Effects of the pesticides maneb and carbaryl on the phylloplane microflora of lettuce

Transcription

1 Effects of the pesticides maneb and carbaryl on the phylloplane microflora of lettuce JULIEN MERCIER AND R. D. REELEDER' Department of Plant Science, Macdonald College of McGill University, Ste-Anne-de-Bellevue, Qut., Canada H9X ICO Received July 3 1, 1986 Accepted November 24, 1986 MERC~ER J., and REELEDER, R. D Effects of the pesticides rnaneb and carbaryl on the phylloplane microflora of lettuce. Can. J. Microbiol. 33: The effects of the fungicide maneb and the insecticide carbaryl on the microflora of lettuce leaves were studied in field plots in 1983 and In 1983, populations of filamentous fungi were significantly, but temporarily, reduced by a single application of maneb. This treatment did not reduce populations in 1984; however, two additional applications of maneb did result in a significant decline in fungal populations when compared with unsprayed plants. Maneb had no effect on bacterial populations in either year. Carbaryl altered neither bacterial nor fungal populations. Pesticide applications did not change the composition of the microflora. Phoma and Fusarium spp. were the fungi most frequently isolated from lettuce leaves. MERCIER J., et REELEDER, R. D Effects of the pesticides maneb and carbaryl on the phylloplane microflora of lettuce. Can. J. Microbiol. 33 : Les effets du fongicide 'maneb' et de I'insecticide 'carbaryl' sur les rnicroflores des feuilles de laitue ont CtC CtudiCs en parcelles de champ en 1983 et En 1983, les populations de champignons filamenteux furent significativement, mais temporairement, rcduites par un seul traitement au maneb. En 1984, ce traitement n'a pas reduit les populations; toutefois, deux applications additionnelles ont causc un dcclin significatif des populations fongiques, par comparaison aux plants tcmoins. Au cours de ces deux annces, le maneb n'a pas eu d'effet sur les populations bacteriennes. D'autre part, le carbaryl n'a altcre ni les populations bactkriennes, ni les populations fongiques. De plus. les applications de pesticides n'ont pas changc la composition des microflores. Les Phoma et Fusarium spp. furent les champignons les plus frcquemment isolcs des feuilles de laitue. [Traduit par la revue] Biological control of diseases using phylloplane-inhabiting antagonists is receiving increasing attention as an alternative to fungicide application (Blakeman and Fokkema 1982). Naturally occurring leaf surface microorganisms can antagonize pathogens and reduce infection levels. Under field situations, Fokkema et al. (1983) showed that a well-developed microflora suppressed development of populations of Cochliobolus sativus (Ito & Kuribayashi) Drechsler ex Dastur and Septoria nodorum Berk. Cochliobolus sativus caused greater degrees of necrosis on rye plants when microfloral populations were reduced by benomyl applications (Fokkema et al. 1975). Newhook (195 1) and Wood (195 1) reported that naturally occurring saprophytes on lettuce (Lactuca sativa L.) could be inhibitory to Botrytis cinerea Pers. ex Fr. Lettuce crops are sprayed frequently with various pesticides and the elimination or reduction of these leaf surface microbial populations might decrease natural biological control, resulting in increases in disease severity. Our objective was to evaluate the impact of pesticide applications on the phylloplane microflora of lettuce. An insecticide (carbaryl) and a fungicide (maneb) were evaluated for such effects in the field in 1983 and Carbaryl is widely used in Canada to control several insect pests of lettuce, whereas maneb and related fungicides are used to control downy mildew. Materials and methods Field plots Lettuce, cv. Ithaca, was seeded in a plot located at Macdonald College, Ste-Anne-de-Bellevue, QuCbec, using a randomized complete block design with six replicates of each treatment. Plots of 2 x 2 m, each consisting of four rows, were separated from the neighbouring plots by three guard rows. Rainfall data were collected from a meteorological station located nearby. There were three different treatments in 1983: a single application of carbaryl (Sevin, 85% wettable powder (WP) at a rate of 3.25 kg active '~uthor to whom all correspondence should be addressed. ingredient (a.i.)/ha (1.3 g/plot); a single application of maneb (Wilson Maneb, 80% WP) at a rate of 2.2 kg a.i./ha (0.9 g/plot); and a control treatment which received no pesticide applications. In 1984, a fourth treatment consisting of three separate applications of maneb (each at 2.2 kg a.i./ha) was added. Pesticide applications were made using a manually pumped compressed-air sprayer. In 1983, the plots were seeded on May 25 and leaf samples (see below) were collected on July 11, 18, 25, and August 2 with pesticides applied on July 14. In 1984, the plots were seeded on May 28 with samples collected on July 3, 12, 17, and 25. Pesticides were applied on July 9, and for the treatment consisting of three applications of maneb, the additional applications were made on July 1 3 and 2 1. Collection and processing of leaf samples On each sampling date, three plants were selected at random from each plot, placed in separate plastic bags, and then held in a cooler until processing. Using a sterile cork borer, 5 discs (1 crn diameter) were cut from the third, fourth, or fifth leaf of each plant and the resulting 15 discs were combined together for further processing. Fungi were recovered on malt extract agar (Difco), amended with 100 mg tetracycline/l, using the three methods described below. Three replicate plates were prepared per plot sample, for each of the following methods. Leaf washing Five leaf discs were placed in 125-ml flasks containing 25 ml of sterile phosphate buffer, ph 7.0, and shaken for 1 h at 200 rpm. Then, 0.25 ml of the suspension was pipetted onto each of three plates of malt extract agar (MEA) and spreadover the surface of the agar with a bent glass rod. Cultures were incubated at room temperature (22-24"C), under ambient light conditions, for 7-14 days, and then held at 5OC until identification of the cultures could be made. Bacteria were recovered from leaf discs using the leaf washing method as described above for fungi. Nutrient agar (NA; Difco), without added antibiotics, was used rather than MEA. Plates were inverted, and the cultures were incubated at 22-24OC. After 2 or 3 days, plates were transferred to a 5OC cold room and held there until selection of colonies could be cavied out (see below). Leaf imprints One leaf disc was placed on each MEA plate and then removed from

2 MERCIER AND REELEDER 213 the plate after 12 or 24 h of incubation at C. In 1984, only the 24-h period was used. Cultures observed growing from the disc imprints over a 7- to 14-day period were transferred to fresh MEA plates. Cultures were incubated and stored as indicated above for the leaf washing procedure. Washed discs Discs used for the leaf washings (see above) were retrieved and washed twice more in 50 ml of sterile water, containing a drop of Tween 80, for 10 min on a shaker at 200 rpm. Each of the 15 discs was placed in a separate plate of MEA. Colonies growing from the disc imprints were transferred and stored as indicated above for the leaf imprints. This method was used in 1984 only. Composition of the microflora Three randomly selected colonies of filamentous fungi were identified from each plate treated with leaf washings. For the other methods, identification was attempted for all the colonies recovered. Exposing the cultures to near ultraviolet light to stimulate sporulation was often necessary for identification. Three bacterial colonies were selected at random from NA plates and streaked onto fresh plates of NA. Cultures were purified by restreaking colonies and subculturing onto NA. Cultures were then classified as Gram positive or negative and oxidase positive or negative based on results from Gram stain and oxidase tests. The oxidase reaction was determined by smearing bacteria on filter paper previously treated with a solution of 1 % tetramethyl-p-phenylenediamine dihydrochloride (Smilbert and Krieg 198 1). The Gram stain was performed following the method of Hucker (Smilbert and Krieg 198 1). A number of Gram-positive cultures were subjected to a 80 C heat test to determine their spore-forming ability (Smilbert and Krieg ). Gram-negative strains were tested for oxidation and fermentation of glucose (Smilbert and Krieg 198 1). Populations of filamentous fungi and bacteria were estimated by counting the total number of colonies growing on each of the leaf wash plates. Populations were expressed as the number of colony-forming units (cfu) per plate. Yeast presence was noted but their populations were not included in the population estimates as their colonies often coalesced. Population values obtained from the Petri plates were converted into cfu/cm2 of leaf tissue and these raw population data were then transformed to loglo equivalents to obtain normal distributions of data. The log-transformed population data were analysed with the analysis of variance procedure and Duncan's multiple range test was used to separate treatment means. Results Fungal Jora In 1983, populations of filamentous fungi were negatively affected by maneb (Fig. I). The effect of this fungicide was demonstrated on the second sampling when population levels were 70 cfu/cm2 in maneb-treated plots, as opposed to 126 cfu/ cm2 for the control. This reduction was not permanent, however, and, on the third and fourth sampling dates (July 25 and August 2), population levels in maneb-treated plots were not significantly different from those of the control treatment. Carbaryl did not cause any reduction in populations during the sampling period. The control and carbaryl-treated plants had very stable population levels of approximately 125 cfu/cm2 for the first two samplings. The population mean for all treatments in 1983 (Table 1) was quite stable for the first three sampling dates, but by the fourth date, it had reached a significantly higher level (236 cfu/cm2). In 1984, the first application of maneb (July 9) did not have any effect on the fungal population (Fig. I). The population levels in the control and maneb (single application) treatments were similar on the four sampling dates. The second and third applications of maneb (July 13 and July 21) in the multiple applications treatment resulted, however, in significant reduc- (A) 1983 mcont amaneb mcarb =man x3 Sample period Fic. 1. Populations of filamentous fungi recovered from lettuce leaf discs (leaf washing method) in (A) 1983 and (B) cont, control treatment; maneb, single maneb application treatment (1983, July 14; 1984, July 9); carb, carbaryl application (1983, July 14; 1984, July 9); man X 3, multiple applications of maneb ( 1984 only, July 9, 1 3, and 21). Sampling periods 1-4 were July 7, 18,25, and August 2 in 1983, and July 3, 12, 17, and 25 in Common letters at the tops of the columns indicate that populations for a given year are not significantly different from one another (P = 0.05, Duncan's multiple range test). TABLE 1. Lettuce leaf microflora populations (expressed as cfu/cm2) -- Filamentous Precipitation fungi Bacteria (mm)* NOTE: Values listed are means of all plots for a given date. For each year and each class of microorganism, means followed by the same letter are not significantly different according to Duncan's multiple range test, P = *Rainfall accumulation, in mm, since date of previous sampling. tin 1983, samples I through 4 were collected on July 7, 18, 25, and August 2, respectively. Corresponding dates for 1984 were July 3, 12, 17, and 25. tions in the number of colonies recovered from the July 17 and July 25 samples, respectively. The reduction was especially dramatic on the fourth sampling (July 25) where populations were 10 cfu/cm2 for the multiple application treatment versus 62 cfu/cm2 for the control. Leaf surface populations in carbaryl-treated plots were generally equal to or higher than those in the control plots (Fig. I), although the differences were not significant. The population mean for all treatments at the first sampling in

3 CAN. J. MICROBIOL. VOL. 33, 1987 TABLE 2. Composition of lettuce leaf mycoflora, 1983 July 7 July 18 July 25 Aug. 2 Fungus recovered LW LI LW LI LW LI LW LI Alternaria spp. Aureobasidium pullulans Botrytis cinerea Cladosporium spp. Epicoccum purpurascens Fusarium spp. Gliocladium spp. Myrothecium spp. Phoma spp. Mucorales Trichoderma spp. Penicillium spp. Unknown Miscellaneous NOTE: LW, data derived from results of leaf washing procedure; LI, data derived from results of leaf imprint procedures, where discs were in contact with agar for 24 h. Data are given as percentage of the total fungi isolated, for each isolation method and date. TABLE 3. Composition of lettuce leaf mycoflora, July 3 July 12 July 17 July 25 Fungus recovered LW LI LW LI LW LI LW LI Alternaria spp. Aureobasidium pullulans Botrytis cinerea Cladosporium spp. Epicoccum purpurascens Fusarium spp. Gliocladium spp. Myrothecium spp. Phoma spp. Mucorales Trichoderma spp. Penicillium spp. Unknown Miscellaneous Sterile mycelium NOTE: See footnote to Table 2 for details was 66 cfu/cm2 and there were significant increases on the second and third sampling dates (Table 1 ). Populations levels, however, declined significantly on the fourth date. In 1984, a total of only 1.0 mm of rainfall as recorded during the interval between the third and fourth sampling dates (July 17 and 25, respectively). From 150 to 220 fungal isolates were identified for each sampling method on each date (Tables 2 and 3). There appeared to be few major differences between pesticide treatments in the composition of the mycroflora and, therefore, results for all treatments were combined together. Plating out of washed discs gave results similar to that of the leaf imprint test and these data have not, therefore, been included in Table 3. The 12-h leaf imprint procedure gave results similar to those of the 24-h procedure and therefore only the 24-h data are presented in Table 2. Phoma (mainly P. medicaginis Malbr. and Roum.) and Fusarium (mainly F. equiseti (Corda Sacc.) were the filamentous fungi most consistently recovered regardless of the isolation method used. The occurrence of other genera varied, depending upon the date of sampling and the method used. Yeasts, Aureobasidium pullulans (de Bary) Am., and Cladosporium Link ex Fries spp. were major components with the leaf washing method. Yeasts were represented mainly by Rhodotorula glutinis (Fres. ) Harrison and Bullera Derx sp. Alternaria Nees ex Fr. spp. was present on each date and was most often isolated from leaf imprints and washed discs. Soil-borne fungi such as Gliocladium Corda, Myrothecium Tode ex Fr., Trichoderma Pers. ex Fr. and mucoraceous fungi occurred sporadically. A fungus producing white, sterile mycelium was present in small numbers in 1984 but was absent in Fungi such as Aureobasidium pullulans and Cladosporium spp. contributed more to the total fungal population in 1984 than in Bacterial flora Bacterial populations did not seem to be affected by any of the pesticide treatments. For all treatments, the population levels

4 MERCIER AND REELEDER 215 were quite stable in 1983 with a mean between 623 and 669 cfu/cm2 for the three first sampling dates (Table I). Only on the fourth date was a significant increase observed (1254 cfu/cm2). In 1984, population levels were about 380 to 540 cfu/cm2 except on the third date when they peaked to an average of 1823 cfu/cm2 and then dropped back again on the fourth sampling date (Table I). The composition of the bacterial community varied considerably for each sampling date in both years. Gram-positive or Gram-negative (oxidase negative and positive) types dominated on different dates. Actinomycetes were always present on each date with variable importance. No trends in bacterial group importance were observed with respect to time or pesticide treatment nor did the treatments alter the overall bacterial counts. Of the Gram-positive isolates, 25 out of the 39 tested (64%) were able to form spores and were thus classified as Bacillus. For the Gram-negative bacteria, 133 isolates (8 1 %) were found to be nonfermentative, with only 31 (19%) capable of fermenting glucose. Discussion Maneb altered fungal populations on lettuce leaves. The amount of alteration depends on weather conditions and number of applications. In 1983, 6.5 mm of rain fell between the date of pesticide application and the date of the second sampling. In contrast, in 1984, 23.2 mm of rain fell in the interval between the first application of maneb and the second sampling. This difference in rainfall accumulation could explain the different effects of a single application of maneb in 1983 and It is possible that the absence of effect in 1984 was due to removal of maneb by rain and a rapid recolonization of leaves. Microorganisms killed can be replaced over a short period of time if environmental conditions are favourable. Soil splashed onto leaves could resupply leaves with those components of the leaf microflora which originate from soil populations (e.g., Phoma and Fusarium). Kuthubutheen and Pugh (1 978) found spores of phylloplane fungi treated with captan or dicloran to be viable after washing. Periods of rain could thus restore the germinability of phylloplane fungal spores and at the same time remove fungicide residues from the leaves. In 1983, the eventual recovery in the population (after the initial decrease following maneb application) could have similar causes. In 1984, repeated applications of maneb caused population levels to fall progressively. The decrease was especially large on the last sampling which followed a hot dry spell. This type of weather did not permit washing of the fungicide from the leaves, and subsequent recolonization of the leaves was therefore suppressed. Overall, the results obtained agree well with those of Bainbridge and Dickinson (1972) who studied the leaf microflora of potato. They found filamentous fungi to be slightly affected by maneb. They also found that the bacterial flora were not affected. The effects of this fungicide appear to be quite unspecific, since no obvious changes in the microflora composition resulted from maneb application. Shifts in microflora composition have been reported, such as an increase in the importance of white yeasts (Warren 1974) or the selective elimination of certain taxa (Andrews and Kenerley 1978; Fokkema et al. 1975; McKenzie 1971; Warren 1974). Carbaryl did not have any effect on the bacterial or fungal flora. It may even have caused a stimulation of the microflora, but no statistically significant effects were observed. No particular component of the microflora appeared to be stimulated or negatively affected. Almost nothing is known about possible effects of insecticides on phylloplane microorganisms (Andrews 1981). This is unfortunate since insecticides are widely used in crop protection, and destruction or stimulation of the microflora could have an influence on disease incidence. This investigation of the effects of carbaryl appears to be the first one dealing directly with the impact of insecticides on phylloplane microflora. However, there are many other insecticides currently in use and it cannot be assumed that they are all harmless to the microflora. There was no consistent trend in respect to changes in population over time. However, increases or decreases in the fungal and bacterial population levels usually occurred at about the same time. Increases were observed on the fourth sampling in 1983 and on the second and third samplings in 1984, all of which were made after several rainy days. The decrease observed at the fourth sampling in 1984 was after an extremely dry and hot week. Meteorological conditions are known to be important factors in determining the population level of the microflora (Hirano and Upper 1983). An examination of the composition of the microflora over the different sampling dates fails to indicate any definite pattern of succession. The microflora of lettuce is apparently composed of a number of soilborne and airborne types. The proximity of lettuce leaves (in particular, the older leaves which were sampled during this study) to the ground may explain why soilborne fungi were so common. Generally, the microflora constituents are similar to those Newhook (1951) isolated from dead lettuce tissues. Among the fungi recovered, Phoma and Fusarium were clearly the dominating types. These fungi are often reported to be present on plants, but they are rarely reported as important parts of the microflora. Various Phoma species were present on beet seed (Byford and Gambogi 1985). Norse (1972) found Phoma sporulating on the surface of tobacco leaves. Since Phoma, Alternaria, and Fusarium were isolated frequently from washed discs, it appears that either their spores adhere strongly to the leaf surface or they exist as hyphae on the surface. It is also possible that they exhibit endophytic growth in the leaves. Lettuce leaves disappear rapidly when they reach senescence and these fungi could play an important role in their decomposition. Overall, only a few fungal genera were found in significant numbers. The bacterial flora was found to be very variable for each date, indicating that weather conditions probably played a greater role than the pesticides studied in affecting the composition of the flora. Unpigmented and nonfermentative types dominated in the Gram-negative groups, while the Grampositive group appeared to have a large proportion of spore formers, which were assigned to the genus Bacillus. In conclusion, the results reported here indicate that single applications of maneb have temporary, variable effects on phylloplane populations. The magnitude of these effects is probably determined largely by weather conditions. Multiple applications of maneb are more effective in suppressing populations. The insecticide carbaryl does not affect populations of lettuce leaf microflora. Thus, the effects of pesticides on phylloplane flora depend both upon the type of pesticide being used and the number of applications made. ANDREWS, J. H Effect of pesticides on non-target microorganisms on leaves. In Microbial ecology of the phylloplane. Edited by J. P. Blakeman. Academic Press, London. pp

5 2 16 CAN. J. MICROBIOL. VOL. 33, 1987 ANDREWS, J. H., and KENERLEY, C. M The effects of a pesticide program on non-target epiphytic microbial populations of apple leaves. Can. J. Microbiol. 24: BAINBRIDGE, A., and DICKINSON, C. H Effect of fungicides on the microflora of potato leaves. Trans. Br. Mycol. Soc. 59: BLAKEMAN, J. P., and FOKKEMA, N. J Potential for biological control of plant disease on the phylloplane. Annu. Rev. Phytopathol. 20: BYFORD, W. J., and GAMBOGI, P Phoma and other fungi on beet seed. Trans. Br. Mycol. Soc. 84: FOKKEMA, N. J., RIPHAGEN, I., POOT, R. J., and DE JONG, C Aphid honeydew, a potential stimulant of Cochliobolus sativus and Septoria nodorum and the competitive role of saprophytic mycoflora. Trans. Br. Mycol. Soc. 81: FOKKEMA, N. J., VAN DE LAAR, J. A. J., NELIS-BLOMBERG, A. L., and SCHIPPERS, B The buffering capacity of the natural mycoflora of rye leaves to infection by Cochliobolus sativus, and its susceptibility to benomyl. Neth. J. Plant Pathol. 81: HIRANO, S. S., and UPPER, C. D Ecology and epidemiology of foliar bacterial plant pathogens. Annu. Rev. Phytopathol. 21: KUTHUBUTHEEN, A. J., and PUGH, G. J. F Effects of fungicides on physiology of phylloplane fungi. Trans. Br. Mycol. SOC. 71: MCKENZIE, E. H. C Seasonal changes in fungal spore numbers in ryegrass - white clover pasture, and the effects of benomyl on pasture fungi. N.Z.J. Agric. Res. 14: NEWHOOK, F. J Microbiological control of Botrytis cinerea Pers. 11. Antogonism by fungi and actinomycetes. Ann. Appl. Biol. 38: NORSE, D Fungal populations of tobacco leaves and their effect on the growth of Alternaria longipes. Trans. Br. Mycol. Soc. 59: SMILBERT, R. M., and KRIEG, N. R General characterization. In Manual of methods for general bacteriology. Edited by P. Gerhardt. American Society for Microbiology, Washington, DC. pp WARREN, R. C Differential effects of fungicides on phylloplane fungi isolated from oak. Trans. Br. Mycol. Soc. 62: WOOD, R. K. S The control of diseases of lettuce by the use of antagonistic organisms. I. The control of Botrytis cinerea Pers. Ann. Appl. Biol. 38: