DESCRIPTION TECHNICAL FIELD

This invention is related to the field of biological control by saprophytic yeast of fungal diseases of agricultural commodities.

BACKGROUND ART Postharvest spoilage of agricultural commodities such as fruits and vegetables has been estimated to result in losses of approximately 24% of the crop in the United States and up to 50% of the crop worldwide. Preharvest losses of agricultural commodities are also significant.

Much of the pre- and postharvest loss is due to fungal diseases, such as molds and rots. Often infection is initiated by injuries made at harvest or by mechanical wounds to the surface of the agricultural commodity during processing. Chemical fungicides are the principal means of controlling postharvest losses due to fungal disease. The traditional method of fungicide use is to treat fruit after harvest and before storage with chemical fungicides. However, fungicide-tolerant strains are present in most packing houses, rendering chemical fungicides less effective or totally ineffective. For this reason, multiple fungicides are sometimes used simultaneously to improve control of fungal pathogen. There are several important fungal pathogens, e.g., Mucor spp., for which there is currently no effective postharvest fungicide.

An additional problem with the use of chemical fungicides is the fact that many are carcinogenic or environmentally hazardous. At a time when synthetic pesticide use is being curtailed, there is clearly an urgent need to develop safe, new and effective methods of controlling postharvest diseases of agricultural commodities that are safe, environmentally benign, and effective.

There have been numerous reports of attempts to control fungal diseases of agricultural commodities using microbial biological control agents, as reviewed in Wisniewski and Wilson, HortScience, 27:94-98, 1992. Ideally, such biological control agents are naturally occurring saprophytic microorganisms which do not produce antifungai compounds, do not grow at human body temperature, and are consistent in controlling the target disease. Patents and patent applications describing attempts to use bacteria and yeasts for biological control of fungal disease of agricultural commodities include U.S. Patents No. 5,314,691 (Coffey et al ), 5,270,059 (Janisiwicz et al ), 5,266,316 (Elad et al.), 5,244,680 (Roberts), 5,238,690 (Elad et al.), and 5,041,384 (Wilson and Chalutz), and WO 92/18009 (Shanmuganathan) and WO 91/01641 (Wilson et al ).

However, despite a significant amount of research, there is no commercially available biological control agent that has proved consistently effective in controlling fungal disease of

agricultural commodities See, e g , Wisniewski and Wilson, HortScience, 27 94-98, 1992, and Wilson et al , Plant Dis 78 837-844, 1994

Most researchers have used a "silver bullet" approach in the selection of microbial biological control agents, identifying a single microbe for biological control of fungal disease rather than documenϋng the diverse microflora on fruit or vegetable surfaces grown in different locations, then testing the efficacy of these microbial colonists in controlling postharvest diseases (Spurr, "The microbial ecology of fruit and vegetable surfaces Its relationship to postharvest biocontrol," in Wilson and Wisniewski (eds ), Biological Control of Postharvest Diseases Theory and Practices, CRC Press, Boca Raton, FL, pp 11-23, 1994) There is a need for an efficient method for the recovery of diverse microflora, saprophytic yeast or bacterial colonists that effectively control postharvest diseases of fruits, including microbes that have a very low population on the plant surface

SUMMARY OF THE INVENTION The present invention provides yeast strains that are highly effective in reducing the incidence and/or seventy of fungal disease of agricultural commodities, such as fruit, caused by a number of important fungal pathogens

Accordingly, it is an object of the invention to provide compositions for biological control of a fungal disease of an agricultural commodity compnsing an effective amount of one or more biologically pure yeast strains selected from the group consisϋng of Cryptococcus tnfirmo- mimatus (preferably Cryptococcus infirmo-mmiatus Phaff and Fell isolate YY-6), Cryptococcus albtdus isolate HRB-2, Cryptococcus laurentti (Kufferath) Skinner isolate HRA-5, Rhodotorula aurantiaca isolate YCL-5, Rhodotorula glut mis Harrison isolate HRA-3, Rhodotorula glu tints Harrison isolate HRA-4, and Rhodotorula gluttms Harrison isolate HRB-6 Such compositions provide surpnsmgly effective biological control of fungal disease caused by a vaπety of important fungal pathogens of agricultural commodities such as fruit (e g , pear, apple, or cherries) These fungal pathogens include but are not limited to Penicilltum spp , Botrytts spp , Mucor spp , Pezicula spp , and Phtalophora spp , and Moniltnea spp

It is also an object of the invention to provide methods for controlling a fungal disease of an agricultural commodity compnsing applying to the agncultural commodity an effective amount of such a composition

The present invention also encompasses a method of controlling the incidence or seventy of disease caused by a Phtalophora spp compnsing applying to an agncultural commodity an effective amount of a composition compnsing a biologically pure strain of Rhodotorula spp or Cryptococcus infirmo-mtniatus

It has also been discovered that a combination of a microbial biological control agent, including the yeast strains listed above, in combination with low levels of a chemical fungicide is

suφnsingly synergistic in controlling the incidence and seventy of a vanety of fungal diseases An effective amount of such a combination may compnse an amount of the chemical fungicide about 25% or less of the normal amount of the fungicide, 1 e , an amount that would be effective if the chemical fungicide were to be applied without the microbial biological control agent, preferably about 10% or less of the normal amount, more preferably about 5% or less, and most preferably about 3% or less of the chemical fungicide

It is a further object of the invention to provide novel methods for isolating from agncultural commodities a highly diverse population of microbes, e g , bacteπa, yeast, and filamentous fungi, that are useful for the biological control of fungal disease Such methods compnse the steps of providing an undamaged whole agncultural commodity, such as a fruit, that has preferably not been treated with a biocide (e g , pesticide, fungicide, etc ), submersing the agncultural commodity in a nutnent-poor aqueous solution, sonicating the submersed agncultural commodity for about one to five minutes (preferably for about five minutes), thereby producing a sonicate, plating the sonicate on a nutnent-poor media under conditions appropnate for growth of the microbe, identifying the microbe, and testing the microbe for efficacy in controlling the incidence and or seventy of fungal disease Preferably the sonication step is combined of follows a step of shaking the submersed agncultural commodity, e g , for about one to five minutes In this way one may isolate, for example, strains of Cryptococcus tnfirmo-mimatus, Cryptococcus albtdus, Cryptococcus laurenttt, Rhodotorula auranttaca, and Rhodotorula glutints in addition to those descπbed below

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 shows the growth and population dynamics of Cryptococcus tnfirmo- mimatus and Cryptococcus laurenttt in wounds of Golden Delicious apple fruits stored at (A) 0°C, (B) 5°C, (C) 10°C, and (D) 20°C Bars represent the standard deviation

Figure 2 shows the growth and population dynamics of Cryptococcus laurenttt, Rhodotorula glutints, and Cryptococcus tnfirmo-mimatus in wounds on pears stored at -1°C Bars represent standard deviations of the means

Figure 3 shows the growth and population dynamics of Cryptococcus laurentu, Rhodotorula glutints, and Cryptococcus tnfirmo-mimatus in wounds on pears stored at (A) 5°C, (B)

10°C, and (C) 20°C Bars represent standard deviations of the means

MODES FOR CARRYING OUT THE INVENTION

Biocontrol of Fungus Infections of Agncultural Commodities with Saprophytic Yeast Applied Alone or in Combination With A Chemical Fungicide

For biological control of fungal pathogens of agncultural commodities, 1 e , reducing the incidence and/or seventy of disease caused by such fungal pathogens, Cryptococcus albidus

(preferably strain HRB-2), Cryptococcus tnfirmo-mimatus strain YY-6, Cryptococcus laurentu

(preferably strain HRA-5), Rhodotorula aurantiaca (preferably strain YCL-5), and Rhodotorula glutims (preferably strains HRA-3, HRA-4 and HRB-6) are preferred These yeast strains are effective against a wide vanety of fungal pathogens of agncultural commodities It has been discovered that a wide vanety of microbial biological control agents are surprisingly synergistic in controlling fungal pathogens of agncultural commodities when used together with low doses of well known chemical fungicides Although the discussion below focuses on the use of strains of Cryptococcus and Rhodotorula either alone or in combination with a chemical fungicide, this discovery extends to combinations of low doses of chemical fungicides with any microbial biological control agent effective against fungal pathogens of agncultural commodities

This discovery has important practical implications Natural populations of fungal pathogens of agncultural commodities commonly compnse a mix of sensitive and insensitive strains The spores of sensitive strains will not germinate or growth will be slowed by a chemical fungicide Thus, the inoculum dose of the pathogen will be lowered and the yeast will compete more effectively With fungicide-insensitive pathogens, even the high rate of fungicide will not be effective, but the yeast in the combination will give effective biological control

The present invention provides "biologically pure" strains of microbial biological control agents, including saprophytic yeasts such as Cryptococcus spp and Rhodotorula spp, and compositions compnsing such biologically pure strains A biologically pure strain of a microbe is one isolated from its natural environment (e g , the surface of a fruit) and which is not contaminated by other microbial strains Preferably, the biologically pure strain originates from a single cell of the microbe and thus represents a clonal population

Biological control of Penicillium spp At least 11 species of Penicillium have been isolated from pome fruits naturally infected with blue mold, but Penicillium expansum is the most common and economically important species Blue mold, also known as soft rot and wet rot, is the most important postharvest disease of apples and is also important on other fruits, including pears, for example

The present invention provides strains of naturally occurnng saprophytic yeast effective in reducing the incidence and severity of fungal disease of agricultural commodities caused by Penicillium spp , including but not limited to Penicillium expansum For reducing the incidence and seventy of disease caused by Penicillium spp , Cryptococcus tnfirmo-mimatus (preferably strain

YY-6), Cryptococcus laurenttt (preferably strain HRA-5), Rhodotorula aurantiaca (preferably strain YCL-5), and Rhodotorula glutints (preferably strains HRA-4 and HRB-6) are prefened, particularly for control of blue mold Most preferred are Cryptococcus tnfirmo-mimatus isolate YY-6, Cryptococcus laurentn strain HRA-5, and Rhodotorula glutints strain HRB-6 Disease caused by Penicillium may be treated by application of one or more of these yeasts alone or in combination with a low dose of an appropnate chemical fungicide, as defined below For example, for control of blue mold, about 3% of the normal dose of thiabendazole may be used in combination with Cryptococcus laurenttt strain HRA-5, Rhodotorula aurantiaca strain YCL-5, and Rhodotorula glutmts strain HRB-6 as descnbed in greater detail below Biological control of Mucor spp The present invention provides strains of naturalh occurnng saprophytic yeast effective in reducing the incidence and seventy of fungal disease of agncultural commodities caused by Mucor spp , including but not limited to Mucor pirtformis Mucor rot, which is caused pnmanly by Mucor pmformis E Fischer, occurs less consistently than blue mold and gray mold, although in special situations it can cause severe losses of apples and pears All fungicides currently registered for postharvest treatment of pome fruits are ineffective against Mucor pirtformis

For reducing the incidence and seventy of diseases of agncultural commodities caused by Mucor spp , Cryptococcus tnfirmo-mimatus (preferably strain YY-6), Cryptococcus laurentn (preferably strain HRA-5), and Rhodotorula glutmts (preferably strain HRB-6) are prefened, particularly for control of Mucor rot Most prefened is Cryptococcus mfirmo-miniatus isolate YY- 6 Biocontrol of Botrytis spp The present invention provides strains of naturally occurnng saprophytic yeast effective in reducing the incidence and seventy of fungal disease of agncultural commodities caused by Botrytis spp , including but not limited to gray mold, caused b> Botrytis cinerea Pers Gray mold is the most important postharvest disease of pears and is second to blue mold in importance on apples Also known as cluster rot or nest rot, gray mold can cause large losses because of its ability to spread from infected to adjacent healthy fruit dunng storage The disease develops more rapidly dunng cold storage temperature than any other postharvest decay except Mucor rot

For reducing the incidence and seventy of diseases of agncultural commodities caused by Botrytis spp , Cryptococcus infirmo-mimatus (preferably strain YY-6), Cryptococcus laurentn (preferably strain HRA-5), and Rhodotorula gluttms (preferably strain HRB-6) are prefened, particularly for control of gray mold It is prefened that the treatment combine one or more strains of the yeasts of the invention with a low dose of a chemical fungicide known in the art to be effective against Botrytis spp , e g , thiabendazole, thiram, or dichloran Biocontrol of Phtalophora spp The present invention provides strains of natural.} occurnng saprophytic yeast effective in reducing the incidence and seventy of fungal disease of agncultural commodities caused by Phtalophora spp , including but not limited to Phtalophora

malorum, which causes side rot Benziπudazole fungicides such as thiabendazole are ineffective in reducing the incidence and seventy of side rot

For reducing the incidence and seventy of diseases of agncultural commodities caused by Phtalophora spp , Cryptococcus tnfirmo-mimatus (preferably strain YY-6), Cryptococcus laurentn (preferably strain HRA-5), and Rhodotorula glutims (preferably strain HRB-6) are prefened, particularly for control of side rot The treatment may employ one or more strains of the yeasts of the invention alone or in combination with a low dose of a chemical fungicide

Biocontrol of Pezicula spp The present invention provides strains of naturally occurnng saprophytic yeast effective in reducing the incidence and seventy of fungal disease of agncultural commodities caused by Pezicula spp , including but not limited to Pezicula malicorticis, which causes bull's-eye rot

For reducing the incidence and seventy of diseases of agncultural commodities caused by Pezicula spp , Cryptococcus infirmo-mimatus (preferably strain YY-6), Cryptococcus laurentn (preferably strain HRA-5), and Rhodotorula glutmts (preferably strain HRB-6) are prefened, particularly for control of bull's-eye rot The treatment may employ one or more strains of the yeasts of the invention alone or in combination with a low dose of a chemical fungicide

Biocontrol of Momltnea spp The present invention provides strains of naturally occurnng saprophytic yeast effective in reducing the incidence and seventy of fungal disease of agncultural commodities caused by Momltnea spp , including but not limited to Momltnea fructtcola, which causes brown rot of chernes

For reducing the incidence and seventy of diseases of agncultural commodities caused by Momltnea spp , Cryptococcus laurentn (preferably strain HRA-5) is prefened, if used alone, or, in combination with a chemical fungicide, such as iprodione, Cryptococcus laurentn (preferably strain HRA-5) or Cryptococcus infirmo-mimatus (preferably strain YY-6) are prefened, particularly for control of brown rot The treatment may employ one or more strains of the yeasts of the invention alone or in combination with a low dose of a chemical fungicide

Preparation and Application of the Yeasts of the Invention for Biological Control of Fungal Disease of Agncultural Commodities

Growth of yeast The yeast strains of the present invention are grown under aerobic conditions at any temperature satisfactory for growth of the organism, e g , from about 10°C to about

30°C The prefened temperature is 20-25°C The pH of the nutnent medium is preferably about neutral, 1 e , approximately pH 5 8 to 7 2 The incubation time is that time necessary for the organisms to reach a stationary phase of growth Incubation time is preferably from about 48-72 hours The yeast strains may be grown by any of the methods known in the art for such yeast

For small scale fermentation, conventional shaker flasks are prefened For large scale fermentation, fermentation tanks are prefened Agitation and/or aeration is preferably supplied to

the inoculated liquid medium. Following incubation, the organisms are harvested by conventional methods, e.g., centrifugation or filtering. Cultures or harvested cells may be stored by conventional means, e.g., by freeze drying after addition of a cryoprotectant.

Amount of veast in the compositions of the invention. The compositions of this invention are generally provided in an amount effective to treat and/or prevent fungal disease of agricultural commodities. An "effective amount" of a composition of the present invention is an amount of the composition which reduces the incidence or severity of a fungal disease when applied to the agricultural commodity, preferably by 50% or more as compared with controls.

As will be apparent to one skilled in the art, effective concentrations may vary depending upon factors including: the strain of yeast employed, the type, age, and ripeness of agricultural commodity; and the type and concentration of fungal pathogens affecting the agricultural commodity; temperature and humidity. Exemplary concentrations range from about 1 x 10 ⁴ to 1 x 10 ¹² colony forming units per milliliter (CFU/ml). Preferably, compositions comprise in excess of 1 x 10 ⁷ CFU/ml. Modes of application of the compositions of the present invention. The compositions of the present invention may be provided in any of the standard forms known in the art. These fungal biocontrol compositions may be in a solid or liquid form. Solid compositions may be in the form of lyophilized yeast in the form of dusts, granules, or wettable powders. Lyophilized cultures may be readily re-suspended in aqueous solutions for application to agricultural commodities. Liquid compositions may be in the form of aqueous or non-aqueous media, in solutions, suspensions, dispersions, dispensions, or concentrated form, e.g., a slurry or paste. Compositions comprising yeast of the present invention may be applied by any method known in the art, including but not limited to spraying, dipping, drenching, brushing, or misting. In addition, the organisms of the invention may be incorporated into waxes, wraps or other protective coatings used in processing the agricultural commodities.

Fungal pathogens. The yeasts of the present invention are useful for the biological control of the incidence and severity of pre- and postharvest disease caused by a variety of plant pathogens, including those which cause disease in pome fruits such as pear and apple. Examples of plant pathogens against which the yeast of the invention are useful include, but are not limited to, Penicillium spp. (e.g., Penicillium expansum), Botrytis spp. (e.g., Botrytis cinereά), Mucor spp.

(e.g., Mucor piriformis), Pezicula spp. (e.g., Pezicula malicorticis), Phialophora spp. (e.g., Phtalophora malorum), and Monilinea spp., e.g., Monilinea fructicola.

These fungi affect a wide variety of agricultural commodities. For a listing of fungal plant pathogens and their distribution and host range see, e.g., Fan et al., eds., Fungi on Plants and Plant Products in the United States, American Phytopathological Society, St. Paul, MN, 1989 (No. 5 in the series Contributions from the U.S. National Fungus Collections).

Agncultural commodities The yeasts of the present invention are useful in controlling plant pathogens on a vanety agncultural commodities including, but not limited to fruits, vegetables, cereals, grains (e g , wheat, corn, sorghum, soybean, and barley), nuts (e g , peanuts, almonds, and pecans), seeds, floral bulbs, nursery seedlings, and silage Examples of fruits include but are not limited to pome fruits (e g , apples and pears), stone fruits (e g , peaches, nectannes, apncots, plums, and cherries), citrus fruits (e g , grapefruit, orange, lemon, kumquat, lime, mandannes and pummelo), grapes, tomatoes, persimmons, strawbemes, and papayas Apples which may be treated in accordance with this invention include Granny Smith, Red and Golden Delicious, Jonathan, Gala, Fuji, Newton, Macintosh, and other well known apple strains Examples of pears include d'Anjou, Packham's Tnumph, William's Bon Chretian, and Beune Bosc In addition to unprocessed agncultural commodities, the present invention may be utilized with processed agncultural commodities including, for example, raisins, prunes, figs, dned apncots, and dates

Fruit treated in accordance with the methods of this invention may be stored at standard fruit storage temperatures, such as 0°C, 4°C, and room temperature, free of the effects of fungal infection or with reduced incidence and/or seventy of infection Fruit treated in accordance with this invention may also be stored in a controlled atmosphere (such as 2 5% oxygen and 2 5% carbon dioxide)

The agncultural commodity may be treated at any time before or after harvest The preferred time of treatment is after harvest and pπor to storage or shipment

Fungicides The microbial biological control agents of the present invention, such as the saprophytic yeast strains disclosed herein, may be used to control the incidence and seventy of fungal disease in agncultural commodities in combination with one or more well known chemical fungicides, including but not limited to benzamidazoles (including benomyl, carbendazim, and thiabendazole), thiram, dichloran, vinclozohn, φrodione, procymidon, dichlorfluanide, tebuconazole, prochloraz, fenethanil, diethefencarb, metomeclan, chlorothaloml, and mixtures of these fungicides

If a microbial biological control agent is applied in combination with a chemical fungicide used in combination, it is prefened that "low levels" of the chemical fungicide be used A "low dose" of a chemical fungicide is defined as about 25% or less of the amount or dosage of the chemical fungicide commonly used for control of a particular fungal disease when applied alone, more preferably about 10% or less, even more preferably about 5% or less, and most preferably about 3% or less

The chemical fungicide may be applied in a composition compnsing the yeast and the chemical fungicide or the chemical fungicide may separately applied, either before or after the application of the yeast, using any well known method of application, including but not limited to spraying, dipping, drenching, misting, and incorporation into packaging matenals Benziπudazole

and dicarboxiπude fungicides, for example, are commonly applied as postharvest dips, drenches or line sprays

Combination of the yeasts of the invention with each other and with other biological control agents The yeast strains of the present invention may be used individually or in combination with one or more other yeast strains of the mvenϋon The yeast strains of the present invention may also be used in combination with other microbes used for the biological control of fungal or other diseases of agncultural commod ties in an amount compatible with the effectiveness of a yeast strain of the present invention The yeast strains of the present invention and the additional biological control agent may be applied to the agncultural commodity at the same time as part of a single composition or at different times, either before or after application of a yeast of the present invention

Such additional biological control agents include, but not limited to Acremomum breve, Bacillus subtihs, Candida parapstlosis, Candida guilliermondit (Ptchta guilltermondn), Cryptococcus albidus, Cryptococcus flavus, Cryptococcus laurentn, Debaryomyces hansen , Enterobacter aerogenes, Hansemaspora uvarum, Myrothecium roridum, Pseudomonas cepacia, Pseudomonas syrmgae, Pseudomonas gladioli, Rhodotorula glutmts, Rhodotorula mucilagmosa, Sporobolomyces roseus, Trichoderma harzianum, and Trichoderma pseudokoningit

Denvatives and mutants This invention extends to denvatives and mutants of the yeast strains of the invention which are likewise effective in the biological control of fungal disease of agncultural commodities Such denvatives or mutants may be prepared according to standard microbiological or recombinant DNA methods, such as mutagenesis with chemical mutagens (e g , nitrosoguanidine and ethanemethylsulphonate), or irradiation (e g , ultraviolet radiation, infrared radiation, and irradiation with X-rays or gamma rays or a source emitting alpha or beta particles), spontaneous mutation, mutagenesis using recombinant DNA techniques, and transformation with heterologous nucleic acids, including recombinant plasmid or viral vectors Denvatives or mutants may differ from the parent strains in respect of one or more morphological, biochemical, physiological, or other charactenstics, but will retain the capacity to inhibit growth of fungal pathogens

Carπers. excipients. and other additives The yeast strains of the present invention may be present in compositions compnsing one or more agnculturally acceptable carπers, excipients, or other matenals as are known in the art, agnculturally compatible, and compatible with the viability, growth, and effectiveness of the yeast strains of the present invention in biological control of fungal pathogens

The term "earners" includes, for example, (1) a gel or gum based earner (e g , xanthan gum) or (2) a water based earner (e g , water, buffer solutions, carbohydrate containing solutions, and saline solutions), (3) an oil based earner (e g , "Fresh Mark" or "Fresh Wax 58P" (which is a paste wax for peaches, plums and nectannes, containing - white oil, paraffin wax,

- 10 -

petrolatum and oleic acid) both from Fresh Mark (Chemical Corporation, Orlando, FL), (4) a wax based earner (e g , including wax coatings typically used on citrus fruit and apples, for example "Bntex 551" or "Bπtex 559," both from Broshar (Chemicals) Ltd ,

Kefar-Saba, Israel), (5) a powdered earner ingredient to provide the composition in powdered form, and in which the mιcroorgaιusm(s) are dispersed and thus diluted to a desired concentration in the powdered composition (e g , starch and talc), and (6) a mixture of the foregoing

The term "excipient" refers to conventional additives, such as surfactants and wetting agents (e g , Tween 20 and Tπton X-100), anϋoxidants, nutnents, emulsifiers, spreading agents, suspending agents, sticking agents, anti-scald agents (e g , diphenylamine or ethoxyquin), preservatives, chemical pesticides and the like The compositions of the invention may also compnse a source of calcium, such as calcium chlonde or other non-toxic calcium source in an amount from 0 1 to 50% (v/v) See WO 92/18009 (Shanmuganathan) Preservatives may include, for example (a) a gum, e g , a natural gum, such as guar gum, locust bean gum, karaya gum, tragacanth gum or preferably xanthan gum, (b) methyl cellulose, (c) silica gel, and (d) mixtures of the foregoing

Isolation of Saprophytic Yeast Strains Effective in Fungal Biocontrol

The present invention provides methods for isolating diverse microflora and accurately estimating the population size and diversity of cultivable yeasts and bacteria on the surface of fruit at harvest and in cold storage Moreover, these methods are useful for obtaimng microorganisms on the surface of fruit that are effective in the biocontrol of fungal pathogens, such as those responsible for postharvest losses of fruit

Plant surfaces are colomzed by diverse microbial flora At harvest, the microbial community on the fruit surface includes bacteπa, fungi, yeasts, and other microorganisms (Andrews and Kenerley, Can J Microbiol 24 1058-1072, 1978, Andrews and Kenerley, Can J Microbiol 25 1331-1344, 1979, Spun, 1994) Some microorganisms casually occupy plant surfaces and are unable to multiply (Hirano and Upper, Annu Rev Phytopathol 21 243-269, 1983) These microorganisms soon disappear unless replenished Other microorganisms are epiphytic residents able to grow and survive on plant surfaces This microbial population plays an important role in the development of postharvest rots of fruits and vegetables (Chalutz and Wilson, Plant Dis 74 134-137, 1990, Spun, 1994) Disease incidence has been positively conelated to the size of the epiphytic population of plant pathogenic bactena (Rouse et al , Phytopathol 75 505-509, 1985)

Yeasts colonize plant surfaces or wounds for a long penod under dry conditions, produce extracellular polysacchandes that enhance their survivability, rapidly multiply and use available nutrient, and are impacted minimally by pesticides (Wisniewski and Wilson, HortScience 27 94-98, 1992)

A number of methods have been described for removing saprophytic microorganisms from plant surfaces. Swabbing has been reported to be less efficient than pulping for removal of yeasts from apple fruit surfaces (Marshall and Walkey, Food Res. 16:448-456, 1951) and is also less efficient than shaking or sonication for removal of fungal conidia from the surface of grape or plum fruits. Swabbing introduces the additional problem of removing organisms from the swabs (Walters, J. Appl. Bacteriol. 30: 56-65, 1967). Other methods for removal of microorganisms include: washing samples in water (Le Roux et al., Phytophylactica 5:51-54, 1973; El-Din et al., Zentralbl. Mikrobiol. 141:488-492, 1986; Kamra and Madan, Microbios. Letters 34:79-85, 1987; Roberts, Phytopathol. 80: 526-530, 1990), water-Tween (Guerzoni and Marchetti, Appl. Environ. Microbiol. 87:571- 576, 1986) or phosphate buffer (Janisiewicz, Phytopathol. 77:481-485, 1987). This washing step is normally followed by vortexing or shaking the suspension, then plating the suspension on agar media. Blending of strawberries (Buhagiar and Barnett, J. Appl. Bact. 34:727-739, 1971) and grapes (Singh and Kainsa, Indian Phytopathol. 36:72-76, 1981) in phosphate buffer has been used to study the microflora of fruit. Ultrasound has also been employed to facilitate harvesting and quantification of plant epiphytic microorganisms (Martini et al., Can. J. Microbiol. 26:856-859, 1980; Guerzoni and Marchetti, Appl. Environ. Microbiol. 87:571-576, 1986). It has been reported that ultrasonication resulted in dislodgement of higher numbers of yeasts species from plant surface and consistently resulted in 100-200 percent more yeast species than other methods (Martini et al., Can. J. Microbiol. 26:856-859, 1980). However, none of these reports document the optimum time of sonication that is necessary for the detachment of most microorganisms from plant surfaces and for isolation of the most diverse microbial flora.

For standard procedures for isolation and cultivation of yeast strains, see also Phaff, Miller and Mrak, The Life of Yeasts, 2nd edition, Harvard University Press, 1978; and Devenport, Outline Guide to Media and Methods for Studying Yeasts and Yeast-like Organisms, in Biology and

Activity of Yeasts, A.P. London, 1980. Yeast strains may be grown on nutrient agar (e.g., potato dextrose agar). Yeast strains can be readily typed according to standard procedures, such as are described in Barnet et al., Yeasts: Characteristics and Identification, Cambridge University Press, 1983. The methods of the present invention optimize the recovery of a diverse microbial population, particularly those useful for fungal biocontrol, from fruit or other agricultural commodities.

Briefly stated, a healthy, mature agricultural commodity, e.g., a fruit, preferably free from punctures or other types of damage and preferably not treated with a bactericide, fungicide or insecticide, is randomly selected. Fruit that has not been treated for pest control, such as from a neglected orchard, is prefened in order to ensure a more diverse microbial population. If the fruit must be stored prior to microbial isolation, storage at a low temperature, e.g., at approximately 5°C,

IS prefened in order to preserve natural microbial populations, storage at a higher temperature is expected to favor bactena and adversely affect yeasts It is prefened that the fruit be stored for 24 hours or less

The fruit, along with its stem, is partially or fully submersed in an aqueous buffer, e g , SPBT (0005 M stenle potassium phosphate buffer, pH 7 0, 0 005% Tween 80) A low-nutnent medium simulates the low nutrient status of fruit surfaces, whereas high nutrient media may cause fast-growing yeasts or bactena to overgrow the plates or actually prove toxic to some yeasts

Microbes on the surface of the fruit are dislodged, by shaking, followed by sonication, preferably for one to five minutes, most preferably for five minutes It has been determined that this combination treatment is highly effective in detaching tightly bound microbes and optimizes the recovery of a diverse population of bactena, yeast, and filamentous fungi

Culturable populations of yeasts, bactena and filamentous fungi are plated on an appropnate media, preferably a nutnent-poor media, under conditions appropnate for growth of the microbe The plates are carefully examined by visual inspection for differences in the moφhological and biochemical charactenstics of the colonies in order to identify and select the most diverse microbial populations

Nutnent-poor media are prefened for plating the microbes, e g , diluted YM agar and diluted nutnent broth agar for the isolation of yeasts and bactena, respectively Colony charactenstics are more stable and better defined on nutnent-poor agar media than on nutnent-nch media Often, different strains or species will have subtle differences in colony moφhology which can be observed on nutnent-deficient media, but not on nutnent-nch media Moreover, fast- growing colonies of a few yeasts or bactena are less likely to overgrow nutnent-poor media plates and prevent the development of slower-growing isolates Pigmentation and macro- and micro- moφhological charactenstics of yeast and bactenal colonies have been found to be stable as long as the composition of the agar medium and the incubation temperature are not changed

In order to isolate diverse bactenal populations, the media is preferably supplemented with agents that prevent fungal growth without interfering with bactenal growth, e g , cyclohexamide, for the isolation of yeasts and filamentous fungi, the media is supplemented with an agent that restricts radial growth of filamentous fungi without affecting spore germination, e g , dichloran

In addition to the method of recovery, factors such as temperature, rainfall, fruit matunty, and regional and local geography have been reported to influence epiphytic microbial population of plants Therefore, epiphytic yeast and bactenal population size on fruits from different locations of the Pacific Northwest are different This simplified recovery method provides a powerful means to isolate naturally occurnng bactena, yeast, and filamentous fungi that are effective biological control agents This

method can also be used to study the population dynamics of yeasts and/or bacteria associated with the surface of an agricultural commodity such as pear or apple fruits.

The invention will be better understood by reference to the following examples, which are intended to merely illustrate the best mode now known for practicing the invention. The scope of the invention is not to be considered limited thereto, however.

EXAMPLES

EXAMPLE 1 : Improved method for obtaining microbes colonizing the surface of fruit and for studying their microbial ecology

Healthy mature Golden Delicious apple fruits (Maius domestica Borkh), free from punctures or other types of damage, were randomly selected from an unsprayed orchard. Twenty fruits were collected in each polythene bag. Three bags, each representing a replicate, were placed on ice during transit from the field to the laboratory. The samples were kept in an incubator at 5°C and processed within a day of sampling and after 4 and 8 weeks of storage.

To determine population size and microbial diversity in associated yeasts, bacteria, and filamentous fungi, three apple fruits, one from each bag, were sampled. Each fruit, along with its stem, was submersed in 600 ml beakers containing 200 ml SPBT (0.005 M sterile phosphate buffer, pH 7.0, 0.005% Tween 80). The beakers were placed on a rotaiy shaker (150 φm) for 5 min., after which the beakers containing the fruits were placed in an ultrasonic bath having an output of 270 watts and frequency of 43 KHz (model Q140, L & H Manufacturing Company, Kearny, NJ). During sonication, the fruits were gently rotated every 30 sec using a sterile stainless steel spatula. A 5.0 ml sample of buffer was removed from each beaker immediately before sonication and after 1, 2, 5, 10 and 15 min of sonication. The samples were vortexed for 1 min using Vari-Whirlmixer (VWR Scientific, WA) and three sequential 10-fold dilutions were made in

SPBT by adding a 0.5 ml of wash buffer to 4.5 ml of SPBT.

Cultivable populations of yeasts, bacteria and filamentous fungi were recovered by spreading 0.1 ml of suspension from each replicate onto duplicate plates of agar media. For bacterial populations and diversity, the samples were plated on diluted nutrient broth agar (Chand et al., Appl. Environ. Microbiol. 58:3374-3379, 1992) supplemented with cycloheximide (100 mg/ml) to prevent fungal growth (Hirano et al., Appl. Environ. Microbiol. 44:695-700, 1982). For yeasts and filamentous fungi, the samples were plated on diluted YM agar (DYMA), consisting of 4.1 g of Bacto YM broth (Difco Laboratories, Detroit, ML), 18 g Bacto-Agar, 100 mg chloramphemicol and 2 mg dichloran per liter of medium. Dichloran was added to the medium to restrict the radial growth of filamentous fungi without affecting spore germination (Byrde and Willetts, The Brown

Rot Fungi of Fruits: Their Biology and Control, Pergamon Press Ltd., Oxford, England, 1977; Pitt and Hocking, Fungi in Food Spoilage. Academic Press, New York, 1985). The inoculated plates

were incubated at 20±1°C and observed daily The number of bactenal colony forming units (CFU) per ml of wash buffer was estimated after 14 days of incubation (Chand et al , Appl Environ Microbiol 58 3374-3379, 1992) Yeast and filamemtous fungal CFU per ml of wash buffer were estimated after 7 days of incubation The surface area of each fruit was measured by peeling its skin with a potato peeler and determimng the total surface area (cm ² ) of skin with an area meter (Li-Cor model no LI-3000, Lincoln, NE) The number of CFU cm ² of each fruit was calculated from counts and surface area data

Colony charactenstics (pigmentation, size, opacity, form, elevation, margin, smoothness, texture, and spreading nature) of different types of bactena have been descnbed (Chand et al , Appl Environ Microbiol 58 3374-3379, 1992, Smibert and Kneg, "General Charactenzation," p 409-443 in Gerhardt et al (eds ), Manual of Methods for General Bactenology, A encan Society for Microbiology, Washington, DC, 1981) and were used to characteπze colomes Pigmentation and macro- and micro-moφhological charactenstics of yeast colonies were used to charactenze diverse yeasts (Guerzone and Marchetti, Appl Environ Microbiol 87 571-576, 1986, Martini et al , Can J Microbiol 26 856-859, 1980, Sasaki and Yoshida, J Facul Agr Hokkaido Unrv , Sapporo 51 194-220, 1959) Total counts of filamentous fungi were recorded without charactenzing their colomes, because different fungi can have similar colony charactenstics

To isolate the epiphytic yeasts, the fruits were treated as descnbed in the above expeπment, except that the volume of SPBT in each beaker was 200 ml and the fruits were sonicated for 5 mm The wash buffer of five fruits, representing a replicate, was combined and plated on DYMA plates, as descnbed above After 7 days of incubation, yeast colomes having diverse colony charactenstics were characteπzed and enumerated Yeast colomes having similar and/or different colony charactenstics were lifted and streaked on YM agar (Difco Laboratones, Detroit, MI ) for purification Purified yeast isolates were stored at -70 to -75°C (El-Dm et al ,

Zentralbl Mikrobiol 141 488-492, 1986)

Using the above descnbed approach, the epiphytic yeast microflora of mature pear fruits from three locations in Oregon (Hood River, Cascade Locks, Medford) and one location in Washington (Yakima) were also studied There were three replicates, each consisting of five randomly selected fruits from each location per cultivar (Table 4)

The yeasts were identified by Central Bureau for Fungal Culture (CBS), Baarn, the Netherlands, based on moφhological charactenstics, fermentation of sugars, and growth on different carbon and mϋogen sources

The lowest population sizes of yeast, bactenal, and filamentous fungi were recovered when samples were not sonicated (Table 1) When the samples were sonicated for 1 nun, the recovery of yeasts and bactena increased substantially The increase in counts of culturable cells

was slower from 1 to 5 nun of sonication Sonication for 10 or 15 nun resulted in little increase of recovery of yeast or bactenal CFU cm ^"2 of fruit surface Recovery of filamentous fungi, however continued to increase even at the longest sonication time Filamentous fungi adhere to the plant surface more tightly than saprophytic bactena and yeasts (Nicholson, "Adhesion of fungi to plant cuticle," pp 74-89 in Roberts and Aist (eds ), Infection Processes in Fungi, Rockefeller Foundation, New York, 1984, Nicholson and Epstein, "Adhesion of fungi to plant surface," pp 3-23 in Cole and Hoch (eds ), The Fungal Spore and Disease Initiation in Plants and Animals, Plenum, New York, 1991) and more time apparently was required to dislodge them from the fruit surface When ultrasonic waves travel through SPBT, they create a scrubbing action that is capable of cleaning up to 16 times more efficiently than hand scrubbing the fruits Microscopic examination of unsonicated fruit washings revealed the presence of clumps of yeast and bactenal cells Sonication for 5 nun separated these clumps into individual cells Probably for this reason the numbers of moφhologically distinct isolates of yeasts and bactena did not increase after 5 nun of sonication

In general, the epiphytic yeast population declined dunng two month storage of fruits at 5°C (Table 1) The population size of epiphytic bactena changed very little dunng storage and no definite pattern of population change was recorded for filamentous fimgi

The lowest and highest number of moφhologically distinct colomes of yeasts and bactena were recovered when samples were not somcated or somcated for 5 nun, respectively (Table 2) Increase in the sonication time from 5 to 10 or 15 nun did not result in any further increase in the diversity of isolates from the fruits

Diluted YM agar and diluted nutrient broth agar were used for the isolation of yeasts and bactena, respectively, because colony charactenstics have been reported to be more stable and better defined on nutnent-poor agar media than on nutnent-nch media (Chand et al , Appl

Environ Microbiol 58 3374-3379, 1992, Hattoπ, Rep Inst Agπc Res Tohoku Univ , 27 23-30, 1976, Waksman, Pπnciples of Soil Microbiology, The Williams & Wilkins Co , Baltimore, 1932) Occasionally, the buffer wash of the fruits was also plated onto nutnent-nch media (full strength YM agar and nutnent agar for yeasts and bactena, respectively) In such cases it was observed that the pigmentation and other colony charactenstics of yeasts and bactena were always better defined on nutnent-poor agar-media Additionally, fast-growing colomes of a few yeasts or bactena overgrew the nutnent-nch media plates and prevented the development of slow-growing isolates It was also observed that pigmentation and macro- and πucro-moφhological charactenstics of yeast and bactenal colomes were stable as long as the composition of the agar medium and the incubation temperature were not changed

The yeast and bactenal population on the surface of unsprayed mature apple fruits at harvest was determined to be approximately 80 x 10 ³ and 9 5 x 10 ⁴ CFU cm , respectively

It was observed that when two yeast isolates showed only minor differences in one of the colony characters such as shade of pigmentation or colony margin, they were identified as the same species; such variability is probably related to strain differences. The isolates that were moφhologically distinct from each other were always identified as different species of the same genus or yeasts of different genera. Thus, moφhological characteristics may be used to select the most diverse yeast colonists of plant surfaces.

The surface of pear fruit was found to be colonized by at least 31 moφhologically different strains of 10 yeast and yeast-like fungal species.

This simplified recovery method provides a powerful means to isolate yeasts and bacteria which are effective biocontrol agents, as discussed below. This method can also be used to study the population dynamics of yeasts and/or bacteria associated with the surface of fruit such as pear or apple fruits.

Table 1. Effect of sonication on population size of microorgansims recovered from surface of Golden Delicious apple fruits.

Storage Sonication Colony forming units (xl0 ³ )/cm ² of fruit time time (min) ^b surface"

(weeks)

Yeast Bacteria Filamentous fungi

0 0 0 1 1..44 ±± 00..55 14.2 ± 3.0 0.1 ± 0.0

1 4.0 ± 1.7 49.6 ± 11.9 0.3 ± 0.1

2 4.9 ± 1.3 56.7 ± 8.6 0.5 ± 0.1 5 5 7 7..77 ±± 11..77 73.1 ± 8.9 0.5 ± 0.1

10 7.9 ± 1.7 91.2 ± 7.6 0.8 ± 0.1

15 8.3 ± 2.1 95.4 ± 6.0 1.3 ± 0.3

0 1.9 ± 0.7 31.4 ± 6.5 1.7 ± 0.9 1 1 4 4..00 ± ± 00..88 60.2 ± 18.6 2.3 ± 1.2

2 4.7 ± 1.0 74.5 ± 22.6 2.8 ± 1.1

5 5.4 ± 0.8 95.5 ± 37.4 3.0 ± 1.1

10 5.8 ± 1.0 99.6 ± 35.3 3.8 ± 1.2

15 6.1 ± 1.3 93.3 ± 35.2 4.0 ± 1.4

0 1.1 ± 0.3 44.5 ± 19.0 1.0 ± 0.4

1 2.0 ± 0.8 71.7 ± 32.0 1.1 ± 0.3

2 2.2 ± 0.9 79.2 ± 36.8 1.6 ± 0.4

5 3.5 ± 1.3 104.3 ± 52.4 1.7 ± 0.4 1 100 3 3..22 ±± 11..11 115.2 ± 54.8 1.6 ± 0.5

15 3.4 + 1.4 118.6 ± 54.0 2.4 ± 0.9

"Each value is the average of three replications ± standard error. ^b Each fruit was placed in a beaker (600 ml) containing 200 ml of sterile phosphate buffer tween and placed on to a rotary shaker (150 rpm) before sonication. The wash buffer was plated on diluted YM agar and diluted nutrient broth agar plates for yeasts and bacteria, respectively.

Table 2. Effect of sonication and storage on the isolation of morphologically distinct yeast and bacterial isolates from the surface of Golden Delicious apple fruits*.

Number of moφhologically distinct isolates

Yeast Bacteria

Storage Sonication Min. Max. Average Min. Max. Average time time (min)

(weeks)

0 0 2 3 2.3 2 3 2.3

1 2 5 3.3 2 6 3.4

2 3 6 4.3 3 7 4.3

5 5 7 5.7 5 9 6.3

10 5 7 5.7 5 9 6.3

15 5 7 5.7 5 9 6.7

4 0 3 4 3.3 3 4 3.7

1 4 5 4.3 4 6 4.7

2 4 5 4.7 5 7 5.7

5 6 8 6.7 6 8 6.7

10 6 7 6.3 6 8 6.7

15 6 8 6.7 6 8 6.7

8 0 3 4 3.7 2 3 2.7

1 4 5 4.3 4 4 4.0

2 4 6 5.0 5 5 5.0

5 7 8 7.3 5 6 5.3

10 6 8 6.7 5 6 5.3

15 7 8 7.3 5 6 5.3 "Each fruit was placed in a beaker (600 ml) containing 200 ml of sterile phosphate tween and placed on to a rotary shaker (150 rpm) for 5 min before sonication. The wash buffer was plated on diluted YM agar and diluted nutrient broth agar for yeasts and bacteria, respectively.

Table 3. Population size of yeasts isolated from pear cultivars in the Pacific Northwest*.

Number of moφhologically distinct yeast isolates

Location Cultivar Yeast cfu\cm ² Min Max Average of fruit surface

Hood River Anjou 355 ± 33 5 6 5.7

Hood River Bartlett 990 ± 233 6 7 6.7

Medford Bosc 4115 ± 188 3 4 3.7

Cascade Locks Bartlett 6430 ± 616 6 7 6.7

Yakima Bartlett 7366 ± 2300 6 7 6.7

"Fruits (five) of each replicate (three replicates per location) were shaken for 5 min in SPBT and sonicated for five minutes before plating on DYMA medium.

Table 4. Yeast colonists of pears in different locations of the Pacific Northwest.

Relative abundance (% cfu) of yeasts'

Cascade

Hood River Locks Medford Yakima

Yeast Anjou Bartlett Bartlett Bosc Bartlett

Aureobasidium 59(2) 47(2) 39(1) 81(2) 52(2) pullulans

Cryptococcus albidus — 37(1) 21(1) 14(1) 25(2)

Cryptococcus infirmo- miniatus .. 1(1) — — 1(1)

Cryptococcus laurentii 13(1) 10(1) — — 19(1)

Debaryomyces hansenii — — 11(1) — —

Rhodotorula -- — 7(1) — — aurantiaca

Rhodotorula — — 14(1) ~ — fujisanensis

Rhodotorula glutinis 22(2) 5(2) — 5(1) 3(1)

Rhodotorula minuta 6(1) — 1(1) — —

Sporobolomyces roseus — — 7(1) — —

"Fruits (five) of each replicate (three replicates per location) were shaken for 5 min in SPBT and sonicated for 5 min before plating on to DYMA medium. Value in parenthesis represents the number of moφhologically different isolates.

EXAMPLE 2 Control of postharvest pear disease using natural saprophytic veast colonists and their integration with low dosage of thiabendazole

Saprophytic yeast colonists of mature pear fruits isolated and identified as descnbed above were tested for their effectiveness in controlling blue mold, gray mold, Mucor rot, bull's-eye rot, and side rot of pear fruit

Fruits Pear fruits, cv d'Anjou and Bosc, grown at the Mid-Columbia Agricultural

Research and Extension Center in Hood River, Oregon, were harvested At harvest, fruit firmness was 60 ± 2 and 62 ± 2 Newtons and soluble solids were 12 0 ± 2 and 14 ± 1 5, respectively, as determined with a UC pressure test penetrometer (Ametek, Belhngham, WA) Soluble solids were measured using a hand refractometer (VWR Scientific, Los Angeles, CA)

Fruits were air stored in cardboard boxes lined with perforated polythene bags at -1°C until use For different expeπments, the fruits were removed from cold storage, surface sterilized with 0 105% sodium hypochloπte (Clorox, Oakland, CA) for 2 nun, nnsed with running tap water, then air dπed Two puncture wounds (6 mm in diameter and 3 mm deep) were made about 3 5 cm apart on each fruit midway along the calyx-stem end axis After wounding, fruits were inoculated within 30 nun In all expeπments d'Anjou pears were used, except for the side rot tπal, in which

Bosc pears were used Side rot is a problem pnmaπly on Bosc pears (Sugar, "The disease cycle of side rot of pear, caused by Phtalophora malorum," Ph D thesis, Oregon State University, 1990)

Biocontrol yeasts Thirty-one moφhologically distinct strains of yeasts isolated from the surface of pear fruits by the method descnbed above were stored at -70°C These strains were activated by dispensing 70 ml of suspension into 70 ml of yeast malt dextrose broth (YMDB, consisting of 3 0 g Bacto yeast extract, 3 0 g Difco malt extract, 50 g Bacto peptone and 100 g

Bacto dextrose per liter) in a 200 ml flask The yeast suspensions were incubated on a rotary shaker

(125 rpm) for 48 h at 23 ± 1°C The suspension (0 1 ml per Petn dish) was spread on yeast malt dextrose agar (YMDA) medium (composition was similar to YMDB, 18 0 g Bacto agar liter ^'1 was added to solidify the medium) After 48 h of incubation at 23 ± 1° C, yeast growth from a Petn dish of each strain was removed with a stenle rubber policeman, suspended in 50 ml of stenle distilled water (SDW), and vortexed for 1 to 3 nun The yeast cells were centnfuged at 1 0 x 10 ⁴ φm for 15 min and the supernatant was discarded The cells were resuspended in SDW and the concentration was adjusted to 80 to 12 0 x 10 ⁵ cells per ml (70% transmittance at 550 nm using a Spectronιc-20 spectrophotometer (Bausch and Lomb Optical Co , Rochester, NY)

From the initial screemng of 31 yeast strains, four yeast strains that gave the best control of blue mold of pears were selected Cryptococcus laurentn strain HRA-5 , Rhodotorula aurantiaca strain YCL5 and Rhodotorula glutmts strains HRA-4 and HRB-6 The efficacy of these yeast strains in controlling blue mold was further tested using 1 6 to 2 4 x 10 ⁶ cells per ml (50% transmittance at 550 nm)

Of these fours strains, two strains, Cryptococcus laurentn strain HRA-5 and Rhodotorula glutmts strain HRB-6, were selected for further testing for control of blue mold when used alone or combined with a low dose of the fungicide thiabendazole (Mertect 340F, 15 mg ai ml ') Testing was conducted at 5, 10 and 20°C, using 1 5 - 2 0 x 10 ⁷ yeast cells per ml (6% transmittance at 550 nm)

Cryptococcus laurentn strain HRA-5, Rhodotorula glutmts strain HRB-6, and Cryptococcus mfirmo-mimatus strain YY-6 were screened for control of gray mold, Mucor rot, bull's-eye rot, and side rot of pear using 4 0 to 8 0 x 10 ⁷ yeast cells per ml (2% transmittance at 550 nm) Cryptococcus mfirmo-mimatus strain YY-6 was included because it showed the greatest effectiveness in controlling blue mold on Golden Delicious apples

Pathogens Penicillium expansum (strain 46), Botrytis cinerea (strain 62), Mucor pirtformis (strain 57, ATCC # 60988), Phtalophora malorum (strain 47) and Pezicula malicorticis (strain 14) were grown on potato-dextrose agar at 22 ± 1°C After 7 to 8 days of incubation, a spore suspension of Penicillium expansum was made in SDW containing 0005% Tween-80, filtered through a double layer of stenle muslin cloth, somcated for 5 min (Model T14, L & R

Manufactunng Co , Kearny, NJ) to break spore chains into individual spores, and vortexed for 1 nun to assure uniform mixing For making the suspension of B cmerea, the same procedure was used, except that 14-day -old cultures were used Spore suspensions of Mucor pirtformis and other two pathogens were made in SDW (without Tween-80 and sonication) after 7 and 14 days of incubation, respectively Spore concentrations were determined with a hemacytometer and suspensions were used within 2-3 h The concentrations of Penicillium expansum, Botrytis cmerea, Mucor pirtformis, Phtalophora malorum and Pezicula malicorticis used were 5 0 x 10 ³ , 2 0 x 10 ³ , 20 x 10 ³ , 4 0 x and 40 x 10 ³ , respectively

Fruit treatment Wounded fruits were placed on cardboard trays and each wound was inoculated with 50 ml of the fungus spore suspension The pathogemcity of 31 strains of yeasts on pears and their efficacy in controlling blue mold was tested by inoculating a treatment mixture (yeast alone, yeast mixed with P expansum, and P expansum alone as control) into wounds on five pear fruit per treatment Inoculated fruits were randomized, placed in cardboard boxes, incubated for 4 days at 5°C, then incubated for 6-7 days at 22°C before recording disease incidence (% of inoculated wounds that developed rot) and seventy (average rot diameter of diseased wound sites)

Four yeasts that showed good potential in controlling blue mold were studied further The efficac} of these yeasts in controlling blue mold was further tested using the method descnbed above, except that each treatment had three replicates of five fruits each

Cryptococcus laurentn and Rhodotorula glutmts were studied further for controlling blue mold at 5, 10 and 20°C Treatment mixtures (l e , yeast mixed with a pathogen, yeast integrated with a low concentration of the fungicide thiabendazole [15 mg ml '] and mixed with a pathogen, a mixture of the pathogen and a low or high concentration (525 mg ml ', as commercialh

recommended) of thiabendazole, and the pathogen alone as control] were applied in the wounds of pear fruits There were three replications of ten fruits per treatment Trays with inoculated fruits were randomized and placed in cardboard boxes lined with perforated polythene bags Disease incidence and disease seventy of blue mold were recorded after 6 to 7, 19 to 20 and 38 to 40 days of incubation at 5, 10 and 22°C storage, respectively

As descnbed above, treatment mixtures of yeasts with Botrytis cmerea, Mucor ptriformis, Pezicula malicorticis and Phtalophora malorum were made and applied into the wounds on pears

For Mucor rot and gray mold, there were three replications of three fruits per treatment for fruits incubated for 6 to 7 days at 22 ± 1°C There were 20 fruits per replication for fruits incubated for 30 and 60 days at -1° C The expenments were repeated three times with three replications of 20 fruits per treatment with Cryptococcus tnfirmo-mimatus at 22 ± 1 ° C For bull's- eye rot and side rots there were three replications of three fruits per treatment, and treated fruits were incubated for 30 days at 10 ± 1° C before recording disease data Statistical analysis Disease incidence was determined as the percentage of inoculated wounds that developed lesions Percent data were subjected to arcsine-square root transformation (Steel and Tome, Pnnciples and Procedures of Statistics, McGraw Hill, New York, 1980) before analysis of vanance, and means were separated by Duncan's multiple range test (P=0 05)

Results All yeasts and yeast-like fungal strains reduced rot diameter of diseased wounds, but only seven (approximately 23%) of the strains (Cryptococcus albidus strain HRB-2, C mfirmo-mimatus strain YY-6, C laurent strain HRA-5, Rhodotorula aurantiaca strain YCL-5, and R glutims strains HRA-3, HRA-4 and HRB-6) reduced disease incidence of blue mold on pears (Table 5) C laurentn strain HRA-5, R aurantiaca strain YCL5, and R glutints strains HRA-4 and HRB-6 were most effective in reducing disease incidence and seventy Control of blue mold of pears using C mfirmo-mimatus, R aurantiaca and R glutmts has not been previously reported

(Wisniewski and Wilson, HortScience 27 94-98, 1992) It is important to note that most of the yeast strains, particularly those in the genus Cryptococcus, were not effective biocontrol agents in

None of the yeasts caused any disease in the inoculated fruits, even when the fruits were incubated for 15 days at 22 ± 2°C The lesion diameter seldom exceeded 10 mm, even when inoculated fruits were incubated for 13-15 days at 22 °C

C laurentn strain HRA-5 and R glutims strain HRB-6 were most effective in reducing disease incidence and significantly reduced the seventy of blue mold on inoculated fruits (Table 6) None of the fruit inoculated with a yeast suspension alone developed disease C laurentn and R glutmts performed better for blue mold control of pears than in preliminary tests presented in Table 5 This may be due to higher numbers of yeast cells per ml in the inoculation mixture than were used in the preliminary tests Similarly, the efficacy of a yeast (Candida sp ) in

control ng blue and gray mold of Golden Delicious apples has been reported to increase with increased concentration of yeast cells (McLaughlin et al , Phytopathol 80 456-461, 1990)

Based on these results, C laurentn strain HRA-5 and R glutims strain HRB-6 were tested alone and in combination with a low dose (15 mg ml ^"1 ) of thiabendazole for the control of blue mold at 5, 10 and 20°C The yeasts alone or in combination with thiabendazole significantly reduced blue mold incidence and seventy at all temperatures (Tables 7 and 8)

The combination of yeasts with a low dose of thiabendazole gave significantly better disease control at all temperatures than the low dose of thiabendazole alone and was comparable to disease control achieved using a high dose of thiabendazole Using this approach, thiabendazole residues can be reduced by about 97% or more and still achieve excellent control of blue mold This approach is especially attractive to reduce thiabendazole residues on food Under packinghouse conditions, P expansum populations consist of benzimidazole-sensitive and insensitive strains (Spotts and Cervantes, Plant Dis 70 106-108, 1986) Under such conditions, for example, the yeast-thiabendazole combination protects against blue mold better than thiabendazole alone, even at high concentrations

The efficacy of C laurentn, R glutmts and C tnfirmo-mimatus alone and combined with a low dose of thiabendazole against Mucor rot was determined All of these yeasts alone or when combined with a low dose of thiabendazole significantly reduced disease incidence and seventy at -1 and 20° C Thiabendazole alone (low and high concentrations) failed to provide am protection against Mucor rot (Table 9) Benzimidazole fungicides have been reported to be ineffective in controlling Mucor rot (Maculates and Spotts, Plant Dis 74 537-543, 1990) In most cases, a combination of a yeast and a low dose of thiabendazole was not significantly better in controlling disease than yeast alone, probably due to lnsensitivity of Mucor pirtformis to benzimidazole fungicides C infirmo-mimatus was significantly better than all other treatments in reducing the incidence of Mucor rot, disease incidence was reduced from 100% to 42% at -1°C and from 100% to 11% at 20°C (Table 10) The effectiveness of these yeasts against Mucor rot, especially Cryptococcus infirmo-mimatus, is especially significant since all fungicides cunently registered for postharvest treatment of pome fruits are ineffective against Mucor pirtformis (Maculates and Spotts, Plant Dis 74 537-543, 1990)

The efficacy of C mfirmo-mimatus was tested in three different trials to confirm its effectiveness in the biological control of Mucor rot (Table 10) In all trials C infirmo-mimatus significantly reduced disease incidence and seventy The reduction in disease incidence varied from 62 to 89 percent, whereas the diameters of diseased wounds were reduced from 40 3 to 5 3, 28 3 to 8 8 and 31 3 to 86 in tnals 1, 2 and 3, respectively This is the first report documenting the biological control potential of C mfirmo-mimatus against Mucor rot

All treatments significantly reduced gray mold incidence and seventy on pears at - 1°C. When yeasts were integrated with a low dose of thiabendazole, the reduction in gray mold incidence was not significantly different from the fruits treated with a high dose of thiabendazole alone and stored at -1 and 20°C (Table 11). Fruits with apparently healthy wounds after 60 days of storage at -1°C were ripened for 5 days at 20°C, and more than 95% of the wounds remained healthy. It usually takes 45 to 60 days at -1°C for pear wounds to heal. Probably due to wound healing, gray mold incidence did not increase. In the case of Mucor rot, approximately 50% of apparently healthy wounds (after 30 days of fruit storage at -1°C) developed rot when incubated for five days at 20°C, indicating survival of the pathogen and partial wound healing. All yeasts were more effective in reducing disease severity when integrated with a low dose of thiabendazole than alone. For the first time we report the control of gray mold of pears using three yeast C. laurentn, R. glutinis and C. infirmo-mimatus.

Side rot of Bosc pears was completely controlled by all yeasts alone or when integrated with a low dose of thiabendazole (Table 12). Thiabendazole (15 and 525 mg ml ¹ ) was ineffective in reducing disease incidence and severity due to insensitivity of P. malorum to benzimidazole fungicides (Sugar et al., Plant Dis. 78:791-795, 1994). This is the first report of control of side rot of pears with R. glutinis or C. infirmo-miniatus.

All yeasts alone or when integrated with a low dose of thiabendazole significantly reduced the incidence and severity of bull's-eye rot (Table 12). Yeasts, when integrated with a low dose of thiabendazole, completely controlled bull's-eye rot, and control was not significantly different than using the high dose of thiabendazole (Table 12). There are apparently no previously published reports on the control of bull's-eye rot using biological control agents.

It is important to note that most of the yeast strains tested, particularly those in the genus Cryptococcus, were determined to be ineffective biocontrol agents. There are significant differences among strains of even a single species of yeast in their effectiveness in controlling fungal disease.

Table 5

Efficacy* of yeasts and yeast-like fungal colonists of pears, isolated from different locations of the Pacific Northwest of the United States, in controlling blue mold of d'Anjou pears.

Yeast Pathoge

Treatment Location* Cultivar isolate -nicity ^y DP DS*

Aureobasidium HR Anjou HRA1 + 100 19.5 pullulans

HRA2 + 100 17.8

Bartlett HRB1 + 100 19.0

HRB4 + 100 20.3

CL Bartlett YCL1 + 100 20.9

M Bosc YMB1 + 100 26.7

YMB3 + 100 27.4

Y Bartlett YY1 + 100 25.2

YY4 + 100 21.5

Crvptococcus HR Bartlett HRB2 - 90 16.4 albidus

CL Bartlett YCL2 - 100 19.4

M Bosc YMB2 - 100 23.1

Y Bartlett YY2 - 100 20.5

YY8 - 100 19.9

Cryptococcus HR Bartlett HRB7 - 100 17.5 infirmo-miniatus

Y Bartlett YY6 - 90 19.9

YY7 - 100 25.7

Crvptococcus HR Anjou HRA5 - 80 13.2 laurentii

HRB3 - 100 21.8

Y Bartlett YY3 - 100 24.8

Debarvomvces CL Bartlett YCL - 100 23.9 hansenii

RhodptQryla CL Bartlett YCL5 - 50 15.8 aurantiaca

^' 26a

Yeast Pathoge

Treatment Location* Cultivar isolate -nicity ^y DF DS ^Z

Rhodotorula CL Bartlett YCL3 100 25.9 fujisanensis

Rhodotorula HR Anjou HRA3 90 16.1 glutinis

HRA4 90 12.1

Bartlett HRB5 100 21.9

HRB6 80 14.0

M Bosc YMB4 100 25.3

Rhodo or l HR Bartlett HRA6 100 17.6 minuta

CL Bartlett YCL7 100 21.0

Sporobolomyces CL Bartlett YCL6 100 21.9 roseus

Penicillium 100 35.5 expansum alone

"The wounds were treated with a spore suspension (5.0 x 10 ³ ml ^"1 ) of Penicillium expansum alone or a mixture of yeast suspension (8.0-12.0 x 10 ^s cells ml ^"1 ) and Penicillium expansum. Data were recorded after 4 days of incubation at 5°C followed by 6-7 days of incubation at 22 +. 2°C.

ΗR, Hood River (Oregon); CL, Cascade Locks (Oregon); M, Medford (Oregon) and Y, Yakima (Washington). ^y +, yeast alone caused dry rot at the inoculated wound sites. -, yeast alone caused no disease at the inoculated wound sites.

^γ + , yeast alone caused dry rot at inoculated wound sites. -, yeast alone caused no disease at inoculated wound sites. ^z Disease incidence (DI) is the percent of treated wounds that developed rot. Disease severity (DS) is rot diameter (mm) of diseased wounds.

-27-

Table 6

Efficacy ^y of 4 yeast isolates in controlling blue mold of d'Anjou pears.

Treatment Isolate DF DS ^Z

C. laurentii HRA5 68a 14.8a

R. aurantiaca YCL5 92bc 16.7a

R. glutinis HRA4 82ab 15.2a R. glutinis HRB6 68a 15.8a

Penicillium expansum alone 46 100c 31.1b

'The wounds were treated with a spore suspension (5.0 x 10 ³ spores ml ^"1 ) of Penicillium expansum or a mixture of yeast suspension (8.0-12.0 x 10 ⁵ ml ^*1 ) and Penicillium expansum. Data were recorded after 6-7 days of incubation at 22 +_ 2° C.

*Disease incidence (DI) is the percent of treated wounds that developed rot. Disease severity (DS) is rot diameter (mm) of diseased wounds. Within a column, values with the same letters are not significantly different at P = 0.05 according to Duncan's multiple range test.

-28-

Table 7

Efficacy ^y of Crvptococcus laurentii (isolate HRA5) and Rhodotorula glutinis (isolate HRB6) alone or in the presence of a reduced rate of thiabendazole (TBZ) in controlling incidence of blue mold of pears caused by Penicillium expansum.

Diseased wounds (%) ^z

5° C 10° C 20° C

Treatment 1993 1994 1993 1994 1993 1994

C. laurentii 0a 13b 7b 43b 47b 50b

C. laurentii + TBZ 0a 0a 0a 7a 0a 7a

R, l inis 7b 10b 17c 33b 53b 47b

R. glutinis + TBZ 0a 0a 0a 3a 3a 3a

TBZ (15 μg ml ^"1 ) 7b 40c 13bc 43b 73c 93c

TBZ (525 μg ml ^"1 ) 0a 0a 0a 3a 0a 3a

P. expansum alone 53c 93d 83d 100c 100c 100c

HTie wounds were treated with a spore suspension of P. expansum (5.0 x 10 ³ ml ^"1 ) alone, a mixture of TBZ (15 and 525 μg ml ^"1 ) and L expansum. or yeast suspension (2.0-3.0 x 10 ⁷ cells ml ^"1 ) and P. expansum with a reduced rate of TBZ or without it. Data were recorded after 35-40, 17-20, and 6-8 days of incubation at 5, 10, and 20° C, respectively.

^z Within a column, values with the same letters are not significantly different at P = 0.05 according to Duncan's multiple range test.

-29-

Table 8

Efficacy ^y of Cryptococcus laurentii (isolate HRA5) and Rhodotorula glutinis (isolate HRB6) alone or in the presence of a reduced rate of thiabendazole (TBZ) in controlling severity of blue mold on pears caused by Penicillium expansum.

Lesion diameter (mm)'

5° C 10° C 20° C Treatment 1993 1994 1993 1994 1993 1994

C. laurentii NI 7.6a 14.7a 15.6b 13.0b 9.4a

C. laurentii + TBZ NI NI NI 5.3a NI 5.7a

R. glutinis 4.4a 4.8a 26.3ab 18.5b 13.6b 10.2a

R. glutinis + TBZ NI NI NI 3.6a 2.3a 4.3a

TBZ (15 μg ml ^"1 ) 6.3a 11.6b 39.0b 19.8b 30.3c 24.5b

TBZ (525 μg ml ^"1 ) NI NI NI 33a NI 5.6a

P. expansum alone 23.5b 26.5c 40.7c 37.2c 33.3c 32.0b

Hhe wounds were treated with a spore suspension of P. expansum (5.0 x 10 ³ ml ^"1 ), a mixture of TBZ (15 and 525 μg ml ^"1 ) and P. expansum. or yeast suspension (2.0-3.0 x 10 ⁷ cells ml" ¹ ) and P. expansum with a reduced rate of TBZ or without it. Data were recorded after 35-40, 17-20, and 6- 8 days of incubation at 5, 10, and 20° C, respectively.

^z Within a column, values with the same letters are not significantly different at P = 0.05 according to Duncan's multiple range test.

-30-

Table 9

Efficacy ^y of Cryptococcus laurentii (isolate HRA5), Rhodotorula glutinis. (isolate HRB6), and Crvptococcus infirmo-miniatus (isolate YY6) alone or in the presence of a reduced rate of thiabendazole (TBZ) in controlling Mucor rot caused by Mucor piriformis at -1 and 20°C

-1° C 20° C

Treatment DF DS ^Z DI DS

C. laurentii 92cd 17.7c 82bc 20.9b

C. laurentii + TBZ 83bc 15.4bc 72b 26.9bc

R. glutinis 70b 14.9bc 83bc 26.0bc

R. glutinis + TBZ 76b 16.6c 83bc 25.2bc

C. infirmo-miniatus 49a 12.5b 11a 5.3a

C. infirmo-miniatus + TBZ 42a 7.9a 11a 7.7a

TBZ (15 μg ml ^"1 ) 97de 29.0d 94c 31.7cd

TBZ (525 μg ml ^"1 ) 98de 28.4d 100c 32.2cd

Mucor Diriformis alone lOOe 29.8d 100c 40.6d

The wounds were treated with a spore suspension (2.0 x 10 ³ ml ^"1 ) of

M. piriformis. a mixture of TBZ (15 and 525 μg ml ^'1 ) and M. piriformis. or yeast suspension (1.4-1.6 x 10 ⁸ cells ml ^*1 ) and M. piriformis mixed with a reduced rate of TBZ or without it. Data were recorded after 30 and 6-7 days of incubation at -1 and 22° C, respectively.

^z Disease incidence (DI) is the percent of treated wounds that developed rot. Disease severity (DS) is rot diameter (mm) of diseased wounds. Within a column, values with the same letters are not significantly different at P = 0.05 according to Duncan's multiple range test.

-31-

Table 10

Efficacy ^y of Crvptococcus infirmo-miniatus (isolate YY6) in controlling Mucor rot caused by Mucor piriformis at 22°C.

Trial 1 Trial 2 Trial 3

Treatment DI ^Z DS ^Z DI DS DI DS

C. infirmo-miniatus 11a 5.3a 38a 8.8a 23a 8.6a Mucor piriformis alone 100b 40.3b 100b 28.3b 100b 31.3b

The wounds were treated with a spore suspension (2.0 x 10 ³ ml ^"1 ) of Mucor piriformis or with a mixture of yeast suspension (1.4-1.6 x 10 ⁸ cells ml ^*1 ) and Mi piriformis. Disease data were recorded after 6-7 days of incubation at 22° C.

^z Disease incidence (DI) is the percent of treated wounds that developed rot. Disease severity (DS) is rot diameter (mm) of diseased wounds. Within a column, values with the same letters are not significantly different at P = 0.05 according to Duncan's multiple range test.

-32-

Table 11

Effϊcacy ^y of Crvptococcus laurentii (isolate HRA5), Rhodotorula glutinis (isolate HRB6), and Cryptococcus infirmo-miniatus (isolate YY6) alone or in the presence of a reduced rate of thiabendazole (TBZ) in controlling gray mold caused by Botrytis cinerea at -1 and 20°C.

-1° C 20° C Treatment DF DS ^Z DI DS

C. laurentii 65c 34.3cd 44c 19.4b

C. laurentii + TBZ 3a 22.0bcd 17ab 7.0a

R. glutinis 57c 36.0d 44c 14.4ab

R. glutinis + TBZ 2a 14.7ab 17ab 12.0ab C. infirmo-miniatus 65c 30.9bcd 22b 9.0a

C. infirmo-miniatus + TBZ 3a 17.1abc llab 5.6a

TBZ (15 μg ml ^"1 ) 35b 33.1cd lOOd 35.8c

TBZ (525 μg ml ^*1 ) 0a NI 6a 6.6a

Botrvtis cinerea alone lOOd 83.6e lOOd 40.7c

The wounds were treated with a spore suspension (2.0 x 10 ³ ml ^'1 ) of β. cinerea. a mixture of TBZ (15 and 525 μg ml ^"1 ) and B. cinerea. or yeast suspension (1.4-1.6 x 10 ⁸ cells ml ^"1 ) and B. cinerea with a reduced rate of TBZ or without it. Data were recorded after 60 and 6-7 days of incubation at -1 and 22° C.

^z Disease incidence (DI) is the percent of treated wounds that developed rot. Disease severity (DS) is rot diameter (mm) of diseased wounds. No visual infection (NI) at inoculated wound sites. Within a column, values with the same letters are not significantly different at P = 0.05 according to Duncan's multiple range test.

-33-

Table 12

Efficacy of Cryptococcus laurentii (isolate HRA5), Rhodotorula lutinis (isolate HRB6), and Cryptococcus infirmo-miniatus (isolate YY6) alone or in the presence of a reduced rate of thiabendazole (TBZ) in controlling side rot of Bosc pears and bull's-eye rot of d'Anjou pears caused by Phialophora malorum and Pezicula malicorticis. respectively.

P, malorum P. malicorticis

Treatment DF DS ^Z DI DS

C. laurentii 5a 2.0a lib 5.3ab

C. laurentii + TBZ 5a 1.7a 0a NI

R. glutinis 0a NI 5ab 2.7a

R. glutinis + TBZ 0a NI 0a NI

C. infirmo-miniatus 0a NI lib 4.3ab

C. infirmo-miniatus + TBZ 0a NI 0a NI

TBZ (15 μg ml ^'1 ) 72b 7.4b 56c 8.6bc

TBZ (525 μg ml ^'1 ) 72b 7.7b 0a NI

Fungal pathogen alone 80b 8.8b lOOd 12.2c

The wounds were treated with a fungal (4.0 x 10 ³ spores ml ^"1 ) pathogen alone, a mixture of TBZ (15 and 525 μg ml ^"1 ) and a fungal pathogen, or yeast suspension (1.4-1.6 x 10 ⁸ cells ml ^*1 ) and a fungal pathogen with a reduced rate of TBZ or without it. Data were recorded after 30 days of incubation at 10° C.

^z Disease incidence (DI) is the percent of treated wounds that developed rot. Disease severity (DS) is rot diameter (mm) of diseased wounds. No visual infection (NI) at inoculated wound sites. Within a column, values with the same letters are not significantly different at P = 0.05 according to Duncan's multiple range test.

- 34 -

EXAMPLE 3: Control of postharvest blue mold of apples using natural saprophytic yeast colonists Fruits. Apple fruits (cv. Golden Delicious) grown at the Mid-Columbia Agricultural Research and Extension Center in Hood River, Oregon, were harvested in the second week of September during 1993 and 1994. At harvest, fruit firmness was 55 ± 2 and 50 ± 2 Newtons (0.225 lb. per Newton) and percent soluble solids were 12.0 ± 2 and 13 ± 1.5, in 1993 and 1995, respectively. Firmness was determined with a UC pressure test penetrometer (A etek, Bellingham. WA) and soluble solids using a hand refractometer (VWR Scientific, Los Angeles, CA). Fruits were stored at 0°C in cardboard boxes lined with perforated polythene bags until use.

On each apple, two puncture wounds (6 mm in diameter and 3 mm deep) were made about 3.5 cm apart midway along the calyx-stem axis. After wounding, apple fruits were inoculated within 30 min.

Biocontrol yeasts. Six yeast strains (two strains of Cryptococcus spp. and four strains of Rhodotorula spp.) isolated from the surface of pear fruits and shown to control blue mold of pear as described above were selected for the apple studies. Yeasts were stored at -70°C and were activated by dispensing 70 ml of suspension into 70 ml of yeast malt dextrose broth (YMDE; consisting of 3.0 g Bacto yeast extract, 3.0 g Difco malt extract, 5.0 g Bacto peptone and 10.0 g Bacto dextrose liter ^'1 of medium) in a 200 ml flask. Yeast suspensions were incubated on a rotary shaker at 125 rpm for 48 hours at 23 ± 1°C The suspension (0.1 ml per Petri dish) was spread on yeast malt dextrose agar (YMDA) medium (similar to YMDB, except 18.0 g Bacto agar liter ^"1 was added to solidify the medium). After 48 hours of incubation at 23 ± 1°C, yeast growth was removed with a sterile rubber policeman, suspended into 50 ml of sterile distilled water (SDW), and vortexed for 1-3 minutes. The yeast cells were centrifuged at 1.0 x 10" rpm for 15 minutes and the supernatant was discarded. The cells were resuspended in SDW, and the concentration was adjusted to 1.6 - 2.4 x 10 ⁶ cells per ml (50% transmittance at 550 nm using a spectrophotometer model Spectronic 20, Bausch and Lomb Optical Co., Rochester, NY).

From the initial screening of six yeast strains, two yeast strains, Cryptococcus infirmo-miniatus strain YY6 and C. laurentii strain HRA5, gave the best control of blue mold of apples. The efficacy of either of these yeasts alone or when integrated with a low dose of the fungicide thiabendazole (TBZ, Mertect 340F, 15 mg ai ml ^'1 ) was tested for control of blue mold at 5, 10 and 20°C using 2.0 - 3.0 x 10 ⁷ yeast cells per ml

(6% transmittance at 550 nm).

Pathogens. Penicillium expansum strain 46 was grown on potato dextrose agar at 22 ± 1°C. After 7-8 days of incubation, a spore suspension was made in SDW containing 0.005% Tween-80 (SDWT), filtered through a double layer of sterile muslin cloth, sonicated for 5 minutes (Model T14, L&R Manufacturing Co., Kearny, NJ) to break spore chains into individual snores, and vortexed for one minute to assure uniform mixing. Spore concentrations were determined with a hemacytometer, and suspensions were used within two to three hours.

- 35 -

During the course of this investigation the concentration of Penicillium expansum used was 5.0 x 10 ³ spores/ml.

Fruit treatment. The efficacy of six strains of yeasts for control of blue mold was tested by inoculating wounded apple fruits with 50 ml per wound of yeast alone, yeast mixed with P. expansum, and P. expansum alone as control. There were five replications of five fruits per treatment. Trays with inoculated fruits were randomized and placed in cardboard boxes lined with perforated polythene bags. Fruits were incubated for 7-8 days at 22°C before recording disease incidence (percentage of inoculated wounds that developed rot) and severity (average rot diameter of diseased wound sites). Two yeasts, Cryptococcus infirmo-miniatus and C. laurentii, showed the greatest potential for control of blue mold and were selected for further studies. The efficacy of these two yeasts for control of blue mold at 5, 10 and 20°C was tested. The following treatment mixtures were applied in the wounds of apple fruits: P. expansum alone, a mixture of TBZ (15 and 525 mg ml ^'1 and P. expansum, or yeast suspension and P. expansum with the low rate of TBZ or without it. There were three replications of five fruits per treatment. Trays with inoculated fruits were randomized and placed in cardboard boxes lined with perforated polythene bags. Incidence and severity of blue mold were recorded after 6-8, 17-20 and 35-40 days of incubation at 5, 10 and 20°C, respectively.

Population studies of yeasts in wounds. The ability of yeast cells to survive and multiply in apple wounds was studied to determine if these yeasts were effective colonizers. Fruits were wounded, then inoculated with 50 ml of a suspension containing 2 x 10 ⁷ yeast cells ml ^'1 . Fruits were incubated in plastic boxes at 0, 5, 10 and 20°C. Viable cell concentrations of the inoculum were determined by a modified dilution plate frequency technique (Chand et al., Appl. Environ. Microbiol. 58:3374-3379, 1992). Individual fruits served as one replicate in a randomized complete block design, and three replicates were sampled at each time and temperature. For the fruits incubated at 5, 10 and 20°C, the wounds were sampled 0, 1, 2, 3, 4 and 7 days after inoculation, whereas at 0°C, samples were removed 0, 3, 10, 15 and 30 days after inoculation. Samples were taken using a cork borer (10 mm internal diameter) and macerated in distilled sterile water. Serial dilutions were plated in duplicate on YMDA using the modified dilution plate frequency technique (Chand et al., 1992) and incubated at 22 ± 2°C for 4 days when yeast populations were determined and expressed as logio CFU per wound.

Statistical analysis. Disease incidence was the percentage of inoculated wounds that developed lesions. Percent data were subjected to arsine-square root transformation (Steel and Torrie, Principles and Procedures of Statistics, p. 158, McGraw-Hill, New York, 1980) before analysis of variance, and means were separated by Duncan's multiple range test (P=0.05). Disease severity was the rot diameter of diseased wound sites and was statistically analyzed with the method above for incidence.

- 36 -

Results All six yeasts significantly reduced both incidence and seventy of blue mold on apple fruit (Table 13) When both of these measures of disease were considered, C tnfirmo- mimatus and C laurentii were the most effective of the six yeasts, and these two were selected for further study Control of blue mold of apple with C tnfirmo-mimatus and C laurentii was evaluated at 5, 10, and 20°C At 5°C, both yeasts alone and with TBZ reduced the incidence of blue mold, and there were no significant differences (P=0 05) between biocontrol treatments (Table 14) At 10°C, C laurentii alone and TBZ at 15 mg ml ^] were less effecUve than other treatments, differences were significant (P=005) in 1993, but not in 1994 At 20°C, a treatment with either yeast alone or either yeast together with a low dose of TBZ alone was less effecUve than a treatment employing one of the yeasts together with TBZ or a high dose of TBZ alone However, either yeast alone or either yeast together with a low dose of TBZ provided a significant reduction of incidence of blue mold when compared with the control (P expansum alone) (Table 14) Disease incidence was greater in 1993 than in 1994 The effect of yeast treatments on seventy of blue mold was similar to the effects on incidence (Table 15)

The population dynamics of the yeasts in apple wounds was studied at 0, 5, 10, and 20°C At 0°C, the populaUons of C mfirmo-mimatus and C laurentii increased in logio umts from about 6 6 to a maximum of about 7 8 per wound in 10 days (Figure 1A) This was followed by a gradual decline to about 7 6-7 7 log umts by day 30 A similar pattern was observed at temperatures between 5 ^C C and 20°C the populaUons of the yeasts increased about 1 4 log umts to a maximum of 74 in the first two days (Figures IB, 1C, and ID) Yeast populaUons declined over the next five days to about 7 1 log umts per wound

The yeasts C infirmo-mimatus and C laurentii were thus shown to be effective in the biocontrol of blue mold of apple at 5°C, but were less effecϋve at 10°C or 20°C when the yeasts were used alone When C mfirmo-mimatus or C laurentii were combined with TBZ at 15 mg ml ' , excellent control was achieved at all temperatures and was equivalent to the full, labeled rate of TBZ The low rates of TBZ and iprodione in this study represent reductions of 97 and 98%, respectively, of the full labeled rates (Willet et al , Postharvest Pomology Newsletter 7 1-15, 1989) Thus, a significant reducfion in fungicide residues is possible with these combinauon treatments Biocontrol over a wide temperature range is important with apples, because fruit are stored at 0°C or slightly warmer (U S Dept of Agnculture, Agriculture Research Service, The commercial storage of fruits, vegetables, and florist and nursery stocks, pp 23-26, Agπc Handbook 291, 1968), then held at warmer temperatures in terminal markets

Fruit matuπty also effects suscepUbility to decay (Pierson et al , Market diseases of apples, pears, and quinces, U S Dept of Agnculture Handbook 376, U S Department of

Agnculture, Agncultural Research Service, 1971, at p 2) Apple fruit appeared to be considerably

- 37 -

more susceptible in 1993 than in 1994 (Table 14) Nevertheless, the combination treatments of yeast with the low rate of TBZ controlled blue mold in both years

In addition to varying temperatures and fruit matuπty levels, apple fruit also are infected in the postharvest environment by a vanety of pathogens (Pierson et al , 1971) As descnbed above, C firmo-mimatus and C laurentii control gray mold, bull's-eye rot, Mucor rot, and side rot of pear These pathogens also affect apple fruit, and apples can be similarly treated with the yeast strains of the present invention alone or in combination with appropnate chemical fungicides

Resistance of P expansum to benzimidazoles has been reported (Wicks, Plant Dis Reptr 61 447-449, 1977) Because the mode of action of C mfirmo-mimatus and C laurentii does not involve antibiosis, as discussed above, it is unlikely that resistance will develop to these yeasts

-38-

Table 13. Efficacy ^y of 6 yeast strains in controlling blue mold of Golden Delicious apples.

Treatment Strain DI ² DS ²

C. infirmo-miniatus YY6 8a 3.6a

£_ laure ii HRA5 12a 3.7a R, aurantiaca YCL5 36ab 10.3b R_ glutinis HRA3 36ab 11.8b R. glutinis HRA4 32ab 10.5b R. glutinis HRB6 36ab 8.6ab

Penicillium expansum alone 100c 31.1c

The wounds were treated with a spore suspension (5.0 x 10 ³ spores ml ^"1 ) of Penicillium expansum or a mixture of yeast suspension (1.6 - 2.4 x 10 ⁶ ml ^"1 ) and Penicillium expansum. Data were recorded after 7-8 days of incubation at 22 ± 2° C. ^z Disease incidence (DI) is the percent of treated wounds that developed rot. Disease severity (DS) is rot diameter (mm) of diseased wounds. Within a column, values with the same letters are not significantly different at P = 0.05 according to Duncan's multiple range test.

-39-

Table 14. Efficacy ^y of Cryptococcus infirmo-miniatus (strain YY6) and Cryptococcus laurentii (strain HRA5) alone or in the presence of a reduced rate (15 μg ml ^"1 ) of thiabendazole (TBZ) in controlling incidence of blue mold of Golden Delicious apples caused by Penicillium expansum.

Diseased wounds (%) ²

5°C 10°C 20°C

Treatments 1993 1994 1993 1994 1993 1994

C. infirmo-miniatus 0a 0a 3a 3a 37bc 7a

C. infirmo-miniatus + TBZ 0a 0a 0a 0a 3a 0a

C, lauremii 3ab 3ab 37c 10a 56c 17b

C. laurentii + TBZ 0a 0a 3a 0a 3a 0a

TBZ (15 ppm ai) 7b 7b 13b 10a 23b 17b

TBZ (525 ppm ai) 0a 0a 0a 0a 3a 0a expansum 43c 33c 97d 67b lOOd 87c

The wounds were treated with a spore suspension of H expansum (5.0 x 10 ³ ml ^"1 ) alone, a mixture of TBZ (15 and 525 ppm ai) and £. expansum. or yeast suspension (2.0 - 3.0 x 10 ⁷ ml ^"1 ) and P. expansum with a reduced rate of TBZ or without it. There were three replicates of 5 fruits each per treatment. Data were recorded after 35-40, 17-20, and 6-8 days of incubation at 5, 10, and 20°C, respectively. ^z Within a column, values with the same letters are not significantly different at P = 0.05 according to Duncan's multiple range test.

-40-

Table 15 Efficacy ^y of Crvptococcus infirmo-miniatus (strain YY6) and Cryptococcus laurentii (strain HRA5) alone or in the presence of a reduced rate (15 μg ml ^"1 ) of thiabendazole (TBZ) in controlling severity of blue mold of Golden Delicious apples caused by Penicillium expansum.

Lesion diameter (mm) ²

5° C 10° C 20° C

Treatments 1993 1994 1993 1994 1993 1994

C. infirmo-miniatus NI NI 7.7a 2.0a 16.6b 5.3a

C. infirmo-miniatus NI NI NI NI 3.3a NI + TBZ

C, lauremii 6.6a 3.3a 21.5b 7.8a 19.9b 10.8b

£. laurgntϋ + TBZ NI NI 3.3a NI 4.7a NI

TBZ (15 ppm ai) 5.7a 7.8a 12.3ab 9.2a 20.8b 14.3c

TBZ (525 ppm ai) NI NI NI NI 4.3a NI

P. expansum 36.2b 36.2b 33.7c 29.7b 34.3c 30.0d

NI, no visual infection at inoculated wound sites.

The wounds were treated with a spore suspension of P. expansum (5.0 x 10 ³ ml ^"1 ) alone, a mixture of TBZ (15 and 525 ppm ai) and P. expansum. or yeast suspension (2.0-3.0 x 10 ⁷ ml ^"1 ) and P. expansum with a reduced rate of TBZ or without it. There were three replicates of 5 fruits each per treatment. Data were recorded after 35-40, 17-20, and 6-8 days of incubation at 5, 10, and 20°C, respectively.

^z Within a column, values with the same letters are not significantly different at P = 0.05 according to Duncan's multiple range test.

- 41 -

EXAMPLE 4 Control of postharvest brown rot of chernes using natural saprophytic yeast colonists

Fruits Sweet cherry fruits (cv Lambert) were harvested on 15 June 1994 at commercial matunty, stored at -1° C and used within 20 days of harvest Healthy-looking apple and cherry fruits were removed from cold storage for different expenments, surface sten zed with 0 105% sodium hypochlonte (The Clorox Co , Oakland, CA) for 2 minutes, rinsed with tap water, then air-dned Cherry fruits were not wounded

Biocontrol yeasts The efficacy of Cryptococcus infirmo-miniatus strain YY6 and C laurent strain HRA5 alone or when integrated with a low dose of the fungicide φrodione (Rovral 50 WP, 20 mg ai ml ^'1 ) was tested for control of brown rot of sweet cherry using 2 0 - 3 0 x 10 ⁷ yeast cells per millihter (6% transmittance at 550 nm)

Pathogens For Mom lima fructicola, diseased cherry fruits with extensive spomlauon of the pathogen on their surface were collected from an orchard at the Mid-Columbia Agncultural Research and Extension Center Spores of M fructicola were removed by dipping 15 diseased fruits in 50 ml of SDWT, and the resulting suspension was somcated for 3 minutes and vortexed for 1 minute Spore concentrations were determined with a hemacytometer, and suspensions were used within 2-3 hours The concentration of Momlinia fructicola used was 1 0 x 10 ⁴ spores/ml Fruit treatment Sweet cherry fruits were treated as descnbed above with the following M fructicola alone, a mixture of lprodione (20 and 1175 mg ml ^"1 ) and M fructicola, or a yeast suspension and M fructicola with the low rate of lprodione or without it There were three replicates of 50 cherry fruits each Fruits were transferred into nylon nets and immersed for one minute into treatment suspensions Fruits were drained for two minutes and stored in perforated plastic bags at 2 8°C Disease was evaluated after 20 days

Statistical analysis Disease incidence was the percentage of inoculated wounds that developed lesions Percent data were subjected to arsine-square root transformation (Steel and Tome, 1980) before analysis of vanance, and means were separated by Duncan's multiple range test

(P=005) Disease seventy was the rot diameter of diseased wound sites and was statistically analyzed with the method above for incidence

Results The reduction of brown rot of sweet cherry was not significant with C infirmo-mimatus, C laurentn, or the low rate of lpodione (Table 16) However, when either yeast was combined with the low rate of lprodione, significant (P=0 05) control was achieved, and the level of control was equivalent to that obtained with the full dose of φrodione

Resistance of M fructicola to lprodione has been reported (Elmer and Gaunt, Crop Protection 12 83-88, 1993) Because the mode of action of C mfirmo-mimatus and C laurentii does not involve antibiosis, as discussed above, it is unlikely that resistance will develop to these yeasts

Postharvest losses of sweet cherry to fungal decay occur despite application of fungicides and other recommended measures (Willett et al , Postharvest Pomology Newsl 7 1-15,

- 42 -

1989). After the opening of the Japanese market in 1978 to U.S. cherries, the volume of exported cherries has steadily increased. At the same time, the need to reduce postharvest losses also has increased. Alternatives to fungicides are needed because Japan has not established a regulatory tolerance for postharvest fungicide residues (Dugan and Roberts, Phytopathol. 84: 1031-1036, 1994). This is the first report of control of brown rot of sweet cherry with naturally-occurring saprophytic yeasts.

-43-

Table 16. Efficacy ^y of Cryptococcus infirmo-miniatus (strain YY6) and Crvptococcus laurentii (strain HRA5) alone or in the presence of a reduced rates (20 μg ml ^'1 ) of iprodione in controlling disease incidence (%) of brown rot of sweet cherries (cv. Lambert) caused by Monilinia fructicola.

Treatments Diseased Fruits ( ) ^z

£. infirmo-miniatus 14.7bc

£_ infirmo-miniatus + iprodione 2.0a £. laurentii 8.0abc 1 laurentii + iprodione 0.7a

Iprodione (20 μg per ml ^"1 ) 5.9abc

Iprodione (1175 μg per ml ^"1 ) 1.3a

Monilinia fructicola 17.5c Ηealthy cherry fruits (cv. Lambert) were surface-sterilized, washed in running tap water and immersed in the suspension of M. fructicola (1.0 x 10 ⁴ per ml) alone, a mixture of iprodione (20 and 1175 μg ml" ¹ ) and M. fructicola. or a yeast cell suspension (4.0 - 6.0 x 10 ⁷ per ml) and M _» fructicola with a reduced rate (20 μg ml ^"1 ) of iprodione or without it. There were three replicates of 50 fruits each per treatment. After treatments the fruits were allowed to drain for 2 min before storing them in perforated polythene bags. Treatments were randomized and placed in cardboard boxes lined with perforated polythene bags. Disease incidence was recorded after 20 days of incubation at 2.8°C. ^z Within a column, values with the same letters are not significantly different at P = 0.05 according to Duncan's multiple range test.

.44 .

EXAMPLE 5: Population studies of yeasts in pear wounds

The ability of yeast cells to survive and multiply in wounds was studied to determine if these yeasts were effective colonizers of pear wounds. Fruits were wounded, inoculated with yeast cells, and held in plastic boxes at -1, 5, 10 and 20°C Viable cell concentrations of the inoculum were determined by a modified dilution plate frequency technique (Chand et al., Appl. Environ.

Microbiol. 58:3374-3379, 1992). Individual fruits served as one replicate in a randomized complete block design and three replicates were sampled at each sampling time and temperature. For the fruits incubated at 5, 10 and 20°C, the wounds were sampled 0, 1, 2, 3, 4 and 7 days after inoculation, whereas at -1°C, the wound samples were removed 0, 3, 10, 15, 30 and 45 days after inoculation. Samples were taken using a cork borer (10 mm internal diameter) and macerated in distilled sterile water (Roberts, Phytopathol. 80: 1051, 1990). Serial dilutions were plated in duplicate onto YMDA using the modified dilution plate frequency technique (Chand et al., Appl. Environ. Microbiol., 58:3374-3379, 1992) and incubated at 22 ± 2°C for four days when yeast populations were counted and expressed as logio CFU per wound. All yeasts colonized and multiplied in pear wounds at -1, 5, 10 or 20°C (Figures 2 and

3). Yeast populations in wounds increased by approximately 1.3 log units over the initial populations within 10 days at

-1°C storage temperature, then started declining slowly (Figure 2). After 45 days of storage, the population of yeasts in wounds was approximately 1.0 log unit higher than the population at the time of inoculation. At storage temperatures of 5, 10 and 20°C, the maximum yeast populations, approximately 1.7 log units over the initial populations, were reached within two days of storage after inoculation (Figure 3). After reaching the highest population in the wounds, there was faster decline as the storage temperature increased from 5 to 20°C

EXAMPLE 6: Mode of antifungai action of C. laurentii, R. elutinis and C. infirmo-mimatus strains To determine if extracellular metabolites in culture broth were responsible for antagonism of C, laurentii, R. glutinis and C. infirmo-miniatus against P. expansum, yeasts were grown in YMDB at 22 ± 1°C as described above. Yeast cells and culture broth were separated after two and seven days of incubation by centrifugation (8.0 x 10 ³ rpm min ^"1 ). The supernatant was decanted then filter sterilized using a 0.2 mm polycarbonate membrane filter. Yeast pellets were washed twice with 50 ml of SDW (pH 6.8) and a cell suspension was made in SDW (1.5 to 2.0 x 10 ¹ ml ^"1 ). A few microliters of concentrated spore suspension of P. expansum were added into the above supernatant and yeast cell suspensions so that the P. expansum spore concentration in the inoculation mixtures as 5.0 x 10 ³ ml ^"1 . Fifty microliters of inoculation mixture was added into each puncture on surface sterilized, punctured fruits. There were three replications of five fruits per treatment. Treatments were randomized and incubated for 6 to 8 days at 20°C.

- 45 -

All wounds treated with P expansum in water or yeast supernatants developed rots, indicating the absence of antifungai compounds in the culture of yeasts Lesions that developed from wounds treated with a mixture of culture filtrate of the yeasts and P expansum had 43 ± 3 mm diameter whereas the average diameter of lesions treated with P expansum in water was 33 5 ± 3 5 5 mm Wounds treated with double-washed cell suspensions of C laurent , R glutims and C mfirmo-mimatus reduced disease incidence to 45 ± 5, 46 ± 4, 50 ± 6 percent, respectively, and the average diameter of diseased wounds was significantly smaller than controls

To detect the production of antifungai metabolites by these yeasts, they were streaked on to YMDA in petn dishes, and plates were incubated at

10 22 ± 2°C After 2, 4 and 7 days of incubation, a concentrated spore suspension of P expansum (5 0 to 7 0 x 10 ⁷ ml ^"1 ) was atomized onto the plates Plates were incubated at 22 ± 2°C and observed daily for zones of inhibition for seven days

Zones of inhibition did not develop around yeast growth on Petn plates, indicating that no antifungai compounds were produced by any of the three yeasts

15 Because of their rapid growth rate at all temperature and lack of antifungai compounds, it is likely that fungal control was due to nutrient competition However, induced resistance (I e , the yeast or a yeast metabolite activates a resistance mechanism in the treated agncultural commodity), m vivo production of some enzymes (I e , the yeast produces metabolites in wounds of the fruit that are not produced in culture or on an artificial medium), and hyperparasitism

20 (the yeast directly parasitizes the pathogen by attacking spores and mycelia) may also play a role in control of postharvest diseases of pears by these saprophytic yeasts

None of the three yeast strains used grew at 37°C, indicating that they cannot cause infection in the human body

While this invention has been specifically descnbed with reference to Cryptococcus

25 mfirmo-mimatus ssp , particularly Phaff and Fell isolate YY-6, Cryptococcus albidus isolate HRB-2, Cryptococcus laurentii (Kufferath) Skinner isolate HRA-5, Rhodotorula aurantiaca isolate YCL-5, and Rhodotorula gluti s Harnson isolates HRA-3, HRA-4, and HRB-6, it is to be understood that the invention is not so limited Neither is the invention to be limited to the specific fungal pathogens descnbed above

30 All publications and published patent documents cited in this specification are incorporated herein by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference

Yeast strains useful in the practice of the present invention are available from the Mid-Columbia Agncultural Research and Extension Center, Hood River, Oregon Cryptococcus

35 laurentii strain HRA-5, Rhodotorula glutinis strain HRB-6, and Cryptococcus mfirmo-mimatus strain YY-6 have been deposited with the Agncultural Research Service Culture Collection (NRRL),

- 46 -

Peoria, Illinois under accession numbers NRRL Y-21411, NRRL Y-21412, and NRRL Y-21413, respectively.

This invention has been detailed both by example and by direct description. It should be apparent that one having ordinary skill in this art would be able to surmise equivalents to the invention as described in the claims which follow but which would be within the spirit of the description above. Those equivalents are to be included within the scope of this invention.