Antifungal Substances Produced by Penicillium frequentans and Their Relationship to the Biocontrol of Monilinia laxa

Ecology and Epidemiology Antifungal Substances Produced by Penicilliumfrequentans and Their Relationship to the Biocontrol of Monilinia laxa A. De Cal, E. M.-Sagasta, and P. Melgarejo Department of Plant Pathology, E.T.S. Ingenieros Agr6nomos, Madrid, Spain; Department of Plant Protection, C.I.T.-I.N.I.A., Madrid, Spain. This research was supported in part by a grant of the Comisi6n Asesora de Investigaci6n Cientifica y T6cnica (Proyecto No. 0375-84). A. De Cal was in receipt of a Beca del Plan de Formaci6n de Personal Investigador from the Ministerio de Educaci6n y Ciencia, Spain. We thank Professor F. Garcia-Arenal and Professor A. Fraile for critically reading the manuscript. Accepted for publication 16 December 1987. ABSTRACT De Cal, A., M.-Sagasta, E., and Melgarejo, P. 1988. Antifungal substances produced by Penicilliumfrequentans and their relationship to the biocontrol of Monilinia laxa. Phytopathology 78:888-893. Antibiotic substances produced by an isolate of Penicilliumfrequentans, germination of spores and the germ tube growth of M. laxa were 0.31 a component of the resident mycoflora of peach twigs previously found to mg/ ml and 0.02 mg/ ml, respectively, for antibiotic A and 4.84 mg/ ml and antagonize Monilinia laxa, were produced in potato-dextrose broth and 0.13 mg/ml, respectively, for antibiotic B. Both antibiotics were isolated and partially purified from the cell-free medium by liquid:liquid temperature-stable, while the activity of antibiotic A was lowered at basic partitions. Two active substances were partially characterized by thin-layer pH. Antibiotic B was also more stable over time at room temperature chromatography, infrared absorbance, and qualitative analysis and were than antibiotic A, conserving 30% of its activity against the germ tubes of named antibiotic A and antibiotic B. Both antibiotics were active against a M. laxa after 70 days. Both antibiotics showed significant inhibition wide range of plant pathogenic fungi including M. laxa and several (p •<0.05) of M. laxa on peach twigs. components of the mycoflora of peach twigs. The ED 5o values for the Additional key words: antibiosis, brown rot disease, competition, phyllosphere. Monilinia laxa (Aderh et Ruhl) Honey causes brown rot disease Whatman No. I filter paper in a Biichner funnel and by of stone fruit crops with economically important losses. The centrifugation at 15,000 g for 15 min. The culture filtrate was then control of the disease is not satisfactory, and its treatment presents sterilized by filtration through 0.22-jm Millipore membranes. The significant problems (16). Biological control of the fungus may be toxicity of these crude filtrates toward the germination of spores an alternative to chemicals. Penicilliumfrequentans Westling is a and the germ tube growth of M. laxa was bioassayed. component of the resident mycoflora of peach twigs and flowers in Isolation and partial purification of antibiotics. Crude central Spain. The potential for biocontrol of M. laxa with P. preparations of antibiotics were obtained from crude filtrates of frequentans has recently been shown both in the laboratory (9) and 20-day-old stationary cultures of P. frequentans by liquid:liquid in peach tree orchards (10). Based on preliminary data, an partitions as described by Birkinshaw (1) with modifications (Fig. antibiosis mechanism was postulated for the antagonism between 1). At each purification step, relative antibiotic activity was P. frequentans and M. laxa (11). bioassayed on the germination of spores and the germ tube growth The purpose of the present study was to investigate the of M. laxa and Cladosporium cucumerinum Ellis & Arth. C. production of antifungal compounds by the isolate of P. cucumerinum was bioassayed for antibiotic activity as it allows the frequentans isolated by us from peach twigs (9) and its involvement localization of active bands in chromatograms, since M. laxa in the antagonistic action of P. frequentans against M. laxa. cannot grow over the silica gel plates. Partial characterization of antibiotics. The active substances MATERIALS AND METHODS were partially characterized by thin-layer chromatography (TLC), infrared absorbance (IR), and qualitative analysis. Silica gel plates Fungal cultures. The 909 isolate of P.frequentans was originally were developed in equilibration tanks containing acetone:water taken from the phyllosphere of peach twigs (9) at an experimental (9:1, v/v) and examined under UV light. The plates were then orchard in Madrid, Spain. The monosporic isolate of M. laxa was sprayed with SbCl 3 saturated in chloroform and heated to 100 C collected from a commercial apricot orchard in Almonacid de la for 10 min (6,12). Duplicate plates were bioassayed with C. Sierra (Zaragoza, Spain). Both fungi were stored on potatocucumerinum. The IR spectra were obtained in chloroform with a dextrose agar (PDA) slants at 5 C and were grown on PDA at Hilger D 209 (Chemical Faculty, Universidad Complutense, 20-25 C in the dark for conidial and mycelial production. Madrid, Spain) spectrometer. Production of antifungal compounds by P. frequentans. To test Qualitative analyses were carried out as described in Curtis et al for optimal production of antifungal compounds by the (4) and Birkinshaw (1) by assaying several specific chemical antagonist, conical flasks (250 cm 3) containing 150 ml of potatoreactions: reduction of AgNO 3 in NH 3 solution, decoloration of dextrose broth (PDB) were each inoculated with three disks of Br2 in C1 4C solution, precipitation with NaHSO 3, and Brady's test mycelium, 8 mm in diameter, cut from the edges of actively (12). growing colonies of P. frequentans. Flasks were incubated in the Bioassays. I) In vitro: The toxicity of the crude filtrates, the dark at 25 ± 1 C either in a stationary incubator or in a rotary different fractions from the purification procedure (Fig. 1) and of shaker incubator at 150 rpm for 10, 20, and 30 days. Media were the aqueous solutions of crude antibiotics on the germination of separated from the mycelia and spores by filtration through spores, and the growth of germ tubes of M. laxa and C. cucumerinum was assayed in sterile Czapek broth (2 g of KNO 3, 0.5 g of MgSO 4 "7 H2 0, 1 g of KH 2 PO 4, and 30 g of sacarose in © 1988 The American Phytopathological Society 1,000 ml of distilled water). Sterile glass slides were placed on glass 888 PHYTOPATHOLOGY 15-mm-diameter petri dishes lined with moist filter paper. On a Effects of time, temperature, and pH on antibiotic activity. To slide, a 15-Al droplet of spore suspension on Czapek broth was test the relative loss of activity of the crude antibiotics against M. mixed with a 30-e1 droplet of each treatment. Spore concentration laxa with time, aqueous solutions (8.4 mg/ ml) were maintained in and incubation time were I X 106 spores per milliliter and 8 hr for the laboratory at room temperature for different time periods. M. laxa and 1 X 106 spores per milliliter and 24 hr for C. To test the influence of temperature and pH on the activity of the cucumerinum. Controls and concentrations of extracted fractions antibiotics, crude antibiotic A (1.5 mg/ml) or crude antibiotic B were fixed in each experiment. Results were analyzed by analysis of (6.8 mg/ml) was maintained for 30 min at temperatures ranging variance and Duncan's multiple range test. Percentages of from 25 to 120 C, or their pH was adjusted from pH I to 10 with 0.1 inhibition were treated by arc sine transformation to achieve N HCl or 0.1 N NaOH before performing bioassay against M. laxa. homogenization of variances prior to statistical analysis. The Antifungal spectrum of antibiotics. The crude antibiotics were results of the relative toxicity of different concentration levels of tested for activity against the fungi listed in Table I by the method crude antibiotics on M. laxa were processed by the probit-analysis described in McKeen et al (8) with modifications: 1OOAI1 of 105_106 method (5). The effective doses (DE) were calculated in milligrams spore suspensions, or when necessary, fragmented mycelial per milliliter. A test was done for parallelism according to the suspensions of the fungi were spread on the PDA surface before relative potency estimation method (5). When potency ratios of placing a sterile Whatman No. I filter paper disk (15 mm in two lines were very similar, a studentized range test was performed diameter) imbibed with 30 /1 of diethyl ether solutions of each (15) to confirm whether both lines were different, antibiotic (5.47 mg/ml) on the center of the plate. All fungi were The toxicity of compounds was also tested on C. cucumerinum tested four times. Control plates contained 30 Il1 of diethyl ether in to locate active bands in chromatograms. The plates were sprayed the filter paper. Plates were incubated in the dark at 25 ± I C. with a dense spore suspension of the fungus in Czapek broth and Inhibition zones of the treated plates were measured after clearly incubated for 48-72 hr at 25 ± I C in the dark under high humidity visible fungal lawns had covered the control plates. conditions. 2) In vivo: Healthy peach twigs, collected from active trees, RESULTS surface sterilized as recommended by Sauer and Burroughs (13), were wounded by gently abrading their cuticle. Then, 5-Al-droplets Production of antifungal compounds by P. frequentans. Isolate of different treatments (aqueous solutions of crude antibiotic A, 909 of P. frequentans, incubated in stationary regime in PDB, 1.48 mg/ ml; crude antibiotic B, 6.7 mg/ ml; benomyl, 0.25 mg/ ml; produced antifungal compounds that inhibited (p < 0.05) the spores of P. frequentans, I X 109 spores per milliliter, or distilled germination of spores and the germ tube growth of M. laxa (Table 2). water in the controls) were applied to the wounds. Twigs were The production ofthese compounds was detected at 10 days by the incubated for 8 hr at 20-25 C under high humidity conditions. inhibition of the germ tube growth of M. laxa. Maximal Droplets were then collected and mounted with lactophenol blue concentration of antifungal compounds in the filtrates was to determine the toxicity of treatments to spore germination and achieved at 20 days of incubation, when crude filtrates strongly germ tube growth of M. laxa. inhibited both the germination of spores and the germ tube growth All bioassays had four replicates, and independent assays were of the fungus. At 30 days, inhibition of M. laxa by the crude done at least twice. Fifty conidia or 25 germ tubes were considered filtrates was decreased. No antifungal effect was observed in crude in each replicate. A spore was considered germinated when its germ filtrates of the fungus when grown in shaking system. tube was 4 ym. Isolation and partial purification of antibiotics. Bioassays of the Crude filtrate Partition with chloroform (1:1) Twice (1) Aqueous fraction (2) Interface (3) Organic fraction Partition with distilled water (5:1) pH 11 with 1N-NaOH (4) Organic fraction Aqueous fraction (Crude antibiotic B) Acidification to pH 3-4 with lN-HC1 Partition with diethyl ether (5) Organic fraction (6) Aqueous fraction (Crude antibiotic A) Fig. 1. Procedure of isolation and purification of antibiotics. Vol. 78, No. 7,1988 889 relative antibiotic activity of different fractions (Fig. 1) obtained the crude filtrate (Table 3). Only fractions 1 and 2 were inactive from the extraction of the crude filtrate of 20-day-old stationary (Table 3). cultures of P.frequentans showed that fractions 3, 4, and 5 were the Inhibition of germination of spores and germ tube growth of C. most toxic (p < 0.05) to M. laxa, and they were more active than cucumerinum followed the same pattern as that of M. laxa (data not shown). Partial characterization of antibiotics. Fractions 3, 4, and 5 TABLE 1. Antibiotic spectrum of antibioticsa (Fig. 1), the most toxic towards M. laxa (Table 3), were partially characterized by specific chemical reactions and IR spectra. Inhibition zonec Chemical reactions showed the presence of 1-3 dienes in Fungi" Antibiotic A Antibiotic B Control fractions 3, 4, and 5 (decoloration of Br 2 in CC1 4 solution and a continuous purpureal-violet band parallel to the front at Rfvalues Pathogens of 0.96-0.98 on TLC silica gel plates treated with SbCI 3 saturated Ascochytapisi Lib.d+ - in chloroform). Acetone groups were also detected (Brady's test Botrytis cinerea Pers. ex Fr.' ... ++ - positive and obtaining of a white precipitate with NaHS03) in Cladosporium cucumerinum Ellis & Arth + ++ - these fractions. However, only fractions 3 and 5 reduced AgNO3 in Fusarium oxysporum Schlecht. NH 3 solution, a test specific for the aldehyde group. Based on the emend. Snyd. & Hansd + + - intensity of these reactions, fraction 3 was judged to contain a Fusarium solani (Mart.) Appel & mixture of fractions 4 and 5. Wollenw. emend. Snyd. & Hans 4 +++ .++ + Fractions 4 and 5 were further characterized by IR spectroscopy Monilinia laxa (Aderh et Ruhl) (data not shown). There were strong similarities between the Honey ++ ++ - spectra of the two fractions. Bands due to alcoholic hydroxyl Penicillium expansum Link ex F.S. groups occurred at 3,400 cm1 in both spectra, being more intense Rhizoctonia solani Kiihnd _ .. in fraction 4. Also, a band due to C = 0 stretching modes occurred Verticillium dahliae Klebahn' ++ ++ - in the 1,600-1,650 region. However, marked differences appeared in the region 1,000-1,200 cm-1. In this region absorptions due to Components of the mycoflora of peach twigs vibrational modes involving deformation of alcoholic OH groups Alternaria alternata (Fr.) Keissler, --- Aspergillusflavus Link ex Fries ... ... -occur, and while three fairly strong bands are present in both Cladosporium herbarum (Pers.) spectra (near 1,080 cm-, near 1,110 cm-', and near 1,150 cm1) Link ex S.F. Gray + + - variations in this frequency range were apparent. Epicoccum nigrum Link + ++ - Thus, two different active compounds were mainly present in Mucorpusillus Lindt - - - fractions 4 and 5. We named antibiotic B the one in fraction 4 and Nigrospora sphaerica (Sacc.) Mason --- antibiotic A the one in fraction 5. Nitchkea sp. ... ... - Antifungal properties. The germination of spores and the Penicillium purpurogenum Stoll --- growth of germ tubes of M. laxa were affected by antibiotics A and Phoma glomerata (Corda) Wr. & Hochapf ... ... + B. Figure 2A and B shows the probit-log concentration lines and Rhizopus stolonifer (Ehrenb. ex the ED 50 values for each antibiotic to the germination of spores and Fr.) Lind. _ -- the germ tube growth of M. laxa. We found quite different values Sordaria sp. --- for the slopes of the four probit lines, and it is convenient to test for Trichoderma viride Pers. +++ ... - parallelism to compare the lines. The test for parallelism showed Trichothecium roseum (Pers.) Link + ++ - that data could be re2Presented by parallel probit-log concentration a30 l ofcrudeantibioticAandcrudeantibioticBdissolvedindiethylether regression lines (X values for parallelism not significant with to give a concentration of 5.47 mg/ ml was assayed. The assay medium was 1 df in all cases). Figure 2C-F shows: antibiotic B is more toxic potato-dextrose agar. Control treatment was 30 j I of diethyl ether. A (p •< 0.05) to the germination of spores of M. laxa than antibiotic 15-mm-diameter Whatman No. I filter paper imbibed with test solutions A, while the germ tube growth of M. laxa is more sensitive (p was placed on the center of plates. 1 0.05) to antibiotic A than to antibiotic B, and the toxicity of To produce a lawn in the assay plate, 100 #1 of 105-106 spore suspensions antibiotic A to the germ tube growth of M. laxa is significantly or fragmented mycelial suspensions per milliliter were used as inoculum. greater (p •< 0.05) than to the germination of spores of this fungus, _: no inhibition, +: little inhibition, ++: medium inhibition, ++: while the toxicity of antibiotic B is similar for spores and germ tube high inhibition. w th. dCulture obtained from J. Tello, Department of Plant Protection, I.N.I.A., growth. Madrid, Spain. The antibiotics were also toxic to M. laxa in detached peach eCulture obtained from I. Palaz6n, Department of Plant Protection, twigs under laboratory conditions (Table 4). Both antibiotics S.I.A., Zaragoza, Spain. significantly inhibited (p <0.05) the spore germination and the TABLE 2. Toxicity of culture filtrates of Penicilliumfrequentans on Monilinia laxa Crude filtrates Germination from different Spores Germ tube growth age cultures Incubation germinatedy Inhibition' Germ tube lengthy Inhibitionz (days) system (%) (%) (Am) (%) Control (potatodextrose broth) 85 ± 3.8 "" 80 ± 5.8 "" 10 Stationary 83 ± 4.6 ... 65 ± 6.3 19 a Shaking 86 ± 3.9 ... 78 ± 1.7 "" 20 Stationary 37 ± 1.5 57 a 4 ± 0.0 95 b S haking 80 ± 4.5 "" 83 ± 5.8 "" 30 Stationary 62 ± 3.8 27 b 31 ± 2.7 62 c Shaking 87 ± 2.7 ... 76 ± 3.4 "" 'Incubation systems were in a stationary incubator or in a rotary shaker incubator at 150 rpm. 'Mean and standard deviation of four replicates; 50 conidia or 25 germ tubes were considered in each replicate. 'All values shown were different from that of control (p < 0.05). Means followed by the same letters are not significantly different by Duncan's multiple range test (p < 0.05). 890 PHYTOPATHOLOGY germ tube growth of the pathogen, and their activity was similar to compounds present in the crude filtrates of P. frequentans, isolate that obtained with the commercial fungicide benomyl. However, a 909. This was shown by testing the biological activity of the crude concentrated spore suspension of P. frequentans inhibited (p filtrates and of the fractions 3, 4, and 5 from the extraction •<0.05) M. laxa but to a lesser degree (Table 4). procedure (Fig. 1) against C. cucumerinum on TLC silica gel The loss of activity of both antibiotics against M. laxa began to plates. Only active bands parallel to the front at Rf values of be significant (p < 0.05) at 14 days (Fig. 3). At this time, the activity 0.96-0.98 were observed. These bands were also the only ones that of antibiotic A and antibiotic B on the germination of spores of the we observed in duplicate plates when sprayed with SbC1 3 . fungus decreased by 51 and 44%, respectively, and by 28 and 29%, The partial characterization of the antibiotics shows that respectively, on the growth of germ tubes. At 70 days, antibiotic A antibiotics A and B should be two substances closely related in had lost almost all of' its activity (90-95%), while antibiotic B structure with 1-3 dienes, acetone, and alcoholic hydroxyl groups conserved 30% of its activity on the germ tube growth of M. laxa in their molecules, being the presence of an aldehyde group in (Fig. 3). antibiotic A the main difference between them. An antifungal and Activity of the antibiotics was not affected by changes of weakly antibacterial substance called frequentin was described as a temperature in a water bath or after autoclaving (data not shown). product of an unusual strain of P.frequentans found abundantly in Antibiotic B was also not affected by changes of pH, while acid heath soils (4). Recently, another isolate of P. frequentans maximum inhibition (95-100%) of spore germination and germ from olive fruits was shown to produce frequentin and was highly tube growth of M. laxa by antibiotic A was obtained at pH 1-5. Its antagonistic to Geotrichum candidum (12). In liquid cultures of activity then decreased significantly (p <0.05), showing at pH 7, 64 these P. frequentans isolates, another secondary metabolite (2), and 66% of inhibition of spore germination and germ tube growth, palitantin, was described, with no appreciable antifungal activity. respectively, and 13 and 26%, respectively, at pH 10. Birkinshaw (1) obtained crude frequentin and palitantin by a The spectrum of action of the antibiotics is quite wide, since both liquid:liquid partition procedure similar to that followed here to inhibited the growth of 14 out of 22 fungi tested (Table 1). The antibiotics A and B (Fig. 1). The IR spectra of antibiotics A and B fungi tested included important pathogens and the inhibition of has similarities to those given in Curtis and Duncanson (3) for growth obtained with the antibiotics against these fungi was frequentin and palitantin, respectively. Thus, antibiotics A and B similar to that against M. laxa (Table 1). The other fungi tested seem to be frequentin and palitantin, respectively, although at were components of the mycoflora of peach twigs (9). Some of present we cannot assure this and further work on their them were also inhibited by both antibiotics while others were not purification and characterization now in progress is needed. (Table 1). Whatever the antibiotics were, we demonstrate here their toxicity to M. laxa and 14 other fungi (Fig. 2, Table 1). The DISCUSSION inhibition of the germination of spores of M. laxa by antibiotic B was significantly higher than by antibiotic A (Fig. 2). Antibiotic B The production of antifungal compounds against M. laxa by P. also presents an uncommon characteristic in being as toxic to frequentans in PDB in stationary regime begins about 10 days after spores as to germ tube growth of M. laxa. The ED 50 values of the incubation, at which time the crude filtrate is toxic to the germ tube antibiotics for the germination of spores and for the germ tube growth of the pathogen (Table 2). Production continues until growth of our isolate of M. laxa are higher than the ED 50 values of approximately 20 days, when inhibition reaches a maximum. benomyl to M. fructicola (I pg/ml) (14) or vinclozolin and Then, production may or may not stop but, in any case, the activity iprodione to M. laxa (5 ,g/ ml) (7). However, we cannot forget that of the crude filtrate decreases. This can be attributed to a we are working with crude antibiotics and their purified forms diminished rate of synthesis, or to an increased rate of metabolism could be more or less active. of the antibiotics by the fungus, or a natural breakdown as in The antibiotics were also toxic to M. laxa in detached peach Figure 3. Alternatively, the change in activity of the crude filtrate twigs under laboratory conditions in the same degree as benomyl could also be related to changes in pH, as we have observed that the (Table 3). However, under the same conditions, spores of P. pH of the filtrate of 20-day-old cultures is approximately 5, and frequentans inhibited the pathogens to a lesser degree, probably that of 30-day-old cultures (and of shaking cultures) is higher than because they have not had time enough to produce the antibiotics. 7. Also, we have shown that the toxicity of antibiotic A is pH We can postulate that the mechanism of antibiosis should be dependent. Oxidative reactions could also explain the lack of efficient in the biocontrol of M. laxa by P.frequentans. However, activity of filtrates from shaking cultures. our evidence for the role of antibiosis in nature is obviously Antibiotics A and B were the only significant antifungal tentative, pending the technically difficult task of isolating the antibiotics at inhibitory concentration levels from fungal treated twigs. This work is also in progress. TABLE 3. Toxicity of different compounds on Monilinia laxa Germination Germ tube growth Spores Germ tube TABLE 4. Toxicity of different treatments on Monilinia laxa in peach twigs germinated' Inhibition' length' Inhibition' Germination Germ tube growth Compoundx (%) (%) (Am) (%) Control 80 ± 3.6 "" 32.3 ± 2.8 "" Spores Germ tube Culture filtrate 52 ± 2.0 34 a 12.9 ± 3.3 60 a Treatment germinated Inhibitin length nhibitin Fractions 1 82 ± 2.0 ... 35.7 ± 4.3 ... _MM__ _ _ _ _ _ M 2 81 ± 1.7 ... 28.2 ± 1.7 Control 88 ± 3.27 ... 63.39 ± 7.55 "" 3 2 ± 0.5 97 c 0.0 ± 0.0 100 c Antibiotic A 17 ± 6.53 89 a 19.63 ± 8.22 69 a 4 2 ± 0.4 97 c 0.0 ± 0.0 100 c Antibiotic B 34 ± 2.00 61 a 41.40 ± 1.58 50 a 5 31 ± 3.5 61 d 9.2 ± 0.5 71 d Spore'suspension 6 70 ± 2.2 12 e 18.3 + 4.6 43 e of Penicillium 'Compounds are: culture filtrate and different fractions obtained from the frequentans 70 ± 6.39 21 b 55.98 ± 2.89 12 b liquid: liquid partition procedure described in Figure 1. Concentration of Benomyl 24 ± 8.41 72 a 11.50 ± 1.02 82 a compounds in the crude filtrate were maintained in fractions. Control was 'Treatments are : Antibiotic A (1.48 mg/ml), Antibiotic B (6.7 mg/ml), sterile distilled water. benomyl (0.25 mg/ml), and aqueous spore suspension of P. frequentans 'Mean and standard deviation of four replicates; 50 conidia or 25 germ (I X 109 spores/ml). tubes were considered in each replicate. 'Mean and standard deviation of four replicates; 50 conidia or 25 germ 'All reported differences from control are significant (p <,0.05). Means tubes were considered in each replicate. followed by the same letters are not significantly different (p < 0.05) by 'Means followed by the same letters in each column are not significantly Duncan multiple range test. different (p •<0.05) by Duncan's multiple range test. Vol. 78, No. 7,1988 891 I I I I I I I I I I I I I a) / b) / 5/ /- // / // / !/// -/ ~~~ED50=0.31 /E5 =00 _ /B". Y=l1.49x +3. 98 B:Y =2.57x+ 7. 2 8 ED50= 4.84 ED50 =0.135 o I I I I - c)/ // / /d)/// / / d 5/ / H/ // -/ / A-Y /=A1.43x+l1.52 /A: Y 1.47x +6.93 ED50= 265.8 / ED50O=.05 SB:Y= 1.43x + 4.01 / BY= 1.47x + 3.99 X ED5O =4.87 ED50 =4.83 0I II I I I I I I I I I I I _ e f) // // 6' 6' 6' 6' 6' 6' S'Y=1.36x4 6.77 S Y= 1.57x+3.96 ED50 =0.05 S ED50 = 4.57 G G:Y=1.36x+ 1.64 G GY= 1.57x +3.93 ED50 =295. I ED50 =4.79 0 1 I I I I I I -2 0 2 4 -2 0 2 4 Log Concentration (mg/mI) Fig. 2. Toxicity of antibiotics to Monilinia laxa represented as probit-log concentration regression lines and ED 50 (mg/ml). -A, --- B. A, Toxicity of antibiotic A and antibiotic B to germination of spores of M. laxa. B, Toxicity of antibiotic A and antibiotic B to germ tube growth of M. laxa. C and D, Parallel regression lines fitted for toxicity of the antibiotics to the germination of spores and to the germ tube growth of M. laxa, respectively. E and F, Parallel regression lines fitted for toxicity of antibiotic A and antibiotic B (respectively) to the germination of spores (G) and to the germ tube growth (S) of M. laxa. X2 for parallelism not significant in all cases. Significant differences (p < 0.05) exist among lines in cells c, d, and e. 892 PHYTOPATHOLOGY I I I I I I I I Another important aspect concerning the mechanisms involved 100 oo? 0) in the biocontrol of M. laxa by P.frequentans in peach twigs is the relationship among the microorganisms present in the phyllosphere of twigs. Thus, we have tested the activity of the 80 T. antibiotics on several fungal components of the mycoflora of peach twigs (9). Some of them were inhibited by antibiotics A and B; this 60may favor the growth and development of P. frequentans and, 60-"thus, enhanced the biocontrol of M. laxa, if antibiotics were produced by the antagonist in the twigs. 40LITERATURE CITED o 1. Birkinshaw, J. H. 1952. Studies in biochemistry of micro-organisms. -: 2088. Palitantin. Part 2. Further derivatives and degradation products. m- -Biochem. J. 51:271-276. 2. Chaplen, P., and Thomas, R. 1960. Studies in the biosynthesis of -fungal metabolites. The biosynthesis of palitantin. Biochem. J. 77:91-96. 100b) 3. Curtis, P. J., and Duncanson, L. A. 1952. A structural relationship 0 between frequentin and palitantin. Biochem. J. 51:276-278. 04. Curtis, P.J., Hemming, H. G.,and Smith, W. K. 1951. Frequentin: An antibiotic produced by some strains of Penicillium frequentans 80Westling. Nature (London) 167:557-558. 5. Finney, D. J. 1971. Probit Analysis. 3rd ed. Cambridge University Press, Cambridge, UK. 333 pp. 60 6. Harborne, J. B. 1973. Phytochemical Methods. A Guide to Modern Techniques of Plant Analysis. Chapman and Hall, London. 278 pp. 7. Katan, T., and Shabi, E. S. 1981. Resistance to dicarboximide 40 -fungicides in laboratory isolates of Monilinia laxa. Neth. J. Plant Pathol. 87:242. 8. McKeen, C. D., Reilly, C. C., and Pusey, P. L. 1986. Production and 20 partial characterization of antifungal substances antagonistic to Moniliniafructicola from Bacillus subtilis. Phytopathology 76:136-139. 9. Melgarejo, P., Carrillo, R., and M.-Sagasta, E. 1985. Mycoflora of 1I I peach twigs and flowers and its possible significance in biological 0 10 20 30 40 50 60 70 control of Monilinia laxa. Trans. Br. Mycol. Soc. 85:313-317. 10. Melgarejo, P., Carrillo, R., and M.-Sagasta, E. 1986. Potential for D a y ~ sbiological control of Monilinia laxa in peach twigs. Crop Prot. 5:422-426. Fig. 3. Loss of activity of antibiotic A (a) and antibiotic B (b) with time. 11. Melgarejo, P., and M.-Sagasta, E. 1984. Fungal antagonism in relation Points and brackets represent mean and standard error of four replicates of to peaches. Pages 127-136 in: Antimicrobials and Agriculture. M. W. percentage of inhibition of spore germination (--o--) and of germ tube Woodbine, ed. Butterworths, London. 583 pp. growth (--o--) of Monilinia laxa. Significant differences are reported 12. M.-Sagasta, I. 1986. Antagonismo de Penicilliumfrequentans text. Westling sobre Penicillium crustrosum Thom y Geotrichum candidum Link ex Leman, aislados de aceitunas almacenadas en troje. Tesina de Licenciatura en Ciencias Quimicas. Universidad Complutense de Because antibiotics appear to have a role in the biocontrol of M. Madrid, Spain. 140 pp. laxa by P. frequentans their stability is crucial. Both antibiotics 13. Sauer, D. B., and Burroughs, R. 1986. Disinfectation of seed surfaces were shown to be stable at a wide range of temperatures. As a rule, with sodium hypochlorite. Phytopathology 76:745-749. antibiotic B is more stable and less influenced by the external 14. Shabi, E., Ogawa, J. M. 1981. Benomyl-resistant and sensitive monoascosporic isolates of Moniliniafructicola. Neth. J. Plant Pathol. media than antibiotic A, conserving 30% of its activity on germ 87:252. tube growth after 70 days at room temperature as compared with 15. Snedecor, J. W., and Cochran, W. G. 1971. Statistical Methods. 10% for antibiotic A. In addition, pH changes did not affect Compahia Editorial Continental S. A. Mexico. 703 pp. antibiotic B while the activity of antibiotic A decreased in alkaline 16. Zehr, E. I. 1982. Control of brown rot in peach orchards. Plant Dis. conditions. 66:1101-1105. Vol. 78, No. 7,1988 893