogenic Hyphomycete, Paecilomyces fumosoroseus. Mycopathologia 1998; 143: 33–46 25. Wurms K, Labbé C, Benhamou A, and B,langer RR. Effects of Milsana and Benzothiazole on the ultrastructure of powdery mildew haustoria on cucumber. Phytopathology 1999; 89(9): 728–736. 26. Gomez and Gomez, 1984. 27. Yang, 1995. 28. Newton et al. 1997. 29. Landa Z, Bohatá, A. Compatibility of entomogenous fungus Paecilomyces fumosoroseus with natural insecticides based on azadirachtin and Neem Oil. Collection of Scientific Papers, Faculty of Agriculture in Ceske Budejovice, Series for Crop Science, 1999 16(2): 99–106. 30. Barnett HL, Binder FL. The fungal host-parasite relationship. Annu Rev Phytopathol 1973; 11: 273–292. 31. Lebeda A. Methods of testing vegetable crops for resistance to plant pathogens 1986; 1: 87–90. 32. Gams, W. CBS Course of mycology 3rd edition. Centraal Bureau voor Schimmel Cultures, Baarn, Institute of the Royal Netherlands Academy of Arts and Science 1987; p. 120. 33. Sundheim L. Control of cucumber powdery mildew by the hyperperasite Ampelomyces quisqualis and fungicides. Plant Pathol 1982; 31: 209–214. 34. Urquhart JE, Menzies JG, Punja ZK. Growth and biological control activity of Tilletiopsis species against powdery mildew (Sphaerotheca fuliginea) on greenhouse cucumber. Phytopathology 1994; 84: 341–351. Address for correspondence: M. Kavkova, Faculty of Biology, Department of Botony, University of South Bohemia, Branisovska 31, 37005 Ceske Budejovice, Czechia E-mail: kavkova@hotmail.com